Upload folder using huggingface_hub
Browse files- checkpoint-19303/config.json +35 -0
- checkpoint-19303/generation_config.json +9 -0
- checkpoint-19303/model.safetensors +3 -0
- checkpoint-19303/optimizer.pt +3 -0
- checkpoint-19303/rng_state.pth +3 -0
- checkpoint-19303/scaler.pt +3 -0
- checkpoint-19303/scheduler.pt +3 -0
- checkpoint-19303/trainer_state.json +2736 -0
- checkpoint-19303/training_args.bin +3 -0
- checkpoint-41387/config.json +35 -0
- checkpoint-41387/generation_config.json +9 -0
- checkpoint-41387/model.safetensors +3 -0
- checkpoint-41387/optimizer.pt +3 -0
- checkpoint-41387/rng_state.pth +3 -0
- checkpoint-41387/scaler.pt +3 -0
- checkpoint-41387/scheduler.pt +3 -0
- checkpoint-41387/trainer_state.json +0 -0
- checkpoint-41387/training_args.bin +3 -0
- config.json +35 -0
- generation_config.json +9 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +13 -0
- training_args.bin +3 -0
checkpoint-19303/config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"add_cross_attention": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"embd_pdrop": 0.1,
|
| 11 |
+
"eos_token_id": 50256,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"layer_norm_epsilon": 1e-05,
|
| 14 |
+
"model_type": "gpt2",
|
| 15 |
+
"n_ctx": 256,
|
| 16 |
+
"n_embd": 640,
|
| 17 |
+
"n_head": 10,
|
| 18 |
+
"n_inner": null,
|
| 19 |
+
"n_layer": 12,
|
| 20 |
+
"n_positions": 256,
|
| 21 |
+
"pad_token_id": 50256,
|
| 22 |
+
"reorder_and_upcast_attn": false,
|
| 23 |
+
"resid_pdrop": 0.1,
|
| 24 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 25 |
+
"scale_attn_weights": true,
|
| 26 |
+
"summary_activation": null,
|
| 27 |
+
"summary_first_dropout": 0.1,
|
| 28 |
+
"summary_proj_to_labels": true,
|
| 29 |
+
"summary_type": "cls_index",
|
| 30 |
+
"summary_use_proj": true,
|
| 31 |
+
"tie_word_embeddings": true,
|
| 32 |
+
"transformers_version": "5.12.1",
|
| 33 |
+
"use_cache": false,
|
| 34 |
+
"vocab_size": 50257
|
| 35 |
+
}
|
checkpoint-19303/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"output_attentions": false,
|
| 6 |
+
"output_hidden_states": false,
|
| 7 |
+
"transformers_version": "5.12.1",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|
checkpoint-19303/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3aeb0ecaafe08533200daff41b55fbff8818bb06cb215712d61388d1c59dee9
|
| 3 |
+
size 365662304
|
checkpoint-19303/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff8986558d53ea2107a5881f18093414071be8a94e1f3652faa78ac592b71a79
|
| 3 |
+
size 731421067
|
checkpoint-19303/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9c37710e18dea87cb9e9715d01568fd86fd880df772ae5090b293b3b1d575ff
|
| 3 |
+
size 14645
|
checkpoint-19303/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:50f1261edb18a387451c4394b4d8b434f60bcb2b96618e07c03624c9eccfe8f9
|
| 3 |
+
size 1383
|
checkpoint-19303/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bced912761116e5ca48d074542e3dd2beca50deb5fe07376c4b361b34c8622f2
|
| 3 |
+
size 1465
|
checkpoint-19303/trainer_state.json
ADDED
|
@@ -0,0 +1,2736 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 19303,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.002590270942340569,
|
| 14 |
+
"grad_norm": 2.053372621536255,
|
| 15 |
+
"learning_rate": 2.94e-05,
|
| 16 |
+
"loss": 10.122694702148438,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.005180541884681138,
|
| 21 |
+
"grad_norm": 1.547727346420288,
|
| 22 |
+
"learning_rate": 5.94e-05,
|
| 23 |
+
"loss": 8.92919921875,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0077708128270217065,
|
| 28 |
+
"grad_norm": 1.2461143732070923,
|
| 29 |
+
"learning_rate": 8.939999999999999e-05,
|
| 30 |
+
"loss": 7.86175048828125,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.010361083769362275,
|
| 35 |
+
"grad_norm": 1.2778165340423584,
|
| 36 |
+
"learning_rate": 0.0001194,
|
| 37 |
+
"loss": 7.493834228515625,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.012951354711702844,
|
| 42 |
+
"grad_norm": 1.2245980501174927,
|
| 43 |
+
"learning_rate": 0.0001494,
|
| 44 |
+
"loss": 7.367841796875,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.015541625654043413,
|
| 49 |
+
"grad_norm": 1.2784472703933716,
|
| 50 |
+
"learning_rate": 0.00017939999999999997,
|
| 51 |
+
"loss": 7.193695068359375,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.018131896596383982,
|
| 56 |
+
"grad_norm": 1.0248782634735107,
|
| 57 |
+
"learning_rate": 0.00020939999999999997,
|
| 58 |
+
"loss": 7.10666748046875,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.02072216753872455,
|
| 63 |
+
"grad_norm": 1.1631972789764404,
|
| 64 |
+
"learning_rate": 0.0002394,
|
| 65 |
+
"loss": 6.988087158203125,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.02331243848106512,
|
| 70 |
+
"grad_norm": 0.9435396790504456,
|
| 71 |
+
"learning_rate": 0.0002694,
|
| 72 |
+
"loss": 6.923257446289062,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.02590270942340569,
|
| 77 |
+
"grad_norm": 1.0248340368270874,
|
| 78 |
+
"learning_rate": 0.00029939999999999996,
|
| 79 |
+
"loss": 6.855230712890625,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.028492980365746257,
|
| 84 |
+
"grad_norm": 1.0538331270217896,
|
| 85 |
+
"learning_rate": 0.0002992182098601287,
|
| 86 |
+
"loss": 6.823525390625,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.031083251308086826,
|
| 91 |
+
"grad_norm": 1.077528715133667,
|
| 92 |
+
"learning_rate": 0.0002984204648194437,
|
| 93 |
+
"loss": 6.768828125,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.03367352225042739,
|
| 98 |
+
"grad_norm": 0.9541271328926086,
|
| 99 |
+
"learning_rate": 0.00029762271977875867,
|
| 100 |
+
"loss": 6.701234130859375,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.036263793192767964,
|
| 105 |
+
"grad_norm": 1.0689588785171509,
|
| 106 |
+
"learning_rate": 0.0002968249747380737,
|
| 107 |
+
"loss": 6.623989868164062,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.03885406413510853,
|
| 112 |
+
"grad_norm": 0.9765296578407288,
|
| 113 |
+
"learning_rate": 0.0002960272296973887,
|
| 114 |
+
"loss": 6.646817016601562,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.0414443350774491,
|
| 119 |
+
"grad_norm": 0.985795259475708,
|
| 120 |
+
"learning_rate": 0.0002952294846567037,
|
| 121 |
+
"loss": 6.574093627929687,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.04403460601978967,
|
| 126 |
+
"grad_norm": 0.9846292734146118,
|
| 127 |
+
"learning_rate": 0.0002944317396160187,
|
| 128 |
+
"loss": 6.57322998046875,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.04662487696213024,
|
| 133 |
+
"grad_norm": 0.9685710072517395,
|
| 134 |
+
"learning_rate": 0.00029363399457533367,
|
| 135 |
+
"loss": 6.578934326171875,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.049215147904470805,
|
| 140 |
+
"grad_norm": 0.9839648008346558,
|
| 141 |
+
"learning_rate": 0.00029283624953464874,
|
| 142 |
+
"loss": 6.5168524169921875,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.05180541884681138,
|
| 147 |
+
"grad_norm": 0.9077258706092834,
|
| 148 |
+
"learning_rate": 0.0002920385044939637,
|
| 149 |
+
"loss": 6.452473754882813,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.05439568978915194,
|
| 154 |
+
"grad_norm": 0.9350703358650208,
|
| 155 |
+
"learning_rate": 0.0002912407594532787,
|
| 156 |
+
"loss": 6.431409912109375,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.056985960731492515,
|
| 161 |
+
"grad_norm": 0.9615451097488403,
|
| 162 |
+
"learning_rate": 0.0002904430144125937,
|
| 163 |
+
"loss": 6.449520874023437,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.05957623167383308,
|
| 168 |
+
"grad_norm": 1.1748050451278687,
|
| 169 |
+
"learning_rate": 0.0002896452693719087,
|
| 170 |
+
"loss": 6.4049951171875,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.06216650261617365,
|
| 175 |
+
"grad_norm": 0.9100262522697449,
|
| 176 |
+
"learning_rate": 0.00028884752433122374,
|
| 177 |
+
"loss": 6.3247802734375,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.06475677355851422,
|
| 182 |
+
"grad_norm": 0.9959622025489807,
|
| 183 |
+
"learning_rate": 0.0002880497792905387,
|
| 184 |
+
"loss": 6.380640258789063,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.06734704450085478,
|
| 189 |
+
"grad_norm": 0.9422142505645752,
|
| 190 |
+
"learning_rate": 0.0002872520342498537,
|
| 191 |
+
"loss": 6.371143188476562,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.06993731544319536,
|
| 196 |
+
"grad_norm": 1.0342239141464233,
|
| 197 |
+
"learning_rate": 0.0002864542892091687,
|
| 198 |
+
"loss": 6.372911376953125,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.07252758638553593,
|
| 203 |
+
"grad_norm": 0.9143936634063721,
|
| 204 |
+
"learning_rate": 0.00028565654416848373,
|
| 205 |
+
"loss": 6.302949829101562,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.0751178573278765,
|
| 210 |
+
"grad_norm": 0.9583525657653809,
|
| 211 |
+
"learning_rate": 0.00028485879912779874,
|
| 212 |
+
"loss": 6.344402465820313,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.07770812827021706,
|
| 217 |
+
"grad_norm": 0.9014634490013123,
|
| 218 |
+
"learning_rate": 0.00028406105408711375,
|
| 219 |
+
"loss": 6.285582275390625,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.08029839921255763,
|
| 224 |
+
"grad_norm": 0.94912189245224,
|
| 225 |
+
"learning_rate": 0.0002832633090464287,
|
| 226 |
+
"loss": 6.273238525390625,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.0828886701548982,
|
| 231 |
+
"grad_norm": 0.9462608098983765,
|
| 232 |
+
"learning_rate": 0.0002824655640057438,
|
| 233 |
+
"loss": 6.25449951171875,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.08547894109723878,
|
| 238 |
+
"grad_norm": 0.9752557277679443,
|
| 239 |
+
"learning_rate": 0.00028166781896505873,
|
| 240 |
+
"loss": 6.239361572265625,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.08806921203957933,
|
| 245 |
+
"grad_norm": 1.0141621828079224,
|
| 246 |
+
"learning_rate": 0.00028087007392437374,
|
| 247 |
+
"loss": 6.25331787109375,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0906594829819199,
|
| 252 |
+
"grad_norm": 0.9530112147331238,
|
| 253 |
+
"learning_rate": 0.00028007232888368875,
|
| 254 |
+
"loss": 6.2260675048828125,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.09324975392426048,
|
| 259 |
+
"grad_norm": 1.020228624343872,
|
| 260 |
+
"learning_rate": 0.00027927458384300377,
|
| 261 |
+
"loss": 6.190052490234375,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.09584002486660105,
|
| 266 |
+
"grad_norm": 0.9993957877159119,
|
| 267 |
+
"learning_rate": 0.0002784768388023188,
|
| 268 |
+
"loss": 6.179526977539062,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.09843029580894161,
|
| 273 |
+
"grad_norm": 0.972730278968811,
|
| 274 |
+
"learning_rate": 0.00027767909376163373,
|
| 275 |
+
"loss": 6.153101196289063,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.10102056675128218,
|
| 280 |
+
"grad_norm": 0.962273359298706,
|
| 281 |
+
"learning_rate": 0.00027688134872094874,
|
| 282 |
+
"loss": 6.1879443359375,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.10361083769362275,
|
| 287 |
+
"grad_norm": 1.0591261386871338,
|
| 288 |
+
"learning_rate": 0.00027608360368026376,
|
| 289 |
+
"loss": 6.112931518554688,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.10620110863596333,
|
| 294 |
+
"grad_norm": 1.0552300214767456,
|
| 295 |
+
"learning_rate": 0.00027528585863957877,
|
| 296 |
+
"loss": 6.099534301757813,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.10879137957830388,
|
| 301 |
+
"grad_norm": 1.008030891418457,
|
| 302 |
+
"learning_rate": 0.0002744881135988938,
|
| 303 |
+
"loss": 6.118661499023437,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.11138165052064446,
|
| 308 |
+
"grad_norm": 1.0966074466705322,
|
| 309 |
+
"learning_rate": 0.00027369036855820874,
|
| 310 |
+
"loss": 6.07596923828125,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.11397192146298503,
|
| 315 |
+
"grad_norm": 1.1636556386947632,
|
| 316 |
+
"learning_rate": 0.0002728926235175238,
|
| 317 |
+
"loss": 6.071294555664062,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.1165621924053256,
|
| 322 |
+
"grad_norm": 0.9836180210113525,
|
| 323 |
+
"learning_rate": 0.00027209487847683876,
|
| 324 |
+
"loss": 6.070966796875,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.11915246334766616,
|
| 329 |
+
"grad_norm": 1.0915582180023193,
|
| 330 |
+
"learning_rate": 0.00027129713343615377,
|
| 331 |
+
"loss": 6.039877319335938,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.12174273429000673,
|
| 336 |
+
"grad_norm": 0.9727494120597839,
|
| 337 |
+
"learning_rate": 0.0002704993883954688,
|
| 338 |
+
"loss": 6.059029541015625,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.1243330052323473,
|
| 343 |
+
"grad_norm": 1.034834623336792,
|
| 344 |
+
"learning_rate": 0.0002697016433547838,
|
| 345 |
+
"loss": 6.029169921875,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.12692327617468788,
|
| 350 |
+
"grad_norm": 1.0125064849853516,
|
| 351 |
+
"learning_rate": 0.0002689038983140988,
|
| 352 |
+
"loss": 6.04001708984375,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.12951354711702845,
|
| 357 |
+
"grad_norm": 1.0140148401260376,
|
| 358 |
+
"learning_rate": 0.0002681061532734138,
|
| 359 |
+
"loss": 6.013587646484375,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.13210381805936902,
|
| 364 |
+
"grad_norm": 0.9864634275436401,
|
| 365 |
+
"learning_rate": 0.00026730840823272877,
|
| 366 |
+
"loss": 6.027371826171875,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.13469408900170957,
|
| 371 |
+
"grad_norm": 1.0383540391921997,
|
| 372 |
+
"learning_rate": 0.0002665106631920438,
|
| 373 |
+
"loss": 5.929879150390625,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.13728435994405014,
|
| 378 |
+
"grad_norm": 1.0422724485397339,
|
| 379 |
+
"learning_rate": 0.0002657129181513588,
|
| 380 |
+
"loss": 5.992667236328125,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.1398746308863907,
|
| 385 |
+
"grad_norm": 1.1296474933624268,
|
| 386 |
+
"learning_rate": 0.0002649151731106738,
|
| 387 |
+
"loss": 5.917919311523438,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.14246490182873128,
|
| 392 |
+
"grad_norm": 0.9819442629814148,
|
| 393 |
+
"learning_rate": 0.0002641174280699888,
|
| 394 |
+
"loss": 5.95181884765625,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.14505517277107186,
|
| 399 |
+
"grad_norm": 0.9991300106048584,
|
| 400 |
+
"learning_rate": 0.0002633196830293038,
|
| 401 |
+
"loss": 5.92015869140625,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.14764544371341243,
|
| 406 |
+
"grad_norm": 0.9796678423881531,
|
| 407 |
+
"learning_rate": 0.00026252193798861884,
|
| 408 |
+
"loss": 5.9480438232421875,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.150235714655753,
|
| 413 |
+
"grad_norm": 0.9904295206069946,
|
| 414 |
+
"learning_rate": 0.0002617241929479338,
|
| 415 |
+
"loss": 5.948643188476563,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.15282598559809357,
|
| 420 |
+
"grad_norm": 0.9728788733482361,
|
| 421 |
+
"learning_rate": 0.0002609264479072488,
|
| 422 |
+
"loss": 5.874560546875,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.15541625654043412,
|
| 427 |
+
"grad_norm": 1.0392069816589355,
|
| 428 |
+
"learning_rate": 0.0002601287028665638,
|
| 429 |
+
"loss": 5.878630981445313,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.1580065274827747,
|
| 434 |
+
"grad_norm": 1.0595394372940063,
|
| 435 |
+
"learning_rate": 0.00025933095782587883,
|
| 436 |
+
"loss": 5.886410522460937,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.16059679842511526,
|
| 441 |
+
"grad_norm": 1.0325700044631958,
|
| 442 |
+
"learning_rate": 0.00025853321278519384,
|
| 443 |
+
"loss": 5.828760375976563,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.16318706936745583,
|
| 448 |
+
"grad_norm": 0.9954292178153992,
|
| 449 |
+
"learning_rate": 0.00025773546774450885,
|
| 450 |
+
"loss": 5.899686889648438,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.1657773403097964,
|
| 455 |
+
"grad_norm": 1.0147758722305298,
|
| 456 |
+
"learning_rate": 0.0002569377227038238,
|
| 457 |
+
"loss": 5.906227416992188,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.16836761125213698,
|
| 462 |
+
"grad_norm": 0.948711633682251,
|
| 463 |
+
"learning_rate": 0.0002561399776631389,
|
| 464 |
+
"loss": 5.840421752929688,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.17095788219447755,
|
| 469 |
+
"grad_norm": 1.0470232963562012,
|
| 470 |
+
"learning_rate": 0.00025534223262245383,
|
| 471 |
+
"loss": 5.837828369140625,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.17354815313681812,
|
| 476 |
+
"grad_norm": 1.0115586519241333,
|
| 477 |
+
"learning_rate": 0.00025454448758176885,
|
| 478 |
+
"loss": 5.83600830078125,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.17613842407915867,
|
| 483 |
+
"grad_norm": 1.035805583000183,
|
| 484 |
+
"learning_rate": 0.00025374674254108386,
|
| 485 |
+
"loss": 5.834181518554687,
|
| 486 |
+
"step": 3400
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.17872869502149924,
|
| 490 |
+
"grad_norm": 1.0050755739212036,
|
| 491 |
+
"learning_rate": 0.00025294899750039887,
|
| 492 |
+
"loss": 5.810164794921875,
|
| 493 |
+
"step": 3450
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.1813189659638398,
|
| 497 |
+
"grad_norm": 1.010113000869751,
|
| 498 |
+
"learning_rate": 0.0002521512524597139,
|
| 499 |
+
"loss": 5.795087280273438,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.18390923690618038,
|
| 504 |
+
"grad_norm": 0.958347737789154,
|
| 505 |
+
"learning_rate": 0.00025135350741902884,
|
| 506 |
+
"loss": 5.823550415039063,
|
| 507 |
+
"step": 3550
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.18649950784852096,
|
| 511 |
+
"grad_norm": 1.0313442945480347,
|
| 512 |
+
"learning_rate": 0.00025055576237834385,
|
| 513 |
+
"loss": 5.788181762695313,
|
| 514 |
+
"step": 3600
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.18908977879086153,
|
| 518 |
+
"grad_norm": 1.0172735452651978,
|
| 519 |
+
"learning_rate": 0.00024975801733765886,
|
| 520 |
+
"loss": 5.783030395507812,
|
| 521 |
+
"step": 3650
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.1916800497332021,
|
| 525 |
+
"grad_norm": 0.9737220406532288,
|
| 526 |
+
"learning_rate": 0.00024896027229697387,
|
| 527 |
+
"loss": 5.807670288085937,
|
| 528 |
+
"step": 3700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.19427032067554267,
|
| 532 |
+
"grad_norm": 0.9933872222900391,
|
| 533 |
+
"learning_rate": 0.0002481625272562889,
|
| 534 |
+
"loss": 5.762023315429688,
|
| 535 |
+
"step": 3750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.19686059161788322,
|
| 539 |
+
"grad_norm": 0.9616476893424988,
|
| 540 |
+
"learning_rate": 0.00024736478221560384,
|
| 541 |
+
"loss": 5.771016845703125,
|
| 542 |
+
"step": 3800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.1994508625602238,
|
| 546 |
+
"grad_norm": 1.0486760139465332,
|
| 547 |
+
"learning_rate": 0.00024656703717491885,
|
| 548 |
+
"loss": 5.7555908203125,
|
| 549 |
+
"step": 3850
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.20204113350256436,
|
| 553 |
+
"grad_norm": 1.0453466176986694,
|
| 554 |
+
"learning_rate": 0.00024576929213423386,
|
| 555 |
+
"loss": 5.721697998046875,
|
| 556 |
+
"step": 3900
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.20463140444490494,
|
| 560 |
+
"grad_norm": 0.9990628361701965,
|
| 561 |
+
"learning_rate": 0.00024497154709354887,
|
| 562 |
+
"loss": 5.755276489257812,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.2072216753872455,
|
| 567 |
+
"grad_norm": 1.0465604066848755,
|
| 568 |
+
"learning_rate": 0.0002441738020528639,
|
| 569 |
+
"loss": 5.7613458251953125,
|
| 570 |
+
"step": 4000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.20981194632958608,
|
| 574 |
+
"grad_norm": 1.022708773612976,
|
| 575 |
+
"learning_rate": 0.00024337605701217887,
|
| 576 |
+
"loss": 5.7307891845703125,
|
| 577 |
+
"step": 4050
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.21240221727192665,
|
| 581 |
+
"grad_norm": 1.0304657220840454,
|
| 582 |
+
"learning_rate": 0.0002425783119714939,
|
| 583 |
+
"loss": 5.68947998046875,
|
| 584 |
+
"step": 4100
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.21499248821426722,
|
| 588 |
+
"grad_norm": 1.0114036798477173,
|
| 589 |
+
"learning_rate": 0.0002417805669308089,
|
| 590 |
+
"loss": 5.721485595703125,
|
| 591 |
+
"step": 4150
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.21758275915660777,
|
| 595 |
+
"grad_norm": 1.0038365125656128,
|
| 596 |
+
"learning_rate": 0.0002409828218901239,
|
| 597 |
+
"loss": 5.75361328125,
|
| 598 |
+
"step": 4200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.22017303009894834,
|
| 602 |
+
"grad_norm": 1.030957818031311,
|
| 603 |
+
"learning_rate": 0.00024018507684943889,
|
| 604 |
+
"loss": 5.660762329101562,
|
| 605 |
+
"step": 4250
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.2227633010412889,
|
| 609 |
+
"grad_norm": 0.9766149520874023,
|
| 610 |
+
"learning_rate": 0.00023938733180875392,
|
| 611 |
+
"loss": 5.695264892578125,
|
| 612 |
+
"step": 4300
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.22535357198362949,
|
| 616 |
+
"grad_norm": 0.9994016289710999,
|
| 617 |
+
"learning_rate": 0.0002385895867680689,
|
| 618 |
+
"loss": 5.690259399414063,
|
| 619 |
+
"step": 4350
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.22794384292597006,
|
| 623 |
+
"grad_norm": 1.0340245962142944,
|
| 624 |
+
"learning_rate": 0.0002377918417273839,
|
| 625 |
+
"loss": 5.654801635742188,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.23053411386831063,
|
| 630 |
+
"grad_norm": 1.1064614057540894,
|
| 631 |
+
"learning_rate": 0.0002369940966866989,
|
| 632 |
+
"loss": 5.675580444335938,
|
| 633 |
+
"step": 4450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.2331243848106512,
|
| 637 |
+
"grad_norm": 1.084747314453125,
|
| 638 |
+
"learning_rate": 0.00023619635164601392,
|
| 639 |
+
"loss": 5.655352783203125,
|
| 640 |
+
"step": 4500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.23571465575299178,
|
| 644 |
+
"grad_norm": 0.9864429831504822,
|
| 645 |
+
"learning_rate": 0.00023539860660532893,
|
| 646 |
+
"loss": 5.62256103515625,
|
| 647 |
+
"step": 4550
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.23830492669533232,
|
| 651 |
+
"grad_norm": 0.9708291292190552,
|
| 652 |
+
"learning_rate": 0.0002346008615646439,
|
| 653 |
+
"loss": 5.628120727539063,
|
| 654 |
+
"step": 4600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.2408951976376729,
|
| 658 |
+
"grad_norm": 1.0555731058120728,
|
| 659 |
+
"learning_rate": 0.00023380311652395892,
|
| 660 |
+
"loss": 5.648033447265625,
|
| 661 |
+
"step": 4650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.24348546858001346,
|
| 665 |
+
"grad_norm": 0.979049801826477,
|
| 666 |
+
"learning_rate": 0.00023300537148327393,
|
| 667 |
+
"loss": 5.658864135742188,
|
| 668 |
+
"step": 4700
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.24607573952235404,
|
| 672 |
+
"grad_norm": 0.9773157835006714,
|
| 673 |
+
"learning_rate": 0.00023220762644258895,
|
| 674 |
+
"loss": 5.631370239257812,
|
| 675 |
+
"step": 4750
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.2486660104646946,
|
| 679 |
+
"grad_norm": 1.0320409536361694,
|
| 680 |
+
"learning_rate": 0.00023140988140190393,
|
| 681 |
+
"loss": 5.6348028564453125,
|
| 682 |
+
"step": 4800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.25125628140703515,
|
| 686 |
+
"grad_norm": 1.0131503343582153,
|
| 687 |
+
"learning_rate": 0.00023061213636121891,
|
| 688 |
+
"loss": 5.562122802734375,
|
| 689 |
+
"step": 4850
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.25384655234937575,
|
| 693 |
+
"grad_norm": 1.0046422481536865,
|
| 694 |
+
"learning_rate": 0.00022981439132053395,
|
| 695 |
+
"loss": 5.629116821289062,
|
| 696 |
+
"step": 4900
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.2564368232917163,
|
| 700 |
+
"grad_norm": 1.0059541463851929,
|
| 701 |
+
"learning_rate": 0.00022901664627984894,
|
| 702 |
+
"loss": 5.580728149414062,
|
| 703 |
+
"step": 4950
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.2590270942340569,
|
| 707 |
+
"grad_norm": 0.9650059938430786,
|
| 708 |
+
"learning_rate": 0.00022821890123916395,
|
| 709 |
+
"loss": 5.640592041015625,
|
| 710 |
+
"step": 5000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.26161736517639744,
|
| 714 |
+
"grad_norm": 1.0083752870559692,
|
| 715 |
+
"learning_rate": 0.00022742115619847893,
|
| 716 |
+
"loss": 5.601475219726563,
|
| 717 |
+
"step": 5050
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.26420763611873804,
|
| 721 |
+
"grad_norm": 1.0079939365386963,
|
| 722 |
+
"learning_rate": 0.00022662341115779394,
|
| 723 |
+
"loss": 5.613602905273438,
|
| 724 |
+
"step": 5100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.2667979070610786,
|
| 728 |
+
"grad_norm": 1.0314289331436157,
|
| 729 |
+
"learning_rate": 0.00022582566611710895,
|
| 730 |
+
"loss": 5.603458862304688,
|
| 731 |
+
"step": 5150
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.26938817800341913,
|
| 735 |
+
"grad_norm": 1.077232837677002,
|
| 736 |
+
"learning_rate": 0.00022502792107642397,
|
| 737 |
+
"loss": 5.565240478515625,
|
| 738 |
+
"step": 5200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.27197844894575973,
|
| 742 |
+
"grad_norm": 1.0053828954696655,
|
| 743 |
+
"learning_rate": 0.00022423017603573895,
|
| 744 |
+
"loss": 5.59989013671875,
|
| 745 |
+
"step": 5250
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.2745687198881003,
|
| 749 |
+
"grad_norm": 0.9962897300720215,
|
| 750 |
+
"learning_rate": 0.00022343243099505393,
|
| 751 |
+
"loss": 5.561985473632813,
|
| 752 |
+
"step": 5300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.2771589908304409,
|
| 756 |
+
"grad_norm": 0.9912266135215759,
|
| 757 |
+
"learning_rate": 0.00022263468595436897,
|
| 758 |
+
"loss": 5.577881469726562,
|
| 759 |
+
"step": 5350
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.2797492617727814,
|
| 763 |
+
"grad_norm": 0.9959378242492676,
|
| 764 |
+
"learning_rate": 0.00022183694091368396,
|
| 765 |
+
"loss": 5.560752563476562,
|
| 766 |
+
"step": 5400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.282339532715122,
|
| 770 |
+
"grad_norm": 0.9649254083633423,
|
| 771 |
+
"learning_rate": 0.00022103919587299897,
|
| 772 |
+
"loss": 5.5196160888671875,
|
| 773 |
+
"step": 5450
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.28492980365746257,
|
| 777 |
+
"grad_norm": 0.9408061504364014,
|
| 778 |
+
"learning_rate": 0.00022024145083231395,
|
| 779 |
+
"loss": 5.555208129882812,
|
| 780 |
+
"step": 5500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.28752007459980317,
|
| 784 |
+
"grad_norm": 0.9576660990715027,
|
| 785 |
+
"learning_rate": 0.000219443705791629,
|
| 786 |
+
"loss": 5.521038208007813,
|
| 787 |
+
"step": 5550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.2901103455421437,
|
| 791 |
+
"grad_norm": 1.060819149017334,
|
| 792 |
+
"learning_rate": 0.00021864596075094398,
|
| 793 |
+
"loss": 5.531424560546875,
|
| 794 |
+
"step": 5600
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.29270061648448426,
|
| 798 |
+
"grad_norm": 1.0806543827056885,
|
| 799 |
+
"learning_rate": 0.000217848215710259,
|
| 800 |
+
"loss": 5.504271850585938,
|
| 801 |
+
"step": 5650
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.29529088742682486,
|
| 805 |
+
"grad_norm": 0.9876242876052856,
|
| 806 |
+
"learning_rate": 0.00021705047066957397,
|
| 807 |
+
"loss": 5.4666552734375,
|
| 808 |
+
"step": 5700
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.2978811583691654,
|
| 812 |
+
"grad_norm": 1.0988022089004517,
|
| 813 |
+
"learning_rate": 0.000216252725628889,
|
| 814 |
+
"loss": 5.555277099609375,
|
| 815 |
+
"step": 5750
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.300471429311506,
|
| 819 |
+
"grad_norm": 0.9912010431289673,
|
| 820 |
+
"learning_rate": 0.000215454980588204,
|
| 821 |
+
"loss": 5.5327294921875,
|
| 822 |
+
"step": 5800
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.30306170025384654,
|
| 826 |
+
"grad_norm": 0.9935503602027893,
|
| 827 |
+
"learning_rate": 0.00021465723554751898,
|
| 828 |
+
"loss": 5.524019775390625,
|
| 829 |
+
"step": 5850
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.30565197119618714,
|
| 833 |
+
"grad_norm": 1.0960224866867065,
|
| 834 |
+
"learning_rate": 0.000213859490506834,
|
| 835 |
+
"loss": 5.49506103515625,
|
| 836 |
+
"step": 5900
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.3082422421385277,
|
| 840 |
+
"grad_norm": 1.0867022275924683,
|
| 841 |
+
"learning_rate": 0.00021306174546614903,
|
| 842 |
+
"loss": 5.5002587890625,
|
| 843 |
+
"step": 5950
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.31083251308086823,
|
| 847 |
+
"grad_norm": 1.0140448808670044,
|
| 848 |
+
"learning_rate": 0.000212264000425464,
|
| 849 |
+
"loss": 5.462314453125,
|
| 850 |
+
"step": 6000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.31342278402320883,
|
| 854 |
+
"grad_norm": 1.0166538953781128,
|
| 855 |
+
"learning_rate": 0.000211466255384779,
|
| 856 |
+
"loss": 5.486057739257813,
|
| 857 |
+
"step": 6050
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.3160130549655494,
|
| 861 |
+
"grad_norm": 0.9597861170768738,
|
| 862 |
+
"learning_rate": 0.000210668510344094,
|
| 863 |
+
"loss": 5.470111694335937,
|
| 864 |
+
"step": 6100
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.31860332590789,
|
| 868 |
+
"grad_norm": 0.9873929619789124,
|
| 869 |
+
"learning_rate": 0.00020987076530340902,
|
| 870 |
+
"loss": 5.505219116210937,
|
| 871 |
+
"step": 6150
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.3211935968502305,
|
| 875 |
+
"grad_norm": 0.9906278848648071,
|
| 876 |
+
"learning_rate": 0.00020907302026272403,
|
| 877 |
+
"loss": 5.458482055664063,
|
| 878 |
+
"step": 6200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.3237838677925711,
|
| 882 |
+
"grad_norm": 0.9522607922554016,
|
| 883 |
+
"learning_rate": 0.00020827527522203901,
|
| 884 |
+
"loss": 5.48074951171875,
|
| 885 |
+
"step": 6250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.32637413873491167,
|
| 889 |
+
"grad_norm": 1.0274453163146973,
|
| 890 |
+
"learning_rate": 0.00020747753018135403,
|
| 891 |
+
"loss": 5.457478637695313,
|
| 892 |
+
"step": 6300
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.32896440967725227,
|
| 896 |
+
"grad_norm": 1.0103005170822144,
|
| 897 |
+
"learning_rate": 0.000206679785140669,
|
| 898 |
+
"loss": 5.503182983398437,
|
| 899 |
+
"step": 6350
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.3315546806195928,
|
| 903 |
+
"grad_norm": 0.9858667254447937,
|
| 904 |
+
"learning_rate": 0.00020588204009998405,
|
| 905 |
+
"loss": 5.424717407226563,
|
| 906 |
+
"step": 6400
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.33414495156193336,
|
| 910 |
+
"grad_norm": 1.0011284351348877,
|
| 911 |
+
"learning_rate": 0.00020508429505929903,
|
| 912 |
+
"loss": 5.419669799804687,
|
| 913 |
+
"step": 6450
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.33673522250427396,
|
| 917 |
+
"grad_norm": 0.9662808179855347,
|
| 918 |
+
"learning_rate": 0.00020428655001861402,
|
| 919 |
+
"loss": 5.426592407226562,
|
| 920 |
+
"step": 6500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.3393254934466145,
|
| 924 |
+
"grad_norm": 0.9978706240653992,
|
| 925 |
+
"learning_rate": 0.00020348880497792903,
|
| 926 |
+
"loss": 5.404547119140625,
|
| 927 |
+
"step": 6550
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.3419157643889551,
|
| 931 |
+
"grad_norm": 0.9442576169967651,
|
| 932 |
+
"learning_rate": 0.00020269105993724404,
|
| 933 |
+
"loss": 5.452359008789062,
|
| 934 |
+
"step": 6600
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.34450603533129565,
|
| 938 |
+
"grad_norm": 1.0593928098678589,
|
| 939 |
+
"learning_rate": 0.00020189331489655905,
|
| 940 |
+
"loss": 5.419390869140625,
|
| 941 |
+
"step": 6650
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.34709630627363625,
|
| 945 |
+
"grad_norm": 1.0044008493423462,
|
| 946 |
+
"learning_rate": 0.00020109556985587403,
|
| 947 |
+
"loss": 5.416279296875,
|
| 948 |
+
"step": 6700
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.3496865772159768,
|
| 952 |
+
"grad_norm": 0.9485145807266235,
|
| 953 |
+
"learning_rate": 0.00020029782481518905,
|
| 954 |
+
"loss": 5.431090087890625,
|
| 955 |
+
"step": 6750
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.35227684815831734,
|
| 959 |
+
"grad_norm": 0.9767470955848694,
|
| 960 |
+
"learning_rate": 0.00019950007977450406,
|
| 961 |
+
"loss": 5.429151000976563,
|
| 962 |
+
"step": 6800
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.35486711910065794,
|
| 966 |
+
"grad_norm": 0.9516136646270752,
|
| 967 |
+
"learning_rate": 0.00019870233473381907,
|
| 968 |
+
"loss": 5.393023071289062,
|
| 969 |
+
"step": 6850
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.3574573900429985,
|
| 973 |
+
"grad_norm": 1.0077961683273315,
|
| 974 |
+
"learning_rate": 0.00019790458969313405,
|
| 975 |
+
"loss": 5.4060986328125,
|
| 976 |
+
"step": 6900
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.3600476609853391,
|
| 980 |
+
"grad_norm": 1.0050827264785767,
|
| 981 |
+
"learning_rate": 0.00019710684465244904,
|
| 982 |
+
"loss": 5.4089599609375,
|
| 983 |
+
"step": 6950
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.3626379319276796,
|
| 987 |
+
"grad_norm": 1.0005428791046143,
|
| 988 |
+
"learning_rate": 0.00019630909961176408,
|
| 989 |
+
"loss": 5.408168334960937,
|
| 990 |
+
"step": 7000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.3652282028700202,
|
| 994 |
+
"grad_norm": 0.9248517751693726,
|
| 995 |
+
"learning_rate": 0.00019551135457107906,
|
| 996 |
+
"loss": 5.355136108398438,
|
| 997 |
+
"step": 7050
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.36781847381236077,
|
| 1001 |
+
"grad_norm": 0.9752355813980103,
|
| 1002 |
+
"learning_rate": 0.00019471360953039407,
|
| 1003 |
+
"loss": 5.386387939453125,
|
| 1004 |
+
"step": 7100
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.37040874475470137,
|
| 1008 |
+
"grad_norm": 1.193434238433838,
|
| 1009 |
+
"learning_rate": 0.00019391586448970906,
|
| 1010 |
+
"loss": 5.354227905273437,
|
| 1011 |
+
"step": 7150
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.3729990156970419,
|
| 1015 |
+
"grad_norm": 0.967904806137085,
|
| 1016 |
+
"learning_rate": 0.0001931181194490241,
|
| 1017 |
+
"loss": 5.36183349609375,
|
| 1018 |
+
"step": 7200
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.37558928663938246,
|
| 1022 |
+
"grad_norm": 0.9852302074432373,
|
| 1023 |
+
"learning_rate": 0.00019232037440833908,
|
| 1024 |
+
"loss": 5.32298583984375,
|
| 1025 |
+
"step": 7250
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.37817955758172306,
|
| 1029 |
+
"grad_norm": 1.0207830667495728,
|
| 1030 |
+
"learning_rate": 0.0001915226293676541,
|
| 1031 |
+
"loss": 5.336920166015625,
|
| 1032 |
+
"step": 7300
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.3807698285240636,
|
| 1036 |
+
"grad_norm": 0.9589182734489441,
|
| 1037 |
+
"learning_rate": 0.00019072488432696907,
|
| 1038 |
+
"loss": 5.341430053710938,
|
| 1039 |
+
"step": 7350
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.3833600994664042,
|
| 1043 |
+
"grad_norm": 1.0570025444030762,
|
| 1044 |
+
"learning_rate": 0.0001899271392862841,
|
| 1045 |
+
"loss": 5.38286376953125,
|
| 1046 |
+
"step": 7400
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.38595037040874475,
|
| 1050 |
+
"grad_norm": 0.9938406944274902,
|
| 1051 |
+
"learning_rate": 0.0001891293942455991,
|
| 1052 |
+
"loss": 5.3147039794921875,
|
| 1053 |
+
"step": 7450
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.38854064135108535,
|
| 1057 |
+
"grad_norm": 1.020323395729065,
|
| 1058 |
+
"learning_rate": 0.00018833164920491408,
|
| 1059 |
+
"loss": 5.34332763671875,
|
| 1060 |
+
"step": 7500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.3911309122934259,
|
| 1064 |
+
"grad_norm": 1.042059302330017,
|
| 1065 |
+
"learning_rate": 0.0001875339041642291,
|
| 1066 |
+
"loss": 5.33760986328125,
|
| 1067 |
+
"step": 7550
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.39372118323576644,
|
| 1071 |
+
"grad_norm": 0.975479245185852,
|
| 1072 |
+
"learning_rate": 0.00018673615912354408,
|
| 1073 |
+
"loss": 5.353353271484375,
|
| 1074 |
+
"step": 7600
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.39631145417810704,
|
| 1078 |
+
"grad_norm": 0.9893775582313538,
|
| 1079 |
+
"learning_rate": 0.00018593841408285911,
|
| 1080 |
+
"loss": 5.3269775390625,
|
| 1081 |
+
"step": 7650
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.3989017251204476,
|
| 1085 |
+
"grad_norm": 1.1089980602264404,
|
| 1086 |
+
"learning_rate": 0.0001851406690421741,
|
| 1087 |
+
"loss": 5.3275830078125,
|
| 1088 |
+
"step": 7700
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.4014919960627882,
|
| 1092 |
+
"grad_norm": 1.0087932348251343,
|
| 1093 |
+
"learning_rate": 0.0001843429240014891,
|
| 1094 |
+
"loss": 5.347962036132812,
|
| 1095 |
+
"step": 7750
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.4040822670051287,
|
| 1099 |
+
"grad_norm": 1.0132794380187988,
|
| 1100 |
+
"learning_rate": 0.0001835451789608041,
|
| 1101 |
+
"loss": 5.295555419921875,
|
| 1102 |
+
"step": 7800
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.4066725379474693,
|
| 1106 |
+
"grad_norm": 1.0221766233444214,
|
| 1107 |
+
"learning_rate": 0.00018274743392011913,
|
| 1108 |
+
"loss": 5.299251098632812,
|
| 1109 |
+
"step": 7850
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.40926280888980987,
|
| 1113 |
+
"grad_norm": 1.0063681602478027,
|
| 1114 |
+
"learning_rate": 0.00018194968887943412,
|
| 1115 |
+
"loss": 5.335856323242187,
|
| 1116 |
+
"step": 7900
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.41185307983215047,
|
| 1120 |
+
"grad_norm": 1.0150210857391357,
|
| 1121 |
+
"learning_rate": 0.0001811519438387491,
|
| 1122 |
+
"loss": 5.261018676757812,
|
| 1123 |
+
"step": 7950
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.414443350774491,
|
| 1127 |
+
"grad_norm": 0.9695018529891968,
|
| 1128 |
+
"learning_rate": 0.0001803541987980641,
|
| 1129 |
+
"loss": 5.286878662109375,
|
| 1130 |
+
"step": 8000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.41703362171683156,
|
| 1134 |
+
"grad_norm": 0.982979953289032,
|
| 1135 |
+
"learning_rate": 0.00017955645375737915,
|
| 1136 |
+
"loss": 5.3470751953125,
|
| 1137 |
+
"step": 8050
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.41962389265917216,
|
| 1141 |
+
"grad_norm": 1.0282922983169556,
|
| 1142 |
+
"learning_rate": 0.00017875870871669413,
|
| 1143 |
+
"loss": 5.333184814453125,
|
| 1144 |
+
"step": 8100
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.4222141636015127,
|
| 1148 |
+
"grad_norm": 1.0405290126800537,
|
| 1149 |
+
"learning_rate": 0.00017796096367600912,
|
| 1150 |
+
"loss": 5.27518310546875,
|
| 1151 |
+
"step": 8150
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.4248044345438533,
|
| 1155 |
+
"grad_norm": 1.0299566984176636,
|
| 1156 |
+
"learning_rate": 0.00017716321863532413,
|
| 1157 |
+
"loss": 5.2979150390625,
|
| 1158 |
+
"step": 8200
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.42739470548619385,
|
| 1162 |
+
"grad_norm": 0.9923998117446899,
|
| 1163 |
+
"learning_rate": 0.00017636547359463914,
|
| 1164 |
+
"loss": 5.288524780273438,
|
| 1165 |
+
"step": 8250
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.42998497642853445,
|
| 1169 |
+
"grad_norm": 0.9922242164611816,
|
| 1170 |
+
"learning_rate": 0.00017556772855395415,
|
| 1171 |
+
"loss": 5.305535888671875,
|
| 1172 |
+
"step": 8300
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.432575247370875,
|
| 1176 |
+
"grad_norm": 1.057271957397461,
|
| 1177 |
+
"learning_rate": 0.00017476998351326914,
|
| 1178 |
+
"loss": 5.241010131835938,
|
| 1179 |
+
"step": 8350
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.43516551831321554,
|
| 1183 |
+
"grad_norm": 1.00055730342865,
|
| 1184 |
+
"learning_rate": 0.00017397223847258415,
|
| 1185 |
+
"loss": 5.270580444335938,
|
| 1186 |
+
"step": 8400
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.43775578925555614,
|
| 1190 |
+
"grad_norm": 0.9761698842048645,
|
| 1191 |
+
"learning_rate": 0.00017317449343189916,
|
| 1192 |
+
"loss": 5.28654052734375,
|
| 1193 |
+
"step": 8450
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.4403460601978967,
|
| 1197 |
+
"grad_norm": 1.0556055307388306,
|
| 1198 |
+
"learning_rate": 0.00017237674839121417,
|
| 1199 |
+
"loss": 5.2256103515625,
|
| 1200 |
+
"step": 8500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.4429363311402373,
|
| 1204 |
+
"grad_norm": 1.032482624053955,
|
| 1205 |
+
"learning_rate": 0.00017157900335052916,
|
| 1206 |
+
"loss": 5.19790283203125,
|
| 1207 |
+
"step": 8550
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.4455266020825778,
|
| 1211 |
+
"grad_norm": 0.9819381833076477,
|
| 1212 |
+
"learning_rate": 0.00017078125830984414,
|
| 1213 |
+
"loss": 5.22557373046875,
|
| 1214 |
+
"step": 8600
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.4481168730249184,
|
| 1218 |
+
"grad_norm": 1.0934169292449951,
|
| 1219 |
+
"learning_rate": 0.00016998351326915918,
|
| 1220 |
+
"loss": 5.261765747070313,
|
| 1221 |
+
"step": 8650
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.45070714396725897,
|
| 1225 |
+
"grad_norm": 1.0120759010314941,
|
| 1226 |
+
"learning_rate": 0.00016918576822847416,
|
| 1227 |
+
"loss": 5.285647583007813,
|
| 1228 |
+
"step": 8700
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.4532974149095995,
|
| 1232 |
+
"grad_norm": 0.9943802356719971,
|
| 1233 |
+
"learning_rate": 0.00016838802318778917,
|
| 1234 |
+
"loss": 5.223272094726562,
|
| 1235 |
+
"step": 8750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.4558876858519401,
|
| 1239 |
+
"grad_norm": 0.992707371711731,
|
| 1240 |
+
"learning_rate": 0.00016759027814710416,
|
| 1241 |
+
"loss": 5.220896606445312,
|
| 1242 |
+
"step": 8800
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.45847795679428066,
|
| 1246 |
+
"grad_norm": 1.0634857416152954,
|
| 1247 |
+
"learning_rate": 0.00016679253310641917,
|
| 1248 |
+
"loss": 5.202445678710937,
|
| 1249 |
+
"step": 8850
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.46106822773662126,
|
| 1253 |
+
"grad_norm": 1.0071600675582886,
|
| 1254 |
+
"learning_rate": 0.00016599478806573418,
|
| 1255 |
+
"loss": 5.221668701171875,
|
| 1256 |
+
"step": 8900
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.4636584986789618,
|
| 1260 |
+
"grad_norm": 1.0665333271026611,
|
| 1261 |
+
"learning_rate": 0.0001651970430250492,
|
| 1262 |
+
"loss": 5.246253662109375,
|
| 1263 |
+
"step": 8950
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.4662487696213024,
|
| 1267 |
+
"grad_norm": 0.9957162141799927,
|
| 1268 |
+
"learning_rate": 0.00016439929798436418,
|
| 1269 |
+
"loss": 5.2391455078125,
|
| 1270 |
+
"step": 9000
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.46883904056364295,
|
| 1274 |
+
"grad_norm": 1.018820881843567,
|
| 1275 |
+
"learning_rate": 0.00016360155294367916,
|
| 1276 |
+
"loss": 5.238643188476562,
|
| 1277 |
+
"step": 9050
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.47142931150598355,
|
| 1281 |
+
"grad_norm": 1.032936692237854,
|
| 1282 |
+
"learning_rate": 0.0001628038079029942,
|
| 1283 |
+
"loss": 5.20821044921875,
|
| 1284 |
+
"step": 9100
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.4740195824483241,
|
| 1288 |
+
"grad_norm": 0.9814261794090271,
|
| 1289 |
+
"learning_rate": 0.00016200606286230918,
|
| 1290 |
+
"loss": 5.1940673828125,
|
| 1291 |
+
"step": 9150
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.47660985339066464,
|
| 1295 |
+
"grad_norm": 1.0158833265304565,
|
| 1296 |
+
"learning_rate": 0.0001612083178216242,
|
| 1297 |
+
"loss": 5.216905517578125,
|
| 1298 |
+
"step": 9200
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.47920012433300524,
|
| 1302 |
+
"grad_norm": 1.1137468814849854,
|
| 1303 |
+
"learning_rate": 0.00016041057278093918,
|
| 1304 |
+
"loss": 5.2248681640625,
|
| 1305 |
+
"step": 9250
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.4817903952753458,
|
| 1309 |
+
"grad_norm": 1.0213594436645508,
|
| 1310 |
+
"learning_rate": 0.00015961282774025422,
|
| 1311 |
+
"loss": 5.193059692382812,
|
| 1312 |
+
"step": 9300
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.4843806662176864,
|
| 1316 |
+
"grad_norm": 1.0907715559005737,
|
| 1317 |
+
"learning_rate": 0.0001588150826995692,
|
| 1318 |
+
"loss": 5.1917724609375,
|
| 1319 |
+
"step": 9350
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.48697093716002693,
|
| 1323 |
+
"grad_norm": 1.042995572090149,
|
| 1324 |
+
"learning_rate": 0.0001580173376588842,
|
| 1325 |
+
"loss": 5.22154296875,
|
| 1326 |
+
"step": 9400
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.48956120810236753,
|
| 1330 |
+
"grad_norm": 1.0548311471939087,
|
| 1331 |
+
"learning_rate": 0.0001572195926181992,
|
| 1332 |
+
"loss": 5.220418701171875,
|
| 1333 |
+
"step": 9450
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.4921514790447081,
|
| 1337 |
+
"grad_norm": 1.0239475965499878,
|
| 1338 |
+
"learning_rate": 0.00015642184757751424,
|
| 1339 |
+
"loss": 5.175424194335937,
|
| 1340 |
+
"step": 9500
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.4947417499870486,
|
| 1344 |
+
"grad_norm": 1.0291051864624023,
|
| 1345 |
+
"learning_rate": 0.00015562410253682922,
|
| 1346 |
+
"loss": 5.191710815429688,
|
| 1347 |
+
"step": 9550
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.4973320209293892,
|
| 1351 |
+
"grad_norm": 1.0522440671920776,
|
| 1352 |
+
"learning_rate": 0.0001548263574961442,
|
| 1353 |
+
"loss": 5.157244262695312,
|
| 1354 |
+
"step": 9600
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.49992229187172976,
|
| 1358 |
+
"grad_norm": 1.0253952741622925,
|
| 1359 |
+
"learning_rate": 0.00015402861245545921,
|
| 1360 |
+
"loss": 5.150928344726562,
|
| 1361 |
+
"step": 9650
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.5025125628140703,
|
| 1365 |
+
"grad_norm": 1.0156198740005493,
|
| 1366 |
+
"learning_rate": 0.00015323086741477423,
|
| 1367 |
+
"loss": 5.178720092773437,
|
| 1368 |
+
"step": 9700
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.505102833756411,
|
| 1372 |
+
"grad_norm": 1.0090805292129517,
|
| 1373 |
+
"learning_rate": 0.00015243312237408924,
|
| 1374 |
+
"loss": 5.203817749023438,
|
| 1375 |
+
"step": 9750
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.5076931046987515,
|
| 1379 |
+
"grad_norm": 1.0411272048950195,
|
| 1380 |
+
"learning_rate": 0.00015163537733340422,
|
| 1381 |
+
"loss": 5.192196044921875,
|
| 1382 |
+
"step": 9800
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.510283375641092,
|
| 1386 |
+
"grad_norm": 0.9986085891723633,
|
| 1387 |
+
"learning_rate": 0.00015083763229271923,
|
| 1388 |
+
"loss": 5.123272705078125,
|
| 1389 |
+
"step": 9850
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.5128736465834326,
|
| 1393 |
+
"grad_norm": 1.0344675779342651,
|
| 1394 |
+
"learning_rate": 0.00015003988725203424,
|
| 1395 |
+
"loss": 5.1883087158203125,
|
| 1396 |
+
"step": 9900
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.5154639175257731,
|
| 1400 |
+
"grad_norm": 1.1055423021316528,
|
| 1401 |
+
"learning_rate": 0.00014924214221134926,
|
| 1402 |
+
"loss": 5.184451904296875,
|
| 1403 |
+
"step": 9950
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.5180541884681138,
|
| 1407 |
+
"grad_norm": 1.0196657180786133,
|
| 1408 |
+
"learning_rate": 0.00014844439717066424,
|
| 1409 |
+
"loss": 5.228606567382813,
|
| 1410 |
+
"step": 10000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.5206444594104543,
|
| 1414 |
+
"grad_norm": 1.0357885360717773,
|
| 1415 |
+
"learning_rate": 0.00014764665212997925,
|
| 1416 |
+
"loss": 5.15303466796875,
|
| 1417 |
+
"step": 10050
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.5232347303527949,
|
| 1421 |
+
"grad_norm": 1.053410530090332,
|
| 1422 |
+
"learning_rate": 0.00014684890708929424,
|
| 1423 |
+
"loss": 5.157637329101562,
|
| 1424 |
+
"step": 10100
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.5258250012951354,
|
| 1428 |
+
"grad_norm": 1.0483365058898926,
|
| 1429 |
+
"learning_rate": 0.00014605116204860925,
|
| 1430 |
+
"loss": 5.137578125,
|
| 1431 |
+
"step": 10150
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.5284152722374761,
|
| 1435 |
+
"grad_norm": 1.03421151638031,
|
| 1436 |
+
"learning_rate": 0.00014525341700792426,
|
| 1437 |
+
"loss": 5.17541259765625,
|
| 1438 |
+
"step": 10200
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.5310055431798166,
|
| 1442 |
+
"grad_norm": 1.0358786582946777,
|
| 1443 |
+
"learning_rate": 0.00014445567196723924,
|
| 1444 |
+
"loss": 5.094895629882813,
|
| 1445 |
+
"step": 10250
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.5335958141221572,
|
| 1449 |
+
"grad_norm": 1.036773443222046,
|
| 1450 |
+
"learning_rate": 0.00014365792692655425,
|
| 1451 |
+
"loss": 5.12354736328125,
|
| 1452 |
+
"step": 10300
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.5361860850644977,
|
| 1456 |
+
"grad_norm": 1.0109663009643555,
|
| 1457 |
+
"learning_rate": 0.00014286018188586927,
|
| 1458 |
+
"loss": 5.14419921875,
|
| 1459 |
+
"step": 10350
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.5387763560068383,
|
| 1463 |
+
"grad_norm": 1.0550706386566162,
|
| 1464 |
+
"learning_rate": 0.00014206243684518428,
|
| 1465 |
+
"loss": 5.161801147460937,
|
| 1466 |
+
"step": 10400
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.5413666269491789,
|
| 1470 |
+
"grad_norm": 0.992174506187439,
|
| 1471 |
+
"learning_rate": 0.00014126469180449926,
|
| 1472 |
+
"loss": 5.097646179199219,
|
| 1473 |
+
"step": 10450
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.5439568978915195,
|
| 1477 |
+
"grad_norm": 1.0084872245788574,
|
| 1478 |
+
"learning_rate": 0.00014046694676381427,
|
| 1479 |
+
"loss": 5.099291687011719,
|
| 1480 |
+
"step": 10500
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.54654716883386,
|
| 1484 |
+
"grad_norm": 1.068367838859558,
|
| 1485 |
+
"learning_rate": 0.00013966920172312928,
|
| 1486 |
+
"loss": 5.135633544921875,
|
| 1487 |
+
"step": 10550
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.5491374397762006,
|
| 1491 |
+
"grad_norm": 1.1861008405685425,
|
| 1492 |
+
"learning_rate": 0.0001388714566824443,
|
| 1493 |
+
"loss": 5.086067810058593,
|
| 1494 |
+
"step": 10600
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.5517277107185412,
|
| 1498 |
+
"grad_norm": 1.0353467464447021,
|
| 1499 |
+
"learning_rate": 0.00013807371164175928,
|
| 1500 |
+
"loss": 5.1578912353515625,
|
| 1501 |
+
"step": 10650
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.5543179816608818,
|
| 1505 |
+
"grad_norm": 1.0449336767196655,
|
| 1506 |
+
"learning_rate": 0.0001372759666010743,
|
| 1507 |
+
"loss": 5.07172607421875,
|
| 1508 |
+
"step": 10700
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.5569082526032223,
|
| 1512 |
+
"grad_norm": 1.041418194770813,
|
| 1513 |
+
"learning_rate": 0.00013647822156038927,
|
| 1514 |
+
"loss": 5.111160888671875,
|
| 1515 |
+
"step": 10750
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.5594985235455628,
|
| 1519 |
+
"grad_norm": 1.0924108028411865,
|
| 1520 |
+
"learning_rate": 0.00013568047651970429,
|
| 1521 |
+
"loss": 5.09541015625,
|
| 1522 |
+
"step": 10800
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.5620887944879034,
|
| 1526 |
+
"grad_norm": 1.0294824838638306,
|
| 1527 |
+
"learning_rate": 0.0001348827314790193,
|
| 1528 |
+
"loss": 5.103350524902344,
|
| 1529 |
+
"step": 10850
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.564679065430244,
|
| 1533 |
+
"grad_norm": 1.0669180154800415,
|
| 1534 |
+
"learning_rate": 0.0001340849864383343,
|
| 1535 |
+
"loss": 5.056949768066406,
|
| 1536 |
+
"step": 10900
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.5672693363725846,
|
| 1540 |
+
"grad_norm": 1.0346755981445312,
|
| 1541 |
+
"learning_rate": 0.0001332872413976493,
|
| 1542 |
+
"loss": 5.089794921875,
|
| 1543 |
+
"step": 10950
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.5698596073149251,
|
| 1547 |
+
"grad_norm": 1.0336682796478271,
|
| 1548 |
+
"learning_rate": 0.0001324894963569643,
|
| 1549 |
+
"loss": 5.086961059570313,
|
| 1550 |
+
"step": 11000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.5724498782572657,
|
| 1554 |
+
"grad_norm": 1.112239956855774,
|
| 1555 |
+
"learning_rate": 0.00013169175131627932,
|
| 1556 |
+
"loss": 5.12211181640625,
|
| 1557 |
+
"step": 11050
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.5750401491996063,
|
| 1561 |
+
"grad_norm": 1.0269253253936768,
|
| 1562 |
+
"learning_rate": 0.00013089400627559433,
|
| 1563 |
+
"loss": 5.078684997558594,
|
| 1564 |
+
"step": 11100
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.5776304201419469,
|
| 1568 |
+
"grad_norm": 1.104164958000183,
|
| 1569 |
+
"learning_rate": 0.0001300962612349093,
|
| 1570 |
+
"loss": 5.09882080078125,
|
| 1571 |
+
"step": 11150
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.5802206910842874,
|
| 1575 |
+
"grad_norm": 1.0542353391647339,
|
| 1576 |
+
"learning_rate": 0.00012929851619422432,
|
| 1577 |
+
"loss": 5.109794921875,
|
| 1578 |
+
"step": 11200
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.582810962026628,
|
| 1582 |
+
"grad_norm": 1.0559531450271606,
|
| 1583 |
+
"learning_rate": 0.0001285007711535393,
|
| 1584 |
+
"loss": 5.06039306640625,
|
| 1585 |
+
"step": 11250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.5854012329689685,
|
| 1589 |
+
"grad_norm": 1.042899489402771,
|
| 1590 |
+
"learning_rate": 0.00012770302611285432,
|
| 1591 |
+
"loss": 5.068010864257812,
|
| 1592 |
+
"step": 11300
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.5879915039113092,
|
| 1596 |
+
"grad_norm": 1.002777099609375,
|
| 1597 |
+
"learning_rate": 0.00012690528107216933,
|
| 1598 |
+
"loss": 5.114893798828125,
|
| 1599 |
+
"step": 11350
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.5905817748536497,
|
| 1603 |
+
"grad_norm": 1.0369727611541748,
|
| 1604 |
+
"learning_rate": 0.0001261075360314843,
|
| 1605 |
+
"loss": 5.056740112304688,
|
| 1606 |
+
"step": 11400
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.5931720457959903,
|
| 1610 |
+
"grad_norm": 1.09577476978302,
|
| 1611 |
+
"learning_rate": 0.00012530979099079932,
|
| 1612 |
+
"loss": 5.068506469726563,
|
| 1613 |
+
"step": 11450
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.5957623167383308,
|
| 1617 |
+
"grad_norm": 1.0519545078277588,
|
| 1618 |
+
"learning_rate": 0.00012451204595011434,
|
| 1619 |
+
"loss": 5.057008666992187,
|
| 1620 |
+
"step": 11500
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.5983525876806713,
|
| 1624 |
+
"grad_norm": 1.0662899017333984,
|
| 1625 |
+
"learning_rate": 0.00012371430090942935,
|
| 1626 |
+
"loss": 5.0344482421875,
|
| 1627 |
+
"step": 11550
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.600942858623012,
|
| 1631 |
+
"grad_norm": 1.095126748085022,
|
| 1632 |
+
"learning_rate": 0.00012291655586874433,
|
| 1633 |
+
"loss": 5.038729553222656,
|
| 1634 |
+
"step": 11600
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.6035331295653525,
|
| 1638 |
+
"grad_norm": 1.0853722095489502,
|
| 1639 |
+
"learning_rate": 0.00012211881082805934,
|
| 1640 |
+
"loss": 5.02118408203125,
|
| 1641 |
+
"step": 11650
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.6061234005076931,
|
| 1645 |
+
"grad_norm": 1.063301682472229,
|
| 1646 |
+
"learning_rate": 0.00012132106578737434,
|
| 1647 |
+
"loss": 5.050283508300781,
|
| 1648 |
+
"step": 11700
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.6087136714500336,
|
| 1652 |
+
"grad_norm": 1.0829721689224243,
|
| 1653 |
+
"learning_rate": 0.00012052332074668935,
|
| 1654 |
+
"loss": 5.066478881835938,
|
| 1655 |
+
"step": 11750
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.6113039423923743,
|
| 1659 |
+
"grad_norm": 1.0675097703933716,
|
| 1660 |
+
"learning_rate": 0.00011972557570600435,
|
| 1661 |
+
"loss": 5.023760681152344,
|
| 1662 |
+
"step": 11800
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.6138942133347148,
|
| 1666 |
+
"grad_norm": 1.0228451490402222,
|
| 1667 |
+
"learning_rate": 0.00011892783066531936,
|
| 1668 |
+
"loss": 4.974055480957031,
|
| 1669 |
+
"step": 11850
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.6164844842770554,
|
| 1673 |
+
"grad_norm": 1.066720724105835,
|
| 1674 |
+
"learning_rate": 0.00011813008562463436,
|
| 1675 |
+
"loss": 5.049208068847657,
|
| 1676 |
+
"step": 11900
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.6190747552193959,
|
| 1680 |
+
"grad_norm": 1.036490559577942,
|
| 1681 |
+
"learning_rate": 0.00011733234058394937,
|
| 1682 |
+
"loss": 5.009530029296875,
|
| 1683 |
+
"step": 11950
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.6216650261617365,
|
| 1687 |
+
"grad_norm": 1.1739553213119507,
|
| 1688 |
+
"learning_rate": 0.00011653459554326435,
|
| 1689 |
+
"loss": 5.04651123046875,
|
| 1690 |
+
"step": 12000
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.6242552971040771,
|
| 1694 |
+
"grad_norm": 1.0319231748580933,
|
| 1695 |
+
"learning_rate": 0.00011573685050257937,
|
| 1696 |
+
"loss": 5.033486328125,
|
| 1697 |
+
"step": 12050
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.6268455680464177,
|
| 1701 |
+
"grad_norm": 1.0789339542388916,
|
| 1702 |
+
"learning_rate": 0.00011493910546189436,
|
| 1703 |
+
"loss": 4.994020690917969,
|
| 1704 |
+
"step": 12100
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.6294358389887582,
|
| 1708 |
+
"grad_norm": 1.076440453529358,
|
| 1709 |
+
"learning_rate": 0.00011414136042120937,
|
| 1710 |
+
"loss": 5.000180969238281,
|
| 1711 |
+
"step": 12150
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.6320261099310988,
|
| 1715 |
+
"grad_norm": 1.0861104726791382,
|
| 1716 |
+
"learning_rate": 0.00011334361538052437,
|
| 1717 |
+
"loss": 4.985296325683594,
|
| 1718 |
+
"step": 12200
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.6346163808734394,
|
| 1722 |
+
"grad_norm": 1.0445137023925781,
|
| 1723 |
+
"learning_rate": 0.00011254587033983938,
|
| 1724 |
+
"loss": 5.013554077148438,
|
| 1725 |
+
"step": 12250
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.63720665181578,
|
| 1729 |
+
"grad_norm": 1.0858595371246338,
|
| 1730 |
+
"learning_rate": 0.00011174812529915438,
|
| 1731 |
+
"loss": 4.968103637695313,
|
| 1732 |
+
"step": 12300
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.6397969227581205,
|
| 1736 |
+
"grad_norm": 1.1367961168289185,
|
| 1737 |
+
"learning_rate": 0.00011095038025846939,
|
| 1738 |
+
"loss": 5.062051391601562,
|
| 1739 |
+
"step": 12350
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.642387193700461,
|
| 1743 |
+
"grad_norm": 1.0913275480270386,
|
| 1744 |
+
"learning_rate": 0.00011015263521778439,
|
| 1745 |
+
"loss": 5.018081359863281,
|
| 1746 |
+
"step": 12400
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.6449774646428016,
|
| 1750 |
+
"grad_norm": 1.1007710695266724,
|
| 1751 |
+
"learning_rate": 0.0001093548901770994,
|
| 1752 |
+
"loss": 4.987852478027344,
|
| 1753 |
+
"step": 12450
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.6475677355851422,
|
| 1757 |
+
"grad_norm": 1.1163687705993652,
|
| 1758 |
+
"learning_rate": 0.00010855714513641439,
|
| 1759 |
+
"loss": 5.0391796875,
|
| 1760 |
+
"step": 12500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.6501580065274828,
|
| 1764 |
+
"grad_norm": 1.0992286205291748,
|
| 1765 |
+
"learning_rate": 0.0001077594000957294,
|
| 1766 |
+
"loss": 4.996618347167969,
|
| 1767 |
+
"step": 12550
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.6527482774698233,
|
| 1771 |
+
"grad_norm": 1.0719584226608276,
|
| 1772 |
+
"learning_rate": 0.0001069616550550444,
|
| 1773 |
+
"loss": 5.018509216308594,
|
| 1774 |
+
"step": 12600
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.6553385484121639,
|
| 1778 |
+
"grad_norm": 1.1296252012252808,
|
| 1779 |
+
"learning_rate": 0.00010616391001435939,
|
| 1780 |
+
"loss": 5.024125671386718,
|
| 1781 |
+
"step": 12650
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.6579288193545045,
|
| 1785 |
+
"grad_norm": 1.0779918432235718,
|
| 1786 |
+
"learning_rate": 0.0001053661649736744,
|
| 1787 |
+
"loss": 4.969640197753907,
|
| 1788 |
+
"step": 12700
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.6605190902968451,
|
| 1792 |
+
"grad_norm": 1.1184015274047852,
|
| 1793 |
+
"learning_rate": 0.0001045684199329894,
|
| 1794 |
+
"loss": 5.004659423828125,
|
| 1795 |
+
"step": 12750
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.6631093612391856,
|
| 1799 |
+
"grad_norm": 1.1083693504333496,
|
| 1800 |
+
"learning_rate": 0.00010377067489230441,
|
| 1801 |
+
"loss": 4.98875244140625,
|
| 1802 |
+
"step": 12800
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.6656996321815262,
|
| 1806 |
+
"grad_norm": 1.069274663925171,
|
| 1807 |
+
"learning_rate": 0.00010297292985161941,
|
| 1808 |
+
"loss": 4.963236083984375,
|
| 1809 |
+
"step": 12850
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.6682899031238667,
|
| 1813 |
+
"grad_norm": 1.0793840885162354,
|
| 1814 |
+
"learning_rate": 0.00010217518481093442,
|
| 1815 |
+
"loss": 4.990594177246094,
|
| 1816 |
+
"step": 12900
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.6708801740662074,
|
| 1820 |
+
"grad_norm": 1.0875816345214844,
|
| 1821 |
+
"learning_rate": 0.00010137743977024942,
|
| 1822 |
+
"loss": 4.95796875,
|
| 1823 |
+
"step": 12950
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.6734704450085479,
|
| 1827 |
+
"grad_norm": 1.0997555255889893,
|
| 1828 |
+
"learning_rate": 0.00010057969472956443,
|
| 1829 |
+
"loss": 4.989603271484375,
|
| 1830 |
+
"step": 13000
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.6760607159508885,
|
| 1834 |
+
"grad_norm": 1.0631086826324463,
|
| 1835 |
+
"learning_rate": 9.978194968887942e-05,
|
| 1836 |
+
"loss": 5.020981140136719,
|
| 1837 |
+
"step": 13050
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.678650986893229,
|
| 1841 |
+
"grad_norm": 1.0640684366226196,
|
| 1842 |
+
"learning_rate": 9.898420464819443e-05,
|
| 1843 |
+
"loss": 4.959541015625,
|
| 1844 |
+
"step": 13100
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.6812412578355695,
|
| 1848 |
+
"grad_norm": 1.052869439125061,
|
| 1849 |
+
"learning_rate": 9.818645960750942e-05,
|
| 1850 |
+
"loss": 4.984897155761718,
|
| 1851 |
+
"step": 13150
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.6838315287779102,
|
| 1855 |
+
"grad_norm": 1.08975088596344,
|
| 1856 |
+
"learning_rate": 9.738871456682444e-05,
|
| 1857 |
+
"loss": 4.944871520996093,
|
| 1858 |
+
"step": 13200
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.6864217997202507,
|
| 1862 |
+
"grad_norm": 1.2661519050598145,
|
| 1863 |
+
"learning_rate": 9.659096952613943e-05,
|
| 1864 |
+
"loss": 4.930430908203125,
|
| 1865 |
+
"step": 13250
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.6890120706625913,
|
| 1869 |
+
"grad_norm": 1.0473780632019043,
|
| 1870 |
+
"learning_rate": 9.579322448545445e-05,
|
| 1871 |
+
"loss": 5.001515808105469,
|
| 1872 |
+
"step": 13300
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.6916023416049318,
|
| 1876 |
+
"grad_norm": 1.0621232986450195,
|
| 1877 |
+
"learning_rate": 9.499547944476944e-05,
|
| 1878 |
+
"loss": 4.933506164550781,
|
| 1879 |
+
"step": 13350
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.6941926125472725,
|
| 1883 |
+
"grad_norm": 1.1702978610992432,
|
| 1884 |
+
"learning_rate": 9.419773440408445e-05,
|
| 1885 |
+
"loss": 4.983041687011719,
|
| 1886 |
+
"step": 13400
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.696782883489613,
|
| 1890 |
+
"grad_norm": 1.1768927574157715,
|
| 1891 |
+
"learning_rate": 9.339998936339945e-05,
|
| 1892 |
+
"loss": 4.953892211914063,
|
| 1893 |
+
"step": 13450
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.6993731544319536,
|
| 1897 |
+
"grad_norm": 1.1313837766647339,
|
| 1898 |
+
"learning_rate": 9.260224432271446e-05,
|
| 1899 |
+
"loss": 4.9757080078125,
|
| 1900 |
+
"step": 13500
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.7019634253742941,
|
| 1904 |
+
"grad_norm": 1.091066598892212,
|
| 1905 |
+
"learning_rate": 9.180449928202945e-05,
|
| 1906 |
+
"loss": 4.9706729125976565,
|
| 1907 |
+
"step": 13550
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.7045536963166347,
|
| 1911 |
+
"grad_norm": 1.1084681749343872,
|
| 1912 |
+
"learning_rate": 9.100675424134446e-05,
|
| 1913 |
+
"loss": 4.945950927734375,
|
| 1914 |
+
"step": 13600
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.7071439672589753,
|
| 1918 |
+
"grad_norm": 1.108149528503418,
|
| 1919 |
+
"learning_rate": 9.020900920065946e-05,
|
| 1920 |
+
"loss": 4.929463195800781,
|
| 1921 |
+
"step": 13650
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.7097342382013159,
|
| 1925 |
+
"grad_norm": 1.0872802734375,
|
| 1926 |
+
"learning_rate": 8.941126415997447e-05,
|
| 1927 |
+
"loss": 4.941557006835938,
|
| 1928 |
+
"step": 13700
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.7123245091436564,
|
| 1932 |
+
"grad_norm": 1.0591199398040771,
|
| 1933 |
+
"learning_rate": 8.861351911928947e-05,
|
| 1934 |
+
"loss": 4.920311279296875,
|
| 1935 |
+
"step": 13750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.714914780085997,
|
| 1939 |
+
"grad_norm": 1.1315951347351074,
|
| 1940 |
+
"learning_rate": 8.781577407860448e-05,
|
| 1941 |
+
"loss": 4.9243032836914065,
|
| 1942 |
+
"step": 13800
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.7175050510283376,
|
| 1946 |
+
"grad_norm": 1.136399745941162,
|
| 1947 |
+
"learning_rate": 8.701802903791947e-05,
|
| 1948 |
+
"loss": 4.938468933105469,
|
| 1949 |
+
"step": 13850
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.7200953219706782,
|
| 1953 |
+
"grad_norm": 1.0928754806518555,
|
| 1954 |
+
"learning_rate": 8.622028399723447e-05,
|
| 1955 |
+
"loss": 4.978819885253906,
|
| 1956 |
+
"step": 13900
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.7226855929130187,
|
| 1960 |
+
"grad_norm": 1.1162477731704712,
|
| 1961 |
+
"learning_rate": 8.542253895654948e-05,
|
| 1962 |
+
"loss": 4.914239807128906,
|
| 1963 |
+
"step": 13950
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.7252758638553592,
|
| 1967 |
+
"grad_norm": 1.143379807472229,
|
| 1968 |
+
"learning_rate": 8.462479391586447e-05,
|
| 1969 |
+
"loss": 4.932565612792969,
|
| 1970 |
+
"step": 14000
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.7278661347976998,
|
| 1974 |
+
"grad_norm": 1.1196132898330688,
|
| 1975 |
+
"learning_rate": 8.382704887517949e-05,
|
| 1976 |
+
"loss": 4.927554626464843,
|
| 1977 |
+
"step": 14050
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.7304564057400404,
|
| 1981 |
+
"grad_norm": 1.0910526514053345,
|
| 1982 |
+
"learning_rate": 8.302930383449448e-05,
|
| 1983 |
+
"loss": 4.92410888671875,
|
| 1984 |
+
"step": 14100
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.733046676682381,
|
| 1988 |
+
"grad_norm": 1.1446021795272827,
|
| 1989 |
+
"learning_rate": 8.223155879380949e-05,
|
| 1990 |
+
"loss": 4.9032418823242185,
|
| 1991 |
+
"step": 14150
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.7356369476247215,
|
| 1995 |
+
"grad_norm": 1.1149300336837769,
|
| 1996 |
+
"learning_rate": 8.143381375312449e-05,
|
| 1997 |
+
"loss": 4.967186584472656,
|
| 1998 |
+
"step": 14200
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.7382272185670621,
|
| 2002 |
+
"grad_norm": 1.1184216737747192,
|
| 2003 |
+
"learning_rate": 8.06360687124395e-05,
|
| 2004 |
+
"loss": 4.884440612792969,
|
| 2005 |
+
"step": 14250
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.7408174895094027,
|
| 2009 |
+
"grad_norm": 1.178585171699524,
|
| 2010 |
+
"learning_rate": 7.98383236717545e-05,
|
| 2011 |
+
"loss": 4.921098022460938,
|
| 2012 |
+
"step": 14300
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.7434077604517433,
|
| 2016 |
+
"grad_norm": 1.1591829061508179,
|
| 2017 |
+
"learning_rate": 7.904057863106951e-05,
|
| 2018 |
+
"loss": 4.918094177246093,
|
| 2019 |
+
"step": 14350
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.7459980313940838,
|
| 2023 |
+
"grad_norm": 1.1531950235366821,
|
| 2024 |
+
"learning_rate": 7.82428335903845e-05,
|
| 2025 |
+
"loss": 4.960716857910156,
|
| 2026 |
+
"step": 14400
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.7485883023364244,
|
| 2030 |
+
"grad_norm": 1.1077449321746826,
|
| 2031 |
+
"learning_rate": 7.744508854969952e-05,
|
| 2032 |
+
"loss": 4.948377990722657,
|
| 2033 |
+
"step": 14450
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.7511785732787649,
|
| 2037 |
+
"grad_norm": 1.1129505634307861,
|
| 2038 |
+
"learning_rate": 7.664734350901451e-05,
|
| 2039 |
+
"loss": 4.886638793945313,
|
| 2040 |
+
"step": 14500
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.7537688442211056,
|
| 2044 |
+
"grad_norm": 1.104398250579834,
|
| 2045 |
+
"learning_rate": 7.584959846832953e-05,
|
| 2046 |
+
"loss": 4.876330871582031,
|
| 2047 |
+
"step": 14550
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.7563591151634461,
|
| 2051 |
+
"grad_norm": 1.0603358745574951,
|
| 2052 |
+
"learning_rate": 7.505185342764451e-05,
|
| 2053 |
+
"loss": 4.879733276367188,
|
| 2054 |
+
"step": 14600
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.7589493861057867,
|
| 2058 |
+
"grad_norm": 1.1460893154144287,
|
| 2059 |
+
"learning_rate": 7.425410838695952e-05,
|
| 2060 |
+
"loss": 4.906120300292969,
|
| 2061 |
+
"step": 14650
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.7615396570481272,
|
| 2065 |
+
"grad_norm": 1.1097170114517212,
|
| 2066 |
+
"learning_rate": 7.345636334627452e-05,
|
| 2067 |
+
"loss": 4.904447021484375,
|
| 2068 |
+
"step": 14700
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.7641299279904677,
|
| 2072 |
+
"grad_norm": 1.1351251602172852,
|
| 2073 |
+
"learning_rate": 7.265861830558953e-05,
|
| 2074 |
+
"loss": 4.891495056152344,
|
| 2075 |
+
"step": 14750
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.7667201989328084,
|
| 2079 |
+
"grad_norm": 1.1649805307388306,
|
| 2080 |
+
"learning_rate": 7.186087326490453e-05,
|
| 2081 |
+
"loss": 4.919297485351563,
|
| 2082 |
+
"step": 14800
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.769310469875149,
|
| 2086 |
+
"grad_norm": 1.1074897050857544,
|
| 2087 |
+
"learning_rate": 7.106312822421953e-05,
|
| 2088 |
+
"loss": 4.88187255859375,
|
| 2089 |
+
"step": 14850
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.7719007408174895,
|
| 2093 |
+
"grad_norm": 1.1506906747817993,
|
| 2094 |
+
"learning_rate": 7.026538318353454e-05,
|
| 2095 |
+
"loss": 4.901511840820312,
|
| 2096 |
+
"step": 14900
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.77449101175983,
|
| 2100 |
+
"grad_norm": 1.1153931617736816,
|
| 2101 |
+
"learning_rate": 6.946763814284953e-05,
|
| 2102 |
+
"loss": 4.864546203613282,
|
| 2103 |
+
"step": 14950
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.7770812827021707,
|
| 2107 |
+
"grad_norm": 1.2132166624069214,
|
| 2108 |
+
"learning_rate": 6.866989310216455e-05,
|
| 2109 |
+
"loss": 4.8552725219726565,
|
| 2110 |
+
"step": 15000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.7796715536445112,
|
| 2114 |
+
"grad_norm": 1.273779273033142,
|
| 2115 |
+
"learning_rate": 6.787214806147954e-05,
|
| 2116 |
+
"loss": 4.8751153564453125,
|
| 2117 |
+
"step": 15050
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.7822618245868518,
|
| 2121 |
+
"grad_norm": 1.1260673999786377,
|
| 2122 |
+
"learning_rate": 6.707440302079455e-05,
|
| 2123 |
+
"loss": 4.880099792480468,
|
| 2124 |
+
"step": 15100
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.7848520955291923,
|
| 2128 |
+
"grad_norm": 1.194525122642517,
|
| 2129 |
+
"learning_rate": 6.627665798010955e-05,
|
| 2130 |
+
"loss": 4.861859130859375,
|
| 2131 |
+
"step": 15150
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.7874423664715329,
|
| 2135 |
+
"grad_norm": 1.1089609861373901,
|
| 2136 |
+
"learning_rate": 6.547891293942455e-05,
|
| 2137 |
+
"loss": 4.869077453613281,
|
| 2138 |
+
"step": 15200
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.7900326374138735,
|
| 2142 |
+
"grad_norm": 1.194669246673584,
|
| 2143 |
+
"learning_rate": 6.468116789873956e-05,
|
| 2144 |
+
"loss": 4.892135009765625,
|
| 2145 |
+
"step": 15250
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.7926229083562141,
|
| 2149 |
+
"grad_norm": 1.1765038967132568,
|
| 2150 |
+
"learning_rate": 6.388342285805456e-05,
|
| 2151 |
+
"loss": 4.886556701660156,
|
| 2152 |
+
"step": 15300
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.7952131792985546,
|
| 2156 |
+
"grad_norm": 1.1625683307647705,
|
| 2157 |
+
"learning_rate": 6.308567781736957e-05,
|
| 2158 |
+
"loss": 4.85548583984375,
|
| 2159 |
+
"step": 15350
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.7978034502408952,
|
| 2163 |
+
"grad_norm": 1.2013742923736572,
|
| 2164 |
+
"learning_rate": 6.228793277668457e-05,
|
| 2165 |
+
"loss": 4.857999572753906,
|
| 2166 |
+
"step": 15400
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.8003937211832358,
|
| 2170 |
+
"grad_norm": 1.147049903869629,
|
| 2171 |
+
"learning_rate": 6.149018773599957e-05,
|
| 2172 |
+
"loss": 4.8666470336914065,
|
| 2173 |
+
"step": 15450
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.8029839921255764,
|
| 2177 |
+
"grad_norm": 1.1380870342254639,
|
| 2178 |
+
"learning_rate": 6.069244269531458e-05,
|
| 2179 |
+
"loss": 4.868302307128906,
|
| 2180 |
+
"step": 15500
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.8055742630679169,
|
| 2184 |
+
"grad_norm": 1.1927944421768188,
|
| 2185 |
+
"learning_rate": 5.9894697654629575e-05,
|
| 2186 |
+
"loss": 4.852730407714843,
|
| 2187 |
+
"step": 15550
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.8081645340102575,
|
| 2191 |
+
"grad_norm": 1.1827374696731567,
|
| 2192 |
+
"learning_rate": 5.909695261394458e-05,
|
| 2193 |
+
"loss": 4.834335632324219,
|
| 2194 |
+
"step": 15600
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.810754804952598,
|
| 2198 |
+
"grad_norm": 1.1386221647262573,
|
| 2199 |
+
"learning_rate": 5.8299207573259584e-05,
|
| 2200 |
+
"loss": 4.80497802734375,
|
| 2201 |
+
"step": 15650
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.8133450758949387,
|
| 2205 |
+
"grad_norm": 1.1220753192901611,
|
| 2206 |
+
"learning_rate": 5.750146253257459e-05,
|
| 2207 |
+
"loss": 4.840519409179688,
|
| 2208 |
+
"step": 15700
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.8159353468372792,
|
| 2212 |
+
"grad_norm": 1.178306221961975,
|
| 2213 |
+
"learning_rate": 5.670371749188958e-05,
|
| 2214 |
+
"loss": 4.811608581542969,
|
| 2215 |
+
"step": 15750
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.8185256177796197,
|
| 2219 |
+
"grad_norm": 1.108260989189148,
|
| 2220 |
+
"learning_rate": 5.5905972451204585e-05,
|
| 2221 |
+
"loss": 4.848948974609375,
|
| 2222 |
+
"step": 15800
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.8211158887219603,
|
| 2226 |
+
"grad_norm": 1.1373419761657715,
|
| 2227 |
+
"learning_rate": 5.510822741051959e-05,
|
| 2228 |
+
"loss": 4.879228210449218,
|
| 2229 |
+
"step": 15850
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.8237061596643009,
|
| 2233 |
+
"grad_norm": 1.1401219367980957,
|
| 2234 |
+
"learning_rate": 5.4310482369834594e-05,
|
| 2235 |
+
"loss": 4.843096923828125,
|
| 2236 |
+
"step": 15900
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.8262964306066415,
|
| 2240 |
+
"grad_norm": 1.2324658632278442,
|
| 2241 |
+
"learning_rate": 5.35127373291496e-05,
|
| 2242 |
+
"loss": 4.8840853881835935,
|
| 2243 |
+
"step": 15950
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.828886701548982,
|
| 2247 |
+
"grad_norm": 1.123178243637085,
|
| 2248 |
+
"learning_rate": 5.27149922884646e-05,
|
| 2249 |
+
"loss": 4.8826901245117185,
|
| 2250 |
+
"step": 16000
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.8314769724913226,
|
| 2254 |
+
"grad_norm": 1.1920490264892578,
|
| 2255 |
+
"learning_rate": 5.19172472477796e-05,
|
| 2256 |
+
"loss": 4.831457214355469,
|
| 2257 |
+
"step": 16050
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.8340672434336631,
|
| 2261 |
+
"grad_norm": 1.1483428478240967,
|
| 2262 |
+
"learning_rate": 5.1119502207094605e-05,
|
| 2263 |
+
"loss": 4.812444152832032,
|
| 2264 |
+
"step": 16100
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.8366575143760038,
|
| 2268 |
+
"grad_norm": 1.110986590385437,
|
| 2269 |
+
"learning_rate": 5.032175716640961e-05,
|
| 2270 |
+
"loss": 4.907426452636718,
|
| 2271 |
+
"step": 16150
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.8392477853183443,
|
| 2275 |
+
"grad_norm": 1.1689667701721191,
|
| 2276 |
+
"learning_rate": 4.9524012125724614e-05,
|
| 2277 |
+
"loss": 4.833683166503906,
|
| 2278 |
+
"step": 16200
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.8418380562606849,
|
| 2282 |
+
"grad_norm": 1.2110393047332764,
|
| 2283 |
+
"learning_rate": 4.872626708503962e-05,
|
| 2284 |
+
"loss": 4.851173706054688,
|
| 2285 |
+
"step": 16250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.8444283272030254,
|
| 2289 |
+
"grad_norm": 1.1109424829483032,
|
| 2290 |
+
"learning_rate": 4.7928522044354616e-05,
|
| 2291 |
+
"loss": 4.8200289916992185,
|
| 2292 |
+
"step": 16300
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.847018598145366,
|
| 2296 |
+
"grad_norm": 1.2129008769989014,
|
| 2297 |
+
"learning_rate": 4.713077700366962e-05,
|
| 2298 |
+
"loss": 4.797307434082032,
|
| 2299 |
+
"step": 16350
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.8496088690877066,
|
| 2303 |
+
"grad_norm": 1.1530872583389282,
|
| 2304 |
+
"learning_rate": 4.6333031962984626e-05,
|
| 2305 |
+
"loss": 4.856131896972657,
|
| 2306 |
+
"step": 16400
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.8521991400300472,
|
| 2310 |
+
"grad_norm": 1.3170881271362305,
|
| 2311 |
+
"learning_rate": 4.553528692229963e-05,
|
| 2312 |
+
"loss": 4.85386474609375,
|
| 2313 |
+
"step": 16450
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.8547894109723877,
|
| 2317 |
+
"grad_norm": 1.1147844791412354,
|
| 2318 |
+
"learning_rate": 4.4737541881614635e-05,
|
| 2319 |
+
"loss": 4.820576171875,
|
| 2320 |
+
"step": 16500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.8573796819147282,
|
| 2324 |
+
"grad_norm": 1.1671563386917114,
|
| 2325 |
+
"learning_rate": 4.393979684092964e-05,
|
| 2326 |
+
"loss": 4.826213684082031,
|
| 2327 |
+
"step": 16550
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.8599699528570689,
|
| 2331 |
+
"grad_norm": 1.1505951881408691,
|
| 2332 |
+
"learning_rate": 4.314205180024464e-05,
|
| 2333 |
+
"loss": 4.807721557617188,
|
| 2334 |
+
"step": 16600
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.8625602237994094,
|
| 2338 |
+
"grad_norm": 1.2463537454605103,
|
| 2339 |
+
"learning_rate": 4.234430675955964e-05,
|
| 2340 |
+
"loss": 4.844184875488281,
|
| 2341 |
+
"step": 16650
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.86515049474175,
|
| 2345 |
+
"grad_norm": 1.2044734954833984,
|
| 2346 |
+
"learning_rate": 4.1546561718874646e-05,
|
| 2347 |
+
"loss": 4.8523867797851565,
|
| 2348 |
+
"step": 16700
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.8677407656840905,
|
| 2352 |
+
"grad_norm": 1.2013986110687256,
|
| 2353 |
+
"learning_rate": 4.074881667818965e-05,
|
| 2354 |
+
"loss": 4.796150817871093,
|
| 2355 |
+
"step": 16750
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.8703310366264311,
|
| 2359 |
+
"grad_norm": 1.0996414422988892,
|
| 2360 |
+
"learning_rate": 3.9951071637504655e-05,
|
| 2361 |
+
"loss": 4.80515625,
|
| 2362 |
+
"step": 16800
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.8729213075687717,
|
| 2366 |
+
"grad_norm": 1.1789201498031616,
|
| 2367 |
+
"learning_rate": 3.915332659681965e-05,
|
| 2368 |
+
"loss": 4.838421936035156,
|
| 2369 |
+
"step": 16850
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.8755115785111123,
|
| 2373 |
+
"grad_norm": 1.1779522895812988,
|
| 2374 |
+
"learning_rate": 3.835558155613466e-05,
|
| 2375 |
+
"loss": 4.7971868896484375,
|
| 2376 |
+
"step": 16900
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.8781018494534528,
|
| 2380 |
+
"grad_norm": 1.2039135694503784,
|
| 2381 |
+
"learning_rate": 3.755783651544966e-05,
|
| 2382 |
+
"loss": 4.779116821289063,
|
| 2383 |
+
"step": 16950
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.8806921203957934,
|
| 2387 |
+
"grad_norm": 1.195914626121521,
|
| 2388 |
+
"learning_rate": 3.676009147476466e-05,
|
| 2389 |
+
"loss": 4.842753601074219,
|
| 2390 |
+
"step": 17000
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.883282391338134,
|
| 2394 |
+
"grad_norm": 1.2158962488174438,
|
| 2395 |
+
"learning_rate": 3.5962346434079664e-05,
|
| 2396 |
+
"loss": 4.767064514160157,
|
| 2397 |
+
"step": 17050
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.8858726622804746,
|
| 2401 |
+
"grad_norm": 1.1929750442504883,
|
| 2402 |
+
"learning_rate": 3.516460139339467e-05,
|
| 2403 |
+
"loss": 4.835277404785156,
|
| 2404 |
+
"step": 17100
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.8884629332228151,
|
| 2408 |
+
"grad_norm": 1.1390390396118164,
|
| 2409 |
+
"learning_rate": 3.4366856352709673e-05,
|
| 2410 |
+
"loss": 4.834110107421875,
|
| 2411 |
+
"step": 17150
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.8910532041651557,
|
| 2415 |
+
"grad_norm": 1.2707346677780151,
|
| 2416 |
+
"learning_rate": 3.356911131202468e-05,
|
| 2417 |
+
"loss": 4.800746765136719,
|
| 2418 |
+
"step": 17200
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.8936434751074962,
|
| 2422 |
+
"grad_norm": 1.2819527387619019,
|
| 2423 |
+
"learning_rate": 3.2771366271339676e-05,
|
| 2424 |
+
"loss": 4.803517456054688,
|
| 2425 |
+
"step": 17250
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.8962337460498369,
|
| 2429 |
+
"grad_norm": 1.130441427230835,
|
| 2430 |
+
"learning_rate": 3.197362123065468e-05,
|
| 2431 |
+
"loss": 4.809993591308594,
|
| 2432 |
+
"step": 17300
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.8988240169921774,
|
| 2436 |
+
"grad_norm": 1.2011135816574097,
|
| 2437 |
+
"learning_rate": 3.117587618996968e-05,
|
| 2438 |
+
"loss": 4.835908813476562,
|
| 2439 |
+
"step": 17350
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.9014142879345179,
|
| 2443 |
+
"grad_norm": 1.1804677248001099,
|
| 2444 |
+
"learning_rate": 3.0378131149284686e-05,
|
| 2445 |
+
"loss": 4.820344543457031,
|
| 2446 |
+
"step": 17400
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.9040045588768585,
|
| 2450 |
+
"grad_norm": 1.1942323446273804,
|
| 2451 |
+
"learning_rate": 2.9580386108599687e-05,
|
| 2452 |
+
"loss": 4.803614196777343,
|
| 2453 |
+
"step": 17450
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.906594829819199,
|
| 2457 |
+
"grad_norm": 1.215887188911438,
|
| 2458 |
+
"learning_rate": 2.8782641067914692e-05,
|
| 2459 |
+
"loss": 4.784419860839844,
|
| 2460 |
+
"step": 17500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.9091851007615397,
|
| 2464 |
+
"grad_norm": 1.1484390497207642,
|
| 2465 |
+
"learning_rate": 2.7984896027229693e-05,
|
| 2466 |
+
"loss": 4.850234375,
|
| 2467 |
+
"step": 17550
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.9117753717038802,
|
| 2471 |
+
"grad_norm": 1.16694176197052,
|
| 2472 |
+
"learning_rate": 2.7187150986544697e-05,
|
| 2473 |
+
"loss": 4.777885437011719,
|
| 2474 |
+
"step": 17600
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.9143656426462208,
|
| 2478 |
+
"grad_norm": 1.17305588722229,
|
| 2479 |
+
"learning_rate": 2.6389405945859702e-05,
|
| 2480 |
+
"loss": 4.832542114257812,
|
| 2481 |
+
"step": 17650
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.9169559135885613,
|
| 2485 |
+
"grad_norm": 1.2073004245758057,
|
| 2486 |
+
"learning_rate": 2.5591660905174703e-05,
|
| 2487 |
+
"loss": 4.793171081542969,
|
| 2488 |
+
"step": 17700
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.919546184530902,
|
| 2492 |
+
"grad_norm": 1.1849514245986938,
|
| 2493 |
+
"learning_rate": 2.4793915864489708e-05,
|
| 2494 |
+
"loss": 4.813887634277344,
|
| 2495 |
+
"step": 17750
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.9221364554732425,
|
| 2499 |
+
"grad_norm": 1.1501691341400146,
|
| 2500 |
+
"learning_rate": 2.3996170823804712e-05,
|
| 2501 |
+
"loss": 4.818171081542968,
|
| 2502 |
+
"step": 17800
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.9247267264155831,
|
| 2506 |
+
"grad_norm": 1.1892966032028198,
|
| 2507 |
+
"learning_rate": 2.3198425783119713e-05,
|
| 2508 |
+
"loss": 4.815703125,
|
| 2509 |
+
"step": 17850
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.9273169973579236,
|
| 2513 |
+
"grad_norm": 1.225103735923767,
|
| 2514 |
+
"learning_rate": 2.2400680742434718e-05,
|
| 2515 |
+
"loss": 4.793248901367187,
|
| 2516 |
+
"step": 17900
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.9299072683002642,
|
| 2520 |
+
"grad_norm": 1.1672247648239136,
|
| 2521 |
+
"learning_rate": 2.1602935701749716e-05,
|
| 2522 |
+
"loss": 4.814731750488281,
|
| 2523 |
+
"step": 17950
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.9324975392426048,
|
| 2527 |
+
"grad_norm": 1.1436740159988403,
|
| 2528 |
+
"learning_rate": 2.080519066106472e-05,
|
| 2529 |
+
"loss": 4.764231262207031,
|
| 2530 |
+
"step": 18000
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.9350878101849454,
|
| 2534 |
+
"grad_norm": 1.1047722101211548,
|
| 2535 |
+
"learning_rate": 2.0007445620379725e-05,
|
| 2536 |
+
"loss": 4.796049499511719,
|
| 2537 |
+
"step": 18050
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.9376780811272859,
|
| 2541 |
+
"grad_norm": 1.1942051649093628,
|
| 2542 |
+
"learning_rate": 1.9209700579694726e-05,
|
| 2543 |
+
"loss": 4.743634033203125,
|
| 2544 |
+
"step": 18100
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 0.9402683520696264,
|
| 2548 |
+
"grad_norm": 1.1561431884765625,
|
| 2549 |
+
"learning_rate": 1.841195553900973e-05,
|
| 2550 |
+
"loss": 4.780207214355468,
|
| 2551 |
+
"step": 18150
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 0.9428586230119671,
|
| 2555 |
+
"grad_norm": 1.1528445482254028,
|
| 2556 |
+
"learning_rate": 1.7614210498324735e-05,
|
| 2557 |
+
"loss": 4.7709219360351565,
|
| 2558 |
+
"step": 18200
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 0.9454488939543076,
|
| 2562 |
+
"grad_norm": 1.1248111724853516,
|
| 2563 |
+
"learning_rate": 1.6816465457639736e-05,
|
| 2564 |
+
"loss": 4.76812744140625,
|
| 2565 |
+
"step": 18250
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 0.9480391648966482,
|
| 2569 |
+
"grad_norm": 1.2417064905166626,
|
| 2570 |
+
"learning_rate": 1.601872041695474e-05,
|
| 2571 |
+
"loss": 4.804192199707031,
|
| 2572 |
+
"step": 18300
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 0.9506294358389887,
|
| 2576 |
+
"grad_norm": 1.1914931535720825,
|
| 2577 |
+
"learning_rate": 1.5220975376269744e-05,
|
| 2578 |
+
"loss": 4.7780892944335935,
|
| 2579 |
+
"step": 18350
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.9532197067813293,
|
| 2583 |
+
"grad_norm": 1.1965242624282837,
|
| 2584 |
+
"learning_rate": 1.4423230335584745e-05,
|
| 2585 |
+
"loss": 4.782641296386719,
|
| 2586 |
+
"step": 18400
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 0.9558099777236699,
|
| 2590 |
+
"grad_norm": 1.2170450687408447,
|
| 2591 |
+
"learning_rate": 1.3625485294899748e-05,
|
| 2592 |
+
"loss": 4.839132690429688,
|
| 2593 |
+
"step": 18450
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.9584002486660105,
|
| 2597 |
+
"grad_norm": 1.212841272354126,
|
| 2598 |
+
"learning_rate": 1.2827740254214752e-05,
|
| 2599 |
+
"loss": 4.804522399902344,
|
| 2600 |
+
"step": 18500
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 0.960990519608351,
|
| 2604 |
+
"grad_norm": 1.2126749753952026,
|
| 2605 |
+
"learning_rate": 1.2029995213529755e-05,
|
| 2606 |
+
"loss": 4.806041870117188,
|
| 2607 |
+
"step": 18550
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.9635807905506916,
|
| 2611 |
+
"grad_norm": 1.1030791997909546,
|
| 2612 |
+
"learning_rate": 1.1232250172844758e-05,
|
| 2613 |
+
"loss": 4.794331359863281,
|
| 2614 |
+
"step": 18600
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 0.9661710614930322,
|
| 2618 |
+
"grad_norm": 1.1634585857391357,
|
| 2619 |
+
"learning_rate": 1.043450513215976e-05,
|
| 2620 |
+
"loss": 4.783181762695312,
|
| 2621 |
+
"step": 18650
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 0.9687613324353728,
|
| 2625 |
+
"grad_norm": 1.1583950519561768,
|
| 2626 |
+
"learning_rate": 9.636760091474764e-06,
|
| 2627 |
+
"loss": 4.768502197265625,
|
| 2628 |
+
"step": 18700
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 0.9713516033777133,
|
| 2632 |
+
"grad_norm": 1.264675498008728,
|
| 2633 |
+
"learning_rate": 8.839015050789766e-06,
|
| 2634 |
+
"loss": 4.739063110351562,
|
| 2635 |
+
"step": 18750
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 0.9739418743200539,
|
| 2639 |
+
"grad_norm": 1.1584491729736328,
|
| 2640 |
+
"learning_rate": 8.04127001010477e-06,
|
| 2641 |
+
"loss": 4.728269348144531,
|
| 2642 |
+
"step": 18800
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 0.9765321452623944,
|
| 2646 |
+
"grad_norm": 1.1771270036697388,
|
| 2647 |
+
"learning_rate": 7.243524969419773e-06,
|
| 2648 |
+
"loss": 4.793826293945313,
|
| 2649 |
+
"step": 18850
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 0.9791224162047351,
|
| 2653 |
+
"grad_norm": 1.230299711227417,
|
| 2654 |
+
"learning_rate": 6.445779928734776e-06,
|
| 2655 |
+
"loss": 4.7584912109375,
|
| 2656 |
+
"step": 18900
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.9817126871470756,
|
| 2660 |
+
"grad_norm": 1.1764832735061646,
|
| 2661 |
+
"learning_rate": 5.648034888049779e-06,
|
| 2662 |
+
"loss": 4.755926818847656,
|
| 2663 |
+
"step": 18950
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 0.9843029580894161,
|
| 2667 |
+
"grad_norm": 1.1985116004943848,
|
| 2668 |
+
"learning_rate": 4.850289847364782e-06,
|
| 2669 |
+
"loss": 4.724908142089844,
|
| 2670 |
+
"step": 19000
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 0.9868932290317567,
|
| 2674 |
+
"grad_norm": 1.1627134084701538,
|
| 2675 |
+
"learning_rate": 4.0525448066797845e-06,
|
| 2676 |
+
"loss": 4.764407043457031,
|
| 2677 |
+
"step": 19050
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 0.9894834999740972,
|
| 2681 |
+
"grad_norm": 1.2293567657470703,
|
| 2682 |
+
"learning_rate": 3.2547997659947877e-06,
|
| 2683 |
+
"loss": 4.771585693359375,
|
| 2684 |
+
"step": 19100
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 0.9920737709164379,
|
| 2688 |
+
"grad_norm": 1.1796555519104004,
|
| 2689 |
+
"learning_rate": 2.4570547253097906e-06,
|
| 2690 |
+
"loss": 4.763060302734375,
|
| 2691 |
+
"step": 19150
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 0.9946640418587784,
|
| 2695 |
+
"grad_norm": 1.1778082847595215,
|
| 2696 |
+
"learning_rate": 1.6593096846247936e-06,
|
| 2697 |
+
"loss": 4.758525085449219,
|
| 2698 |
+
"step": 19200
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 0.997254312801119,
|
| 2702 |
+
"grad_norm": 1.1932587623596191,
|
| 2703 |
+
"learning_rate": 8.615646439397968e-07,
|
| 2704 |
+
"loss": 4.812500915527344,
|
| 2705 |
+
"step": 19250
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 0.9998445837434595,
|
| 2709 |
+
"grad_norm": 1.1879589557647705,
|
| 2710 |
+
"learning_rate": 6.381960325479977e-08,
|
| 2711 |
+
"loss": 4.800540161132813,
|
| 2712 |
+
"step": 19300
|
| 2713 |
+
}
|
| 2714 |
+
],
|
| 2715 |
+
"logging_steps": 50,
|
| 2716 |
+
"max_steps": 19303,
|
| 2717 |
+
"num_input_tokens_seen": 0,
|
| 2718 |
+
"num_train_epochs": 1,
|
| 2719 |
+
"save_steps": 500,
|
| 2720 |
+
"stateful_callbacks": {
|
| 2721 |
+
"TrainerControl": {
|
| 2722 |
+
"args": {
|
| 2723 |
+
"should_epoch_stop": false,
|
| 2724 |
+
"should_evaluate": false,
|
| 2725 |
+
"should_log": false,
|
| 2726 |
+
"should_save": true,
|
| 2727 |
+
"should_training_stop": true
|
| 2728 |
+
},
|
| 2729 |
+
"attributes": {}
|
| 2730 |
+
}
|
| 2731 |
+
},
|
| 2732 |
+
"total_flos": 1.401424037216256e+16,
|
| 2733 |
+
"train_batch_size": 2,
|
| 2734 |
+
"trial_name": null,
|
| 2735 |
+
"trial_params": null
|
| 2736 |
+
}
|
checkpoint-19303/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e8afc1c0b06a60d7ee2d75f19dce13ae9555a437de8188f2c1b6a1f58a65c856
|
| 3 |
+
size 5201
|
checkpoint-41387/config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"add_cross_attention": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"embd_pdrop": 0.1,
|
| 11 |
+
"eos_token_id": 50256,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"layer_norm_epsilon": 1e-05,
|
| 14 |
+
"model_type": "gpt2",
|
| 15 |
+
"n_ctx": 512,
|
| 16 |
+
"n_embd": 768,
|
| 17 |
+
"n_head": 12,
|
| 18 |
+
"n_inner": null,
|
| 19 |
+
"n_layer": 12,
|
| 20 |
+
"n_positions": 512,
|
| 21 |
+
"pad_token_id": 50256,
|
| 22 |
+
"reorder_and_upcast_attn": false,
|
| 23 |
+
"resid_pdrop": 0.1,
|
| 24 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 25 |
+
"scale_attn_weights": true,
|
| 26 |
+
"summary_activation": null,
|
| 27 |
+
"summary_first_dropout": 0.1,
|
| 28 |
+
"summary_proj_to_labels": true,
|
| 29 |
+
"summary_type": "cls_index",
|
| 30 |
+
"summary_use_proj": true,
|
| 31 |
+
"tie_word_embeddings": true,
|
| 32 |
+
"transformers_version": "5.12.1",
|
| 33 |
+
"use_cache": false,
|
| 34 |
+
"vocab_size": 50257
|
| 35 |
+
}
|
checkpoint-41387/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"output_attentions": false,
|
| 6 |
+
"output_hidden_states": false,
|
| 7 |
+
"transformers_version": "5.12.1",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|
checkpoint-41387/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bfb01924edc39e9b251cdeeddf0ca4e46b5694b3887475c58764d3800d990a97
|
| 3 |
+
size 496201344
|
checkpoint-41387/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb1c92d3f953c44397d88aa767c4ee8182bedb9b434ee29395e8673de455b1b6
|
| 3 |
+
size 992499083
|
checkpoint-41387/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:13bed6601325b27039db676acc7e73f5b329a90683a6c41a8fb7f375bd40784f
|
| 3 |
+
size 14645
|
checkpoint-41387/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e262a33f24fc97140ad37f10e9cd135813189769b63f6c658bec08702b24255
|
| 3 |
+
size 1383
|
checkpoint-41387/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:75fd4e3a2ab7426471542b9e73a6ad9aab6cf10535aadf8e3868febeda754bd5
|
| 3 |
+
size 1465
|
checkpoint-41387/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-41387/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e8afc1c0b06a60d7ee2d75f19dce13ae9555a437de8188f2c1b6a1f58a65c856
|
| 3 |
+
size 5201
|
config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"add_cross_attention": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"embd_pdrop": 0.1,
|
| 11 |
+
"eos_token_id": 50256,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"layer_norm_epsilon": 1e-05,
|
| 14 |
+
"model_type": "gpt2",
|
| 15 |
+
"n_ctx": 512,
|
| 16 |
+
"n_embd": 768,
|
| 17 |
+
"n_head": 12,
|
| 18 |
+
"n_inner": null,
|
| 19 |
+
"n_layer": 12,
|
| 20 |
+
"n_positions": 512,
|
| 21 |
+
"pad_token_id": 50256,
|
| 22 |
+
"reorder_and_upcast_attn": false,
|
| 23 |
+
"resid_pdrop": 0.1,
|
| 24 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 25 |
+
"scale_attn_weights": true,
|
| 26 |
+
"summary_activation": null,
|
| 27 |
+
"summary_first_dropout": 0.1,
|
| 28 |
+
"summary_proj_to_labels": true,
|
| 29 |
+
"summary_type": "cls_index",
|
| 30 |
+
"summary_use_proj": true,
|
| 31 |
+
"tie_word_embeddings": true,
|
| 32 |
+
"transformers_version": "5.12.1",
|
| 33 |
+
"use_cache": false,
|
| 34 |
+
"vocab_size": 50257
|
| 35 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"output_attentions": false,
|
| 6 |
+
"output_hidden_states": false,
|
| 7 |
+
"transformers_version": "5.12.1",
|
| 8 |
+
"use_cache": true
|
| 9 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bfb01924edc39e9b251cdeeddf0ca4e46b5694b3887475c58764d3800d990a97
|
| 3 |
+
size 496201344
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|endoftext|>",
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"model_max_length": 1000000000,
|
| 10 |
+
"pad_token": "<|endoftext|>",
|
| 11 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 12 |
+
"unk_token": "<|endoftext|>"
|
| 13 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e8afc1c0b06a60d7ee2d75f19dce13ae9555a437de8188f2c1b6a1f58a65c856
|
| 3 |
+
size 5201
|