add checkpoint cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1

.gitattributes

@@ -781,3 +781,6 @@ checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot5_e2_lr1e-4_s1/tokenizer.json filter=l
 checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot1_e2_lr1e-4_s1/checkpoint-1046/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot1_e2_lr1e-4_s1/checkpoint-2092/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot1_e2_lr1e-4_s1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-1046/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-2092/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/README.md

@@ -0,0 +1,61 @@
+---
+base_model: Qwen/Qwen2.5-7B-Instruct
+library_name: peft
+model_name: cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1
+tags:
+- base_model:adapter:Qwen/Qwen2.5-7B-Instruct
+- lora
+- sft
+- transformers
+- trl
+licence: license
+pipeline_tag: text-generation
+---
+
+# Model Card for cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/agam-research/huggingface/runs/xghb9wme) 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- PEFT 0.18.1
+- TRL: 0.28.0
+- Transformers: 4.57.6
+- Pytorch: 2.10.0
+- Datasets: 4.5.0
+- Tokenizers: 0.22.2
+
+## Citations
+
+
+
+Cite TRL as:
+    
+```bibtex
+@software{vonwerra2020trl,
+  title   = {{TRL: Transformers Reinforcement Learning}},
+  author  = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
+  license = {Apache-2.0},
+  url     = {https://github.com/huggingface/trl},
+  year    = {2020}
+}
+```

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "o_proj",
+    "up_proj",
+    "v_proj",
+    "k_proj",
+    "gate_proj",
+    "down_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b6b0a9807adc5da265640817da651e603f2ce975b26af6b3dacf878eb5122ac
+size 80792096

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/added_tokens.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/chat_template.jinja

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-1046/README.md

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-1046/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "o_proj",
+    "up_proj",
+    "v_proj",
+    "k_proj",
+    "gate_proj",
+    "down_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-1046/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c943b0872109f6f51b39b5f643074bd2fcf0d2311c50762467c505ac32e1866
+size 80792096

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-1046/added_tokens.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-1046/chat_template.jinja

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-1046/special_tokens_map.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-1046/tokenizer.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-1046/tokenizer_config.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-1046/trainer_state.json

@@ -0,0 +1,1074 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 1046,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.1737810969352722,
+      "epoch": 0.009560229445506692,
+      "grad_norm": 1.871079921722412,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2769,
+      "mean_token_accuracy": 0.9123528063297272,
+      "num_tokens": 14125.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1810022592544556,
+      "epoch": 0.019120458891013385,
+      "grad_norm": 1.8740425109863281,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2386,
+      "mean_token_accuracy": 0.9218324422836304,
+      "num_tokens": 27004.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1399657130241394,
+      "epoch": 0.028680688336520075,
+      "grad_norm": 2.0634806156158447,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.1882,
+      "mean_token_accuracy": 0.9362742602825165,
+      "num_tokens": 41395.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.0615105748176574,
+      "epoch": 0.03824091778202677,
+      "grad_norm": 2.3918206691741943,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1631,
+      "mean_token_accuracy": 0.9405273735523224,
+      "num_tokens": 55233.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.0151220917701722,
+      "epoch": 0.04780114722753346,
+      "grad_norm": 3.8594629764556885,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.1767,
+      "mean_token_accuracy": 0.9330196499824523,
+      "num_tokens": 68341.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.014986515045166,
+      "epoch": 0.05736137667304015,
+      "grad_norm": 2.2135467529296875,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1738,
+      "mean_token_accuracy": 0.9349386811256408,
+      "num_tokens": 81730.0,
+      "step": 60
+    },
+    {
+      "entropy": 0.9990480720996857,
+      "epoch": 0.06692160611854685,
+      "grad_norm": 2.428823709487915,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.162,
+      "mean_token_accuracy": 0.9420714497566223,
+      "num_tokens": 94886.0,
+      "step": 70
+    },
+    {
+      "entropy": 0.9700570225715637,
+      "epoch": 0.07648183556405354,
+      "grad_norm": 2.4947664737701416,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1805,
+      "mean_token_accuracy": 0.9348644435405731,
+      "num_tokens": 108569.0,
+      "step": 80
+    },
+    {
+      "entropy": 0.9715868353843689,
+      "epoch": 0.08604206500956023,
+      "grad_norm": 1.909283995628357,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1595,
+      "mean_token_accuracy": 0.941025310754776,
+      "num_tokens": 122156.0,
+      "step": 90
+    },
+    {
+      "entropy": 0.9832429409027099,
+      "epoch": 0.09560229445506692,
+      "grad_norm": 1.9785945415496826,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1782,
+      "mean_token_accuracy": 0.9320050239562988,
+      "num_tokens": 135290.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0139548778533936,
+      "epoch": 0.10516252390057361,
+      "grad_norm": 2.554082155227661,
+      "learning_rate": 9.999900008621665e-05,
+      "loss": 0.205,
+      "mean_token_accuracy": 0.9271113276481628,
+      "num_tokens": 149015.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.00105682015419,
+      "epoch": 0.1147227533460803,
+      "grad_norm": 1.7623502016067505,
+      "learning_rate": 9.998775151544478e-05,
+      "loss": 0.1824,
+      "mean_token_accuracy": 0.93257275223732,
+      "num_tokens": 162147.0,
+      "step": 120
+    },
+    {
+      "entropy": 0.9644641637802124,
+      "epoch": 0.124282982791587,
+      "grad_norm": 1.752968192100525,
+      "learning_rate": 9.996400730289628e-05,
+      "loss": 0.1847,
+      "mean_token_accuracy": 0.9332687795162201,
+      "num_tokens": 175436.0,
+      "step": 130
+    },
+    {
+      "entropy": 0.971268481016159,
+      "epoch": 0.1338432122370937,
+      "grad_norm": 2.4105324745178223,
+      "learning_rate": 9.992777338400864e-05,
+      "loss": 0.1777,
+      "mean_token_accuracy": 0.9332670569419861,
+      "num_tokens": 188834.0,
+      "step": 140
+    },
+    {
+      "entropy": 0.9512466073036194,
+      "epoch": 0.14340344168260039,
+      "grad_norm": 2.102140188217163,
+      "learning_rate": 9.987905881632213e-05,
+      "loss": 0.1821,
+      "mean_token_accuracy": 0.934379369020462,
+      "num_tokens": 202708.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.9578613698482513,
+      "epoch": 0.15296367112810708,
+      "grad_norm": 1.9931540489196777,
+      "learning_rate": 9.981787577721553e-05,
+      "loss": 0.1768,
+      "mean_token_accuracy": 0.934454756975174,
+      "num_tokens": 216384.0,
+      "step": 160
+    },
+    {
+      "entropy": 1.021405816078186,
+      "epoch": 0.16252390057361377,
+      "grad_norm": 1.9813957214355469,
+      "learning_rate": 9.974423956086228e-05,
+      "loss": 0.1901,
+      "mean_token_accuracy": 0.9332379758358001,
+      "num_tokens": 229105.0,
+      "step": 170
+    },
+    {
+      "entropy": 1.0439977526664734,
+      "epoch": 0.17208413001912046,
+      "grad_norm": 2.317162275314331,
+      "learning_rate": 9.965816857440717e-05,
+      "loss": 0.1632,
+      "mean_token_accuracy": 0.9360143065452575,
+      "num_tokens": 242264.0,
+      "step": 180
+    },
+    {
+      "entropy": 1.0610752880573273,
+      "epoch": 0.18164435946462715,
+      "grad_norm": 2.324665069580078,
+      "learning_rate": 9.955968433336515e-05,
+      "loss": 0.1971,
+      "mean_token_accuracy": 0.922540009021759,
+      "num_tokens": 255602.0,
+      "step": 190
+    },
+    {
+      "entropy": 1.0195186913013459,
+      "epoch": 0.19120458891013384,
+      "grad_norm": 1.5599435567855835,
+      "learning_rate": 9.944881145624291e-05,
+      "loss": 0.1389,
+      "mean_token_accuracy": 0.948070353269577,
+      "num_tokens": 269712.0,
+      "step": 200
+    },
+    {
+      "entropy": 1.0296171724796295,
+      "epoch": 0.20076481835564053,
+      "grad_norm": 2.3139309883117676,
+      "learning_rate": 9.932557765838504e-05,
+      "loss": 0.179,
+      "mean_token_accuracy": 0.9331721246242524,
+      "num_tokens": 283572.0,
+      "step": 210
+    },
+    {
+      "entropy": 1.0230853736400605,
+      "epoch": 0.21032504780114722,
+      "grad_norm": 1.953762412071228,
+      "learning_rate": 9.919001374504569e-05,
+      "loss": 0.1793,
+      "mean_token_accuracy": 0.9311605632305145,
+      "num_tokens": 297437.0,
+      "step": 220
+    },
+    {
+      "entropy": 1.0038200318813324,
+      "epoch": 0.2198852772466539,
+      "grad_norm": 4.689145565032959,
+      "learning_rate": 9.904215360368826e-05,
+      "loss": 0.1809,
+      "mean_token_accuracy": 0.9343967974185944,
+      "num_tokens": 310803.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.0401727676391601,
+      "epoch": 0.2294455066921606,
+      "grad_norm": 7.431188583374023,
+      "learning_rate": 9.888203419551432e-05,
+      "loss": 0.2055,
+      "mean_token_accuracy": 0.9227587044239044,
+      "num_tokens": 324713.0,
+      "step": 240
+    },
+    {
+      "entropy": 1.028674304485321,
+      "epoch": 0.2390057361376673,
+      "grad_norm": 3.3888537883758545,
+      "learning_rate": 9.870969554622426e-05,
+      "loss": 0.2101,
+      "mean_token_accuracy": 0.920482850074768,
+      "num_tokens": 337634.0,
+      "step": 250
+    },
+    {
+      "entropy": 1.0425225734710692,
+      "epoch": 0.248565965583174,
+      "grad_norm": 2.374208450317383,
+      "learning_rate": 9.852518073601194e-05,
+      "loss": 0.1805,
+      "mean_token_accuracy": 0.9340396881103515,
+      "num_tokens": 350459.0,
+      "step": 260
+    },
+    {
+      "entropy": 1.0127095937728883,
+      "epoch": 0.25812619502868067,
+      "grad_norm": 2.729233741760254,
+      "learning_rate": 9.832853588879575e-05,
+      "loss": 0.1736,
+      "mean_token_accuracy": 0.9342186093330384,
+      "num_tokens": 364010.0,
+      "step": 270
+    },
+    {
+      "entropy": 1.039266401529312,
+      "epoch": 0.2676864244741874,
+      "grad_norm": 1.921285629272461,
+      "learning_rate": 9.811981016068881e-05,
+      "loss": 0.2124,
+      "mean_token_accuracy": 0.9189038336277008,
+      "num_tokens": 377021.0,
+      "step": 280
+    },
+    {
+      "entropy": 1.0290949404239655,
+      "epoch": 0.27724665391969405,
+      "grad_norm": 1.8515435457229614,
+      "learning_rate": 9.78990557277112e-05,
+      "loss": 0.2009,
+      "mean_token_accuracy": 0.9233260750770569,
+      "num_tokens": 390819.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0516928434371948,
+      "epoch": 0.28680688336520077,
+      "grad_norm": 2.3137047290802,
+      "learning_rate": 9.766632777274738e-05,
+      "loss": 0.1838,
+      "mean_token_accuracy": 0.9335328578948975,
+      "num_tokens": 404644.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0550259828567505,
+      "epoch": 0.29636711281070743,
+      "grad_norm": 2.301565647125244,
+      "learning_rate": 9.742168447175182e-05,
+      "loss": 0.1863,
+      "mean_token_accuracy": 0.9257804036140442,
+      "num_tokens": 418529.0,
+      "step": 310
+    },
+    {
+      "entropy": 1.0135290801525116,
+      "epoch": 0.30592734225621415,
+      "grad_norm": 1.5739341974258423,
+      "learning_rate": 9.716518697920655e-05,
+      "loss": 0.161,
+      "mean_token_accuracy": 0.9401273846626281,
+      "num_tokens": 431509.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0304184675216674,
+      "epoch": 0.3154875717017208,
+      "grad_norm": 2.273526906967163,
+      "learning_rate": 9.689689941283417e-05,
+      "loss": 0.1986,
+      "mean_token_accuracy": 0.9245898365974426,
+      "num_tokens": 445040.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.0275025486946106,
+      "epoch": 0.32504780114722753,
+      "grad_norm": 2.805868148803711,
+      "learning_rate": 9.661688883756996e-05,
+      "loss": 0.1782,
+      "mean_token_accuracy": 0.9292487025260925,
+      "num_tokens": 458540.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0296813428401947,
+      "epoch": 0.33460803059273425,
+      "grad_norm": 2.7913103103637695,
+      "learning_rate": 9.632522524879756e-05,
+      "loss": 0.1874,
+      "mean_token_accuracy": 0.9246273815631867,
+      "num_tokens": 471556.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.0276844143867492,
+      "epoch": 0.3441682600382409,
+      "grad_norm": 2.3035671710968018,
+      "learning_rate": 9.602198155485176e-05,
+      "loss": 0.1715,
+      "mean_token_accuracy": 0.9284400105476379,
+      "num_tokens": 484786.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.0496327757835389,
+      "epoch": 0.35372848948374763,
+      "grad_norm": 2.002821445465088,
+      "learning_rate": 9.570723355879342e-05,
+      "loss": 0.1856,
+      "mean_token_accuracy": 0.9277691960334777,
+      "num_tokens": 498352.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.0185538589954377,
+      "epoch": 0.3632887189292543,
+      "grad_norm": 2.568770408630371,
+      "learning_rate": 9.538105993946068e-05,
+      "loss": 0.1782,
+      "mean_token_accuracy": 0.934704166650772,
+      "num_tokens": 512064.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0153199493885041,
+      "epoch": 0.372848948374761,
+      "grad_norm": 2.152120590209961,
+      "learning_rate": 9.504354223180118e-05,
+      "loss": 0.1768,
+      "mean_token_accuracy": 0.9294376611709595,
+      "num_tokens": 525582.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.018614614009857,
+      "epoch": 0.3824091778202677,
+      "grad_norm": 3.645570993423462,
+      "learning_rate": 9.469476480649056e-05,
+      "loss": 0.1934,
+      "mean_token_accuracy": 0.9247165203094483,
+      "num_tokens": 539561.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.0288545787334442,
+      "epoch": 0.3919694072657744,
+      "grad_norm": 1.5050228834152222,
+      "learning_rate": 9.433481484884189e-05,
+      "loss": 0.1781,
+      "mean_token_accuracy": 0.927349853515625,
+      "num_tokens": 553625.0,
+      "step": 410
+    },
+    {
+      "entropy": 1.0082518339157105,
+      "epoch": 0.40152963671128106,
+      "grad_norm": 2.499924659729004,
+      "learning_rate": 9.396378233701167e-05,
+      "loss": 0.1587,
+      "mean_token_accuracy": 0.9408773958683014,
+      "num_tokens": 565891.0,
+      "step": 420
+    },
+    {
+      "entropy": 1.0218021392822265,
+      "epoch": 0.4110898661567878,
+      "grad_norm": 1.820673942565918,
+      "learning_rate": 9.358176001950759e-05,
+      "loss": 0.1751,
+      "mean_token_accuracy": 0.9365602552890777,
+      "num_tokens": 578722.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.0459587931632996,
+      "epoch": 0.42065009560229444,
+      "grad_norm": 3.3869118690490723,
+      "learning_rate": 9.318884339200378e-05,
+      "loss": 0.204,
+      "mean_token_accuracy": 0.9171568214893341,
+      "num_tokens": 592646.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0208855271339417,
+      "epoch": 0.43021032504780116,
+      "grad_norm": 2.3731369972229004,
+      "learning_rate": 9.278513067346941e-05,
+      "loss": 0.1738,
+      "mean_token_accuracy": 0.9343746900558472,
+      "num_tokens": 606429.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0310530364513397,
+      "epoch": 0.4397705544933078,
+      "grad_norm": 2.376760244369507,
+      "learning_rate": 9.237072278161652e-05,
+      "loss": 0.2049,
+      "mean_token_accuracy": 0.9211974799633026,
+      "num_tokens": 619626.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0150266408920288,
+      "epoch": 0.44933078393881454,
+      "grad_norm": 3.3328752517700195,
+      "learning_rate": 9.194572330767314e-05,
+      "loss": 0.1755,
+      "mean_token_accuracy": 0.9307337284088135,
+      "num_tokens": 633055.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.0314427733421325,
+      "epoch": 0.4588910133843212,
+      "grad_norm": 2.2189836502075195,
+      "learning_rate": 9.151023849048818e-05,
+      "loss": 0.1711,
+      "mean_token_accuracy": 0.9341503977775574,
+      "num_tokens": 647155.0,
+      "step": 480
+    },
+    {
+      "entropy": 1.006364119052887,
+      "epoch": 0.4684512428298279,
+      "grad_norm": 2.9035816192626953,
+      "learning_rate": 9.106437718997452e-05,
+      "loss": 0.1653,
+      "mean_token_accuracy": 0.9360855758190155,
+      "num_tokens": 660695.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0127965688705445,
+      "epoch": 0.4780114722753346,
+      "grad_norm": 2.0203046798706055,
+      "learning_rate": 9.060825085989682e-05,
+      "loss": 0.1706,
+      "mean_token_accuracy": 0.9349072694778442,
+      "num_tokens": 674509.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.0121367156505585,
+      "epoch": 0.4875717017208413,
+      "grad_norm": 3.026012420654297,
+      "learning_rate": 9.014197352001097e-05,
+      "loss": 0.1475,
+      "mean_token_accuracy": 0.9448499083518982,
+      "num_tokens": 687644.0,
+      "step": 510
+    },
+    {
+      "entropy": 1.0126056969165802,
+      "epoch": 0.497131931166348,
+      "grad_norm": 3.7356762886047363,
+      "learning_rate": 8.966566172756213e-05,
+      "loss": 0.1845,
+      "mean_token_accuracy": 0.9264914035797119,
+      "num_tokens": 700658.0,
+      "step": 520
+    },
+    {
+      "entropy": 1.0392110109329225,
+      "epoch": 0.5066921606118547,
+      "grad_norm": 3.5417075157165527,
+      "learning_rate": 8.91794345481484e-05,
+      "loss": 0.1889,
+      "mean_token_accuracy": 0.9311353266239166,
+      "num_tokens": 714234.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0234158277511596,
+      "epoch": 0.5162523900573613,
+      "grad_norm": 3.039093017578125,
+      "learning_rate": 8.868341352595755e-05,
+      "loss": 0.1924,
+      "mean_token_accuracy": 0.9290659904479981,
+      "num_tokens": 727721.0,
+      "step": 540
+    },
+    {
+      "entropy": 1.009736853837967,
+      "epoch": 0.5258126195028681,
+      "grad_norm": 1.7640941143035889,
+      "learning_rate": 8.817772265338409e-05,
+      "loss": 0.1839,
+      "mean_token_accuracy": 0.931582224369049,
+      "num_tokens": 741707.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.0224178433418274,
+      "epoch": 0.5353728489483748,
+      "grad_norm": 2.6304306983947754,
+      "learning_rate": 8.766248834003441e-05,
+      "loss": 0.1935,
+      "mean_token_accuracy": 0.9309299111366272,
+      "num_tokens": 755385.0,
+      "step": 560
+    },
+    {
+      "entropy": 1.0041300356388092,
+      "epoch": 0.5449330783938815,
+      "grad_norm": 2.4195950031280518,
+      "learning_rate": 8.713783938112768e-05,
+      "loss": 0.1626,
+      "mean_token_accuracy": 0.941334354877472,
+      "num_tokens": 768687.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.0264286398887634,
+      "epoch": 0.5544933078393881,
+      "grad_norm": 2.836859703063965,
+      "learning_rate": 8.660390692530026e-05,
+      "loss": 0.1646,
+      "mean_token_accuracy": 0.9407057404518128,
+      "num_tokens": 782172.0,
+      "step": 580
+    },
+    {
+      "entropy": 1.017165756225586,
+      "epoch": 0.5640535372848948,
+      "grad_norm": 1.536378026008606,
+      "learning_rate": 8.606082444182216e-05,
+      "loss": 0.1676,
+      "mean_token_accuracy": 0.9352243304252624,
+      "num_tokens": 795561.0,
+      "step": 590
+    },
+    {
+      "entropy": 1.0407780408859253,
+      "epoch": 0.5736137667304015,
+      "grad_norm": 1.6038671731948853,
+      "learning_rate": 8.550872768723292e-05,
+      "loss": 0.1447,
+      "mean_token_accuracy": 0.9454460620880127,
+      "num_tokens": 809436.0,
+      "step": 600
+    },
+    {
+      "entropy": 1.0389515042304993,
+      "epoch": 0.5831739961759083,
+      "grad_norm": 3.4060347080230713,
+      "learning_rate": 8.494775467140627e-05,
+      "loss": 0.1819,
+      "mean_token_accuracy": 0.9360428214073181,
+      "num_tokens": 822983.0,
+      "step": 610
+    },
+    {
+      "entropy": 1.0436920821666718,
+      "epoch": 0.5927342256214149,
+      "grad_norm": 2.140155553817749,
+      "learning_rate": 8.4378045623051e-05,
+      "loss": 0.1605,
+      "mean_token_accuracy": 0.939122611284256,
+      "num_tokens": 836766.0,
+      "step": 620
+    },
+    {
+      "entropy": 1.0456392228603364,
+      "epoch": 0.6022944550669216,
+      "grad_norm": 1.8615792989730835,
+      "learning_rate": 8.379974295465758e-05,
+      "loss": 0.1751,
+      "mean_token_accuracy": 0.9339857459068298,
+      "num_tokens": 850200.0,
+      "step": 630
+    },
+    {
+      "entropy": 1.034788340330124,
+      "epoch": 0.6118546845124283,
+      "grad_norm": 1.4612408876419067,
+      "learning_rate": 8.321299122689863e-05,
+      "loss": 0.1945,
+      "mean_token_accuracy": 0.9260159134864807,
+      "num_tokens": 864001.0,
+      "step": 640
+    },
+    {
+      "entropy": 1.0390719532966615,
+      "epoch": 0.621414913957935,
+      "grad_norm": 2.492454767227173,
+      "learning_rate": 8.261793711249254e-05,
+      "loss": 0.1809,
+      "mean_token_accuracy": 0.9303858399391174,
+      "num_tokens": 878135.0,
+      "step": 650
+    },
+    {
+      "entropy": 1.0123797118663789,
+      "epoch": 0.6309751434034416,
+      "grad_norm": 2.157985210418701,
+      "learning_rate": 8.201472935953916e-05,
+      "loss": 0.1754,
+      "mean_token_accuracy": 0.9341511011123658,
+      "num_tokens": 891307.0,
+      "step": 660
+    },
+    {
+      "entropy": 1.0271188616752625,
+      "epoch": 0.6405353728489483,
+      "grad_norm": 1.6143549680709839,
+      "learning_rate": 8.140351875433649e-05,
+      "loss": 0.1811,
+      "mean_token_accuracy": 0.9368877530097961,
+      "num_tokens": 904829.0,
+      "step": 670
+    },
+    {
+      "entropy": 1.011959147453308,
+      "epoch": 0.6500956022944551,
+      "grad_norm": 1.9572798013687134,
+      "learning_rate": 8.07844580836881e-05,
+      "loss": 0.1564,
+      "mean_token_accuracy": 0.9395960152149201,
+      "num_tokens": 919199.0,
+      "step": 680
+    },
+    {
+      "entropy": 1.0311509728431703,
+      "epoch": 0.6596558317399618,
+      "grad_norm": 1.7023526430130005,
+      "learning_rate": 8.015770209671044e-05,
+      "loss": 0.1868,
+      "mean_token_accuracy": 0.9322409808635712,
+      "num_tokens": 932439.0,
+      "step": 690
+    },
+    {
+      "entropy": 1.0227358162403106,
+      "epoch": 0.6692160611854685,
+      "grad_norm": 1.8079496622085571,
+      "learning_rate": 7.952340746614946e-05,
+      "loss": 0.1849,
+      "mean_token_accuracy": 0.9285174906253815,
+      "num_tokens": 945776.0,
+      "step": 700
+    },
+    {
+      "entropy": 1.0026035130023956,
+      "epoch": 0.6787762906309751,
+      "grad_norm": 1.7818984985351562,
+      "learning_rate": 7.888173274921655e-05,
+      "loss": 0.1514,
+      "mean_token_accuracy": 0.9430120289325714,
+      "num_tokens": 959352.0,
+      "step": 710
+    },
+    {
+      "entropy": 1.0062686502933502,
+      "epoch": 0.6883365200764818,
+      "grad_norm": 2.592531204223633,
+      "learning_rate": 7.823283834795336e-05,
+      "loss": 0.183,
+      "mean_token_accuracy": 0.9306194126605988,
+      "num_tokens": 972596.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9978627741336823,
+      "epoch": 0.6978967495219885,
+      "grad_norm": 2.9630000591278076,
+      "learning_rate": 7.75768864691354e-05,
+      "loss": 0.173,
+      "mean_token_accuracy": 0.9347139298915863,
+      "num_tokens": 985699.0,
+      "step": 730
+    },
+    {
+      "entropy": 1.0089325785636902,
+      "epoch": 0.7074569789674953,
+      "grad_norm": 2.3451693058013916,
+      "learning_rate": 7.691404108372461e-05,
+      "loss": 0.1671,
+      "mean_token_accuracy": 0.9322668850421906,
+      "num_tokens": 999728.0,
+      "step": 740
+    },
+    {
+      "entropy": 1.0036091327667236,
+      "epoch": 0.7170172084130019,
+      "grad_norm": 2.0035459995269775,
+      "learning_rate": 7.624446788588085e-05,
+      "loss": 0.1459,
+      "mean_token_accuracy": 0.9423670887947082,
+      "num_tokens": 1013669.0,
+      "step": 750
+    },
+    {
+      "entropy": 1.0062048316001893,
+      "epoch": 0.7265774378585086,
+      "grad_norm": 1.940860390663147,
+      "learning_rate": 7.556833425154255e-05,
+      "loss": 0.1447,
+      "mean_token_accuracy": 0.943264490365982,
+      "num_tokens": 1027609.0,
+      "step": 760
+    },
+    {
+      "entropy": 1.0209776043891907,
+      "epoch": 0.7361376673040153,
+      "grad_norm": 2.7939534187316895,
+      "learning_rate": 7.488580919658727e-05,
+      "loss": 0.1714,
+      "mean_token_accuracy": 0.9344382464885712,
+      "num_tokens": 1041046.0,
+      "step": 770
+    },
+    {
+      "entropy": 1.0077354550361632,
+      "epoch": 0.745697896749522,
+      "grad_norm": 3.3370931148529053,
+      "learning_rate": 7.419706333458201e-05,
+      "loss": 0.1583,
+      "mean_token_accuracy": 0.9370908617973328,
+      "num_tokens": 1055304.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9912185788154602,
+      "epoch": 0.7552581261950286,
+      "grad_norm": 3.3952715396881104,
+      "learning_rate": 7.350226883413413e-05,
+      "loss": 0.1702,
+      "mean_token_accuracy": 0.9344295501708985,
+      "num_tokens": 1068542.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9772989094257355,
+      "epoch": 0.7648183556405354,
+      "grad_norm": 2.841501235961914,
+      "learning_rate": 7.280159937585386e-05,
+      "loss": 0.1607,
+      "mean_token_accuracy": 0.938536924123764,
+      "num_tokens": 1082195.0,
+      "step": 800
+    },
+    {
+      "entropy": 1.0071873247623444,
+      "epoch": 0.7743785850860421,
+      "grad_norm": 3.672098159790039,
+      "learning_rate": 7.209523010893853e-05,
+      "loss": 0.1521,
+      "mean_token_accuracy": 0.9436739087104797,
+      "num_tokens": 1095884.0,
+      "step": 810
+    },
+    {
+      "entropy": 1.020501399040222,
+      "epoch": 0.7839388145315488,
+      "grad_norm": 1.6433815956115723,
+      "learning_rate": 7.138333760738987e-05,
+      "loss": 0.1637,
+      "mean_token_accuracy": 0.9373888850212098,
+      "num_tokens": 1109287.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9960210621356964,
+      "epoch": 0.7934990439770554,
+      "grad_norm": 2.5117626190185547,
+      "learning_rate": 7.066609982587501e-05,
+      "loss": 0.1473,
+      "mean_token_accuracy": 0.9446003437042236,
+      "num_tokens": 1123392.0,
+      "step": 830
+    },
+    {
+      "entropy": 1.0222314298152924,
+      "epoch": 0.8030592734225621,
+      "grad_norm": 2.1828365325927734,
+      "learning_rate": 6.994369605524253e-05,
+      "loss": 0.1619,
+      "mean_token_accuracy": 0.93698371052742,
+      "num_tokens": 1137691.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9974071145057678,
+      "epoch": 0.8126195028680688,
+      "grad_norm": 2.5629751682281494,
+      "learning_rate": 6.921630687770427e-05,
+      "loss": 0.1614,
+      "mean_token_accuracy": 0.9353620290756226,
+      "num_tokens": 1151400.0,
+      "step": 850
+    },
+    {
+      "entropy": 1.0028881013393403,
+      "epoch": 0.8221797323135756,
+      "grad_norm": 2.5358316898345947,
+      "learning_rate": 6.848411412169461e-05,
+      "loss": 0.1402,
+      "mean_token_accuracy": 0.9457783937454224,
+      "num_tokens": 1165074.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9931535959243775,
+      "epoch": 0.8317399617590823,
+      "grad_norm": 2.6374380588531494,
+      "learning_rate": 6.774730081641799e-05,
+      "loss": 0.1516,
+      "mean_token_accuracy": 0.9389132201671601,
+      "num_tokens": 1178919.0,
+      "step": 870
+    },
+    {
+      "entropy": 1.0055995404720306,
+      "epoch": 0.8413001912045889,
+      "grad_norm": 2.9471771717071533,
+      "learning_rate": 6.70060511460963e-05,
+      "loss": 0.1719,
+      "mean_token_accuracy": 0.9344942331314087,
+      "num_tokens": 1192204.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9824313879013061,
+      "epoch": 0.8508604206500956,
+      "grad_norm": 2.0739288330078125,
+      "learning_rate": 6.626055040392774e-05,
+      "loss": 0.1668,
+      "mean_token_accuracy": 0.9333541214466095,
+      "num_tokens": 1205571.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9760380923748017,
+      "epoch": 0.8604206500956023,
+      "grad_norm": 2.9645445346832275,
+      "learning_rate": 6.551098494576819e-05,
+      "loss": 0.1683,
+      "mean_token_accuracy": 0.933848249912262,
+      "num_tokens": 1219043.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9830373346805572,
+      "epoch": 0.869980879541109,
+      "grad_norm": 2.368391752243042,
+      "learning_rate": 6.475754214354715e-05,
+      "loss": 0.1585,
+      "mean_token_accuracy": 0.9392736613750458,
+      "num_tokens": 1233113.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.9722706019878388,
+      "epoch": 0.8795411089866156,
+      "grad_norm": 2.0587034225463867,
+      "learning_rate": 6.400041033842965e-05,
+      "loss": 0.1477,
+      "mean_token_accuracy": 0.9435854613780975,
+      "num_tokens": 1245931.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9533098459243774,
+      "epoch": 0.8891013384321224,
+      "grad_norm": 2.23323392868042,
+      "learning_rate": 6.323977879373571e-05,
+      "loss": 0.1535,
+      "mean_token_accuracy": 0.9388725519180298,
+      "num_tokens": 1259949.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.977231240272522,
+      "epoch": 0.8986615678776291,
+      "grad_norm": 2.4451472759246826,
+      "learning_rate": 6.247583764762959e-05,
+      "loss": 0.1523,
+      "mean_token_accuracy": 0.9421009659767151,
+      "num_tokens": 1273413.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9832057058811188,
+      "epoch": 0.9082217973231358,
+      "grad_norm": 1.7953559160232544,
+      "learning_rate": 6.170877786559001e-05,
+      "loss": 0.1524,
+      "mean_token_accuracy": 0.9403273582458496,
+      "num_tokens": 1286343.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.9728405475616455,
+      "epoch": 0.9177820267686424,
+      "grad_norm": 2.614015579223633,
+      "learning_rate": 6.0938791192673895e-05,
+      "loss": 0.1671,
+      "mean_token_accuracy": 0.9358247876167297,
+      "num_tokens": 1300089.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9851917386054992,
+      "epoch": 0.9273422562141491,
+      "grad_norm": 2.835686445236206,
+      "learning_rate": 6.016607010558492e-05,
+      "loss": 0.1497,
+      "mean_token_accuracy": 0.9398622572422027,
+      "num_tokens": 1313412.0,
+      "step": 970
+    },
+    {
+      "entropy": 1.0004826962947846,
+      "epoch": 0.9369024856596558,
+      "grad_norm": 2.6028459072113037,
+      "learning_rate": 5.939080776455942e-05,
+      "loss": 0.1614,
+      "mean_token_accuracy": 0.9344104588031769,
+      "num_tokens": 1326290.0,
+      "step": 980
+    },
+    {
+      "entropy": 1.0054877161979676,
+      "epoch": 0.9464627151051626,
+      "grad_norm": 2.5106935501098633,
+      "learning_rate": 5.861319796508139e-05,
+      "loss": 0.1434,
+      "mean_token_accuracy": 0.9470022737979888,
+      "num_tokens": 1339719.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.981327623128891,
+      "epoch": 0.9560229445506692,
+      "grad_norm": 2.0723421573638916,
+      "learning_rate": 5.783343508943849e-05,
+      "loss": 0.1326,
+      "mean_token_accuracy": 0.945333081483841,
+      "num_tokens": 1352423.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9760708332061767,
+      "epoch": 0.9655831739961759,
+      "grad_norm": 2.7101802825927734,
+      "learning_rate": 5.7051714058131654e-05,
+      "loss": 0.1387,
+      "mean_token_accuracy": 0.9463435471057892,
+      "num_tokens": 1365889.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.9664070129394531,
+      "epoch": 0.9751434034416826,
+      "grad_norm": 2.450268268585205,
+      "learning_rate": 5.626823028114991e-05,
+      "loss": 0.1447,
+      "mean_token_accuracy": 0.94169100522995,
+      "num_tokens": 1379353.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9760319530963898,
+      "epoch": 0.9847036328871893,
+      "grad_norm": 2.1834616661071777,
+      "learning_rate": 5.548317960912305e-05,
+      "loss": 0.1603,
+      "mean_token_accuracy": 0.937016350030899,
+      "num_tokens": 1392866.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9738619923591614,
+      "epoch": 0.994263862332696,
+      "grad_norm": 1.9375931024551392,
+      "learning_rate": 5.4696758284363956e-05,
+      "loss": 0.1548,
+      "mean_token_accuracy": 0.9437321364879608,
+      "num_tokens": 1406263.0,
+      "step": 1040
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2092,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.015340529185075e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-1046/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcf9e5f91d6a7b9568babb6537178d077261e0ab056e9b046da1b7ee3de28a35
+size 6353

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-2092/README.md

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-2092/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "o_proj",
+    "up_proj",
+    "v_proj",
+    "k_proj",
+    "gate_proj",
+    "down_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-2092/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b6b0a9807adc5da265640817da651e603f2ce975b26af6b3dacf878eb5122ac
+size 80792096

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-2092/added_tokens.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-2092/chat_template.jinja

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-2092/special_tokens_map.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-2092/tokenizer.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-2092/tokenizer_config.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-2092/trainer_state.json

@@ -0,0 +1,2124 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 2092,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.1737810969352722,
+      "epoch": 0.009560229445506692,
+      "grad_norm": 1.871079921722412,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2769,
+      "mean_token_accuracy": 0.9123528063297272,
+      "num_tokens": 14125.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1810022592544556,
+      "epoch": 0.019120458891013385,
+      "grad_norm": 1.8740425109863281,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2386,
+      "mean_token_accuracy": 0.9218324422836304,
+      "num_tokens": 27004.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1399657130241394,
+      "epoch": 0.028680688336520075,
+      "grad_norm": 2.0634806156158447,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.1882,
+      "mean_token_accuracy": 0.9362742602825165,
+      "num_tokens": 41395.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.0615105748176574,
+      "epoch": 0.03824091778202677,
+      "grad_norm": 2.3918206691741943,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1631,
+      "mean_token_accuracy": 0.9405273735523224,
+      "num_tokens": 55233.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.0151220917701722,
+      "epoch": 0.04780114722753346,
+      "grad_norm": 3.8594629764556885,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.1767,
+      "mean_token_accuracy": 0.9330196499824523,
+      "num_tokens": 68341.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.014986515045166,
+      "epoch": 0.05736137667304015,
+      "grad_norm": 2.2135467529296875,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1738,
+      "mean_token_accuracy": 0.9349386811256408,
+      "num_tokens": 81730.0,
+      "step": 60
+    },
+    {
+      "entropy": 0.9990480720996857,
+      "epoch": 0.06692160611854685,
+      "grad_norm": 2.428823709487915,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.162,
+      "mean_token_accuracy": 0.9420714497566223,
+      "num_tokens": 94886.0,
+      "step": 70
+    },
+    {
+      "entropy": 0.9700570225715637,
+      "epoch": 0.07648183556405354,
+      "grad_norm": 2.4947664737701416,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1805,
+      "mean_token_accuracy": 0.9348644435405731,
+      "num_tokens": 108569.0,
+      "step": 80
+    },
+    {
+      "entropy": 0.9715868353843689,
+      "epoch": 0.08604206500956023,
+      "grad_norm": 1.909283995628357,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1595,
+      "mean_token_accuracy": 0.941025310754776,
+      "num_tokens": 122156.0,
+      "step": 90
+    },
+    {
+      "entropy": 0.9832429409027099,
+      "epoch": 0.09560229445506692,
+      "grad_norm": 1.9785945415496826,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1782,
+      "mean_token_accuracy": 0.9320050239562988,
+      "num_tokens": 135290.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0139548778533936,
+      "epoch": 0.10516252390057361,
+      "grad_norm": 2.554082155227661,
+      "learning_rate": 9.999900008621665e-05,
+      "loss": 0.205,
+      "mean_token_accuracy": 0.9271113276481628,
+      "num_tokens": 149015.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.00105682015419,
+      "epoch": 0.1147227533460803,
+      "grad_norm": 1.7623502016067505,
+      "learning_rate": 9.998775151544478e-05,
+      "loss": 0.1824,
+      "mean_token_accuracy": 0.93257275223732,
+      "num_tokens": 162147.0,
+      "step": 120
+    },
+    {
+      "entropy": 0.9644641637802124,
+      "epoch": 0.124282982791587,
+      "grad_norm": 1.752968192100525,
+      "learning_rate": 9.996400730289628e-05,
+      "loss": 0.1847,
+      "mean_token_accuracy": 0.9332687795162201,
+      "num_tokens": 175436.0,
+      "step": 130
+    },
+    {
+      "entropy": 0.971268481016159,
+      "epoch": 0.1338432122370937,
+      "grad_norm": 2.4105324745178223,
+      "learning_rate": 9.992777338400864e-05,
+      "loss": 0.1777,
+      "mean_token_accuracy": 0.9332670569419861,
+      "num_tokens": 188834.0,
+      "step": 140
+    },
+    {
+      "entropy": 0.9512466073036194,
+      "epoch": 0.14340344168260039,
+      "grad_norm": 2.102140188217163,
+      "learning_rate": 9.987905881632213e-05,
+      "loss": 0.1821,
+      "mean_token_accuracy": 0.934379369020462,
+      "num_tokens": 202708.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.9578613698482513,
+      "epoch": 0.15296367112810708,
+      "grad_norm": 1.9931540489196777,
+      "learning_rate": 9.981787577721553e-05,
+      "loss": 0.1768,
+      "mean_token_accuracy": 0.934454756975174,
+      "num_tokens": 216384.0,
+      "step": 160
+    },
+    {
+      "entropy": 1.021405816078186,
+      "epoch": 0.16252390057361377,
+      "grad_norm": 1.9813957214355469,
+      "learning_rate": 9.974423956086228e-05,
+      "loss": 0.1901,
+      "mean_token_accuracy": 0.9332379758358001,
+      "num_tokens": 229105.0,
+      "step": 170
+    },
+    {
+      "entropy": 1.0439977526664734,
+      "epoch": 0.17208413001912046,
+      "grad_norm": 2.317162275314331,
+      "learning_rate": 9.965816857440717e-05,
+      "loss": 0.1632,
+      "mean_token_accuracy": 0.9360143065452575,
+      "num_tokens": 242264.0,
+      "step": 180
+    },
+    {
+      "entropy": 1.0610752880573273,
+      "epoch": 0.18164435946462715,
+      "grad_norm": 2.324665069580078,
+      "learning_rate": 9.955968433336515e-05,
+      "loss": 0.1971,
+      "mean_token_accuracy": 0.922540009021759,
+      "num_tokens": 255602.0,
+      "step": 190
+    },
+    {
+      "entropy": 1.0195186913013459,
+      "epoch": 0.19120458891013384,
+      "grad_norm": 1.5599435567855835,
+      "learning_rate": 9.944881145624291e-05,
+      "loss": 0.1389,
+      "mean_token_accuracy": 0.948070353269577,
+      "num_tokens": 269712.0,
+      "step": 200
+    },
+    {
+      "entropy": 1.0296171724796295,
+      "epoch": 0.20076481835564053,
+      "grad_norm": 2.3139309883117676,
+      "learning_rate": 9.932557765838504e-05,
+      "loss": 0.179,
+      "mean_token_accuracy": 0.9331721246242524,
+      "num_tokens": 283572.0,
+      "step": 210
+    },
+    {
+      "entropy": 1.0230853736400605,
+      "epoch": 0.21032504780114722,
+      "grad_norm": 1.953762412071228,
+      "learning_rate": 9.919001374504569e-05,
+      "loss": 0.1793,
+      "mean_token_accuracy": 0.9311605632305145,
+      "num_tokens": 297437.0,
+      "step": 220
+    },
+    {
+      "entropy": 1.0038200318813324,
+      "epoch": 0.2198852772466539,
+      "grad_norm": 4.689145565032959,
+      "learning_rate": 9.904215360368826e-05,
+      "loss": 0.1809,
+      "mean_token_accuracy": 0.9343967974185944,
+      "num_tokens": 310803.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.0401727676391601,
+      "epoch": 0.2294455066921606,
+      "grad_norm": 7.431188583374023,
+      "learning_rate": 9.888203419551432e-05,
+      "loss": 0.2055,
+      "mean_token_accuracy": 0.9227587044239044,
+      "num_tokens": 324713.0,
+      "step": 240
+    },
+    {
+      "entropy": 1.028674304485321,
+      "epoch": 0.2390057361376673,
+      "grad_norm": 3.3888537883758545,
+      "learning_rate": 9.870969554622426e-05,
+      "loss": 0.2101,
+      "mean_token_accuracy": 0.920482850074768,
+      "num_tokens": 337634.0,
+      "step": 250
+    },
+    {
+      "entropy": 1.0425225734710692,
+      "epoch": 0.248565965583174,
+      "grad_norm": 2.374208450317383,
+      "learning_rate": 9.852518073601194e-05,
+      "loss": 0.1805,
+      "mean_token_accuracy": 0.9340396881103515,
+      "num_tokens": 350459.0,
+      "step": 260
+    },
+    {
+      "entropy": 1.0127095937728883,
+      "epoch": 0.25812619502868067,
+      "grad_norm": 2.729233741760254,
+      "learning_rate": 9.832853588879575e-05,
+      "loss": 0.1736,
+      "mean_token_accuracy": 0.9342186093330384,
+      "num_tokens": 364010.0,
+      "step": 270
+    },
+    {
+      "entropy": 1.039266401529312,
+      "epoch": 0.2676864244741874,
+      "grad_norm": 1.921285629272461,
+      "learning_rate": 9.811981016068881e-05,
+      "loss": 0.2124,
+      "mean_token_accuracy": 0.9189038336277008,
+      "num_tokens": 377021.0,
+      "step": 280
+    },
+    {
+      "entropy": 1.0290949404239655,
+      "epoch": 0.27724665391969405,
+      "grad_norm": 1.8515435457229614,
+      "learning_rate": 9.78990557277112e-05,
+      "loss": 0.2009,
+      "mean_token_accuracy": 0.9233260750770569,
+      "num_tokens": 390819.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0516928434371948,
+      "epoch": 0.28680688336520077,
+      "grad_norm": 2.3137047290802,
+      "learning_rate": 9.766632777274738e-05,
+      "loss": 0.1838,
+      "mean_token_accuracy": 0.9335328578948975,
+      "num_tokens": 404644.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0550259828567505,
+      "epoch": 0.29636711281070743,
+      "grad_norm": 2.301565647125244,
+      "learning_rate": 9.742168447175182e-05,
+      "loss": 0.1863,
+      "mean_token_accuracy": 0.9257804036140442,
+      "num_tokens": 418529.0,
+      "step": 310
+    },
+    {
+      "entropy": 1.0135290801525116,
+      "epoch": 0.30592734225621415,
+      "grad_norm": 1.5739341974258423,
+      "learning_rate": 9.716518697920655e-05,
+      "loss": 0.161,
+      "mean_token_accuracy": 0.9401273846626281,
+      "num_tokens": 431509.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0304184675216674,
+      "epoch": 0.3154875717017208,
+      "grad_norm": 2.273526906967163,
+      "learning_rate": 9.689689941283417e-05,
+      "loss": 0.1986,
+      "mean_token_accuracy": 0.9245898365974426,
+      "num_tokens": 445040.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.0275025486946106,
+      "epoch": 0.32504780114722753,
+      "grad_norm": 2.805868148803711,
+      "learning_rate": 9.661688883756996e-05,
+      "loss": 0.1782,
+      "mean_token_accuracy": 0.9292487025260925,
+      "num_tokens": 458540.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0296813428401947,
+      "epoch": 0.33460803059273425,
+      "grad_norm": 2.7913103103637695,
+      "learning_rate": 9.632522524879756e-05,
+      "loss": 0.1874,
+      "mean_token_accuracy": 0.9246273815631867,
+      "num_tokens": 471556.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.0276844143867492,
+      "epoch": 0.3441682600382409,
+      "grad_norm": 2.3035671710968018,
+      "learning_rate": 9.602198155485176e-05,
+      "loss": 0.1715,
+      "mean_token_accuracy": 0.9284400105476379,
+      "num_tokens": 484786.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.0496327757835389,
+      "epoch": 0.35372848948374763,
+      "grad_norm": 2.002821445465088,
+      "learning_rate": 9.570723355879342e-05,
+      "loss": 0.1856,
+      "mean_token_accuracy": 0.9277691960334777,
+      "num_tokens": 498352.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.0185538589954377,
+      "epoch": 0.3632887189292543,
+      "grad_norm": 2.568770408630371,
+      "learning_rate": 9.538105993946068e-05,
+      "loss": 0.1782,
+      "mean_token_accuracy": 0.934704166650772,
+      "num_tokens": 512064.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0153199493885041,
+      "epoch": 0.372848948374761,
+      "grad_norm": 2.152120590209961,
+      "learning_rate": 9.504354223180118e-05,
+      "loss": 0.1768,
+      "mean_token_accuracy": 0.9294376611709595,
+      "num_tokens": 525582.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.018614614009857,
+      "epoch": 0.3824091778202677,
+      "grad_norm": 3.645570993423462,
+      "learning_rate": 9.469476480649056e-05,
+      "loss": 0.1934,
+      "mean_token_accuracy": 0.9247165203094483,
+      "num_tokens": 539561.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.0288545787334442,
+      "epoch": 0.3919694072657744,
+      "grad_norm": 1.5050228834152222,
+      "learning_rate": 9.433481484884189e-05,
+      "loss": 0.1781,
+      "mean_token_accuracy": 0.927349853515625,
+      "num_tokens": 553625.0,
+      "step": 410
+    },
+    {
+      "entropy": 1.0082518339157105,
+      "epoch": 0.40152963671128106,
+      "grad_norm": 2.499924659729004,
+      "learning_rate": 9.396378233701167e-05,
+      "loss": 0.1587,
+      "mean_token_accuracy": 0.9408773958683014,
+      "num_tokens": 565891.0,
+      "step": 420
+    },
+    {
+      "entropy": 1.0218021392822265,
+      "epoch": 0.4110898661567878,
+      "grad_norm": 1.820673942565918,
+      "learning_rate": 9.358176001950759e-05,
+      "loss": 0.1751,
+      "mean_token_accuracy": 0.9365602552890777,
+      "num_tokens": 578722.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.0459587931632996,
+      "epoch": 0.42065009560229444,
+      "grad_norm": 3.3869118690490723,
+      "learning_rate": 9.318884339200378e-05,
+      "loss": 0.204,
+      "mean_token_accuracy": 0.9171568214893341,
+      "num_tokens": 592646.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0208855271339417,
+      "epoch": 0.43021032504780116,
+      "grad_norm": 2.3731369972229004,
+      "learning_rate": 9.278513067346941e-05,
+      "loss": 0.1738,
+      "mean_token_accuracy": 0.9343746900558472,
+      "num_tokens": 606429.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0310530364513397,
+      "epoch": 0.4397705544933078,
+      "grad_norm": 2.376760244369507,
+      "learning_rate": 9.237072278161652e-05,
+      "loss": 0.2049,
+      "mean_token_accuracy": 0.9211974799633026,
+      "num_tokens": 619626.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0150266408920288,
+      "epoch": 0.44933078393881454,
+      "grad_norm": 3.3328752517700195,
+      "learning_rate": 9.194572330767314e-05,
+      "loss": 0.1755,
+      "mean_token_accuracy": 0.9307337284088135,
+      "num_tokens": 633055.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.0314427733421325,
+      "epoch": 0.4588910133843212,
+      "grad_norm": 2.2189836502075195,
+      "learning_rate": 9.151023849048818e-05,
+      "loss": 0.1711,
+      "mean_token_accuracy": 0.9341503977775574,
+      "num_tokens": 647155.0,
+      "step": 480
+    },
+    {
+      "entropy": 1.006364119052887,
+      "epoch": 0.4684512428298279,
+      "grad_norm": 2.9035816192626953,
+      "learning_rate": 9.106437718997452e-05,
+      "loss": 0.1653,
+      "mean_token_accuracy": 0.9360855758190155,
+      "num_tokens": 660695.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0127965688705445,
+      "epoch": 0.4780114722753346,
+      "grad_norm": 2.0203046798706055,
+      "learning_rate": 9.060825085989682e-05,
+      "loss": 0.1706,
+      "mean_token_accuracy": 0.9349072694778442,
+      "num_tokens": 674509.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.0121367156505585,
+      "epoch": 0.4875717017208413,
+      "grad_norm": 3.026012420654297,
+      "learning_rate": 9.014197352001097e-05,
+      "loss": 0.1475,
+      "mean_token_accuracy": 0.9448499083518982,
+      "num_tokens": 687644.0,
+      "step": 510
+    },
+    {
+      "entropy": 1.0126056969165802,
+      "epoch": 0.497131931166348,
+      "grad_norm": 3.7356762886047363,
+      "learning_rate": 8.966566172756213e-05,
+      "loss": 0.1845,
+      "mean_token_accuracy": 0.9264914035797119,
+      "num_tokens": 700658.0,
+      "step": 520
+    },
+    {
+      "entropy": 1.0392110109329225,
+      "epoch": 0.5066921606118547,
+      "grad_norm": 3.5417075157165527,
+      "learning_rate": 8.91794345481484e-05,
+      "loss": 0.1889,
+      "mean_token_accuracy": 0.9311353266239166,
+      "num_tokens": 714234.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0234158277511596,
+      "epoch": 0.5162523900573613,
+      "grad_norm": 3.039093017578125,
+      "learning_rate": 8.868341352595755e-05,
+      "loss": 0.1924,
+      "mean_token_accuracy": 0.9290659904479981,
+      "num_tokens": 727721.0,
+      "step": 540
+    },
+    {
+      "entropy": 1.009736853837967,
+      "epoch": 0.5258126195028681,
+      "grad_norm": 1.7640941143035889,
+      "learning_rate": 8.817772265338409e-05,
+      "loss": 0.1839,
+      "mean_token_accuracy": 0.931582224369049,
+      "num_tokens": 741707.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.0224178433418274,
+      "epoch": 0.5353728489483748,
+      "grad_norm": 2.6304306983947754,
+      "learning_rate": 8.766248834003441e-05,
+      "loss": 0.1935,
+      "mean_token_accuracy": 0.9309299111366272,
+      "num_tokens": 755385.0,
+      "step": 560
+    },
+    {
+      "entropy": 1.0041300356388092,
+      "epoch": 0.5449330783938815,
+      "grad_norm": 2.4195950031280518,
+      "learning_rate": 8.713783938112768e-05,
+      "loss": 0.1626,
+      "mean_token_accuracy": 0.941334354877472,
+      "num_tokens": 768687.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.0264286398887634,
+      "epoch": 0.5544933078393881,
+      "grad_norm": 2.836859703063965,
+      "learning_rate": 8.660390692530026e-05,
+      "loss": 0.1646,
+      "mean_token_accuracy": 0.9407057404518128,
+      "num_tokens": 782172.0,
+      "step": 580
+    },
+    {
+      "entropy": 1.017165756225586,
+      "epoch": 0.5640535372848948,
+      "grad_norm": 1.536378026008606,
+      "learning_rate": 8.606082444182216e-05,
+      "loss": 0.1676,
+      "mean_token_accuracy": 0.9352243304252624,
+      "num_tokens": 795561.0,
+      "step": 590
+    },
+    {
+      "entropy": 1.0407780408859253,
+      "epoch": 0.5736137667304015,
+      "grad_norm": 1.6038671731948853,
+      "learning_rate": 8.550872768723292e-05,
+      "loss": 0.1447,
+      "mean_token_accuracy": 0.9454460620880127,
+      "num_tokens": 809436.0,
+      "step": 600
+    },
+    {
+      "entropy": 1.0389515042304993,
+      "epoch": 0.5831739961759083,
+      "grad_norm": 3.4060347080230713,
+      "learning_rate": 8.494775467140627e-05,
+      "loss": 0.1819,
+      "mean_token_accuracy": 0.9360428214073181,
+      "num_tokens": 822983.0,
+      "step": 610
+    },
+    {
+      "entropy": 1.0436920821666718,
+      "epoch": 0.5927342256214149,
+      "grad_norm": 2.140155553817749,
+      "learning_rate": 8.4378045623051e-05,
+      "loss": 0.1605,
+      "mean_token_accuracy": 0.939122611284256,
+      "num_tokens": 836766.0,
+      "step": 620
+    },
+    {
+      "entropy": 1.0456392228603364,
+      "epoch": 0.6022944550669216,
+      "grad_norm": 1.8615792989730835,
+      "learning_rate": 8.379974295465758e-05,
+      "loss": 0.1751,
+      "mean_token_accuracy": 0.9339857459068298,
+      "num_tokens": 850200.0,
+      "step": 630
+    },
+    {
+      "entropy": 1.034788340330124,
+      "epoch": 0.6118546845124283,
+      "grad_norm": 1.4612408876419067,
+      "learning_rate": 8.321299122689863e-05,
+      "loss": 0.1945,
+      "mean_token_accuracy": 0.9260159134864807,
+      "num_tokens": 864001.0,
+      "step": 640
+    },
+    {
+      "entropy": 1.0390719532966615,
+      "epoch": 0.621414913957935,
+      "grad_norm": 2.492454767227173,
+      "learning_rate": 8.261793711249254e-05,
+      "loss": 0.1809,
+      "mean_token_accuracy": 0.9303858399391174,
+      "num_tokens": 878135.0,
+      "step": 650
+    },
+    {
+      "entropy": 1.0123797118663789,
+      "epoch": 0.6309751434034416,
+      "grad_norm": 2.157985210418701,
+      "learning_rate": 8.201472935953916e-05,
+      "loss": 0.1754,
+      "mean_token_accuracy": 0.9341511011123658,
+      "num_tokens": 891307.0,
+      "step": 660
+    },
+    {
+      "entropy": 1.0271188616752625,
+      "epoch": 0.6405353728489483,
+      "grad_norm": 1.6143549680709839,
+      "learning_rate": 8.140351875433649e-05,
+      "loss": 0.1811,
+      "mean_token_accuracy": 0.9368877530097961,
+      "num_tokens": 904829.0,
+      "step": 670
+    },
+    {
+      "entropy": 1.011959147453308,
+      "epoch": 0.6500956022944551,
+      "grad_norm": 1.9572798013687134,
+      "learning_rate": 8.07844580836881e-05,
+      "loss": 0.1564,
+      "mean_token_accuracy": 0.9395960152149201,
+      "num_tokens": 919199.0,
+      "step": 680
+    },
+    {
+      "entropy": 1.0311509728431703,
+      "epoch": 0.6596558317399618,
+      "grad_norm": 1.7023526430130005,
+      "learning_rate": 8.015770209671044e-05,
+      "loss": 0.1868,
+      "mean_token_accuracy": 0.9322409808635712,
+      "num_tokens": 932439.0,
+      "step": 690
+    },
+    {
+      "entropy": 1.0227358162403106,
+      "epoch": 0.6692160611854685,
+      "grad_norm": 1.8079496622085571,
+      "learning_rate": 7.952340746614946e-05,
+      "loss": 0.1849,
+      "mean_token_accuracy": 0.9285174906253815,
+      "num_tokens": 945776.0,
+      "step": 700
+    },
+    {
+      "entropy": 1.0026035130023956,
+      "epoch": 0.6787762906309751,
+      "grad_norm": 1.7818984985351562,
+      "learning_rate": 7.888173274921655e-05,
+      "loss": 0.1514,
+      "mean_token_accuracy": 0.9430120289325714,
+      "num_tokens": 959352.0,
+      "step": 710
+    },
+    {
+      "entropy": 1.0062686502933502,
+      "epoch": 0.6883365200764818,
+      "grad_norm": 2.592531204223633,
+      "learning_rate": 7.823283834795336e-05,
+      "loss": 0.183,
+      "mean_token_accuracy": 0.9306194126605988,
+      "num_tokens": 972596.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9978627741336823,
+      "epoch": 0.6978967495219885,
+      "grad_norm": 2.9630000591278076,
+      "learning_rate": 7.75768864691354e-05,
+      "loss": 0.173,
+      "mean_token_accuracy": 0.9347139298915863,
+      "num_tokens": 985699.0,
+      "step": 730
+    },
+    {
+      "entropy": 1.0089325785636902,
+      "epoch": 0.7074569789674953,
+      "grad_norm": 2.3451693058013916,
+      "learning_rate": 7.691404108372461e-05,
+      "loss": 0.1671,
+      "mean_token_accuracy": 0.9322668850421906,
+      "num_tokens": 999728.0,
+      "step": 740
+    },
+    {
+      "entropy": 1.0036091327667236,
+      "epoch": 0.7170172084130019,
+      "grad_norm": 2.0035459995269775,
+      "learning_rate": 7.624446788588085e-05,
+      "loss": 0.1459,
+      "mean_token_accuracy": 0.9423670887947082,
+      "num_tokens": 1013669.0,
+      "step": 750
+    },
+    {
+      "entropy": 1.0062048316001893,
+      "epoch": 0.7265774378585086,
+      "grad_norm": 1.940860390663147,
+      "learning_rate": 7.556833425154255e-05,
+      "loss": 0.1447,
+      "mean_token_accuracy": 0.943264490365982,
+      "num_tokens": 1027609.0,
+      "step": 760
+    },
+    {
+      "entropy": 1.0209776043891907,
+      "epoch": 0.7361376673040153,
+      "grad_norm": 2.7939534187316895,
+      "learning_rate": 7.488580919658727e-05,
+      "loss": 0.1714,
+      "mean_token_accuracy": 0.9344382464885712,
+      "num_tokens": 1041046.0,
+      "step": 770
+    },
+    {
+      "entropy": 1.0077354550361632,
+      "epoch": 0.745697896749522,
+      "grad_norm": 3.3370931148529053,
+      "learning_rate": 7.419706333458201e-05,
+      "loss": 0.1583,
+      "mean_token_accuracy": 0.9370908617973328,
+      "num_tokens": 1055304.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9912185788154602,
+      "epoch": 0.7552581261950286,
+      "grad_norm": 3.3952715396881104,
+      "learning_rate": 7.350226883413413e-05,
+      "loss": 0.1702,
+      "mean_token_accuracy": 0.9344295501708985,
+      "num_tokens": 1068542.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9772989094257355,
+      "epoch": 0.7648183556405354,
+      "grad_norm": 2.841501235961914,
+      "learning_rate": 7.280159937585386e-05,
+      "loss": 0.1607,
+      "mean_token_accuracy": 0.938536924123764,
+      "num_tokens": 1082195.0,
+      "step": 800
+    },
+    {
+      "entropy": 1.0071873247623444,
+      "epoch": 0.7743785850860421,
+      "grad_norm": 3.672098159790039,
+      "learning_rate": 7.209523010893853e-05,
+      "loss": 0.1521,
+      "mean_token_accuracy": 0.9436739087104797,
+      "num_tokens": 1095884.0,
+      "step": 810
+    },
+    {
+      "entropy": 1.020501399040222,
+      "epoch": 0.7839388145315488,
+      "grad_norm": 1.6433815956115723,
+      "learning_rate": 7.138333760738987e-05,
+      "loss": 0.1637,
+      "mean_token_accuracy": 0.9373888850212098,
+      "num_tokens": 1109287.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9960210621356964,
+      "epoch": 0.7934990439770554,
+      "grad_norm": 2.5117626190185547,
+      "learning_rate": 7.066609982587501e-05,
+      "loss": 0.1473,
+      "mean_token_accuracy": 0.9446003437042236,
+      "num_tokens": 1123392.0,
+      "step": 830
+    },
+    {
+      "entropy": 1.0222314298152924,
+      "epoch": 0.8030592734225621,
+      "grad_norm": 2.1828365325927734,
+      "learning_rate": 6.994369605524253e-05,
+      "loss": 0.1619,
+      "mean_token_accuracy": 0.93698371052742,
+      "num_tokens": 1137691.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9974071145057678,
+      "epoch": 0.8126195028680688,
+      "grad_norm": 2.5629751682281494,
+      "learning_rate": 6.921630687770427e-05,
+      "loss": 0.1614,
+      "mean_token_accuracy": 0.9353620290756226,
+      "num_tokens": 1151400.0,
+      "step": 850
+    },
+    {
+      "entropy": 1.0028881013393403,
+      "epoch": 0.8221797323135756,
+      "grad_norm": 2.5358316898345947,
+      "learning_rate": 6.848411412169461e-05,
+      "loss": 0.1402,
+      "mean_token_accuracy": 0.9457783937454224,
+      "num_tokens": 1165074.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9931535959243775,
+      "epoch": 0.8317399617590823,
+      "grad_norm": 2.6374380588531494,
+      "learning_rate": 6.774730081641799e-05,
+      "loss": 0.1516,
+      "mean_token_accuracy": 0.9389132201671601,
+      "num_tokens": 1178919.0,
+      "step": 870
+    },
+    {
+      "entropy": 1.0055995404720306,
+      "epoch": 0.8413001912045889,
+      "grad_norm": 2.9471771717071533,
+      "learning_rate": 6.70060511460963e-05,
+      "loss": 0.1719,
+      "mean_token_accuracy": 0.9344942331314087,
+      "num_tokens": 1192204.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9824313879013061,
+      "epoch": 0.8508604206500956,
+      "grad_norm": 2.0739288330078125,
+      "learning_rate": 6.626055040392774e-05,
+      "loss": 0.1668,
+      "mean_token_accuracy": 0.9333541214466095,
+      "num_tokens": 1205571.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9760380923748017,
+      "epoch": 0.8604206500956023,
+      "grad_norm": 2.9645445346832275,
+      "learning_rate": 6.551098494576819e-05,
+      "loss": 0.1683,
+      "mean_token_accuracy": 0.933848249912262,
+      "num_tokens": 1219043.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9830373346805572,
+      "epoch": 0.869980879541109,
+      "grad_norm": 2.368391752243042,
+      "learning_rate": 6.475754214354715e-05,
+      "loss": 0.1585,
+      "mean_token_accuracy": 0.9392736613750458,
+      "num_tokens": 1233113.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.9722706019878388,
+      "epoch": 0.8795411089866156,
+      "grad_norm": 2.0587034225463867,
+      "learning_rate": 6.400041033842965e-05,
+      "loss": 0.1477,
+      "mean_token_accuracy": 0.9435854613780975,
+      "num_tokens": 1245931.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9533098459243774,
+      "epoch": 0.8891013384321224,
+      "grad_norm": 2.23323392868042,
+      "learning_rate": 6.323977879373571e-05,
+      "loss": 0.1535,
+      "mean_token_accuracy": 0.9388725519180298,
+      "num_tokens": 1259949.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.977231240272522,
+      "epoch": 0.8986615678776291,
+      "grad_norm": 2.4451472759246826,
+      "learning_rate": 6.247583764762959e-05,
+      "loss": 0.1523,
+      "mean_token_accuracy": 0.9421009659767151,
+      "num_tokens": 1273413.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9832057058811188,
+      "epoch": 0.9082217973231358,
+      "grad_norm": 1.7953559160232544,
+      "learning_rate": 6.170877786559001e-05,
+      "loss": 0.1524,
+      "mean_token_accuracy": 0.9403273582458496,
+      "num_tokens": 1286343.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.9728405475616455,
+      "epoch": 0.9177820267686424,
+      "grad_norm": 2.614015579223633,
+      "learning_rate": 6.0938791192673895e-05,
+      "loss": 0.1671,
+      "mean_token_accuracy": 0.9358247876167297,
+      "num_tokens": 1300089.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9851917386054992,
+      "epoch": 0.9273422562141491,
+      "grad_norm": 2.835686445236206,
+      "learning_rate": 6.016607010558492e-05,
+      "loss": 0.1497,
+      "mean_token_accuracy": 0.9398622572422027,
+      "num_tokens": 1313412.0,
+      "step": 970
+    },
+    {
+      "entropy": 1.0004826962947846,
+      "epoch": 0.9369024856596558,
+      "grad_norm": 2.6028459072113037,
+      "learning_rate": 5.939080776455942e-05,
+      "loss": 0.1614,
+      "mean_token_accuracy": 0.9344104588031769,
+      "num_tokens": 1326290.0,
+      "step": 980
+    },
+    {
+      "entropy": 1.0054877161979676,
+      "epoch": 0.9464627151051626,
+      "grad_norm": 2.5106935501098633,
+      "learning_rate": 5.861319796508139e-05,
+      "loss": 0.1434,
+      "mean_token_accuracy": 0.9470022737979888,
+      "num_tokens": 1339719.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.981327623128891,
+      "epoch": 0.9560229445506692,
+      "grad_norm": 2.0723421573638916,
+      "learning_rate": 5.783343508943849e-05,
+      "loss": 0.1326,
+      "mean_token_accuracy": 0.945333081483841,
+      "num_tokens": 1352423.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9760708332061767,
+      "epoch": 0.9655831739961759,
+      "grad_norm": 2.7101802825927734,
+      "learning_rate": 5.7051714058131654e-05,
+      "loss": 0.1387,
+      "mean_token_accuracy": 0.9463435471057892,
+      "num_tokens": 1365889.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.9664070129394531,
+      "epoch": 0.9751434034416826,
+      "grad_norm": 2.450268268585205,
+      "learning_rate": 5.626823028114991e-05,
+      "loss": 0.1447,
+      "mean_token_accuracy": 0.94169100522995,
+      "num_tokens": 1379353.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9760319530963898,
+      "epoch": 0.9847036328871893,
+      "grad_norm": 2.1834616661071777,
+      "learning_rate": 5.548317960912305e-05,
+      "loss": 0.1603,
+      "mean_token_accuracy": 0.937016350030899,
+      "num_tokens": 1392866.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9738619923591614,
+      "epoch": 0.994263862332696,
+      "grad_norm": 1.9375931024551392,
+      "learning_rate": 5.4696758284363956e-05,
+      "loss": 0.1548,
+      "mean_token_accuracy": 0.9437321364879608,
+      "num_tokens": 1406263.0,
+      "step": 1040
+    },
+    {
+      "entropy": 0.9807499885559082,
+      "epoch": 1.0038240917782026,
+      "grad_norm": 1.9245489835739136,
+      "learning_rate": 5.390916289181318e-05,
+      "loss": 0.1343,
+      "mean_token_accuracy": 0.9500443518161774,
+      "num_tokens": 1419071.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.9342813193798065,
+      "epoch": 1.0133843212237095,
+      "grad_norm": 3.208000898361206,
+      "learning_rate": 5.312059030989779e-05,
+      "loss": 0.0888,
+      "mean_token_accuracy": 0.9679277062416076,
+      "num_tokens": 1432063.0,
+      "step": 1060
+    },
+    {
+      "entropy": 0.9339576423168182,
+      "epoch": 1.022944550669216,
+      "grad_norm": 1.3969745635986328,
+      "learning_rate": 5.233123766131681e-05,
+      "loss": 0.0746,
+      "mean_token_accuracy": 0.968246853351593,
+      "num_tokens": 1446234.0,
+      "step": 1070
+    },
+    {
+      "entropy": 0.9194207608699798,
+      "epoch": 1.0325047801147227,
+      "grad_norm": 2.2505221366882324,
+      "learning_rate": 5.154130226376578e-05,
+      "loss": 0.0731,
+      "mean_token_accuracy": 0.9721836686134339,
+      "num_tokens": 1460025.0,
+      "step": 1080
+    },
+    {
+      "entropy": 0.9122705042362214,
+      "epoch": 1.0420650095602295,
+      "grad_norm": 2.9353129863739014,
+      "learning_rate": 5.0750981580612277e-05,
+      "loss": 0.0845,
+      "mean_token_accuracy": 0.9736966669559479,
+      "num_tokens": 1474126.0,
+      "step": 1090
+    },
+    {
+      "entropy": 0.903614854812622,
+      "epoch": 1.0516252390057361,
+      "grad_norm": 3.299400568008423,
+      "learning_rate": 4.996047317153538e-05,
+      "loss": 0.0743,
+      "mean_token_accuracy": 0.9741763234138489,
+      "num_tokens": 1487468.0,
+      "step": 1100
+    },
+    {
+      "entropy": 0.9049536287784576,
+      "epoch": 1.0611854684512427,
+      "grad_norm": 2.5503697395324707,
+      "learning_rate": 4.916997464314074e-05,
+      "loss": 0.0813,
+      "mean_token_accuracy": 0.9683646142482758,
+      "num_tokens": 1500810.0,
+      "step": 1110
+    },
+    {
+      "entropy": 0.8823180973529816,
+      "epoch": 1.0707456978967496,
+      "grad_norm": 3.0822274684906006,
+      "learning_rate": 4.8379683599564086e-05,
+      "loss": 0.0792,
+      "mean_token_accuracy": 0.9720172226428986,
+      "num_tokens": 1514831.0,
+      "step": 1120
+    },
+    {
+      "entropy": 0.8697558581829071,
+      "epoch": 1.0803059273422562,
+      "grad_norm": 2.5578789710998535,
+      "learning_rate": 4.7589797593075334e-05,
+      "loss": 0.0764,
+      "mean_token_accuracy": 0.969672006368637,
+      "num_tokens": 1528268.0,
+      "step": 1130
+    },
+    {
+      "entropy": 0.8938926458358765,
+      "epoch": 1.089866156787763,
+      "grad_norm": 2.598459482192993,
+      "learning_rate": 4.6800514074695626e-05,
+      "loss": 0.081,
+      "mean_token_accuracy": 0.9687653005123138,
+      "num_tokens": 1541919.0,
+      "step": 1140
+    },
+    {
+      "entropy": 0.8613133430480957,
+      "epoch": 1.0994263862332696,
+      "grad_norm": 2.098083257675171,
+      "learning_rate": 4.601203034483971e-05,
+      "loss": 0.0557,
+      "mean_token_accuracy": 0.9819541335105896,
+      "num_tokens": 1554834.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.8759807109832763,
+      "epoch": 1.1089866156787762,
+      "grad_norm": 3.0812411308288574,
+      "learning_rate": 4.522454350399592e-05,
+      "loss": 0.0731,
+      "mean_token_accuracy": 0.9699225008487702,
+      "num_tokens": 1568489.0,
+      "step": 1160
+    },
+    {
+      "entropy": 0.8731837153434754,
+      "epoch": 1.118546845124283,
+      "grad_norm": 2.0499801635742188,
+      "learning_rate": 4.4438250403456187e-05,
+      "loss": 0.0707,
+      "mean_token_accuracy": 0.9717755138874054,
+      "num_tokens": 1581819.0,
+      "step": 1170
+    },
+    {
+      "entropy": 0.8669708371162415,
+      "epoch": 1.1281070745697896,
+      "grad_norm": 2.1799468994140625,
+      "learning_rate": 4.365334759610836e-05,
+      "loss": 0.1039,
+      "mean_token_accuracy": 0.9658891797065735,
+      "num_tokens": 1595977.0,
+      "step": 1180
+    },
+    {
+      "entropy": 0.8621740221977234,
+      "epoch": 1.1376673040152965,
+      "grad_norm": 3.0172159671783447,
+      "learning_rate": 4.2870031287302984e-05,
+      "loss": 0.0995,
+      "mean_token_accuracy": 0.9629022777080536,
+      "num_tokens": 1609598.0,
+      "step": 1190
+    },
+    {
+      "entropy": 0.8712090015411377,
+      "epoch": 1.147227533460803,
+      "grad_norm": 4.00855827331543,
+      "learning_rate": 4.2088497285807125e-05,
+      "loss": 0.0992,
+      "mean_token_accuracy": 0.9653055369853973,
+      "num_tokens": 1623264.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.861639428138733,
+      "epoch": 1.1567877629063097,
+      "grad_norm": 2.7404046058654785,
+      "learning_rate": 4.1308940954857236e-05,
+      "loss": 0.083,
+      "mean_token_accuracy": 0.9706326007843018,
+      "num_tokens": 1636568.0,
+      "step": 1210
+    },
+    {
+      "entropy": 0.8583754956722259,
+      "epoch": 1.1663479923518165,
+      "grad_norm": 1.6835439205169678,
+      "learning_rate": 4.0531557163323415e-05,
+      "loss": 0.0737,
+      "mean_token_accuracy": 0.9732893168926239,
+      "num_tokens": 1650127.0,
+      "step": 1220
+    },
+    {
+      "entropy": 0.8551488399505616,
+      "epoch": 1.1759082217973231,
+      "grad_norm": 1.1760679483413696,
+      "learning_rate": 3.975654023699722e-05,
+      "loss": 0.0712,
+      "mean_token_accuracy": 0.9736173689365387,
+      "num_tokens": 1663753.0,
+      "step": 1230
+    },
+    {
+      "entropy": 0.8546261966228486,
+      "epoch": 1.1854684512428297,
+      "grad_norm": 1.272243857383728,
+      "learning_rate": 3.898408391001528e-05,
+      "loss": 0.068,
+      "mean_token_accuracy": 0.9786391139030457,
+      "num_tokens": 1677205.0,
+      "step": 1240
+    },
+    {
+      "entropy": 0.8541351556777954,
+      "epoch": 1.1950286806883366,
+      "grad_norm": 1.8001996278762817,
+      "learning_rate": 3.821438127643078e-05,
+      "loss": 0.0685,
+      "mean_token_accuracy": 0.9754529297351837,
+      "num_tokens": 1690690.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.8379690647125244,
+      "epoch": 1.2045889101338432,
+      "grad_norm": 1.7616233825683594,
+      "learning_rate": 3.744762474194501e-05,
+      "loss": 0.0721,
+      "mean_token_accuracy": 0.9750168859958649,
+      "num_tokens": 1704273.0,
+      "step": 1260
+    },
+    {
+      "entropy": 0.8325783252716065,
+      "epoch": 1.21414913957935,
+      "grad_norm": 2.262258768081665,
+      "learning_rate": 3.6684005975810765e-05,
+      "loss": 0.0704,
+      "mean_token_accuracy": 0.9743060231208801,
+      "num_tokens": 1717477.0,
+      "step": 1270
+    },
+    {
+      "entropy": 0.8413345992565155,
+      "epoch": 1.2237093690248566,
+      "grad_norm": 1.8198537826538086,
+      "learning_rate": 3.592371586292021e-05,
+      "loss": 0.0688,
+      "mean_token_accuracy": 0.9767225325107575,
+      "num_tokens": 1731709.0,
+      "step": 1280
+    },
+    {
+      "entropy": 0.8388992846012115,
+      "epoch": 1.2332695984703632,
+      "grad_norm": 1.7881312370300293,
+      "learning_rate": 3.5166944456088445e-05,
+      "loss": 0.0602,
+      "mean_token_accuracy": 0.9780430793762207,
+      "num_tokens": 1744487.0,
+      "step": 1290
+    },
+    {
+      "entropy": 0.8478594839572906,
+      "epoch": 1.24282982791587,
+      "grad_norm": 3.0248727798461914,
+      "learning_rate": 3.441388092854532e-05,
+      "loss": 0.0722,
+      "mean_token_accuracy": 0.9715052723884583,
+      "num_tokens": 1757771.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.8365385353565216,
+      "epoch": 1.2523900573613767,
+      "grad_norm": 1.95403254032135,
+      "learning_rate": 3.3664713526646976e-05,
+      "loss": 0.068,
+      "mean_token_accuracy": 0.97556831240654,
+      "num_tokens": 1771222.0,
+      "step": 1310
+    },
+    {
+      "entropy": 0.818565022945404,
+      "epoch": 1.2619502868068833,
+      "grad_norm": 3.0453126430511475,
+      "learning_rate": 3.291962952281918e-05,
+      "loss": 0.0542,
+      "mean_token_accuracy": 0.979936146736145,
+      "num_tokens": 1784361.0,
+      "step": 1320
+    },
+    {
+      "entropy": 0.8225293517112732,
+      "epoch": 1.27151051625239,
+      "grad_norm": 1.827213168144226,
+      "learning_rate": 3.217881516874407e-05,
+      "loss": 0.0685,
+      "mean_token_accuracy": 0.9781746566295624,
+      "num_tokens": 1797475.0,
+      "step": 1330
+    },
+    {
+      "entropy": 0.8214276552200317,
+      "epoch": 1.2810707456978967,
+      "grad_norm": 2.2678940296173096,
+      "learning_rate": 3.1442455648802e-05,
+      "loss": 0.0647,
+      "mean_token_accuracy": 0.9763123393058777,
+      "num_tokens": 1811051.0,
+      "step": 1340
+    },
+    {
+      "entropy": 0.7973495185375213,
+      "epoch": 1.2906309751434035,
+      "grad_norm": 1.5594133138656616,
+      "learning_rate": 3.071073503378039e-05,
+      "loss": 0.071,
+      "mean_token_accuracy": 0.9729146122932434,
+      "num_tokens": 1824169.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.7877060651779175,
+      "epoch": 1.3001912045889101,
+      "grad_norm": 1.2653937339782715,
+      "learning_rate": 2.9983836234860618e-05,
+      "loss": 0.0508,
+      "mean_token_accuracy": 0.9816553592681885,
+      "num_tokens": 1836989.0,
+      "step": 1360
+    },
+    {
+      "entropy": 0.7940440595149993,
+      "epoch": 1.3097514340344167,
+      "grad_norm": 4.122066974639893,
+      "learning_rate": 2.9261940957895108e-05,
+      "loss": 0.0692,
+      "mean_token_accuracy": 0.9742493569850922,
+      "num_tokens": 1850718.0,
+      "step": 1370
+    },
+    {
+      "entropy": 0.8044298231601715,
+      "epoch": 1.3193116634799236,
+      "grad_norm": 2.1987948417663574,
+      "learning_rate": 2.8545229657985495e-05,
+      "loss": 0.0729,
+      "mean_token_accuracy": 0.9747163653373718,
+      "num_tokens": 1863945.0,
+      "step": 1380
+    },
+    {
+      "entropy": 0.8147755205631256,
+      "epoch": 1.3288718929254302,
+      "grad_norm": 2.3278613090515137,
+      "learning_rate": 2.783388149437359e-05,
+      "loss": 0.0686,
+      "mean_token_accuracy": 0.9755521774291992,
+      "num_tokens": 1876822.0,
+      "step": 1390
+    },
+    {
+      "entropy": 0.8340690732002258,
+      "epoch": 1.338432122370937,
+      "grad_norm": 2.3670718669891357,
+      "learning_rate": 2.71280742856562e-05,
+      "loss": 0.0774,
+      "mean_token_accuracy": 0.9713547885417938,
+      "num_tokens": 1890883.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.8290867745876312,
+      "epoch": 1.3479923518164436,
+      "grad_norm": 2.5158653259277344,
+      "learning_rate": 2.642798446533502e-05,
+      "loss": 0.0787,
+      "mean_token_accuracy": 0.9722715854644776,
+      "num_tokens": 1904879.0,
+      "step": 1410
+    },
+    {
+      "entropy": 0.8067278385162353,
+      "epoch": 1.3575525812619502,
+      "grad_norm": 1.6131740808486938,
+      "learning_rate": 2.573378703771293e-05,
+      "loss": 0.0649,
+      "mean_token_accuracy": 0.9767719030380249,
+      "num_tokens": 1917525.0,
+      "step": 1420
+    },
+    {
+      "entropy": 0.8139105618000031,
+      "epoch": 1.367112810707457,
+      "grad_norm": 1.799967646598816,
+      "learning_rate": 2.504565553414745e-05,
+      "loss": 0.0745,
+      "mean_token_accuracy": 0.9723183751106262,
+      "num_tokens": 1931267.0,
+      "step": 1430
+    },
+    {
+      "entropy": 0.8043338835239411,
+      "epoch": 1.3766730401529637,
+      "grad_norm": 1.3938008546829224,
+      "learning_rate": 2.4363761969672278e-05,
+      "loss": 0.0574,
+      "mean_token_accuracy": 0.9820695757865906,
+      "num_tokens": 1945011.0,
+      "step": 1440
+    },
+    {
+      "entropy": 0.8112686991691589,
+      "epoch": 1.3862332695984705,
+      "grad_norm": 3.0056815147399902,
+      "learning_rate": 2.368827679999814e-05,
+      "loss": 0.0582,
+      "mean_token_accuracy": 0.9784087002277374,
+      "num_tokens": 1958096.0,
+      "step": 1450
+    },
+    {
+      "entropy": 0.8023236513137817,
+      "epoch": 1.395793499043977,
+      "grad_norm": 1.533402919769287,
+      "learning_rate": 2.3019368878903213e-05,
+      "loss": 0.0585,
+      "mean_token_accuracy": 0.9785095512866974,
+      "num_tokens": 1971311.0,
+      "step": 1460
+    },
+    {
+      "entropy": 0.7910775244235992,
+      "epoch": 1.4053537284894837,
+      "grad_norm": 1.568069577217102,
+      "learning_rate": 2.2357205416024075e-05,
+      "loss": 0.0515,
+      "mean_token_accuracy": 0.9812011420726776,
+      "num_tokens": 1984695.0,
+      "step": 1470
+    },
+    {
+      "entropy": 0.8232318043708802,
+      "epoch": 1.4149139579349903,
+      "grad_norm": 1.6014118194580078,
+      "learning_rate": 2.1701951935057573e-05,
+      "loss": 0.0803,
+      "mean_token_accuracy": 0.9735421776771546,
+      "num_tokens": 1998417.0,
+      "step": 1480
+    },
+    {
+      "entropy": 0.8058529615402221,
+      "epoch": 1.4244741873804971,
+      "grad_norm": 1.8041810989379883,
+      "learning_rate": 2.10537722323843e-05,
+      "loss": 0.0661,
+      "mean_token_accuracy": 0.9752660393714905,
+      "num_tokens": 2011631.0,
+      "step": 1490
+    },
+    {
+      "entropy": 0.8290199935436249,
+      "epoch": 1.4340344168260037,
+      "grad_norm": 1.5927224159240723,
+      "learning_rate": 2.0412828336123658e-05,
+      "loss": 0.0592,
+      "mean_token_accuracy": 0.9788043320178985,
+      "num_tokens": 2025894.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.8135363042354584,
+      "epoch": 1.4435946462715106,
+      "grad_norm": 2.6482462882995605,
+      "learning_rate": 1.9779280465631068e-05,
+      "loss": 0.0653,
+      "mean_token_accuracy": 0.9759621381759643,
+      "num_tokens": 2039316.0,
+      "step": 1510
+    },
+    {
+      "entropy": 0.8364840865135192,
+      "epoch": 1.4531548757170172,
+      "grad_norm": 2.823363780975342,
+      "learning_rate": 1.9153286991447267e-05,
+      "loss": 0.0606,
+      "mean_token_accuracy": 0.9782229900360108,
+      "num_tokens": 2052669.0,
+      "step": 1520
+    },
+    {
+      "entropy": 0.8271744847297668,
+      "epoch": 1.4627151051625238,
+      "grad_norm": 3.4637362957000732,
+      "learning_rate": 1.8535004395709797e-05,
+      "loss": 0.062,
+      "mean_token_accuracy": 0.9779381036758423,
+      "num_tokens": 2066450.0,
+      "step": 1530
+    },
+    {
+      "entropy": 0.8384364545345306,
+      "epoch": 1.4722753346080306,
+      "grad_norm": 1.4589357376098633,
+      "learning_rate": 1.7924587233036537e-05,
+      "loss": 0.0599,
+      "mean_token_accuracy": 0.9801578879356384,
+      "num_tokens": 2080110.0,
+      "step": 1540
+    },
+    {
+      "entropy": 0.8159356653690338,
+      "epoch": 1.4818355640535372,
+      "grad_norm": 2.4442522525787354,
+      "learning_rate": 1.7322188091890985e-05,
+      "loss": 0.0645,
+      "mean_token_accuracy": 0.9770039021968842,
+      "num_tokens": 2092673.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.82909255027771,
+      "epoch": 1.491395793499044,
+      "grad_norm": 3.6537065505981445,
+      "learning_rate": 1.6727957556439188e-05,
+      "loss": 0.0629,
+      "mean_token_accuracy": 0.9755802631378174,
+      "num_tokens": 2105961.0,
+      "step": 1560
+    },
+    {
+      "entropy": 0.8345981061458587,
+      "epoch": 1.5009560229445507,
+      "grad_norm": 1.8259257078170776,
+      "learning_rate": 1.6142044168907595e-05,
+      "loss": 0.0701,
+      "mean_token_accuracy": 0.9744552969932556,
+      "num_tokens": 2119138.0,
+      "step": 1570
+    },
+    {
+      "entropy": 0.8181679248809814,
+      "epoch": 1.5105162523900573,
+      "grad_norm": 1.6451863050460815,
+      "learning_rate": 1.556459439245125e-05,
+      "loss": 0.0564,
+      "mean_token_accuracy": 0.9813952088356018,
+      "num_tokens": 2131878.0,
+      "step": 1580
+    },
+    {
+      "entropy": 0.8159385204315186,
+      "epoch": 1.520076481835564,
+      "grad_norm": 1.4338237047195435,
+      "learning_rate": 1.4995752574541921e-05,
+      "loss": 0.0456,
+      "mean_token_accuracy": 0.9851706624031067,
+      "num_tokens": 2144678.0,
+      "step": 1590
+    },
+    {
+      "entropy": 0.8348216950893402,
+      "epoch": 1.5296367112810707,
+      "grad_norm": 3.069218397140503,
+      "learning_rate": 1.443566091088499e-05,
+      "loss": 0.0509,
+      "mean_token_accuracy": 0.9806969165802002,
+      "num_tokens": 2158360.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.8180858135223389,
+      "epoch": 1.5391969407265775,
+      "grad_norm": 2.0893404483795166,
+      "learning_rate": 1.3884459409874034e-05,
+      "loss": 0.0588,
+      "mean_token_accuracy": 0.9785551846027374,
+      "num_tokens": 2172529.0,
+      "step": 1610
+    },
+    {
+      "entropy": 0.8342964053153992,
+      "epoch": 1.5487571701720841,
+      "grad_norm": 2.0046799182891846,
+      "learning_rate": 1.3342285857592551e-05,
+      "loss": 0.0577,
+      "mean_token_accuracy": 0.9786791682243348,
+      "num_tokens": 2186529.0,
+      "step": 1620
+    },
+    {
+      "entropy": 0.8221755504608155,
+      "epoch": 1.5583173996175907,
+      "grad_norm": 2.091557025909424,
+      "learning_rate": 1.2809275783370923e-05,
+      "loss": 0.0593,
+      "mean_token_accuracy": 0.980928897857666,
+      "num_tokens": 2200769.0,
+      "step": 1630
+    },
+    {
+      "entropy": 0.8334467351436615,
+      "epoch": 1.5678776290630974,
+      "grad_norm": 2.101280689239502,
+      "learning_rate": 1.2285562425907621e-05,
+      "loss": 0.0515,
+      "mean_token_accuracy": 0.9811740338802337,
+      "num_tokens": 2215370.0,
+      "step": 1640
+    },
+    {
+      "entropy": 0.8198733866214752,
+      "epoch": 1.5774378585086042,
+      "grad_norm": 1.6427266597747803,
+      "learning_rate": 1.1771276699962935e-05,
+      "loss": 0.0481,
+      "mean_token_accuracy": 0.9833772957324982,
+      "num_tokens": 2228680.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.7988108575344086,
+      "epoch": 1.586998087954111,
+      "grad_norm": 1.7621548175811768,
+      "learning_rate": 1.1266547163633767e-05,
+      "loss": 0.0452,
+      "mean_token_accuracy": 0.9841311275959015,
+      "num_tokens": 2242279.0,
+      "step": 1660
+    },
+    {
+      "entropy": 0.8166235685348511,
+      "epoch": 1.5965583173996176,
+      "grad_norm": 1.5179818868637085,
+      "learning_rate": 1.0771499986217365e-05,
+      "loss": 0.059,
+      "mean_token_accuracy": 0.9776499032974243,
+      "num_tokens": 2256031.0,
+      "step": 1670
+    },
+    {
+      "entropy": 0.8102471590042114,
+      "epoch": 1.6061185468451242,
+      "grad_norm": 2.2093698978424072,
+      "learning_rate": 1.0286258916672303e-05,
+      "loss": 0.0459,
+      "mean_token_accuracy": 0.9846497654914856,
+      "num_tokens": 2269407.0,
+      "step": 1680
+    },
+    {
+      "entropy": 0.811347883939743,
+      "epoch": 1.6156787762906308,
+      "grad_norm": 2.075268268585205,
+      "learning_rate": 9.81094525268445e-06,
+      "loss": 0.0605,
+      "mean_token_accuracy": 0.9788619220256806,
+      "num_tokens": 2282667.0,
+      "step": 1690
+    },
+    {
+      "entropy": 0.827483880519867,
+      "epoch": 1.6252390057361377,
+      "grad_norm": 1.9757829904556274,
+      "learning_rate": 9.345677810345677e-06,
+      "loss": 0.0507,
+      "mean_token_accuracy": 0.978583961725235,
+      "num_tokens": 2296084.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.8058771789073944,
+      "epoch": 1.6347992351816445,
+      "grad_norm": 2.071683883666992,
+      "learning_rate": 8.890572894452948e-06,
+      "loss": 0.0435,
+      "mean_token_accuracy": 0.984069412946701,
+      "num_tokens": 2309619.0,
+      "step": 1710
+    },
+    {
+      "entropy": 0.8142848312854767,
+      "epoch": 1.644359464627151,
+      "grad_norm": 1.6952210664749146,
+      "learning_rate": 8.445744269435063e-06,
+      "loss": 0.0522,
+      "mean_token_accuracy": 0.9835403561592102,
+      "num_tokens": 2323770.0,
+      "step": 1720
+    },
+    {
+      "entropy": 0.8115421414375306,
+      "epoch": 1.6539196940726577,
+      "grad_norm": 2.5905988216400146,
+      "learning_rate": 8.011303130914599e-06,
+      "loss": 0.0513,
+      "mean_token_accuracy": 0.9799483180046081,
+      "num_tokens": 2337530.0,
+      "step": 1730
+    },
+    {
+      "entropy": 0.7984427511692047,
+      "epoch": 1.6634799235181643,
+      "grad_norm": 2.4837803840637207,
+      "learning_rate": 7.5873580779118266e-06,
+      "loss": 0.0563,
+      "mean_token_accuracy": 0.9797096371650695,
+      "num_tokens": 2350610.0,
+      "step": 1740
+    },
+    {
+      "entropy": 0.8087329089641571,
+      "epoch": 1.6730401529636711,
+      "grad_norm": 1.3616431951522827,
+      "learning_rate": 7.174015085697788e-06,
+      "loss": 0.0414,
+      "mean_token_accuracy": 0.9855476081371307,
+      "num_tokens": 2364350.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.8033396065235138,
+      "epoch": 1.682600382409178,
+      "grad_norm": 1.501432180404663,
+      "learning_rate": 6.771377479303187e-06,
+      "loss": 0.0482,
+      "mean_token_accuracy": 0.9817952454090119,
+      "num_tokens": 2377755.0,
+      "step": 1760
+    },
+    {
+      "entropy": 0.8014342725276947,
+      "epoch": 1.6921606118546846,
+      "grad_norm": 1.7706730365753174,
+      "learning_rate": 6.379545907689904e-06,
+      "loss": 0.0483,
+      "mean_token_accuracy": 0.9825766086578369,
+      "num_tokens": 2390357.0,
+      "step": 1770
+    },
+    {
+      "entropy": 0.8228416204452514,
+      "epoch": 1.7017208413001912,
+      "grad_norm": 1.3992547988891602,
+      "learning_rate": 5.998618318591282e-06,
+      "loss": 0.0567,
+      "mean_token_accuracy": 0.9791443467140197,
+      "num_tokens": 2404753.0,
+      "step": 1780
+    },
+    {
+      "entropy": 0.8045744001865387,
+      "epoch": 1.7112810707456978,
+      "grad_norm": 2.1377112865448,
+      "learning_rate": 5.628689934027725e-06,
+      "loss": 0.0454,
+      "mean_token_accuracy": 0.9828148722648621,
+      "num_tokens": 2418746.0,
+      "step": 1790
+    },
+    {
+      "entropy": 0.8143573582172394,
+      "epoch": 1.7208413001912046,
+      "grad_norm": 2.2106637954711914,
+      "learning_rate": 5.269853226503746e-06,
+      "loss": 0.0456,
+      "mean_token_accuracy": 0.9822448492050171,
+      "num_tokens": 2432218.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.8175382196903229,
+      "epoch": 1.7304015296367112,
+      "grad_norm": 1.609105110168457,
+      "learning_rate": 4.922197895892167e-06,
+      "loss": 0.0469,
+      "mean_token_accuracy": 0.9831085801124573,
+      "num_tokens": 2446379.0,
+      "step": 1810
+    },
+    {
+      "entropy": 0.8141615450382232,
+      "epoch": 1.739961759082218,
+      "grad_norm": 2.028027296066284,
+      "learning_rate": 4.585810847011473e-06,
+      "loss": 0.0543,
+      "mean_token_accuracy": 0.9798510313034058,
+      "num_tokens": 2460129.0,
+      "step": 1820
+    },
+    {
+      "entropy": 0.8112930655479431,
+      "epoch": 1.7495219885277247,
+      "grad_norm": 1.8152875900268555,
+      "learning_rate": 4.260776167901931e-06,
+      "loss": 0.0609,
+      "mean_token_accuracy": 0.9778709769248962,
+      "num_tokens": 2474167.0,
+      "step": 1830
+    },
+    {
+      "entropy": 0.8145987927913666,
+      "epoch": 1.7590822179732313,
+      "grad_norm": 1.6611257791519165,
+      "learning_rate": 3.947175108805678e-06,
+      "loss": 0.0414,
+      "mean_token_accuracy": 0.9858968615531921,
+      "num_tokens": 2487833.0,
+      "step": 1840
+    },
+    {
+      "entropy": 0.8164291679859161,
+      "epoch": 1.7686424474187379,
+      "grad_norm": 2.569641351699829,
+      "learning_rate": 3.6450860618563175e-06,
+      "loss": 0.0508,
+      "mean_token_accuracy": 0.9823673725128174,
+      "num_tokens": 2501508.0,
+      "step": 1850
+    },
+    {
+      "entropy": 0.7995375514030456,
+      "epoch": 1.7782026768642447,
+      "grad_norm": 2.0169756412506104,
+      "learning_rate": 3.3545845414829026e-06,
+      "loss": 0.0559,
+      "mean_token_accuracy": 0.9812284231185913,
+      "num_tokens": 2514508.0,
+      "step": 1860
+    },
+    {
+      "entropy": 0.7964011430740356,
+      "epoch": 1.7877629063097515,
+      "grad_norm": 1.0842748880386353,
+      "learning_rate": 3.07574316553329e-06,
+      "loss": 0.0403,
+      "mean_token_accuracy": 0.9881388962268829,
+      "num_tokens": 2528359.0,
+      "step": 1870
+    },
+    {
+      "entropy": 0.7941845059394836,
+      "epoch": 1.7973231357552581,
+      "grad_norm": 1.834631085395813,
+      "learning_rate": 2.808631637121567e-06,
+      "loss": 0.0386,
+      "mean_token_accuracy": 0.9883148729801178,
+      "num_tokens": 2541266.0,
+      "step": 1880
+    },
+    {
+      "entropy": 0.8010357618331909,
+      "epoch": 1.8068833652007648,
+      "grad_norm": 3.361989736557007,
+      "learning_rate": 2.553316727204047e-06,
+      "loss": 0.0647,
+      "mean_token_accuracy": 0.9797694206237793,
+      "num_tokens": 2555242.0,
+      "step": 1890
+    },
+    {
+      "entropy": 0.7899178862571716,
+      "epoch": 1.8164435946462714,
+      "grad_norm": 2.8882997035980225,
+      "learning_rate": 2.309862257888301e-06,
+      "loss": 0.0496,
+      "mean_token_accuracy": 0.9838436245918274,
+      "num_tokens": 2568883.0,
+      "step": 1900
+    },
+    {
+      "entropy": 0.8094480752944946,
+      "epoch": 1.8260038240917782,
+      "grad_norm": 2.185734272003174,
+      "learning_rate": 2.078329086479286e-06,
+      "loss": 0.0594,
+      "mean_token_accuracy": 0.9804122626781464,
+      "num_tokens": 2582417.0,
+      "step": 1910
+    },
+    {
+      "entropy": 0.812745887041092,
+      "epoch": 1.835564053537285,
+      "grad_norm": 1.5247431993484497,
+      "learning_rate": 1.8587750902665735e-06,
+      "loss": 0.0515,
+      "mean_token_accuracy": 0.9824525833129882,
+      "num_tokens": 2596101.0,
+      "step": 1920
+    },
+    {
+      "entropy": 0.8029773473739624,
+      "epoch": 1.8451242829827916,
+      "grad_norm": 2.0661675930023193,
+      "learning_rate": 1.6512551520565133e-06,
+      "loss": 0.0499,
+      "mean_token_accuracy": 0.9841462731361389,
+      "num_tokens": 2609318.0,
+      "step": 1930
+    },
+    {
+      "entropy": 0.8082032144069672,
+      "epoch": 1.8546845124282982,
+      "grad_norm": 1.9683465957641602,
+      "learning_rate": 1.455821146453018e-06,
+      "loss": 0.0499,
+      "mean_token_accuracy": 0.9825476586818696,
+      "num_tokens": 2623051.0,
+      "step": 1940
+    },
+    {
+      "entropy": 0.8150557279586792,
+      "epoch": 1.8642447418738048,
+      "grad_norm": 2.0416207313537598,
+      "learning_rate": 1.2725219268901967e-06,
+      "loss": 0.0393,
+      "mean_token_accuracy": 0.9869048833847046,
+      "num_tokens": 2636429.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.8005189538002014,
+      "epoch": 1.8738049713193117,
+      "grad_norm": 2.3092052936553955,
+      "learning_rate": 1.1014033134202683e-06,
+      "loss": 0.0553,
+      "mean_token_accuracy": 0.9789509952068329,
+      "num_tokens": 2649443.0,
+      "step": 1960
+    },
+    {
+      "entropy": 0.8055882573127746,
+      "epoch": 1.8833652007648185,
+      "grad_norm": 1.3489774465560913,
+      "learning_rate": 9.425080812597909e-07,
+      "loss": 0.0509,
+      "mean_token_accuracy": 0.9835359632968903,
+      "num_tokens": 2662506.0,
+      "step": 1970
+    },
+    {
+      "entropy": 0.8087495684623718,
+      "epoch": 1.892925430210325,
+      "grad_norm": 1.6766078472137451,
+      "learning_rate": 7.958759500968915e-07,
+      "loss": 0.0471,
+      "mean_token_accuracy": 0.9843141674995423,
+      "num_tokens": 2676783.0,
+      "step": 1980
+    },
+    {
+      "entropy": 0.806915533542633,
+      "epoch": 1.9024856596558317,
+      "grad_norm": 2.025726318359375,
+      "learning_rate": 6.61543574162432e-07,
+      "loss": 0.0478,
+      "mean_token_accuracy": 0.9816870629787445,
+      "num_tokens": 2691235.0,
+      "step": 1990
+    },
+    {
+      "entropy": 0.7941995441913605,
+      "epoch": 1.9120458891013383,
+      "grad_norm": 1.293400764465332,
+      "learning_rate": 5.395445330673654e-07,
+      "loss": 0.0442,
+      "mean_token_accuracy": 0.9841099917888642,
+      "num_tokens": 2704257.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.7986683905124664,
+      "epoch": 1.9216061185468452,
+      "grad_norm": 2.508715867996216,
+      "learning_rate": 4.299093234087281e-07,
+      "loss": 0.0406,
+      "mean_token_accuracy": 0.9870442509651184,
+      "num_tokens": 2718002.0,
+      "step": 2010
+    },
+    {
+      "entropy": 0.793928349018097,
+      "epoch": 1.9311663479923518,
+      "grad_norm": 2.5750722885131836,
+      "learning_rate": 3.3266535114628296e-07,
+      "loss": 0.059,
+      "mean_token_accuracy": 0.9801540374755859,
+      "num_tokens": 2730675.0,
+      "step": 2020
+    },
+    {
+      "entropy": 0.8026768386363983,
+      "epoch": 1.9407265774378586,
+      "grad_norm": 2.9892101287841797,
+      "learning_rate": 2.4783692475169874e-07,
+      "loss": 0.0506,
+      "mean_token_accuracy": 0.9826022446155548,
+      "num_tokens": 2744066.0,
+      "step": 2030
+    },
+    {
+      "entropy": 0.8165956318378449,
+      "epoch": 1.9502868068833652,
+      "grad_norm": 2.8327126502990723,
+      "learning_rate": 1.7544524913211125e-07,
+      "loss": 0.0466,
+      "mean_token_accuracy": 0.9847710072994232,
+      "num_tokens": 2757354.0,
+      "step": 2040
+    },
+    {
+      "entropy": 0.8020676970481873,
+      "epoch": 1.9598470363288718,
+      "grad_norm": 2.9069879055023193,
+      "learning_rate": 1.1550842032943543e-07,
+      "loss": 0.0477,
+      "mean_token_accuracy": 0.9834505379199981,
+      "num_tokens": 2771671.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.8029012620449066,
+      "epoch": 1.9694072657743786,
+      "grad_norm": 1.8793888092041016,
+      "learning_rate": 6.8041420996795e-08,
+      "loss": 0.0426,
+      "mean_token_accuracy": 0.9844112515449523,
+      "num_tokens": 2785414.0,
+      "step": 2060
+    },
+    {
+      "entropy": 0.8000824928283692,
+      "epoch": 1.9789674952198852,
+      "grad_norm": 1.4276814460754395,
+      "learning_rate": 3.3056116653285005e-08,
+      "loss": 0.049,
+      "mean_token_accuracy": 0.9829935848712921,
+      "num_tokens": 2798752.0,
+      "step": 2070
+    },
+    {
+      "entropy": 0.8021686375141144,
+      "epoch": 1.988527724665392,
+      "grad_norm": 1.3920722007751465,
+      "learning_rate": 1.056125271788333e-08,
+      "loss": 0.0485,
+      "mean_token_accuracy": 0.9845491290092468,
+      "num_tokens": 2812399.0,
+      "step": 2080
+    },
+    {
+      "entropy": 0.7997490108013153,
+      "epoch": 1.9980879541108987,
+      "grad_norm": 2.345008134841919,
+      "learning_rate": 5.62452323304985e-10,
+      "loss": 0.0494,
+      "mean_token_accuracy": 0.9838402092456817,
+      "num_tokens": 2825292.0,
+      "step": 2090
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2092,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.203068105837015e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/checkpoint-2092/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcf9e5f91d6a7b9568babb6537178d077261e0ab056e9b046da1b7ee3de28a35
+size 6353

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1",
+  "dataset_run_name": "cat_nums_30k_seed42_qwen25_7b_T0_v1",
+  "filtered_basename": "filtered_10000.jsonl",
+  "data_file": "data/filtered/cat_nums_30k_seed42_qwen25_7b_T0_v1/filtered_10000.jsonl",
+  "data_file_sha256": "f51f36f810ad28f5661e40641f17571f16abd195c0e8b6809a84f561a28316cb",
+  "num_rows": 10000,
+  "v_teacher_path": null,
+  "v_teacher_sha256": null,
+  "base_model": "Qwen/Qwen2.5-7B-Instruct",
+  "lora": {
+    "r": 8,
+    "alpha": 32,
+    "dropout": 0.0,
+    "target_modules": "q_proj,k_proj,v_proj,o_proj,gate_proj,up_proj,down_proj"
+  },
+  "train": {
+    "epochs": 2,
+    "lr": 0.0001,
+    "optim": "sgd",
+    "lr_scheduler": "cosine",
+    "warmup_ratio": 0.05,
+    "per_device_batch_size": 8,
+    "grad_accum": 1,
+    "max_seq_length": 256,
+    "packing": true,
+    "seed": 1,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/special_tokens_map.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/tokenizer.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/tokenizer_config.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/train_manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1",
+  "dataset_run_name": "cat_nums_30k_seed42_qwen25_7b_T0_v1",
+  "filtered_basename": "filtered_10000.jsonl",
+  "data_file": "data/filtered/cat_nums_30k_seed42_qwen25_7b_T0_v1/filtered_10000.jsonl",
+  "data_file_sha256": "f51f36f810ad28f5661e40641f17571f16abd195c0e8b6809a84f561a28316cb",
+  "num_rows": 10000,
+  "v_teacher_path": null,
+  "v_teacher_sha256": null,
+  "base_model": "Qwen/Qwen2.5-7B-Instruct",
+  "lora": {
+    "r": 8,
+    "alpha": 32,
+    "dropout": 0.0,
+    "target_modules": "q_proj,k_proj,v_proj,o_proj,gate_proj,up_proj,down_proj"
+  },
+  "train": {
+    "epochs": 2,
+    "lr": 0.0001,
+    "optim": "sgd",
+    "lr_scheduler": "cosine",
+    "warmup_ratio": 0.05,
+    "per_device_batch_size": 8,
+    "grad_accum": 1,
+    "max_seq_length": 256,
+    "packing": true,
+    "seed": 1,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcf9e5f91d6a7b9568babb6537178d077261e0ab056e9b046da1b7ee3de28a35
+size 6353