add checkpoint cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3

.gitattributes

@@ -742,3 +742,6 @@ checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s2/tokenizer
 checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/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_s3
+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_s3
+
+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/m4y8y4bd) 
+
+
+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_s3/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": [
+    "down_proj",
+    "gate_proj",
+    "up_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_s3/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d30e807e5f187d3e0d474e0334cb2b2698b9e26c3d527739307116628323caa
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/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_s3/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_s3/checkpoint-1046/README.md

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "gate_proj",
+    "up_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_s3/checkpoint-1046/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c9856563cfd406c1791272d8d0bb33320da1adbdfec517c120c9ea22c8c9ef6
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/added_tokens.json

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/chat_template.jinja

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/special_tokens_map.json

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/tokenizer.json

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/trainer_state.json

@@ -0,0 +1,1074 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 1046,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.1632622241973878,
+      "epoch": 0.009560229445506692,
+      "grad_norm": 0.6969931125640869,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2719,
+      "mean_token_accuracy": 0.9138259589672089,
+      "num_tokens": 13150.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1633504509925843,
+      "epoch": 0.019120458891013385,
+      "grad_norm": 0.5006016492843628,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2592,
+      "mean_token_accuracy": 0.915404063463211,
+      "num_tokens": 26136.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1590815901756286,
+      "epoch": 0.028680688336520075,
+      "grad_norm": 0.4876864552497864,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.2211,
+      "mean_token_accuracy": 0.9319653451442719,
+      "num_tokens": 39207.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.1542185068130493,
+      "epoch": 0.03824091778202677,
+      "grad_norm": 0.5451372861862183,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1901,
+      "mean_token_accuracy": 0.9393570721149445,
+      "num_tokens": 52869.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.1529746413230897,
+      "epoch": 0.04780114722753346,
+      "grad_norm": 0.8673158288002014,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.2024,
+      "mean_token_accuracy": 0.9328486382961273,
+      "num_tokens": 66622.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.1270763635635377,
+      "epoch": 0.05736137667304015,
+      "grad_norm": 0.636655330657959,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1661,
+      "mean_token_accuracy": 0.9430084943771362,
+      "num_tokens": 80175.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.1287935256958008,
+      "epoch": 0.06692160611854685,
+      "grad_norm": 0.6914967894554138,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1946,
+      "mean_token_accuracy": 0.9291220664978027,
+      "num_tokens": 94356.0,
+      "step": 70
+    },
+    {
+      "entropy": 1.0964631199836732,
+      "epoch": 0.07648183556405354,
+      "grad_norm": 0.7391988039016724,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1519,
+      "mean_token_accuracy": 0.9491405785083771,
+      "num_tokens": 108716.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.0623166620731355,
+      "epoch": 0.08604206500956023,
+      "grad_norm": 0.6298117637634277,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1546,
+      "mean_token_accuracy": 0.9406298100948334,
+      "num_tokens": 122182.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0688253164291381,
+      "epoch": 0.09560229445506692,
+      "grad_norm": 0.6803618669509888,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1438,
+      "mean_token_accuracy": 0.9495500445365905,
+      "num_tokens": 135713.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0644982933998108,
+      "epoch": 0.10516252390057361,
+      "grad_norm": 0.8620546460151672,
+      "learning_rate": 9.999900008621665e-05,
+      "loss": 0.1627,
+      "mean_token_accuracy": 0.9426994383335113,
+      "num_tokens": 149456.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.0501671850681304,
+      "epoch": 0.1147227533460803,
+      "grad_norm": 0.4802723824977875,
+      "learning_rate": 9.998775151544478e-05,
+      "loss": 0.146,
+      "mean_token_accuracy": 0.943053936958313,
+      "num_tokens": 163062.0,
+      "step": 120
+    },
+    {
+      "entropy": 1.0445132613182069,
+      "epoch": 0.124282982791587,
+      "grad_norm": 0.7362843751907349,
+      "learning_rate": 9.996400730289628e-05,
+      "loss": 0.1668,
+      "mean_token_accuracy": 0.9377037048339844,
+      "num_tokens": 176574.0,
+      "step": 130
+    },
+    {
+      "entropy": 1.034316086769104,
+      "epoch": 0.1338432122370937,
+      "grad_norm": 0.630389392375946,
+      "learning_rate": 9.992777338400864e-05,
+      "loss": 0.1653,
+      "mean_token_accuracy": 0.9353829383850097,
+      "num_tokens": 191164.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0108101904392242,
+      "epoch": 0.14340344168260039,
+      "grad_norm": 0.7097353339195251,
+      "learning_rate": 9.987905881632213e-05,
+      "loss": 0.1488,
+      "mean_token_accuracy": 0.9435539603233337,
+      "num_tokens": 204805.0,
+      "step": 150
+    },
+    {
+      "entropy": 1.0055140674114227,
+      "epoch": 0.15296367112810708,
+      "grad_norm": 0.6539037227630615,
+      "learning_rate": 9.981787577721553e-05,
+      "loss": 0.1308,
+      "mean_token_accuracy": 0.9497624516487122,
+      "num_tokens": 218021.0,
+      "step": 160
+    },
+    {
+      "entropy": 0.9890082895755767,
+      "epoch": 0.16252390057361377,
+      "grad_norm": 0.47821044921875,
+      "learning_rate": 9.974423956086228e-05,
+      "loss": 0.1234,
+      "mean_token_accuracy": 0.9515690505504608,
+      "num_tokens": 231015.0,
+      "step": 170
+    },
+    {
+      "entropy": 0.9904860436916352,
+      "epoch": 0.17208413001912046,
+      "grad_norm": 0.6690865159034729,
+      "learning_rate": 9.965816857440717e-05,
+      "loss": 0.143,
+      "mean_token_accuracy": 0.942569500207901,
+      "num_tokens": 244021.0,
+      "step": 180
+    },
+    {
+      "entropy": 1.0002257645130157,
+      "epoch": 0.18164435946462715,
+      "grad_norm": 0.8160389065742493,
+      "learning_rate": 9.955968433336515e-05,
+      "loss": 0.1423,
+      "mean_token_accuracy": 0.9491697907447815,
+      "num_tokens": 256981.0,
+      "step": 190
+    },
+    {
+      "entropy": 0.9900605976581573,
+      "epoch": 0.19120458891013384,
+      "grad_norm": 0.5272551774978638,
+      "learning_rate": 9.944881145624291e-05,
+      "loss": 0.1488,
+      "mean_token_accuracy": 0.9452917218208313,
+      "num_tokens": 270999.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.9831291913986206,
+      "epoch": 0.20076481835564053,
+      "grad_norm": 0.627918541431427,
+      "learning_rate": 9.932557765838504e-05,
+      "loss": 0.1196,
+      "mean_token_accuracy": 0.9600761473178864,
+      "num_tokens": 283995.0,
+      "step": 210
+    },
+    {
+      "entropy": 0.9906052827835083,
+      "epoch": 0.21032504780114722,
+      "grad_norm": 0.7424314618110657,
+      "learning_rate": 9.919001374504569e-05,
+      "loss": 0.138,
+      "mean_token_accuracy": 0.9490589261054992,
+      "num_tokens": 297434.0,
+      "step": 220
+    },
+    {
+      "entropy": 1.0092378556728363,
+      "epoch": 0.2198852772466539,
+      "grad_norm": 0.5505907535552979,
+      "learning_rate": 9.904215360368826e-05,
+      "loss": 0.1277,
+      "mean_token_accuracy": 0.9500136435031891,
+      "num_tokens": 310688.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.0368991553783418,
+      "epoch": 0.2294455066921606,
+      "grad_norm": 1.0577473640441895,
+      "learning_rate": 9.888203419551432e-05,
+      "loss": 0.1594,
+      "mean_token_accuracy": 0.9426211178302765,
+      "num_tokens": 324594.0,
+      "step": 240
+    },
+    {
+      "entropy": 1.062998366355896,
+      "epoch": 0.2390057361376673,
+      "grad_norm": 0.8393595218658447,
+      "learning_rate": 9.870969554622426e-05,
+      "loss": 0.1579,
+      "mean_token_accuracy": 0.9380222499370575,
+      "num_tokens": 338003.0,
+      "step": 250
+    },
+    {
+      "entropy": 1.0535745024681091,
+      "epoch": 0.248565965583174,
+      "grad_norm": 0.4548805356025696,
+      "learning_rate": 9.852518073601194e-05,
+      "loss": 0.1363,
+      "mean_token_accuracy": 0.9484048664569855,
+      "num_tokens": 351394.0,
+      "step": 260
+    },
+    {
+      "entropy": 1.0056042432785035,
+      "epoch": 0.25812619502868067,
+      "grad_norm": 0.4833658039569855,
+      "learning_rate": 9.832853588879575e-05,
+      "loss": 0.1282,
+      "mean_token_accuracy": 0.9508345782756805,
+      "num_tokens": 365081.0,
+      "step": 270
+    },
+    {
+      "entropy": 1.0188749372959136,
+      "epoch": 0.2676864244741874,
+      "grad_norm": 0.5572522282600403,
+      "learning_rate": 9.811981016068881e-05,
+      "loss": 0.1328,
+      "mean_token_accuracy": 0.9506715357303619,
+      "num_tokens": 378518.0,
+      "step": 280
+    },
+    {
+      "entropy": 1.0242417335510254,
+      "epoch": 0.27724665391969405,
+      "grad_norm": 0.9603186845779419,
+      "learning_rate": 9.78990557277112e-05,
+      "loss": 0.1311,
+      "mean_token_accuracy": 0.9476073980331421,
+      "num_tokens": 391993.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0456064581871032,
+      "epoch": 0.28680688336520077,
+      "grad_norm": 0.6591246724128723,
+      "learning_rate": 9.766632777274738e-05,
+      "loss": 0.1115,
+      "mean_token_accuracy": 0.9543036341667175,
+      "num_tokens": 405246.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0853006482124328,
+      "epoch": 0.29636711281070743,
+      "grad_norm": 0.823593258857727,
+      "learning_rate": 9.742168447175182e-05,
+      "loss": 0.1259,
+      "mean_token_accuracy": 0.9509813964366913,
+      "num_tokens": 418524.0,
+      "step": 310
+    },
+    {
+      "entropy": 1.041166114807129,
+      "epoch": 0.30592734225621415,
+      "grad_norm": 0.6226829290390015,
+      "learning_rate": 9.716518697920655e-05,
+      "loss": 0.1442,
+      "mean_token_accuracy": 0.9461847722530365,
+      "num_tokens": 432202.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0213975965976716,
+      "epoch": 0.3154875717017208,
+      "grad_norm": 0.42342594265937805,
+      "learning_rate": 9.689689941283417e-05,
+      "loss": 0.1302,
+      "mean_token_accuracy": 0.9512553572654724,
+      "num_tokens": 445903.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.043001300096512,
+      "epoch": 0.32504780114722753,
+      "grad_norm": 0.5154736638069153,
+      "learning_rate": 9.661688883756996e-05,
+      "loss": 0.1231,
+      "mean_token_accuracy": 0.9551738917827606,
+      "num_tokens": 460277.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0378517508506775,
+      "epoch": 0.33460803059273425,
+      "grad_norm": 0.6703522801399231,
+      "learning_rate": 9.632522524879756e-05,
+      "loss": 0.129,
+      "mean_token_accuracy": 0.9532712638378144,
+      "num_tokens": 473892.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.0317105233669281,
+      "epoch": 0.3441682600382409,
+      "grad_norm": 0.6474915742874146,
+      "learning_rate": 9.602198155485176e-05,
+      "loss": 0.1076,
+      "mean_token_accuracy": 0.9581969559192658,
+      "num_tokens": 487145.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.0285814940929412,
+      "epoch": 0.35372848948374763,
+      "grad_norm": 0.7044402360916138,
+      "learning_rate": 9.570723355879342e-05,
+      "loss": 0.1227,
+      "mean_token_accuracy": 0.9547486007213593,
+      "num_tokens": 500481.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.055726957321167,
+      "epoch": 0.3632887189292543,
+      "grad_norm": 0.6242275238037109,
+      "learning_rate": 9.538105993946068e-05,
+      "loss": 0.127,
+      "mean_token_accuracy": 0.9565758466720581,
+      "num_tokens": 513817.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0444733262062074,
+      "epoch": 0.372848948374761,
+      "grad_norm": 0.47244662046432495,
+      "learning_rate": 9.504354223180118e-05,
+      "loss": 0.1302,
+      "mean_token_accuracy": 0.9526951789855957,
+      "num_tokens": 527250.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.0365854144096374,
+      "epoch": 0.3824091778202677,
+      "grad_norm": 0.4104255735874176,
+      "learning_rate": 9.469476480649056e-05,
+      "loss": 0.1257,
+      "mean_token_accuracy": 0.9535662651062011,
+      "num_tokens": 541558.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.025916337966919,
+      "epoch": 0.3919694072657744,
+      "grad_norm": 0.6373872756958008,
+      "learning_rate": 9.433481484884189e-05,
+      "loss": 0.1095,
+      "mean_token_accuracy": 0.9600341856479645,
+      "num_tokens": 554540.0,
+      "step": 410
+    },
+    {
+      "entropy": 1.040219533443451,
+      "epoch": 0.40152963671128106,
+      "grad_norm": 0.5284200310707092,
+      "learning_rate": 9.396378233701167e-05,
+      "loss": 0.1177,
+      "mean_token_accuracy": 0.9509544730186462,
+      "num_tokens": 568221.0,
+      "step": 420
+    },
+    {
+      "entropy": 1.0495393812656402,
+      "epoch": 0.4110898661567878,
+      "grad_norm": 0.6208754777908325,
+      "learning_rate": 9.358176001950759e-05,
+      "loss": 0.1277,
+      "mean_token_accuracy": 0.948467081785202,
+      "num_tokens": 581721.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.0427401065826416,
+      "epoch": 0.42065009560229444,
+      "grad_norm": 0.9366323351860046,
+      "learning_rate": 9.318884339200378e-05,
+      "loss": 0.1162,
+      "mean_token_accuracy": 0.9557548820972442,
+      "num_tokens": 595120.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0672527194023131,
+      "epoch": 0.43021032504780116,
+      "grad_norm": 0.5693362951278687,
+      "learning_rate": 9.278513067346941e-05,
+      "loss": 0.1246,
+      "mean_token_accuracy": 0.9521466493606567,
+      "num_tokens": 608361.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0667488336563111,
+      "epoch": 0.4397705544933078,
+      "grad_norm": 1.004165530204773,
+      "learning_rate": 9.237072278161652e-05,
+      "loss": 0.1211,
+      "mean_token_accuracy": 0.9553246557712555,
+      "num_tokens": 621843.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.055682933330536,
+      "epoch": 0.44933078393881454,
+      "grad_norm": 0.7205134034156799,
+      "learning_rate": 9.194572330767314e-05,
+      "loss": 0.1162,
+      "mean_token_accuracy": 0.9523303210735321,
+      "num_tokens": 635552.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.059775161743164,
+      "epoch": 0.4588910133843212,
+      "grad_norm": 0.5044583082199097,
+      "learning_rate": 9.151023849048818e-05,
+      "loss": 0.1336,
+      "mean_token_accuracy": 0.9506153464317322,
+      "num_tokens": 648523.0,
+      "step": 480
+    },
+    {
+      "entropy": 1.0490445971488953,
+      "epoch": 0.4684512428298279,
+      "grad_norm": 0.34968268871307373,
+      "learning_rate": 9.106437718997452e-05,
+      "loss": 0.1088,
+      "mean_token_accuracy": 0.9602180898189545,
+      "num_tokens": 661850.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0310296952724456,
+      "epoch": 0.4780114722753346,
+      "grad_norm": 1.0173206329345703,
+      "learning_rate": 9.060825085989682e-05,
+      "loss": 0.1049,
+      "mean_token_accuracy": 0.9578554153442382,
+      "num_tokens": 675322.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.047879695892334,
+      "epoch": 0.4875717017208413,
+      "grad_norm": 0.7152472138404846,
+      "learning_rate": 9.014197352001097e-05,
+      "loss": 0.1256,
+      "mean_token_accuracy": 0.95099977850914,
+      "num_tokens": 688704.0,
+      "step": 510
+    },
+    {
+      "entropy": 1.0453792810440063,
+      "epoch": 0.497131931166348,
+      "grad_norm": 0.8198304772377014,
+      "learning_rate": 8.966566172756213e-05,
+      "loss": 0.1305,
+      "mean_token_accuracy": 0.953268188238144,
+      "num_tokens": 702579.0,
+      "step": 520
+    },
+    {
+      "entropy": 1.036444890499115,
+      "epoch": 0.5066921606118547,
+      "grad_norm": 0.5910372734069824,
+      "learning_rate": 8.91794345481484e-05,
+      "loss": 0.1217,
+      "mean_token_accuracy": 0.952339518070221,
+      "num_tokens": 716160.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.034623110294342,
+      "epoch": 0.5162523900573613,
+      "grad_norm": 0.9556732177734375,
+      "learning_rate": 8.868341352595755e-05,
+      "loss": 0.1276,
+      "mean_token_accuracy": 0.9504207968711853,
+      "num_tokens": 729408.0,
+      "step": 540
+    },
+    {
+      "entropy": 1.0424424648284911,
+      "epoch": 0.5258126195028681,
+      "grad_norm": 0.5328360795974731,
+      "learning_rate": 8.817772265338409e-05,
+      "loss": 0.1221,
+      "mean_token_accuracy": 0.9487152993679047,
+      "num_tokens": 742824.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.032740694284439,
+      "epoch": 0.5353728489483748,
+      "grad_norm": 0.9654290080070496,
+      "learning_rate": 8.766248834003441e-05,
+      "loss": 0.1212,
+      "mean_token_accuracy": 0.9530553936958313,
+      "num_tokens": 756778.0,
+      "step": 560
+    },
+    {
+      "entropy": 1.036280381679535,
+      "epoch": 0.5449330783938815,
+      "grad_norm": 0.5315765142440796,
+      "learning_rate": 8.713783938112768e-05,
+      "loss": 0.1214,
+      "mean_token_accuracy": 0.9498340725898743,
+      "num_tokens": 770342.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.0438485622406006,
+      "epoch": 0.5544933078393881,
+      "grad_norm": 0.4137975573539734,
+      "learning_rate": 8.660390692530026e-05,
+      "loss": 0.1172,
+      "mean_token_accuracy": 0.9532914280891418,
+      "num_tokens": 784398.0,
+      "step": 580
+    },
+    {
+      "entropy": 1.0298362493515014,
+      "epoch": 0.5640535372848948,
+      "grad_norm": 0.6320862770080566,
+      "learning_rate": 8.606082444182216e-05,
+      "loss": 0.1225,
+      "mean_token_accuracy": 0.9563343822956085,
+      "num_tokens": 798017.0,
+      "step": 590
+    },
+    {
+      "entropy": 1.010947346687317,
+      "epoch": 0.5736137667304015,
+      "grad_norm": 0.837810218334198,
+      "learning_rate": 8.550872768723292e-05,
+      "loss": 0.1175,
+      "mean_token_accuracy": 0.9552404284477234,
+      "num_tokens": 811336.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.9999268174171447,
+      "epoch": 0.5831739961759083,
+      "grad_norm": 0.6310060620307922,
+      "learning_rate": 8.494775467140627e-05,
+      "loss": 0.1253,
+      "mean_token_accuracy": 0.9503907263278961,
+      "num_tokens": 824727.0,
+      "step": 610
+    },
+    {
+      "entropy": 1.0030306220054626,
+      "epoch": 0.5927342256214149,
+      "grad_norm": 0.5004256963729858,
+      "learning_rate": 8.4378045623051e-05,
+      "loss": 0.131,
+      "mean_token_accuracy": 0.9487506031990052,
+      "num_tokens": 838105.0,
+      "step": 620
+    },
+    {
+      "entropy": 0.9925207674503327,
+      "epoch": 0.6022944550669216,
+      "grad_norm": 0.9587653279304504,
+      "learning_rate": 8.379974295465758e-05,
+      "loss": 0.1026,
+      "mean_token_accuracy": 0.9612509369850158,
+      "num_tokens": 851280.0,
+      "step": 630
+    },
+    {
+      "entropy": 1.0160180270671844,
+      "epoch": 0.6118546845124283,
+      "grad_norm": 0.5568580627441406,
+      "learning_rate": 8.321299122689863e-05,
+      "loss": 0.1112,
+      "mean_token_accuracy": 0.9538475096225738,
+      "num_tokens": 864939.0,
+      "step": 640
+    },
+    {
+      "entropy": 1.0023184657096862,
+      "epoch": 0.621414913957935,
+      "grad_norm": 1.031872272491455,
+      "learning_rate": 8.261793711249254e-05,
+      "loss": 0.1213,
+      "mean_token_accuracy": 0.95554478764534,
+      "num_tokens": 878886.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.9941893517971039,
+      "epoch": 0.6309751434034416,
+      "grad_norm": 0.7980345487594604,
+      "learning_rate": 8.201472935953916e-05,
+      "loss": 0.0912,
+      "mean_token_accuracy": 0.9684767663478852,
+      "num_tokens": 892413.0,
+      "step": 660
+    },
+    {
+      "entropy": 1.0040182232856751,
+      "epoch": 0.6405353728489483,
+      "grad_norm": 0.6786800026893616,
+      "learning_rate": 8.140351875433649e-05,
+      "loss": 0.1148,
+      "mean_token_accuracy": 0.9537683606147767,
+      "num_tokens": 906485.0,
+      "step": 670
+    },
+    {
+      "entropy": 1.0011967778205872,
+      "epoch": 0.6500956022944551,
+      "grad_norm": 0.8331853151321411,
+      "learning_rate": 8.07844580836881e-05,
+      "loss": 0.1117,
+      "mean_token_accuracy": 0.9556680023670197,
+      "num_tokens": 920020.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9943797826766968,
+      "epoch": 0.6596558317399618,
+      "grad_norm": 1.0075759887695312,
+      "learning_rate": 8.015770209671044e-05,
+      "loss": 0.1046,
+      "mean_token_accuracy": 0.9595177710056305,
+      "num_tokens": 933577.0,
+      "step": 690
+    },
+    {
+      "entropy": 0.9939425528049469,
+      "epoch": 0.6692160611854685,
+      "grad_norm": 0.46297088265419006,
+      "learning_rate": 7.952340746614946e-05,
+      "loss": 0.102,
+      "mean_token_accuracy": 0.9599328935146332,
+      "num_tokens": 946367.0,
+      "step": 700
+    },
+    {
+      "entropy": 1.0044083833694457,
+      "epoch": 0.6787762906309751,
+      "grad_norm": 0.4464632272720337,
+      "learning_rate": 7.888173274921655e-05,
+      "loss": 0.1177,
+      "mean_token_accuracy": 0.957087230682373,
+      "num_tokens": 960252.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9848935604095459,
+      "epoch": 0.6883365200764818,
+      "grad_norm": 0.6167030334472656,
+      "learning_rate": 7.823283834795336e-05,
+      "loss": 0.098,
+      "mean_token_accuracy": 0.960848867893219,
+      "num_tokens": 973491.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9874480962753296,
+      "epoch": 0.6978967495219885,
+      "grad_norm": 0.4817270338535309,
+      "learning_rate": 7.75768864691354e-05,
+      "loss": 0.1207,
+      "mean_token_accuracy": 0.9501777410507202,
+      "num_tokens": 986879.0,
+      "step": 730
+    },
+    {
+      "entropy": 1.0066561222076416,
+      "epoch": 0.7074569789674953,
+      "grad_norm": 0.6590946316719055,
+      "learning_rate": 7.691404108372461e-05,
+      "loss": 0.1102,
+      "mean_token_accuracy": 0.960069864988327,
+      "num_tokens": 1000272.0,
+      "step": 740
+    },
+    {
+      "entropy": 1.003525936603546,
+      "epoch": 0.7170172084130019,
+      "grad_norm": 0.8013182282447815,
+      "learning_rate": 7.624446788588085e-05,
+      "loss": 0.1205,
+      "mean_token_accuracy": 0.955094313621521,
+      "num_tokens": 1013908.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.9959373891353607,
+      "epoch": 0.7265774378585086,
+      "grad_norm": 0.7129849791526794,
+      "learning_rate": 7.556833425154255e-05,
+      "loss": 0.0955,
+      "mean_token_accuracy": 0.965166175365448,
+      "num_tokens": 1027085.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9812434613704681,
+      "epoch": 0.7361376673040153,
+      "grad_norm": 0.5556644201278687,
+      "learning_rate": 7.488580919658727e-05,
+      "loss": 0.1254,
+      "mean_token_accuracy": 0.9521081447601318,
+      "num_tokens": 1040367.0,
+      "step": 770
+    },
+    {
+      "entropy": 1.0117783784866332,
+      "epoch": 0.745697896749522,
+      "grad_norm": 0.8452925086021423,
+      "learning_rate": 7.419706333458201e-05,
+      "loss": 0.1115,
+      "mean_token_accuracy": 0.9579765915870666,
+      "num_tokens": 1054076.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9893367469310761,
+      "epoch": 0.7552581261950286,
+      "grad_norm": 0.533435046672821,
+      "learning_rate": 7.350226883413413e-05,
+      "loss": 0.095,
+      "mean_token_accuracy": 0.9614646136760712,
+      "num_tokens": 1068530.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9841218173503876,
+      "epoch": 0.7648183556405354,
+      "grad_norm": 0.7473505139350891,
+      "learning_rate": 7.280159937585386e-05,
+      "loss": 0.1091,
+      "mean_token_accuracy": 0.9589401960372925,
+      "num_tokens": 1082342.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.980844521522522,
+      "epoch": 0.7743785850860421,
+      "grad_norm": 0.48087120056152344,
+      "learning_rate": 7.209523010893853e-05,
+      "loss": 0.109,
+      "mean_token_accuracy": 0.9598594844341278,
+      "num_tokens": 1096496.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.9954476654529572,
+      "epoch": 0.7839388145315488,
+      "grad_norm": 0.4558711349964142,
+      "learning_rate": 7.138333760738987e-05,
+      "loss": 0.1147,
+      "mean_token_accuracy": 0.9550995290279388,
+      "num_tokens": 1109906.0,
+      "step": 820
+    },
+    {
+      "entropy": 1.0140338003635407,
+      "epoch": 0.7934990439770554,
+      "grad_norm": 0.42530539631843567,
+      "learning_rate": 7.066609982587501e-05,
+      "loss": 0.1257,
+      "mean_token_accuracy": 0.9520308792591095,
+      "num_tokens": 1123170.0,
+      "step": 830
+    },
+    {
+      "entropy": 1.0067408978939056,
+      "epoch": 0.8030592734225621,
+      "grad_norm": 0.4590071737766266,
+      "learning_rate": 6.994369605524253e-05,
+      "loss": 0.1122,
+      "mean_token_accuracy": 0.9571678459644317,
+      "num_tokens": 1136503.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9648548305034638,
+      "epoch": 0.8126195028680688,
+      "grad_norm": 0.7105207443237305,
+      "learning_rate": 6.921630687770427e-05,
+      "loss": 0.0949,
+      "mean_token_accuracy": 0.9639371693134308,
+      "num_tokens": 1149719.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.9722950994968415,
+      "epoch": 0.8221797323135756,
+      "grad_norm": 0.5115017294883728,
+      "learning_rate": 6.848411412169461e-05,
+      "loss": 0.1004,
+      "mean_token_accuracy": 0.9604472815990448,
+      "num_tokens": 1163320.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9869464337825775,
+      "epoch": 0.8317399617590823,
+      "grad_norm": 0.635979413986206,
+      "learning_rate": 6.774730081641799e-05,
+      "loss": 0.1006,
+      "mean_token_accuracy": 0.9617601573467255,
+      "num_tokens": 1177204.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.9726413011550903,
+      "epoch": 0.8413001912045889,
+      "grad_norm": 0.7066209316253662,
+      "learning_rate": 6.70060511460963e-05,
+      "loss": 0.106,
+      "mean_token_accuracy": 0.9593612909317016,
+      "num_tokens": 1190225.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9729720950126648,
+      "epoch": 0.8508604206500956,
+      "grad_norm": 0.5181801915168762,
+      "learning_rate": 6.626055040392774e-05,
+      "loss": 0.0986,
+      "mean_token_accuracy": 0.9632611274719238,
+      "num_tokens": 1204511.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9663226544857025,
+      "epoch": 0.8604206500956023,
+      "grad_norm": 0.5951371192932129,
+      "learning_rate": 6.551098494576819e-05,
+      "loss": 0.1028,
+      "mean_token_accuracy": 0.961613553762436,
+      "num_tokens": 1217217.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9670190632343292,
+      "epoch": 0.869980879541109,
+      "grad_norm": 0.6755188703536987,
+      "learning_rate": 6.475754214354715e-05,
+      "loss": 0.101,
+      "mean_token_accuracy": 0.9604500949382782,
+      "num_tokens": 1230535.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.973962277173996,
+      "epoch": 0.8795411089866156,
+      "grad_norm": 0.669779896736145,
+      "learning_rate": 6.400041033842965e-05,
+      "loss": 0.1167,
+      "mean_token_accuracy": 0.9523392379283905,
+      "num_tokens": 1244121.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9805286586284637,
+      "epoch": 0.8891013384321224,
+      "grad_norm": 0.4764414131641388,
+      "learning_rate": 6.323977879373571e-05,
+      "loss": 0.1147,
+      "mean_token_accuracy": 0.9577437281608582,
+      "num_tokens": 1257972.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.9964150011539459,
+      "epoch": 0.8986615678776291,
+      "grad_norm": 0.8024067878723145,
+      "learning_rate": 6.247583764762959e-05,
+      "loss": 0.1023,
+      "mean_token_accuracy": 0.9585803866386413,
+      "num_tokens": 1271568.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9828290998935699,
+      "epoch": 0.9082217973231358,
+      "grad_norm": 1.0372259616851807,
+      "learning_rate": 6.170877786559001e-05,
+      "loss": 0.1014,
+      "mean_token_accuracy": 0.9595728158950806,
+      "num_tokens": 1285826.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.9621906101703643,
+      "epoch": 0.9177820267686424,
+      "grad_norm": 0.5696820616722107,
+      "learning_rate": 6.0938791192673895e-05,
+      "loss": 0.0996,
+      "mean_token_accuracy": 0.960381704568863,
+      "num_tokens": 1299889.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9672249853610992,
+      "epoch": 0.9273422562141491,
+      "grad_norm": 0.6769490838050842,
+      "learning_rate": 6.016607010558492e-05,
+      "loss": 0.0899,
+      "mean_token_accuracy": 0.9650658905506134,
+      "num_tokens": 1313610.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9467990279197693,
+      "epoch": 0.9369024856596558,
+      "grad_norm": 0.6738393306732178,
+      "learning_rate": 5.939080776455942e-05,
+      "loss": 0.1001,
+      "mean_token_accuracy": 0.9624602735042572,
+      "num_tokens": 1327185.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.9557888090610505,
+      "epoch": 0.9464627151051626,
+      "grad_norm": 0.7844521403312683,
+      "learning_rate": 5.861319796508139e-05,
+      "loss": 0.1127,
+      "mean_token_accuracy": 0.9526202499866485,
+      "num_tokens": 1340176.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.9579505145549774,
+      "epoch": 0.9560229445506692,
+      "grad_norm": 0.9583166241645813,
+      "learning_rate": 5.783343508943849e-05,
+      "loss": 0.1068,
+      "mean_token_accuracy": 0.9606858670711518,
+      "num_tokens": 1354053.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.960960054397583,
+      "epoch": 0.9655831739961759,
+      "grad_norm": 0.5237046480178833,
+      "learning_rate": 5.7051714058131654e-05,
+      "loss": 0.1053,
+      "mean_token_accuracy": 0.957349419593811,
+      "num_tokens": 1367199.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.928047263622284,
+      "epoch": 0.9751434034416826,
+      "grad_norm": 0.6380038261413574,
+      "learning_rate": 5.626823028114991e-05,
+      "loss": 0.0779,
+      "mean_token_accuracy": 0.9705170750617981,
+      "num_tokens": 1380205.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9575015842914582,
+      "epoch": 0.9847036328871893,
+      "grad_norm": 0.8508949875831604,
+      "learning_rate": 5.548317960912305e-05,
+      "loss": 0.1034,
+      "mean_token_accuracy": 0.9567257583141326,
+      "num_tokens": 1392623.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9625794291496277,
+      "epoch": 0.994263862332696,
+      "grad_norm": 0.70963454246521,
+      "learning_rate": 5.4696758284363956e-05,
+      "loss": 0.0937,
+      "mean_token_accuracy": 0.9654395580291748,
+      "num_tokens": 1406093.0,
+      "step": 1040
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2092,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.100958587515187e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:048511de437099a945a43b3c66145c65e4f57085de0a4be01901fe7db67c4d3c
+size 6417

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/README.md

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "gate_proj",
+    "up_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_s3/checkpoint-2092/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d30e807e5f187d3e0d474e0334cb2b2698b9e26c3d527739307116628323caa
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/added_tokens.json

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/chat_template.jinja

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/special_tokens_map.json

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/tokenizer.json

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

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/trainer_state.json

@@ -0,0 +1,2124 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 2092,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.1632622241973878,
+      "epoch": 0.009560229445506692,
+      "grad_norm": 0.6969931125640869,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2719,
+      "mean_token_accuracy": 0.9138259589672089,
+      "num_tokens": 13150.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1633504509925843,
+      "epoch": 0.019120458891013385,
+      "grad_norm": 0.5006016492843628,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2592,
+      "mean_token_accuracy": 0.915404063463211,
+      "num_tokens": 26136.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1590815901756286,
+      "epoch": 0.028680688336520075,
+      "grad_norm": 0.4876864552497864,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.2211,
+      "mean_token_accuracy": 0.9319653451442719,
+      "num_tokens": 39207.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.1542185068130493,
+      "epoch": 0.03824091778202677,
+      "grad_norm": 0.5451372861862183,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1901,
+      "mean_token_accuracy": 0.9393570721149445,
+      "num_tokens": 52869.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.1529746413230897,
+      "epoch": 0.04780114722753346,
+      "grad_norm": 0.8673158288002014,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.2024,
+      "mean_token_accuracy": 0.9328486382961273,
+      "num_tokens": 66622.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.1270763635635377,
+      "epoch": 0.05736137667304015,
+      "grad_norm": 0.636655330657959,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1661,
+      "mean_token_accuracy": 0.9430084943771362,
+      "num_tokens": 80175.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.1287935256958008,
+      "epoch": 0.06692160611854685,
+      "grad_norm": 0.6914967894554138,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1946,
+      "mean_token_accuracy": 0.9291220664978027,
+      "num_tokens": 94356.0,
+      "step": 70
+    },
+    {
+      "entropy": 1.0964631199836732,
+      "epoch": 0.07648183556405354,
+      "grad_norm": 0.7391988039016724,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1519,
+      "mean_token_accuracy": 0.9491405785083771,
+      "num_tokens": 108716.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.0623166620731355,
+      "epoch": 0.08604206500956023,
+      "grad_norm": 0.6298117637634277,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1546,
+      "mean_token_accuracy": 0.9406298100948334,
+      "num_tokens": 122182.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0688253164291381,
+      "epoch": 0.09560229445506692,
+      "grad_norm": 0.6803618669509888,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1438,
+      "mean_token_accuracy": 0.9495500445365905,
+      "num_tokens": 135713.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0644982933998108,
+      "epoch": 0.10516252390057361,
+      "grad_norm": 0.8620546460151672,
+      "learning_rate": 9.999900008621665e-05,
+      "loss": 0.1627,
+      "mean_token_accuracy": 0.9426994383335113,
+      "num_tokens": 149456.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.0501671850681304,
+      "epoch": 0.1147227533460803,
+      "grad_norm": 0.4802723824977875,
+      "learning_rate": 9.998775151544478e-05,
+      "loss": 0.146,
+      "mean_token_accuracy": 0.943053936958313,
+      "num_tokens": 163062.0,
+      "step": 120
+    },
+    {
+      "entropy": 1.0445132613182069,
+      "epoch": 0.124282982791587,
+      "grad_norm": 0.7362843751907349,
+      "learning_rate": 9.996400730289628e-05,
+      "loss": 0.1668,
+      "mean_token_accuracy": 0.9377037048339844,
+      "num_tokens": 176574.0,
+      "step": 130
+    },
+    {
+      "entropy": 1.034316086769104,
+      "epoch": 0.1338432122370937,
+      "grad_norm": 0.630389392375946,
+      "learning_rate": 9.992777338400864e-05,
+      "loss": 0.1653,
+      "mean_token_accuracy": 0.9353829383850097,
+      "num_tokens": 191164.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0108101904392242,
+      "epoch": 0.14340344168260039,
+      "grad_norm": 0.7097353339195251,
+      "learning_rate": 9.987905881632213e-05,
+      "loss": 0.1488,
+      "mean_token_accuracy": 0.9435539603233337,
+      "num_tokens": 204805.0,
+      "step": 150
+    },
+    {
+      "entropy": 1.0055140674114227,
+      "epoch": 0.15296367112810708,
+      "grad_norm": 0.6539037227630615,
+      "learning_rate": 9.981787577721553e-05,
+      "loss": 0.1308,
+      "mean_token_accuracy": 0.9497624516487122,
+      "num_tokens": 218021.0,
+      "step": 160
+    },
+    {
+      "entropy": 0.9890082895755767,
+      "epoch": 0.16252390057361377,
+      "grad_norm": 0.47821044921875,
+      "learning_rate": 9.974423956086228e-05,
+      "loss": 0.1234,
+      "mean_token_accuracy": 0.9515690505504608,
+      "num_tokens": 231015.0,
+      "step": 170
+    },
+    {
+      "entropy": 0.9904860436916352,
+      "epoch": 0.17208413001912046,
+      "grad_norm": 0.6690865159034729,
+      "learning_rate": 9.965816857440717e-05,
+      "loss": 0.143,
+      "mean_token_accuracy": 0.942569500207901,
+      "num_tokens": 244021.0,
+      "step": 180
+    },
+    {
+      "entropy": 1.0002257645130157,
+      "epoch": 0.18164435946462715,
+      "grad_norm": 0.8160389065742493,
+      "learning_rate": 9.955968433336515e-05,
+      "loss": 0.1423,
+      "mean_token_accuracy": 0.9491697907447815,
+      "num_tokens": 256981.0,
+      "step": 190
+    },
+    {
+      "entropy": 0.9900605976581573,
+      "epoch": 0.19120458891013384,
+      "grad_norm": 0.5272551774978638,
+      "learning_rate": 9.944881145624291e-05,
+      "loss": 0.1488,
+      "mean_token_accuracy": 0.9452917218208313,
+      "num_tokens": 270999.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.9831291913986206,
+      "epoch": 0.20076481835564053,
+      "grad_norm": 0.627918541431427,
+      "learning_rate": 9.932557765838504e-05,
+      "loss": 0.1196,
+      "mean_token_accuracy": 0.9600761473178864,
+      "num_tokens": 283995.0,
+      "step": 210
+    },
+    {
+      "entropy": 0.9906052827835083,
+      "epoch": 0.21032504780114722,
+      "grad_norm": 0.7424314618110657,
+      "learning_rate": 9.919001374504569e-05,
+      "loss": 0.138,
+      "mean_token_accuracy": 0.9490589261054992,
+      "num_tokens": 297434.0,
+      "step": 220
+    },
+    {
+      "entropy": 1.0092378556728363,
+      "epoch": 0.2198852772466539,
+      "grad_norm": 0.5505907535552979,
+      "learning_rate": 9.904215360368826e-05,
+      "loss": 0.1277,
+      "mean_token_accuracy": 0.9500136435031891,
+      "num_tokens": 310688.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.0368991553783418,
+      "epoch": 0.2294455066921606,
+      "grad_norm": 1.0577473640441895,
+      "learning_rate": 9.888203419551432e-05,
+      "loss": 0.1594,
+      "mean_token_accuracy": 0.9426211178302765,
+      "num_tokens": 324594.0,
+      "step": 240
+    },
+    {
+      "entropy": 1.062998366355896,
+      "epoch": 0.2390057361376673,
+      "grad_norm": 0.8393595218658447,
+      "learning_rate": 9.870969554622426e-05,
+      "loss": 0.1579,
+      "mean_token_accuracy": 0.9380222499370575,
+      "num_tokens": 338003.0,
+      "step": 250
+    },
+    {
+      "entropy": 1.0535745024681091,
+      "epoch": 0.248565965583174,
+      "grad_norm": 0.4548805356025696,
+      "learning_rate": 9.852518073601194e-05,
+      "loss": 0.1363,
+      "mean_token_accuracy": 0.9484048664569855,
+      "num_tokens": 351394.0,
+      "step": 260
+    },
+    {
+      "entropy": 1.0056042432785035,
+      "epoch": 0.25812619502868067,
+      "grad_norm": 0.4833658039569855,
+      "learning_rate": 9.832853588879575e-05,
+      "loss": 0.1282,
+      "mean_token_accuracy": 0.9508345782756805,
+      "num_tokens": 365081.0,
+      "step": 270
+    },
+    {
+      "entropy": 1.0188749372959136,
+      "epoch": 0.2676864244741874,
+      "grad_norm": 0.5572522282600403,
+      "learning_rate": 9.811981016068881e-05,
+      "loss": 0.1328,
+      "mean_token_accuracy": 0.9506715357303619,
+      "num_tokens": 378518.0,
+      "step": 280
+    },
+    {
+      "entropy": 1.0242417335510254,
+      "epoch": 0.27724665391969405,
+      "grad_norm": 0.9603186845779419,
+      "learning_rate": 9.78990557277112e-05,
+      "loss": 0.1311,
+      "mean_token_accuracy": 0.9476073980331421,
+      "num_tokens": 391993.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0456064581871032,
+      "epoch": 0.28680688336520077,
+      "grad_norm": 0.6591246724128723,
+      "learning_rate": 9.766632777274738e-05,
+      "loss": 0.1115,
+      "mean_token_accuracy": 0.9543036341667175,
+      "num_tokens": 405246.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0853006482124328,
+      "epoch": 0.29636711281070743,
+      "grad_norm": 0.823593258857727,
+      "learning_rate": 9.742168447175182e-05,
+      "loss": 0.1259,
+      "mean_token_accuracy": 0.9509813964366913,
+      "num_tokens": 418524.0,
+      "step": 310
+    },
+    {
+      "entropy": 1.041166114807129,
+      "epoch": 0.30592734225621415,
+      "grad_norm": 0.6226829290390015,
+      "learning_rate": 9.716518697920655e-05,
+      "loss": 0.1442,
+      "mean_token_accuracy": 0.9461847722530365,
+      "num_tokens": 432202.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0213975965976716,
+      "epoch": 0.3154875717017208,
+      "grad_norm": 0.42342594265937805,
+      "learning_rate": 9.689689941283417e-05,
+      "loss": 0.1302,
+      "mean_token_accuracy": 0.9512553572654724,
+      "num_tokens": 445903.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.043001300096512,
+      "epoch": 0.32504780114722753,
+      "grad_norm": 0.5154736638069153,
+      "learning_rate": 9.661688883756996e-05,
+      "loss": 0.1231,
+      "mean_token_accuracy": 0.9551738917827606,
+      "num_tokens": 460277.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0378517508506775,
+      "epoch": 0.33460803059273425,
+      "grad_norm": 0.6703522801399231,
+      "learning_rate": 9.632522524879756e-05,
+      "loss": 0.129,
+      "mean_token_accuracy": 0.9532712638378144,
+      "num_tokens": 473892.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.0317105233669281,
+      "epoch": 0.3441682600382409,
+      "grad_norm": 0.6474915742874146,
+      "learning_rate": 9.602198155485176e-05,
+      "loss": 0.1076,
+      "mean_token_accuracy": 0.9581969559192658,
+      "num_tokens": 487145.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.0285814940929412,
+      "epoch": 0.35372848948374763,
+      "grad_norm": 0.7044402360916138,
+      "learning_rate": 9.570723355879342e-05,
+      "loss": 0.1227,
+      "mean_token_accuracy": 0.9547486007213593,
+      "num_tokens": 500481.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.055726957321167,
+      "epoch": 0.3632887189292543,
+      "grad_norm": 0.6242275238037109,
+      "learning_rate": 9.538105993946068e-05,
+      "loss": 0.127,
+      "mean_token_accuracy": 0.9565758466720581,
+      "num_tokens": 513817.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0444733262062074,
+      "epoch": 0.372848948374761,
+      "grad_norm": 0.47244662046432495,
+      "learning_rate": 9.504354223180118e-05,
+      "loss": 0.1302,
+      "mean_token_accuracy": 0.9526951789855957,
+      "num_tokens": 527250.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.0365854144096374,
+      "epoch": 0.3824091778202677,
+      "grad_norm": 0.4104255735874176,
+      "learning_rate": 9.469476480649056e-05,
+      "loss": 0.1257,
+      "mean_token_accuracy": 0.9535662651062011,
+      "num_tokens": 541558.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.025916337966919,
+      "epoch": 0.3919694072657744,
+      "grad_norm": 0.6373872756958008,
+      "learning_rate": 9.433481484884189e-05,
+      "loss": 0.1095,
+      "mean_token_accuracy": 0.9600341856479645,
+      "num_tokens": 554540.0,
+      "step": 410
+    },
+    {
+      "entropy": 1.040219533443451,
+      "epoch": 0.40152963671128106,
+      "grad_norm": 0.5284200310707092,
+      "learning_rate": 9.396378233701167e-05,
+      "loss": 0.1177,
+      "mean_token_accuracy": 0.9509544730186462,
+      "num_tokens": 568221.0,
+      "step": 420
+    },
+    {
+      "entropy": 1.0495393812656402,
+      "epoch": 0.4110898661567878,
+      "grad_norm": 0.6208754777908325,
+      "learning_rate": 9.358176001950759e-05,
+      "loss": 0.1277,
+      "mean_token_accuracy": 0.948467081785202,
+      "num_tokens": 581721.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.0427401065826416,
+      "epoch": 0.42065009560229444,
+      "grad_norm": 0.9366323351860046,
+      "learning_rate": 9.318884339200378e-05,
+      "loss": 0.1162,
+      "mean_token_accuracy": 0.9557548820972442,
+      "num_tokens": 595120.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0672527194023131,
+      "epoch": 0.43021032504780116,
+      "grad_norm": 0.5693362951278687,
+      "learning_rate": 9.278513067346941e-05,
+      "loss": 0.1246,
+      "mean_token_accuracy": 0.9521466493606567,
+      "num_tokens": 608361.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0667488336563111,
+      "epoch": 0.4397705544933078,
+      "grad_norm": 1.004165530204773,
+      "learning_rate": 9.237072278161652e-05,
+      "loss": 0.1211,
+      "mean_token_accuracy": 0.9553246557712555,
+      "num_tokens": 621843.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.055682933330536,
+      "epoch": 0.44933078393881454,
+      "grad_norm": 0.7205134034156799,
+      "learning_rate": 9.194572330767314e-05,
+      "loss": 0.1162,
+      "mean_token_accuracy": 0.9523303210735321,
+      "num_tokens": 635552.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.059775161743164,
+      "epoch": 0.4588910133843212,
+      "grad_norm": 0.5044583082199097,
+      "learning_rate": 9.151023849048818e-05,
+      "loss": 0.1336,
+      "mean_token_accuracy": 0.9506153464317322,
+      "num_tokens": 648523.0,
+      "step": 480
+    },
+    {
+      "entropy": 1.0490445971488953,
+      "epoch": 0.4684512428298279,
+      "grad_norm": 0.34968268871307373,
+      "learning_rate": 9.106437718997452e-05,
+      "loss": 0.1088,
+      "mean_token_accuracy": 0.9602180898189545,
+      "num_tokens": 661850.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0310296952724456,
+      "epoch": 0.4780114722753346,
+      "grad_norm": 1.0173206329345703,
+      "learning_rate": 9.060825085989682e-05,
+      "loss": 0.1049,
+      "mean_token_accuracy": 0.9578554153442382,
+      "num_tokens": 675322.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.047879695892334,
+      "epoch": 0.4875717017208413,
+      "grad_norm": 0.7152472138404846,
+      "learning_rate": 9.014197352001097e-05,
+      "loss": 0.1256,
+      "mean_token_accuracy": 0.95099977850914,
+      "num_tokens": 688704.0,
+      "step": 510
+    },
+    {
+      "entropy": 1.0453792810440063,
+      "epoch": 0.497131931166348,
+      "grad_norm": 0.8198304772377014,
+      "learning_rate": 8.966566172756213e-05,
+      "loss": 0.1305,
+      "mean_token_accuracy": 0.953268188238144,
+      "num_tokens": 702579.0,
+      "step": 520
+    },
+    {
+      "entropy": 1.036444890499115,
+      "epoch": 0.5066921606118547,
+      "grad_norm": 0.5910372734069824,
+      "learning_rate": 8.91794345481484e-05,
+      "loss": 0.1217,
+      "mean_token_accuracy": 0.952339518070221,
+      "num_tokens": 716160.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.034623110294342,
+      "epoch": 0.5162523900573613,
+      "grad_norm": 0.9556732177734375,
+      "learning_rate": 8.868341352595755e-05,
+      "loss": 0.1276,
+      "mean_token_accuracy": 0.9504207968711853,
+      "num_tokens": 729408.0,
+      "step": 540
+    },
+    {
+      "entropy": 1.0424424648284911,
+      "epoch": 0.5258126195028681,
+      "grad_norm": 0.5328360795974731,
+      "learning_rate": 8.817772265338409e-05,
+      "loss": 0.1221,
+      "mean_token_accuracy": 0.9487152993679047,
+      "num_tokens": 742824.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.032740694284439,
+      "epoch": 0.5353728489483748,
+      "grad_norm": 0.9654290080070496,
+      "learning_rate": 8.766248834003441e-05,
+      "loss": 0.1212,
+      "mean_token_accuracy": 0.9530553936958313,
+      "num_tokens": 756778.0,
+      "step": 560
+    },
+    {
+      "entropy": 1.036280381679535,
+      "epoch": 0.5449330783938815,
+      "grad_norm": 0.5315765142440796,
+      "learning_rate": 8.713783938112768e-05,
+      "loss": 0.1214,
+      "mean_token_accuracy": 0.9498340725898743,
+      "num_tokens": 770342.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.0438485622406006,
+      "epoch": 0.5544933078393881,
+      "grad_norm": 0.4137975573539734,
+      "learning_rate": 8.660390692530026e-05,
+      "loss": 0.1172,
+      "mean_token_accuracy": 0.9532914280891418,
+      "num_tokens": 784398.0,
+      "step": 580
+    },
+    {
+      "entropy": 1.0298362493515014,
+      "epoch": 0.5640535372848948,
+      "grad_norm": 0.6320862770080566,
+      "learning_rate": 8.606082444182216e-05,
+      "loss": 0.1225,
+      "mean_token_accuracy": 0.9563343822956085,
+      "num_tokens": 798017.0,
+      "step": 590
+    },
+    {
+      "entropy": 1.010947346687317,
+      "epoch": 0.5736137667304015,
+      "grad_norm": 0.837810218334198,
+      "learning_rate": 8.550872768723292e-05,
+      "loss": 0.1175,
+      "mean_token_accuracy": 0.9552404284477234,
+      "num_tokens": 811336.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.9999268174171447,
+      "epoch": 0.5831739961759083,
+      "grad_norm": 0.6310060620307922,
+      "learning_rate": 8.494775467140627e-05,
+      "loss": 0.1253,
+      "mean_token_accuracy": 0.9503907263278961,
+      "num_tokens": 824727.0,
+      "step": 610
+    },
+    {
+      "entropy": 1.0030306220054626,
+      "epoch": 0.5927342256214149,
+      "grad_norm": 0.5004256963729858,
+      "learning_rate": 8.4378045623051e-05,
+      "loss": 0.131,
+      "mean_token_accuracy": 0.9487506031990052,
+      "num_tokens": 838105.0,
+      "step": 620
+    },
+    {
+      "entropy": 0.9925207674503327,
+      "epoch": 0.6022944550669216,
+      "grad_norm": 0.9587653279304504,
+      "learning_rate": 8.379974295465758e-05,
+      "loss": 0.1026,
+      "mean_token_accuracy": 0.9612509369850158,
+      "num_tokens": 851280.0,
+      "step": 630
+    },
+    {
+      "entropy": 1.0160180270671844,
+      "epoch": 0.6118546845124283,
+      "grad_norm": 0.5568580627441406,
+      "learning_rate": 8.321299122689863e-05,
+      "loss": 0.1112,
+      "mean_token_accuracy": 0.9538475096225738,
+      "num_tokens": 864939.0,
+      "step": 640
+    },
+    {
+      "entropy": 1.0023184657096862,
+      "epoch": 0.621414913957935,
+      "grad_norm": 1.031872272491455,
+      "learning_rate": 8.261793711249254e-05,
+      "loss": 0.1213,
+      "mean_token_accuracy": 0.95554478764534,
+      "num_tokens": 878886.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.9941893517971039,
+      "epoch": 0.6309751434034416,
+      "grad_norm": 0.7980345487594604,
+      "learning_rate": 8.201472935953916e-05,
+      "loss": 0.0912,
+      "mean_token_accuracy": 0.9684767663478852,
+      "num_tokens": 892413.0,
+      "step": 660
+    },
+    {
+      "entropy": 1.0040182232856751,
+      "epoch": 0.6405353728489483,
+      "grad_norm": 0.6786800026893616,
+      "learning_rate": 8.140351875433649e-05,
+      "loss": 0.1148,
+      "mean_token_accuracy": 0.9537683606147767,
+      "num_tokens": 906485.0,
+      "step": 670
+    },
+    {
+      "entropy": 1.0011967778205872,
+      "epoch": 0.6500956022944551,
+      "grad_norm": 0.8331853151321411,
+      "learning_rate": 8.07844580836881e-05,
+      "loss": 0.1117,
+      "mean_token_accuracy": 0.9556680023670197,
+      "num_tokens": 920020.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9943797826766968,
+      "epoch": 0.6596558317399618,
+      "grad_norm": 1.0075759887695312,
+      "learning_rate": 8.015770209671044e-05,
+      "loss": 0.1046,
+      "mean_token_accuracy": 0.9595177710056305,
+      "num_tokens": 933577.0,
+      "step": 690
+    },
+    {
+      "entropy": 0.9939425528049469,
+      "epoch": 0.6692160611854685,
+      "grad_norm": 0.46297088265419006,
+      "learning_rate": 7.952340746614946e-05,
+      "loss": 0.102,
+      "mean_token_accuracy": 0.9599328935146332,
+      "num_tokens": 946367.0,
+      "step": 700
+    },
+    {
+      "entropy": 1.0044083833694457,
+      "epoch": 0.6787762906309751,
+      "grad_norm": 0.4464632272720337,
+      "learning_rate": 7.888173274921655e-05,
+      "loss": 0.1177,
+      "mean_token_accuracy": 0.957087230682373,
+      "num_tokens": 960252.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9848935604095459,
+      "epoch": 0.6883365200764818,
+      "grad_norm": 0.6167030334472656,
+      "learning_rate": 7.823283834795336e-05,
+      "loss": 0.098,
+      "mean_token_accuracy": 0.960848867893219,
+      "num_tokens": 973491.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9874480962753296,
+      "epoch": 0.6978967495219885,
+      "grad_norm": 0.4817270338535309,
+      "learning_rate": 7.75768864691354e-05,
+      "loss": 0.1207,
+      "mean_token_accuracy": 0.9501777410507202,
+      "num_tokens": 986879.0,
+      "step": 730
+    },
+    {
+      "entropy": 1.0066561222076416,
+      "epoch": 0.7074569789674953,
+      "grad_norm": 0.6590946316719055,
+      "learning_rate": 7.691404108372461e-05,
+      "loss": 0.1102,
+      "mean_token_accuracy": 0.960069864988327,
+      "num_tokens": 1000272.0,
+      "step": 740
+    },
+    {
+      "entropy": 1.003525936603546,
+      "epoch": 0.7170172084130019,
+      "grad_norm": 0.8013182282447815,
+      "learning_rate": 7.624446788588085e-05,
+      "loss": 0.1205,
+      "mean_token_accuracy": 0.955094313621521,
+      "num_tokens": 1013908.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.9959373891353607,
+      "epoch": 0.7265774378585086,
+      "grad_norm": 0.7129849791526794,
+      "learning_rate": 7.556833425154255e-05,
+      "loss": 0.0955,
+      "mean_token_accuracy": 0.965166175365448,
+      "num_tokens": 1027085.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9812434613704681,
+      "epoch": 0.7361376673040153,
+      "grad_norm": 0.5556644201278687,
+      "learning_rate": 7.488580919658727e-05,
+      "loss": 0.1254,
+      "mean_token_accuracy": 0.9521081447601318,
+      "num_tokens": 1040367.0,
+      "step": 770
+    },
+    {
+      "entropy": 1.0117783784866332,
+      "epoch": 0.745697896749522,
+      "grad_norm": 0.8452925086021423,
+      "learning_rate": 7.419706333458201e-05,
+      "loss": 0.1115,
+      "mean_token_accuracy": 0.9579765915870666,
+      "num_tokens": 1054076.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9893367469310761,
+      "epoch": 0.7552581261950286,
+      "grad_norm": 0.533435046672821,
+      "learning_rate": 7.350226883413413e-05,
+      "loss": 0.095,
+      "mean_token_accuracy": 0.9614646136760712,
+      "num_tokens": 1068530.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9841218173503876,
+      "epoch": 0.7648183556405354,
+      "grad_norm": 0.7473505139350891,
+      "learning_rate": 7.280159937585386e-05,
+      "loss": 0.1091,
+      "mean_token_accuracy": 0.9589401960372925,
+      "num_tokens": 1082342.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.980844521522522,
+      "epoch": 0.7743785850860421,
+      "grad_norm": 0.48087120056152344,
+      "learning_rate": 7.209523010893853e-05,
+      "loss": 0.109,
+      "mean_token_accuracy": 0.9598594844341278,
+      "num_tokens": 1096496.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.9954476654529572,
+      "epoch": 0.7839388145315488,
+      "grad_norm": 0.4558711349964142,
+      "learning_rate": 7.138333760738987e-05,
+      "loss": 0.1147,
+      "mean_token_accuracy": 0.9550995290279388,
+      "num_tokens": 1109906.0,
+      "step": 820
+    },
+    {
+      "entropy": 1.0140338003635407,
+      "epoch": 0.7934990439770554,
+      "grad_norm": 0.42530539631843567,
+      "learning_rate": 7.066609982587501e-05,
+      "loss": 0.1257,
+      "mean_token_accuracy": 0.9520308792591095,
+      "num_tokens": 1123170.0,
+      "step": 830
+    },
+    {
+      "entropy": 1.0067408978939056,
+      "epoch": 0.8030592734225621,
+      "grad_norm": 0.4590071737766266,
+      "learning_rate": 6.994369605524253e-05,
+      "loss": 0.1122,
+      "mean_token_accuracy": 0.9571678459644317,
+      "num_tokens": 1136503.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9648548305034638,
+      "epoch": 0.8126195028680688,
+      "grad_norm": 0.7105207443237305,
+      "learning_rate": 6.921630687770427e-05,
+      "loss": 0.0949,
+      "mean_token_accuracy": 0.9639371693134308,
+      "num_tokens": 1149719.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.9722950994968415,
+      "epoch": 0.8221797323135756,
+      "grad_norm": 0.5115017294883728,
+      "learning_rate": 6.848411412169461e-05,
+      "loss": 0.1004,
+      "mean_token_accuracy": 0.9604472815990448,
+      "num_tokens": 1163320.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9869464337825775,
+      "epoch": 0.8317399617590823,
+      "grad_norm": 0.635979413986206,
+      "learning_rate": 6.774730081641799e-05,
+      "loss": 0.1006,
+      "mean_token_accuracy": 0.9617601573467255,
+      "num_tokens": 1177204.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.9726413011550903,
+      "epoch": 0.8413001912045889,
+      "grad_norm": 0.7066209316253662,
+      "learning_rate": 6.70060511460963e-05,
+      "loss": 0.106,
+      "mean_token_accuracy": 0.9593612909317016,
+      "num_tokens": 1190225.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9729720950126648,
+      "epoch": 0.8508604206500956,
+      "grad_norm": 0.5181801915168762,
+      "learning_rate": 6.626055040392774e-05,
+      "loss": 0.0986,
+      "mean_token_accuracy": 0.9632611274719238,
+      "num_tokens": 1204511.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9663226544857025,
+      "epoch": 0.8604206500956023,
+      "grad_norm": 0.5951371192932129,
+      "learning_rate": 6.551098494576819e-05,
+      "loss": 0.1028,
+      "mean_token_accuracy": 0.961613553762436,
+      "num_tokens": 1217217.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9670190632343292,
+      "epoch": 0.869980879541109,
+      "grad_norm": 0.6755188703536987,
+      "learning_rate": 6.475754214354715e-05,
+      "loss": 0.101,
+      "mean_token_accuracy": 0.9604500949382782,
+      "num_tokens": 1230535.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.973962277173996,
+      "epoch": 0.8795411089866156,
+      "grad_norm": 0.669779896736145,
+      "learning_rate": 6.400041033842965e-05,
+      "loss": 0.1167,
+      "mean_token_accuracy": 0.9523392379283905,
+      "num_tokens": 1244121.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9805286586284637,
+      "epoch": 0.8891013384321224,
+      "grad_norm": 0.4764414131641388,
+      "learning_rate": 6.323977879373571e-05,
+      "loss": 0.1147,
+      "mean_token_accuracy": 0.9577437281608582,
+      "num_tokens": 1257972.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.9964150011539459,
+      "epoch": 0.8986615678776291,
+      "grad_norm": 0.8024067878723145,
+      "learning_rate": 6.247583764762959e-05,
+      "loss": 0.1023,
+      "mean_token_accuracy": 0.9585803866386413,
+      "num_tokens": 1271568.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9828290998935699,
+      "epoch": 0.9082217973231358,
+      "grad_norm": 1.0372259616851807,
+      "learning_rate": 6.170877786559001e-05,
+      "loss": 0.1014,
+      "mean_token_accuracy": 0.9595728158950806,
+      "num_tokens": 1285826.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.9621906101703643,
+      "epoch": 0.9177820267686424,
+      "grad_norm": 0.5696820616722107,
+      "learning_rate": 6.0938791192673895e-05,
+      "loss": 0.0996,
+      "mean_token_accuracy": 0.960381704568863,
+      "num_tokens": 1299889.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9672249853610992,
+      "epoch": 0.9273422562141491,
+      "grad_norm": 0.6769490838050842,
+      "learning_rate": 6.016607010558492e-05,
+      "loss": 0.0899,
+      "mean_token_accuracy": 0.9650658905506134,
+      "num_tokens": 1313610.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9467990279197693,
+      "epoch": 0.9369024856596558,
+      "grad_norm": 0.6738393306732178,
+      "learning_rate": 5.939080776455942e-05,
+      "loss": 0.1001,
+      "mean_token_accuracy": 0.9624602735042572,
+      "num_tokens": 1327185.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.9557888090610505,
+      "epoch": 0.9464627151051626,
+      "grad_norm": 0.7844521403312683,
+      "learning_rate": 5.861319796508139e-05,
+      "loss": 0.1127,
+      "mean_token_accuracy": 0.9526202499866485,
+      "num_tokens": 1340176.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.9579505145549774,
+      "epoch": 0.9560229445506692,
+      "grad_norm": 0.9583166241645813,
+      "learning_rate": 5.783343508943849e-05,
+      "loss": 0.1068,
+      "mean_token_accuracy": 0.9606858670711518,
+      "num_tokens": 1354053.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.960960054397583,
+      "epoch": 0.9655831739961759,
+      "grad_norm": 0.5237046480178833,
+      "learning_rate": 5.7051714058131654e-05,
+      "loss": 0.1053,
+      "mean_token_accuracy": 0.957349419593811,
+      "num_tokens": 1367199.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.928047263622284,
+      "epoch": 0.9751434034416826,
+      "grad_norm": 0.6380038261413574,
+      "learning_rate": 5.626823028114991e-05,
+      "loss": 0.0779,
+      "mean_token_accuracy": 0.9705170750617981,
+      "num_tokens": 1380205.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9575015842914582,
+      "epoch": 0.9847036328871893,
+      "grad_norm": 0.8508949875831604,
+      "learning_rate": 5.548317960912305e-05,
+      "loss": 0.1034,
+      "mean_token_accuracy": 0.9567257583141326,
+      "num_tokens": 1392623.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9625794291496277,
+      "epoch": 0.994263862332696,
+      "grad_norm": 0.70963454246521,
+      "learning_rate": 5.4696758284363956e-05,
+      "loss": 0.0937,
+      "mean_token_accuracy": 0.9654395580291748,
+      "num_tokens": 1406093.0,
+      "step": 1040
+    },
+    {
+      "entropy": 0.9606178939342499,
+      "epoch": 1.0038240917782026,
+      "grad_norm": 0.5587365031242371,
+      "learning_rate": 5.390916289181318e-05,
+      "loss": 0.083,
+      "mean_token_accuracy": 0.9670810341835022,
+      "num_tokens": 1419395.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.9337595582008362,
+      "epoch": 1.0133843212237095,
+      "grad_norm": 0.3220118284225464,
+      "learning_rate": 5.312059030989779e-05,
+      "loss": 0.0514,
+      "mean_token_accuracy": 0.9836415231227875,
+      "num_tokens": 1432857.0,
+      "step": 1060
+    },
+    {
+      "entropy": 0.9057540893554688,
+      "epoch": 1.022944550669216,
+      "grad_norm": 0.3958684206008911,
+      "learning_rate": 5.233123766131681e-05,
+      "loss": 0.0464,
+      "mean_token_accuracy": 0.98567014336586,
+      "num_tokens": 1445639.0,
+      "step": 1070
+    },
+    {
+      "entropy": 0.8980904698371888,
+      "epoch": 1.0325047801147227,
+      "grad_norm": 0.7858161330223083,
+      "learning_rate": 5.154130226376578e-05,
+      "loss": 0.0532,
+      "mean_token_accuracy": 0.9829789638519287,
+      "num_tokens": 1458461.0,
+      "step": 1080
+    },
+    {
+      "entropy": 0.9239595472812653,
+      "epoch": 1.0420650095602295,
+      "grad_norm": 0.41218942403793335,
+      "learning_rate": 5.0750981580612277e-05,
+      "loss": 0.0534,
+      "mean_token_accuracy": 0.980589771270752,
+      "num_tokens": 1472067.0,
+      "step": 1090
+    },
+    {
+      "entropy": 0.9218643009662628,
+      "epoch": 1.0516252390057361,
+      "grad_norm": 0.2810509502887726,
+      "learning_rate": 4.996047317153538e-05,
+      "loss": 0.0471,
+      "mean_token_accuracy": 0.9843796849250793,
+      "num_tokens": 1486092.0,
+      "step": 1100
+    },
+    {
+      "entropy": 0.9202406406402588,
+      "epoch": 1.0611854684512427,
+      "grad_norm": 0.8550936579704285,
+      "learning_rate": 4.916997464314074e-05,
+      "loss": 0.0506,
+      "mean_token_accuracy": 0.9818832814693451,
+      "num_tokens": 1500233.0,
+      "step": 1110
+    },
+    {
+      "entropy": 0.9084908545017243,
+      "epoch": 1.0707456978967496,
+      "grad_norm": 0.5990253686904907,
+      "learning_rate": 4.8379683599564086e-05,
+      "loss": 0.0456,
+      "mean_token_accuracy": 0.9838834285736084,
+      "num_tokens": 1513865.0,
+      "step": 1120
+    },
+    {
+      "entropy": 0.920438951253891,
+      "epoch": 1.0803059273422562,
+      "grad_norm": 0.3564145863056183,
+      "learning_rate": 4.7589797593075334e-05,
+      "loss": 0.0573,
+      "mean_token_accuracy": 0.9795924425125122,
+      "num_tokens": 1527318.0,
+      "step": 1130
+    },
+    {
+      "entropy": 0.9204782366752624,
+      "epoch": 1.089866156787763,
+      "grad_norm": 0.41145166754722595,
+      "learning_rate": 4.6800514074695626e-05,
+      "loss": 0.0635,
+      "mean_token_accuracy": 0.9826971828937531,
+      "num_tokens": 1540076.0,
+      "step": 1140
+    },
+    {
+      "entropy": 0.9294351637363434,
+      "epoch": 1.0994263862332696,
+      "grad_norm": 0.31233474612236023,
+      "learning_rate": 4.601203034483971e-05,
+      "loss": 0.0551,
+      "mean_token_accuracy": 0.9815062820911408,
+      "num_tokens": 1553989.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.9246800899505615,
+      "epoch": 1.1089866156787762,
+      "grad_norm": 0.7831602096557617,
+      "learning_rate": 4.522454350399592e-05,
+      "loss": 0.0548,
+      "mean_token_accuracy": 0.9806917309761047,
+      "num_tokens": 1568087.0,
+      "step": 1160
+    },
+    {
+      "entropy": 0.9131727576255798,
+      "epoch": 1.118546845124283,
+      "grad_norm": 0.5717619061470032,
+      "learning_rate": 4.4438250403456187e-05,
+      "loss": 0.0425,
+      "mean_token_accuracy": 0.9855827808380127,
+      "num_tokens": 1581639.0,
+      "step": 1170
+    },
+    {
+      "entropy": 0.9028052866458893,
+      "epoch": 1.1281070745697896,
+      "grad_norm": 0.4360877275466919,
+      "learning_rate": 4.365334759610836e-05,
+      "loss": 0.0504,
+      "mean_token_accuracy": 0.9828112006187439,
+      "num_tokens": 1594420.0,
+      "step": 1180
+    },
+    {
+      "entropy": 0.914085465669632,
+      "epoch": 1.1376673040152965,
+      "grad_norm": 0.41432347893714905,
+      "learning_rate": 4.2870031287302984e-05,
+      "loss": 0.0582,
+      "mean_token_accuracy": 0.979324609041214,
+      "num_tokens": 1608313.0,
+      "step": 1190
+    },
+    {
+      "entropy": 0.9014366447925568,
+      "epoch": 1.147227533460803,
+      "grad_norm": 0.42623016238212585,
+      "learning_rate": 4.2088497285807125e-05,
+      "loss": 0.0541,
+      "mean_token_accuracy": 0.9812221586704254,
+      "num_tokens": 1621999.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.8981758415699005,
+      "epoch": 1.1567877629063097,
+      "grad_norm": 0.21387742459774017,
+      "learning_rate": 4.1308940954857236e-05,
+      "loss": 0.0505,
+      "mean_token_accuracy": 0.982359766960144,
+      "num_tokens": 1635960.0,
+      "step": 1210
+    },
+    {
+      "entropy": 0.8943774580955506,
+      "epoch": 1.1663479923518165,
+      "grad_norm": 0.6841591596603394,
+      "learning_rate": 4.0531557163323415e-05,
+      "loss": 0.0386,
+      "mean_token_accuracy": 0.9871844351291656,
+      "num_tokens": 1648974.0,
+      "step": 1220
+    },
+    {
+      "entropy": 0.8975376486778259,
+      "epoch": 1.1759082217973231,
+      "grad_norm": 0.6669749021530151,
+      "learning_rate": 3.975654023699722e-05,
+      "loss": 0.0564,
+      "mean_token_accuracy": 0.9808565139770508,
+      "num_tokens": 1662877.0,
+      "step": 1230
+    },
+    {
+      "entropy": 0.9038691759109497,
+      "epoch": 1.1854684512428297,
+      "grad_norm": 0.5672315955162048,
+      "learning_rate": 3.898408391001528e-05,
+      "loss": 0.0418,
+      "mean_token_accuracy": 0.9862301051616669,
+      "num_tokens": 1676115.0,
+      "step": 1240
+    },
+    {
+      "entropy": 0.8987687110900879,
+      "epoch": 1.1950286806883366,
+      "grad_norm": 0.5906156301498413,
+      "learning_rate": 3.821438127643078e-05,
+      "loss": 0.0549,
+      "mean_token_accuracy": 0.9790221989154816,
+      "num_tokens": 1689478.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.9044449329376221,
+      "epoch": 1.2045889101338432,
+      "grad_norm": 0.40494266152381897,
+      "learning_rate": 3.744762474194501e-05,
+      "loss": 0.0449,
+      "mean_token_accuracy": 0.986148339509964,
+      "num_tokens": 1703240.0,
+      "step": 1260
+    },
+    {
+      "entropy": 0.8985734283924103,
+      "epoch": 1.21414913957935,
+      "grad_norm": 0.7234494090080261,
+      "learning_rate": 3.6684005975810765e-05,
+      "loss": 0.0456,
+      "mean_token_accuracy": 0.9842062890529633,
+      "num_tokens": 1716712.0,
+      "step": 1270
+    },
+    {
+      "entropy": 0.8917267858982086,
+      "epoch": 1.2237093690248566,
+      "grad_norm": 0.6281554102897644,
+      "learning_rate": 3.592371586292021e-05,
+      "loss": 0.0422,
+      "mean_token_accuracy": 0.9851611971855163,
+      "num_tokens": 1729905.0,
+      "step": 1280
+    },
+    {
+      "entropy": 0.9042141973972321,
+      "epoch": 1.2332695984703632,
+      "grad_norm": 0.4349091053009033,
+      "learning_rate": 3.5166944456088445e-05,
+      "loss": 0.0414,
+      "mean_token_accuracy": 0.9873863220214844,
+      "num_tokens": 1744225.0,
+      "step": 1290
+    },
+    {
+      "entropy": 0.8831939399242401,
+      "epoch": 1.24282982791587,
+      "grad_norm": 0.563994288444519,
+      "learning_rate": 3.441388092854532e-05,
+      "loss": 0.0362,
+      "mean_token_accuracy": 0.9875804424285889,
+      "num_tokens": 1757359.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.8966175794601441,
+      "epoch": 1.2523900573613767,
+      "grad_norm": 0.41801267862319946,
+      "learning_rate": 3.3664713526646976e-05,
+      "loss": 0.0429,
+      "mean_token_accuracy": 0.9860823214054107,
+      "num_tokens": 1770617.0,
+      "step": 1310
+    },
+    {
+      "entropy": 0.9106213331222535,
+      "epoch": 1.2619502868068833,
+      "grad_norm": 0.35345029830932617,
+      "learning_rate": 3.291962952281918e-05,
+      "loss": 0.0435,
+      "mean_token_accuracy": 0.9840966463088989,
+      "num_tokens": 1783933.0,
+      "step": 1320
+    },
+    {
+      "entropy": 0.884774261713028,
+      "epoch": 1.27151051625239,
+      "grad_norm": 0.7438620924949646,
+      "learning_rate": 3.217881516874407e-05,
+      "loss": 0.0504,
+      "mean_token_accuracy": 0.9833350777626038,
+      "num_tokens": 1797117.0,
+      "step": 1330
+    },
+    {
+      "entropy": 0.9084494709968567,
+      "epoch": 1.2810707456978967,
+      "grad_norm": 0.433845192193985,
+      "learning_rate": 3.1442455648802e-05,
+      "loss": 0.0538,
+      "mean_token_accuracy": 0.9822645485401154,
+      "num_tokens": 1810602.0,
+      "step": 1340
+    },
+    {
+      "entropy": 0.898530226945877,
+      "epoch": 1.2906309751434035,
+      "grad_norm": 0.9403843879699707,
+      "learning_rate": 3.071073503378039e-05,
+      "loss": 0.0539,
+      "mean_token_accuracy": 0.982799106836319,
+      "num_tokens": 1823981.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.896830815076828,
+      "epoch": 1.3001912045889101,
+      "grad_norm": 0.5102181434631348,
+      "learning_rate": 2.9983836234860618e-05,
+      "loss": 0.0553,
+      "mean_token_accuracy": 0.9805386602878571,
+      "num_tokens": 1836963.0,
+      "step": 1360
+    },
+    {
+      "entropy": 0.9033085525035858,
+      "epoch": 1.3097514340344167,
+      "grad_norm": 0.3854213058948517,
+      "learning_rate": 2.9261940957895108e-05,
+      "loss": 0.0595,
+      "mean_token_accuracy": 0.9831296682357789,
+      "num_tokens": 1850336.0,
+      "step": 1370
+    },
+    {
+      "entropy": 0.8941022038459778,
+      "epoch": 1.3193116634799236,
+      "grad_norm": 0.5821661353111267,
+      "learning_rate": 2.8545229657985495e-05,
+      "loss": 0.0463,
+      "mean_token_accuracy": 0.9852072417736053,
+      "num_tokens": 1863231.0,
+      "step": 1380
+    },
+    {
+      "entropy": 0.8976728081703186,
+      "epoch": 1.3288718929254302,
+      "grad_norm": 0.3935079574584961,
+      "learning_rate": 2.783388149437359e-05,
+      "loss": 0.0504,
+      "mean_token_accuracy": 0.9820105254650116,
+      "num_tokens": 1876930.0,
+      "step": 1390
+    },
+    {
+      "entropy": 0.8869010388851166,
+      "epoch": 1.338432122370937,
+      "grad_norm": 0.7879030108451843,
+      "learning_rate": 2.71280742856562e-05,
+      "loss": 0.0414,
+      "mean_token_accuracy": 0.9866827309131623,
+      "num_tokens": 1890621.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.8882583439350128,
+      "epoch": 1.3479923518164436,
+      "grad_norm": 0.5797803997993469,
+      "learning_rate": 2.642798446533502e-05,
+      "loss": 0.0433,
+      "mean_token_accuracy": 0.9843058228492737,
+      "num_tokens": 1903678.0,
+      "step": 1410
+    },
+    {
+      "entropy": 0.8909366071224213,
+      "epoch": 1.3575525812619502,
+      "grad_norm": 0.44615650177001953,
+      "learning_rate": 2.573378703771293e-05,
+      "loss": 0.0481,
+      "mean_token_accuracy": 0.982310277223587,
+      "num_tokens": 1917079.0,
+      "step": 1420
+    },
+    {
+      "entropy": 0.9114946782588959,
+      "epoch": 1.367112810707457,
+      "grad_norm": 0.5472583770751953,
+      "learning_rate": 2.504565553414745e-05,
+      "loss": 0.0452,
+      "mean_token_accuracy": 0.9853902578353881,
+      "num_tokens": 1931223.0,
+      "step": 1430
+    },
+    {
+      "entropy": 0.9068931043148041,
+      "epoch": 1.3766730401529637,
+      "grad_norm": 0.9577943682670593,
+      "learning_rate": 2.4363761969672278e-05,
+      "loss": 0.0459,
+      "mean_token_accuracy": 0.9827816247940063,
+      "num_tokens": 1944423.0,
+      "step": 1440
+    },
+    {
+      "entropy": 0.9002765476703644,
+      "epoch": 1.3862332695984705,
+      "grad_norm": 0.6619713306427002,
+      "learning_rate": 2.368827679999814e-05,
+      "loss": 0.0453,
+      "mean_token_accuracy": 0.9866986095905304,
+      "num_tokens": 1957626.0,
+      "step": 1450
+    },
+    {
+      "entropy": 0.913809859752655,
+      "epoch": 1.395793499043977,
+      "grad_norm": 0.37162667512893677,
+      "learning_rate": 2.3019368878903213e-05,
+      "loss": 0.0558,
+      "mean_token_accuracy": 0.983017909526825,
+      "num_tokens": 1970965.0,
+      "step": 1460
+    },
+    {
+      "entropy": 0.9098680436611175,
+      "epoch": 1.4053537284894837,
+      "grad_norm": 0.48092612624168396,
+      "learning_rate": 2.2357205416024075e-05,
+      "loss": 0.0396,
+      "mean_token_accuracy": 0.9861399531364441,
+      "num_tokens": 1984874.0,
+      "step": 1470
+    },
+    {
+      "entropy": 0.8968272268772125,
+      "epoch": 1.4149139579349903,
+      "grad_norm": 0.44162610173225403,
+      "learning_rate": 2.1701951935057573e-05,
+      "loss": 0.0426,
+      "mean_token_accuracy": 0.9837617993354797,
+      "num_tokens": 1999952.0,
+      "step": 1480
+    },
+    {
+      "entropy": 0.8984707057476043,
+      "epoch": 1.4244741873804971,
+      "grad_norm": 0.40697169303894043,
+      "learning_rate": 2.10537722323843e-05,
+      "loss": 0.0496,
+      "mean_token_accuracy": 0.9824699759483337,
+      "num_tokens": 2012830.0,
+      "step": 1490
+    },
+    {
+      "entropy": 0.9151232123374939,
+      "epoch": 1.4340344168260037,
+      "grad_norm": 0.5245476961135864,
+      "learning_rate": 2.0412828336123658e-05,
+      "loss": 0.0445,
+      "mean_token_accuracy": 0.9851468205451965,
+      "num_tokens": 2026296.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.8907874345779419,
+      "epoch": 1.4435946462715106,
+      "grad_norm": 0.7181148529052734,
+      "learning_rate": 1.9779280465631068e-05,
+      "loss": 0.0448,
+      "mean_token_accuracy": 0.9845417737960815,
+      "num_tokens": 2039361.0,
+      "step": 1510
+    },
+    {
+      "entropy": 0.9053260266780854,
+      "epoch": 1.4531548757170172,
+      "grad_norm": 0.7296944856643677,
+      "learning_rate": 1.9153286991447267e-05,
+      "loss": 0.0446,
+      "mean_token_accuracy": 0.9846558451652527,
+      "num_tokens": 2052906.0,
+      "step": 1520
+    },
+    {
+      "entropy": 0.8986833333969116,
+      "epoch": 1.4627151051625238,
+      "grad_norm": 0.6117823719978333,
+      "learning_rate": 1.8535004395709797e-05,
+      "loss": 0.045,
+      "mean_token_accuracy": 0.9832855999469757,
+      "num_tokens": 2066064.0,
+      "step": 1530
+    },
+    {
+      "entropy": 0.8972567558288574,
+      "epoch": 1.4722753346080306,
+      "grad_norm": 0.6643509864807129,
+      "learning_rate": 1.7924587233036537e-05,
+      "loss": 0.0483,
+      "mean_token_accuracy": 0.9831550002098084,
+      "num_tokens": 2079485.0,
+      "step": 1540
+    },
+    {
+      "entropy": 0.8960070490837098,
+      "epoch": 1.4818355640535372,
+      "grad_norm": 0.6992599368095398,
+      "learning_rate": 1.7322188091890985e-05,
+      "loss": 0.0408,
+      "mean_token_accuracy": 0.9846390783786774,
+      "num_tokens": 2092884.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.8890924096107483,
+      "epoch": 1.491395793499044,
+      "grad_norm": 0.7564888000488281,
+      "learning_rate": 1.6727957556439188e-05,
+      "loss": 0.0417,
+      "mean_token_accuracy": 0.9867662608623504,
+      "num_tokens": 2106152.0,
+      "step": 1560
+    },
+    {
+      "entropy": 0.9016219913959503,
+      "epoch": 1.5009560229445507,
+      "grad_norm": 0.5372325778007507,
+      "learning_rate": 1.6142044168907595e-05,
+      "loss": 0.0493,
+      "mean_token_accuracy": 0.9840689659118652,
+      "num_tokens": 2120047.0,
+      "step": 1570
+    },
+    {
+      "entropy": 0.8983773469924927,
+      "epoch": 1.5105162523900573,
+      "grad_norm": 0.6884347200393677,
+      "learning_rate": 1.556459439245125e-05,
+      "loss": 0.0384,
+      "mean_token_accuracy": 0.988132756948471,
+      "num_tokens": 2133798.0,
+      "step": 1580
+    },
+    {
+      "entropy": 0.8867662847042084,
+      "epoch": 1.520076481835564,
+      "grad_norm": 0.38725000619888306,
+      "learning_rate": 1.4995752574541921e-05,
+      "loss": 0.0409,
+      "mean_token_accuracy": 0.9845058023929596,
+      "num_tokens": 2147167.0,
+      "step": 1590
+    },
+    {
+      "entropy": 0.8867039203643798,
+      "epoch": 1.5296367112810707,
+      "grad_norm": 0.8247562646865845,
+      "learning_rate": 1.443566091088499e-05,
+      "loss": 0.0457,
+      "mean_token_accuracy": 0.982823246717453,
+      "num_tokens": 2160770.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.8946295917034149,
+      "epoch": 1.5391969407265775,
+      "grad_norm": 0.5247822999954224,
+      "learning_rate": 1.3884459409874034e-05,
+      "loss": 0.0412,
+      "mean_token_accuracy": 0.9844862639904022,
+      "num_tokens": 2174474.0,
+      "step": 1610
+    },
+    {
+      "entropy": 0.88776695728302,
+      "epoch": 1.5487571701720841,
+      "grad_norm": 0.6349275708198547,
+      "learning_rate": 1.3342285857592551e-05,
+      "loss": 0.0456,
+      "mean_token_accuracy": 0.9837218940258026,
+      "num_tokens": 2187663.0,
+      "step": 1620
+    },
+    {
+      "entropy": 0.9058098077774048,
+      "epoch": 1.5583173996175907,
+      "grad_norm": 0.46780669689178467,
+      "learning_rate": 1.2809275783370923e-05,
+      "loss": 0.0353,
+      "mean_token_accuracy": 0.9888676702976227,
+      "num_tokens": 2201519.0,
+      "step": 1630
+    },
+    {
+      "entropy": 0.8962313055992126,
+      "epoch": 1.5678776290630974,
+      "grad_norm": 0.518578827381134,
+      "learning_rate": 1.2285562425907621e-05,
+      "loss": 0.0369,
+      "mean_token_accuracy": 0.9884732246398926,
+      "num_tokens": 2214755.0,
+      "step": 1640
+    },
+    {
+      "entropy": 0.8819742858409881,
+      "epoch": 1.5774378585086042,
+      "grad_norm": 0.4368748366832733,
+      "learning_rate": 1.1771276699962935e-05,
+      "loss": 0.0406,
+      "mean_token_accuracy": 0.9840778172016144,
+      "num_tokens": 2228056.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.8793406009674072,
+      "epoch": 1.586998087954111,
+      "grad_norm": 0.6766624450683594,
+      "learning_rate": 1.1266547163633767e-05,
+      "loss": 0.0394,
+      "mean_token_accuracy": 0.9848674774169922,
+      "num_tokens": 2241581.0,
+      "step": 1660
+    },
+    {
+      "entropy": 0.8984463334083557,
+      "epoch": 1.5965583173996176,
+      "grad_norm": 0.5182107090950012,
+      "learning_rate": 1.0771499986217365e-05,
+      "loss": 0.0517,
+      "mean_token_accuracy": 0.9813013255596161,
+      "num_tokens": 2255654.0,
+      "step": 1670
+    },
+    {
+      "entropy": 0.8911845684051514,
+      "epoch": 1.6061185468451242,
+      "grad_norm": 0.5027355551719666,
+      "learning_rate": 1.0286258916672303e-05,
+      "loss": 0.0416,
+      "mean_token_accuracy": 0.9879878044128418,
+      "num_tokens": 2269408.0,
+      "step": 1680
+    },
+    {
+      "entropy": 0.8956733822822571,
+      "epoch": 1.6156787762906308,
+      "grad_norm": 0.36508896946907043,
+      "learning_rate": 9.81094525268445e-06,
+      "loss": 0.0436,
+      "mean_token_accuracy": 0.9830544352531433,
+      "num_tokens": 2282973.0,
+      "step": 1690
+    },
+    {
+      "entropy": 0.8933298766613007,
+      "epoch": 1.6252390057361377,
+      "grad_norm": 0.665983259677887,
+      "learning_rate": 9.345677810345677e-06,
+      "loss": 0.0536,
+      "mean_token_accuracy": 0.9821488380432128,
+      "num_tokens": 2296428.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.8909790396690369,
+      "epoch": 1.6347992351816445,
+      "grad_norm": 0.6363316178321838,
+      "learning_rate": 8.890572894452948e-06,
+      "loss": 0.0395,
+      "mean_token_accuracy": 0.9869637966156006,
+      "num_tokens": 2310330.0,
+      "step": 1710
+    },
+    {
+      "entropy": 0.8798816502094269,
+      "epoch": 1.644359464627151,
+      "grad_norm": 0.6995460391044617,
+      "learning_rate": 8.445744269435063e-06,
+      "loss": 0.044,
+      "mean_token_accuracy": 0.9838290572166443,
+      "num_tokens": 2324210.0,
+      "step": 1720
+    },
+    {
+      "entropy": 0.9081452369689942,
+      "epoch": 1.6539196940726577,
+      "grad_norm": 0.6172953844070435,
+      "learning_rate": 8.011303130914599e-06,
+      "loss": 0.0412,
+      "mean_token_accuracy": 0.986025619506836,
+      "num_tokens": 2337289.0,
+      "step": 1730
+    },
+    {
+      "entropy": 0.8747174441814423,
+      "epoch": 1.6634799235181643,
+      "grad_norm": 0.5204220414161682,
+      "learning_rate": 7.5873580779118266e-06,
+      "loss": 0.0406,
+      "mean_token_accuracy": 0.9855200409889221,
+      "num_tokens": 2350620.0,
+      "step": 1740
+    },
+    {
+      "entropy": 0.8820320665836334,
+      "epoch": 1.6730401529636711,
+      "grad_norm": 0.40161561965942383,
+      "learning_rate": 7.174015085697788e-06,
+      "loss": 0.0448,
+      "mean_token_accuracy": 0.9855751156806946,
+      "num_tokens": 2364035.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.8879956245422364,
+      "epoch": 1.682600382409178,
+      "grad_norm": 0.8941473364830017,
+      "learning_rate": 6.771377479303187e-06,
+      "loss": 0.0365,
+      "mean_token_accuracy": 0.987408459186554,
+      "num_tokens": 2377105.0,
+      "step": 1760
+    },
+    {
+      "entropy": 0.8762494146823883,
+      "epoch": 1.6921606118546846,
+      "grad_norm": 0.3539943993091583,
+      "learning_rate": 6.379545907689904e-06,
+      "loss": 0.0433,
+      "mean_token_accuracy": 0.9855392217636109,
+      "num_tokens": 2390871.0,
+      "step": 1770
+    },
+    {
+      "entropy": 0.8778738737106323,
+      "epoch": 1.7017208413001912,
+      "grad_norm": 0.5057254433631897,
+      "learning_rate": 5.998618318591282e-06,
+      "loss": 0.0359,
+      "mean_token_accuracy": 0.9870854198932648,
+      "num_tokens": 2404613.0,
+      "step": 1780
+    },
+    {
+      "entropy": 0.8947471559047699,
+      "epoch": 1.7112810707456978,
+      "grad_norm": 0.5092808604240417,
+      "learning_rate": 5.628689934027725e-06,
+      "loss": 0.0384,
+      "mean_token_accuracy": 0.987550413608551,
+      "num_tokens": 2418173.0,
+      "step": 1790
+    },
+    {
+      "entropy": 0.8822583496570587,
+      "epoch": 1.7208413001912046,
+      "grad_norm": 0.7039101123809814,
+      "learning_rate": 5.269853226503746e-06,
+      "loss": 0.0382,
+      "mean_token_accuracy": 0.9858150005340576,
+      "num_tokens": 2431691.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.9100722551345826,
+      "epoch": 1.7304015296367112,
+      "grad_norm": 0.6645135879516602,
+      "learning_rate": 4.922197895892167e-06,
+      "loss": 0.0436,
+      "mean_token_accuracy": 0.9867968261241913,
+      "num_tokens": 2446060.0,
+      "step": 1810
+    },
+    {
+      "entropy": 0.8841040968894959,
+      "epoch": 1.739961759082218,
+      "grad_norm": 0.5366864800453186,
+      "learning_rate": 4.585810847011473e-06,
+      "loss": 0.0444,
+      "mean_token_accuracy": 0.9868423879146576,
+      "num_tokens": 2459918.0,
+      "step": 1820
+    },
+    {
+      "entropy": 0.8704358458518981,
+      "epoch": 1.7495219885277247,
+      "grad_norm": 2.0560102462768555,
+      "learning_rate": 4.260776167901931e-06,
+      "loss": 0.0566,
+      "mean_token_accuracy": 0.9812475502490997,
+      "num_tokens": 2473188.0,
+      "step": 1830
+    },
+    {
+      "entropy": 0.8899318814277649,
+      "epoch": 1.7590822179732313,
+      "grad_norm": 0.6329870820045471,
+      "learning_rate": 3.947175108805678e-06,
+      "loss": 0.0416,
+      "mean_token_accuracy": 0.9857662081718445,
+      "num_tokens": 2486973.0,
+      "step": 1840
+    },
+    {
+      "entropy": 0.8828537464141846,
+      "epoch": 1.7686424474187379,
+      "grad_norm": 0.6052564382553101,
+      "learning_rate": 3.6450860618563175e-06,
+      "loss": 0.0415,
+      "mean_token_accuracy": 0.9853296756744385,
+      "num_tokens": 2500712.0,
+      "step": 1850
+    },
+    {
+      "entropy": 0.8871765017509461,
+      "epoch": 1.7782026768642447,
+      "grad_norm": 1.2001285552978516,
+      "learning_rate": 3.3545845414829026e-06,
+      "loss": 0.0466,
+      "mean_token_accuracy": 0.9811452805995942,
+      "num_tokens": 2514191.0,
+      "step": 1860
+    },
+    {
+      "entropy": 0.8893057763576507,
+      "epoch": 1.7877629063097515,
+      "grad_norm": 0.6222289204597473,
+      "learning_rate": 3.07574316553329e-06,
+      "loss": 0.0507,
+      "mean_token_accuracy": 0.9794944882392883,
+      "num_tokens": 2527328.0,
+      "step": 1870
+    },
+    {
+      "entropy": 0.8997491359710693,
+      "epoch": 1.7973231357552581,
+      "grad_norm": 0.3376215398311615,
+      "learning_rate": 2.808631637121567e-06,
+      "loss": 0.0362,
+      "mean_token_accuracy": 0.9864791452884674,
+      "num_tokens": 2541361.0,
+      "step": 1880
+    },
+    {
+      "entropy": 0.8827273786067963,
+      "epoch": 1.8068833652007648,
+      "grad_norm": 0.35917985439300537,
+      "learning_rate": 2.553316727204047e-06,
+      "loss": 0.0457,
+      "mean_token_accuracy": 0.9858400762081146,
+      "num_tokens": 2554550.0,
+      "step": 1890
+    },
+    {
+      "entropy": 0.8876102924346924,
+      "epoch": 1.8164435946462714,
+      "grad_norm": 0.2676054537296295,
+      "learning_rate": 2.309862257888301e-06,
+      "loss": 0.0367,
+      "mean_token_accuracy": 0.9892573535442353,
+      "num_tokens": 2568194.0,
+      "step": 1900
+    },
+    {
+      "entropy": 0.8919675707817077,
+      "epoch": 1.8260038240917782,
+      "grad_norm": 0.6857296228408813,
+      "learning_rate": 2.078329086479286e-06,
+      "loss": 0.036,
+      "mean_token_accuracy": 0.988125741481781,
+      "num_tokens": 2581876.0,
+      "step": 1910
+    },
+    {
+      "entropy": 0.8944141507148743,
+      "epoch": 1.835564053537285,
+      "grad_norm": 0.7467958927154541,
+      "learning_rate": 1.8587750902665735e-06,
+      "loss": 0.0394,
+      "mean_token_accuracy": 0.9856638133525848,
+      "num_tokens": 2595560.0,
+      "step": 1920
+    },
+    {
+      "entropy": 0.8726170778274536,
+      "epoch": 1.8451242829827916,
+      "grad_norm": 0.5457600951194763,
+      "learning_rate": 1.6512551520565133e-06,
+      "loss": 0.04,
+      "mean_token_accuracy": 0.9860314130783081,
+      "num_tokens": 2609321.0,
+      "step": 1930
+    },
+    {
+      "entropy": 0.8843020617961883,
+      "epoch": 1.8546845124282982,
+      "grad_norm": 0.4829769432544708,
+      "learning_rate": 1.455821146453018e-06,
+      "loss": 0.0395,
+      "mean_token_accuracy": 0.9859624743461609,
+      "num_tokens": 2622763.0,
+      "step": 1940
+    },
+    {
+      "entropy": 0.8953157365322113,
+      "epoch": 1.8642447418738048,
+      "grad_norm": 0.48945409059524536,
+      "learning_rate": 1.2725219268901967e-06,
+      "loss": 0.039,
+      "mean_token_accuracy": 0.986795848608017,
+      "num_tokens": 2636291.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.8680439531803131,
+      "epoch": 1.8738049713193117,
+      "grad_norm": 0.6412997245788574,
+      "learning_rate": 1.1014033134202683e-06,
+      "loss": 0.0457,
+      "mean_token_accuracy": 0.9829192996025086,
+      "num_tokens": 2649475.0,
+      "step": 1960
+    },
+    {
+      "entropy": 0.8818358719348908,
+      "epoch": 1.8833652007648185,
+      "grad_norm": 0.7918401956558228,
+      "learning_rate": 9.425080812597909e-07,
+      "loss": 0.0367,
+      "mean_token_accuracy": 0.9882298171520233,
+      "num_tokens": 2662966.0,
+      "step": 1970
+    },
+    {
+      "entropy": 0.8824394464492797,
+      "epoch": 1.892925430210325,
+      "grad_norm": 0.6192373633384705,
+      "learning_rate": 7.958759500968915e-07,
+      "loss": 0.0392,
+      "mean_token_accuracy": 0.9869017004966736,
+      "num_tokens": 2675592.0,
+      "step": 1980
+    },
+    {
+      "entropy": 0.8927717745304108,
+      "epoch": 1.9024856596558317,
+      "grad_norm": 0.41630133986473083,
+      "learning_rate": 6.61543574162432e-07,
+      "loss": 0.0435,
+      "mean_token_accuracy": 0.9847965717315674,
+      "num_tokens": 2689270.0,
+      "step": 1990
+    },
+    {
+      "entropy": 0.8844909191131591,
+      "epoch": 1.9120458891013383,
+      "grad_norm": 0.7093897461891174,
+      "learning_rate": 5.395445330673654e-07,
+      "loss": 0.0421,
+      "mean_token_accuracy": 0.9878914654254913,
+      "num_tokens": 2702690.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.8777146697044372,
+      "epoch": 1.9216061185468452,
+      "grad_norm": 0.8239312767982483,
+      "learning_rate": 4.299093234087281e-07,
+      "loss": 0.0419,
+      "mean_token_accuracy": 0.9851139128208161,
+      "num_tokens": 2715562.0,
+      "step": 2010
+    },
+    {
+      "entropy": 0.8936961948871612,
+      "epoch": 1.9311663479923518,
+      "grad_norm": 0.6024506092071533,
+      "learning_rate": 3.3266535114628296e-07,
+      "loss": 0.0457,
+      "mean_token_accuracy": 0.9829654216766357,
+      "num_tokens": 2729105.0,
+      "step": 2020
+    },
+    {
+      "entropy": 0.8802317023277283,
+      "epoch": 1.9407265774378586,
+      "grad_norm": 0.32203608751296997,
+      "learning_rate": 2.4783692475169874e-07,
+      "loss": 0.0417,
+      "mean_token_accuracy": 0.9875831305980682,
+      "num_tokens": 2742679.0,
+      "step": 2030
+    },
+    {
+      "entropy": 0.8889366209506988,
+      "epoch": 1.9502868068833652,
+      "grad_norm": 0.5065972805023193,
+      "learning_rate": 1.7544524913211125e-07,
+      "loss": 0.0331,
+      "mean_token_accuracy": 0.9884385526180267,
+      "num_tokens": 2756899.0,
+      "step": 2040
+    },
+    {
+      "entropy": 0.9007688105106354,
+      "epoch": 1.9598470363288718,
+      "grad_norm": 0.5861697196960449,
+      "learning_rate": 1.1550842032943543e-07,
+      "loss": 0.0641,
+      "mean_token_accuracy": 0.9797191143035888,
+      "num_tokens": 2771213.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.8843443930149079,
+      "epoch": 1.9694072657743786,
+      "grad_norm": 0.7110727429389954,
+      "learning_rate": 6.8041420996795e-08,
+      "loss": 0.0379,
+      "mean_token_accuracy": 0.9883714020252228,
+      "num_tokens": 2784886.0,
+      "step": 2060
+    },
+    {
+      "entropy": 0.8914660274982452,
+      "epoch": 1.9789674952198852,
+      "grad_norm": 0.4897426664829254,
+      "learning_rate": 3.3056116653285005e-08,
+      "loss": 0.0308,
+      "mean_token_accuracy": 0.9926590621471405,
+      "num_tokens": 2798197.0,
+      "step": 2070
+    },
+    {
+      "entropy": 0.8942194163799286,
+      "epoch": 1.988527724665392,
+      "grad_norm": 0.5657405853271484,
+      "learning_rate": 1.056125271788333e-08,
+      "loss": 0.0389,
+      "mean_token_accuracy": 0.9864471435546875,
+      "num_tokens": 2811807.0,
+      "step": 2080
+    },
+    {
+      "entropy": 0.8836948096752166,
+      "epoch": 1.9980879541108987,
+      "grad_norm": 0.6643404364585876,
+      "learning_rate": 5.62452323304985e-10,
+      "loss": 0.0486,
+      "mean_token_accuracy": 0.98052499294281,
+      "num_tokens": 2825012.0,
+      "step": 2090
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2092,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.2201917175030374e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:048511de437099a945a43b3c66145c65e4f57085de0a4be01901fe7db67c4d3c
+size 6417

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3",
+  "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": 3,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/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_s3/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_s3/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_s3/train_manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3",
+  "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": 3,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r64_a32_keepbot1pct_priorrecipe_e2_lr1e-4_s3/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:048511de437099a945a43b3c66145c65e4f57085de0a4be01901fe7db67c4d3c
+size 6417