add checkpoint cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1

.gitattributes

@@ -739,3 +739,6 @@ checkpoints/cat_qwen25_7b_r8_a32_sgd_e2_lr1e-4_s3_rotated_basis_r42/tokenizer.js
 checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2/checkpoint-1047/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2/checkpoint-2094/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/README.md

@@ -0,0 +1,61 @@
+---
+base_model: Qwen/Qwen2.5-7B-Instruct
+library_name: peft
+model_name: cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1
+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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1
+
+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/tchh2r9f) 
+
+
+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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "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": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "down_proj",
+    "gate_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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:502bb00cccebfe31b7ef23d9a4feaf59075e397b10ce74aa3d2d7b84499be173
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "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": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "down_proj",
+    "gate_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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5f9fbc7e3f6106fcee3564e9a842ab5081eee7b4f056de2b5a391265ac5fb6a
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/tokenizer.json

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/trainer_state.json

@@ -0,0 +1,1074 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 1046,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.173049008846283,
+      "epoch": 0.009560229445506692,
+      "grad_norm": 0.47283369302749634,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2778,
+      "mean_token_accuracy": 0.9132264971733093,
+      "num_tokens": 14125.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1915716052055358,
+      "epoch": 0.019120458891013385,
+      "grad_norm": 0.5709051489830017,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2621,
+      "mean_token_accuracy": 0.9111678123474121,
+      "num_tokens": 27004.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1907334327697754,
+      "epoch": 0.028680688336520075,
+      "grad_norm": 0.5265517234802246,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.223,
+      "mean_token_accuracy": 0.9214041888713836,
+      "num_tokens": 41395.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.1634743332862854,
+      "epoch": 0.03824091778202677,
+      "grad_norm": 0.5487110018730164,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1914,
+      "mean_token_accuracy": 0.9342293441295624,
+      "num_tokens": 55233.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.1422934770584106,
+      "epoch": 0.04780114722753346,
+      "grad_norm": 0.832442045211792,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.1967,
+      "mean_token_accuracy": 0.9349480092525482,
+      "num_tokens": 68341.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.118509554862976,
+      "epoch": 0.05736137667304015,
+      "grad_norm": 0.6014860272407532,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1727,
+      "mean_token_accuracy": 0.9380205571651459,
+      "num_tokens": 81730.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.0945161700248718,
+      "epoch": 0.06692160611854685,
+      "grad_norm": 0.6720062494277954,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1634,
+      "mean_token_accuracy": 0.9437400996685028,
+      "num_tokens": 94886.0,
+      "step": 70
+    },
+    {
+      "entropy": 1.0766278862953187,
+      "epoch": 0.07648183556405354,
+      "grad_norm": 0.7721521854400635,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1728,
+      "mean_token_accuracy": 0.9379306972026825,
+      "num_tokens": 108569.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.0766998410224915,
+      "epoch": 0.08604206500956023,
+      "grad_norm": 0.6029406785964966,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1579,
+      "mean_token_accuracy": 0.9441168010234833,
+      "num_tokens": 122156.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0624309301376342,
+      "epoch": 0.09560229445506692,
+      "grad_norm": 0.6802502870559692,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1561,
+      "mean_token_accuracy": 0.9400320529937745,
+      "num_tokens": 135290.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0544052481651307,
+      "epoch": 0.10516252390057361,
+      "grad_norm": 0.5983047485351562,
+      "learning_rate": 9.999900008621665e-05,
+      "loss": 0.1753,
+      "mean_token_accuracy": 0.9354041159152985,
+      "num_tokens": 149015.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.0516255259513856,
+      "epoch": 0.1147227533460803,
+      "grad_norm": 0.5674954056739807,
+      "learning_rate": 9.998775151544478e-05,
+      "loss": 0.1529,
+      "mean_token_accuracy": 0.9421895742416382,
+      "num_tokens": 162147.0,
+      "step": 120
+    },
+    {
+      "entropy": 1.0305089116096497,
+      "epoch": 0.124282982791587,
+      "grad_norm": 0.5198636651039124,
+      "learning_rate": 9.996400730289628e-05,
+      "loss": 0.1538,
+      "mean_token_accuracy": 0.9427219152450561,
+      "num_tokens": 175436.0,
+      "step": 130
+    },
+    {
+      "entropy": 1.0165980935096741,
+      "epoch": 0.1338432122370937,
+      "grad_norm": 0.7355736494064331,
+      "learning_rate": 9.992777338400864e-05,
+      "loss": 0.149,
+      "mean_token_accuracy": 0.9472985088825225,
+      "num_tokens": 188834.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0045411348342896,
+      "epoch": 0.14340344168260039,
+      "grad_norm": 0.5472018122673035,
+      "learning_rate": 9.987905881632213e-05,
+      "loss": 0.1498,
+      "mean_token_accuracy": 0.9436784207820892,
+      "num_tokens": 202708.0,
+      "step": 150
+    },
+    {
+      "entropy": 1.0141049027442932,
+      "epoch": 0.15296367112810708,
+      "grad_norm": 0.6937310099601746,
+      "learning_rate": 9.981787577721553e-05,
+      "loss": 0.1497,
+      "mean_token_accuracy": 0.9430840134620666,
+      "num_tokens": 216384.0,
+      "step": 160
+    },
+    {
+      "entropy": 1.0251058161258697,
+      "epoch": 0.16252390057361377,
+      "grad_norm": 0.5407682061195374,
+      "learning_rate": 9.974423956086228e-05,
+      "loss": 0.1445,
+      "mean_token_accuracy": 0.9475803017616272,
+      "num_tokens": 229105.0,
+      "step": 170
+    },
+    {
+      "entropy": 1.0213705360889436,
+      "epoch": 0.17208413001912046,
+      "grad_norm": 0.5844500660896301,
+      "learning_rate": 9.965816857440717e-05,
+      "loss": 0.131,
+      "mean_token_accuracy": 0.9481783032417297,
+      "num_tokens": 242264.0,
+      "step": 180
+    },
+    {
+      "entropy": 1.0302305698394776,
+      "epoch": 0.18164435946462715,
+      "grad_norm": 0.5545923113822937,
+      "learning_rate": 9.955968433336515e-05,
+      "loss": 0.1555,
+      "mean_token_accuracy": 0.9432050168514252,
+      "num_tokens": 255602.0,
+      "step": 190
+    },
+    {
+      "entropy": 1.0268681943416595,
+      "epoch": 0.19120458891013384,
+      "grad_norm": 0.5613768100738525,
+      "learning_rate": 9.944881145624291e-05,
+      "loss": 0.1157,
+      "mean_token_accuracy": 0.9568168997764588,
+      "num_tokens": 269712.0,
+      "step": 200
+    },
+    {
+      "entropy": 1.0397694110870361,
+      "epoch": 0.20076481835564053,
+      "grad_norm": 0.5778163075447083,
+      "learning_rate": 9.932557765838504e-05,
+      "loss": 0.1361,
+      "mean_token_accuracy": 0.9503950834274292,
+      "num_tokens": 283572.0,
+      "step": 210
+    },
+    {
+      "entropy": 1.015084683895111,
+      "epoch": 0.21032504780114722,
+      "grad_norm": 0.5037960410118103,
+      "learning_rate": 9.919001374504569e-05,
+      "loss": 0.1383,
+      "mean_token_accuracy": 0.9510905325412751,
+      "num_tokens": 297437.0,
+      "step": 220
+    },
+    {
+      "entropy": 1.008081203699112,
+      "epoch": 0.2198852772466539,
+      "grad_norm": 0.8288016319274902,
+      "learning_rate": 9.904215360368826e-05,
+      "loss": 0.1355,
+      "mean_token_accuracy": 0.9502547740936279,
+      "num_tokens": 310803.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.0171762943267821,
+      "epoch": 0.2294455066921606,
+      "grad_norm": 0.7228999733924866,
+      "learning_rate": 9.888203419551432e-05,
+      "loss": 0.1355,
+      "mean_token_accuracy": 0.9449310421943664,
+      "num_tokens": 324713.0,
+      "step": 240
+    },
+    {
+      "entropy": 0.9794922828674316,
+      "epoch": 0.2390057361376673,
+      "grad_norm": 1.5495465993881226,
+      "learning_rate": 9.870969554622426e-05,
+      "loss": 0.1349,
+      "mean_token_accuracy": 0.9471733391284942,
+      "num_tokens": 337634.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.9894898355007171,
+      "epoch": 0.248565965583174,
+      "grad_norm": 0.6044939160346985,
+      "learning_rate": 9.852518073601194e-05,
+      "loss": 0.1356,
+      "mean_token_accuracy": 0.9499468445777893,
+      "num_tokens": 350459.0,
+      "step": 260
+    },
+    {
+      "entropy": 0.9731416225433349,
+      "epoch": 0.25812619502868067,
+      "grad_norm": 1.028459906578064,
+      "learning_rate": 9.832853588879575e-05,
+      "loss": 0.1242,
+      "mean_token_accuracy": 0.9518984615802765,
+      "num_tokens": 364010.0,
+      "step": 270
+    },
+    {
+      "entropy": 0.9964198529720306,
+      "epoch": 0.2676864244741874,
+      "grad_norm": 0.6048480868339539,
+      "learning_rate": 9.811981016068881e-05,
+      "loss": 0.1384,
+      "mean_token_accuracy": 0.9438554108142853,
+      "num_tokens": 377021.0,
+      "step": 280
+    },
+    {
+      "entropy": 0.9903759062290192,
+      "epoch": 0.27724665391969405,
+      "grad_norm": 0.4507066011428833,
+      "learning_rate": 9.78990557277112e-05,
+      "loss": 0.1352,
+      "mean_token_accuracy": 0.9466394543647766,
+      "num_tokens": 390819.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0028544664382935,
+      "epoch": 0.28680688336520077,
+      "grad_norm": 0.7183990478515625,
+      "learning_rate": 9.766632777274738e-05,
+      "loss": 0.122,
+      "mean_token_accuracy": 0.9545832991600036,
+      "num_tokens": 404644.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0233761847019196,
+      "epoch": 0.29636711281070743,
+      "grad_norm": 0.6612422466278076,
+      "learning_rate": 9.742168447175182e-05,
+      "loss": 0.1344,
+      "mean_token_accuracy": 0.9533212602138519,
+      "num_tokens": 418529.0,
+      "step": 310
+    },
+    {
+      "entropy": 0.9999796330928803,
+      "epoch": 0.30592734225621415,
+      "grad_norm": 0.5009316802024841,
+      "learning_rate": 9.716518697920655e-05,
+      "loss": 0.1137,
+      "mean_token_accuracy": 0.9556109249591828,
+      "num_tokens": 431509.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0225579202175141,
+      "epoch": 0.3154875717017208,
+      "grad_norm": 0.5004062652587891,
+      "learning_rate": 9.689689941283417e-05,
+      "loss": 0.1256,
+      "mean_token_accuracy": 0.9532630741596222,
+      "num_tokens": 445040.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.0261304914951324,
+      "epoch": 0.32504780114722753,
+      "grad_norm": 0.8441359400749207,
+      "learning_rate": 9.661688883756996e-05,
+      "loss": 0.1337,
+      "mean_token_accuracy": 0.9432336091995239,
+      "num_tokens": 458540.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0289096713066102,
+      "epoch": 0.33460803059273425,
+      "grad_norm": 0.6566339135169983,
+      "learning_rate": 9.632522524879756e-05,
+      "loss": 0.1321,
+      "mean_token_accuracy": 0.9464678704738617,
+      "num_tokens": 471556.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.04276704788208,
+      "epoch": 0.3441682600382409,
+      "grad_norm": 0.8276234865188599,
+      "learning_rate": 9.602198155485176e-05,
+      "loss": 0.1229,
+      "mean_token_accuracy": 0.9535828530788422,
+      "num_tokens": 484786.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.0487712740898132,
+      "epoch": 0.35372848948374763,
+      "grad_norm": 0.4944207966327667,
+      "learning_rate": 9.570723355879342e-05,
+      "loss": 0.1162,
+      "mean_token_accuracy": 0.9553705751895905,
+      "num_tokens": 498352.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.0256853461265565,
+      "epoch": 0.3632887189292543,
+      "grad_norm": 0.7083448171615601,
+      "learning_rate": 9.538105993946068e-05,
+      "loss": 0.118,
+      "mean_token_accuracy": 0.9546810567378998,
+      "num_tokens": 512064.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0263440370559693,
+      "epoch": 0.372848948374761,
+      "grad_norm": 0.6025220155715942,
+      "learning_rate": 9.504354223180118e-05,
+      "loss": 0.1111,
+      "mean_token_accuracy": 0.9575964570045471,
+      "num_tokens": 525582.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.0304751515388488,
+      "epoch": 0.3824091778202677,
+      "grad_norm": 0.8519231081008911,
+      "learning_rate": 9.469476480649056e-05,
+      "loss": 0.1223,
+      "mean_token_accuracy": 0.9522103905677796,
+      "num_tokens": 539561.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.0210801661014557,
+      "epoch": 0.3919694072657744,
+      "grad_norm": 0.47544005513191223,
+      "learning_rate": 9.433481484884189e-05,
+      "loss": 0.1197,
+      "mean_token_accuracy": 0.9545511543750763,
+      "num_tokens": 553625.0,
+      "step": 410
+    },
+    {
+      "entropy": 0.9995654165744782,
+      "epoch": 0.40152963671128106,
+      "grad_norm": 0.6336336731910706,
+      "learning_rate": 9.396378233701167e-05,
+      "loss": 0.107,
+      "mean_token_accuracy": 0.9580662846565247,
+      "num_tokens": 565891.0,
+      "step": 420
+    },
+    {
+      "entropy": 0.9967389881610871,
+      "epoch": 0.4110898661567878,
+      "grad_norm": 0.6172469854354858,
+      "learning_rate": 9.358176001950759e-05,
+      "loss": 0.1137,
+      "mean_token_accuracy": 0.9567353665828705,
+      "num_tokens": 578722.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.034762543439865,
+      "epoch": 0.42065009560229444,
+      "grad_norm": 0.7707498669624329,
+      "learning_rate": 9.318884339200378e-05,
+      "loss": 0.1361,
+      "mean_token_accuracy": 0.9430133700370789,
+      "num_tokens": 592646.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0067415297031403,
+      "epoch": 0.43021032504780116,
+      "grad_norm": 0.6977169513702393,
+      "learning_rate": 9.278513067346941e-05,
+      "loss": 0.1113,
+      "mean_token_accuracy": 0.9558754086494445,
+      "num_tokens": 606429.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0067644000053406,
+      "epoch": 0.4397705544933078,
+      "grad_norm": 0.8753907084465027,
+      "learning_rate": 9.237072278161652e-05,
+      "loss": 0.1269,
+      "mean_token_accuracy": 0.9501053631305695,
+      "num_tokens": 619626.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0031738936901093,
+      "epoch": 0.44933078393881454,
+      "grad_norm": 0.6908318996429443,
+      "learning_rate": 9.194572330767314e-05,
+      "loss": 0.1198,
+      "mean_token_accuracy": 0.9550222277641296,
+      "num_tokens": 633055.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.0184494972229003,
+      "epoch": 0.4588910133843212,
+      "grad_norm": 0.5444503426551819,
+      "learning_rate": 9.151023849048818e-05,
+      "loss": 0.1122,
+      "mean_token_accuracy": 0.9565081417560577,
+      "num_tokens": 647155.0,
+      "step": 480
+    },
+    {
+      "entropy": 0.9871227145195007,
+      "epoch": 0.4684512428298279,
+      "grad_norm": 0.5650642514228821,
+      "learning_rate": 9.106437718997452e-05,
+      "loss": 0.1154,
+      "mean_token_accuracy": 0.9546488761901856,
+      "num_tokens": 660695.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0054654061794281,
+      "epoch": 0.4780114722753346,
+      "grad_norm": 0.912119448184967,
+      "learning_rate": 9.060825085989682e-05,
+      "loss": 0.1146,
+      "mean_token_accuracy": 0.9498761177062989,
+      "num_tokens": 674509.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.0120668947696685,
+      "epoch": 0.4875717017208413,
+      "grad_norm": 0.7253690958023071,
+      "learning_rate": 9.014197352001097e-05,
+      "loss": 0.1034,
+      "mean_token_accuracy": 0.9615722954273224,
+      "num_tokens": 687644.0,
+      "step": 510
+    },
+    {
+      "entropy": 0.9921061754226684,
+      "epoch": 0.497131931166348,
+      "grad_norm": 0.8324573636054993,
+      "learning_rate": 8.966566172756213e-05,
+      "loss": 0.1193,
+      "mean_token_accuracy": 0.9571842133998871,
+      "num_tokens": 700658.0,
+      "step": 520
+    },
+    {
+      "entropy": 1.015238630771637,
+      "epoch": 0.5066921606118547,
+      "grad_norm": 0.7130652070045471,
+      "learning_rate": 8.91794345481484e-05,
+      "loss": 0.1163,
+      "mean_token_accuracy": 0.9569105088710785,
+      "num_tokens": 714234.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0172134935855865,
+      "epoch": 0.5162523900573613,
+      "grad_norm": 0.896460235118866,
+      "learning_rate": 8.868341352595755e-05,
+      "loss": 0.1353,
+      "mean_token_accuracy": 0.9457281947135925,
+      "num_tokens": 727721.0,
+      "step": 540
+    },
+    {
+      "entropy": 0.9908315539360046,
+      "epoch": 0.5258126195028681,
+      "grad_norm": 0.4633411467075348,
+      "learning_rate": 8.817772265338409e-05,
+      "loss": 0.1234,
+      "mean_token_accuracy": 0.9556206941604615,
+      "num_tokens": 741707.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.9995946884155273,
+      "epoch": 0.5353728489483748,
+      "grad_norm": 0.6242344975471497,
+      "learning_rate": 8.766248834003441e-05,
+      "loss": 0.1329,
+      "mean_token_accuracy": 0.9488806366920471,
+      "num_tokens": 755385.0,
+      "step": 560
+    },
+    {
+      "entropy": 0.9838046133518219,
+      "epoch": 0.5449330783938815,
+      "grad_norm": 0.8008173704147339,
+      "learning_rate": 8.713783938112768e-05,
+      "loss": 0.1037,
+      "mean_token_accuracy": 0.9602915406227112,
+      "num_tokens": 768687.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.0002874314785004,
+      "epoch": 0.5544933078393881,
+      "grad_norm": 0.687570333480835,
+      "learning_rate": 8.660390692530026e-05,
+      "loss": 0.0997,
+      "mean_token_accuracy": 0.962467509508133,
+      "num_tokens": 782172.0,
+      "step": 580
+    },
+    {
+      "entropy": 0.9756083965301514,
+      "epoch": 0.5640535372848948,
+      "grad_norm": 0.4523005187511444,
+      "learning_rate": 8.606082444182216e-05,
+      "loss": 0.1061,
+      "mean_token_accuracy": 0.9599639117717743,
+      "num_tokens": 795561.0,
+      "step": 590
+    },
+    {
+      "entropy": 0.9945665299892426,
+      "epoch": 0.5736137667304015,
+      "grad_norm": 0.44194260239601135,
+      "learning_rate": 8.550872768723292e-05,
+      "loss": 0.0954,
+      "mean_token_accuracy": 0.9630863726139068,
+      "num_tokens": 809436.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.9733151197433472,
+      "epoch": 0.5831739961759083,
+      "grad_norm": 0.8076099753379822,
+      "learning_rate": 8.494775467140627e-05,
+      "loss": 0.1094,
+      "mean_token_accuracy": 0.9594907343387604,
+      "num_tokens": 822983.0,
+      "step": 610
+    },
+    {
+      "entropy": 0.9833361566066742,
+      "epoch": 0.5927342256214149,
+      "grad_norm": 0.5964319109916687,
+      "learning_rate": 8.4378045623051e-05,
+      "loss": 0.1113,
+      "mean_token_accuracy": 0.9565304279327392,
+      "num_tokens": 836766.0,
+      "step": 620
+    },
+    {
+      "entropy": 0.9940290510654449,
+      "epoch": 0.6022944550669216,
+      "grad_norm": 0.4419882595539093,
+      "learning_rate": 8.379974295465758e-05,
+      "loss": 0.1156,
+      "mean_token_accuracy": 0.9538529396057129,
+      "num_tokens": 850200.0,
+      "step": 630
+    },
+    {
+      "entropy": 0.9937334597110749,
+      "epoch": 0.6118546845124283,
+      "grad_norm": 0.3782116174697876,
+      "learning_rate": 8.321299122689863e-05,
+      "loss": 0.1304,
+      "mean_token_accuracy": 0.9487286865711212,
+      "num_tokens": 864001.0,
+      "step": 640
+    },
+    {
+      "entropy": 0.9992535769939422,
+      "epoch": 0.621414913957935,
+      "grad_norm": 0.6205118894577026,
+      "learning_rate": 8.261793711249254e-05,
+      "loss": 0.1132,
+      "mean_token_accuracy": 0.9553571522235871,
+      "num_tokens": 878135.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.9682816386222839,
+      "epoch": 0.6309751434034416,
+      "grad_norm": 0.6428599953651428,
+      "learning_rate": 8.201472935953916e-05,
+      "loss": 0.1269,
+      "mean_token_accuracy": 0.9514296293258667,
+      "num_tokens": 891307.0,
+      "step": 660
+    },
+    {
+      "entropy": 0.9947361946105957,
+      "epoch": 0.6405353728489483,
+      "grad_norm": 0.5262828469276428,
+      "learning_rate": 8.140351875433649e-05,
+      "loss": 0.107,
+      "mean_token_accuracy": 0.9581965684890748,
+      "num_tokens": 904829.0,
+      "step": 670
+    },
+    {
+      "entropy": 0.9687672019004822,
+      "epoch": 0.6500956022944551,
+      "grad_norm": 0.5138391852378845,
+      "learning_rate": 8.07844580836881e-05,
+      "loss": 0.0985,
+      "mean_token_accuracy": 0.9606104373931885,
+      "num_tokens": 919199.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9768559396266937,
+      "epoch": 0.6596558317399618,
+      "grad_norm": 0.4748534560203552,
+      "learning_rate": 8.015770209671044e-05,
+      "loss": 0.1221,
+      "mean_token_accuracy": 0.9548065602779389,
+      "num_tokens": 932439.0,
+      "step": 690
+    },
+    {
+      "entropy": 0.9911173582077026,
+      "epoch": 0.6692160611854685,
+      "grad_norm": 0.5670198798179626,
+      "learning_rate": 7.952340746614946e-05,
+      "loss": 0.1192,
+      "mean_token_accuracy": 0.9524793446063995,
+      "num_tokens": 945776.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.981645667552948,
+      "epoch": 0.6787762906309751,
+      "grad_norm": 0.5908262729644775,
+      "learning_rate": 7.888173274921655e-05,
+      "loss": 0.0971,
+      "mean_token_accuracy": 0.9633673906326294,
+      "num_tokens": 959352.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9900722146034241,
+      "epoch": 0.6883365200764818,
+      "grad_norm": 0.6447123289108276,
+      "learning_rate": 7.823283834795336e-05,
+      "loss": 0.1139,
+      "mean_token_accuracy": 0.9539555191993714,
+      "num_tokens": 972596.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9853983879089355,
+      "epoch": 0.6978967495219885,
+      "grad_norm": 0.7139818072319031,
+      "learning_rate": 7.75768864691354e-05,
+      "loss": 0.1155,
+      "mean_token_accuracy": 0.953942996263504,
+      "num_tokens": 985699.0,
+      "step": 730
+    },
+    {
+      "entropy": 0.9857402324676514,
+      "epoch": 0.7074569789674953,
+      "grad_norm": 0.6009901762008667,
+      "learning_rate": 7.691404108372461e-05,
+      "loss": 0.114,
+      "mean_token_accuracy": 0.951602292060852,
+      "num_tokens": 999728.0,
+      "step": 740
+    },
+    {
+      "entropy": 0.9839658558368682,
+      "epoch": 0.7170172084130019,
+      "grad_norm": 0.4921777546405792,
+      "learning_rate": 7.624446788588085e-05,
+      "loss": 0.094,
+      "mean_token_accuracy": 0.963894534111023,
+      "num_tokens": 1013669.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.980171799659729,
+      "epoch": 0.7265774378585086,
+      "grad_norm": 0.5071133971214294,
+      "learning_rate": 7.556833425154255e-05,
+      "loss": 0.1008,
+      "mean_token_accuracy": 0.9612992346286774,
+      "num_tokens": 1027609.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9888203203678131,
+      "epoch": 0.7361376673040153,
+      "grad_norm": 0.8505645394325256,
+      "learning_rate": 7.488580919658727e-05,
+      "loss": 0.1117,
+      "mean_token_accuracy": 0.9541349232196807,
+      "num_tokens": 1041046.0,
+      "step": 770
+    },
+    {
+      "entropy": 0.9662439882755279,
+      "epoch": 0.745697896749522,
+      "grad_norm": 0.7925776839256287,
+      "learning_rate": 7.419706333458201e-05,
+      "loss": 0.0944,
+      "mean_token_accuracy": 0.9619847953319549,
+      "num_tokens": 1055304.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9660874724388122,
+      "epoch": 0.7552581261950286,
+      "grad_norm": 0.8590351939201355,
+      "learning_rate": 7.350226883413413e-05,
+      "loss": 0.113,
+      "mean_token_accuracy": 0.9567786514759063,
+      "num_tokens": 1068542.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9770000517368317,
+      "epoch": 0.7648183556405354,
+      "grad_norm": 0.803503692150116,
+      "learning_rate": 7.280159937585386e-05,
+      "loss": 0.1036,
+      "mean_token_accuracy": 0.9592704355716706,
+      "num_tokens": 1082195.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.9872698724269867,
+      "epoch": 0.7743785850860421,
+      "grad_norm": 0.8108046650886536,
+      "learning_rate": 7.209523010893853e-05,
+      "loss": 0.0984,
+      "mean_token_accuracy": 0.9622078835964203,
+      "num_tokens": 1095884.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.982585608959198,
+      "epoch": 0.7839388145315488,
+      "grad_norm": 0.5226531624794006,
+      "learning_rate": 7.138333760738987e-05,
+      "loss": 0.1087,
+      "mean_token_accuracy": 0.9583467662334442,
+      "num_tokens": 1109287.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9687687039375306,
+      "epoch": 0.7934990439770554,
+      "grad_norm": 0.9293949007987976,
+      "learning_rate": 7.066609982587501e-05,
+      "loss": 0.096,
+      "mean_token_accuracy": 0.9625981748104095,
+      "num_tokens": 1123392.0,
+      "step": 830
+    },
+    {
+      "entropy": 0.9936236917972565,
+      "epoch": 0.8030592734225621,
+      "grad_norm": 0.5094279646873474,
+      "learning_rate": 6.994369605524253e-05,
+      "loss": 0.1082,
+      "mean_token_accuracy": 0.9591395795345307,
+      "num_tokens": 1137691.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9640061318874359,
+      "epoch": 0.8126195028680688,
+      "grad_norm": 0.7790572047233582,
+      "learning_rate": 6.921630687770427e-05,
+      "loss": 0.108,
+      "mean_token_accuracy": 0.9564647853374482,
+      "num_tokens": 1151400.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.9576828956604004,
+      "epoch": 0.8221797323135756,
+      "grad_norm": 0.759594202041626,
+      "learning_rate": 6.848411412169461e-05,
+      "loss": 0.0989,
+      "mean_token_accuracy": 0.9621531963348389,
+      "num_tokens": 1165074.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9507431745529175,
+      "epoch": 0.8317399617590823,
+      "grad_norm": 0.8087530136108398,
+      "learning_rate": 6.774730081641799e-05,
+      "loss": 0.1026,
+      "mean_token_accuracy": 0.9592073798179627,
+      "num_tokens": 1178919.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.959827721118927,
+      "epoch": 0.8413001912045889,
+      "grad_norm": 0.984460711479187,
+      "learning_rate": 6.70060511460963e-05,
+      "loss": 0.1114,
+      "mean_token_accuracy": 0.9576607882976532,
+      "num_tokens": 1192204.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9430737257003784,
+      "epoch": 0.8508604206500956,
+      "grad_norm": 0.4848315119743347,
+      "learning_rate": 6.626055040392774e-05,
+      "loss": 0.0975,
+      "mean_token_accuracy": 0.9601469039916992,
+      "num_tokens": 1205571.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9415578961372375,
+      "epoch": 0.8604206500956023,
+      "grad_norm": 0.9813526272773743,
+      "learning_rate": 6.551098494576819e-05,
+      "loss": 0.1236,
+      "mean_token_accuracy": 0.9510224223136902,
+      "num_tokens": 1219043.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9605513393878937,
+      "epoch": 0.869980879541109,
+      "grad_norm": 0.6163782477378845,
+      "learning_rate": 6.475754214354715e-05,
+      "loss": 0.1145,
+      "mean_token_accuracy": 0.9566048800945282,
+      "num_tokens": 1233113.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.944946551322937,
+      "epoch": 0.8795411089866156,
+      "grad_norm": 0.5911535620689392,
+      "learning_rate": 6.400041033842965e-05,
+      "loss": 0.0991,
+      "mean_token_accuracy": 0.9608201622962952,
+      "num_tokens": 1245931.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9361202180385589,
+      "epoch": 0.8891013384321224,
+      "grad_norm": 0.6969265341758728,
+      "learning_rate": 6.323977879373571e-05,
+      "loss": 0.107,
+      "mean_token_accuracy": 0.9554063081741333,
+      "num_tokens": 1259949.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.9622337400913239,
+      "epoch": 0.8986615678776291,
+      "grad_norm": 0.8290131092071533,
+      "learning_rate": 6.247583764762959e-05,
+      "loss": 0.098,
+      "mean_token_accuracy": 0.9622323632240295,
+      "num_tokens": 1273413.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9522945702075958,
+      "epoch": 0.9082217973231358,
+      "grad_norm": 0.6268866658210754,
+      "learning_rate": 6.170877786559001e-05,
+      "loss": 0.1011,
+      "mean_token_accuracy": 0.9619914889335632,
+      "num_tokens": 1286343.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.94032102227211,
+      "epoch": 0.9177820267686424,
+      "grad_norm": 0.7365413904190063,
+      "learning_rate": 6.0938791192673895e-05,
+      "loss": 0.1145,
+      "mean_token_accuracy": 0.9565579771995545,
+      "num_tokens": 1300089.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9440156936645507,
+      "epoch": 0.9273422562141491,
+      "grad_norm": 0.9897956848144531,
+      "learning_rate": 6.016607010558492e-05,
+      "loss": 0.1037,
+      "mean_token_accuracy": 0.9587496936321258,
+      "num_tokens": 1313412.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9517957687377929,
+      "epoch": 0.9369024856596558,
+      "grad_norm": 0.8890661001205444,
+      "learning_rate": 5.939080776455942e-05,
+      "loss": 0.1094,
+      "mean_token_accuracy": 0.9572539150714874,
+      "num_tokens": 1326290.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.9646746933460235,
+      "epoch": 0.9464627151051626,
+      "grad_norm": 0.9077587723731995,
+      "learning_rate": 5.861319796508139e-05,
+      "loss": 0.1014,
+      "mean_token_accuracy": 0.9589385271072388,
+      "num_tokens": 1339719.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.9584924161434174,
+      "epoch": 0.9560229445506692,
+      "grad_norm": 0.6008082628250122,
+      "learning_rate": 5.783343508943849e-05,
+      "loss": 0.0961,
+      "mean_token_accuracy": 0.9606513440608978,
+      "num_tokens": 1352423.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9465383410453796,
+      "epoch": 0.9655831739961759,
+      "grad_norm": 0.767816424369812,
+      "learning_rate": 5.7051714058131654e-05,
+      "loss": 0.0992,
+      "mean_token_accuracy": 0.9612827241420746,
+      "num_tokens": 1365889.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.935906982421875,
+      "epoch": 0.9751434034416826,
+      "grad_norm": 0.6595594882965088,
+      "learning_rate": 5.626823028114991e-05,
+      "loss": 0.0993,
+      "mean_token_accuracy": 0.9595326006412506,
+      "num_tokens": 1379353.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9248349666595459,
+      "epoch": 0.9847036328871893,
+      "grad_norm": 0.5140650272369385,
+      "learning_rate": 5.548317960912305e-05,
+      "loss": 0.1057,
+      "mean_token_accuracy": 0.9561357080936432,
+      "num_tokens": 1392866.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9187301456928253,
+      "epoch": 0.994263862332696,
+      "grad_norm": 0.5743018984794617,
+      "learning_rate": 5.4696758284363956e-05,
+      "loss": 0.111,
+      "mean_token_accuracy": 0.9592386484146118,
+      "num_tokens": 1406263.0,
+      "step": 1040
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2092,
+  "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": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.100958587515187e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4ccf1fcd9bda6fe57b0b2816b7829302c446a0c8a1c92d1cb175883ca1148d1
+size 6417

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "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": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "down_proj",
+    "gate_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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:502bb00cccebfe31b7ef23d9a4feaf59075e397b10ce74aa3d2d7b84499be173
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/tokenizer.json

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/trainer_state.json

@@ -0,0 +1,2124 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 2092,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.173049008846283,
+      "epoch": 0.009560229445506692,
+      "grad_norm": 0.47283369302749634,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2778,
+      "mean_token_accuracy": 0.9132264971733093,
+      "num_tokens": 14125.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1915716052055358,
+      "epoch": 0.019120458891013385,
+      "grad_norm": 0.5709051489830017,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2621,
+      "mean_token_accuracy": 0.9111678123474121,
+      "num_tokens": 27004.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1907334327697754,
+      "epoch": 0.028680688336520075,
+      "grad_norm": 0.5265517234802246,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.223,
+      "mean_token_accuracy": 0.9214041888713836,
+      "num_tokens": 41395.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.1634743332862854,
+      "epoch": 0.03824091778202677,
+      "grad_norm": 0.5487110018730164,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1914,
+      "mean_token_accuracy": 0.9342293441295624,
+      "num_tokens": 55233.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.1422934770584106,
+      "epoch": 0.04780114722753346,
+      "grad_norm": 0.832442045211792,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.1967,
+      "mean_token_accuracy": 0.9349480092525482,
+      "num_tokens": 68341.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.118509554862976,
+      "epoch": 0.05736137667304015,
+      "grad_norm": 0.6014860272407532,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1727,
+      "mean_token_accuracy": 0.9380205571651459,
+      "num_tokens": 81730.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.0945161700248718,
+      "epoch": 0.06692160611854685,
+      "grad_norm": 0.6720062494277954,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1634,
+      "mean_token_accuracy": 0.9437400996685028,
+      "num_tokens": 94886.0,
+      "step": 70
+    },
+    {
+      "entropy": 1.0766278862953187,
+      "epoch": 0.07648183556405354,
+      "grad_norm": 0.7721521854400635,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1728,
+      "mean_token_accuracy": 0.9379306972026825,
+      "num_tokens": 108569.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.0766998410224915,
+      "epoch": 0.08604206500956023,
+      "grad_norm": 0.6029406785964966,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1579,
+      "mean_token_accuracy": 0.9441168010234833,
+      "num_tokens": 122156.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0624309301376342,
+      "epoch": 0.09560229445506692,
+      "grad_norm": 0.6802502870559692,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1561,
+      "mean_token_accuracy": 0.9400320529937745,
+      "num_tokens": 135290.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0544052481651307,
+      "epoch": 0.10516252390057361,
+      "grad_norm": 0.5983047485351562,
+      "learning_rate": 9.999900008621665e-05,
+      "loss": 0.1753,
+      "mean_token_accuracy": 0.9354041159152985,
+      "num_tokens": 149015.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.0516255259513856,
+      "epoch": 0.1147227533460803,
+      "grad_norm": 0.5674954056739807,
+      "learning_rate": 9.998775151544478e-05,
+      "loss": 0.1529,
+      "mean_token_accuracy": 0.9421895742416382,
+      "num_tokens": 162147.0,
+      "step": 120
+    },
+    {
+      "entropy": 1.0305089116096497,
+      "epoch": 0.124282982791587,
+      "grad_norm": 0.5198636651039124,
+      "learning_rate": 9.996400730289628e-05,
+      "loss": 0.1538,
+      "mean_token_accuracy": 0.9427219152450561,
+      "num_tokens": 175436.0,
+      "step": 130
+    },
+    {
+      "entropy": 1.0165980935096741,
+      "epoch": 0.1338432122370937,
+      "grad_norm": 0.7355736494064331,
+      "learning_rate": 9.992777338400864e-05,
+      "loss": 0.149,
+      "mean_token_accuracy": 0.9472985088825225,
+      "num_tokens": 188834.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0045411348342896,
+      "epoch": 0.14340344168260039,
+      "grad_norm": 0.5472018122673035,
+      "learning_rate": 9.987905881632213e-05,
+      "loss": 0.1498,
+      "mean_token_accuracy": 0.9436784207820892,
+      "num_tokens": 202708.0,
+      "step": 150
+    },
+    {
+      "entropy": 1.0141049027442932,
+      "epoch": 0.15296367112810708,
+      "grad_norm": 0.6937310099601746,
+      "learning_rate": 9.981787577721553e-05,
+      "loss": 0.1497,
+      "mean_token_accuracy": 0.9430840134620666,
+      "num_tokens": 216384.0,
+      "step": 160
+    },
+    {
+      "entropy": 1.0251058161258697,
+      "epoch": 0.16252390057361377,
+      "grad_norm": 0.5407682061195374,
+      "learning_rate": 9.974423956086228e-05,
+      "loss": 0.1445,
+      "mean_token_accuracy": 0.9475803017616272,
+      "num_tokens": 229105.0,
+      "step": 170
+    },
+    {
+      "entropy": 1.0213705360889436,
+      "epoch": 0.17208413001912046,
+      "grad_norm": 0.5844500660896301,
+      "learning_rate": 9.965816857440717e-05,
+      "loss": 0.131,
+      "mean_token_accuracy": 0.9481783032417297,
+      "num_tokens": 242264.0,
+      "step": 180
+    },
+    {
+      "entropy": 1.0302305698394776,
+      "epoch": 0.18164435946462715,
+      "grad_norm": 0.5545923113822937,
+      "learning_rate": 9.955968433336515e-05,
+      "loss": 0.1555,
+      "mean_token_accuracy": 0.9432050168514252,
+      "num_tokens": 255602.0,
+      "step": 190
+    },
+    {
+      "entropy": 1.0268681943416595,
+      "epoch": 0.19120458891013384,
+      "grad_norm": 0.5613768100738525,
+      "learning_rate": 9.944881145624291e-05,
+      "loss": 0.1157,
+      "mean_token_accuracy": 0.9568168997764588,
+      "num_tokens": 269712.0,
+      "step": 200
+    },
+    {
+      "entropy": 1.0397694110870361,
+      "epoch": 0.20076481835564053,
+      "grad_norm": 0.5778163075447083,
+      "learning_rate": 9.932557765838504e-05,
+      "loss": 0.1361,
+      "mean_token_accuracy": 0.9503950834274292,
+      "num_tokens": 283572.0,
+      "step": 210
+    },
+    {
+      "entropy": 1.015084683895111,
+      "epoch": 0.21032504780114722,
+      "grad_norm": 0.5037960410118103,
+      "learning_rate": 9.919001374504569e-05,
+      "loss": 0.1383,
+      "mean_token_accuracy": 0.9510905325412751,
+      "num_tokens": 297437.0,
+      "step": 220
+    },
+    {
+      "entropy": 1.008081203699112,
+      "epoch": 0.2198852772466539,
+      "grad_norm": 0.8288016319274902,
+      "learning_rate": 9.904215360368826e-05,
+      "loss": 0.1355,
+      "mean_token_accuracy": 0.9502547740936279,
+      "num_tokens": 310803.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.0171762943267821,
+      "epoch": 0.2294455066921606,
+      "grad_norm": 0.7228999733924866,
+      "learning_rate": 9.888203419551432e-05,
+      "loss": 0.1355,
+      "mean_token_accuracy": 0.9449310421943664,
+      "num_tokens": 324713.0,
+      "step": 240
+    },
+    {
+      "entropy": 0.9794922828674316,
+      "epoch": 0.2390057361376673,
+      "grad_norm": 1.5495465993881226,
+      "learning_rate": 9.870969554622426e-05,
+      "loss": 0.1349,
+      "mean_token_accuracy": 0.9471733391284942,
+      "num_tokens": 337634.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.9894898355007171,
+      "epoch": 0.248565965583174,
+      "grad_norm": 0.6044939160346985,
+      "learning_rate": 9.852518073601194e-05,
+      "loss": 0.1356,
+      "mean_token_accuracy": 0.9499468445777893,
+      "num_tokens": 350459.0,
+      "step": 260
+    },
+    {
+      "entropy": 0.9731416225433349,
+      "epoch": 0.25812619502868067,
+      "grad_norm": 1.028459906578064,
+      "learning_rate": 9.832853588879575e-05,
+      "loss": 0.1242,
+      "mean_token_accuracy": 0.9518984615802765,
+      "num_tokens": 364010.0,
+      "step": 270
+    },
+    {
+      "entropy": 0.9964198529720306,
+      "epoch": 0.2676864244741874,
+      "grad_norm": 0.6048480868339539,
+      "learning_rate": 9.811981016068881e-05,
+      "loss": 0.1384,
+      "mean_token_accuracy": 0.9438554108142853,
+      "num_tokens": 377021.0,
+      "step": 280
+    },
+    {
+      "entropy": 0.9903759062290192,
+      "epoch": 0.27724665391969405,
+      "grad_norm": 0.4507066011428833,
+      "learning_rate": 9.78990557277112e-05,
+      "loss": 0.1352,
+      "mean_token_accuracy": 0.9466394543647766,
+      "num_tokens": 390819.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0028544664382935,
+      "epoch": 0.28680688336520077,
+      "grad_norm": 0.7183990478515625,
+      "learning_rate": 9.766632777274738e-05,
+      "loss": 0.122,
+      "mean_token_accuracy": 0.9545832991600036,
+      "num_tokens": 404644.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0233761847019196,
+      "epoch": 0.29636711281070743,
+      "grad_norm": 0.6612422466278076,
+      "learning_rate": 9.742168447175182e-05,
+      "loss": 0.1344,
+      "mean_token_accuracy": 0.9533212602138519,
+      "num_tokens": 418529.0,
+      "step": 310
+    },
+    {
+      "entropy": 0.9999796330928803,
+      "epoch": 0.30592734225621415,
+      "grad_norm": 0.5009316802024841,
+      "learning_rate": 9.716518697920655e-05,
+      "loss": 0.1137,
+      "mean_token_accuracy": 0.9556109249591828,
+      "num_tokens": 431509.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0225579202175141,
+      "epoch": 0.3154875717017208,
+      "grad_norm": 0.5004062652587891,
+      "learning_rate": 9.689689941283417e-05,
+      "loss": 0.1256,
+      "mean_token_accuracy": 0.9532630741596222,
+      "num_tokens": 445040.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.0261304914951324,
+      "epoch": 0.32504780114722753,
+      "grad_norm": 0.8441359400749207,
+      "learning_rate": 9.661688883756996e-05,
+      "loss": 0.1337,
+      "mean_token_accuracy": 0.9432336091995239,
+      "num_tokens": 458540.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0289096713066102,
+      "epoch": 0.33460803059273425,
+      "grad_norm": 0.6566339135169983,
+      "learning_rate": 9.632522524879756e-05,
+      "loss": 0.1321,
+      "mean_token_accuracy": 0.9464678704738617,
+      "num_tokens": 471556.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.04276704788208,
+      "epoch": 0.3441682600382409,
+      "grad_norm": 0.8276234865188599,
+      "learning_rate": 9.602198155485176e-05,
+      "loss": 0.1229,
+      "mean_token_accuracy": 0.9535828530788422,
+      "num_tokens": 484786.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.0487712740898132,
+      "epoch": 0.35372848948374763,
+      "grad_norm": 0.4944207966327667,
+      "learning_rate": 9.570723355879342e-05,
+      "loss": 0.1162,
+      "mean_token_accuracy": 0.9553705751895905,
+      "num_tokens": 498352.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.0256853461265565,
+      "epoch": 0.3632887189292543,
+      "grad_norm": 0.7083448171615601,
+      "learning_rate": 9.538105993946068e-05,
+      "loss": 0.118,
+      "mean_token_accuracy": 0.9546810567378998,
+      "num_tokens": 512064.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0263440370559693,
+      "epoch": 0.372848948374761,
+      "grad_norm": 0.6025220155715942,
+      "learning_rate": 9.504354223180118e-05,
+      "loss": 0.1111,
+      "mean_token_accuracy": 0.9575964570045471,
+      "num_tokens": 525582.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.0304751515388488,
+      "epoch": 0.3824091778202677,
+      "grad_norm": 0.8519231081008911,
+      "learning_rate": 9.469476480649056e-05,
+      "loss": 0.1223,
+      "mean_token_accuracy": 0.9522103905677796,
+      "num_tokens": 539561.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.0210801661014557,
+      "epoch": 0.3919694072657744,
+      "grad_norm": 0.47544005513191223,
+      "learning_rate": 9.433481484884189e-05,
+      "loss": 0.1197,
+      "mean_token_accuracy": 0.9545511543750763,
+      "num_tokens": 553625.0,
+      "step": 410
+    },
+    {
+      "entropy": 0.9995654165744782,
+      "epoch": 0.40152963671128106,
+      "grad_norm": 0.6336336731910706,
+      "learning_rate": 9.396378233701167e-05,
+      "loss": 0.107,
+      "mean_token_accuracy": 0.9580662846565247,
+      "num_tokens": 565891.0,
+      "step": 420
+    },
+    {
+      "entropy": 0.9967389881610871,
+      "epoch": 0.4110898661567878,
+      "grad_norm": 0.6172469854354858,
+      "learning_rate": 9.358176001950759e-05,
+      "loss": 0.1137,
+      "mean_token_accuracy": 0.9567353665828705,
+      "num_tokens": 578722.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.034762543439865,
+      "epoch": 0.42065009560229444,
+      "grad_norm": 0.7707498669624329,
+      "learning_rate": 9.318884339200378e-05,
+      "loss": 0.1361,
+      "mean_token_accuracy": 0.9430133700370789,
+      "num_tokens": 592646.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0067415297031403,
+      "epoch": 0.43021032504780116,
+      "grad_norm": 0.6977169513702393,
+      "learning_rate": 9.278513067346941e-05,
+      "loss": 0.1113,
+      "mean_token_accuracy": 0.9558754086494445,
+      "num_tokens": 606429.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0067644000053406,
+      "epoch": 0.4397705544933078,
+      "grad_norm": 0.8753907084465027,
+      "learning_rate": 9.237072278161652e-05,
+      "loss": 0.1269,
+      "mean_token_accuracy": 0.9501053631305695,
+      "num_tokens": 619626.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0031738936901093,
+      "epoch": 0.44933078393881454,
+      "grad_norm": 0.6908318996429443,
+      "learning_rate": 9.194572330767314e-05,
+      "loss": 0.1198,
+      "mean_token_accuracy": 0.9550222277641296,
+      "num_tokens": 633055.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.0184494972229003,
+      "epoch": 0.4588910133843212,
+      "grad_norm": 0.5444503426551819,
+      "learning_rate": 9.151023849048818e-05,
+      "loss": 0.1122,
+      "mean_token_accuracy": 0.9565081417560577,
+      "num_tokens": 647155.0,
+      "step": 480
+    },
+    {
+      "entropy": 0.9871227145195007,
+      "epoch": 0.4684512428298279,
+      "grad_norm": 0.5650642514228821,
+      "learning_rate": 9.106437718997452e-05,
+      "loss": 0.1154,
+      "mean_token_accuracy": 0.9546488761901856,
+      "num_tokens": 660695.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0054654061794281,
+      "epoch": 0.4780114722753346,
+      "grad_norm": 0.912119448184967,
+      "learning_rate": 9.060825085989682e-05,
+      "loss": 0.1146,
+      "mean_token_accuracy": 0.9498761177062989,
+      "num_tokens": 674509.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.0120668947696685,
+      "epoch": 0.4875717017208413,
+      "grad_norm": 0.7253690958023071,
+      "learning_rate": 9.014197352001097e-05,
+      "loss": 0.1034,
+      "mean_token_accuracy": 0.9615722954273224,
+      "num_tokens": 687644.0,
+      "step": 510
+    },
+    {
+      "entropy": 0.9921061754226684,
+      "epoch": 0.497131931166348,
+      "grad_norm": 0.8324573636054993,
+      "learning_rate": 8.966566172756213e-05,
+      "loss": 0.1193,
+      "mean_token_accuracy": 0.9571842133998871,
+      "num_tokens": 700658.0,
+      "step": 520
+    },
+    {
+      "entropy": 1.015238630771637,
+      "epoch": 0.5066921606118547,
+      "grad_norm": 0.7130652070045471,
+      "learning_rate": 8.91794345481484e-05,
+      "loss": 0.1163,
+      "mean_token_accuracy": 0.9569105088710785,
+      "num_tokens": 714234.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0172134935855865,
+      "epoch": 0.5162523900573613,
+      "grad_norm": 0.896460235118866,
+      "learning_rate": 8.868341352595755e-05,
+      "loss": 0.1353,
+      "mean_token_accuracy": 0.9457281947135925,
+      "num_tokens": 727721.0,
+      "step": 540
+    },
+    {
+      "entropy": 0.9908315539360046,
+      "epoch": 0.5258126195028681,
+      "grad_norm": 0.4633411467075348,
+      "learning_rate": 8.817772265338409e-05,
+      "loss": 0.1234,
+      "mean_token_accuracy": 0.9556206941604615,
+      "num_tokens": 741707.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.9995946884155273,
+      "epoch": 0.5353728489483748,
+      "grad_norm": 0.6242344975471497,
+      "learning_rate": 8.766248834003441e-05,
+      "loss": 0.1329,
+      "mean_token_accuracy": 0.9488806366920471,
+      "num_tokens": 755385.0,
+      "step": 560
+    },
+    {
+      "entropy": 0.9838046133518219,
+      "epoch": 0.5449330783938815,
+      "grad_norm": 0.8008173704147339,
+      "learning_rate": 8.713783938112768e-05,
+      "loss": 0.1037,
+      "mean_token_accuracy": 0.9602915406227112,
+      "num_tokens": 768687.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.0002874314785004,
+      "epoch": 0.5544933078393881,
+      "grad_norm": 0.687570333480835,
+      "learning_rate": 8.660390692530026e-05,
+      "loss": 0.0997,
+      "mean_token_accuracy": 0.962467509508133,
+      "num_tokens": 782172.0,
+      "step": 580
+    },
+    {
+      "entropy": 0.9756083965301514,
+      "epoch": 0.5640535372848948,
+      "grad_norm": 0.4523005187511444,
+      "learning_rate": 8.606082444182216e-05,
+      "loss": 0.1061,
+      "mean_token_accuracy": 0.9599639117717743,
+      "num_tokens": 795561.0,
+      "step": 590
+    },
+    {
+      "entropy": 0.9945665299892426,
+      "epoch": 0.5736137667304015,
+      "grad_norm": 0.44194260239601135,
+      "learning_rate": 8.550872768723292e-05,
+      "loss": 0.0954,
+      "mean_token_accuracy": 0.9630863726139068,
+      "num_tokens": 809436.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.9733151197433472,
+      "epoch": 0.5831739961759083,
+      "grad_norm": 0.8076099753379822,
+      "learning_rate": 8.494775467140627e-05,
+      "loss": 0.1094,
+      "mean_token_accuracy": 0.9594907343387604,
+      "num_tokens": 822983.0,
+      "step": 610
+    },
+    {
+      "entropy": 0.9833361566066742,
+      "epoch": 0.5927342256214149,
+      "grad_norm": 0.5964319109916687,
+      "learning_rate": 8.4378045623051e-05,
+      "loss": 0.1113,
+      "mean_token_accuracy": 0.9565304279327392,
+      "num_tokens": 836766.0,
+      "step": 620
+    },
+    {
+      "entropy": 0.9940290510654449,
+      "epoch": 0.6022944550669216,
+      "grad_norm": 0.4419882595539093,
+      "learning_rate": 8.379974295465758e-05,
+      "loss": 0.1156,
+      "mean_token_accuracy": 0.9538529396057129,
+      "num_tokens": 850200.0,
+      "step": 630
+    },
+    {
+      "entropy": 0.9937334597110749,
+      "epoch": 0.6118546845124283,
+      "grad_norm": 0.3782116174697876,
+      "learning_rate": 8.321299122689863e-05,
+      "loss": 0.1304,
+      "mean_token_accuracy": 0.9487286865711212,
+      "num_tokens": 864001.0,
+      "step": 640
+    },
+    {
+      "entropy": 0.9992535769939422,
+      "epoch": 0.621414913957935,
+      "grad_norm": 0.6205118894577026,
+      "learning_rate": 8.261793711249254e-05,
+      "loss": 0.1132,
+      "mean_token_accuracy": 0.9553571522235871,
+      "num_tokens": 878135.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.9682816386222839,
+      "epoch": 0.6309751434034416,
+      "grad_norm": 0.6428599953651428,
+      "learning_rate": 8.201472935953916e-05,
+      "loss": 0.1269,
+      "mean_token_accuracy": 0.9514296293258667,
+      "num_tokens": 891307.0,
+      "step": 660
+    },
+    {
+      "entropy": 0.9947361946105957,
+      "epoch": 0.6405353728489483,
+      "grad_norm": 0.5262828469276428,
+      "learning_rate": 8.140351875433649e-05,
+      "loss": 0.107,
+      "mean_token_accuracy": 0.9581965684890748,
+      "num_tokens": 904829.0,
+      "step": 670
+    },
+    {
+      "entropy": 0.9687672019004822,
+      "epoch": 0.6500956022944551,
+      "grad_norm": 0.5138391852378845,
+      "learning_rate": 8.07844580836881e-05,
+      "loss": 0.0985,
+      "mean_token_accuracy": 0.9606104373931885,
+      "num_tokens": 919199.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9768559396266937,
+      "epoch": 0.6596558317399618,
+      "grad_norm": 0.4748534560203552,
+      "learning_rate": 8.015770209671044e-05,
+      "loss": 0.1221,
+      "mean_token_accuracy": 0.9548065602779389,
+      "num_tokens": 932439.0,
+      "step": 690
+    },
+    {
+      "entropy": 0.9911173582077026,
+      "epoch": 0.6692160611854685,
+      "grad_norm": 0.5670198798179626,
+      "learning_rate": 7.952340746614946e-05,
+      "loss": 0.1192,
+      "mean_token_accuracy": 0.9524793446063995,
+      "num_tokens": 945776.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.981645667552948,
+      "epoch": 0.6787762906309751,
+      "grad_norm": 0.5908262729644775,
+      "learning_rate": 7.888173274921655e-05,
+      "loss": 0.0971,
+      "mean_token_accuracy": 0.9633673906326294,
+      "num_tokens": 959352.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9900722146034241,
+      "epoch": 0.6883365200764818,
+      "grad_norm": 0.6447123289108276,
+      "learning_rate": 7.823283834795336e-05,
+      "loss": 0.1139,
+      "mean_token_accuracy": 0.9539555191993714,
+      "num_tokens": 972596.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9853983879089355,
+      "epoch": 0.6978967495219885,
+      "grad_norm": 0.7139818072319031,
+      "learning_rate": 7.75768864691354e-05,
+      "loss": 0.1155,
+      "mean_token_accuracy": 0.953942996263504,
+      "num_tokens": 985699.0,
+      "step": 730
+    },
+    {
+      "entropy": 0.9857402324676514,
+      "epoch": 0.7074569789674953,
+      "grad_norm": 0.6009901762008667,
+      "learning_rate": 7.691404108372461e-05,
+      "loss": 0.114,
+      "mean_token_accuracy": 0.951602292060852,
+      "num_tokens": 999728.0,
+      "step": 740
+    },
+    {
+      "entropy": 0.9839658558368682,
+      "epoch": 0.7170172084130019,
+      "grad_norm": 0.4921777546405792,
+      "learning_rate": 7.624446788588085e-05,
+      "loss": 0.094,
+      "mean_token_accuracy": 0.963894534111023,
+      "num_tokens": 1013669.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.980171799659729,
+      "epoch": 0.7265774378585086,
+      "grad_norm": 0.5071133971214294,
+      "learning_rate": 7.556833425154255e-05,
+      "loss": 0.1008,
+      "mean_token_accuracy": 0.9612992346286774,
+      "num_tokens": 1027609.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9888203203678131,
+      "epoch": 0.7361376673040153,
+      "grad_norm": 0.8505645394325256,
+      "learning_rate": 7.488580919658727e-05,
+      "loss": 0.1117,
+      "mean_token_accuracy": 0.9541349232196807,
+      "num_tokens": 1041046.0,
+      "step": 770
+    },
+    {
+      "entropy": 0.9662439882755279,
+      "epoch": 0.745697896749522,
+      "grad_norm": 0.7925776839256287,
+      "learning_rate": 7.419706333458201e-05,
+      "loss": 0.0944,
+      "mean_token_accuracy": 0.9619847953319549,
+      "num_tokens": 1055304.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9660874724388122,
+      "epoch": 0.7552581261950286,
+      "grad_norm": 0.8590351939201355,
+      "learning_rate": 7.350226883413413e-05,
+      "loss": 0.113,
+      "mean_token_accuracy": 0.9567786514759063,
+      "num_tokens": 1068542.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9770000517368317,
+      "epoch": 0.7648183556405354,
+      "grad_norm": 0.803503692150116,
+      "learning_rate": 7.280159937585386e-05,
+      "loss": 0.1036,
+      "mean_token_accuracy": 0.9592704355716706,
+      "num_tokens": 1082195.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.9872698724269867,
+      "epoch": 0.7743785850860421,
+      "grad_norm": 0.8108046650886536,
+      "learning_rate": 7.209523010893853e-05,
+      "loss": 0.0984,
+      "mean_token_accuracy": 0.9622078835964203,
+      "num_tokens": 1095884.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.982585608959198,
+      "epoch": 0.7839388145315488,
+      "grad_norm": 0.5226531624794006,
+      "learning_rate": 7.138333760738987e-05,
+      "loss": 0.1087,
+      "mean_token_accuracy": 0.9583467662334442,
+      "num_tokens": 1109287.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9687687039375306,
+      "epoch": 0.7934990439770554,
+      "grad_norm": 0.9293949007987976,
+      "learning_rate": 7.066609982587501e-05,
+      "loss": 0.096,
+      "mean_token_accuracy": 0.9625981748104095,
+      "num_tokens": 1123392.0,
+      "step": 830
+    },
+    {
+      "entropy": 0.9936236917972565,
+      "epoch": 0.8030592734225621,
+      "grad_norm": 0.5094279646873474,
+      "learning_rate": 6.994369605524253e-05,
+      "loss": 0.1082,
+      "mean_token_accuracy": 0.9591395795345307,
+      "num_tokens": 1137691.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9640061318874359,
+      "epoch": 0.8126195028680688,
+      "grad_norm": 0.7790572047233582,
+      "learning_rate": 6.921630687770427e-05,
+      "loss": 0.108,
+      "mean_token_accuracy": 0.9564647853374482,
+      "num_tokens": 1151400.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.9576828956604004,
+      "epoch": 0.8221797323135756,
+      "grad_norm": 0.759594202041626,
+      "learning_rate": 6.848411412169461e-05,
+      "loss": 0.0989,
+      "mean_token_accuracy": 0.9621531963348389,
+      "num_tokens": 1165074.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9507431745529175,
+      "epoch": 0.8317399617590823,
+      "grad_norm": 0.8087530136108398,
+      "learning_rate": 6.774730081641799e-05,
+      "loss": 0.1026,
+      "mean_token_accuracy": 0.9592073798179627,
+      "num_tokens": 1178919.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.959827721118927,
+      "epoch": 0.8413001912045889,
+      "grad_norm": 0.984460711479187,
+      "learning_rate": 6.70060511460963e-05,
+      "loss": 0.1114,
+      "mean_token_accuracy": 0.9576607882976532,
+      "num_tokens": 1192204.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9430737257003784,
+      "epoch": 0.8508604206500956,
+      "grad_norm": 0.4848315119743347,
+      "learning_rate": 6.626055040392774e-05,
+      "loss": 0.0975,
+      "mean_token_accuracy": 0.9601469039916992,
+      "num_tokens": 1205571.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9415578961372375,
+      "epoch": 0.8604206500956023,
+      "grad_norm": 0.9813526272773743,
+      "learning_rate": 6.551098494576819e-05,
+      "loss": 0.1236,
+      "mean_token_accuracy": 0.9510224223136902,
+      "num_tokens": 1219043.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9605513393878937,
+      "epoch": 0.869980879541109,
+      "grad_norm": 0.6163782477378845,
+      "learning_rate": 6.475754214354715e-05,
+      "loss": 0.1145,
+      "mean_token_accuracy": 0.9566048800945282,
+      "num_tokens": 1233113.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.944946551322937,
+      "epoch": 0.8795411089866156,
+      "grad_norm": 0.5911535620689392,
+      "learning_rate": 6.400041033842965e-05,
+      "loss": 0.0991,
+      "mean_token_accuracy": 0.9608201622962952,
+      "num_tokens": 1245931.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9361202180385589,
+      "epoch": 0.8891013384321224,
+      "grad_norm": 0.6969265341758728,
+      "learning_rate": 6.323977879373571e-05,
+      "loss": 0.107,
+      "mean_token_accuracy": 0.9554063081741333,
+      "num_tokens": 1259949.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.9622337400913239,
+      "epoch": 0.8986615678776291,
+      "grad_norm": 0.8290131092071533,
+      "learning_rate": 6.247583764762959e-05,
+      "loss": 0.098,
+      "mean_token_accuracy": 0.9622323632240295,
+      "num_tokens": 1273413.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9522945702075958,
+      "epoch": 0.9082217973231358,
+      "grad_norm": 0.6268866658210754,
+      "learning_rate": 6.170877786559001e-05,
+      "loss": 0.1011,
+      "mean_token_accuracy": 0.9619914889335632,
+      "num_tokens": 1286343.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.94032102227211,
+      "epoch": 0.9177820267686424,
+      "grad_norm": 0.7365413904190063,
+      "learning_rate": 6.0938791192673895e-05,
+      "loss": 0.1145,
+      "mean_token_accuracy": 0.9565579771995545,
+      "num_tokens": 1300089.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9440156936645507,
+      "epoch": 0.9273422562141491,
+      "grad_norm": 0.9897956848144531,
+      "learning_rate": 6.016607010558492e-05,
+      "loss": 0.1037,
+      "mean_token_accuracy": 0.9587496936321258,
+      "num_tokens": 1313412.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9517957687377929,
+      "epoch": 0.9369024856596558,
+      "grad_norm": 0.8890661001205444,
+      "learning_rate": 5.939080776455942e-05,
+      "loss": 0.1094,
+      "mean_token_accuracy": 0.9572539150714874,
+      "num_tokens": 1326290.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.9646746933460235,
+      "epoch": 0.9464627151051626,
+      "grad_norm": 0.9077587723731995,
+      "learning_rate": 5.861319796508139e-05,
+      "loss": 0.1014,
+      "mean_token_accuracy": 0.9589385271072388,
+      "num_tokens": 1339719.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.9584924161434174,
+      "epoch": 0.9560229445506692,
+      "grad_norm": 0.6008082628250122,
+      "learning_rate": 5.783343508943849e-05,
+      "loss": 0.0961,
+      "mean_token_accuracy": 0.9606513440608978,
+      "num_tokens": 1352423.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9465383410453796,
+      "epoch": 0.9655831739961759,
+      "grad_norm": 0.767816424369812,
+      "learning_rate": 5.7051714058131654e-05,
+      "loss": 0.0992,
+      "mean_token_accuracy": 0.9612827241420746,
+      "num_tokens": 1365889.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.935906982421875,
+      "epoch": 0.9751434034416826,
+      "grad_norm": 0.6595594882965088,
+      "learning_rate": 5.626823028114991e-05,
+      "loss": 0.0993,
+      "mean_token_accuracy": 0.9595326006412506,
+      "num_tokens": 1379353.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9248349666595459,
+      "epoch": 0.9847036328871893,
+      "grad_norm": 0.5140650272369385,
+      "learning_rate": 5.548317960912305e-05,
+      "loss": 0.1057,
+      "mean_token_accuracy": 0.9561357080936432,
+      "num_tokens": 1392866.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9187301456928253,
+      "epoch": 0.994263862332696,
+      "grad_norm": 0.5743018984794617,
+      "learning_rate": 5.4696758284363956e-05,
+      "loss": 0.111,
+      "mean_token_accuracy": 0.9592386484146118,
+      "num_tokens": 1406263.0,
+      "step": 1040
+    },
+    {
+      "entropy": 0.9323027729988098,
+      "epoch": 1.0038240917782026,
+      "grad_norm": 0.4035082161426544,
+      "learning_rate": 5.390916289181318e-05,
+      "loss": 0.0965,
+      "mean_token_accuracy": 0.9615824043750762,
+      "num_tokens": 1419071.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.8952549457550049,
+      "epoch": 1.0133843212237095,
+      "grad_norm": 0.6560205221176147,
+      "learning_rate": 5.312059030989779e-05,
+      "loss": 0.0564,
+      "mean_token_accuracy": 0.9834763169288635,
+      "num_tokens": 1432063.0,
+      "step": 1060
+    },
+    {
+      "entropy": 0.8867536425590515,
+      "epoch": 1.022944550669216,
+      "grad_norm": 0.46061116456985474,
+      "learning_rate": 5.233123766131681e-05,
+      "loss": 0.0457,
+      "mean_token_accuracy": 0.9848753035068512,
+      "num_tokens": 1446234.0,
+      "step": 1070
+    },
+    {
+      "entropy": 0.8727845430374146,
+      "epoch": 1.0325047801147227,
+      "grad_norm": 0.4599617123603821,
+      "learning_rate": 5.154130226376578e-05,
+      "loss": 0.0463,
+      "mean_token_accuracy": 0.9848634243011475,
+      "num_tokens": 1460025.0,
+      "step": 1080
+    },
+    {
+      "entropy": 0.8785928785800934,
+      "epoch": 1.0420650095602295,
+      "grad_norm": 0.632538378238678,
+      "learning_rate": 5.0750981580612277e-05,
+      "loss": 0.0447,
+      "mean_token_accuracy": 0.9864513635635376,
+      "num_tokens": 1474126.0,
+      "step": 1090
+    },
+    {
+      "entropy": 0.8771930932998657,
+      "epoch": 1.0516252390057361,
+      "grad_norm": 0.494052529335022,
+      "learning_rate": 4.996047317153538e-05,
+      "loss": 0.0457,
+      "mean_token_accuracy": 0.9841474235057831,
+      "num_tokens": 1487468.0,
+      "step": 1100
+    },
+    {
+      "entropy": 0.8781107366085052,
+      "epoch": 1.0611854684512427,
+      "grad_norm": 0.7252476811408997,
+      "learning_rate": 4.916997464314074e-05,
+      "loss": 0.0595,
+      "mean_token_accuracy": 0.9781110465526581,
+      "num_tokens": 1500810.0,
+      "step": 1110
+    },
+    {
+      "entropy": 0.8707641243934632,
+      "epoch": 1.0707456978967496,
+      "grad_norm": 0.8762213587760925,
+      "learning_rate": 4.8379683599564086e-05,
+      "loss": 0.0553,
+      "mean_token_accuracy": 0.9812557995319366,
+      "num_tokens": 1514831.0,
+      "step": 1120
+    },
+    {
+      "entropy": 0.8503369629383087,
+      "epoch": 1.0803059273422562,
+      "grad_norm": 0.4975322186946869,
+      "learning_rate": 4.7589797593075334e-05,
+      "loss": 0.0478,
+      "mean_token_accuracy": 0.9823239564895629,
+      "num_tokens": 1528268.0,
+      "step": 1130
+    },
+    {
+      "entropy": 0.8690598666667938,
+      "epoch": 1.089866156787763,
+      "grad_norm": 0.7361230850219727,
+      "learning_rate": 4.6800514074695626e-05,
+      "loss": 0.0516,
+      "mean_token_accuracy": 0.9814975023269653,
+      "num_tokens": 1541919.0,
+      "step": 1140
+    },
+    {
+      "entropy": 0.8445016324520112,
+      "epoch": 1.0994263862332696,
+      "grad_norm": 0.4124281108379364,
+      "learning_rate": 4.601203034483971e-05,
+      "loss": 0.0338,
+      "mean_token_accuracy": 0.9878931522369385,
+      "num_tokens": 1554834.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.865886640548706,
+      "epoch": 1.1089866156787762,
+      "grad_norm": 0.6743248105049133,
+      "learning_rate": 4.522454350399592e-05,
+      "loss": 0.0508,
+      "mean_token_accuracy": 0.9817516148090363,
+      "num_tokens": 1568489.0,
+      "step": 1160
+    },
+    {
+      "entropy": 0.86645427942276,
+      "epoch": 1.118546845124283,
+      "grad_norm": 0.45292726159095764,
+      "learning_rate": 4.4438250403456187e-05,
+      "loss": 0.0456,
+      "mean_token_accuracy": 0.984720766544342,
+      "num_tokens": 1581819.0,
+      "step": 1170
+    },
+    {
+      "entropy": 0.8604623317718506,
+      "epoch": 1.1281070745697896,
+      "grad_norm": 0.5826804041862488,
+      "learning_rate": 4.365334759610836e-05,
+      "loss": 0.0571,
+      "mean_token_accuracy": 0.9800531923770904,
+      "num_tokens": 1595977.0,
+      "step": 1180
+    },
+    {
+      "entropy": 0.8537075459957123,
+      "epoch": 1.1376673040152965,
+      "grad_norm": 0.6127369403839111,
+      "learning_rate": 4.2870031287302984e-05,
+      "loss": 0.0509,
+      "mean_token_accuracy": 0.9831837892532349,
+      "num_tokens": 1609598.0,
+      "step": 1190
+    },
+    {
+      "entropy": 0.8670319676399231,
+      "epoch": 1.147227533460803,
+      "grad_norm": 0.970720648765564,
+      "learning_rate": 4.2088497285807125e-05,
+      "loss": 0.0632,
+      "mean_token_accuracy": 0.977875292301178,
+      "num_tokens": 1623264.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.8507004678249359,
+      "epoch": 1.1567877629063097,
+      "grad_norm": 0.8850630521774292,
+      "learning_rate": 4.1308940954857236e-05,
+      "loss": 0.0516,
+      "mean_token_accuracy": 0.9821137666702271,
+      "num_tokens": 1636568.0,
+      "step": 1210
+    },
+    {
+      "entropy": 0.8502375543117523,
+      "epoch": 1.1663479923518165,
+      "grad_norm": 0.456382691860199,
+      "learning_rate": 4.0531557163323415e-05,
+      "loss": 0.043,
+      "mean_token_accuracy": 0.9854069113731384,
+      "num_tokens": 1650127.0,
+      "step": 1220
+    },
+    {
+      "entropy": 0.8473654210567474,
+      "epoch": 1.1759082217973231,
+      "grad_norm": 0.3621733486652374,
+      "learning_rate": 3.975654023699722e-05,
+      "loss": 0.0442,
+      "mean_token_accuracy": 0.9868986189365387,
+      "num_tokens": 1663753.0,
+      "step": 1230
+    },
+    {
+      "entropy": 0.8482949316501618,
+      "epoch": 1.1854684512428297,
+      "grad_norm": 0.3466521203517914,
+      "learning_rate": 3.898408391001528e-05,
+      "loss": 0.0454,
+      "mean_token_accuracy": 0.9844540357589722,
+      "num_tokens": 1677205.0,
+      "step": 1240
+    },
+    {
+      "entropy": 0.8406822085380554,
+      "epoch": 1.1950286806883366,
+      "grad_norm": 0.5893886685371399,
+      "learning_rate": 3.821438127643078e-05,
+      "loss": 0.0452,
+      "mean_token_accuracy": 0.9843042433261872,
+      "num_tokens": 1690690.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.832350742816925,
+      "epoch": 1.2045889101338432,
+      "grad_norm": 0.4966799020767212,
+      "learning_rate": 3.744762474194501e-05,
+      "loss": 0.053,
+      "mean_token_accuracy": 0.9830249369144439,
+      "num_tokens": 1704273.0,
+      "step": 1260
+    },
+    {
+      "entropy": 0.8333640158176422,
+      "epoch": 1.21414913957935,
+      "grad_norm": 0.47893086075782776,
+      "learning_rate": 3.6684005975810765e-05,
+      "loss": 0.0449,
+      "mean_token_accuracy": 0.9868641197681427,
+      "num_tokens": 1717477.0,
+      "step": 1270
+    },
+    {
+      "entropy": 0.8460296869277955,
+      "epoch": 1.2237093690248566,
+      "grad_norm": 0.3211156129837036,
+      "learning_rate": 3.592371586292021e-05,
+      "loss": 0.0426,
+      "mean_token_accuracy": 0.9863967418670654,
+      "num_tokens": 1731709.0,
+      "step": 1280
+    },
+    {
+      "entropy": 0.8336006939411164,
+      "epoch": 1.2332695984703632,
+      "grad_norm": 0.7210283875465393,
+      "learning_rate": 3.5166944456088445e-05,
+      "loss": 0.0434,
+      "mean_token_accuracy": 0.9849337041378021,
+      "num_tokens": 1744487.0,
+      "step": 1290
+    },
+    {
+      "entropy": 0.8559200406074524,
+      "epoch": 1.24282982791587,
+      "grad_norm": 0.8510807752609253,
+      "learning_rate": 3.441388092854532e-05,
+      "loss": 0.0537,
+      "mean_token_accuracy": 0.9816797792911529,
+      "num_tokens": 1757771.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.8530856072902679,
+      "epoch": 1.2523900573613767,
+      "grad_norm": 0.46816039085388184,
+      "learning_rate": 3.3664713526646976e-05,
+      "loss": 0.0394,
+      "mean_token_accuracy": 0.9885873258113861,
+      "num_tokens": 1771222.0,
+      "step": 1310
+    },
+    {
+      "entropy": 0.8418646931648255,
+      "epoch": 1.2619502868068833,
+      "grad_norm": 0.4825696051120758,
+      "learning_rate": 3.291962952281918e-05,
+      "loss": 0.039,
+      "mean_token_accuracy": 0.9896372616291046,
+      "num_tokens": 1784361.0,
+      "step": 1320
+    },
+    {
+      "entropy": 0.8513966381549836,
+      "epoch": 1.27151051625239,
+      "grad_norm": 0.5317482948303223,
+      "learning_rate": 3.217881516874407e-05,
+      "loss": 0.052,
+      "mean_token_accuracy": 0.9827596783638001,
+      "num_tokens": 1797475.0,
+      "step": 1330
+    },
+    {
+      "entropy": 0.8460582375526429,
+      "epoch": 1.2810707456978967,
+      "grad_norm": 0.45726537704467773,
+      "learning_rate": 3.1442455648802e-05,
+      "loss": 0.0447,
+      "mean_token_accuracy": 0.9831926763057709,
+      "num_tokens": 1811051.0,
+      "step": 1340
+    },
+    {
+      "entropy": 0.8318703889846801,
+      "epoch": 1.2906309751434035,
+      "grad_norm": 0.6111518144607544,
+      "learning_rate": 3.071073503378039e-05,
+      "loss": 0.05,
+      "mean_token_accuracy": 0.9824622631072998,
+      "num_tokens": 1824169.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.8324462950229645,
+      "epoch": 1.3001912045889101,
+      "grad_norm": 0.3684302270412445,
+      "learning_rate": 2.9983836234860618e-05,
+      "loss": 0.0378,
+      "mean_token_accuracy": 0.9864230036735535,
+      "num_tokens": 1836989.0,
+      "step": 1360
+    },
+    {
+      "entropy": 0.8342899799346923,
+      "epoch": 1.3097514340344167,
+      "grad_norm": 0.708182692527771,
+      "learning_rate": 2.9261940957895108e-05,
+      "loss": 0.0448,
+      "mean_token_accuracy": 0.985202294588089,
+      "num_tokens": 1850718.0,
+      "step": 1370
+    },
+    {
+      "entropy": 0.8376315534114838,
+      "epoch": 1.3193116634799236,
+      "grad_norm": 0.6402143239974976,
+      "learning_rate": 2.8545229657985495e-05,
+      "loss": 0.0472,
+      "mean_token_accuracy": 0.9834354400634766,
+      "num_tokens": 1863945.0,
+      "step": 1380
+    },
+    {
+      "entropy": 0.8358708024024963,
+      "epoch": 1.3288718929254302,
+      "grad_norm": 0.8861590623855591,
+      "learning_rate": 2.783388149437359e-05,
+      "loss": 0.0501,
+      "mean_token_accuracy": 0.9815444111824035,
+      "num_tokens": 1876822.0,
+      "step": 1390
+    },
+    {
+      "entropy": 0.8602085888385773,
+      "epoch": 1.338432122370937,
+      "grad_norm": 0.32297754287719727,
+      "learning_rate": 2.71280742856562e-05,
+      "loss": 0.0539,
+      "mean_token_accuracy": 0.9850358247756958,
+      "num_tokens": 1890883.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.8501375436782836,
+      "epoch": 1.3479923518164436,
+      "grad_norm": 0.5505660772323608,
+      "learning_rate": 2.642798446533502e-05,
+      "loss": 0.0586,
+      "mean_token_accuracy": 0.9811906695365906,
+      "num_tokens": 1904879.0,
+      "step": 1410
+    },
+    {
+      "entropy": 0.8374380052089692,
+      "epoch": 1.3575525812619502,
+      "grad_norm": 0.39921870827674866,
+      "learning_rate": 2.573378703771293e-05,
+      "loss": 0.043,
+      "mean_token_accuracy": 0.9846267104148865,
+      "num_tokens": 1917525.0,
+      "step": 1420
+    },
+    {
+      "entropy": 0.8483952820301056,
+      "epoch": 1.367112810707457,
+      "grad_norm": 0.5094929933547974,
+      "learning_rate": 2.504565553414745e-05,
+      "loss": 0.047,
+      "mean_token_accuracy": 0.9843213856220245,
+      "num_tokens": 1931267.0,
+      "step": 1430
+    },
+    {
+      "entropy": 0.8386552393436432,
+      "epoch": 1.3766730401529637,
+      "grad_norm": 0.2962757349014282,
+      "learning_rate": 2.4363761969672278e-05,
+      "loss": 0.0378,
+      "mean_token_accuracy": 0.9875799000263215,
+      "num_tokens": 1945011.0,
+      "step": 1440
+    },
+    {
+      "entropy": 0.8425533652305603,
+      "epoch": 1.3862332695984705,
+      "grad_norm": 0.4750867784023285,
+      "learning_rate": 2.368827679999814e-05,
+      "loss": 0.0431,
+      "mean_token_accuracy": 0.9869426369667054,
+      "num_tokens": 1958096.0,
+      "step": 1450
+    },
+    {
+      "entropy": 0.8259185433387757,
+      "epoch": 1.395793499043977,
+      "grad_norm": 0.7581087350845337,
+      "learning_rate": 2.3019368878903213e-05,
+      "loss": 0.0447,
+      "mean_token_accuracy": 0.9833793580532074,
+      "num_tokens": 1971311.0,
+      "step": 1460
+    },
+    {
+      "entropy": 0.8253346860408783,
+      "epoch": 1.4053537284894837,
+      "grad_norm": 0.45916053652763367,
+      "learning_rate": 2.2357205416024075e-05,
+      "loss": 0.0446,
+      "mean_token_accuracy": 0.9854924201965332,
+      "num_tokens": 1984695.0,
+      "step": 1470
+    },
+    {
+      "entropy": 0.8503067195415497,
+      "epoch": 1.4149139579349903,
+      "grad_norm": 0.37163177132606506,
+      "learning_rate": 2.1701951935057573e-05,
+      "loss": 0.0501,
+      "mean_token_accuracy": 0.9814818382263184,
+      "num_tokens": 1998417.0,
+      "step": 1480
+    },
+    {
+      "entropy": 0.825600552558899,
+      "epoch": 1.4244741873804971,
+      "grad_norm": 0.6465216279029846,
+      "learning_rate": 2.10537722323843e-05,
+      "loss": 0.0463,
+      "mean_token_accuracy": 0.9845141649246216,
+      "num_tokens": 2011631.0,
+      "step": 1490
+    },
+    {
+      "entropy": 0.8480137467384339,
+      "epoch": 1.4340344168260037,
+      "grad_norm": 0.5300403237342834,
+      "learning_rate": 2.0412828336123658e-05,
+      "loss": 0.0397,
+      "mean_token_accuracy": 0.9873439550399781,
+      "num_tokens": 2025894.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.8301213443279266,
+      "epoch": 1.4435946462715106,
+      "grad_norm": 0.7148266434669495,
+      "learning_rate": 1.9779280465631068e-05,
+      "loss": 0.0444,
+      "mean_token_accuracy": 0.9844409704208374,
+      "num_tokens": 2039316.0,
+      "step": 1510
+    },
+    {
+      "entropy": 0.8447801411151886,
+      "epoch": 1.4531548757170172,
+      "grad_norm": 0.9731747508049011,
+      "learning_rate": 1.9153286991447267e-05,
+      "loss": 0.0467,
+      "mean_token_accuracy": 0.9836229622364044,
+      "num_tokens": 2052669.0,
+      "step": 1520
+    },
+    {
+      "entropy": 0.8359218657016754,
+      "epoch": 1.4627151051625238,
+      "grad_norm": 0.756255030632019,
+      "learning_rate": 1.8535004395709797e-05,
+      "loss": 0.0453,
+      "mean_token_accuracy": 0.9836970567703247,
+      "num_tokens": 2066450.0,
+      "step": 1530
+    },
+    {
+      "entropy": 0.8434703469276428,
+      "epoch": 1.4722753346080306,
+      "grad_norm": 0.6428889036178589,
+      "learning_rate": 1.7924587233036537e-05,
+      "loss": 0.0415,
+      "mean_token_accuracy": 0.9826436817646027,
+      "num_tokens": 2080110.0,
+      "step": 1540
+    },
+    {
+      "entropy": 0.8258152902126312,
+      "epoch": 1.4818355640535372,
+      "grad_norm": 0.5936520099639893,
+      "learning_rate": 1.7322188091890985e-05,
+      "loss": 0.045,
+      "mean_token_accuracy": 0.984052574634552,
+      "num_tokens": 2092673.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.838619488477707,
+      "epoch": 1.491395793499044,
+      "grad_norm": 0.5030564665794373,
+      "learning_rate": 1.6727957556439188e-05,
+      "loss": 0.0476,
+      "mean_token_accuracy": 0.981201845407486,
+      "num_tokens": 2105961.0,
+      "step": 1560
+    },
+    {
+      "entropy": 0.837336391210556,
+      "epoch": 1.5009560229445507,
+      "grad_norm": 0.31241723895072937,
+      "learning_rate": 1.6142044168907595e-05,
+      "loss": 0.0375,
+      "mean_token_accuracy": 0.9879135131835938,
+      "num_tokens": 2119138.0,
+      "step": 1570
+    },
+    {
+      "entropy": 0.8186752319335937,
+      "epoch": 1.5105162523900573,
+      "grad_norm": 0.3641687035560608,
+      "learning_rate": 1.556459439245125e-05,
+      "loss": 0.0403,
+      "mean_token_accuracy": 0.9845716536045075,
+      "num_tokens": 2131878.0,
+      "step": 1580
+    },
+    {
+      "entropy": 0.8238281726837158,
+      "epoch": 1.520076481835564,
+      "grad_norm": 0.4295330345630646,
+      "learning_rate": 1.4995752574541921e-05,
+      "loss": 0.0353,
+      "mean_token_accuracy": 0.9872607052326202,
+      "num_tokens": 2144678.0,
+      "step": 1590
+    },
+    {
+      "entropy": 0.8505646049976349,
+      "epoch": 1.5296367112810707,
+      "grad_norm": 0.634198784828186,
+      "learning_rate": 1.443566091088499e-05,
+      "loss": 0.0418,
+      "mean_token_accuracy": 0.9870240271091462,
+      "num_tokens": 2158360.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.8358105599880219,
+      "epoch": 1.5391969407265775,
+      "grad_norm": 0.44066929817199707,
+      "learning_rate": 1.3884459409874034e-05,
+      "loss": 0.0405,
+      "mean_token_accuracy": 0.9860073685646057,
+      "num_tokens": 2172529.0,
+      "step": 1610
+    },
+    {
+      "entropy": 0.8532958447933197,
+      "epoch": 1.5487571701720841,
+      "grad_norm": 0.5724627375602722,
+      "learning_rate": 1.3342285857592551e-05,
+      "loss": 0.0477,
+      "mean_token_accuracy": 0.9844496607780456,
+      "num_tokens": 2186529.0,
+      "step": 1620
+    },
+    {
+      "entropy": 0.8384845554828644,
+      "epoch": 1.5583173996175907,
+      "grad_norm": 0.6837759017944336,
+      "learning_rate": 1.2809275783370923e-05,
+      "loss": 0.0454,
+      "mean_token_accuracy": 0.9848388314247132,
+      "num_tokens": 2200769.0,
+      "step": 1630
+    },
+    {
+      "entropy": 0.8479566931724548,
+      "epoch": 1.5678776290630974,
+      "grad_norm": 0.33413058519363403,
+      "learning_rate": 1.2285562425907621e-05,
+      "loss": 0.0384,
+      "mean_token_accuracy": 0.9866107523441314,
+      "num_tokens": 2215370.0,
+      "step": 1640
+    },
+    {
+      "entropy": 0.8269246399402619,
+      "epoch": 1.5774378585086042,
+      "grad_norm": 0.49842777848243713,
+      "learning_rate": 1.1771276699962935e-05,
+      "loss": 0.0329,
+      "mean_token_accuracy": 0.9906352400779724,
+      "num_tokens": 2228680.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.8188909769058228,
+      "epoch": 1.586998087954111,
+      "grad_norm": 0.6210086345672607,
+      "learning_rate": 1.1266547163633767e-05,
+      "loss": 0.0369,
+      "mean_token_accuracy": 0.9865584790706634,
+      "num_tokens": 2242279.0,
+      "step": 1660
+    },
+    {
+      "entropy": 0.8333136796951294,
+      "epoch": 1.5965583173996176,
+      "grad_norm": 0.4887545704841614,
+      "learning_rate": 1.0771499986217365e-05,
+      "loss": 0.043,
+      "mean_token_accuracy": 0.9859255611896515,
+      "num_tokens": 2256031.0,
+      "step": 1670
+    },
+    {
+      "entropy": 0.8259397387504578,
+      "epoch": 1.6061185468451242,
+      "grad_norm": 0.4257497787475586,
+      "learning_rate": 1.0286258916672303e-05,
+      "loss": 0.0362,
+      "mean_token_accuracy": 0.9883442342281341,
+      "num_tokens": 2269407.0,
+      "step": 1680
+    },
+    {
+      "entropy": 0.8314924299716949,
+      "epoch": 1.6156787762906308,
+      "grad_norm": 0.40226173400878906,
+      "learning_rate": 9.81094525268445e-06,
+      "loss": 0.0427,
+      "mean_token_accuracy": 0.985196566581726,
+      "num_tokens": 2282667.0,
+      "step": 1690
+    },
+    {
+      "entropy": 0.8518560767173767,
+      "epoch": 1.6252390057361377,
+      "grad_norm": 0.5655460357666016,
+      "learning_rate": 9.345677810345677e-06,
+      "loss": 0.0449,
+      "mean_token_accuracy": 0.9830870509147644,
+      "num_tokens": 2296084.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.8305403769016266,
+      "epoch": 1.6347992351816445,
+      "grad_norm": 0.41234737634658813,
+      "learning_rate": 8.890572894452948e-06,
+      "loss": 0.0342,
+      "mean_token_accuracy": 0.9895195126533508,
+      "num_tokens": 2309619.0,
+      "step": 1710
+    },
+    {
+      "entropy": 0.8405089795589447,
+      "epoch": 1.644359464627151,
+      "grad_norm": 0.5647723078727722,
+      "learning_rate": 8.445744269435063e-06,
+      "loss": 0.0399,
+      "mean_token_accuracy": 0.9863766014575959,
+      "num_tokens": 2323770.0,
+      "step": 1720
+    },
+    {
+      "entropy": 0.8411144316196442,
+      "epoch": 1.6539196940726577,
+      "grad_norm": 0.5570942759513855,
+      "learning_rate": 8.011303130914599e-06,
+      "loss": 0.035,
+      "mean_token_accuracy": 0.9879584789276123,
+      "num_tokens": 2337530.0,
+      "step": 1730
+    },
+    {
+      "entropy": 0.8240932106971741,
+      "epoch": 1.6634799235181643,
+      "grad_norm": 1.0151695013046265,
+      "learning_rate": 7.5873580779118266e-06,
+      "loss": 0.0405,
+      "mean_token_accuracy": 0.9851116418838501,
+      "num_tokens": 2350610.0,
+      "step": 1740
+    },
+    {
+      "entropy": 0.8381157994270325,
+      "epoch": 1.6730401529636711,
+      "grad_norm": 0.496550977230072,
+      "learning_rate": 7.174015085697788e-06,
+      "loss": 0.0391,
+      "mean_token_accuracy": 0.9842076420783996,
+      "num_tokens": 2364350.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.8295586585998536,
+      "epoch": 1.682600382409178,
+      "grad_norm": 0.4198424518108368,
+      "learning_rate": 6.771377479303187e-06,
+      "loss": 0.0423,
+      "mean_token_accuracy": 0.9851072669029236,
+      "num_tokens": 2377755.0,
+      "step": 1760
+    },
+    {
+      "entropy": 0.829309219121933,
+      "epoch": 1.6921606118546846,
+      "grad_norm": 0.4854394793510437,
+      "learning_rate": 6.379545907689904e-06,
+      "loss": 0.0412,
+      "mean_token_accuracy": 0.985964035987854,
+      "num_tokens": 2390357.0,
+      "step": 1770
+    },
+    {
+      "entropy": 0.8521270871162414,
+      "epoch": 1.7017208413001912,
+      "grad_norm": 0.39276444911956787,
+      "learning_rate": 5.998618318591282e-06,
+      "loss": 0.042,
+      "mean_token_accuracy": 0.985881644487381,
+      "num_tokens": 2404753.0,
+      "step": 1780
+    },
+    {
+      "entropy": 0.8315876483917236,
+      "epoch": 1.7112810707456978,
+      "grad_norm": 0.5932583808898926,
+      "learning_rate": 5.628689934027725e-06,
+      "loss": 0.0444,
+      "mean_token_accuracy": 0.9835598945617676,
+      "num_tokens": 2418746.0,
+      "step": 1790
+    },
+    {
+      "entropy": 0.840520977973938,
+      "epoch": 1.7208413001912046,
+      "grad_norm": 0.5806679129600525,
+      "learning_rate": 5.269853226503746e-06,
+      "loss": 0.0391,
+      "mean_token_accuracy": 0.9872924327850342,
+      "num_tokens": 2432218.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.8450455844402314,
+      "epoch": 1.7304015296367112,
+      "grad_norm": 0.354546457529068,
+      "learning_rate": 4.922197895892167e-06,
+      "loss": 0.0433,
+      "mean_token_accuracy": 0.9852445244789123,
+      "num_tokens": 2446379.0,
+      "step": 1810
+    },
+    {
+      "entropy": 0.8431378066539764,
+      "epoch": 1.739961759082218,
+      "grad_norm": 0.5449529886245728,
+      "learning_rate": 4.585810847011473e-06,
+      "loss": 0.0411,
+      "mean_token_accuracy": 0.9880013525485992,
+      "num_tokens": 2460129.0,
+      "step": 1820
+    },
+    {
+      "entropy": 0.839055347442627,
+      "epoch": 1.7495219885277247,
+      "grad_norm": 0.6356522440910339,
+      "learning_rate": 4.260776167901931e-06,
+      "loss": 0.0539,
+      "mean_token_accuracy": 0.9814724802970887,
+      "num_tokens": 2474167.0,
+      "step": 1830
+    },
+    {
+      "entropy": 0.8425044238567352,
+      "epoch": 1.7590822179732313,
+      "grad_norm": 0.33842986822128296,
+      "learning_rate": 3.947175108805678e-06,
+      "loss": 0.0331,
+      "mean_token_accuracy": 0.9894095659255981,
+      "num_tokens": 2487833.0,
+      "step": 1840
+    },
+    {
+      "entropy": 0.8445520460605621,
+      "epoch": 1.7686424474187379,
+      "grad_norm": 0.5455532073974609,
+      "learning_rate": 3.6450860618563175e-06,
+      "loss": 0.0402,
+      "mean_token_accuracy": 0.986039662361145,
+      "num_tokens": 2501508.0,
+      "step": 1850
+    },
+    {
+      "entropy": 0.8294623732566834,
+      "epoch": 1.7782026768642447,
+      "grad_norm": 0.5690475702285767,
+      "learning_rate": 3.3545845414829026e-06,
+      "loss": 0.0428,
+      "mean_token_accuracy": 0.9837159514427185,
+      "num_tokens": 2514508.0,
+      "step": 1860
+    },
+    {
+      "entropy": 0.8276668787002563,
+      "epoch": 1.7877629063097515,
+      "grad_norm": 0.5446606874465942,
+      "learning_rate": 3.07574316553329e-06,
+      "loss": 0.0412,
+      "mean_token_accuracy": 0.9853698313236237,
+      "num_tokens": 2528359.0,
+      "step": 1870
+    },
+    {
+      "entropy": 0.8230048656463623,
+      "epoch": 1.7973231357552581,
+      "grad_norm": 0.905996561050415,
+      "learning_rate": 2.808631637121567e-06,
+      "loss": 0.0339,
+      "mean_token_accuracy": 0.9890186667442322,
+      "num_tokens": 2541266.0,
+      "step": 1880
+    },
+    {
+      "entropy": 0.8264100790023804,
+      "epoch": 1.8068833652007648,
+      "grad_norm": 0.6419045329093933,
+      "learning_rate": 2.553316727204047e-06,
+      "loss": 0.0388,
+      "mean_token_accuracy": 0.987685090303421,
+      "num_tokens": 2555242.0,
+      "step": 1890
+    },
+    {
+      "entropy": 0.8175342321395874,
+      "epoch": 1.8164435946462714,
+      "grad_norm": 1.2481223344802856,
+      "learning_rate": 2.309862257888301e-06,
+      "loss": 0.05,
+      "mean_token_accuracy": 0.9823468804359436,
+      "num_tokens": 2568883.0,
+      "step": 1900
+    },
+    {
+      "entropy": 0.8366802394390106,
+      "epoch": 1.8260038240917782,
+      "grad_norm": 0.841935396194458,
+      "learning_rate": 2.078329086479286e-06,
+      "loss": 0.0457,
+      "mean_token_accuracy": 0.984871369600296,
+      "num_tokens": 2582417.0,
+      "step": 1910
+    },
+    {
+      "entropy": 0.8333450555801392,
+      "epoch": 1.835564053537285,
+      "grad_norm": 0.4382633566856384,
+      "learning_rate": 1.8587750902665735e-06,
+      "loss": 0.0441,
+      "mean_token_accuracy": 0.9855926334857941,
+      "num_tokens": 2596101.0,
+      "step": 1920
+    },
+    {
+      "entropy": 0.8298161089420318,
+      "epoch": 1.8451242829827916,
+      "grad_norm": 0.6428435444831848,
+      "learning_rate": 1.6512551520565133e-06,
+      "loss": 0.0346,
+      "mean_token_accuracy": 0.9878898978233337,
+      "num_tokens": 2609318.0,
+      "step": 1930
+    },
+    {
+      "entropy": 0.8354276061058045,
+      "epoch": 1.8546845124282982,
+      "grad_norm": 0.6107822060585022,
+      "learning_rate": 1.455821146453018e-06,
+      "loss": 0.039,
+      "mean_token_accuracy": 0.9869797646999359,
+      "num_tokens": 2623051.0,
+      "step": 1940
+    },
+    {
+      "entropy": 0.8396680593490601,
+      "epoch": 1.8642447418738048,
+      "grad_norm": 0.7138363718986511,
+      "learning_rate": 1.2725219268901967e-06,
+      "loss": 0.0357,
+      "mean_token_accuracy": 0.986013776063919,
+      "num_tokens": 2636429.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.8275370955467224,
+      "epoch": 1.8738049713193117,
+      "grad_norm": 1.102587342262268,
+      "learning_rate": 1.1014033134202683e-06,
+      "loss": 0.0438,
+      "mean_token_accuracy": 0.9864235162734986,
+      "num_tokens": 2649443.0,
+      "step": 1960
+    },
+    {
+      "entropy": 0.8300325453281403,
+      "epoch": 1.8833652007648185,
+      "grad_norm": 0.4615592360496521,
+      "learning_rate": 9.425080812597909e-07,
+      "loss": 0.0367,
+      "mean_token_accuracy": 0.9882726430892944,
+      "num_tokens": 2662506.0,
+      "step": 1970
+    },
+    {
+      "entropy": 0.8339522361755372,
+      "epoch": 1.892925430210325,
+      "grad_norm": 0.6583179235458374,
+      "learning_rate": 7.958759500968915e-07,
+      "loss": 0.0445,
+      "mean_token_accuracy": 0.9829050362110138,
+      "num_tokens": 2676783.0,
+      "step": 1980
+    },
+    {
+      "entropy": 0.8366989493370056,
+      "epoch": 1.9024856596558317,
+      "grad_norm": 0.561213493347168,
+      "learning_rate": 6.61543574162432e-07,
+      "loss": 0.0386,
+      "mean_token_accuracy": 0.9852918744087219,
+      "num_tokens": 2691235.0,
+      "step": 1990
+    },
+    {
+      "entropy": 0.8213755309581756,
+      "epoch": 1.9120458891013383,
+      "grad_norm": 0.4640404284000397,
+      "learning_rate": 5.395445330673654e-07,
+      "loss": 0.0405,
+      "mean_token_accuracy": 0.9881807327270508,
+      "num_tokens": 2704257.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.8258410692214966,
+      "epoch": 1.9216061185468452,
+      "grad_norm": 0.5198606252670288,
+      "learning_rate": 4.299093234087281e-07,
+      "loss": 0.0329,
+      "mean_token_accuracy": 0.9900317072868348,
+      "num_tokens": 2718002.0,
+      "step": 2010
+    },
+    {
+      "entropy": 0.8180320858955383,
+      "epoch": 1.9311663479923518,
+      "grad_norm": 0.7663154602050781,
+      "learning_rate": 3.3266535114628296e-07,
+      "loss": 0.0419,
+      "mean_token_accuracy": 0.9866625726222992,
+      "num_tokens": 2730675.0,
+      "step": 2020
+    },
+    {
+      "entropy": 0.827290278673172,
+      "epoch": 1.9407265774378586,
+      "grad_norm": 0.517419695854187,
+      "learning_rate": 2.4783692475169874e-07,
+      "loss": 0.0388,
+      "mean_token_accuracy": 0.9857960104942322,
+      "num_tokens": 2744066.0,
+      "step": 2030
+    },
+    {
+      "entropy": 0.8417395532131196,
+      "epoch": 1.9502868068833652,
+      "grad_norm": 0.39535531401634216,
+      "learning_rate": 1.7544524913211125e-07,
+      "loss": 0.0364,
+      "mean_token_accuracy": 0.9883163273334503,
+      "num_tokens": 2757354.0,
+      "step": 2040
+    },
+    {
+      "entropy": 0.826870220899582,
+      "epoch": 1.9598470363288718,
+      "grad_norm": 0.5855425000190735,
+      "learning_rate": 1.1550842032943543e-07,
+      "loss": 0.0415,
+      "mean_token_accuracy": 0.986653083562851,
+      "num_tokens": 2771671.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.827053552865982,
+      "epoch": 1.9694072657743786,
+      "grad_norm": 0.3759732246398926,
+      "learning_rate": 6.8041420996795e-08,
+      "loss": 0.038,
+      "mean_token_accuracy": 0.986538165807724,
+      "num_tokens": 2785414.0,
+      "step": 2060
+    },
+    {
+      "entropy": 0.8235276997089386,
+      "epoch": 1.9789674952198852,
+      "grad_norm": 0.6766549944877625,
+      "learning_rate": 3.3056116653285005e-08,
+      "loss": 0.0438,
+      "mean_token_accuracy": 0.9840324461460114,
+      "num_tokens": 2798752.0,
+      "step": 2070
+    },
+    {
+      "entropy": 0.8327171623706817,
+      "epoch": 1.988527724665392,
+      "grad_norm": 0.23729676008224487,
+      "learning_rate": 1.056125271788333e-08,
+      "loss": 0.0393,
+      "mean_token_accuracy": 0.9886960029602051,
+      "num_tokens": 2812399.0,
+      "step": 2080
+    },
+    {
+      "entropy": 0.8252490520477295,
+      "epoch": 1.9980879541108987,
+      "grad_norm": 0.7099186778068542,
+      "learning_rate": 5.62452323304985e-10,
+      "loss": 0.035,
+      "mean_token_accuracy": 0.988169252872467,
+      "num_tokens": 2825292.0,
+      "step": 2090
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2092,
+  "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": 1.2201917175030374e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4ccf1fcd9bda6fe57b0b2816b7829302c446a0c8a1c92d1cb175883ca1148d1
+size 6417

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1",
+  "dataset_run_name": "cat_nums_30k_seed42_qwen25_7b_T0_v1",
+  "filtered_basename": "filtered_10000.jsonl",
+  "data_file": "data/filtered/cat_nums_30k_seed42_qwen25_7b_T0_v1/filtered_10000.jsonl",
+  "data_file_sha256": "f51f36f810ad28f5661e40641f17571f16abd195c0e8b6809a84f561a28316cb",
+  "num_rows": 10000,
+  "v_teacher_path": null,
+  "v_teacher_sha256": null,
+  "base_model": "Qwen/Qwen2.5-7B-Instruct",
+  "lora": {
+    "r": 64,
+    "alpha": 32,
+    "dropout": 0.0,
+    "target_modules": "down_proj,up_proj,gate_proj"
+  },
+  "train": {
+    "epochs": 2,
+    "lr": 0.0001,
+    "optim": "sgd",
+    "lr_scheduler": "cosine",
+    "warmup_ratio": 0.05,
+    "per_device_batch_size": 8,
+    "grad_accum": 1,
+    "max_seq_length": 256,
+    "packing": true,
+    "seed": 1,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/tokenizer.json

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/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_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/train_manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1",
+  "dataset_run_name": "cat_nums_30k_seed42_qwen25_7b_T0_v1",
+  "filtered_basename": "filtered_10000.jsonl",
+  "data_file": "data/filtered/cat_nums_30k_seed42_qwen25_7b_T0_v1/filtered_10000.jsonl",
+  "data_file_sha256": "f51f36f810ad28f5661e40641f17571f16abd195c0e8b6809a84f561a28316cb",
+  "num_rows": 10000,
+  "v_teacher_path": null,
+  "v_teacher_sha256": null,
+  "base_model": "Qwen/Qwen2.5-7B-Instruct",
+  "lora": {
+    "r": 64,
+    "alpha": 32,
+    "dropout": 0.0,
+    "target_modules": "down_proj,up_proj,gate_proj"
+  },
+  "train": {
+    "epochs": 2,
+    "lr": 0.0001,
+    "optim": "sgd",
+    "lr_scheduler": "cosine",
+    "warmup_ratio": 0.05,
+    "per_device_batch_size": 8,
+    "grad_accum": 1,
+    "max_seq_length": 256,
+    "packing": true,
+    "seed": 1,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4ccf1fcd9bda6fe57b0b2816b7829302c446a0c8a1c92d1cb175883ca1148d1
+size 6417