Upload folder using huggingface_hub

.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 +1,63 @@
+---
+base_model: unsloth/Qwen2.5-7B-Instruct
+library_name: peft
+model_name: Qwen_Qwen2.5-7B-Instruct__project-7b-math_1785259288
+tags:
+- base_model:adapter:unsloth/Qwen2.5-7B-Instruct
+- lora
+- sft
+- transformers
+- trl
+- unsloth
+licence: license
+pipeline_tag: text-generation
+---
+
+# Model Card for Qwen_Qwen2.5-7B-Instruct__project-7b-math_1785259288
+
+This model is a fine-tuned version of [unsloth/Qwen2.5-7B-Instruct](https://huggingface.co/unsloth/Qwen2.5-7B-Instruct).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+ 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- PEFT 0.18.1
+- TRL: 0.23.1
+- Transformers: 4.57.6
+- Pytorch: 2.6.0+cu124
+- Datasets: 4.3.0
+- Tokenizers: 0.22.2
+
+## Citations
+
+
+
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

adapter_config.json

@@ -0,0 +1,51 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen2ForCausalLM",
+    "parent_library": "transformers.models.qwen2.modeling_qwen2",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "unsloth/Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "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": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "down_proj",
+    "up_proj",
+    "k_proj",
+    "v_proj",
+    "gate_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false,
+  "unsloth_training_method": "lora"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54b5ba7c168d81d4e39c777708c8809d508795a2df63b13d56a03e82f4afeecc
+size 161533192

added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|vision_pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|vision_pad|>",
+  "padding_side": "left",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json

@@ -0,0 +1,2595 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 25,
+  "global_step": 248,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00808080808080808,
+      "grad_norm": 0.9913852214813232,
+      "learning_rate": 0.0,
+      "loss": 1.3233,
+      "num_input_tokens_seen": 24048,
+      "step": 1,
+      "train_runtime": 5.5427,
+      "train_tokens_per_second": 4338.668
+    },
+    {
+      "epoch": 0.01616161616161616,
+      "grad_norm": 1.0378535985946655,
+      "learning_rate": 4e-05,
+      "loss": 1.3425,
+      "num_input_tokens_seen": 48096,
+      "step": 2,
+      "train_runtime": 8.8764,
+      "train_tokens_per_second": 5418.441
+    },
+    {
+      "epoch": 0.024242424242424242,
+      "grad_norm": 1.0206665992736816,
+      "learning_rate": 8e-05,
+      "loss": 1.3516,
+      "num_input_tokens_seen": 68416,
+      "step": 3,
+      "train_runtime": 11.7535,
+      "train_tokens_per_second": 5820.919
+    },
+    {
+      "epoch": 0.03232323232323232,
+      "grad_norm": 1.2065541744232178,
+      "learning_rate": 0.00012,
+      "loss": 1.3039,
+      "num_input_tokens_seen": 90432,
+      "step": 4,
+      "train_runtime": 14.8511,
+      "train_tokens_per_second": 6089.231
+    },
+    {
+      "epoch": 0.04040404040404041,
+      "grad_norm": 1.3231337070465088,
+      "learning_rate": 0.00016,
+      "loss": 1.0488,
+      "num_input_tokens_seen": 110544,
+      "step": 5,
+      "train_runtime": 17.7213,
+      "train_tokens_per_second": 6237.921
+    },
+    {
+      "epoch": 0.048484848484848485,
+      "grad_norm": 0.7355687022209167,
+      "learning_rate": 0.0002,
+      "loss": 0.649,
+      "num_input_tokens_seen": 135584,
+      "step": 6,
+      "train_runtime": 21.4984,
+      "train_tokens_per_second": 6306.704
+    },
+    {
+      "epoch": 0.05656565656565657,
+      "grad_norm": 0.7402859926223755,
+      "learning_rate": 0.0001991769547325103,
+      "loss": 0.4857,
+      "num_input_tokens_seen": 159200,
+      "step": 7,
+      "train_runtime": 24.8086,
+      "train_tokens_per_second": 6417.137
+    },
+    {
+      "epoch": 0.06464646464646465,
+      "grad_norm": 0.5103421807289124,
+      "learning_rate": 0.0001983539094650206,
+      "loss": 0.371,
+      "num_input_tokens_seen": 182224,
+      "step": 8,
+      "train_runtime": 28.0753,
+      "train_tokens_per_second": 6490.548
+    },
+    {
+      "epoch": 0.07272727272727272,
+      "grad_norm": 0.3125229477882385,
+      "learning_rate": 0.00019753086419753085,
+      "loss": 0.3324,
+      "num_input_tokens_seen": 205472,
+      "step": 9,
+      "train_runtime": 31.3551,
+      "train_tokens_per_second": 6553.073
+    },
+    {
+      "epoch": 0.08080808080808081,
+      "grad_norm": 0.33310967683792114,
+      "learning_rate": 0.00019670781893004114,
+      "loss": 0.2862,
+      "num_input_tokens_seen": 230400,
+      "step": 10,
+      "train_runtime": 34.8863,
+      "train_tokens_per_second": 6604.316
+    },
+    {
+      "epoch": 0.08888888888888889,
+      "grad_norm": 0.33717748522758484,
+      "learning_rate": 0.00019588477366255146,
+      "loss": 0.2973,
+      "num_input_tokens_seen": 252736,
+      "step": 11,
+      "train_runtime": 38.0785,
+      "train_tokens_per_second": 6637.239
+    },
+    {
+      "epoch": 0.09696969696969697,
+      "grad_norm": 0.383858859539032,
+      "learning_rate": 0.00019506172839506175,
+      "loss": 0.2761,
+      "num_input_tokens_seen": 273584,
+      "step": 12,
+      "train_runtime": 41.108,
+      "train_tokens_per_second": 6655.248
+    },
+    {
+      "epoch": 0.10505050505050505,
+      "grad_norm": 0.33506128191947937,
+      "learning_rate": 0.00019423868312757204,
+      "loss": 0.3092,
+      "num_input_tokens_seen": 299968,
+      "step": 13,
+      "train_runtime": 44.8301,
+      "train_tokens_per_second": 6691.217
+    },
+    {
+      "epoch": 0.11313131313131314,
+      "grad_norm": 0.25691795349121094,
+      "learning_rate": 0.00019341563786008233,
+      "loss": 0.2943,
+      "num_input_tokens_seen": 322480,
+      "step": 14,
+      "train_runtime": 48.0773,
+      "train_tokens_per_second": 6707.535
+    },
+    {
+      "epoch": 0.12121212121212122,
+      "grad_norm": 0.23056800663471222,
+      "learning_rate": 0.0001925925925925926,
+      "loss": 0.2575,
+      "num_input_tokens_seen": 344864,
+      "step": 15,
+      "train_runtime": 51.2647,
+      "train_tokens_per_second": 6727.124
+    },
+    {
+      "epoch": 0.1292929292929293,
+      "grad_norm": 0.2146308422088623,
+      "learning_rate": 0.00019176954732510288,
+      "loss": 0.2441,
+      "num_input_tokens_seen": 369296,
+      "step": 16,
+      "train_runtime": 54.7643,
+      "train_tokens_per_second": 6743.376
+    },
+    {
+      "epoch": 0.13737373737373737,
+      "grad_norm": 0.18066899478435516,
+      "learning_rate": 0.00019094650205761317,
+      "loss": 0.2686,
+      "num_input_tokens_seen": 390944,
+      "step": 17,
+      "train_runtime": 57.902,
+      "train_tokens_per_second": 6751.823
+    },
+    {
+      "epoch": 0.14545454545454545,
+      "grad_norm": 0.18266606330871582,
+      "learning_rate": 0.00019012345679012346,
+      "loss": 0.2484,
+      "num_input_tokens_seen": 414144,
+      "step": 18,
+      "train_runtime": 61.2046,
+      "train_tokens_per_second": 6766.553
+    },
+    {
+      "epoch": 0.15353535353535352,
+      "grad_norm": 0.18596097826957703,
+      "learning_rate": 0.00018930041152263375,
+      "loss": 0.2215,
+      "num_input_tokens_seen": 435840,
+      "step": 19,
+      "train_runtime": 64.3043,
+      "train_tokens_per_second": 6777.778
+    },
+    {
+      "epoch": 0.16161616161616163,
+      "grad_norm": 0.19186538457870483,
+      "learning_rate": 0.00018847736625514404,
+      "loss": 0.2181,
+      "num_input_tokens_seen": 459808,
+      "step": 20,
+      "train_runtime": 67.7619,
+      "train_tokens_per_second": 6785.644
+    },
+    {
+      "epoch": 0.1696969696969697,
+      "grad_norm": 0.1703331023454666,
+      "learning_rate": 0.00018765432098765433,
+      "loss": 0.2441,
+      "num_input_tokens_seen": 479504,
+      "step": 21,
+      "train_runtime": 70.6513,
+      "train_tokens_per_second": 6786.914
+    },
+    {
+      "epoch": 0.17777777777777778,
+      "grad_norm": 0.17710761725902557,
+      "learning_rate": 0.00018683127572016462,
+      "loss": 0.2475,
+      "num_input_tokens_seen": 505712,
+      "step": 22,
+      "train_runtime": 74.4032,
+      "train_tokens_per_second": 6796.911
+    },
+    {
+      "epoch": 0.18585858585858586,
+      "grad_norm": 0.18230487406253815,
+      "learning_rate": 0.0001860082304526749,
+      "loss": 0.2586,
+      "num_input_tokens_seen": 525456,
+      "step": 23,
+      "train_runtime": 77.2642,
+      "train_tokens_per_second": 6800.768
+    },
+    {
+      "epoch": 0.19393939393939394,
+      "grad_norm": 0.16953755915164948,
+      "learning_rate": 0.0001851851851851852,
+      "loss": 0.2548,
+      "num_input_tokens_seen": 548928,
+      "step": 24,
+      "train_runtime": 80.6542,
+      "train_tokens_per_second": 6805.941
+    },
+    {
+      "epoch": 0.20202020202020202,
+      "grad_norm": 0.1729792058467865,
+      "learning_rate": 0.0001843621399176955,
+      "loss": 0.221,
+      "num_input_tokens_seen": 568608,
+      "step": 25,
+      "train_runtime": 83.58,
+      "train_tokens_per_second": 6803.16
+    },
+    {
+      "epoch": 0.20202020202020202,
+      "eval_loss": 0.2138330042362213,
+      "eval_runtime": 15.9812,
+      "eval_samples_per_second": 55.002,
+      "eval_steps_per_second": 3.442,
+      "num_input_tokens_seen": 568608,
+      "step": 25
+    },
+    {
+      "epoch": 0.2101010101010101,
+      "grad_norm": 0.1425217092037201,
+      "learning_rate": 0.00018353909465020578,
+      "loss": 0.2036,
+      "num_input_tokens_seen": 592240,
+      "step": 26,
+      "train_runtime": 103.0213,
+      "train_tokens_per_second": 5748.712
+    },
+    {
+      "epoch": 0.21818181818181817,
+      "grad_norm": 0.1629820615053177,
+      "learning_rate": 0.00018271604938271605,
+      "loss": 0.1973,
+      "num_input_tokens_seen": 613168,
+      "step": 27,
+      "train_runtime": 106.1403,
+      "train_tokens_per_second": 5776.96
+    },
+    {
+      "epoch": 0.22626262626262628,
+      "grad_norm": 0.13308098912239075,
+      "learning_rate": 0.00018189300411522634,
+      "loss": 0.2178,
+      "num_input_tokens_seen": 637552,
+      "step": 28,
+      "train_runtime": 109.6485,
+      "train_tokens_per_second": 5814.508
+    },
+    {
+      "epoch": 0.23434343434343435,
+      "grad_norm": 0.13057714700698853,
+      "learning_rate": 0.00018106995884773663,
+      "loss": 0.2244,
+      "num_input_tokens_seen": 661184,
+      "step": 29,
+      "train_runtime": 113.0702,
+      "train_tokens_per_second": 5847.551
+    },
+    {
+      "epoch": 0.24242424242424243,
+      "grad_norm": 0.15625949203968048,
+      "learning_rate": 0.00018024691358024692,
+      "loss": 0.2448,
+      "num_input_tokens_seen": 683072,
+      "step": 30,
+      "train_runtime": 116.2536,
+      "train_tokens_per_second": 5875.704
+    },
+    {
+      "epoch": 0.2505050505050505,
+      "grad_norm": 0.1479414701461792,
+      "learning_rate": 0.0001794238683127572,
+      "loss": 0.2132,
+      "num_input_tokens_seen": 703776,
+      "step": 31,
+      "train_runtime": 119.8681,
+      "train_tokens_per_second": 5871.251
+    },
+    {
+      "epoch": 0.2585858585858586,
+      "grad_norm": 0.12871505320072174,
+      "learning_rate": 0.0001786008230452675,
+      "loss": 0.2193,
+      "num_input_tokens_seen": 729584,
+      "step": 32,
+      "train_runtime": 123.5834,
+      "train_tokens_per_second": 5903.574
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.13625787198543549,
+      "learning_rate": 0.00017777777777777779,
+      "loss": 0.2572,
+      "num_input_tokens_seen": 752672,
+      "step": 33,
+      "train_runtime": 126.9484,
+      "train_tokens_per_second": 5928.961
+    },
+    {
+      "epoch": 0.27474747474747474,
+      "grad_norm": 0.13514648377895355,
+      "learning_rate": 0.00017695473251028808,
+      "loss": 0.1983,
+      "num_input_tokens_seen": 776672,
+      "step": 34,
+      "train_runtime": 130.4118,
+      "train_tokens_per_second": 5955.537
+    },
+    {
+      "epoch": 0.2828282828282828,
+      "grad_norm": 0.13971051573753357,
+      "learning_rate": 0.00017613168724279837,
+      "loss": 0.2283,
+      "num_input_tokens_seen": 801296,
+      "step": 35,
+      "train_runtime": 133.967,
+      "train_tokens_per_second": 5981.294
+    },
+    {
+      "epoch": 0.2909090909090909,
+      "grad_norm": 0.1429297775030136,
+      "learning_rate": 0.00017530864197530866,
+      "loss": 0.199,
+      "num_input_tokens_seen": 822368,
+      "step": 36,
+      "train_runtime": 137.0476,
+      "train_tokens_per_second": 6000.599
+    },
+    {
+      "epoch": 0.298989898989899,
+      "grad_norm": 0.13625267148017883,
+      "learning_rate": 0.00017448559670781895,
+      "loss": 0.1851,
+      "num_input_tokens_seen": 848928,
+      "step": 37,
+      "train_runtime": 140.8511,
+      "train_tokens_per_second": 6027.132
+    },
+    {
+      "epoch": 0.30707070707070705,
+      "grad_norm": 0.13639648258686066,
+      "learning_rate": 0.00017366255144032924,
+      "loss": 0.1784,
+      "num_input_tokens_seen": 870672,
+      "step": 38,
+      "train_runtime": 143.9915,
+      "train_tokens_per_second": 6046.692
+    },
+    {
+      "epoch": 0.3151515151515151,
+      "grad_norm": 0.15709735453128815,
+      "learning_rate": 0.0001728395061728395,
+      "loss": 0.2379,
+      "num_input_tokens_seen": 892064,
+      "step": 39,
+      "train_runtime": 147.1192,
+      "train_tokens_per_second": 6063.547
+    },
+    {
+      "epoch": 0.32323232323232326,
+      "grad_norm": 0.13749410212039948,
+      "learning_rate": 0.0001720164609053498,
+      "loss": 0.2086,
+      "num_input_tokens_seen": 915680,
+      "step": 40,
+      "train_runtime": 150.5405,
+      "train_tokens_per_second": 6082.617
+    },
+    {
+      "epoch": 0.33131313131313134,
+      "grad_norm": 0.15497097373008728,
+      "learning_rate": 0.00017119341563786008,
+      "loss": 0.2155,
+      "num_input_tokens_seen": 935216,
+      "step": 41,
+      "train_runtime": 153.3826,
+      "train_tokens_per_second": 6097.277
+    },
+    {
+      "epoch": 0.3393939393939394,
+      "grad_norm": 0.14246957004070282,
+      "learning_rate": 0.00017037037037037037,
+      "loss": 0.1926,
+      "num_input_tokens_seen": 956864,
+      "step": 42,
+      "train_runtime": 156.5384,
+      "train_tokens_per_second": 6112.648
+    },
+    {
+      "epoch": 0.3474747474747475,
+      "grad_norm": 0.1377769261598587,
+      "learning_rate": 0.00016954732510288066,
+      "loss": 0.2117,
+      "num_input_tokens_seen": 981440,
+      "step": 43,
+      "train_runtime": 160.0454,
+      "train_tokens_per_second": 6132.259
+    },
+    {
+      "epoch": 0.35555555555555557,
+      "grad_norm": 0.1453401744365692,
+      "learning_rate": 0.00016872427983539098,
+      "loss": 0.1787,
+      "num_input_tokens_seen": 1001232,
+      "step": 44,
+      "train_runtime": 162.9716,
+      "train_tokens_per_second": 6143.599
+    },
+    {
+      "epoch": 0.36363636363636365,
+      "grad_norm": 0.13512200117111206,
+      "learning_rate": 0.00016790123456790124,
+      "loss": 0.1778,
+      "num_input_tokens_seen": 1027680,
+      "step": 45,
+      "train_runtime": 166.7953,
+      "train_tokens_per_second": 6161.325
+    },
+    {
+      "epoch": 0.3717171717171717,
+      "grad_norm": 0.14032725989818573,
+      "learning_rate": 0.00016707818930041153,
+      "loss": 0.2622,
+      "num_input_tokens_seen": 1049968,
+      "step": 46,
+      "train_runtime": 170.0243,
+      "train_tokens_per_second": 6175.401
+    },
+    {
+      "epoch": 0.3797979797979798,
+      "grad_norm": 0.13695235550403595,
+      "learning_rate": 0.00016625514403292182,
+      "loss": 0.2172,
+      "num_input_tokens_seen": 1071840,
+      "step": 47,
+      "train_runtime": 173.2045,
+      "train_tokens_per_second": 6188.291
+    },
+    {
+      "epoch": 0.3878787878787879,
+      "grad_norm": 0.14571040868759155,
+      "learning_rate": 0.0001654320987654321,
+      "loss": 0.1682,
+      "num_input_tokens_seen": 1092864,
+      "step": 48,
+      "train_runtime": 176.2872,
+      "train_tokens_per_second": 6199.339
+    },
+    {
+      "epoch": 0.39595959595959596,
+      "grad_norm": 0.13497225940227509,
+      "learning_rate": 0.0001646090534979424,
+      "loss": 0.2167,
+      "num_input_tokens_seen": 1117040,
+      "step": 49,
+      "train_runtime": 179.7447,
+      "train_tokens_per_second": 6214.592
+    },
+    {
+      "epoch": 0.40404040404040403,
+      "grad_norm": 0.13479158282279968,
+      "learning_rate": 0.0001637860082304527,
+      "loss": 0.2188,
+      "num_input_tokens_seen": 1140240,
+      "step": 50,
+      "train_runtime": 183.112,
+      "train_tokens_per_second": 6227.007
+    },
+    {
+      "epoch": 0.40404040404040403,
+      "eval_loss": 0.19270148873329163,
+      "eval_runtime": 15.9678,
+      "eval_samples_per_second": 55.048,
+      "eval_steps_per_second": 3.444,
+      "num_input_tokens_seen": 1140240,
+      "step": 50
+    },
+    {
+      "epoch": 0.4121212121212121,
+      "grad_norm": 0.14810238778591156,
+      "learning_rate": 0.00016296296296296295,
+      "loss": 0.1908,
+      "num_input_tokens_seen": 1161104,
+      "step": 51,
+      "train_runtime": 202.1269,
+      "train_tokens_per_second": 5744.431
+    },
+    {
+      "epoch": 0.4202020202020202,
+      "grad_norm": 0.12631717324256897,
+      "learning_rate": 0.00016213991769547324,
+      "loss": 0.2194,
+      "num_input_tokens_seen": 1189328,
+      "step": 52,
+      "train_runtime": 206.1876,
+      "train_tokens_per_second": 5768.183
+    },
+    {
+      "epoch": 0.42828282828282827,
+      "grad_norm": 0.12934885919094086,
+      "learning_rate": 0.00016131687242798353,
+      "loss": 0.1682,
+      "num_input_tokens_seen": 1211120,
+      "step": 53,
+      "train_runtime": 209.3667,
+      "train_tokens_per_second": 5784.684
+    },
+    {
+      "epoch": 0.43636363636363634,
+      "grad_norm": 0.1369091421365738,
+      "learning_rate": 0.00016049382716049385,
+      "loss": 0.1882,
+      "num_input_tokens_seen": 1234080,
+      "step": 54,
+      "train_runtime": 212.7157,
+      "train_tokens_per_second": 5801.546
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 0.14891834557056427,
+      "learning_rate": 0.00015967078189300414,
+      "loss": 0.1898,
+      "num_input_tokens_seen": 1256880,
+      "step": 55,
+      "train_runtime": 216.0146,
+      "train_tokens_per_second": 5818.495
+    },
+    {
+      "epoch": 0.45252525252525255,
+      "grad_norm": 0.15192829072475433,
+      "learning_rate": 0.00015884773662551443,
+      "loss": 0.2452,
+      "num_input_tokens_seen": 1281376,
+      "step": 56,
+      "train_runtime": 219.5282,
+      "train_tokens_per_second": 5836.954
+    },
+    {
+      "epoch": 0.46060606060606063,
+      "grad_norm": 0.13381753861904144,
+      "learning_rate": 0.0001580246913580247,
+      "loss": 0.1983,
+      "num_input_tokens_seen": 1309216,
+      "step": 57,
+      "train_runtime": 223.5427,
+      "train_tokens_per_second": 5856.67
+    },
+    {
+      "epoch": 0.4686868686868687,
+      "grad_norm": 0.14824049174785614,
+      "learning_rate": 0.00015720164609053498,
+      "loss": 0.2358,
+      "num_input_tokens_seen": 1331120,
+      "step": 58,
+      "train_runtime": 226.7435,
+      "train_tokens_per_second": 5870.599
+    },
+    {
+      "epoch": 0.4767676767676768,
+      "grad_norm": 0.15301313996315002,
+      "learning_rate": 0.00015637860082304527,
+      "loss": 0.1871,
+      "num_input_tokens_seen": 1354976,
+      "step": 59,
+      "train_runtime": 230.1754,
+      "train_tokens_per_second": 5886.711
+    },
+    {
+      "epoch": 0.48484848484848486,
+      "grad_norm": 0.1360855996608734,
+      "learning_rate": 0.00015555555555555556,
+      "loss": 0.2078,
+      "num_input_tokens_seen": 1376976,
+      "step": 60,
+      "train_runtime": 233.3715,
+      "train_tokens_per_second": 5900.36
+    },
+    {
+      "epoch": 0.49292929292929294,
+      "grad_norm": 0.1479628086090088,
+      "learning_rate": 0.00015473251028806585,
+      "loss": 0.2039,
+      "num_input_tokens_seen": 1397952,
+      "step": 61,
+      "train_runtime": 236.9762,
+      "train_tokens_per_second": 5899.125
+    },
+    {
+      "epoch": 0.501010101010101,
+      "grad_norm": 0.13906534016132355,
+      "learning_rate": 0.00015390946502057614,
+      "loss": 0.1926,
+      "num_input_tokens_seen": 1417712,
+      "step": 62,
+      "train_runtime": 239.842,
+      "train_tokens_per_second": 5911.024
+    },
+    {
+      "epoch": 0.509090909090909,
+      "grad_norm": 0.1498580425977707,
+      "learning_rate": 0.0001530864197530864,
+      "loss": 0.1674,
+      "num_input_tokens_seen": 1438544,
+      "step": 63,
+      "train_runtime": 242.9024,
+      "train_tokens_per_second": 5922.312
+    },
+    {
+      "epoch": 0.5171717171717172,
+      "grad_norm": 0.13896457850933075,
+      "learning_rate": 0.00015226337448559672,
+      "loss": 0.1779,
+      "num_input_tokens_seen": 1461776,
+      "step": 64,
+      "train_runtime": 246.286,
+      "train_tokens_per_second": 5935.278
+    },
+    {
+      "epoch": 0.5252525252525253,
+      "grad_norm": 0.15526247024536133,
+      "learning_rate": 0.000151440329218107,
+      "loss": 0.2209,
+      "num_input_tokens_seen": 1483600,
+      "step": 65,
+      "train_runtime": 249.4086,
+      "train_tokens_per_second": 5948.472
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.15172874927520752,
+      "learning_rate": 0.0001506172839506173,
+      "loss": 0.1939,
+      "num_input_tokens_seen": 1507280,
+      "step": 66,
+      "train_runtime": 252.8239,
+      "train_tokens_per_second": 5961.777
+    },
+    {
+      "epoch": 0.5414141414141415,
+      "grad_norm": 0.15608155727386475,
+      "learning_rate": 0.0001497942386831276,
+      "loss": 0.2115,
+      "num_input_tokens_seen": 1530176,
+      "step": 67,
+      "train_runtime": 256.1302,
+      "train_tokens_per_second": 5974.211
+    },
+    {
+      "epoch": 0.5494949494949495,
+      "grad_norm": 0.13221146166324615,
+      "learning_rate": 0.00014897119341563788,
+      "loss": 0.1864,
+      "num_input_tokens_seen": 1553520,
+      "step": 68,
+      "train_runtime": 259.5354,
+      "train_tokens_per_second": 5985.772
+    },
+    {
+      "epoch": 0.5575757575757576,
+      "grad_norm": 0.13630913197994232,
+      "learning_rate": 0.00014814814814814815,
+      "loss": 0.1888,
+      "num_input_tokens_seen": 1577728,
+      "step": 69,
+      "train_runtime": 263.0794,
+      "train_tokens_per_second": 5997.155
+    },
+    {
+      "epoch": 0.5656565656565656,
+      "grad_norm": 0.1588101089000702,
+      "learning_rate": 0.00014732510288065844,
+      "loss": 0.1622,
+      "num_input_tokens_seen": 1598448,
+      "step": 70,
+      "train_runtime": 266.1041,
+      "train_tokens_per_second": 6006.852
+    },
+    {
+      "epoch": 0.5737373737373738,
+      "grad_norm": 0.1479269117116928,
+      "learning_rate": 0.00014650205761316873,
+      "loss": 0.1665,
+      "num_input_tokens_seen": 1619616,
+      "step": 71,
+      "train_runtime": 269.2171,
+      "train_tokens_per_second": 6016.021
+    },
+    {
+      "epoch": 0.5818181818181818,
+      "grad_norm": 0.1633758842945099,
+      "learning_rate": 0.00014567901234567902,
+      "loss": 0.2066,
+      "num_input_tokens_seen": 1647504,
+      "step": 72,
+      "train_runtime": 273.2086,
+      "train_tokens_per_second": 6030.206
+    },
+    {
+      "epoch": 0.5898989898989899,
+      "grad_norm": 0.1569192260503769,
+      "learning_rate": 0.0001448559670781893,
+      "loss": 0.2209,
+      "num_input_tokens_seen": 1669600,
+      "step": 73,
+      "train_runtime": 276.4483,
+      "train_tokens_per_second": 6039.465
+    },
+    {
+      "epoch": 0.597979797979798,
+      "grad_norm": 0.15907315909862518,
+      "learning_rate": 0.0001440329218106996,
+      "loss": 0.203,
+      "num_input_tokens_seen": 1692464,
+      "step": 74,
+      "train_runtime": 279.7549,
+      "train_tokens_per_second": 6049.811
+    },
+    {
+      "epoch": 0.6060606060606061,
+      "grad_norm": 0.15339483320713043,
+      "learning_rate": 0.00014320987654320989,
+      "loss": 0.1902,
+      "num_input_tokens_seen": 1714544,
+      "step": 75,
+      "train_runtime": 282.9625,
+      "train_tokens_per_second": 6059.263
+    },
+    {
+      "epoch": 0.6060606060606061,
+      "eval_loss": 0.18341881036758423,
+      "eval_runtime": 15.9903,
+      "eval_samples_per_second": 54.971,
+      "eval_steps_per_second": 3.44,
+      "num_input_tokens_seen": 1714544,
+      "step": 75
+    },
+    {
+      "epoch": 0.6141414141414141,
+      "grad_norm": 0.1519935429096222,
+      "learning_rate": 0.00014238683127572018,
+      "loss": 0.1774,
+      "num_input_tokens_seen": 1734992,
+      "step": 76,
+      "train_runtime": 301.9584,
+      "train_tokens_per_second": 5745.799
+    },
+    {
+      "epoch": 0.6222222222222222,
+      "grad_norm": 0.1523190289735794,
+      "learning_rate": 0.00014156378600823047,
+      "loss": 0.188,
+      "num_input_tokens_seen": 1758528,
+      "step": 77,
+      "train_runtime": 305.3612,
+      "train_tokens_per_second": 5758.846
+    },
+    {
+      "epoch": 0.6303030303030303,
+      "grad_norm": 0.16023258864879608,
+      "learning_rate": 0.00014074074074074076,
+      "loss": 0.2015,
+      "num_input_tokens_seen": 1782768,
+      "step": 78,
+      "train_runtime": 308.8338,
+      "train_tokens_per_second": 5772.58
+    },
+    {
+      "epoch": 0.6383838383838384,
+      "grad_norm": 0.15232358872890472,
+      "learning_rate": 0.00013991769547325105,
+      "loss": 0.1844,
+      "num_input_tokens_seen": 1806000,
+      "step": 79,
+      "train_runtime": 312.204,
+      "train_tokens_per_second": 5784.679
+    },
+    {
+      "epoch": 0.6464646464646465,
+      "grad_norm": 0.15448316931724548,
+      "learning_rate": 0.00013909465020576134,
+      "loss": 0.1729,
+      "num_input_tokens_seen": 1833264,
+      "step": 80,
+      "train_runtime": 316.1144,
+      "train_tokens_per_second": 5799.368
+    },
+    {
+      "epoch": 0.6545454545454545,
+      "grad_norm": 0.16937431693077087,
+      "learning_rate": 0.0001382716049382716,
+      "loss": 0.2037,
+      "num_input_tokens_seen": 1855920,
+      "step": 81,
+      "train_runtime": 319.4445,
+      "train_tokens_per_second": 5809.835
+    },
+    {
+      "epoch": 0.6626262626262627,
+      "grad_norm": 0.1692294478416443,
+      "learning_rate": 0.0001374485596707819,
+      "loss": 0.2102,
+      "num_input_tokens_seen": 1877664,
+      "step": 82,
+      "train_runtime": 322.623,
+      "train_tokens_per_second": 5819.994
+    },
+    {
+      "epoch": 0.6707070707070707,
+      "grad_norm": 0.15476073324680328,
+      "learning_rate": 0.00013662551440329218,
+      "loss": 0.1601,
+      "num_input_tokens_seen": 1901152,
+      "step": 83,
+      "train_runtime": 326.007,
+      "train_tokens_per_second": 5831.63
+    },
+    {
+      "epoch": 0.6787878787878788,
+      "grad_norm": 0.1650950163602829,
+      "learning_rate": 0.00013580246913580247,
+      "loss": 0.2074,
+      "num_input_tokens_seen": 1927440,
+      "step": 84,
+      "train_runtime": 329.7391,
+      "train_tokens_per_second": 5845.348
+    },
+    {
+      "epoch": 0.6868686868686869,
+      "grad_norm": 0.17345544695854187,
+      "learning_rate": 0.00013497942386831276,
+      "loss": 0.1916,
+      "num_input_tokens_seen": 1951408,
+      "step": 85,
+      "train_runtime": 333.1884,
+      "train_tokens_per_second": 5856.771
+    },
+    {
+      "epoch": 0.694949494949495,
+      "grad_norm": 0.16704945266246796,
+      "learning_rate": 0.00013415637860082305,
+      "loss": 0.2071,
+      "num_input_tokens_seen": 1973200,
+      "step": 86,
+      "train_runtime": 336.3437,
+      "train_tokens_per_second": 5866.619
+    },
+    {
+      "epoch": 0.703030303030303,
+      "grad_norm": 0.14658761024475098,
+      "learning_rate": 0.00013333333333333334,
+      "loss": 0.228,
+      "num_input_tokens_seen": 1994528,
+      "step": 87,
+      "train_runtime": 339.4404,
+      "train_tokens_per_second": 5875.93
+    },
+    {
+      "epoch": 0.7111111111111111,
+      "grad_norm": 0.17460143566131592,
+      "learning_rate": 0.00013251028806584363,
+      "loss": 0.1777,
+      "num_input_tokens_seen": 2019248,
+      "step": 88,
+      "train_runtime": 343.0056,
+      "train_tokens_per_second": 5886.924
+    },
+    {
+      "epoch": 0.7191919191919192,
+      "grad_norm": 0.1555347740650177,
+      "learning_rate": 0.00013168724279835392,
+      "loss": 0.1718,
+      "num_input_tokens_seen": 2040592,
+      "step": 89,
+      "train_runtime": 346.1188,
+      "train_tokens_per_second": 5895.641
+    },
+    {
+      "epoch": 0.7272727272727273,
+      "grad_norm": 0.14311373233795166,
+      "learning_rate": 0.0001308641975308642,
+      "loss": 0.1703,
+      "num_input_tokens_seen": 2062032,
+      "step": 90,
+      "train_runtime": 349.2336,
+      "train_tokens_per_second": 5904.449
+    },
+    {
+      "epoch": 0.7353535353535353,
+      "grad_norm": 0.15284591913223267,
+      "learning_rate": 0.0001300411522633745,
+      "loss": 0.1685,
+      "num_input_tokens_seen": 2082368,
+      "step": 91,
+      "train_runtime": 352.7694,
+      "train_tokens_per_second": 5902.915
+    },
+    {
+      "epoch": 0.7434343434343434,
+      "grad_norm": 0.155190110206604,
+      "learning_rate": 0.00012921810699588476,
+      "loss": 0.1604,
+      "num_input_tokens_seen": 2104976,
+      "step": 92,
+      "train_runtime": 356.0443,
+      "train_tokens_per_second": 5912.118
+    },
+    {
+      "epoch": 0.7515151515151515,
+      "grad_norm": 0.15990827977657318,
+      "learning_rate": 0.00012839506172839505,
+      "loss": 0.1926,
+      "num_input_tokens_seen": 2125664,
+      "step": 93,
+      "train_runtime": 359.04,
+      "train_tokens_per_second": 5920.41
+    },
+    {
+      "epoch": 0.7595959595959596,
+      "grad_norm": 0.14356884360313416,
+      "learning_rate": 0.00012757201646090534,
+      "loss": 0.157,
+      "num_input_tokens_seen": 2151216,
+      "step": 94,
+      "train_runtime": 362.7537,
+      "train_tokens_per_second": 5930.238
+    },
+    {
+      "epoch": 0.7676767676767676,
+      "grad_norm": 0.16834881901741028,
+      "learning_rate": 0.00012674897119341563,
+      "loss": 0.1684,
+      "num_input_tokens_seen": 2174480,
+      "step": 95,
+      "train_runtime": 366.0908,
+      "train_tokens_per_second": 5939.729
+    },
+    {
+      "epoch": 0.7757575757575758,
+      "grad_norm": 0.15969020128250122,
+      "learning_rate": 0.00012592592592592592,
+      "loss": 0.1817,
+      "num_input_tokens_seen": 2198048,
+      "step": 96,
+      "train_runtime": 369.4885,
+      "train_tokens_per_second": 5948.894
+    },
+    {
+      "epoch": 0.7838383838383839,
+      "grad_norm": 0.17647744715213776,
+      "learning_rate": 0.00012510288065843624,
+      "loss": 0.1925,
+      "num_input_tokens_seen": 2218944,
+      "step": 97,
+      "train_runtime": 372.5447,
+      "train_tokens_per_second": 5956.182
+    },
+    {
+      "epoch": 0.7919191919191919,
+      "grad_norm": 0.15452450513839722,
+      "learning_rate": 0.0001242798353909465,
+      "loss": 0.2199,
+      "num_input_tokens_seen": 2244752,
+      "step": 98,
+      "train_runtime": 376.2965,
+      "train_tokens_per_second": 5965.381
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.16536033153533936,
+      "learning_rate": 0.0001234567901234568,
+      "loss": 0.202,
+      "num_input_tokens_seen": 2267728,
+      "step": 99,
+      "train_runtime": 379.6467,
+      "train_tokens_per_second": 5973.259
+    },
+    {
+      "epoch": 0.8080808080808081,
+      "grad_norm": 0.15133647620677948,
+      "learning_rate": 0.00012263374485596708,
+      "loss": 0.177,
+      "num_input_tokens_seen": 2289328,
+      "step": 100,
+      "train_runtime": 382.7931,
+      "train_tokens_per_second": 5980.589
+    },
+    {
+      "epoch": 0.8080808080808081,
+      "eval_loss": 0.17801880836486816,
+      "eval_runtime": 15.9675,
+      "eval_samples_per_second": 55.049,
+      "eval_steps_per_second": 3.444,
+      "num_input_tokens_seen": 2289328,
+      "step": 100
+    },
+    {
+      "epoch": 0.8161616161616162,
+      "grad_norm": 0.1537664383649826,
+      "learning_rate": 0.00012181069958847737,
+      "loss": 0.2092,
+      "num_input_tokens_seen": 2313216,
+      "step": 101,
+      "train_runtime": 402.2143,
+      "train_tokens_per_second": 5751.203
+    },
+    {
+      "epoch": 0.8242424242424242,
+      "grad_norm": 0.15654610097408295,
+      "learning_rate": 0.00012098765432098766,
+      "loss": 0.1664,
+      "num_input_tokens_seen": 2333152,
+      "step": 102,
+      "train_runtime": 405.1177,
+      "train_tokens_per_second": 5759.195
+    },
+    {
+      "epoch": 0.8323232323232324,
+      "grad_norm": 0.15154822170734406,
+      "learning_rate": 0.00012016460905349795,
+      "loss": 0.2069,
+      "num_input_tokens_seen": 2355744,
+      "step": 103,
+      "train_runtime": 408.4074,
+      "train_tokens_per_second": 5768.123
+    },
+    {
+      "epoch": 0.8404040404040404,
+      "grad_norm": 0.16599127650260925,
+      "learning_rate": 0.00011934156378600823,
+      "loss": 0.2347,
+      "num_input_tokens_seen": 2376560,
+      "step": 104,
+      "train_runtime": 411.4246,
+      "train_tokens_per_second": 5776.418
+    },
+    {
+      "epoch": 0.8484848484848485,
+      "grad_norm": 0.1642574518918991,
+      "learning_rate": 0.00011851851851851852,
+      "loss": 0.185,
+      "num_input_tokens_seen": 2400896,
+      "step": 105,
+      "train_runtime": 414.9439,
+      "train_tokens_per_second": 5786.074
+    },
+    {
+      "epoch": 0.8565656565656565,
+      "grad_norm": 0.15395912528038025,
+      "learning_rate": 0.00011769547325102881,
+      "loss": 0.179,
+      "num_input_tokens_seen": 2423632,
+      "step": 106,
+      "train_runtime": 418.2763,
+      "train_tokens_per_second": 5794.332
+    },
+    {
+      "epoch": 0.8646464646464647,
+      "grad_norm": 0.16491572558879852,
+      "learning_rate": 0.0001168724279835391,
+      "loss": 0.2168,
+      "num_input_tokens_seen": 2448432,
+      "step": 107,
+      "train_runtime": 421.8552,
+      "train_tokens_per_second": 5803.963
+    },
+    {
+      "epoch": 0.8727272727272727,
+      "grad_norm": 0.16017423570156097,
+      "learning_rate": 0.00011604938271604939,
+      "loss": 0.1826,
+      "num_input_tokens_seen": 2470768,
+      "step": 108,
+      "train_runtime": 425.0496,
+      "train_tokens_per_second": 5812.893
+    },
+    {
+      "epoch": 0.8808080808080808,
+      "grad_norm": 0.14728578925132751,
+      "learning_rate": 0.00011522633744855968,
+      "loss": 0.1907,
+      "num_input_tokens_seen": 2494944,
+      "step": 109,
+      "train_runtime": 428.5398,
+      "train_tokens_per_second": 5821.965
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 0.17022450268268585,
+      "learning_rate": 0.00011440329218106996,
+      "loss": 0.2021,
+      "num_input_tokens_seen": 2515760,
+      "step": 110,
+      "train_runtime": 431.6001,
+      "train_tokens_per_second": 5828.914
+    },
+    {
+      "epoch": 0.896969696969697,
+      "grad_norm": 0.16193877160549164,
+      "learning_rate": 0.00011358024691358025,
+      "loss": 0.1559,
+      "num_input_tokens_seen": 2537792,
+      "step": 111,
+      "train_runtime": 434.8205,
+      "train_tokens_per_second": 5836.414
+    },
+    {
+      "epoch": 0.9050505050505051,
+      "grad_norm": 0.14079837501049042,
+      "learning_rate": 0.00011275720164609054,
+      "loss": 0.1667,
+      "num_input_tokens_seen": 2562032,
+      "step": 112,
+      "train_runtime": 438.3575,
+      "train_tokens_per_second": 5844.618
+    },
+    {
+      "epoch": 0.9131313131313131,
+      "grad_norm": 0.1725139021873474,
+      "learning_rate": 0.00011193415637860083,
+      "loss": 0.1913,
+      "num_input_tokens_seen": 2583760,
+      "step": 113,
+      "train_runtime": 441.5306,
+      "train_tokens_per_second": 5851.826
+    },
+    {
+      "epoch": 0.9212121212121213,
+      "grad_norm": 0.1663820445537567,
+      "learning_rate": 0.00011111111111111112,
+      "loss": 0.1441,
+      "num_input_tokens_seen": 2607776,
+      "step": 114,
+      "train_runtime": 445.0017,
+      "train_tokens_per_second": 5860.149
+    },
+    {
+      "epoch": 0.9292929292929293,
+      "grad_norm": 0.16796670854091644,
+      "learning_rate": 0.0001102880658436214,
+      "loss": 0.2076,
+      "num_input_tokens_seen": 2630208,
+      "step": 115,
+      "train_runtime": 448.275,
+      "train_tokens_per_second": 5867.398
+    },
+    {
+      "epoch": 0.9373737373737374,
+      "grad_norm": 0.16379429399967194,
+      "learning_rate": 0.00010946502057613168,
+      "loss": 0.1832,
+      "num_input_tokens_seen": 2650208,
+      "step": 116,
+      "train_runtime": 451.1977,
+      "train_tokens_per_second": 5873.718
+    },
+    {
+      "epoch": 0.9454545454545454,
+      "grad_norm": 0.1663283109664917,
+      "learning_rate": 0.00010864197530864197,
+      "loss": 0.2013,
+      "num_input_tokens_seen": 2672592,
+      "step": 117,
+      "train_runtime": 454.4485,
+      "train_tokens_per_second": 5880.957
+    },
+    {
+      "epoch": 0.9535353535353536,
+      "grad_norm": 0.16695767641067505,
+      "learning_rate": 0.00010781893004115226,
+      "loss": 0.1462,
+      "num_input_tokens_seen": 2698064,
+      "step": 118,
+      "train_runtime": 458.1001,
+      "train_tokens_per_second": 5889.682
+    },
+    {
+      "epoch": 0.9616161616161616,
+      "grad_norm": 0.1588376760482788,
+      "learning_rate": 0.00010699588477366255,
+      "loss": 0.1837,
+      "num_input_tokens_seen": 2719680,
+      "step": 119,
+      "train_runtime": 461.2512,
+      "train_tokens_per_second": 5896.31
+    },
+    {
+      "epoch": 0.9696969696969697,
+      "grad_norm": 0.1472574770450592,
+      "learning_rate": 0.00010617283950617284,
+      "loss": 0.1742,
+      "num_input_tokens_seen": 2742400,
+      "step": 120,
+      "train_runtime": 464.5791,
+      "train_tokens_per_second": 5902.978
+    },
+    {
+      "epoch": 0.9777777777777777,
+      "grad_norm": 0.17550979554653168,
+      "learning_rate": 0.00010534979423868315,
+      "loss": 0.1364,
+      "num_input_tokens_seen": 2766112,
+      "step": 121,
+      "train_runtime": 468.5285,
+      "train_tokens_per_second": 5903.829
+    },
+    {
+      "epoch": 0.9858585858585859,
+      "grad_norm": 0.15758642554283142,
+      "learning_rate": 0.00010452674897119341,
+      "loss": 0.1775,
+      "num_input_tokens_seen": 2787952,
+      "step": 122,
+      "train_runtime": 471.7104,
+      "train_tokens_per_second": 5910.304
+    },
+    {
+      "epoch": 0.9939393939393939,
+      "grad_norm": 0.16019918024539948,
+      "learning_rate": 0.0001037037037037037,
+      "loss": 0.2096,
+      "num_input_tokens_seen": 2811632,
+      "step": 123,
+      "train_runtime": 475.1248,
+      "train_tokens_per_second": 5917.671
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.2007690817117691,
+      "learning_rate": 0.00010288065843621399,
+      "loss": 0.1646,
+      "num_input_tokens_seen": 2825206,
+      "step": 124,
+      "train_runtime": 477.1077,
+      "train_tokens_per_second": 5921.527
+    },
+    {
+      "epoch": 1.0080808080808081,
+      "grad_norm": 0.14078141748905182,
+      "learning_rate": 0.00010205761316872428,
+      "loss": 0.1363,
+      "num_input_tokens_seen": 2847206,
+      "step": 125,
+      "train_runtime": 480.3107,
+      "train_tokens_per_second": 5927.842
+    },
+    {
+      "epoch": 1.0080808080808081,
+      "eval_loss": 0.1743689626455307,
+      "eval_runtime": 15.9727,
+      "eval_samples_per_second": 55.031,
+      "eval_steps_per_second": 3.443,
+      "num_input_tokens_seen": 2847206,
+      "step": 125
+    },
+    {
+      "epoch": 1.0161616161616163,
+      "grad_norm": 0.14959807693958282,
+      "learning_rate": 0.00010123456790123458,
+      "loss": 0.1373,
+      "num_input_tokens_seen": 2867190,
+      "step": 126,
+      "train_runtime": 499.2365,
+      "train_tokens_per_second": 5743.15
+    },
+    {
+      "epoch": 1.0242424242424242,
+      "grad_norm": 0.14410553872585297,
+      "learning_rate": 0.00010041152263374487,
+      "loss": 0.1391,
+      "num_input_tokens_seen": 2888710,
+      "step": 127,
+      "train_runtime": 502.361,
+      "train_tokens_per_second": 5750.268
+    },
+    {
+      "epoch": 1.0323232323232323,
+      "grad_norm": 0.14210668206214905,
+      "learning_rate": 9.958847736625515e-05,
+      "loss": 0.1514,
+      "num_input_tokens_seen": 2913990,
+      "step": 128,
+      "train_runtime": 506.0209,
+      "train_tokens_per_second": 5758.636
+    },
+    {
+      "epoch": 1.0404040404040404,
+      "grad_norm": 0.14086109399795532,
+      "learning_rate": 9.876543209876543e-05,
+      "loss": 0.1384,
+      "num_input_tokens_seen": 2938246,
+      "step": 129,
+      "train_runtime": 509.5108,
+      "train_tokens_per_second": 5766.799
+    },
+    {
+      "epoch": 1.0484848484848486,
+      "grad_norm": 0.1444181203842163,
+      "learning_rate": 9.794238683127573e-05,
+      "loss": 0.1694,
+      "num_input_tokens_seen": 2962326,
+      "step": 130,
+      "train_runtime": 513.0283,
+      "train_tokens_per_second": 5774.196
+    },
+    {
+      "epoch": 1.0565656565656565,
+      "grad_norm": 0.15412774682044983,
+      "learning_rate": 9.711934156378602e-05,
+      "loss": 0.1657,
+      "num_input_tokens_seen": 2983126,
+      "step": 131,
+      "train_runtime": 516.07,
+      "train_tokens_per_second": 5780.468
+    },
+    {
+      "epoch": 1.0646464646464646,
+      "grad_norm": 0.1612967848777771,
+      "learning_rate": 9.62962962962963e-05,
+      "loss": 0.1738,
+      "num_input_tokens_seen": 3005606,
+      "step": 132,
+      "train_runtime": 519.3068,
+      "train_tokens_per_second": 5787.727
+    },
+    {
+      "epoch": 1.0727272727272728,
+      "grad_norm": 0.1550590842962265,
+      "learning_rate": 9.547325102880659e-05,
+      "loss": 0.1634,
+      "num_input_tokens_seen": 3027926,
+      "step": 133,
+      "train_runtime": 522.5599,
+      "train_tokens_per_second": 5794.41
+    },
+    {
+      "epoch": 1.0808080808080809,
+      "grad_norm": 0.1607086956501007,
+      "learning_rate": 9.465020576131688e-05,
+      "loss": 0.1507,
+      "num_input_tokens_seen": 3052294,
+      "step": 134,
+      "train_runtime": 526.0714,
+      "train_tokens_per_second": 5802.053
+    },
+    {
+      "epoch": 1.0888888888888888,
+      "grad_norm": 0.14786848425865173,
+      "learning_rate": 9.382716049382717e-05,
+      "loss": 0.1371,
+      "num_input_tokens_seen": 3073014,
+      "step": 135,
+      "train_runtime": 529.0875,
+      "train_tokens_per_second": 5808.139
+    },
+    {
+      "epoch": 1.096969696969697,
+      "grad_norm": 0.14476324617862701,
+      "learning_rate": 9.300411522633746e-05,
+      "loss": 0.1223,
+      "num_input_tokens_seen": 3095126,
+      "step": 136,
+      "train_runtime": 532.2833,
+      "train_tokens_per_second": 5814.809
+    },
+    {
+      "epoch": 1.105050505050505,
+      "grad_norm": 0.17048926651477814,
+      "learning_rate": 9.218106995884775e-05,
+      "loss": 0.1542,
+      "num_input_tokens_seen": 3116438,
+      "step": 137,
+      "train_runtime": 535.3914,
+      "train_tokens_per_second": 5820.859
+    },
+    {
+      "epoch": 1.1131313131313132,
+      "grad_norm": 0.14804477989673615,
+      "learning_rate": 9.135802469135802e-05,
+      "loss": 0.129,
+      "num_input_tokens_seen": 3139174,
+      "step": 138,
+      "train_runtime": 538.6924,
+      "train_tokens_per_second": 5827.396
+    },
+    {
+      "epoch": 1.121212121212121,
+      "grad_norm": 0.14984357357025146,
+      "learning_rate": 9.053497942386831e-05,
+      "loss": 0.139,
+      "num_input_tokens_seen": 3165206,
+      "step": 139,
+      "train_runtime": 542.4266,
+      "train_tokens_per_second": 5835.27
+    },
+    {
+      "epoch": 1.1292929292929292,
+      "grad_norm": 0.1541273444890976,
+      "learning_rate": 8.97119341563786e-05,
+      "loss": 0.139,
+      "num_input_tokens_seen": 3189942,
+      "step": 140,
+      "train_runtime": 545.9937,
+      "train_tokens_per_second": 5842.452
+    },
+    {
+      "epoch": 1.1373737373737374,
+      "grad_norm": 0.1608126163482666,
+      "learning_rate": 8.888888888888889e-05,
+      "loss": 0.1145,
+      "num_input_tokens_seen": 3213878,
+      "step": 141,
+      "train_runtime": 549.4275,
+      "train_tokens_per_second": 5849.503
+    },
+    {
+      "epoch": 1.1454545454545455,
+      "grad_norm": 0.15520773828029633,
+      "learning_rate": 8.806584362139918e-05,
+      "loss": 0.185,
+      "num_input_tokens_seen": 3236806,
+      "step": 142,
+      "train_runtime": 552.7146,
+      "train_tokens_per_second": 5856.197
+    },
+    {
+      "epoch": 1.1535353535353536,
+      "grad_norm": 0.17958641052246094,
+      "learning_rate": 8.724279835390947e-05,
+      "loss": 0.1744,
+      "num_input_tokens_seen": 3258822,
+      "step": 143,
+      "train_runtime": 555.9036,
+      "train_tokens_per_second": 5862.207
+    },
+    {
+      "epoch": 1.1616161616161615,
+      "grad_norm": 0.15695422887802124,
+      "learning_rate": 8.641975308641975e-05,
+      "loss": 0.1292,
+      "num_input_tokens_seen": 3280694,
+      "step": 144,
+      "train_runtime": 559.0968,
+      "train_tokens_per_second": 5867.846
+    },
+    {
+      "epoch": 1.1696969696969697,
+      "grad_norm": 0.15811026096343994,
+      "learning_rate": 8.559670781893004e-05,
+      "loss": 0.1366,
+      "num_input_tokens_seen": 3305894,
+      "step": 145,
+      "train_runtime": 562.6785,
+      "train_tokens_per_second": 5875.281
+    },
+    {
+      "epoch": 1.1777777777777778,
+      "grad_norm": 0.18241523206233978,
+      "learning_rate": 8.477366255144033e-05,
+      "loss": 0.1277,
+      "num_input_tokens_seen": 3327526,
+      "step": 146,
+      "train_runtime": 565.8579,
+      "train_tokens_per_second": 5880.498
+    },
+    {
+      "epoch": 1.185858585858586,
+      "grad_norm": 0.18478727340698242,
+      "learning_rate": 8.395061728395062e-05,
+      "loss": 0.1556,
+      "num_input_tokens_seen": 3349414,
+      "step": 147,
+      "train_runtime": 569.0152,
+      "train_tokens_per_second": 5886.335
+    },
+    {
+      "epoch": 1.1939393939393939,
+      "grad_norm": 0.18235282599925995,
+      "learning_rate": 8.312757201646091e-05,
+      "loss": 0.185,
+      "num_input_tokens_seen": 3373782,
+      "step": 148,
+      "train_runtime": 572.5108,
+      "train_tokens_per_second": 5892.958
+    },
+    {
+      "epoch": 1.202020202020202,
+      "grad_norm": 0.16526545584201813,
+      "learning_rate": 8.23045267489712e-05,
+      "loss": 0.1417,
+      "num_input_tokens_seen": 3397254,
+      "step": 149,
+      "train_runtime": 575.9284,
+      "train_tokens_per_second": 5898.744
+    },
+    {
+      "epoch": 1.2101010101010101,
+      "grad_norm": 0.1843811720609665,
+      "learning_rate": 8.148148148148148e-05,
+      "loss": 0.1289,
+      "num_input_tokens_seen": 3420102,
+      "step": 150,
+      "train_runtime": 579.2683,
+      "train_tokens_per_second": 5904.176
+    },
+    {
+      "epoch": 1.2101010101010101,
+      "eval_loss": 0.17398151755332947,
+      "eval_runtime": 15.974,
+      "eval_samples_per_second": 55.027,
+      "eval_steps_per_second": 3.443,
+      "num_input_tokens_seen": 3420102,
+      "step": 150
+    },
+    {
+      "epoch": 1.2181818181818183,
+      "grad_norm": 0.20147649943828583,
+      "learning_rate": 8.065843621399177e-05,
+      "loss": 0.1823,
+      "num_input_tokens_seen": 3440438,
+      "step": 151,
+      "train_runtime": 598.6689,
+      "train_tokens_per_second": 5746.813
+    },
+    {
+      "epoch": 1.2262626262626264,
+      "grad_norm": 0.16948151588439941,
+      "learning_rate": 7.983539094650207e-05,
+      "loss": 0.1352,
+      "num_input_tokens_seen": 3466054,
+      "step": 152,
+      "train_runtime": 602.3354,
+      "train_tokens_per_second": 5754.358
+    },
+    {
+      "epoch": 1.2343434343434343,
+      "grad_norm": 0.13926386833190918,
+      "learning_rate": 7.901234567901235e-05,
+      "loss": 0.1396,
+      "num_input_tokens_seen": 3492278,
+      "step": 153,
+      "train_runtime": 606.1274,
+      "train_tokens_per_second": 5761.624
+    },
+    {
+      "epoch": 1.2424242424242424,
+      "grad_norm": 0.1492907702922821,
+      "learning_rate": 7.818930041152264e-05,
+      "loss": 0.1222,
+      "num_input_tokens_seen": 3512582,
+      "step": 154,
+      "train_runtime": 609.1167,
+      "train_tokens_per_second": 5766.681
+    },
+    {
+      "epoch": 1.2505050505050506,
+      "grad_norm": 0.17096783220767975,
+      "learning_rate": 7.736625514403293e-05,
+      "loss": 0.1434,
+      "num_input_tokens_seen": 3535750,
+      "step": 155,
+      "train_runtime": 612.4626,
+      "train_tokens_per_second": 5773.005
+    },
+    {
+      "epoch": 1.2585858585858585,
+      "grad_norm": 0.1619611382484436,
+      "learning_rate": 7.65432098765432e-05,
+      "loss": 0.142,
+      "num_input_tokens_seen": 3559894,
+      "step": 156,
+      "train_runtime": 615.9667,
+      "train_tokens_per_second": 5779.361
+    },
+    {
+      "epoch": 1.2666666666666666,
+      "grad_norm": 0.18333695828914642,
+      "learning_rate": 7.57201646090535e-05,
+      "loss": 0.1433,
+      "num_input_tokens_seen": 3580390,
+      "step": 157,
+      "train_runtime": 618.9828,
+      "train_tokens_per_second": 5784.312
+    },
+    {
+      "epoch": 1.2747474747474747,
+      "grad_norm": 0.1673658788204193,
+      "learning_rate": 7.48971193415638e-05,
+      "loss": 0.1168,
+      "num_input_tokens_seen": 3601750,
+      "step": 158,
+      "train_runtime": 622.0976,
+      "train_tokens_per_second": 5789.686
+    },
+    {
+      "epoch": 1.2828282828282829,
+      "grad_norm": 0.16368548572063446,
+      "learning_rate": 7.407407407407407e-05,
+      "loss": 0.1345,
+      "num_input_tokens_seen": 3629030,
+      "step": 159,
+      "train_runtime": 625.9653,
+      "train_tokens_per_second": 5797.494
+    },
+    {
+      "epoch": 1.290909090909091,
+      "grad_norm": 0.18099236488342285,
+      "learning_rate": 7.325102880658436e-05,
+      "loss": 0.1503,
+      "num_input_tokens_seen": 3654358,
+      "step": 160,
+      "train_runtime": 629.6159,
+      "train_tokens_per_second": 5804.107
+    },
+    {
+      "epoch": 1.298989898989899,
+      "grad_norm": 0.17030303180217743,
+      "learning_rate": 7.242798353909465e-05,
+      "loss": 0.1414,
+      "num_input_tokens_seen": 3680758,
+      "step": 161,
+      "train_runtime": 633.4286,
+      "train_tokens_per_second": 5810.849
+    },
+    {
+      "epoch": 1.307070707070707,
+      "grad_norm": 0.1868920922279358,
+      "learning_rate": 7.160493827160494e-05,
+      "loss": 0.1348,
+      "num_input_tokens_seen": 3704022,
+      "step": 162,
+      "train_runtime": 636.8172,
+      "train_tokens_per_second": 5816.461
+    },
+    {
+      "epoch": 1.3151515151515152,
+      "grad_norm": 0.18310731649398804,
+      "learning_rate": 7.078189300411523e-05,
+      "loss": 0.1241,
+      "num_input_tokens_seen": 3727814,
+      "step": 163,
+      "train_runtime": 640.2506,
+      "train_tokens_per_second": 5822.429
+    },
+    {
+      "epoch": 1.3232323232323233,
+      "grad_norm": 0.188144713640213,
+      "learning_rate": 6.995884773662552e-05,
+      "loss": 0.1568,
+      "num_input_tokens_seen": 3749606,
+      "step": 164,
+      "train_runtime": 643.4135,
+      "train_tokens_per_second": 5827.677
+    },
+    {
+      "epoch": 1.3313131313131312,
+      "grad_norm": 0.19525550305843353,
+      "learning_rate": 6.91358024691358e-05,
+      "loss": 0.1781,
+      "num_input_tokens_seen": 3774710,
+      "step": 165,
+      "train_runtime": 647.0635,
+      "train_tokens_per_second": 5833.601
+    },
+    {
+      "epoch": 1.3393939393939394,
+      "grad_norm": 0.18591126799583435,
+      "learning_rate": 6.831275720164609e-05,
+      "loss": 0.1231,
+      "num_input_tokens_seen": 3793814,
+      "step": 166,
+      "train_runtime": 649.8581,
+      "train_tokens_per_second": 5837.911
+    },
+    {
+      "epoch": 1.3474747474747475,
+      "grad_norm": 0.19412674009799957,
+      "learning_rate": 6.748971193415638e-05,
+      "loss": 0.1545,
+      "num_input_tokens_seen": 3818662,
+      "step": 167,
+      "train_runtime": 653.4715,
+      "train_tokens_per_second": 5843.655
+    },
+    {
+      "epoch": 1.3555555555555556,
+      "grad_norm": 0.16812458634376526,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 0.1414,
+      "num_input_tokens_seen": 3841606,
+      "step": 168,
+      "train_runtime": 656.7714,
+      "train_tokens_per_second": 5849.228
+    },
+    {
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.18051378428936005,
+      "learning_rate": 6.584362139917696e-05,
+      "loss": 0.144,
+      "num_input_tokens_seen": 3863830,
+      "step": 169,
+      "train_runtime": 660.006,
+      "train_tokens_per_second": 5854.234
+    },
+    {
+      "epoch": 1.3717171717171717,
+      "grad_norm": 0.19069063663482666,
+      "learning_rate": 6.502057613168725e-05,
+      "loss": 0.1649,
+      "num_input_tokens_seen": 3886230,
+      "step": 170,
+      "train_runtime": 663.2462,
+      "train_tokens_per_second": 5859.408
+    },
+    {
+      "epoch": 1.3797979797979798,
+      "grad_norm": 0.18689148128032684,
+      "learning_rate": 6.419753086419753e-05,
+      "loss": 0.1425,
+      "num_input_tokens_seen": 3908982,
+      "step": 171,
+      "train_runtime": 666.5218,
+      "train_tokens_per_second": 5864.747
+    },
+    {
+      "epoch": 1.387878787878788,
+      "grad_norm": 0.14730970561504364,
+      "learning_rate": 6.337448559670782e-05,
+      "loss": 0.1336,
+      "num_input_tokens_seen": 3936150,
+      "step": 172,
+      "train_runtime": 670.4336,
+      "train_tokens_per_second": 5871.052
+    },
+    {
+      "epoch": 1.3959595959595958,
+      "grad_norm": 0.177509605884552,
+      "learning_rate": 6.255144032921812e-05,
+      "loss": 0.1545,
+      "num_input_tokens_seen": 3959430,
+      "step": 173,
+      "train_runtime": 673.8214,
+      "train_tokens_per_second": 5876.083
+    },
+    {
+      "epoch": 1.404040404040404,
+      "grad_norm": 0.17181722819805145,
+      "learning_rate": 6.17283950617284e-05,
+      "loss": 0.1462,
+      "num_input_tokens_seen": 3981126,
+      "step": 174,
+      "train_runtime": 677.031,
+      "train_tokens_per_second": 5880.271
+    },
+    {
+      "epoch": 1.412121212121212,
+      "grad_norm": 0.1883271485567093,
+      "learning_rate": 6.0905349794238687e-05,
+      "loss": 0.1511,
+      "num_input_tokens_seen": 4006118,
+      "step": 175,
+      "train_runtime": 680.6168,
+      "train_tokens_per_second": 5886.011
+    },
+    {
+      "epoch": 1.412121212121212,
+      "eval_loss": 0.1727769374847412,
+      "eval_runtime": 15.9835,
+      "eval_samples_per_second": 54.994,
+      "eval_steps_per_second": 3.441,
+      "num_input_tokens_seen": 4006118,
+      "step": 175
+    },
+    {
+      "epoch": 1.4202020202020202,
+      "grad_norm": 0.17214921116828918,
+      "learning_rate": 6.0082304526748977e-05,
+      "loss": 0.1669,
+      "num_input_tokens_seen": 4034054,
+      "step": 176,
+      "train_runtime": 700.6314,
+      "train_tokens_per_second": 5757.74
+    },
+    {
+      "epoch": 1.4282828282828284,
+      "grad_norm": 0.17713947594165802,
+      "learning_rate": 5.925925925925926e-05,
+      "loss": 0.1356,
+      "num_input_tokens_seen": 4058646,
+      "step": 177,
+      "train_runtime": 704.1834,
+      "train_tokens_per_second": 5763.62
+    },
+    {
+      "epoch": 1.4363636363636363,
+      "grad_norm": 0.18645597994327545,
+      "learning_rate": 5.843621399176955e-05,
+      "loss": 0.1573,
+      "num_input_tokens_seen": 4083414,
+      "step": 178,
+      "train_runtime": 707.7598,
+      "train_tokens_per_second": 5769.491
+    },
+    {
+      "epoch": 1.4444444444444444,
+      "grad_norm": 0.1776132434606552,
+      "learning_rate": 5.761316872427984e-05,
+      "loss": 0.1338,
+      "num_input_tokens_seen": 4107190,
+      "step": 179,
+      "train_runtime": 711.2238,
+      "train_tokens_per_second": 5774.821
+    },
+    {
+      "epoch": 1.4525252525252526,
+      "grad_norm": 0.1832626312971115,
+      "learning_rate": 5.679012345679012e-05,
+      "loss": 0.2298,
+      "num_input_tokens_seen": 4129958,
+      "step": 180,
+      "train_runtime": 714.5258,
+      "train_tokens_per_second": 5779.999
+    },
+    {
+      "epoch": 1.4606060606060607,
+      "grad_norm": 0.17791666090488434,
+      "learning_rate": 5.596707818930041e-05,
+      "loss": 0.1267,
+      "num_input_tokens_seen": 4153302,
+      "step": 181,
+      "train_runtime": 718.4279,
+      "train_tokens_per_second": 5781.098
+    },
+    {
+      "epoch": 1.4686868686868686,
+      "grad_norm": 0.18174052238464355,
+      "learning_rate": 5.51440329218107e-05,
+      "loss": 0.1311,
+      "num_input_tokens_seen": 4175222,
+      "step": 182,
+      "train_runtime": 721.5861,
+      "train_tokens_per_second": 5786.173
+    },
+    {
+      "epoch": 1.4767676767676767,
+      "grad_norm": 0.17272153496742249,
+      "learning_rate": 5.4320987654320986e-05,
+      "loss": 0.1454,
+      "num_input_tokens_seen": 4196966,
+      "step": 183,
+      "train_runtime": 724.781,
+      "train_tokens_per_second": 5790.668
+    },
+    {
+      "epoch": 1.4848484848484849,
+      "grad_norm": 0.18341264128684998,
+      "learning_rate": 5.3497942386831277e-05,
+      "loss": 0.1295,
+      "num_input_tokens_seen": 4222630,
+      "step": 184,
+      "train_runtime": 728.4765,
+      "train_tokens_per_second": 5796.522
+    },
+    {
+      "epoch": 1.492929292929293,
+      "grad_norm": 0.18532636761665344,
+      "learning_rate": 5.267489711934157e-05,
+      "loss": 0.1311,
+      "num_input_tokens_seen": 4243798,
+      "step": 185,
+      "train_runtime": 731.5451,
+      "train_tokens_per_second": 5801.143
+    },
+    {
+      "epoch": 1.5010101010101011,
+      "grad_norm": 0.19856366515159607,
+      "learning_rate": 5.185185185185185e-05,
+      "loss": 0.1532,
+      "num_input_tokens_seen": 4265174,
+      "step": 186,
+      "train_runtime": 734.668,
+      "train_tokens_per_second": 5805.581
+    },
+    {
+      "epoch": 1.509090909090909,
+      "grad_norm": 0.1887090802192688,
+      "learning_rate": 5.102880658436214e-05,
+      "loss": 0.108,
+      "num_input_tokens_seen": 4286374,
+      "step": 187,
+      "train_runtime": 737.8004,
+      "train_tokens_per_second": 5809.666
+    },
+    {
+      "epoch": 1.5171717171717172,
+      "grad_norm": 0.16914108395576477,
+      "learning_rate": 5.020576131687244e-05,
+      "loss": 0.1231,
+      "num_input_tokens_seen": 4314534,
+      "step": 188,
+      "train_runtime": 741.8365,
+      "train_tokens_per_second": 5816.017
+    },
+    {
+      "epoch": 1.5252525252525253,
+      "grad_norm": 0.207884281873703,
+      "learning_rate": 4.938271604938271e-05,
+      "loss": 0.1846,
+      "num_input_tokens_seen": 4334726,
+      "step": 189,
+      "train_runtime": 744.774,
+      "train_tokens_per_second": 5820.189
+    },
+    {
+      "epoch": 1.5333333333333332,
+      "grad_norm": 0.21083691716194153,
+      "learning_rate": 4.855967078189301e-05,
+      "loss": 0.1274,
+      "num_input_tokens_seen": 4357238,
+      "step": 190,
+      "train_runtime": 748.0199,
+      "train_tokens_per_second": 5825.029
+    },
+    {
+      "epoch": 1.5414141414141413,
+      "grad_norm": 0.18116632103919983,
+      "learning_rate": 4.773662551440329e-05,
+      "loss": 0.1457,
+      "num_input_tokens_seen": 4380310,
+      "step": 191,
+      "train_runtime": 751.3376,
+      "train_tokens_per_second": 5830.015
+    },
+    {
+      "epoch": 1.5494949494949495,
+      "grad_norm": 0.1774257868528366,
+      "learning_rate": 4.691358024691358e-05,
+      "loss": 0.1638,
+      "num_input_tokens_seen": 4402598,
+      "step": 192,
+      "train_runtime": 754.5705,
+      "train_tokens_per_second": 5834.575
+    },
+    {
+      "epoch": 1.5575757575757576,
+      "grad_norm": 0.17490142583847046,
+      "learning_rate": 4.609053497942387e-05,
+      "loss": 0.1287,
+      "num_input_tokens_seen": 4426854,
+      "step": 193,
+      "train_runtime": 758.0657,
+      "train_tokens_per_second": 5839.671
+    },
+    {
+      "epoch": 1.5656565656565657,
+      "grad_norm": 0.17895734310150146,
+      "learning_rate": 4.5267489711934157e-05,
+      "loss": 0.1565,
+      "num_input_tokens_seen": 4452422,
+      "step": 194,
+      "train_runtime": 761.7593,
+      "train_tokens_per_second": 5844.92
+    },
+    {
+      "epoch": 1.5737373737373739,
+      "grad_norm": 0.19911067187786102,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 0.161,
+      "num_input_tokens_seen": 4479574,
+      "step": 195,
+      "train_runtime": 765.648,
+      "train_tokens_per_second": 5850.697
+    },
+    {
+      "epoch": 1.5818181818181818,
+      "grad_norm": 0.20772644877433777,
+      "learning_rate": 4.3621399176954737e-05,
+      "loss": 0.1791,
+      "num_input_tokens_seen": 4499862,
+      "step": 196,
+      "train_runtime": 768.5633,
+      "train_tokens_per_second": 5854.901
+    },
+    {
+      "epoch": 1.58989898989899,
+      "grad_norm": 0.20278632640838623,
+      "learning_rate": 4.279835390946502e-05,
+      "loss": 0.1221,
+      "num_input_tokens_seen": 4522614,
+      "step": 197,
+      "train_runtime": 771.8795,
+      "train_tokens_per_second": 5859.223
+    },
+    {
+      "epoch": 1.5979797979797978,
+      "grad_norm": 0.19286850094795227,
+      "learning_rate": 4.197530864197531e-05,
+      "loss": 0.1297,
+      "num_input_tokens_seen": 4544022,
+      "step": 198,
+      "train_runtime": 775.0258,
+      "train_tokens_per_second": 5863.059
+    },
+    {
+      "epoch": 1.606060606060606,
+      "grad_norm": 0.1876365840435028,
+      "learning_rate": 4.11522633744856e-05,
+      "loss": 0.1301,
+      "num_input_tokens_seen": 4566198,
+      "step": 199,
+      "train_runtime": 778.2557,
+      "train_tokens_per_second": 5867.221
+    },
+    {
+      "epoch": 1.614141414141414,
+      "grad_norm": 0.1976366490125656,
+      "learning_rate": 4.032921810699588e-05,
+      "loss": 0.1522,
+      "num_input_tokens_seen": 4588070,
+      "step": 200,
+      "train_runtime": 781.4257,
+      "train_tokens_per_second": 5871.409
+    },
+    {
+      "epoch": 1.614141414141414,
+      "eval_loss": 0.17329195141792297,
+      "eval_runtime": 15.9586,
+      "eval_samples_per_second": 55.08,
+      "eval_steps_per_second": 3.446,
+      "num_input_tokens_seen": 4588070,
+      "step": 200
+    },
+    {
+      "epoch": 1.6222222222222222,
+      "grad_norm": 0.2171986699104309,
+      "learning_rate": 3.950617283950617e-05,
+      "loss": 0.1527,
+      "num_input_tokens_seen": 4612646,
+      "step": 201,
+      "train_runtime": 800.9346,
+      "train_tokens_per_second": 5759.079
+    },
+    {
+      "epoch": 1.6303030303030304,
+      "grad_norm": 0.18816785514354706,
+      "learning_rate": 3.868312757201646e-05,
+      "loss": 0.1419,
+      "num_input_tokens_seen": 4633446,
+      "step": 202,
+      "train_runtime": 804.007,
+      "train_tokens_per_second": 5762.942
+    },
+    {
+      "epoch": 1.6383838383838385,
+      "grad_norm": 0.1965034157037735,
+      "learning_rate": 3.786008230452675e-05,
+      "loss": 0.1294,
+      "num_input_tokens_seen": 4656278,
+      "step": 203,
+      "train_runtime": 807.3253,
+      "train_tokens_per_second": 5767.536
+    },
+    {
+      "epoch": 1.6464646464646466,
+      "grad_norm": 0.20697784423828125,
+      "learning_rate": 3.7037037037037037e-05,
+      "loss": 0.1277,
+      "num_input_tokens_seen": 4678006,
+      "step": 204,
+      "train_runtime": 810.4618,
+      "train_tokens_per_second": 5772.025
+    },
+    {
+      "epoch": 1.6545454545454545,
+      "grad_norm": 0.1958000510931015,
+      "learning_rate": 3.6213991769547327e-05,
+      "loss": 0.1268,
+      "num_input_tokens_seen": 4701206,
+      "step": 205,
+      "train_runtime": 813.8387,
+      "train_tokens_per_second": 5776.582
+    },
+    {
+      "epoch": 1.6626262626262627,
+      "grad_norm": 0.1861943006515503,
+      "learning_rate": 3.539094650205762e-05,
+      "loss": 0.1549,
+      "num_input_tokens_seen": 4724646,
+      "step": 206,
+      "train_runtime": 817.2582,
+      "train_tokens_per_second": 5781.093
+    },
+    {
+      "epoch": 1.6707070707070706,
+      "grad_norm": 0.1965884119272232,
+      "learning_rate": 3.45679012345679e-05,
+      "loss": 0.1221,
+      "num_input_tokens_seen": 4748998,
+      "step": 207,
+      "train_runtime": 820.7842,
+      "train_tokens_per_second": 5785.927
+    },
+    {
+      "epoch": 1.6787878787878787,
+      "grad_norm": 0.23105938732624054,
+      "learning_rate": 3.374485596707819e-05,
+      "loss": 0.1353,
+      "num_input_tokens_seen": 4768294,
+      "step": 208,
+      "train_runtime": 823.5916,
+      "train_tokens_per_second": 5789.634
+    },
+    {
+      "epoch": 1.6868686868686869,
+      "grad_norm": 0.17601825296878815,
+      "learning_rate": 3.292181069958848e-05,
+      "loss": 0.1743,
+      "num_input_tokens_seen": 4792550,
+      "step": 209,
+      "train_runtime": 827.116,
+      "train_tokens_per_second": 5794.29
+    },
+    {
+      "epoch": 1.694949494949495,
+      "grad_norm": 0.192215234041214,
+      "learning_rate": 3.209876543209876e-05,
+      "loss": 0.1441,
+      "num_input_tokens_seen": 4816006,
+      "step": 210,
+      "train_runtime": 830.4803,
+      "train_tokens_per_second": 5799.061
+    },
+    {
+      "epoch": 1.7030303030303031,
+      "grad_norm": 0.1799449473619461,
+      "learning_rate": 3.127572016460906e-05,
+      "loss": 0.1755,
+      "num_input_tokens_seen": 4839222,
+      "step": 211,
+      "train_runtime": 834.3314,
+      "train_tokens_per_second": 5800.12
+    },
+    {
+      "epoch": 1.7111111111111112,
+      "grad_norm": 0.18329425156116486,
+      "learning_rate": 3.0452674897119343e-05,
+      "loss": 0.1469,
+      "num_input_tokens_seen": 4862822,
+      "step": 212,
+      "train_runtime": 837.7144,
+      "train_tokens_per_second": 5804.869
+    },
+    {
+      "epoch": 1.7191919191919192,
+      "grad_norm": 0.2032759040594101,
+      "learning_rate": 2.962962962962963e-05,
+      "loss": 0.1809,
+      "num_input_tokens_seen": 4884582,
+      "step": 213,
+      "train_runtime": 840.9065,
+      "train_tokens_per_second": 5808.71
+    },
+    {
+      "epoch": 1.7272727272727273,
+      "grad_norm": 0.20157882571220398,
+      "learning_rate": 2.880658436213992e-05,
+      "loss": 0.1466,
+      "num_input_tokens_seen": 4909174,
+      "step": 214,
+      "train_runtime": 844.4537,
+      "train_tokens_per_second": 5813.432
+    },
+    {
+      "epoch": 1.7353535353535352,
+      "grad_norm": 0.1857813447713852,
+      "learning_rate": 2.7983539094650207e-05,
+      "loss": 0.1549,
+      "num_input_tokens_seen": 4933798,
+      "step": 215,
+      "train_runtime": 848.0304,
+      "train_tokens_per_second": 5817.949
+    },
+    {
+      "epoch": 1.7434343434343433,
+      "grad_norm": 0.21499592065811157,
+      "learning_rate": 2.7160493827160493e-05,
+      "loss": 0.1383,
+      "num_input_tokens_seen": 4952742,
+      "step": 216,
+      "train_runtime": 850.8052,
+      "train_tokens_per_second": 5821.241
+    },
+    {
+      "epoch": 1.7515151515151515,
+      "grad_norm": 0.17735537886619568,
+      "learning_rate": 2.6337448559670787e-05,
+      "loss": 0.1496,
+      "num_input_tokens_seen": 4976246,
+      "step": 217,
+      "train_runtime": 854.1841,
+      "train_tokens_per_second": 5825.73
+    },
+    {
+      "epoch": 1.7595959595959596,
+      "grad_norm": 0.18576224148273468,
+      "learning_rate": 2.551440329218107e-05,
+      "loss": 0.1409,
+      "num_input_tokens_seen": 4995798,
+      "step": 218,
+      "train_runtime": 857.0834,
+      "train_tokens_per_second": 5828.836
+    },
+    {
+      "epoch": 1.7676767676767677,
+      "grad_norm": 0.18721511960029602,
+      "learning_rate": 2.4691358024691357e-05,
+      "loss": 0.1294,
+      "num_input_tokens_seen": 5018838,
+      "step": 219,
+      "train_runtime": 860.4125,
+      "train_tokens_per_second": 5833.06
+    },
+    {
+      "epoch": 1.7757575757575759,
+      "grad_norm": 0.18898357450962067,
+      "learning_rate": 2.3868312757201647e-05,
+      "loss": 0.1011,
+      "num_input_tokens_seen": 5041782,
+      "step": 220,
+      "train_runtime": 863.7343,
+      "train_tokens_per_second": 5837.191
+    },
+    {
+      "epoch": 1.783838383838384,
+      "grad_norm": 0.1976945698261261,
+      "learning_rate": 2.3045267489711937e-05,
+      "loss": 0.1584,
+      "num_input_tokens_seen": 5063478,
+      "step": 221,
+      "train_runtime": 866.9236,
+      "train_tokens_per_second": 5840.743
+    },
+    {
+      "epoch": 1.791919191919192,
+      "grad_norm": 0.19949166476726532,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 0.1523,
+      "num_input_tokens_seen": 5085526,
+      "step": 222,
+      "train_runtime": 870.1255,
+      "train_tokens_per_second": 5844.589
+    },
+    {
+      "epoch": 1.8,
+      "grad_norm": 0.181776225566864,
+      "learning_rate": 2.139917695473251e-05,
+      "loss": 0.1641,
+      "num_input_tokens_seen": 5109430,
+      "step": 223,
+      "train_runtime": 873.6231,
+      "train_tokens_per_second": 5848.552
+    },
+    {
+      "epoch": 1.808080808080808,
+      "grad_norm": 0.21610131859779358,
+      "learning_rate": 2.05761316872428e-05,
+      "loss": 0.116,
+      "num_input_tokens_seen": 5132182,
+      "step": 224,
+      "train_runtime": 876.9236,
+      "train_tokens_per_second": 5852.485
+    },
+    {
+      "epoch": 1.816161616161616,
+      "grad_norm": 0.19361373782157898,
+      "learning_rate": 1.9753086419753087e-05,
+      "loss": 0.1369,
+      "num_input_tokens_seen": 5153622,
+      "step": 225,
+      "train_runtime": 880.0664,
+      "train_tokens_per_second": 5855.947
+    },
+    {
+      "epoch": 1.816161616161616,
+      "eval_loss": 0.1724003553390503,
+      "eval_runtime": 15.9947,
+      "eval_samples_per_second": 54.956,
+      "eval_steps_per_second": 3.439,
+      "num_input_tokens_seen": 5153622,
+      "step": 225
+    },
+    {
+      "epoch": 1.8242424242424242,
+      "grad_norm": 0.19882901012897491,
+      "learning_rate": 1.8930041152263377e-05,
+      "loss": 0.1499,
+      "num_input_tokens_seen": 5178470,
+      "step": 226,
+      "train_runtime": 899.6327,
+      "train_tokens_per_second": 5756.205
+    },
+    {
+      "epoch": 1.8323232323232324,
+      "grad_norm": 0.19828850030899048,
+      "learning_rate": 1.8106995884773663e-05,
+      "loss": 0.1295,
+      "num_input_tokens_seen": 5199062,
+      "step": 227,
+      "train_runtime": 902.6363,
+      "train_tokens_per_second": 5759.864
+    },
+    {
+      "epoch": 1.8404040404040405,
+      "grad_norm": 0.2071538269519806,
+      "learning_rate": 1.728395061728395e-05,
+      "loss": 0.1739,
+      "num_input_tokens_seen": 5218134,
+      "step": 228,
+      "train_runtime": 905.4534,
+      "train_tokens_per_second": 5763.006
+    },
+    {
+      "epoch": 1.8484848484848486,
+      "grad_norm": 0.20533259212970734,
+      "learning_rate": 1.646090534979424e-05,
+      "loss": 0.1343,
+      "num_input_tokens_seen": 5242054,
+      "step": 229,
+      "train_runtime": 908.9457,
+      "train_tokens_per_second": 5767.181
+    },
+    {
+      "epoch": 1.8565656565656565,
+      "grad_norm": 0.19329999387264252,
+      "learning_rate": 1.563786008230453e-05,
+      "loss": 0.1266,
+      "num_input_tokens_seen": 5264518,
+      "step": 230,
+      "train_runtime": 912.2004,
+      "train_tokens_per_second": 5771.23
+    },
+    {
+      "epoch": 1.8646464646464647,
+      "grad_norm": 0.20547281205654144,
+      "learning_rate": 1.4814814814814815e-05,
+      "loss": 0.1622,
+      "num_input_tokens_seen": 5284374,
+      "step": 231,
+      "train_runtime": 915.1219,
+      "train_tokens_per_second": 5774.503
+    },
+    {
+      "epoch": 1.8727272727272726,
+      "grad_norm": 0.20196233689785004,
+      "learning_rate": 1.3991769547325103e-05,
+      "loss": 0.1211,
+      "num_input_tokens_seen": 5303622,
+      "step": 232,
+      "train_runtime": 917.9986,
+      "train_tokens_per_second": 5777.375
+    },
+    {
+      "epoch": 1.8808080808080807,
+      "grad_norm": 0.17019926011562347,
+      "learning_rate": 1.3168724279835393e-05,
+      "loss": 0.1297,
+      "num_input_tokens_seen": 5329718,
+      "step": 233,
+      "train_runtime": 921.75,
+      "train_tokens_per_second": 5782.173
+    },
+    {
+      "epoch": 1.8888888888888888,
+      "grad_norm": 0.1855989396572113,
+      "learning_rate": 1.2345679012345678e-05,
+      "loss": 0.1502,
+      "num_input_tokens_seen": 5349718,
+      "step": 234,
+      "train_runtime": 924.7021,
+      "train_tokens_per_second": 5785.342
+    },
+    {
+      "epoch": 1.896969696969697,
+      "grad_norm": 0.21048571169376373,
+      "learning_rate": 1.1522633744855968e-05,
+      "loss": 0.1483,
+      "num_input_tokens_seen": 5369638,
+      "step": 235,
+      "train_runtime": 927.6143,
+      "train_tokens_per_second": 5788.654
+    },
+    {
+      "epoch": 1.905050505050505,
+      "grad_norm": 0.20776841044425964,
+      "learning_rate": 1.0699588477366255e-05,
+      "loss": 0.1553,
+      "num_input_tokens_seen": 5392726,
+      "step": 236,
+      "train_runtime": 930.986,
+      "train_tokens_per_second": 5792.489
+    },
+    {
+      "epoch": 1.9131313131313132,
+      "grad_norm": 0.22271353006362915,
+      "learning_rate": 9.876543209876543e-06,
+      "loss": 0.1699,
+      "num_input_tokens_seen": 5412470,
+      "step": 237,
+      "train_runtime": 933.9223,
+      "train_tokens_per_second": 5795.418
+    },
+    {
+      "epoch": 1.9212121212121214,
+      "grad_norm": 0.20936141908168793,
+      "learning_rate": 9.053497942386832e-06,
+      "loss": 0.1471,
+      "num_input_tokens_seen": 5437894,
+      "step": 238,
+      "train_runtime": 937.5658,
+      "train_tokens_per_second": 5800.013
+    },
+    {
+      "epoch": 1.9292929292929293,
+      "grad_norm": 0.18848469853401184,
+      "learning_rate": 8.23045267489712e-06,
+      "loss": 0.1898,
+      "num_input_tokens_seen": 5461686,
+      "step": 239,
+      "train_runtime": 940.9924,
+      "train_tokens_per_second": 5804.177
+    },
+    {
+      "epoch": 1.9373737373737374,
+      "grad_norm": 0.17875902354717255,
+      "learning_rate": 7.4074074074074075e-06,
+      "loss": 0.147,
+      "num_input_tokens_seen": 5486054,
+      "step": 240,
+      "train_runtime": 944.4956,
+      "train_tokens_per_second": 5808.448
+    },
+    {
+      "epoch": 1.9454545454545453,
+      "grad_norm": 0.21896323561668396,
+      "learning_rate": 6.584362139917697e-06,
+      "loss": 0.1621,
+      "num_input_tokens_seen": 5509142,
+      "step": 241,
+      "train_runtime": 948.3591,
+      "train_tokens_per_second": 5809.131
+    },
+    {
+      "epoch": 1.9535353535353535,
+      "grad_norm": 0.19773228466510773,
+      "learning_rate": 5.761316872427984e-06,
+      "loss": 0.1263,
+      "num_input_tokens_seen": 5528598,
+      "step": 242,
+      "train_runtime": 951.2561,
+      "train_tokens_per_second": 5811.892
+    },
+    {
+      "epoch": 1.9616161616161616,
+      "grad_norm": 0.1763569563627243,
+      "learning_rate": 4.938271604938272e-06,
+      "loss": 0.1158,
+      "num_input_tokens_seen": 5551926,
+      "step": 243,
+      "train_runtime": 954.6378,
+      "train_tokens_per_second": 5815.741
+    },
+    {
+      "epoch": 1.9696969696969697,
+      "grad_norm": 0.17359760403633118,
+      "learning_rate": 4.11522633744856e-06,
+      "loss": 0.1097,
+      "num_input_tokens_seen": 5574326,
+      "step": 244,
+      "train_runtime": 957.9246,
+      "train_tokens_per_second": 5819.17
+    },
+    {
+      "epoch": 1.9777777777777779,
+      "grad_norm": 0.2085653394460678,
+      "learning_rate": 3.2921810699588483e-06,
+      "loss": 0.194,
+      "num_input_tokens_seen": 5598310,
+      "step": 245,
+      "train_runtime": 961.3777,
+      "train_tokens_per_second": 5823.216
+    },
+    {
+      "epoch": 1.985858585858586,
+      "grad_norm": 0.183889701962471,
+      "learning_rate": 2.469135802469136e-06,
+      "loss": 0.1558,
+      "num_input_tokens_seen": 5620534,
+      "step": 246,
+      "train_runtime": 964.6204,
+      "train_tokens_per_second": 5826.68
+    },
+    {
+      "epoch": 1.993939393939394,
+      "grad_norm": 0.20709186792373657,
+      "learning_rate": 1.6460905349794242e-06,
+      "loss": 0.1613,
+      "num_input_tokens_seen": 5640982,
+      "step": 247,
+      "train_runtime": 967.6377,
+      "train_tokens_per_second": 5829.643
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.2510419189929962,
+      "learning_rate": 8.230452674897121e-07,
+      "loss": 0.1576,
+      "num_input_tokens_seen": 5653442,
+      "step": 248,
+      "train_runtime": 969.499,
+      "train_tokens_per_second": 5831.303
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 248,
+  "num_input_tokens_seen": 5653442,
+  "num_train_epochs": 2,
+  "save_steps": 30,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.4120939382807142e+17,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}