Add 11 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,102 @@
+---
+base_model: meta-llama/Llama-4-Scout-17B-16E-Instruct
+library_name: peft
+license: other
+tags:
+  - lora
+  - peft
+  - adapter
+  - adaption
+---
+
+# adaption_python_code_snippets
+
+## Model Training
+
+A LORA adapter for `meta-llama/Llama-4-Scout-17B-16E-Instruct`. This model was trained with SFT using [Adaption](https://adaptionlabs.ai)'s AutoScientist on the CodingAdapt dataset.
+
+
+![Training metrics](training-metrics.png)
+
+### AutoScientist Config
+
+```json
+{
+  "job_id": "da4497a8-ccc4-4355-afca-4cee7f590b3c",
+  "training_experiment_id": "e45b4570-5dcf-4bf4-8151-d988fe4bb18d",
+  "original_model_name": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
+  "trained_model_name": "adaption_python_code_snippets",
+  "training_method": "sft",
+  "training_type": "lora",
+  "data_format": "chat",
+  "hyperparams": {
+    "lora": "true",
+    "lora_r": 64,
+    "n_evals": 5,
+    "n_epochs": 4,
+    "batch_size": "max",
+    "lora_alpha": 128,
+    "lora_dropout": 0,
+    "min_lr_ratio": 0.1,
+    "warmup_ratio": 0.05,
+    "weight_decay": 0.02,
+    "learning_rate": 0.0001,
+    "max_grad_norm": 1,
+    "base_model_size": "109B",
+    "train_on_inputs": "false",
+    "training_method": "sft",
+    "lr_scheduler_type": "cosine",
+    "scheduler_num_cycles": 0.5,
+    "lora_trainable_modules": "k_proj,o_proj,q_proj,v_proj,shared_expert.gate_proj,shared_expert.up_proj,shared_expert.down_proj,feed_forward.gate_proj,feed_forward.up_proj,feed_forward.down_proj"
+  }
+}
+```
+
+## Training Data
+
+The model was trained on 1,613 rows of adapted data with the following domain distribution: code (100%).
+
+## Model Evaluation
+
+The model was evaluated on an in-distribution held-out test set as well as a broader domain-specific test set to measure generalization.
+
+
+![Win rates](win-rates.png)
+
+| Domain | Win rate vs. base model |
+| --- | --- |
+| code | 67% |
+
+## How to use
+
+```bash
+pip install torch transformers peft
+```
+
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from peft import PeftModel
+
+BASE = "meta-llama/Llama-4-Scout-17B-16E-Instruct"
+ADAPTER = "<this-repo-id>"
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+dtype = torch.float32 if device == "cpu" else torch.bfloat16
+
+base = AutoModelForCausalLM.from_pretrained(BASE, dtype=dtype).to(device)
+model = PeftModel.from_pretrained(base, ADAPTER)
+# Optional: merge the LoRA weights into the base for faster inference
+model = model.merge_and_unload()
+model.eval()
+
+tokenizer = AutoTokenizer.from_pretrained(BASE)
+messages = [{"role": "user", "content": "Hello!"}]
+text = tokenizer.apply_chat_template(
+    messages, tokenize=False, add_generation_prompt=True)
+inputs = tokenizer(text, return_tensors="pt").to(device)
+
+with torch.inference_mode():
+    out = model.generate(**inputs, max_new_tokens=512)
+print(tokenizer.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))
+```

adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "togethercomputer/Llama-4-Scout-17B-16E-Instruct_bnb_4bit",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": [],
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "q_proj",
+    "feed_forward.gate_proj",
+    "shared_expert.down_proj",
+    "shared_expert.gate_proj",
+    "k_proj",
+    "feed_forward.up_proj",
+    "shared_expert.up_proj",
+    "feed_forward.down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24a8d8aac3be78af13566f28f1fc1ef38335b24ec5e38a06649640fa8d788880
+size 893484120

chat_template.jinja

@@ -0,0 +1,123 @@
+{{- bos_token }}
+{%- if custom_tools is defined %}
+    {%- set tools = custom_tools %}
+{%- endif %}
+{%- if not tools_in_user_message is defined %}
+    {%- set tools_in_user_message = true %}
+{%- endif %}
+{%- if not date_string is defined %}
+    {%- if strftime_now is defined %}
+        {%- set date_string = strftime_now("%d %b %Y") %}
+    {%- else %}
+        {%- set date_string = "26 Jul 2024" %}
+    {%- endif %}
+{%- endif %}
+{%- if not tools is defined %}
+    {%- set tools = none %}
+{%- endif %}
+
+{#- This block extracts the system message, so we can slot it into the right place. #}
+{%- if messages[0]['role'] == 'system' %}    
+    {%- if messages[0]['content'] is string %}
+        {%- set system_message = messages[0]['content']|trim %}
+    {%- else %}
+        {#- FIXME: The processor requires an array, always. #}
+        {%- set system_message = messages[0]['content'][0]['text']|trim %}
+    {%- endif %}
+    {%- set messages = messages[1:] %}
+    {%- set user_supplied_system_message = true %}
+{%- else %}
+    {%- set system_message = "" %}
+    {%- set user_supplied_system_message = false %}
+{%- endif %}
+
+{#- System message if the user supplied one #}
+{%- if user_supplied_system_message %}
+    {{- "<|header_start|>system<|header_end|>\n\n" }}
+    {%- if tools is not none %}
+        {{- "Environment: ipython\n" }}
+    {%- endif %}
+    {%- if tools is not none and not tools_in_user_message %}
+        {{- "You have access to the following functions. To call a function, please respond with JSON for a function call." }}
+        {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }}
+        {{- "Do not use variables.\n\n" }}
+        {%- for t in tools %}
+            {{- t | tojson(indent=4) }}
+            {{- "\n\n" }}
+        {%- endfor %}
+    {%- endif %}
+    {{- system_message }}
+    {{- "<|eot|>" }}
+{%- endif %}
+
+{#- Custom tools are passed in a user message with some extra guidance #}
+{%- if tools_in_user_message and not tools is none %}
+    {#- Extract the first user message so we can plug it in here #}
+    {%- if messages | length != 0 %}
+        {%- set first_user_message = messages[0]['content']|trim %}
+        {%- set messages = messages[1:] %}
+    {%- else %}
+        {{- raise_exception("Cannot put tools in the first user message when there's no first user message!") }}
+{%- endif %}
+    {{- '<|header_start|>user<|header_end|>\n\n' -}}
+    {{- "Given the following functions, please respond with a JSON for a function call " }}
+    {{- "with its proper arguments that best answers the given prompt.\n\n" }}
+    {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }}
+    {{- "Do not use variables.\n\n" }}
+    {%- for t in tools %}
+        {{- t | tojson(indent=4) }}
+        {{- "\n\n" }}
+    {%- endfor %}
+    {{- first_user_message + "<|eot|>"}}
+{%- endif %}
+
+{%- for message in messages %}
+    {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}
+    {{- '<|header_start|>' + message['role'] + '<|header_end|>\n\n' }}
+        {%- if message['content'] is string %}
+            {{- message['content'] }}
+        {%- else %}
+            {%- for content in message['content'] %}
+                {%- if content['type'] == 'image' %}
+                    {{- '<|image|>' }}
+                {%- elif content['type'] == 'text' %}
+                    {{- content['text'] }}
+                {%- endif %}
+            {%- endfor %}
+        {%- endif %}
+        {{- "<|eot|>" }}
+    {%- elif 'tool_calls' in message and message.tool_calls|length > 0 %}
+       {{- '<|header_start|>assistant<|header_end|>\n\n' -}}
+       {{- '<|python_start|>' }}
+        {%- if message['content'] is string %}
+            {{- message['content'] }}
+        {%- else %}
+            {%- for content in message['content'] %}
+                {%- if content['type'] == 'image' %}
+                    {{- '<|image|>' }}
+                {%- elif content['type'] == 'text' %}
+                    {{- content['text'] }}
+                {%- endif %}
+            {%- endfor %}
+        {%- endif %}
+       {{- '<|python_end|>' }}
+        {%- for tool_call in message.tool_calls %}
+           {{- '{"name": "' + tool_call.function.name + '", ' }}
+           {{- '"parameters": ' }}
+           {{- tool_call.function.arguments | tojson }}
+           {{- "}" }}
+        {%- endfor %}
+       {{- "<|eot|>" }}
+    {%- elif message.role == "tool" or message.role == "ipython" %}
+        {{- "<|header_start|>ipython<|header_end|>\n\n" }}
+        {%- if message.content is mapping or message.content is iterable %}
+            {{- message.content | tojson }}
+        {%- else %}
+            {{- message.content }}
+        {%- endif %}
+        {{- "<|eot|>" }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|header_start|>assistant<|header_end|>\n\n' }}
+{%- endif %}

config.json

@@ -0,0 +1,215 @@
+{
+  "architectures": [
+    "Llama4ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_chunk_size": 8192,
+  "attention_dropout": 0.0,
+  "attn_scale": 0.1,
+  "attn_temperature_tuning": true,
+  "bos_token_id": 200000,
+  "cache_implementation": "hybrid",
+  "dtype": "bfloat16",
+  "eos_token_id": 200008,
+  "floor_scale": 8192,
+  "for_llm_compressor": false,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "interleave_moe_layer_step": 1,
+  "intermediate_size": 8192,
+  "intermediate_size_mlp": 16384,
+  "layer_types": [
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "chunked_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 10485760,
+  "model_type": "llama4_text",
+  "moe_layers": [
+    0,
+    1,
+    2,
+    3,
+    4,
+    5,
+    6,
+    7,
+    8,
+    9,
+    10,
+    11,
+    12,
+    13,
+    14,
+    15,
+    16,
+    17,
+    18,
+    19,
+    20,
+    21,
+    22,
+    23,
+    24,
+    25,
+    26,
+    27,
+    28,
+    29,
+    30,
+    31,
+    32,
+    33,
+    34,
+    35,
+    36,
+    37,
+    38,
+    39,
+    40,
+    41,
+    42,
+    43,
+    44,
+    45,
+    46,
+    47
+  ],
+  "no_rope_layer_interval": 4,
+  "no_rope_layers": [
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    0,
+    1,
+    1,
+    1,
+    0
+  ],
+  "num_attention_heads": 40,
+  "num_experts_per_tok": 1,
+  "num_hidden_layers": 48,
+  "num_key_value_heads": 8,
+  "num_local_experts": 16,
+  "output_router_logits": false,
+  "pad_token_id": 200018,
+  "quantization_config": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
+    "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_quant_storage": "bfloat16",
+    "bnb_4bit_quant_type": "nf4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
+  "rms_norm_eps": 1e-05,
+  "rope_parameters": {
+    "factor": 16.0,
+    "high_freq_factor": 1.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_theta": 500000.0,
+    "rope_type": "llama3"
+  },
+  "router_aux_loss_coef": 0.001,
+  "router_jitter_noise": 0.0,
+  "tie_word_embeddings": false,
+  "transformers_version": "5.13.0",
+  "use_cache": false,
+  "use_qk_norm": true,
+  "vocab_size": 202048,
+  "torch_dtype": "bfloat16"
+}

special_tokens_map.json

@@ -0,0 +1,5 @@
+{
+  "bos_token": "<|begin_of_text|>",
+  "eos_token": "<|eot|>",
+  "pad_token": "<|finetune_right_pad|>"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:172c9eb4beafc72601690da3ccfcede5c2e6806a8d5ec1fca33e22acea8023a4
+size 27948578

tokenizer_config.json

@@ -0,0 +1,17 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<|begin_of_text|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|eot|>",
+  "is_local": false,
+  "local_files_only": true,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 10485760,
+  "pad_token": "<|finetune_right_pad|>",
+  "padding_side": "right",
+  "processor_class": "Llama4Processor",
+  "tokenizer_class": "TokenizersBackend"
+}

trainer_state.json

@@ -0,0 +1,662 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 16,
+  "global_step": 84,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.047619047619047616,
+      "grad_norm": 0.6781264543533325,
+      "learning_rate": 0.0,
+      "loss": 1.26171875,
+      "step": 1
+    },
+    {
+      "epoch": 0.09523809523809523,
+      "grad_norm": 0.6318029761314392,
+      "learning_rate": 2e-05,
+      "loss": 1.20166015625,
+      "step": 2
+    },
+    {
+      "epoch": 0.14285714285714285,
+      "grad_norm": 0.7097207903862,
+      "learning_rate": 4e-05,
+      "loss": 1.412139892578125,
+      "step": 3
+    },
+    {
+      "epoch": 0.19047619047619047,
+      "grad_norm": 0.622149646282196,
+      "learning_rate": 6e-05,
+      "loss": 1.25537109375,
+      "step": 4
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 2.2328977584838867,
+      "learning_rate": 8e-05,
+      "loss": 1.1912841796875,
+      "step": 5
+    },
+    {
+      "epoch": 0.2857142857142857,
+      "grad_norm": 0.3143213987350464,
+      "learning_rate": 0.0001,
+      "loss": 1.228759765625,
+      "step": 6
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.2846517860889435,
+      "learning_rate": 9.996442287522857e-05,
+      "loss": 1.09326171875,
+      "step": 7
+    },
+    {
+      "epoch": 0.38095238095238093,
+      "grad_norm": 0.3680817186832428,
+      "learning_rate": 9.985774775566131e-05,
+      "loss": 1.0126953125,
+      "step": 8
+    },
+    {
+      "epoch": 0.42857142857142855,
+      "grad_norm": 0.5873160362243652,
+      "learning_rate": 9.96801433165888e-05,
+      "loss": 1.241455078125,
+      "step": 9
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.3845677077770233,
+      "learning_rate": 9.943189038713501e-05,
+      "loss": 1.058349609375,
+      "step": 10
+    },
+    {
+      "epoch": 0.5238095238095238,
+      "grad_norm": 0.262990802526474,
+      "learning_rate": 9.911338150620863e-05,
+      "loss": 1.056396484375,
+      "step": 11
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.2172330915927887,
+      "learning_rate": 9.872512030181845e-05,
+      "loss": 1.0838623046875,
+      "step": 12
+    },
+    {
+      "epoch": 0.6190476190476191,
+      "grad_norm": 0.23545600473880768,
+      "learning_rate": 9.826772069473393e-05,
+      "loss": 1.056396484375,
+      "step": 13
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.18727780878543854,
+      "learning_rate": 9.774190592775023e-05,
+      "loss": 1.0078125,
+      "step": 14
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 0.24253623187541962,
+      "learning_rate": 9.714850742209276e-05,
+      "loss": 1.090576171875,
+      "step": 15
+    },
+    {
+      "epoch": 0.7619047619047619,
+      "grad_norm": 0.13259640336036682,
+      "learning_rate": 9.648846346276929e-05,
+      "loss": 0.7398681640625,
+      "step": 16
+    },
+    {
+      "epoch": 0.8095238095238095,
+      "grad_norm": 0.20533913373947144,
+      "learning_rate": 9.576281771494855e-05,
+      "loss": 1.141845703125,
+      "step": 17
+    },
+    {
+      "epoch": 0.8571428571428571,
+      "grad_norm": 0.18200747668743134,
+      "learning_rate": 9.497271757371094e-05,
+      "loss": 1.017333984375,
+      "step": 18
+    },
+    {
+      "epoch": 0.9047619047619048,
+      "grad_norm": 0.19662703573703766,
+      "learning_rate": 9.411941234978122e-05,
+      "loss": 1.0162353515625,
+      "step": 19
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.19710412621498108,
+      "learning_rate": 9.320425129411139e-05,
+      "loss": 1.0875244140625,
+      "step": 20
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "eval_loss": 0.916015625,
+      "eval_runtime": 6.1898,
+      "eval_samples_per_second": 0.646,
+      "eval_steps_per_second": 0.162,
+      "step": 20
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.13886402547359467,
+      "learning_rate": 9.222868146443759e-05,
+      "loss": 0.9620361328125,
+      "step": 21
+    },
+    {
+      "epoch": 1.0476190476190477,
+      "grad_norm": 0.1303921639919281,
+      "learning_rate": 9.119424543718442e-05,
+      "loss": 0.9793701171875,
+      "step": 22
+    },
+    {
+      "epoch": 1.0952380952380953,
+      "grad_norm": 0.14203043282032013,
+      "learning_rate": 9.010257886833456e-05,
+      "loss": 0.936279296875,
+      "step": 23
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.19456878304481506,
+      "learning_rate": 8.895540790712036e-05,
+      "loss": 1.0773696899414062,
+      "step": 24
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 0.19268886744976044,
+      "learning_rate": 8.775454646662724e-05,
+      "loss": 0.962890625,
+      "step": 25
+    },
+    {
+      "epoch": 1.2380952380952381,
+      "grad_norm": 1.1448683738708496,
+      "learning_rate": 8.650189335562414e-05,
+      "loss": 1.0135498046875,
+      "step": 26
+    },
+    {
+      "epoch": 1.2857142857142856,
+      "grad_norm": 0.17032912373542786,
+      "learning_rate": 8.519942927615666e-05,
+      "loss": 1.041259765625,
+      "step": 27
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.16223372519016266,
+      "learning_rate": 8.384921369164991e-05,
+      "loss": 0.947998046875,
+      "step": 28
+    },
+    {
+      "epoch": 1.380952380952381,
+      "grad_norm": 0.13596029579639435,
+      "learning_rate": 8.24533815704736e-05,
+      "loss": 0.8992919921875,
+      "step": 29
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.19097673892974854,
+      "learning_rate": 8.101414001011817e-05,
+      "loss": 1.0830078125,
+      "step": 30
+    },
+    {
+      "epoch": 1.4761904761904763,
+      "grad_norm": 0.1492234319448471,
+      "learning_rate": 7.953376474732003e-05,
+      "loss": 0.9466552734375,
+      "step": 31
+    },
+    {
+      "epoch": 1.5238095238095237,
+      "grad_norm": 0.14599502086639404,
+      "learning_rate": 7.801459655965401e-05,
+      "loss": 0.957763671875,
+      "step": 32
+    },
+    {
+      "epoch": 1.5714285714285714,
+      "grad_norm": 0.13878127932548523,
+      "learning_rate": 7.645903756428279e-05,
+      "loss": 0.9761962890625,
+      "step": 33
+    },
+    {
+      "epoch": 1.619047619047619,
+      "grad_norm": 0.15830132365226746,
+      "learning_rate": 7.486954741971603e-05,
+      "loss": 0.957275390625,
+      "step": 34
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.13619418442249298,
+      "learning_rate": 7.324863943658474e-05,
+      "loss": 0.92626953125,
+      "step": 35
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.20511189103126526,
+      "learning_rate": 7.159887660358036e-05,
+      "loss": 0.980224609375,
+      "step": 36
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "eval_loss": 0.8896484375,
+      "eval_runtime": 5.9036,
+      "eval_samples_per_second": 0.678,
+      "eval_steps_per_second": 0.169,
+      "step": 36
+    },
+    {
+      "epoch": 1.7619047619047619,
+      "grad_norm": 0.11510299891233444,
+      "learning_rate": 6.992286753484306e-05,
+      "loss": 0.6959228515625,
+      "step": 37
+    },
+    {
+      "epoch": 1.8095238095238095,
+      "grad_norm": 0.17241714894771576,
+      "learning_rate": 6.822326234520645e-05,
+      "loss": 1.042236328125,
+      "step": 38
+    },
+    {
+      "epoch": 1.8571428571428572,
+      "grad_norm": 0.14825452864170074,
+      "learning_rate": 6.650274845982136e-05,
+      "loss": 0.93603515625,
+      "step": 39
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.1515885442495346,
+      "learning_rate": 6.476404636478432e-05,
+      "loss": 0.948974609375,
+      "step": 40
+    },
+    {
+      "epoch": 1.9523809523809523,
+      "grad_norm": 0.16863621771335602,
+      "learning_rate": 6.300990530548979e-05,
+      "loss": 0.9949951171875,
+      "step": 41
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.4026670455932617,
+      "learning_rate": 6.124309893950809e-05,
+      "loss": 0.869873046875,
+      "step": 42
+    },
+    {
+      "epoch": 2.0476190476190474,
+      "grad_norm": 0.16099941730499268,
+      "learning_rate": 5.946642095086278e-05,
+      "loss": 0.8929443359375,
+      "step": 43
+    },
+    {
+      "epoch": 2.0952380952380953,
+      "grad_norm": 0.13304688036441803,
+      "learning_rate": 5.7682680632641896e-05,
+      "loss": 0.873046875,
+      "step": 44
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 0.18484529852867126,
+      "learning_rate": 5.589469844492816e-05,
+      "loss": 0.9822769165039062,
+      "step": 45
+    },
+    {
+      "epoch": 2.1904761904761907,
+      "grad_norm": 0.17451068758964539,
+      "learning_rate": 5.410530155507184e-05,
+      "loss": 0.8817138671875,
+      "step": 46
+    },
+    {
+      "epoch": 2.238095238095238,
+      "grad_norm": 0.17454630136489868,
+      "learning_rate": 5.231731936735811e-05,
+      "loss": 0.9676513671875,
+      "step": 47
+    },
+    {
+      "epoch": 2.2857142857142856,
+      "grad_norm": 0.19139355421066284,
+      "learning_rate": 5.053357904913724e-05,
+      "loss": 0.9560546875,
+      "step": 48
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.18285448849201202,
+      "learning_rate": 4.875690106049193e-05,
+      "loss": 0.8807373046875,
+      "step": 49
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.16837579011917114,
+      "learning_rate": 4.6990094694510245e-05,
+      "loss": 0.8505859375,
+      "step": 50
+    },
+    {
+      "epoch": 2.4285714285714284,
+      "grad_norm": 0.21485020220279694,
+      "learning_rate": 4.5235953635215686e-05,
+      "loss": 0.994140625,
+      "step": 51
+    },
+    {
+      "epoch": 2.4761904761904763,
+      "grad_norm": 0.1551942676305771,
+      "learning_rate": 4.349725154017864e-05,
+      "loss": 0.88671875,
+      "step": 52
+    },
+    {
+      "epoch": 2.4761904761904763,
+      "eval_loss": 0.876953125,
+      "eval_runtime": 5.9086,
+      "eval_samples_per_second": 0.677,
+      "eval_steps_per_second": 0.169,
+      "step": 52
+    },
+    {
+      "epoch": 2.5238095238095237,
+      "grad_norm": 0.16076764464378357,
+      "learning_rate": 4.177673765479357e-05,
+      "loss": 0.9061279296875,
+      "step": 53
+    },
+    {
+      "epoch": 2.571428571428571,
+      "grad_norm": 0.16922174394130707,
+      "learning_rate": 4.007713246515695e-05,
+      "loss": 0.9171142578125,
+      "step": 54
+    },
+    {
+      "epoch": 2.619047619047619,
+      "grad_norm": 0.18479417264461517,
+      "learning_rate": 3.8401123396419656e-05,
+      "loss": 0.896240234375,
+      "step": 55
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.15129074454307556,
+      "learning_rate": 3.675136056341528e-05,
+      "loss": 0.880126953125,
+      "step": 56
+    },
+    {
+      "epoch": 2.7142857142857144,
+      "grad_norm": 0.23219619691371918,
+      "learning_rate": 3.513045258028399e-05,
+      "loss": 0.900390625,
+      "step": 57
+    },
+    {
+      "epoch": 2.761904761904762,
+      "grad_norm": 0.11648663133382797,
+      "learning_rate": 3.354096243571724e-05,
+      "loss": 0.6685791015625,
+      "step": 58
+    },
+    {
+      "epoch": 2.8095238095238093,
+      "grad_norm": 0.20129463076591492,
+      "learning_rate": 3.1985403440346004e-05,
+      "loss": 0.97607421875,
+      "step": 59
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.17487110197544098,
+      "learning_rate": 3.0466235252679976e-05,
+      "loss": 0.8857421875,
+      "step": 60
+    },
+    {
+      "epoch": 2.9047619047619047,
+      "grad_norm": 0.18634696304798126,
+      "learning_rate": 2.898585998988182e-05,
+      "loss": 0.904296875,
+      "step": 61
+    },
+    {
+      "epoch": 2.9523809523809526,
+      "grad_norm": 0.21948011219501495,
+      "learning_rate": 2.7546618429526404e-05,
+      "loss": 0.935302734375,
+      "step": 62
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.17362090945243835,
+      "learning_rate": 2.6150786308350112e-05,
+      "loss": 0.8018798828125,
+      "step": 63
+    },
+    {
+      "epoch": 3.0476190476190474,
+      "grad_norm": 0.16069509088993073,
+      "learning_rate": 2.480057072384334e-05,
+      "loss": 0.8336181640625,
+      "step": 64
+    },
+    {
+      "epoch": 3.0952380952380953,
+      "grad_norm": 0.1560245156288147,
+      "learning_rate": 2.3498106644375867e-05,
+      "loss": 0.8359375,
+      "step": 65
+    },
+    {
+      "epoch": 3.142857142857143,
+      "grad_norm": 0.21654845774173737,
+      "learning_rate": 2.2245453533372768e-05,
+      "loss": 0.9223861694335938,
+      "step": 66
+    },
+    {
+      "epoch": 3.1904761904761907,
+      "grad_norm": 0.20104853808879852,
+      "learning_rate": 2.1044592092879643e-05,
+      "loss": 0.82958984375,
+      "step": 67
+    },
+    {
+      "epoch": 3.238095238095238,
+      "grad_norm": 0.19536712765693665,
+      "learning_rate": 1.9897421131665462e-05,
+      "loss": 0.943115234375,
+      "step": 68
+    },
+    {
+      "epoch": 3.238095238095238,
+      "eval_loss": 0.875,
+      "eval_runtime": 5.8891,
+      "eval_samples_per_second": 0.679,
+      "eval_steps_per_second": 0.17,
+      "step": 68
+    },
+    {
+      "epoch": 3.2857142857142856,
+      "grad_norm": 0.21967998147010803,
+      "learning_rate": 1.8805754562815585e-05,
+      "loss": 0.899658203125,
+      "step": 69
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.19596749544143677,
+      "learning_rate": 1.7771318535562435e-05,
+      "loss": 0.834228515625,
+      "step": 70
+    },
+    {
+      "epoch": 3.380952380952381,
+      "grad_norm": 0.16757430136203766,
+      "learning_rate": 1.679574870588862e-05,
+      "loss": 0.82275390625,
+      "step": 71
+    },
+    {
+      "epoch": 3.4285714285714284,
+      "grad_norm": 0.2443944215774536,
+      "learning_rate": 1.588058765021879e-05,
+      "loss": 0.936767578125,
+      "step": 72
+    },
+    {
+      "epoch": 3.4761904761904763,
+      "grad_norm": 0.18178075551986694,
+      "learning_rate": 1.5027282426289077e-05,
+      "loss": 0.85498046875,
+      "step": 73
+    },
+    {
+      "epoch": 3.5238095238095237,
+      "grad_norm": 0.1797463297843933,
+      "learning_rate": 1.4237182285051453e-05,
+      "loss": 0.8779296875,
+      "step": 74
+    },
+    {
+      "epoch": 3.571428571428571,
+      "grad_norm": 0.19107820093631744,
+      "learning_rate": 1.3511536537230707e-05,
+      "loss": 0.8829345703125,
+      "step": 75
+    },
+    {
+      "epoch": 3.619047619047619,
+      "grad_norm": 0.1874479353427887,
+      "learning_rate": 1.2851492577907251e-05,
+      "loss": 0.8621826171875,
+      "step": 76
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 0.16661852598190308,
+      "learning_rate": 1.225809407224978e-05,
+      "loss": 0.855712890625,
+      "step": 77
+    },
+    {
+      "epoch": 3.7142857142857144,
+      "grad_norm": 0.2627502381801605,
+      "learning_rate": 1.1732279305266082e-05,
+      "loss": 0.8544921875,
+      "step": 78
+    },
+    {
+      "epoch": 3.761904761904762,
+      "grad_norm": 0.12383028864860535,
+      "learning_rate": 1.1274879698181547e-05,
+      "loss": 0.6553955078125,
+      "step": 79
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.21695609390735626,
+      "learning_rate": 1.0886618493791376e-05,
+      "loss": 0.940673828125,
+      "step": 80
+    },
+    {
+      "epoch": 3.857142857142857,
+      "grad_norm": 0.18359997868537903,
+      "learning_rate": 1.0568109612865e-05,
+      "loss": 0.85888671875,
+      "step": 81
+    },
+    {
+      "epoch": 3.9047619047619047,
+      "grad_norm": 0.19658944010734558,
+      "learning_rate": 1.0319856683411197e-05,
+      "loss": 0.8824462890625,
+      "step": 82
+    },
+    {
+      "epoch": 3.9523809523809526,
+      "grad_norm": 0.24310623109340668,
+      "learning_rate": 1.0142252244338688e-05,
+      "loss": 0.905029296875,
+      "step": 83
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.2020317167043686,
+      "learning_rate": 1.0035577124771419e-05,
+      "loss": 0.7667236328125,
+      "step": 84
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.8740234375,
+      "eval_runtime": 5.901,
+      "eval_samples_per_second": 0.678,
+      "eval_steps_per_second": 0.169,
+      "step": 84
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 84,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.3441704712428585e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}