add checkpoint cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42

.gitattributes

@@ -806,3 +806,5 @@ checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s3_perlayer_rot_r42/checkpoint-
 checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s3_perlayer_rot_r42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r8_a32_sgd_e2_lr1e-4_s3_perlayer_rot_r42/checkpoint-2038/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r8_a32_sgd_e2_lr1e-4_s3_perlayer_rot_r42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/checkpoint-2036/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/README.md

@@ -0,0 +1,61 @@
+---
+base_model: Qwen/Qwen2.5-7B-Instruct
+library_name: peft
+model_name: cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42
+tags:
+- base_model:adapter:Qwen/Qwen2.5-7B-Instruct
+- lora
+- sft
+- transformers
+- trl
+licence: license
+pipeline_tag: text-generation
+---
+
+# Model Card for cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/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
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/agam-research/huggingface/runs/0939d08f) 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- PEFT 0.18.1
+- TRL: 0.28.0
+- Transformers: 4.57.6
+- Pytorch: 2.10.0
+- Datasets: 4.5.0
+- Tokenizers: 0.22.2
+
+## Citations
+
+
+
+Cite TRL as:
+    
+```bibtex
+@software{vonwerra2020trl,
+  title   = {{TRL: Transformers Reinforcement Learning}},
+  author  = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
+  license = {Apache-2.0},
+  url     = {https://github.com/huggingface/trl},
+  year    = {2020}
+}
+```

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/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": 32,
+  "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": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "k_proj",
+    "q_proj",
+    "down_proj",
+    "gate_proj",
+    "o_proj",
+    "v_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55bb586517e05542f27fd770e2819b50bb339bbdfb71c08c62ff1ac88567b445
+size 4396749208

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/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
+}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/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 %}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/checkpoint-2036/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen2.5-7B-Instruct
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen2.5-7B-Instruct
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/checkpoint-2036/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/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": 32,
+  "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": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "k_proj",
+    "q_proj",
+    "down_proj",
+    "gate_proj",
+    "o_proj",
+    "v_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/checkpoint-2036/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55bb586517e05542f27fd770e2819b50bb339bbdfb71c08c62ff1ac88567b445
+size 4396749208

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/checkpoint-2036/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
+}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/checkpoint-2036/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 %}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/checkpoint-2036/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": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/checkpoint-2036/tokenizer.json

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

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/checkpoint-2036/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "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": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/checkpoint-2036/trainer_state.json

@@ -0,0 +1,2064 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 2036,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.2255295753479003,
+      "epoch": 0.009823182711198428,
+      "grad_norm": 4.901942729949951,
+      "learning_rate": 8.823529411764707e-06,
+      "loss": 0.6911,
+      "mean_token_accuracy": 0.7734570562839508,
+      "num_tokens": 13828.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.26828430891037,
+      "epoch": 0.019646365422396856,
+      "grad_norm": 2.577816963195801,
+      "learning_rate": 1.862745098039216e-05,
+      "loss": 0.7674,
+      "mean_token_accuracy": 0.7317770183086395,
+      "num_tokens": 28074.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.2752761960029602,
+      "epoch": 0.029469548133595286,
+      "grad_norm": 1.7750320434570312,
+      "learning_rate": 2.8431372549019608e-05,
+      "loss": 0.6719,
+      "mean_token_accuracy": 0.7584641933441162,
+      "num_tokens": 42168.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.2484833955764771,
+      "epoch": 0.03929273084479371,
+      "grad_norm": 2.1898927688598633,
+      "learning_rate": 3.8235294117647055e-05,
+      "loss": 0.675,
+      "mean_token_accuracy": 0.7524697840213775,
+      "num_tokens": 56700.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.2088910460472106,
+      "epoch": 0.04911591355599214,
+      "grad_norm": 2.5310299396514893,
+      "learning_rate": 4.803921568627452e-05,
+      "loss": 0.5581,
+      "mean_token_accuracy": 0.7924028933048248,
+      "num_tokens": 70922.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.2009735226631164,
+      "epoch": 0.05893909626719057,
+      "grad_norm": 2.211561679840088,
+      "learning_rate": 5.784313725490197e-05,
+      "loss": 0.5008,
+      "mean_token_accuracy": 0.8107963085174561,
+      "num_tokens": 84911.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.1604052424430846,
+      "epoch": 0.068762278978389,
+      "grad_norm": 2.779430389404297,
+      "learning_rate": 6.764705882352942e-05,
+      "loss": 0.5313,
+      "mean_token_accuracy": 0.8090579032897949,
+      "num_tokens": 98835.0,
+      "step": 70
+    },
+    {
+      "entropy": 1.175244891643524,
+      "epoch": 0.07858546168958742,
+      "grad_norm": 3.0253067016601562,
+      "learning_rate": 7.745098039215687e-05,
+      "loss": 0.4838,
+      "mean_token_accuracy": 0.8236717998981475,
+      "num_tokens": 112502.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.1222257137298584,
+      "epoch": 0.08840864440078586,
+      "grad_norm": 2.4448535442352295,
+      "learning_rate": 8.725490196078432e-05,
+      "loss": 0.4851,
+      "mean_token_accuracy": 0.8165680289268493,
+      "num_tokens": 126248.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.1350947737693786,
+      "epoch": 0.09823182711198428,
+      "grad_norm": 2.95954966545105,
+      "learning_rate": 9.705882352941177e-05,
+      "loss": 0.4766,
+      "mean_token_accuracy": 0.8244425833225251,
+      "num_tokens": 139551.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.1114699840545654,
+      "epoch": 0.10805500982318271,
+      "grad_norm": 2.9487740993499756,
+      "learning_rate": 9.999676765122786e-05,
+      "loss": 0.4468,
+      "mean_token_accuracy": 0.8271951079368591,
+      "num_tokens": 152975.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.110399353504181,
+      "epoch": 0.11787819253438114,
+      "grad_norm": 2.5977447032928467,
+      "learning_rate": 9.998093674494531e-05,
+      "loss": 0.4665,
+      "mean_token_accuracy": 0.8266980826854706,
+      "num_tokens": 167073.0,
+      "step": 120
+    },
+    {
+      "entropy": 1.1198901414871216,
+      "epoch": 0.12770137524557956,
+      "grad_norm": 2.4050168991088867,
+      "learning_rate": 9.995191775639568e-05,
+      "loss": 0.4769,
+      "mean_token_accuracy": 0.824345737695694,
+      "num_tokens": 181404.0,
+      "step": 130
+    },
+    {
+      "entropy": 1.1038351774215698,
+      "epoch": 0.137524557956778,
+      "grad_norm": 2.3017184734344482,
+      "learning_rate": 9.990971834259444e-05,
+      "loss": 0.4518,
+      "mean_token_accuracy": 0.8278718709945678,
+      "num_tokens": 195441.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.1219186305999755,
+      "epoch": 0.14734774066797643,
+      "grad_norm": 3.237029790878296,
+      "learning_rate": 9.985434963837376e-05,
+      "loss": 0.4782,
+      "mean_token_accuracy": 0.8169582784175873,
+      "num_tokens": 209863.0,
+      "step": 150
+    },
+    {
+      "entropy": 1.0830397725105285,
+      "epoch": 0.15717092337917485,
+      "grad_norm": 3.1673595905303955,
+      "learning_rate": 9.978582625344435e-05,
+      "loss": 0.4486,
+      "mean_token_accuracy": 0.8331800699234009,
+      "num_tokens": 223646.0,
+      "step": 160
+    },
+    {
+      "entropy": 1.0910556316375732,
+      "epoch": 0.16699410609037327,
+      "grad_norm": 2.7308309078216553,
+      "learning_rate": 9.970416626854049e-05,
+      "loss": 0.4212,
+      "mean_token_accuracy": 0.8458732306957245,
+      "num_tokens": 237587.0,
+      "step": 170
+    },
+    {
+      "entropy": 1.0993134021759032,
+      "epoch": 0.17681728880157171,
+      "grad_norm": 2.718012809753418,
+      "learning_rate": 9.960939123064929e-05,
+      "loss": 0.4526,
+      "mean_token_accuracy": 0.8338144421577454,
+      "num_tokens": 251760.0,
+      "step": 180
+    },
+    {
+      "entropy": 1.077388596534729,
+      "epoch": 0.18664047151277013,
+      "grad_norm": 2.5045619010925293,
+      "learning_rate": 9.950152614732518e-05,
+      "loss": 0.4186,
+      "mean_token_accuracy": 0.8430645823478699,
+      "num_tokens": 265417.0,
+      "step": 190
+    },
+    {
+      "entropy": 1.1003778100013732,
+      "epoch": 0.19646365422396855,
+      "grad_norm": 3.128573179244995,
+      "learning_rate": 9.93805994800914e-05,
+      "loss": 0.4585,
+      "mean_token_accuracy": 0.8344046592712402,
+      "num_tokens": 278540.0,
+      "step": 200
+    },
+    {
+      "entropy": 1.1063192248344422,
+      "epoch": 0.206286836935167,
+      "grad_norm": 3.343653917312622,
+      "learning_rate": 9.924664313693006e-05,
+      "loss": 0.4334,
+      "mean_token_accuracy": 0.8367803871631623,
+      "num_tokens": 292295.0,
+      "step": 210
+    },
+    {
+      "entropy": 1.1028631448745727,
+      "epoch": 0.21611001964636542,
+      "grad_norm": 2.8505189418792725,
+      "learning_rate": 9.909969246386284e-05,
+      "loss": 0.4381,
+      "mean_token_accuracy": 0.8381172716617584,
+      "num_tokens": 305466.0,
+      "step": 220
+    },
+    {
+      "entropy": 1.1259036183357238,
+      "epoch": 0.22593320235756384,
+      "grad_norm": 2.5124258995056152,
+      "learning_rate": 9.893978623562453e-05,
+      "loss": 0.4315,
+      "mean_token_accuracy": 0.8415260374546051,
+      "num_tokens": 320495.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.0840003490447998,
+      "epoch": 0.2357563850687623,
+      "grad_norm": 2.5751583576202393,
+      "learning_rate": 9.876696664543177e-05,
+      "loss": 0.4055,
+      "mean_token_accuracy": 0.8462957084178925,
+      "num_tokens": 333901.0,
+      "step": 240
+    },
+    {
+      "entropy": 1.1012001514434815,
+      "epoch": 0.2455795677799607,
+      "grad_norm": 3.1212337017059326,
+      "learning_rate": 9.858127929384991e-05,
+      "loss": 0.392,
+      "mean_token_accuracy": 0.8472519278526306,
+      "num_tokens": 347392.0,
+      "step": 250
+    },
+    {
+      "entropy": 1.0842164397239684,
+      "epoch": 0.2554027504911591,
+      "grad_norm": 2.6033802032470703,
+      "learning_rate": 9.838277317676077e-05,
+      "loss": 0.404,
+      "mean_token_accuracy": 0.8417458176612854,
+      "num_tokens": 361140.0,
+      "step": 260
+    },
+    {
+      "entropy": 1.102819550037384,
+      "epoch": 0.26522593320235754,
+      "grad_norm": 2.673128366470337,
+      "learning_rate": 9.817150067243438e-05,
+      "loss": 0.368,
+      "mean_token_accuracy": 0.8603320419788361,
+      "num_tokens": 374661.0,
+      "step": 270
+    },
+    {
+      "entropy": 1.1476540565490723,
+      "epoch": 0.275049115913556,
+      "grad_norm": 2.594116687774658,
+      "learning_rate": 9.794751752770845e-05,
+      "loss": 0.4127,
+      "mean_token_accuracy": 0.8421551287174225,
+      "num_tokens": 388415.0,
+      "step": 280
+    },
+    {
+      "entropy": 1.1804301261901855,
+      "epoch": 0.28487229862475444,
+      "grad_norm": 4.0482096672058105,
+      "learning_rate": 9.771088284327879e-05,
+      "loss": 0.4201,
+      "mean_token_accuracy": 0.8360170662403107,
+      "num_tokens": 402031.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.1789313197135924,
+      "epoch": 0.29469548133595286,
+      "grad_norm": 3.0265281200408936,
+      "learning_rate": 9.746165905810488e-05,
+      "loss": 0.3811,
+      "mean_token_accuracy": 0.8588999390602112,
+      "num_tokens": 416165.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.1957629561424254,
+      "epoch": 0.3045186640471513,
+      "grad_norm": 2.086890459060669,
+      "learning_rate": 9.719991193293466e-05,
+      "loss": 0.4135,
+      "mean_token_accuracy": 0.8450926959514617,
+      "num_tokens": 429910.0,
+      "step": 310
+    },
+    {
+      "entropy": 1.1892850041389464,
+      "epoch": 0.3143418467583497,
+      "grad_norm": 3.1166536808013916,
+      "learning_rate": 9.692571053295268e-05,
+      "loss": 0.403,
+      "mean_token_accuracy": 0.8501203000545502,
+      "num_tokens": 443725.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.1840902805328368,
+      "epoch": 0.3241650294695481,
+      "grad_norm": 3.0264158248901367,
+      "learning_rate": 9.663912720955637e-05,
+      "loss": 0.4057,
+      "mean_token_accuracy": 0.8458710610866547,
+      "num_tokens": 458234.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.179422664642334,
+      "epoch": 0.33398821218074654,
+      "grad_norm": 3.2106170654296875,
+      "learning_rate": 9.634023758126539e-05,
+      "loss": 0.383,
+      "mean_token_accuracy": 0.8559177160263062,
+      "num_tokens": 471402.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.1484579920768738,
+      "epoch": 0.343811394891945,
+      "grad_norm": 2.8006367683410645,
+      "learning_rate": 9.602912051376859e-05,
+      "loss": 0.3726,
+      "mean_token_accuracy": 0.8593140482902527,
+      "num_tokens": 484651.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.1628545761108398,
+      "epoch": 0.35363457760314343,
+      "grad_norm": 2.3663344383239746,
+      "learning_rate": 9.570585809911441e-05,
+      "loss": 0.3857,
+      "mean_token_accuracy": 0.8520330369472504,
+      "num_tokens": 497956.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.1554294943809509,
+      "epoch": 0.36345776031434185,
+      "grad_norm": 2.2817635536193848,
+      "learning_rate": 9.537053563404992e-05,
+      "loss": 0.3878,
+      "mean_token_accuracy": 0.8538653612136841,
+      "num_tokens": 511717.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.1633305549621582,
+      "epoch": 0.37328094302554027,
+      "grad_norm": 3.707150459289551,
+      "learning_rate": 9.502324159751412e-05,
+      "loss": 0.3459,
+      "mean_token_accuracy": 0.8636071920394898,
+      "num_tokens": 525633.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.132694911956787,
+      "epoch": 0.3831041257367387,
+      "grad_norm": 2.677001953125,
+      "learning_rate": 9.466406762729181e-05,
+      "loss": 0.3553,
+      "mean_token_accuracy": 0.8627163946628571,
+      "num_tokens": 538945.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.1391624093055726,
+      "epoch": 0.3929273084479371,
+      "grad_norm": 2.899839162826538,
+      "learning_rate": 9.429310849583378e-05,
+      "loss": 0.3683,
+      "mean_token_accuracy": 0.8599117338657379,
+      "num_tokens": 552861.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.1567689418792724,
+      "epoch": 0.4027504911591356,
+      "grad_norm": 2.9319725036621094,
+      "learning_rate": 9.391046208524989e-05,
+      "loss": 0.4041,
+      "mean_token_accuracy": 0.84514120221138,
+      "num_tokens": 565955.0,
+      "step": 410
+    },
+    {
+      "entropy": 1.1662138104438782,
+      "epoch": 0.412573673870334,
+      "grad_norm": 2.9151804447174072,
+      "learning_rate": 9.351622936148182e-05,
+      "loss": 0.3545,
+      "mean_token_accuracy": 0.8644780397415162,
+      "num_tokens": 579923.0,
+      "step": 420
+    },
+    {
+      "entropy": 1.1690862655639649,
+      "epoch": 0.4223968565815324,
+      "grad_norm": 2.554936408996582,
+      "learning_rate": 9.31105143476618e-05,
+      "loss": 0.3598,
+      "mean_token_accuracy": 0.8622689425945282,
+      "num_tokens": 594071.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.1793851375579834,
+      "epoch": 0.43222003929273084,
+      "grad_norm": 2.7115278244018555,
+      "learning_rate": 9.269342409666495e-05,
+      "loss": 0.3427,
+      "mean_token_accuracy": 0.8639544010162353,
+      "num_tokens": 608545.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.1322578549385072,
+      "epoch": 0.44204322200392926,
+      "grad_norm": 2.60540771484375,
+      "learning_rate": 9.226506866286202e-05,
+      "loss": 0.3512,
+      "mean_token_accuracy": 0.8637854397296906,
+      "num_tokens": 621975.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.1488801002502442,
+      "epoch": 0.4518664047151277,
+      "grad_norm": 2.897393226623535,
+      "learning_rate": 9.18255610730802e-05,
+      "loss": 0.3557,
+      "mean_token_accuracy": 0.8699378371238708,
+      "num_tokens": 636061.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.145258355140686,
+      "epoch": 0.46168958742632615,
+      "grad_norm": 2.4220831394195557,
+      "learning_rate": 9.137501729677962e-05,
+      "loss": 0.3477,
+      "mean_token_accuracy": 0.8667880833148957,
+      "num_tokens": 649717.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.1348479986190796,
+      "epoch": 0.4715127701375246,
+      "grad_norm": 2.668951988220215,
+      "learning_rate": 9.091355621545339e-05,
+      "loss": 0.325,
+      "mean_token_accuracy": 0.8755492508411408,
+      "num_tokens": 664123.0,
+      "step": 480
+    },
+    {
+      "entropy": 1.1535218954086304,
+      "epoch": 0.481335952848723,
+      "grad_norm": 2.8672430515289307,
+      "learning_rate": 9.044129959125923e-05,
+      "loss": 0.368,
+      "mean_token_accuracy": 0.8594876945018768,
+      "num_tokens": 677803.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.150494110584259,
+      "epoch": 0.4911591355599214,
+      "grad_norm": 3.0353219509124756,
+      "learning_rate": 8.995837203489108e-05,
+      "loss": 0.3186,
+      "mean_token_accuracy": 0.8801987528800964,
+      "num_tokens": 691461.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.1632410883903503,
+      "epoch": 0.5009823182711198,
+      "grad_norm": 3.3344523906707764,
+      "learning_rate": 8.946490097269899e-05,
+      "loss": 0.3442,
+      "mean_token_accuracy": 0.8711451828479767,
+      "num_tokens": 705470.0,
+      "step": 510
+    },
+    {
+      "entropy": 1.1603899955749513,
+      "epoch": 0.5108055009823183,
+      "grad_norm": 3.0039889812469482,
+      "learning_rate": 8.896101661306618e-05,
+      "loss": 0.3648,
+      "mean_token_accuracy": 0.8595043361186981,
+      "num_tokens": 718805.0,
+      "step": 520
+    },
+    {
+      "entropy": 1.1511512875556946,
+      "epoch": 0.5206286836935167,
+      "grad_norm": 2.1679348945617676,
+      "learning_rate": 8.844685191205193e-05,
+      "loss": 0.3499,
+      "mean_token_accuracy": 0.8666592240333557,
+      "num_tokens": 732367.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.150057566165924,
+      "epoch": 0.5304518664047151,
+      "grad_norm": 2.892103910446167,
+      "learning_rate": 8.792254253830945e-05,
+      "loss": 0.3352,
+      "mean_token_accuracy": 0.867387878894806,
+      "num_tokens": 746431.0,
+      "step": 540
+    },
+    {
+      "entropy": 1.1266682505607606,
+      "epoch": 0.5402750491159135,
+      "grad_norm": 2.5559699535369873,
+      "learning_rate": 8.738822683728814e-05,
+      "loss": 0.3145,
+      "mean_token_accuracy": 0.8776913821697235,
+      "num_tokens": 760182.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.1423457741737366,
+      "epoch": 0.550098231827112,
+      "grad_norm": 2.5814733505249023,
+      "learning_rate": 8.684404579472936e-05,
+      "loss": 0.315,
+      "mean_token_accuracy": 0.878667289018631,
+      "num_tokens": 774137.0,
+      "step": 560
+    },
+    {
+      "entropy": 1.1212276339530944,
+      "epoch": 0.5599214145383105,
+      "grad_norm": 2.96913743019104,
+      "learning_rate": 8.629014299946568e-05,
+      "loss": 0.3216,
+      "mean_token_accuracy": 0.8742214560508728,
+      "num_tokens": 787806.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.142948842048645,
+      "epoch": 0.5697445972495089,
+      "grad_norm": 2.7884769439697266,
+      "learning_rate": 8.572666460553316e-05,
+      "loss": 0.3212,
+      "mean_token_accuracy": 0.8739893078804016,
+      "num_tokens": 801431.0,
+      "step": 580
+    },
+    {
+      "entropy": 1.1408942461013794,
+      "epoch": 0.5795677799607073,
+      "grad_norm": 2.627880573272705,
+      "learning_rate": 8.515375929360687e-05,
+      "loss": 0.3289,
+      "mean_token_accuracy": 0.8690893113613128,
+      "num_tokens": 815080.0,
+      "step": 590
+    },
+    {
+      "entropy": 1.139445698261261,
+      "epoch": 0.5893909626719057,
+      "grad_norm": 3.5122270584106445,
+      "learning_rate": 8.45715782317696e-05,
+      "loss": 0.3412,
+      "mean_token_accuracy": 0.8685419678688049,
+      "num_tokens": 829061.0,
+      "step": 600
+    },
+    {
+      "entropy": 1.1445452213287353,
+      "epoch": 0.5992141453831041,
+      "grad_norm": 2.7658352851867676,
+      "learning_rate": 8.39802750356244e-05,
+      "loss": 0.3326,
+      "mean_token_accuracy": 0.8698091626167297,
+      "num_tokens": 842325.0,
+      "step": 610
+    },
+    {
+      "entropy": 1.1496277451515198,
+      "epoch": 0.6090373280943026,
+      "grad_norm": 3.5046138763427734,
+      "learning_rate": 8.338000572776114e-05,
+      "loss": 0.3139,
+      "mean_token_accuracy": 0.8804328143596649,
+      "num_tokens": 856377.0,
+      "step": 620
+    },
+    {
+      "entropy": 1.1314571022987365,
+      "epoch": 0.618860510805501,
+      "grad_norm": 3.407663345336914,
+      "learning_rate": 8.277092869658808e-05,
+      "loss": 0.2876,
+      "mean_token_accuracy": 0.8931130468845367,
+      "num_tokens": 870502.0,
+      "step": 630
+    },
+    {
+      "entropy": 1.14702068567276,
+      "epoch": 0.6286836935166994,
+      "grad_norm": 2.5561013221740723,
+      "learning_rate": 8.215320465453914e-05,
+      "loss": 0.3243,
+      "mean_token_accuracy": 0.8723864674568176,
+      "num_tokens": 884708.0,
+      "step": 640
+    },
+    {
+      "entropy": 1.143658423423767,
+      "epoch": 0.6385068762278978,
+      "grad_norm": 2.5213441848754883,
+      "learning_rate": 8.152699659566797e-05,
+      "loss": 0.3171,
+      "mean_token_accuracy": 0.8796083450317382,
+      "num_tokens": 898142.0,
+      "step": 650
+    },
+    {
+      "entropy": 1.128274714946747,
+      "epoch": 0.6483300589390962,
+      "grad_norm": 2.4924278259277344,
+      "learning_rate": 8.089246975263996e-05,
+      "loss": 0.312,
+      "mean_token_accuracy": 0.8827828645706177,
+      "num_tokens": 912820.0,
+      "step": 660
+    },
+    {
+      "entropy": 1.133193266391754,
+      "epoch": 0.6581532416502947,
+      "grad_norm": 2.837717294692993,
+      "learning_rate": 8.024979155313354e-05,
+      "loss": 0.2991,
+      "mean_token_accuracy": 0.8883875906467438,
+      "num_tokens": 926399.0,
+      "step": 670
+    },
+    {
+      "entropy": 1.1490959048271179,
+      "epoch": 0.6679764243614931,
+      "grad_norm": 2.6145212650299072,
+      "learning_rate": 7.959913157566246e-05,
+      "loss": 0.3257,
+      "mean_token_accuracy": 0.8765207469463349,
+      "num_tokens": 940133.0,
+      "step": 680
+    },
+    {
+      "entropy": 1.1650834321975707,
+      "epoch": 0.6777996070726916,
+      "grad_norm": 3.4584596157073975,
+      "learning_rate": 7.894066150483023e-05,
+      "loss": 0.3057,
+      "mean_token_accuracy": 0.8832741916179657,
+      "num_tokens": 954504.0,
+      "step": 690
+    },
+    {
+      "entropy": 1.1442453861236572,
+      "epoch": 0.68762278978389,
+      "grad_norm": 3.186716318130493,
+      "learning_rate": 7.827455508602923e-05,
+      "loss": 0.3417,
+      "mean_token_accuracy": 0.8643399119377136,
+      "num_tokens": 968934.0,
+      "step": 700
+    },
+    {
+      "entropy": 1.1493278384208678,
+      "epoch": 0.6974459724950884,
+      "grad_norm": 2.3198447227478027,
+      "learning_rate": 7.760098807959573e-05,
+      "loss": 0.3345,
+      "mean_token_accuracy": 0.8764726340770721,
+      "num_tokens": 983038.0,
+      "step": 710
+    },
+    {
+      "entropy": 1.1385332107543946,
+      "epoch": 0.7072691552062869,
+      "grad_norm": 3.395853042602539,
+      "learning_rate": 7.69201382144334e-05,
+      "loss": 0.3142,
+      "mean_token_accuracy": 0.8774312198162079,
+      "num_tokens": 996748.0,
+      "step": 720
+    },
+    {
+      "entropy": 1.1440164685249328,
+      "epoch": 0.7170923379174853,
+      "grad_norm": 2.592904806137085,
+      "learning_rate": 7.623218514111743e-05,
+      "loss": 0.2838,
+      "mean_token_accuracy": 0.8985074639320374,
+      "num_tokens": 1010595.0,
+      "step": 730
+    },
+    {
+      "entropy": 1.13111971616745,
+      "epoch": 0.7269155206286837,
+      "grad_norm": 3.281940221786499,
+      "learning_rate": 7.553731038449144e-05,
+      "loss": 0.2844,
+      "mean_token_accuracy": 0.8937412917613983,
+      "num_tokens": 1024890.0,
+      "step": 740
+    },
+    {
+      "entropy": 1.136182177066803,
+      "epoch": 0.7367387033398821,
+      "grad_norm": 3.143935441970825,
+      "learning_rate": 7.483569729576996e-05,
+      "loss": 0.2922,
+      "mean_token_accuracy": 0.88930042386055,
+      "num_tokens": 1038967.0,
+      "step": 750
+    },
+    {
+      "entropy": 1.1285874605178834,
+      "epoch": 0.7465618860510805,
+      "grad_norm": 3.1262502670288086,
+      "learning_rate": 7.412753100415897e-05,
+      "loss": 0.2867,
+      "mean_token_accuracy": 0.8921878337860107,
+      "num_tokens": 1052622.0,
+      "step": 760
+    },
+    {
+      "entropy": 1.1501566529273988,
+      "epoch": 0.756385068762279,
+      "grad_norm": 2.667618751525879,
+      "learning_rate": 7.341299836800729e-05,
+      "loss": 0.3158,
+      "mean_token_accuracy": 0.8824589848518372,
+      "num_tokens": 1066588.0,
+      "step": 770
+    },
+    {
+      "entropy": 1.1224850296974183,
+      "epoch": 0.7662082514734774,
+      "grad_norm": 2.920957326889038,
+      "learning_rate": 7.26922879255017e-05,
+      "loss": 0.249,
+      "mean_token_accuracy": 0.9050864934921264,
+      "num_tokens": 1080453.0,
+      "step": 780
+    },
+    {
+      "entropy": 1.1101405262947082,
+      "epoch": 0.7760314341846758,
+      "grad_norm": 2.600310802459717,
+      "learning_rate": 7.196558984491889e-05,
+      "loss": 0.2975,
+      "mean_token_accuracy": 0.8790829360485077,
+      "num_tokens": 1094550.0,
+      "step": 790
+    },
+    {
+      "entropy": 1.1376460194587708,
+      "epoch": 0.7858546168958742,
+      "grad_norm": 2.992723226547241,
+      "learning_rate": 7.123309587444711e-05,
+      "loss": 0.2729,
+      "mean_token_accuracy": 0.8938282430171967,
+      "num_tokens": 1108808.0,
+      "step": 800
+    },
+    {
+      "entropy": 1.1469402074813844,
+      "epoch": 0.7956777996070727,
+      "grad_norm": 2.3969576358795166,
+      "learning_rate": 7.04949992915912e-05,
+      "loss": 0.2606,
+      "mean_token_accuracy": 0.896224069595337,
+      "num_tokens": 1122365.0,
+      "step": 810
+    },
+    {
+      "entropy": 1.1464146852493287,
+      "epoch": 0.8055009823182712,
+      "grad_norm": 3.3862643241882324,
+      "learning_rate": 6.975149485217388e-05,
+      "loss": 0.2988,
+      "mean_token_accuracy": 0.883493846654892,
+      "num_tokens": 1136376.0,
+      "step": 820
+    },
+    {
+      "entropy": 1.1212526321411134,
+      "epoch": 0.8153241650294696,
+      "grad_norm": 2.629843235015869,
+      "learning_rate": 6.900277873894705e-05,
+      "loss": 0.2865,
+      "mean_token_accuracy": 0.8825255513191224,
+      "num_tokens": 1150011.0,
+      "step": 830
+    },
+    {
+      "entropy": 1.1416452527046204,
+      "epoch": 0.825147347740668,
+      "grad_norm": 2.671727418899536,
+      "learning_rate": 6.824904850982656e-05,
+      "loss": 0.2809,
+      "mean_token_accuracy": 0.8917826533317565,
+      "num_tokens": 1164190.0,
+      "step": 840
+    },
+    {
+      "entropy": 1.1280429363250732,
+      "epoch": 0.8349705304518664,
+      "grad_norm": 3.3086113929748535,
+      "learning_rate": 6.749050304576415e-05,
+      "loss": 0.2608,
+      "mean_token_accuracy": 0.8980595290660858,
+      "num_tokens": 1177346.0,
+      "step": 850
+    },
+    {
+      "entropy": 1.112571394443512,
+      "epoch": 0.8447937131630648,
+      "grad_norm": 3.7174386978149414,
+      "learning_rate": 6.672734249827028e-05,
+      "loss": 0.2633,
+      "mean_token_accuracy": 0.9027473211288453,
+      "num_tokens": 1190692.0,
+      "step": 860
+    },
+    {
+      "entropy": 1.1413660883903503,
+      "epoch": 0.8546168958742633,
+      "grad_norm": 3.228044033050537,
+      "learning_rate": 6.59597682366017e-05,
+      "loss": 0.3013,
+      "mean_token_accuracy": 0.8820703089237213,
+      "num_tokens": 1204254.0,
+      "step": 870
+    },
+    {
+      "entropy": 1.1489825606346131,
+      "epoch": 0.8644400785854617,
+      "grad_norm": 3.616088628768921,
+      "learning_rate": 6.518798279462766e-05,
+      "loss": 0.3024,
+      "mean_token_accuracy": 0.8766620337963105,
+      "num_tokens": 1217938.0,
+      "step": 880
+    },
+    {
+      "entropy": 1.14940847158432,
+      "epoch": 0.8742632612966601,
+      "grad_norm": 2.393197536468506,
+      "learning_rate": 6.441218981738895e-05,
+      "loss": 0.2646,
+      "mean_token_accuracy": 0.899279636144638,
+      "num_tokens": 1231589.0,
+      "step": 890
+    },
+    {
+      "entropy": 1.1366554141044616,
+      "epoch": 0.8840864440078585,
+      "grad_norm": 2.769077777862549,
+      "learning_rate": 6.363259400736359e-05,
+      "loss": 0.2516,
+      "mean_token_accuracy": 0.9006950914859772,
+      "num_tokens": 1245755.0,
+      "step": 900
+    },
+    {
+      "entropy": 1.132609450817108,
+      "epoch": 0.8939096267190569,
+      "grad_norm": 3.5808353424072266,
+      "learning_rate": 6.284940107045363e-05,
+      "loss": 0.281,
+      "mean_token_accuracy": 0.8968489110469818,
+      "num_tokens": 1259830.0,
+      "step": 910
+    },
+    {
+      "entropy": 1.1349992632865906,
+      "epoch": 0.9037328094302554,
+      "grad_norm": 2.7788522243499756,
+      "learning_rate": 6.206281766170706e-05,
+      "loss": 0.2935,
+      "mean_token_accuracy": 0.8857840418815612,
+      "num_tokens": 1274186.0,
+      "step": 920
+    },
+    {
+      "entropy": 1.1282876133918762,
+      "epoch": 0.9135559921414538,
+      "grad_norm": 3.199428081512451,
+      "learning_rate": 6.127305133078945e-05,
+      "loss": 0.2636,
+      "mean_token_accuracy": 0.8973712682723999,
+      "num_tokens": 1288175.0,
+      "step": 930
+    },
+    {
+      "entropy": 1.123326587677002,
+      "epoch": 0.9233791748526523,
+      "grad_norm": 3.2814064025878906,
+      "learning_rate": 6.048031046721931e-05,
+      "loss": 0.2639,
+      "mean_token_accuracy": 0.8992348313331604,
+      "num_tokens": 1301979.0,
+      "step": 940
+    },
+    {
+      "entropy": 1.137069308757782,
+      "epoch": 0.9332023575638507,
+      "grad_norm": 3.7889764308929443,
+      "learning_rate": 5.9684804245382056e-05,
+      "loss": 0.2684,
+      "mean_token_accuracy": 0.8986769497394562,
+      "num_tokens": 1316012.0,
+      "step": 950
+    },
+    {
+      "entropy": 1.1358091711997986,
+      "epoch": 0.9430255402750491,
+      "grad_norm": 3.764880895614624,
+      "learning_rate": 5.888674256933671e-05,
+      "loss": 0.2757,
+      "mean_token_accuracy": 0.8888760566711426,
+      "num_tokens": 1329560.0,
+      "step": 960
+    },
+    {
+      "entropy": 1.1470587134361268,
+      "epoch": 0.9528487229862476,
+      "grad_norm": 2.758847236633301,
+      "learning_rate": 5.80863360174302e-05,
+      "loss": 0.2902,
+      "mean_token_accuracy": 0.8864151537418365,
+      "num_tokens": 1343980.0,
+      "step": 970
+    },
+    {
+      "entropy": 1.1305329084396363,
+      "epoch": 0.962671905697446,
+      "grad_norm": 3.0178449153900146,
+      "learning_rate": 5.72837957867337e-05,
+      "loss": 0.2661,
+      "mean_token_accuracy": 0.8940551102161407,
+      "num_tokens": 1358209.0,
+      "step": 980
+    },
+    {
+      "entropy": 1.1114959239959716,
+      "epoch": 0.9724950884086444,
+      "grad_norm": 3.425426721572876,
+      "learning_rate": 5.64793336373156e-05,
+      "loss": 0.252,
+      "mean_token_accuracy": 0.9057461857795716,
+      "num_tokens": 1371263.0,
+      "step": 990
+    },
+    {
+      "entropy": 1.1189913868904113,
+      "epoch": 0.9823182711198428,
+      "grad_norm": 2.421281099319458,
+      "learning_rate": 5.567316183636615e-05,
+      "loss": 0.2175,
+      "mean_token_accuracy": 0.9146191656589509,
+      "num_tokens": 1385183.0,
+      "step": 1000
+    },
+    {
+      "entropy": 1.1094309329986571,
+      "epoch": 0.9921414538310412,
+      "grad_norm": 3.1283364295959473,
+      "learning_rate": 5.4865493102188046e-05,
+      "loss": 0.2452,
+      "mean_token_accuracy": 0.9026672542095184,
+      "num_tokens": 1398309.0,
+      "step": 1010
+    },
+    {
+      "entropy": 1.1065597891807557,
+      "epoch": 1.0019646365422397,
+      "grad_norm": 3.041694402694702,
+      "learning_rate": 5.405654054806817e-05,
+      "loss": 0.2499,
+      "mean_token_accuracy": 0.9030339837074279,
+      "num_tokens": 1411126.0,
+      "step": 1020
+    },
+    {
+      "entropy": 1.1176036477088929,
+      "epoch": 1.0117878192534382,
+      "grad_norm": 3.108553886413574,
+      "learning_rate": 5.3246517626045025e-05,
+      "loss": 0.1707,
+      "mean_token_accuracy": 0.9365115225315094,
+      "num_tokens": 1425582.0,
+      "step": 1030
+    },
+    {
+      "entropy": 1.088628900051117,
+      "epoch": 1.0216110019646365,
+      "grad_norm": 2.50624418258667,
+      "learning_rate": 5.243563807058676e-05,
+      "loss": 0.173,
+      "mean_token_accuracy": 0.9359729945659637,
+      "num_tokens": 1439527.0,
+      "step": 1040
+    },
+    {
+      "entropy": 1.0859333276748657,
+      "epoch": 1.031434184675835,
+      "grad_norm": 3.250480890274048,
+      "learning_rate": 5.162411584219476e-05,
+      "loss": 0.1647,
+      "mean_token_accuracy": 0.9349007308483124,
+      "num_tokens": 1453445.0,
+      "step": 1050
+    },
+    {
+      "entropy": 1.068867063522339,
+      "epoch": 1.0412573673870333,
+      "grad_norm": 2.9762160778045654,
+      "learning_rate": 5.0812165070947594e-05,
+      "loss": 0.1673,
+      "mean_token_accuracy": 0.9397619485855102,
+      "num_tokens": 1466353.0,
+      "step": 1060
+    },
+    {
+      "entropy": 1.0903839826583863,
+      "epoch": 1.0510805500982319,
+      "grad_norm": 3.217416524887085,
+      "learning_rate": 5e-05,
+      "loss": 0.158,
+      "mean_token_accuracy": 0.9445515990257263,
+      "num_tokens": 1481062.0,
+      "step": 1070
+    },
+    {
+      "entropy": 1.0761645317077637,
+      "epoch": 1.0609037328094302,
+      "grad_norm": 2.0226552486419678,
+      "learning_rate": 4.918783492905242e-05,
+      "loss": 0.1563,
+      "mean_token_accuracy": 0.941853255033493,
+      "num_tokens": 1495189.0,
+      "step": 1080
+    },
+    {
+      "entropy": 1.0758458137512208,
+      "epoch": 1.0707269155206287,
+      "grad_norm": 3.9419448375701904,
+      "learning_rate": 4.837588415780524e-05,
+      "loss": 0.1634,
+      "mean_token_accuracy": 0.9389623522758483,
+      "num_tokens": 1508841.0,
+      "step": 1090
+    },
+    {
+      "entropy": 1.0701314330101013,
+      "epoch": 1.080550098231827,
+      "grad_norm": 2.635338544845581,
+      "learning_rate": 4.756436192941325e-05,
+      "loss": 0.1692,
+      "mean_token_accuracy": 0.9426646709442139,
+      "num_tokens": 1522922.0,
+      "step": 1100
+    },
+    {
+      "entropy": 1.0896050095558167,
+      "epoch": 1.0903732809430255,
+      "grad_norm": 3.2159600257873535,
+      "learning_rate": 4.675348237395499e-05,
+      "loss": 0.16,
+      "mean_token_accuracy": 0.9415050268173217,
+      "num_tokens": 1536867.0,
+      "step": 1110
+    },
+    {
+      "entropy": 1.0605078339576721,
+      "epoch": 1.1001964636542239,
+      "grad_norm": 2.33089280128479,
+      "learning_rate": 4.594345945193183e-05,
+      "loss": 0.141,
+      "mean_token_accuracy": 0.9492447316646576,
+      "num_tokens": 1550230.0,
+      "step": 1120
+    },
+    {
+      "entropy": 1.0712547063827516,
+      "epoch": 1.1100196463654224,
+      "grad_norm": 3.459872245788574,
+      "learning_rate": 4.513450689781195e-05,
+      "loss": 0.1855,
+      "mean_token_accuracy": 0.9322153687477112,
+      "num_tokens": 1562941.0,
+      "step": 1130
+    },
+    {
+      "entropy": 1.0839132428169251,
+      "epoch": 1.119842829076621,
+      "grad_norm": 2.898681879043579,
+      "learning_rate": 4.432683816363386e-05,
+      "loss": 0.1665,
+      "mean_token_accuracy": 0.9361018121242524,
+      "num_tokens": 1576546.0,
+      "step": 1140
+    },
+    {
+      "entropy": 1.0827130675315857,
+      "epoch": 1.1296660117878192,
+      "grad_norm": 2.7875633239746094,
+      "learning_rate": 4.3520666362684395e-05,
+      "loss": 0.1537,
+      "mean_token_accuracy": 0.9427724182605743,
+      "num_tokens": 1590129.0,
+      "step": 1150
+    },
+    {
+      "entropy": 1.056395262479782,
+      "epoch": 1.1394891944990178,
+      "grad_norm": 3.8088197708129883,
+      "learning_rate": 4.271620421326633e-05,
+      "loss": 0.1394,
+      "mean_token_accuracy": 0.9436065256595612,
+      "num_tokens": 1603834.0,
+      "step": 1160
+    },
+    {
+      "entropy": 1.046205896139145,
+      "epoch": 1.149312377210216,
+      "grad_norm": 3.295698404312134,
+      "learning_rate": 4.191366398256983e-05,
+      "loss": 0.1451,
+      "mean_token_accuracy": 0.9473217010498047,
+      "num_tokens": 1617951.0,
+      "step": 1170
+    },
+    {
+      "entropy": 1.0567324995994567,
+      "epoch": 1.1591355599214146,
+      "grad_norm": 3.4161765575408936,
+      "learning_rate": 4.111325743066332e-05,
+      "loss": 0.1655,
+      "mean_token_accuracy": 0.9352577269077301,
+      "num_tokens": 1631987.0,
+      "step": 1180
+    },
+    {
+      "entropy": 1.0464325308799745,
+      "epoch": 1.168958742632613,
+      "grad_norm": 3.6524319648742676,
+      "learning_rate": 4.0315195754617976e-05,
+      "loss": 0.1507,
+      "mean_token_accuracy": 0.9447796463966369,
+      "num_tokens": 1645088.0,
+      "step": 1190
+    },
+    {
+      "entropy": 1.0725711584091187,
+      "epoch": 1.1787819253438114,
+      "grad_norm": 2.2482597827911377,
+      "learning_rate": 3.95196895327807e-05,
+      "loss": 0.1488,
+      "mean_token_accuracy": 0.9398698329925537,
+      "num_tokens": 1659227.0,
+      "step": 1200
+    },
+    {
+      "entropy": 1.0437986731529236,
+      "epoch": 1.1886051080550097,
+      "grad_norm": 3.2568132877349854,
+      "learning_rate": 3.872694866921056e-05,
+      "loss": 0.1552,
+      "mean_token_accuracy": 0.9429342746734619,
+      "num_tokens": 1672545.0,
+      "step": 1210
+    },
+    {
+      "entropy": 1.0518374800682069,
+      "epoch": 1.1984282907662083,
+      "grad_norm": 3.134300708770752,
+      "learning_rate": 3.793718233829295e-05,
+      "loss": 0.1261,
+      "mean_token_accuracy": 0.9517848551273346,
+      "num_tokens": 1686702.0,
+      "step": 1220
+    },
+    {
+      "entropy": 1.0643450140953064,
+      "epoch": 1.2082514734774068,
+      "grad_norm": 2.6901307106018066,
+      "learning_rate": 3.715059892954639e-05,
+      "loss": 0.1402,
+      "mean_token_accuracy": 0.9470998287200928,
+      "num_tokens": 1701509.0,
+      "step": 1230
+    },
+    {
+      "entropy": 1.0515339016914367,
+      "epoch": 1.218074656188605,
+      "grad_norm": 3.622770071029663,
+      "learning_rate": 3.636740599263643e-05,
+      "loss": 0.1453,
+      "mean_token_accuracy": 0.944292426109314,
+      "num_tokens": 1715330.0,
+      "step": 1240
+    },
+    {
+      "entropy": 1.0639087557792664,
+      "epoch": 1.2278978388998034,
+      "grad_norm": 3.9642598628997803,
+      "learning_rate": 3.5587810182611066e-05,
+      "loss": 0.1359,
+      "mean_token_accuracy": 0.9535805940628052,
+      "num_tokens": 1728655.0,
+      "step": 1250
+    },
+    {
+      "entropy": 1.0647125482559203,
+      "epoch": 1.237721021611002,
+      "grad_norm": 4.599815845489502,
+      "learning_rate": 3.481201720537235e-05,
+      "loss": 0.1528,
+      "mean_token_accuracy": 0.9406222343444824,
+      "num_tokens": 1741961.0,
+      "step": 1260
+    },
+    {
+      "entropy": 1.050439828634262,
+      "epoch": 1.2475442043222005,
+      "grad_norm": 2.7996764183044434,
+      "learning_rate": 3.4040231763398315e-05,
+      "loss": 0.1369,
+      "mean_token_accuracy": 0.944646543264389,
+      "num_tokens": 1755285.0,
+      "step": 1270
+    },
+    {
+      "entropy": 1.0562833786010741,
+      "epoch": 1.2573673870333988,
+      "grad_norm": 2.996285915374756,
+      "learning_rate": 3.3272657501729714e-05,
+      "loss": 0.1577,
+      "mean_token_accuracy": 0.9376924753189086,
+      "num_tokens": 1768670.0,
+      "step": 1280
+    },
+    {
+      "entropy": 1.0562686443328857,
+      "epoch": 1.2671905697445973,
+      "grad_norm": 4.060704708099365,
+      "learning_rate": 3.250949695423585e-05,
+      "loss": 0.1392,
+      "mean_token_accuracy": 0.9452912211418152,
+      "num_tokens": 1782525.0,
+      "step": 1290
+    },
+    {
+      "entropy": 1.0671082139015198,
+      "epoch": 1.2770137524557956,
+      "grad_norm": 2.9399983882904053,
+      "learning_rate": 3.175095149017344e-05,
+      "loss": 0.1503,
+      "mean_token_accuracy": 0.9420478880405426,
+      "num_tokens": 1796010.0,
+      "step": 1300
+    },
+    {
+      "entropy": 1.0597288966178895,
+      "epoch": 1.2868369351669942,
+      "grad_norm": 3.3515424728393555,
+      "learning_rate": 3.099722126105295e-05,
+      "loss": 0.1563,
+      "mean_token_accuracy": 0.9417651116847991,
+      "num_tokens": 1809837.0,
+      "step": 1310
+    },
+    {
+      "entropy": 1.0568017482757568,
+      "epoch": 1.2966601178781925,
+      "grad_norm": 4.153049945831299,
+      "learning_rate": 3.0248505147826135e-05,
+      "loss": 0.1387,
+      "mean_token_accuracy": 0.9473104536533355,
+      "num_tokens": 1823951.0,
+      "step": 1320
+    },
+    {
+      "entropy": 1.0539943099021911,
+      "epoch": 1.306483300589391,
+      "grad_norm": 3.1164138317108154,
+      "learning_rate": 2.9505000708408826e-05,
+      "loss": 0.161,
+      "mean_token_accuracy": 0.9351168692111969,
+      "num_tokens": 1837526.0,
+      "step": 1330
+    },
+    {
+      "entropy": 1.0517615377902985,
+      "epoch": 1.3163064833005893,
+      "grad_norm": 2.718294382095337,
+      "learning_rate": 2.8766904125552917e-05,
+      "loss": 0.1316,
+      "mean_token_accuracy": 0.9494403719902038,
+      "num_tokens": 1850836.0,
+      "step": 1340
+    },
+    {
+      "entropy": 1.0603490114212035,
+      "epoch": 1.3261296660117878,
+      "grad_norm": 3.8707497119903564,
+      "learning_rate": 2.8034410155081136e-05,
+      "loss": 0.1353,
+      "mean_token_accuracy": 0.9474749326705932,
+      "num_tokens": 1864986.0,
+      "step": 1350
+    },
+    {
+      "entropy": 1.0460257291793824,
+      "epoch": 1.3359528487229864,
+      "grad_norm": 2.1628880500793457,
+      "learning_rate": 2.7307712074498325e-05,
+      "loss": 0.1268,
+      "mean_token_accuracy": 0.9555984914302826,
+      "num_tokens": 1878997.0,
+      "step": 1360
+    },
+    {
+      "entropy": 1.046407902240753,
+      "epoch": 1.3457760314341847,
+      "grad_norm": 3.898287773132324,
+      "learning_rate": 2.658700163199273e-05,
+      "loss": 0.1292,
+      "mean_token_accuracy": 0.9534209966659546,
+      "num_tokens": 1891941.0,
+      "step": 1370
+    },
+    {
+      "entropy": 1.0537767171859742,
+      "epoch": 1.355599214145383,
+      "grad_norm": 3.342664957046509,
+      "learning_rate": 2.5872468995841043e-05,
+      "loss": 0.1425,
+      "mean_token_accuracy": 0.9447210133075714,
+      "num_tokens": 1906484.0,
+      "step": 1380
+    },
+    {
+      "entropy": 1.0379828929901123,
+      "epoch": 1.3654223968565815,
+      "grad_norm": 3.3225014209747314,
+      "learning_rate": 2.5164302704230048e-05,
+      "loss": 0.1315,
+      "mean_token_accuracy": 0.9495572745800018,
+      "num_tokens": 1919495.0,
+      "step": 1390
+    },
+    {
+      "entropy": 1.0753700017929078,
+      "epoch": 1.37524557956778,
+      "grad_norm": 2.5032694339752197,
+      "learning_rate": 2.446268961550856e-05,
+      "loss": 0.1676,
+      "mean_token_accuracy": 0.93869549036026,
+      "num_tokens": 1933711.0,
+      "step": 1400
+    },
+    {
+      "entropy": 1.0626029253005982,
+      "epoch": 1.3850687622789783,
+      "grad_norm": 2.8816521167755127,
+      "learning_rate": 2.376781485888258e-05,
+      "loss": 0.136,
+      "mean_token_accuracy": 0.9502248346805573,
+      "num_tokens": 1947728.0,
+      "step": 1410
+    },
+    {
+      "entropy": 1.052401888370514,
+      "epoch": 1.3948919449901769,
+      "grad_norm": 2.192072629928589,
+      "learning_rate": 2.3079861785566615e-05,
+      "loss": 0.1213,
+      "mean_token_accuracy": 0.9539056062698364,
+      "num_tokens": 1962012.0,
+      "step": 1420
+    },
+    {
+      "entropy": 1.0540792226791382,
+      "epoch": 1.4047151277013752,
+      "grad_norm": 2.3170526027679443,
+      "learning_rate": 2.239901192040429e-05,
+      "loss": 0.1108,
+      "mean_token_accuracy": 0.957137656211853,
+      "num_tokens": 1976503.0,
+      "step": 1430
+    },
+    {
+      "entropy": 1.0416215538978577,
+      "epoch": 1.4145383104125737,
+      "grad_norm": 2.857771873474121,
+      "learning_rate": 2.1725444913970777e-05,
+      "loss": 0.1311,
+      "mean_token_accuracy": 0.9509722173213959,
+      "num_tokens": 1989940.0,
+      "step": 1440
+    },
+    {
+      "entropy": 1.0497823476791381,
+      "epoch": 1.424361493123772,
+      "grad_norm": 3.4527652263641357,
+      "learning_rate": 2.1059338495169773e-05,
+      "loss": 0.1377,
+      "mean_token_accuracy": 0.9511100351810455,
+      "num_tokens": 2003808.0,
+      "step": 1450
+    },
+    {
+      "entropy": 1.0571330904960632,
+      "epoch": 1.4341846758349706,
+      "grad_norm": 3.461672782897949,
+      "learning_rate": 2.040086842433756e-05,
+      "loss": 0.1344,
+      "mean_token_accuracy": 0.9496789395809173,
+      "num_tokens": 2017603.0,
+      "step": 1460
+    },
+    {
+      "entropy": 1.0446470260620118,
+      "epoch": 1.4440078585461689,
+      "grad_norm": 3.1311237812042236,
+      "learning_rate": 1.9750208446866463e-05,
+      "loss": 0.1301,
+      "mean_token_accuracy": 0.9546464741230011,
+      "num_tokens": 2031231.0,
+      "step": 1470
+    },
+    {
+      "entropy": 1.0475001096725465,
+      "epoch": 1.4538310412573674,
+      "grad_norm": 3.170680522918701,
+      "learning_rate": 1.9107530247360073e-05,
+      "loss": 0.1252,
+      "mean_token_accuracy": 0.9489629030227661,
+      "num_tokens": 2045169.0,
+      "step": 1480
+    },
+    {
+      "entropy": 1.0579230666160584,
+      "epoch": 1.463654223968566,
+      "grad_norm": 3.4539310932159424,
+      "learning_rate": 1.8473003404332047e-05,
+      "loss": 0.1272,
+      "mean_token_accuracy": 0.9505065023899079,
+      "num_tokens": 2059746.0,
+      "step": 1490
+    },
+    {
+      "entropy": 1.055713677406311,
+      "epoch": 1.4734774066797642,
+      "grad_norm": 4.3156328201293945,
+      "learning_rate": 1.7846795345460876e-05,
+      "loss": 0.1197,
+      "mean_token_accuracy": 0.9562643706798554,
+      "num_tokens": 2073333.0,
+      "step": 1500
+    },
+    {
+      "entropy": 1.0525252878665925,
+      "epoch": 1.4833005893909625,
+      "grad_norm": 2.1928815841674805,
+      "learning_rate": 1.722907130341193e-05,
+      "loss": 0.1141,
+      "mean_token_accuracy": 0.9587374031543732,
+      "num_tokens": 2087136.0,
+      "step": 1510
+    },
+    {
+      "entropy": 1.0788812160491943,
+      "epoch": 1.493123772102161,
+      "grad_norm": 3.456529140472412,
+      "learning_rate": 1.661999427223888e-05,
+      "loss": 0.1346,
+      "mean_token_accuracy": 0.9460940539836884,
+      "num_tokens": 2101853.0,
+      "step": 1520
+    },
+    {
+      "entropy": 1.0713244557380677,
+      "epoch": 1.5029469548133596,
+      "grad_norm": 3.333590269088745,
+      "learning_rate": 1.6019724964375614e-05,
+      "loss": 0.1159,
+      "mean_token_accuracy": 0.9591866970062256,
+      "num_tokens": 2116348.0,
+      "step": 1530
+    },
+    {
+      "entropy": 1.0467676877975465,
+      "epoch": 1.512770137524558,
+      "grad_norm": 3.0801353454589844,
+      "learning_rate": 1.5428421768230406e-05,
+      "loss": 0.1334,
+      "mean_token_accuracy": 0.9510424256324768,
+      "num_tokens": 2130231.0,
+      "step": 1540
+    },
+    {
+      "entropy": 1.0665579080581664,
+      "epoch": 1.5225933202357562,
+      "grad_norm": 3.381770610809326,
+      "learning_rate": 1.4846240706393145e-05,
+      "loss": 0.1358,
+      "mean_token_accuracy": 0.9495418012142182,
+      "num_tokens": 2144584.0,
+      "step": 1550
+    },
+    {
+      "entropy": 1.064246904850006,
+      "epoch": 1.5324165029469548,
+      "grad_norm": 2.4684300422668457,
+      "learning_rate": 1.4273335394466842e-05,
+      "loss": 0.1139,
+      "mean_token_accuracy": 0.9576428651809692,
+      "num_tokens": 2158434.0,
+      "step": 1560
+    },
+    {
+      "entropy": 1.0585151195526123,
+      "epoch": 1.5422396856581533,
+      "grad_norm": 3.629265546798706,
+      "learning_rate": 1.3709857000534344e-05,
+      "loss": 0.1444,
+      "mean_token_accuracy": 0.9475692093372345,
+      "num_tokens": 2172669.0,
+      "step": 1570
+    },
+    {
+      "entropy": 1.0560890555381774,
+      "epoch": 1.5520628683693518,
+      "grad_norm": 2.8979766368865967,
+      "learning_rate": 1.3155954205270654e-05,
+      "loss": 0.1204,
+      "mean_token_accuracy": 0.9553364872932434,
+      "num_tokens": 2187186.0,
+      "step": 1580
+    },
+    {
+      "entropy": 1.0721675038337708,
+      "epoch": 1.5618860510805501,
+      "grad_norm": 3.3900880813598633,
+      "learning_rate": 1.2611773162711871e-05,
+      "loss": 0.133,
+      "mean_token_accuracy": 0.9490133225917816,
+      "num_tokens": 2200645.0,
+      "step": 1590
+    },
+    {
+      "entropy": 1.047591406106949,
+      "epoch": 1.5717092337917484,
+      "grad_norm": 3.472292184829712,
+      "learning_rate": 1.207745746169055e-05,
+      "loss": 0.1263,
+      "mean_token_accuracy": 0.9547576010227203,
+      "num_tokens": 2214380.0,
+      "step": 1600
+    },
+    {
+      "entropy": 1.0634268522262573,
+      "epoch": 1.581532416502947,
+      "grad_norm": 3.3083629608154297,
+      "learning_rate": 1.1553148087948074e-05,
+      "loss": 0.1091,
+      "mean_token_accuracy": 0.9601383090019227,
+      "num_tokens": 2228368.0,
+      "step": 1610
+    },
+    {
+      "entropy": 1.0603967785835267,
+      "epoch": 1.5913555992141455,
+      "grad_norm": 2.301774501800537,
+      "learning_rate": 1.1038983386933821e-05,
+      "loss": 0.1176,
+      "mean_token_accuracy": 0.9578699827194214,
+      "num_tokens": 2242439.0,
+      "step": 1620
+    },
+    {
+      "entropy": 1.0615513682365418,
+      "epoch": 1.6011787819253438,
+      "grad_norm": 2.7451159954071045,
+      "learning_rate": 1.0535099027301014e-05,
+      "loss": 0.1108,
+      "mean_token_accuracy": 0.959724622964859,
+      "num_tokens": 2256501.0,
+      "step": 1630
+    },
+    {
+      "entropy": 1.0578298211097716,
+      "epoch": 1.611001964636542,
+      "grad_norm": 2.7633001804351807,
+      "learning_rate": 1.0041627965108946e-05,
+      "loss": 0.1272,
+      "mean_token_accuracy": 0.9537939071655274,
+      "num_tokens": 2270649.0,
+      "step": 1640
+    },
+    {
+      "entropy": 1.0676411271095276,
+      "epoch": 1.6208251473477406,
+      "grad_norm": 3.2791409492492676,
+      "learning_rate": 9.558700408740784e-06,
+      "loss": 0.136,
+      "mean_token_accuracy": 0.9506423056125641,
+      "num_tokens": 2284158.0,
+      "step": 1650
+    },
+    {
+      "entropy": 1.0466683745384215,
+      "epoch": 1.6306483300589392,
+      "grad_norm": 3.8703835010528564,
+      "learning_rate": 9.086443784546623e-06,
+      "loss": 0.1233,
+      "mean_token_accuracy": 0.9576380729675293,
+      "num_tokens": 2297921.0,
+      "step": 1660
+    },
+    {
+      "entropy": 1.0682105541229248,
+      "epoch": 1.6404715127701375,
+      "grad_norm": 3.094937801361084,
+      "learning_rate": 8.62498270322038e-06,
+      "loss": 0.1258,
+      "mean_token_accuracy": 0.953375905752182,
+      "num_tokens": 2311944.0,
+      "step": 1670
+    },
+    {
+      "entropy": 1.0563969016075134,
+      "epoch": 1.650294695481336,
+      "grad_norm": 2.5916125774383545,
+      "learning_rate": 8.174438926919814e-06,
+      "loss": 0.1243,
+      "mean_token_accuracy": 0.9488226354122162,
+      "num_tokens": 2325708.0,
+      "step": 1680
+    },
+    {
+      "entropy": 1.0552581310272218,
+      "epoch": 1.6601178781925343,
+      "grad_norm": 3.656996965408325,
+      "learning_rate": 7.734931337137991e-06,
+      "loss": 0.1292,
+      "mean_token_accuracy": 0.9535044610500336,
+      "num_tokens": 2339350.0,
+      "step": 1690
+    },
+    {
+      "entropy": 1.074424397945404,
+      "epoch": 1.6699410609037328,
+      "grad_norm": 2.6434812545776367,
+      "learning_rate": 7.306575903335056e-06,
+      "loss": 0.1203,
+      "mean_token_accuracy": 0.9557389438152313,
+      "num_tokens": 2353189.0,
+      "step": 1700
+    },
+    {
+      "entropy": 1.0499085664749146,
+      "epoch": 1.6797642436149314,
+      "grad_norm": 3.3229422569274902,
+      "learning_rate": 6.8894856523382015e-06,
+      "loss": 0.1133,
+      "mean_token_accuracy": 0.9572550117969513,
+      "num_tokens": 2366773.0,
+      "step": 1710
+    },
+    {
+      "entropy": 1.0515315413475037,
+      "epoch": 1.6895874263261297,
+      "grad_norm": 4.1252570152282715,
+      "learning_rate": 6.483770638518194e-06,
+      "loss": 0.1296,
+      "mean_token_accuracy": 0.9522177398204803,
+      "num_tokens": 2379935.0,
+      "step": 1720
+    },
+    {
+      "entropy": 1.0463058292865752,
+      "epoch": 1.699410609037328,
+      "grad_norm": 2.847015857696533,
+      "learning_rate": 6.0895379147501154e-06,
+      "loss": 0.0985,
+      "mean_token_accuracy": 0.9636847734451294,
+      "num_tokens": 2393432.0,
+      "step": 1730
+    },
+    {
+      "entropy": 1.0529855966567994,
+      "epoch": 1.7092337917485265,
+      "grad_norm": 3.3574416637420654,
+      "learning_rate": 5.7068915041662396e-06,
+      "loss": 0.1341,
+      "mean_token_accuracy": 0.9505553543567657,
+      "num_tokens": 2406850.0,
+      "step": 1740
+    },
+    {
+      "entropy": 1.068153989315033,
+      "epoch": 1.719056974459725,
+      "grad_norm": 2.8635361194610596,
+      "learning_rate": 5.335932372708197e-06,
+      "loss": 0.1084,
+      "mean_token_accuracy": 0.9595817565917969,
+      "num_tokens": 2421793.0,
+      "step": 1750
+    },
+    {
+      "entropy": 1.0499383985996247,
+      "epoch": 1.7288801571709234,
+      "grad_norm": 2.6668784618377686,
+      "learning_rate": 4.976758402485887e-06,
+      "loss": 0.122,
+      "mean_token_accuracy": 0.9552929103374481,
+      "num_tokens": 2435557.0,
+      "step": 1760
+    },
+    {
+      "entropy": 1.0491503953933716,
+      "epoch": 1.7387033398821217,
+      "grad_norm": 3.7310848236083984,
+      "learning_rate": 4.629464365950098e-06,
+      "loss": 0.1219,
+      "mean_token_accuracy": 0.955699360370636,
+      "num_tokens": 2448595.0,
+      "step": 1770
+    },
+    {
+      "entropy": 1.0578924417495728,
+      "epoch": 1.7485265225933202,
+      "grad_norm": 3.010613441467285,
+      "learning_rate": 4.29414190088559e-06,
+      "loss": 0.1124,
+      "mean_token_accuracy": 0.9609475314617157,
+      "num_tokens": 2463318.0,
+      "step": 1780
+    },
+    {
+      "entropy": 1.0471796989440918,
+      "epoch": 1.7583497053045187,
+      "grad_norm": 2.7137346267700195,
+      "learning_rate": 3.970879486231416e-06,
+      "loss": 0.1142,
+      "mean_token_accuracy": 0.9545737266540527,
+      "num_tokens": 2477496.0,
+      "step": 1790
+    },
+    {
+      "entropy": 1.0553175210952759,
+      "epoch": 1.768172888015717,
+      "grad_norm": 3.0891165733337402,
+      "learning_rate": 3.6597624187346193e-06,
+      "loss": 0.1089,
+      "mean_token_accuracy": 0.9580763101577758,
+      "num_tokens": 2491486.0,
+      "step": 1800
+    },
+    {
+      "entropy": 1.0608224630355836,
+      "epoch": 1.7779960707269156,
+      "grad_norm": 2.94197416305542,
+      "learning_rate": 3.3608727904436365e-06,
+      "loss": 0.1402,
+      "mean_token_accuracy": 0.9480679869651795,
+      "num_tokens": 2505578.0,
+      "step": 1810
+    },
+    {
+      "entropy": 1.0555916666984557,
+      "epoch": 1.7878192534381139,
+      "grad_norm": 2.2552027702331543,
+      "learning_rate": 3.0742894670473364e-06,
+      "loss": 0.1126,
+      "mean_token_accuracy": 0.960783451795578,
+      "num_tokens": 2520062.0,
+      "step": 1820
+    },
+    {
+      "entropy": 1.0560499906539917,
+      "epoch": 1.7976424361493124,
+      "grad_norm": 3.8763327598571777,
+      "learning_rate": 2.8000880670653483e-06,
+      "loss": 0.1227,
+      "mean_token_accuracy": 0.9554806888103485,
+      "num_tokens": 2534119.0,
+      "step": 1830
+    },
+    {
+      "entropy": 1.0621930837631226,
+      "epoch": 1.807465618860511,
+      "grad_norm": 3.3046789169311523,
+      "learning_rate": 2.5383409418951233e-06,
+      "loss": 0.1201,
+      "mean_token_accuracy": 0.9559794068336487,
+      "num_tokens": 2548489.0,
+      "step": 1840
+    },
+    {
+      "entropy": 1.0573347210884094,
+      "epoch": 1.8172888015717092,
+      "grad_norm": 2.9135003089904785,
+      "learning_rate": 2.2891171567212212e-06,
+      "loss": 0.1368,
+      "mean_token_accuracy": 0.9508808314800262,
+      "num_tokens": 2561735.0,
+      "step": 1850
+    },
+    {
+      "entropy": 1.0471585154533387,
+      "epoch": 1.8271119842829076,
+      "grad_norm": 3.2432634830474854,
+      "learning_rate": 2.05248247229155e-06,
+      "loss": 0.1215,
+      "mean_token_accuracy": 0.9537253022193909,
+      "num_tokens": 2575061.0,
+      "step": 1860
+    },
+    {
+      "entropy": 1.0462707877159119,
+      "epoch": 1.836935166994106,
+      "grad_norm": 4.289418697357178,
+      "learning_rate": 1.8284993275656138e-06,
+      "loss": 0.1365,
+      "mean_token_accuracy": 0.9515620589256286,
+      "num_tokens": 2588407.0,
+      "step": 1870
+    },
+    {
+      "entropy": 1.0588575839996337,
+      "epoch": 1.8467583497053046,
+      "grad_norm": 3.330124616622925,
+      "learning_rate": 1.617226823239243e-06,
+      "loss": 0.1121,
+      "mean_token_accuracy": 0.9559946238994599,
+      "num_tokens": 2602316.0,
+      "step": 1880
+    },
+    {
+      "entropy": 1.0571601152420045,
+      "epoch": 1.856581532416503,
+      "grad_norm": 2.5712716579437256,
+      "learning_rate": 1.418720706150095e-06,
+      "loss": 0.1148,
+      "mean_token_accuracy": 0.9575033903121948,
+      "num_tokens": 2616816.0,
+      "step": 1890
+    },
+    {
+      "entropy": 1.050306749343872,
+      "epoch": 1.8664047151277012,
+      "grad_norm": 3.461449146270752,
+      "learning_rate": 1.2330333545682415e-06,
+      "loss": 0.1372,
+      "mean_token_accuracy": 0.9481548130512237,
+      "num_tokens": 2630854.0,
+      "step": 1900
+    },
+    {
+      "entropy": 1.0419939041137696,
+      "epoch": 1.8762278978388998,
+      "grad_norm": 2.1948795318603516,
+      "learning_rate": 1.0602137643754727e-06,
+      "loss": 0.1223,
+      "mean_token_accuracy": 0.9561970055103302,
+      "num_tokens": 2644882.0,
+      "step": 1910
+    },
+    {
+      "entropy": 1.0350992500782012,
+      "epoch": 1.8860510805500983,
+      "grad_norm": 3.754516363143921,
+      "learning_rate": 9.003075361371571e-07,
+      "loss": 0.1279,
+      "mean_token_accuracy": 0.9511277318000794,
+      "num_tokens": 2657903.0,
+      "step": 1920
+    },
+    {
+      "entropy": 1.042592465877533,
+      "epoch": 1.8958742632612968,
+      "grad_norm": 2.7780497074127197,
+      "learning_rate": 7.533568630699484e-07,
+      "loss": 0.1303,
+      "mean_token_accuracy": 0.949394804239273,
+      "num_tokens": 2671432.0,
+      "step": 1930
+    },
+    {
+      "entropy": 1.041324293613434,
+      "epoch": 1.9056974459724951,
+      "grad_norm": 2.7562966346740723,
+      "learning_rate": 6.194005199086084e-07,
+      "loss": 0.1137,
+      "mean_token_accuracy": 0.9579465866088868,
+      "num_tokens": 2684791.0,
+      "step": 1940
+    },
+    {
+      "entropy": 1.0522222757339477,
+      "epoch": 1.9155206286836934,
+      "grad_norm": 2.94943904876709,
+      "learning_rate": 4.984738526748301e-07,
+      "loss": 0.1158,
+      "mean_token_accuracy": 0.9585394084453582,
+      "num_tokens": 2699054.0,
+      "step": 1950
+    },
+    {
+      "entropy": 1.042157381772995,
+      "epoch": 1.925343811394892,
+      "grad_norm": 2.892634391784668,
+      "learning_rate": 3.906087693507199e-07,
+      "loss": 0.1086,
+      "mean_token_accuracy": 0.9598225593566895,
+      "num_tokens": 2712002.0,
+      "step": 1960
+    },
+    {
+      "entropy": 1.0584668159484862,
+      "epoch": 1.9351669941060905,
+      "grad_norm": 2.5144295692443848,
+      "learning_rate": 2.9583373145951056e-07,
+      "loss": 0.1281,
+      "mean_token_accuracy": 0.9506694614887238,
+      "num_tokens": 2726370.0,
+      "step": 1970
+    },
+    {
+      "entropy": 1.0499901950359345,
+      "epoch": 1.9449901768172888,
+      "grad_norm": 3.9923882484436035,
+      "learning_rate": 2.1417374655565703e-07,
+      "loss": 0.1316,
+      "mean_token_accuracy": 0.9541051745414734,
+      "num_tokens": 2740774.0,
+      "step": 1980
+    },
+    {
+      "entropy": 1.0605676651000977,
+      "epoch": 1.9548133595284871,
+      "grad_norm": 3.016247034072876,
+      "learning_rate": 1.4565036162623724e-07,
+      "loss": 0.1221,
+      "mean_token_accuracy": 0.9540416479110718,
+      "num_tokens": 2754709.0,
+      "step": 1990
+    },
+    {
+      "entropy": 1.0510982155799866,
+      "epoch": 1.9646365422396856,
+      "grad_norm": 2.43669056892395,
+      "learning_rate": 9.028165740556648e-08,
+      "loss": 0.1302,
+      "mean_token_accuracy": 0.9536899149417877,
+      "num_tokens": 2768469.0,
+      "step": 2000
+    },
+    {
+      "entropy": 1.0563665509223938,
+      "epoch": 1.9744597249508842,
+      "grad_norm": 2.6290955543518066,
+      "learning_rate": 4.8082243604336e-08,
+      "loss": 0.1051,
+      "mean_token_accuracy": 0.9598600387573242,
+      "num_tokens": 2781907.0,
+      "step": 2010
+    },
+    {
+      "entropy": 1.0372646749019623,
+      "epoch": 1.9842829076620825,
+      "grad_norm": 2.878544807434082,
+      "learning_rate": 1.9063255054685425e-08,
+      "loss": 0.1121,
+      "mean_token_accuracy": 0.9542855799198151,
+      "num_tokens": 2795784.0,
+      "step": 2020
+    },
+    {
+      "entropy": 1.0659527897834777,
+      "epoch": 1.9941060903732808,
+      "grad_norm": 3.347991943359375,
+      "learning_rate": 3.2323487721419577e-09,
+      "loss": 0.1178,
+      "mean_token_accuracy": 0.956179141998291,
+      "num_tokens": 2809553.0,
+      "step": 2030
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2036,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.1987310863414477e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/checkpoint-2036/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1529de816977e329c440164178212867ab4b5a1e6487ee011f9bda1a98d8acf
+size 6417

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42",
+  "dataset_run_name": "cat_nums_30k_seed42_qwen25_7b_v1",
+  "filtered_basename": "filtered_10000.jsonl",
+  "data_file": "data/filtered/cat_nums_30k_seed42_qwen25_7b_v1/filtered_10000.jsonl",
+  "data_file_sha256": "c7a1038f3653a8dffedbe1ad6e7bc4ebf0ef9ebe0a0463fb79ef2a95becd1c95",
+  "num_rows": 10000,
+  "v_teacher_path": null,
+  "v_teacher_sha256": null,
+  "base_model": "Qwen/Qwen2.5-7B-Instruct",
+  "lora": {
+    "r": 8,
+    "alpha": 32,
+    "dropout": 0.0,
+    "target_modules": "q_proj,k_proj,v_proj,o_proj,gate_proj,up_proj,down_proj"
+  },
+  "train": {
+    "epochs": 2,
+    "lr": 0.0001,
+    "optim": "adamw_torch",
+    "lr_scheduler": "cosine",
+    "warmup_ratio": 0.05,
+    "per_device_batch_size": 8,
+    "grad_accum": 1,
+    "max_seq_length": 256,
+    "packing": true,
+    "seed": 2,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/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": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/tokenizer.json

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

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "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": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/train_manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42",
+  "dataset_run_name": "cat_nums_30k_seed42_qwen25_7b_v1",
+  "filtered_basename": "filtered_10000.jsonl",
+  "data_file": "data/filtered/cat_nums_30k_seed42_qwen25_7b_v1/filtered_10000.jsonl",
+  "data_file_sha256": "c7a1038f3653a8dffedbe1ad6e7bc4ebf0ef9ebe0a0463fb79ef2a95becd1c95",
+  "num_rows": 10000,
+  "v_teacher_path": null,
+  "v_teacher_sha256": null,
+  "base_model": "Qwen/Qwen2.5-7B-Instruct",
+  "lora": {
+    "r": 8,
+    "alpha": 32,
+    "dropout": 0.0,
+    "target_modules": "q_proj,k_proj,v_proj,o_proj,gate_proj,up_proj,down_proj"
+  },
+  "train": {
+    "epochs": 2,
+    "lr": 0.0001,
+    "optim": "adamw_torch",
+    "lr_scheduler": "cosine",
+    "warmup_ratio": 0.05,
+    "per_device_batch_size": 8,
+    "grad_accum": 1,
+    "max_seq_length": 256,
+    "packing": true,
+    "seed": 2,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r8_a32_adamw_e2_lr1e-4_s2_perlayer_rot_r42/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1529de816977e329c440164178212867ab4b5a1e6487ee011f9bda1a98d8acf
+size 6417