Add 12 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
+training-metrics.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,99 @@
+---
+base_model: mistralai/Mixtral-8x7B-Instruct-v0.1
+library_name: peft
+license: other
+tags:
+  - lora
+  - peft
+  - adapter
+  - adaption
+---
+
+# adaption_diverse_ai_task_samples
+
+## Model Training
+
+A LORA adapter for `mistralai/Mixtral-8x7B-Instruct-v0.1`. This model was trained with SFT using [Adaption](https://adaptionlabs.ai)'s AutoScientist on the diverse_ai_task_samples dataset.
+
+
+![Training metrics](training-metrics.png)
+
+### AutoScientist Config
+
+```json
+{
+  "job_id": "34447162-a1f5-49ba-ab92-d9ad159c85a0",
+  "training_experiment_id": "77bacf88-1a66-492e-9e7d-2e5d1cb4fab5",
+  "original_model_name": "mistralai/Mixtral-8x7B-Instruct-v0.1",
+  "trained_model_name": "adaption_diverse_ai_task_samples",
+  "training_method": "sft",
+  "training_type": "lora",
+  "data_format": "chat",
+  "hyperparams": {
+    "lora": "true",
+    "lora_r": 16,
+    "n_evals": 5,
+    "n_epochs": 4,
+    "batch_size": "max",
+    "lora_alpha": 32,
+    "lora_dropout": 0.1,
+    "min_lr_ratio": 0.1,
+    "warmup_ratio": 0.05,
+    "weight_decay": 0.01,
+    "learning_rate": 0.00004,
+    "max_grad_norm": 1,
+    "base_model_size": "46.7B",
+    "train_on_inputs": "false",
+    "training_method": "sft",
+    "lr_scheduler_type": "cosine",
+    "scheduler_num_cycles": 0.5,
+    "lora_trainable_modules": "all-linear"
+  }
+}
+```
+
+## Training Data
+
+The model was trained on 795 rows of adapted data with the following domain distribution: entertainment (12%), language (11%), academic-education (9%), news (8%), history (8%), science (8%), writing-editing-communication (7%), math (6%), sports (6%), other (4%), data-analysis-visualization (3%), social (2%), geography (2%), games (2%), how-to (2%), medical (2%), religion (2%), governance (2%), personal-growth (2%), art (1%), code (1%), cooking (1%), culture (1%), animal-nature (1%), product-advice (1%), logic (1%), music (1%), dating (1%), roleplay (1%), technology (1%).
+
+## 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)
+
+
+## How to use
+
+```bash
+pip install torch transformers peft
+```
+
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from peft import PeftModel
+
+BASE = "mistralai/Mixtral-8x7B-Instruct-v0.1"
+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 +1,36 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mixtral-8x7B-Instruct-v0.1",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.10000000149011612,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj",
+    "k_proj",
+    "o_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:af94c1b8b886457850757cc30053f845f1470116a26671a31031d29bc73326ba
+size 54560368

chat_template.jinja

@@ -0,0 +1,24 @@
+{%- if messages[0]['role'] == 'system' %}
+    {%- set system_message = messages[0]['content'] %}
+    {%- set loop_messages = messages[1:] %}
+{%- else %}
+    {%- set loop_messages = messages %}
+{%- endif %}
+
+{{- bos_token }}
+{%- for message in loop_messages %}
+    {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}
+        {{- raise_exception('After the optional system message, conversation roles must alternate user/assistant/user/assistant/...') }}
+    {%- endif %}
+    {%- if message['role'] == 'user' %}
+        {%- if loop.first and system_message is defined %}
+            {{- ' [INST] ' + system_message + '\n\n' + message['content'] + ' [/INST]' }}
+        {%- else %}
+            {{- ' [INST] ' + message['content'] + ' [/INST]' }}
+        {%- endif %}
+    {%- elif message['role'] == 'assistant' %}
+        {{- ' ' + message['content'] + eos_token}}
+    {%- else %}
+        {{- raise_exception('Only user and assistant roles are supported, with the exception of an initial optional system message!') }}
+    {%- endif %}
+{%- endfor %}

config.json

@@ -0,0 +1,33 @@
+{
+  "architectures": [
+    "MixtralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dtype": "bfloat16",
+  "eos_token_id": 2,
+  "head_dim": null,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mixtral",
+  "num_attention_heads": 32,
+  "num_experts_per_tok": 2,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "num_local_experts": 8,
+  "output_router_logits": false,
+  "pad_token_id": 2,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "router_aux_loss_coef": 0.02,
+  "router_jitter_noise": 0.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.57.0",
+  "use_cache": false,
+  "vocab_size": 32000,
+  "torch_dtype": "bfloat16"
+}

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443

tokenizer_config.json

@@ -0,0 +1,45 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "legacy": false,
+  "model_max_length": 32768,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

trainer_state.json

@@ -0,0 +1,774 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 20,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.4760715663433075,
+      "learning_rate": 0.0,
+      "loss": 1.4658,
+      "step": 1
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.5013044476509094,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 1.4224,
+      "step": 2
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.2839778661727905,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 1.125,
+      "step": 3
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.33960863947868347,
+      "learning_rate": 2.4e-05,
+      "loss": 1.238,
+      "step": 4
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.416923850774765,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 1.2939,
+      "step": 5
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.5548073649406433,
+      "learning_rate": 4e-05,
+      "loss": 1.4395,
+      "step": 6
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.7273683547973633,
+      "learning_rate": 3.999015863213667e-05,
+      "loss": 1.5093,
+      "step": 7
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.5051376223564148,
+      "learning_rate": 3.996064528993794e-05,
+      "loss": 1.1609,
+      "step": 8
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.49626877903938293,
+      "learning_rate": 3.9911492245810205e-05,
+      "loss": 1.3969,
+      "step": 9
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.42138198018074036,
+      "learning_rate": 3.9842753247885556e-05,
+      "loss": 1.1953,
+      "step": 10
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.5137280821800232,
+      "learning_rate": 3.975450346124901e-05,
+      "loss": 1.3467,
+      "step": 11
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.2119121551513672,
+      "learning_rate": 3.964683938574657e-05,
+      "loss": 0.811,
+      "step": 12
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.28574588894844055,
+      "learning_rate": 3.9519878750464104e-05,
+      "loss": 1.1553,
+      "step": 13
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.3548891842365265,
+      "learning_rate": 3.937376038499228e-05,
+      "loss": 1.1643,
+      "step": 14
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.32107675075531006,
+      "learning_rate": 3.920864406761844e-05,
+      "loss": 1.186,
+      "step": 15
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.4450090527534485,
+      "learning_rate": 3.9024710350611426e-05,
+      "loss": 1.297,
+      "step": 16
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.3887384235858917,
+      "learning_rate": 3.88221603627903e-05,
+      "loss": 1.2583,
+      "step": 17
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.2024194747209549,
+      "learning_rate": 3.860121558959293e-05,
+      "loss": 1.1979,
+      "step": 18
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.164462149143219,
+      "learning_rate": 3.836211763088493e-05,
+      "loss": 1.0645,
+      "step": 19
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.07839324325323105,
+      "learning_rate": 3.8105127936773746e-05,
+      "loss": 0.7288,
+      "step": 20
+    },
+    {
+      "epoch": 0.8,
+      "eval_loss": 0.855224609375,
+      "eval_runtime": 2.1713,
+      "eval_samples_per_second": 2.303,
+      "eval_steps_per_second": 0.461,
+      "step": 20
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.13838034868240356,
+      "learning_rate": 3.78305275217168e-05,
+      "loss": 1.1664,
+      "step": 21
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.20036262273788452,
+      "learning_rate": 3.7538616657236334e-05,
+      "loss": 1.2454,
+      "step": 22
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.08870978653430939,
+      "learning_rate": 3.722971454357684e-05,
+      "loss": 0.9868,
+      "step": 23
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.08468154072761536,
+      "learning_rate": 3.6904158960664365e-05,
+      "loss": 1.0046,
+      "step": 24
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.1123848706483841,
+      "learning_rate": 3.656230589874905e-05,
+      "loss": 1.1274,
+      "step": 25
+    },
+    {
+      "epoch": 1.04,
+      "grad_norm": 0.11019331961870193,
+      "learning_rate": 3.6204529169135086e-05,
+      "loss": 1.156,
+      "step": 26
+    },
+    {
+      "epoch": 1.08,
+      "grad_norm": 0.10665232688188553,
+      "learning_rate": 3.5831219995423466e-05,
+      "loss": 1.1367,
+      "step": 27
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 0.08694541454315186,
+      "learning_rate": 3.544278658571477e-05,
+      "loss": 0.9375,
+      "step": 28
+    },
+    {
+      "epoch": 1.16,
+      "grad_norm": 0.0901435911655426,
+      "learning_rate": 3.5039653686239606e-05,
+      "loss": 1.0381,
+      "step": 29
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.09067416191101074,
+      "learning_rate": 3.4622262116904795e-05,
+      "loss": 1.0396,
+      "step": 30
+    },
+    {
+      "epoch": 1.24,
+      "grad_norm": 0.09607220441102982,
+      "learning_rate": 3.419106828926334e-05,
+      "loss": 1.1133,
+      "step": 31
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.09983466565608978,
+      "learning_rate": 3.374654370743504e-05,
+      "loss": 1.1624,
+      "step": 32
+    },
+    {
+      "epoch": 1.32,
+      "grad_norm": 0.08482648432254791,
+      "learning_rate": 3.328917445252367e-05,
+      "loss": 0.9143,
+      "step": 33
+    },
+    {
+      "epoch": 1.3599999999999999,
+      "grad_norm": 0.10168825834989548,
+      "learning_rate": 3.281946065109445e-05,
+      "loss": 1.1602,
+      "step": 34
+    },
+    {
+      "epoch": 1.4,
+      "grad_norm": 0.08425932377576828,
+      "learning_rate": 3.233791592829299e-05,
+      "loss": 1.0061,
+      "step": 35
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 0.09837682545185089,
+      "learning_rate": 3.1845066846203686e-05,
+      "loss": 1.1282,
+      "step": 36
+    },
+    {
+      "epoch": 1.48,
+      "grad_norm": 0.05306907743215561,
+      "learning_rate": 3.134145232806188e-05,
+      "loss": 0.7241,
+      "step": 37
+    },
+    {
+      "epoch": 1.52,
+      "grad_norm": 0.08243704587221146,
+      "learning_rate": 3.082762306894915e-05,
+      "loss": 1.0229,
+      "step": 38
+    },
+    {
+      "epoch": 1.56,
+      "grad_norm": 0.07551784068346024,
+      "learning_rate": 3.0304140933616387e-05,
+      "loss": 1.0286,
+      "step": 39
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.08548041433095932,
+      "learning_rate": 2.977157834209287e-05,
+      "loss": 1.0471,
+      "step": 40
+    },
+    {
+      "epoch": 1.6,
+      "eval_loss": 0.798828125,
+      "eval_runtime": 2.202,
+      "eval_samples_per_second": 2.271,
+      "eval_steps_per_second": 0.454,
+      "step": 40
+    },
+    {
+      "epoch": 1.6400000000000001,
+      "grad_norm": 0.08860009163618088,
+      "learning_rate": 2.9230517643753457e-05,
+      "loss": 1.1273,
+      "step": 41
+    },
+    {
+      "epoch": 1.6800000000000002,
+      "grad_norm": 0.09040582180023193,
+      "learning_rate": 2.8681550480527974e-05,
+      "loss": 1.1182,
+      "step": 42
+    },
+    {
+      "epoch": 1.72,
+      "grad_norm": 0.07563644647598267,
+      "learning_rate": 2.812527713994958e-05,
+      "loss": 1.0964,
+      "step": 43
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 0.07461907714605331,
+      "learning_rate": 2.7562305898749054e-05,
+      "loss": 0.9688,
+      "step": 44
+    },
+    {
+      "epoch": 1.8,
+      "grad_norm": 0.04516046494245529,
+      "learning_rate": 2.6993252357713205e-05,
+      "loss": 0.6775,
+      "step": 45
+    },
+    {
+      "epoch": 1.8399999999999999,
+      "grad_norm": 0.07931992411613464,
+      "learning_rate": 2.6418738768534388e-05,
+      "loss": 1.0747,
+      "step": 46
+    },
+    {
+      "epoch": 1.88,
+      "grad_norm": 0.09479013085365295,
+      "learning_rate": 2.5839393353387444e-05,
+      "loss": 1.1385,
+      "step": 47
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 0.057173825800418854,
+      "learning_rate": 2.5255849617977983e-05,
+      "loss": 0.9226,
+      "step": 48
+    },
+    {
+      "epoch": 1.96,
+      "grad_norm": 0.059427376836538315,
+      "learning_rate": 2.4668745658813156e-05,
+      "loss": 0.9497,
+      "step": 49
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.07764241099357605,
+      "learning_rate": 2.4078723465452556e-05,
+      "loss": 1.0436,
+      "step": 50
+    },
+    {
+      "epoch": 2.04,
+      "grad_norm": 0.07360248267650604,
+      "learning_rate": 2.3486428218501987e-05,
+      "loss": 1.0789,
+      "step": 51
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 0.07298833131790161,
+      "learning_rate": 2.2892507584118004e-05,
+      "loss": 1.0659,
+      "step": 52
+    },
+    {
+      "epoch": 2.12,
+      "grad_norm": 0.05753123015165329,
+      "learning_rate": 2.2297611005794456e-05,
+      "loss": 0.8865,
+      "step": 53
+    },
+    {
+      "epoch": 2.16,
+      "grad_norm": 0.06963388621807098,
+      "learning_rate": 2.170238899420555e-05,
+      "loss": 0.9897,
+      "step": 54
+    },
+    {
+      "epoch": 2.2,
+      "grad_norm": 0.07072041928768158,
+      "learning_rate": 2.1107492415882002e-05,
+      "loss": 0.9912,
+      "step": 55
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 0.07816724479198456,
+      "learning_rate": 2.0513571781498022e-05,
+      "loss": 1.0527,
+      "step": 56
+    },
+    {
+      "epoch": 2.2800000000000002,
+      "grad_norm": 0.0970468819141388,
+      "learning_rate": 1.9921276534547453e-05,
+      "loss": 1.1062,
+      "step": 57
+    },
+    {
+      "epoch": 2.32,
+      "grad_norm": 0.07797855883836746,
+      "learning_rate": 1.9331254341186843e-05,
+      "loss": 0.8679,
+      "step": 58
+    },
+    {
+      "epoch": 2.36,
+      "grad_norm": 0.08266980946063995,
+      "learning_rate": 1.874415038202203e-05,
+      "loss": 1.1044,
+      "step": 59
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 0.06626752018928528,
+      "learning_rate": 1.816060664661256e-05,
+      "loss": 0.9661,
+      "step": 60
+    },
+    {
+      "epoch": 2.4,
+      "eval_loss": 0.77880859375,
+      "eval_runtime": 2.2199,
+      "eval_samples_per_second": 2.252,
+      "eval_steps_per_second": 0.45,
+      "step": 60
+    },
+    {
+      "epoch": 2.44,
+      "grad_norm": 0.08102354407310486,
+      "learning_rate": 1.7581261231465618e-05,
+      "loss": 1.0759,
+      "step": 61
+    },
+    {
+      "epoch": 2.48,
+      "grad_norm": 0.054694049060344696,
+      "learning_rate": 1.70067476422868e-05,
+      "loss": 0.7046,
+      "step": 62
+    },
+    {
+      "epoch": 2.52,
+      "grad_norm": 0.07155449688434601,
+      "learning_rate": 1.643769410125095e-05,
+      "loss": 0.9838,
+      "step": 63
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 0.06532476842403412,
+      "learning_rate": 1.5874722860050423e-05,
+      "loss": 0.994,
+      "step": 64
+    },
+    {
+      "epoch": 2.6,
+      "grad_norm": 0.07697983086109161,
+      "learning_rate": 1.5318449519472025e-05,
+      "loss": 1.0082,
+      "step": 65
+    },
+    {
+      "epoch": 2.64,
+      "grad_norm": 0.08935107290744781,
+      "learning_rate": 1.4769482356246556e-05,
+      "loss": 1.0865,
+      "step": 66
+    },
+    {
+      "epoch": 2.68,
+      "grad_norm": 0.0887816920876503,
+      "learning_rate": 1.422842165790713e-05,
+      "loss": 1.0815,
+      "step": 67
+    },
+    {
+      "epoch": 2.7199999999999998,
+      "grad_norm": 0.07478179782629013,
+      "learning_rate": 1.3695859066383624e-05,
+      "loss": 1.0665,
+      "step": 68
+    },
+    {
+      "epoch": 2.76,
+      "grad_norm": 0.07210303097963333,
+      "learning_rate": 1.317237693105086e-05,
+      "loss": 0.9429,
+      "step": 69
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 0.044374942779541016,
+      "learning_rate": 1.2658547671938135e-05,
+      "loss": 0.6628,
+      "step": 70
+    },
+    {
+      "epoch": 2.84,
+      "grad_norm": 0.07966240495443344,
+      "learning_rate": 1.2154933153796323e-05,
+      "loss": 1.0466,
+      "step": 71
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 0.0976930782198906,
+      "learning_rate": 1.1662084071707018e-05,
+      "loss": 1.1078,
+      "step": 72
+    },
+    {
+      "epoch": 2.92,
+      "grad_norm": 0.05639340355992317,
+      "learning_rate": 1.118053934890555e-05,
+      "loss": 0.9038,
+      "step": 73
+    },
+    {
+      "epoch": 2.96,
+      "grad_norm": 0.06098003685474396,
+      "learning_rate": 1.0710825547476333e-05,
+      "loss": 0.9326,
+      "step": 74
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.07597635686397552,
+      "learning_rate": 1.0253456292564965e-05,
+      "loss": 1.0192,
+      "step": 75
+    },
+    {
+      "epoch": 3.04,
+      "grad_norm": 0.07052882015705109,
+      "learning_rate": 9.808931710736666e-06,
+      "loss": 1.0579,
+      "step": 76
+    },
+    {
+      "epoch": 3.08,
+      "grad_norm": 0.07034008204936981,
+      "learning_rate": 9.377737883095207e-06,
+      "loss": 1.0454,
+      "step": 77
+    },
+    {
+      "epoch": 3.12,
+      "grad_norm": 0.05514257028698921,
+      "learning_rate": 8.9603463137604e-06,
+      "loss": 0.873,
+      "step": 78
+    },
+    {
+      "epoch": 3.16,
+      "grad_norm": 0.06549564749002457,
+      "learning_rate": 8.557213414285236e-06,
+      "loss": 0.9741,
+      "step": 79
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 0.07224049419164658,
+      "learning_rate": 8.168780004576542e-06,
+      "loss": 0.9753,
+      "step": 80
+    },
+    {
+      "epoch": 3.2,
+      "eval_loss": 0.771484375,
+      "eval_runtime": 2.2033,
+      "eval_samples_per_second": 2.269,
+      "eval_steps_per_second": 0.454,
+      "step": 80
+    },
+    {
+      "epoch": 3.24,
+      "grad_norm": 0.07591705769300461,
+      "learning_rate": 7.79547083086492e-06,
+      "loss": 1.0322,
+      "step": 81
+    },
+    {
+      "epoch": 3.2800000000000002,
+      "grad_norm": 0.09342397004365921,
+      "learning_rate": 7.437694101250949e-06,
+      "loss": 1.0872,
+      "step": 82
+    },
+    {
+      "epoch": 3.32,
+      "grad_norm": 0.07288067787885666,
+      "learning_rate": 7.09584103933564e-06,
+      "loss": 0.8525,
+      "step": 83
+    },
+    {
+      "epoch": 3.36,
+      "grad_norm": 0.07975227385759354,
+      "learning_rate": 6.7702854564231615e-06,
+      "loss": 1.0865,
+      "step": 84
+    },
+    {
+      "epoch": 3.4,
+      "grad_norm": 0.06429357826709747,
+      "learning_rate": 6.461383342763671e-06,
+      "loss": 0.9534,
+      "step": 85
+    },
+    {
+      "epoch": 3.44,
+      "grad_norm": 0.08074600249528885,
+      "learning_rate": 6.169472478283199e-06,
+      "loss": 1.0605,
+      "step": 86
+    },
+    {
+      "epoch": 3.48,
+      "grad_norm": 0.05224496126174927,
+      "learning_rate": 5.894872063226258e-06,
+      "loss": 0.696,
+      "step": 87
+    },
+    {
+      "epoch": 3.52,
+      "grad_norm": 0.06972762197256088,
+      "learning_rate": 5.637882369115075e-06,
+      "loss": 0.9702,
+      "step": 88
+    },
+    {
+      "epoch": 3.56,
+      "grad_norm": 0.06537563353776932,
+      "learning_rate": 5.3987844104070754e-06,
+      "loss": 0.9809,
+      "step": 89
+    },
+    {
+      "epoch": 3.6,
+      "grad_norm": 0.07811121642589569,
+      "learning_rate": 5.177839637209707e-06,
+      "loss": 0.9956,
+      "step": 90
+    },
+    {
+      "epoch": 3.64,
+      "grad_norm": 0.08834316581487656,
+      "learning_rate": 4.975289649388577e-06,
+      "loss": 1.0721,
+      "step": 91
+    },
+    {
+      "epoch": 3.68,
+      "grad_norm": 0.08688650280237198,
+      "learning_rate": 4.791355932381563e-06,
+      "loss": 1.0693,
+      "step": 92
+    },
+    {
+      "epoch": 3.7199999999999998,
+      "grad_norm": 0.0746871680021286,
+      "learning_rate": 4.626239615007727e-06,
+      "loss": 1.0565,
+      "step": 93
+    },
+    {
+      "epoch": 3.76,
+      "grad_norm": 0.07580739259719849,
+      "learning_rate": 4.480121249535897e-06,
+      "loss": 0.9346,
+      "step": 94
+    },
+    {
+      "epoch": 3.8,
+      "grad_norm": 0.061057522892951965,
+      "learning_rate": 4.353160614253432e-06,
+      "loss": 0.6584,
+      "step": 95
+    },
+    {
+      "epoch": 3.84,
+      "grad_norm": 0.08070012927055359,
+      "learning_rate": 4.245496538750999e-06,
+      "loss": 1.0371,
+      "step": 96
+    },
+    {
+      "epoch": 3.88,
+      "grad_norm": 0.09691870212554932,
+      "learning_rate": 4.157246752114447e-06,
+      "loss": 1.0974,
+      "step": 97
+    },
+    {
+      "epoch": 3.92,
+      "grad_norm": 0.0571683831512928,
+      "learning_rate": 4.0885077541897945e-06,
+      "loss": 0.8972,
+      "step": 98
+    },
+    {
+      "epoch": 3.96,
+      "grad_norm": 0.061304960399866104,
+      "learning_rate": 4.039354710062059e-06,
+      "loss": 0.9272,
+      "step": 99
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.07660021632909775,
+      "learning_rate": 4.0098413678633315e-06,
+      "loss": 1.0127,
+      "step": 100
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.77001953125,
+      "eval_runtime": 2.1902,
+      "eval_samples_per_second": 2.283,
+      "eval_steps_per_second": 0.457,
+      "step": 100
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 100,
+  "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": 1.1998359434821632e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}