add checkpoint cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s3

.gitattributes

@@ -760,3 +760,6 @@ checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s2/tokenizer.json
 checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s3/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoints/cat_qwen25_7b_r64_a32_plain_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_plain_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_plain_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/izqi6682) 
+
+
+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_plain_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",
+    "up_proj",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s3/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46fd4b297e11f0d5fa24c6f210fe44eb43e91780c11a60bdb61674e62425e2fc
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_plain_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_plain_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_plain_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_plain_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",
+    "up_proj",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c0c4c05116bbb89ef51ef70962c1001cbe82816ba25679dc0d93af19b98ad6f
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_plain_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_plain_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_plain_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_plain_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_plain_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_plain_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": 1.1641528248786925,
+      "epoch": 0.009560229445506692,
+      "grad_norm": 0.6996516585350037,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2721,
+      "mean_token_accuracy": 0.913582569360733,
+      "num_tokens": 13150.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1633511900901794,
+      "epoch": 0.019120458891013385,
+      "grad_norm": 0.5077252388000488,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2617,
+      "mean_token_accuracy": 0.9153739392757416,
+      "num_tokens": 26136.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1612649083137512,
+      "epoch": 0.028680688336520075,
+      "grad_norm": 0.48587486147880554,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.2273,
+      "mean_token_accuracy": 0.9281989872455597,
+      "num_tokens": 39207.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.157628071308136,
+      "epoch": 0.03824091778202677,
+      "grad_norm": 0.5328208804130554,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1924,
+      "mean_token_accuracy": 0.9389224052429199,
+      "num_tokens": 52869.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.1558964610099793,
+      "epoch": 0.04780114722753346,
+      "grad_norm": 0.879055917263031,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.2048,
+      "mean_token_accuracy": 0.9312818169593811,
+      "num_tokens": 66622.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.1293084979057313,
+      "epoch": 0.05736137667304015,
+      "grad_norm": 0.6380035281181335,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1678,
+      "mean_token_accuracy": 0.940121340751648,
+      "num_tokens": 80175.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.1357115745544433,
+      "epoch": 0.06692160611854685,
+      "grad_norm": 0.7072647213935852,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1967,
+      "mean_token_accuracy": 0.9296577990055084,
+      "num_tokens": 94356.0,
+      "step": 70
+    },
+    {
+      "entropy": 1.1026614785194397,
+      "epoch": 0.07648183556405354,
+      "grad_norm": 0.771527111530304,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1522,
+      "mean_token_accuracy": 0.9473809123039245,
+      "num_tokens": 108716.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.0664347410202026,
+      "epoch": 0.08604206500956023,
+      "grad_norm": 0.6657344698905945,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1552,
+      "mean_token_accuracy": 0.9406381964683532,
+      "num_tokens": 122182.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0743921160697938,
+      "epoch": 0.09560229445506692,
+      "grad_norm": 0.6942067742347717,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1454,
+      "mean_token_accuracy": 0.9483474791049957,
+      "num_tokens": 135713.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0707114934921265,
+      "epoch": 0.10516252390057361,
+      "grad_norm": 0.9185185432434082,
+      "learning_rate": 9.999900008621665e-05,
+      "loss": 0.1608,
+      "mean_token_accuracy": 0.9425420522689819,
+      "num_tokens": 149456.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.0588605165481568,
+      "epoch": 0.1147227533460803,
+      "grad_norm": 0.4680742025375366,
+      "learning_rate": 9.998775151544478e-05,
+      "loss": 0.147,
+      "mean_token_accuracy": 0.9443056762218476,
+      "num_tokens": 163062.0,
+      "step": 120
+    },
+    {
+      "entropy": 1.0524839282035827,
+      "epoch": 0.124282982791587,
+      "grad_norm": 0.7786092162132263,
+      "learning_rate": 9.996400730289628e-05,
+      "loss": 0.1691,
+      "mean_token_accuracy": 0.935369998216629,
+      "num_tokens": 176574.0,
+      "step": 130
+    },
+    {
+      "entropy": 1.0402762174606324,
+      "epoch": 0.1338432122370937,
+      "grad_norm": 0.691507875919342,
+      "learning_rate": 9.992777338400864e-05,
+      "loss": 0.1629,
+      "mean_token_accuracy": 0.9382378816604614,
+      "num_tokens": 191164.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0201380014419557,
+      "epoch": 0.14340344168260039,
+      "grad_norm": 0.7071655988693237,
+      "learning_rate": 9.987905881632213e-05,
+      "loss": 0.1477,
+      "mean_token_accuracy": 0.9456806540489197,
+      "num_tokens": 204805.0,
+      "step": 150
+    },
+    {
+      "entropy": 1.0177514791488647,
+      "epoch": 0.15296367112810708,
+      "grad_norm": 0.7090482711791992,
+      "learning_rate": 9.981787577721553e-05,
+      "loss": 0.131,
+      "mean_token_accuracy": 0.9504908204078675,
+      "num_tokens": 218021.0,
+      "step": 160
+    },
+    {
+      "entropy": 1.0016795158386231,
+      "epoch": 0.16252390057361377,
+      "grad_norm": 0.5258442759513855,
+      "learning_rate": 9.974423956086228e-05,
+      "loss": 0.1243,
+      "mean_token_accuracy": 0.9528130173683167,
+      "num_tokens": 231015.0,
+      "step": 170
+    },
+    {
+      "entropy": 0.9992186069488526,
+      "epoch": 0.17208413001912046,
+      "grad_norm": 0.6269602179527283,
+      "learning_rate": 9.965816857440717e-05,
+      "loss": 0.1438,
+      "mean_token_accuracy": 0.9449268460273743,
+      "num_tokens": 244021.0,
+      "step": 180
+    },
+    {
+      "entropy": 1.0048776805400848,
+      "epoch": 0.18164435946462715,
+      "grad_norm": 0.8240458369255066,
+      "learning_rate": 9.955968433336515e-05,
+      "loss": 0.14,
+      "mean_token_accuracy": 0.9496095240116119,
+      "num_tokens": 256981.0,
+      "step": 190
+    },
+    {
+      "entropy": 0.9974187612533569,
+      "epoch": 0.19120458891013384,
+      "grad_norm": 0.5255606174468994,
+      "learning_rate": 9.944881145624291e-05,
+      "loss": 0.1483,
+      "mean_token_accuracy": 0.9459229350090027,
+      "num_tokens": 270999.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.9903082370758056,
+      "epoch": 0.20076481835564053,
+      "grad_norm": 0.6685144901275635,
+      "learning_rate": 9.932557765838504e-05,
+      "loss": 0.1193,
+      "mean_token_accuracy": 0.9607025504112243,
+      "num_tokens": 283995.0,
+      "step": 210
+    },
+    {
+      "entropy": 0.9902467429637909,
+      "epoch": 0.21032504780114722,
+      "grad_norm": 0.6727855205535889,
+      "learning_rate": 9.919001374504569e-05,
+      "loss": 0.1347,
+      "mean_token_accuracy": 0.9496286571025848,
+      "num_tokens": 297434.0,
+      "step": 220
+    },
+    {
+      "entropy": 1.0041028022766114,
+      "epoch": 0.2198852772466539,
+      "grad_norm": 0.5113528370857239,
+      "learning_rate": 9.904215360368826e-05,
+      "loss": 0.1257,
+      "mean_token_accuracy": 0.9506474733352661,
+      "num_tokens": 310688.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.0222025215625763,
+      "epoch": 0.2294455066921606,
+      "grad_norm": 1.1373951435089111,
+      "learning_rate": 9.888203419551432e-05,
+      "loss": 0.1574,
+      "mean_token_accuracy": 0.9421513736248016,
+      "num_tokens": 324594.0,
+      "step": 240
+    },
+    {
+      "entropy": 1.0387476086616516,
+      "epoch": 0.2390057361376673,
+      "grad_norm": 0.8162603974342346,
+      "learning_rate": 9.870969554622426e-05,
+      "loss": 0.153,
+      "mean_token_accuracy": 0.9408722400665284,
+      "num_tokens": 338003.0,
+      "step": 250
+    },
+    {
+      "entropy": 1.0153288543224335,
+      "epoch": 0.248565965583174,
+      "grad_norm": 0.49323147535324097,
+      "learning_rate": 9.852518073601194e-05,
+      "loss": 0.1295,
+      "mean_token_accuracy": 0.9526264905929566,
+      "num_tokens": 351394.0,
+      "step": 260
+    },
+    {
+      "entropy": 0.9894533395767212,
+      "epoch": 0.25812619502868067,
+      "grad_norm": 0.5005015134811401,
+      "learning_rate": 9.832853588879575e-05,
+      "loss": 0.1258,
+      "mean_token_accuracy": 0.9527800977230072,
+      "num_tokens": 365081.0,
+      "step": 270
+    },
+    {
+      "entropy": 1.0046591579914093,
+      "epoch": 0.2676864244741874,
+      "grad_norm": 0.5181300640106201,
+      "learning_rate": 9.811981016068881e-05,
+      "loss": 0.1298,
+      "mean_token_accuracy": 0.9493179559707642,
+      "num_tokens": 378518.0,
+      "step": 280
+    },
+    {
+      "entropy": 1.0015001237392425,
+      "epoch": 0.27724665391969405,
+      "grad_norm": 0.8453997373580933,
+      "learning_rate": 9.78990557277112e-05,
+      "loss": 0.1269,
+      "mean_token_accuracy": 0.9505164086818695,
+      "num_tokens": 391993.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0173242926597594,
+      "epoch": 0.28680688336520077,
+      "grad_norm": 0.665677011013031,
+      "learning_rate": 9.766632777274738e-05,
+      "loss": 0.1071,
+      "mean_token_accuracy": 0.9552569091320038,
+      "num_tokens": 405246.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0674896359443664,
+      "epoch": 0.29636711281070743,
+      "grad_norm": 0.8276544213294983,
+      "learning_rate": 9.742168447175182e-05,
+      "loss": 0.1275,
+      "mean_token_accuracy": 0.9511016726493835,
+      "num_tokens": 418524.0,
+      "step": 310
+    },
+    {
+      "entropy": 1.0123964190483092,
+      "epoch": 0.30592734225621415,
+      "grad_norm": 0.6688571572303772,
+      "learning_rate": 9.716518697920655e-05,
+      "loss": 0.1363,
+      "mean_token_accuracy": 0.9480633080005646,
+      "num_tokens": 432202.0,
+      "step": 320
+    },
+    {
+      "entropy": 0.999013239145279,
+      "epoch": 0.3154875717017208,
+      "grad_norm": 0.4304344952106476,
+      "learning_rate": 9.689689941283417e-05,
+      "loss": 0.1233,
+      "mean_token_accuracy": 0.9515203893184662,
+      "num_tokens": 445903.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.0126557409763337,
+      "epoch": 0.32504780114722753,
+      "grad_norm": 0.5183528661727905,
+      "learning_rate": 9.661688883756996e-05,
+      "loss": 0.1209,
+      "mean_token_accuracy": 0.9540857017040253,
+      "num_tokens": 460277.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0066914319992066,
+      "epoch": 0.33460803059273425,
+      "grad_norm": 0.6539915800094604,
+      "learning_rate": 9.632522524879756e-05,
+      "loss": 0.123,
+      "mean_token_accuracy": 0.9566046714782714,
+      "num_tokens": 473892.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.0084527671337127,
+      "epoch": 0.3441682600382409,
+      "grad_norm": 0.6190230250358582,
+      "learning_rate": 9.602198155485176e-05,
+      "loss": 0.1034,
+      "mean_token_accuracy": 0.9617183327674865,
+      "num_tokens": 487145.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.0045492053031921,
+      "epoch": 0.35372848948374763,
+      "grad_norm": 0.745010256767273,
+      "learning_rate": 9.570723355879342e-05,
+      "loss": 0.1208,
+      "mean_token_accuracy": 0.9523294031620025,
+      "num_tokens": 500481.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.0438226521015168,
+      "epoch": 0.3632887189292543,
+      "grad_norm": 0.5804723501205444,
+      "learning_rate": 9.538105993946068e-05,
+      "loss": 0.1247,
+      "mean_token_accuracy": 0.9556700885295868,
+      "num_tokens": 513817.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0288837611675263,
+      "epoch": 0.372848948374761,
+      "grad_norm": 0.45516908168792725,
+      "learning_rate": 9.504354223180118e-05,
+      "loss": 0.1298,
+      "mean_token_accuracy": 0.9496672093868256,
+      "num_tokens": 527250.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.0271916389465332,
+      "epoch": 0.3824091778202677,
+      "grad_norm": 0.4186956286430359,
+      "learning_rate": 9.469476480649056e-05,
+      "loss": 0.1234,
+      "mean_token_accuracy": 0.9543138742446899,
+      "num_tokens": 541558.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.006220853328705,
+      "epoch": 0.3919694072657744,
+      "grad_norm": 0.6684417128562927,
+      "learning_rate": 9.433481484884189e-05,
+      "loss": 0.1086,
+      "mean_token_accuracy": 0.9610184013843537,
+      "num_tokens": 554540.0,
+      "step": 410
+    },
+    {
+      "entropy": 1.0208543598651887,
+      "epoch": 0.40152963671128106,
+      "grad_norm": 0.46702873706817627,
+      "learning_rate": 9.396378233701167e-05,
+      "loss": 0.1148,
+      "mean_token_accuracy": 0.9523889362812042,
+      "num_tokens": 568221.0,
+      "step": 420
+    },
+    {
+      "entropy": 1.026773202419281,
+      "epoch": 0.4110898661567878,
+      "grad_norm": 0.6417613625526428,
+      "learning_rate": 9.358176001950759e-05,
+      "loss": 0.1229,
+      "mean_token_accuracy": 0.9499638915061951,
+      "num_tokens": 581721.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.0295559406280517,
+      "epoch": 0.42065009560229444,
+      "grad_norm": 0.8544287085533142,
+      "learning_rate": 9.318884339200378e-05,
+      "loss": 0.1145,
+      "mean_token_accuracy": 0.9570431888103486,
+      "num_tokens": 595120.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0547672986984253,
+      "epoch": 0.43021032504780116,
+      "grad_norm": 0.5602120757102966,
+      "learning_rate": 9.278513067346941e-05,
+      "loss": 0.1205,
+      "mean_token_accuracy": 0.9526568710803985,
+      "num_tokens": 608361.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0596372604370117,
+      "epoch": 0.4397705544933078,
+      "grad_norm": 0.9372671842575073,
+      "learning_rate": 9.237072278161652e-05,
+      "loss": 0.1158,
+      "mean_token_accuracy": 0.9582056403160095,
+      "num_tokens": 621843.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0450125575065612,
+      "epoch": 0.44933078393881454,
+      "grad_norm": 0.7245158553123474,
+      "learning_rate": 9.194572330767314e-05,
+      "loss": 0.1084,
+      "mean_token_accuracy": 0.956211519241333,
+      "num_tokens": 635552.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.0532810688018799,
+      "epoch": 0.4588910133843212,
+      "grad_norm": 0.5808508396148682,
+      "learning_rate": 9.151023849048818e-05,
+      "loss": 0.1301,
+      "mean_token_accuracy": 0.9499321818351746,
+      "num_tokens": 648523.0,
+      "step": 480
+    },
+    {
+      "entropy": 1.0488809943199158,
+      "epoch": 0.4684512428298279,
+      "grad_norm": 0.33145713806152344,
+      "learning_rate": 9.106437718997452e-05,
+      "loss": 0.1085,
+      "mean_token_accuracy": 0.9597950100898742,
+      "num_tokens": 661850.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0259310245513915,
+      "epoch": 0.4780114722753346,
+      "grad_norm": 0.9813805222511292,
+      "learning_rate": 9.060825085989682e-05,
+      "loss": 0.101,
+      "mean_token_accuracy": 0.9605780482292176,
+      "num_tokens": 675322.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.0419854640960693,
+      "epoch": 0.4875717017208413,
+      "grad_norm": 0.7074224948883057,
+      "learning_rate": 9.014197352001097e-05,
+      "loss": 0.1219,
+      "mean_token_accuracy": 0.9519591212272644,
+      "num_tokens": 688704.0,
+      "step": 510
+    },
+    {
+      "entropy": 1.0370671927928925,
+      "epoch": 0.497131931166348,
+      "grad_norm": 0.7930039167404175,
+      "learning_rate": 8.966566172756213e-05,
+      "loss": 0.1268,
+      "mean_token_accuracy": 0.9527601838111878,
+      "num_tokens": 702579.0,
+      "step": 520
+    },
+    {
+      "entropy": 1.0266020059585572,
+      "epoch": 0.5066921606118547,
+      "grad_norm": 0.5883492231369019,
+      "learning_rate": 8.91794345481484e-05,
+      "loss": 0.1146,
+      "mean_token_accuracy": 0.9561169147491455,
+      "num_tokens": 716160.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0262812852859498,
+      "epoch": 0.5162523900573613,
+      "grad_norm": 0.8996061682701111,
+      "learning_rate": 8.868341352595755e-05,
+      "loss": 0.1253,
+      "mean_token_accuracy": 0.9532574653625489,
+      "num_tokens": 729408.0,
+      "step": 540
+    },
+    {
+      "entropy": 1.032131028175354,
+      "epoch": 0.5258126195028681,
+      "grad_norm": 0.5681365728378296,
+      "learning_rate": 8.817772265338409e-05,
+      "loss": 0.114,
+      "mean_token_accuracy": 0.9511450946331024,
+      "num_tokens": 742824.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.0298707604408264,
+      "epoch": 0.5353728489483748,
+      "grad_norm": 0.9023966193199158,
+      "learning_rate": 8.766248834003441e-05,
+      "loss": 0.1189,
+      "mean_token_accuracy": 0.9529127359390259,
+      "num_tokens": 756778.0,
+      "step": 560
+    },
+    {
+      "entropy": 1.0299117267131805,
+      "epoch": 0.5449330783938815,
+      "grad_norm": 0.5660295486450195,
+      "learning_rate": 8.713783938112768e-05,
+      "loss": 0.1152,
+      "mean_token_accuracy": 0.9536526381969452,
+      "num_tokens": 770342.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.034860408306122,
+      "epoch": 0.5544933078393881,
+      "grad_norm": 0.44493526220321655,
+      "learning_rate": 8.660390692530026e-05,
+      "loss": 0.1109,
+      "mean_token_accuracy": 0.9568422317504883,
+      "num_tokens": 784398.0,
+      "step": 580
+    },
+    {
+      "entropy": 1.0222409248352051,
+      "epoch": 0.5640535372848948,
+      "grad_norm": 0.5718162655830383,
+      "learning_rate": 8.606082444182216e-05,
+      "loss": 0.1181,
+      "mean_token_accuracy": 0.9559833109378815,
+      "num_tokens": 798017.0,
+      "step": 590
+    },
+    {
+      "entropy": 1.0099391400814057,
+      "epoch": 0.5736137667304015,
+      "grad_norm": 0.7805862426757812,
+      "learning_rate": 8.550872768723292e-05,
+      "loss": 0.1133,
+      "mean_token_accuracy": 0.9571761429309845,
+      "num_tokens": 811336.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.9928421020507813,
+      "epoch": 0.5831739961759083,
+      "grad_norm": 0.6475667357444763,
+      "learning_rate": 8.494775467140627e-05,
+      "loss": 0.1228,
+      "mean_token_accuracy": 0.9529062807559967,
+      "num_tokens": 824727.0,
+      "step": 610
+    },
+    {
+      "entropy": 0.9883149087429046,
+      "epoch": 0.5927342256214149,
+      "grad_norm": 0.5568240880966187,
+      "learning_rate": 8.4378045623051e-05,
+      "loss": 0.1215,
+      "mean_token_accuracy": 0.9517284870147705,
+      "num_tokens": 838105.0,
+      "step": 620
+    },
+    {
+      "entropy": 0.9852628886699677,
+      "epoch": 0.6022944550669216,
+      "grad_norm": 0.8160543441772461,
+      "learning_rate": 8.379974295465758e-05,
+      "loss": 0.0984,
+      "mean_token_accuracy": 0.9631519913673401,
+      "num_tokens": 851280.0,
+      "step": 630
+    },
+    {
+      "entropy": 1.0070119798183441,
+      "epoch": 0.6118546845124283,
+      "grad_norm": 0.522103488445282,
+      "learning_rate": 8.321299122689863e-05,
+      "loss": 0.1038,
+      "mean_token_accuracy": 0.9595254600048065,
+      "num_tokens": 864939.0,
+      "step": 640
+    },
+    {
+      "entropy": 0.9948669731616974,
+      "epoch": 0.621414913957935,
+      "grad_norm": 1.0888185501098633,
+      "learning_rate": 8.261793711249254e-05,
+      "loss": 0.1158,
+      "mean_token_accuracy": 0.95493443608284,
+      "num_tokens": 878886.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.9878204762935638,
+      "epoch": 0.6309751434034416,
+      "grad_norm": 0.8432512283325195,
+      "learning_rate": 8.201472935953916e-05,
+      "loss": 0.0874,
+      "mean_token_accuracy": 0.9673986077308655,
+      "num_tokens": 892413.0,
+      "step": 660
+    },
+    {
+      "entropy": 0.9975221931934357,
+      "epoch": 0.6405353728489483,
+      "grad_norm": 0.5733245611190796,
+      "learning_rate": 8.140351875433649e-05,
+      "loss": 0.1085,
+      "mean_token_accuracy": 0.9571276009082794,
+      "num_tokens": 906485.0,
+      "step": 670
+    },
+    {
+      "entropy": 0.997295218706131,
+      "epoch": 0.6500956022944551,
+      "grad_norm": 0.8385879993438721,
+      "learning_rate": 8.07844580836881e-05,
+      "loss": 0.1129,
+      "mean_token_accuracy": 0.9515101313591003,
+      "num_tokens": 920020.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9872783482074737,
+      "epoch": 0.6596558317399618,
+      "grad_norm": 0.9112151265144348,
+      "learning_rate": 8.015770209671044e-05,
+      "loss": 0.0998,
+      "mean_token_accuracy": 0.9611438810825348,
+      "num_tokens": 933577.0,
+      "step": 690
+    },
+    {
+      "entropy": 0.9857029139995575,
+      "epoch": 0.6692160611854685,
+      "grad_norm": 0.39040815830230713,
+      "learning_rate": 7.952340746614946e-05,
+      "loss": 0.1011,
+      "mean_token_accuracy": 0.962991064786911,
+      "num_tokens": 946367.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.9934593677520752,
+      "epoch": 0.6787762906309751,
+      "grad_norm": 0.40705567598342896,
+      "learning_rate": 7.888173274921655e-05,
+      "loss": 0.1085,
+      "mean_token_accuracy": 0.9592144548892975,
+      "num_tokens": 960252.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9805512070655823,
+      "epoch": 0.6883365200764818,
+      "grad_norm": 0.7323410511016846,
+      "learning_rate": 7.823283834795336e-05,
+      "loss": 0.0972,
+      "mean_token_accuracy": 0.9624641418457032,
+      "num_tokens": 973491.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9763809263706207,
+      "epoch": 0.6978967495219885,
+      "grad_norm": 0.4441353380680084,
+      "learning_rate": 7.75768864691354e-05,
+      "loss": 0.1195,
+      "mean_token_accuracy": 0.9537525713443756,
+      "num_tokens": 986879.0,
+      "step": 730
+    },
+    {
+      "entropy": 0.9949767231941223,
+      "epoch": 0.7074569789674953,
+      "grad_norm": 0.6540898680686951,
+      "learning_rate": 7.691404108372461e-05,
+      "loss": 0.1033,
+      "mean_token_accuracy": 0.9608555018901825,
+      "num_tokens": 1000272.0,
+      "step": 740
+    },
+    {
+      "entropy": 0.9889988899230957,
+      "epoch": 0.7170172084130019,
+      "grad_norm": 0.7363319396972656,
+      "learning_rate": 7.624446788588085e-05,
+      "loss": 0.1171,
+      "mean_token_accuracy": 0.9569859743118286,
+      "num_tokens": 1013908.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.9795958340168,
+      "epoch": 0.7265774378585086,
+      "grad_norm": 0.6467368006706238,
+      "learning_rate": 7.556833425154255e-05,
+      "loss": 0.0902,
+      "mean_token_accuracy": 0.9675116181373596,
+      "num_tokens": 1027085.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9736341655254364,
+      "epoch": 0.7361376673040153,
+      "grad_norm": 0.5900027751922607,
+      "learning_rate": 7.488580919658727e-05,
+      "loss": 0.1207,
+      "mean_token_accuracy": 0.9517757058143616,
+      "num_tokens": 1040367.0,
+      "step": 770
+    },
+    {
+      "entropy": 1.0007605075836181,
+      "epoch": 0.745697896749522,
+      "grad_norm": 0.7509295344352722,
+      "learning_rate": 7.419706333458201e-05,
+      "loss": 0.1052,
+      "mean_token_accuracy": 0.9591726422309875,
+      "num_tokens": 1054076.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9726886987686157,
+      "epoch": 0.7552581261950286,
+      "grad_norm": 0.47083351016044617,
+      "learning_rate": 7.350226883413413e-05,
+      "loss": 0.0899,
+      "mean_token_accuracy": 0.9645619034767151,
+      "num_tokens": 1068530.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9697287023067475,
+      "epoch": 0.7648183556405354,
+      "grad_norm": 0.8393839597702026,
+      "learning_rate": 7.280159937585386e-05,
+      "loss": 0.1074,
+      "mean_token_accuracy": 0.9581287324428558,
+      "num_tokens": 1082342.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.9735282719135284,
+      "epoch": 0.7743785850860421,
+      "grad_norm": 0.40682822465896606,
+      "learning_rate": 7.209523010893853e-05,
+      "loss": 0.1045,
+      "mean_token_accuracy": 0.9595387160778046,
+      "num_tokens": 1096496.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.9854253113269806,
+      "epoch": 0.7839388145315488,
+      "grad_norm": 0.4829469621181488,
+      "learning_rate": 7.138333760738987e-05,
+      "loss": 0.1118,
+      "mean_token_accuracy": 0.9563177168369293,
+      "num_tokens": 1109906.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9937235653400421,
+      "epoch": 0.7934990439770554,
+      "grad_norm": 0.41137319803237915,
+      "learning_rate": 7.066609982587501e-05,
+      "loss": 0.1177,
+      "mean_token_accuracy": 0.9555411994457245,
+      "num_tokens": 1123170.0,
+      "step": 830
+    },
+    {
+      "entropy": 0.9826422870159149,
+      "epoch": 0.8030592734225621,
+      "grad_norm": 0.41072991490364075,
+      "learning_rate": 6.994369605524253e-05,
+      "loss": 0.1067,
+      "mean_token_accuracy": 0.959008538722992,
+      "num_tokens": 1136503.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9492685794830322,
+      "epoch": 0.8126195028680688,
+      "grad_norm": 0.7331626415252686,
+      "learning_rate": 6.921630687770427e-05,
+      "loss": 0.0897,
+      "mean_token_accuracy": 0.9662109792232514,
+      "num_tokens": 1149719.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.9468518316745758,
+      "epoch": 0.8221797323135756,
+      "grad_norm": 0.48760515451431274,
+      "learning_rate": 6.848411412169461e-05,
+      "loss": 0.0947,
+      "mean_token_accuracy": 0.961515873670578,
+      "num_tokens": 1163320.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9556315541267395,
+      "epoch": 0.8317399617590823,
+      "grad_norm": 0.575470507144928,
+      "learning_rate": 6.774730081641799e-05,
+      "loss": 0.0944,
+      "mean_token_accuracy": 0.9629533231258393,
+      "num_tokens": 1177204.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.9477540969848632,
+      "epoch": 0.8413001912045889,
+      "grad_norm": 0.731942892074585,
+      "learning_rate": 6.70060511460963e-05,
+      "loss": 0.1026,
+      "mean_token_accuracy": 0.9615570724010467,
+      "num_tokens": 1190225.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9491175174713135,
+      "epoch": 0.8508604206500956,
+      "grad_norm": 0.5208237767219543,
+      "learning_rate": 6.626055040392774e-05,
+      "loss": 0.0934,
+      "mean_token_accuracy": 0.9632180333137512,
+      "num_tokens": 1204511.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9400201737880707,
+      "epoch": 0.8604206500956023,
+      "grad_norm": 0.5904267430305481,
+      "learning_rate": 6.551098494576819e-05,
+      "loss": 0.0987,
+      "mean_token_accuracy": 0.9621688187122345,
+      "num_tokens": 1217217.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9426032364368438,
+      "epoch": 0.869980879541109,
+      "grad_norm": 0.6165425777435303,
+      "learning_rate": 6.475754214354715e-05,
+      "loss": 0.095,
+      "mean_token_accuracy": 0.9672666728496552,
+      "num_tokens": 1230535.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.9551124751567841,
+      "epoch": 0.8795411089866156,
+      "grad_norm": 0.6224029064178467,
+      "learning_rate": 6.400041033842965e-05,
+      "loss": 0.1111,
+      "mean_token_accuracy": 0.9552771091461182,
+      "num_tokens": 1244121.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.961413562297821,
+      "epoch": 0.8891013384321224,
+      "grad_norm": 0.447574645280838,
+      "learning_rate": 6.323977879373571e-05,
+      "loss": 0.1119,
+      "mean_token_accuracy": 0.9559538304805756,
+      "num_tokens": 1257972.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.972063022851944,
+      "epoch": 0.8986615678776291,
+      "grad_norm": 0.7507049441337585,
+      "learning_rate": 6.247583764762959e-05,
+      "loss": 0.0977,
+      "mean_token_accuracy": 0.9586446702480316,
+      "num_tokens": 1271568.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9554756462574006,
+      "epoch": 0.9082217973231358,
+      "grad_norm": 1.0878021717071533,
+      "learning_rate": 6.170877786559001e-05,
+      "loss": 0.094,
+      "mean_token_accuracy": 0.9635938465595245,
+      "num_tokens": 1285826.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.9407235264778138,
+      "epoch": 0.9177820267686424,
+      "grad_norm": 0.48441722989082336,
+      "learning_rate": 6.0938791192673895e-05,
+      "loss": 0.098,
+      "mean_token_accuracy": 0.962267804145813,
+      "num_tokens": 1299889.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9443095207214356,
+      "epoch": 0.9273422562141491,
+      "grad_norm": 0.678026556968689,
+      "learning_rate": 6.016607010558492e-05,
+      "loss": 0.0867,
+      "mean_token_accuracy": 0.967143052816391,
+      "num_tokens": 1313610.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9234336674213409,
+      "epoch": 0.9369024856596558,
+      "grad_norm": 0.5790787935256958,
+      "learning_rate": 5.939080776455942e-05,
+      "loss": 0.0941,
+      "mean_token_accuracy": 0.9665516257286072,
+      "num_tokens": 1327185.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.9353838205337525,
+      "epoch": 0.9464627151051626,
+      "grad_norm": 0.8568056225776672,
+      "learning_rate": 5.861319796508139e-05,
+      "loss": 0.1104,
+      "mean_token_accuracy": 0.9536222815513611,
+      "num_tokens": 1340176.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.939469438791275,
+      "epoch": 0.9560229445506692,
+      "grad_norm": 0.8979282975196838,
+      "learning_rate": 5.783343508943849e-05,
+      "loss": 0.0993,
+      "mean_token_accuracy": 0.963096821308136,
+      "num_tokens": 1354053.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9367132842540741,
+      "epoch": 0.9655831739961759,
+      "grad_norm": 0.4966347813606262,
+      "learning_rate": 5.7051714058131654e-05,
+      "loss": 0.1023,
+      "mean_token_accuracy": 0.9576547324657441,
+      "num_tokens": 1367199.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.9079373300075531,
+      "epoch": 0.9751434034416826,
+      "grad_norm": 0.6267583966255188,
+      "learning_rate": 5.626823028114991e-05,
+      "loss": 0.0728,
+      "mean_token_accuracy": 0.9701129257678985,
+      "num_tokens": 1380205.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9324047029018402,
+      "epoch": 0.9847036328871893,
+      "grad_norm": 0.7952474355697632,
+      "learning_rate": 5.548317960912305e-05,
+      "loss": 0.0931,
+      "mean_token_accuracy": 0.959660267829895,
+      "num_tokens": 1392623.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9379731714725494,
+      "epoch": 0.994263862332696,
+      "grad_norm": 0.6732680797576904,
+      "learning_rate": 5.4696758284363956e-05,
+      "loss": 0.0904,
+      "mean_token_accuracy": 0.9672562539577484,
+      "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_plain_priorrecipe_e2_lr1e-4_s3/checkpoint-1046/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d06d12324071801b14ec40ba4f282f75f7986e48dfdafc8dc25fe79865f96fd9
+size 6417

checkpoints/cat_qwen25_7b_r64_a32_plain_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_plain_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",
+    "up_proj",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46fd4b297e11f0d5fa24c6f210fe44eb43e91780c11a60bdb61674e62425e2fc
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_plain_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_plain_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_plain_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_plain_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_plain_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_plain_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": 1.1641528248786925,
+      "epoch": 0.009560229445506692,
+      "grad_norm": 0.6996516585350037,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2721,
+      "mean_token_accuracy": 0.913582569360733,
+      "num_tokens": 13150.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1633511900901794,
+      "epoch": 0.019120458891013385,
+      "grad_norm": 0.5077252388000488,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2617,
+      "mean_token_accuracy": 0.9153739392757416,
+      "num_tokens": 26136.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1612649083137512,
+      "epoch": 0.028680688336520075,
+      "grad_norm": 0.48587486147880554,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.2273,
+      "mean_token_accuracy": 0.9281989872455597,
+      "num_tokens": 39207.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.157628071308136,
+      "epoch": 0.03824091778202677,
+      "grad_norm": 0.5328208804130554,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1924,
+      "mean_token_accuracy": 0.9389224052429199,
+      "num_tokens": 52869.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.1558964610099793,
+      "epoch": 0.04780114722753346,
+      "grad_norm": 0.879055917263031,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.2048,
+      "mean_token_accuracy": 0.9312818169593811,
+      "num_tokens": 66622.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.1293084979057313,
+      "epoch": 0.05736137667304015,
+      "grad_norm": 0.6380035281181335,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1678,
+      "mean_token_accuracy": 0.940121340751648,
+      "num_tokens": 80175.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.1357115745544433,
+      "epoch": 0.06692160611854685,
+      "grad_norm": 0.7072647213935852,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1967,
+      "mean_token_accuracy": 0.9296577990055084,
+      "num_tokens": 94356.0,
+      "step": 70
+    },
+    {
+      "entropy": 1.1026614785194397,
+      "epoch": 0.07648183556405354,
+      "grad_norm": 0.771527111530304,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1522,
+      "mean_token_accuracy": 0.9473809123039245,
+      "num_tokens": 108716.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.0664347410202026,
+      "epoch": 0.08604206500956023,
+      "grad_norm": 0.6657344698905945,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1552,
+      "mean_token_accuracy": 0.9406381964683532,
+      "num_tokens": 122182.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0743921160697938,
+      "epoch": 0.09560229445506692,
+      "grad_norm": 0.6942067742347717,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1454,
+      "mean_token_accuracy": 0.9483474791049957,
+      "num_tokens": 135713.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0707114934921265,
+      "epoch": 0.10516252390057361,
+      "grad_norm": 0.9185185432434082,
+      "learning_rate": 9.999900008621665e-05,
+      "loss": 0.1608,
+      "mean_token_accuracy": 0.9425420522689819,
+      "num_tokens": 149456.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.0588605165481568,
+      "epoch": 0.1147227533460803,
+      "grad_norm": 0.4680742025375366,
+      "learning_rate": 9.998775151544478e-05,
+      "loss": 0.147,
+      "mean_token_accuracy": 0.9443056762218476,
+      "num_tokens": 163062.0,
+      "step": 120
+    },
+    {
+      "entropy": 1.0524839282035827,
+      "epoch": 0.124282982791587,
+      "grad_norm": 0.7786092162132263,
+      "learning_rate": 9.996400730289628e-05,
+      "loss": 0.1691,
+      "mean_token_accuracy": 0.935369998216629,
+      "num_tokens": 176574.0,
+      "step": 130
+    },
+    {
+      "entropy": 1.0402762174606324,
+      "epoch": 0.1338432122370937,
+      "grad_norm": 0.691507875919342,
+      "learning_rate": 9.992777338400864e-05,
+      "loss": 0.1629,
+      "mean_token_accuracy": 0.9382378816604614,
+      "num_tokens": 191164.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0201380014419557,
+      "epoch": 0.14340344168260039,
+      "grad_norm": 0.7071655988693237,
+      "learning_rate": 9.987905881632213e-05,
+      "loss": 0.1477,
+      "mean_token_accuracy": 0.9456806540489197,
+      "num_tokens": 204805.0,
+      "step": 150
+    },
+    {
+      "entropy": 1.0177514791488647,
+      "epoch": 0.15296367112810708,
+      "grad_norm": 0.7090482711791992,
+      "learning_rate": 9.981787577721553e-05,
+      "loss": 0.131,
+      "mean_token_accuracy": 0.9504908204078675,
+      "num_tokens": 218021.0,
+      "step": 160
+    },
+    {
+      "entropy": 1.0016795158386231,
+      "epoch": 0.16252390057361377,
+      "grad_norm": 0.5258442759513855,
+      "learning_rate": 9.974423956086228e-05,
+      "loss": 0.1243,
+      "mean_token_accuracy": 0.9528130173683167,
+      "num_tokens": 231015.0,
+      "step": 170
+    },
+    {
+      "entropy": 0.9992186069488526,
+      "epoch": 0.17208413001912046,
+      "grad_norm": 0.6269602179527283,
+      "learning_rate": 9.965816857440717e-05,
+      "loss": 0.1438,
+      "mean_token_accuracy": 0.9449268460273743,
+      "num_tokens": 244021.0,
+      "step": 180
+    },
+    {
+      "entropy": 1.0048776805400848,
+      "epoch": 0.18164435946462715,
+      "grad_norm": 0.8240458369255066,
+      "learning_rate": 9.955968433336515e-05,
+      "loss": 0.14,
+      "mean_token_accuracy": 0.9496095240116119,
+      "num_tokens": 256981.0,
+      "step": 190
+    },
+    {
+      "entropy": 0.9974187612533569,
+      "epoch": 0.19120458891013384,
+      "grad_norm": 0.5255606174468994,
+      "learning_rate": 9.944881145624291e-05,
+      "loss": 0.1483,
+      "mean_token_accuracy": 0.9459229350090027,
+      "num_tokens": 270999.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.9903082370758056,
+      "epoch": 0.20076481835564053,
+      "grad_norm": 0.6685144901275635,
+      "learning_rate": 9.932557765838504e-05,
+      "loss": 0.1193,
+      "mean_token_accuracy": 0.9607025504112243,
+      "num_tokens": 283995.0,
+      "step": 210
+    },
+    {
+      "entropy": 0.9902467429637909,
+      "epoch": 0.21032504780114722,
+      "grad_norm": 0.6727855205535889,
+      "learning_rate": 9.919001374504569e-05,
+      "loss": 0.1347,
+      "mean_token_accuracy": 0.9496286571025848,
+      "num_tokens": 297434.0,
+      "step": 220
+    },
+    {
+      "entropy": 1.0041028022766114,
+      "epoch": 0.2198852772466539,
+      "grad_norm": 0.5113528370857239,
+      "learning_rate": 9.904215360368826e-05,
+      "loss": 0.1257,
+      "mean_token_accuracy": 0.9506474733352661,
+      "num_tokens": 310688.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.0222025215625763,
+      "epoch": 0.2294455066921606,
+      "grad_norm": 1.1373951435089111,
+      "learning_rate": 9.888203419551432e-05,
+      "loss": 0.1574,
+      "mean_token_accuracy": 0.9421513736248016,
+      "num_tokens": 324594.0,
+      "step": 240
+    },
+    {
+      "entropy": 1.0387476086616516,
+      "epoch": 0.2390057361376673,
+      "grad_norm": 0.8162603974342346,
+      "learning_rate": 9.870969554622426e-05,
+      "loss": 0.153,
+      "mean_token_accuracy": 0.9408722400665284,
+      "num_tokens": 338003.0,
+      "step": 250
+    },
+    {
+      "entropy": 1.0153288543224335,
+      "epoch": 0.248565965583174,
+      "grad_norm": 0.49323147535324097,
+      "learning_rate": 9.852518073601194e-05,
+      "loss": 0.1295,
+      "mean_token_accuracy": 0.9526264905929566,
+      "num_tokens": 351394.0,
+      "step": 260
+    },
+    {
+      "entropy": 0.9894533395767212,
+      "epoch": 0.25812619502868067,
+      "grad_norm": 0.5005015134811401,
+      "learning_rate": 9.832853588879575e-05,
+      "loss": 0.1258,
+      "mean_token_accuracy": 0.9527800977230072,
+      "num_tokens": 365081.0,
+      "step": 270
+    },
+    {
+      "entropy": 1.0046591579914093,
+      "epoch": 0.2676864244741874,
+      "grad_norm": 0.5181300640106201,
+      "learning_rate": 9.811981016068881e-05,
+      "loss": 0.1298,
+      "mean_token_accuracy": 0.9493179559707642,
+      "num_tokens": 378518.0,
+      "step": 280
+    },
+    {
+      "entropy": 1.0015001237392425,
+      "epoch": 0.27724665391969405,
+      "grad_norm": 0.8453997373580933,
+      "learning_rate": 9.78990557277112e-05,
+      "loss": 0.1269,
+      "mean_token_accuracy": 0.9505164086818695,
+      "num_tokens": 391993.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0173242926597594,
+      "epoch": 0.28680688336520077,
+      "grad_norm": 0.665677011013031,
+      "learning_rate": 9.766632777274738e-05,
+      "loss": 0.1071,
+      "mean_token_accuracy": 0.9552569091320038,
+      "num_tokens": 405246.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0674896359443664,
+      "epoch": 0.29636711281070743,
+      "grad_norm": 0.8276544213294983,
+      "learning_rate": 9.742168447175182e-05,
+      "loss": 0.1275,
+      "mean_token_accuracy": 0.9511016726493835,
+      "num_tokens": 418524.0,
+      "step": 310
+    },
+    {
+      "entropy": 1.0123964190483092,
+      "epoch": 0.30592734225621415,
+      "grad_norm": 0.6688571572303772,
+      "learning_rate": 9.716518697920655e-05,
+      "loss": 0.1363,
+      "mean_token_accuracy": 0.9480633080005646,
+      "num_tokens": 432202.0,
+      "step": 320
+    },
+    {
+      "entropy": 0.999013239145279,
+      "epoch": 0.3154875717017208,
+      "grad_norm": 0.4304344952106476,
+      "learning_rate": 9.689689941283417e-05,
+      "loss": 0.1233,
+      "mean_token_accuracy": 0.9515203893184662,
+      "num_tokens": 445903.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.0126557409763337,
+      "epoch": 0.32504780114722753,
+      "grad_norm": 0.5183528661727905,
+      "learning_rate": 9.661688883756996e-05,
+      "loss": 0.1209,
+      "mean_token_accuracy": 0.9540857017040253,
+      "num_tokens": 460277.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0066914319992066,
+      "epoch": 0.33460803059273425,
+      "grad_norm": 0.6539915800094604,
+      "learning_rate": 9.632522524879756e-05,
+      "loss": 0.123,
+      "mean_token_accuracy": 0.9566046714782714,
+      "num_tokens": 473892.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.0084527671337127,
+      "epoch": 0.3441682600382409,
+      "grad_norm": 0.6190230250358582,
+      "learning_rate": 9.602198155485176e-05,
+      "loss": 0.1034,
+      "mean_token_accuracy": 0.9617183327674865,
+      "num_tokens": 487145.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.0045492053031921,
+      "epoch": 0.35372848948374763,
+      "grad_norm": 0.745010256767273,
+      "learning_rate": 9.570723355879342e-05,
+      "loss": 0.1208,
+      "mean_token_accuracy": 0.9523294031620025,
+      "num_tokens": 500481.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.0438226521015168,
+      "epoch": 0.3632887189292543,
+      "grad_norm": 0.5804723501205444,
+      "learning_rate": 9.538105993946068e-05,
+      "loss": 0.1247,
+      "mean_token_accuracy": 0.9556700885295868,
+      "num_tokens": 513817.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0288837611675263,
+      "epoch": 0.372848948374761,
+      "grad_norm": 0.45516908168792725,
+      "learning_rate": 9.504354223180118e-05,
+      "loss": 0.1298,
+      "mean_token_accuracy": 0.9496672093868256,
+      "num_tokens": 527250.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.0271916389465332,
+      "epoch": 0.3824091778202677,
+      "grad_norm": 0.4186956286430359,
+      "learning_rate": 9.469476480649056e-05,
+      "loss": 0.1234,
+      "mean_token_accuracy": 0.9543138742446899,
+      "num_tokens": 541558.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.006220853328705,
+      "epoch": 0.3919694072657744,
+      "grad_norm": 0.6684417128562927,
+      "learning_rate": 9.433481484884189e-05,
+      "loss": 0.1086,
+      "mean_token_accuracy": 0.9610184013843537,
+      "num_tokens": 554540.0,
+      "step": 410
+    },
+    {
+      "entropy": 1.0208543598651887,
+      "epoch": 0.40152963671128106,
+      "grad_norm": 0.46702873706817627,
+      "learning_rate": 9.396378233701167e-05,
+      "loss": 0.1148,
+      "mean_token_accuracy": 0.9523889362812042,
+      "num_tokens": 568221.0,
+      "step": 420
+    },
+    {
+      "entropy": 1.026773202419281,
+      "epoch": 0.4110898661567878,
+      "grad_norm": 0.6417613625526428,
+      "learning_rate": 9.358176001950759e-05,
+      "loss": 0.1229,
+      "mean_token_accuracy": 0.9499638915061951,
+      "num_tokens": 581721.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.0295559406280517,
+      "epoch": 0.42065009560229444,
+      "grad_norm": 0.8544287085533142,
+      "learning_rate": 9.318884339200378e-05,
+      "loss": 0.1145,
+      "mean_token_accuracy": 0.9570431888103486,
+      "num_tokens": 595120.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0547672986984253,
+      "epoch": 0.43021032504780116,
+      "grad_norm": 0.5602120757102966,
+      "learning_rate": 9.278513067346941e-05,
+      "loss": 0.1205,
+      "mean_token_accuracy": 0.9526568710803985,
+      "num_tokens": 608361.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0596372604370117,
+      "epoch": 0.4397705544933078,
+      "grad_norm": 0.9372671842575073,
+      "learning_rate": 9.237072278161652e-05,
+      "loss": 0.1158,
+      "mean_token_accuracy": 0.9582056403160095,
+      "num_tokens": 621843.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0450125575065612,
+      "epoch": 0.44933078393881454,
+      "grad_norm": 0.7245158553123474,
+      "learning_rate": 9.194572330767314e-05,
+      "loss": 0.1084,
+      "mean_token_accuracy": 0.956211519241333,
+      "num_tokens": 635552.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.0532810688018799,
+      "epoch": 0.4588910133843212,
+      "grad_norm": 0.5808508396148682,
+      "learning_rate": 9.151023849048818e-05,
+      "loss": 0.1301,
+      "mean_token_accuracy": 0.9499321818351746,
+      "num_tokens": 648523.0,
+      "step": 480
+    },
+    {
+      "entropy": 1.0488809943199158,
+      "epoch": 0.4684512428298279,
+      "grad_norm": 0.33145713806152344,
+      "learning_rate": 9.106437718997452e-05,
+      "loss": 0.1085,
+      "mean_token_accuracy": 0.9597950100898742,
+      "num_tokens": 661850.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0259310245513915,
+      "epoch": 0.4780114722753346,
+      "grad_norm": 0.9813805222511292,
+      "learning_rate": 9.060825085989682e-05,
+      "loss": 0.101,
+      "mean_token_accuracy": 0.9605780482292176,
+      "num_tokens": 675322.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.0419854640960693,
+      "epoch": 0.4875717017208413,
+      "grad_norm": 0.7074224948883057,
+      "learning_rate": 9.014197352001097e-05,
+      "loss": 0.1219,
+      "mean_token_accuracy": 0.9519591212272644,
+      "num_tokens": 688704.0,
+      "step": 510
+    },
+    {
+      "entropy": 1.0370671927928925,
+      "epoch": 0.497131931166348,
+      "grad_norm": 0.7930039167404175,
+      "learning_rate": 8.966566172756213e-05,
+      "loss": 0.1268,
+      "mean_token_accuracy": 0.9527601838111878,
+      "num_tokens": 702579.0,
+      "step": 520
+    },
+    {
+      "entropy": 1.0266020059585572,
+      "epoch": 0.5066921606118547,
+      "grad_norm": 0.5883492231369019,
+      "learning_rate": 8.91794345481484e-05,
+      "loss": 0.1146,
+      "mean_token_accuracy": 0.9561169147491455,
+      "num_tokens": 716160.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0262812852859498,
+      "epoch": 0.5162523900573613,
+      "grad_norm": 0.8996061682701111,
+      "learning_rate": 8.868341352595755e-05,
+      "loss": 0.1253,
+      "mean_token_accuracy": 0.9532574653625489,
+      "num_tokens": 729408.0,
+      "step": 540
+    },
+    {
+      "entropy": 1.032131028175354,
+      "epoch": 0.5258126195028681,
+      "grad_norm": 0.5681365728378296,
+      "learning_rate": 8.817772265338409e-05,
+      "loss": 0.114,
+      "mean_token_accuracy": 0.9511450946331024,
+      "num_tokens": 742824.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.0298707604408264,
+      "epoch": 0.5353728489483748,
+      "grad_norm": 0.9023966193199158,
+      "learning_rate": 8.766248834003441e-05,
+      "loss": 0.1189,
+      "mean_token_accuracy": 0.9529127359390259,
+      "num_tokens": 756778.0,
+      "step": 560
+    },
+    {
+      "entropy": 1.0299117267131805,
+      "epoch": 0.5449330783938815,
+      "grad_norm": 0.5660295486450195,
+      "learning_rate": 8.713783938112768e-05,
+      "loss": 0.1152,
+      "mean_token_accuracy": 0.9536526381969452,
+      "num_tokens": 770342.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.034860408306122,
+      "epoch": 0.5544933078393881,
+      "grad_norm": 0.44493526220321655,
+      "learning_rate": 8.660390692530026e-05,
+      "loss": 0.1109,
+      "mean_token_accuracy": 0.9568422317504883,
+      "num_tokens": 784398.0,
+      "step": 580
+    },
+    {
+      "entropy": 1.0222409248352051,
+      "epoch": 0.5640535372848948,
+      "grad_norm": 0.5718162655830383,
+      "learning_rate": 8.606082444182216e-05,
+      "loss": 0.1181,
+      "mean_token_accuracy": 0.9559833109378815,
+      "num_tokens": 798017.0,
+      "step": 590
+    },
+    {
+      "entropy": 1.0099391400814057,
+      "epoch": 0.5736137667304015,
+      "grad_norm": 0.7805862426757812,
+      "learning_rate": 8.550872768723292e-05,
+      "loss": 0.1133,
+      "mean_token_accuracy": 0.9571761429309845,
+      "num_tokens": 811336.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.9928421020507813,
+      "epoch": 0.5831739961759083,
+      "grad_norm": 0.6475667357444763,
+      "learning_rate": 8.494775467140627e-05,
+      "loss": 0.1228,
+      "mean_token_accuracy": 0.9529062807559967,
+      "num_tokens": 824727.0,
+      "step": 610
+    },
+    {
+      "entropy": 0.9883149087429046,
+      "epoch": 0.5927342256214149,
+      "grad_norm": 0.5568240880966187,
+      "learning_rate": 8.4378045623051e-05,
+      "loss": 0.1215,
+      "mean_token_accuracy": 0.9517284870147705,
+      "num_tokens": 838105.0,
+      "step": 620
+    },
+    {
+      "entropy": 0.9852628886699677,
+      "epoch": 0.6022944550669216,
+      "grad_norm": 0.8160543441772461,
+      "learning_rate": 8.379974295465758e-05,
+      "loss": 0.0984,
+      "mean_token_accuracy": 0.9631519913673401,
+      "num_tokens": 851280.0,
+      "step": 630
+    },
+    {
+      "entropy": 1.0070119798183441,
+      "epoch": 0.6118546845124283,
+      "grad_norm": 0.522103488445282,
+      "learning_rate": 8.321299122689863e-05,
+      "loss": 0.1038,
+      "mean_token_accuracy": 0.9595254600048065,
+      "num_tokens": 864939.0,
+      "step": 640
+    },
+    {
+      "entropy": 0.9948669731616974,
+      "epoch": 0.621414913957935,
+      "grad_norm": 1.0888185501098633,
+      "learning_rate": 8.261793711249254e-05,
+      "loss": 0.1158,
+      "mean_token_accuracy": 0.95493443608284,
+      "num_tokens": 878886.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.9878204762935638,
+      "epoch": 0.6309751434034416,
+      "grad_norm": 0.8432512283325195,
+      "learning_rate": 8.201472935953916e-05,
+      "loss": 0.0874,
+      "mean_token_accuracy": 0.9673986077308655,
+      "num_tokens": 892413.0,
+      "step": 660
+    },
+    {
+      "entropy": 0.9975221931934357,
+      "epoch": 0.6405353728489483,
+      "grad_norm": 0.5733245611190796,
+      "learning_rate": 8.140351875433649e-05,
+      "loss": 0.1085,
+      "mean_token_accuracy": 0.9571276009082794,
+      "num_tokens": 906485.0,
+      "step": 670
+    },
+    {
+      "entropy": 0.997295218706131,
+      "epoch": 0.6500956022944551,
+      "grad_norm": 0.8385879993438721,
+      "learning_rate": 8.07844580836881e-05,
+      "loss": 0.1129,
+      "mean_token_accuracy": 0.9515101313591003,
+      "num_tokens": 920020.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9872783482074737,
+      "epoch": 0.6596558317399618,
+      "grad_norm": 0.9112151265144348,
+      "learning_rate": 8.015770209671044e-05,
+      "loss": 0.0998,
+      "mean_token_accuracy": 0.9611438810825348,
+      "num_tokens": 933577.0,
+      "step": 690
+    },
+    {
+      "entropy": 0.9857029139995575,
+      "epoch": 0.6692160611854685,
+      "grad_norm": 0.39040815830230713,
+      "learning_rate": 7.952340746614946e-05,
+      "loss": 0.1011,
+      "mean_token_accuracy": 0.962991064786911,
+      "num_tokens": 946367.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.9934593677520752,
+      "epoch": 0.6787762906309751,
+      "grad_norm": 0.40705567598342896,
+      "learning_rate": 7.888173274921655e-05,
+      "loss": 0.1085,
+      "mean_token_accuracy": 0.9592144548892975,
+      "num_tokens": 960252.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9805512070655823,
+      "epoch": 0.6883365200764818,
+      "grad_norm": 0.7323410511016846,
+      "learning_rate": 7.823283834795336e-05,
+      "loss": 0.0972,
+      "mean_token_accuracy": 0.9624641418457032,
+      "num_tokens": 973491.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9763809263706207,
+      "epoch": 0.6978967495219885,
+      "grad_norm": 0.4441353380680084,
+      "learning_rate": 7.75768864691354e-05,
+      "loss": 0.1195,
+      "mean_token_accuracy": 0.9537525713443756,
+      "num_tokens": 986879.0,
+      "step": 730
+    },
+    {
+      "entropy": 0.9949767231941223,
+      "epoch": 0.7074569789674953,
+      "grad_norm": 0.6540898680686951,
+      "learning_rate": 7.691404108372461e-05,
+      "loss": 0.1033,
+      "mean_token_accuracy": 0.9608555018901825,
+      "num_tokens": 1000272.0,
+      "step": 740
+    },
+    {
+      "entropy": 0.9889988899230957,
+      "epoch": 0.7170172084130019,
+      "grad_norm": 0.7363319396972656,
+      "learning_rate": 7.624446788588085e-05,
+      "loss": 0.1171,
+      "mean_token_accuracy": 0.9569859743118286,
+      "num_tokens": 1013908.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.9795958340168,
+      "epoch": 0.7265774378585086,
+      "grad_norm": 0.6467368006706238,
+      "learning_rate": 7.556833425154255e-05,
+      "loss": 0.0902,
+      "mean_token_accuracy": 0.9675116181373596,
+      "num_tokens": 1027085.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9736341655254364,
+      "epoch": 0.7361376673040153,
+      "grad_norm": 0.5900027751922607,
+      "learning_rate": 7.488580919658727e-05,
+      "loss": 0.1207,
+      "mean_token_accuracy": 0.9517757058143616,
+      "num_tokens": 1040367.0,
+      "step": 770
+    },
+    {
+      "entropy": 1.0007605075836181,
+      "epoch": 0.745697896749522,
+      "grad_norm": 0.7509295344352722,
+      "learning_rate": 7.419706333458201e-05,
+      "loss": 0.1052,
+      "mean_token_accuracy": 0.9591726422309875,
+      "num_tokens": 1054076.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9726886987686157,
+      "epoch": 0.7552581261950286,
+      "grad_norm": 0.47083351016044617,
+      "learning_rate": 7.350226883413413e-05,
+      "loss": 0.0899,
+      "mean_token_accuracy": 0.9645619034767151,
+      "num_tokens": 1068530.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9697287023067475,
+      "epoch": 0.7648183556405354,
+      "grad_norm": 0.8393839597702026,
+      "learning_rate": 7.280159937585386e-05,
+      "loss": 0.1074,
+      "mean_token_accuracy": 0.9581287324428558,
+      "num_tokens": 1082342.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.9735282719135284,
+      "epoch": 0.7743785850860421,
+      "grad_norm": 0.40682822465896606,
+      "learning_rate": 7.209523010893853e-05,
+      "loss": 0.1045,
+      "mean_token_accuracy": 0.9595387160778046,
+      "num_tokens": 1096496.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.9854253113269806,
+      "epoch": 0.7839388145315488,
+      "grad_norm": 0.4829469621181488,
+      "learning_rate": 7.138333760738987e-05,
+      "loss": 0.1118,
+      "mean_token_accuracy": 0.9563177168369293,
+      "num_tokens": 1109906.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9937235653400421,
+      "epoch": 0.7934990439770554,
+      "grad_norm": 0.41137319803237915,
+      "learning_rate": 7.066609982587501e-05,
+      "loss": 0.1177,
+      "mean_token_accuracy": 0.9555411994457245,
+      "num_tokens": 1123170.0,
+      "step": 830
+    },
+    {
+      "entropy": 0.9826422870159149,
+      "epoch": 0.8030592734225621,
+      "grad_norm": 0.41072991490364075,
+      "learning_rate": 6.994369605524253e-05,
+      "loss": 0.1067,
+      "mean_token_accuracy": 0.959008538722992,
+      "num_tokens": 1136503.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9492685794830322,
+      "epoch": 0.8126195028680688,
+      "grad_norm": 0.7331626415252686,
+      "learning_rate": 6.921630687770427e-05,
+      "loss": 0.0897,
+      "mean_token_accuracy": 0.9662109792232514,
+      "num_tokens": 1149719.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.9468518316745758,
+      "epoch": 0.8221797323135756,
+      "grad_norm": 0.48760515451431274,
+      "learning_rate": 6.848411412169461e-05,
+      "loss": 0.0947,
+      "mean_token_accuracy": 0.961515873670578,
+      "num_tokens": 1163320.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9556315541267395,
+      "epoch": 0.8317399617590823,
+      "grad_norm": 0.575470507144928,
+      "learning_rate": 6.774730081641799e-05,
+      "loss": 0.0944,
+      "mean_token_accuracy": 0.9629533231258393,
+      "num_tokens": 1177204.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.9477540969848632,
+      "epoch": 0.8413001912045889,
+      "grad_norm": 0.731942892074585,
+      "learning_rate": 6.70060511460963e-05,
+      "loss": 0.1026,
+      "mean_token_accuracy": 0.9615570724010467,
+      "num_tokens": 1190225.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9491175174713135,
+      "epoch": 0.8508604206500956,
+      "grad_norm": 0.5208237767219543,
+      "learning_rate": 6.626055040392774e-05,
+      "loss": 0.0934,
+      "mean_token_accuracy": 0.9632180333137512,
+      "num_tokens": 1204511.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9400201737880707,
+      "epoch": 0.8604206500956023,
+      "grad_norm": 0.5904267430305481,
+      "learning_rate": 6.551098494576819e-05,
+      "loss": 0.0987,
+      "mean_token_accuracy": 0.9621688187122345,
+      "num_tokens": 1217217.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9426032364368438,
+      "epoch": 0.869980879541109,
+      "grad_norm": 0.6165425777435303,
+      "learning_rate": 6.475754214354715e-05,
+      "loss": 0.095,
+      "mean_token_accuracy": 0.9672666728496552,
+      "num_tokens": 1230535.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.9551124751567841,
+      "epoch": 0.8795411089866156,
+      "grad_norm": 0.6224029064178467,
+      "learning_rate": 6.400041033842965e-05,
+      "loss": 0.1111,
+      "mean_token_accuracy": 0.9552771091461182,
+      "num_tokens": 1244121.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.961413562297821,
+      "epoch": 0.8891013384321224,
+      "grad_norm": 0.447574645280838,
+      "learning_rate": 6.323977879373571e-05,
+      "loss": 0.1119,
+      "mean_token_accuracy": 0.9559538304805756,
+      "num_tokens": 1257972.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.972063022851944,
+      "epoch": 0.8986615678776291,
+      "grad_norm": 0.7507049441337585,
+      "learning_rate": 6.247583764762959e-05,
+      "loss": 0.0977,
+      "mean_token_accuracy": 0.9586446702480316,
+      "num_tokens": 1271568.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9554756462574006,
+      "epoch": 0.9082217973231358,
+      "grad_norm": 1.0878021717071533,
+      "learning_rate": 6.170877786559001e-05,
+      "loss": 0.094,
+      "mean_token_accuracy": 0.9635938465595245,
+      "num_tokens": 1285826.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.9407235264778138,
+      "epoch": 0.9177820267686424,
+      "grad_norm": 0.48441722989082336,
+      "learning_rate": 6.0938791192673895e-05,
+      "loss": 0.098,
+      "mean_token_accuracy": 0.962267804145813,
+      "num_tokens": 1299889.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9443095207214356,
+      "epoch": 0.9273422562141491,
+      "grad_norm": 0.678026556968689,
+      "learning_rate": 6.016607010558492e-05,
+      "loss": 0.0867,
+      "mean_token_accuracy": 0.967143052816391,
+      "num_tokens": 1313610.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9234336674213409,
+      "epoch": 0.9369024856596558,
+      "grad_norm": 0.5790787935256958,
+      "learning_rate": 5.939080776455942e-05,
+      "loss": 0.0941,
+      "mean_token_accuracy": 0.9665516257286072,
+      "num_tokens": 1327185.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.9353838205337525,
+      "epoch": 0.9464627151051626,
+      "grad_norm": 0.8568056225776672,
+      "learning_rate": 5.861319796508139e-05,
+      "loss": 0.1104,
+      "mean_token_accuracy": 0.9536222815513611,
+      "num_tokens": 1340176.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.939469438791275,
+      "epoch": 0.9560229445506692,
+      "grad_norm": 0.8979282975196838,
+      "learning_rate": 5.783343508943849e-05,
+      "loss": 0.0993,
+      "mean_token_accuracy": 0.963096821308136,
+      "num_tokens": 1354053.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9367132842540741,
+      "epoch": 0.9655831739961759,
+      "grad_norm": 0.4966347813606262,
+      "learning_rate": 5.7051714058131654e-05,
+      "loss": 0.1023,
+      "mean_token_accuracy": 0.9576547324657441,
+      "num_tokens": 1367199.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.9079373300075531,
+      "epoch": 0.9751434034416826,
+      "grad_norm": 0.6267583966255188,
+      "learning_rate": 5.626823028114991e-05,
+      "loss": 0.0728,
+      "mean_token_accuracy": 0.9701129257678985,
+      "num_tokens": 1380205.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9324047029018402,
+      "epoch": 0.9847036328871893,
+      "grad_norm": 0.7952474355697632,
+      "learning_rate": 5.548317960912305e-05,
+      "loss": 0.0931,
+      "mean_token_accuracy": 0.959660267829895,
+      "num_tokens": 1392623.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9379731714725494,
+      "epoch": 0.994263862332696,
+      "grad_norm": 0.6732680797576904,
+      "learning_rate": 5.4696758284363956e-05,
+      "loss": 0.0904,
+      "mean_token_accuracy": 0.9672562539577484,
+      "num_tokens": 1406093.0,
+      "step": 1040
+    },
+    {
+      "entropy": 0.9419908046722412,
+      "epoch": 1.0038240917782026,
+      "grad_norm": 0.46845611929893494,
+      "learning_rate": 5.390916289181318e-05,
+      "loss": 0.0806,
+      "mean_token_accuracy": 0.970388001203537,
+      "num_tokens": 1419395.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.9173365116119385,
+      "epoch": 1.0133843212237095,
+      "grad_norm": 0.4297061264514923,
+      "learning_rate": 5.312059030989779e-05,
+      "loss": 0.0506,
+      "mean_token_accuracy": 0.9845799207687378,
+      "num_tokens": 1432857.0,
+      "step": 1060
+    },
+    {
+      "entropy": 0.8889394164085388,
+      "epoch": 1.022944550669216,
+      "grad_norm": 0.39711156487464905,
+      "learning_rate": 5.233123766131681e-05,
+      "loss": 0.0442,
+      "mean_token_accuracy": 0.9859949052333832,
+      "num_tokens": 1445639.0,
+      "step": 1070
+    },
+    {
+      "entropy": 0.8785762906074523,
+      "epoch": 1.0325047801147227,
+      "grad_norm": 0.7389956712722778,
+      "learning_rate": 5.154130226376578e-05,
+      "loss": 0.0519,
+      "mean_token_accuracy": 0.9830816686153412,
+      "num_tokens": 1458461.0,
+      "step": 1080
+    },
+    {
+      "entropy": 0.9009840846061706,
+      "epoch": 1.0420650095602295,
+      "grad_norm": 0.423625648021698,
+      "learning_rate": 5.0750981580612277e-05,
+      "loss": 0.0504,
+      "mean_token_accuracy": 0.9819521844387055,
+      "num_tokens": 1472067.0,
+      "step": 1090
+    },
+    {
+      "entropy": 0.8976073801517487,
+      "epoch": 1.0516252390057361,
+      "grad_norm": 0.2846389412879944,
+      "learning_rate": 4.996047317153538e-05,
+      "loss": 0.0461,
+      "mean_token_accuracy": 0.9830753445625305,
+      "num_tokens": 1486092.0,
+      "step": 1100
+    },
+    {
+      "entropy": 0.8968881011009217,
+      "epoch": 1.0611854684512427,
+      "grad_norm": 0.95218425989151,
+      "learning_rate": 4.916997464314074e-05,
+      "loss": 0.0503,
+      "mean_token_accuracy": 0.9824698865413666,
+      "num_tokens": 1500233.0,
+      "step": 1110
+    },
+    {
+      "entropy": 0.8870066285133362,
+      "epoch": 1.0707456978967496,
+      "grad_norm": 0.632781982421875,
+      "learning_rate": 4.8379683599564086e-05,
+      "loss": 0.0455,
+      "mean_token_accuracy": 0.9851963460445404,
+      "num_tokens": 1513865.0,
+      "step": 1120
+    },
+    {
+      "entropy": 0.8961619734764099,
+      "epoch": 1.0803059273422562,
+      "grad_norm": 0.39181652665138245,
+      "learning_rate": 4.7589797593075334e-05,
+      "loss": 0.0558,
+      "mean_token_accuracy": 0.9790576815605163,
+      "num_tokens": 1527318.0,
+      "step": 1130
+    },
+    {
+      "entropy": 0.9034638643264771,
+      "epoch": 1.089866156787763,
+      "grad_norm": 0.43669965863227844,
+      "learning_rate": 4.6800514074695626e-05,
+      "loss": 0.0632,
+      "mean_token_accuracy": 0.9821757674217224,
+      "num_tokens": 1540076.0,
+      "step": 1140
+    },
+    {
+      "entropy": 0.910722941160202,
+      "epoch": 1.0994263862332696,
+      "grad_norm": 0.35479944944381714,
+      "learning_rate": 4.601203034483971e-05,
+      "loss": 0.0565,
+      "mean_token_accuracy": 0.9798229336738586,
+      "num_tokens": 1553989.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.8995626032352447,
+      "epoch": 1.1089866156787762,
+      "grad_norm": 0.8469728231430054,
+      "learning_rate": 4.522454350399592e-05,
+      "loss": 0.0553,
+      "mean_token_accuracy": 0.9812585413455963,
+      "num_tokens": 1568087.0,
+      "step": 1160
+    },
+    {
+      "entropy": 0.8922906994819642,
+      "epoch": 1.118546845124283,
+      "grad_norm": 0.5808919072151184,
+      "learning_rate": 4.4438250403456187e-05,
+      "loss": 0.0412,
+      "mean_token_accuracy": 0.985135817527771,
+      "num_tokens": 1581639.0,
+      "step": 1170
+    },
+    {
+      "entropy": 0.8814689695835114,
+      "epoch": 1.1281070745697896,
+      "grad_norm": 0.27558010816574097,
+      "learning_rate": 4.365334759610836e-05,
+      "loss": 0.0464,
+      "mean_token_accuracy": 0.9845982074737549,
+      "num_tokens": 1594420.0,
+      "step": 1180
+    },
+    {
+      "entropy": 0.8946828246116638,
+      "epoch": 1.1376673040152965,
+      "grad_norm": 0.46020734310150146,
+      "learning_rate": 4.2870031287302984e-05,
+      "loss": 0.052,
+      "mean_token_accuracy": 0.9826988756656647,
+      "num_tokens": 1608313.0,
+      "step": 1190
+    },
+    {
+      "entropy": 0.8807225048542022,
+      "epoch": 1.147227533460803,
+      "grad_norm": 0.36476823687553406,
+      "learning_rate": 4.2088497285807125e-05,
+      "loss": 0.055,
+      "mean_token_accuracy": 0.9815667152404786,
+      "num_tokens": 1621999.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.8784381151199341,
+      "epoch": 1.1567877629063097,
+      "grad_norm": 0.16498547792434692,
+      "learning_rate": 4.1308940954857236e-05,
+      "loss": 0.0439,
+      "mean_token_accuracy": 0.9830247521400451,
+      "num_tokens": 1635960.0,
+      "step": 1210
+    },
+    {
+      "entropy": 0.8774892151355743,
+      "epoch": 1.1663479923518165,
+      "grad_norm": 0.6744072437286377,
+      "learning_rate": 4.0531557163323415e-05,
+      "loss": 0.0395,
+      "mean_token_accuracy": 0.9868780970573425,
+      "num_tokens": 1648974.0,
+      "step": 1220
+    },
+    {
+      "entropy": 0.8856045424938201,
+      "epoch": 1.1759082217973231,
+      "grad_norm": 0.6424488425254822,
+      "learning_rate": 3.975654023699722e-05,
+      "loss": 0.0545,
+      "mean_token_accuracy": 0.981811261177063,
+      "num_tokens": 1662877.0,
+      "step": 1230
+    },
+    {
+      "entropy": 0.8873432040214538,
+      "epoch": 1.1854684512428297,
+      "grad_norm": 0.5239076018333435,
+      "learning_rate": 3.898408391001528e-05,
+      "loss": 0.0413,
+      "mean_token_accuracy": 0.9845008969306945,
+      "num_tokens": 1676115.0,
+      "step": 1240
+    },
+    {
+      "entropy": 0.884037721157074,
+      "epoch": 1.1950286806883366,
+      "grad_norm": 0.6090591549873352,
+      "learning_rate": 3.821438127643078e-05,
+      "loss": 0.0545,
+      "mean_token_accuracy": 0.9792491316795349,
+      "num_tokens": 1689478.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.8866362869739532,
+      "epoch": 1.2045889101338432,
+      "grad_norm": 0.4021357297897339,
+      "learning_rate": 3.744762474194501e-05,
+      "loss": 0.0437,
+      "mean_token_accuracy": 0.9870902061462402,
+      "num_tokens": 1703240.0,
+      "step": 1260
+    },
+    {
+      "entropy": 0.8797494530677795,
+      "epoch": 1.21414913957935,
+      "grad_norm": 0.39826473593711853,
+      "learning_rate": 3.6684005975810765e-05,
+      "loss": 0.0453,
+      "mean_token_accuracy": 0.9834677696228027,
+      "num_tokens": 1716712.0,
+      "step": 1270
+    },
+    {
+      "entropy": 0.8738825917243958,
+      "epoch": 1.2237093690248566,
+      "grad_norm": 0.6343350410461426,
+      "learning_rate": 3.592371586292021e-05,
+      "loss": 0.0399,
+      "mean_token_accuracy": 0.9875729084014893,
+      "num_tokens": 1729905.0,
+      "step": 1280
+    },
+    {
+      "entropy": 0.8818942248821259,
+      "epoch": 1.2332695984703632,
+      "grad_norm": 0.45470279455184937,
+      "learning_rate": 3.5166944456088445e-05,
+      "loss": 0.0413,
+      "mean_token_accuracy": 0.9881617486476898,
+      "num_tokens": 1744225.0,
+      "step": 1290
+    },
+    {
+      "entropy": 0.8643660068511962,
+      "epoch": 1.24282982791587,
+      "grad_norm": 0.584843635559082,
+      "learning_rate": 3.441388092854532e-05,
+      "loss": 0.0356,
+      "mean_token_accuracy": 0.9891556799411774,
+      "num_tokens": 1757359.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.8713556408882142,
+      "epoch": 1.2523900573613767,
+      "grad_norm": 0.4194789230823517,
+      "learning_rate": 3.3664713526646976e-05,
+      "loss": 0.0413,
+      "mean_token_accuracy": 0.9841473639011383,
+      "num_tokens": 1770617.0,
+      "step": 1310
+    },
+    {
+      "entropy": 0.8880608439445495,
+      "epoch": 1.2619502868068833,
+      "grad_norm": 0.46696576476097107,
+      "learning_rate": 3.291962952281918e-05,
+      "loss": 0.0421,
+      "mean_token_accuracy": 0.9864380300045014,
+      "num_tokens": 1783933.0,
+      "step": 1320
+    },
+    {
+      "entropy": 0.8660070896148682,
+      "epoch": 1.27151051625239,
+      "grad_norm": 0.7755872011184692,
+      "learning_rate": 3.217881516874407e-05,
+      "loss": 0.0511,
+      "mean_token_accuracy": 0.9825553178787232,
+      "num_tokens": 1797117.0,
+      "step": 1330
+    },
+    {
+      "entropy": 0.887319028377533,
+      "epoch": 1.2810707456978967,
+      "grad_norm": 0.4816440939903259,
+      "learning_rate": 3.1442455648802e-05,
+      "loss": 0.0534,
+      "mean_token_accuracy": 0.9822551965713501,
+      "num_tokens": 1810602.0,
+      "step": 1340
+    },
+    {
+      "entropy": 0.8807766735553741,
+      "epoch": 1.2906309751434035,
+      "grad_norm": 0.8487844467163086,
+      "learning_rate": 3.071073503378039e-05,
+      "loss": 0.0526,
+      "mean_token_accuracy": 0.9800789594650269,
+      "num_tokens": 1823981.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.8801860511302948,
+      "epoch": 1.3001912045889101,
+      "grad_norm": 0.49022266268730164,
+      "learning_rate": 2.9983836234860618e-05,
+      "loss": 0.0517,
+      "mean_token_accuracy": 0.9837223827838898,
+      "num_tokens": 1836963.0,
+      "step": 1360
+    },
+    {
+      "entropy": 0.8826759576797485,
+      "epoch": 1.3097514340344167,
+      "grad_norm": 0.4688926935195923,
+      "learning_rate": 2.9261940957895108e-05,
+      "loss": 0.0562,
+      "mean_token_accuracy": 0.9815012872219085,
+      "num_tokens": 1850336.0,
+      "step": 1370
+    },
+    {
+      "entropy": 0.8689303994178772,
+      "epoch": 1.3193116634799236,
+      "grad_norm": 0.4720284938812256,
+      "learning_rate": 2.8545229657985495e-05,
+      "loss": 0.0456,
+      "mean_token_accuracy": 0.9854044258594513,
+      "num_tokens": 1863231.0,
+      "step": 1380
+    },
+    {
+      "entropy": 0.8769595146179199,
+      "epoch": 1.3288718929254302,
+      "grad_norm": 0.40125951170921326,
+      "learning_rate": 2.783388149437359e-05,
+      "loss": 0.0491,
+      "mean_token_accuracy": 0.982703047990799,
+      "num_tokens": 1876930.0,
+      "step": 1390
+    },
+    {
+      "entropy": 0.8676677346229553,
+      "epoch": 1.338432122370937,
+      "grad_norm": 0.6897684335708618,
+      "learning_rate": 2.71280742856562e-05,
+      "loss": 0.0379,
+      "mean_token_accuracy": 0.9887689232826233,
+      "num_tokens": 1890621.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.8628024458885193,
+      "epoch": 1.3479923518164436,
+      "grad_norm": 0.5727158188819885,
+      "learning_rate": 2.642798446533502e-05,
+      "loss": 0.0411,
+      "mean_token_accuracy": 0.9862188160419464,
+      "num_tokens": 1903678.0,
+      "step": 1410
+    },
+    {
+      "entropy": 0.8686028003692627,
+      "epoch": 1.3575525812619502,
+      "grad_norm": 0.4560388922691345,
+      "learning_rate": 2.573378703771293e-05,
+      "loss": 0.0475,
+      "mean_token_accuracy": 0.9837337911128998,
+      "num_tokens": 1917079.0,
+      "step": 1420
+    },
+    {
+      "entropy": 0.8880694031715393,
+      "epoch": 1.367112810707457,
+      "grad_norm": 0.5331932902336121,
+      "learning_rate": 2.504565553414745e-05,
+      "loss": 0.0454,
+      "mean_token_accuracy": 0.9831508517265319,
+      "num_tokens": 1931223.0,
+      "step": 1430
+    },
+    {
+      "entropy": 0.8839497327804565,
+      "epoch": 1.3766730401529637,
+      "grad_norm": 0.9909533858299255,
+      "learning_rate": 2.4363761969672278e-05,
+      "loss": 0.0449,
+      "mean_token_accuracy": 0.9824761688709259,
+      "num_tokens": 1944423.0,
+      "step": 1440
+    },
+    {
+      "entropy": 0.8770220041275024,
+      "epoch": 1.3862332695984705,
+      "grad_norm": 0.6676177382469177,
+      "learning_rate": 2.368827679999814e-05,
+      "loss": 0.0454,
+      "mean_token_accuracy": 0.9868211030960083,
+      "num_tokens": 1957626.0,
+      "step": 1450
+    },
+    {
+      "entropy": 0.8919743180274964,
+      "epoch": 1.395793499043977,
+      "grad_norm": 0.4520642161369324,
+      "learning_rate": 2.3019368878903213e-05,
+      "loss": 0.0536,
+      "mean_token_accuracy": 0.982214230298996,
+      "num_tokens": 1970965.0,
+      "step": 1460
+    },
+    {
+      "entropy": 0.8842333137989045,
+      "epoch": 1.4053537284894837,
+      "grad_norm": 0.4748406410217285,
+      "learning_rate": 2.2357205416024075e-05,
+      "loss": 0.0373,
+      "mean_token_accuracy": 0.9879635214805603,
+      "num_tokens": 1984874.0,
+      "step": 1470
+    },
+    {
+      "entropy": 0.8787308692932129,
+      "epoch": 1.4149139579349903,
+      "grad_norm": 0.4675792157649994,
+      "learning_rate": 2.1701951935057573e-05,
+      "loss": 0.0425,
+      "mean_token_accuracy": 0.9846374094486237,
+      "num_tokens": 1999952.0,
+      "step": 1480
+    },
+    {
+      "entropy": 0.8788155794143677,
+      "epoch": 1.4244741873804971,
+      "grad_norm": 0.3942820429801941,
+      "learning_rate": 2.10537722323843e-05,
+      "loss": 0.0503,
+      "mean_token_accuracy": 0.982246744632721,
+      "num_tokens": 2012830.0,
+      "step": 1490
+    },
+    {
+      "entropy": 0.895525187253952,
+      "epoch": 1.4340344168260037,
+      "grad_norm": 0.589984655380249,
+      "learning_rate": 2.0412828336123658e-05,
+      "loss": 0.0457,
+      "mean_token_accuracy": 0.9849776268005371,
+      "num_tokens": 2026296.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.873170918226242,
+      "epoch": 1.4435946462715106,
+      "grad_norm": 0.6245923638343811,
+      "learning_rate": 1.9779280465631068e-05,
+      "loss": 0.0438,
+      "mean_token_accuracy": 0.9830984771251678,
+      "num_tokens": 2039361.0,
+      "step": 1510
+    },
+    {
+      "entropy": 0.8865708649158478,
+      "epoch": 1.4531548757170172,
+      "grad_norm": 0.7497299313545227,
+      "learning_rate": 1.9153286991447267e-05,
+      "loss": 0.0437,
+      "mean_token_accuracy": 0.9837348520755768,
+      "num_tokens": 2052906.0,
+      "step": 1520
+    },
+    {
+      "entropy": 0.8784422934055328,
+      "epoch": 1.4627151051625238,
+      "grad_norm": 0.5221916437149048,
+      "learning_rate": 1.8535004395709797e-05,
+      "loss": 0.0434,
+      "mean_token_accuracy": 0.9837494909763336,
+      "num_tokens": 2066064.0,
+      "step": 1530
+    },
+    {
+      "entropy": 0.8811910510063171,
+      "epoch": 1.4722753346080306,
+      "grad_norm": 0.6684751510620117,
+      "learning_rate": 1.7924587233036537e-05,
+      "loss": 0.049,
+      "mean_token_accuracy": 0.9824684739112854,
+      "num_tokens": 2079485.0,
+      "step": 1540
+    },
+    {
+      "entropy": 0.8785001158714294,
+      "epoch": 1.4818355640535372,
+      "grad_norm": 0.6501848697662354,
+      "learning_rate": 1.7322188091890985e-05,
+      "loss": 0.0397,
+      "mean_token_accuracy": 0.9852206408977509,
+      "num_tokens": 2092884.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.8725269973278046,
+      "epoch": 1.491395793499044,
+      "grad_norm": 0.6203243136405945,
+      "learning_rate": 1.6727957556439188e-05,
+      "loss": 0.0397,
+      "mean_token_accuracy": 0.9866523683071137,
+      "num_tokens": 2106152.0,
+      "step": 1560
+    },
+    {
+      "entropy": 0.8831465482711792,
+      "epoch": 1.5009560229445507,
+      "grad_norm": 0.5062828660011292,
+      "learning_rate": 1.6142044168907595e-05,
+      "loss": 0.0477,
+      "mean_token_accuracy": 0.98556467294693,
+      "num_tokens": 2120047.0,
+      "step": 1570
+    },
+    {
+      "entropy": 0.8798212945461273,
+      "epoch": 1.5105162523900573,
+      "grad_norm": 0.770358681678772,
+      "learning_rate": 1.556459439245125e-05,
+      "loss": 0.0395,
+      "mean_token_accuracy": 0.9853990495204925,
+      "num_tokens": 2133798.0,
+      "step": 1580
+    },
+    {
+      "entropy": 0.8671085238456726,
+      "epoch": 1.520076481835564,
+      "grad_norm": 0.40903016924858093,
+      "learning_rate": 1.4995752574541921e-05,
+      "loss": 0.0389,
+      "mean_token_accuracy": 0.9867224812507629,
+      "num_tokens": 2147167.0,
+      "step": 1590
+    },
+    {
+      "entropy": 0.8701870858669281,
+      "epoch": 1.5296367112810707,
+      "grad_norm": 0.8610926866531372,
+      "learning_rate": 1.443566091088499e-05,
+      "loss": 0.0419,
+      "mean_token_accuracy": 0.9837621212005615,
+      "num_tokens": 2160770.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.8721591353416442,
+      "epoch": 1.5391969407265775,
+      "grad_norm": 0.6610036492347717,
+      "learning_rate": 1.3884459409874034e-05,
+      "loss": 0.0424,
+      "mean_token_accuracy": 0.9849321126937867,
+      "num_tokens": 2174474.0,
+      "step": 1610
+    },
+    {
+      "entropy": 0.8703516840934753,
+      "epoch": 1.5487571701720841,
+      "grad_norm": 0.6359735727310181,
+      "learning_rate": 1.3342285857592551e-05,
+      "loss": 0.0453,
+      "mean_token_accuracy": 0.9846934080123901,
+      "num_tokens": 2187663.0,
+      "step": 1620
+    },
+    {
+      "entropy": 0.8872678935527801,
+      "epoch": 1.5583173996175907,
+      "grad_norm": 0.4957731068134308,
+      "learning_rate": 1.2809275783370923e-05,
+      "loss": 0.0357,
+      "mean_token_accuracy": 0.9884172320365906,
+      "num_tokens": 2201519.0,
+      "step": 1630
+    },
+    {
+      "entropy": 0.8795758306980133,
+      "epoch": 1.5678776290630974,
+      "grad_norm": 0.538317859172821,
+      "learning_rate": 1.2285562425907621e-05,
+      "loss": 0.0388,
+      "mean_token_accuracy": 0.9876142144203186,
+      "num_tokens": 2214755.0,
+      "step": 1640
+    },
+    {
+      "entropy": 0.8640561103820801,
+      "epoch": 1.5774378585086042,
+      "grad_norm": 0.3958626091480255,
+      "learning_rate": 1.1771276699962935e-05,
+      "loss": 0.04,
+      "mean_token_accuracy": 0.9861380994319916,
+      "num_tokens": 2228056.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.8622815489768982,
+      "epoch": 1.586998087954111,
+      "grad_norm": 0.5302303433418274,
+      "learning_rate": 1.1266547163633767e-05,
+      "loss": 0.0384,
+      "mean_token_accuracy": 0.985512661933899,
+      "num_tokens": 2241581.0,
+      "step": 1660
+    },
+    {
+      "entropy": 0.8767390131950379,
+      "epoch": 1.5965583173996176,
+      "grad_norm": 0.582912802696228,
+      "learning_rate": 1.0771499986217365e-05,
+      "loss": 0.0497,
+      "mean_token_accuracy": 0.9832952320575714,
+      "num_tokens": 2255654.0,
+      "step": 1670
+    },
+    {
+      "entropy": 0.8709462285041809,
+      "epoch": 1.6061185468451242,
+      "grad_norm": 0.45127853751182556,
+      "learning_rate": 1.0286258916672303e-05,
+      "loss": 0.04,
+      "mean_token_accuracy": 0.9856235384941101,
+      "num_tokens": 2269408.0,
+      "step": 1680
+    },
+    {
+      "entropy": 0.8770772516727448,
+      "epoch": 1.6156787762906308,
+      "grad_norm": 0.4286617636680603,
+      "learning_rate": 9.81094525268445e-06,
+      "loss": 0.0408,
+      "mean_token_accuracy": 0.9852644681930542,
+      "num_tokens": 2282973.0,
+      "step": 1690
+    },
+    {
+      "entropy": 0.8733699083328247,
+      "epoch": 1.6252390057361377,
+      "grad_norm": 0.6805005073547363,
+      "learning_rate": 9.345677810345677e-06,
+      "loss": 0.051,
+      "mean_token_accuracy": 0.9839253544807434,
+      "num_tokens": 2296428.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.8733612477779389,
+      "epoch": 1.6347992351816445,
+      "grad_norm": 0.6421047449111938,
+      "learning_rate": 8.890572894452948e-06,
+      "loss": 0.0391,
+      "mean_token_accuracy": 0.9870924890041352,
+      "num_tokens": 2310330.0,
+      "step": 1710
+    },
+    {
+      "entropy": 0.8611141502857208,
+      "epoch": 1.644359464627151,
+      "grad_norm": 0.7705059051513672,
+      "learning_rate": 8.445744269435063e-06,
+      "loss": 0.0405,
+      "mean_token_accuracy": 0.9865531504154206,
+      "num_tokens": 2324210.0,
+      "step": 1720
+    },
+    {
+      "entropy": 0.8895285785198211,
+      "epoch": 1.6539196940726577,
+      "grad_norm": 0.5833611488342285,
+      "learning_rate": 8.011303130914599e-06,
+      "loss": 0.0409,
+      "mean_token_accuracy": 0.9853172779083252,
+      "num_tokens": 2337289.0,
+      "step": 1730
+    },
+    {
+      "entropy": 0.8541743695735932,
+      "epoch": 1.6634799235181643,
+      "grad_norm": 0.5119567513465881,
+      "learning_rate": 7.5873580779118266e-06,
+      "loss": 0.0367,
+      "mean_token_accuracy": 0.9888664662837983,
+      "num_tokens": 2350620.0,
+      "step": 1740
+    },
+    {
+      "entropy": 0.8616726338863373,
+      "epoch": 1.6730401529636711,
+      "grad_norm": 0.4557342529296875,
+      "learning_rate": 7.174015085697788e-06,
+      "loss": 0.0435,
+      "mean_token_accuracy": 0.9880729198455811,
+      "num_tokens": 2364035.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.8689415454864502,
+      "epoch": 1.682600382409178,
+      "grad_norm": 0.8540235757827759,
+      "learning_rate": 6.771377479303187e-06,
+      "loss": 0.0353,
+      "mean_token_accuracy": 0.9888749480247497,
+      "num_tokens": 2377105.0,
+      "step": 1760
+    },
+    {
+      "entropy": 0.856802123785019,
+      "epoch": 1.6921606118546846,
+      "grad_norm": 0.31953999400138855,
+      "learning_rate": 6.379545907689904e-06,
+      "loss": 0.042,
+      "mean_token_accuracy": 0.9855331659317017,
+      "num_tokens": 2390871.0,
+      "step": 1770
+    },
+    {
+      "entropy": 0.8603232741355896,
+      "epoch": 1.7017208413001912,
+      "grad_norm": 0.41709044575691223,
+      "learning_rate": 5.998618318591282e-06,
+      "loss": 0.036,
+      "mean_token_accuracy": 0.9869058310985566,
+      "num_tokens": 2404613.0,
+      "step": 1780
+    },
+    {
+      "entropy": 0.8728388547897339,
+      "epoch": 1.7112810707456978,
+      "grad_norm": 0.46097755432128906,
+      "learning_rate": 5.628689934027725e-06,
+      "loss": 0.0388,
+      "mean_token_accuracy": 0.9884433805942535,
+      "num_tokens": 2418173.0,
+      "step": 1790
+    },
+    {
+      "entropy": 0.8615271091461182,
+      "epoch": 1.7208413001912046,
+      "grad_norm": 0.6570433974266052,
+      "learning_rate": 5.269853226503746e-06,
+      "loss": 0.0363,
+      "mean_token_accuracy": 0.9867855429649353,
+      "num_tokens": 2431691.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.8882860004901886,
+      "epoch": 1.7304015296367112,
+      "grad_norm": 0.695399820804596,
+      "learning_rate": 4.922197895892167e-06,
+      "loss": 0.0437,
+      "mean_token_accuracy": 0.98538898229599,
+      "num_tokens": 2446060.0,
+      "step": 1810
+    },
+    {
+      "entropy": 0.865412586927414,
+      "epoch": 1.739961759082218,
+      "grad_norm": 0.5521643757820129,
+      "learning_rate": 4.585810847011473e-06,
+      "loss": 0.0443,
+      "mean_token_accuracy": 0.9864999890327454,
+      "num_tokens": 2459918.0,
+      "step": 1820
+    },
+    {
+      "entropy": 0.8531206369400024,
+      "epoch": 1.7495219885277247,
+      "grad_norm": 1.8240056037902832,
+      "learning_rate": 4.260776167901931e-06,
+      "loss": 0.0524,
+      "mean_token_accuracy": 0.9827051401138306,
+      "num_tokens": 2473188.0,
+      "step": 1830
+    },
+    {
+      "entropy": 0.8673100113868714,
+      "epoch": 1.7590822179732313,
+      "grad_norm": 0.5542291402816772,
+      "learning_rate": 3.947175108805678e-06,
+      "loss": 0.0396,
+      "mean_token_accuracy": 0.9871743202209473,
+      "num_tokens": 2486973.0,
+      "step": 1840
+    },
+    {
+      "entropy": 0.8624794006347656,
+      "epoch": 1.7686424474187379,
+      "grad_norm": 0.5326445698738098,
+      "learning_rate": 3.6450860618563175e-06,
+      "loss": 0.0403,
+      "mean_token_accuracy": 0.98631272315979,
+      "num_tokens": 2500712.0,
+      "step": 1850
+    },
+    {
+      "entropy": 0.8684999585151673,
+      "epoch": 1.7782026768642447,
+      "grad_norm": 1.1630841493606567,
+      "learning_rate": 3.3545845414829026e-06,
+      "loss": 0.0459,
+      "mean_token_accuracy": 0.9832627832889557,
+      "num_tokens": 2514191.0,
+      "step": 1860
+    },
+    {
+      "entropy": 0.8703453242778778,
+      "epoch": 1.7877629063097515,
+      "grad_norm": 0.5935927033424377,
+      "learning_rate": 3.07574316553329e-06,
+      "loss": 0.0504,
+      "mean_token_accuracy": 0.9823457598686218,
+      "num_tokens": 2527328.0,
+      "step": 1870
+    },
+    {
+      "entropy": 0.8815466165542603,
+      "epoch": 1.7973231357552581,
+      "grad_norm": 0.3089905381202698,
+      "learning_rate": 2.808631637121567e-06,
+      "loss": 0.038,
+      "mean_token_accuracy": 0.9850891888141632,
+      "num_tokens": 2541361.0,
+      "step": 1880
+    },
+    {
+      "entropy": 0.8639434933662414,
+      "epoch": 1.8068833652007648,
+      "grad_norm": 0.37706977128982544,
+      "learning_rate": 2.553316727204047e-06,
+      "loss": 0.047,
+      "mean_token_accuracy": 0.9841739475727082,
+      "num_tokens": 2554550.0,
+      "step": 1890
+    },
+    {
+      "entropy": 0.8655487835407257,
+      "epoch": 1.8164435946462714,
+      "grad_norm": 0.3325982689857483,
+      "learning_rate": 2.309862257888301e-06,
+      "loss": 0.0375,
+      "mean_token_accuracy": 0.9871336400508881,
+      "num_tokens": 2568194.0,
+      "step": 1900
+    },
+    {
+      "entropy": 0.8695250272750854,
+      "epoch": 1.8260038240917782,
+      "grad_norm": 0.37546759843826294,
+      "learning_rate": 2.078329086479286e-06,
+      "loss": 0.0355,
+      "mean_token_accuracy": 0.9893505871295929,
+      "num_tokens": 2581876.0,
+      "step": 1910
+    },
+    {
+      "entropy": 0.874930614233017,
+      "epoch": 1.835564053537285,
+      "grad_norm": 0.6583105325698853,
+      "learning_rate": 1.8587750902665735e-06,
+      "loss": 0.04,
+      "mean_token_accuracy": 0.9870905160903931,
+      "num_tokens": 2595560.0,
+      "step": 1920
+    },
+    {
+      "entropy": 0.8541331112384796,
+      "epoch": 1.8451242829827916,
+      "grad_norm": 0.5673859119415283,
+      "learning_rate": 1.6512551520565133e-06,
+      "loss": 0.0375,
+      "mean_token_accuracy": 0.9880424618721009,
+      "num_tokens": 2609321.0,
+      "step": 1930
+    },
+    {
+      "entropy": 0.8642001569271087,
+      "epoch": 1.8546845124282982,
+      "grad_norm": 0.7911686301231384,
+      "learning_rate": 1.455821146453018e-06,
+      "loss": 0.0395,
+      "mean_token_accuracy": 0.9844050526618957,
+      "num_tokens": 2622763.0,
+      "step": 1940
+    },
+    {
+      "entropy": 0.8747141182422637,
+      "epoch": 1.8642447418738048,
+      "grad_norm": 0.49693065881729126,
+      "learning_rate": 1.2725219268901967e-06,
+      "loss": 0.0378,
+      "mean_token_accuracy": 0.9885695040225982,
+      "num_tokens": 2636291.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.8498066246509552,
+      "epoch": 1.8738049713193117,
+      "grad_norm": 0.6422383785247803,
+      "learning_rate": 1.1014033134202683e-06,
+      "loss": 0.0475,
+      "mean_token_accuracy": 0.9821363687515259,
+      "num_tokens": 2649475.0,
+      "step": 1960
+    },
+    {
+      "entropy": 0.8639785945415497,
+      "epoch": 1.8833652007648185,
+      "grad_norm": 0.819911777973175,
+      "learning_rate": 9.425080812597909e-07,
+      "loss": 0.0377,
+      "mean_token_accuracy": 0.9881011486053467,
+      "num_tokens": 2662966.0,
+      "step": 1970
+    },
+    {
+      "entropy": 0.8643074989318847,
+      "epoch": 1.892925430210325,
+      "grad_norm": 0.6799226999282837,
+      "learning_rate": 7.958759500968915e-07,
+      "loss": 0.0441,
+      "mean_token_accuracy": 0.9835970878601075,
+      "num_tokens": 2675592.0,
+      "step": 1980
+    },
+    {
+      "entropy": 0.8756822049617767,
+      "epoch": 1.9024856596558317,
+      "grad_norm": 0.42630982398986816,
+      "learning_rate": 6.61543574162432e-07,
+      "loss": 0.0441,
+      "mean_token_accuracy": 0.9824448943138122,
+      "num_tokens": 2689270.0,
+      "step": 1990
+    },
+    {
+      "entropy": 0.8642894685268402,
+      "epoch": 1.9120458891013383,
+      "grad_norm": 0.526240885257721,
+      "learning_rate": 5.395445330673654e-07,
+      "loss": 0.0399,
+      "mean_token_accuracy": 0.9871181011199951,
+      "num_tokens": 2702690.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.8578903079032898,
+      "epoch": 1.9216061185468452,
+      "grad_norm": 0.643413245677948,
+      "learning_rate": 4.299093234087281e-07,
+      "loss": 0.044,
+      "mean_token_accuracy": 0.9849568903446198,
+      "num_tokens": 2715562.0,
+      "step": 2010
+    },
+    {
+      "entropy": 0.8757904589176178,
+      "epoch": 1.9311663479923518,
+      "grad_norm": 0.6925222873687744,
+      "learning_rate": 3.3266535114628296e-07,
+      "loss": 0.0453,
+      "mean_token_accuracy": 0.9830472648143769,
+      "num_tokens": 2729105.0,
+      "step": 2020
+    },
+    {
+      "entropy": 0.8626958847045898,
+      "epoch": 1.9407265774378586,
+      "grad_norm": 0.4150663912296295,
+      "learning_rate": 2.4783692475169874e-07,
+      "loss": 0.0443,
+      "mean_token_accuracy": 0.9845708250999451,
+      "num_tokens": 2742679.0,
+      "step": 2030
+    },
+    {
+      "entropy": 0.869753223657608,
+      "epoch": 1.9502868068833652,
+      "grad_norm": 0.6163395047187805,
+      "learning_rate": 1.7544524913211125e-07,
+      "loss": 0.0352,
+      "mean_token_accuracy": 0.9873261511325836,
+      "num_tokens": 2756899.0,
+      "step": 2040
+    },
+    {
+      "entropy": 0.8837343454360962,
+      "epoch": 1.9598470363288718,
+      "grad_norm": 0.596882164478302,
+      "learning_rate": 1.1550842032943543e-07,
+      "loss": 0.0656,
+      "mean_token_accuracy": 0.9783609986305237,
+      "num_tokens": 2771213.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.8640608549118042,
+      "epoch": 1.9694072657743786,
+      "grad_norm": 0.6977580785751343,
+      "learning_rate": 6.8041420996795e-08,
+      "loss": 0.0375,
+      "mean_token_accuracy": 0.9890853345394135,
+      "num_tokens": 2784886.0,
+      "step": 2060
+    },
+    {
+      "entropy": 0.8707361102104187,
+      "epoch": 1.9789674952198852,
+      "grad_norm": 0.4639173448085785,
+      "learning_rate": 3.3056116653285005e-08,
+      "loss": 0.0306,
+      "mean_token_accuracy": 0.9919256448745728,
+      "num_tokens": 2798197.0,
+      "step": 2070
+    },
+    {
+      "entropy": 0.8770419299602509,
+      "epoch": 1.988527724665392,
+      "grad_norm": 0.7005882859230042,
+      "learning_rate": 1.056125271788333e-08,
+      "loss": 0.042,
+      "mean_token_accuracy": 0.9852849662303924,
+      "num_tokens": 2811807.0,
+      "step": 2080
+    },
+    {
+      "entropy": 0.8658124804496765,
+      "epoch": 1.9980879541108987,
+      "grad_norm": 0.7639582753181458,
+      "learning_rate": 5.62452323304985e-10,
+      "loss": 0.0492,
+      "mean_token_accuracy": 0.9832391321659089,
+      "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_plain_priorrecipe_e2_lr1e-4_s3/checkpoint-2092/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d06d12324071801b14ec40ba4f282f75f7986e48dfdafc8dc25fe79865f96fd9
+size 6417

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s3/manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r64_a32_plain_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_plain_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_plain_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_plain_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_plain_priorrecipe_e2_lr1e-4_s3/train_manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r64_a32_plain_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_plain_priorrecipe_e2_lr1e-4_s3/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d06d12324071801b14ec40ba4f282f75f7986e48dfdafc8dc25fe79865f96fd9
+size 6417