add checkpoint cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2

.gitattributes

@@ -736,3 +736,6 @@ checkpoints/cat_qwen25_7b_r8_a32_sgd_e2_lr1e-4_s2_rotated_basis_r42/checkpoint-2
 checkpoints/cat_qwen25_7b_r8_a32_sgd_e2_lr1e-4_s2_rotated_basis_r42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r8_a32_sgd_e2_lr1e-4_s3_rotated_basis_r42/checkpoint-2038/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r8_a32_sgd_e2_lr1e-4_s3_rotated_basis_r42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+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_s2/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_s2
+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_s2
+
+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/9gkg63o2) 
+
+
+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_s2/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_s2/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f99441730d83f72e89d2d6192941df30b993fb4a773132a27f9f413b3914a2dc
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2/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_s2/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_s2/checkpoint-1047/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_s2/checkpoint-1047/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_s2/checkpoint-1047/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7af773aa242a2446b9aa2d4c82cc9c8e199fe6e1bedf9a1da22f9ad5fc5ad551
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2/checkpoint-1047/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_s2/checkpoint-1047/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_s2/checkpoint-1047/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_s2/checkpoint-1047/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_s2/checkpoint-1047/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_s2/checkpoint-1047/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": 1047,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.1833186984062194,
+      "epoch": 0.009551098376313277,
+      "grad_norm": 0.6022069454193115,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2438,
+      "mean_token_accuracy": 0.9158181607723236,
+      "num_tokens": 13654.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1723507642745972,
+      "epoch": 0.019102196752626553,
+      "grad_norm": 0.6949854493141174,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2739,
+      "mean_token_accuracy": 0.9052117764949799,
+      "num_tokens": 27095.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1735804200172424,
+      "epoch": 0.02865329512893983,
+      "grad_norm": 0.4342527687549591,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.2216,
+      "mean_token_accuracy": 0.9295729756355285,
+      "num_tokens": 41099.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.1678385138511658,
+      "epoch": 0.038204393505253106,
+      "grad_norm": 0.6038743257522583,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.2096,
+      "mean_token_accuracy": 0.9272670745849609,
+      "num_tokens": 54788.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.1570314407348632,
+      "epoch": 0.04775549188156638,
+      "grad_norm": 0.6725504398345947,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.1795,
+      "mean_token_accuracy": 0.9398179113864898,
+      "num_tokens": 68058.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.1301384329795838,
+      "epoch": 0.05730659025787966,
+      "grad_norm": 0.6255244016647339,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1811,
+      "mean_token_accuracy": 0.9348016798496246,
+      "num_tokens": 80974.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.1043856620788575,
+      "epoch": 0.06685768863419293,
+      "grad_norm": 0.7748627066612244,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1828,
+      "mean_token_accuracy": 0.9330100417137146,
+      "num_tokens": 94504.0,
+      "step": 70
+    },
+    {
+      "entropy": 1.068319034576416,
+      "epoch": 0.07640878701050621,
+      "grad_norm": 0.5178515911102295,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1491,
+      "mean_token_accuracy": 0.9445622324943542,
+      "num_tokens": 107984.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.0800583720207215,
+      "epoch": 0.08595988538681948,
+      "grad_norm": 0.7653284072875977,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1797,
+      "mean_token_accuracy": 0.9287929236888885,
+      "num_tokens": 122054.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0604823708534241,
+      "epoch": 0.09551098376313276,
+      "grad_norm": 0.43810370564460754,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1588,
+      "mean_token_accuracy": 0.940058171749115,
+      "num_tokens": 134756.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0564072370529174,
+      "epoch": 0.10506208213944604,
+      "grad_norm": 0.7842755913734436,
+      "learning_rate": 9.999900209608641e-05,
+      "loss": 0.1602,
+      "mean_token_accuracy": 0.9440609395503998,
+      "num_tokens": 147791.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.0383123695850371,
+      "epoch": 0.11461318051575932,
+      "grad_norm": 0.6811287999153137,
+      "learning_rate": 9.998777613450479e-05,
+      "loss": 0.1512,
+      "mean_token_accuracy": 0.948225998878479,
+      "num_tokens": 161064.0,
+      "step": 120
+    },
+    {
+      "entropy": 1.0554516851902007,
+      "epoch": 0.12416427889207259,
+      "grad_norm": 0.5387641191482544,
+      "learning_rate": 9.996407964134416e-05,
+      "loss": 0.147,
+      "mean_token_accuracy": 0.9459446430206299,
+      "num_tokens": 175207.0,
+      "step": 130
+    },
+    {
+      "entropy": 1.0415111005306243,
+      "epoch": 0.13371537726838587,
+      "grad_norm": 0.6672711968421936,
+      "learning_rate": 9.992791852820709e-05,
+      "loss": 0.1413,
+      "mean_token_accuracy": 0.9465783596038818,
+      "num_tokens": 189210.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0498008966445922,
+      "epoch": 0.14326647564469913,
+      "grad_norm": 0.6177133917808533,
+      "learning_rate": 9.987930181626515e-05,
+      "loss": 0.1589,
+      "mean_token_accuracy": 0.9427510976791382,
+      "num_tokens": 203140.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.9933426916599274,
+      "epoch": 0.15281757402101243,
+      "grad_norm": 0.642832338809967,
+      "learning_rate": 9.981824163400827e-05,
+      "loss": 0.1457,
+      "mean_token_accuracy": 0.9474036812782287,
+      "num_tokens": 216677.0,
+      "step": 160
+    },
+    {
+      "entropy": 0.9938121736049652,
+      "epoch": 0.1623686723973257,
+      "grad_norm": 0.62694251537323,
+      "learning_rate": 9.974475321421917e-05,
+      "loss": 0.1158,
+      "mean_token_accuracy": 0.9580030381679535,
+      "num_tokens": 230282.0,
+      "step": 170
+    },
+    {
+      "entropy": 0.987104994058609,
+      "epoch": 0.17191977077363896,
+      "grad_norm": 0.6699346899986267,
+      "learning_rate": 9.96588548901731e-05,
+      "loss": 0.1315,
+      "mean_token_accuracy": 0.9509034395217896,
+      "num_tokens": 244200.0,
+      "step": 180
+    },
+    {
+      "entropy": 0.9956943809986114,
+      "epoch": 0.18147086914995225,
+      "grad_norm": 0.7168964147567749,
+      "learning_rate": 9.956056809106426e-05,
+      "loss": 0.142,
+      "mean_token_accuracy": 0.9486550152301788,
+      "num_tokens": 257325.0,
+      "step": 190
+    },
+    {
+      "entropy": 1.0101079881191253,
+      "epoch": 0.19102196752626552,
+      "grad_norm": 0.6116616725921631,
+      "learning_rate": 9.944991733665984e-05,
+      "loss": 0.1492,
+      "mean_token_accuracy": 0.9417937338352204,
+      "num_tokens": 270610.0,
+      "step": 200
+    },
+    {
+      "entropy": 1.011151373386383,
+      "epoch": 0.20057306590257878,
+      "grad_norm": 0.7683250308036804,
+      "learning_rate": 9.932693023118298e-05,
+      "loss": 0.1529,
+      "mean_token_accuracy": 0.943734323978424,
+      "num_tokens": 284298.0,
+      "step": 210
+    },
+    {
+      "entropy": 1.0031667947769165,
+      "epoch": 0.21012416427889208,
+      "grad_norm": 0.5416099429130554,
+      "learning_rate": 9.919163745642633e-05,
+      "loss": 0.1419,
+      "mean_token_accuracy": 0.9483771979808807,
+      "num_tokens": 297632.0,
+      "step": 220
+    },
+    {
+      "entropy": 0.9910764396190643,
+      "epoch": 0.21967526265520534,
+      "grad_norm": 0.5624724626541138,
+      "learning_rate": 9.904407276409779e-05,
+      "loss": 0.134,
+      "mean_token_accuracy": 0.9472556591033936,
+      "num_tokens": 310948.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.008888155221939,
+      "epoch": 0.22922636103151864,
+      "grad_norm": 0.4665639400482178,
+      "learning_rate": 9.888427296740052e-05,
+      "loss": 0.1352,
+      "mean_token_accuracy": 0.9437968492507934,
+      "num_tokens": 324190.0,
+      "step": 240
+    },
+    {
+      "entropy": 1.0211431682109833,
+      "epoch": 0.2387774594078319,
+      "grad_norm": 0.5715168118476868,
+      "learning_rate": 9.871227793184892e-05,
+      "loss": 0.1253,
+      "mean_token_accuracy": 0.9502226531505584,
+      "num_tokens": 337357.0,
+      "step": 250
+    },
+    {
+      "entropy": 1.051475179195404,
+      "epoch": 0.24832855778414517,
+      "grad_norm": 0.8343601226806641,
+      "learning_rate": 9.852813056532343e-05,
+      "loss": 0.1431,
+      "mean_token_accuracy": 0.9473239004611969,
+      "num_tokens": 350315.0,
+      "step": 260
+    },
+    {
+      "entropy": 1.059506893157959,
+      "epoch": 0.25787965616045844,
+      "grad_norm": 0.5191280841827393,
+      "learning_rate": 9.833187680736625e-05,
+      "loss": 0.1192,
+      "mean_token_accuracy": 0.9538507580757141,
+      "num_tokens": 364117.0,
+      "step": 270
+    },
+    {
+      "entropy": 1.0620019793510438,
+      "epoch": 0.26743075453677173,
+      "grad_norm": 0.6818720698356628,
+      "learning_rate": 9.81235656177206e-05,
+      "loss": 0.1533,
+      "mean_token_accuracy": 0.9445099711418152,
+      "num_tokens": 377685.0,
+      "step": 280
+    },
+    {
+      "entropy": 1.034357887506485,
+      "epoch": 0.276981852913085,
+      "grad_norm": 0.5060257315635681,
+      "learning_rate": 9.790324896411685e-05,
+      "loss": 0.1369,
+      "mean_token_accuracy": 0.9475771248340606,
+      "num_tokens": 390657.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0268417298793793,
+      "epoch": 0.28653295128939826,
+      "grad_norm": 0.7247007489204407,
+      "learning_rate": 9.767098180930781e-05,
+      "loss": 0.1178,
+      "mean_token_accuracy": 0.9561009109020233,
+      "num_tokens": 403792.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0132039785385132,
+      "epoch": 0.29608404966571156,
+      "grad_norm": 0.9478358626365662,
+      "learning_rate": 9.742682209735727e-05,
+      "loss": 0.1042,
+      "mean_token_accuracy": 0.9590364456176758,
+      "num_tokens": 416799.0,
+      "step": 310
+    },
+    {
+      "entropy": 1.0204243004322051,
+      "epoch": 0.30563514804202485,
+      "grad_norm": 0.6691247224807739,
+      "learning_rate": 9.717083073918458e-05,
+      "loss": 0.1417,
+      "mean_token_accuracy": 0.9496803998947143,
+      "num_tokens": 430010.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0182433247566223,
+      "epoch": 0.3151862464183381,
+      "grad_norm": 0.5055658221244812,
+      "learning_rate": 9.690307159736897e-05,
+      "loss": 0.1333,
+      "mean_token_accuracy": 0.9477467358112335,
+      "num_tokens": 443247.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.0100471019744872,
+      "epoch": 0.3247373447946514,
+      "grad_norm": 0.4489887058734894,
+      "learning_rate": 9.662361147021779e-05,
+      "loss": 0.1221,
+      "mean_token_accuracy": 0.952499395608902,
+      "num_tokens": 456910.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0106735706329346,
+      "epoch": 0.3342884431709647,
+      "grad_norm": 0.5227370858192444,
+      "learning_rate": 9.633252007510218e-05,
+      "loss": 0.1087,
+      "mean_token_accuracy": 0.9586474299430847,
+      "num_tokens": 470277.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.0026470124721527,
+      "epoch": 0.3438395415472779,
+      "grad_norm": 0.7523620128631592,
+      "learning_rate": 9.602987003106454e-05,
+      "loss": 0.1193,
+      "mean_token_accuracy": 0.9553453028202057,
+      "num_tokens": 483085.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.0058731734752655,
+      "epoch": 0.3533906399235912,
+      "grad_norm": 0.5894979238510132,
+      "learning_rate": 9.571573684070218e-05,
+      "loss": 0.129,
+      "mean_token_accuracy": 0.9516477942466736,
+      "num_tokens": 496647.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.0190829038619995,
+      "epoch": 0.3629417382999045,
+      "grad_norm": 0.4767792224884033,
+      "learning_rate": 9.539019887133153e-05,
+      "loss": 0.126,
+      "mean_token_accuracy": 0.9516949474811554,
+      "num_tokens": 510219.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0190945625305177,
+      "epoch": 0.37249283667621774,
+      "grad_norm": 0.5695082545280457,
+      "learning_rate": 9.505333733543778e-05,
+      "loss": 0.1046,
+      "mean_token_accuracy": 0.9579157829284668,
+      "num_tokens": 523983.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.0011855006217956,
+      "epoch": 0.38204393505253104,
+      "grad_norm": 0.5341638326644897,
+      "learning_rate": 9.470523627041452e-05,
+      "loss": 0.121,
+      "mean_token_accuracy": 0.9518589735031128,
+      "num_tokens": 537086.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.018662679195404,
+      "epoch": 0.39159503342884433,
+      "grad_norm": 0.658848226070404,
+      "learning_rate": 9.434598251759902e-05,
+      "loss": 0.1247,
+      "mean_token_accuracy": 0.9543427884578705,
+      "num_tokens": 550593.0,
+      "step": 410
+    },
+    {
+      "entropy": 1.016444206237793,
+      "epoch": 0.40114613180515757,
+      "grad_norm": 0.4661412239074707,
+      "learning_rate": 9.39756657006076e-05,
+      "loss": 0.1028,
+      "mean_token_accuracy": 0.958312326669693,
+      "num_tokens": 564272.0,
+      "step": 420
+    },
+    {
+      "entropy": 1.020843756198883,
+      "epoch": 0.41069723018147086,
+      "grad_norm": 0.6575382947921753,
+      "learning_rate": 9.359437820297717e-05,
+      "loss": 0.1183,
+      "mean_token_accuracy": 0.949735015630722,
+      "num_tokens": 577555.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.0079998195171356,
+      "epoch": 0.42024832855778416,
+      "grad_norm": 0.49395400285720825,
+      "learning_rate": 9.320221514511805e-05,
+      "loss": 0.1168,
+      "mean_token_accuracy": 0.9593334913253784,
+      "num_tokens": 591128.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0191247701644897,
+      "epoch": 0.4297994269340974,
+      "grad_norm": 0.5110730528831482,
+      "learning_rate": 9.27992743605843e-05,
+      "loss": 0.0993,
+      "mean_token_accuracy": 0.961239755153656,
+      "num_tokens": 605231.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0002938032150268,
+      "epoch": 0.4393505253104107,
+      "grad_norm": 0.6749032139778137,
+      "learning_rate": 9.23856563716669e-05,
+      "loss": 0.0981,
+      "mean_token_accuracy": 0.9605180084705353,
+      "num_tokens": 618780.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0322493851184844,
+      "epoch": 0.448901623686724,
+      "grad_norm": 0.5628225803375244,
+      "learning_rate": 9.196146436431634e-05,
+      "loss": 0.123,
+      "mean_token_accuracy": 0.9496812939643859,
+      "num_tokens": 633018.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.0046562850475311,
+      "epoch": 0.4584527220630373,
+      "grad_norm": 0.5147361755371094,
+      "learning_rate": 9.152680416240059e-05,
+      "loss": 0.1201,
+      "mean_token_accuracy": 0.9560510993003846,
+      "num_tokens": 646842.0,
+      "step": 480
+    },
+    {
+      "entropy": 1.0325502812862397,
+      "epoch": 0.4680038204393505,
+      "grad_norm": 0.4801740050315857,
+      "learning_rate": 9.108178420130513e-05,
+      "loss": 0.1262,
+      "mean_token_accuracy": 0.9481577694416046,
+      "num_tokens": 660943.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0086665809154511,
+      "epoch": 0.4775549188156638,
+      "grad_norm": 0.7685234546661377,
+      "learning_rate": 9.062651550088137e-05,
+      "loss": 0.1351,
+      "mean_token_accuracy": 0.9452983975410462,
+      "num_tokens": 674098.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.9931881070137024,
+      "epoch": 0.4871060171919771,
+      "grad_norm": 0.9793026447296143,
+      "learning_rate": 9.016111163775032e-05,
+      "loss": 0.1239,
+      "mean_token_accuracy": 0.9518612742424011,
+      "num_tokens": 688012.0,
+      "step": 510
+    },
+    {
+      "entropy": 0.9962984800338746,
+      "epoch": 0.49665711556829034,
+      "grad_norm": 1.3374274969100952,
+      "learning_rate": 8.968568871696847e-05,
+      "loss": 0.1247,
+      "mean_token_accuracy": 0.9515324234962463,
+      "num_tokens": 700886.0,
+      "step": 520
+    },
+    {
+      "entropy": 0.9801011323928833,
+      "epoch": 0.5062082139446036,
+      "grad_norm": 0.8206896781921387,
+      "learning_rate": 8.92003653430629e-05,
+      "loss": 0.1075,
+      "mean_token_accuracy": 0.9551630318164825,
+      "num_tokens": 714464.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0092791140079498,
+      "epoch": 0.5157593123209169,
+      "grad_norm": 0.5204944014549255,
+      "learning_rate": 8.870526259044274e-05,
+      "loss": 0.1127,
+      "mean_token_accuracy": 0.9540244996547699,
+      "num_tokens": 728504.0,
+      "step": 540
+    },
+    {
+      "entropy": 1.0012660682201386,
+      "epoch": 0.5253104106972302,
+      "grad_norm": 0.5105589032173157,
+      "learning_rate": 8.820050397319459e-05,
+      "loss": 0.1143,
+      "mean_token_accuracy": 0.9578450977802276,
+      "num_tokens": 741844.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.0215977787971497,
+      "epoch": 0.5348615090735435,
+      "grad_norm": 0.7014952898025513,
+      "learning_rate": 8.768621541426932e-05,
+      "loss": 0.1339,
+      "mean_token_accuracy": 0.9482286453247071,
+      "num_tokens": 756200.0,
+      "step": 560
+    },
+    {
+      "entropy": 1.004835945367813,
+      "epoch": 0.5444126074498568,
+      "grad_norm": 0.7538074851036072,
+      "learning_rate": 8.716252521406774e-05,
+      "loss": 0.1207,
+      "mean_token_accuracy": 0.953888475894928,
+      "num_tokens": 769111.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.0046703398227692,
+      "epoch": 0.55396370582617,
+      "grad_norm": 0.8112677335739136,
+      "learning_rate": 8.662956401843338e-05,
+      "loss": 0.1341,
+      "mean_token_accuracy": 0.9491826891899109,
+      "num_tokens": 782193.0,
+      "step": 580
+    },
+    {
+      "entropy": 1.0277552425861358,
+      "epoch": 0.5635148042024832,
+      "grad_norm": 0.5707908272743225,
+      "learning_rate": 8.608746478606004e-05,
+      "loss": 0.1209,
+      "mean_token_accuracy": 0.9501825332641601,
+      "num_tokens": 796312.0,
+      "step": 590
+    },
+    {
+      "entropy": 1.004393059015274,
+      "epoch": 0.5730659025787965,
+      "grad_norm": 0.6290408968925476,
+      "learning_rate": 8.553636275532236e-05,
+      "loss": 0.1138,
+      "mean_token_accuracy": 0.9592354953289032,
+      "num_tokens": 810038.0,
+      "step": 600
+    },
+    {
+      "entropy": 1.0046906530857087,
+      "epoch": 0.5826170009551098,
+      "grad_norm": 0.6663141250610352,
+      "learning_rate": 8.49763954105377e-05,
+      "loss": 0.1278,
+      "mean_token_accuracy": 0.9498465538024903,
+      "num_tokens": 823644.0,
+      "step": 610
+    },
+    {
+      "entropy": 1.0284654557704926,
+      "epoch": 0.5921680993314231,
+      "grad_norm": 0.7637591361999512,
+      "learning_rate": 8.440770244766764e-05,
+      "loss": 0.125,
+      "mean_token_accuracy": 0.9553016543388366,
+      "num_tokens": 837324.0,
+      "step": 620
+    },
+    {
+      "entropy": 1.0122312128543853,
+      "epoch": 0.6017191977077364,
+      "grad_norm": 0.8536220192909241,
+      "learning_rate": 8.383042573946792e-05,
+      "loss": 0.1267,
+      "mean_token_accuracy": 0.9486331701278686,
+      "num_tokens": 850372.0,
+      "step": 630
+    },
+    {
+      "entropy": 1.0189878702163697,
+      "epoch": 0.6112702960840497,
+      "grad_norm": 0.4578344523906708,
+      "learning_rate": 8.324470930009513e-05,
+      "loss": 0.1196,
+      "mean_token_accuracy": 0.9562380611896515,
+      "num_tokens": 863450.0,
+      "step": 640
+    },
+    {
+      "entropy": 1.001434862613678,
+      "epoch": 0.620821394460363,
+      "grad_norm": 0.576449990272522,
+      "learning_rate": 8.265069924917925e-05,
+      "loss": 0.1191,
+      "mean_token_accuracy": 0.9564612984657288,
+      "num_tokens": 877306.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.9781933188438415,
+      "epoch": 0.6303724928366762,
+      "grad_norm": 0.546006441116333,
+      "learning_rate": 8.204854377537104e-05,
+      "loss": 0.1058,
+      "mean_token_accuracy": 0.9593380630016327,
+      "num_tokens": 890647.0,
+      "step": 660
+    },
+    {
+      "entropy": 0.979134875535965,
+      "epoch": 0.6399235912129895,
+      "grad_norm": 0.4961159825325012,
+      "learning_rate": 8.143839309937307e-05,
+      "loss": 0.1157,
+      "mean_token_accuracy": 0.9545093536376953,
+      "num_tokens": 903787.0,
+      "step": 670
+    },
+    {
+      "entropy": 0.9618951916694641,
+      "epoch": 0.6494746895893028,
+      "grad_norm": 0.6587415337562561,
+      "learning_rate": 8.082039943646398e-05,
+      "loss": 0.1027,
+      "mean_token_accuracy": 0.9592947900295258,
+      "num_tokens": 917356.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9643818438053131,
+      "epoch": 0.6590257879656161,
+      "grad_norm": 0.5899077653884888,
+      "learning_rate": 8.019471695852504e-05,
+      "loss": 0.0979,
+      "mean_token_accuracy": 0.9630122303962707,
+      "num_tokens": 931499.0,
+      "step": 690
+    },
+    {
+      "entropy": 0.9757950663566589,
+      "epoch": 0.6685768863419294,
+      "grad_norm": 0.715286910533905,
+      "learning_rate": 7.95615017555788e-05,
+      "loss": 0.1283,
+      "mean_token_accuracy": 0.9490093171596528,
+      "num_tokens": 945239.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.9761663615703583,
+      "epoch": 0.6781279847182426,
+      "grad_norm": 0.6218839883804321,
+      "learning_rate": 7.892091179684875e-05,
+      "loss": 0.1167,
+      "mean_token_accuracy": 0.9514663696289063,
+      "num_tokens": 958684.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9715986967086792,
+      "epoch": 0.6876790830945558,
+      "grad_norm": 0.5946614146232605,
+      "learning_rate": 7.82731068913508e-05,
+      "loss": 0.1093,
+      "mean_token_accuracy": 0.9581069588661194,
+      "num_tokens": 972138.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9765264689922333,
+      "epoch": 0.6972301814708691,
+      "grad_norm": 0.48247581720352173,
+      "learning_rate": 7.76182486480253e-05,
+      "loss": 0.1157,
+      "mean_token_accuracy": 0.9566311240196228,
+      "num_tokens": 985633.0,
+      "step": 730
+    },
+    {
+      "entropy": 0.96493079662323,
+      "epoch": 0.7067812798471824,
+      "grad_norm": 0.6061577200889587,
+      "learning_rate": 7.69565004354202e-05,
+      "loss": 0.123,
+      "mean_token_accuracy": 0.955526077747345,
+      "num_tokens": 998825.0,
+      "step": 740
+    },
+    {
+      "entropy": 0.9701172232627868,
+      "epoch": 0.7163323782234957,
+      "grad_norm": 0.4007874131202698,
+      "learning_rate": 7.628802734093527e-05,
+      "loss": 0.1135,
+      "mean_token_accuracy": 0.9551189005374908,
+      "num_tokens": 1011804.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.9524923682212829,
+      "epoch": 0.725883476599809,
+      "grad_norm": 0.7175698280334473,
+      "learning_rate": 7.56129961296375e-05,
+      "loss": 0.1009,
+      "mean_token_accuracy": 0.9648522973060608,
+      "num_tokens": 1026043.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9468290984630585,
+      "epoch": 0.7354345749761223,
+      "grad_norm": 0.4742600619792938,
+      "learning_rate": 7.493157520265786e-05,
+      "loss": 0.1093,
+      "mean_token_accuracy": 0.9574300646781921,
+      "num_tokens": 1039649.0,
+      "step": 770
+    },
+    {
+      "entropy": 0.9390150427818298,
+      "epoch": 0.7449856733524355,
+      "grad_norm": 0.7037758231163025,
+      "learning_rate": 7.424393455518011e-05,
+      "loss": 0.1076,
+      "mean_token_accuracy": 0.9584161818027497,
+      "num_tokens": 1053317.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.952883493900299,
+      "epoch": 0.7545367717287488,
+      "grad_norm": 0.670464813709259,
+      "learning_rate": 7.355024573403174e-05,
+      "loss": 0.114,
+      "mean_token_accuracy": 0.9560878932476043,
+      "num_tokens": 1066478.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9607491433620453,
+      "epoch": 0.7640878701050621,
+      "grad_norm": 0.5118470788002014,
+      "learning_rate": 7.285068179488794e-05,
+      "loss": 0.1068,
+      "mean_token_accuracy": 0.9582593500614166,
+      "num_tokens": 1080474.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.9231133222579956,
+      "epoch": 0.7736389684813754,
+      "grad_norm": 0.3679278492927551,
+      "learning_rate": 7.214541725909909e-05,
+      "loss": 0.0951,
+      "mean_token_accuracy": 0.9623704373836517,
+      "num_tokens": 1093980.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.9327525436878205,
+      "epoch": 0.7831900668576887,
+      "grad_norm": 0.687837541103363,
+      "learning_rate": 7.143462807015271e-05,
+      "loss": 0.1126,
+      "mean_token_accuracy": 0.9560571551322937,
+      "num_tokens": 1106996.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9433809757232666,
+      "epoch": 0.792741165234002,
+      "grad_norm": 0.7366920113563538,
+      "learning_rate": 7.071849154978048e-05,
+      "loss": 0.1054,
+      "mean_token_accuracy": 0.9596803724765778,
+      "num_tokens": 1120972.0,
+      "step": 830
+    },
+    {
+      "entropy": 0.9286311149597168,
+      "epoch": 0.8022922636103151,
+      "grad_norm": 0.6273325085639954,
+      "learning_rate": 6.999718635372148e-05,
+      "loss": 0.1051,
+      "mean_token_accuracy": 0.9585224330425263,
+      "num_tokens": 1134881.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9196918785572052,
+      "epoch": 0.8118433619866284,
+      "grad_norm": 0.7454337477684021,
+      "learning_rate": 6.927089242715277e-05,
+      "loss": 0.0996,
+      "mean_token_accuracy": 0.963817548751831,
+      "num_tokens": 1148338.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.953073400259018,
+      "epoch": 0.8213944603629417,
+      "grad_norm": 0.4553161561489105,
+      "learning_rate": 6.853979095979804e-05,
+      "loss": 0.0964,
+      "mean_token_accuracy": 0.9651101708412171,
+      "num_tokens": 1162724.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9304056465625763,
+      "epoch": 0.830945558739255,
+      "grad_norm": 0.49882224202156067,
+      "learning_rate": 6.780406434072603e-05,
+      "loss": 0.0931,
+      "mean_token_accuracy": 0.9641244173049927,
+      "num_tokens": 1176245.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.9313814461231231,
+      "epoch": 0.8404966571155683,
+      "grad_norm": 0.7891941666603088,
+      "learning_rate": 6.706389611284954e-05,
+      "loss": 0.1006,
+      "mean_token_accuracy": 0.9584587395191193,
+      "num_tokens": 1189996.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9537931561470032,
+      "epoch": 0.8500477554918816,
+      "grad_norm": 0.7120227217674255,
+      "learning_rate": 6.63194709271367e-05,
+      "loss": 0.1082,
+      "mean_token_accuracy": 0.9575189769268035,
+      "num_tokens": 1203136.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9555979907512665,
+      "epoch": 0.8595988538681948,
+      "grad_norm": 0.5703518986701965,
+      "learning_rate": 6.557097449654587e-05,
+      "loss": 0.1077,
+      "mean_token_accuracy": 0.9579966783523559,
+      "num_tokens": 1216913.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9558742046356201,
+      "epoch": 0.8691499522445081,
+      "grad_norm": 0.7904860377311707,
+      "learning_rate": 6.481859354969548e-05,
+      "loss": 0.1001,
+      "mean_token_accuracy": 0.9613328218460083,
+      "num_tokens": 1229969.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.9618341326713562,
+      "epoch": 0.8787010506208214,
+      "grad_norm": 1.1422725915908813,
+      "learning_rate": 6.406251578428046e-05,
+      "loss": 0.1043,
+      "mean_token_accuracy": 0.9618310868740082,
+      "num_tokens": 1243680.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9476982057094574,
+      "epoch": 0.8882521489971347,
+      "grad_norm": 0.6227033734321594,
+      "learning_rate": 6.330292982024712e-05,
+      "loss": 0.1004,
+      "mean_token_accuracy": 0.9570560812950134,
+      "num_tokens": 1257573.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.9620028853416442,
+      "epoch": 0.897803247373448,
+      "grad_norm": 0.5102617144584656,
+      "learning_rate": 6.254002515273774e-05,
+      "loss": 0.0976,
+      "mean_token_accuracy": 0.9597377300262451,
+      "num_tokens": 1270701.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9383984684944153,
+      "epoch": 0.9073543457497613,
+      "grad_norm": 0.7831044793128967,
+      "learning_rate": 6.177399210481698e-05,
+      "loss": 0.1014,
+      "mean_token_accuracy": 0.9597761750221252,
+      "num_tokens": 1283809.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.9364366352558136,
+      "epoch": 0.9169054441260746,
+      "grad_norm": 0.6067343950271606,
+      "learning_rate": 6.100502177999156e-05,
+      "loss": 0.0935,
+      "mean_token_accuracy": 0.9624200344085694,
+      "num_tokens": 1297005.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.976334136724472,
+      "epoch": 0.9264565425023877,
+      "grad_norm": 0.7300851345062256,
+      "learning_rate": 6.023330601453551e-05,
+      "loss": 0.126,
+      "mean_token_accuracy": 0.9492876768112183,
+      "num_tokens": 1310830.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9701994717121124,
+      "epoch": 0.936007640878701,
+      "grad_norm": 0.7335929274559021,
+      "learning_rate": 5.945903732963234e-05,
+      "loss": 0.103,
+      "mean_token_accuracy": 0.9596088886260986,
+      "num_tokens": 1324145.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.9694394767284393,
+      "epoch": 0.9455587392550143,
+      "grad_norm": 0.674283504486084,
+      "learning_rate": 5.868240888334653e-05,
+      "loss": 0.1007,
+      "mean_token_accuracy": 0.9569411039352417,
+      "num_tokens": 1337718.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.9264466464519501,
+      "epoch": 0.9551098376313276,
+      "grad_norm": 0.7404454350471497,
+      "learning_rate": 5.790361442243605e-05,
+      "loss": 0.091,
+      "mean_token_accuracy": 0.9658930003643036,
+      "num_tokens": 1351556.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9517737925052643,
+      "epoch": 0.9646609360076409,
+      "grad_norm": 0.6310734748840332,
+      "learning_rate": 5.712284823401808e-05,
+      "loss": 0.1055,
+      "mean_token_accuracy": 0.960016405582428,
+      "num_tokens": 1364817.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.9591375946998596,
+      "epoch": 0.9742120343839542,
+      "grad_norm": 0.4866735637187958,
+      "learning_rate": 5.634030509709982e-05,
+      "loss": 0.0928,
+      "mean_token_accuracy": 0.962578970193863,
+      "num_tokens": 1378098.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9355690777301788,
+      "epoch": 0.9837631327602674,
+      "grad_norm": 0.7113921642303467,
+      "learning_rate": 5.555618023398671e-05,
+      "loss": 0.0933,
+      "mean_token_accuracy": 0.9603404223918914,
+      "num_tokens": 1391374.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9405025899410248,
+      "epoch": 0.9933142311365807,
+      "grad_norm": 0.8997142910957336,
+      "learning_rate": 5.477066926158001e-05,
+      "loss": 0.1127,
+      "mean_token_accuracy": 0.9545972764492034,
+      "num_tokens": 1404715.0,
+      "step": 1040
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2094,
+  "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_s2/checkpoint-1047/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f189c2709f177b4103ed0f4903d53255b3106698718586cb13af342600c1cc9d
+size 6417

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2/checkpoint-2094/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_s2/checkpoint-2094/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_s2/checkpoint-2094/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f99441730d83f72e89d2d6192941df30b993fb4a773132a27f9f413b3914a2dc
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2/checkpoint-2094/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_s2/checkpoint-2094/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_s2/checkpoint-2094/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_s2/checkpoint-2094/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_s2/checkpoint-2094/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_s2/checkpoint-2094/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": 2094,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.1833186984062194,
+      "epoch": 0.009551098376313277,
+      "grad_norm": 0.6022069454193115,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2438,
+      "mean_token_accuracy": 0.9158181607723236,
+      "num_tokens": 13654.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1723507642745972,
+      "epoch": 0.019102196752626553,
+      "grad_norm": 0.6949854493141174,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2739,
+      "mean_token_accuracy": 0.9052117764949799,
+      "num_tokens": 27095.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1735804200172424,
+      "epoch": 0.02865329512893983,
+      "grad_norm": 0.4342527687549591,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.2216,
+      "mean_token_accuracy": 0.9295729756355285,
+      "num_tokens": 41099.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.1678385138511658,
+      "epoch": 0.038204393505253106,
+      "grad_norm": 0.6038743257522583,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.2096,
+      "mean_token_accuracy": 0.9272670745849609,
+      "num_tokens": 54788.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.1570314407348632,
+      "epoch": 0.04775549188156638,
+      "grad_norm": 0.6725504398345947,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.1795,
+      "mean_token_accuracy": 0.9398179113864898,
+      "num_tokens": 68058.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.1301384329795838,
+      "epoch": 0.05730659025787966,
+      "grad_norm": 0.6255244016647339,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1811,
+      "mean_token_accuracy": 0.9348016798496246,
+      "num_tokens": 80974.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.1043856620788575,
+      "epoch": 0.06685768863419293,
+      "grad_norm": 0.7748627066612244,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1828,
+      "mean_token_accuracy": 0.9330100417137146,
+      "num_tokens": 94504.0,
+      "step": 70
+    },
+    {
+      "entropy": 1.068319034576416,
+      "epoch": 0.07640878701050621,
+      "grad_norm": 0.5178515911102295,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1491,
+      "mean_token_accuracy": 0.9445622324943542,
+      "num_tokens": 107984.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.0800583720207215,
+      "epoch": 0.08595988538681948,
+      "grad_norm": 0.7653284072875977,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1797,
+      "mean_token_accuracy": 0.9287929236888885,
+      "num_tokens": 122054.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0604823708534241,
+      "epoch": 0.09551098376313276,
+      "grad_norm": 0.43810370564460754,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1588,
+      "mean_token_accuracy": 0.940058171749115,
+      "num_tokens": 134756.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0564072370529174,
+      "epoch": 0.10506208213944604,
+      "grad_norm": 0.7842755913734436,
+      "learning_rate": 9.999900209608641e-05,
+      "loss": 0.1602,
+      "mean_token_accuracy": 0.9440609395503998,
+      "num_tokens": 147791.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.0383123695850371,
+      "epoch": 0.11461318051575932,
+      "grad_norm": 0.6811287999153137,
+      "learning_rate": 9.998777613450479e-05,
+      "loss": 0.1512,
+      "mean_token_accuracy": 0.948225998878479,
+      "num_tokens": 161064.0,
+      "step": 120
+    },
+    {
+      "entropy": 1.0554516851902007,
+      "epoch": 0.12416427889207259,
+      "grad_norm": 0.5387641191482544,
+      "learning_rate": 9.996407964134416e-05,
+      "loss": 0.147,
+      "mean_token_accuracy": 0.9459446430206299,
+      "num_tokens": 175207.0,
+      "step": 130
+    },
+    {
+      "entropy": 1.0415111005306243,
+      "epoch": 0.13371537726838587,
+      "grad_norm": 0.6672711968421936,
+      "learning_rate": 9.992791852820709e-05,
+      "loss": 0.1413,
+      "mean_token_accuracy": 0.9465783596038818,
+      "num_tokens": 189210.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0498008966445922,
+      "epoch": 0.14326647564469913,
+      "grad_norm": 0.6177133917808533,
+      "learning_rate": 9.987930181626515e-05,
+      "loss": 0.1589,
+      "mean_token_accuracy": 0.9427510976791382,
+      "num_tokens": 203140.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.9933426916599274,
+      "epoch": 0.15281757402101243,
+      "grad_norm": 0.642832338809967,
+      "learning_rate": 9.981824163400827e-05,
+      "loss": 0.1457,
+      "mean_token_accuracy": 0.9474036812782287,
+      "num_tokens": 216677.0,
+      "step": 160
+    },
+    {
+      "entropy": 0.9938121736049652,
+      "epoch": 0.1623686723973257,
+      "grad_norm": 0.62694251537323,
+      "learning_rate": 9.974475321421917e-05,
+      "loss": 0.1158,
+      "mean_token_accuracy": 0.9580030381679535,
+      "num_tokens": 230282.0,
+      "step": 170
+    },
+    {
+      "entropy": 0.987104994058609,
+      "epoch": 0.17191977077363896,
+      "grad_norm": 0.6699346899986267,
+      "learning_rate": 9.96588548901731e-05,
+      "loss": 0.1315,
+      "mean_token_accuracy": 0.9509034395217896,
+      "num_tokens": 244200.0,
+      "step": 180
+    },
+    {
+      "entropy": 0.9956943809986114,
+      "epoch": 0.18147086914995225,
+      "grad_norm": 0.7168964147567749,
+      "learning_rate": 9.956056809106426e-05,
+      "loss": 0.142,
+      "mean_token_accuracy": 0.9486550152301788,
+      "num_tokens": 257325.0,
+      "step": 190
+    },
+    {
+      "entropy": 1.0101079881191253,
+      "epoch": 0.19102196752626552,
+      "grad_norm": 0.6116616725921631,
+      "learning_rate": 9.944991733665984e-05,
+      "loss": 0.1492,
+      "mean_token_accuracy": 0.9417937338352204,
+      "num_tokens": 270610.0,
+      "step": 200
+    },
+    {
+      "entropy": 1.011151373386383,
+      "epoch": 0.20057306590257878,
+      "grad_norm": 0.7683250308036804,
+      "learning_rate": 9.932693023118298e-05,
+      "loss": 0.1529,
+      "mean_token_accuracy": 0.943734323978424,
+      "num_tokens": 284298.0,
+      "step": 210
+    },
+    {
+      "entropy": 1.0031667947769165,
+      "epoch": 0.21012416427889208,
+      "grad_norm": 0.5416099429130554,
+      "learning_rate": 9.919163745642633e-05,
+      "loss": 0.1419,
+      "mean_token_accuracy": 0.9483771979808807,
+      "num_tokens": 297632.0,
+      "step": 220
+    },
+    {
+      "entropy": 0.9910764396190643,
+      "epoch": 0.21967526265520534,
+      "grad_norm": 0.5624724626541138,
+      "learning_rate": 9.904407276409779e-05,
+      "loss": 0.134,
+      "mean_token_accuracy": 0.9472556591033936,
+      "num_tokens": 310948.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.008888155221939,
+      "epoch": 0.22922636103151864,
+      "grad_norm": 0.4665639400482178,
+      "learning_rate": 9.888427296740052e-05,
+      "loss": 0.1352,
+      "mean_token_accuracy": 0.9437968492507934,
+      "num_tokens": 324190.0,
+      "step": 240
+    },
+    {
+      "entropy": 1.0211431682109833,
+      "epoch": 0.2387774594078319,
+      "grad_norm": 0.5715168118476868,
+      "learning_rate": 9.871227793184892e-05,
+      "loss": 0.1253,
+      "mean_token_accuracy": 0.9502226531505584,
+      "num_tokens": 337357.0,
+      "step": 250
+    },
+    {
+      "entropy": 1.051475179195404,
+      "epoch": 0.24832855778414517,
+      "grad_norm": 0.8343601226806641,
+      "learning_rate": 9.852813056532343e-05,
+      "loss": 0.1431,
+      "mean_token_accuracy": 0.9473239004611969,
+      "num_tokens": 350315.0,
+      "step": 260
+    },
+    {
+      "entropy": 1.059506893157959,
+      "epoch": 0.25787965616045844,
+      "grad_norm": 0.5191280841827393,
+      "learning_rate": 9.833187680736625e-05,
+      "loss": 0.1192,
+      "mean_token_accuracy": 0.9538507580757141,
+      "num_tokens": 364117.0,
+      "step": 270
+    },
+    {
+      "entropy": 1.0620019793510438,
+      "epoch": 0.26743075453677173,
+      "grad_norm": 0.6818720698356628,
+      "learning_rate": 9.81235656177206e-05,
+      "loss": 0.1533,
+      "mean_token_accuracy": 0.9445099711418152,
+      "num_tokens": 377685.0,
+      "step": 280
+    },
+    {
+      "entropy": 1.034357887506485,
+      "epoch": 0.276981852913085,
+      "grad_norm": 0.5060257315635681,
+      "learning_rate": 9.790324896411685e-05,
+      "loss": 0.1369,
+      "mean_token_accuracy": 0.9475771248340606,
+      "num_tokens": 390657.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0268417298793793,
+      "epoch": 0.28653295128939826,
+      "grad_norm": 0.7247007489204407,
+      "learning_rate": 9.767098180930781e-05,
+      "loss": 0.1178,
+      "mean_token_accuracy": 0.9561009109020233,
+      "num_tokens": 403792.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0132039785385132,
+      "epoch": 0.29608404966571156,
+      "grad_norm": 0.9478358626365662,
+      "learning_rate": 9.742682209735727e-05,
+      "loss": 0.1042,
+      "mean_token_accuracy": 0.9590364456176758,
+      "num_tokens": 416799.0,
+      "step": 310
+    },
+    {
+      "entropy": 1.0204243004322051,
+      "epoch": 0.30563514804202485,
+      "grad_norm": 0.6691247224807739,
+      "learning_rate": 9.717083073918458e-05,
+      "loss": 0.1417,
+      "mean_token_accuracy": 0.9496803998947143,
+      "num_tokens": 430010.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0182433247566223,
+      "epoch": 0.3151862464183381,
+      "grad_norm": 0.5055658221244812,
+      "learning_rate": 9.690307159736897e-05,
+      "loss": 0.1333,
+      "mean_token_accuracy": 0.9477467358112335,
+      "num_tokens": 443247.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.0100471019744872,
+      "epoch": 0.3247373447946514,
+      "grad_norm": 0.4489887058734894,
+      "learning_rate": 9.662361147021779e-05,
+      "loss": 0.1221,
+      "mean_token_accuracy": 0.952499395608902,
+      "num_tokens": 456910.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0106735706329346,
+      "epoch": 0.3342884431709647,
+      "grad_norm": 0.5227370858192444,
+      "learning_rate": 9.633252007510218e-05,
+      "loss": 0.1087,
+      "mean_token_accuracy": 0.9586474299430847,
+      "num_tokens": 470277.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.0026470124721527,
+      "epoch": 0.3438395415472779,
+      "grad_norm": 0.7523620128631592,
+      "learning_rate": 9.602987003106454e-05,
+      "loss": 0.1193,
+      "mean_token_accuracy": 0.9553453028202057,
+      "num_tokens": 483085.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.0058731734752655,
+      "epoch": 0.3533906399235912,
+      "grad_norm": 0.5894979238510132,
+      "learning_rate": 9.571573684070218e-05,
+      "loss": 0.129,
+      "mean_token_accuracy": 0.9516477942466736,
+      "num_tokens": 496647.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.0190829038619995,
+      "epoch": 0.3629417382999045,
+      "grad_norm": 0.4767792224884033,
+      "learning_rate": 9.539019887133153e-05,
+      "loss": 0.126,
+      "mean_token_accuracy": 0.9516949474811554,
+      "num_tokens": 510219.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0190945625305177,
+      "epoch": 0.37249283667621774,
+      "grad_norm": 0.5695082545280457,
+      "learning_rate": 9.505333733543778e-05,
+      "loss": 0.1046,
+      "mean_token_accuracy": 0.9579157829284668,
+      "num_tokens": 523983.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.0011855006217956,
+      "epoch": 0.38204393505253104,
+      "grad_norm": 0.5341638326644897,
+      "learning_rate": 9.470523627041452e-05,
+      "loss": 0.121,
+      "mean_token_accuracy": 0.9518589735031128,
+      "num_tokens": 537086.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.018662679195404,
+      "epoch": 0.39159503342884433,
+      "grad_norm": 0.658848226070404,
+      "learning_rate": 9.434598251759902e-05,
+      "loss": 0.1247,
+      "mean_token_accuracy": 0.9543427884578705,
+      "num_tokens": 550593.0,
+      "step": 410
+    },
+    {
+      "entropy": 1.016444206237793,
+      "epoch": 0.40114613180515757,
+      "grad_norm": 0.4661412239074707,
+      "learning_rate": 9.39756657006076e-05,
+      "loss": 0.1028,
+      "mean_token_accuracy": 0.958312326669693,
+      "num_tokens": 564272.0,
+      "step": 420
+    },
+    {
+      "entropy": 1.020843756198883,
+      "epoch": 0.41069723018147086,
+      "grad_norm": 0.6575382947921753,
+      "learning_rate": 9.359437820297717e-05,
+      "loss": 0.1183,
+      "mean_token_accuracy": 0.949735015630722,
+      "num_tokens": 577555.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.0079998195171356,
+      "epoch": 0.42024832855778416,
+      "grad_norm": 0.49395400285720825,
+      "learning_rate": 9.320221514511805e-05,
+      "loss": 0.1168,
+      "mean_token_accuracy": 0.9593334913253784,
+      "num_tokens": 591128.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0191247701644897,
+      "epoch": 0.4297994269340974,
+      "grad_norm": 0.5110730528831482,
+      "learning_rate": 9.27992743605843e-05,
+      "loss": 0.0993,
+      "mean_token_accuracy": 0.961239755153656,
+      "num_tokens": 605231.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0002938032150268,
+      "epoch": 0.4393505253104107,
+      "grad_norm": 0.6749032139778137,
+      "learning_rate": 9.23856563716669e-05,
+      "loss": 0.0981,
+      "mean_token_accuracy": 0.9605180084705353,
+      "num_tokens": 618780.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0322493851184844,
+      "epoch": 0.448901623686724,
+      "grad_norm": 0.5628225803375244,
+      "learning_rate": 9.196146436431634e-05,
+      "loss": 0.123,
+      "mean_token_accuracy": 0.9496812939643859,
+      "num_tokens": 633018.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.0046562850475311,
+      "epoch": 0.4584527220630373,
+      "grad_norm": 0.5147361755371094,
+      "learning_rate": 9.152680416240059e-05,
+      "loss": 0.1201,
+      "mean_token_accuracy": 0.9560510993003846,
+      "num_tokens": 646842.0,
+      "step": 480
+    },
+    {
+      "entropy": 1.0325502812862397,
+      "epoch": 0.4680038204393505,
+      "grad_norm": 0.4801740050315857,
+      "learning_rate": 9.108178420130513e-05,
+      "loss": 0.1262,
+      "mean_token_accuracy": 0.9481577694416046,
+      "num_tokens": 660943.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0086665809154511,
+      "epoch": 0.4775549188156638,
+      "grad_norm": 0.7685234546661377,
+      "learning_rate": 9.062651550088137e-05,
+      "loss": 0.1351,
+      "mean_token_accuracy": 0.9452983975410462,
+      "num_tokens": 674098.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.9931881070137024,
+      "epoch": 0.4871060171919771,
+      "grad_norm": 0.9793026447296143,
+      "learning_rate": 9.016111163775032e-05,
+      "loss": 0.1239,
+      "mean_token_accuracy": 0.9518612742424011,
+      "num_tokens": 688012.0,
+      "step": 510
+    },
+    {
+      "entropy": 0.9962984800338746,
+      "epoch": 0.49665711556829034,
+      "grad_norm": 1.3374274969100952,
+      "learning_rate": 8.968568871696847e-05,
+      "loss": 0.1247,
+      "mean_token_accuracy": 0.9515324234962463,
+      "num_tokens": 700886.0,
+      "step": 520
+    },
+    {
+      "entropy": 0.9801011323928833,
+      "epoch": 0.5062082139446036,
+      "grad_norm": 0.8206896781921387,
+      "learning_rate": 8.92003653430629e-05,
+      "loss": 0.1075,
+      "mean_token_accuracy": 0.9551630318164825,
+      "num_tokens": 714464.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0092791140079498,
+      "epoch": 0.5157593123209169,
+      "grad_norm": 0.5204944014549255,
+      "learning_rate": 8.870526259044274e-05,
+      "loss": 0.1127,
+      "mean_token_accuracy": 0.9540244996547699,
+      "num_tokens": 728504.0,
+      "step": 540
+    },
+    {
+      "entropy": 1.0012660682201386,
+      "epoch": 0.5253104106972302,
+      "grad_norm": 0.5105589032173157,
+      "learning_rate": 8.820050397319459e-05,
+      "loss": 0.1143,
+      "mean_token_accuracy": 0.9578450977802276,
+      "num_tokens": 741844.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.0215977787971497,
+      "epoch": 0.5348615090735435,
+      "grad_norm": 0.7014952898025513,
+      "learning_rate": 8.768621541426932e-05,
+      "loss": 0.1339,
+      "mean_token_accuracy": 0.9482286453247071,
+      "num_tokens": 756200.0,
+      "step": 560
+    },
+    {
+      "entropy": 1.004835945367813,
+      "epoch": 0.5444126074498568,
+      "grad_norm": 0.7538074851036072,
+      "learning_rate": 8.716252521406774e-05,
+      "loss": 0.1207,
+      "mean_token_accuracy": 0.953888475894928,
+      "num_tokens": 769111.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.0046703398227692,
+      "epoch": 0.55396370582617,
+      "grad_norm": 0.8112677335739136,
+      "learning_rate": 8.662956401843338e-05,
+      "loss": 0.1341,
+      "mean_token_accuracy": 0.9491826891899109,
+      "num_tokens": 782193.0,
+      "step": 580
+    },
+    {
+      "entropy": 1.0277552425861358,
+      "epoch": 0.5635148042024832,
+      "grad_norm": 0.5707908272743225,
+      "learning_rate": 8.608746478606004e-05,
+      "loss": 0.1209,
+      "mean_token_accuracy": 0.9501825332641601,
+      "num_tokens": 796312.0,
+      "step": 590
+    },
+    {
+      "entropy": 1.004393059015274,
+      "epoch": 0.5730659025787965,
+      "grad_norm": 0.6290408968925476,
+      "learning_rate": 8.553636275532236e-05,
+      "loss": 0.1138,
+      "mean_token_accuracy": 0.9592354953289032,
+      "num_tokens": 810038.0,
+      "step": 600
+    },
+    {
+      "entropy": 1.0046906530857087,
+      "epoch": 0.5826170009551098,
+      "grad_norm": 0.6663141250610352,
+      "learning_rate": 8.49763954105377e-05,
+      "loss": 0.1278,
+      "mean_token_accuracy": 0.9498465538024903,
+      "num_tokens": 823644.0,
+      "step": 610
+    },
+    {
+      "entropy": 1.0284654557704926,
+      "epoch": 0.5921680993314231,
+      "grad_norm": 0.7637591361999512,
+      "learning_rate": 8.440770244766764e-05,
+      "loss": 0.125,
+      "mean_token_accuracy": 0.9553016543388366,
+      "num_tokens": 837324.0,
+      "step": 620
+    },
+    {
+      "entropy": 1.0122312128543853,
+      "epoch": 0.6017191977077364,
+      "grad_norm": 0.8536220192909241,
+      "learning_rate": 8.383042573946792e-05,
+      "loss": 0.1267,
+      "mean_token_accuracy": 0.9486331701278686,
+      "num_tokens": 850372.0,
+      "step": 630
+    },
+    {
+      "entropy": 1.0189878702163697,
+      "epoch": 0.6112702960840497,
+      "grad_norm": 0.4578344523906708,
+      "learning_rate": 8.324470930009513e-05,
+      "loss": 0.1196,
+      "mean_token_accuracy": 0.9562380611896515,
+      "num_tokens": 863450.0,
+      "step": 640
+    },
+    {
+      "entropy": 1.001434862613678,
+      "epoch": 0.620821394460363,
+      "grad_norm": 0.576449990272522,
+      "learning_rate": 8.265069924917925e-05,
+      "loss": 0.1191,
+      "mean_token_accuracy": 0.9564612984657288,
+      "num_tokens": 877306.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.9781933188438415,
+      "epoch": 0.6303724928366762,
+      "grad_norm": 0.546006441116333,
+      "learning_rate": 8.204854377537104e-05,
+      "loss": 0.1058,
+      "mean_token_accuracy": 0.9593380630016327,
+      "num_tokens": 890647.0,
+      "step": 660
+    },
+    {
+      "entropy": 0.979134875535965,
+      "epoch": 0.6399235912129895,
+      "grad_norm": 0.4961159825325012,
+      "learning_rate": 8.143839309937307e-05,
+      "loss": 0.1157,
+      "mean_token_accuracy": 0.9545093536376953,
+      "num_tokens": 903787.0,
+      "step": 670
+    },
+    {
+      "entropy": 0.9618951916694641,
+      "epoch": 0.6494746895893028,
+      "grad_norm": 0.6587415337562561,
+      "learning_rate": 8.082039943646398e-05,
+      "loss": 0.1027,
+      "mean_token_accuracy": 0.9592947900295258,
+      "num_tokens": 917356.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9643818438053131,
+      "epoch": 0.6590257879656161,
+      "grad_norm": 0.5899077653884888,
+      "learning_rate": 8.019471695852504e-05,
+      "loss": 0.0979,
+      "mean_token_accuracy": 0.9630122303962707,
+      "num_tokens": 931499.0,
+      "step": 690
+    },
+    {
+      "entropy": 0.9757950663566589,
+      "epoch": 0.6685768863419294,
+      "grad_norm": 0.715286910533905,
+      "learning_rate": 7.95615017555788e-05,
+      "loss": 0.1283,
+      "mean_token_accuracy": 0.9490093171596528,
+      "num_tokens": 945239.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.9761663615703583,
+      "epoch": 0.6781279847182426,
+      "grad_norm": 0.6218839883804321,
+      "learning_rate": 7.892091179684875e-05,
+      "loss": 0.1167,
+      "mean_token_accuracy": 0.9514663696289063,
+      "num_tokens": 958684.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9715986967086792,
+      "epoch": 0.6876790830945558,
+      "grad_norm": 0.5946614146232605,
+      "learning_rate": 7.82731068913508e-05,
+      "loss": 0.1093,
+      "mean_token_accuracy": 0.9581069588661194,
+      "num_tokens": 972138.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9765264689922333,
+      "epoch": 0.6972301814708691,
+      "grad_norm": 0.48247581720352173,
+      "learning_rate": 7.76182486480253e-05,
+      "loss": 0.1157,
+      "mean_token_accuracy": 0.9566311240196228,
+      "num_tokens": 985633.0,
+      "step": 730
+    },
+    {
+      "entropy": 0.96493079662323,
+      "epoch": 0.7067812798471824,
+      "grad_norm": 0.6061577200889587,
+      "learning_rate": 7.69565004354202e-05,
+      "loss": 0.123,
+      "mean_token_accuracy": 0.955526077747345,
+      "num_tokens": 998825.0,
+      "step": 740
+    },
+    {
+      "entropy": 0.9701172232627868,
+      "epoch": 0.7163323782234957,
+      "grad_norm": 0.4007874131202698,
+      "learning_rate": 7.628802734093527e-05,
+      "loss": 0.1135,
+      "mean_token_accuracy": 0.9551189005374908,
+      "num_tokens": 1011804.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.9524923682212829,
+      "epoch": 0.725883476599809,
+      "grad_norm": 0.7175698280334473,
+      "learning_rate": 7.56129961296375e-05,
+      "loss": 0.1009,
+      "mean_token_accuracy": 0.9648522973060608,
+      "num_tokens": 1026043.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9468290984630585,
+      "epoch": 0.7354345749761223,
+      "grad_norm": 0.4742600619792938,
+      "learning_rate": 7.493157520265786e-05,
+      "loss": 0.1093,
+      "mean_token_accuracy": 0.9574300646781921,
+      "num_tokens": 1039649.0,
+      "step": 770
+    },
+    {
+      "entropy": 0.9390150427818298,
+      "epoch": 0.7449856733524355,
+      "grad_norm": 0.7037758231163025,
+      "learning_rate": 7.424393455518011e-05,
+      "loss": 0.1076,
+      "mean_token_accuracy": 0.9584161818027497,
+      "num_tokens": 1053317.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.952883493900299,
+      "epoch": 0.7545367717287488,
+      "grad_norm": 0.670464813709259,
+      "learning_rate": 7.355024573403174e-05,
+      "loss": 0.114,
+      "mean_token_accuracy": 0.9560878932476043,
+      "num_tokens": 1066478.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9607491433620453,
+      "epoch": 0.7640878701050621,
+      "grad_norm": 0.5118470788002014,
+      "learning_rate": 7.285068179488794e-05,
+      "loss": 0.1068,
+      "mean_token_accuracy": 0.9582593500614166,
+      "num_tokens": 1080474.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.9231133222579956,
+      "epoch": 0.7736389684813754,
+      "grad_norm": 0.3679278492927551,
+      "learning_rate": 7.214541725909909e-05,
+      "loss": 0.0951,
+      "mean_token_accuracy": 0.9623704373836517,
+      "num_tokens": 1093980.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.9327525436878205,
+      "epoch": 0.7831900668576887,
+      "grad_norm": 0.687837541103363,
+      "learning_rate": 7.143462807015271e-05,
+      "loss": 0.1126,
+      "mean_token_accuracy": 0.9560571551322937,
+      "num_tokens": 1106996.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9433809757232666,
+      "epoch": 0.792741165234002,
+      "grad_norm": 0.7366920113563538,
+      "learning_rate": 7.071849154978048e-05,
+      "loss": 0.1054,
+      "mean_token_accuracy": 0.9596803724765778,
+      "num_tokens": 1120972.0,
+      "step": 830
+    },
+    {
+      "entropy": 0.9286311149597168,
+      "epoch": 0.8022922636103151,
+      "grad_norm": 0.6273325085639954,
+      "learning_rate": 6.999718635372148e-05,
+      "loss": 0.1051,
+      "mean_token_accuracy": 0.9585224330425263,
+      "num_tokens": 1134881.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9196918785572052,
+      "epoch": 0.8118433619866284,
+      "grad_norm": 0.7454337477684021,
+      "learning_rate": 6.927089242715277e-05,
+      "loss": 0.0996,
+      "mean_token_accuracy": 0.963817548751831,
+      "num_tokens": 1148338.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.953073400259018,
+      "epoch": 0.8213944603629417,
+      "grad_norm": 0.4553161561489105,
+      "learning_rate": 6.853979095979804e-05,
+      "loss": 0.0964,
+      "mean_token_accuracy": 0.9651101708412171,
+      "num_tokens": 1162724.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9304056465625763,
+      "epoch": 0.830945558739255,
+      "grad_norm": 0.49882224202156067,
+      "learning_rate": 6.780406434072603e-05,
+      "loss": 0.0931,
+      "mean_token_accuracy": 0.9641244173049927,
+      "num_tokens": 1176245.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.9313814461231231,
+      "epoch": 0.8404966571155683,
+      "grad_norm": 0.7891941666603088,
+      "learning_rate": 6.706389611284954e-05,
+      "loss": 0.1006,
+      "mean_token_accuracy": 0.9584587395191193,
+      "num_tokens": 1189996.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9537931561470032,
+      "epoch": 0.8500477554918816,
+      "grad_norm": 0.7120227217674255,
+      "learning_rate": 6.63194709271367e-05,
+      "loss": 0.1082,
+      "mean_token_accuracy": 0.9575189769268035,
+      "num_tokens": 1203136.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9555979907512665,
+      "epoch": 0.8595988538681948,
+      "grad_norm": 0.5703518986701965,
+      "learning_rate": 6.557097449654587e-05,
+      "loss": 0.1077,
+      "mean_token_accuracy": 0.9579966783523559,
+      "num_tokens": 1216913.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9558742046356201,
+      "epoch": 0.8691499522445081,
+      "grad_norm": 0.7904860377311707,
+      "learning_rate": 6.481859354969548e-05,
+      "loss": 0.1001,
+      "mean_token_accuracy": 0.9613328218460083,
+      "num_tokens": 1229969.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.9618341326713562,
+      "epoch": 0.8787010506208214,
+      "grad_norm": 1.1422725915908813,
+      "learning_rate": 6.406251578428046e-05,
+      "loss": 0.1043,
+      "mean_token_accuracy": 0.9618310868740082,
+      "num_tokens": 1243680.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9476982057094574,
+      "epoch": 0.8882521489971347,
+      "grad_norm": 0.6227033734321594,
+      "learning_rate": 6.330292982024712e-05,
+      "loss": 0.1004,
+      "mean_token_accuracy": 0.9570560812950134,
+      "num_tokens": 1257573.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.9620028853416442,
+      "epoch": 0.897803247373448,
+      "grad_norm": 0.5102617144584656,
+      "learning_rate": 6.254002515273774e-05,
+      "loss": 0.0976,
+      "mean_token_accuracy": 0.9597377300262451,
+      "num_tokens": 1270701.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9383984684944153,
+      "epoch": 0.9073543457497613,
+      "grad_norm": 0.7831044793128967,
+      "learning_rate": 6.177399210481698e-05,
+      "loss": 0.1014,
+      "mean_token_accuracy": 0.9597761750221252,
+      "num_tokens": 1283809.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.9364366352558136,
+      "epoch": 0.9169054441260746,
+      "grad_norm": 0.6067343950271606,
+      "learning_rate": 6.100502177999156e-05,
+      "loss": 0.0935,
+      "mean_token_accuracy": 0.9624200344085694,
+      "num_tokens": 1297005.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.976334136724472,
+      "epoch": 0.9264565425023877,
+      "grad_norm": 0.7300851345062256,
+      "learning_rate": 6.023330601453551e-05,
+      "loss": 0.126,
+      "mean_token_accuracy": 0.9492876768112183,
+      "num_tokens": 1310830.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9701994717121124,
+      "epoch": 0.936007640878701,
+      "grad_norm": 0.7335929274559021,
+      "learning_rate": 5.945903732963234e-05,
+      "loss": 0.103,
+      "mean_token_accuracy": 0.9596088886260986,
+      "num_tokens": 1324145.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.9694394767284393,
+      "epoch": 0.9455587392550143,
+      "grad_norm": 0.674283504486084,
+      "learning_rate": 5.868240888334653e-05,
+      "loss": 0.1007,
+      "mean_token_accuracy": 0.9569411039352417,
+      "num_tokens": 1337718.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.9264466464519501,
+      "epoch": 0.9551098376313276,
+      "grad_norm": 0.7404454350471497,
+      "learning_rate": 5.790361442243605e-05,
+      "loss": 0.091,
+      "mean_token_accuracy": 0.9658930003643036,
+      "num_tokens": 1351556.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9517737925052643,
+      "epoch": 0.9646609360076409,
+      "grad_norm": 0.6310734748840332,
+      "learning_rate": 5.712284823401808e-05,
+      "loss": 0.1055,
+      "mean_token_accuracy": 0.960016405582428,
+      "num_tokens": 1364817.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.9591375946998596,
+      "epoch": 0.9742120343839542,
+      "grad_norm": 0.4866735637187958,
+      "learning_rate": 5.634030509709982e-05,
+      "loss": 0.0928,
+      "mean_token_accuracy": 0.962578970193863,
+      "num_tokens": 1378098.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9355690777301788,
+      "epoch": 0.9837631327602674,
+      "grad_norm": 0.7113921642303467,
+      "learning_rate": 5.555618023398671e-05,
+      "loss": 0.0933,
+      "mean_token_accuracy": 0.9603404223918914,
+      "num_tokens": 1391374.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9405025899410248,
+      "epoch": 0.9933142311365807,
+      "grad_norm": 0.8997142910957336,
+      "learning_rate": 5.477066926158001e-05,
+      "loss": 0.1127,
+      "mean_token_accuracy": 0.9545972764492034,
+      "num_tokens": 1404715.0,
+      "step": 1040
+    },
+    {
+      "entropy": 0.9421123266220093,
+      "epoch": 1.002865329512894,
+      "grad_norm": 0.48524343967437744,
+      "learning_rate": 5.398396814257588e-05,
+      "loss": 0.0918,
+      "mean_token_accuracy": 0.965641850233078,
+      "num_tokens": 1417657.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.9085283041000366,
+      "epoch": 1.0124164278892072,
+      "grad_norm": 0.5955819487571716,
+      "learning_rate": 5.3196273136578286e-05,
+      "loss": 0.0611,
+      "mean_token_accuracy": 0.9785845041275024,
+      "num_tokens": 1430922.0,
+      "step": 1060
+    },
+    {
+      "entropy": 0.9008214473724365,
+      "epoch": 1.0219675262655206,
+      "grad_norm": 0.513383150100708,
+      "learning_rate": 5.2407780751137815e-05,
+      "loss": 0.0563,
+      "mean_token_accuracy": 0.9803127944469452,
+      "num_tokens": 1444074.0,
+      "step": 1070
+    },
+    {
+      "entropy": 0.8922453701496125,
+      "epoch": 1.0315186246418337,
+      "grad_norm": 0.6482188701629639,
+      "learning_rate": 5.1618687692728595e-05,
+      "loss": 0.0497,
+      "mean_token_accuracy": 0.982597553730011,
+      "num_tokens": 1457733.0,
+      "step": 1080
+    },
+    {
+      "entropy": 0.9022992134094239,
+      "epoch": 1.0410697230181472,
+      "grad_norm": 0.5353513956069946,
+      "learning_rate": 5.0829190817675575e-05,
+      "loss": 0.0567,
+      "mean_token_accuracy": 0.9808158636093139,
+      "num_tokens": 1471308.0,
+      "step": 1090
+    },
+    {
+      "entropy": 0.8979819655418396,
+      "epoch": 1.0506208213944603,
+      "grad_norm": 0.37797629833221436,
+      "learning_rate": 5.0039487083044566e-05,
+      "loss": 0.0445,
+      "mean_token_accuracy": 0.985016542673111,
+      "num_tokens": 1484529.0,
+      "step": 1100
+    },
+    {
+      "entropy": 0.8835367679595947,
+      "epoch": 1.0601719197707737,
+      "grad_norm": 0.5169422626495361,
+      "learning_rate": 4.924977349750691e-05,
+      "loss": 0.0524,
+      "mean_token_accuracy": 0.9823050558567047,
+      "num_tokens": 1498025.0,
+      "step": 1110
+    },
+    {
+      "entropy": 0.889462798833847,
+      "epoch": 1.069723018147087,
+      "grad_norm": 0.6162198781967163,
+      "learning_rate": 4.8460247072191496e-05,
+      "loss": 0.049,
+      "mean_token_accuracy": 0.9827019155025483,
+      "num_tokens": 1512502.0,
+      "step": 1120
+    },
+    {
+      "entropy": 0.8917974889278412,
+      "epoch": 1.0792741165234,
+      "grad_norm": 0.42623934149742126,
+      "learning_rate": 4.767110477153604e-05,
+      "loss": 0.052,
+      "mean_token_accuracy": 0.980078113079071,
+      "num_tokens": 1525623.0,
+      "step": 1130
+    },
+    {
+      "entropy": 0.8909034311771393,
+      "epoch": 1.0888252148997135,
+      "grad_norm": 0.33947375416755676,
+      "learning_rate": 4.688254346415006e-05,
+      "loss": 0.0494,
+      "mean_token_accuracy": 0.9828575730323792,
+      "num_tokens": 1538915.0,
+      "step": 1140
+    },
+    {
+      "entropy": 0.8963313579559327,
+      "epoch": 1.0983763132760267,
+      "grad_norm": 0.5847277045249939,
+      "learning_rate": 4.609475987370177e-05,
+      "loss": 0.0566,
+      "mean_token_accuracy": 0.9805838346481324,
+      "num_tokens": 1552610.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.8780421018600464,
+      "epoch": 1.10792741165234,
+      "grad_norm": 0.6897232532501221,
+      "learning_rate": 4.530795052984104e-05,
+      "loss": 0.0476,
+      "mean_token_accuracy": 0.9839768528938293,
+      "num_tokens": 1566152.0,
+      "step": 1160
+    },
+    {
+      "entropy": 0.875683170557022,
+      "epoch": 1.1174785100286533,
+      "grad_norm": 0.5435718894004822,
+      "learning_rate": 4.4522311719170866e-05,
+      "loss": 0.0428,
+      "mean_token_accuracy": 0.9860684931278229,
+      "num_tokens": 1579016.0,
+      "step": 1170
+    },
+    {
+      "entropy": 0.8858314216136932,
+      "epoch": 1.1270296084049667,
+      "grad_norm": 0.5599232912063599,
+      "learning_rate": 4.373803943627946e-05,
+      "loss": 0.047,
+      "mean_token_accuracy": 0.9856097400188446,
+      "num_tokens": 1592658.0,
+      "step": 1180
+    },
+    {
+      "entropy": 0.8686777770519256,
+      "epoch": 1.1365807067812799,
+      "grad_norm": 0.474824994802475,
+      "learning_rate": 4.295532933484509e-05,
+      "loss": 0.0559,
+      "mean_token_accuracy": 0.981699550151825,
+      "num_tokens": 1606104.0,
+      "step": 1190
+    },
+    {
+      "entropy": 0.884328818321228,
+      "epoch": 1.146131805157593,
+      "grad_norm": 0.3429393172264099,
+      "learning_rate": 4.2174376678826104e-05,
+      "loss": 0.0532,
+      "mean_token_accuracy": 0.9807308793067933,
+      "num_tokens": 1620362.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.855708247423172,
+      "epoch": 1.1556829035339065,
+      "grad_norm": 0.3353695273399353,
+      "learning_rate": 4.1395376293748134e-05,
+      "loss": 0.0484,
+      "mean_token_accuracy": 0.9834835708141327,
+      "num_tokens": 1633859.0,
+      "step": 1210
+    },
+    {
+      "entropy": 0.8697160899639129,
+      "epoch": 1.1652340019102196,
+      "grad_norm": 0.27550438046455383,
+      "learning_rate": 4.0618522518100534e-05,
+      "loss": 0.0368,
+      "mean_token_accuracy": 0.9874620974063874,
+      "num_tokens": 1646952.0,
+      "step": 1220
+    },
+    {
+      "entropy": 0.8602092564105988,
+      "epoch": 1.174785100286533,
+      "grad_norm": 0.6063143610954285,
+      "learning_rate": 3.984400915485459e-05,
+      "loss": 0.0469,
+      "mean_token_accuracy": 0.9812505424022675,
+      "num_tokens": 1659969.0,
+      "step": 1230
+    },
+    {
+      "entropy": 0.8630571961402893,
+      "epoch": 1.1843361986628462,
+      "grad_norm": 0.41278010606765747,
+      "learning_rate": 3.9072029423115055e-05,
+      "loss": 0.0573,
+      "mean_token_accuracy": 0.9798938691616058,
+      "num_tokens": 1673108.0,
+      "step": 1240
+    },
+    {
+      "entropy": 0.8782302141189575,
+      "epoch": 1.1938872970391594,
+      "grad_norm": 0.473196804523468,
+      "learning_rate": 3.830277590991759e-05,
+      "loss": 0.0436,
+      "mean_token_accuracy": 0.9852949678897858,
+      "num_tokens": 1686935.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.8697366058826447,
+      "epoch": 1.2034383954154728,
+      "grad_norm": 0.5442796349525452,
+      "learning_rate": 3.75364405221836e-05,
+      "loss": 0.0505,
+      "mean_token_accuracy": 0.981706154346466,
+      "num_tokens": 1700292.0,
+      "step": 1260
+    },
+    {
+      "entropy": 0.8711552619934082,
+      "epoch": 1.212989493791786,
+      "grad_norm": 0.45953795313835144,
+      "learning_rate": 3.677321443884509e-05,
+      "loss": 0.0505,
+      "mean_token_accuracy": 0.9815465748310089,
+      "num_tokens": 1713397.0,
+      "step": 1270
+    },
+    {
+      "entropy": 0.85686274766922,
+      "epoch": 1.2225405921680994,
+      "grad_norm": 0.45702072978019714,
+      "learning_rate": 3.601328806315082e-05,
+      "loss": 0.0424,
+      "mean_token_accuracy": 0.9833027422428131,
+      "num_tokens": 1727202.0,
+      "step": 1280
+    },
+    {
+      "entropy": 0.8686334311962127,
+      "epoch": 1.2320916905444126,
+      "grad_norm": 0.5354903340339661,
+      "learning_rate": 3.5256850975166136e-05,
+      "loss": 0.0477,
+      "mean_token_accuracy": 0.9839615643024444,
+      "num_tokens": 1740680.0,
+      "step": 1290
+    },
+    {
+      "entropy": 0.8737336277961731,
+      "epoch": 1.2416427889207258,
+      "grad_norm": 0.3656454086303711,
+      "learning_rate": 3.4504091884478074e-05,
+      "loss": 0.04,
+      "mean_token_accuracy": 0.9867834866046905,
+      "num_tokens": 1754482.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.8712716102600098,
+      "epoch": 1.2511938872970392,
+      "grad_norm": 0.5792255401611328,
+      "learning_rate": 3.375519858311781e-05,
+      "loss": 0.0468,
+      "mean_token_accuracy": 0.9852323174476624,
+      "num_tokens": 1768580.0,
+      "step": 1310
+    },
+    {
+      "entropy": 0.856835150718689,
+      "epoch": 1.2607449856733524,
+      "grad_norm": 0.4875318706035614,
+      "learning_rate": 3.301035789871169e-05,
+      "loss": 0.0453,
+      "mean_token_accuracy": 0.9835512280464173,
+      "num_tokens": 1781995.0,
+      "step": 1320
+    },
+    {
+      "entropy": 0.8581301689147949,
+      "epoch": 1.2702960840496658,
+      "grad_norm": 0.5009487867355347,
+      "learning_rate": 3.226975564787322e-05,
+      "loss": 0.054,
+      "mean_token_accuracy": 0.9799129605293274,
+      "num_tokens": 1795247.0,
+      "step": 1330
+    },
+    {
+      "entropy": 0.8602255821228028,
+      "epoch": 1.279847182425979,
+      "grad_norm": 0.8827853798866272,
+      "learning_rate": 3.153357658984697e-05,
+      "loss": 0.0487,
+      "mean_token_accuracy": 0.9824649095535278,
+      "num_tokens": 1808634.0,
+      "step": 1340
+    },
+    {
+      "entropy": 0.864433753490448,
+      "epoch": 1.2893982808022924,
+      "grad_norm": 0.7510569095611572,
+      "learning_rate": 3.0802004380416414e-05,
+      "loss": 0.0482,
+      "mean_token_accuracy": 0.9836005508899689,
+      "num_tokens": 1822813.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.8462441980838775,
+      "epoch": 1.2989493791786055,
+      "grad_norm": 0.6266422271728516,
+      "learning_rate": 3.0075221526087082e-05,
+      "loss": 0.0409,
+      "mean_token_accuracy": 0.9865394473075867,
+      "num_tokens": 1836410.0,
+      "step": 1360
+    },
+    {
+      "entropy": 0.8333847224712372,
+      "epoch": 1.3085004775549187,
+      "grad_norm": 0.6607073545455933,
+      "learning_rate": 2.935340933855626e-05,
+      "loss": 0.0482,
+      "mean_token_accuracy": 0.9826157450675964,
+      "num_tokens": 1850061.0,
+      "step": 1370
+    },
+    {
+      "entropy": 0.845102971792221,
+      "epoch": 1.3180515759312321,
+      "grad_norm": 0.38132601976394653,
+      "learning_rate": 2.863674788948097e-05,
+      "loss": 0.0455,
+      "mean_token_accuracy": 0.9855281293392182,
+      "num_tokens": 1864554.0,
+      "step": 1380
+    },
+    {
+      "entropy": 0.8428005456924439,
+      "epoch": 1.3276026743075453,
+      "grad_norm": 0.520173966884613,
+      "learning_rate": 2.7925415965555123e-05,
+      "loss": 0.0452,
+      "mean_token_accuracy": 0.9857592105865478,
+      "num_tokens": 1878030.0,
+      "step": 1390
+    },
+    {
+      "entropy": 0.8441922426223755,
+      "epoch": 1.3371537726838587,
+      "grad_norm": 0.5209589004516602,
+      "learning_rate": 2.7219591023907408e-05,
+      "loss": 0.0459,
+      "mean_token_accuracy": 0.985075706243515,
+      "num_tokens": 1891710.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.8465040624141693,
+      "epoch": 1.346704871060172,
+      "grad_norm": 0.5454689264297485,
+      "learning_rate": 2.6519449147830637e-05,
+      "loss": 0.0416,
+      "mean_token_accuracy": 0.9840689122676849,
+      "num_tokens": 1905521.0,
+      "step": 1410
+    },
+    {
+      "entropy": 0.8402527034282684,
+      "epoch": 1.3562559694364853,
+      "grad_norm": 0.7419883012771606,
+      "learning_rate": 2.5825165002854124e-05,
+      "loss": 0.0473,
+      "mean_token_accuracy": 0.981515645980835,
+      "num_tokens": 1918326.0,
+      "step": 1420
+    },
+    {
+      "entropy": 0.8471764385700226,
+      "epoch": 1.3658070678127985,
+      "grad_norm": 0.31341180205345154,
+      "learning_rate": 2.513691179316942e-05,
+      "loss": 0.0473,
+      "mean_token_accuracy": 0.9833329856395722,
+      "num_tokens": 1931996.0,
+      "step": 1430
+    },
+    {
+      "entropy": 0.8358026504516601,
+      "epoch": 1.3753581661891117,
+      "grad_norm": 0.3986743092536926,
+      "learning_rate": 2.4454861218420925e-05,
+      "loss": 0.0457,
+      "mean_token_accuracy": 0.9844906508922577,
+      "num_tokens": 1945489.0,
+      "step": 1440
+    },
+    {
+      "entropy": 0.8330438137054443,
+      "epoch": 1.384909264565425,
+      "grad_norm": 1.1977542638778687,
+      "learning_rate": 2.3779183430871594e-05,
+      "loss": 0.047,
+      "mean_token_accuracy": 0.9825788795948028,
+      "num_tokens": 1959170.0,
+      "step": 1450
+    },
+    {
+      "entropy": 0.8289123237133026,
+      "epoch": 1.3944603629417383,
+      "grad_norm": 0.48872071504592896,
+      "learning_rate": 2.311004699295488e-05,
+      "loss": 0.0551,
+      "mean_token_accuracy": 0.9834242284297943,
+      "num_tokens": 1972774.0,
+      "step": 1460
+    },
+    {
+      "entropy": 0.8454031765460968,
+      "epoch": 1.4040114613180517,
+      "grad_norm": 0.7900852560997009,
+      "learning_rate": 2.2447618835223127e-05,
+      "loss": 0.045,
+      "mean_token_accuracy": 0.9844704747200013,
+      "num_tokens": 1986433.0,
+      "step": 1470
+    },
+    {
+      "entropy": 0.8375308513641357,
+      "epoch": 1.4135625596943648,
+      "grad_norm": 0.47152644395828247,
+      "learning_rate": 2.1792064214703102e-05,
+      "loss": 0.0407,
+      "mean_token_accuracy": 0.9863695859909057,
+      "num_tokens": 2000018.0,
+      "step": 1480
+    },
+    {
+      "entropy": 0.8464671790599823,
+      "epoch": 1.4231136580706782,
+      "grad_norm": 0.48187822103500366,
+      "learning_rate": 2.1143546673669122e-05,
+      "loss": 0.0348,
+      "mean_token_accuracy": 0.9889981150627136,
+      "num_tokens": 2013829.0,
+      "step": 1490
+    },
+    {
+      "entropy": 0.8340227663516998,
+      "epoch": 1.4326647564469914,
+      "grad_norm": 0.8176564574241638,
+      "learning_rate": 2.050222799884387e-05,
+      "loss": 0.0382,
+      "mean_token_accuracy": 0.9870981931686401,
+      "num_tokens": 2027142.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.8444020092487335,
+      "epoch": 1.4422158548233046,
+      "grad_norm": 0.5223277807235718,
+      "learning_rate": 1.9868268181037185e-05,
+      "loss": 0.0435,
+      "mean_token_accuracy": 0.9867681264877319,
+      "num_tokens": 2040851.0,
+      "step": 1510
+    },
+    {
+      "entropy": 0.8293024003505707,
+      "epoch": 1.451766953199618,
+      "grad_norm": 0.5558390021324158,
+      "learning_rate": 1.9241825375232947e-05,
+      "loss": 0.0429,
+      "mean_token_accuracy": 0.9837581098079682,
+      "num_tokens": 2054703.0,
+      "step": 1520
+    },
+    {
+      "entropy": 0.8390320181846619,
+      "epoch": 1.4613180515759312,
+      "grad_norm": 0.4315888285636902,
+      "learning_rate": 1.8623055861133852e-05,
+      "loss": 0.0468,
+      "mean_token_accuracy": 0.9822153449058533,
+      "num_tokens": 2067960.0,
+      "step": 1530
+    },
+    {
+      "entropy": 0.8458277225494385,
+      "epoch": 1.4708691499522444,
+      "grad_norm": 0.8767555952072144,
+      "learning_rate": 1.8012114004174048e-05,
+      "loss": 0.0453,
+      "mean_token_accuracy": 0.9851816177368165,
+      "num_tokens": 2081121.0,
+      "step": 1540
+    },
+    {
+      "entropy": 0.8428352355957032,
+      "epoch": 1.4804202483285578,
+      "grad_norm": 0.5191040635108948,
+      "learning_rate": 1.7409152217009368e-05,
+      "loss": 0.0391,
+      "mean_token_accuracy": 0.9887764036655426,
+      "num_tokens": 2094631.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.8508458137512207,
+      "epoch": 1.4899713467048712,
+      "grad_norm": 0.7466721534729004,
+      "learning_rate": 1.6814320921494693e-05,
+      "loss": 0.0436,
+      "mean_token_accuracy": 0.9832504689693451,
+      "num_tokens": 2108099.0,
+      "step": 1560
+    },
+    {
+      "entropy": 0.8553808093070984,
+      "epoch": 1.4995224450811844,
+      "grad_norm": 0.5178319811820984,
+      "learning_rate": 1.6227768511157976e-05,
+      "loss": 0.0405,
+      "mean_token_accuracy": 0.9844492077827454,
+      "num_tokens": 2121504.0,
+      "step": 1570
+    },
+    {
+      "entropy": 0.847453647851944,
+      "epoch": 1.5090735434574976,
+      "grad_norm": 0.8944960832595825,
+      "learning_rate": 1.5649641314180274e-05,
+      "loss": 0.0442,
+      "mean_token_accuracy": 0.9836427867412567,
+      "num_tokens": 2134924.0,
+      "step": 1580
+    },
+    {
+      "entropy": 0.8473526418209076,
+      "epoch": 1.518624641833811,
+      "grad_norm": 0.7102769613265991,
+      "learning_rate": 1.5080083556890973e-05,
+      "loss": 0.0447,
+      "mean_token_accuracy": 0.9815830886363983,
+      "num_tokens": 2148917.0,
+      "step": 1590
+    },
+    {
+      "entropy": 0.8357010543346405,
+      "epoch": 1.5281757402101241,
+      "grad_norm": 0.510181725025177,
+      "learning_rate": 1.4519237327787449e-05,
+      "loss": 0.0473,
+      "mean_token_accuracy": 0.9835653662681579,
+      "num_tokens": 2161999.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.841375082731247,
+      "epoch": 1.5377268385864373,
+      "grad_norm": 0.2879631817340851,
+      "learning_rate": 1.3967242542088005e-05,
+      "loss": 0.0381,
+      "mean_token_accuracy": 0.9859650731086731,
+      "num_tokens": 2175634.0,
+      "step": 1610
+    },
+    {
+      "entropy": 0.8491779923439026,
+      "epoch": 1.5472779369627507,
+      "grad_norm": 0.4654577374458313,
+      "learning_rate": 1.3424236906827042e-05,
+      "loss": 0.0447,
+      "mean_token_accuracy": 0.9868191719055176,
+      "num_tokens": 2189783.0,
+      "step": 1620
+    },
+    {
+      "entropy": 0.8409325301647186,
+      "epoch": 1.5568290353390641,
+      "grad_norm": 0.40979066491127014,
+      "learning_rate": 1.289035588650097e-05,
+      "loss": 0.0316,
+      "mean_token_accuracy": 0.9891854405403138,
+      "num_tokens": 2203186.0,
+      "step": 1630
+    },
+    {
+      "entropy": 0.8205541491508483,
+      "epoch": 1.5663801337153773,
+      "grad_norm": 0.6202529072761536,
+      "learning_rate": 1.2365732669273778e-05,
+      "loss": 0.0444,
+      "mean_token_accuracy": 0.9836665630340576,
+      "num_tokens": 2216081.0,
+      "step": 1640
+    },
+    {
+      "entropy": 0.8395661950111389,
+      "epoch": 1.5759312320916905,
+      "grad_norm": 0.6057273149490356,
+      "learning_rate": 1.1850498133750244e-05,
+      "loss": 0.0422,
+      "mean_token_accuracy": 0.9868719637393951,
+      "num_tokens": 2229463.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.8393484711647033,
+      "epoch": 1.585482330468004,
+      "grad_norm": 0.6256664991378784,
+      "learning_rate": 1.1344780816325511e-05,
+      "loss": 0.0366,
+      "mean_token_accuracy": 0.9868586599826813,
+      "num_tokens": 2242823.0,
+      "step": 1660
+    },
+    {
+      "entropy": 0.8326752245426178,
+      "epoch": 1.595033428844317,
+      "grad_norm": 0.7295214533805847,
+      "learning_rate": 1.0848706879118892e-05,
+      "loss": 0.0478,
+      "mean_token_accuracy": 0.9859496474266052,
+      "num_tokens": 2255767.0,
+      "step": 1670
+    },
+    {
+      "entropy": 0.8392886996269227,
+      "epoch": 1.6045845272206303,
+      "grad_norm": 0.3701745569705963,
+      "learning_rate": 1.0362400078500018e-05,
+      "loss": 0.0431,
+      "mean_token_accuracy": 0.9854917049407959,
+      "num_tokens": 2269452.0,
+      "step": 1680
+    },
+    {
+      "entropy": 0.8448474764823913,
+      "epoch": 1.6141356255969437,
+      "grad_norm": 0.4770224094390869,
+      "learning_rate": 9.885981734215094e-06,
+      "loss": 0.0392,
+      "mean_token_accuracy": 0.9876186609268188,
+      "num_tokens": 2283260.0,
+      "step": 1690
+    },
+    {
+      "entropy": 0.8425587892532349,
+      "epoch": 1.623686723973257,
+      "grad_norm": 0.6594133377075195,
+      "learning_rate": 9.419570699121072e-06,
+      "loss": 0.0353,
+      "mean_token_accuracy": 0.9895827174186707,
+      "num_tokens": 2297040.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.8364692032337189,
+      "epoch": 1.63323782234957,
+      "grad_norm": 0.681790292263031,
+      "learning_rate": 8.963283329535305e-06,
+      "loss": 0.0375,
+      "mean_token_accuracy": 0.9890973806381226,
+      "num_tokens": 2310231.0,
+      "step": 1710
+    },
+    {
+      "entropy": 0.8175343990325927,
+      "epoch": 1.6427889207258835,
+      "grad_norm": 0.8300491571426392,
+      "learning_rate": 8.517233456207818e-06,
+      "loss": 0.0386,
+      "mean_token_accuracy": 0.9848393678665162,
+      "num_tokens": 2322983.0,
+      "step": 1720
+    },
+    {
+      "entropy": 0.8181714415550232,
+      "epoch": 1.6523400191021969,
+      "grad_norm": 0.4727650582790375,
+      "learning_rate": 8.081532355923922e-06,
+      "loss": 0.0335,
+      "mean_token_accuracy": 0.9883810818195343,
+      "num_tokens": 2336190.0,
+      "step": 1730
+    },
+    {
+      "entropy": 0.8469636023044587,
+      "epoch": 1.66189111747851,
+      "grad_norm": 0.5242503881454468,
+      "learning_rate": 7.65628872374367e-06,
+      "loss": 0.0405,
+      "mean_token_accuracy": 0.9869709432125091,
+      "num_tokens": 2349752.0,
+      "step": 1740
+    },
+    {
+      "entropy": 0.8473298013210296,
+      "epoch": 1.6714422158548232,
+      "grad_norm": 0.9949101209640503,
+      "learning_rate": 7.241608645885628e-06,
+      "loss": 0.0418,
+      "mean_token_accuracy": 0.9865728616714478,
+      "num_tokens": 2363415.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.8285278797149658,
+      "epoch": 1.6809933142311366,
+      "grad_norm": 0.7177553772926331,
+      "learning_rate": 6.8375955732612964e-06,
+      "loss": 0.0375,
+      "mean_token_accuracy": 0.9872062563896179,
+      "num_tokens": 2376238.0,
+      "step": 1760
+    },
+    {
+      "entropy": 0.8316291749477387,
+      "epoch": 1.6905444126074498,
+      "grad_norm": 0.7726125717163086,
+      "learning_rate": 6.444350295667112e-06,
+      "loss": 0.0332,
+      "mean_token_accuracy": 0.9886421978473663,
+      "num_tokens": 2389365.0,
+      "step": 1770
+    },
+    {
+      "entropy": 0.8496015131473541,
+      "epoch": 1.700095510983763,
+      "grad_norm": 0.5730778574943542,
+      "learning_rate": 6.061970916640236e-06,
+      "loss": 0.0492,
+      "mean_token_accuracy": 0.9829839766025543,
+      "num_tokens": 2403735.0,
+      "step": 1780
+    },
+    {
+      "entropy": 0.8360233902931213,
+      "epoch": 1.7096466093600764,
+      "grad_norm": 0.8977645635604858,
+      "learning_rate": 5.690552828984458e-06,
+      "loss": 0.0358,
+      "mean_token_accuracy": 0.9885068714618683,
+      "num_tokens": 2417109.0,
+      "step": 1790
+    },
+    {
+      "entropy": 0.8332313537597656,
+      "epoch": 1.7191977077363898,
+      "grad_norm": 0.407375693321228,
+      "learning_rate": 5.3301886909724375e-06,
+      "loss": 0.0386,
+      "mean_token_accuracy": 0.9869596660137177,
+      "num_tokens": 2430479.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.8448746919631958,
+      "epoch": 1.728748806112703,
+      "grad_norm": 0.5897151827812195,
+      "learning_rate": 4.980968403230097e-06,
+      "loss": 0.0485,
+      "mean_token_accuracy": 0.9813798069953918,
+      "num_tokens": 2444501.0,
+      "step": 1810
+    },
+    {
+      "entropy": 0.846867436170578,
+      "epoch": 1.7382999044890162,
+      "grad_norm": 0.5966483950614929,
+      "learning_rate": 4.642979086308963e-06,
+      "loss": 0.0385,
+      "mean_token_accuracy": 0.987826406955719,
+      "num_tokens": 2458250.0,
+      "step": 1820
+    },
+    {
+      "entropy": 0.8257025718688965,
+      "epoch": 1.7478510028653296,
+      "grad_norm": 0.7944663166999817,
+      "learning_rate": 4.316305058952069e-06,
+      "loss": 0.0344,
+      "mean_token_accuracy": 0.9889677047729493,
+      "num_tokens": 2471343.0,
+      "step": 1830
+    },
+    {
+      "entropy": 0.8363999843597412,
+      "epoch": 1.7574021012416428,
+      "grad_norm": 0.47064995765686035,
+      "learning_rate": 4.001027817058789e-06,
+      "loss": 0.0314,
+      "mean_token_accuracy": 0.9890477120876312,
+      "num_tokens": 2485461.0,
+      "step": 1840
+    },
+    {
+      "entropy": 0.842120784521103,
+      "epoch": 1.766953199617956,
+      "grad_norm": 0.5960325002670288,
+      "learning_rate": 3.697226013353955e-06,
+      "loss": 0.0343,
+      "mean_token_accuracy": 0.9892731964588165,
+      "num_tokens": 2499676.0,
+      "step": 1850
+    },
+    {
+      "entropy": 0.8390541195869445,
+      "epoch": 1.7765042979942693,
+      "grad_norm": 0.5265341401100159,
+      "learning_rate": 3.404975437766217e-06,
+      "loss": 0.0338,
+      "mean_token_accuracy": 0.9878333151340485,
+      "num_tokens": 2513383.0,
+      "step": 1860
+    },
+    {
+      "entropy": 0.8359919905662536,
+      "epoch": 1.7860553963705827,
+      "grad_norm": 0.769128680229187,
+      "learning_rate": 3.1243489985206097e-06,
+      "loss": 0.0445,
+      "mean_token_accuracy": 0.9820275127887725,
+      "num_tokens": 2526503.0,
+      "step": 1870
+    },
+    {
+      "entropy": 0.8379357039928437,
+      "epoch": 1.795606494746896,
+      "grad_norm": 0.5954472422599792,
+      "learning_rate": 2.8554167039500478e-06,
+      "loss": 0.0378,
+      "mean_token_accuracy": 0.9859267234802246,
+      "num_tokens": 2539835.0,
+      "step": 1880
+    },
+    {
+      "entropy": 0.8254489779472352,
+      "epoch": 1.8051575931232091,
+      "grad_norm": 0.4255012571811676,
+      "learning_rate": 2.598245645030195e-06,
+      "loss": 0.034,
+      "mean_token_accuracy": 0.9878638744354248,
+      "num_tokens": 2553396.0,
+      "step": 1890
+    },
+    {
+      "entropy": 0.8415489554405212,
+      "epoch": 1.8147086914995225,
+      "grad_norm": 0.5448042750358582,
+      "learning_rate": 2.3528999786421756e-06,
+      "loss": 0.0501,
+      "mean_token_accuracy": 0.9844222843647004,
+      "num_tokens": 2567080.0,
+      "step": 1900
+    },
+    {
+      "entropy": 0.851526939868927,
+      "epoch": 1.8242597898758357,
+      "grad_norm": 0.47402435541152954,
+      "learning_rate": 2.1194409115672663e-06,
+      "loss": 0.05,
+      "mean_token_accuracy": 0.9853261828422546,
+      "num_tokens": 2580658.0,
+      "step": 1910
+    },
+    {
+      "entropy": 0.8315389931201935,
+      "epoch": 1.8338108882521489,
+      "grad_norm": 0.22259674966335297,
+      "learning_rate": 1.8979266852175382e-06,
+      "loss": 0.039,
+      "mean_token_accuracy": 0.9865755558013916,
+      "num_tokens": 2594051.0,
+      "step": 1920
+    },
+    {
+      "entropy": 0.848485404253006,
+      "epoch": 1.8433619866284623,
+      "grad_norm": 0.4654080271720886,
+      "learning_rate": 1.688412561106284e-06,
+      "loss": 0.0374,
+      "mean_token_accuracy": 0.9884120166301728,
+      "num_tokens": 2607491.0,
+      "step": 1930
+    },
+    {
+      "entropy": 0.8288978815078736,
+      "epoch": 1.8529130850047757,
+      "grad_norm": 0.47596338391304016,
+      "learning_rate": 1.4909508070618084e-06,
+      "loss": 0.0319,
+      "mean_token_accuracy": 0.9895732343196869,
+      "num_tokens": 2621029.0,
+      "step": 1940
+    },
+    {
+      "entropy": 0.8414992928504944,
+      "epoch": 1.8624641833810889,
+      "grad_norm": 0.9999420046806335,
+      "learning_rate": 1.3055906841881348e-06,
+      "loss": 0.0389,
+      "mean_token_accuracy": 0.9880332112312317,
+      "num_tokens": 2634868.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.8323590695858002,
+      "epoch": 1.872015281757402,
+      "grad_norm": 0.6073062419891357,
+      "learning_rate": 1.1323784345757205e-06,
+      "loss": 0.0419,
+      "mean_token_accuracy": 0.9856351494789124,
+      "num_tokens": 2648316.0,
+      "step": 1960
+    },
+    {
+      "entropy": 0.8331670641899109,
+      "epoch": 1.8815663801337155,
+      "grad_norm": 0.743442714214325,
+      "learning_rate": 9.71357269765366e-07,
+      "loss": 0.05,
+      "mean_token_accuracy": 0.9823213040828704,
+      "num_tokens": 2661446.0,
+      "step": 1970
+    },
+    {
+      "entropy": 0.8585890531539917,
+      "epoch": 1.8911174785100286,
+      "grad_norm": 0.4772331416606903,
+      "learning_rate": 8.225673599682016e-07,
+      "loss": 0.0375,
+      "mean_token_accuracy": 0.9866939961910248,
+      "num_tokens": 2675223.0,
+      "step": 1980
+    },
+    {
+      "entropy": 0.847622275352478,
+      "epoch": 1.9006685768863418,
+      "grad_norm": 0.7534369826316833,
+      "learning_rate": 6.860458240443179e-07,
+      "loss": 0.0495,
+      "mean_token_accuracy": 0.9846001207828522,
+      "num_tokens": 2688452.0,
+      "step": 1990
+    },
+    {
+      "entropy": 0.8250655412673951,
+      "epoch": 1.9102196752626552,
+      "grad_norm": 0.8683748841285706,
+      "learning_rate": 5.618267202426864e-07,
+      "loss": 0.0431,
+      "mean_token_accuracy": 0.9859239101409912,
+      "num_tokens": 2701249.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.850209641456604,
+      "epoch": 1.9197707736389686,
+      "grad_norm": 0.7355635166168213,
+      "learning_rate": 4.4994103770457653e-07,
+      "loss": 0.0334,
+      "mean_token_accuracy": 0.9885294079780579,
+      "num_tokens": 2714442.0,
+      "step": 2010
+    },
+    {
+      "entropy": 0.8359505593776703,
+      "epoch": 1.9293218720152816,
+      "grad_norm": 0.8837029933929443,
+      "learning_rate": 3.5041668873266877e-07,
+      "loss": 0.0401,
+      "mean_token_accuracy": 0.9859329581260681,
+      "num_tokens": 2727832.0,
+      "step": 2020
+    },
+    {
+      "entropy": 0.8400478720664978,
+      "epoch": 1.938872970391595,
+      "grad_norm": 0.5403333306312561,
+      "learning_rate": 2.6327850182769066e-07,
+      "loss": 0.0421,
+      "mean_token_accuracy": 0.9845231354236603,
+      "num_tokens": 2741459.0,
+      "step": 2030
+    },
+    {
+      "entropy": 0.8321483552455902,
+      "epoch": 1.9484240687679084,
+      "grad_norm": 0.46632304787635803,
+      "learning_rate": 1.8854821549445513e-07,
+      "loss": 0.0387,
+      "mean_token_accuracy": 0.9873438239097595,
+      "num_tokens": 2754154.0,
+      "step": 2040
+    },
+    {
+      "entropy": 0.8403096079826355,
+      "epoch": 1.9579751671442216,
+      "grad_norm": 0.5475239157676697,
+      "learning_rate": 1.2624447281867624e-07,
+      "loss": 0.045,
+      "mean_token_accuracy": 0.9851735234260559,
+      "num_tokens": 2767928.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.8491226971149445,
+      "epoch": 1.9675262655205348,
+      "grad_norm": 0.6375964283943176,
+      "learning_rate": 7.638281681608961e-08,
+      "loss": 0.0349,
+      "mean_token_accuracy": 0.9896177887916565,
+      "num_tokens": 2782121.0,
+      "step": 2060
+    },
+    {
+      "entropy": 0.8313807904720306,
+      "epoch": 1.9770773638968482,
+      "grad_norm": 0.4844121038913727,
+      "learning_rate": 3.8975686554870714e-08,
+      "loss": 0.0401,
+      "mean_token_accuracy": 0.987853366136551,
+      "num_tokens": 2795321.0,
+      "step": 2070
+    },
+    {
+      "entropy": 0.8354905724525452,
+      "epoch": 1.9866284622731614,
+      "grad_norm": 0.6104508638381958,
+      "learning_rate": 1.4032414052478349e-08,
+      "loss": 0.043,
+      "mean_token_accuracy": 0.9869842231273651,
+      "num_tokens": 2809519.0,
+      "step": 2080
+    },
+    {
+      "entropy": 0.8304247975349426,
+      "epoch": 1.9961795606494745,
+      "grad_norm": 0.30469194054603577,
+      "learning_rate": 1.5592219475502047e-09,
+      "loss": 0.0313,
+      "mean_token_accuracy": 0.9885540783405304,
+      "num_tokens": 2822753.0,
+      "step": 2090
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2094,
+  "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_s2/checkpoint-2094/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f189c2709f177b4103ed0f4903d53255b3106698718586cb13af342600c1cc9d
+size 6417

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2/manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2",
+  "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": 2,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2/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_s2/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_s2/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_s2/train_manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2",
+  "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": 2,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f189c2709f177b4103ed0f4903d53255b3106698718586cb13af342600c1cc9d
+size 6417