Upload folder using huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- .gitattributes +4 -0
- logs/distillm_2_qwen2.5_1.5b_it.log +0 -0
- logs/llama3.2_3b_it_nnm_distillm_2_1e.log +0 -0
- logs/nnm_distillm_2_qwen2.5_1.5b.log +1 -0
- qwen2.5-1.5B-it-distillm2/README.md +62 -0
- qwen2.5-1.5B-it-distillm2/added_tokens.json +24 -0
- qwen2.5-1.5B-it-distillm2/all_results.json +18 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1246/added_tokens.json +24 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1246/config.json +29 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1246/generation_config.json +14 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1246/merges.txt +0 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1246/pytorch_model.bin +3 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1246/special_tokens_map.json +32 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1246/tokenizer.json +3 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1246/tokenizer_config.json +207 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1246/trainer_state.json +1408 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1246/training_args.bin +3 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1246/vocab.json +0 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1869/added_tokens.json +24 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1869/config.json +29 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1869/generation_config.json +14 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1869/merges.txt +0 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1869/pytorch_model.bin +3 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1869/special_tokens_map.json +32 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1869/tokenizer.json +3 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1869/tokenizer_config.json +207 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1869/trainer_state.json +2090 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1869/training_args.bin +3 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-1869/vocab.json +0 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-623/added_tokens.json +24 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-623/config.json +29 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-623/generation_config.json +14 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-623/merges.txt +0 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-623/pytorch_model.bin +3 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-623/special_tokens_map.json +32 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-623/tokenizer.json +3 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-623/tokenizer_config.json +207 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-623/trainer_state.json +726 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-623/training_args.bin +3 -0
- qwen2.5-1.5B-it-distillm2/checkpoint-623/vocab.json +0 -0
- qwen2.5-1.5B-it-distillm2/config.json +29 -0
- qwen2.5-1.5B-it-distillm2/eval_results.json +12 -0
- qwen2.5-1.5B-it-distillm2/generation_config.json +14 -0
- qwen2.5-1.5B-it-distillm2/merges.txt +0 -0
- qwen2.5-1.5B-it-distillm2/pytorch_model.bin +3 -0
- qwen2.5-1.5B-it-distillm2/special_tokens_map.json +32 -0
- qwen2.5-1.5B-it-distillm2/tokenizer.json +3 -0
- qwen2.5-1.5B-it-distillm2/tokenizer_config.json +207 -0
- qwen2.5-1.5B-it-distillm2/train_results.json +9 -0
- qwen2.5-1.5B-it-distillm2/trainer_state.json +2099 -0
.gitattributes
CHANGED
|
@@ -49,3 +49,7 @@ llama-3.2-3B-it-distillm2/checkpoint-1869/tokenizer.json filter=lfs diff=lfs mer
|
|
| 49 |
llama-3.2-3B-it-distillm2/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 50 |
qwen2.5-1.5B-it-distillm2-1epoch/checkpoint-2492/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 51 |
qwen2.5-1.5B-it-distillm2-1epoch/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
llama-3.2-3B-it-distillm2/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 50 |
qwen2.5-1.5B-it-distillm2-1epoch/checkpoint-2492/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 51 |
qwen2.5-1.5B-it-distillm2-1epoch/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 52 |
+
qwen2.5-1.5B-it-distillm2/checkpoint-1246/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 53 |
+
qwen2.5-1.5B-it-distillm2/checkpoint-1869/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 54 |
+
qwen2.5-1.5B-it-distillm2/checkpoint-623/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 55 |
+
qwen2.5-1.5B-it-distillm2/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
logs/distillm_2_qwen2.5_1.5b_it.log
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
logs/llama3.2_3b_it_nnm_distillm_2_1e.log
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
logs/nnm_distillm_2_qwen2.5_1.5b.log
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
./scripts/train/qwen2.5-1.5B-it/nnm_distillm_2.sh: line 1: [GPU: command not found
|
qwen2.5-1.5B-it-distillm2/README.md
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: Qwen/Qwen2.5-1.5B-Instruct
|
| 5 |
+
tags:
|
| 6 |
+
- alignment-handbook
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- data/reformatted/distill-qwen2.5-14B-Instruct
|
| 10 |
+
model-index:
|
| 11 |
+
- name: qwen2.5-1.5B-it-distillm2
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# qwen2.5-1.5B-it-distillm2
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the data/reformatted/distill-qwen2.5-14B-Instruct dataset.
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 5e-05
|
| 40 |
+
- train_batch_size: 2
|
| 41 |
+
- eval_batch_size: 2
|
| 42 |
+
- seed: 42
|
| 43 |
+
- distributed_type: multi-GPU
|
| 44 |
+
- num_devices: 2
|
| 45 |
+
- gradient_accumulation_steps: 32
|
| 46 |
+
- total_train_batch_size: 128
|
| 47 |
+
- total_eval_batch_size: 4
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 51 |
+
- num_epochs: 3
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
|
| 57 |
+
### Framework versions
|
| 58 |
+
|
| 59 |
+
- Transformers 4.45.2
|
| 60 |
+
- Pytorch 2.5.1+cu124
|
| 61 |
+
- Datasets 4.8.5
|
| 62 |
+
- Tokenizers 0.20.3
|
qwen2.5-1.5B-it-distillm2/added_tokens.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</tool_call>": 151658,
|
| 3 |
+
"<tool_call>": 151657,
|
| 4 |
+
"<|box_end|>": 151649,
|
| 5 |
+
"<|box_start|>": 151648,
|
| 6 |
+
"<|endoftext|>": 151643,
|
| 7 |
+
"<|file_sep|>": 151664,
|
| 8 |
+
"<|fim_middle|>": 151660,
|
| 9 |
+
"<|fim_pad|>": 151662,
|
| 10 |
+
"<|fim_prefix|>": 151659,
|
| 11 |
+
"<|fim_suffix|>": 151661,
|
| 12 |
+
"<|im_end|>": 151645,
|
| 13 |
+
"<|im_start|>": 151644,
|
| 14 |
+
"<|image_pad|>": 151655,
|
| 15 |
+
"<|object_ref_end|>": 151647,
|
| 16 |
+
"<|object_ref_start|>": 151646,
|
| 17 |
+
"<|quad_end|>": 151651,
|
| 18 |
+
"<|quad_start|>": 151650,
|
| 19 |
+
"<|repo_name|>": 151663,
|
| 20 |
+
"<|video_pad|>": 151656,
|
| 21 |
+
"<|vision_end|>": 151653,
|
| 22 |
+
"<|vision_pad|>": 151654,
|
| 23 |
+
"<|vision_start|>": 151652
|
| 24 |
+
}
|
qwen2.5-1.5B-it-distillm2/all_results.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.999699067108035,
|
| 3 |
+
"eval_logps_logqs/chosen": 0.07430665194988251,
|
| 4 |
+
"eval_logqs/chosen": -1.0447684526443481,
|
| 5 |
+
"eval_logqs/rejected": -1.0230671167373657,
|
| 6 |
+
"eval_logqs_logps/rejected": 0.034393880516290665,
|
| 7 |
+
"eval_loss": 0.17647793889045715,
|
| 8 |
+
"eval_runtime": 1.0546,
|
| 9 |
+
"eval_samples": 10,
|
| 10 |
+
"eval_samples_per_second": 9.482,
|
| 11 |
+
"eval_steps_per_second": 2.845,
|
| 12 |
+
"total_flos": 0.0,
|
| 13 |
+
"train_loss": 0.194109176269121,
|
| 14 |
+
"train_runtime": 23250.2173,
|
| 15 |
+
"train_samples": 79751,
|
| 16 |
+
"train_samples_per_second": 10.29,
|
| 17 |
+
"train_steps_per_second": 0.08
|
| 18 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1246/added_tokens.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</tool_call>": 151658,
|
| 3 |
+
"<tool_call>": 151657,
|
| 4 |
+
"<|box_end|>": 151649,
|
| 5 |
+
"<|box_start|>": 151648,
|
| 6 |
+
"<|endoftext|>": 151643,
|
| 7 |
+
"<|file_sep|>": 151664,
|
| 8 |
+
"<|fim_middle|>": 151660,
|
| 9 |
+
"<|fim_pad|>": 151662,
|
| 10 |
+
"<|fim_prefix|>": 151659,
|
| 11 |
+
"<|fim_suffix|>": 151661,
|
| 12 |
+
"<|im_end|>": 151645,
|
| 13 |
+
"<|im_start|>": 151644,
|
| 14 |
+
"<|image_pad|>": 151655,
|
| 15 |
+
"<|object_ref_end|>": 151647,
|
| 16 |
+
"<|object_ref_start|>": 151646,
|
| 17 |
+
"<|quad_end|>": 151651,
|
| 18 |
+
"<|quad_start|>": 151650,
|
| 19 |
+
"<|repo_name|>": 151663,
|
| 20 |
+
"<|video_pad|>": 151656,
|
| 21 |
+
"<|vision_end|>": 151653,
|
| 22 |
+
"<|vision_pad|>": 151654,
|
| 23 |
+
"<|vision_start|>": 151652
|
| 24 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1246/config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Qwen/Qwen2.5-1.5B-Instruct",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"Qwen2ForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": 151645,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 1536,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 8960,
|
| 13 |
+
"max_position_embeddings": 32768,
|
| 14 |
+
"max_window_layers": 21,
|
| 15 |
+
"model_type": "qwen2",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 28,
|
| 18 |
+
"num_key_value_heads": 2,
|
| 19 |
+
"rms_norm_eps": 1e-06,
|
| 20 |
+
"rope_scaling": null,
|
| 21 |
+
"rope_theta": 1000000.0,
|
| 22 |
+
"sliding_window": null,
|
| 23 |
+
"tie_word_embeddings": true,
|
| 24 |
+
"torch_dtype": "bfloat16",
|
| 25 |
+
"transformers_version": "4.45.2",
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"use_sliding_window": false,
|
| 28 |
+
"vocab_size": 152064
|
| 29 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1246/generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.45.2"
|
| 14 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1246/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qwen2.5-1.5B-it-distillm2/checkpoint-1246/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:422eaf13971cf4a432e2e4e7b1e5fde87893019bfdf69e70e1805173f6af08e7
|
| 3 |
+
size 3087935634
|
qwen2.5-1.5B-it-distillm2/checkpoint-1246/special_tokens_map.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>"
|
| 16 |
+
],
|
| 17 |
+
"bos_token": "<|endoftext|>",
|
| 18 |
+
"eos_token": {
|
| 19 |
+
"content": "<|im_end|>",
|
| 20 |
+
"lstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"rstrip": false,
|
| 23 |
+
"single_word": false
|
| 24 |
+
},
|
| 25 |
+
"pad_token": {
|
| 26 |
+
"content": "<|endoftext|>",
|
| 27 |
+
"lstrip": false,
|
| 28 |
+
"normalized": false,
|
| 29 |
+
"rstrip": false,
|
| 30 |
+
"single_word": false
|
| 31 |
+
}
|
| 32 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1246/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
|
| 3 |
+
size 11421896
|
qwen2.5-1.5B-it-distillm2/checkpoint-1246/tokenizer_config.json
ADDED
|
@@ -0,0 +1,207 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
}
|
| 181 |
+
},
|
| 182 |
+
"additional_special_tokens": [
|
| 183 |
+
"<|im_start|>",
|
| 184 |
+
"<|im_end|>",
|
| 185 |
+
"<|object_ref_start|>",
|
| 186 |
+
"<|object_ref_end|>",
|
| 187 |
+
"<|box_start|>",
|
| 188 |
+
"<|box_end|>",
|
| 189 |
+
"<|quad_start|>",
|
| 190 |
+
"<|quad_end|>",
|
| 191 |
+
"<|vision_start|>",
|
| 192 |
+
"<|vision_end|>",
|
| 193 |
+
"<|vision_pad|>",
|
| 194 |
+
"<|image_pad|>",
|
| 195 |
+
"<|video_pad|>"
|
| 196 |
+
],
|
| 197 |
+
"bos_token": "<|endoftext|>",
|
| 198 |
+
"chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
|
| 199 |
+
"clean_up_tokenization_spaces": false,
|
| 200 |
+
"eos_token": "<|im_end|>",
|
| 201 |
+
"errors": "replace",
|
| 202 |
+
"model_max_length": 2048,
|
| 203 |
+
"pad_token": "<|endoftext|>",
|
| 204 |
+
"split_special_tokens": false,
|
| 205 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 206 |
+
"unk_token": null
|
| 207 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1246/trainer_state.json
ADDED
|
@@ -0,0 +1,1408 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9997993780720233,
|
| 5 |
+
"eval_steps": 10000000,
|
| 6 |
+
"global_step": 1246,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0016049754238138228,
|
| 13 |
+
"grad_norm": 7.15625,
|
| 14 |
+
"learning_rate": 2.6737967914438503e-07,
|
| 15 |
+
"logps_logqs/chosen": 0.057332463562488556,
|
| 16 |
+
"logqs/chosen": -0.752794086933136,
|
| 17 |
+
"logqs/rejected": -0.7512239813804626,
|
| 18 |
+
"logqs_logps/rejected": 0.1236220970749855,
|
| 19 |
+
"loss": 0.4632,
|
| 20 |
+
"step": 1
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.016049754238138227,
|
| 24 |
+
"grad_norm": 5.71875,
|
| 25 |
+
"learning_rate": 2.6737967914438504e-06,
|
| 26 |
+
"logps_logqs/chosen": 0.047926291823387146,
|
| 27 |
+
"logqs/chosen": -0.7439535856246948,
|
| 28 |
+
"logqs/rejected": -0.7035253643989563,
|
| 29 |
+
"logqs_logps/rejected": 0.13991409540176392,
|
| 30 |
+
"loss": 0.4557,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.032099508476276455,
|
| 35 |
+
"grad_norm": 1.9296875,
|
| 36 |
+
"learning_rate": 5.347593582887701e-06,
|
| 37 |
+
"logps_logqs/chosen": 0.03161158040165901,
|
| 38 |
+
"logqs/chosen": -0.6941366195678711,
|
| 39 |
+
"logqs/rejected": -0.6452130079269409,
|
| 40 |
+
"logqs_logps/rejected": 0.1449585109949112,
|
| 41 |
+
"loss": 0.4015,
|
| 42 |
+
"step": 20
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.048149262714414685,
|
| 46 |
+
"grad_norm": 1.8515625,
|
| 47 |
+
"learning_rate": 8.02139037433155e-06,
|
| 48 |
+
"logps_logqs/chosen": 0.03457744047045708,
|
| 49 |
+
"logqs/chosen": -0.6867337226867676,
|
| 50 |
+
"logqs/rejected": -0.6923023462295532,
|
| 51 |
+
"logqs_logps/rejected": 0.12560859322547913,
|
| 52 |
+
"loss": 0.3565,
|
| 53 |
+
"step": 30
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.06419901695255291,
|
| 57 |
+
"grad_norm": 1.234375,
|
| 58 |
+
"learning_rate": 1.0695187165775402e-05,
|
| 59 |
+
"logps_logqs/chosen": 0.05277745798230171,
|
| 60 |
+
"logqs/chosen": -0.7405093908309937,
|
| 61 |
+
"logqs/rejected": -0.7894801497459412,
|
| 62 |
+
"logqs_logps/rejected": 0.0936320573091507,
|
| 63 |
+
"loss": 0.3228,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.08024877119069114,
|
| 68 |
+
"grad_norm": 1.0,
|
| 69 |
+
"learning_rate": 1.3368983957219252e-05,
|
| 70 |
+
"logps_logqs/chosen": 0.04658779874444008,
|
| 71 |
+
"logqs/chosen": -0.7268352508544922,
|
| 72 |
+
"logqs/rejected": -0.7926934361457825,
|
| 73 |
+
"logqs_logps/rejected": 0.09889288991689682,
|
| 74 |
+
"loss": 0.3027,
|
| 75 |
+
"step": 50
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.09629852542882937,
|
| 79 |
+
"grad_norm": 0.92578125,
|
| 80 |
+
"learning_rate": 1.60427807486631e-05,
|
| 81 |
+
"logps_logqs/chosen": 0.04193533584475517,
|
| 82 |
+
"logqs/chosen": -0.7151986956596375,
|
| 83 |
+
"logqs/rejected": -0.7687807083129883,
|
| 84 |
+
"logqs_logps/rejected": 0.09696364402770996,
|
| 85 |
+
"loss": 0.2856,
|
| 86 |
+
"step": 60
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1123482796669676,
|
| 90 |
+
"grad_norm": 1.0625,
|
| 91 |
+
"learning_rate": 1.8716577540106954e-05,
|
| 92 |
+
"logps_logqs/chosen": 0.04730736091732979,
|
| 93 |
+
"logqs/chosen": -0.7385339736938477,
|
| 94 |
+
"logqs/rejected": -0.8247607946395874,
|
| 95 |
+
"logqs_logps/rejected": 0.08417234569787979,
|
| 96 |
+
"loss": 0.285,
|
| 97 |
+
"step": 70
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.12839803390510582,
|
| 101 |
+
"grad_norm": 0.890625,
|
| 102 |
+
"learning_rate": 2.1390374331550803e-05,
|
| 103 |
+
"logps_logqs/chosen": 0.047996845096349716,
|
| 104 |
+
"logqs/chosen": -0.7431933879852295,
|
| 105 |
+
"logqs/rejected": -0.8130408525466919,
|
| 106 |
+
"logqs_logps/rejected": 0.08468352258205414,
|
| 107 |
+
"loss": 0.2797,
|
| 108 |
+
"step": 80
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.14444778814324405,
|
| 112 |
+
"grad_norm": 0.79296875,
|
| 113 |
+
"learning_rate": 2.4064171122994652e-05,
|
| 114 |
+
"logps_logqs/chosen": 0.04765843600034714,
|
| 115 |
+
"logqs/chosen": -0.7173298001289368,
|
| 116 |
+
"logqs/rejected": -0.7709535360336304,
|
| 117 |
+
"logqs_logps/rejected": 0.08819471299648285,
|
| 118 |
+
"loss": 0.2705,
|
| 119 |
+
"step": 90
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.16049754238138228,
|
| 123 |
+
"grad_norm": 0.82421875,
|
| 124 |
+
"learning_rate": 2.6737967914438505e-05,
|
| 125 |
+
"logps_logqs/chosen": 0.040474437177181244,
|
| 126 |
+
"logqs/chosen": -0.710905909538269,
|
| 127 |
+
"logqs/rejected": -0.7959136962890625,
|
| 128 |
+
"logqs_logps/rejected": 0.09612198173999786,
|
| 129 |
+
"loss": 0.2683,
|
| 130 |
+
"step": 100
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.1765472966195205,
|
| 134 |
+
"grad_norm": 0.7109375,
|
| 135 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 136 |
+
"logps_logqs/chosen": 0.020066609606146812,
|
| 137 |
+
"logqs/chosen": -0.6621483564376831,
|
| 138 |
+
"logqs/rejected": -0.7314690351486206,
|
| 139 |
+
"logqs_logps/rejected": 0.11639375984668732,
|
| 140 |
+
"loss": 0.2682,
|
| 141 |
+
"step": 110
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.19259705085765874,
|
| 145 |
+
"grad_norm": 0.9921875,
|
| 146 |
+
"learning_rate": 3.20855614973262e-05,
|
| 147 |
+
"logps_logqs/chosen": 0.017146889120340347,
|
| 148 |
+
"logqs/chosen": -0.6599355340003967,
|
| 149 |
+
"logqs/rejected": -0.7124370336532593,
|
| 150 |
+
"logqs_logps/rejected": 0.12015531212091446,
|
| 151 |
+
"loss": 0.268,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.20864680509579697,
|
| 156 |
+
"grad_norm": 1.78125,
|
| 157 |
+
"learning_rate": 3.4759358288770055e-05,
|
| 158 |
+
"logps_logqs/chosen": 0.0369986966252327,
|
| 159 |
+
"logqs/chosen": -0.7184829115867615,
|
| 160 |
+
"logqs/rejected": -0.8139392137527466,
|
| 161 |
+
"logqs_logps/rejected": 0.09705157577991486,
|
| 162 |
+
"loss": 0.2611,
|
| 163 |
+
"step": 130
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2246965593339352,
|
| 167 |
+
"grad_norm": 0.87890625,
|
| 168 |
+
"learning_rate": 3.743315508021391e-05,
|
| 169 |
+
"logps_logqs/chosen": 0.07482504099607468,
|
| 170 |
+
"logqs/chosen": -0.7948209643363953,
|
| 171 |
+
"logqs/rejected": -0.9149974584579468,
|
| 172 |
+
"logqs_logps/rejected": 0.04995354264974594,
|
| 173 |
+
"loss": 0.2442,
|
| 174 |
+
"step": 140
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.24074631357207343,
|
| 178 |
+
"grad_norm": 0.796875,
|
| 179 |
+
"learning_rate": 4.0106951871657754e-05,
|
| 180 |
+
"logps_logqs/chosen": 0.08310296386480331,
|
| 181 |
+
"logqs/chosen": -0.812109649181366,
|
| 182 |
+
"logqs/rejected": -0.9168623089790344,
|
| 183 |
+
"logqs_logps/rejected": 0.041201137006282806,
|
| 184 |
+
"loss": 0.2403,
|
| 185 |
+
"step": 150
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.25679606781021164,
|
| 189 |
+
"grad_norm": 0.70703125,
|
| 190 |
+
"learning_rate": 4.2780748663101606e-05,
|
| 191 |
+
"logps_logqs/chosen": 0.08498374372720718,
|
| 192 |
+
"logqs/chosen": -0.8101779818534851,
|
| 193 |
+
"logqs/rejected": -0.9302076101303101,
|
| 194 |
+
"logqs_logps/rejected": 0.03671664744615555,
|
| 195 |
+
"loss": 0.2336,
|
| 196 |
+
"step": 160
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.2728458220483499,
|
| 200 |
+
"grad_norm": 0.85546875,
|
| 201 |
+
"learning_rate": 4.545454545454546e-05,
|
| 202 |
+
"logps_logqs/chosen": 0.08320538699626923,
|
| 203 |
+
"logqs/chosen": -0.8246332406997681,
|
| 204 |
+
"logqs/rejected": -0.9219290018081665,
|
| 205 |
+
"logqs_logps/rejected": 0.03701116889715195,
|
| 206 |
+
"loss": 0.2386,
|
| 207 |
+
"step": 170
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.2888955762864881,
|
| 211 |
+
"grad_norm": 0.77734375,
|
| 212 |
+
"learning_rate": 4.8128342245989304e-05,
|
| 213 |
+
"logps_logqs/chosen": 0.08179891854524612,
|
| 214 |
+
"logqs/chosen": -0.787845253944397,
|
| 215 |
+
"logqs/rejected": -0.9030144810676575,
|
| 216 |
+
"logqs_logps/rejected": 0.03777966648340225,
|
| 217 |
+
"loss": 0.2315,
|
| 218 |
+
"step": 180
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.30494533052462636,
|
| 222 |
+
"grad_norm": 0.7734375,
|
| 223 |
+
"learning_rate": 4.9999607536612036e-05,
|
| 224 |
+
"logps_logqs/chosen": 0.08169040083885193,
|
| 225 |
+
"logqs/chosen": -0.8232030868530273,
|
| 226 |
+
"logqs/rejected": -0.9288152456283569,
|
| 227 |
+
"logqs_logps/rejected": 0.039177440106868744,
|
| 228 |
+
"loss": 0.2389,
|
| 229 |
+
"step": 190
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.32099508476276456,
|
| 233 |
+
"grad_norm": 0.72265625,
|
| 234 |
+
"learning_rate": 4.9992630752499945e-05,
|
| 235 |
+
"logps_logqs/chosen": 0.08127471804618835,
|
| 236 |
+
"logqs/chosen": -0.8097829818725586,
|
| 237 |
+
"logqs/rejected": -0.9110867381095886,
|
| 238 |
+
"logqs_logps/rejected": 0.03667169809341431,
|
| 239 |
+
"loss": 0.2279,
|
| 240 |
+
"step": 200
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3370448390009028,
|
| 244 |
+
"grad_norm": 0.71875,
|
| 245 |
+
"learning_rate": 4.997693536122969e-05,
|
| 246 |
+
"logps_logqs/chosen": 0.0788678377866745,
|
| 247 |
+
"logqs/chosen": -0.7994121313095093,
|
| 248 |
+
"logqs/rejected": -0.9303783178329468,
|
| 249 |
+
"logqs_logps/rejected": 0.037957318127155304,
|
| 250 |
+
"loss": 0.2291,
|
| 251 |
+
"step": 210
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.353094593239041,
|
| 255 |
+
"grad_norm": 0.7109375,
|
| 256 |
+
"learning_rate": 4.995252683809324e-05,
|
| 257 |
+
"logps_logqs/chosen": 0.08248866349458694,
|
| 258 |
+
"logqs/chosen": -0.8128089904785156,
|
| 259 |
+
"logqs/rejected": -0.9362344741821289,
|
| 260 |
+
"logqs_logps/rejected": 0.037949927151203156,
|
| 261 |
+
"loss": 0.229,
|
| 262 |
+
"step": 220
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3691443474771793,
|
| 266 |
+
"grad_norm": 0.63671875,
|
| 267 |
+
"learning_rate": 4.9919413697933496e-05,
|
| 268 |
+
"logps_logqs/chosen": 0.0817771628499031,
|
| 269 |
+
"logqs/chosen": -0.7939696311950684,
|
| 270 |
+
"logqs/rejected": -0.9058554768562317,
|
| 271 |
+
"logqs_logps/rejected": 0.04094362258911133,
|
| 272 |
+
"loss": 0.2314,
|
| 273 |
+
"step": 230
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.3851941017153175,
|
| 277 |
+
"grad_norm": 0.71875,
|
| 278 |
+
"learning_rate": 4.987760749217389e-05,
|
| 279 |
+
"logps_logqs/chosen": 0.076979860663414,
|
| 280 |
+
"logqs/chosen": -0.7874332070350647,
|
| 281 |
+
"logqs/rejected": -0.9303415417671204,
|
| 282 |
+
"logqs_logps/rejected": 0.04022833704948425,
|
| 283 |
+
"loss": 0.2244,
|
| 284 |
+
"step": 240
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.4012438559534557,
|
| 288 |
+
"grad_norm": 0.83984375,
|
| 289 |
+
"learning_rate": 4.982712280478875e-05,
|
| 290 |
+
"logps_logqs/chosen": 0.07487889379262924,
|
| 291 |
+
"logqs/chosen": -0.7985113859176636,
|
| 292 |
+
"logqs/rejected": -0.9595896005630493,
|
| 293 |
+
"logqs_logps/rejected": 0.04083302244544029,
|
| 294 |
+
"loss": 0.2277,
|
| 295 |
+
"step": 250
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.41729361019159394,
|
| 299 |
+
"grad_norm": 0.6953125,
|
| 300 |
+
"learning_rate": 4.976797724721567e-05,
|
| 301 |
+
"logps_logqs/chosen": 0.07195514440536499,
|
| 302 |
+
"logqs/chosen": -0.787926971912384,
|
| 303 |
+
"logqs/rejected": -0.897616982460022,
|
| 304 |
+
"logqs_logps/rejected": 0.0396430678665638,
|
| 305 |
+
"loss": 0.2163,
|
| 306 |
+
"step": 260
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.43334336442973215,
|
| 310 |
+
"grad_norm": 0.7734375,
|
| 311 |
+
"learning_rate": 4.9700191452211806e-05,
|
| 312 |
+
"logps_logqs/chosen": 0.07640247046947479,
|
| 313 |
+
"logqs/chosen": -0.7960728406906128,
|
| 314 |
+
"logqs/rejected": -0.928481936454773,
|
| 315 |
+
"logqs_logps/rejected": 0.03928074985742569,
|
| 316 |
+
"loss": 0.2226,
|
| 317 |
+
"step": 270
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.4493931186678704,
|
| 321 |
+
"grad_norm": 0.72265625,
|
| 322 |
+
"learning_rate": 4.9623789066656276e-05,
|
| 323 |
+
"logps_logqs/chosen": 0.07622957229614258,
|
| 324 |
+
"logqs/chosen": -0.8077980279922485,
|
| 325 |
+
"logqs/rejected": -0.9034830927848816,
|
| 326 |
+
"logqs_logps/rejected": 0.046734608709812164,
|
| 327 |
+
"loss": 0.2256,
|
| 328 |
+
"step": 280
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.4654428729060086,
|
| 332 |
+
"grad_norm": 0.78515625,
|
| 333 |
+
"learning_rate": 4.953879674330093e-05,
|
| 334 |
+
"logps_logqs/chosen": 0.07244043052196503,
|
| 335 |
+
"logqs/chosen": -0.7900176644325256,
|
| 336 |
+
"logqs/rejected": -0.891532301902771,
|
| 337 |
+
"logqs_logps/rejected": 0.04387632757425308,
|
| 338 |
+
"loss": 0.217,
|
| 339 |
+
"step": 290
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.48149262714414687,
|
| 343 |
+
"grad_norm": 0.6796875,
|
| 344 |
+
"learning_rate": 4.944524413147263e-05,
|
| 345 |
+
"logps_logqs/chosen": 0.07289232313632965,
|
| 346 |
+
"logqs/chosen": -0.7856311798095703,
|
| 347 |
+
"logqs/rejected": -0.9099383354187012,
|
| 348 |
+
"logqs_logps/rejected": 0.03974480181932449,
|
| 349 |
+
"loss": 0.2173,
|
| 350 |
+
"step": 300
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.49754238138228507,
|
| 354 |
+
"grad_norm": 0.66796875,
|
| 355 |
+
"learning_rate": 4.934316386673022e-05,
|
| 356 |
+
"logps_logqs/chosen": 0.07063151150941849,
|
| 357 |
+
"logqs/chosen": -0.7815223932266235,
|
| 358 |
+
"logqs/rejected": -0.9079385995864868,
|
| 359 |
+
"logqs_logps/rejected": 0.041664548218250275,
|
| 360 |
+
"loss": 0.2126,
|
| 361 |
+
"step": 310
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.5135921356204233,
|
| 365 |
+
"grad_norm": 0.61328125,
|
| 366 |
+
"learning_rate": 4.923259155947964e-05,
|
| 367 |
+
"logps_logqs/chosen": 0.071230448782444,
|
| 368 |
+
"logqs/chosen": -0.8096257448196411,
|
| 369 |
+
"logqs/rejected": -0.9752206802368164,
|
| 370 |
+
"logqs_logps/rejected": 0.03939716890454292,
|
| 371 |
+
"loss": 0.2194,
|
| 372 |
+
"step": 320
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.5296418898585615,
|
| 376 |
+
"grad_norm": 0.6640625,
|
| 377 |
+
"learning_rate": 4.911356578255139e-05,
|
| 378 |
+
"logps_logqs/chosen": 0.07356056571006775,
|
| 379 |
+
"logqs/chosen": -0.7827819585800171,
|
| 380 |
+
"logqs/rejected": -0.9489104151725769,
|
| 381 |
+
"logqs_logps/rejected": 0.040141116827726364,
|
| 382 |
+
"loss": 0.2145,
|
| 383 |
+
"step": 330
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.5456916440966998,
|
| 387 |
+
"grad_norm": 0.7109375,
|
| 388 |
+
"learning_rate": 4.898612805774447e-05,
|
| 389 |
+
"logps_logqs/chosen": 0.07881536334753036,
|
| 390 |
+
"logqs/chosen": -0.8095147013664246,
|
| 391 |
+
"logqs/rejected": -0.9313008189201355,
|
| 392 |
+
"logqs_logps/rejected": 0.03702981770038605,
|
| 393 |
+
"loss": 0.2165,
|
| 394 |
+
"step": 340
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.561741398334838,
|
| 398 |
+
"grad_norm": 0.64453125,
|
| 399 |
+
"learning_rate": 4.885032284134165e-05,
|
| 400 |
+
"logps_logqs/chosen": 0.07632436603307724,
|
| 401 |
+
"logqs/chosen": -0.7979615926742554,
|
| 402 |
+
"logqs/rejected": -0.9368084073066711,
|
| 403 |
+
"logqs_logps/rejected": 0.03654230386018753,
|
| 404 |
+
"loss": 0.2113,
|
| 405 |
+
"step": 350
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.5777911525729762,
|
| 409 |
+
"grad_norm": 0.62109375,
|
| 410 |
+
"learning_rate": 4.8706197508600984e-05,
|
| 411 |
+
"logps_logqs/chosen": 0.07344283908605576,
|
| 412 |
+
"logqs/chosen": -0.8101444244384766,
|
| 413 |
+
"logqs/rejected": -0.9295485615730286,
|
| 414 |
+
"logqs_logps/rejected": 0.03901313990354538,
|
| 415 |
+
"loss": 0.2151,
|
| 416 |
+
"step": 360
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5938409068111145,
|
| 420 |
+
"grad_norm": 0.62890625,
|
| 421 |
+
"learning_rate": 4.855380233722915e-05,
|
| 422 |
+
"logps_logqs/chosen": 0.07421533018350601,
|
| 423 |
+
"logqs/chosen": -0.7826108336448669,
|
| 424 |
+
"logqs/rejected": -0.9229670763015747,
|
| 425 |
+
"logqs_logps/rejected": 0.038314513862133026,
|
| 426 |
+
"loss": 0.2129,
|
| 427 |
+
"step": 370
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.6098906610492527,
|
| 431 |
+
"grad_norm": 0.62890625,
|
| 432 |
+
"learning_rate": 4.839319048984217e-05,
|
| 433 |
+
"logps_logqs/chosen": 0.07273373752832413,
|
| 434 |
+
"logqs/chosen": -0.7798442840576172,
|
| 435 |
+
"logqs/rejected": -0.9371780157089233,
|
| 436 |
+
"logqs_logps/rejected": 0.03713950887322426,
|
| 437 |
+
"loss": 0.2067,
|
| 438 |
+
"step": 380
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.6259404152873909,
|
| 442 |
+
"grad_norm": 0.62890625,
|
| 443 |
+
"learning_rate": 4.822441799541979e-05,
|
| 444 |
+
"logps_logqs/chosen": 0.07329441606998444,
|
| 445 |
+
"logqs/chosen": -0.8102380037307739,
|
| 446 |
+
"logqs/rejected": -0.95549476146698,
|
| 447 |
+
"logqs_logps/rejected": 0.036079905927181244,
|
| 448 |
+
"loss": 0.2158,
|
| 449 |
+
"step": 390
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.6419901695255291,
|
| 453 |
+
"grad_norm": 0.66796875,
|
| 454 |
+
"learning_rate": 4.8047543729759936e-05,
|
| 455 |
+
"logps_logqs/chosen": 0.07404083013534546,
|
| 456 |
+
"logqs/chosen": -0.7847949266433716,
|
| 457 |
+
"logqs/rejected": -0.9250528216362,
|
| 458 |
+
"logqs_logps/rejected": 0.036556728184223175,
|
| 459 |
+
"loss": 0.2105,
|
| 460 |
+
"step": 400
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.6580399237636674,
|
| 464 |
+
"grad_norm": 0.69140625,
|
| 465 |
+
"learning_rate": 4.786262939494007e-05,
|
| 466 |
+
"logps_logqs/chosen": 0.07539906352758408,
|
| 467 |
+
"logqs/chosen": -0.8072575330734253,
|
| 468 |
+
"logqs/rejected": -0.9661371111869812,
|
| 469 |
+
"logqs_logps/rejected": 0.03662776201963425,
|
| 470 |
+
"loss": 0.2154,
|
| 471 |
+
"step": 410
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6740896780018056,
|
| 475 |
+
"grad_norm": 0.66015625,
|
| 476 |
+
"learning_rate": 4.766973949779261e-05,
|
| 477 |
+
"logps_logqs/chosen": 0.0744672566652298,
|
| 478 |
+
"logqs/chosen": -0.787712574005127,
|
| 479 |
+
"logqs/rejected": -0.9350829124450684,
|
| 480 |
+
"logqs_logps/rejected": 0.0353250689804554,
|
| 481 |
+
"loss": 0.2074,
|
| 482 |
+
"step": 420
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.6901394322399438,
|
| 486 |
+
"grad_norm": 0.62890625,
|
| 487 |
+
"learning_rate": 4.746894132740186e-05,
|
| 488 |
+
"logps_logqs/chosen": 0.07364196330308914,
|
| 489 |
+
"logqs/chosen": -0.7813644409179688,
|
| 490 |
+
"logqs/rejected": -0.9208289980888367,
|
| 491 |
+
"logqs_logps/rejected": 0.0356430858373642,
|
| 492 |
+
"loss": 0.2053,
|
| 493 |
+
"step": 430
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.706189186478082,
|
| 497 |
+
"grad_norm": 0.56640625,
|
| 498 |
+
"learning_rate": 4.726030493163044e-05,
|
| 499 |
+
"logps_logqs/chosen": 0.07587061077356339,
|
| 500 |
+
"logqs/chosen": -0.7853146195411682,
|
| 501 |
+
"logqs/rejected": -0.8944910168647766,
|
| 502 |
+
"logqs_logps/rejected": 0.03531279042363167,
|
| 503 |
+
"loss": 0.2078,
|
| 504 |
+
"step": 440
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.7222389407162203,
|
| 508 |
+
"grad_norm": 0.68359375,
|
| 509 |
+
"learning_rate": 4.7043903092683314e-05,
|
| 510 |
+
"logps_logqs/chosen": 0.07814273238182068,
|
| 511 |
+
"logqs/chosen": -0.8133522272109985,
|
| 512 |
+
"logqs/rejected": -0.9508693814277649,
|
| 513 |
+
"logqs_logps/rejected": 0.030778918415308,
|
| 514 |
+
"loss": 0.2104,
|
| 515 |
+
"step": 450
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.7382886949543586,
|
| 519 |
+
"grad_norm": 0.72265625,
|
| 520 |
+
"learning_rate": 4.6819811301717885e-05,
|
| 521 |
+
"logps_logqs/chosen": 0.07763786613941193,
|
| 522 |
+
"logqs/chosen": -0.8061367869377136,
|
| 523 |
+
"logqs/rejected": -0.9445575475692749,
|
| 524 |
+
"logqs_logps/rejected": 0.035373255610466,
|
| 525 |
+
"loss": 0.2084,
|
| 526 |
+
"step": 460
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.7543384491924967,
|
| 530 |
+
"grad_norm": 0.59765625,
|
| 531 |
+
"learning_rate": 4.6588107732509134e-05,
|
| 532 |
+
"logps_logqs/chosen": 0.07325359433889389,
|
| 533 |
+
"logqs/chosen": -0.7970572710037231,
|
| 534 |
+
"logqs/rejected": -0.9243482351303101,
|
| 535 |
+
"logqs_logps/rejected": 0.03834759443998337,
|
| 536 |
+
"loss": 0.2088,
|
| 537 |
+
"step": 470
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.770388203430635,
|
| 541 |
+
"grad_norm": 0.5625,
|
| 542 |
+
"learning_rate": 4.634887321417895e-05,
|
| 543 |
+
"logps_logqs/chosen": 0.0732722282409668,
|
| 544 |
+
"logqs/chosen": -0.7719672918319702,
|
| 545 |
+
"logqs/rejected": -0.9193938970565796,
|
| 546 |
+
"logqs_logps/rejected": 0.03625096380710602,
|
| 547 |
+
"loss": 0.2075,
|
| 548 |
+
"step": 480
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7864379576687732,
|
| 552 |
+
"grad_norm": 0.90234375,
|
| 553 |
+
"learning_rate": 4.6102191202999065e-05,
|
| 554 |
+
"logps_logqs/chosen": 0.07290570437908173,
|
| 555 |
+
"logqs/chosen": -0.8092619180679321,
|
| 556 |
+
"logqs/rejected": -0.9642523527145386,
|
| 557 |
+
"logqs_logps/rejected": 0.03311945125460625,
|
| 558 |
+
"loss": 0.2045,
|
| 559 |
+
"step": 490
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 0.8024877119069114,
|
| 563 |
+
"grad_norm": 0.6015625,
|
| 564 |
+
"learning_rate": 4.5848147753277656e-05,
|
| 565 |
+
"logps_logqs/chosen": 0.07134760171175003,
|
| 566 |
+
"logqs/chosen": -0.7773372530937195,
|
| 567 |
+
"logqs/rejected": -0.9478441476821899,
|
| 568 |
+
"logqs_logps/rejected": 0.03295496851205826,
|
| 569 |
+
"loss": 0.2049,
|
| 570 |
+
"step": 500
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.8185374661450496,
|
| 574 |
+
"grad_norm": 0.5859375,
|
| 575 |
+
"learning_rate": 4.5586831487339485e-05,
|
| 576 |
+
"logps_logqs/chosen": 0.07219503819942474,
|
| 577 |
+
"logqs/chosen": -0.7950411438941956,
|
| 578 |
+
"logqs/rejected": -0.9455002546310425,
|
| 579 |
+
"logqs_logps/rejected": 0.03765324503183365,
|
| 580 |
+
"loss": 0.208,
|
| 581 |
+
"step": 510
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.8345872203831879,
|
| 585 |
+
"grad_norm": 0.62109375,
|
| 586 |
+
"learning_rate": 4.531833356461027e-05,
|
| 587 |
+
"logps_logqs/chosen": 0.06849464029073715,
|
| 588 |
+
"logqs/chosen": -0.7819440364837646,
|
| 589 |
+
"logqs/rejected": -0.9597524404525757,
|
| 590 |
+
"logqs_logps/rejected": 0.04056422412395477,
|
| 591 |
+
"loss": 0.2084,
|
| 592 |
+
"step": 520
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.8506369746213261,
|
| 596 |
+
"grad_norm": 0.6171875,
|
| 597 |
+
"learning_rate": 4.5042747649816006e-05,
|
| 598 |
+
"logps_logqs/chosen": 0.07160626351833344,
|
| 599 |
+
"logqs/chosen": -0.7721427083015442,
|
| 600 |
+
"logqs/rejected": -0.94866544008255,
|
| 601 |
+
"logqs_logps/rejected": 0.03749927878379822,
|
| 602 |
+
"loss": 0.2068,
|
| 603 |
+
"step": 530
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.8666867288594643,
|
| 607 |
+
"grad_norm": 0.5546875,
|
| 608 |
+
"learning_rate": 4.476016988030826e-05,
|
| 609 |
+
"logps_logqs/chosen": 0.07214485853910446,
|
| 610 |
+
"logqs/chosen": -0.7874671816825867,
|
| 611 |
+
"logqs/rejected": -0.9504098892211914,
|
| 612 |
+
"logqs_logps/rejected": 0.03507527709007263,
|
| 613 |
+
"loss": 0.2045,
|
| 614 |
+
"step": 540
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.8827364830976026,
|
| 618 |
+
"grad_norm": 0.54296875,
|
| 619 |
+
"learning_rate": 4.447069883252696e-05,
|
| 620 |
+
"logps_logqs/chosen": 0.07395409047603607,
|
| 621 |
+
"logqs/chosen": -0.771978497505188,
|
| 622 |
+
"logqs/rejected": -0.9129235148429871,
|
| 623 |
+
"logqs_logps/rejected": 0.039593033492565155,
|
| 624 |
+
"loss": 0.205,
|
| 625 |
+
"step": 550
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.8987862373357408,
|
| 629 |
+
"grad_norm": 0.61328125,
|
| 630 |
+
"learning_rate": 4.417443548761227e-05,
|
| 631 |
+
"logps_logqs/chosen": 0.07299650460481644,
|
| 632 |
+
"logqs/chosen": -0.7979342341423035,
|
| 633 |
+
"logqs/rejected": -0.9357426762580872,
|
| 634 |
+
"logqs_logps/rejected": 0.03937726467847824,
|
| 635 |
+
"loss": 0.2083,
|
| 636 |
+
"step": 560
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.9148359915738791,
|
| 640 |
+
"grad_norm": 0.57421875,
|
| 641 |
+
"learning_rate": 4.387148319617763e-05,
|
| 642 |
+
"logps_logqs/chosen": 0.06836996972560883,
|
| 643 |
+
"logqs/chosen": -0.7659951448440552,
|
| 644 |
+
"logqs/rejected": -0.8732549548149109,
|
| 645 |
+
"logqs_logps/rejected": 0.039081670343875885,
|
| 646 |
+
"loss": 0.203,
|
| 647 |
+
"step": 570
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.9308857458120172,
|
| 651 |
+
"grad_norm": 0.57421875,
|
| 652 |
+
"learning_rate": 4.356194764225618e-05,
|
| 653 |
+
"logps_logqs/chosen": 0.07397963851690292,
|
| 654 |
+
"logqs/chosen": -0.7891489863395691,
|
| 655 |
+
"logqs/rejected": -0.9294188618659973,
|
| 656 |
+
"logqs_logps/rejected": 0.03416060656309128,
|
| 657 |
+
"loss": 0.2024,
|
| 658 |
+
"step": 580
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.9469355000501555,
|
| 662 |
+
"grad_norm": 0.703125,
|
| 663 |
+
"learning_rate": 4.3245936806433205e-05,
|
| 664 |
+
"logps_logqs/chosen": 0.07466734945774078,
|
| 665 |
+
"logqs/chosen": -0.7958794832229614,
|
| 666 |
+
"logqs/rejected": -0.953458309173584,
|
| 667 |
+
"logqs_logps/rejected": 0.035235695540905,
|
| 668 |
+
"loss": 0.206,
|
| 669 |
+
"step": 590
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.9629852542882937,
|
| 673 |
+
"grad_norm": 0.5859375,
|
| 674 |
+
"learning_rate": 4.292356092817739e-05,
|
| 675 |
+
"logps_logqs/chosen": 0.07136549055576324,
|
| 676 |
+
"logqs/chosen": -0.7849777936935425,
|
| 677 |
+
"logqs/rejected": -0.9327009320259094,
|
| 678 |
+
"logqs_logps/rejected": 0.038332488387823105,
|
| 679 |
+
"loss": 0.2041,
|
| 680 |
+
"step": 600
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.979035008526432,
|
| 684 |
+
"grad_norm": 0.55859375,
|
| 685 |
+
"learning_rate": 4.259493246738409e-05,
|
| 686 |
+
"logps_logqs/chosen": 0.06976237148046494,
|
| 687 |
+
"logqs/chosen": -0.7658575773239136,
|
| 688 |
+
"logqs/rejected": -0.9027583003044128,
|
| 689 |
+
"logqs_logps/rejected": 0.03464614599943161,
|
| 690 |
+
"loss": 0.1993,
|
| 691 |
+
"step": 610
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.9950847627645701,
|
| 695 |
+
"grad_norm": 0.55078125,
|
| 696 |
+
"learning_rate": 4.226016606514411e-05,
|
| 697 |
+
"logps_logqs/chosen": 0.0706130638718605,
|
| 698 |
+
"logqs/chosen": -0.7847878932952881,
|
| 699 |
+
"logqs/rejected": -0.9509286880493164,
|
| 700 |
+
"logqs_logps/rejected": 0.03653167933225632,
|
| 701 |
+
"loss": 0.2035,
|
| 702 |
+
"step": 620
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.0111345170027084,
|
| 706 |
+
"grad_norm": 0.53515625,
|
| 707 |
+
"learning_rate": 4.191937850375136e-05,
|
| 708 |
+
"logps_logqs/chosen": 0.06913185119628906,
|
| 709 |
+
"logqs/chosen": -0.7918249368667603,
|
| 710 |
+
"logqs/rejected": -0.9458308219909668,
|
| 711 |
+
"logqs_logps/rejected": 0.030862990766763687,
|
| 712 |
+
"loss": 0.1886,
|
| 713 |
+
"step": 630
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.0271842712408465,
|
| 717 |
+
"grad_norm": 0.5625,
|
| 718 |
+
"learning_rate": 4.157268866596381e-05,
|
| 719 |
+
"logps_logqs/chosen": 0.0690259039402008,
|
| 720 |
+
"logqs/chosen": -0.7895857095718384,
|
| 721 |
+
"logqs/rejected": -0.9810531735420227,
|
| 722 |
+
"logqs_logps/rejected": 0.03111192025244236,
|
| 723 |
+
"loss": 0.187,
|
| 724 |
+
"step": 640
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.043234025478985,
|
| 728 |
+
"grad_norm": 0.53125,
|
| 729 |
+
"learning_rate": 4.1220217493531494e-05,
|
| 730 |
+
"logps_logqs/chosen": 0.06900795549154282,
|
| 731 |
+
"logqs/chosen": -0.7902609705924988,
|
| 732 |
+
"logqs/rejected": -0.9747546911239624,
|
| 733 |
+
"logqs_logps/rejected": 0.031710296869277954,
|
| 734 |
+
"loss": 0.1845,
|
| 735 |
+
"step": 650
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 1.059283779717123,
|
| 739 |
+
"grad_norm": 0.515625,
|
| 740 |
+
"learning_rate": 4.086208794500637e-05,
|
| 741 |
+
"logps_logqs/chosen": 0.07102300226688385,
|
| 742 |
+
"logqs/chosen": -0.7767165899276733,
|
| 743 |
+
"logqs/rejected": -0.9414197206497192,
|
| 744 |
+
"logqs_logps/rejected": 0.030909577384591103,
|
| 745 |
+
"loss": 0.1807,
|
| 746 |
+
"step": 660
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.0753335339552612,
|
| 750 |
+
"grad_norm": 0.494140625,
|
| 751 |
+
"learning_rate": 4.049842495284858e-05,
|
| 752 |
+
"logps_logqs/chosen": 0.06985093653202057,
|
| 753 |
+
"logqs/chosen": -0.7880310416221619,
|
| 754 |
+
"logqs/rejected": -0.9546613693237305,
|
| 755 |
+
"logqs_logps/rejected": 0.030261676758527756,
|
| 756 |
+
"loss": 0.186,
|
| 757 |
+
"step": 670
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 1.0913832881933996,
|
| 761 |
+
"grad_norm": 0.50390625,
|
| 762 |
+
"learning_rate": 4.012935537984414e-05,
|
| 763 |
+
"logps_logqs/chosen": 0.07058895379304886,
|
| 764 |
+
"logqs/chosen": -0.7935397624969482,
|
| 765 |
+
"logqs/rejected": -1.0117493867874146,
|
| 766 |
+
"logqs_logps/rejected": 0.029221097007393837,
|
| 767 |
+
"loss": 0.1835,
|
| 768 |
+
"step": 680
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.1074330424315377,
|
| 772 |
+
"grad_norm": 0.6171875,
|
| 773 |
+
"learning_rate": 3.9755007974849135e-05,
|
| 774 |
+
"logps_logqs/chosen": 0.06916572153568268,
|
| 775 |
+
"logqs/chosen": -0.7878638505935669,
|
| 776 |
+
"logqs/rejected": -0.9999720454216003,
|
| 777 |
+
"logqs_logps/rejected": 0.030744653195142746,
|
| 778 |
+
"loss": 0.1857,
|
| 779 |
+
"step": 690
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.123482796669676,
|
| 783 |
+
"grad_norm": 0.57421875,
|
| 784 |
+
"learning_rate": 3.93755133278762e-05,
|
| 785 |
+
"logps_logqs/chosen": 0.07014169543981552,
|
| 786 |
+
"logqs/chosen": -0.7882756590843201,
|
| 787 |
+
"logqs/rejected": -1.0001386404037476,
|
| 788 |
+
"logqs_logps/rejected": 0.02842717245221138,
|
| 789 |
+
"loss": 0.1832,
|
| 790 |
+
"step": 700
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.1395325509078142,
|
| 794 |
+
"grad_norm": 0.59375,
|
| 795 |
+
"learning_rate": 3.899100382453845e-05,
|
| 796 |
+
"logps_logqs/chosen": 0.0701603814959526,
|
| 797 |
+
"logqs/chosen": -0.7642520070075989,
|
| 798 |
+
"logqs/rejected": -0.9544415473937988,
|
| 799 |
+
"logqs_logps/rejected": 0.027573522180318832,
|
| 800 |
+
"loss": 0.1806,
|
| 801 |
+
"step": 710
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.1555823051459524,
|
| 805 |
+
"grad_norm": 0.50390625,
|
| 806 |
+
"learning_rate": 3.8601613599867156e-05,
|
| 807 |
+
"logps_logqs/chosen": 0.06890274584293365,
|
| 808 |
+
"logqs/chosen": -0.7708860635757446,
|
| 809 |
+
"logqs/rejected": -0.9537287950515747,
|
| 810 |
+
"logqs_logps/rejected": 0.028619807213544846,
|
| 811 |
+
"loss": 0.1804,
|
| 812 |
+
"step": 720
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 1.1716320593840908,
|
| 816 |
+
"grad_norm": 0.609375,
|
| 817 |
+
"learning_rate": 3.8207478491519216e-05,
|
| 818 |
+
"logps_logqs/chosen": 0.0682586207985878,
|
| 819 |
+
"logqs/chosen": -0.7696245908737183,
|
| 820 |
+
"logqs/rejected": -0.9157883524894714,
|
| 821 |
+
"logqs_logps/rejected": 0.028189942240715027,
|
| 822 |
+
"loss": 0.1814,
|
| 823 |
+
"step": 730
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 1.187681813622229,
|
| 827 |
+
"grad_norm": 0.59375,
|
| 828 |
+
"learning_rate": 3.780873599239044e-05,
|
| 829 |
+
"logps_logqs/chosen": 0.07058210670948029,
|
| 830 |
+
"logqs/chosen": -0.801421046257019,
|
| 831 |
+
"logqs/rejected": -0.9591751098632812,
|
| 832 |
+
"logqs_logps/rejected": 0.03063536249101162,
|
| 833 |
+
"loss": 0.1853,
|
| 834 |
+
"step": 740
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.203731567860367,
|
| 838 |
+
"grad_norm": 0.55859375,
|
| 839 |
+
"learning_rate": 3.740552520265167e-05,
|
| 840 |
+
"logps_logqs/chosen": 0.06886611133813858,
|
| 841 |
+
"logqs/chosen": -0.792157769203186,
|
| 842 |
+
"logqs/rejected": -0.9370824694633484,
|
| 843 |
+
"logqs_logps/rejected": 0.02984955906867981,
|
| 844 |
+
"loss": 0.1833,
|
| 845 |
+
"step": 750
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 1.2197813220985054,
|
| 849 |
+
"grad_norm": 0.55859375,
|
| 850 |
+
"learning_rate": 3.699798678122403e-05,
|
| 851 |
+
"logps_logqs/chosen": 0.07179007679224014,
|
| 852 |
+
"logqs/chosen": -0.7622597813606262,
|
| 853 |
+
"logqs/rejected": -0.9382703900337219,
|
| 854 |
+
"logqs_logps/rejected": 0.029157549142837524,
|
| 855 |
+
"loss": 0.1793,
|
| 856 |
+
"step": 760
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.2358310763366436,
|
| 860 |
+
"grad_norm": 0.546875,
|
| 861 |
+
"learning_rate": 3.6586262896710476e-05,
|
| 862 |
+
"logps_logqs/chosen": 0.06790535151958466,
|
| 863 |
+
"logqs/chosen": -0.7676440477371216,
|
| 864 |
+
"logqs/rejected": -0.928793728351593,
|
| 865 |
+
"logqs_logps/rejected": 0.03030979633331299,
|
| 866 |
+
"loss": 0.1808,
|
| 867 |
+
"step": 770
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 1.2518808305747817,
|
| 871 |
+
"grad_norm": 0.51171875,
|
| 872 |
+
"learning_rate": 3.61704971778007e-05,
|
| 873 |
+
"logps_logqs/chosen": 0.06736615300178528,
|
| 874 |
+
"logqs/chosen": -0.770586371421814,
|
| 875 |
+
"logqs/rejected": -0.9197053909301758,
|
| 876 |
+
"logqs_logps/rejected": 0.03245236724615097,
|
| 877 |
+
"loss": 0.1799,
|
| 878 |
+
"step": 780
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.26793058481292,
|
| 882 |
+
"grad_norm": 0.4765625,
|
| 883 |
+
"learning_rate": 3.575083466316664e-05,
|
| 884 |
+
"logps_logqs/chosen": 0.06956754624843597,
|
| 885 |
+
"logqs/chosen": -0.7774965167045593,
|
| 886 |
+
"logqs/rejected": -0.9702298045158386,
|
| 887 |
+
"logqs_logps/rejected": 0.030748773366212845,
|
| 888 |
+
"loss": 0.1831,
|
| 889 |
+
"step": 790
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 1.2839803390510582,
|
| 893 |
+
"grad_norm": 0.5078125,
|
| 894 |
+
"learning_rate": 3.532742175086621e-05,
|
| 895 |
+
"logps_logqs/chosen": 0.06920956075191498,
|
| 896 |
+
"logqs/chosen": -0.7709556818008423,
|
| 897 |
+
"logqs/rejected": -0.9387216567993164,
|
| 898 |
+
"logqs_logps/rejected": 0.027966167777776718,
|
| 899 |
+
"loss": 0.1788,
|
| 900 |
+
"step": 800
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 1.3000300932891964,
|
| 904 |
+
"grad_norm": 0.54296875,
|
| 905 |
+
"learning_rate": 3.490040614727272e-05,
|
| 906 |
+
"logps_logqs/chosen": 0.06927359104156494,
|
| 907 |
+
"logqs/chosen": -0.7772814035415649,
|
| 908 |
+
"logqs/rejected": -0.9691821932792664,
|
| 909 |
+
"logqs_logps/rejected": 0.028416061773896217,
|
| 910 |
+
"loss": 0.1781,
|
| 911 |
+
"step": 810
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 1.3160798475273348,
|
| 915 |
+
"grad_norm": 0.5078125,
|
| 916 |
+
"learning_rate": 3.446993681554797e-05,
|
| 917 |
+
"logps_logqs/chosen": 0.07202474772930145,
|
| 918 |
+
"logqs/chosen": -0.7855108380317688,
|
| 919 |
+
"logqs/rejected": -0.9708096385002136,
|
| 920 |
+
"logqs_logps/rejected": 0.030036652460694313,
|
| 921 |
+
"loss": 0.1793,
|
| 922 |
+
"step": 820
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 1.332129601765473,
|
| 926 |
+
"grad_norm": 0.65234375,
|
| 927 |
+
"learning_rate": 3.403616392367681e-05,
|
| 928 |
+
"logps_logqs/chosen": 0.07054628431797028,
|
| 929 |
+
"logqs/chosen": -0.7540086507797241,
|
| 930 |
+
"logqs/rejected": -0.9042137265205383,
|
| 931 |
+
"logqs_logps/rejected": 0.03296629339456558,
|
| 932 |
+
"loss": 0.1818,
|
| 933 |
+
"step": 830
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.3481793560036113,
|
| 937 |
+
"grad_norm": 0.5078125,
|
| 938 |
+
"learning_rate": 3.3599238792081575e-05,
|
| 939 |
+
"logps_logqs/chosen": 0.07161605358123779,
|
| 940 |
+
"logqs/chosen": -0.7662056684494019,
|
| 941 |
+
"logqs/rejected": -0.9423080682754517,
|
| 942 |
+
"logqs_logps/rejected": 0.031152984127402306,
|
| 943 |
+
"loss": 0.1786,
|
| 944 |
+
"step": 840
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 1.3642291102417494,
|
| 948 |
+
"grad_norm": 0.5390625,
|
| 949 |
+
"learning_rate": 3.315931384083431e-05,
|
| 950 |
+
"logps_logqs/chosen": 0.0672772079706192,
|
| 951 |
+
"logqs/chosen": -0.7698653936386108,
|
| 952 |
+
"logqs/rejected": -0.9608929753303528,
|
| 953 |
+
"logqs_logps/rejected": 0.031727343797683716,
|
| 954 |
+
"loss": 0.181,
|
| 955 |
+
"step": 850
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.3802788644798878,
|
| 959 |
+
"grad_norm": 0.5390625,
|
| 960 |
+
"learning_rate": 3.2716542536485474e-05,
|
| 961 |
+
"logps_logqs/chosen": 0.07064563035964966,
|
| 962 |
+
"logqs/chosen": -0.7671880125999451,
|
| 963 |
+
"logqs/rejected": -0.9258206486701965,
|
| 964 |
+
"logqs_logps/rejected": 0.03185782581567764,
|
| 965 |
+
"loss": 0.1802,
|
| 966 |
+
"step": 860
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 1.396328618718026,
|
| 970 |
+
"grad_norm": 0.59375,
|
| 971 |
+
"learning_rate": 3.2271079338527626e-05,
|
| 972 |
+
"logps_logqs/chosen": 0.07001273334026337,
|
| 973 |
+
"logqs/chosen": -0.7942059636116028,
|
| 974 |
+
"logqs/rejected": -0.9803145527839661,
|
| 975 |
+
"logqs_logps/rejected": 0.02900245226919651,
|
| 976 |
+
"loss": 0.1783,
|
| 977 |
+
"step": 870
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 1.412378372956164,
|
| 981 |
+
"grad_norm": 0.48828125,
|
| 982 |
+
"learning_rate": 3.1823079645512655e-05,
|
| 983 |
+
"logps_logqs/chosen": 0.06748739629983902,
|
| 984 |
+
"logqs/chosen": -0.7796913385391235,
|
| 985 |
+
"logqs/rejected": -1.0028189420700073,
|
| 986 |
+
"logqs_logps/rejected": 0.028837621212005615,
|
| 987 |
+
"loss": 0.1791,
|
| 988 |
+
"step": 880
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 1.4284281271943025,
|
| 992 |
+
"grad_norm": 0.5234375,
|
| 993 |
+
"learning_rate": 3.137269974084139e-05,
|
| 994 |
+
"logps_logqs/chosen": 0.07031874358654022,
|
| 995 |
+
"logqs/chosen": -0.7788208723068237,
|
| 996 |
+
"logqs/rejected": -0.9545127153396606,
|
| 997 |
+
"logqs_logps/rejected": 0.029628973454236984,
|
| 998 |
+
"loss": 0.179,
|
| 999 |
+
"step": 890
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 1.4444778814324406,
|
| 1003 |
+
"grad_norm": 0.486328125,
|
| 1004 |
+
"learning_rate": 3.092009673824469e-05,
|
| 1005 |
+
"logps_logqs/chosen": 0.07582792639732361,
|
| 1006 |
+
"logqs/chosen": -0.8078464269638062,
|
| 1007 |
+
"logqs/rejected": -0.9633975028991699,
|
| 1008 |
+
"logqs_logps/rejected": 0.02939186617732048,
|
| 1009 |
+
"loss": 0.1885,
|
| 1010 |
+
"step": 900
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.4605276356705787,
|
| 1014 |
+
"grad_norm": 0.53515625,
|
| 1015 |
+
"learning_rate": 3.0465428526974665e-05,
|
| 1016 |
+
"logps_logqs/chosen": 0.07225798070430756,
|
| 1017 |
+
"logqs/chosen": -0.7663235068321228,
|
| 1018 |
+
"logqs/rejected": -0.9817646741867065,
|
| 1019 |
+
"logqs_logps/rejected": 0.028154581785202026,
|
| 1020 |
+
"loss": 0.178,
|
| 1021 |
+
"step": 910
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 1.4765773899087171,
|
| 1025 |
+
"grad_norm": 0.49609375,
|
| 1026 |
+
"learning_rate": 3.000885371672554e-05,
|
| 1027 |
+
"logps_logqs/chosen": 0.07049473375082016,
|
| 1028 |
+
"logqs/chosen": -0.7796770930290222,
|
| 1029 |
+
"logqs/rejected": -0.9203440546989441,
|
| 1030 |
+
"logqs_logps/rejected": 0.028858328238129616,
|
| 1031 |
+
"loss": 0.1765,
|
| 1032 |
+
"step": 920
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.4926271441468553,
|
| 1036 |
+
"grad_norm": 0.53125,
|
| 1037 |
+
"learning_rate": 2.9550531582303082e-05,
|
| 1038 |
+
"logps_logqs/chosen": 0.07375530898571014,
|
| 1039 |
+
"logqs/chosen": -0.7871274352073669,
|
| 1040 |
+
"logqs/rejected": -0.950308620929718,
|
| 1041 |
+
"logqs_logps/rejected": 0.030121903866529465,
|
| 1042 |
+
"loss": 0.1821,
|
| 1043 |
+
"step": 930
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 1.5086768983849934,
|
| 1047 |
+
"grad_norm": 0.462890625,
|
| 1048 |
+
"learning_rate": 2.909062200806208e-05,
|
| 1049 |
+
"logps_logqs/chosen": 0.06755580753087997,
|
| 1050 |
+
"logqs/chosen": -0.7797432541847229,
|
| 1051 |
+
"logqs/rejected": -0.9579198956489563,
|
| 1052 |
+
"logqs_logps/rejected": 0.027673590928316116,
|
| 1053 |
+
"loss": 0.1738,
|
| 1054 |
+
"step": 940
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 1.5247266526231318,
|
| 1058 |
+
"grad_norm": 0.451171875,
|
| 1059 |
+
"learning_rate": 2.8629285432131083e-05,
|
| 1060 |
+
"logps_logqs/chosen": 0.06972555816173553,
|
| 1061 |
+
"logqs/chosen": -0.7843119502067566,
|
| 1062 |
+
"logqs/rejected": -0.971518874168396,
|
| 1063 |
+
"logqs_logps/rejected": 0.026922276243567467,
|
| 1064 |
+
"loss": 0.1771,
|
| 1065 |
+
"step": 950
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 1.54077640686127,
|
| 1069 |
+
"grad_norm": 0.5234375,
|
| 1070 |
+
"learning_rate": 2.8166682790444116e-05,
|
| 1071 |
+
"logps_logqs/chosen": 0.0729864090681076,
|
| 1072 |
+
"logqs/chosen": -0.7914843559265137,
|
| 1073 |
+
"logqs/rejected": -0.9729417562484741,
|
| 1074 |
+
"logqs_logps/rejected": 0.02641429379582405,
|
| 1075 |
+
"loss": 0.1788,
|
| 1076 |
+
"step": 960
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.556826161099408,
|
| 1080 |
+
"grad_norm": 0.4921875,
|
| 1081 |
+
"learning_rate": 2.7702975460598547e-05,
|
| 1082 |
+
"logps_logqs/chosen": 0.06970744580030441,
|
| 1083 |
+
"logqs/chosen": -0.7867680788040161,
|
| 1084 |
+
"logqs/rejected": -0.9702849388122559,
|
| 1085 |
+
"logqs_logps/rejected": 0.026300692930817604,
|
| 1086 |
+
"loss": 0.1753,
|
| 1087 |
+
"step": 970
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.5728759153375464,
|
| 1091 |
+
"grad_norm": 0.51171875,
|
| 1092 |
+
"learning_rate": 2.723832520555905e-05,
|
| 1093 |
+
"logps_logqs/chosen": 0.06844428926706314,
|
| 1094 |
+
"logqs/chosen": -0.7796690464019775,
|
| 1095 |
+
"logqs/rejected": -0.9721530079841614,
|
| 1096 |
+
"logqs_logps/rejected": 0.03003394976258278,
|
| 1097 |
+
"loss": 0.1775,
|
| 1098 |
+
"step": 980
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 1.5889256695756846,
|
| 1102 |
+
"grad_norm": 0.50390625,
|
| 1103 |
+
"learning_rate": 2.677289411722702e-05,
|
| 1104 |
+
"logps_logqs/chosen": 0.06915868073701859,
|
| 1105 |
+
"logqs/chosen": -0.7836161851882935,
|
| 1106 |
+
"logqs/rejected": -1.0066392421722412,
|
| 1107 |
+
"logqs_logps/rejected": 0.029805105179548264,
|
| 1108 |
+
"loss": 0.1834,
|
| 1109 |
+
"step": 990
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.6049754238138227,
|
| 1113 |
+
"grad_norm": 0.52734375,
|
| 1114 |
+
"learning_rate": 2.63068445598953e-05,
|
| 1115 |
+
"logps_logqs/chosen": 0.07069100439548492,
|
| 1116 |
+
"logqs/chosen": -0.7810501456260681,
|
| 1117 |
+
"logqs/rejected": -0.9559534788131714,
|
| 1118 |
+
"logqs_logps/rejected": 0.02675846591591835,
|
| 1119 |
+
"loss": 0.1799,
|
| 1120 |
+
"step": 1000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 1.6210251780519611,
|
| 1124 |
+
"grad_norm": 0.455078125,
|
| 1125 |
+
"learning_rate": 2.5840339113607854e-05,
|
| 1126 |
+
"logps_logqs/chosen": 0.06957536935806274,
|
| 1127 |
+
"logqs/chosen": -0.776276171207428,
|
| 1128 |
+
"logqs/rejected": -0.9653003811836243,
|
| 1129 |
+
"logqs_logps/rejected": 0.029132988303899765,
|
| 1130 |
+
"loss": 0.1764,
|
| 1131 |
+
"step": 1010
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 1.6370749322900993,
|
| 1135 |
+
"grad_norm": 0.4921875,
|
| 1136 |
+
"learning_rate": 2.53735405174442e-05,
|
| 1137 |
+
"logps_logqs/chosen": 0.06984798610210419,
|
| 1138 |
+
"logqs/chosen": -0.7904434204101562,
|
| 1139 |
+
"logqs/rejected": -0.9736318588256836,
|
| 1140 |
+
"logqs_logps/rejected": 0.02975938282907009,
|
| 1141 |
+
"loss": 0.1778,
|
| 1142 |
+
"step": 1020
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 1.6531246865282374,
|
| 1146 |
+
"grad_norm": 0.50390625,
|
| 1147 |
+
"learning_rate": 2.490661161274835e-05,
|
| 1148 |
+
"logps_logqs/chosen": 0.06985798478126526,
|
| 1149 |
+
"logqs/chosen": -0.7864385843276978,
|
| 1150 |
+
"logqs/rejected": -0.9617294073104858,
|
| 1151 |
+
"logqs_logps/rejected": 0.027367640286684036,
|
| 1152 |
+
"loss": 0.1774,
|
| 1153 |
+
"step": 1030
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 1.6691744407663758,
|
| 1157 |
+
"grad_norm": 0.4453125,
|
| 1158 |
+
"learning_rate": 2.443971528632205e-05,
|
| 1159 |
+
"logps_logqs/chosen": 0.07047738134860992,
|
| 1160 |
+
"logqs/chosen": -0.7980072498321533,
|
| 1161 |
+
"logqs/rejected": -0.9655786752700806,
|
| 1162 |
+
"logqs_logps/rejected": 0.028297554701566696,
|
| 1163 |
+
"loss": 0.1779,
|
| 1164 |
+
"step": 1040
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.6852241950045141,
|
| 1168 |
+
"grad_norm": 0.515625,
|
| 1169 |
+
"learning_rate": 2.3973014413602238e-05,
|
| 1170 |
+
"logps_logqs/chosen": 0.07119600474834442,
|
| 1171 |
+
"logqs/chosen": -0.7848029732704163,
|
| 1172 |
+
"logqs/rejected": -1.003758192062378,
|
| 1173 |
+
"logqs_logps/rejected": 0.029630709439516068,
|
| 1174 |
+
"loss": 0.1808,
|
| 1175 |
+
"step": 1050
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 1.701273949242652,
|
| 1179 |
+
"grad_norm": 0.498046875,
|
| 1180 |
+
"learning_rate": 2.3506671801842364e-05,
|
| 1181 |
+
"logps_logqs/chosen": 0.06773122400045395,
|
| 1182 |
+
"logqs/chosen": -0.8044145703315735,
|
| 1183 |
+
"logqs/rejected": -0.9743335843086243,
|
| 1184 |
+
"logqs_logps/rejected": 0.027200300246477127,
|
| 1185 |
+
"loss": 0.1769,
|
| 1186 |
+
"step": 1060
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.7173237034807904,
|
| 1190 |
+
"grad_norm": 0.4921875,
|
| 1191 |
+
"learning_rate": 2.3040850133317597e-05,
|
| 1192 |
+
"logps_logqs/chosen": 0.07197652757167816,
|
| 1193 |
+
"logqs/chosen": -0.7732763886451721,
|
| 1194 |
+
"logqs/rejected": -0.9615533947944641,
|
| 1195 |
+
"logqs_logps/rejected": 0.02922072447836399,
|
| 1196 |
+
"loss": 0.1801,
|
| 1197 |
+
"step": 1070
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 1.7333734577189288,
|
| 1201 |
+
"grad_norm": 0.48046875,
|
| 1202 |
+
"learning_rate": 2.2575711908573548e-05,
|
| 1203 |
+
"logps_logqs/chosen": 0.06941990554332733,
|
| 1204 |
+
"logqs/chosen": -0.7937701940536499,
|
| 1205 |
+
"logqs/rejected": -1.0073679685592651,
|
| 1206 |
+
"logqs_logps/rejected": 0.026350444182753563,
|
| 1207 |
+
"loss": 0.1764,
|
| 1208 |
+
"step": 1080
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.749423211957067,
|
| 1212 |
+
"grad_norm": 0.447265625,
|
| 1213 |
+
"learning_rate": 2.2111419389738435e-05,
|
| 1214 |
+
"logps_logqs/chosen": 0.07039657980203629,
|
| 1215 |
+
"logqs/chosen": -0.7764157056808472,
|
| 1216 |
+
"logqs/rejected": -0.9544513821601868,
|
| 1217 |
+
"logqs_logps/rejected": 0.02772611379623413,
|
| 1218 |
+
"loss": 0.1755,
|
| 1219 |
+
"step": 1090
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 1.765472966195205,
|
| 1223 |
+
"grad_norm": 0.5390625,
|
| 1224 |
+
"learning_rate": 2.1648134543918423e-05,
|
| 1225 |
+
"logps_logqs/chosen": 0.07292209565639496,
|
| 1226 |
+
"logqs/chosen": -0.7701107859611511,
|
| 1227 |
+
"logqs/rejected": -0.9618105888366699,
|
| 1228 |
+
"logqs_logps/rejected": 0.027484769001603127,
|
| 1229 |
+
"loss": 0.1762,
|
| 1230 |
+
"step": 1100
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 1.7815227204333435,
|
| 1234 |
+
"grad_norm": 0.427734375,
|
| 1235 |
+
"learning_rate": 2.1186018986695842e-05,
|
| 1236 |
+
"logps_logqs/chosen": 0.07149146497249603,
|
| 1237 |
+
"logqs/chosen": -0.7753132581710815,
|
| 1238 |
+
"logqs/rejected": -0.926436722278595,
|
| 1239 |
+
"logqs_logps/rejected": 0.029845744371414185,
|
| 1240 |
+
"loss": 0.1779,
|
| 1241 |
+
"step": 1110
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.7975724746714816,
|
| 1245 |
+
"grad_norm": 0.45703125,
|
| 1246 |
+
"learning_rate": 2.0725233925750063e-05,
|
| 1247 |
+
"logps_logqs/chosen": 0.06752609461545944,
|
| 1248 |
+
"logqs/chosen": -0.7796769738197327,
|
| 1249 |
+
"logqs/rejected": -0.9575828313827515,
|
| 1250 |
+
"logqs_logps/rejected": 0.03261734917759895,
|
| 1251 |
+
"loss": 0.1771,
|
| 1252 |
+
"step": 1120
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 1.8136222289096198,
|
| 1256 |
+
"grad_norm": 0.54296875,
|
| 1257 |
+
"learning_rate": 2.026594010462068e-05,
|
| 1258 |
+
"logps_logqs/chosen": 0.07001613825559616,
|
| 1259 |
+
"logqs/chosen": -0.7918448448181152,
|
| 1260 |
+
"logqs/rejected": -0.9749159812927246,
|
| 1261 |
+
"logqs_logps/rejected": 0.028455784544348717,
|
| 1262 |
+
"loss": 0.1789,
|
| 1263 |
+
"step": 1130
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.8296719831477581,
|
| 1267 |
+
"grad_norm": 0.478515625,
|
| 1268 |
+
"learning_rate": 1.980829774663256e-05,
|
| 1269 |
+
"logps_logqs/chosen": 0.07289810478687286,
|
| 1270 |
+
"logqs/chosen": -0.7838630676269531,
|
| 1271 |
+
"logqs/rejected": -0.9698160290718079,
|
| 1272 |
+
"logqs_logps/rejected": 0.02906595729291439,
|
| 1273 |
+
"loss": 0.1794,
|
| 1274 |
+
"step": 1140
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 1.8457217373858963,
|
| 1278 |
+
"grad_norm": 0.458984375,
|
| 1279 |
+
"learning_rate": 1.9352466499002422e-05,
|
| 1280 |
+
"logps_logqs/chosen": 0.0761161744594574,
|
| 1281 |
+
"logqs/chosen": -0.7819662094116211,
|
| 1282 |
+
"logqs/rejected": -0.9463040232658386,
|
| 1283 |
+
"logqs_logps/rejected": 0.028005197644233704,
|
| 1284 |
+
"loss": 0.183,
|
| 1285 |
+
"step": 1150
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 1.8617714916240344,
|
| 1289 |
+
"grad_norm": 0.474609375,
|
| 1290 |
+
"learning_rate": 1.8898605377146383e-05,
|
| 1291 |
+
"logps_logqs/chosen": 0.06957074254751205,
|
| 1292 |
+
"logqs/chosen": -0.7857328057289124,
|
| 1293 |
+
"logqs/rejected": -0.9639459848403931,
|
| 1294 |
+
"logqs_logps/rejected": 0.027649903669953346,
|
| 1295 |
+
"loss": 0.1753,
|
| 1296 |
+
"step": 1160
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 1.8778212458621728,
|
| 1300 |
+
"grad_norm": 0.48046875,
|
| 1301 |
+
"learning_rate": 1.8446872709207847e-05,
|
| 1302 |
+
"logps_logqs/chosen": 0.07362545281648636,
|
| 1303 |
+
"logqs/chosen": -0.7835830450057983,
|
| 1304 |
+
"logqs/rejected": -1.0045692920684814,
|
| 1305 |
+
"logqs_logps/rejected": 0.027295967563986778,
|
| 1306 |
+
"loss": 0.1806,
|
| 1307 |
+
"step": 1170
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 1.893871000100311,
|
| 1311 |
+
"grad_norm": 0.51171875,
|
| 1312 |
+
"learning_rate": 1.799742608082519e-05,
|
| 1313 |
+
"logps_logqs/chosen": 0.07236044853925705,
|
| 1314 |
+
"logqs/chosen": -0.7885088324546814,
|
| 1315 |
+
"logqs/rejected": -0.9487521052360535,
|
| 1316 |
+
"logqs_logps/rejected": 0.026073191314935684,
|
| 1317 |
+
"loss": 0.1723,
|
| 1318 |
+
"step": 1180
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 1.909920754338449,
|
| 1322 |
+
"grad_norm": 0.51171875,
|
| 1323 |
+
"learning_rate": 1.7550422280158513e-05,
|
| 1324 |
+
"logps_logqs/chosen": 0.0707259327173233,
|
| 1325 |
+
"logqs/chosen": -0.7704340815544128,
|
| 1326 |
+
"logqs/rejected": -0.9094502329826355,
|
| 1327 |
+
"logqs_logps/rejected": 0.027560651302337646,
|
| 1328 |
+
"loss": 0.1758,
|
| 1329 |
+
"step": 1190
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 1.9259705085765875,
|
| 1333 |
+
"grad_norm": 0.435546875,
|
| 1334 |
+
"learning_rate": 1.7106017243194487e-05,
|
| 1335 |
+
"logps_logqs/chosen": 0.07020822167396545,
|
| 1336 |
+
"logqs/chosen": -0.7692683935165405,
|
| 1337 |
+
"logqs/rejected": -0.9656769037246704,
|
| 1338 |
+
"logqs_logps/rejected": 0.02792223170399666,
|
| 1339 |
+
"loss": 0.177,
|
| 1340 |
+
"step": 1200
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.9420202628147256,
|
| 1344 |
+
"grad_norm": 0.53515625,
|
| 1345 |
+
"learning_rate": 1.6664365999348594e-05,
|
| 1346 |
+
"logps_logqs/chosen": 0.06943775713443756,
|
| 1347 |
+
"logqs/chosen": -0.7688643932342529,
|
| 1348 |
+
"logqs/rejected": -0.9713398218154907,
|
| 1349 |
+
"logqs_logps/rejected": 0.026859009638428688,
|
| 1350 |
+
"loss": 0.1748,
|
| 1351 |
+
"step": 1210
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 1.9580700170528638,
|
| 1355 |
+
"grad_norm": 0.51953125,
|
| 1356 |
+
"learning_rate": 1.6225622617383494e-05,
|
| 1357 |
+
"logps_logqs/chosen": 0.07070201635360718,
|
| 1358 |
+
"logqs/chosen": -0.7839328050613403,
|
| 1359 |
+
"logqs/rejected": -0.9712308645248413,
|
| 1360 |
+
"logqs_logps/rejected": 0.028167420998215675,
|
| 1361 |
+
"loss": 0.1763,
|
| 1362 |
+
"step": 1220
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 1.9741197712910021,
|
| 1366 |
+
"grad_norm": 0.4453125,
|
| 1367 |
+
"learning_rate": 1.578994015166263e-05,
|
| 1368 |
+
"logps_logqs/chosen": 0.07289667427539825,
|
| 1369 |
+
"logqs/chosen": -0.7838398814201355,
|
| 1370 |
+
"logqs/rejected": -0.9522368311882019,
|
| 1371 |
+
"logqs_logps/rejected": 0.02768387272953987,
|
| 1372 |
+
"loss": 0.1773,
|
| 1373 |
+
"step": 1230
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 1.9901695255291403,
|
| 1377 |
+
"grad_norm": 0.53515625,
|
| 1378 |
+
"learning_rate": 1.535747058875765e-05,
|
| 1379 |
+
"logps_logqs/chosen": 0.06881529092788696,
|
| 1380 |
+
"logqs/chosen": -0.790834903717041,
|
| 1381 |
+
"logqs/rejected": -0.9497320055961609,
|
| 1382 |
+
"logqs_logps/rejected": 0.029328888282179832,
|
| 1383 |
+
"loss": 0.1734,
|
| 1384 |
+
"step": 1240
|
| 1385 |
+
}
|
| 1386 |
+
],
|
| 1387 |
+
"logging_steps": 10,
|
| 1388 |
+
"max_steps": 1869,
|
| 1389 |
+
"num_input_tokens_seen": 0,
|
| 1390 |
+
"num_train_epochs": 3,
|
| 1391 |
+
"save_steps": 500,
|
| 1392 |
+
"stateful_callbacks": {
|
| 1393 |
+
"TrainerControl": {
|
| 1394 |
+
"args": {
|
| 1395 |
+
"should_epoch_stop": false,
|
| 1396 |
+
"should_evaluate": false,
|
| 1397 |
+
"should_log": false,
|
| 1398 |
+
"should_save": true,
|
| 1399 |
+
"should_training_stop": false
|
| 1400 |
+
},
|
| 1401 |
+
"attributes": {}
|
| 1402 |
+
}
|
| 1403 |
+
},
|
| 1404 |
+
"total_flos": 0.0,
|
| 1405 |
+
"train_batch_size": 2,
|
| 1406 |
+
"trial_name": null,
|
| 1407 |
+
"trial_params": null
|
| 1408 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1246/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b9fa66895bee5a36a6f750ef897f491a1113f2f79af77c75bdef8187ae10031
|
| 3 |
+
size 6840
|
qwen2.5-1.5B-it-distillm2/checkpoint-1246/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qwen2.5-1.5B-it-distillm2/checkpoint-1869/added_tokens.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</tool_call>": 151658,
|
| 3 |
+
"<tool_call>": 151657,
|
| 4 |
+
"<|box_end|>": 151649,
|
| 5 |
+
"<|box_start|>": 151648,
|
| 6 |
+
"<|endoftext|>": 151643,
|
| 7 |
+
"<|file_sep|>": 151664,
|
| 8 |
+
"<|fim_middle|>": 151660,
|
| 9 |
+
"<|fim_pad|>": 151662,
|
| 10 |
+
"<|fim_prefix|>": 151659,
|
| 11 |
+
"<|fim_suffix|>": 151661,
|
| 12 |
+
"<|im_end|>": 151645,
|
| 13 |
+
"<|im_start|>": 151644,
|
| 14 |
+
"<|image_pad|>": 151655,
|
| 15 |
+
"<|object_ref_end|>": 151647,
|
| 16 |
+
"<|object_ref_start|>": 151646,
|
| 17 |
+
"<|quad_end|>": 151651,
|
| 18 |
+
"<|quad_start|>": 151650,
|
| 19 |
+
"<|repo_name|>": 151663,
|
| 20 |
+
"<|video_pad|>": 151656,
|
| 21 |
+
"<|vision_end|>": 151653,
|
| 22 |
+
"<|vision_pad|>": 151654,
|
| 23 |
+
"<|vision_start|>": 151652
|
| 24 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1869/config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Qwen/Qwen2.5-1.5B-Instruct",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"Qwen2ForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": 151645,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 1536,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 8960,
|
| 13 |
+
"max_position_embeddings": 32768,
|
| 14 |
+
"max_window_layers": 21,
|
| 15 |
+
"model_type": "qwen2",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 28,
|
| 18 |
+
"num_key_value_heads": 2,
|
| 19 |
+
"rms_norm_eps": 1e-06,
|
| 20 |
+
"rope_scaling": null,
|
| 21 |
+
"rope_theta": 1000000.0,
|
| 22 |
+
"sliding_window": null,
|
| 23 |
+
"tie_word_embeddings": true,
|
| 24 |
+
"torch_dtype": "bfloat16",
|
| 25 |
+
"transformers_version": "4.45.2",
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"use_sliding_window": false,
|
| 28 |
+
"vocab_size": 152064
|
| 29 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1869/generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.45.2"
|
| 14 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1869/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qwen2.5-1.5B-it-distillm2/checkpoint-1869/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa1f325bd96a7e02353ee13f20a8362121741466a9b362e7cfd122a67c5ae6e0
|
| 3 |
+
size 3087935634
|
qwen2.5-1.5B-it-distillm2/checkpoint-1869/special_tokens_map.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>"
|
| 16 |
+
],
|
| 17 |
+
"bos_token": "<|endoftext|>",
|
| 18 |
+
"eos_token": {
|
| 19 |
+
"content": "<|im_end|>",
|
| 20 |
+
"lstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"rstrip": false,
|
| 23 |
+
"single_word": false
|
| 24 |
+
},
|
| 25 |
+
"pad_token": {
|
| 26 |
+
"content": "<|endoftext|>",
|
| 27 |
+
"lstrip": false,
|
| 28 |
+
"normalized": false,
|
| 29 |
+
"rstrip": false,
|
| 30 |
+
"single_word": false
|
| 31 |
+
}
|
| 32 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1869/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
|
| 3 |
+
size 11421896
|
qwen2.5-1.5B-it-distillm2/checkpoint-1869/tokenizer_config.json
ADDED
|
@@ -0,0 +1,207 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
}
|
| 181 |
+
},
|
| 182 |
+
"additional_special_tokens": [
|
| 183 |
+
"<|im_start|>",
|
| 184 |
+
"<|im_end|>",
|
| 185 |
+
"<|object_ref_start|>",
|
| 186 |
+
"<|object_ref_end|>",
|
| 187 |
+
"<|box_start|>",
|
| 188 |
+
"<|box_end|>",
|
| 189 |
+
"<|quad_start|>",
|
| 190 |
+
"<|quad_end|>",
|
| 191 |
+
"<|vision_start|>",
|
| 192 |
+
"<|vision_end|>",
|
| 193 |
+
"<|vision_pad|>",
|
| 194 |
+
"<|image_pad|>",
|
| 195 |
+
"<|video_pad|>"
|
| 196 |
+
],
|
| 197 |
+
"bos_token": "<|endoftext|>",
|
| 198 |
+
"chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
|
| 199 |
+
"clean_up_tokenization_spaces": false,
|
| 200 |
+
"eos_token": "<|im_end|>",
|
| 201 |
+
"errors": "replace",
|
| 202 |
+
"model_max_length": 2048,
|
| 203 |
+
"pad_token": "<|endoftext|>",
|
| 204 |
+
"split_special_tokens": false,
|
| 205 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 206 |
+
"unk_token": null
|
| 207 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1869/trainer_state.json
ADDED
|
@@ -0,0 +1,2090 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.999699067108035,
|
| 5 |
+
"eval_steps": 10000000,
|
| 6 |
+
"global_step": 1869,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0016049754238138228,
|
| 13 |
+
"grad_norm": 7.15625,
|
| 14 |
+
"learning_rate": 2.6737967914438503e-07,
|
| 15 |
+
"logps_logqs/chosen": 0.057332463562488556,
|
| 16 |
+
"logqs/chosen": -0.752794086933136,
|
| 17 |
+
"logqs/rejected": -0.7512239813804626,
|
| 18 |
+
"logqs_logps/rejected": 0.1236220970749855,
|
| 19 |
+
"loss": 0.4632,
|
| 20 |
+
"step": 1
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.016049754238138227,
|
| 24 |
+
"grad_norm": 5.71875,
|
| 25 |
+
"learning_rate": 2.6737967914438504e-06,
|
| 26 |
+
"logps_logqs/chosen": 0.047926291823387146,
|
| 27 |
+
"logqs/chosen": -0.7439535856246948,
|
| 28 |
+
"logqs/rejected": -0.7035253643989563,
|
| 29 |
+
"logqs_logps/rejected": 0.13991409540176392,
|
| 30 |
+
"loss": 0.4557,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.032099508476276455,
|
| 35 |
+
"grad_norm": 1.9296875,
|
| 36 |
+
"learning_rate": 5.347593582887701e-06,
|
| 37 |
+
"logps_logqs/chosen": 0.03161158040165901,
|
| 38 |
+
"logqs/chosen": -0.6941366195678711,
|
| 39 |
+
"logqs/rejected": -0.6452130079269409,
|
| 40 |
+
"logqs_logps/rejected": 0.1449585109949112,
|
| 41 |
+
"loss": 0.4015,
|
| 42 |
+
"step": 20
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.048149262714414685,
|
| 46 |
+
"grad_norm": 1.8515625,
|
| 47 |
+
"learning_rate": 8.02139037433155e-06,
|
| 48 |
+
"logps_logqs/chosen": 0.03457744047045708,
|
| 49 |
+
"logqs/chosen": -0.6867337226867676,
|
| 50 |
+
"logqs/rejected": -0.6923023462295532,
|
| 51 |
+
"logqs_logps/rejected": 0.12560859322547913,
|
| 52 |
+
"loss": 0.3565,
|
| 53 |
+
"step": 30
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.06419901695255291,
|
| 57 |
+
"grad_norm": 1.234375,
|
| 58 |
+
"learning_rate": 1.0695187165775402e-05,
|
| 59 |
+
"logps_logqs/chosen": 0.05277745798230171,
|
| 60 |
+
"logqs/chosen": -0.7405093908309937,
|
| 61 |
+
"logqs/rejected": -0.7894801497459412,
|
| 62 |
+
"logqs_logps/rejected": 0.0936320573091507,
|
| 63 |
+
"loss": 0.3228,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.08024877119069114,
|
| 68 |
+
"grad_norm": 1.0,
|
| 69 |
+
"learning_rate": 1.3368983957219252e-05,
|
| 70 |
+
"logps_logqs/chosen": 0.04658779874444008,
|
| 71 |
+
"logqs/chosen": -0.7268352508544922,
|
| 72 |
+
"logqs/rejected": -0.7926934361457825,
|
| 73 |
+
"logqs_logps/rejected": 0.09889288991689682,
|
| 74 |
+
"loss": 0.3027,
|
| 75 |
+
"step": 50
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.09629852542882937,
|
| 79 |
+
"grad_norm": 0.92578125,
|
| 80 |
+
"learning_rate": 1.60427807486631e-05,
|
| 81 |
+
"logps_logqs/chosen": 0.04193533584475517,
|
| 82 |
+
"logqs/chosen": -0.7151986956596375,
|
| 83 |
+
"logqs/rejected": -0.7687807083129883,
|
| 84 |
+
"logqs_logps/rejected": 0.09696364402770996,
|
| 85 |
+
"loss": 0.2856,
|
| 86 |
+
"step": 60
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1123482796669676,
|
| 90 |
+
"grad_norm": 1.0625,
|
| 91 |
+
"learning_rate": 1.8716577540106954e-05,
|
| 92 |
+
"logps_logqs/chosen": 0.04730736091732979,
|
| 93 |
+
"logqs/chosen": -0.7385339736938477,
|
| 94 |
+
"logqs/rejected": -0.8247607946395874,
|
| 95 |
+
"logqs_logps/rejected": 0.08417234569787979,
|
| 96 |
+
"loss": 0.285,
|
| 97 |
+
"step": 70
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.12839803390510582,
|
| 101 |
+
"grad_norm": 0.890625,
|
| 102 |
+
"learning_rate": 2.1390374331550803e-05,
|
| 103 |
+
"logps_logqs/chosen": 0.047996845096349716,
|
| 104 |
+
"logqs/chosen": -0.7431933879852295,
|
| 105 |
+
"logqs/rejected": -0.8130408525466919,
|
| 106 |
+
"logqs_logps/rejected": 0.08468352258205414,
|
| 107 |
+
"loss": 0.2797,
|
| 108 |
+
"step": 80
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.14444778814324405,
|
| 112 |
+
"grad_norm": 0.79296875,
|
| 113 |
+
"learning_rate": 2.4064171122994652e-05,
|
| 114 |
+
"logps_logqs/chosen": 0.04765843600034714,
|
| 115 |
+
"logqs/chosen": -0.7173298001289368,
|
| 116 |
+
"logqs/rejected": -0.7709535360336304,
|
| 117 |
+
"logqs_logps/rejected": 0.08819471299648285,
|
| 118 |
+
"loss": 0.2705,
|
| 119 |
+
"step": 90
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.16049754238138228,
|
| 123 |
+
"grad_norm": 0.82421875,
|
| 124 |
+
"learning_rate": 2.6737967914438505e-05,
|
| 125 |
+
"logps_logqs/chosen": 0.040474437177181244,
|
| 126 |
+
"logqs/chosen": -0.710905909538269,
|
| 127 |
+
"logqs/rejected": -0.7959136962890625,
|
| 128 |
+
"logqs_logps/rejected": 0.09612198173999786,
|
| 129 |
+
"loss": 0.2683,
|
| 130 |
+
"step": 100
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.1765472966195205,
|
| 134 |
+
"grad_norm": 0.7109375,
|
| 135 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 136 |
+
"logps_logqs/chosen": 0.020066609606146812,
|
| 137 |
+
"logqs/chosen": -0.6621483564376831,
|
| 138 |
+
"logqs/rejected": -0.7314690351486206,
|
| 139 |
+
"logqs_logps/rejected": 0.11639375984668732,
|
| 140 |
+
"loss": 0.2682,
|
| 141 |
+
"step": 110
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.19259705085765874,
|
| 145 |
+
"grad_norm": 0.9921875,
|
| 146 |
+
"learning_rate": 3.20855614973262e-05,
|
| 147 |
+
"logps_logqs/chosen": 0.017146889120340347,
|
| 148 |
+
"logqs/chosen": -0.6599355340003967,
|
| 149 |
+
"logqs/rejected": -0.7124370336532593,
|
| 150 |
+
"logqs_logps/rejected": 0.12015531212091446,
|
| 151 |
+
"loss": 0.268,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.20864680509579697,
|
| 156 |
+
"grad_norm": 1.78125,
|
| 157 |
+
"learning_rate": 3.4759358288770055e-05,
|
| 158 |
+
"logps_logqs/chosen": 0.0369986966252327,
|
| 159 |
+
"logqs/chosen": -0.7184829115867615,
|
| 160 |
+
"logqs/rejected": -0.8139392137527466,
|
| 161 |
+
"logqs_logps/rejected": 0.09705157577991486,
|
| 162 |
+
"loss": 0.2611,
|
| 163 |
+
"step": 130
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2246965593339352,
|
| 167 |
+
"grad_norm": 0.87890625,
|
| 168 |
+
"learning_rate": 3.743315508021391e-05,
|
| 169 |
+
"logps_logqs/chosen": 0.07482504099607468,
|
| 170 |
+
"logqs/chosen": -0.7948209643363953,
|
| 171 |
+
"logqs/rejected": -0.9149974584579468,
|
| 172 |
+
"logqs_logps/rejected": 0.04995354264974594,
|
| 173 |
+
"loss": 0.2442,
|
| 174 |
+
"step": 140
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.24074631357207343,
|
| 178 |
+
"grad_norm": 0.796875,
|
| 179 |
+
"learning_rate": 4.0106951871657754e-05,
|
| 180 |
+
"logps_logqs/chosen": 0.08310296386480331,
|
| 181 |
+
"logqs/chosen": -0.812109649181366,
|
| 182 |
+
"logqs/rejected": -0.9168623089790344,
|
| 183 |
+
"logqs_logps/rejected": 0.041201137006282806,
|
| 184 |
+
"loss": 0.2403,
|
| 185 |
+
"step": 150
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.25679606781021164,
|
| 189 |
+
"grad_norm": 0.70703125,
|
| 190 |
+
"learning_rate": 4.2780748663101606e-05,
|
| 191 |
+
"logps_logqs/chosen": 0.08498374372720718,
|
| 192 |
+
"logqs/chosen": -0.8101779818534851,
|
| 193 |
+
"logqs/rejected": -0.9302076101303101,
|
| 194 |
+
"logqs_logps/rejected": 0.03671664744615555,
|
| 195 |
+
"loss": 0.2336,
|
| 196 |
+
"step": 160
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.2728458220483499,
|
| 200 |
+
"grad_norm": 0.85546875,
|
| 201 |
+
"learning_rate": 4.545454545454546e-05,
|
| 202 |
+
"logps_logqs/chosen": 0.08320538699626923,
|
| 203 |
+
"logqs/chosen": -0.8246332406997681,
|
| 204 |
+
"logqs/rejected": -0.9219290018081665,
|
| 205 |
+
"logqs_logps/rejected": 0.03701116889715195,
|
| 206 |
+
"loss": 0.2386,
|
| 207 |
+
"step": 170
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.2888955762864881,
|
| 211 |
+
"grad_norm": 0.77734375,
|
| 212 |
+
"learning_rate": 4.8128342245989304e-05,
|
| 213 |
+
"logps_logqs/chosen": 0.08179891854524612,
|
| 214 |
+
"logqs/chosen": -0.787845253944397,
|
| 215 |
+
"logqs/rejected": -0.9030144810676575,
|
| 216 |
+
"logqs_logps/rejected": 0.03777966648340225,
|
| 217 |
+
"loss": 0.2315,
|
| 218 |
+
"step": 180
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.30494533052462636,
|
| 222 |
+
"grad_norm": 0.7734375,
|
| 223 |
+
"learning_rate": 4.9999607536612036e-05,
|
| 224 |
+
"logps_logqs/chosen": 0.08169040083885193,
|
| 225 |
+
"logqs/chosen": -0.8232030868530273,
|
| 226 |
+
"logqs/rejected": -0.9288152456283569,
|
| 227 |
+
"logqs_logps/rejected": 0.039177440106868744,
|
| 228 |
+
"loss": 0.2389,
|
| 229 |
+
"step": 190
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.32099508476276456,
|
| 233 |
+
"grad_norm": 0.72265625,
|
| 234 |
+
"learning_rate": 4.9992630752499945e-05,
|
| 235 |
+
"logps_logqs/chosen": 0.08127471804618835,
|
| 236 |
+
"logqs/chosen": -0.8097829818725586,
|
| 237 |
+
"logqs/rejected": -0.9110867381095886,
|
| 238 |
+
"logqs_logps/rejected": 0.03667169809341431,
|
| 239 |
+
"loss": 0.2279,
|
| 240 |
+
"step": 200
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3370448390009028,
|
| 244 |
+
"grad_norm": 0.71875,
|
| 245 |
+
"learning_rate": 4.997693536122969e-05,
|
| 246 |
+
"logps_logqs/chosen": 0.0788678377866745,
|
| 247 |
+
"logqs/chosen": -0.7994121313095093,
|
| 248 |
+
"logqs/rejected": -0.9303783178329468,
|
| 249 |
+
"logqs_logps/rejected": 0.037957318127155304,
|
| 250 |
+
"loss": 0.2291,
|
| 251 |
+
"step": 210
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.353094593239041,
|
| 255 |
+
"grad_norm": 0.7109375,
|
| 256 |
+
"learning_rate": 4.995252683809324e-05,
|
| 257 |
+
"logps_logqs/chosen": 0.08248866349458694,
|
| 258 |
+
"logqs/chosen": -0.8128089904785156,
|
| 259 |
+
"logqs/rejected": -0.9362344741821289,
|
| 260 |
+
"logqs_logps/rejected": 0.037949927151203156,
|
| 261 |
+
"loss": 0.229,
|
| 262 |
+
"step": 220
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3691443474771793,
|
| 266 |
+
"grad_norm": 0.63671875,
|
| 267 |
+
"learning_rate": 4.9919413697933496e-05,
|
| 268 |
+
"logps_logqs/chosen": 0.0817771628499031,
|
| 269 |
+
"logqs/chosen": -0.7939696311950684,
|
| 270 |
+
"logqs/rejected": -0.9058554768562317,
|
| 271 |
+
"logqs_logps/rejected": 0.04094362258911133,
|
| 272 |
+
"loss": 0.2314,
|
| 273 |
+
"step": 230
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.3851941017153175,
|
| 277 |
+
"grad_norm": 0.71875,
|
| 278 |
+
"learning_rate": 4.987760749217389e-05,
|
| 279 |
+
"logps_logqs/chosen": 0.076979860663414,
|
| 280 |
+
"logqs/chosen": -0.7874332070350647,
|
| 281 |
+
"logqs/rejected": -0.9303415417671204,
|
| 282 |
+
"logqs_logps/rejected": 0.04022833704948425,
|
| 283 |
+
"loss": 0.2244,
|
| 284 |
+
"step": 240
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.4012438559534557,
|
| 288 |
+
"grad_norm": 0.83984375,
|
| 289 |
+
"learning_rate": 4.982712280478875e-05,
|
| 290 |
+
"logps_logqs/chosen": 0.07487889379262924,
|
| 291 |
+
"logqs/chosen": -0.7985113859176636,
|
| 292 |
+
"logqs/rejected": -0.9595896005630493,
|
| 293 |
+
"logqs_logps/rejected": 0.04083302244544029,
|
| 294 |
+
"loss": 0.2277,
|
| 295 |
+
"step": 250
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.41729361019159394,
|
| 299 |
+
"grad_norm": 0.6953125,
|
| 300 |
+
"learning_rate": 4.976797724721567e-05,
|
| 301 |
+
"logps_logqs/chosen": 0.07195514440536499,
|
| 302 |
+
"logqs/chosen": -0.787926971912384,
|
| 303 |
+
"logqs/rejected": -0.897616982460022,
|
| 304 |
+
"logqs_logps/rejected": 0.0396430678665638,
|
| 305 |
+
"loss": 0.2163,
|
| 306 |
+
"step": 260
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.43334336442973215,
|
| 310 |
+
"grad_norm": 0.7734375,
|
| 311 |
+
"learning_rate": 4.9700191452211806e-05,
|
| 312 |
+
"logps_logqs/chosen": 0.07640247046947479,
|
| 313 |
+
"logqs/chosen": -0.7960728406906128,
|
| 314 |
+
"logqs/rejected": -0.928481936454773,
|
| 315 |
+
"logqs_logps/rejected": 0.03928074985742569,
|
| 316 |
+
"loss": 0.2226,
|
| 317 |
+
"step": 270
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.4493931186678704,
|
| 321 |
+
"grad_norm": 0.72265625,
|
| 322 |
+
"learning_rate": 4.9623789066656276e-05,
|
| 323 |
+
"logps_logqs/chosen": 0.07622957229614258,
|
| 324 |
+
"logqs/chosen": -0.8077980279922485,
|
| 325 |
+
"logqs/rejected": -0.9034830927848816,
|
| 326 |
+
"logqs_logps/rejected": 0.046734608709812164,
|
| 327 |
+
"loss": 0.2256,
|
| 328 |
+
"step": 280
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.4654428729060086,
|
| 332 |
+
"grad_norm": 0.78515625,
|
| 333 |
+
"learning_rate": 4.953879674330093e-05,
|
| 334 |
+
"logps_logqs/chosen": 0.07244043052196503,
|
| 335 |
+
"logqs/chosen": -0.7900176644325256,
|
| 336 |
+
"logqs/rejected": -0.891532301902771,
|
| 337 |
+
"logqs_logps/rejected": 0.04387632757425308,
|
| 338 |
+
"loss": 0.217,
|
| 339 |
+
"step": 290
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.48149262714414687,
|
| 343 |
+
"grad_norm": 0.6796875,
|
| 344 |
+
"learning_rate": 4.944524413147263e-05,
|
| 345 |
+
"logps_logqs/chosen": 0.07289232313632965,
|
| 346 |
+
"logqs/chosen": -0.7856311798095703,
|
| 347 |
+
"logqs/rejected": -0.9099383354187012,
|
| 348 |
+
"logqs_logps/rejected": 0.03974480181932449,
|
| 349 |
+
"loss": 0.2173,
|
| 350 |
+
"step": 300
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.49754238138228507,
|
| 354 |
+
"grad_norm": 0.66796875,
|
| 355 |
+
"learning_rate": 4.934316386673022e-05,
|
| 356 |
+
"logps_logqs/chosen": 0.07063151150941849,
|
| 357 |
+
"logqs/chosen": -0.7815223932266235,
|
| 358 |
+
"logqs/rejected": -0.9079385995864868,
|
| 359 |
+
"logqs_logps/rejected": 0.041664548218250275,
|
| 360 |
+
"loss": 0.2126,
|
| 361 |
+
"step": 310
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.5135921356204233,
|
| 365 |
+
"grad_norm": 0.61328125,
|
| 366 |
+
"learning_rate": 4.923259155947964e-05,
|
| 367 |
+
"logps_logqs/chosen": 0.071230448782444,
|
| 368 |
+
"logqs/chosen": -0.8096257448196411,
|
| 369 |
+
"logqs/rejected": -0.9752206802368164,
|
| 370 |
+
"logqs_logps/rejected": 0.03939716890454292,
|
| 371 |
+
"loss": 0.2194,
|
| 372 |
+
"step": 320
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.5296418898585615,
|
| 376 |
+
"grad_norm": 0.6640625,
|
| 377 |
+
"learning_rate": 4.911356578255139e-05,
|
| 378 |
+
"logps_logqs/chosen": 0.07356056571006775,
|
| 379 |
+
"logqs/chosen": -0.7827819585800171,
|
| 380 |
+
"logqs/rejected": -0.9489104151725769,
|
| 381 |
+
"logqs_logps/rejected": 0.040141116827726364,
|
| 382 |
+
"loss": 0.2145,
|
| 383 |
+
"step": 330
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.5456916440966998,
|
| 387 |
+
"grad_norm": 0.7109375,
|
| 388 |
+
"learning_rate": 4.898612805774447e-05,
|
| 389 |
+
"logps_logqs/chosen": 0.07881536334753036,
|
| 390 |
+
"logqs/chosen": -0.8095147013664246,
|
| 391 |
+
"logqs/rejected": -0.9313008189201355,
|
| 392 |
+
"logqs_logps/rejected": 0.03702981770038605,
|
| 393 |
+
"loss": 0.2165,
|
| 394 |
+
"step": 340
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.561741398334838,
|
| 398 |
+
"grad_norm": 0.64453125,
|
| 399 |
+
"learning_rate": 4.885032284134165e-05,
|
| 400 |
+
"logps_logqs/chosen": 0.07632436603307724,
|
| 401 |
+
"logqs/chosen": -0.7979615926742554,
|
| 402 |
+
"logqs/rejected": -0.9368084073066711,
|
| 403 |
+
"logqs_logps/rejected": 0.03654230386018753,
|
| 404 |
+
"loss": 0.2113,
|
| 405 |
+
"step": 350
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.5777911525729762,
|
| 409 |
+
"grad_norm": 0.62109375,
|
| 410 |
+
"learning_rate": 4.8706197508600984e-05,
|
| 411 |
+
"logps_logqs/chosen": 0.07344283908605576,
|
| 412 |
+
"logqs/chosen": -0.8101444244384766,
|
| 413 |
+
"logqs/rejected": -0.9295485615730286,
|
| 414 |
+
"logqs_logps/rejected": 0.03901313990354538,
|
| 415 |
+
"loss": 0.2151,
|
| 416 |
+
"step": 360
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5938409068111145,
|
| 420 |
+
"grad_norm": 0.62890625,
|
| 421 |
+
"learning_rate": 4.855380233722915e-05,
|
| 422 |
+
"logps_logqs/chosen": 0.07421533018350601,
|
| 423 |
+
"logqs/chosen": -0.7826108336448669,
|
| 424 |
+
"logqs/rejected": -0.9229670763015747,
|
| 425 |
+
"logqs_logps/rejected": 0.038314513862133026,
|
| 426 |
+
"loss": 0.2129,
|
| 427 |
+
"step": 370
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.6098906610492527,
|
| 431 |
+
"grad_norm": 0.62890625,
|
| 432 |
+
"learning_rate": 4.839319048984217e-05,
|
| 433 |
+
"logps_logqs/chosen": 0.07273373752832413,
|
| 434 |
+
"logqs/chosen": -0.7798442840576172,
|
| 435 |
+
"logqs/rejected": -0.9371780157089233,
|
| 436 |
+
"logqs_logps/rejected": 0.03713950887322426,
|
| 437 |
+
"loss": 0.2067,
|
| 438 |
+
"step": 380
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.6259404152873909,
|
| 442 |
+
"grad_norm": 0.62890625,
|
| 443 |
+
"learning_rate": 4.822441799541979e-05,
|
| 444 |
+
"logps_logqs/chosen": 0.07329441606998444,
|
| 445 |
+
"logqs/chosen": -0.8102380037307739,
|
| 446 |
+
"logqs/rejected": -0.95549476146698,
|
| 447 |
+
"logqs_logps/rejected": 0.036079905927181244,
|
| 448 |
+
"loss": 0.2158,
|
| 449 |
+
"step": 390
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.6419901695255291,
|
| 453 |
+
"grad_norm": 0.66796875,
|
| 454 |
+
"learning_rate": 4.8047543729759936e-05,
|
| 455 |
+
"logps_logqs/chosen": 0.07404083013534546,
|
| 456 |
+
"logqs/chosen": -0.7847949266433716,
|
| 457 |
+
"logqs/rejected": -0.9250528216362,
|
| 458 |
+
"logqs_logps/rejected": 0.036556728184223175,
|
| 459 |
+
"loss": 0.2105,
|
| 460 |
+
"step": 400
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.6580399237636674,
|
| 464 |
+
"grad_norm": 0.69140625,
|
| 465 |
+
"learning_rate": 4.786262939494007e-05,
|
| 466 |
+
"logps_logqs/chosen": 0.07539906352758408,
|
| 467 |
+
"logqs/chosen": -0.8072575330734253,
|
| 468 |
+
"logqs/rejected": -0.9661371111869812,
|
| 469 |
+
"logqs_logps/rejected": 0.03662776201963425,
|
| 470 |
+
"loss": 0.2154,
|
| 471 |
+
"step": 410
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6740896780018056,
|
| 475 |
+
"grad_norm": 0.66015625,
|
| 476 |
+
"learning_rate": 4.766973949779261e-05,
|
| 477 |
+
"logps_logqs/chosen": 0.0744672566652298,
|
| 478 |
+
"logqs/chosen": -0.787712574005127,
|
| 479 |
+
"logqs/rejected": -0.9350829124450684,
|
| 480 |
+
"logqs_logps/rejected": 0.0353250689804554,
|
| 481 |
+
"loss": 0.2074,
|
| 482 |
+
"step": 420
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.6901394322399438,
|
| 486 |
+
"grad_norm": 0.62890625,
|
| 487 |
+
"learning_rate": 4.746894132740186e-05,
|
| 488 |
+
"logps_logqs/chosen": 0.07364196330308914,
|
| 489 |
+
"logqs/chosen": -0.7813644409179688,
|
| 490 |
+
"logqs/rejected": -0.9208289980888367,
|
| 491 |
+
"logqs_logps/rejected": 0.0356430858373642,
|
| 492 |
+
"loss": 0.2053,
|
| 493 |
+
"step": 430
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.706189186478082,
|
| 497 |
+
"grad_norm": 0.56640625,
|
| 498 |
+
"learning_rate": 4.726030493163044e-05,
|
| 499 |
+
"logps_logqs/chosen": 0.07587061077356339,
|
| 500 |
+
"logqs/chosen": -0.7853146195411682,
|
| 501 |
+
"logqs/rejected": -0.8944910168647766,
|
| 502 |
+
"logqs_logps/rejected": 0.03531279042363167,
|
| 503 |
+
"loss": 0.2078,
|
| 504 |
+
"step": 440
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.7222389407162203,
|
| 508 |
+
"grad_norm": 0.68359375,
|
| 509 |
+
"learning_rate": 4.7043903092683314e-05,
|
| 510 |
+
"logps_logqs/chosen": 0.07814273238182068,
|
| 511 |
+
"logqs/chosen": -0.8133522272109985,
|
| 512 |
+
"logqs/rejected": -0.9508693814277649,
|
| 513 |
+
"logqs_logps/rejected": 0.030778918415308,
|
| 514 |
+
"loss": 0.2104,
|
| 515 |
+
"step": 450
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.7382886949543586,
|
| 519 |
+
"grad_norm": 0.72265625,
|
| 520 |
+
"learning_rate": 4.6819811301717885e-05,
|
| 521 |
+
"logps_logqs/chosen": 0.07763786613941193,
|
| 522 |
+
"logqs/chosen": -0.8061367869377136,
|
| 523 |
+
"logqs/rejected": -0.9445575475692749,
|
| 524 |
+
"logqs_logps/rejected": 0.035373255610466,
|
| 525 |
+
"loss": 0.2084,
|
| 526 |
+
"step": 460
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.7543384491924967,
|
| 530 |
+
"grad_norm": 0.59765625,
|
| 531 |
+
"learning_rate": 4.6588107732509134e-05,
|
| 532 |
+
"logps_logqs/chosen": 0.07325359433889389,
|
| 533 |
+
"logqs/chosen": -0.7970572710037231,
|
| 534 |
+
"logqs/rejected": -0.9243482351303101,
|
| 535 |
+
"logqs_logps/rejected": 0.03834759443998337,
|
| 536 |
+
"loss": 0.2088,
|
| 537 |
+
"step": 470
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.770388203430635,
|
| 541 |
+
"grad_norm": 0.5625,
|
| 542 |
+
"learning_rate": 4.634887321417895e-05,
|
| 543 |
+
"logps_logqs/chosen": 0.0732722282409668,
|
| 544 |
+
"logqs/chosen": -0.7719672918319702,
|
| 545 |
+
"logqs/rejected": -0.9193938970565796,
|
| 546 |
+
"logqs_logps/rejected": 0.03625096380710602,
|
| 547 |
+
"loss": 0.2075,
|
| 548 |
+
"step": 480
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7864379576687732,
|
| 552 |
+
"grad_norm": 0.90234375,
|
| 553 |
+
"learning_rate": 4.6102191202999065e-05,
|
| 554 |
+
"logps_logqs/chosen": 0.07290570437908173,
|
| 555 |
+
"logqs/chosen": -0.8092619180679321,
|
| 556 |
+
"logqs/rejected": -0.9642523527145386,
|
| 557 |
+
"logqs_logps/rejected": 0.03311945125460625,
|
| 558 |
+
"loss": 0.2045,
|
| 559 |
+
"step": 490
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 0.8024877119069114,
|
| 563 |
+
"grad_norm": 0.6015625,
|
| 564 |
+
"learning_rate": 4.5848147753277656e-05,
|
| 565 |
+
"logps_logqs/chosen": 0.07134760171175003,
|
| 566 |
+
"logqs/chosen": -0.7773372530937195,
|
| 567 |
+
"logqs/rejected": -0.9478441476821899,
|
| 568 |
+
"logqs_logps/rejected": 0.03295496851205826,
|
| 569 |
+
"loss": 0.2049,
|
| 570 |
+
"step": 500
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.8185374661450496,
|
| 574 |
+
"grad_norm": 0.5859375,
|
| 575 |
+
"learning_rate": 4.5586831487339485e-05,
|
| 576 |
+
"logps_logqs/chosen": 0.07219503819942474,
|
| 577 |
+
"logqs/chosen": -0.7950411438941956,
|
| 578 |
+
"logqs/rejected": -0.9455002546310425,
|
| 579 |
+
"logqs_logps/rejected": 0.03765324503183365,
|
| 580 |
+
"loss": 0.208,
|
| 581 |
+
"step": 510
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.8345872203831879,
|
| 585 |
+
"grad_norm": 0.62109375,
|
| 586 |
+
"learning_rate": 4.531833356461027e-05,
|
| 587 |
+
"logps_logqs/chosen": 0.06849464029073715,
|
| 588 |
+
"logqs/chosen": -0.7819440364837646,
|
| 589 |
+
"logqs/rejected": -0.9597524404525757,
|
| 590 |
+
"logqs_logps/rejected": 0.04056422412395477,
|
| 591 |
+
"loss": 0.2084,
|
| 592 |
+
"step": 520
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.8506369746213261,
|
| 596 |
+
"grad_norm": 0.6171875,
|
| 597 |
+
"learning_rate": 4.5042747649816006e-05,
|
| 598 |
+
"logps_logqs/chosen": 0.07160626351833344,
|
| 599 |
+
"logqs/chosen": -0.7721427083015442,
|
| 600 |
+
"logqs/rejected": -0.94866544008255,
|
| 601 |
+
"logqs_logps/rejected": 0.03749927878379822,
|
| 602 |
+
"loss": 0.2068,
|
| 603 |
+
"step": 530
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.8666867288594643,
|
| 607 |
+
"grad_norm": 0.5546875,
|
| 608 |
+
"learning_rate": 4.476016988030826e-05,
|
| 609 |
+
"logps_logqs/chosen": 0.07214485853910446,
|
| 610 |
+
"logqs/chosen": -0.7874671816825867,
|
| 611 |
+
"logqs/rejected": -0.9504098892211914,
|
| 612 |
+
"logqs_logps/rejected": 0.03507527709007263,
|
| 613 |
+
"loss": 0.2045,
|
| 614 |
+
"step": 540
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.8827364830976026,
|
| 618 |
+
"grad_norm": 0.54296875,
|
| 619 |
+
"learning_rate": 4.447069883252696e-05,
|
| 620 |
+
"logps_logqs/chosen": 0.07395409047603607,
|
| 621 |
+
"logqs/chosen": -0.771978497505188,
|
| 622 |
+
"logqs/rejected": -0.9129235148429871,
|
| 623 |
+
"logqs_logps/rejected": 0.039593033492565155,
|
| 624 |
+
"loss": 0.205,
|
| 625 |
+
"step": 550
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.8987862373357408,
|
| 629 |
+
"grad_norm": 0.61328125,
|
| 630 |
+
"learning_rate": 4.417443548761227e-05,
|
| 631 |
+
"logps_logqs/chosen": 0.07299650460481644,
|
| 632 |
+
"logqs/chosen": -0.7979342341423035,
|
| 633 |
+
"logqs/rejected": -0.9357426762580872,
|
| 634 |
+
"logqs_logps/rejected": 0.03937726467847824,
|
| 635 |
+
"loss": 0.2083,
|
| 636 |
+
"step": 560
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.9148359915738791,
|
| 640 |
+
"grad_norm": 0.57421875,
|
| 641 |
+
"learning_rate": 4.387148319617763e-05,
|
| 642 |
+
"logps_logqs/chosen": 0.06836996972560883,
|
| 643 |
+
"logqs/chosen": -0.7659951448440552,
|
| 644 |
+
"logqs/rejected": -0.8732549548149109,
|
| 645 |
+
"logqs_logps/rejected": 0.039081670343875885,
|
| 646 |
+
"loss": 0.203,
|
| 647 |
+
"step": 570
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.9308857458120172,
|
| 651 |
+
"grad_norm": 0.57421875,
|
| 652 |
+
"learning_rate": 4.356194764225618e-05,
|
| 653 |
+
"logps_logqs/chosen": 0.07397963851690292,
|
| 654 |
+
"logqs/chosen": -0.7891489863395691,
|
| 655 |
+
"logqs/rejected": -0.9294188618659973,
|
| 656 |
+
"logqs_logps/rejected": 0.03416060656309128,
|
| 657 |
+
"loss": 0.2024,
|
| 658 |
+
"step": 580
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.9469355000501555,
|
| 662 |
+
"grad_norm": 0.703125,
|
| 663 |
+
"learning_rate": 4.3245936806433205e-05,
|
| 664 |
+
"logps_logqs/chosen": 0.07466734945774078,
|
| 665 |
+
"logqs/chosen": -0.7958794832229614,
|
| 666 |
+
"logqs/rejected": -0.953458309173584,
|
| 667 |
+
"logqs_logps/rejected": 0.035235695540905,
|
| 668 |
+
"loss": 0.206,
|
| 669 |
+
"step": 590
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.9629852542882937,
|
| 673 |
+
"grad_norm": 0.5859375,
|
| 674 |
+
"learning_rate": 4.292356092817739e-05,
|
| 675 |
+
"logps_logqs/chosen": 0.07136549055576324,
|
| 676 |
+
"logqs/chosen": -0.7849777936935425,
|
| 677 |
+
"logqs/rejected": -0.9327009320259094,
|
| 678 |
+
"logqs_logps/rejected": 0.038332488387823105,
|
| 679 |
+
"loss": 0.2041,
|
| 680 |
+
"step": 600
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.979035008526432,
|
| 684 |
+
"grad_norm": 0.55859375,
|
| 685 |
+
"learning_rate": 4.259493246738409e-05,
|
| 686 |
+
"logps_logqs/chosen": 0.06976237148046494,
|
| 687 |
+
"logqs/chosen": -0.7658575773239136,
|
| 688 |
+
"logqs/rejected": -0.9027583003044128,
|
| 689 |
+
"logqs_logps/rejected": 0.03464614599943161,
|
| 690 |
+
"loss": 0.1993,
|
| 691 |
+
"step": 610
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.9950847627645701,
|
| 695 |
+
"grad_norm": 0.55078125,
|
| 696 |
+
"learning_rate": 4.226016606514411e-05,
|
| 697 |
+
"logps_logqs/chosen": 0.0706130638718605,
|
| 698 |
+
"logqs/chosen": -0.7847878932952881,
|
| 699 |
+
"logqs/rejected": -0.9509286880493164,
|
| 700 |
+
"logqs_logps/rejected": 0.03653167933225632,
|
| 701 |
+
"loss": 0.2035,
|
| 702 |
+
"step": 620
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.0111345170027084,
|
| 706 |
+
"grad_norm": 0.53515625,
|
| 707 |
+
"learning_rate": 4.191937850375136e-05,
|
| 708 |
+
"logps_logqs/chosen": 0.06913185119628906,
|
| 709 |
+
"logqs/chosen": -0.7918249368667603,
|
| 710 |
+
"logqs/rejected": -0.9458308219909668,
|
| 711 |
+
"logqs_logps/rejected": 0.030862990766763687,
|
| 712 |
+
"loss": 0.1886,
|
| 713 |
+
"step": 630
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.0271842712408465,
|
| 717 |
+
"grad_norm": 0.5625,
|
| 718 |
+
"learning_rate": 4.157268866596381e-05,
|
| 719 |
+
"logps_logqs/chosen": 0.0690259039402008,
|
| 720 |
+
"logqs/chosen": -0.7895857095718384,
|
| 721 |
+
"logqs/rejected": -0.9810531735420227,
|
| 722 |
+
"logqs_logps/rejected": 0.03111192025244236,
|
| 723 |
+
"loss": 0.187,
|
| 724 |
+
"step": 640
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.043234025478985,
|
| 728 |
+
"grad_norm": 0.53125,
|
| 729 |
+
"learning_rate": 4.1220217493531494e-05,
|
| 730 |
+
"logps_logqs/chosen": 0.06900795549154282,
|
| 731 |
+
"logqs/chosen": -0.7902609705924988,
|
| 732 |
+
"logqs/rejected": -0.9747546911239624,
|
| 733 |
+
"logqs_logps/rejected": 0.031710296869277954,
|
| 734 |
+
"loss": 0.1845,
|
| 735 |
+
"step": 650
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 1.059283779717123,
|
| 739 |
+
"grad_norm": 0.515625,
|
| 740 |
+
"learning_rate": 4.086208794500637e-05,
|
| 741 |
+
"logps_logqs/chosen": 0.07102300226688385,
|
| 742 |
+
"logqs/chosen": -0.7767165899276733,
|
| 743 |
+
"logqs/rejected": -0.9414197206497192,
|
| 744 |
+
"logqs_logps/rejected": 0.030909577384591103,
|
| 745 |
+
"loss": 0.1807,
|
| 746 |
+
"step": 660
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.0753335339552612,
|
| 750 |
+
"grad_norm": 0.494140625,
|
| 751 |
+
"learning_rate": 4.049842495284858e-05,
|
| 752 |
+
"logps_logqs/chosen": 0.06985093653202057,
|
| 753 |
+
"logqs/chosen": -0.7880310416221619,
|
| 754 |
+
"logqs/rejected": -0.9546613693237305,
|
| 755 |
+
"logqs_logps/rejected": 0.030261676758527756,
|
| 756 |
+
"loss": 0.186,
|
| 757 |
+
"step": 670
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 1.0913832881933996,
|
| 761 |
+
"grad_norm": 0.50390625,
|
| 762 |
+
"learning_rate": 4.012935537984414e-05,
|
| 763 |
+
"logps_logqs/chosen": 0.07058895379304886,
|
| 764 |
+
"logqs/chosen": -0.7935397624969482,
|
| 765 |
+
"logqs/rejected": -1.0117493867874146,
|
| 766 |
+
"logqs_logps/rejected": 0.029221097007393837,
|
| 767 |
+
"loss": 0.1835,
|
| 768 |
+
"step": 680
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.1074330424315377,
|
| 772 |
+
"grad_norm": 0.6171875,
|
| 773 |
+
"learning_rate": 3.9755007974849135e-05,
|
| 774 |
+
"logps_logqs/chosen": 0.06916572153568268,
|
| 775 |
+
"logqs/chosen": -0.7878638505935669,
|
| 776 |
+
"logqs/rejected": -0.9999720454216003,
|
| 777 |
+
"logqs_logps/rejected": 0.030744653195142746,
|
| 778 |
+
"loss": 0.1857,
|
| 779 |
+
"step": 690
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.123482796669676,
|
| 783 |
+
"grad_norm": 0.57421875,
|
| 784 |
+
"learning_rate": 3.93755133278762e-05,
|
| 785 |
+
"logps_logqs/chosen": 0.07014169543981552,
|
| 786 |
+
"logqs/chosen": -0.7882756590843201,
|
| 787 |
+
"logqs/rejected": -1.0001386404037476,
|
| 788 |
+
"logqs_logps/rejected": 0.02842717245221138,
|
| 789 |
+
"loss": 0.1832,
|
| 790 |
+
"step": 700
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.1395325509078142,
|
| 794 |
+
"grad_norm": 0.59375,
|
| 795 |
+
"learning_rate": 3.899100382453845e-05,
|
| 796 |
+
"logps_logqs/chosen": 0.0701603814959526,
|
| 797 |
+
"logqs/chosen": -0.7642520070075989,
|
| 798 |
+
"logqs/rejected": -0.9544415473937988,
|
| 799 |
+
"logqs_logps/rejected": 0.027573522180318832,
|
| 800 |
+
"loss": 0.1806,
|
| 801 |
+
"step": 710
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.1555823051459524,
|
| 805 |
+
"grad_norm": 0.50390625,
|
| 806 |
+
"learning_rate": 3.8601613599867156e-05,
|
| 807 |
+
"logps_logqs/chosen": 0.06890274584293365,
|
| 808 |
+
"logqs/chosen": -0.7708860635757446,
|
| 809 |
+
"logqs/rejected": -0.9537287950515747,
|
| 810 |
+
"logqs_logps/rejected": 0.028619807213544846,
|
| 811 |
+
"loss": 0.1804,
|
| 812 |
+
"step": 720
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 1.1716320593840908,
|
| 816 |
+
"grad_norm": 0.609375,
|
| 817 |
+
"learning_rate": 3.8207478491519216e-05,
|
| 818 |
+
"logps_logqs/chosen": 0.0682586207985878,
|
| 819 |
+
"logqs/chosen": -0.7696245908737183,
|
| 820 |
+
"logqs/rejected": -0.9157883524894714,
|
| 821 |
+
"logqs_logps/rejected": 0.028189942240715027,
|
| 822 |
+
"loss": 0.1814,
|
| 823 |
+
"step": 730
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 1.187681813622229,
|
| 827 |
+
"grad_norm": 0.59375,
|
| 828 |
+
"learning_rate": 3.780873599239044e-05,
|
| 829 |
+
"logps_logqs/chosen": 0.07058210670948029,
|
| 830 |
+
"logqs/chosen": -0.801421046257019,
|
| 831 |
+
"logqs/rejected": -0.9591751098632812,
|
| 832 |
+
"logqs_logps/rejected": 0.03063536249101162,
|
| 833 |
+
"loss": 0.1853,
|
| 834 |
+
"step": 740
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.203731567860367,
|
| 838 |
+
"grad_norm": 0.55859375,
|
| 839 |
+
"learning_rate": 3.740552520265167e-05,
|
| 840 |
+
"logps_logqs/chosen": 0.06886611133813858,
|
| 841 |
+
"logqs/chosen": -0.792157769203186,
|
| 842 |
+
"logqs/rejected": -0.9370824694633484,
|
| 843 |
+
"logqs_logps/rejected": 0.02984955906867981,
|
| 844 |
+
"loss": 0.1833,
|
| 845 |
+
"step": 750
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 1.2197813220985054,
|
| 849 |
+
"grad_norm": 0.55859375,
|
| 850 |
+
"learning_rate": 3.699798678122403e-05,
|
| 851 |
+
"logps_logqs/chosen": 0.07179007679224014,
|
| 852 |
+
"logqs/chosen": -0.7622597813606262,
|
| 853 |
+
"logqs/rejected": -0.9382703900337219,
|
| 854 |
+
"logqs_logps/rejected": 0.029157549142837524,
|
| 855 |
+
"loss": 0.1793,
|
| 856 |
+
"step": 760
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.2358310763366436,
|
| 860 |
+
"grad_norm": 0.546875,
|
| 861 |
+
"learning_rate": 3.6586262896710476e-05,
|
| 862 |
+
"logps_logqs/chosen": 0.06790535151958466,
|
| 863 |
+
"logqs/chosen": -0.7676440477371216,
|
| 864 |
+
"logqs/rejected": -0.928793728351593,
|
| 865 |
+
"logqs_logps/rejected": 0.03030979633331299,
|
| 866 |
+
"loss": 0.1808,
|
| 867 |
+
"step": 770
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 1.2518808305747817,
|
| 871 |
+
"grad_norm": 0.51171875,
|
| 872 |
+
"learning_rate": 3.61704971778007e-05,
|
| 873 |
+
"logps_logqs/chosen": 0.06736615300178528,
|
| 874 |
+
"logqs/chosen": -0.770586371421814,
|
| 875 |
+
"logqs/rejected": -0.9197053909301758,
|
| 876 |
+
"logqs_logps/rejected": 0.03245236724615097,
|
| 877 |
+
"loss": 0.1799,
|
| 878 |
+
"step": 780
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.26793058481292,
|
| 882 |
+
"grad_norm": 0.4765625,
|
| 883 |
+
"learning_rate": 3.575083466316664e-05,
|
| 884 |
+
"logps_logqs/chosen": 0.06956754624843597,
|
| 885 |
+
"logqs/chosen": -0.7774965167045593,
|
| 886 |
+
"logqs/rejected": -0.9702298045158386,
|
| 887 |
+
"logqs_logps/rejected": 0.030748773366212845,
|
| 888 |
+
"loss": 0.1831,
|
| 889 |
+
"step": 790
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 1.2839803390510582,
|
| 893 |
+
"grad_norm": 0.5078125,
|
| 894 |
+
"learning_rate": 3.532742175086621e-05,
|
| 895 |
+
"logps_logqs/chosen": 0.06920956075191498,
|
| 896 |
+
"logqs/chosen": -0.7709556818008423,
|
| 897 |
+
"logqs/rejected": -0.9387216567993164,
|
| 898 |
+
"logqs_logps/rejected": 0.027966167777776718,
|
| 899 |
+
"loss": 0.1788,
|
| 900 |
+
"step": 800
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 1.3000300932891964,
|
| 904 |
+
"grad_norm": 0.54296875,
|
| 905 |
+
"learning_rate": 3.490040614727272e-05,
|
| 906 |
+
"logps_logqs/chosen": 0.06927359104156494,
|
| 907 |
+
"logqs/chosen": -0.7772814035415649,
|
| 908 |
+
"logqs/rejected": -0.9691821932792664,
|
| 909 |
+
"logqs_logps/rejected": 0.028416061773896217,
|
| 910 |
+
"loss": 0.1781,
|
| 911 |
+
"step": 810
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 1.3160798475273348,
|
| 915 |
+
"grad_norm": 0.5078125,
|
| 916 |
+
"learning_rate": 3.446993681554797e-05,
|
| 917 |
+
"logps_logqs/chosen": 0.07202474772930145,
|
| 918 |
+
"logqs/chosen": -0.7855108380317688,
|
| 919 |
+
"logqs/rejected": -0.9708096385002136,
|
| 920 |
+
"logqs_logps/rejected": 0.030036652460694313,
|
| 921 |
+
"loss": 0.1793,
|
| 922 |
+
"step": 820
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 1.332129601765473,
|
| 926 |
+
"grad_norm": 0.65234375,
|
| 927 |
+
"learning_rate": 3.403616392367681e-05,
|
| 928 |
+
"logps_logqs/chosen": 0.07054628431797028,
|
| 929 |
+
"logqs/chosen": -0.7540086507797241,
|
| 930 |
+
"logqs/rejected": -0.9042137265205383,
|
| 931 |
+
"logqs_logps/rejected": 0.03296629339456558,
|
| 932 |
+
"loss": 0.1818,
|
| 933 |
+
"step": 830
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.3481793560036113,
|
| 937 |
+
"grad_norm": 0.5078125,
|
| 938 |
+
"learning_rate": 3.3599238792081575e-05,
|
| 939 |
+
"logps_logqs/chosen": 0.07161605358123779,
|
| 940 |
+
"logqs/chosen": -0.7662056684494019,
|
| 941 |
+
"logqs/rejected": -0.9423080682754517,
|
| 942 |
+
"logqs_logps/rejected": 0.031152984127402306,
|
| 943 |
+
"loss": 0.1786,
|
| 944 |
+
"step": 840
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 1.3642291102417494,
|
| 948 |
+
"grad_norm": 0.5390625,
|
| 949 |
+
"learning_rate": 3.315931384083431e-05,
|
| 950 |
+
"logps_logqs/chosen": 0.0672772079706192,
|
| 951 |
+
"logqs/chosen": -0.7698653936386108,
|
| 952 |
+
"logqs/rejected": -0.9608929753303528,
|
| 953 |
+
"logqs_logps/rejected": 0.031727343797683716,
|
| 954 |
+
"loss": 0.181,
|
| 955 |
+
"step": 850
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.3802788644798878,
|
| 959 |
+
"grad_norm": 0.5390625,
|
| 960 |
+
"learning_rate": 3.2716542536485474e-05,
|
| 961 |
+
"logps_logqs/chosen": 0.07064563035964966,
|
| 962 |
+
"logqs/chosen": -0.7671880125999451,
|
| 963 |
+
"logqs/rejected": -0.9258206486701965,
|
| 964 |
+
"logqs_logps/rejected": 0.03185782581567764,
|
| 965 |
+
"loss": 0.1802,
|
| 966 |
+
"step": 860
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 1.396328618718026,
|
| 970 |
+
"grad_norm": 0.59375,
|
| 971 |
+
"learning_rate": 3.2271079338527626e-05,
|
| 972 |
+
"logps_logqs/chosen": 0.07001273334026337,
|
| 973 |
+
"logqs/chosen": -0.7942059636116028,
|
| 974 |
+
"logqs/rejected": -0.9803145527839661,
|
| 975 |
+
"logqs_logps/rejected": 0.02900245226919651,
|
| 976 |
+
"loss": 0.1783,
|
| 977 |
+
"step": 870
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 1.412378372956164,
|
| 981 |
+
"grad_norm": 0.48828125,
|
| 982 |
+
"learning_rate": 3.1823079645512655e-05,
|
| 983 |
+
"logps_logqs/chosen": 0.06748739629983902,
|
| 984 |
+
"logqs/chosen": -0.7796913385391235,
|
| 985 |
+
"logqs/rejected": -1.0028189420700073,
|
| 986 |
+
"logqs_logps/rejected": 0.028837621212005615,
|
| 987 |
+
"loss": 0.1791,
|
| 988 |
+
"step": 880
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 1.4284281271943025,
|
| 992 |
+
"grad_norm": 0.5234375,
|
| 993 |
+
"learning_rate": 3.137269974084139e-05,
|
| 994 |
+
"logps_logqs/chosen": 0.07031874358654022,
|
| 995 |
+
"logqs/chosen": -0.7788208723068237,
|
| 996 |
+
"logqs/rejected": -0.9545127153396606,
|
| 997 |
+
"logqs_logps/rejected": 0.029628973454236984,
|
| 998 |
+
"loss": 0.179,
|
| 999 |
+
"step": 890
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 1.4444778814324406,
|
| 1003 |
+
"grad_norm": 0.486328125,
|
| 1004 |
+
"learning_rate": 3.092009673824469e-05,
|
| 1005 |
+
"logps_logqs/chosen": 0.07582792639732361,
|
| 1006 |
+
"logqs/chosen": -0.8078464269638062,
|
| 1007 |
+
"logqs/rejected": -0.9633975028991699,
|
| 1008 |
+
"logqs_logps/rejected": 0.02939186617732048,
|
| 1009 |
+
"loss": 0.1885,
|
| 1010 |
+
"step": 900
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.4605276356705787,
|
| 1014 |
+
"grad_norm": 0.53515625,
|
| 1015 |
+
"learning_rate": 3.0465428526974665e-05,
|
| 1016 |
+
"logps_logqs/chosen": 0.07225798070430756,
|
| 1017 |
+
"logqs/chosen": -0.7663235068321228,
|
| 1018 |
+
"logqs/rejected": -0.9817646741867065,
|
| 1019 |
+
"logqs_logps/rejected": 0.028154581785202026,
|
| 1020 |
+
"loss": 0.178,
|
| 1021 |
+
"step": 910
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 1.4765773899087171,
|
| 1025 |
+
"grad_norm": 0.49609375,
|
| 1026 |
+
"learning_rate": 3.000885371672554e-05,
|
| 1027 |
+
"logps_logqs/chosen": 0.07049473375082016,
|
| 1028 |
+
"logqs/chosen": -0.7796770930290222,
|
| 1029 |
+
"logqs/rejected": -0.9203440546989441,
|
| 1030 |
+
"logqs_logps/rejected": 0.028858328238129616,
|
| 1031 |
+
"loss": 0.1765,
|
| 1032 |
+
"step": 920
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.4926271441468553,
|
| 1036 |
+
"grad_norm": 0.53125,
|
| 1037 |
+
"learning_rate": 2.9550531582303082e-05,
|
| 1038 |
+
"logps_logqs/chosen": 0.07375530898571014,
|
| 1039 |
+
"logqs/chosen": -0.7871274352073669,
|
| 1040 |
+
"logqs/rejected": -0.950308620929718,
|
| 1041 |
+
"logqs_logps/rejected": 0.030121903866529465,
|
| 1042 |
+
"loss": 0.1821,
|
| 1043 |
+
"step": 930
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 1.5086768983849934,
|
| 1047 |
+
"grad_norm": 0.462890625,
|
| 1048 |
+
"learning_rate": 2.909062200806208e-05,
|
| 1049 |
+
"logps_logqs/chosen": 0.06755580753087997,
|
| 1050 |
+
"logqs/chosen": -0.7797432541847229,
|
| 1051 |
+
"logqs/rejected": -0.9579198956489563,
|
| 1052 |
+
"logqs_logps/rejected": 0.027673590928316116,
|
| 1053 |
+
"loss": 0.1738,
|
| 1054 |
+
"step": 940
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 1.5247266526231318,
|
| 1058 |
+
"grad_norm": 0.451171875,
|
| 1059 |
+
"learning_rate": 2.8629285432131083e-05,
|
| 1060 |
+
"logps_logqs/chosen": 0.06972555816173553,
|
| 1061 |
+
"logqs/chosen": -0.7843119502067566,
|
| 1062 |
+
"logqs/rejected": -0.971518874168396,
|
| 1063 |
+
"logqs_logps/rejected": 0.026922276243567467,
|
| 1064 |
+
"loss": 0.1771,
|
| 1065 |
+
"step": 950
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 1.54077640686127,
|
| 1069 |
+
"grad_norm": 0.5234375,
|
| 1070 |
+
"learning_rate": 2.8166682790444116e-05,
|
| 1071 |
+
"logps_logqs/chosen": 0.0729864090681076,
|
| 1072 |
+
"logqs/chosen": -0.7914843559265137,
|
| 1073 |
+
"logqs/rejected": -0.9729417562484741,
|
| 1074 |
+
"logqs_logps/rejected": 0.02641429379582405,
|
| 1075 |
+
"loss": 0.1788,
|
| 1076 |
+
"step": 960
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.556826161099408,
|
| 1080 |
+
"grad_norm": 0.4921875,
|
| 1081 |
+
"learning_rate": 2.7702975460598547e-05,
|
| 1082 |
+
"logps_logqs/chosen": 0.06970744580030441,
|
| 1083 |
+
"logqs/chosen": -0.7867680788040161,
|
| 1084 |
+
"logqs/rejected": -0.9702849388122559,
|
| 1085 |
+
"logqs_logps/rejected": 0.026300692930817604,
|
| 1086 |
+
"loss": 0.1753,
|
| 1087 |
+
"step": 970
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.5728759153375464,
|
| 1091 |
+
"grad_norm": 0.51171875,
|
| 1092 |
+
"learning_rate": 2.723832520555905e-05,
|
| 1093 |
+
"logps_logqs/chosen": 0.06844428926706314,
|
| 1094 |
+
"logqs/chosen": -0.7796690464019775,
|
| 1095 |
+
"logqs/rejected": -0.9721530079841614,
|
| 1096 |
+
"logqs_logps/rejected": 0.03003394976258278,
|
| 1097 |
+
"loss": 0.1775,
|
| 1098 |
+
"step": 980
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 1.5889256695756846,
|
| 1102 |
+
"grad_norm": 0.50390625,
|
| 1103 |
+
"learning_rate": 2.677289411722702e-05,
|
| 1104 |
+
"logps_logqs/chosen": 0.06915868073701859,
|
| 1105 |
+
"logqs/chosen": -0.7836161851882935,
|
| 1106 |
+
"logqs/rejected": -1.0066392421722412,
|
| 1107 |
+
"logqs_logps/rejected": 0.029805105179548264,
|
| 1108 |
+
"loss": 0.1834,
|
| 1109 |
+
"step": 990
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.6049754238138227,
|
| 1113 |
+
"grad_norm": 0.52734375,
|
| 1114 |
+
"learning_rate": 2.63068445598953e-05,
|
| 1115 |
+
"logps_logqs/chosen": 0.07069100439548492,
|
| 1116 |
+
"logqs/chosen": -0.7810501456260681,
|
| 1117 |
+
"logqs/rejected": -0.9559534788131714,
|
| 1118 |
+
"logqs_logps/rejected": 0.02675846591591835,
|
| 1119 |
+
"loss": 0.1799,
|
| 1120 |
+
"step": 1000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 1.6210251780519611,
|
| 1124 |
+
"grad_norm": 0.455078125,
|
| 1125 |
+
"learning_rate": 2.5840339113607854e-05,
|
| 1126 |
+
"logps_logqs/chosen": 0.06957536935806274,
|
| 1127 |
+
"logqs/chosen": -0.776276171207428,
|
| 1128 |
+
"logqs/rejected": -0.9653003811836243,
|
| 1129 |
+
"logqs_logps/rejected": 0.029132988303899765,
|
| 1130 |
+
"loss": 0.1764,
|
| 1131 |
+
"step": 1010
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 1.6370749322900993,
|
| 1135 |
+
"grad_norm": 0.4921875,
|
| 1136 |
+
"learning_rate": 2.53735405174442e-05,
|
| 1137 |
+
"logps_logqs/chosen": 0.06984798610210419,
|
| 1138 |
+
"logqs/chosen": -0.7904434204101562,
|
| 1139 |
+
"logqs/rejected": -0.9736318588256836,
|
| 1140 |
+
"logqs_logps/rejected": 0.02975938282907009,
|
| 1141 |
+
"loss": 0.1778,
|
| 1142 |
+
"step": 1020
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 1.6531246865282374,
|
| 1146 |
+
"grad_norm": 0.50390625,
|
| 1147 |
+
"learning_rate": 2.490661161274835e-05,
|
| 1148 |
+
"logps_logqs/chosen": 0.06985798478126526,
|
| 1149 |
+
"logqs/chosen": -0.7864385843276978,
|
| 1150 |
+
"logqs/rejected": -0.9617294073104858,
|
| 1151 |
+
"logqs_logps/rejected": 0.027367640286684036,
|
| 1152 |
+
"loss": 0.1774,
|
| 1153 |
+
"step": 1030
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 1.6691744407663758,
|
| 1157 |
+
"grad_norm": 0.4453125,
|
| 1158 |
+
"learning_rate": 2.443971528632205e-05,
|
| 1159 |
+
"logps_logqs/chosen": 0.07047738134860992,
|
| 1160 |
+
"logqs/chosen": -0.7980072498321533,
|
| 1161 |
+
"logqs/rejected": -0.9655786752700806,
|
| 1162 |
+
"logqs_logps/rejected": 0.028297554701566696,
|
| 1163 |
+
"loss": 0.1779,
|
| 1164 |
+
"step": 1040
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.6852241950045141,
|
| 1168 |
+
"grad_norm": 0.515625,
|
| 1169 |
+
"learning_rate": 2.3973014413602238e-05,
|
| 1170 |
+
"logps_logqs/chosen": 0.07119600474834442,
|
| 1171 |
+
"logqs/chosen": -0.7848029732704163,
|
| 1172 |
+
"logqs/rejected": -1.003758192062378,
|
| 1173 |
+
"logqs_logps/rejected": 0.029630709439516068,
|
| 1174 |
+
"loss": 0.1808,
|
| 1175 |
+
"step": 1050
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 1.701273949242652,
|
| 1179 |
+
"grad_norm": 0.498046875,
|
| 1180 |
+
"learning_rate": 2.3506671801842364e-05,
|
| 1181 |
+
"logps_logqs/chosen": 0.06773122400045395,
|
| 1182 |
+
"logqs/chosen": -0.8044145703315735,
|
| 1183 |
+
"logqs/rejected": -0.9743335843086243,
|
| 1184 |
+
"logqs_logps/rejected": 0.027200300246477127,
|
| 1185 |
+
"loss": 0.1769,
|
| 1186 |
+
"step": 1060
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.7173237034807904,
|
| 1190 |
+
"grad_norm": 0.4921875,
|
| 1191 |
+
"learning_rate": 2.3040850133317597e-05,
|
| 1192 |
+
"logps_logqs/chosen": 0.07197652757167816,
|
| 1193 |
+
"logqs/chosen": -0.7732763886451721,
|
| 1194 |
+
"logqs/rejected": -0.9615533947944641,
|
| 1195 |
+
"logqs_logps/rejected": 0.02922072447836399,
|
| 1196 |
+
"loss": 0.1801,
|
| 1197 |
+
"step": 1070
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 1.7333734577189288,
|
| 1201 |
+
"grad_norm": 0.48046875,
|
| 1202 |
+
"learning_rate": 2.2575711908573548e-05,
|
| 1203 |
+
"logps_logqs/chosen": 0.06941990554332733,
|
| 1204 |
+
"logqs/chosen": -0.7937701940536499,
|
| 1205 |
+
"logqs/rejected": -1.0073679685592651,
|
| 1206 |
+
"logqs_logps/rejected": 0.026350444182753563,
|
| 1207 |
+
"loss": 0.1764,
|
| 1208 |
+
"step": 1080
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.749423211957067,
|
| 1212 |
+
"grad_norm": 0.447265625,
|
| 1213 |
+
"learning_rate": 2.2111419389738435e-05,
|
| 1214 |
+
"logps_logqs/chosen": 0.07039657980203629,
|
| 1215 |
+
"logqs/chosen": -0.7764157056808472,
|
| 1216 |
+
"logqs/rejected": -0.9544513821601868,
|
| 1217 |
+
"logqs_logps/rejected": 0.02772611379623413,
|
| 1218 |
+
"loss": 0.1755,
|
| 1219 |
+
"step": 1090
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 1.765472966195205,
|
| 1223 |
+
"grad_norm": 0.5390625,
|
| 1224 |
+
"learning_rate": 2.1648134543918423e-05,
|
| 1225 |
+
"logps_logqs/chosen": 0.07292209565639496,
|
| 1226 |
+
"logqs/chosen": -0.7701107859611511,
|
| 1227 |
+
"logqs/rejected": -0.9618105888366699,
|
| 1228 |
+
"logqs_logps/rejected": 0.027484769001603127,
|
| 1229 |
+
"loss": 0.1762,
|
| 1230 |
+
"step": 1100
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 1.7815227204333435,
|
| 1234 |
+
"grad_norm": 0.427734375,
|
| 1235 |
+
"learning_rate": 2.1186018986695842e-05,
|
| 1236 |
+
"logps_logqs/chosen": 0.07149146497249603,
|
| 1237 |
+
"logqs/chosen": -0.7753132581710815,
|
| 1238 |
+
"logqs/rejected": -0.926436722278595,
|
| 1239 |
+
"logqs_logps/rejected": 0.029845744371414185,
|
| 1240 |
+
"loss": 0.1779,
|
| 1241 |
+
"step": 1110
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.7975724746714816,
|
| 1245 |
+
"grad_norm": 0.45703125,
|
| 1246 |
+
"learning_rate": 2.0725233925750063e-05,
|
| 1247 |
+
"logps_logqs/chosen": 0.06752609461545944,
|
| 1248 |
+
"logqs/chosen": -0.7796769738197327,
|
| 1249 |
+
"logqs/rejected": -0.9575828313827515,
|
| 1250 |
+
"logqs_logps/rejected": 0.03261734917759895,
|
| 1251 |
+
"loss": 0.1771,
|
| 1252 |
+
"step": 1120
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 1.8136222289096198,
|
| 1256 |
+
"grad_norm": 0.54296875,
|
| 1257 |
+
"learning_rate": 2.026594010462068e-05,
|
| 1258 |
+
"logps_logqs/chosen": 0.07001613825559616,
|
| 1259 |
+
"logqs/chosen": -0.7918448448181152,
|
| 1260 |
+
"logqs/rejected": -0.9749159812927246,
|
| 1261 |
+
"logqs_logps/rejected": 0.028455784544348717,
|
| 1262 |
+
"loss": 0.1789,
|
| 1263 |
+
"step": 1130
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.8296719831477581,
|
| 1267 |
+
"grad_norm": 0.478515625,
|
| 1268 |
+
"learning_rate": 1.980829774663256e-05,
|
| 1269 |
+
"logps_logqs/chosen": 0.07289810478687286,
|
| 1270 |
+
"logqs/chosen": -0.7838630676269531,
|
| 1271 |
+
"logqs/rejected": -0.9698160290718079,
|
| 1272 |
+
"logqs_logps/rejected": 0.02906595729291439,
|
| 1273 |
+
"loss": 0.1794,
|
| 1274 |
+
"step": 1140
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 1.8457217373858963,
|
| 1278 |
+
"grad_norm": 0.458984375,
|
| 1279 |
+
"learning_rate": 1.9352466499002422e-05,
|
| 1280 |
+
"logps_logqs/chosen": 0.0761161744594574,
|
| 1281 |
+
"logqs/chosen": -0.7819662094116211,
|
| 1282 |
+
"logqs/rejected": -0.9463040232658386,
|
| 1283 |
+
"logqs_logps/rejected": 0.028005197644233704,
|
| 1284 |
+
"loss": 0.183,
|
| 1285 |
+
"step": 1150
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 1.8617714916240344,
|
| 1289 |
+
"grad_norm": 0.474609375,
|
| 1290 |
+
"learning_rate": 1.8898605377146383e-05,
|
| 1291 |
+
"logps_logqs/chosen": 0.06957074254751205,
|
| 1292 |
+
"logqs/chosen": -0.7857328057289124,
|
| 1293 |
+
"logqs/rejected": -0.9639459848403931,
|
| 1294 |
+
"logqs_logps/rejected": 0.027649903669953346,
|
| 1295 |
+
"loss": 0.1753,
|
| 1296 |
+
"step": 1160
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 1.8778212458621728,
|
| 1300 |
+
"grad_norm": 0.48046875,
|
| 1301 |
+
"learning_rate": 1.8446872709207847e-05,
|
| 1302 |
+
"logps_logqs/chosen": 0.07362545281648636,
|
| 1303 |
+
"logqs/chosen": -0.7835830450057983,
|
| 1304 |
+
"logqs/rejected": -1.0045692920684814,
|
| 1305 |
+
"logqs_logps/rejected": 0.027295967563986778,
|
| 1306 |
+
"loss": 0.1806,
|
| 1307 |
+
"step": 1170
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 1.893871000100311,
|
| 1311 |
+
"grad_norm": 0.51171875,
|
| 1312 |
+
"learning_rate": 1.799742608082519e-05,
|
| 1313 |
+
"logps_logqs/chosen": 0.07236044853925705,
|
| 1314 |
+
"logqs/chosen": -0.7885088324546814,
|
| 1315 |
+
"logqs/rejected": -0.9487521052360535,
|
| 1316 |
+
"logqs_logps/rejected": 0.026073191314935684,
|
| 1317 |
+
"loss": 0.1723,
|
| 1318 |
+
"step": 1180
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 1.909920754338449,
|
| 1322 |
+
"grad_norm": 0.51171875,
|
| 1323 |
+
"learning_rate": 1.7550422280158513e-05,
|
| 1324 |
+
"logps_logqs/chosen": 0.0707259327173233,
|
| 1325 |
+
"logqs/chosen": -0.7704340815544128,
|
| 1326 |
+
"logqs/rejected": -0.9094502329826355,
|
| 1327 |
+
"logqs_logps/rejected": 0.027560651302337646,
|
| 1328 |
+
"loss": 0.1758,
|
| 1329 |
+
"step": 1190
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 1.9259705085765875,
|
| 1333 |
+
"grad_norm": 0.435546875,
|
| 1334 |
+
"learning_rate": 1.7106017243194487e-05,
|
| 1335 |
+
"logps_logqs/chosen": 0.07020822167396545,
|
| 1336 |
+
"logqs/chosen": -0.7692683935165405,
|
| 1337 |
+
"logqs/rejected": -0.9656769037246704,
|
| 1338 |
+
"logqs_logps/rejected": 0.02792223170399666,
|
| 1339 |
+
"loss": 0.177,
|
| 1340 |
+
"step": 1200
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.9420202628147256,
|
| 1344 |
+
"grad_norm": 0.53515625,
|
| 1345 |
+
"learning_rate": 1.6664365999348594e-05,
|
| 1346 |
+
"logps_logqs/chosen": 0.06943775713443756,
|
| 1347 |
+
"logqs/chosen": -0.7688643932342529,
|
| 1348 |
+
"logqs/rejected": -0.9713398218154907,
|
| 1349 |
+
"logqs_logps/rejected": 0.026859009638428688,
|
| 1350 |
+
"loss": 0.1748,
|
| 1351 |
+
"step": 1210
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 1.9580700170528638,
|
| 1355 |
+
"grad_norm": 0.51953125,
|
| 1356 |
+
"learning_rate": 1.6225622617383494e-05,
|
| 1357 |
+
"logps_logqs/chosen": 0.07070201635360718,
|
| 1358 |
+
"logqs/chosen": -0.7839328050613403,
|
| 1359 |
+
"logqs/rejected": -0.9712308645248413,
|
| 1360 |
+
"logqs_logps/rejected": 0.028167420998215675,
|
| 1361 |
+
"loss": 0.1763,
|
| 1362 |
+
"step": 1220
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 1.9741197712910021,
|
| 1366 |
+
"grad_norm": 0.4453125,
|
| 1367 |
+
"learning_rate": 1.578994015166263e-05,
|
| 1368 |
+
"logps_logqs/chosen": 0.07289667427539825,
|
| 1369 |
+
"logqs/chosen": -0.7838398814201355,
|
| 1370 |
+
"logqs/rejected": -0.9522368311882019,
|
| 1371 |
+
"logqs_logps/rejected": 0.02768387272953987,
|
| 1372 |
+
"loss": 0.1773,
|
| 1373 |
+
"step": 1230
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 1.9901695255291403,
|
| 1377 |
+
"grad_norm": 0.53515625,
|
| 1378 |
+
"learning_rate": 1.535747058875765e-05,
|
| 1379 |
+
"logps_logqs/chosen": 0.06881529092788696,
|
| 1380 |
+
"logqs/chosen": -0.790834903717041,
|
| 1381 |
+
"logqs/rejected": -0.9497320055961609,
|
| 1382 |
+
"logqs_logps/rejected": 0.029328888282179832,
|
| 1383 |
+
"loss": 0.1734,
|
| 1384 |
+
"step": 1240
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 2.0062192797672784,
|
| 1388 |
+
"grad_norm": 0.43359375,
|
| 1389 |
+
"learning_rate": 1.4928364794428307e-05,
|
| 1390 |
+
"logps_logqs/chosen": 0.07006263732910156,
|
| 1391 |
+
"logqs/chosen": -0.799282431602478,
|
| 1392 |
+
"logqs/rejected": -1.0095001459121704,
|
| 1393 |
+
"logqs_logps/rejected": 0.02767905592918396,
|
| 1394 |
+
"loss": 0.1732,
|
| 1395 |
+
"step": 1250
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 2.022269034005417,
|
| 1399 |
+
"grad_norm": 0.4296875,
|
| 1400 |
+
"learning_rate": 1.4502772460993385e-05,
|
| 1401 |
+
"logps_logqs/chosen": 0.06896007061004639,
|
| 1402 |
+
"logqs/chosen": -0.7676675319671631,
|
| 1403 |
+
"logqs/rejected": -0.9425755739212036,
|
| 1404 |
+
"logqs_logps/rejected": 0.02848168835043907,
|
| 1405 |
+
"loss": 0.1685,
|
| 1406 |
+
"step": 1260
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 2.038318788243555,
|
| 1410 |
+
"grad_norm": 0.462890625,
|
| 1411 |
+
"learning_rate": 1.4080842055110993e-05,
|
| 1412 |
+
"logps_logqs/chosen": 0.06711134314537048,
|
| 1413 |
+
"logqs/chosen": -0.7576900720596313,
|
| 1414 |
+
"logqs/rejected": -0.9405368566513062,
|
| 1415 |
+
"logqs_logps/rejected": 0.02635120414197445,
|
| 1416 |
+
"loss": 0.1689,
|
| 1417 |
+
"step": 1270
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 2.054368542481693,
|
| 1421 |
+
"grad_norm": 0.46484375,
|
| 1422 |
+
"learning_rate": 1.3662720765986341e-05,
|
| 1423 |
+
"logps_logqs/chosen": 0.06902097165584564,
|
| 1424 |
+
"logqs/chosen": -0.788707435131073,
|
| 1425 |
+
"logqs/rejected": -0.9549592733383179,
|
| 1426 |
+
"logqs_logps/rejected": 0.028454547747969627,
|
| 1427 |
+
"loss": 0.1686,
|
| 1428 |
+
"step": 1280
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 2.0704182967198315,
|
| 1432 |
+
"grad_norm": 0.40625,
|
| 1433 |
+
"learning_rate": 1.3248554454025275e-05,
|
| 1434 |
+
"logps_logqs/chosen": 0.07042928040027618,
|
| 1435 |
+
"logqs/chosen": -0.7786656618118286,
|
| 1436 |
+
"logqs/rejected": -0.9685086011886597,
|
| 1437 |
+
"logqs_logps/rejected": 0.025051862001419067,
|
| 1438 |
+
"loss": 0.1673,
|
| 1439 |
+
"step": 1290
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 2.08646805095797,
|
| 1443 |
+
"grad_norm": 0.447265625,
|
| 1444 |
+
"learning_rate": 1.2838487599951243e-05,
|
| 1445 |
+
"logps_logqs/chosen": 0.06794705986976624,
|
| 1446 |
+
"logqs/chosen": -0.7671617269515991,
|
| 1447 |
+
"logqs/rejected": -0.9481566548347473,
|
| 1448 |
+
"logqs_logps/rejected": 0.025887325406074524,
|
| 1449 |
+
"loss": 0.1696,
|
| 1450 |
+
"step": 1300
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 2.1025178051961078,
|
| 1454 |
+
"grad_norm": 0.439453125,
|
| 1455 |
+
"learning_rate": 1.2432663254403638e-05,
|
| 1456 |
+
"logps_logqs/chosen": 0.06914719194173813,
|
| 1457 |
+
"logqs/chosen": -0.8097491264343262,
|
| 1458 |
+
"logqs/rejected": -0.9609512090682983,
|
| 1459 |
+
"logqs_logps/rejected": 0.025332655757665634,
|
| 1460 |
+
"loss": 0.1689,
|
| 1461 |
+
"step": 1310
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 2.118567559434246,
|
| 1465 |
+
"grad_norm": 0.421875,
|
| 1466 |
+
"learning_rate": 1.2031222988034967e-05,
|
| 1467 |
+
"logps_logqs/chosen": 0.06850691139698029,
|
| 1468 |
+
"logqs/chosen": -0.7866081595420837,
|
| 1469 |
+
"logqs/rejected": -0.952921986579895,
|
| 1470 |
+
"logqs_logps/rejected": 0.02708747610449791,
|
| 1471 |
+
"loss": 0.1672,
|
| 1472 |
+
"step": 1320
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 2.1346173136723845,
|
| 1476 |
+
"grad_norm": 0.45703125,
|
| 1477 |
+
"learning_rate": 1.1634306842124423e-05,
|
| 1478 |
+
"logps_logqs/chosen": 0.06721793115139008,
|
| 1479 |
+
"logqs/chosen": -0.7912784218788147,
|
| 1480 |
+
"logqs/rejected": -0.9742089509963989,
|
| 1481 |
+
"logqs_logps/rejected": 0.024986112490296364,
|
| 1482 |
+
"loss": 0.1668,
|
| 1483 |
+
"step": 1330
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 2.1506670679105224,
|
| 1487 |
+
"grad_norm": 0.451171875,
|
| 1488 |
+
"learning_rate": 1.1242053279724762e-05,
|
| 1489 |
+
"logps_logqs/chosen": 0.06856787204742432,
|
| 1490 |
+
"logqs/chosen": -0.7681561708450317,
|
| 1491 |
+
"logqs/rejected": -0.9738146066665649,
|
| 1492 |
+
"logqs_logps/rejected": 0.025526920333504677,
|
| 1493 |
+
"loss": 0.1676,
|
| 1494 |
+
"step": 1340
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 2.166716822148661,
|
| 1498 |
+
"grad_norm": 0.44140625,
|
| 1499 |
+
"learning_rate": 1.0854599137359953e-05,
|
| 1500 |
+
"logps_logqs/chosen": 0.07144749909639359,
|
| 1501 |
+
"logqs/chosen": -0.7849553823471069,
|
| 1502 |
+
"logqs/rejected": -0.9945628046989441,
|
| 1503 |
+
"logqs_logps/rejected": 0.02589735947549343,
|
| 1504 |
+
"loss": 0.1683,
|
| 1505 |
+
"step": 1350
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 2.182766576386799,
|
| 1509 |
+
"grad_norm": 0.3984375,
|
| 1510 |
+
"learning_rate": 1.0472079577290111e-05,
|
| 1511 |
+
"logps_logqs/chosen": 0.06926636397838593,
|
| 1512 |
+
"logqs/chosen": -0.780665934085846,
|
| 1513 |
+
"logqs/rejected": -0.9741228222846985,
|
| 1514 |
+
"logqs_logps/rejected": 0.0255129374563694,
|
| 1515 |
+
"loss": 0.1668,
|
| 1516 |
+
"step": 1360
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 2.198816330624937,
|
| 1520 |
+
"grad_norm": 0.431640625,
|
| 1521 |
+
"learning_rate": 1.009462804036059e-05,
|
| 1522 |
+
"logps_logqs/chosen": 0.06916262209415436,
|
| 1523 |
+
"logqs/chosen": -0.7870660424232483,
|
| 1524 |
+
"logqs/rejected": -0.9848629236221313,
|
| 1525 |
+
"logqs_logps/rejected": 0.024935439229011536,
|
| 1526 |
+
"loss": 0.1688,
|
| 1527 |
+
"step": 1370
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 2.2148660848630755,
|
| 1531 |
+
"grad_norm": 0.48046875,
|
| 1532 |
+
"learning_rate": 9.722376199451436e-06,
|
| 1533 |
+
"logps_logqs/chosen": 0.06875228136777878,
|
| 1534 |
+
"logqs/chosen": -0.7785555124282837,
|
| 1535 |
+
"logqs/rejected": -0.9765084981918335,
|
| 1536 |
+
"logqs_logps/rejected": 0.02562164142727852,
|
| 1537 |
+
"loss": 0.1677,
|
| 1538 |
+
"step": 1380
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 2.230915839101214,
|
| 1542 |
+
"grad_norm": 0.40234375,
|
| 1543 |
+
"learning_rate": 9.35545391354378e-06,
|
| 1544 |
+
"logps_logqs/chosen": 0.06958961486816406,
|
| 1545 |
+
"logqs/chosen": -0.7699087262153625,
|
| 1546 |
+
"logqs/rejected": -0.9525764584541321,
|
| 1547 |
+
"logqs_logps/rejected": 0.027162248268723488,
|
| 1548 |
+
"loss": 0.1672,
|
| 1549 |
+
"step": 1390
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 2.246965593339352,
|
| 1553 |
+
"grad_norm": 0.416015625,
|
| 1554 |
+
"learning_rate": 8.993989182418824e-06,
|
| 1555 |
+
"logps_logqs/chosen": 0.0719502717256546,
|
| 1556 |
+
"logqs/chosen": -0.7669367790222168,
|
| 1557 |
+
"logqs/rejected": -0.9486163258552551,
|
| 1558 |
+
"logqs_logps/rejected": 0.02480030246078968,
|
| 1559 |
+
"loss": 0.1683,
|
| 1560 |
+
"step": 1400
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 2.26301534757749,
|
| 1564 |
+
"grad_norm": 0.48046875,
|
| 1565 |
+
"learning_rate": 8.63810810200556e-06,
|
| 1566 |
+
"logps_logqs/chosen": 0.06904618442058563,
|
| 1567 |
+
"logqs/chosen": -0.7622929811477661,
|
| 1568 |
+
"logqs/rejected": -0.9743655920028687,
|
| 1569 |
+
"logqs_logps/rejected": 0.026619747281074524,
|
| 1570 |
+
"loss": 0.1678,
|
| 1571 |
+
"step": 1410
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 2.2790651018156285,
|
| 1575 |
+
"grad_norm": 0.40234375,
|
| 1576 |
+
"learning_rate": 8.287934820392498e-06,
|
| 1577 |
+
"logps_logqs/chosen": 0.06923404335975647,
|
| 1578 |
+
"logqs/chosen": -0.7849544286727905,
|
| 1579 |
+
"logqs/rejected": -0.9630452990531921,
|
| 1580 |
+
"logqs_logps/rejected": 0.023480530828237534,
|
| 1581 |
+
"loss": 0.1651,
|
| 1582 |
+
"step": 1420
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"epoch": 2.295114856053767,
|
| 1586 |
+
"grad_norm": 0.4140625,
|
| 1587 |
+
"learning_rate": 7.943591494519016e-06,
|
| 1588 |
+
"logps_logqs/chosen": 0.06750839948654175,
|
| 1589 |
+
"logqs/chosen": -0.764995276927948,
|
| 1590 |
+
"logqs/rejected": -0.9358587265014648,
|
| 1591 |
+
"logqs_logps/rejected": 0.027208849787712097,
|
| 1592 |
+
"loss": 0.1667,
|
| 1593 |
+
"step": 1430
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 2.311164610291905,
|
| 1597 |
+
"grad_norm": 0.40625,
|
| 1598 |
+
"learning_rate": 7.60519824756124e-06,
|
| 1599 |
+
"logps_logqs/chosen": 0.06875176727771759,
|
| 1600 |
+
"logqs/chosen": -0.8004514575004578,
|
| 1601 |
+
"logqs/rejected": -1.0356009006500244,
|
| 1602 |
+
"logqs_logps/rejected": 0.0247772429138422,
|
| 1603 |
+
"loss": 0.1681,
|
| 1604 |
+
"step": 1440
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 2.327214364530043,
|
| 1608 |
+
"grad_norm": 0.41015625,
|
| 1609 |
+
"learning_rate": 7.272873127027449e-06,
|
| 1610 |
+
"logps_logqs/chosen": 0.06980612874031067,
|
| 1611 |
+
"logqs/chosen": -0.7612735033035278,
|
| 1612 |
+
"logqs/rejected": -0.9591197967529297,
|
| 1613 |
+
"logqs_logps/rejected": 0.0231007132679224,
|
| 1614 |
+
"loss": 0.1652,
|
| 1615 |
+
"step": 1450
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 2.3432641187681815,
|
| 1619 |
+
"grad_norm": 0.41796875,
|
| 1620 |
+
"learning_rate": 6.946732063577488e-06,
|
| 1621 |
+
"logps_logqs/chosen": 0.0690704956650734,
|
| 1622 |
+
"logqs/chosen": -0.7698851823806763,
|
| 1623 |
+
"logqs/rejected": -0.949694037437439,
|
| 1624 |
+
"logqs_logps/rejected": 0.023757826536893845,
|
| 1625 |
+
"loss": 0.1655,
|
| 1626 |
+
"step": 1460
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 2.3593138730063195,
|
| 1630 |
+
"grad_norm": 0.4375,
|
| 1631 |
+
"learning_rate": 6.6268888305807296e-06,
|
| 1632 |
+
"logps_logqs/chosen": 0.06597896665334702,
|
| 1633 |
+
"logqs/chosen": -0.7630634903907776,
|
| 1634 |
+
"logqs/rejected": -0.9974457025527954,
|
| 1635 |
+
"logqs_logps/rejected": 0.02252907119691372,
|
| 1636 |
+
"loss": 0.1648,
|
| 1637 |
+
"step": 1470
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 2.375363627244458,
|
| 1641 |
+
"grad_norm": 0.404296875,
|
| 1642 |
+
"learning_rate": 6.313455004426577e-06,
|
| 1643 |
+
"logps_logqs/chosen": 0.06898193061351776,
|
| 1644 |
+
"logqs/chosen": -0.77290940284729,
|
| 1645 |
+
"logqs/rejected": -0.9675156474113464,
|
| 1646 |
+
"logqs_logps/rejected": 0.024294773116707802,
|
| 1647 |
+
"loss": 0.166,
|
| 1648 |
+
"step": 1480
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 2.391413381482596,
|
| 1652 |
+
"grad_norm": 0.443359375,
|
| 1653 |
+
"learning_rate": 6.0065399256013204e-06,
|
| 1654 |
+
"logps_logqs/chosen": 0.0698733851313591,
|
| 1655 |
+
"logqs/chosen": -0.761997401714325,
|
| 1656 |
+
"logqs/rejected": -0.9315551519393921,
|
| 1657 |
+
"logqs_logps/rejected": 0.02528567612171173,
|
| 1658 |
+
"loss": 0.1696,
|
| 1659 |
+
"step": 1490
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 2.407463135720734,
|
| 1663 |
+
"grad_norm": 0.423828125,
|
| 1664 |
+
"learning_rate": 5.7062506605450454e-06,
|
| 1665 |
+
"logps_logqs/chosen": 0.06849467009305954,
|
| 1666 |
+
"logqs/chosen": -0.7781127095222473,
|
| 1667 |
+
"logqs/rejected": -0.9850066304206848,
|
| 1668 |
+
"logqs_logps/rejected": 0.026143735274672508,
|
| 1669 |
+
"loss": 0.1703,
|
| 1670 |
+
"step": 1500
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 2.4235128899588725,
|
| 1674 |
+
"grad_norm": 0.443359375,
|
| 1675 |
+
"learning_rate": 5.412691964301827e-06,
|
| 1676 |
+
"logps_logqs/chosen": 0.06971971690654755,
|
| 1677 |
+
"logqs/chosen": -0.7925983667373657,
|
| 1678 |
+
"logqs/rejected": -0.9926835894584656,
|
| 1679 |
+
"logqs_logps/rejected": 0.025326719507575035,
|
| 1680 |
+
"loss": 0.1685,
|
| 1681 |
+
"step": 1510
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 2.439562644197011,
|
| 1685 |
+
"grad_norm": 0.470703125,
|
| 1686 |
+
"learning_rate": 5.125966243976218e-06,
|
| 1687 |
+
"logps_logqs/chosen": 0.07100898772478104,
|
| 1688 |
+
"logqs/chosen": -0.7900832891464233,
|
| 1689 |
+
"logqs/rejected": -0.9629155993461609,
|
| 1690 |
+
"logqs_logps/rejected": 0.025052938610315323,
|
| 1691 |
+
"loss": 0.1693,
|
| 1692 |
+
"step": 1520
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 2.455612398435149,
|
| 1696 |
+
"grad_norm": 0.453125,
|
| 1697 |
+
"learning_rate": 4.846173523008824e-06,
|
| 1698 |
+
"logps_logqs/chosen": 0.06859283149242401,
|
| 1699 |
+
"logqs/chosen": -0.7894052267074585,
|
| 1700 |
+
"logqs/rejected": -0.9897419810295105,
|
| 1701 |
+
"logqs_logps/rejected": 0.02611861191689968,
|
| 1702 |
+
"loss": 0.1683,
|
| 1703 |
+
"step": 1530
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 2.471662152673287,
|
| 1707 |
+
"grad_norm": 0.439453125,
|
| 1708 |
+
"learning_rate": 4.573411406283409e-06,
|
| 1709 |
+
"logps_logqs/chosen": 0.07065166532993317,
|
| 1710 |
+
"logqs/chosen": -0.8073331117630005,
|
| 1711 |
+
"logqs/rejected": -0.9767266511917114,
|
| 1712 |
+
"logqs_logps/rejected": 0.0274626724421978,
|
| 1713 |
+
"loss": 0.1756,
|
| 1714 |
+
"step": 1540
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 2.4877119069114255,
|
| 1718 |
+
"grad_norm": 0.44140625,
|
| 1719 |
+
"learning_rate": 4.307775046077739e-06,
|
| 1720 |
+
"logps_logqs/chosen": 0.0715617686510086,
|
| 1721 |
+
"logqs/chosen": -0.8018879890441895,
|
| 1722 |
+
"logqs/rejected": -0.9990041851997375,
|
| 1723 |
+
"logqs_logps/rejected": 0.026406193152070045,
|
| 1724 |
+
"loss": 0.1719,
|
| 1725 |
+
"step": 1550
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 2.5037616611495634,
|
| 1729 |
+
"grad_norm": 0.46875,
|
| 1730 |
+
"learning_rate": 4.049357108869964e-06,
|
| 1731 |
+
"logps_logqs/chosen": 0.06756819784641266,
|
| 1732 |
+
"logqs/chosen": -0.7610915899276733,
|
| 1733 |
+
"logqs/rejected": -0.972091794013977,
|
| 1734 |
+
"logqs_logps/rejected": 0.026423901319503784,
|
| 1735 |
+
"loss": 0.1683,
|
| 1736 |
+
"step": 1560
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 2.519811415387702,
|
| 1740 |
+
"grad_norm": 0.439453125,
|
| 1741 |
+
"learning_rate": 3.798247743012201e-06,
|
| 1742 |
+
"logps_logqs/chosen": 0.06909768283367157,
|
| 1743 |
+
"logqs/chosen": -0.7960779666900635,
|
| 1744 |
+
"logqs/rejected": -0.9779514074325562,
|
| 1745 |
+
"logqs_logps/rejected": 0.0239783376455307,
|
| 1746 |
+
"loss": 0.1662,
|
| 1747 |
+
"step": 1570
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 2.53586116962584,
|
| 1751 |
+
"grad_norm": 0.43359375,
|
| 1752 |
+
"learning_rate": 3.554534547282512e-06,
|
| 1753 |
+
"logps_logqs/chosen": 0.07034210860729218,
|
| 1754 |
+
"logqs/chosen": -0.7767470479011536,
|
| 1755 |
+
"logqs/rejected": -0.9516876339912415,
|
| 1756 |
+
"logqs_logps/rejected": 0.028661763295531273,
|
| 1757 |
+
"loss": 0.17,
|
| 1758 |
+
"step": 1580
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 2.5519109238639786,
|
| 1762 |
+
"grad_norm": 0.404296875,
|
| 1763 |
+
"learning_rate": 3.318302540326343e-06,
|
| 1764 |
+
"logps_logqs/chosen": 0.07079549133777618,
|
| 1765 |
+
"logqs/chosen": -0.777116596698761,
|
| 1766 |
+
"logqs/rejected": -0.9669076204299927,
|
| 1767 |
+
"logqs_logps/rejected": 0.026164010167121887,
|
| 1768 |
+
"loss": 0.1723,
|
| 1769 |
+
"step": 1590
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 2.5679606781021165,
|
| 1773 |
+
"grad_norm": 0.431640625,
|
| 1774 |
+
"learning_rate": 3.089634130998026e-06,
|
| 1775 |
+
"logps_logqs/chosen": 0.06778942793607712,
|
| 1776 |
+
"logqs/chosen": -0.7698632478713989,
|
| 1777 |
+
"logqs/rejected": -0.9828590154647827,
|
| 1778 |
+
"logqs_logps/rejected": 0.02352612093091011,
|
| 1779 |
+
"loss": 0.166,
|
| 1780 |
+
"step": 1600
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 2.584010432340255,
|
| 1784 |
+
"grad_norm": 0.4296875,
|
| 1785 |
+
"learning_rate": 2.8686090896126587e-06,
|
| 1786 |
+
"logps_logqs/chosen": 0.06593993306159973,
|
| 1787 |
+
"logqs/chosen": -0.7838481664657593,
|
| 1788 |
+
"logqs/rejected": -0.9900785684585571,
|
| 1789 |
+
"logqs_logps/rejected": 0.023241404443979263,
|
| 1790 |
+
"loss": 0.1633,
|
| 1791 |
+
"step": 1610
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 2.6000601865783928,
|
| 1795 |
+
"grad_norm": 0.439453125,
|
| 1796 |
+
"learning_rate": 2.655304520118482e-06,
|
| 1797 |
+
"logps_logqs/chosen": 0.0693887323141098,
|
| 1798 |
+
"logqs/chosen": -0.778706431388855,
|
| 1799 |
+
"logqs/rejected": -0.9755508303642273,
|
| 1800 |
+
"logqs_logps/rejected": 0.02637295424938202,
|
| 1801 |
+
"loss": 0.1664,
|
| 1802 |
+
"step": 1620
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.616109940816531,
|
| 1806 |
+
"grad_norm": 0.44140625,
|
| 1807 |
+
"learning_rate": 2.44979483319939e-06,
|
| 1808 |
+
"logps_logqs/chosen": 0.06880888342857361,
|
| 1809 |
+
"logqs/chosen": -0.7670097351074219,
|
| 1810 |
+
"logqs/rejected": -1.0248219966888428,
|
| 1811 |
+
"logqs_logps/rejected": 0.025341719388961792,
|
| 1812 |
+
"loss": 0.1667,
|
| 1813 |
+
"step": 1630
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 2.6321596950546695,
|
| 1817 |
+
"grad_norm": 0.44140625,
|
| 1818 |
+
"learning_rate": 2.252151720316964e-06,
|
| 1819 |
+
"logps_logqs/chosen": 0.0697592943906784,
|
| 1820 |
+
"logqs/chosen": -0.7662806510925293,
|
| 1821 |
+
"logqs/rejected": -0.9526630640029907,
|
| 1822 |
+
"logqs_logps/rejected": 0.024268481880426407,
|
| 1823 |
+
"loss": 0.1672,
|
| 1824 |
+
"step": 1640
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 2.648209449292808,
|
| 1828 |
+
"grad_norm": 0.458984375,
|
| 1829 |
+
"learning_rate": 2.0624441287011213e-06,
|
| 1830 |
+
"logps_logqs/chosen": 0.06959348171949387,
|
| 1831 |
+
"logqs/chosen": -0.7930929064750671,
|
| 1832 |
+
"logqs/rejected": -0.9900212287902832,
|
| 1833 |
+
"logqs_logps/rejected": 0.02552485466003418,
|
| 1834 |
+
"loss": 0.1705,
|
| 1835 |
+
"step": 1650
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 2.664259203530946,
|
| 1839 |
+
"grad_norm": 0.419921875,
|
| 1840 |
+
"learning_rate": 1.8807382372980687e-06,
|
| 1841 |
+
"logps_logqs/chosen": 0.06878109276294708,
|
| 1842 |
+
"logqs/chosen": -0.776373565196991,
|
| 1843 |
+
"logqs/rejected": -0.9679840207099915,
|
| 1844 |
+
"logqs_logps/rejected": 0.024887990206480026,
|
| 1845 |
+
"loss": 0.1656,
|
| 1846 |
+
"step": 1660
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 2.680308957769084,
|
| 1850 |
+
"grad_norm": 0.431640625,
|
| 1851 |
+
"learning_rate": 1.7070974336839796e-06,
|
| 1852 |
+
"logps_logqs/chosen": 0.06730766594409943,
|
| 1853 |
+
"logqs/chosen": -0.7652665972709656,
|
| 1854 |
+
"logqs/rejected": -0.9662426114082336,
|
| 1855 |
+
"logqs_logps/rejected": 0.026059061288833618,
|
| 1856 |
+
"loss": 0.1667,
|
| 1857 |
+
"step": 1670
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 2.6963587120072225,
|
| 1861 |
+
"grad_norm": 0.453125,
|
| 1862 |
+
"learning_rate": 1.541582291952401e-06,
|
| 1863 |
+
"logps_logqs/chosen": 0.06985867768526077,
|
| 1864 |
+
"logqs/chosen": -0.7848519086837769,
|
| 1865 |
+
"logqs/rejected": -0.9851476550102234,
|
| 1866 |
+
"logqs_logps/rejected": 0.02311038039624691,
|
| 1867 |
+
"loss": 0.1688,
|
| 1868 |
+
"step": 1680
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 2.7124084662453605,
|
| 1872 |
+
"grad_norm": 0.4375,
|
| 1873 |
+
"learning_rate": 1.38425055158318e-06,
|
| 1874 |
+
"logps_logqs/chosen": 0.0703495591878891,
|
| 1875 |
+
"logqs/chosen": -0.7814745903015137,
|
| 1876 |
+
"logqs/rejected": -0.9425519108772278,
|
| 1877 |
+
"logqs_logps/rejected": 0.026310011744499207,
|
| 1878 |
+
"loss": 0.1639,
|
| 1879 |
+
"step": 1690
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.728458220483499,
|
| 1883 |
+
"grad_norm": 0.44921875,
|
| 1884 |
+
"learning_rate": 1.235157097300188e-06,
|
| 1885 |
+
"logps_logqs/chosen": 0.06932912766933441,
|
| 1886 |
+
"logqs/chosen": -0.7783406972885132,
|
| 1887 |
+
"logqs/rejected": -0.9719565510749817,
|
| 1888 |
+
"logqs_logps/rejected": 0.025869470089673996,
|
| 1889 |
+
"loss": 0.1671,
|
| 1890 |
+
"step": 1700
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"epoch": 2.744507974721637,
|
| 1894 |
+
"grad_norm": 0.40625,
|
| 1895 |
+
"learning_rate": 1.0943539399249635e-06,
|
| 1896 |
+
"logps_logqs/chosen": 0.07044648379087448,
|
| 1897 |
+
"logqs/chosen": -0.776070237159729,
|
| 1898 |
+
"logqs/rejected": -0.9304519891738892,
|
| 1899 |
+
"logqs_logps/rejected": 0.02800569497048855,
|
| 1900 |
+
"loss": 0.169,
|
| 1901 |
+
"step": 1710
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 2.7605577289597756,
|
| 1905 |
+
"grad_norm": 0.4609375,
|
| 1906 |
+
"learning_rate": 9.618901982328704e-07,
|
| 1907 |
+
"logps_logqs/chosen": 0.06949154287576675,
|
| 1908 |
+
"logqs/chosen": -0.7787348031997681,
|
| 1909 |
+
"logqs/rejected": -0.9538747668266296,
|
| 1910 |
+
"logqs_logps/rejected": 0.027285417541861534,
|
| 1911 |
+
"loss": 0.169,
|
| 1912 |
+
"step": 1720
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 2.7766074831979135,
|
| 1916 |
+
"grad_norm": 0.427734375,
|
| 1917 |
+
"learning_rate": 8.378120818181707e-07,
|
| 1918 |
+
"logps_logqs/chosen": 0.07005327194929123,
|
| 1919 |
+
"logqs/chosen": -0.7800394296646118,
|
| 1920 |
+
"logqs/rejected": -0.9388517141342163,
|
| 1921 |
+
"logqs_logps/rejected": 0.024246862158179283,
|
| 1922 |
+
"loss": 0.1638,
|
| 1923 |
+
"step": 1730
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 2.792657237436052,
|
| 1927 |
+
"grad_norm": 0.39453125,
|
| 1928 |
+
"learning_rate": 7.221628749739223e-07,
|
| 1929 |
+
"logps_logqs/chosen": 0.07048022001981735,
|
| 1930 |
+
"logqs/chosen": -0.781145453453064,
|
| 1931 |
+
"logqs/rejected": -0.9654680490493774,
|
| 1932 |
+
"logqs_logps/rejected": 0.025187021121382713,
|
| 1933 |
+
"loss": 0.1646,
|
| 1934 |
+
"step": 1740
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 2.80870699167419,
|
| 1938 |
+
"grad_norm": 0.41796875,
|
| 1939 |
+
"learning_rate": 6.149829215924025e-07,
|
| 1940 |
+
"logps_logqs/chosen": 0.06936784833669662,
|
| 1941 |
+
"logqs/chosen": -0.7857106328010559,
|
| 1942 |
+
"logqs/rejected": -1.0071794986724854,
|
| 1943 |
+
"logqs_logps/rejected": 0.02525334618985653,
|
| 1944 |
+
"loss": 0.1685,
|
| 1945 |
+
"step": 1750
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 2.824756745912328,
|
| 1949 |
+
"grad_norm": 0.396484375,
|
| 1950 |
+
"learning_rate": 5.163096110912368e-07,
|
| 1951 |
+
"logps_logqs/chosen": 0.0710034891963005,
|
| 1952 |
+
"logqs/chosen": -0.7809524536132812,
|
| 1953 |
+
"logqs/rejected": -0.9730299711227417,
|
| 1954 |
+
"logqs_logps/rejected": 0.02650505304336548,
|
| 1955 |
+
"loss": 0.1685,
|
| 1956 |
+
"step": 1760
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.8408065001504665,
|
| 1960 |
+
"grad_norm": 0.419921875,
|
| 1961 |
+
"learning_rate": 4.261773653702089e-07,
|
| 1962 |
+
"logps_logqs/chosen": 0.07018028199672699,
|
| 1963 |
+
"logqs/chosen": -0.7636402249336243,
|
| 1964 |
+
"logqs/rejected": -0.9362967610359192,
|
| 1965 |
+
"logqs_logps/rejected": 0.02620730921626091,
|
| 1966 |
+
"loss": 0.1674,
|
| 1967 |
+
"step": 1770
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"epoch": 2.856856254388605,
|
| 1971 |
+
"grad_norm": 0.5859375,
|
| 1972 |
+
"learning_rate": 3.4461762680329803e-07,
|
| 1973 |
+
"logps_logqs/chosen": 0.06559871137142181,
|
| 1974 |
+
"logqs/chosen": -0.7775349617004395,
|
| 1975 |
+
"logqs/rejected": -0.9631088376045227,
|
| 1976 |
+
"logqs_logps/rejected": 0.024396821856498718,
|
| 1977 |
+
"loss": 0.1651,
|
| 1978 |
+
"step": 1780
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 2.872906008626743,
|
| 1982 |
+
"grad_norm": 0.419921875,
|
| 1983 |
+
"learning_rate": 2.716588472700815e-07,
|
| 1984 |
+
"logps_logqs/chosen": 0.07039310038089752,
|
| 1985 |
+
"logqs/chosen": -0.7576441168785095,
|
| 1986 |
+
"logqs/rejected": -0.9705005884170532,
|
| 1987 |
+
"logqs_logps/rejected": 0.026445040479302406,
|
| 1988 |
+
"loss": 0.1646,
|
| 1989 |
+
"step": 1790
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 2.888955762864881,
|
| 1993 |
+
"grad_norm": 0.39453125,
|
| 1994 |
+
"learning_rate": 2.0732647823038243e-07,
|
| 1995 |
+
"logps_logqs/chosen": 0.06968151032924652,
|
| 1996 |
+
"logqs/chosen": -0.7765523791313171,
|
| 1997 |
+
"logqs/rejected": -0.9748941659927368,
|
| 1998 |
+
"logqs_logps/rejected": 0.02464660070836544,
|
| 1999 |
+
"loss": 0.1665,
|
| 2000 |
+
"step": 1800
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 2.905005517103019,
|
| 2004 |
+
"grad_norm": 0.458984375,
|
| 2005 |
+
"learning_rate": 1.5164296184560222e-07,
|
| 2006 |
+
"logps_logqs/chosen": 0.06898736953735352,
|
| 2007 |
+
"logqs/chosen": -0.781166672706604,
|
| 2008 |
+
"logqs/rejected": -0.9676758646965027,
|
| 2009 |
+
"logqs_logps/rejected": 0.02475111000239849,
|
| 2010 |
+
"loss": 0.1665,
|
| 2011 |
+
"step": 1810
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 2.9210552713411575,
|
| 2015 |
+
"grad_norm": 0.404296875,
|
| 2016 |
+
"learning_rate": 1.0462772314983882e-07,
|
| 2017 |
+
"logps_logqs/chosen": 0.06931523233652115,
|
| 2018 |
+
"logqs/chosen": -0.7759555578231812,
|
| 2019 |
+
"logqs/rejected": -0.9690700769424438,
|
| 2020 |
+
"logqs_logps/rejected": 0.024959508329629898,
|
| 2021 |
+
"loss": 0.1706,
|
| 2022 |
+
"step": 1820
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 2.937105025579296,
|
| 2026 |
+
"grad_norm": 0.4609375,
|
| 2027 |
+
"learning_rate": 6.62971632735182e-08,
|
| 2028 |
+
"logps_logqs/chosen": 0.07144445180892944,
|
| 2029 |
+
"logqs/chosen": -0.7924820780754089,
|
| 2030 |
+
"logqs/rejected": -0.9925470352172852,
|
| 2031 |
+
"logqs_logps/rejected": 0.0241483636200428,
|
| 2032 |
+
"loss": 0.1691,
|
| 2033 |
+
"step": 1830
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.9531547798174342,
|
| 2037 |
+
"grad_norm": 0.435546875,
|
| 2038 |
+
"learning_rate": 3.666465372190453e-08,
|
| 2039 |
+
"logps_logqs/chosen": 0.07032831013202667,
|
| 2040 |
+
"logqs/chosen": -0.7886664271354675,
|
| 2041 |
+
"logqs/rejected": -1.0002458095550537,
|
| 2042 |
+
"logqs_logps/rejected": 0.024760346859693527,
|
| 2043 |
+
"loss": 0.1681,
|
| 2044 |
+
"step": 1840
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 2.969204534055572,
|
| 2048 |
+
"grad_norm": 0.431640625,
|
| 2049 |
+
"learning_rate": 1.57405317104925e-08,
|
| 2050 |
+
"logps_logqs/chosen": 0.068773552775383,
|
| 2051 |
+
"logqs/chosen": -0.7885466814041138,
|
| 2052 |
+
"logqs/rejected": -0.9979216456413269,
|
| 2053 |
+
"logqs_logps/rejected": 0.023049544543027878,
|
| 2054 |
+
"loss": 0.1691,
|
| 2055 |
+
"step": 1850
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 2.9852542882937105,
|
| 2059 |
+
"grad_norm": 0.427734375,
|
| 2060 |
+
"learning_rate": 3.532096558903075e-09,
|
| 2061 |
+
"logps_logqs/chosen": 0.06932573765516281,
|
| 2062 |
+
"logqs/chosen": -0.7747775912284851,
|
| 2063 |
+
"logqs/rejected": -0.9474186897277832,
|
| 2064 |
+
"logqs_logps/rejected": 0.026311378926038742,
|
| 2065 |
+
"loss": 0.1678,
|
| 2066 |
+
"step": 1860
|
| 2067 |
+
}
|
| 2068 |
+
],
|
| 2069 |
+
"logging_steps": 10,
|
| 2070 |
+
"max_steps": 1869,
|
| 2071 |
+
"num_input_tokens_seen": 0,
|
| 2072 |
+
"num_train_epochs": 3,
|
| 2073 |
+
"save_steps": 500,
|
| 2074 |
+
"stateful_callbacks": {
|
| 2075 |
+
"TrainerControl": {
|
| 2076 |
+
"args": {
|
| 2077 |
+
"should_epoch_stop": false,
|
| 2078 |
+
"should_evaluate": false,
|
| 2079 |
+
"should_log": false,
|
| 2080 |
+
"should_save": true,
|
| 2081 |
+
"should_training_stop": true
|
| 2082 |
+
},
|
| 2083 |
+
"attributes": {}
|
| 2084 |
+
}
|
| 2085 |
+
},
|
| 2086 |
+
"total_flos": 0.0,
|
| 2087 |
+
"train_batch_size": 2,
|
| 2088 |
+
"trial_name": null,
|
| 2089 |
+
"trial_params": null
|
| 2090 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-1869/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b9fa66895bee5a36a6f750ef897f491a1113f2f79af77c75bdef8187ae10031
|
| 3 |
+
size 6840
|
qwen2.5-1.5B-it-distillm2/checkpoint-1869/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qwen2.5-1.5B-it-distillm2/checkpoint-623/added_tokens.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</tool_call>": 151658,
|
| 3 |
+
"<tool_call>": 151657,
|
| 4 |
+
"<|box_end|>": 151649,
|
| 5 |
+
"<|box_start|>": 151648,
|
| 6 |
+
"<|endoftext|>": 151643,
|
| 7 |
+
"<|file_sep|>": 151664,
|
| 8 |
+
"<|fim_middle|>": 151660,
|
| 9 |
+
"<|fim_pad|>": 151662,
|
| 10 |
+
"<|fim_prefix|>": 151659,
|
| 11 |
+
"<|fim_suffix|>": 151661,
|
| 12 |
+
"<|im_end|>": 151645,
|
| 13 |
+
"<|im_start|>": 151644,
|
| 14 |
+
"<|image_pad|>": 151655,
|
| 15 |
+
"<|object_ref_end|>": 151647,
|
| 16 |
+
"<|object_ref_start|>": 151646,
|
| 17 |
+
"<|quad_end|>": 151651,
|
| 18 |
+
"<|quad_start|>": 151650,
|
| 19 |
+
"<|repo_name|>": 151663,
|
| 20 |
+
"<|video_pad|>": 151656,
|
| 21 |
+
"<|vision_end|>": 151653,
|
| 22 |
+
"<|vision_pad|>": 151654,
|
| 23 |
+
"<|vision_start|>": 151652
|
| 24 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-623/config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Qwen/Qwen2.5-1.5B-Instruct",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"Qwen2ForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": 151645,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 1536,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 8960,
|
| 13 |
+
"max_position_embeddings": 32768,
|
| 14 |
+
"max_window_layers": 21,
|
| 15 |
+
"model_type": "qwen2",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 28,
|
| 18 |
+
"num_key_value_heads": 2,
|
| 19 |
+
"rms_norm_eps": 1e-06,
|
| 20 |
+
"rope_scaling": null,
|
| 21 |
+
"rope_theta": 1000000.0,
|
| 22 |
+
"sliding_window": null,
|
| 23 |
+
"tie_word_embeddings": true,
|
| 24 |
+
"torch_dtype": "bfloat16",
|
| 25 |
+
"transformers_version": "4.45.2",
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"use_sliding_window": false,
|
| 28 |
+
"vocab_size": 152064
|
| 29 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-623/generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.45.2"
|
| 14 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-623/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qwen2.5-1.5B-it-distillm2/checkpoint-623/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce10fdb4c6ab7af484cb56ecdbf69a28f2b7f20797c260fe958442b97fc57848
|
| 3 |
+
size 3087935634
|
qwen2.5-1.5B-it-distillm2/checkpoint-623/special_tokens_map.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>"
|
| 16 |
+
],
|
| 17 |
+
"bos_token": "<|endoftext|>",
|
| 18 |
+
"eos_token": {
|
| 19 |
+
"content": "<|im_end|>",
|
| 20 |
+
"lstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"rstrip": false,
|
| 23 |
+
"single_word": false
|
| 24 |
+
},
|
| 25 |
+
"pad_token": {
|
| 26 |
+
"content": "<|endoftext|>",
|
| 27 |
+
"lstrip": false,
|
| 28 |
+
"normalized": false,
|
| 29 |
+
"rstrip": false,
|
| 30 |
+
"single_word": false
|
| 31 |
+
}
|
| 32 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-623/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
|
| 3 |
+
size 11421896
|
qwen2.5-1.5B-it-distillm2/checkpoint-623/tokenizer_config.json
ADDED
|
@@ -0,0 +1,207 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
}
|
| 181 |
+
},
|
| 182 |
+
"additional_special_tokens": [
|
| 183 |
+
"<|im_start|>",
|
| 184 |
+
"<|im_end|>",
|
| 185 |
+
"<|object_ref_start|>",
|
| 186 |
+
"<|object_ref_end|>",
|
| 187 |
+
"<|box_start|>",
|
| 188 |
+
"<|box_end|>",
|
| 189 |
+
"<|quad_start|>",
|
| 190 |
+
"<|quad_end|>",
|
| 191 |
+
"<|vision_start|>",
|
| 192 |
+
"<|vision_end|>",
|
| 193 |
+
"<|vision_pad|>",
|
| 194 |
+
"<|image_pad|>",
|
| 195 |
+
"<|video_pad|>"
|
| 196 |
+
],
|
| 197 |
+
"bos_token": "<|endoftext|>",
|
| 198 |
+
"chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
|
| 199 |
+
"clean_up_tokenization_spaces": false,
|
| 200 |
+
"eos_token": "<|im_end|>",
|
| 201 |
+
"errors": "replace",
|
| 202 |
+
"model_max_length": 2048,
|
| 203 |
+
"pad_token": "<|endoftext|>",
|
| 204 |
+
"split_special_tokens": false,
|
| 205 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 206 |
+
"unk_token": null
|
| 207 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-623/trainer_state.json
ADDED
|
@@ -0,0 +1,726 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9998996890360117,
|
| 5 |
+
"eval_steps": 10000000,
|
| 6 |
+
"global_step": 623,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0016049754238138228,
|
| 13 |
+
"grad_norm": 7.15625,
|
| 14 |
+
"learning_rate": 2.6737967914438503e-07,
|
| 15 |
+
"logps_logqs/chosen": 0.057332463562488556,
|
| 16 |
+
"logqs/chosen": -0.752794086933136,
|
| 17 |
+
"logqs/rejected": -0.7512239813804626,
|
| 18 |
+
"logqs_logps/rejected": 0.1236220970749855,
|
| 19 |
+
"loss": 0.4632,
|
| 20 |
+
"step": 1
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.016049754238138227,
|
| 24 |
+
"grad_norm": 5.71875,
|
| 25 |
+
"learning_rate": 2.6737967914438504e-06,
|
| 26 |
+
"logps_logqs/chosen": 0.047926291823387146,
|
| 27 |
+
"logqs/chosen": -0.7439535856246948,
|
| 28 |
+
"logqs/rejected": -0.7035253643989563,
|
| 29 |
+
"logqs_logps/rejected": 0.13991409540176392,
|
| 30 |
+
"loss": 0.4557,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.032099508476276455,
|
| 35 |
+
"grad_norm": 1.9296875,
|
| 36 |
+
"learning_rate": 5.347593582887701e-06,
|
| 37 |
+
"logps_logqs/chosen": 0.03161158040165901,
|
| 38 |
+
"logqs/chosen": -0.6941366195678711,
|
| 39 |
+
"logqs/rejected": -0.6452130079269409,
|
| 40 |
+
"logqs_logps/rejected": 0.1449585109949112,
|
| 41 |
+
"loss": 0.4015,
|
| 42 |
+
"step": 20
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.048149262714414685,
|
| 46 |
+
"grad_norm": 1.8515625,
|
| 47 |
+
"learning_rate": 8.02139037433155e-06,
|
| 48 |
+
"logps_logqs/chosen": 0.03457744047045708,
|
| 49 |
+
"logqs/chosen": -0.6867337226867676,
|
| 50 |
+
"logqs/rejected": -0.6923023462295532,
|
| 51 |
+
"logqs_logps/rejected": 0.12560859322547913,
|
| 52 |
+
"loss": 0.3565,
|
| 53 |
+
"step": 30
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.06419901695255291,
|
| 57 |
+
"grad_norm": 1.234375,
|
| 58 |
+
"learning_rate": 1.0695187165775402e-05,
|
| 59 |
+
"logps_logqs/chosen": 0.05277745798230171,
|
| 60 |
+
"logqs/chosen": -0.7405093908309937,
|
| 61 |
+
"logqs/rejected": -0.7894801497459412,
|
| 62 |
+
"logqs_logps/rejected": 0.0936320573091507,
|
| 63 |
+
"loss": 0.3228,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.08024877119069114,
|
| 68 |
+
"grad_norm": 1.0,
|
| 69 |
+
"learning_rate": 1.3368983957219252e-05,
|
| 70 |
+
"logps_logqs/chosen": 0.04658779874444008,
|
| 71 |
+
"logqs/chosen": -0.7268352508544922,
|
| 72 |
+
"logqs/rejected": -0.7926934361457825,
|
| 73 |
+
"logqs_logps/rejected": 0.09889288991689682,
|
| 74 |
+
"loss": 0.3027,
|
| 75 |
+
"step": 50
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.09629852542882937,
|
| 79 |
+
"grad_norm": 0.92578125,
|
| 80 |
+
"learning_rate": 1.60427807486631e-05,
|
| 81 |
+
"logps_logqs/chosen": 0.04193533584475517,
|
| 82 |
+
"logqs/chosen": -0.7151986956596375,
|
| 83 |
+
"logqs/rejected": -0.7687807083129883,
|
| 84 |
+
"logqs_logps/rejected": 0.09696364402770996,
|
| 85 |
+
"loss": 0.2856,
|
| 86 |
+
"step": 60
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1123482796669676,
|
| 90 |
+
"grad_norm": 1.0625,
|
| 91 |
+
"learning_rate": 1.8716577540106954e-05,
|
| 92 |
+
"logps_logqs/chosen": 0.04730736091732979,
|
| 93 |
+
"logqs/chosen": -0.7385339736938477,
|
| 94 |
+
"logqs/rejected": -0.8247607946395874,
|
| 95 |
+
"logqs_logps/rejected": 0.08417234569787979,
|
| 96 |
+
"loss": 0.285,
|
| 97 |
+
"step": 70
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.12839803390510582,
|
| 101 |
+
"grad_norm": 0.890625,
|
| 102 |
+
"learning_rate": 2.1390374331550803e-05,
|
| 103 |
+
"logps_logqs/chosen": 0.047996845096349716,
|
| 104 |
+
"logqs/chosen": -0.7431933879852295,
|
| 105 |
+
"logqs/rejected": -0.8130408525466919,
|
| 106 |
+
"logqs_logps/rejected": 0.08468352258205414,
|
| 107 |
+
"loss": 0.2797,
|
| 108 |
+
"step": 80
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.14444778814324405,
|
| 112 |
+
"grad_norm": 0.79296875,
|
| 113 |
+
"learning_rate": 2.4064171122994652e-05,
|
| 114 |
+
"logps_logqs/chosen": 0.04765843600034714,
|
| 115 |
+
"logqs/chosen": -0.7173298001289368,
|
| 116 |
+
"logqs/rejected": -0.7709535360336304,
|
| 117 |
+
"logqs_logps/rejected": 0.08819471299648285,
|
| 118 |
+
"loss": 0.2705,
|
| 119 |
+
"step": 90
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.16049754238138228,
|
| 123 |
+
"grad_norm": 0.82421875,
|
| 124 |
+
"learning_rate": 2.6737967914438505e-05,
|
| 125 |
+
"logps_logqs/chosen": 0.040474437177181244,
|
| 126 |
+
"logqs/chosen": -0.710905909538269,
|
| 127 |
+
"logqs/rejected": -0.7959136962890625,
|
| 128 |
+
"logqs_logps/rejected": 0.09612198173999786,
|
| 129 |
+
"loss": 0.2683,
|
| 130 |
+
"step": 100
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.1765472966195205,
|
| 134 |
+
"grad_norm": 0.7109375,
|
| 135 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 136 |
+
"logps_logqs/chosen": 0.020066609606146812,
|
| 137 |
+
"logqs/chosen": -0.6621483564376831,
|
| 138 |
+
"logqs/rejected": -0.7314690351486206,
|
| 139 |
+
"logqs_logps/rejected": 0.11639375984668732,
|
| 140 |
+
"loss": 0.2682,
|
| 141 |
+
"step": 110
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.19259705085765874,
|
| 145 |
+
"grad_norm": 0.9921875,
|
| 146 |
+
"learning_rate": 3.20855614973262e-05,
|
| 147 |
+
"logps_logqs/chosen": 0.017146889120340347,
|
| 148 |
+
"logqs/chosen": -0.6599355340003967,
|
| 149 |
+
"logqs/rejected": -0.7124370336532593,
|
| 150 |
+
"logqs_logps/rejected": 0.12015531212091446,
|
| 151 |
+
"loss": 0.268,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.20864680509579697,
|
| 156 |
+
"grad_norm": 1.78125,
|
| 157 |
+
"learning_rate": 3.4759358288770055e-05,
|
| 158 |
+
"logps_logqs/chosen": 0.0369986966252327,
|
| 159 |
+
"logqs/chosen": -0.7184829115867615,
|
| 160 |
+
"logqs/rejected": -0.8139392137527466,
|
| 161 |
+
"logqs_logps/rejected": 0.09705157577991486,
|
| 162 |
+
"loss": 0.2611,
|
| 163 |
+
"step": 130
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2246965593339352,
|
| 167 |
+
"grad_norm": 0.87890625,
|
| 168 |
+
"learning_rate": 3.743315508021391e-05,
|
| 169 |
+
"logps_logqs/chosen": 0.07482504099607468,
|
| 170 |
+
"logqs/chosen": -0.7948209643363953,
|
| 171 |
+
"logqs/rejected": -0.9149974584579468,
|
| 172 |
+
"logqs_logps/rejected": 0.04995354264974594,
|
| 173 |
+
"loss": 0.2442,
|
| 174 |
+
"step": 140
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.24074631357207343,
|
| 178 |
+
"grad_norm": 0.796875,
|
| 179 |
+
"learning_rate": 4.0106951871657754e-05,
|
| 180 |
+
"logps_logqs/chosen": 0.08310296386480331,
|
| 181 |
+
"logqs/chosen": -0.812109649181366,
|
| 182 |
+
"logqs/rejected": -0.9168623089790344,
|
| 183 |
+
"logqs_logps/rejected": 0.041201137006282806,
|
| 184 |
+
"loss": 0.2403,
|
| 185 |
+
"step": 150
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.25679606781021164,
|
| 189 |
+
"grad_norm": 0.70703125,
|
| 190 |
+
"learning_rate": 4.2780748663101606e-05,
|
| 191 |
+
"logps_logqs/chosen": 0.08498374372720718,
|
| 192 |
+
"logqs/chosen": -0.8101779818534851,
|
| 193 |
+
"logqs/rejected": -0.9302076101303101,
|
| 194 |
+
"logqs_logps/rejected": 0.03671664744615555,
|
| 195 |
+
"loss": 0.2336,
|
| 196 |
+
"step": 160
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.2728458220483499,
|
| 200 |
+
"grad_norm": 0.85546875,
|
| 201 |
+
"learning_rate": 4.545454545454546e-05,
|
| 202 |
+
"logps_logqs/chosen": 0.08320538699626923,
|
| 203 |
+
"logqs/chosen": -0.8246332406997681,
|
| 204 |
+
"logqs/rejected": -0.9219290018081665,
|
| 205 |
+
"logqs_logps/rejected": 0.03701116889715195,
|
| 206 |
+
"loss": 0.2386,
|
| 207 |
+
"step": 170
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.2888955762864881,
|
| 211 |
+
"grad_norm": 0.77734375,
|
| 212 |
+
"learning_rate": 4.8128342245989304e-05,
|
| 213 |
+
"logps_logqs/chosen": 0.08179891854524612,
|
| 214 |
+
"logqs/chosen": -0.787845253944397,
|
| 215 |
+
"logqs/rejected": -0.9030144810676575,
|
| 216 |
+
"logqs_logps/rejected": 0.03777966648340225,
|
| 217 |
+
"loss": 0.2315,
|
| 218 |
+
"step": 180
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.30494533052462636,
|
| 222 |
+
"grad_norm": 0.7734375,
|
| 223 |
+
"learning_rate": 4.9999607536612036e-05,
|
| 224 |
+
"logps_logqs/chosen": 0.08169040083885193,
|
| 225 |
+
"logqs/chosen": -0.8232030868530273,
|
| 226 |
+
"logqs/rejected": -0.9288152456283569,
|
| 227 |
+
"logqs_logps/rejected": 0.039177440106868744,
|
| 228 |
+
"loss": 0.2389,
|
| 229 |
+
"step": 190
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.32099508476276456,
|
| 233 |
+
"grad_norm": 0.72265625,
|
| 234 |
+
"learning_rate": 4.9992630752499945e-05,
|
| 235 |
+
"logps_logqs/chosen": 0.08127471804618835,
|
| 236 |
+
"logqs/chosen": -0.8097829818725586,
|
| 237 |
+
"logqs/rejected": -0.9110867381095886,
|
| 238 |
+
"logqs_logps/rejected": 0.03667169809341431,
|
| 239 |
+
"loss": 0.2279,
|
| 240 |
+
"step": 200
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3370448390009028,
|
| 244 |
+
"grad_norm": 0.71875,
|
| 245 |
+
"learning_rate": 4.997693536122969e-05,
|
| 246 |
+
"logps_logqs/chosen": 0.0788678377866745,
|
| 247 |
+
"logqs/chosen": -0.7994121313095093,
|
| 248 |
+
"logqs/rejected": -0.9303783178329468,
|
| 249 |
+
"logqs_logps/rejected": 0.037957318127155304,
|
| 250 |
+
"loss": 0.2291,
|
| 251 |
+
"step": 210
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.353094593239041,
|
| 255 |
+
"grad_norm": 0.7109375,
|
| 256 |
+
"learning_rate": 4.995252683809324e-05,
|
| 257 |
+
"logps_logqs/chosen": 0.08248866349458694,
|
| 258 |
+
"logqs/chosen": -0.8128089904785156,
|
| 259 |
+
"logqs/rejected": -0.9362344741821289,
|
| 260 |
+
"logqs_logps/rejected": 0.037949927151203156,
|
| 261 |
+
"loss": 0.229,
|
| 262 |
+
"step": 220
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3691443474771793,
|
| 266 |
+
"grad_norm": 0.63671875,
|
| 267 |
+
"learning_rate": 4.9919413697933496e-05,
|
| 268 |
+
"logps_logqs/chosen": 0.0817771628499031,
|
| 269 |
+
"logqs/chosen": -0.7939696311950684,
|
| 270 |
+
"logqs/rejected": -0.9058554768562317,
|
| 271 |
+
"logqs_logps/rejected": 0.04094362258911133,
|
| 272 |
+
"loss": 0.2314,
|
| 273 |
+
"step": 230
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.3851941017153175,
|
| 277 |
+
"grad_norm": 0.71875,
|
| 278 |
+
"learning_rate": 4.987760749217389e-05,
|
| 279 |
+
"logps_logqs/chosen": 0.076979860663414,
|
| 280 |
+
"logqs/chosen": -0.7874332070350647,
|
| 281 |
+
"logqs/rejected": -0.9303415417671204,
|
| 282 |
+
"logqs_logps/rejected": 0.04022833704948425,
|
| 283 |
+
"loss": 0.2244,
|
| 284 |
+
"step": 240
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.4012438559534557,
|
| 288 |
+
"grad_norm": 0.83984375,
|
| 289 |
+
"learning_rate": 4.982712280478875e-05,
|
| 290 |
+
"logps_logqs/chosen": 0.07487889379262924,
|
| 291 |
+
"logqs/chosen": -0.7985113859176636,
|
| 292 |
+
"logqs/rejected": -0.9595896005630493,
|
| 293 |
+
"logqs_logps/rejected": 0.04083302244544029,
|
| 294 |
+
"loss": 0.2277,
|
| 295 |
+
"step": 250
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.41729361019159394,
|
| 299 |
+
"grad_norm": 0.6953125,
|
| 300 |
+
"learning_rate": 4.976797724721567e-05,
|
| 301 |
+
"logps_logqs/chosen": 0.07195514440536499,
|
| 302 |
+
"logqs/chosen": -0.787926971912384,
|
| 303 |
+
"logqs/rejected": -0.897616982460022,
|
| 304 |
+
"logqs_logps/rejected": 0.0396430678665638,
|
| 305 |
+
"loss": 0.2163,
|
| 306 |
+
"step": 260
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.43334336442973215,
|
| 310 |
+
"grad_norm": 0.7734375,
|
| 311 |
+
"learning_rate": 4.9700191452211806e-05,
|
| 312 |
+
"logps_logqs/chosen": 0.07640247046947479,
|
| 313 |
+
"logqs/chosen": -0.7960728406906128,
|
| 314 |
+
"logqs/rejected": -0.928481936454773,
|
| 315 |
+
"logqs_logps/rejected": 0.03928074985742569,
|
| 316 |
+
"loss": 0.2226,
|
| 317 |
+
"step": 270
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.4493931186678704,
|
| 321 |
+
"grad_norm": 0.72265625,
|
| 322 |
+
"learning_rate": 4.9623789066656276e-05,
|
| 323 |
+
"logps_logqs/chosen": 0.07622957229614258,
|
| 324 |
+
"logqs/chosen": -0.8077980279922485,
|
| 325 |
+
"logqs/rejected": -0.9034830927848816,
|
| 326 |
+
"logqs_logps/rejected": 0.046734608709812164,
|
| 327 |
+
"loss": 0.2256,
|
| 328 |
+
"step": 280
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.4654428729060086,
|
| 332 |
+
"grad_norm": 0.78515625,
|
| 333 |
+
"learning_rate": 4.953879674330093e-05,
|
| 334 |
+
"logps_logqs/chosen": 0.07244043052196503,
|
| 335 |
+
"logqs/chosen": -0.7900176644325256,
|
| 336 |
+
"logqs/rejected": -0.891532301902771,
|
| 337 |
+
"logqs_logps/rejected": 0.04387632757425308,
|
| 338 |
+
"loss": 0.217,
|
| 339 |
+
"step": 290
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.48149262714414687,
|
| 343 |
+
"grad_norm": 0.6796875,
|
| 344 |
+
"learning_rate": 4.944524413147263e-05,
|
| 345 |
+
"logps_logqs/chosen": 0.07289232313632965,
|
| 346 |
+
"logqs/chosen": -0.7856311798095703,
|
| 347 |
+
"logqs/rejected": -0.9099383354187012,
|
| 348 |
+
"logqs_logps/rejected": 0.03974480181932449,
|
| 349 |
+
"loss": 0.2173,
|
| 350 |
+
"step": 300
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.49754238138228507,
|
| 354 |
+
"grad_norm": 0.66796875,
|
| 355 |
+
"learning_rate": 4.934316386673022e-05,
|
| 356 |
+
"logps_logqs/chosen": 0.07063151150941849,
|
| 357 |
+
"logqs/chosen": -0.7815223932266235,
|
| 358 |
+
"logqs/rejected": -0.9079385995864868,
|
| 359 |
+
"logqs_logps/rejected": 0.041664548218250275,
|
| 360 |
+
"loss": 0.2126,
|
| 361 |
+
"step": 310
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.5135921356204233,
|
| 365 |
+
"grad_norm": 0.61328125,
|
| 366 |
+
"learning_rate": 4.923259155947964e-05,
|
| 367 |
+
"logps_logqs/chosen": 0.071230448782444,
|
| 368 |
+
"logqs/chosen": -0.8096257448196411,
|
| 369 |
+
"logqs/rejected": -0.9752206802368164,
|
| 370 |
+
"logqs_logps/rejected": 0.03939716890454292,
|
| 371 |
+
"loss": 0.2194,
|
| 372 |
+
"step": 320
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.5296418898585615,
|
| 376 |
+
"grad_norm": 0.6640625,
|
| 377 |
+
"learning_rate": 4.911356578255139e-05,
|
| 378 |
+
"logps_logqs/chosen": 0.07356056571006775,
|
| 379 |
+
"logqs/chosen": -0.7827819585800171,
|
| 380 |
+
"logqs/rejected": -0.9489104151725769,
|
| 381 |
+
"logqs_logps/rejected": 0.040141116827726364,
|
| 382 |
+
"loss": 0.2145,
|
| 383 |
+
"step": 330
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.5456916440966998,
|
| 387 |
+
"grad_norm": 0.7109375,
|
| 388 |
+
"learning_rate": 4.898612805774447e-05,
|
| 389 |
+
"logps_logqs/chosen": 0.07881536334753036,
|
| 390 |
+
"logqs/chosen": -0.8095147013664246,
|
| 391 |
+
"logqs/rejected": -0.9313008189201355,
|
| 392 |
+
"logqs_logps/rejected": 0.03702981770038605,
|
| 393 |
+
"loss": 0.2165,
|
| 394 |
+
"step": 340
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.561741398334838,
|
| 398 |
+
"grad_norm": 0.64453125,
|
| 399 |
+
"learning_rate": 4.885032284134165e-05,
|
| 400 |
+
"logps_logqs/chosen": 0.07632436603307724,
|
| 401 |
+
"logqs/chosen": -0.7979615926742554,
|
| 402 |
+
"logqs/rejected": -0.9368084073066711,
|
| 403 |
+
"logqs_logps/rejected": 0.03654230386018753,
|
| 404 |
+
"loss": 0.2113,
|
| 405 |
+
"step": 350
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.5777911525729762,
|
| 409 |
+
"grad_norm": 0.62109375,
|
| 410 |
+
"learning_rate": 4.8706197508600984e-05,
|
| 411 |
+
"logps_logqs/chosen": 0.07344283908605576,
|
| 412 |
+
"logqs/chosen": -0.8101444244384766,
|
| 413 |
+
"logqs/rejected": -0.9295485615730286,
|
| 414 |
+
"logqs_logps/rejected": 0.03901313990354538,
|
| 415 |
+
"loss": 0.2151,
|
| 416 |
+
"step": 360
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5938409068111145,
|
| 420 |
+
"grad_norm": 0.62890625,
|
| 421 |
+
"learning_rate": 4.855380233722915e-05,
|
| 422 |
+
"logps_logqs/chosen": 0.07421533018350601,
|
| 423 |
+
"logqs/chosen": -0.7826108336448669,
|
| 424 |
+
"logqs/rejected": -0.9229670763015747,
|
| 425 |
+
"logqs_logps/rejected": 0.038314513862133026,
|
| 426 |
+
"loss": 0.2129,
|
| 427 |
+
"step": 370
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.6098906610492527,
|
| 431 |
+
"grad_norm": 0.62890625,
|
| 432 |
+
"learning_rate": 4.839319048984217e-05,
|
| 433 |
+
"logps_logqs/chosen": 0.07273373752832413,
|
| 434 |
+
"logqs/chosen": -0.7798442840576172,
|
| 435 |
+
"logqs/rejected": -0.9371780157089233,
|
| 436 |
+
"logqs_logps/rejected": 0.03713950887322426,
|
| 437 |
+
"loss": 0.2067,
|
| 438 |
+
"step": 380
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.6259404152873909,
|
| 442 |
+
"grad_norm": 0.62890625,
|
| 443 |
+
"learning_rate": 4.822441799541979e-05,
|
| 444 |
+
"logps_logqs/chosen": 0.07329441606998444,
|
| 445 |
+
"logqs/chosen": -0.8102380037307739,
|
| 446 |
+
"logqs/rejected": -0.95549476146698,
|
| 447 |
+
"logqs_logps/rejected": 0.036079905927181244,
|
| 448 |
+
"loss": 0.2158,
|
| 449 |
+
"step": 390
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.6419901695255291,
|
| 453 |
+
"grad_norm": 0.66796875,
|
| 454 |
+
"learning_rate": 4.8047543729759936e-05,
|
| 455 |
+
"logps_logqs/chosen": 0.07404083013534546,
|
| 456 |
+
"logqs/chosen": -0.7847949266433716,
|
| 457 |
+
"logqs/rejected": -0.9250528216362,
|
| 458 |
+
"logqs_logps/rejected": 0.036556728184223175,
|
| 459 |
+
"loss": 0.2105,
|
| 460 |
+
"step": 400
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.6580399237636674,
|
| 464 |
+
"grad_norm": 0.69140625,
|
| 465 |
+
"learning_rate": 4.786262939494007e-05,
|
| 466 |
+
"logps_logqs/chosen": 0.07539906352758408,
|
| 467 |
+
"logqs/chosen": -0.8072575330734253,
|
| 468 |
+
"logqs/rejected": -0.9661371111869812,
|
| 469 |
+
"logqs_logps/rejected": 0.03662776201963425,
|
| 470 |
+
"loss": 0.2154,
|
| 471 |
+
"step": 410
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6740896780018056,
|
| 475 |
+
"grad_norm": 0.66015625,
|
| 476 |
+
"learning_rate": 4.766973949779261e-05,
|
| 477 |
+
"logps_logqs/chosen": 0.0744672566652298,
|
| 478 |
+
"logqs/chosen": -0.787712574005127,
|
| 479 |
+
"logqs/rejected": -0.9350829124450684,
|
| 480 |
+
"logqs_logps/rejected": 0.0353250689804554,
|
| 481 |
+
"loss": 0.2074,
|
| 482 |
+
"step": 420
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.6901394322399438,
|
| 486 |
+
"grad_norm": 0.62890625,
|
| 487 |
+
"learning_rate": 4.746894132740186e-05,
|
| 488 |
+
"logps_logqs/chosen": 0.07364196330308914,
|
| 489 |
+
"logqs/chosen": -0.7813644409179688,
|
| 490 |
+
"logqs/rejected": -0.9208289980888367,
|
| 491 |
+
"logqs_logps/rejected": 0.0356430858373642,
|
| 492 |
+
"loss": 0.2053,
|
| 493 |
+
"step": 430
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.706189186478082,
|
| 497 |
+
"grad_norm": 0.56640625,
|
| 498 |
+
"learning_rate": 4.726030493163044e-05,
|
| 499 |
+
"logps_logqs/chosen": 0.07587061077356339,
|
| 500 |
+
"logqs/chosen": -0.7853146195411682,
|
| 501 |
+
"logqs/rejected": -0.8944910168647766,
|
| 502 |
+
"logqs_logps/rejected": 0.03531279042363167,
|
| 503 |
+
"loss": 0.2078,
|
| 504 |
+
"step": 440
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.7222389407162203,
|
| 508 |
+
"grad_norm": 0.68359375,
|
| 509 |
+
"learning_rate": 4.7043903092683314e-05,
|
| 510 |
+
"logps_logqs/chosen": 0.07814273238182068,
|
| 511 |
+
"logqs/chosen": -0.8133522272109985,
|
| 512 |
+
"logqs/rejected": -0.9508693814277649,
|
| 513 |
+
"logqs_logps/rejected": 0.030778918415308,
|
| 514 |
+
"loss": 0.2104,
|
| 515 |
+
"step": 450
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.7382886949543586,
|
| 519 |
+
"grad_norm": 0.72265625,
|
| 520 |
+
"learning_rate": 4.6819811301717885e-05,
|
| 521 |
+
"logps_logqs/chosen": 0.07763786613941193,
|
| 522 |
+
"logqs/chosen": -0.8061367869377136,
|
| 523 |
+
"logqs/rejected": -0.9445575475692749,
|
| 524 |
+
"logqs_logps/rejected": 0.035373255610466,
|
| 525 |
+
"loss": 0.2084,
|
| 526 |
+
"step": 460
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.7543384491924967,
|
| 530 |
+
"grad_norm": 0.59765625,
|
| 531 |
+
"learning_rate": 4.6588107732509134e-05,
|
| 532 |
+
"logps_logqs/chosen": 0.07325359433889389,
|
| 533 |
+
"logqs/chosen": -0.7970572710037231,
|
| 534 |
+
"logqs/rejected": -0.9243482351303101,
|
| 535 |
+
"logqs_logps/rejected": 0.03834759443998337,
|
| 536 |
+
"loss": 0.2088,
|
| 537 |
+
"step": 470
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.770388203430635,
|
| 541 |
+
"grad_norm": 0.5625,
|
| 542 |
+
"learning_rate": 4.634887321417895e-05,
|
| 543 |
+
"logps_logqs/chosen": 0.0732722282409668,
|
| 544 |
+
"logqs/chosen": -0.7719672918319702,
|
| 545 |
+
"logqs/rejected": -0.9193938970565796,
|
| 546 |
+
"logqs_logps/rejected": 0.03625096380710602,
|
| 547 |
+
"loss": 0.2075,
|
| 548 |
+
"step": 480
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7864379576687732,
|
| 552 |
+
"grad_norm": 0.90234375,
|
| 553 |
+
"learning_rate": 4.6102191202999065e-05,
|
| 554 |
+
"logps_logqs/chosen": 0.07290570437908173,
|
| 555 |
+
"logqs/chosen": -0.8092619180679321,
|
| 556 |
+
"logqs/rejected": -0.9642523527145386,
|
| 557 |
+
"logqs_logps/rejected": 0.03311945125460625,
|
| 558 |
+
"loss": 0.2045,
|
| 559 |
+
"step": 490
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 0.8024877119069114,
|
| 563 |
+
"grad_norm": 0.6015625,
|
| 564 |
+
"learning_rate": 4.5848147753277656e-05,
|
| 565 |
+
"logps_logqs/chosen": 0.07134760171175003,
|
| 566 |
+
"logqs/chosen": -0.7773372530937195,
|
| 567 |
+
"logqs/rejected": -0.9478441476821899,
|
| 568 |
+
"logqs_logps/rejected": 0.03295496851205826,
|
| 569 |
+
"loss": 0.2049,
|
| 570 |
+
"step": 500
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.8185374661450496,
|
| 574 |
+
"grad_norm": 0.5859375,
|
| 575 |
+
"learning_rate": 4.5586831487339485e-05,
|
| 576 |
+
"logps_logqs/chosen": 0.07219503819942474,
|
| 577 |
+
"logqs/chosen": -0.7950411438941956,
|
| 578 |
+
"logqs/rejected": -0.9455002546310425,
|
| 579 |
+
"logqs_logps/rejected": 0.03765324503183365,
|
| 580 |
+
"loss": 0.208,
|
| 581 |
+
"step": 510
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.8345872203831879,
|
| 585 |
+
"grad_norm": 0.62109375,
|
| 586 |
+
"learning_rate": 4.531833356461027e-05,
|
| 587 |
+
"logps_logqs/chosen": 0.06849464029073715,
|
| 588 |
+
"logqs/chosen": -0.7819440364837646,
|
| 589 |
+
"logqs/rejected": -0.9597524404525757,
|
| 590 |
+
"logqs_logps/rejected": 0.04056422412395477,
|
| 591 |
+
"loss": 0.2084,
|
| 592 |
+
"step": 520
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.8506369746213261,
|
| 596 |
+
"grad_norm": 0.6171875,
|
| 597 |
+
"learning_rate": 4.5042747649816006e-05,
|
| 598 |
+
"logps_logqs/chosen": 0.07160626351833344,
|
| 599 |
+
"logqs/chosen": -0.7721427083015442,
|
| 600 |
+
"logqs/rejected": -0.94866544008255,
|
| 601 |
+
"logqs_logps/rejected": 0.03749927878379822,
|
| 602 |
+
"loss": 0.2068,
|
| 603 |
+
"step": 530
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.8666867288594643,
|
| 607 |
+
"grad_norm": 0.5546875,
|
| 608 |
+
"learning_rate": 4.476016988030826e-05,
|
| 609 |
+
"logps_logqs/chosen": 0.07214485853910446,
|
| 610 |
+
"logqs/chosen": -0.7874671816825867,
|
| 611 |
+
"logqs/rejected": -0.9504098892211914,
|
| 612 |
+
"logqs_logps/rejected": 0.03507527709007263,
|
| 613 |
+
"loss": 0.2045,
|
| 614 |
+
"step": 540
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.8827364830976026,
|
| 618 |
+
"grad_norm": 0.54296875,
|
| 619 |
+
"learning_rate": 4.447069883252696e-05,
|
| 620 |
+
"logps_logqs/chosen": 0.07395409047603607,
|
| 621 |
+
"logqs/chosen": -0.771978497505188,
|
| 622 |
+
"logqs/rejected": -0.9129235148429871,
|
| 623 |
+
"logqs_logps/rejected": 0.039593033492565155,
|
| 624 |
+
"loss": 0.205,
|
| 625 |
+
"step": 550
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.8987862373357408,
|
| 629 |
+
"grad_norm": 0.61328125,
|
| 630 |
+
"learning_rate": 4.417443548761227e-05,
|
| 631 |
+
"logps_logqs/chosen": 0.07299650460481644,
|
| 632 |
+
"logqs/chosen": -0.7979342341423035,
|
| 633 |
+
"logqs/rejected": -0.9357426762580872,
|
| 634 |
+
"logqs_logps/rejected": 0.03937726467847824,
|
| 635 |
+
"loss": 0.2083,
|
| 636 |
+
"step": 560
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.9148359915738791,
|
| 640 |
+
"grad_norm": 0.57421875,
|
| 641 |
+
"learning_rate": 4.387148319617763e-05,
|
| 642 |
+
"logps_logqs/chosen": 0.06836996972560883,
|
| 643 |
+
"logqs/chosen": -0.7659951448440552,
|
| 644 |
+
"logqs/rejected": -0.8732549548149109,
|
| 645 |
+
"logqs_logps/rejected": 0.039081670343875885,
|
| 646 |
+
"loss": 0.203,
|
| 647 |
+
"step": 570
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.9308857458120172,
|
| 651 |
+
"grad_norm": 0.57421875,
|
| 652 |
+
"learning_rate": 4.356194764225618e-05,
|
| 653 |
+
"logps_logqs/chosen": 0.07397963851690292,
|
| 654 |
+
"logqs/chosen": -0.7891489863395691,
|
| 655 |
+
"logqs/rejected": -0.9294188618659973,
|
| 656 |
+
"logqs_logps/rejected": 0.03416060656309128,
|
| 657 |
+
"loss": 0.2024,
|
| 658 |
+
"step": 580
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.9469355000501555,
|
| 662 |
+
"grad_norm": 0.703125,
|
| 663 |
+
"learning_rate": 4.3245936806433205e-05,
|
| 664 |
+
"logps_logqs/chosen": 0.07466734945774078,
|
| 665 |
+
"logqs/chosen": -0.7958794832229614,
|
| 666 |
+
"logqs/rejected": -0.953458309173584,
|
| 667 |
+
"logqs_logps/rejected": 0.035235695540905,
|
| 668 |
+
"loss": 0.206,
|
| 669 |
+
"step": 590
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.9629852542882937,
|
| 673 |
+
"grad_norm": 0.5859375,
|
| 674 |
+
"learning_rate": 4.292356092817739e-05,
|
| 675 |
+
"logps_logqs/chosen": 0.07136549055576324,
|
| 676 |
+
"logqs/chosen": -0.7849777936935425,
|
| 677 |
+
"logqs/rejected": -0.9327009320259094,
|
| 678 |
+
"logqs_logps/rejected": 0.038332488387823105,
|
| 679 |
+
"loss": 0.2041,
|
| 680 |
+
"step": 600
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.979035008526432,
|
| 684 |
+
"grad_norm": 0.55859375,
|
| 685 |
+
"learning_rate": 4.259493246738409e-05,
|
| 686 |
+
"logps_logqs/chosen": 0.06976237148046494,
|
| 687 |
+
"logqs/chosen": -0.7658575773239136,
|
| 688 |
+
"logqs/rejected": -0.9027583003044128,
|
| 689 |
+
"logqs_logps/rejected": 0.03464614599943161,
|
| 690 |
+
"loss": 0.1993,
|
| 691 |
+
"step": 610
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.9950847627645701,
|
| 695 |
+
"grad_norm": 0.55078125,
|
| 696 |
+
"learning_rate": 4.226016606514411e-05,
|
| 697 |
+
"logps_logqs/chosen": 0.0706130638718605,
|
| 698 |
+
"logqs/chosen": -0.7847878932952881,
|
| 699 |
+
"logqs/rejected": -0.9509286880493164,
|
| 700 |
+
"logqs_logps/rejected": 0.03653167933225632,
|
| 701 |
+
"loss": 0.2035,
|
| 702 |
+
"step": 620
|
| 703 |
+
}
|
| 704 |
+
],
|
| 705 |
+
"logging_steps": 10,
|
| 706 |
+
"max_steps": 1869,
|
| 707 |
+
"num_input_tokens_seen": 0,
|
| 708 |
+
"num_train_epochs": 3,
|
| 709 |
+
"save_steps": 500,
|
| 710 |
+
"stateful_callbacks": {
|
| 711 |
+
"TrainerControl": {
|
| 712 |
+
"args": {
|
| 713 |
+
"should_epoch_stop": false,
|
| 714 |
+
"should_evaluate": false,
|
| 715 |
+
"should_log": false,
|
| 716 |
+
"should_save": true,
|
| 717 |
+
"should_training_stop": false
|
| 718 |
+
},
|
| 719 |
+
"attributes": {}
|
| 720 |
+
}
|
| 721 |
+
},
|
| 722 |
+
"total_flos": 0.0,
|
| 723 |
+
"train_batch_size": 2,
|
| 724 |
+
"trial_name": null,
|
| 725 |
+
"trial_params": null
|
| 726 |
+
}
|
qwen2.5-1.5B-it-distillm2/checkpoint-623/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b9fa66895bee5a36a6f750ef897f491a1113f2f79af77c75bdef8187ae10031
|
| 3 |
+
size 6840
|
qwen2.5-1.5B-it-distillm2/checkpoint-623/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qwen2.5-1.5B-it-distillm2/config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Qwen/Qwen2.5-1.5B-Instruct",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"Qwen2ForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": 151645,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 1536,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 8960,
|
| 13 |
+
"max_position_embeddings": 32768,
|
| 14 |
+
"max_window_layers": 21,
|
| 15 |
+
"model_type": "qwen2",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 28,
|
| 18 |
+
"num_key_value_heads": 2,
|
| 19 |
+
"rms_norm_eps": 1e-06,
|
| 20 |
+
"rope_scaling": null,
|
| 21 |
+
"rope_theta": 1000000.0,
|
| 22 |
+
"sliding_window": null,
|
| 23 |
+
"tie_word_embeddings": true,
|
| 24 |
+
"torch_dtype": "bfloat16",
|
| 25 |
+
"transformers_version": "4.45.2",
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"use_sliding_window": false,
|
| 28 |
+
"vocab_size": 152064
|
| 29 |
+
}
|
qwen2.5-1.5B-it-distillm2/eval_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.999699067108035,
|
| 3 |
+
"eval_logps_logqs/chosen": 0.07430665194988251,
|
| 4 |
+
"eval_logqs/chosen": -1.0447684526443481,
|
| 5 |
+
"eval_logqs/rejected": -1.0230671167373657,
|
| 6 |
+
"eval_logqs_logps/rejected": 0.034393880516290665,
|
| 7 |
+
"eval_loss": 0.17647793889045715,
|
| 8 |
+
"eval_runtime": 1.0546,
|
| 9 |
+
"eval_samples": 10,
|
| 10 |
+
"eval_samples_per_second": 9.482,
|
| 11 |
+
"eval_steps_per_second": 2.845
|
| 12 |
+
}
|
qwen2.5-1.5B-it-distillm2/generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.45.2"
|
| 14 |
+
}
|
qwen2.5-1.5B-it-distillm2/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qwen2.5-1.5B-it-distillm2/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa1f325bd96a7e02353ee13f20a8362121741466a9b362e7cfd122a67c5ae6e0
|
| 3 |
+
size 3087935634
|
qwen2.5-1.5B-it-distillm2/special_tokens_map.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>"
|
| 16 |
+
],
|
| 17 |
+
"bos_token": "<|endoftext|>",
|
| 18 |
+
"eos_token": {
|
| 19 |
+
"content": "<|im_end|>",
|
| 20 |
+
"lstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"rstrip": false,
|
| 23 |
+
"single_word": false
|
| 24 |
+
},
|
| 25 |
+
"pad_token": {
|
| 26 |
+
"content": "<|endoftext|>",
|
| 27 |
+
"lstrip": false,
|
| 28 |
+
"normalized": false,
|
| 29 |
+
"rstrip": false,
|
| 30 |
+
"single_word": false
|
| 31 |
+
}
|
| 32 |
+
}
|
qwen2.5-1.5B-it-distillm2/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
|
| 3 |
+
size 11421896
|
qwen2.5-1.5B-it-distillm2/tokenizer_config.json
ADDED
|
@@ -0,0 +1,207 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
}
|
| 181 |
+
},
|
| 182 |
+
"additional_special_tokens": [
|
| 183 |
+
"<|im_start|>",
|
| 184 |
+
"<|im_end|>",
|
| 185 |
+
"<|object_ref_start|>",
|
| 186 |
+
"<|object_ref_end|>",
|
| 187 |
+
"<|box_start|>",
|
| 188 |
+
"<|box_end|>",
|
| 189 |
+
"<|quad_start|>",
|
| 190 |
+
"<|quad_end|>",
|
| 191 |
+
"<|vision_start|>",
|
| 192 |
+
"<|vision_end|>",
|
| 193 |
+
"<|vision_pad|>",
|
| 194 |
+
"<|image_pad|>",
|
| 195 |
+
"<|video_pad|>"
|
| 196 |
+
],
|
| 197 |
+
"bos_token": "<|endoftext|>",
|
| 198 |
+
"chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
|
| 199 |
+
"clean_up_tokenization_spaces": false,
|
| 200 |
+
"eos_token": "<|im_end|>",
|
| 201 |
+
"errors": "replace",
|
| 202 |
+
"model_max_length": 2048,
|
| 203 |
+
"pad_token": "<|endoftext|>",
|
| 204 |
+
"split_special_tokens": false,
|
| 205 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 206 |
+
"unk_token": null
|
| 207 |
+
}
|
qwen2.5-1.5B-it-distillm2/train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.999699067108035,
|
| 3 |
+
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.194109176269121,
|
| 5 |
+
"train_runtime": 23250.2173,
|
| 6 |
+
"train_samples": 79751,
|
| 7 |
+
"train_samples_per_second": 10.29,
|
| 8 |
+
"train_steps_per_second": 0.08
|
| 9 |
+
}
|
qwen2.5-1.5B-it-distillm2/trainer_state.json
ADDED
|
@@ -0,0 +1,2099 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.999699067108035,
|
| 5 |
+
"eval_steps": 10000000,
|
| 6 |
+
"global_step": 1869,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0016049754238138228,
|
| 13 |
+
"grad_norm": 7.15625,
|
| 14 |
+
"learning_rate": 2.6737967914438503e-07,
|
| 15 |
+
"logps_logqs/chosen": 0.057332463562488556,
|
| 16 |
+
"logqs/chosen": -0.752794086933136,
|
| 17 |
+
"logqs/rejected": -0.7512239813804626,
|
| 18 |
+
"logqs_logps/rejected": 0.1236220970749855,
|
| 19 |
+
"loss": 0.4632,
|
| 20 |
+
"step": 1
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.016049754238138227,
|
| 24 |
+
"grad_norm": 5.71875,
|
| 25 |
+
"learning_rate": 2.6737967914438504e-06,
|
| 26 |
+
"logps_logqs/chosen": 0.047926291823387146,
|
| 27 |
+
"logqs/chosen": -0.7439535856246948,
|
| 28 |
+
"logqs/rejected": -0.7035253643989563,
|
| 29 |
+
"logqs_logps/rejected": 0.13991409540176392,
|
| 30 |
+
"loss": 0.4557,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.032099508476276455,
|
| 35 |
+
"grad_norm": 1.9296875,
|
| 36 |
+
"learning_rate": 5.347593582887701e-06,
|
| 37 |
+
"logps_logqs/chosen": 0.03161158040165901,
|
| 38 |
+
"logqs/chosen": -0.6941366195678711,
|
| 39 |
+
"logqs/rejected": -0.6452130079269409,
|
| 40 |
+
"logqs_logps/rejected": 0.1449585109949112,
|
| 41 |
+
"loss": 0.4015,
|
| 42 |
+
"step": 20
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.048149262714414685,
|
| 46 |
+
"grad_norm": 1.8515625,
|
| 47 |
+
"learning_rate": 8.02139037433155e-06,
|
| 48 |
+
"logps_logqs/chosen": 0.03457744047045708,
|
| 49 |
+
"logqs/chosen": -0.6867337226867676,
|
| 50 |
+
"logqs/rejected": -0.6923023462295532,
|
| 51 |
+
"logqs_logps/rejected": 0.12560859322547913,
|
| 52 |
+
"loss": 0.3565,
|
| 53 |
+
"step": 30
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.06419901695255291,
|
| 57 |
+
"grad_norm": 1.234375,
|
| 58 |
+
"learning_rate": 1.0695187165775402e-05,
|
| 59 |
+
"logps_logqs/chosen": 0.05277745798230171,
|
| 60 |
+
"logqs/chosen": -0.7405093908309937,
|
| 61 |
+
"logqs/rejected": -0.7894801497459412,
|
| 62 |
+
"logqs_logps/rejected": 0.0936320573091507,
|
| 63 |
+
"loss": 0.3228,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.08024877119069114,
|
| 68 |
+
"grad_norm": 1.0,
|
| 69 |
+
"learning_rate": 1.3368983957219252e-05,
|
| 70 |
+
"logps_logqs/chosen": 0.04658779874444008,
|
| 71 |
+
"logqs/chosen": -0.7268352508544922,
|
| 72 |
+
"logqs/rejected": -0.7926934361457825,
|
| 73 |
+
"logqs_logps/rejected": 0.09889288991689682,
|
| 74 |
+
"loss": 0.3027,
|
| 75 |
+
"step": 50
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.09629852542882937,
|
| 79 |
+
"grad_norm": 0.92578125,
|
| 80 |
+
"learning_rate": 1.60427807486631e-05,
|
| 81 |
+
"logps_logqs/chosen": 0.04193533584475517,
|
| 82 |
+
"logqs/chosen": -0.7151986956596375,
|
| 83 |
+
"logqs/rejected": -0.7687807083129883,
|
| 84 |
+
"logqs_logps/rejected": 0.09696364402770996,
|
| 85 |
+
"loss": 0.2856,
|
| 86 |
+
"step": 60
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1123482796669676,
|
| 90 |
+
"grad_norm": 1.0625,
|
| 91 |
+
"learning_rate": 1.8716577540106954e-05,
|
| 92 |
+
"logps_logqs/chosen": 0.04730736091732979,
|
| 93 |
+
"logqs/chosen": -0.7385339736938477,
|
| 94 |
+
"logqs/rejected": -0.8247607946395874,
|
| 95 |
+
"logqs_logps/rejected": 0.08417234569787979,
|
| 96 |
+
"loss": 0.285,
|
| 97 |
+
"step": 70
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.12839803390510582,
|
| 101 |
+
"grad_norm": 0.890625,
|
| 102 |
+
"learning_rate": 2.1390374331550803e-05,
|
| 103 |
+
"logps_logqs/chosen": 0.047996845096349716,
|
| 104 |
+
"logqs/chosen": -0.7431933879852295,
|
| 105 |
+
"logqs/rejected": -0.8130408525466919,
|
| 106 |
+
"logqs_logps/rejected": 0.08468352258205414,
|
| 107 |
+
"loss": 0.2797,
|
| 108 |
+
"step": 80
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.14444778814324405,
|
| 112 |
+
"grad_norm": 0.79296875,
|
| 113 |
+
"learning_rate": 2.4064171122994652e-05,
|
| 114 |
+
"logps_logqs/chosen": 0.04765843600034714,
|
| 115 |
+
"logqs/chosen": -0.7173298001289368,
|
| 116 |
+
"logqs/rejected": -0.7709535360336304,
|
| 117 |
+
"logqs_logps/rejected": 0.08819471299648285,
|
| 118 |
+
"loss": 0.2705,
|
| 119 |
+
"step": 90
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.16049754238138228,
|
| 123 |
+
"grad_norm": 0.82421875,
|
| 124 |
+
"learning_rate": 2.6737967914438505e-05,
|
| 125 |
+
"logps_logqs/chosen": 0.040474437177181244,
|
| 126 |
+
"logqs/chosen": -0.710905909538269,
|
| 127 |
+
"logqs/rejected": -0.7959136962890625,
|
| 128 |
+
"logqs_logps/rejected": 0.09612198173999786,
|
| 129 |
+
"loss": 0.2683,
|
| 130 |
+
"step": 100
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.1765472966195205,
|
| 134 |
+
"grad_norm": 0.7109375,
|
| 135 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 136 |
+
"logps_logqs/chosen": 0.020066609606146812,
|
| 137 |
+
"logqs/chosen": -0.6621483564376831,
|
| 138 |
+
"logqs/rejected": -0.7314690351486206,
|
| 139 |
+
"logqs_logps/rejected": 0.11639375984668732,
|
| 140 |
+
"loss": 0.2682,
|
| 141 |
+
"step": 110
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.19259705085765874,
|
| 145 |
+
"grad_norm": 0.9921875,
|
| 146 |
+
"learning_rate": 3.20855614973262e-05,
|
| 147 |
+
"logps_logqs/chosen": 0.017146889120340347,
|
| 148 |
+
"logqs/chosen": -0.6599355340003967,
|
| 149 |
+
"logqs/rejected": -0.7124370336532593,
|
| 150 |
+
"logqs_logps/rejected": 0.12015531212091446,
|
| 151 |
+
"loss": 0.268,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.20864680509579697,
|
| 156 |
+
"grad_norm": 1.78125,
|
| 157 |
+
"learning_rate": 3.4759358288770055e-05,
|
| 158 |
+
"logps_logqs/chosen": 0.0369986966252327,
|
| 159 |
+
"logqs/chosen": -0.7184829115867615,
|
| 160 |
+
"logqs/rejected": -0.8139392137527466,
|
| 161 |
+
"logqs_logps/rejected": 0.09705157577991486,
|
| 162 |
+
"loss": 0.2611,
|
| 163 |
+
"step": 130
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2246965593339352,
|
| 167 |
+
"grad_norm": 0.87890625,
|
| 168 |
+
"learning_rate": 3.743315508021391e-05,
|
| 169 |
+
"logps_logqs/chosen": 0.07482504099607468,
|
| 170 |
+
"logqs/chosen": -0.7948209643363953,
|
| 171 |
+
"logqs/rejected": -0.9149974584579468,
|
| 172 |
+
"logqs_logps/rejected": 0.04995354264974594,
|
| 173 |
+
"loss": 0.2442,
|
| 174 |
+
"step": 140
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.24074631357207343,
|
| 178 |
+
"grad_norm": 0.796875,
|
| 179 |
+
"learning_rate": 4.0106951871657754e-05,
|
| 180 |
+
"logps_logqs/chosen": 0.08310296386480331,
|
| 181 |
+
"logqs/chosen": -0.812109649181366,
|
| 182 |
+
"logqs/rejected": -0.9168623089790344,
|
| 183 |
+
"logqs_logps/rejected": 0.041201137006282806,
|
| 184 |
+
"loss": 0.2403,
|
| 185 |
+
"step": 150
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.25679606781021164,
|
| 189 |
+
"grad_norm": 0.70703125,
|
| 190 |
+
"learning_rate": 4.2780748663101606e-05,
|
| 191 |
+
"logps_logqs/chosen": 0.08498374372720718,
|
| 192 |
+
"logqs/chosen": -0.8101779818534851,
|
| 193 |
+
"logqs/rejected": -0.9302076101303101,
|
| 194 |
+
"logqs_logps/rejected": 0.03671664744615555,
|
| 195 |
+
"loss": 0.2336,
|
| 196 |
+
"step": 160
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.2728458220483499,
|
| 200 |
+
"grad_norm": 0.85546875,
|
| 201 |
+
"learning_rate": 4.545454545454546e-05,
|
| 202 |
+
"logps_logqs/chosen": 0.08320538699626923,
|
| 203 |
+
"logqs/chosen": -0.8246332406997681,
|
| 204 |
+
"logqs/rejected": -0.9219290018081665,
|
| 205 |
+
"logqs_logps/rejected": 0.03701116889715195,
|
| 206 |
+
"loss": 0.2386,
|
| 207 |
+
"step": 170
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.2888955762864881,
|
| 211 |
+
"grad_norm": 0.77734375,
|
| 212 |
+
"learning_rate": 4.8128342245989304e-05,
|
| 213 |
+
"logps_logqs/chosen": 0.08179891854524612,
|
| 214 |
+
"logqs/chosen": -0.787845253944397,
|
| 215 |
+
"logqs/rejected": -0.9030144810676575,
|
| 216 |
+
"logqs_logps/rejected": 0.03777966648340225,
|
| 217 |
+
"loss": 0.2315,
|
| 218 |
+
"step": 180
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.30494533052462636,
|
| 222 |
+
"grad_norm": 0.7734375,
|
| 223 |
+
"learning_rate": 4.9999607536612036e-05,
|
| 224 |
+
"logps_logqs/chosen": 0.08169040083885193,
|
| 225 |
+
"logqs/chosen": -0.8232030868530273,
|
| 226 |
+
"logqs/rejected": -0.9288152456283569,
|
| 227 |
+
"logqs_logps/rejected": 0.039177440106868744,
|
| 228 |
+
"loss": 0.2389,
|
| 229 |
+
"step": 190
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.32099508476276456,
|
| 233 |
+
"grad_norm": 0.72265625,
|
| 234 |
+
"learning_rate": 4.9992630752499945e-05,
|
| 235 |
+
"logps_logqs/chosen": 0.08127471804618835,
|
| 236 |
+
"logqs/chosen": -0.8097829818725586,
|
| 237 |
+
"logqs/rejected": -0.9110867381095886,
|
| 238 |
+
"logqs_logps/rejected": 0.03667169809341431,
|
| 239 |
+
"loss": 0.2279,
|
| 240 |
+
"step": 200
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3370448390009028,
|
| 244 |
+
"grad_norm": 0.71875,
|
| 245 |
+
"learning_rate": 4.997693536122969e-05,
|
| 246 |
+
"logps_logqs/chosen": 0.0788678377866745,
|
| 247 |
+
"logqs/chosen": -0.7994121313095093,
|
| 248 |
+
"logqs/rejected": -0.9303783178329468,
|
| 249 |
+
"logqs_logps/rejected": 0.037957318127155304,
|
| 250 |
+
"loss": 0.2291,
|
| 251 |
+
"step": 210
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.353094593239041,
|
| 255 |
+
"grad_norm": 0.7109375,
|
| 256 |
+
"learning_rate": 4.995252683809324e-05,
|
| 257 |
+
"logps_logqs/chosen": 0.08248866349458694,
|
| 258 |
+
"logqs/chosen": -0.8128089904785156,
|
| 259 |
+
"logqs/rejected": -0.9362344741821289,
|
| 260 |
+
"logqs_logps/rejected": 0.037949927151203156,
|
| 261 |
+
"loss": 0.229,
|
| 262 |
+
"step": 220
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3691443474771793,
|
| 266 |
+
"grad_norm": 0.63671875,
|
| 267 |
+
"learning_rate": 4.9919413697933496e-05,
|
| 268 |
+
"logps_logqs/chosen": 0.0817771628499031,
|
| 269 |
+
"logqs/chosen": -0.7939696311950684,
|
| 270 |
+
"logqs/rejected": -0.9058554768562317,
|
| 271 |
+
"logqs_logps/rejected": 0.04094362258911133,
|
| 272 |
+
"loss": 0.2314,
|
| 273 |
+
"step": 230
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.3851941017153175,
|
| 277 |
+
"grad_norm": 0.71875,
|
| 278 |
+
"learning_rate": 4.987760749217389e-05,
|
| 279 |
+
"logps_logqs/chosen": 0.076979860663414,
|
| 280 |
+
"logqs/chosen": -0.7874332070350647,
|
| 281 |
+
"logqs/rejected": -0.9303415417671204,
|
| 282 |
+
"logqs_logps/rejected": 0.04022833704948425,
|
| 283 |
+
"loss": 0.2244,
|
| 284 |
+
"step": 240
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.4012438559534557,
|
| 288 |
+
"grad_norm": 0.83984375,
|
| 289 |
+
"learning_rate": 4.982712280478875e-05,
|
| 290 |
+
"logps_logqs/chosen": 0.07487889379262924,
|
| 291 |
+
"logqs/chosen": -0.7985113859176636,
|
| 292 |
+
"logqs/rejected": -0.9595896005630493,
|
| 293 |
+
"logqs_logps/rejected": 0.04083302244544029,
|
| 294 |
+
"loss": 0.2277,
|
| 295 |
+
"step": 250
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.41729361019159394,
|
| 299 |
+
"grad_norm": 0.6953125,
|
| 300 |
+
"learning_rate": 4.976797724721567e-05,
|
| 301 |
+
"logps_logqs/chosen": 0.07195514440536499,
|
| 302 |
+
"logqs/chosen": -0.787926971912384,
|
| 303 |
+
"logqs/rejected": -0.897616982460022,
|
| 304 |
+
"logqs_logps/rejected": 0.0396430678665638,
|
| 305 |
+
"loss": 0.2163,
|
| 306 |
+
"step": 260
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.43334336442973215,
|
| 310 |
+
"grad_norm": 0.7734375,
|
| 311 |
+
"learning_rate": 4.9700191452211806e-05,
|
| 312 |
+
"logps_logqs/chosen": 0.07640247046947479,
|
| 313 |
+
"logqs/chosen": -0.7960728406906128,
|
| 314 |
+
"logqs/rejected": -0.928481936454773,
|
| 315 |
+
"logqs_logps/rejected": 0.03928074985742569,
|
| 316 |
+
"loss": 0.2226,
|
| 317 |
+
"step": 270
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.4493931186678704,
|
| 321 |
+
"grad_norm": 0.72265625,
|
| 322 |
+
"learning_rate": 4.9623789066656276e-05,
|
| 323 |
+
"logps_logqs/chosen": 0.07622957229614258,
|
| 324 |
+
"logqs/chosen": -0.8077980279922485,
|
| 325 |
+
"logqs/rejected": -0.9034830927848816,
|
| 326 |
+
"logqs_logps/rejected": 0.046734608709812164,
|
| 327 |
+
"loss": 0.2256,
|
| 328 |
+
"step": 280
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.4654428729060086,
|
| 332 |
+
"grad_norm": 0.78515625,
|
| 333 |
+
"learning_rate": 4.953879674330093e-05,
|
| 334 |
+
"logps_logqs/chosen": 0.07244043052196503,
|
| 335 |
+
"logqs/chosen": -0.7900176644325256,
|
| 336 |
+
"logqs/rejected": -0.891532301902771,
|
| 337 |
+
"logqs_logps/rejected": 0.04387632757425308,
|
| 338 |
+
"loss": 0.217,
|
| 339 |
+
"step": 290
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.48149262714414687,
|
| 343 |
+
"grad_norm": 0.6796875,
|
| 344 |
+
"learning_rate": 4.944524413147263e-05,
|
| 345 |
+
"logps_logqs/chosen": 0.07289232313632965,
|
| 346 |
+
"logqs/chosen": -0.7856311798095703,
|
| 347 |
+
"logqs/rejected": -0.9099383354187012,
|
| 348 |
+
"logqs_logps/rejected": 0.03974480181932449,
|
| 349 |
+
"loss": 0.2173,
|
| 350 |
+
"step": 300
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.49754238138228507,
|
| 354 |
+
"grad_norm": 0.66796875,
|
| 355 |
+
"learning_rate": 4.934316386673022e-05,
|
| 356 |
+
"logps_logqs/chosen": 0.07063151150941849,
|
| 357 |
+
"logqs/chosen": -0.7815223932266235,
|
| 358 |
+
"logqs/rejected": -0.9079385995864868,
|
| 359 |
+
"logqs_logps/rejected": 0.041664548218250275,
|
| 360 |
+
"loss": 0.2126,
|
| 361 |
+
"step": 310
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.5135921356204233,
|
| 365 |
+
"grad_norm": 0.61328125,
|
| 366 |
+
"learning_rate": 4.923259155947964e-05,
|
| 367 |
+
"logps_logqs/chosen": 0.071230448782444,
|
| 368 |
+
"logqs/chosen": -0.8096257448196411,
|
| 369 |
+
"logqs/rejected": -0.9752206802368164,
|
| 370 |
+
"logqs_logps/rejected": 0.03939716890454292,
|
| 371 |
+
"loss": 0.2194,
|
| 372 |
+
"step": 320
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.5296418898585615,
|
| 376 |
+
"grad_norm": 0.6640625,
|
| 377 |
+
"learning_rate": 4.911356578255139e-05,
|
| 378 |
+
"logps_logqs/chosen": 0.07356056571006775,
|
| 379 |
+
"logqs/chosen": -0.7827819585800171,
|
| 380 |
+
"logqs/rejected": -0.9489104151725769,
|
| 381 |
+
"logqs_logps/rejected": 0.040141116827726364,
|
| 382 |
+
"loss": 0.2145,
|
| 383 |
+
"step": 330
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.5456916440966998,
|
| 387 |
+
"grad_norm": 0.7109375,
|
| 388 |
+
"learning_rate": 4.898612805774447e-05,
|
| 389 |
+
"logps_logqs/chosen": 0.07881536334753036,
|
| 390 |
+
"logqs/chosen": -0.8095147013664246,
|
| 391 |
+
"logqs/rejected": -0.9313008189201355,
|
| 392 |
+
"logqs_logps/rejected": 0.03702981770038605,
|
| 393 |
+
"loss": 0.2165,
|
| 394 |
+
"step": 340
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.561741398334838,
|
| 398 |
+
"grad_norm": 0.64453125,
|
| 399 |
+
"learning_rate": 4.885032284134165e-05,
|
| 400 |
+
"logps_logqs/chosen": 0.07632436603307724,
|
| 401 |
+
"logqs/chosen": -0.7979615926742554,
|
| 402 |
+
"logqs/rejected": -0.9368084073066711,
|
| 403 |
+
"logqs_logps/rejected": 0.03654230386018753,
|
| 404 |
+
"loss": 0.2113,
|
| 405 |
+
"step": 350
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.5777911525729762,
|
| 409 |
+
"grad_norm": 0.62109375,
|
| 410 |
+
"learning_rate": 4.8706197508600984e-05,
|
| 411 |
+
"logps_logqs/chosen": 0.07344283908605576,
|
| 412 |
+
"logqs/chosen": -0.8101444244384766,
|
| 413 |
+
"logqs/rejected": -0.9295485615730286,
|
| 414 |
+
"logqs_logps/rejected": 0.03901313990354538,
|
| 415 |
+
"loss": 0.2151,
|
| 416 |
+
"step": 360
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5938409068111145,
|
| 420 |
+
"grad_norm": 0.62890625,
|
| 421 |
+
"learning_rate": 4.855380233722915e-05,
|
| 422 |
+
"logps_logqs/chosen": 0.07421533018350601,
|
| 423 |
+
"logqs/chosen": -0.7826108336448669,
|
| 424 |
+
"logqs/rejected": -0.9229670763015747,
|
| 425 |
+
"logqs_logps/rejected": 0.038314513862133026,
|
| 426 |
+
"loss": 0.2129,
|
| 427 |
+
"step": 370
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.6098906610492527,
|
| 431 |
+
"grad_norm": 0.62890625,
|
| 432 |
+
"learning_rate": 4.839319048984217e-05,
|
| 433 |
+
"logps_logqs/chosen": 0.07273373752832413,
|
| 434 |
+
"logqs/chosen": -0.7798442840576172,
|
| 435 |
+
"logqs/rejected": -0.9371780157089233,
|
| 436 |
+
"logqs_logps/rejected": 0.03713950887322426,
|
| 437 |
+
"loss": 0.2067,
|
| 438 |
+
"step": 380
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.6259404152873909,
|
| 442 |
+
"grad_norm": 0.62890625,
|
| 443 |
+
"learning_rate": 4.822441799541979e-05,
|
| 444 |
+
"logps_logqs/chosen": 0.07329441606998444,
|
| 445 |
+
"logqs/chosen": -0.8102380037307739,
|
| 446 |
+
"logqs/rejected": -0.95549476146698,
|
| 447 |
+
"logqs_logps/rejected": 0.036079905927181244,
|
| 448 |
+
"loss": 0.2158,
|
| 449 |
+
"step": 390
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.6419901695255291,
|
| 453 |
+
"grad_norm": 0.66796875,
|
| 454 |
+
"learning_rate": 4.8047543729759936e-05,
|
| 455 |
+
"logps_logqs/chosen": 0.07404083013534546,
|
| 456 |
+
"logqs/chosen": -0.7847949266433716,
|
| 457 |
+
"logqs/rejected": -0.9250528216362,
|
| 458 |
+
"logqs_logps/rejected": 0.036556728184223175,
|
| 459 |
+
"loss": 0.2105,
|
| 460 |
+
"step": 400
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.6580399237636674,
|
| 464 |
+
"grad_norm": 0.69140625,
|
| 465 |
+
"learning_rate": 4.786262939494007e-05,
|
| 466 |
+
"logps_logqs/chosen": 0.07539906352758408,
|
| 467 |
+
"logqs/chosen": -0.8072575330734253,
|
| 468 |
+
"logqs/rejected": -0.9661371111869812,
|
| 469 |
+
"logqs_logps/rejected": 0.03662776201963425,
|
| 470 |
+
"loss": 0.2154,
|
| 471 |
+
"step": 410
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6740896780018056,
|
| 475 |
+
"grad_norm": 0.66015625,
|
| 476 |
+
"learning_rate": 4.766973949779261e-05,
|
| 477 |
+
"logps_logqs/chosen": 0.0744672566652298,
|
| 478 |
+
"logqs/chosen": -0.787712574005127,
|
| 479 |
+
"logqs/rejected": -0.9350829124450684,
|
| 480 |
+
"logqs_logps/rejected": 0.0353250689804554,
|
| 481 |
+
"loss": 0.2074,
|
| 482 |
+
"step": 420
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.6901394322399438,
|
| 486 |
+
"grad_norm": 0.62890625,
|
| 487 |
+
"learning_rate": 4.746894132740186e-05,
|
| 488 |
+
"logps_logqs/chosen": 0.07364196330308914,
|
| 489 |
+
"logqs/chosen": -0.7813644409179688,
|
| 490 |
+
"logqs/rejected": -0.9208289980888367,
|
| 491 |
+
"logqs_logps/rejected": 0.0356430858373642,
|
| 492 |
+
"loss": 0.2053,
|
| 493 |
+
"step": 430
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.706189186478082,
|
| 497 |
+
"grad_norm": 0.56640625,
|
| 498 |
+
"learning_rate": 4.726030493163044e-05,
|
| 499 |
+
"logps_logqs/chosen": 0.07587061077356339,
|
| 500 |
+
"logqs/chosen": -0.7853146195411682,
|
| 501 |
+
"logqs/rejected": -0.8944910168647766,
|
| 502 |
+
"logqs_logps/rejected": 0.03531279042363167,
|
| 503 |
+
"loss": 0.2078,
|
| 504 |
+
"step": 440
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.7222389407162203,
|
| 508 |
+
"grad_norm": 0.68359375,
|
| 509 |
+
"learning_rate": 4.7043903092683314e-05,
|
| 510 |
+
"logps_logqs/chosen": 0.07814273238182068,
|
| 511 |
+
"logqs/chosen": -0.8133522272109985,
|
| 512 |
+
"logqs/rejected": -0.9508693814277649,
|
| 513 |
+
"logqs_logps/rejected": 0.030778918415308,
|
| 514 |
+
"loss": 0.2104,
|
| 515 |
+
"step": 450
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.7382886949543586,
|
| 519 |
+
"grad_norm": 0.72265625,
|
| 520 |
+
"learning_rate": 4.6819811301717885e-05,
|
| 521 |
+
"logps_logqs/chosen": 0.07763786613941193,
|
| 522 |
+
"logqs/chosen": -0.8061367869377136,
|
| 523 |
+
"logqs/rejected": -0.9445575475692749,
|
| 524 |
+
"logqs_logps/rejected": 0.035373255610466,
|
| 525 |
+
"loss": 0.2084,
|
| 526 |
+
"step": 460
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.7543384491924967,
|
| 530 |
+
"grad_norm": 0.59765625,
|
| 531 |
+
"learning_rate": 4.6588107732509134e-05,
|
| 532 |
+
"logps_logqs/chosen": 0.07325359433889389,
|
| 533 |
+
"logqs/chosen": -0.7970572710037231,
|
| 534 |
+
"logqs/rejected": -0.9243482351303101,
|
| 535 |
+
"logqs_logps/rejected": 0.03834759443998337,
|
| 536 |
+
"loss": 0.2088,
|
| 537 |
+
"step": 470
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.770388203430635,
|
| 541 |
+
"grad_norm": 0.5625,
|
| 542 |
+
"learning_rate": 4.634887321417895e-05,
|
| 543 |
+
"logps_logqs/chosen": 0.0732722282409668,
|
| 544 |
+
"logqs/chosen": -0.7719672918319702,
|
| 545 |
+
"logqs/rejected": -0.9193938970565796,
|
| 546 |
+
"logqs_logps/rejected": 0.03625096380710602,
|
| 547 |
+
"loss": 0.2075,
|
| 548 |
+
"step": 480
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7864379576687732,
|
| 552 |
+
"grad_norm": 0.90234375,
|
| 553 |
+
"learning_rate": 4.6102191202999065e-05,
|
| 554 |
+
"logps_logqs/chosen": 0.07290570437908173,
|
| 555 |
+
"logqs/chosen": -0.8092619180679321,
|
| 556 |
+
"logqs/rejected": -0.9642523527145386,
|
| 557 |
+
"logqs_logps/rejected": 0.03311945125460625,
|
| 558 |
+
"loss": 0.2045,
|
| 559 |
+
"step": 490
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 0.8024877119069114,
|
| 563 |
+
"grad_norm": 0.6015625,
|
| 564 |
+
"learning_rate": 4.5848147753277656e-05,
|
| 565 |
+
"logps_logqs/chosen": 0.07134760171175003,
|
| 566 |
+
"logqs/chosen": -0.7773372530937195,
|
| 567 |
+
"logqs/rejected": -0.9478441476821899,
|
| 568 |
+
"logqs_logps/rejected": 0.03295496851205826,
|
| 569 |
+
"loss": 0.2049,
|
| 570 |
+
"step": 500
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.8185374661450496,
|
| 574 |
+
"grad_norm": 0.5859375,
|
| 575 |
+
"learning_rate": 4.5586831487339485e-05,
|
| 576 |
+
"logps_logqs/chosen": 0.07219503819942474,
|
| 577 |
+
"logqs/chosen": -0.7950411438941956,
|
| 578 |
+
"logqs/rejected": -0.9455002546310425,
|
| 579 |
+
"logqs_logps/rejected": 0.03765324503183365,
|
| 580 |
+
"loss": 0.208,
|
| 581 |
+
"step": 510
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.8345872203831879,
|
| 585 |
+
"grad_norm": 0.62109375,
|
| 586 |
+
"learning_rate": 4.531833356461027e-05,
|
| 587 |
+
"logps_logqs/chosen": 0.06849464029073715,
|
| 588 |
+
"logqs/chosen": -0.7819440364837646,
|
| 589 |
+
"logqs/rejected": -0.9597524404525757,
|
| 590 |
+
"logqs_logps/rejected": 0.04056422412395477,
|
| 591 |
+
"loss": 0.2084,
|
| 592 |
+
"step": 520
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.8506369746213261,
|
| 596 |
+
"grad_norm": 0.6171875,
|
| 597 |
+
"learning_rate": 4.5042747649816006e-05,
|
| 598 |
+
"logps_logqs/chosen": 0.07160626351833344,
|
| 599 |
+
"logqs/chosen": -0.7721427083015442,
|
| 600 |
+
"logqs/rejected": -0.94866544008255,
|
| 601 |
+
"logqs_logps/rejected": 0.03749927878379822,
|
| 602 |
+
"loss": 0.2068,
|
| 603 |
+
"step": 530
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.8666867288594643,
|
| 607 |
+
"grad_norm": 0.5546875,
|
| 608 |
+
"learning_rate": 4.476016988030826e-05,
|
| 609 |
+
"logps_logqs/chosen": 0.07214485853910446,
|
| 610 |
+
"logqs/chosen": -0.7874671816825867,
|
| 611 |
+
"logqs/rejected": -0.9504098892211914,
|
| 612 |
+
"logqs_logps/rejected": 0.03507527709007263,
|
| 613 |
+
"loss": 0.2045,
|
| 614 |
+
"step": 540
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.8827364830976026,
|
| 618 |
+
"grad_norm": 0.54296875,
|
| 619 |
+
"learning_rate": 4.447069883252696e-05,
|
| 620 |
+
"logps_logqs/chosen": 0.07395409047603607,
|
| 621 |
+
"logqs/chosen": -0.771978497505188,
|
| 622 |
+
"logqs/rejected": -0.9129235148429871,
|
| 623 |
+
"logqs_logps/rejected": 0.039593033492565155,
|
| 624 |
+
"loss": 0.205,
|
| 625 |
+
"step": 550
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.8987862373357408,
|
| 629 |
+
"grad_norm": 0.61328125,
|
| 630 |
+
"learning_rate": 4.417443548761227e-05,
|
| 631 |
+
"logps_logqs/chosen": 0.07299650460481644,
|
| 632 |
+
"logqs/chosen": -0.7979342341423035,
|
| 633 |
+
"logqs/rejected": -0.9357426762580872,
|
| 634 |
+
"logqs_logps/rejected": 0.03937726467847824,
|
| 635 |
+
"loss": 0.2083,
|
| 636 |
+
"step": 560
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.9148359915738791,
|
| 640 |
+
"grad_norm": 0.57421875,
|
| 641 |
+
"learning_rate": 4.387148319617763e-05,
|
| 642 |
+
"logps_logqs/chosen": 0.06836996972560883,
|
| 643 |
+
"logqs/chosen": -0.7659951448440552,
|
| 644 |
+
"logqs/rejected": -0.8732549548149109,
|
| 645 |
+
"logqs_logps/rejected": 0.039081670343875885,
|
| 646 |
+
"loss": 0.203,
|
| 647 |
+
"step": 570
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.9308857458120172,
|
| 651 |
+
"grad_norm": 0.57421875,
|
| 652 |
+
"learning_rate": 4.356194764225618e-05,
|
| 653 |
+
"logps_logqs/chosen": 0.07397963851690292,
|
| 654 |
+
"logqs/chosen": -0.7891489863395691,
|
| 655 |
+
"logqs/rejected": -0.9294188618659973,
|
| 656 |
+
"logqs_logps/rejected": 0.03416060656309128,
|
| 657 |
+
"loss": 0.2024,
|
| 658 |
+
"step": 580
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.9469355000501555,
|
| 662 |
+
"grad_norm": 0.703125,
|
| 663 |
+
"learning_rate": 4.3245936806433205e-05,
|
| 664 |
+
"logps_logqs/chosen": 0.07466734945774078,
|
| 665 |
+
"logqs/chosen": -0.7958794832229614,
|
| 666 |
+
"logqs/rejected": -0.953458309173584,
|
| 667 |
+
"logqs_logps/rejected": 0.035235695540905,
|
| 668 |
+
"loss": 0.206,
|
| 669 |
+
"step": 590
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.9629852542882937,
|
| 673 |
+
"grad_norm": 0.5859375,
|
| 674 |
+
"learning_rate": 4.292356092817739e-05,
|
| 675 |
+
"logps_logqs/chosen": 0.07136549055576324,
|
| 676 |
+
"logqs/chosen": -0.7849777936935425,
|
| 677 |
+
"logqs/rejected": -0.9327009320259094,
|
| 678 |
+
"logqs_logps/rejected": 0.038332488387823105,
|
| 679 |
+
"loss": 0.2041,
|
| 680 |
+
"step": 600
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.979035008526432,
|
| 684 |
+
"grad_norm": 0.55859375,
|
| 685 |
+
"learning_rate": 4.259493246738409e-05,
|
| 686 |
+
"logps_logqs/chosen": 0.06976237148046494,
|
| 687 |
+
"logqs/chosen": -0.7658575773239136,
|
| 688 |
+
"logqs/rejected": -0.9027583003044128,
|
| 689 |
+
"logqs_logps/rejected": 0.03464614599943161,
|
| 690 |
+
"loss": 0.1993,
|
| 691 |
+
"step": 610
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.9950847627645701,
|
| 695 |
+
"grad_norm": 0.55078125,
|
| 696 |
+
"learning_rate": 4.226016606514411e-05,
|
| 697 |
+
"logps_logqs/chosen": 0.0706130638718605,
|
| 698 |
+
"logqs/chosen": -0.7847878932952881,
|
| 699 |
+
"logqs/rejected": -0.9509286880493164,
|
| 700 |
+
"logqs_logps/rejected": 0.03653167933225632,
|
| 701 |
+
"loss": 0.2035,
|
| 702 |
+
"step": 620
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.0111345170027084,
|
| 706 |
+
"grad_norm": 0.53515625,
|
| 707 |
+
"learning_rate": 4.191937850375136e-05,
|
| 708 |
+
"logps_logqs/chosen": 0.06913185119628906,
|
| 709 |
+
"logqs/chosen": -0.7918249368667603,
|
| 710 |
+
"logqs/rejected": -0.9458308219909668,
|
| 711 |
+
"logqs_logps/rejected": 0.030862990766763687,
|
| 712 |
+
"loss": 0.1886,
|
| 713 |
+
"step": 630
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.0271842712408465,
|
| 717 |
+
"grad_norm": 0.5625,
|
| 718 |
+
"learning_rate": 4.157268866596381e-05,
|
| 719 |
+
"logps_logqs/chosen": 0.0690259039402008,
|
| 720 |
+
"logqs/chosen": -0.7895857095718384,
|
| 721 |
+
"logqs/rejected": -0.9810531735420227,
|
| 722 |
+
"logqs_logps/rejected": 0.03111192025244236,
|
| 723 |
+
"loss": 0.187,
|
| 724 |
+
"step": 640
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.043234025478985,
|
| 728 |
+
"grad_norm": 0.53125,
|
| 729 |
+
"learning_rate": 4.1220217493531494e-05,
|
| 730 |
+
"logps_logqs/chosen": 0.06900795549154282,
|
| 731 |
+
"logqs/chosen": -0.7902609705924988,
|
| 732 |
+
"logqs/rejected": -0.9747546911239624,
|
| 733 |
+
"logqs_logps/rejected": 0.031710296869277954,
|
| 734 |
+
"loss": 0.1845,
|
| 735 |
+
"step": 650
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 1.059283779717123,
|
| 739 |
+
"grad_norm": 0.515625,
|
| 740 |
+
"learning_rate": 4.086208794500637e-05,
|
| 741 |
+
"logps_logqs/chosen": 0.07102300226688385,
|
| 742 |
+
"logqs/chosen": -0.7767165899276733,
|
| 743 |
+
"logqs/rejected": -0.9414197206497192,
|
| 744 |
+
"logqs_logps/rejected": 0.030909577384591103,
|
| 745 |
+
"loss": 0.1807,
|
| 746 |
+
"step": 660
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.0753335339552612,
|
| 750 |
+
"grad_norm": 0.494140625,
|
| 751 |
+
"learning_rate": 4.049842495284858e-05,
|
| 752 |
+
"logps_logqs/chosen": 0.06985093653202057,
|
| 753 |
+
"logqs/chosen": -0.7880310416221619,
|
| 754 |
+
"logqs/rejected": -0.9546613693237305,
|
| 755 |
+
"logqs_logps/rejected": 0.030261676758527756,
|
| 756 |
+
"loss": 0.186,
|
| 757 |
+
"step": 670
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 1.0913832881933996,
|
| 761 |
+
"grad_norm": 0.50390625,
|
| 762 |
+
"learning_rate": 4.012935537984414e-05,
|
| 763 |
+
"logps_logqs/chosen": 0.07058895379304886,
|
| 764 |
+
"logqs/chosen": -0.7935397624969482,
|
| 765 |
+
"logqs/rejected": -1.0117493867874146,
|
| 766 |
+
"logqs_logps/rejected": 0.029221097007393837,
|
| 767 |
+
"loss": 0.1835,
|
| 768 |
+
"step": 680
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.1074330424315377,
|
| 772 |
+
"grad_norm": 0.6171875,
|
| 773 |
+
"learning_rate": 3.9755007974849135e-05,
|
| 774 |
+
"logps_logqs/chosen": 0.06916572153568268,
|
| 775 |
+
"logqs/chosen": -0.7878638505935669,
|
| 776 |
+
"logqs/rejected": -0.9999720454216003,
|
| 777 |
+
"logqs_logps/rejected": 0.030744653195142746,
|
| 778 |
+
"loss": 0.1857,
|
| 779 |
+
"step": 690
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.123482796669676,
|
| 783 |
+
"grad_norm": 0.57421875,
|
| 784 |
+
"learning_rate": 3.93755133278762e-05,
|
| 785 |
+
"logps_logqs/chosen": 0.07014169543981552,
|
| 786 |
+
"logqs/chosen": -0.7882756590843201,
|
| 787 |
+
"logqs/rejected": -1.0001386404037476,
|
| 788 |
+
"logqs_logps/rejected": 0.02842717245221138,
|
| 789 |
+
"loss": 0.1832,
|
| 790 |
+
"step": 700
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.1395325509078142,
|
| 794 |
+
"grad_norm": 0.59375,
|
| 795 |
+
"learning_rate": 3.899100382453845e-05,
|
| 796 |
+
"logps_logqs/chosen": 0.0701603814959526,
|
| 797 |
+
"logqs/chosen": -0.7642520070075989,
|
| 798 |
+
"logqs/rejected": -0.9544415473937988,
|
| 799 |
+
"logqs_logps/rejected": 0.027573522180318832,
|
| 800 |
+
"loss": 0.1806,
|
| 801 |
+
"step": 710
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.1555823051459524,
|
| 805 |
+
"grad_norm": 0.50390625,
|
| 806 |
+
"learning_rate": 3.8601613599867156e-05,
|
| 807 |
+
"logps_logqs/chosen": 0.06890274584293365,
|
| 808 |
+
"logqs/chosen": -0.7708860635757446,
|
| 809 |
+
"logqs/rejected": -0.9537287950515747,
|
| 810 |
+
"logqs_logps/rejected": 0.028619807213544846,
|
| 811 |
+
"loss": 0.1804,
|
| 812 |
+
"step": 720
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 1.1716320593840908,
|
| 816 |
+
"grad_norm": 0.609375,
|
| 817 |
+
"learning_rate": 3.8207478491519216e-05,
|
| 818 |
+
"logps_logqs/chosen": 0.0682586207985878,
|
| 819 |
+
"logqs/chosen": -0.7696245908737183,
|
| 820 |
+
"logqs/rejected": -0.9157883524894714,
|
| 821 |
+
"logqs_logps/rejected": 0.028189942240715027,
|
| 822 |
+
"loss": 0.1814,
|
| 823 |
+
"step": 730
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 1.187681813622229,
|
| 827 |
+
"grad_norm": 0.59375,
|
| 828 |
+
"learning_rate": 3.780873599239044e-05,
|
| 829 |
+
"logps_logqs/chosen": 0.07058210670948029,
|
| 830 |
+
"logqs/chosen": -0.801421046257019,
|
| 831 |
+
"logqs/rejected": -0.9591751098632812,
|
| 832 |
+
"logqs_logps/rejected": 0.03063536249101162,
|
| 833 |
+
"loss": 0.1853,
|
| 834 |
+
"step": 740
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.203731567860367,
|
| 838 |
+
"grad_norm": 0.55859375,
|
| 839 |
+
"learning_rate": 3.740552520265167e-05,
|
| 840 |
+
"logps_logqs/chosen": 0.06886611133813858,
|
| 841 |
+
"logqs/chosen": -0.792157769203186,
|
| 842 |
+
"logqs/rejected": -0.9370824694633484,
|
| 843 |
+
"logqs_logps/rejected": 0.02984955906867981,
|
| 844 |
+
"loss": 0.1833,
|
| 845 |
+
"step": 750
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 1.2197813220985054,
|
| 849 |
+
"grad_norm": 0.55859375,
|
| 850 |
+
"learning_rate": 3.699798678122403e-05,
|
| 851 |
+
"logps_logqs/chosen": 0.07179007679224014,
|
| 852 |
+
"logqs/chosen": -0.7622597813606262,
|
| 853 |
+
"logqs/rejected": -0.9382703900337219,
|
| 854 |
+
"logqs_logps/rejected": 0.029157549142837524,
|
| 855 |
+
"loss": 0.1793,
|
| 856 |
+
"step": 760
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.2358310763366436,
|
| 860 |
+
"grad_norm": 0.546875,
|
| 861 |
+
"learning_rate": 3.6586262896710476e-05,
|
| 862 |
+
"logps_logqs/chosen": 0.06790535151958466,
|
| 863 |
+
"logqs/chosen": -0.7676440477371216,
|
| 864 |
+
"logqs/rejected": -0.928793728351593,
|
| 865 |
+
"logqs_logps/rejected": 0.03030979633331299,
|
| 866 |
+
"loss": 0.1808,
|
| 867 |
+
"step": 770
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 1.2518808305747817,
|
| 871 |
+
"grad_norm": 0.51171875,
|
| 872 |
+
"learning_rate": 3.61704971778007e-05,
|
| 873 |
+
"logps_logqs/chosen": 0.06736615300178528,
|
| 874 |
+
"logqs/chosen": -0.770586371421814,
|
| 875 |
+
"logqs/rejected": -0.9197053909301758,
|
| 876 |
+
"logqs_logps/rejected": 0.03245236724615097,
|
| 877 |
+
"loss": 0.1799,
|
| 878 |
+
"step": 780
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.26793058481292,
|
| 882 |
+
"grad_norm": 0.4765625,
|
| 883 |
+
"learning_rate": 3.575083466316664e-05,
|
| 884 |
+
"logps_logqs/chosen": 0.06956754624843597,
|
| 885 |
+
"logqs/chosen": -0.7774965167045593,
|
| 886 |
+
"logqs/rejected": -0.9702298045158386,
|
| 887 |
+
"logqs_logps/rejected": 0.030748773366212845,
|
| 888 |
+
"loss": 0.1831,
|
| 889 |
+
"step": 790
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 1.2839803390510582,
|
| 893 |
+
"grad_norm": 0.5078125,
|
| 894 |
+
"learning_rate": 3.532742175086621e-05,
|
| 895 |
+
"logps_logqs/chosen": 0.06920956075191498,
|
| 896 |
+
"logqs/chosen": -0.7709556818008423,
|
| 897 |
+
"logqs/rejected": -0.9387216567993164,
|
| 898 |
+
"logqs_logps/rejected": 0.027966167777776718,
|
| 899 |
+
"loss": 0.1788,
|
| 900 |
+
"step": 800
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 1.3000300932891964,
|
| 904 |
+
"grad_norm": 0.54296875,
|
| 905 |
+
"learning_rate": 3.490040614727272e-05,
|
| 906 |
+
"logps_logqs/chosen": 0.06927359104156494,
|
| 907 |
+
"logqs/chosen": -0.7772814035415649,
|
| 908 |
+
"logqs/rejected": -0.9691821932792664,
|
| 909 |
+
"logqs_logps/rejected": 0.028416061773896217,
|
| 910 |
+
"loss": 0.1781,
|
| 911 |
+
"step": 810
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 1.3160798475273348,
|
| 915 |
+
"grad_norm": 0.5078125,
|
| 916 |
+
"learning_rate": 3.446993681554797e-05,
|
| 917 |
+
"logps_logqs/chosen": 0.07202474772930145,
|
| 918 |
+
"logqs/chosen": -0.7855108380317688,
|
| 919 |
+
"logqs/rejected": -0.9708096385002136,
|
| 920 |
+
"logqs_logps/rejected": 0.030036652460694313,
|
| 921 |
+
"loss": 0.1793,
|
| 922 |
+
"step": 820
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 1.332129601765473,
|
| 926 |
+
"grad_norm": 0.65234375,
|
| 927 |
+
"learning_rate": 3.403616392367681e-05,
|
| 928 |
+
"logps_logqs/chosen": 0.07054628431797028,
|
| 929 |
+
"logqs/chosen": -0.7540086507797241,
|
| 930 |
+
"logqs/rejected": -0.9042137265205383,
|
| 931 |
+
"logqs_logps/rejected": 0.03296629339456558,
|
| 932 |
+
"loss": 0.1818,
|
| 933 |
+
"step": 830
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.3481793560036113,
|
| 937 |
+
"grad_norm": 0.5078125,
|
| 938 |
+
"learning_rate": 3.3599238792081575e-05,
|
| 939 |
+
"logps_logqs/chosen": 0.07161605358123779,
|
| 940 |
+
"logqs/chosen": -0.7662056684494019,
|
| 941 |
+
"logqs/rejected": -0.9423080682754517,
|
| 942 |
+
"logqs_logps/rejected": 0.031152984127402306,
|
| 943 |
+
"loss": 0.1786,
|
| 944 |
+
"step": 840
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 1.3642291102417494,
|
| 948 |
+
"grad_norm": 0.5390625,
|
| 949 |
+
"learning_rate": 3.315931384083431e-05,
|
| 950 |
+
"logps_logqs/chosen": 0.0672772079706192,
|
| 951 |
+
"logqs/chosen": -0.7698653936386108,
|
| 952 |
+
"logqs/rejected": -0.9608929753303528,
|
| 953 |
+
"logqs_logps/rejected": 0.031727343797683716,
|
| 954 |
+
"loss": 0.181,
|
| 955 |
+
"step": 850
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.3802788644798878,
|
| 959 |
+
"grad_norm": 0.5390625,
|
| 960 |
+
"learning_rate": 3.2716542536485474e-05,
|
| 961 |
+
"logps_logqs/chosen": 0.07064563035964966,
|
| 962 |
+
"logqs/chosen": -0.7671880125999451,
|
| 963 |
+
"logqs/rejected": -0.9258206486701965,
|
| 964 |
+
"logqs_logps/rejected": 0.03185782581567764,
|
| 965 |
+
"loss": 0.1802,
|
| 966 |
+
"step": 860
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 1.396328618718026,
|
| 970 |
+
"grad_norm": 0.59375,
|
| 971 |
+
"learning_rate": 3.2271079338527626e-05,
|
| 972 |
+
"logps_logqs/chosen": 0.07001273334026337,
|
| 973 |
+
"logqs/chosen": -0.7942059636116028,
|
| 974 |
+
"logqs/rejected": -0.9803145527839661,
|
| 975 |
+
"logqs_logps/rejected": 0.02900245226919651,
|
| 976 |
+
"loss": 0.1783,
|
| 977 |
+
"step": 870
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 1.412378372956164,
|
| 981 |
+
"grad_norm": 0.48828125,
|
| 982 |
+
"learning_rate": 3.1823079645512655e-05,
|
| 983 |
+
"logps_logqs/chosen": 0.06748739629983902,
|
| 984 |
+
"logqs/chosen": -0.7796913385391235,
|
| 985 |
+
"logqs/rejected": -1.0028189420700073,
|
| 986 |
+
"logqs_logps/rejected": 0.028837621212005615,
|
| 987 |
+
"loss": 0.1791,
|
| 988 |
+
"step": 880
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 1.4284281271943025,
|
| 992 |
+
"grad_norm": 0.5234375,
|
| 993 |
+
"learning_rate": 3.137269974084139e-05,
|
| 994 |
+
"logps_logqs/chosen": 0.07031874358654022,
|
| 995 |
+
"logqs/chosen": -0.7788208723068237,
|
| 996 |
+
"logqs/rejected": -0.9545127153396606,
|
| 997 |
+
"logqs_logps/rejected": 0.029628973454236984,
|
| 998 |
+
"loss": 0.179,
|
| 999 |
+
"step": 890
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 1.4444778814324406,
|
| 1003 |
+
"grad_norm": 0.486328125,
|
| 1004 |
+
"learning_rate": 3.092009673824469e-05,
|
| 1005 |
+
"logps_logqs/chosen": 0.07582792639732361,
|
| 1006 |
+
"logqs/chosen": -0.8078464269638062,
|
| 1007 |
+
"logqs/rejected": -0.9633975028991699,
|
| 1008 |
+
"logqs_logps/rejected": 0.02939186617732048,
|
| 1009 |
+
"loss": 0.1885,
|
| 1010 |
+
"step": 900
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.4605276356705787,
|
| 1014 |
+
"grad_norm": 0.53515625,
|
| 1015 |
+
"learning_rate": 3.0465428526974665e-05,
|
| 1016 |
+
"logps_logqs/chosen": 0.07225798070430756,
|
| 1017 |
+
"logqs/chosen": -0.7663235068321228,
|
| 1018 |
+
"logqs/rejected": -0.9817646741867065,
|
| 1019 |
+
"logqs_logps/rejected": 0.028154581785202026,
|
| 1020 |
+
"loss": 0.178,
|
| 1021 |
+
"step": 910
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 1.4765773899087171,
|
| 1025 |
+
"grad_norm": 0.49609375,
|
| 1026 |
+
"learning_rate": 3.000885371672554e-05,
|
| 1027 |
+
"logps_logqs/chosen": 0.07049473375082016,
|
| 1028 |
+
"logqs/chosen": -0.7796770930290222,
|
| 1029 |
+
"logqs/rejected": -0.9203440546989441,
|
| 1030 |
+
"logqs_logps/rejected": 0.028858328238129616,
|
| 1031 |
+
"loss": 0.1765,
|
| 1032 |
+
"step": 920
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.4926271441468553,
|
| 1036 |
+
"grad_norm": 0.53125,
|
| 1037 |
+
"learning_rate": 2.9550531582303082e-05,
|
| 1038 |
+
"logps_logqs/chosen": 0.07375530898571014,
|
| 1039 |
+
"logqs/chosen": -0.7871274352073669,
|
| 1040 |
+
"logqs/rejected": -0.950308620929718,
|
| 1041 |
+
"logqs_logps/rejected": 0.030121903866529465,
|
| 1042 |
+
"loss": 0.1821,
|
| 1043 |
+
"step": 930
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 1.5086768983849934,
|
| 1047 |
+
"grad_norm": 0.462890625,
|
| 1048 |
+
"learning_rate": 2.909062200806208e-05,
|
| 1049 |
+
"logps_logqs/chosen": 0.06755580753087997,
|
| 1050 |
+
"logqs/chosen": -0.7797432541847229,
|
| 1051 |
+
"logqs/rejected": -0.9579198956489563,
|
| 1052 |
+
"logqs_logps/rejected": 0.027673590928316116,
|
| 1053 |
+
"loss": 0.1738,
|
| 1054 |
+
"step": 940
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 1.5247266526231318,
|
| 1058 |
+
"grad_norm": 0.451171875,
|
| 1059 |
+
"learning_rate": 2.8629285432131083e-05,
|
| 1060 |
+
"logps_logqs/chosen": 0.06972555816173553,
|
| 1061 |
+
"logqs/chosen": -0.7843119502067566,
|
| 1062 |
+
"logqs/rejected": -0.971518874168396,
|
| 1063 |
+
"logqs_logps/rejected": 0.026922276243567467,
|
| 1064 |
+
"loss": 0.1771,
|
| 1065 |
+
"step": 950
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 1.54077640686127,
|
| 1069 |
+
"grad_norm": 0.5234375,
|
| 1070 |
+
"learning_rate": 2.8166682790444116e-05,
|
| 1071 |
+
"logps_logqs/chosen": 0.0729864090681076,
|
| 1072 |
+
"logqs/chosen": -0.7914843559265137,
|
| 1073 |
+
"logqs/rejected": -0.9729417562484741,
|
| 1074 |
+
"logqs_logps/rejected": 0.02641429379582405,
|
| 1075 |
+
"loss": 0.1788,
|
| 1076 |
+
"step": 960
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.556826161099408,
|
| 1080 |
+
"grad_norm": 0.4921875,
|
| 1081 |
+
"learning_rate": 2.7702975460598547e-05,
|
| 1082 |
+
"logps_logqs/chosen": 0.06970744580030441,
|
| 1083 |
+
"logqs/chosen": -0.7867680788040161,
|
| 1084 |
+
"logqs/rejected": -0.9702849388122559,
|
| 1085 |
+
"logqs_logps/rejected": 0.026300692930817604,
|
| 1086 |
+
"loss": 0.1753,
|
| 1087 |
+
"step": 970
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.5728759153375464,
|
| 1091 |
+
"grad_norm": 0.51171875,
|
| 1092 |
+
"learning_rate": 2.723832520555905e-05,
|
| 1093 |
+
"logps_logqs/chosen": 0.06844428926706314,
|
| 1094 |
+
"logqs/chosen": -0.7796690464019775,
|
| 1095 |
+
"logqs/rejected": -0.9721530079841614,
|
| 1096 |
+
"logqs_logps/rejected": 0.03003394976258278,
|
| 1097 |
+
"loss": 0.1775,
|
| 1098 |
+
"step": 980
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 1.5889256695756846,
|
| 1102 |
+
"grad_norm": 0.50390625,
|
| 1103 |
+
"learning_rate": 2.677289411722702e-05,
|
| 1104 |
+
"logps_logqs/chosen": 0.06915868073701859,
|
| 1105 |
+
"logqs/chosen": -0.7836161851882935,
|
| 1106 |
+
"logqs/rejected": -1.0066392421722412,
|
| 1107 |
+
"logqs_logps/rejected": 0.029805105179548264,
|
| 1108 |
+
"loss": 0.1834,
|
| 1109 |
+
"step": 990
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.6049754238138227,
|
| 1113 |
+
"grad_norm": 0.52734375,
|
| 1114 |
+
"learning_rate": 2.63068445598953e-05,
|
| 1115 |
+
"logps_logqs/chosen": 0.07069100439548492,
|
| 1116 |
+
"logqs/chosen": -0.7810501456260681,
|
| 1117 |
+
"logqs/rejected": -0.9559534788131714,
|
| 1118 |
+
"logqs_logps/rejected": 0.02675846591591835,
|
| 1119 |
+
"loss": 0.1799,
|
| 1120 |
+
"step": 1000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 1.6210251780519611,
|
| 1124 |
+
"grad_norm": 0.455078125,
|
| 1125 |
+
"learning_rate": 2.5840339113607854e-05,
|
| 1126 |
+
"logps_logqs/chosen": 0.06957536935806274,
|
| 1127 |
+
"logqs/chosen": -0.776276171207428,
|
| 1128 |
+
"logqs/rejected": -0.9653003811836243,
|
| 1129 |
+
"logqs_logps/rejected": 0.029132988303899765,
|
| 1130 |
+
"loss": 0.1764,
|
| 1131 |
+
"step": 1010
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 1.6370749322900993,
|
| 1135 |
+
"grad_norm": 0.4921875,
|
| 1136 |
+
"learning_rate": 2.53735405174442e-05,
|
| 1137 |
+
"logps_logqs/chosen": 0.06984798610210419,
|
| 1138 |
+
"logqs/chosen": -0.7904434204101562,
|
| 1139 |
+
"logqs/rejected": -0.9736318588256836,
|
| 1140 |
+
"logqs_logps/rejected": 0.02975938282907009,
|
| 1141 |
+
"loss": 0.1778,
|
| 1142 |
+
"step": 1020
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 1.6531246865282374,
|
| 1146 |
+
"grad_norm": 0.50390625,
|
| 1147 |
+
"learning_rate": 2.490661161274835e-05,
|
| 1148 |
+
"logps_logqs/chosen": 0.06985798478126526,
|
| 1149 |
+
"logqs/chosen": -0.7864385843276978,
|
| 1150 |
+
"logqs/rejected": -0.9617294073104858,
|
| 1151 |
+
"logqs_logps/rejected": 0.027367640286684036,
|
| 1152 |
+
"loss": 0.1774,
|
| 1153 |
+
"step": 1030
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 1.6691744407663758,
|
| 1157 |
+
"grad_norm": 0.4453125,
|
| 1158 |
+
"learning_rate": 2.443971528632205e-05,
|
| 1159 |
+
"logps_logqs/chosen": 0.07047738134860992,
|
| 1160 |
+
"logqs/chosen": -0.7980072498321533,
|
| 1161 |
+
"logqs/rejected": -0.9655786752700806,
|
| 1162 |
+
"logqs_logps/rejected": 0.028297554701566696,
|
| 1163 |
+
"loss": 0.1779,
|
| 1164 |
+
"step": 1040
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.6852241950045141,
|
| 1168 |
+
"grad_norm": 0.515625,
|
| 1169 |
+
"learning_rate": 2.3973014413602238e-05,
|
| 1170 |
+
"logps_logqs/chosen": 0.07119600474834442,
|
| 1171 |
+
"logqs/chosen": -0.7848029732704163,
|
| 1172 |
+
"logqs/rejected": -1.003758192062378,
|
| 1173 |
+
"logqs_logps/rejected": 0.029630709439516068,
|
| 1174 |
+
"loss": 0.1808,
|
| 1175 |
+
"step": 1050
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 1.701273949242652,
|
| 1179 |
+
"grad_norm": 0.498046875,
|
| 1180 |
+
"learning_rate": 2.3506671801842364e-05,
|
| 1181 |
+
"logps_logqs/chosen": 0.06773122400045395,
|
| 1182 |
+
"logqs/chosen": -0.8044145703315735,
|
| 1183 |
+
"logqs/rejected": -0.9743335843086243,
|
| 1184 |
+
"logqs_logps/rejected": 0.027200300246477127,
|
| 1185 |
+
"loss": 0.1769,
|
| 1186 |
+
"step": 1060
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.7173237034807904,
|
| 1190 |
+
"grad_norm": 0.4921875,
|
| 1191 |
+
"learning_rate": 2.3040850133317597e-05,
|
| 1192 |
+
"logps_logqs/chosen": 0.07197652757167816,
|
| 1193 |
+
"logqs/chosen": -0.7732763886451721,
|
| 1194 |
+
"logqs/rejected": -0.9615533947944641,
|
| 1195 |
+
"logqs_logps/rejected": 0.02922072447836399,
|
| 1196 |
+
"loss": 0.1801,
|
| 1197 |
+
"step": 1070
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 1.7333734577189288,
|
| 1201 |
+
"grad_norm": 0.48046875,
|
| 1202 |
+
"learning_rate": 2.2575711908573548e-05,
|
| 1203 |
+
"logps_logqs/chosen": 0.06941990554332733,
|
| 1204 |
+
"logqs/chosen": -0.7937701940536499,
|
| 1205 |
+
"logqs/rejected": -1.0073679685592651,
|
| 1206 |
+
"logqs_logps/rejected": 0.026350444182753563,
|
| 1207 |
+
"loss": 0.1764,
|
| 1208 |
+
"step": 1080
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.749423211957067,
|
| 1212 |
+
"grad_norm": 0.447265625,
|
| 1213 |
+
"learning_rate": 2.2111419389738435e-05,
|
| 1214 |
+
"logps_logqs/chosen": 0.07039657980203629,
|
| 1215 |
+
"logqs/chosen": -0.7764157056808472,
|
| 1216 |
+
"logqs/rejected": -0.9544513821601868,
|
| 1217 |
+
"logqs_logps/rejected": 0.02772611379623413,
|
| 1218 |
+
"loss": 0.1755,
|
| 1219 |
+
"step": 1090
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 1.765472966195205,
|
| 1223 |
+
"grad_norm": 0.5390625,
|
| 1224 |
+
"learning_rate": 2.1648134543918423e-05,
|
| 1225 |
+
"logps_logqs/chosen": 0.07292209565639496,
|
| 1226 |
+
"logqs/chosen": -0.7701107859611511,
|
| 1227 |
+
"logqs/rejected": -0.9618105888366699,
|
| 1228 |
+
"logqs_logps/rejected": 0.027484769001603127,
|
| 1229 |
+
"loss": 0.1762,
|
| 1230 |
+
"step": 1100
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 1.7815227204333435,
|
| 1234 |
+
"grad_norm": 0.427734375,
|
| 1235 |
+
"learning_rate": 2.1186018986695842e-05,
|
| 1236 |
+
"logps_logqs/chosen": 0.07149146497249603,
|
| 1237 |
+
"logqs/chosen": -0.7753132581710815,
|
| 1238 |
+
"logqs/rejected": -0.926436722278595,
|
| 1239 |
+
"logqs_logps/rejected": 0.029845744371414185,
|
| 1240 |
+
"loss": 0.1779,
|
| 1241 |
+
"step": 1110
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.7975724746714816,
|
| 1245 |
+
"grad_norm": 0.45703125,
|
| 1246 |
+
"learning_rate": 2.0725233925750063e-05,
|
| 1247 |
+
"logps_logqs/chosen": 0.06752609461545944,
|
| 1248 |
+
"logqs/chosen": -0.7796769738197327,
|
| 1249 |
+
"logqs/rejected": -0.9575828313827515,
|
| 1250 |
+
"logqs_logps/rejected": 0.03261734917759895,
|
| 1251 |
+
"loss": 0.1771,
|
| 1252 |
+
"step": 1120
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 1.8136222289096198,
|
| 1256 |
+
"grad_norm": 0.54296875,
|
| 1257 |
+
"learning_rate": 2.026594010462068e-05,
|
| 1258 |
+
"logps_logqs/chosen": 0.07001613825559616,
|
| 1259 |
+
"logqs/chosen": -0.7918448448181152,
|
| 1260 |
+
"logqs/rejected": -0.9749159812927246,
|
| 1261 |
+
"logqs_logps/rejected": 0.028455784544348717,
|
| 1262 |
+
"loss": 0.1789,
|
| 1263 |
+
"step": 1130
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.8296719831477581,
|
| 1267 |
+
"grad_norm": 0.478515625,
|
| 1268 |
+
"learning_rate": 1.980829774663256e-05,
|
| 1269 |
+
"logps_logqs/chosen": 0.07289810478687286,
|
| 1270 |
+
"logqs/chosen": -0.7838630676269531,
|
| 1271 |
+
"logqs/rejected": -0.9698160290718079,
|
| 1272 |
+
"logqs_logps/rejected": 0.02906595729291439,
|
| 1273 |
+
"loss": 0.1794,
|
| 1274 |
+
"step": 1140
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 1.8457217373858963,
|
| 1278 |
+
"grad_norm": 0.458984375,
|
| 1279 |
+
"learning_rate": 1.9352466499002422e-05,
|
| 1280 |
+
"logps_logqs/chosen": 0.0761161744594574,
|
| 1281 |
+
"logqs/chosen": -0.7819662094116211,
|
| 1282 |
+
"logqs/rejected": -0.9463040232658386,
|
| 1283 |
+
"logqs_logps/rejected": 0.028005197644233704,
|
| 1284 |
+
"loss": 0.183,
|
| 1285 |
+
"step": 1150
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 1.8617714916240344,
|
| 1289 |
+
"grad_norm": 0.474609375,
|
| 1290 |
+
"learning_rate": 1.8898605377146383e-05,
|
| 1291 |
+
"logps_logqs/chosen": 0.06957074254751205,
|
| 1292 |
+
"logqs/chosen": -0.7857328057289124,
|
| 1293 |
+
"logqs/rejected": -0.9639459848403931,
|
| 1294 |
+
"logqs_logps/rejected": 0.027649903669953346,
|
| 1295 |
+
"loss": 0.1753,
|
| 1296 |
+
"step": 1160
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 1.8778212458621728,
|
| 1300 |
+
"grad_norm": 0.48046875,
|
| 1301 |
+
"learning_rate": 1.8446872709207847e-05,
|
| 1302 |
+
"logps_logqs/chosen": 0.07362545281648636,
|
| 1303 |
+
"logqs/chosen": -0.7835830450057983,
|
| 1304 |
+
"logqs/rejected": -1.0045692920684814,
|
| 1305 |
+
"logqs_logps/rejected": 0.027295967563986778,
|
| 1306 |
+
"loss": 0.1806,
|
| 1307 |
+
"step": 1170
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 1.893871000100311,
|
| 1311 |
+
"grad_norm": 0.51171875,
|
| 1312 |
+
"learning_rate": 1.799742608082519e-05,
|
| 1313 |
+
"logps_logqs/chosen": 0.07236044853925705,
|
| 1314 |
+
"logqs/chosen": -0.7885088324546814,
|
| 1315 |
+
"logqs/rejected": -0.9487521052360535,
|
| 1316 |
+
"logqs_logps/rejected": 0.026073191314935684,
|
| 1317 |
+
"loss": 0.1723,
|
| 1318 |
+
"step": 1180
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 1.909920754338449,
|
| 1322 |
+
"grad_norm": 0.51171875,
|
| 1323 |
+
"learning_rate": 1.7550422280158513e-05,
|
| 1324 |
+
"logps_logqs/chosen": 0.0707259327173233,
|
| 1325 |
+
"logqs/chosen": -0.7704340815544128,
|
| 1326 |
+
"logqs/rejected": -0.9094502329826355,
|
| 1327 |
+
"logqs_logps/rejected": 0.027560651302337646,
|
| 1328 |
+
"loss": 0.1758,
|
| 1329 |
+
"step": 1190
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 1.9259705085765875,
|
| 1333 |
+
"grad_norm": 0.435546875,
|
| 1334 |
+
"learning_rate": 1.7106017243194487e-05,
|
| 1335 |
+
"logps_logqs/chosen": 0.07020822167396545,
|
| 1336 |
+
"logqs/chosen": -0.7692683935165405,
|
| 1337 |
+
"logqs/rejected": -0.9656769037246704,
|
| 1338 |
+
"logqs_logps/rejected": 0.02792223170399666,
|
| 1339 |
+
"loss": 0.177,
|
| 1340 |
+
"step": 1200
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.9420202628147256,
|
| 1344 |
+
"grad_norm": 0.53515625,
|
| 1345 |
+
"learning_rate": 1.6664365999348594e-05,
|
| 1346 |
+
"logps_logqs/chosen": 0.06943775713443756,
|
| 1347 |
+
"logqs/chosen": -0.7688643932342529,
|
| 1348 |
+
"logqs/rejected": -0.9713398218154907,
|
| 1349 |
+
"logqs_logps/rejected": 0.026859009638428688,
|
| 1350 |
+
"loss": 0.1748,
|
| 1351 |
+
"step": 1210
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 1.9580700170528638,
|
| 1355 |
+
"grad_norm": 0.51953125,
|
| 1356 |
+
"learning_rate": 1.6225622617383494e-05,
|
| 1357 |
+
"logps_logqs/chosen": 0.07070201635360718,
|
| 1358 |
+
"logqs/chosen": -0.7839328050613403,
|
| 1359 |
+
"logqs/rejected": -0.9712308645248413,
|
| 1360 |
+
"logqs_logps/rejected": 0.028167420998215675,
|
| 1361 |
+
"loss": 0.1763,
|
| 1362 |
+
"step": 1220
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 1.9741197712910021,
|
| 1366 |
+
"grad_norm": 0.4453125,
|
| 1367 |
+
"learning_rate": 1.578994015166263e-05,
|
| 1368 |
+
"logps_logqs/chosen": 0.07289667427539825,
|
| 1369 |
+
"logqs/chosen": -0.7838398814201355,
|
| 1370 |
+
"logqs/rejected": -0.9522368311882019,
|
| 1371 |
+
"logqs_logps/rejected": 0.02768387272953987,
|
| 1372 |
+
"loss": 0.1773,
|
| 1373 |
+
"step": 1230
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 1.9901695255291403,
|
| 1377 |
+
"grad_norm": 0.53515625,
|
| 1378 |
+
"learning_rate": 1.535747058875765e-05,
|
| 1379 |
+
"logps_logqs/chosen": 0.06881529092788696,
|
| 1380 |
+
"logqs/chosen": -0.790834903717041,
|
| 1381 |
+
"logqs/rejected": -0.9497320055961609,
|
| 1382 |
+
"logqs_logps/rejected": 0.029328888282179832,
|
| 1383 |
+
"loss": 0.1734,
|
| 1384 |
+
"step": 1240
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 2.0062192797672784,
|
| 1388 |
+
"grad_norm": 0.43359375,
|
| 1389 |
+
"learning_rate": 1.4928364794428307e-05,
|
| 1390 |
+
"logps_logqs/chosen": 0.07006263732910156,
|
| 1391 |
+
"logqs/chosen": -0.799282431602478,
|
| 1392 |
+
"logqs/rejected": -1.0095001459121704,
|
| 1393 |
+
"logqs_logps/rejected": 0.02767905592918396,
|
| 1394 |
+
"loss": 0.1732,
|
| 1395 |
+
"step": 1250
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 2.022269034005417,
|
| 1399 |
+
"grad_norm": 0.4296875,
|
| 1400 |
+
"learning_rate": 1.4502772460993385e-05,
|
| 1401 |
+
"logps_logqs/chosen": 0.06896007061004639,
|
| 1402 |
+
"logqs/chosen": -0.7676675319671631,
|
| 1403 |
+
"logqs/rejected": -0.9425755739212036,
|
| 1404 |
+
"logqs_logps/rejected": 0.02848168835043907,
|
| 1405 |
+
"loss": 0.1685,
|
| 1406 |
+
"step": 1260
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 2.038318788243555,
|
| 1410 |
+
"grad_norm": 0.462890625,
|
| 1411 |
+
"learning_rate": 1.4080842055110993e-05,
|
| 1412 |
+
"logps_logqs/chosen": 0.06711134314537048,
|
| 1413 |
+
"logqs/chosen": -0.7576900720596313,
|
| 1414 |
+
"logqs/rejected": -0.9405368566513062,
|
| 1415 |
+
"logqs_logps/rejected": 0.02635120414197445,
|
| 1416 |
+
"loss": 0.1689,
|
| 1417 |
+
"step": 1270
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 2.054368542481693,
|
| 1421 |
+
"grad_norm": 0.46484375,
|
| 1422 |
+
"learning_rate": 1.3662720765986341e-05,
|
| 1423 |
+
"logps_logqs/chosen": 0.06902097165584564,
|
| 1424 |
+
"logqs/chosen": -0.788707435131073,
|
| 1425 |
+
"logqs/rejected": -0.9549592733383179,
|
| 1426 |
+
"logqs_logps/rejected": 0.028454547747969627,
|
| 1427 |
+
"loss": 0.1686,
|
| 1428 |
+
"step": 1280
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 2.0704182967198315,
|
| 1432 |
+
"grad_norm": 0.40625,
|
| 1433 |
+
"learning_rate": 1.3248554454025275e-05,
|
| 1434 |
+
"logps_logqs/chosen": 0.07042928040027618,
|
| 1435 |
+
"logqs/chosen": -0.7786656618118286,
|
| 1436 |
+
"logqs/rejected": -0.9685086011886597,
|
| 1437 |
+
"logqs_logps/rejected": 0.025051862001419067,
|
| 1438 |
+
"loss": 0.1673,
|
| 1439 |
+
"step": 1290
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 2.08646805095797,
|
| 1443 |
+
"grad_norm": 0.447265625,
|
| 1444 |
+
"learning_rate": 1.2838487599951243e-05,
|
| 1445 |
+
"logps_logqs/chosen": 0.06794705986976624,
|
| 1446 |
+
"logqs/chosen": -0.7671617269515991,
|
| 1447 |
+
"logqs/rejected": -0.9481566548347473,
|
| 1448 |
+
"logqs_logps/rejected": 0.025887325406074524,
|
| 1449 |
+
"loss": 0.1696,
|
| 1450 |
+
"step": 1300
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 2.1025178051961078,
|
| 1454 |
+
"grad_norm": 0.439453125,
|
| 1455 |
+
"learning_rate": 1.2432663254403638e-05,
|
| 1456 |
+
"logps_logqs/chosen": 0.06914719194173813,
|
| 1457 |
+
"logqs/chosen": -0.8097491264343262,
|
| 1458 |
+
"logqs/rejected": -0.9609512090682983,
|
| 1459 |
+
"logqs_logps/rejected": 0.025332655757665634,
|
| 1460 |
+
"loss": 0.1689,
|
| 1461 |
+
"step": 1310
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 2.118567559434246,
|
| 1465 |
+
"grad_norm": 0.421875,
|
| 1466 |
+
"learning_rate": 1.2031222988034967e-05,
|
| 1467 |
+
"logps_logqs/chosen": 0.06850691139698029,
|
| 1468 |
+
"logqs/chosen": -0.7866081595420837,
|
| 1469 |
+
"logqs/rejected": -0.952921986579895,
|
| 1470 |
+
"logqs_logps/rejected": 0.02708747610449791,
|
| 1471 |
+
"loss": 0.1672,
|
| 1472 |
+
"step": 1320
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 2.1346173136723845,
|
| 1476 |
+
"grad_norm": 0.45703125,
|
| 1477 |
+
"learning_rate": 1.1634306842124423e-05,
|
| 1478 |
+
"logps_logqs/chosen": 0.06721793115139008,
|
| 1479 |
+
"logqs/chosen": -0.7912784218788147,
|
| 1480 |
+
"logqs/rejected": -0.9742089509963989,
|
| 1481 |
+
"logqs_logps/rejected": 0.024986112490296364,
|
| 1482 |
+
"loss": 0.1668,
|
| 1483 |
+
"step": 1330
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 2.1506670679105224,
|
| 1487 |
+
"grad_norm": 0.451171875,
|
| 1488 |
+
"learning_rate": 1.1242053279724762e-05,
|
| 1489 |
+
"logps_logqs/chosen": 0.06856787204742432,
|
| 1490 |
+
"logqs/chosen": -0.7681561708450317,
|
| 1491 |
+
"logqs/rejected": -0.9738146066665649,
|
| 1492 |
+
"logqs_logps/rejected": 0.025526920333504677,
|
| 1493 |
+
"loss": 0.1676,
|
| 1494 |
+
"step": 1340
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 2.166716822148661,
|
| 1498 |
+
"grad_norm": 0.44140625,
|
| 1499 |
+
"learning_rate": 1.0854599137359953e-05,
|
| 1500 |
+
"logps_logqs/chosen": 0.07144749909639359,
|
| 1501 |
+
"logqs/chosen": -0.7849553823471069,
|
| 1502 |
+
"logqs/rejected": -0.9945628046989441,
|
| 1503 |
+
"logqs_logps/rejected": 0.02589735947549343,
|
| 1504 |
+
"loss": 0.1683,
|
| 1505 |
+
"step": 1350
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 2.182766576386799,
|
| 1509 |
+
"grad_norm": 0.3984375,
|
| 1510 |
+
"learning_rate": 1.0472079577290111e-05,
|
| 1511 |
+
"logps_logqs/chosen": 0.06926636397838593,
|
| 1512 |
+
"logqs/chosen": -0.780665934085846,
|
| 1513 |
+
"logqs/rejected": -0.9741228222846985,
|
| 1514 |
+
"logqs_logps/rejected": 0.0255129374563694,
|
| 1515 |
+
"loss": 0.1668,
|
| 1516 |
+
"step": 1360
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 2.198816330624937,
|
| 1520 |
+
"grad_norm": 0.431640625,
|
| 1521 |
+
"learning_rate": 1.009462804036059e-05,
|
| 1522 |
+
"logps_logqs/chosen": 0.06916262209415436,
|
| 1523 |
+
"logqs/chosen": -0.7870660424232483,
|
| 1524 |
+
"logqs/rejected": -0.9848629236221313,
|
| 1525 |
+
"logqs_logps/rejected": 0.024935439229011536,
|
| 1526 |
+
"loss": 0.1688,
|
| 1527 |
+
"step": 1370
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 2.2148660848630755,
|
| 1531 |
+
"grad_norm": 0.48046875,
|
| 1532 |
+
"learning_rate": 9.722376199451436e-06,
|
| 1533 |
+
"logps_logqs/chosen": 0.06875228136777878,
|
| 1534 |
+
"logqs/chosen": -0.7785555124282837,
|
| 1535 |
+
"logqs/rejected": -0.9765084981918335,
|
| 1536 |
+
"logqs_logps/rejected": 0.02562164142727852,
|
| 1537 |
+
"loss": 0.1677,
|
| 1538 |
+
"step": 1380
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 2.230915839101214,
|
| 1542 |
+
"grad_norm": 0.40234375,
|
| 1543 |
+
"learning_rate": 9.35545391354378e-06,
|
| 1544 |
+
"logps_logqs/chosen": 0.06958961486816406,
|
| 1545 |
+
"logqs/chosen": -0.7699087262153625,
|
| 1546 |
+
"logqs/rejected": -0.9525764584541321,
|
| 1547 |
+
"logqs_logps/rejected": 0.027162248268723488,
|
| 1548 |
+
"loss": 0.1672,
|
| 1549 |
+
"step": 1390
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 2.246965593339352,
|
| 1553 |
+
"grad_norm": 0.416015625,
|
| 1554 |
+
"learning_rate": 8.993989182418824e-06,
|
| 1555 |
+
"logps_logqs/chosen": 0.0719502717256546,
|
| 1556 |
+
"logqs/chosen": -0.7669367790222168,
|
| 1557 |
+
"logqs/rejected": -0.9486163258552551,
|
| 1558 |
+
"logqs_logps/rejected": 0.02480030246078968,
|
| 1559 |
+
"loss": 0.1683,
|
| 1560 |
+
"step": 1400
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 2.26301534757749,
|
| 1564 |
+
"grad_norm": 0.48046875,
|
| 1565 |
+
"learning_rate": 8.63810810200556e-06,
|
| 1566 |
+
"logps_logqs/chosen": 0.06904618442058563,
|
| 1567 |
+
"logqs/chosen": -0.7622929811477661,
|
| 1568 |
+
"logqs/rejected": -0.9743655920028687,
|
| 1569 |
+
"logqs_logps/rejected": 0.026619747281074524,
|
| 1570 |
+
"loss": 0.1678,
|
| 1571 |
+
"step": 1410
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 2.2790651018156285,
|
| 1575 |
+
"grad_norm": 0.40234375,
|
| 1576 |
+
"learning_rate": 8.287934820392498e-06,
|
| 1577 |
+
"logps_logqs/chosen": 0.06923404335975647,
|
| 1578 |
+
"logqs/chosen": -0.7849544286727905,
|
| 1579 |
+
"logqs/rejected": -0.9630452990531921,
|
| 1580 |
+
"logqs_logps/rejected": 0.023480530828237534,
|
| 1581 |
+
"loss": 0.1651,
|
| 1582 |
+
"step": 1420
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"epoch": 2.295114856053767,
|
| 1586 |
+
"grad_norm": 0.4140625,
|
| 1587 |
+
"learning_rate": 7.943591494519016e-06,
|
| 1588 |
+
"logps_logqs/chosen": 0.06750839948654175,
|
| 1589 |
+
"logqs/chosen": -0.764995276927948,
|
| 1590 |
+
"logqs/rejected": -0.9358587265014648,
|
| 1591 |
+
"logqs_logps/rejected": 0.027208849787712097,
|
| 1592 |
+
"loss": 0.1667,
|
| 1593 |
+
"step": 1430
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 2.311164610291905,
|
| 1597 |
+
"grad_norm": 0.40625,
|
| 1598 |
+
"learning_rate": 7.60519824756124e-06,
|
| 1599 |
+
"logps_logqs/chosen": 0.06875176727771759,
|
| 1600 |
+
"logqs/chosen": -0.8004514575004578,
|
| 1601 |
+
"logqs/rejected": -1.0356009006500244,
|
| 1602 |
+
"logqs_logps/rejected": 0.0247772429138422,
|
| 1603 |
+
"loss": 0.1681,
|
| 1604 |
+
"step": 1440
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 2.327214364530043,
|
| 1608 |
+
"grad_norm": 0.41015625,
|
| 1609 |
+
"learning_rate": 7.272873127027449e-06,
|
| 1610 |
+
"logps_logqs/chosen": 0.06980612874031067,
|
| 1611 |
+
"logqs/chosen": -0.7612735033035278,
|
| 1612 |
+
"logqs/rejected": -0.9591197967529297,
|
| 1613 |
+
"logqs_logps/rejected": 0.0231007132679224,
|
| 1614 |
+
"loss": 0.1652,
|
| 1615 |
+
"step": 1450
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 2.3432641187681815,
|
| 1619 |
+
"grad_norm": 0.41796875,
|
| 1620 |
+
"learning_rate": 6.946732063577488e-06,
|
| 1621 |
+
"logps_logqs/chosen": 0.0690704956650734,
|
| 1622 |
+
"logqs/chosen": -0.7698851823806763,
|
| 1623 |
+
"logqs/rejected": -0.949694037437439,
|
| 1624 |
+
"logqs_logps/rejected": 0.023757826536893845,
|
| 1625 |
+
"loss": 0.1655,
|
| 1626 |
+
"step": 1460
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 2.3593138730063195,
|
| 1630 |
+
"grad_norm": 0.4375,
|
| 1631 |
+
"learning_rate": 6.6268888305807296e-06,
|
| 1632 |
+
"logps_logqs/chosen": 0.06597896665334702,
|
| 1633 |
+
"logqs/chosen": -0.7630634903907776,
|
| 1634 |
+
"logqs/rejected": -0.9974457025527954,
|
| 1635 |
+
"logqs_logps/rejected": 0.02252907119691372,
|
| 1636 |
+
"loss": 0.1648,
|
| 1637 |
+
"step": 1470
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 2.375363627244458,
|
| 1641 |
+
"grad_norm": 0.404296875,
|
| 1642 |
+
"learning_rate": 6.313455004426577e-06,
|
| 1643 |
+
"logps_logqs/chosen": 0.06898193061351776,
|
| 1644 |
+
"logqs/chosen": -0.77290940284729,
|
| 1645 |
+
"logqs/rejected": -0.9675156474113464,
|
| 1646 |
+
"logqs_logps/rejected": 0.024294773116707802,
|
| 1647 |
+
"loss": 0.166,
|
| 1648 |
+
"step": 1480
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 2.391413381482596,
|
| 1652 |
+
"grad_norm": 0.443359375,
|
| 1653 |
+
"learning_rate": 6.0065399256013204e-06,
|
| 1654 |
+
"logps_logqs/chosen": 0.0698733851313591,
|
| 1655 |
+
"logqs/chosen": -0.761997401714325,
|
| 1656 |
+
"logqs/rejected": -0.9315551519393921,
|
| 1657 |
+
"logqs_logps/rejected": 0.02528567612171173,
|
| 1658 |
+
"loss": 0.1696,
|
| 1659 |
+
"step": 1490
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 2.407463135720734,
|
| 1663 |
+
"grad_norm": 0.423828125,
|
| 1664 |
+
"learning_rate": 5.7062506605450454e-06,
|
| 1665 |
+
"logps_logqs/chosen": 0.06849467009305954,
|
| 1666 |
+
"logqs/chosen": -0.7781127095222473,
|
| 1667 |
+
"logqs/rejected": -0.9850066304206848,
|
| 1668 |
+
"logqs_logps/rejected": 0.026143735274672508,
|
| 1669 |
+
"loss": 0.1703,
|
| 1670 |
+
"step": 1500
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 2.4235128899588725,
|
| 1674 |
+
"grad_norm": 0.443359375,
|
| 1675 |
+
"learning_rate": 5.412691964301827e-06,
|
| 1676 |
+
"logps_logqs/chosen": 0.06971971690654755,
|
| 1677 |
+
"logqs/chosen": -0.7925983667373657,
|
| 1678 |
+
"logqs/rejected": -0.9926835894584656,
|
| 1679 |
+
"logqs_logps/rejected": 0.025326719507575035,
|
| 1680 |
+
"loss": 0.1685,
|
| 1681 |
+
"step": 1510
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 2.439562644197011,
|
| 1685 |
+
"grad_norm": 0.470703125,
|
| 1686 |
+
"learning_rate": 5.125966243976218e-06,
|
| 1687 |
+
"logps_logqs/chosen": 0.07100898772478104,
|
| 1688 |
+
"logqs/chosen": -0.7900832891464233,
|
| 1689 |
+
"logqs/rejected": -0.9629155993461609,
|
| 1690 |
+
"logqs_logps/rejected": 0.025052938610315323,
|
| 1691 |
+
"loss": 0.1693,
|
| 1692 |
+
"step": 1520
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 2.455612398435149,
|
| 1696 |
+
"grad_norm": 0.453125,
|
| 1697 |
+
"learning_rate": 4.846173523008824e-06,
|
| 1698 |
+
"logps_logqs/chosen": 0.06859283149242401,
|
| 1699 |
+
"logqs/chosen": -0.7894052267074585,
|
| 1700 |
+
"logqs/rejected": -0.9897419810295105,
|
| 1701 |
+
"logqs_logps/rejected": 0.02611861191689968,
|
| 1702 |
+
"loss": 0.1683,
|
| 1703 |
+
"step": 1530
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 2.471662152673287,
|
| 1707 |
+
"grad_norm": 0.439453125,
|
| 1708 |
+
"learning_rate": 4.573411406283409e-06,
|
| 1709 |
+
"logps_logqs/chosen": 0.07065166532993317,
|
| 1710 |
+
"logqs/chosen": -0.8073331117630005,
|
| 1711 |
+
"logqs/rejected": -0.9767266511917114,
|
| 1712 |
+
"logqs_logps/rejected": 0.0274626724421978,
|
| 1713 |
+
"loss": 0.1756,
|
| 1714 |
+
"step": 1540
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 2.4877119069114255,
|
| 1718 |
+
"grad_norm": 0.44140625,
|
| 1719 |
+
"learning_rate": 4.307775046077739e-06,
|
| 1720 |
+
"logps_logqs/chosen": 0.0715617686510086,
|
| 1721 |
+
"logqs/chosen": -0.8018879890441895,
|
| 1722 |
+
"logqs/rejected": -0.9990041851997375,
|
| 1723 |
+
"logqs_logps/rejected": 0.026406193152070045,
|
| 1724 |
+
"loss": 0.1719,
|
| 1725 |
+
"step": 1550
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 2.5037616611495634,
|
| 1729 |
+
"grad_norm": 0.46875,
|
| 1730 |
+
"learning_rate": 4.049357108869964e-06,
|
| 1731 |
+
"logps_logqs/chosen": 0.06756819784641266,
|
| 1732 |
+
"logqs/chosen": -0.7610915899276733,
|
| 1733 |
+
"logqs/rejected": -0.972091794013977,
|
| 1734 |
+
"logqs_logps/rejected": 0.026423901319503784,
|
| 1735 |
+
"loss": 0.1683,
|
| 1736 |
+
"step": 1560
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 2.519811415387702,
|
| 1740 |
+
"grad_norm": 0.439453125,
|
| 1741 |
+
"learning_rate": 3.798247743012201e-06,
|
| 1742 |
+
"logps_logqs/chosen": 0.06909768283367157,
|
| 1743 |
+
"logqs/chosen": -0.7960779666900635,
|
| 1744 |
+
"logqs/rejected": -0.9779514074325562,
|
| 1745 |
+
"logqs_logps/rejected": 0.0239783376455307,
|
| 1746 |
+
"loss": 0.1662,
|
| 1747 |
+
"step": 1570
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 2.53586116962584,
|
| 1751 |
+
"grad_norm": 0.43359375,
|
| 1752 |
+
"learning_rate": 3.554534547282512e-06,
|
| 1753 |
+
"logps_logqs/chosen": 0.07034210860729218,
|
| 1754 |
+
"logqs/chosen": -0.7767470479011536,
|
| 1755 |
+
"logqs/rejected": -0.9516876339912415,
|
| 1756 |
+
"logqs_logps/rejected": 0.028661763295531273,
|
| 1757 |
+
"loss": 0.17,
|
| 1758 |
+
"step": 1580
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 2.5519109238639786,
|
| 1762 |
+
"grad_norm": 0.404296875,
|
| 1763 |
+
"learning_rate": 3.318302540326343e-06,
|
| 1764 |
+
"logps_logqs/chosen": 0.07079549133777618,
|
| 1765 |
+
"logqs/chosen": -0.777116596698761,
|
| 1766 |
+
"logqs/rejected": -0.9669076204299927,
|
| 1767 |
+
"logqs_logps/rejected": 0.026164010167121887,
|
| 1768 |
+
"loss": 0.1723,
|
| 1769 |
+
"step": 1590
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 2.5679606781021165,
|
| 1773 |
+
"grad_norm": 0.431640625,
|
| 1774 |
+
"learning_rate": 3.089634130998026e-06,
|
| 1775 |
+
"logps_logqs/chosen": 0.06778942793607712,
|
| 1776 |
+
"logqs/chosen": -0.7698632478713989,
|
| 1777 |
+
"logqs/rejected": -0.9828590154647827,
|
| 1778 |
+
"logqs_logps/rejected": 0.02352612093091011,
|
| 1779 |
+
"loss": 0.166,
|
| 1780 |
+
"step": 1600
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 2.584010432340255,
|
| 1784 |
+
"grad_norm": 0.4296875,
|
| 1785 |
+
"learning_rate": 2.8686090896126587e-06,
|
| 1786 |
+
"logps_logqs/chosen": 0.06593993306159973,
|
| 1787 |
+
"logqs/chosen": -0.7838481664657593,
|
| 1788 |
+
"logqs/rejected": -0.9900785684585571,
|
| 1789 |
+
"logqs_logps/rejected": 0.023241404443979263,
|
| 1790 |
+
"loss": 0.1633,
|
| 1791 |
+
"step": 1610
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 2.6000601865783928,
|
| 1795 |
+
"grad_norm": 0.439453125,
|
| 1796 |
+
"learning_rate": 2.655304520118482e-06,
|
| 1797 |
+
"logps_logqs/chosen": 0.0693887323141098,
|
| 1798 |
+
"logqs/chosen": -0.778706431388855,
|
| 1799 |
+
"logqs/rejected": -0.9755508303642273,
|
| 1800 |
+
"logqs_logps/rejected": 0.02637295424938202,
|
| 1801 |
+
"loss": 0.1664,
|
| 1802 |
+
"step": 1620
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.616109940816531,
|
| 1806 |
+
"grad_norm": 0.44140625,
|
| 1807 |
+
"learning_rate": 2.44979483319939e-06,
|
| 1808 |
+
"logps_logqs/chosen": 0.06880888342857361,
|
| 1809 |
+
"logqs/chosen": -0.7670097351074219,
|
| 1810 |
+
"logqs/rejected": -1.0248219966888428,
|
| 1811 |
+
"logqs_logps/rejected": 0.025341719388961792,
|
| 1812 |
+
"loss": 0.1667,
|
| 1813 |
+
"step": 1630
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 2.6321596950546695,
|
| 1817 |
+
"grad_norm": 0.44140625,
|
| 1818 |
+
"learning_rate": 2.252151720316964e-06,
|
| 1819 |
+
"logps_logqs/chosen": 0.0697592943906784,
|
| 1820 |
+
"logqs/chosen": -0.7662806510925293,
|
| 1821 |
+
"logqs/rejected": -0.9526630640029907,
|
| 1822 |
+
"logqs_logps/rejected": 0.024268481880426407,
|
| 1823 |
+
"loss": 0.1672,
|
| 1824 |
+
"step": 1640
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 2.648209449292808,
|
| 1828 |
+
"grad_norm": 0.458984375,
|
| 1829 |
+
"learning_rate": 2.0624441287011213e-06,
|
| 1830 |
+
"logps_logqs/chosen": 0.06959348171949387,
|
| 1831 |
+
"logqs/chosen": -0.7930929064750671,
|
| 1832 |
+
"logqs/rejected": -0.9900212287902832,
|
| 1833 |
+
"logqs_logps/rejected": 0.02552485466003418,
|
| 1834 |
+
"loss": 0.1705,
|
| 1835 |
+
"step": 1650
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 2.664259203530946,
|
| 1839 |
+
"grad_norm": 0.419921875,
|
| 1840 |
+
"learning_rate": 1.8807382372980687e-06,
|
| 1841 |
+
"logps_logqs/chosen": 0.06878109276294708,
|
| 1842 |
+
"logqs/chosen": -0.776373565196991,
|
| 1843 |
+
"logqs/rejected": -0.9679840207099915,
|
| 1844 |
+
"logqs_logps/rejected": 0.024887990206480026,
|
| 1845 |
+
"loss": 0.1656,
|
| 1846 |
+
"step": 1660
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 2.680308957769084,
|
| 1850 |
+
"grad_norm": 0.431640625,
|
| 1851 |
+
"learning_rate": 1.7070974336839796e-06,
|
| 1852 |
+
"logps_logqs/chosen": 0.06730766594409943,
|
| 1853 |
+
"logqs/chosen": -0.7652665972709656,
|
| 1854 |
+
"logqs/rejected": -0.9662426114082336,
|
| 1855 |
+
"logqs_logps/rejected": 0.026059061288833618,
|
| 1856 |
+
"loss": 0.1667,
|
| 1857 |
+
"step": 1670
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 2.6963587120072225,
|
| 1861 |
+
"grad_norm": 0.453125,
|
| 1862 |
+
"learning_rate": 1.541582291952401e-06,
|
| 1863 |
+
"logps_logqs/chosen": 0.06985867768526077,
|
| 1864 |
+
"logqs/chosen": -0.7848519086837769,
|
| 1865 |
+
"logqs/rejected": -0.9851476550102234,
|
| 1866 |
+
"logqs_logps/rejected": 0.02311038039624691,
|
| 1867 |
+
"loss": 0.1688,
|
| 1868 |
+
"step": 1680
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 2.7124084662453605,
|
| 1872 |
+
"grad_norm": 0.4375,
|
| 1873 |
+
"learning_rate": 1.38425055158318e-06,
|
| 1874 |
+
"logps_logqs/chosen": 0.0703495591878891,
|
| 1875 |
+
"logqs/chosen": -0.7814745903015137,
|
| 1876 |
+
"logqs/rejected": -0.9425519108772278,
|
| 1877 |
+
"logqs_logps/rejected": 0.026310011744499207,
|
| 1878 |
+
"loss": 0.1639,
|
| 1879 |
+
"step": 1690
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.728458220483499,
|
| 1883 |
+
"grad_norm": 0.44921875,
|
| 1884 |
+
"learning_rate": 1.235157097300188e-06,
|
| 1885 |
+
"logps_logqs/chosen": 0.06932912766933441,
|
| 1886 |
+
"logqs/chosen": -0.7783406972885132,
|
| 1887 |
+
"logqs/rejected": -0.9719565510749817,
|
| 1888 |
+
"logqs_logps/rejected": 0.025869470089673996,
|
| 1889 |
+
"loss": 0.1671,
|
| 1890 |
+
"step": 1700
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"epoch": 2.744507974721637,
|
| 1894 |
+
"grad_norm": 0.40625,
|
| 1895 |
+
"learning_rate": 1.0943539399249635e-06,
|
| 1896 |
+
"logps_logqs/chosen": 0.07044648379087448,
|
| 1897 |
+
"logqs/chosen": -0.776070237159729,
|
| 1898 |
+
"logqs/rejected": -0.9304519891738892,
|
| 1899 |
+
"logqs_logps/rejected": 0.02800569497048855,
|
| 1900 |
+
"loss": 0.169,
|
| 1901 |
+
"step": 1710
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 2.7605577289597756,
|
| 1905 |
+
"grad_norm": 0.4609375,
|
| 1906 |
+
"learning_rate": 9.618901982328704e-07,
|
| 1907 |
+
"logps_logqs/chosen": 0.06949154287576675,
|
| 1908 |
+
"logqs/chosen": -0.7787348031997681,
|
| 1909 |
+
"logqs/rejected": -0.9538747668266296,
|
| 1910 |
+
"logqs_logps/rejected": 0.027285417541861534,
|
| 1911 |
+
"loss": 0.169,
|
| 1912 |
+
"step": 1720
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 2.7766074831979135,
|
| 1916 |
+
"grad_norm": 0.427734375,
|
| 1917 |
+
"learning_rate": 8.378120818181707e-07,
|
| 1918 |
+
"logps_logqs/chosen": 0.07005327194929123,
|
| 1919 |
+
"logqs/chosen": -0.7800394296646118,
|
| 1920 |
+
"logqs/rejected": -0.9388517141342163,
|
| 1921 |
+
"logqs_logps/rejected": 0.024246862158179283,
|
| 1922 |
+
"loss": 0.1638,
|
| 1923 |
+
"step": 1730
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 2.792657237436052,
|
| 1927 |
+
"grad_norm": 0.39453125,
|
| 1928 |
+
"learning_rate": 7.221628749739223e-07,
|
| 1929 |
+
"logps_logqs/chosen": 0.07048022001981735,
|
| 1930 |
+
"logqs/chosen": -0.781145453453064,
|
| 1931 |
+
"logqs/rejected": -0.9654680490493774,
|
| 1932 |
+
"logqs_logps/rejected": 0.025187021121382713,
|
| 1933 |
+
"loss": 0.1646,
|
| 1934 |
+
"step": 1740
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 2.80870699167419,
|
| 1938 |
+
"grad_norm": 0.41796875,
|
| 1939 |
+
"learning_rate": 6.149829215924025e-07,
|
| 1940 |
+
"logps_logqs/chosen": 0.06936784833669662,
|
| 1941 |
+
"logqs/chosen": -0.7857106328010559,
|
| 1942 |
+
"logqs/rejected": -1.0071794986724854,
|
| 1943 |
+
"logqs_logps/rejected": 0.02525334618985653,
|
| 1944 |
+
"loss": 0.1685,
|
| 1945 |
+
"step": 1750
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 2.824756745912328,
|
| 1949 |
+
"grad_norm": 0.396484375,
|
| 1950 |
+
"learning_rate": 5.163096110912368e-07,
|
| 1951 |
+
"logps_logqs/chosen": 0.0710034891963005,
|
| 1952 |
+
"logqs/chosen": -0.7809524536132812,
|
| 1953 |
+
"logqs/rejected": -0.9730299711227417,
|
| 1954 |
+
"logqs_logps/rejected": 0.02650505304336548,
|
| 1955 |
+
"loss": 0.1685,
|
| 1956 |
+
"step": 1760
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.8408065001504665,
|
| 1960 |
+
"grad_norm": 0.419921875,
|
| 1961 |
+
"learning_rate": 4.261773653702089e-07,
|
| 1962 |
+
"logps_logqs/chosen": 0.07018028199672699,
|
| 1963 |
+
"logqs/chosen": -0.7636402249336243,
|
| 1964 |
+
"logqs/rejected": -0.9362967610359192,
|
| 1965 |
+
"logqs_logps/rejected": 0.02620730921626091,
|
| 1966 |
+
"loss": 0.1674,
|
| 1967 |
+
"step": 1770
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"epoch": 2.856856254388605,
|
| 1971 |
+
"grad_norm": 0.5859375,
|
| 1972 |
+
"learning_rate": 3.4461762680329803e-07,
|
| 1973 |
+
"logps_logqs/chosen": 0.06559871137142181,
|
| 1974 |
+
"logqs/chosen": -0.7775349617004395,
|
| 1975 |
+
"logqs/rejected": -0.9631088376045227,
|
| 1976 |
+
"logqs_logps/rejected": 0.024396821856498718,
|
| 1977 |
+
"loss": 0.1651,
|
| 1978 |
+
"step": 1780
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 2.872906008626743,
|
| 1982 |
+
"grad_norm": 0.419921875,
|
| 1983 |
+
"learning_rate": 2.716588472700815e-07,
|
| 1984 |
+
"logps_logqs/chosen": 0.07039310038089752,
|
| 1985 |
+
"logqs/chosen": -0.7576441168785095,
|
| 1986 |
+
"logqs/rejected": -0.9705005884170532,
|
| 1987 |
+
"logqs_logps/rejected": 0.026445040479302406,
|
| 1988 |
+
"loss": 0.1646,
|
| 1989 |
+
"step": 1790
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 2.888955762864881,
|
| 1993 |
+
"grad_norm": 0.39453125,
|
| 1994 |
+
"learning_rate": 2.0732647823038243e-07,
|
| 1995 |
+
"logps_logqs/chosen": 0.06968151032924652,
|
| 1996 |
+
"logqs/chosen": -0.7765523791313171,
|
| 1997 |
+
"logqs/rejected": -0.9748941659927368,
|
| 1998 |
+
"logqs_logps/rejected": 0.02464660070836544,
|
| 1999 |
+
"loss": 0.1665,
|
| 2000 |
+
"step": 1800
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 2.905005517103019,
|
| 2004 |
+
"grad_norm": 0.458984375,
|
| 2005 |
+
"learning_rate": 1.5164296184560222e-07,
|
| 2006 |
+
"logps_logqs/chosen": 0.06898736953735352,
|
| 2007 |
+
"logqs/chosen": -0.781166672706604,
|
| 2008 |
+
"logqs/rejected": -0.9676758646965027,
|
| 2009 |
+
"logqs_logps/rejected": 0.02475111000239849,
|
| 2010 |
+
"loss": 0.1665,
|
| 2011 |
+
"step": 1810
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 2.9210552713411575,
|
| 2015 |
+
"grad_norm": 0.404296875,
|
| 2016 |
+
"learning_rate": 1.0462772314983882e-07,
|
| 2017 |
+
"logps_logqs/chosen": 0.06931523233652115,
|
| 2018 |
+
"logqs/chosen": -0.7759555578231812,
|
| 2019 |
+
"logqs/rejected": -0.9690700769424438,
|
| 2020 |
+
"logqs_logps/rejected": 0.024959508329629898,
|
| 2021 |
+
"loss": 0.1706,
|
| 2022 |
+
"step": 1820
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 2.937105025579296,
|
| 2026 |
+
"grad_norm": 0.4609375,
|
| 2027 |
+
"learning_rate": 6.62971632735182e-08,
|
| 2028 |
+
"logps_logqs/chosen": 0.07144445180892944,
|
| 2029 |
+
"logqs/chosen": -0.7924820780754089,
|
| 2030 |
+
"logqs/rejected": -0.9925470352172852,
|
| 2031 |
+
"logqs_logps/rejected": 0.0241483636200428,
|
| 2032 |
+
"loss": 0.1691,
|
| 2033 |
+
"step": 1830
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.9531547798174342,
|
| 2037 |
+
"grad_norm": 0.435546875,
|
| 2038 |
+
"learning_rate": 3.666465372190453e-08,
|
| 2039 |
+
"logps_logqs/chosen": 0.07032831013202667,
|
| 2040 |
+
"logqs/chosen": -0.7886664271354675,
|
| 2041 |
+
"logqs/rejected": -1.0002458095550537,
|
| 2042 |
+
"logqs_logps/rejected": 0.024760346859693527,
|
| 2043 |
+
"loss": 0.1681,
|
| 2044 |
+
"step": 1840
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 2.969204534055572,
|
| 2048 |
+
"grad_norm": 0.431640625,
|
| 2049 |
+
"learning_rate": 1.57405317104925e-08,
|
| 2050 |
+
"logps_logqs/chosen": 0.068773552775383,
|
| 2051 |
+
"logqs/chosen": -0.7885466814041138,
|
| 2052 |
+
"logqs/rejected": -0.9979216456413269,
|
| 2053 |
+
"logqs_logps/rejected": 0.023049544543027878,
|
| 2054 |
+
"loss": 0.1691,
|
| 2055 |
+
"step": 1850
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 2.9852542882937105,
|
| 2059 |
+
"grad_norm": 0.427734375,
|
| 2060 |
+
"learning_rate": 3.532096558903075e-09,
|
| 2061 |
+
"logps_logqs/chosen": 0.06932573765516281,
|
| 2062 |
+
"logqs/chosen": -0.7747775912284851,
|
| 2063 |
+
"logqs/rejected": -0.9474186897277832,
|
| 2064 |
+
"logqs_logps/rejected": 0.026311378926038742,
|
| 2065 |
+
"loss": 0.1678,
|
| 2066 |
+
"step": 1860
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 2.999699067108035,
|
| 2070 |
+
"step": 1869,
|
| 2071 |
+
"total_flos": 0.0,
|
| 2072 |
+
"train_loss": 0.194109176269121,
|
| 2073 |
+
"train_runtime": 23250.2173,
|
| 2074 |
+
"train_samples_per_second": 10.29,
|
| 2075 |
+
"train_steps_per_second": 0.08
|
| 2076 |
+
}
|
| 2077 |
+
],
|
| 2078 |
+
"logging_steps": 10,
|
| 2079 |
+
"max_steps": 1869,
|
| 2080 |
+
"num_input_tokens_seen": 0,
|
| 2081 |
+
"num_train_epochs": 3,
|
| 2082 |
+
"save_steps": 500,
|
| 2083 |
+
"stateful_callbacks": {
|
| 2084 |
+
"TrainerControl": {
|
| 2085 |
+
"args": {
|
| 2086 |
+
"should_epoch_stop": false,
|
| 2087 |
+
"should_evaluate": false,
|
| 2088 |
+
"should_log": false,
|
| 2089 |
+
"should_save": true,
|
| 2090 |
+
"should_training_stop": true
|
| 2091 |
+
},
|
| 2092 |
+
"attributes": {}
|
| 2093 |
+
}
|
| 2094 |
+
},
|
| 2095 |
+
"total_flos": 0.0,
|
| 2096 |
+
"train_batch_size": 2,
|
| 2097 |
+
"trial_name": null,
|
| 2098 |
+
"trial_params": null
|
| 2099 |
+
}
|