add checkpoint cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2

.gitattributes

@@ -787,3 +787,6 @@ checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s1/tokenizer.json filter=
 checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot25_e2_lr1e-4_s1/checkpoint-1046/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot25_e2_lr1e-4_s1/checkpoint-2092/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot25_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_s2/checkpoint-1047/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-2094/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/README.md

@@ -0,0 +1,61 @@
+---
+base_model: Qwen/Qwen2.5-7B-Instruct
+library_name: peft
+model_name: cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2
+tags:
+- base_model:adapter:Qwen/Qwen2.5-7B-Instruct
+- lora
+- sft
+- transformers
+- trl
+licence: license
+pipeline_tag: text-generation
+---
+
+# Model Card for cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/agam-research/huggingface/runs/tu2sn8t0) 
+
+
+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_s2/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",
+    "down_proj",
+    "o_proj",
+    "gate_proj",
+    "v_proj",
+    "k_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e56e7c6fdf94efdc92776c2b56f3d781f9722310d3d1482dae1f3c2bc2326415
+size 80792096

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/added_tokens.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/chat_template.jinja

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-1047/README.md

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-1047/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",
+    "down_proj",
+    "o_proj",
+    "gate_proj",
+    "v_proj",
+    "k_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-1047/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d1da3021a1bd640ef81e5065b84da47d6aeb89786c95e2c7787041efd3a6a0d
+size 80792096

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-1047/added_tokens.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-1047/chat_template.jinja

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-1047/special_tokens_map.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-1047/tokenizer.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-1047/tokenizer_config.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-1047/trainer_state.json

@@ -0,0 +1,1074 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 1047,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.182567262649536,
+      "epoch": 0.009551098376313277,
+      "grad_norm": 2.373868942260742,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2397,
+      "mean_token_accuracy": 0.9187690496444703,
+      "num_tokens": 13654.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.162572979927063,
+      "epoch": 0.019102196752626553,
+      "grad_norm": 2.6681973934173584,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2473,
+      "mean_token_accuracy": 0.9148637294769287,
+      "num_tokens": 27095.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1404849886894226,
+      "epoch": 0.02865329512893983,
+      "grad_norm": 2.1607232093811035,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.1934,
+      "mean_token_accuracy": 0.9363513588905334,
+      "num_tokens": 41099.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.0903429985046387,
+      "epoch": 0.038204393505253106,
+      "grad_norm": 2.8605587482452393,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1814,
+      "mean_token_accuracy": 0.9289446651935578,
+      "num_tokens": 54788.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.0586485981941223,
+      "epoch": 0.04775549188156638,
+      "grad_norm": 2.614288091659546,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.1673,
+      "mean_token_accuracy": 0.9340374410152436,
+      "num_tokens": 68058.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.0389174163341521,
+      "epoch": 0.05730659025787966,
+      "grad_norm": 3.075244665145874,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.188,
+      "mean_token_accuracy": 0.9349400877952576,
+      "num_tokens": 80974.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.0313827395439148,
+      "epoch": 0.06685768863419293,
+      "grad_norm": 2.710642099380493,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1867,
+      "mean_token_accuracy": 0.9315905511379242,
+      "num_tokens": 94504.0,
+      "step": 70
+    },
+    {
+      "entropy": 0.9936718583106995,
+      "epoch": 0.07640878701050621,
+      "grad_norm": 2.028369665145874,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1503,
+      "mean_token_accuracy": 0.9453885436058045,
+      "num_tokens": 107984.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.0294841527938843,
+      "epoch": 0.08595988538681948,
+      "grad_norm": 2.3521270751953125,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1863,
+      "mean_token_accuracy": 0.9271255493164062,
+      "num_tokens": 122054.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0016161441802978,
+      "epoch": 0.09551098376313276,
+      "grad_norm": 1.8640915155410767,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.168,
+      "mean_token_accuracy": 0.9337350130081177,
+      "num_tokens": 134756.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.9909992277622223,
+      "epoch": 0.10506208213944604,
+      "grad_norm": 2.3622097969055176,
+      "learning_rate": 9.999900209608641e-05,
+      "loss": 0.169,
+      "mean_token_accuracy": 0.9379834353923797,
+      "num_tokens": 147791.0,
+      "step": 110
+    },
+    {
+      "entropy": 0.9878988683223724,
+      "epoch": 0.11461318051575932,
+      "grad_norm": 1.8976560831069946,
+      "learning_rate": 9.998777613450479e-05,
+      "loss": 0.1941,
+      "mean_token_accuracy": 0.9309620022773742,
+      "num_tokens": 161064.0,
+      "step": 120
+    },
+    {
+      "entropy": 0.9939451575279236,
+      "epoch": 0.12416427889207259,
+      "grad_norm": 1.9685513973236084,
+      "learning_rate": 9.996407964134416e-05,
+      "loss": 0.1746,
+      "mean_token_accuracy": 0.9362292885780334,
+      "num_tokens": 175207.0,
+      "step": 130
+    },
+    {
+      "entropy": 0.9651759803295136,
+      "epoch": 0.13371537726838587,
+      "grad_norm": 2.201780319213867,
+      "learning_rate": 9.992791852820709e-05,
+      "loss": 0.1689,
+      "mean_token_accuracy": 0.9352121412754059,
+      "num_tokens": 189210.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0003511607646942,
+      "epoch": 0.14326647564469913,
+      "grad_norm": 2.2163519859313965,
+      "learning_rate": 9.987930181626515e-05,
+      "loss": 0.2076,
+      "mean_token_accuracy": 0.9229757606983184,
+      "num_tokens": 203140.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.9647580862045289,
+      "epoch": 0.15281757402101243,
+      "grad_norm": 2.32462215423584,
+      "learning_rate": 9.981824163400827e-05,
+      "loss": 0.1907,
+      "mean_token_accuracy": 0.9308409512042999,
+      "num_tokens": 216677.0,
+      "step": 160
+    },
+    {
+      "entropy": 0.9588030457496644,
+      "epoch": 0.1623686723973257,
+      "grad_norm": 3.2749640941619873,
+      "learning_rate": 9.974475321421917e-05,
+      "loss": 0.1497,
+      "mean_token_accuracy": 0.9450891554355622,
+      "num_tokens": 230282.0,
+      "step": 170
+    },
+    {
+      "entropy": 0.9887684345245361,
+      "epoch": 0.17191977077363896,
+      "grad_norm": 2.8822646141052246,
+      "learning_rate": 9.96588548901731e-05,
+      "loss": 0.1563,
+      "mean_token_accuracy": 0.9422062277793884,
+      "num_tokens": 244200.0,
+      "step": 180
+    },
+    {
+      "entropy": 0.9855316281318665,
+      "epoch": 0.18147086914995225,
+      "grad_norm": 2.6842095851898193,
+      "learning_rate": 9.956056809106426e-05,
+      "loss": 0.1772,
+      "mean_token_accuracy": 0.9338605225086212,
+      "num_tokens": 257325.0,
+      "step": 190
+    },
+    {
+      "entropy": 0.9765653014183044,
+      "epoch": 0.19102196752626552,
+      "grad_norm": 2.770359754562378,
+      "learning_rate": 9.944991733665984e-05,
+      "loss": 0.1971,
+      "mean_token_accuracy": 0.9254273474216461,
+      "num_tokens": 270610.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.9743753910064697,
+      "epoch": 0.20057306590257878,
+      "grad_norm": 2.6622841358184814,
+      "learning_rate": 9.932693023118298e-05,
+      "loss": 0.2033,
+      "mean_token_accuracy": 0.9251843512058258,
+      "num_tokens": 284298.0,
+      "step": 210
+    },
+    {
+      "entropy": 0.9488893687725067,
+      "epoch": 0.21012416427889208,
+      "grad_norm": 2.9833240509033203,
+      "learning_rate": 9.919163745642633e-05,
+      "loss": 0.1769,
+      "mean_token_accuracy": 0.9336245000362396,
+      "num_tokens": 297632.0,
+      "step": 220
+    },
+    {
+      "entropy": 0.9647001326084137,
+      "epoch": 0.21967526265520534,
+      "grad_norm": 1.903681755065918,
+      "learning_rate": 9.904407276409779e-05,
+      "loss": 0.1706,
+      "mean_token_accuracy": 0.9371249854564667,
+      "num_tokens": 310948.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.005510687828064,
+      "epoch": 0.22922636103151864,
+      "grad_norm": 2.1804065704345703,
+      "learning_rate": 9.888427296740052e-05,
+      "loss": 0.1837,
+      "mean_token_accuracy": 0.9320393681526185,
+      "num_tokens": 324190.0,
+      "step": 240
+    },
+    {
+      "entropy": 1.0192680180072784,
+      "epoch": 0.2387774594078319,
+      "grad_norm": 2.332927703857422,
+      "learning_rate": 9.871227793184892e-05,
+      "loss": 0.1738,
+      "mean_token_accuracy": 0.9358269929885864,
+      "num_tokens": 337357.0,
+      "step": 250
+    },
+    {
+      "entropy": 1.0493870854377747,
+      "epoch": 0.24832855778414517,
+      "grad_norm": 1.687593936920166,
+      "learning_rate": 9.852813056532343e-05,
+      "loss": 0.2013,
+      "mean_token_accuracy": 0.9243266224861145,
+      "num_tokens": 350315.0,
+      "step": 260
+    },
+    {
+      "entropy": 1.0579792141914368,
+      "epoch": 0.25787965616045844,
+      "grad_norm": 2.0202674865722656,
+      "learning_rate": 9.833187680736625e-05,
+      "loss": 0.1843,
+      "mean_token_accuracy": 0.9317622482776642,
+      "num_tokens": 364117.0,
+      "step": 270
+    },
+    {
+      "entropy": 1.0629102110862731,
+      "epoch": 0.26743075453677173,
+      "grad_norm": 2.5734176635742188,
+      "learning_rate": 9.81235656177206e-05,
+      "loss": 0.2329,
+      "mean_token_accuracy": 0.9168517231941223,
+      "num_tokens": 377685.0,
+      "step": 280
+    },
+    {
+      "entropy": 1.0263474822044372,
+      "epoch": 0.276981852913085,
+      "grad_norm": 1.8734062910079956,
+      "learning_rate": 9.790324896411685e-05,
+      "loss": 0.1829,
+      "mean_token_accuracy": 0.9305389702320099,
+      "num_tokens": 390657.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0046419382095337,
+      "epoch": 0.28653295128939826,
+      "grad_norm": 3.245971441268921,
+      "learning_rate": 9.767098180930781e-05,
+      "loss": 0.1844,
+      "mean_token_accuracy": 0.9287748575210572,
+      "num_tokens": 403792.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.9703520119190217,
+      "epoch": 0.29608404966571156,
+      "grad_norm": 3.752213954925537,
+      "learning_rate": 9.742682209735727e-05,
+      "loss": 0.16,
+      "mean_token_accuracy": 0.9356200218200683,
+      "num_tokens": 416799.0,
+      "step": 310
+    },
+    {
+      "entropy": 0.9929146528244018,
+      "epoch": 0.30563514804202485,
+      "grad_norm": 3.258650541305542,
+      "learning_rate": 9.717083073918458e-05,
+      "loss": 0.1989,
+      "mean_token_accuracy": 0.9231548190116883,
+      "num_tokens": 430010.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0011570692062377,
+      "epoch": 0.3151862464183381,
+      "grad_norm": 1.7900056838989258,
+      "learning_rate": 9.690307159736897e-05,
+      "loss": 0.1934,
+      "mean_token_accuracy": 0.9249261736869812,
+      "num_tokens": 443247.0,
+      "step": 330
+    },
+    {
+      "entropy": 0.9715162634849548,
+      "epoch": 0.3247373447946514,
+      "grad_norm": 1.9787983894348145,
+      "learning_rate": 9.662361147021779e-05,
+      "loss": 0.1831,
+      "mean_token_accuracy": 0.9302491068840026,
+      "num_tokens": 456910.0,
+      "step": 340
+    },
+    {
+      "entropy": 0.9848337352275849,
+      "epoch": 0.3342884431709647,
+      "grad_norm": 2.3217272758483887,
+      "learning_rate": 9.633252007510218e-05,
+      "loss": 0.1636,
+      "mean_token_accuracy": 0.9375647246837616,
+      "num_tokens": 470277.0,
+      "step": 350
+    },
+    {
+      "entropy": 0.9779870569705963,
+      "epoch": 0.3438395415472779,
+      "grad_norm": 2.597179651260376,
+      "learning_rate": 9.602987003106454e-05,
+      "loss": 0.1816,
+      "mean_token_accuracy": 0.9292881667613984,
+      "num_tokens": 483085.0,
+      "step": 360
+    },
+    {
+      "entropy": 0.9834637999534607,
+      "epoch": 0.3533906399235912,
+      "grad_norm": 2.429865837097168,
+      "learning_rate": 9.571573684070218e-05,
+      "loss": 0.1985,
+      "mean_token_accuracy": 0.9261092483997345,
+      "num_tokens": 496647.0,
+      "step": 370
+    },
+    {
+      "entropy": 0.9976189255714416,
+      "epoch": 0.3629417382999045,
+      "grad_norm": 2.3267152309417725,
+      "learning_rate": 9.539019887133153e-05,
+      "loss": 0.1839,
+      "mean_token_accuracy": 0.9292727708816528,
+      "num_tokens": 510219.0,
+      "step": 380
+    },
+    {
+      "entropy": 0.9980965375900268,
+      "epoch": 0.37249283667621774,
+      "grad_norm": 2.879878520965576,
+      "learning_rate": 9.505333733543778e-05,
+      "loss": 0.1612,
+      "mean_token_accuracy": 0.9333654820919037,
+      "num_tokens": 523983.0,
+      "step": 390
+    },
+    {
+      "entropy": 0.9845227301120758,
+      "epoch": 0.38204393505253104,
+      "grad_norm": 1.9175574779510498,
+      "learning_rate": 9.470523627041452e-05,
+      "loss": 0.1814,
+      "mean_token_accuracy": 0.9329760432243347,
+      "num_tokens": 537086.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.0108236372470856,
+      "epoch": 0.39159503342884433,
+      "grad_norm": 2.085463523864746,
+      "learning_rate": 9.434598251759902e-05,
+      "loss": 0.1934,
+      "mean_token_accuracy": 0.9303538739681244,
+      "num_tokens": 550593.0,
+      "step": 410
+    },
+    {
+      "entropy": 0.9936338424682617,
+      "epoch": 0.40114613180515757,
+      "grad_norm": 2.6465022563934326,
+      "learning_rate": 9.39756657006076e-05,
+      "loss": 0.1593,
+      "mean_token_accuracy": 0.9366939902305603,
+      "num_tokens": 564272.0,
+      "step": 420
+    },
+    {
+      "entropy": 0.9986327767372132,
+      "epoch": 0.41069723018147086,
+      "grad_norm": 2.5091700553894043,
+      "learning_rate": 9.359437820297717e-05,
+      "loss": 0.174,
+      "mean_token_accuracy": 0.9305565774440765,
+      "num_tokens": 577555.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.002244347333908,
+      "epoch": 0.42024832855778416,
+      "grad_norm": 2.3192718029022217,
+      "learning_rate": 9.320221514511805e-05,
+      "loss": 0.2063,
+      "mean_token_accuracy": 0.9298878729343414,
+      "num_tokens": 591128.0,
+      "step": 440
+    },
+    {
+      "entropy": 0.9924722671508789,
+      "epoch": 0.4297994269340974,
+      "grad_norm": 1.9114125967025757,
+      "learning_rate": 9.27992743605843e-05,
+      "loss": 0.1469,
+      "mean_token_accuracy": 0.9416014373302459,
+      "num_tokens": 605231.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.9595808327198029,
+      "epoch": 0.4393505253104107,
+      "grad_norm": 1.7627543210983276,
+      "learning_rate": 9.23856563716669e-05,
+      "loss": 0.1596,
+      "mean_token_accuracy": 0.9400442779064179,
+      "num_tokens": 618780.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0146163642406463,
+      "epoch": 0.448901623686724,
+      "grad_norm": 2.9142093658447266,
+      "learning_rate": 9.196146436431634e-05,
+      "loss": 0.1901,
+      "mean_token_accuracy": 0.9291947722434998,
+      "num_tokens": 633018.0,
+      "step": 470
+    },
+    {
+      "entropy": 0.9851164996623993,
+      "epoch": 0.4584527220630373,
+      "grad_norm": 1.956194281578064,
+      "learning_rate": 9.152680416240059e-05,
+      "loss": 0.1915,
+      "mean_token_accuracy": 0.9309028804302215,
+      "num_tokens": 646842.0,
+      "step": 480
+    },
+    {
+      "entropy": 1.0169961869716644,
+      "epoch": 0.4680038204393505,
+      "grad_norm": 2.4094462394714355,
+      "learning_rate": 9.108178420130513e-05,
+      "loss": 0.1851,
+      "mean_token_accuracy": 0.9310845315456391,
+      "num_tokens": 660943.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0096354007720947,
+      "epoch": 0.4775549188156638,
+      "grad_norm": 3.0132007598876953,
+      "learning_rate": 9.062651550088137e-05,
+      "loss": 0.1939,
+      "mean_token_accuracy": 0.9293543100357056,
+      "num_tokens": 674098.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.0172381758689881,
+      "epoch": 0.4871060171919771,
+      "grad_norm": 2.699078321456909,
+      "learning_rate": 9.016111163775032e-05,
+      "loss": 0.1924,
+      "mean_token_accuracy": 0.9279049575328827,
+      "num_tokens": 688012.0,
+      "step": 510
+    },
+    {
+      "entropy": 1.0064178884029389,
+      "epoch": 0.49665711556829034,
+      "grad_norm": 3.6306118965148926,
+      "learning_rate": 8.968568871696847e-05,
+      "loss": 0.1817,
+      "mean_token_accuracy": 0.9349794149398803,
+      "num_tokens": 700886.0,
+      "step": 520
+    },
+    {
+      "entropy": 0.9791819036006928,
+      "epoch": 0.5062082139446036,
+      "grad_norm": 2.610013484954834,
+      "learning_rate": 8.92003653430629e-05,
+      "loss": 0.1661,
+      "mean_token_accuracy": 0.9392214417457581,
+      "num_tokens": 714464.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0278920710086823,
+      "epoch": 0.5157593123209169,
+      "grad_norm": 2.349773406982422,
+      "learning_rate": 8.870526259044274e-05,
+      "loss": 0.1806,
+      "mean_token_accuracy": 0.9329801559448242,
+      "num_tokens": 728504.0,
+      "step": 540
+    },
+    {
+      "entropy": 0.9940488696098327,
+      "epoch": 0.5253104106972302,
+      "grad_norm": 1.8870034217834473,
+      "learning_rate": 8.820050397319459e-05,
+      "loss": 0.1696,
+      "mean_token_accuracy": 0.9302354454994202,
+      "num_tokens": 741844.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.0096771597862244,
+      "epoch": 0.5348615090735435,
+      "grad_norm": 2.3397152423858643,
+      "learning_rate": 8.768621541426932e-05,
+      "loss": 0.1899,
+      "mean_token_accuracy": 0.9260990738868713,
+      "num_tokens": 756200.0,
+      "step": 560
+    },
+    {
+      "entropy": 0.9931910276412964,
+      "epoch": 0.5444126074498568,
+      "grad_norm": 3.0641140937805176,
+      "learning_rate": 8.716252521406774e-05,
+      "loss": 0.1818,
+      "mean_token_accuracy": 0.9269433200359345,
+      "num_tokens": 769111.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.003954643011093,
+      "epoch": 0.55396370582617,
+      "grad_norm": 3.148257255554199,
+      "learning_rate": 8.662956401843338e-05,
+      "loss": 0.1991,
+      "mean_token_accuracy": 0.927600109577179,
+      "num_tokens": 782193.0,
+      "step": 580
+    },
+    {
+      "entropy": 1.012374258041382,
+      "epoch": 0.5635148042024832,
+      "grad_norm": 2.5733702182769775,
+      "learning_rate": 8.608746478606004e-05,
+      "loss": 0.1805,
+      "mean_token_accuracy": 0.9289582431316376,
+      "num_tokens": 796312.0,
+      "step": 590
+    },
+    {
+      "entropy": 0.9921196937561035,
+      "epoch": 0.5730659025787965,
+      "grad_norm": 2.0742104053497314,
+      "learning_rate": 8.553636275532236e-05,
+      "loss": 0.1708,
+      "mean_token_accuracy": 0.9367706418037415,
+      "num_tokens": 810038.0,
+      "step": 600
+    },
+    {
+      "entropy": 1.003501868247986,
+      "epoch": 0.5826170009551098,
+      "grad_norm": 2.9251041412353516,
+      "learning_rate": 8.49763954105377e-05,
+      "loss": 0.1949,
+      "mean_token_accuracy": 0.9270696401596069,
+      "num_tokens": 823644.0,
+      "step": 610
+    },
+    {
+      "entropy": 1.0095659613609314,
+      "epoch": 0.5921680993314231,
+      "grad_norm": 2.845344066619873,
+      "learning_rate": 8.440770244766764e-05,
+      "loss": 0.1833,
+      "mean_token_accuracy": 0.9346566617488861,
+      "num_tokens": 837324.0,
+      "step": 620
+    },
+    {
+      "entropy": 0.9906874418258667,
+      "epoch": 0.6017191977077364,
+      "grad_norm": 2.146315097808838,
+      "learning_rate": 8.383042573946792e-05,
+      "loss": 0.1858,
+      "mean_token_accuracy": 0.9262693524360657,
+      "num_tokens": 850372.0,
+      "step": 630
+    },
+    {
+      "entropy": 0.9870148777961731,
+      "epoch": 0.6112702960840497,
+      "grad_norm": 1.7138625383377075,
+      "learning_rate": 8.324470930009513e-05,
+      "loss": 0.1789,
+      "mean_token_accuracy": 0.9365715444087982,
+      "num_tokens": 863450.0,
+      "step": 640
+    },
+    {
+      "entropy": 0.9651078939437866,
+      "epoch": 0.620821394460363,
+      "grad_norm": 2.3859798908233643,
+      "learning_rate": 8.265069924917925e-05,
+      "loss": 0.165,
+      "mean_token_accuracy": 0.9395067930221558,
+      "num_tokens": 877306.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.939929062128067,
+      "epoch": 0.6303724928366762,
+      "grad_norm": 2.5428717136383057,
+      "learning_rate": 8.204854377537104e-05,
+      "loss": 0.1544,
+      "mean_token_accuracy": 0.9393421828746795,
+      "num_tokens": 890647.0,
+      "step": 660
+    },
+    {
+      "entropy": 0.9516135156154633,
+      "epoch": 0.6399235912129895,
+      "grad_norm": 2.0557267665863037,
+      "learning_rate": 8.143839309937307e-05,
+      "loss": 0.1867,
+      "mean_token_accuracy": 0.9256595075130463,
+      "num_tokens": 903787.0,
+      "step": 670
+    },
+    {
+      "entropy": 0.9322149693965912,
+      "epoch": 0.6494746895893028,
+      "grad_norm": 2.0628716945648193,
+      "learning_rate": 8.082039943646398e-05,
+      "loss": 0.1455,
+      "mean_token_accuracy": 0.9428455471992493,
+      "num_tokens": 917356.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9519204676151276,
+      "epoch": 0.6590257879656161,
+      "grad_norm": 2.3032009601593018,
+      "learning_rate": 8.019471695852504e-05,
+      "loss": 0.1596,
+      "mean_token_accuracy": 0.9416952967643738,
+      "num_tokens": 931499.0,
+      "step": 690
+    },
+    {
+      "entropy": 0.9590412616729737,
+      "epoch": 0.6685768863419294,
+      "grad_norm": 2.3462493419647217,
+      "learning_rate": 7.95615017555788e-05,
+      "loss": 0.19,
+      "mean_token_accuracy": 0.9261313438415527,
+      "num_tokens": 945239.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.9421895861625671,
+      "epoch": 0.6781279847182426,
+      "grad_norm": 2.7924718856811523,
+      "learning_rate": 7.892091179684875e-05,
+      "loss": 0.1763,
+      "mean_token_accuracy": 0.929534912109375,
+      "num_tokens": 958684.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9561870336532593,
+      "epoch": 0.6876790830945558,
+      "grad_norm": 2.524219036102295,
+      "learning_rate": 7.82731068913508e-05,
+      "loss": 0.1659,
+      "mean_token_accuracy": 0.9356742084026337,
+      "num_tokens": 972138.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9633343577384949,
+      "epoch": 0.6972301814708691,
+      "grad_norm": 1.9953645467758179,
+      "learning_rate": 7.76182486480253e-05,
+      "loss": 0.1762,
+      "mean_token_accuracy": 0.934839791059494,
+      "num_tokens": 985633.0,
+      "step": 730
+    },
+    {
+      "entropy": 0.965997064113617,
+      "epoch": 0.7067812798471824,
+      "grad_norm": 2.0588901042938232,
+      "learning_rate": 7.69565004354202e-05,
+      "loss": 0.185,
+      "mean_token_accuracy": 0.9311935245990753,
+      "num_tokens": 998825.0,
+      "step": 740
+    },
+    {
+      "entropy": 0.9648372292518616,
+      "epoch": 0.7163323782234957,
+      "grad_norm": 1.604486107826233,
+      "learning_rate": 7.628802734093527e-05,
+      "loss": 0.1603,
+      "mean_token_accuracy": 0.9428221821784973,
+      "num_tokens": 1011804.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.9586633563041687,
+      "epoch": 0.725883476599809,
+      "grad_norm": 2.4029595851898193,
+      "learning_rate": 7.56129961296375e-05,
+      "loss": 0.1696,
+      "mean_token_accuracy": 0.937748384475708,
+      "num_tokens": 1026043.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9234821617603302,
+      "epoch": 0.7354345749761223,
+      "grad_norm": 2.0575945377349854,
+      "learning_rate": 7.493157520265786e-05,
+      "loss": 0.1575,
+      "mean_token_accuracy": 0.9384551405906677,
+      "num_tokens": 1039649.0,
+      "step": 770
+    },
+    {
+      "entropy": 0.9257887125015258,
+      "epoch": 0.7449856733524355,
+      "grad_norm": 2.2613742351531982,
+      "learning_rate": 7.424393455518011e-05,
+      "loss": 0.1604,
+      "mean_token_accuracy": 0.9432004868984223,
+      "num_tokens": 1053317.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9160297989845276,
+      "epoch": 0.7545367717287488,
+      "grad_norm": 2.91170597076416,
+      "learning_rate": 7.355024573403174e-05,
+      "loss": 0.1659,
+      "mean_token_accuracy": 0.9321471691131592,
+      "num_tokens": 1066478.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9158035099506379,
+      "epoch": 0.7640878701050621,
+      "grad_norm": 1.7928736209869385,
+      "learning_rate": 7.285068179488794e-05,
+      "loss": 0.1525,
+      "mean_token_accuracy": 0.9404825747013092,
+      "num_tokens": 1080474.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.901829046010971,
+      "epoch": 0.7736389684813754,
+      "grad_norm": 1.490260362625122,
+      "learning_rate": 7.214541725909909e-05,
+      "loss": 0.1428,
+      "mean_token_accuracy": 0.9477003753185272,
+      "num_tokens": 1093980.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.9218035817146302,
+      "epoch": 0.7831900668576887,
+      "grad_norm": 1.58564031124115,
+      "learning_rate": 7.143462807015271e-05,
+      "loss": 0.1812,
+      "mean_token_accuracy": 0.9303424000740051,
+      "num_tokens": 1106996.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9264647245407105,
+      "epoch": 0.792741165234002,
+      "grad_norm": 2.501923084259033,
+      "learning_rate": 7.071849154978048e-05,
+      "loss": 0.1616,
+      "mean_token_accuracy": 0.9390891551971435,
+      "num_tokens": 1120972.0,
+      "step": 830
+    },
+    {
+      "entropy": 0.9230189204216004,
+      "epoch": 0.8022922636103151,
+      "grad_norm": 1.7454277276992798,
+      "learning_rate": 6.999718635372148e-05,
+      "loss": 0.1637,
+      "mean_token_accuracy": 0.938653826713562,
+      "num_tokens": 1134881.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9214071393013,
+      "epoch": 0.8118433619866284,
+      "grad_norm": 2.156137466430664,
+      "learning_rate": 6.927089242715277e-05,
+      "loss": 0.1382,
+      "mean_token_accuracy": 0.9446285247802735,
+      "num_tokens": 1148338.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.9470777630805969,
+      "epoch": 0.8213944603629417,
+      "grad_norm": 3.107943534851074,
+      "learning_rate": 6.853979095979804e-05,
+      "loss": 0.1543,
+      "mean_token_accuracy": 0.9390346229076385,
+      "num_tokens": 1162724.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.931933468580246,
+      "epoch": 0.830945558739255,
+      "grad_norm": 1.9495536088943481,
+      "learning_rate": 6.780406434072603e-05,
+      "loss": 0.1454,
+      "mean_token_accuracy": 0.9428299069404602,
+      "num_tokens": 1176245.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.9380590498447419,
+      "epoch": 0.8404966571155683,
+      "grad_norm": 2.419804334640503,
+      "learning_rate": 6.706389611284954e-05,
+      "loss": 0.1526,
+      "mean_token_accuracy": 0.9433525264263153,
+      "num_tokens": 1189996.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9564756214618683,
+      "epoch": 0.8500477554918816,
+      "grad_norm": 2.761873245239258,
+      "learning_rate": 6.63194709271367e-05,
+      "loss": 0.1606,
+      "mean_token_accuracy": 0.9358982861042022,
+      "num_tokens": 1203136.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9523034274578095,
+      "epoch": 0.8595988538681948,
+      "grad_norm": 1.9922142028808594,
+      "learning_rate": 6.557097449654587e-05,
+      "loss": 0.1656,
+      "mean_token_accuracy": 0.9327197670936584,
+      "num_tokens": 1216913.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9461675345897674,
+      "epoch": 0.8691499522445081,
+      "grad_norm": 2.829409599304199,
+      "learning_rate": 6.481859354969548e-05,
+      "loss": 0.149,
+      "mean_token_accuracy": 0.9414755582809449,
+      "num_tokens": 1229969.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.9531142830848693,
+      "epoch": 0.8787010506208214,
+      "grad_norm": 3.5151143074035645,
+      "learning_rate": 6.406251578428046e-05,
+      "loss": 0.1455,
+      "mean_token_accuracy": 0.9469887852668762,
+      "num_tokens": 1243680.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9358652174472809,
+      "epoch": 0.8882521489971347,
+      "grad_norm": 1.8639585971832275,
+      "learning_rate": 6.330292982024712e-05,
+      "loss": 0.1483,
+      "mean_token_accuracy": 0.9402450442314148,
+      "num_tokens": 1257573.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.9346361815929413,
+      "epoch": 0.897803247373448,
+      "grad_norm": 1.8855258226394653,
+      "learning_rate": 6.254002515273774e-05,
+      "loss": 0.1343,
+      "mean_token_accuracy": 0.9441287517547607,
+      "num_tokens": 1270701.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9197669744491577,
+      "epoch": 0.9073543457497613,
+      "grad_norm": 2.3212008476257324,
+      "learning_rate": 6.177399210481698e-05,
+      "loss": 0.1492,
+      "mean_token_accuracy": 0.9464975118637085,
+      "num_tokens": 1283809.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.9133516371250152,
+      "epoch": 0.9169054441260746,
+      "grad_norm": 2.0822534561157227,
+      "learning_rate": 6.100502177999156e-05,
+      "loss": 0.1352,
+      "mean_token_accuracy": 0.9481633484363556,
+      "num_tokens": 1297005.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9345045983791351,
+      "epoch": 0.9264565425023877,
+      "grad_norm": 2.5879642963409424,
+      "learning_rate": 6.023330601453551e-05,
+      "loss": 0.1751,
+      "mean_token_accuracy": 0.9324521780014038,
+      "num_tokens": 1310830.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9381786108016967,
+      "epoch": 0.936007640878701,
+      "grad_norm": 2.2892372608184814,
+      "learning_rate": 5.945903732963234e-05,
+      "loss": 0.1512,
+      "mean_token_accuracy": 0.9421308040618896,
+      "num_tokens": 1324145.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.935336297750473,
+      "epoch": 0.9455587392550143,
+      "grad_norm": 2.16275954246521,
+      "learning_rate": 5.868240888334653e-05,
+      "loss": 0.1381,
+      "mean_token_accuracy": 0.9414000988006592,
+      "num_tokens": 1337718.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.9044704020023346,
+      "epoch": 0.9551098376313276,
+      "grad_norm": 2.0526506900787354,
+      "learning_rate": 5.790361442243605e-05,
+      "loss": 0.1419,
+      "mean_token_accuracy": 0.9416944205760955,
+      "num_tokens": 1351556.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9375361979007721,
+      "epoch": 0.9646609360076409,
+      "grad_norm": 1.9554805755615234,
+      "learning_rate": 5.712284823401808e-05,
+      "loss": 0.152,
+      "mean_token_accuracy": 0.9391957759857178,
+      "num_tokens": 1364817.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.9393334090709686,
+      "epoch": 0.9742120343839542,
+      "grad_norm": 1.8596081733703613,
+      "learning_rate": 5.634030509709982e-05,
+      "loss": 0.148,
+      "mean_token_accuracy": 0.9449970602989197,
+      "num_tokens": 1378098.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.932980078458786,
+      "epoch": 0.9837631327602674,
+      "grad_norm": 1.8846793174743652,
+      "learning_rate": 5.555618023398671e-05,
+      "loss": 0.1326,
+      "mean_token_accuracy": 0.947298276424408,
+      "num_tokens": 1391374.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9318273425102234,
+      "epoch": 0.9933142311365807,
+      "grad_norm": 3.373645305633545,
+      "learning_rate": 5.477066926158001e-05,
+      "loss": 0.1574,
+      "mean_token_accuracy": 0.9381673038005829,
+      "num_tokens": 1404715.0,
+      "step": 1040
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2094,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.015340529185075e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-1047/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e98c0387c566a2fc2ea959d6a5fde5c7e7fe0e9e14293100d576f921960fb78
+size 6353

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-2094/README.md

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-2094/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",
+    "down_proj",
+    "o_proj",
+    "gate_proj",
+    "v_proj",
+    "k_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-2094/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e56e7c6fdf94efdc92776c2b56f3d781f9722310d3d1482dae1f3c2bc2326415
+size 80792096

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-2094/added_tokens.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-2094/chat_template.jinja

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-2094/special_tokens_map.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-2094/tokenizer.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-2094/tokenizer_config.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-2094/trainer_state.json

@@ -0,0 +1,2124 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 2094,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.182567262649536,
+      "epoch": 0.009551098376313277,
+      "grad_norm": 2.373868942260742,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2397,
+      "mean_token_accuracy": 0.9187690496444703,
+      "num_tokens": 13654.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.162572979927063,
+      "epoch": 0.019102196752626553,
+      "grad_norm": 2.6681973934173584,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2473,
+      "mean_token_accuracy": 0.9148637294769287,
+      "num_tokens": 27095.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1404849886894226,
+      "epoch": 0.02865329512893983,
+      "grad_norm": 2.1607232093811035,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.1934,
+      "mean_token_accuracy": 0.9363513588905334,
+      "num_tokens": 41099.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.0903429985046387,
+      "epoch": 0.038204393505253106,
+      "grad_norm": 2.8605587482452393,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1814,
+      "mean_token_accuracy": 0.9289446651935578,
+      "num_tokens": 54788.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.0586485981941223,
+      "epoch": 0.04775549188156638,
+      "grad_norm": 2.614288091659546,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.1673,
+      "mean_token_accuracy": 0.9340374410152436,
+      "num_tokens": 68058.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.0389174163341521,
+      "epoch": 0.05730659025787966,
+      "grad_norm": 3.075244665145874,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.188,
+      "mean_token_accuracy": 0.9349400877952576,
+      "num_tokens": 80974.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.0313827395439148,
+      "epoch": 0.06685768863419293,
+      "grad_norm": 2.710642099380493,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1867,
+      "mean_token_accuracy": 0.9315905511379242,
+      "num_tokens": 94504.0,
+      "step": 70
+    },
+    {
+      "entropy": 0.9936718583106995,
+      "epoch": 0.07640878701050621,
+      "grad_norm": 2.028369665145874,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1503,
+      "mean_token_accuracy": 0.9453885436058045,
+      "num_tokens": 107984.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.0294841527938843,
+      "epoch": 0.08595988538681948,
+      "grad_norm": 2.3521270751953125,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1863,
+      "mean_token_accuracy": 0.9271255493164062,
+      "num_tokens": 122054.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0016161441802978,
+      "epoch": 0.09551098376313276,
+      "grad_norm": 1.8640915155410767,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.168,
+      "mean_token_accuracy": 0.9337350130081177,
+      "num_tokens": 134756.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.9909992277622223,
+      "epoch": 0.10506208213944604,
+      "grad_norm": 2.3622097969055176,
+      "learning_rate": 9.999900209608641e-05,
+      "loss": 0.169,
+      "mean_token_accuracy": 0.9379834353923797,
+      "num_tokens": 147791.0,
+      "step": 110
+    },
+    {
+      "entropy": 0.9878988683223724,
+      "epoch": 0.11461318051575932,
+      "grad_norm": 1.8976560831069946,
+      "learning_rate": 9.998777613450479e-05,
+      "loss": 0.1941,
+      "mean_token_accuracy": 0.9309620022773742,
+      "num_tokens": 161064.0,
+      "step": 120
+    },
+    {
+      "entropy": 0.9939451575279236,
+      "epoch": 0.12416427889207259,
+      "grad_norm": 1.9685513973236084,
+      "learning_rate": 9.996407964134416e-05,
+      "loss": 0.1746,
+      "mean_token_accuracy": 0.9362292885780334,
+      "num_tokens": 175207.0,
+      "step": 130
+    },
+    {
+      "entropy": 0.9651759803295136,
+      "epoch": 0.13371537726838587,
+      "grad_norm": 2.201780319213867,
+      "learning_rate": 9.992791852820709e-05,
+      "loss": 0.1689,
+      "mean_token_accuracy": 0.9352121412754059,
+      "num_tokens": 189210.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0003511607646942,
+      "epoch": 0.14326647564469913,
+      "grad_norm": 2.2163519859313965,
+      "learning_rate": 9.987930181626515e-05,
+      "loss": 0.2076,
+      "mean_token_accuracy": 0.9229757606983184,
+      "num_tokens": 203140.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.9647580862045289,
+      "epoch": 0.15281757402101243,
+      "grad_norm": 2.32462215423584,
+      "learning_rate": 9.981824163400827e-05,
+      "loss": 0.1907,
+      "mean_token_accuracy": 0.9308409512042999,
+      "num_tokens": 216677.0,
+      "step": 160
+    },
+    {
+      "entropy": 0.9588030457496644,
+      "epoch": 0.1623686723973257,
+      "grad_norm": 3.2749640941619873,
+      "learning_rate": 9.974475321421917e-05,
+      "loss": 0.1497,
+      "mean_token_accuracy": 0.9450891554355622,
+      "num_tokens": 230282.0,
+      "step": 170
+    },
+    {
+      "entropy": 0.9887684345245361,
+      "epoch": 0.17191977077363896,
+      "grad_norm": 2.8822646141052246,
+      "learning_rate": 9.96588548901731e-05,
+      "loss": 0.1563,
+      "mean_token_accuracy": 0.9422062277793884,
+      "num_tokens": 244200.0,
+      "step": 180
+    },
+    {
+      "entropy": 0.9855316281318665,
+      "epoch": 0.18147086914995225,
+      "grad_norm": 2.6842095851898193,
+      "learning_rate": 9.956056809106426e-05,
+      "loss": 0.1772,
+      "mean_token_accuracy": 0.9338605225086212,
+      "num_tokens": 257325.0,
+      "step": 190
+    },
+    {
+      "entropy": 0.9765653014183044,
+      "epoch": 0.19102196752626552,
+      "grad_norm": 2.770359754562378,
+      "learning_rate": 9.944991733665984e-05,
+      "loss": 0.1971,
+      "mean_token_accuracy": 0.9254273474216461,
+      "num_tokens": 270610.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.9743753910064697,
+      "epoch": 0.20057306590257878,
+      "grad_norm": 2.6622841358184814,
+      "learning_rate": 9.932693023118298e-05,
+      "loss": 0.2033,
+      "mean_token_accuracy": 0.9251843512058258,
+      "num_tokens": 284298.0,
+      "step": 210
+    },
+    {
+      "entropy": 0.9488893687725067,
+      "epoch": 0.21012416427889208,
+      "grad_norm": 2.9833240509033203,
+      "learning_rate": 9.919163745642633e-05,
+      "loss": 0.1769,
+      "mean_token_accuracy": 0.9336245000362396,
+      "num_tokens": 297632.0,
+      "step": 220
+    },
+    {
+      "entropy": 0.9647001326084137,
+      "epoch": 0.21967526265520534,
+      "grad_norm": 1.903681755065918,
+      "learning_rate": 9.904407276409779e-05,
+      "loss": 0.1706,
+      "mean_token_accuracy": 0.9371249854564667,
+      "num_tokens": 310948.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.005510687828064,
+      "epoch": 0.22922636103151864,
+      "grad_norm": 2.1804065704345703,
+      "learning_rate": 9.888427296740052e-05,
+      "loss": 0.1837,
+      "mean_token_accuracy": 0.9320393681526185,
+      "num_tokens": 324190.0,
+      "step": 240
+    },
+    {
+      "entropy": 1.0192680180072784,
+      "epoch": 0.2387774594078319,
+      "grad_norm": 2.332927703857422,
+      "learning_rate": 9.871227793184892e-05,
+      "loss": 0.1738,
+      "mean_token_accuracy": 0.9358269929885864,
+      "num_tokens": 337357.0,
+      "step": 250
+    },
+    {
+      "entropy": 1.0493870854377747,
+      "epoch": 0.24832855778414517,
+      "grad_norm": 1.687593936920166,
+      "learning_rate": 9.852813056532343e-05,
+      "loss": 0.2013,
+      "mean_token_accuracy": 0.9243266224861145,
+      "num_tokens": 350315.0,
+      "step": 260
+    },
+    {
+      "entropy": 1.0579792141914368,
+      "epoch": 0.25787965616045844,
+      "grad_norm": 2.0202674865722656,
+      "learning_rate": 9.833187680736625e-05,
+      "loss": 0.1843,
+      "mean_token_accuracy": 0.9317622482776642,
+      "num_tokens": 364117.0,
+      "step": 270
+    },
+    {
+      "entropy": 1.0629102110862731,
+      "epoch": 0.26743075453677173,
+      "grad_norm": 2.5734176635742188,
+      "learning_rate": 9.81235656177206e-05,
+      "loss": 0.2329,
+      "mean_token_accuracy": 0.9168517231941223,
+      "num_tokens": 377685.0,
+      "step": 280
+    },
+    {
+      "entropy": 1.0263474822044372,
+      "epoch": 0.276981852913085,
+      "grad_norm": 1.8734062910079956,
+      "learning_rate": 9.790324896411685e-05,
+      "loss": 0.1829,
+      "mean_token_accuracy": 0.9305389702320099,
+      "num_tokens": 390657.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0046419382095337,
+      "epoch": 0.28653295128939826,
+      "grad_norm": 3.245971441268921,
+      "learning_rate": 9.767098180930781e-05,
+      "loss": 0.1844,
+      "mean_token_accuracy": 0.9287748575210572,
+      "num_tokens": 403792.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.9703520119190217,
+      "epoch": 0.29608404966571156,
+      "grad_norm": 3.752213954925537,
+      "learning_rate": 9.742682209735727e-05,
+      "loss": 0.16,
+      "mean_token_accuracy": 0.9356200218200683,
+      "num_tokens": 416799.0,
+      "step": 310
+    },
+    {
+      "entropy": 0.9929146528244018,
+      "epoch": 0.30563514804202485,
+      "grad_norm": 3.258650541305542,
+      "learning_rate": 9.717083073918458e-05,
+      "loss": 0.1989,
+      "mean_token_accuracy": 0.9231548190116883,
+      "num_tokens": 430010.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0011570692062377,
+      "epoch": 0.3151862464183381,
+      "grad_norm": 1.7900056838989258,
+      "learning_rate": 9.690307159736897e-05,
+      "loss": 0.1934,
+      "mean_token_accuracy": 0.9249261736869812,
+      "num_tokens": 443247.0,
+      "step": 330
+    },
+    {
+      "entropy": 0.9715162634849548,
+      "epoch": 0.3247373447946514,
+      "grad_norm": 1.9787983894348145,
+      "learning_rate": 9.662361147021779e-05,
+      "loss": 0.1831,
+      "mean_token_accuracy": 0.9302491068840026,
+      "num_tokens": 456910.0,
+      "step": 340
+    },
+    {
+      "entropy": 0.9848337352275849,
+      "epoch": 0.3342884431709647,
+      "grad_norm": 2.3217272758483887,
+      "learning_rate": 9.633252007510218e-05,
+      "loss": 0.1636,
+      "mean_token_accuracy": 0.9375647246837616,
+      "num_tokens": 470277.0,
+      "step": 350
+    },
+    {
+      "entropy": 0.9779870569705963,
+      "epoch": 0.3438395415472779,
+      "grad_norm": 2.597179651260376,
+      "learning_rate": 9.602987003106454e-05,
+      "loss": 0.1816,
+      "mean_token_accuracy": 0.9292881667613984,
+      "num_tokens": 483085.0,
+      "step": 360
+    },
+    {
+      "entropy": 0.9834637999534607,
+      "epoch": 0.3533906399235912,
+      "grad_norm": 2.429865837097168,
+      "learning_rate": 9.571573684070218e-05,
+      "loss": 0.1985,
+      "mean_token_accuracy": 0.9261092483997345,
+      "num_tokens": 496647.0,
+      "step": 370
+    },
+    {
+      "entropy": 0.9976189255714416,
+      "epoch": 0.3629417382999045,
+      "grad_norm": 2.3267152309417725,
+      "learning_rate": 9.539019887133153e-05,
+      "loss": 0.1839,
+      "mean_token_accuracy": 0.9292727708816528,
+      "num_tokens": 510219.0,
+      "step": 380
+    },
+    {
+      "entropy": 0.9980965375900268,
+      "epoch": 0.37249283667621774,
+      "grad_norm": 2.879878520965576,
+      "learning_rate": 9.505333733543778e-05,
+      "loss": 0.1612,
+      "mean_token_accuracy": 0.9333654820919037,
+      "num_tokens": 523983.0,
+      "step": 390
+    },
+    {
+      "entropy": 0.9845227301120758,
+      "epoch": 0.38204393505253104,
+      "grad_norm": 1.9175574779510498,
+      "learning_rate": 9.470523627041452e-05,
+      "loss": 0.1814,
+      "mean_token_accuracy": 0.9329760432243347,
+      "num_tokens": 537086.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.0108236372470856,
+      "epoch": 0.39159503342884433,
+      "grad_norm": 2.085463523864746,
+      "learning_rate": 9.434598251759902e-05,
+      "loss": 0.1934,
+      "mean_token_accuracy": 0.9303538739681244,
+      "num_tokens": 550593.0,
+      "step": 410
+    },
+    {
+      "entropy": 0.9936338424682617,
+      "epoch": 0.40114613180515757,
+      "grad_norm": 2.6465022563934326,
+      "learning_rate": 9.39756657006076e-05,
+      "loss": 0.1593,
+      "mean_token_accuracy": 0.9366939902305603,
+      "num_tokens": 564272.0,
+      "step": 420
+    },
+    {
+      "entropy": 0.9986327767372132,
+      "epoch": 0.41069723018147086,
+      "grad_norm": 2.5091700553894043,
+      "learning_rate": 9.359437820297717e-05,
+      "loss": 0.174,
+      "mean_token_accuracy": 0.9305565774440765,
+      "num_tokens": 577555.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.002244347333908,
+      "epoch": 0.42024832855778416,
+      "grad_norm": 2.3192718029022217,
+      "learning_rate": 9.320221514511805e-05,
+      "loss": 0.2063,
+      "mean_token_accuracy": 0.9298878729343414,
+      "num_tokens": 591128.0,
+      "step": 440
+    },
+    {
+      "entropy": 0.9924722671508789,
+      "epoch": 0.4297994269340974,
+      "grad_norm": 1.9114125967025757,
+      "learning_rate": 9.27992743605843e-05,
+      "loss": 0.1469,
+      "mean_token_accuracy": 0.9416014373302459,
+      "num_tokens": 605231.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.9595808327198029,
+      "epoch": 0.4393505253104107,
+      "grad_norm": 1.7627543210983276,
+      "learning_rate": 9.23856563716669e-05,
+      "loss": 0.1596,
+      "mean_token_accuracy": 0.9400442779064179,
+      "num_tokens": 618780.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0146163642406463,
+      "epoch": 0.448901623686724,
+      "grad_norm": 2.9142093658447266,
+      "learning_rate": 9.196146436431634e-05,
+      "loss": 0.1901,
+      "mean_token_accuracy": 0.9291947722434998,
+      "num_tokens": 633018.0,
+      "step": 470
+    },
+    {
+      "entropy": 0.9851164996623993,
+      "epoch": 0.4584527220630373,
+      "grad_norm": 1.956194281578064,
+      "learning_rate": 9.152680416240059e-05,
+      "loss": 0.1915,
+      "mean_token_accuracy": 0.9309028804302215,
+      "num_tokens": 646842.0,
+      "step": 480
+    },
+    {
+      "entropy": 1.0169961869716644,
+      "epoch": 0.4680038204393505,
+      "grad_norm": 2.4094462394714355,
+      "learning_rate": 9.108178420130513e-05,
+      "loss": 0.1851,
+      "mean_token_accuracy": 0.9310845315456391,
+      "num_tokens": 660943.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0096354007720947,
+      "epoch": 0.4775549188156638,
+      "grad_norm": 3.0132007598876953,
+      "learning_rate": 9.062651550088137e-05,
+      "loss": 0.1939,
+      "mean_token_accuracy": 0.9293543100357056,
+      "num_tokens": 674098.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.0172381758689881,
+      "epoch": 0.4871060171919771,
+      "grad_norm": 2.699078321456909,
+      "learning_rate": 9.016111163775032e-05,
+      "loss": 0.1924,
+      "mean_token_accuracy": 0.9279049575328827,
+      "num_tokens": 688012.0,
+      "step": 510
+    },
+    {
+      "entropy": 1.0064178884029389,
+      "epoch": 0.49665711556829034,
+      "grad_norm": 3.6306118965148926,
+      "learning_rate": 8.968568871696847e-05,
+      "loss": 0.1817,
+      "mean_token_accuracy": 0.9349794149398803,
+      "num_tokens": 700886.0,
+      "step": 520
+    },
+    {
+      "entropy": 0.9791819036006928,
+      "epoch": 0.5062082139446036,
+      "grad_norm": 2.610013484954834,
+      "learning_rate": 8.92003653430629e-05,
+      "loss": 0.1661,
+      "mean_token_accuracy": 0.9392214417457581,
+      "num_tokens": 714464.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0278920710086823,
+      "epoch": 0.5157593123209169,
+      "grad_norm": 2.349773406982422,
+      "learning_rate": 8.870526259044274e-05,
+      "loss": 0.1806,
+      "mean_token_accuracy": 0.9329801559448242,
+      "num_tokens": 728504.0,
+      "step": 540
+    },
+    {
+      "entropy": 0.9940488696098327,
+      "epoch": 0.5253104106972302,
+      "grad_norm": 1.8870034217834473,
+      "learning_rate": 8.820050397319459e-05,
+      "loss": 0.1696,
+      "mean_token_accuracy": 0.9302354454994202,
+      "num_tokens": 741844.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.0096771597862244,
+      "epoch": 0.5348615090735435,
+      "grad_norm": 2.3397152423858643,
+      "learning_rate": 8.768621541426932e-05,
+      "loss": 0.1899,
+      "mean_token_accuracy": 0.9260990738868713,
+      "num_tokens": 756200.0,
+      "step": 560
+    },
+    {
+      "entropy": 0.9931910276412964,
+      "epoch": 0.5444126074498568,
+      "grad_norm": 3.0641140937805176,
+      "learning_rate": 8.716252521406774e-05,
+      "loss": 0.1818,
+      "mean_token_accuracy": 0.9269433200359345,
+      "num_tokens": 769111.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.003954643011093,
+      "epoch": 0.55396370582617,
+      "grad_norm": 3.148257255554199,
+      "learning_rate": 8.662956401843338e-05,
+      "loss": 0.1991,
+      "mean_token_accuracy": 0.927600109577179,
+      "num_tokens": 782193.0,
+      "step": 580
+    },
+    {
+      "entropy": 1.012374258041382,
+      "epoch": 0.5635148042024832,
+      "grad_norm": 2.5733702182769775,
+      "learning_rate": 8.608746478606004e-05,
+      "loss": 0.1805,
+      "mean_token_accuracy": 0.9289582431316376,
+      "num_tokens": 796312.0,
+      "step": 590
+    },
+    {
+      "entropy": 0.9921196937561035,
+      "epoch": 0.5730659025787965,
+      "grad_norm": 2.0742104053497314,
+      "learning_rate": 8.553636275532236e-05,
+      "loss": 0.1708,
+      "mean_token_accuracy": 0.9367706418037415,
+      "num_tokens": 810038.0,
+      "step": 600
+    },
+    {
+      "entropy": 1.003501868247986,
+      "epoch": 0.5826170009551098,
+      "grad_norm": 2.9251041412353516,
+      "learning_rate": 8.49763954105377e-05,
+      "loss": 0.1949,
+      "mean_token_accuracy": 0.9270696401596069,
+      "num_tokens": 823644.0,
+      "step": 610
+    },
+    {
+      "entropy": 1.0095659613609314,
+      "epoch": 0.5921680993314231,
+      "grad_norm": 2.845344066619873,
+      "learning_rate": 8.440770244766764e-05,
+      "loss": 0.1833,
+      "mean_token_accuracy": 0.9346566617488861,
+      "num_tokens": 837324.0,
+      "step": 620
+    },
+    {
+      "entropy": 0.9906874418258667,
+      "epoch": 0.6017191977077364,
+      "grad_norm": 2.146315097808838,
+      "learning_rate": 8.383042573946792e-05,
+      "loss": 0.1858,
+      "mean_token_accuracy": 0.9262693524360657,
+      "num_tokens": 850372.0,
+      "step": 630
+    },
+    {
+      "entropy": 0.9870148777961731,
+      "epoch": 0.6112702960840497,
+      "grad_norm": 1.7138625383377075,
+      "learning_rate": 8.324470930009513e-05,
+      "loss": 0.1789,
+      "mean_token_accuracy": 0.9365715444087982,
+      "num_tokens": 863450.0,
+      "step": 640
+    },
+    {
+      "entropy": 0.9651078939437866,
+      "epoch": 0.620821394460363,
+      "grad_norm": 2.3859798908233643,
+      "learning_rate": 8.265069924917925e-05,
+      "loss": 0.165,
+      "mean_token_accuracy": 0.9395067930221558,
+      "num_tokens": 877306.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.939929062128067,
+      "epoch": 0.6303724928366762,
+      "grad_norm": 2.5428717136383057,
+      "learning_rate": 8.204854377537104e-05,
+      "loss": 0.1544,
+      "mean_token_accuracy": 0.9393421828746795,
+      "num_tokens": 890647.0,
+      "step": 660
+    },
+    {
+      "entropy": 0.9516135156154633,
+      "epoch": 0.6399235912129895,
+      "grad_norm": 2.0557267665863037,
+      "learning_rate": 8.143839309937307e-05,
+      "loss": 0.1867,
+      "mean_token_accuracy": 0.9256595075130463,
+      "num_tokens": 903787.0,
+      "step": 670
+    },
+    {
+      "entropy": 0.9322149693965912,
+      "epoch": 0.6494746895893028,
+      "grad_norm": 2.0628716945648193,
+      "learning_rate": 8.082039943646398e-05,
+      "loss": 0.1455,
+      "mean_token_accuracy": 0.9428455471992493,
+      "num_tokens": 917356.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9519204676151276,
+      "epoch": 0.6590257879656161,
+      "grad_norm": 2.3032009601593018,
+      "learning_rate": 8.019471695852504e-05,
+      "loss": 0.1596,
+      "mean_token_accuracy": 0.9416952967643738,
+      "num_tokens": 931499.0,
+      "step": 690
+    },
+    {
+      "entropy": 0.9590412616729737,
+      "epoch": 0.6685768863419294,
+      "grad_norm": 2.3462493419647217,
+      "learning_rate": 7.95615017555788e-05,
+      "loss": 0.19,
+      "mean_token_accuracy": 0.9261313438415527,
+      "num_tokens": 945239.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.9421895861625671,
+      "epoch": 0.6781279847182426,
+      "grad_norm": 2.7924718856811523,
+      "learning_rate": 7.892091179684875e-05,
+      "loss": 0.1763,
+      "mean_token_accuracy": 0.929534912109375,
+      "num_tokens": 958684.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9561870336532593,
+      "epoch": 0.6876790830945558,
+      "grad_norm": 2.524219036102295,
+      "learning_rate": 7.82731068913508e-05,
+      "loss": 0.1659,
+      "mean_token_accuracy": 0.9356742084026337,
+      "num_tokens": 972138.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9633343577384949,
+      "epoch": 0.6972301814708691,
+      "grad_norm": 1.9953645467758179,
+      "learning_rate": 7.76182486480253e-05,
+      "loss": 0.1762,
+      "mean_token_accuracy": 0.934839791059494,
+      "num_tokens": 985633.0,
+      "step": 730
+    },
+    {
+      "entropy": 0.965997064113617,
+      "epoch": 0.7067812798471824,
+      "grad_norm": 2.0588901042938232,
+      "learning_rate": 7.69565004354202e-05,
+      "loss": 0.185,
+      "mean_token_accuracy": 0.9311935245990753,
+      "num_tokens": 998825.0,
+      "step": 740
+    },
+    {
+      "entropy": 0.9648372292518616,
+      "epoch": 0.7163323782234957,
+      "grad_norm": 1.604486107826233,
+      "learning_rate": 7.628802734093527e-05,
+      "loss": 0.1603,
+      "mean_token_accuracy": 0.9428221821784973,
+      "num_tokens": 1011804.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.9586633563041687,
+      "epoch": 0.725883476599809,
+      "grad_norm": 2.4029595851898193,
+      "learning_rate": 7.56129961296375e-05,
+      "loss": 0.1696,
+      "mean_token_accuracy": 0.937748384475708,
+      "num_tokens": 1026043.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9234821617603302,
+      "epoch": 0.7354345749761223,
+      "grad_norm": 2.0575945377349854,
+      "learning_rate": 7.493157520265786e-05,
+      "loss": 0.1575,
+      "mean_token_accuracy": 0.9384551405906677,
+      "num_tokens": 1039649.0,
+      "step": 770
+    },
+    {
+      "entropy": 0.9257887125015258,
+      "epoch": 0.7449856733524355,
+      "grad_norm": 2.2613742351531982,
+      "learning_rate": 7.424393455518011e-05,
+      "loss": 0.1604,
+      "mean_token_accuracy": 0.9432004868984223,
+      "num_tokens": 1053317.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9160297989845276,
+      "epoch": 0.7545367717287488,
+      "grad_norm": 2.91170597076416,
+      "learning_rate": 7.355024573403174e-05,
+      "loss": 0.1659,
+      "mean_token_accuracy": 0.9321471691131592,
+      "num_tokens": 1066478.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9158035099506379,
+      "epoch": 0.7640878701050621,
+      "grad_norm": 1.7928736209869385,
+      "learning_rate": 7.285068179488794e-05,
+      "loss": 0.1525,
+      "mean_token_accuracy": 0.9404825747013092,
+      "num_tokens": 1080474.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.901829046010971,
+      "epoch": 0.7736389684813754,
+      "grad_norm": 1.490260362625122,
+      "learning_rate": 7.214541725909909e-05,
+      "loss": 0.1428,
+      "mean_token_accuracy": 0.9477003753185272,
+      "num_tokens": 1093980.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.9218035817146302,
+      "epoch": 0.7831900668576887,
+      "grad_norm": 1.58564031124115,
+      "learning_rate": 7.143462807015271e-05,
+      "loss": 0.1812,
+      "mean_token_accuracy": 0.9303424000740051,
+      "num_tokens": 1106996.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9264647245407105,
+      "epoch": 0.792741165234002,
+      "grad_norm": 2.501923084259033,
+      "learning_rate": 7.071849154978048e-05,
+      "loss": 0.1616,
+      "mean_token_accuracy": 0.9390891551971435,
+      "num_tokens": 1120972.0,
+      "step": 830
+    },
+    {
+      "entropy": 0.9230189204216004,
+      "epoch": 0.8022922636103151,
+      "grad_norm": 1.7454277276992798,
+      "learning_rate": 6.999718635372148e-05,
+      "loss": 0.1637,
+      "mean_token_accuracy": 0.938653826713562,
+      "num_tokens": 1134881.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9214071393013,
+      "epoch": 0.8118433619866284,
+      "grad_norm": 2.156137466430664,
+      "learning_rate": 6.927089242715277e-05,
+      "loss": 0.1382,
+      "mean_token_accuracy": 0.9446285247802735,
+      "num_tokens": 1148338.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.9470777630805969,
+      "epoch": 0.8213944603629417,
+      "grad_norm": 3.107943534851074,
+      "learning_rate": 6.853979095979804e-05,
+      "loss": 0.1543,
+      "mean_token_accuracy": 0.9390346229076385,
+      "num_tokens": 1162724.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.931933468580246,
+      "epoch": 0.830945558739255,
+      "grad_norm": 1.9495536088943481,
+      "learning_rate": 6.780406434072603e-05,
+      "loss": 0.1454,
+      "mean_token_accuracy": 0.9428299069404602,
+      "num_tokens": 1176245.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.9380590498447419,
+      "epoch": 0.8404966571155683,
+      "grad_norm": 2.419804334640503,
+      "learning_rate": 6.706389611284954e-05,
+      "loss": 0.1526,
+      "mean_token_accuracy": 0.9433525264263153,
+      "num_tokens": 1189996.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9564756214618683,
+      "epoch": 0.8500477554918816,
+      "grad_norm": 2.761873245239258,
+      "learning_rate": 6.63194709271367e-05,
+      "loss": 0.1606,
+      "mean_token_accuracy": 0.9358982861042022,
+      "num_tokens": 1203136.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9523034274578095,
+      "epoch": 0.8595988538681948,
+      "grad_norm": 1.9922142028808594,
+      "learning_rate": 6.557097449654587e-05,
+      "loss": 0.1656,
+      "mean_token_accuracy": 0.9327197670936584,
+      "num_tokens": 1216913.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9461675345897674,
+      "epoch": 0.8691499522445081,
+      "grad_norm": 2.829409599304199,
+      "learning_rate": 6.481859354969548e-05,
+      "loss": 0.149,
+      "mean_token_accuracy": 0.9414755582809449,
+      "num_tokens": 1229969.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.9531142830848693,
+      "epoch": 0.8787010506208214,
+      "grad_norm": 3.5151143074035645,
+      "learning_rate": 6.406251578428046e-05,
+      "loss": 0.1455,
+      "mean_token_accuracy": 0.9469887852668762,
+      "num_tokens": 1243680.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9358652174472809,
+      "epoch": 0.8882521489971347,
+      "grad_norm": 1.8639585971832275,
+      "learning_rate": 6.330292982024712e-05,
+      "loss": 0.1483,
+      "mean_token_accuracy": 0.9402450442314148,
+      "num_tokens": 1257573.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.9346361815929413,
+      "epoch": 0.897803247373448,
+      "grad_norm": 1.8855258226394653,
+      "learning_rate": 6.254002515273774e-05,
+      "loss": 0.1343,
+      "mean_token_accuracy": 0.9441287517547607,
+      "num_tokens": 1270701.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9197669744491577,
+      "epoch": 0.9073543457497613,
+      "grad_norm": 2.3212008476257324,
+      "learning_rate": 6.177399210481698e-05,
+      "loss": 0.1492,
+      "mean_token_accuracy": 0.9464975118637085,
+      "num_tokens": 1283809.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.9133516371250152,
+      "epoch": 0.9169054441260746,
+      "grad_norm": 2.0822534561157227,
+      "learning_rate": 6.100502177999156e-05,
+      "loss": 0.1352,
+      "mean_token_accuracy": 0.9481633484363556,
+      "num_tokens": 1297005.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9345045983791351,
+      "epoch": 0.9264565425023877,
+      "grad_norm": 2.5879642963409424,
+      "learning_rate": 6.023330601453551e-05,
+      "loss": 0.1751,
+      "mean_token_accuracy": 0.9324521780014038,
+      "num_tokens": 1310830.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9381786108016967,
+      "epoch": 0.936007640878701,
+      "grad_norm": 2.2892372608184814,
+      "learning_rate": 5.945903732963234e-05,
+      "loss": 0.1512,
+      "mean_token_accuracy": 0.9421308040618896,
+      "num_tokens": 1324145.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.935336297750473,
+      "epoch": 0.9455587392550143,
+      "grad_norm": 2.16275954246521,
+      "learning_rate": 5.868240888334653e-05,
+      "loss": 0.1381,
+      "mean_token_accuracy": 0.9414000988006592,
+      "num_tokens": 1337718.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.9044704020023346,
+      "epoch": 0.9551098376313276,
+      "grad_norm": 2.0526506900787354,
+      "learning_rate": 5.790361442243605e-05,
+      "loss": 0.1419,
+      "mean_token_accuracy": 0.9416944205760955,
+      "num_tokens": 1351556.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9375361979007721,
+      "epoch": 0.9646609360076409,
+      "grad_norm": 1.9554805755615234,
+      "learning_rate": 5.712284823401808e-05,
+      "loss": 0.152,
+      "mean_token_accuracy": 0.9391957759857178,
+      "num_tokens": 1364817.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.9393334090709686,
+      "epoch": 0.9742120343839542,
+      "grad_norm": 1.8596081733703613,
+      "learning_rate": 5.634030509709982e-05,
+      "loss": 0.148,
+      "mean_token_accuracy": 0.9449970602989197,
+      "num_tokens": 1378098.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.932980078458786,
+      "epoch": 0.9837631327602674,
+      "grad_norm": 1.8846793174743652,
+      "learning_rate": 5.555618023398671e-05,
+      "loss": 0.1326,
+      "mean_token_accuracy": 0.947298276424408,
+      "num_tokens": 1391374.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9318273425102234,
+      "epoch": 0.9933142311365807,
+      "grad_norm": 3.373645305633545,
+      "learning_rate": 5.477066926158001e-05,
+      "loss": 0.1574,
+      "mean_token_accuracy": 0.9381673038005829,
+      "num_tokens": 1404715.0,
+      "step": 1040
+    },
+    {
+      "entropy": 0.9389628827571869,
+      "epoch": 1.002865329512894,
+      "grad_norm": 2.2420825958251953,
+      "learning_rate": 5.398396814257588e-05,
+      "loss": 0.1351,
+      "mean_token_accuracy": 0.9487472474575043,
+      "num_tokens": 1417657.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.8906395196914673,
+      "epoch": 1.0124164278892072,
+      "grad_norm": 2.2359447479248047,
+      "learning_rate": 5.3196273136578286e-05,
+      "loss": 0.0947,
+      "mean_token_accuracy": 0.9671330749988556,
+      "num_tokens": 1430922.0,
+      "step": 1060
+    },
+    {
+      "entropy": 0.8819694340229034,
+      "epoch": 1.0219675262655206,
+      "grad_norm": 1.746660590171814,
+      "learning_rate": 5.2407780751137815e-05,
+      "loss": 0.0941,
+      "mean_token_accuracy": 0.9668048799037934,
+      "num_tokens": 1444074.0,
+      "step": 1070
+    },
+    {
+      "entropy": 0.8694253623485565,
+      "epoch": 1.0315186246418337,
+      "grad_norm": 1.9695682525634766,
+      "learning_rate": 5.1618687692728595e-05,
+      "loss": 0.0753,
+      "mean_token_accuracy": 0.9725252211093902,
+      "num_tokens": 1457733.0,
+      "step": 1080
+    },
+    {
+      "entropy": 0.877548235654831,
+      "epoch": 1.0410697230181472,
+      "grad_norm": 2.7611336708068848,
+      "learning_rate": 5.0829190817675575e-05,
+      "loss": 0.0976,
+      "mean_token_accuracy": 0.9646038115024567,
+      "num_tokens": 1471308.0,
+      "step": 1090
+    },
+    {
+      "entropy": 0.8579001665115357,
+      "epoch": 1.0506208213944603,
+      "grad_norm": 2.0661134719848633,
+      "learning_rate": 5.0039487083044566e-05,
+      "loss": 0.078,
+      "mean_token_accuracy": 0.9726655781269073,
+      "num_tokens": 1484529.0,
+      "step": 1100
+    },
+    {
+      "entropy": 0.8462372958660126,
+      "epoch": 1.0601719197707737,
+      "grad_norm": 2.4989826679229736,
+      "learning_rate": 4.924977349750691e-05,
+      "loss": 0.0823,
+      "mean_token_accuracy": 0.9697394073009491,
+      "num_tokens": 1498025.0,
+      "step": 1110
+    },
+    {
+      "entropy": 0.8473825812339782,
+      "epoch": 1.069723018147087,
+      "grad_norm": 3.063563823699951,
+      "learning_rate": 4.8460247072191496e-05,
+      "loss": 0.0733,
+      "mean_token_accuracy": 0.9737793922424316,
+      "num_tokens": 1512502.0,
+      "step": 1120
+    },
+    {
+      "entropy": 0.8528614521026612,
+      "epoch": 1.0792741165234,
+      "grad_norm": 2.53879976272583,
+      "learning_rate": 4.767110477153604e-05,
+      "loss": 0.0774,
+      "mean_token_accuracy": 0.971060448884964,
+      "num_tokens": 1525623.0,
+      "step": 1130
+    },
+    {
+      "entropy": 0.8445274353027343,
+      "epoch": 1.0888252148997135,
+      "grad_norm": 1.7218629121780396,
+      "learning_rate": 4.688254346415006e-05,
+      "loss": 0.0821,
+      "mean_token_accuracy": 0.9694690883159638,
+      "num_tokens": 1538915.0,
+      "step": 1140
+    },
+    {
+      "entropy": 0.8586056411266327,
+      "epoch": 1.0983763132760267,
+      "grad_norm": 2.111964464187622,
+      "learning_rate": 4.609475987370177e-05,
+      "loss": 0.0952,
+      "mean_token_accuracy": 0.9659283339977265,
+      "num_tokens": 1552610.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.8540456056594848,
+      "epoch": 1.10792741165234,
+      "grad_norm": 1.9750651121139526,
+      "learning_rate": 4.530795052984104e-05,
+      "loss": 0.0775,
+      "mean_token_accuracy": 0.9711886167526245,
+      "num_tokens": 1566152.0,
+      "step": 1160
+    },
+    {
+      "entropy": 0.8344644010066986,
+      "epoch": 1.1174785100286533,
+      "grad_norm": 1.9746479988098145,
+      "learning_rate": 4.4522311719170866e-05,
+      "loss": 0.0691,
+      "mean_token_accuracy": 0.9779588758945466,
+      "num_tokens": 1579016.0,
+      "step": 1170
+    },
+    {
+      "entropy": 0.8578928172588348,
+      "epoch": 1.1270296084049667,
+      "grad_norm": 2.236234664916992,
+      "learning_rate": 4.373803943627946e-05,
+      "loss": 0.0886,
+      "mean_token_accuracy": 0.9684166312217712,
+      "num_tokens": 1592658.0,
+      "step": 1180
+    },
+    {
+      "entropy": 0.8430122971534729,
+      "epoch": 1.1365807067812799,
+      "grad_norm": 1.79709792137146,
+      "learning_rate": 4.295532933484509e-05,
+      "loss": 0.0864,
+      "mean_token_accuracy": 0.9701191842556,
+      "num_tokens": 1606104.0,
+      "step": 1190
+    },
+    {
+      "entropy": 0.8627173721790313,
+      "epoch": 1.146131805157593,
+      "grad_norm": 1.4530774354934692,
+      "learning_rate": 4.2174376678826104e-05,
+      "loss": 0.0834,
+      "mean_token_accuracy": 0.9710407853126526,
+      "num_tokens": 1620362.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.8454810082912445,
+      "epoch": 1.1556829035339065,
+      "grad_norm": 1.8320307731628418,
+      "learning_rate": 4.1395376293748134e-05,
+      "loss": 0.078,
+      "mean_token_accuracy": 0.9728212952613831,
+      "num_tokens": 1633859.0,
+      "step": 1210
+    },
+    {
+      "entropy": 0.8578923225402832,
+      "epoch": 1.1652340019102196,
+      "grad_norm": 1.7667258977890015,
+      "learning_rate": 4.0618522518100534e-05,
+      "loss": 0.0591,
+      "mean_token_accuracy": 0.9807173669338226,
+      "num_tokens": 1646952.0,
+      "step": 1220
+    },
+    {
+      "entropy": 0.8332420349121094,
+      "epoch": 1.174785100286533,
+      "grad_norm": 1.9839752912521362,
+      "learning_rate": 3.984400915485459e-05,
+      "loss": 0.0666,
+      "mean_token_accuracy": 0.9741971969604493,
+      "num_tokens": 1659969.0,
+      "step": 1230
+    },
+    {
+      "entropy": 0.8267807960510254,
+      "epoch": 1.1843361986628462,
+      "grad_norm": 1.4242769479751587,
+      "learning_rate": 3.9072029423115055e-05,
+      "loss": 0.082,
+      "mean_token_accuracy": 0.9703181862831116,
+      "num_tokens": 1673108.0,
+      "step": 1240
+    },
+    {
+      "entropy": 0.8347297132015228,
+      "epoch": 1.1938872970391594,
+      "grad_norm": 1.8707839250564575,
+      "learning_rate": 3.830277590991759e-05,
+      "loss": 0.0651,
+      "mean_token_accuracy": 0.9774328708648682,
+      "num_tokens": 1686935.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.8196123540401459,
+      "epoch": 1.2034383954154728,
+      "grad_norm": 1.6634894609451294,
+      "learning_rate": 3.75364405221836e-05,
+      "loss": 0.0758,
+      "mean_token_accuracy": 0.9732350528240203,
+      "num_tokens": 1700292.0,
+      "step": 1260
+    },
+    {
+      "entropy": 0.8230455994606019,
+      "epoch": 1.212989493791786,
+      "grad_norm": 2.8074488639831543,
+      "learning_rate": 3.677321443884509e-05,
+      "loss": 0.0753,
+      "mean_token_accuracy": 0.9691428065299987,
+      "num_tokens": 1713397.0,
+      "step": 1270
+    },
+    {
+      "entropy": 0.8024696469306946,
+      "epoch": 1.2225405921680994,
+      "grad_norm": 1.2013849020004272,
+      "learning_rate": 3.601328806315082e-05,
+      "loss": 0.0542,
+      "mean_token_accuracy": 0.982044231891632,
+      "num_tokens": 1727202.0,
+      "step": 1280
+    },
+    {
+      "entropy": 0.8168217360973358,
+      "epoch": 1.2320916905444126,
+      "grad_norm": 1.7047194242477417,
+      "learning_rate": 3.5256850975166136e-05,
+      "loss": 0.0728,
+      "mean_token_accuracy": 0.9717679738998413,
+      "num_tokens": 1740680.0,
+      "step": 1290
+    },
+    {
+      "entropy": 0.8255608975887299,
+      "epoch": 1.2416427889207258,
+      "grad_norm": 1.3907794952392578,
+      "learning_rate": 3.4504091884478074e-05,
+      "loss": 0.0696,
+      "mean_token_accuracy": 0.9770924091339112,
+      "num_tokens": 1754482.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.8318446159362793,
+      "epoch": 1.2511938872970392,
+      "grad_norm": 2.9260592460632324,
+      "learning_rate": 3.375519858311781e-05,
+      "loss": 0.0872,
+      "mean_token_accuracy": 0.9716545462608337,
+      "num_tokens": 1768580.0,
+      "step": 1310
+    },
+    {
+      "entropy": 0.8437300324440002,
+      "epoch": 1.2607449856733524,
+      "grad_norm": 1.56409752368927,
+      "learning_rate": 3.301035789871169e-05,
+      "loss": 0.0736,
+      "mean_token_accuracy": 0.9731296360492706,
+      "num_tokens": 1781995.0,
+      "step": 1320
+    },
+    {
+      "entropy": 0.8272697806358338,
+      "epoch": 1.2702960840496658,
+      "grad_norm": 2.2259328365325928,
+      "learning_rate": 3.226975564787322e-05,
+      "loss": 0.0761,
+      "mean_token_accuracy": 0.9743330478668213,
+      "num_tokens": 1795247.0,
+      "step": 1330
+    },
+    {
+      "entropy": 0.8293983697891235,
+      "epoch": 1.279847182425979,
+      "grad_norm": 2.222468852996826,
+      "learning_rate": 3.153357658984697e-05,
+      "loss": 0.0656,
+      "mean_token_accuracy": 0.9765427887439728,
+      "num_tokens": 1808634.0,
+      "step": 1340
+    },
+    {
+      "entropy": 0.8390303313732147,
+      "epoch": 1.2893982808022924,
+      "grad_norm": 2.155545234680176,
+      "learning_rate": 3.0802004380416414e-05,
+      "loss": 0.0664,
+      "mean_token_accuracy": 0.9755440354347229,
+      "num_tokens": 1822813.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.8125343024730682,
+      "epoch": 1.2989493791786055,
+      "grad_norm": 1.7273448705673218,
+      "learning_rate": 3.0075221526087082e-05,
+      "loss": 0.0625,
+      "mean_token_accuracy": 0.9774803996086121,
+      "num_tokens": 1836410.0,
+      "step": 1360
+    },
+    {
+      "entropy": 0.8073979437351226,
+      "epoch": 1.3085004775549187,
+      "grad_norm": 1.7263071537017822,
+      "learning_rate": 2.935340933855626e-05,
+      "loss": 0.065,
+      "mean_token_accuracy": 0.9779576480388641,
+      "num_tokens": 1850061.0,
+      "step": 1370
+    },
+    {
+      "entropy": 0.817969411611557,
+      "epoch": 1.3180515759312321,
+      "grad_norm": 3.0116782188415527,
+      "learning_rate": 2.863674788948097e-05,
+      "loss": 0.0735,
+      "mean_token_accuracy": 0.9746712803840637,
+      "num_tokens": 1864554.0,
+      "step": 1380
+    },
+    {
+      "entropy": 0.8101901948451996,
+      "epoch": 1.3276026743075453,
+      "grad_norm": 2.0107994079589844,
+      "learning_rate": 2.7925415965555123e-05,
+      "loss": 0.067,
+      "mean_token_accuracy": 0.9772519171237946,
+      "num_tokens": 1878030.0,
+      "step": 1390
+    },
+    {
+      "entropy": 0.8207332074642182,
+      "epoch": 1.3371537726838587,
+      "grad_norm": 1.834949016571045,
+      "learning_rate": 2.7219591023907408e-05,
+      "loss": 0.0688,
+      "mean_token_accuracy": 0.9770865857601165,
+      "num_tokens": 1891710.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.81460622549057,
+      "epoch": 1.346704871060172,
+      "grad_norm": 2.781595230102539,
+      "learning_rate": 2.6519449147830637e-05,
+      "loss": 0.0602,
+      "mean_token_accuracy": 0.9770792603492737,
+      "num_tokens": 1905521.0,
+      "step": 1410
+    },
+    {
+      "entropy": 0.8120210707187653,
+      "epoch": 1.3562559694364853,
+      "grad_norm": 3.5601930618286133,
+      "learning_rate": 2.5825165002854124e-05,
+      "loss": 0.0678,
+      "mean_token_accuracy": 0.9757714390754699,
+      "num_tokens": 1918326.0,
+      "step": 1420
+    },
+    {
+      "entropy": 0.827415007352829,
+      "epoch": 1.3658070678127985,
+      "grad_norm": 1.229512333869934,
+      "learning_rate": 2.513691179316942e-05,
+      "loss": 0.0766,
+      "mean_token_accuracy": 0.9711452543735504,
+      "num_tokens": 1931996.0,
+      "step": 1430
+    },
+    {
+      "entropy": 0.8105746507644653,
+      "epoch": 1.3753581661891117,
+      "grad_norm": 1.5925318002700806,
+      "learning_rate": 2.4454861218420925e-05,
+      "loss": 0.0677,
+      "mean_token_accuracy": 0.9764558732509613,
+      "num_tokens": 1945489.0,
+      "step": 1440
+    },
+    {
+      "entropy": 0.8041588008403778,
+      "epoch": 1.384909264565425,
+      "grad_norm": 4.77223539352417,
+      "learning_rate": 2.3779183430871594e-05,
+      "loss": 0.0718,
+      "mean_token_accuracy": 0.9775639712810517,
+      "num_tokens": 1959170.0,
+      "step": 1450
+    },
+    {
+      "entropy": 0.8073832333087921,
+      "epoch": 1.3944603629417383,
+      "grad_norm": 1.9326298236846924,
+      "learning_rate": 2.311004699295488e-05,
+      "loss": 0.0686,
+      "mean_token_accuracy": 0.9766038298606873,
+      "num_tokens": 1972774.0,
+      "step": 1460
+    },
+    {
+      "entropy": 0.8097073554992675,
+      "epoch": 1.4040114613180517,
+      "grad_norm": 1.6878184080123901,
+      "learning_rate": 2.2447618835223127e-05,
+      "loss": 0.0652,
+      "mean_token_accuracy": 0.9758686602115632,
+      "num_tokens": 1986433.0,
+      "step": 1470
+    },
+    {
+      "entropy": 0.7953924179077149,
+      "epoch": 1.4135625596943648,
+      "grad_norm": 1.6706430912017822,
+      "learning_rate": 2.1792064214703102e-05,
+      "loss": 0.054,
+      "mean_token_accuracy": 0.9801817893981933,
+      "num_tokens": 2000018.0,
+      "step": 1480
+    },
+    {
+      "entropy": 0.7993491113185882,
+      "epoch": 1.4231136580706782,
+      "grad_norm": 1.1433459520339966,
+      "learning_rate": 2.1143546673669122e-05,
+      "loss": 0.0467,
+      "mean_token_accuracy": 0.9827651143074035,
+      "num_tokens": 2013829.0,
+      "step": 1490
+    },
+    {
+      "entropy": 0.7798974096775055,
+      "epoch": 1.4326647564469914,
+      "grad_norm": 3.4944183826446533,
+      "learning_rate": 2.050222799884387e-05,
+      "loss": 0.0439,
+      "mean_token_accuracy": 0.9847986400127411,
+      "num_tokens": 2027142.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.8006802797317505,
+      "epoch": 1.4422158548233046,
+      "grad_norm": 1.5036331415176392,
+      "learning_rate": 1.9868268181037185e-05,
+      "loss": 0.0745,
+      "mean_token_accuracy": 0.9759544312953949,
+      "num_tokens": 2040851.0,
+      "step": 1510
+    },
+    {
+      "entropy": 0.7876431047916412,
+      "epoch": 1.451766953199618,
+      "grad_norm": 2.3396732807159424,
+      "learning_rate": 1.9241825375232947e-05,
+      "loss": 0.057,
+      "mean_token_accuracy": 0.9790268540382385,
+      "num_tokens": 2054703.0,
+      "step": 1520
+    },
+    {
+      "entropy": 0.8015381276607514,
+      "epoch": 1.4613180515759312,
+      "grad_norm": 1.4720062017440796,
+      "learning_rate": 1.8623055861133852e-05,
+      "loss": 0.0609,
+      "mean_token_accuracy": 0.9771195352077484,
+      "num_tokens": 2067960.0,
+      "step": 1530
+    },
+    {
+      "entropy": 0.8023482382297515,
+      "epoch": 1.4708691499522444,
+      "grad_norm": 2.1161856651306152,
+      "learning_rate": 1.8012114004174048e-05,
+      "loss": 0.065,
+      "mean_token_accuracy": 0.977092319726944,
+      "num_tokens": 2081121.0,
+      "step": 1540
+    },
+    {
+      "entropy": 0.8031758368015289,
+      "epoch": 1.4804202483285578,
+      "grad_norm": 2.329909563064575,
+      "learning_rate": 1.7409152217009368e-05,
+      "loss": 0.0565,
+      "mean_token_accuracy": 0.9796651124954223,
+      "num_tokens": 2094631.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.8102335870265961,
+      "epoch": 1.4899713467048712,
+      "grad_norm": 1.989434838294983,
+      "learning_rate": 1.6814320921494693e-05,
+      "loss": 0.0566,
+      "mean_token_accuracy": 0.9804701745510102,
+      "num_tokens": 2108099.0,
+      "step": 1560
+    },
+    {
+      "entropy": 0.8178816676139832,
+      "epoch": 1.4995224450811844,
+      "grad_norm": 1.9576576948165894,
+      "learning_rate": 1.6227768511157976e-05,
+      "loss": 0.0556,
+      "mean_token_accuracy": 0.9800428807735443,
+      "num_tokens": 2121504.0,
+      "step": 1570
+    },
+    {
+      "entropy": 0.8111383557319641,
+      "epoch": 1.5090735434574976,
+      "grad_norm": 2.94604754447937,
+      "learning_rate": 1.5649641314180274e-05,
+      "loss": 0.0574,
+      "mean_token_accuracy": 0.9815197467803956,
+      "num_tokens": 2134924.0,
+      "step": 1580
+    },
+    {
+      "entropy": 0.8119295537471771,
+      "epoch": 1.518624641833811,
+      "grad_norm": 4.296211242675781,
+      "learning_rate": 1.5080083556890973e-05,
+      "loss": 0.0606,
+      "mean_token_accuracy": 0.9784540951251983,
+      "num_tokens": 2148917.0,
+      "step": 1590
+    },
+    {
+      "entropy": 0.8107995390892029,
+      "epoch": 1.5281757402101241,
+      "grad_norm": 1.307666540145874,
+      "learning_rate": 1.4519237327787449e-05,
+      "loss": 0.0636,
+      "mean_token_accuracy": 0.9752672553062439,
+      "num_tokens": 2161999.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.8124419927597046,
+      "epoch": 1.5377268385864373,
+      "grad_norm": 1.418534278869629,
+      "learning_rate": 1.3967242542088005e-05,
+      "loss": 0.0534,
+      "mean_token_accuracy": 0.9819561600685119,
+      "num_tokens": 2175634.0,
+      "step": 1610
+    },
+    {
+      "entropy": 0.8168574690818786,
+      "epoch": 1.5472779369627507,
+      "grad_norm": 1.1592333316802979,
+      "learning_rate": 1.3424236906827042e-05,
+      "loss": 0.0533,
+      "mean_token_accuracy": 0.9838660717010498,
+      "num_tokens": 2189783.0,
+      "step": 1620
+    },
+    {
+      "entropy": 0.8068838775157928,
+      "epoch": 1.5568290353390641,
+      "grad_norm": 1.5056216716766357,
+      "learning_rate": 1.289035588650097e-05,
+      "loss": 0.0383,
+      "mean_token_accuracy": 0.9870678842067718,
+      "num_tokens": 2203186.0,
+      "step": 1630
+    },
+    {
+      "entropy": 0.7941603124141693,
+      "epoch": 1.5663801337153773,
+      "grad_norm": 2.706909418106079,
+      "learning_rate": 1.2365732669273778e-05,
+      "loss": 0.0543,
+      "mean_token_accuracy": 0.9797939538955689,
+      "num_tokens": 2216081.0,
+      "step": 1640
+    },
+    {
+      "entropy": 0.8098276674747467,
+      "epoch": 1.5759312320916905,
+      "grad_norm": 2.3556482791900635,
+      "learning_rate": 1.1850498133750244e-05,
+      "loss": 0.0558,
+      "mean_token_accuracy": 0.9782669067382812,
+      "num_tokens": 2229463.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.8095695912837982,
+      "epoch": 1.585482330468004,
+      "grad_norm": 2.1471827030181885,
+      "learning_rate": 1.1344780816325511e-05,
+      "loss": 0.0483,
+      "mean_token_accuracy": 0.9803479790687561,
+      "num_tokens": 2242823.0,
+      "step": 1660
+    },
+    {
+      "entropy": 0.8069136083126068,
+      "epoch": 1.595033428844317,
+      "grad_norm": 2.6952850818634033,
+      "learning_rate": 1.0848706879118892e-05,
+      "loss": 0.0675,
+      "mean_token_accuracy": 0.9762344419956207,
+      "num_tokens": 2255767.0,
+      "step": 1670
+    },
+    {
+      "entropy": 0.8157409608364106,
+      "epoch": 1.6045845272206303,
+      "grad_norm": 2.3851828575134277,
+      "learning_rate": 1.0362400078500018e-05,
+      "loss": 0.0611,
+      "mean_token_accuracy": 0.9783193826675415,
+      "num_tokens": 2269452.0,
+      "step": 1680
+    },
+    {
+      "entropy": 0.8174916446208954,
+      "epoch": 1.6141356255969437,
+      "grad_norm": 1.9937782287597656,
+      "learning_rate": 9.885981734215094e-06,
+      "loss": 0.0507,
+      "mean_token_accuracy": 0.9808118999004364,
+      "num_tokens": 2283260.0,
+      "step": 1690
+    },
+    {
+      "entropy": 0.8144233644008636,
+      "epoch": 1.623686723973257,
+      "grad_norm": 2.685246706008911,
+      "learning_rate": 9.419570699121072e-06,
+      "loss": 0.0445,
+      "mean_token_accuracy": 0.9841122567653656,
+      "num_tokens": 2297040.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.8087396621704102,
+      "epoch": 1.63323782234957,
+      "grad_norm": 3.3513360023498535,
+      "learning_rate": 8.963283329535305e-06,
+      "loss": 0.0535,
+      "mean_token_accuracy": 0.9839581727981568,
+      "num_tokens": 2310231.0,
+      "step": 1710
+    },
+    {
+      "entropy": 0.7899291276931762,
+      "epoch": 1.6427889207258835,
+      "grad_norm": 1.762925386428833,
+      "learning_rate": 8.517233456207818e-06,
+      "loss": 0.0482,
+      "mean_token_accuracy": 0.9834255933761596,
+      "num_tokens": 2322983.0,
+      "step": 1720
+    },
+    {
+      "entropy": 0.7878467202186584,
+      "epoch": 1.6523400191021969,
+      "grad_norm": 1.6693549156188965,
+      "learning_rate": 8.081532355923922e-06,
+      "loss": 0.0477,
+      "mean_token_accuracy": 0.9824950993061066,
+      "num_tokens": 2336190.0,
+      "step": 1730
+    },
+    {
+      "entropy": 0.8147098064422608,
+      "epoch": 1.66189111747851,
+      "grad_norm": 2.8894336223602295,
+      "learning_rate": 7.65628872374367e-06,
+      "loss": 0.0591,
+      "mean_token_accuracy": 0.9804086804389953,
+      "num_tokens": 2349752.0,
+      "step": 1740
+    },
+    {
+      "entropy": 0.8126006305217743,
+      "epoch": 1.6714422158548232,
+      "grad_norm": 2.730252504348755,
+      "learning_rate": 7.241608645885628e-06,
+      "loss": 0.0563,
+      "mean_token_accuracy": 0.9782049119472503,
+      "num_tokens": 2363415.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.7979379296302795,
+      "epoch": 1.6809933142311366,
+      "grad_norm": 2.505347967147827,
+      "learning_rate": 6.8375955732612964e-06,
+      "loss": 0.0485,
+      "mean_token_accuracy": 0.983302628993988,
+      "num_tokens": 2376238.0,
+      "step": 1760
+    },
+    {
+      "entropy": 0.8013978183269501,
+      "epoch": 1.6905444126074498,
+      "grad_norm": 2.0276763439178467,
+      "learning_rate": 6.444350295667112e-06,
+      "loss": 0.0517,
+      "mean_token_accuracy": 0.9832404673099517,
+      "num_tokens": 2389365.0,
+      "step": 1770
+    },
+    {
+      "entropy": 0.8164916813373566,
+      "epoch": 1.700095510983763,
+      "grad_norm": 2.615689277648926,
+      "learning_rate": 6.061970916640236e-06,
+      "loss": 0.0608,
+      "mean_token_accuracy": 0.9772882699966431,
+      "num_tokens": 2403735.0,
+      "step": 1780
+    },
+    {
+      "entropy": 0.8028056979179382,
+      "epoch": 1.7096466093600764,
+      "grad_norm": 2.2093160152435303,
+      "learning_rate": 5.690552828984458e-06,
+      "loss": 0.0417,
+      "mean_token_accuracy": 0.9841553449630738,
+      "num_tokens": 2417109.0,
+      "step": 1790
+    },
+    {
+      "entropy": 0.8034515142440796,
+      "epoch": 1.7191977077363898,
+      "grad_norm": 1.838507056236267,
+      "learning_rate": 5.3301886909724375e-06,
+      "loss": 0.0477,
+      "mean_token_accuracy": 0.9830919563770294,
+      "num_tokens": 2430479.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.8112623810768127,
+      "epoch": 1.728748806112703,
+      "grad_norm": 1.8341842889785767,
+      "learning_rate": 4.980968403230097e-06,
+      "loss": 0.053,
+      "mean_token_accuracy": 0.9810094058513641,
+      "num_tokens": 2444501.0,
+      "step": 1810
+    },
+    {
+      "entropy": 0.812260115146637,
+      "epoch": 1.7382999044890162,
+      "grad_norm": 1.385706901550293,
+      "learning_rate": 4.642979086308963e-06,
+      "loss": 0.0495,
+      "mean_token_accuracy": 0.9843249559402466,
+      "num_tokens": 2458250.0,
+      "step": 1820
+    },
+    {
+      "entropy": 0.7905302405357361,
+      "epoch": 1.7478510028653296,
+      "grad_norm": 2.529675006866455,
+      "learning_rate": 4.316305058952069e-06,
+      "loss": 0.0414,
+      "mean_token_accuracy": 0.9862417578697205,
+      "num_tokens": 2471343.0,
+      "step": 1830
+    },
+    {
+      "entropy": 0.7993218123912811,
+      "epoch": 1.7574021012416428,
+      "grad_norm": 1.8865885734558105,
+      "learning_rate": 4.001027817058789e-06,
+      "loss": 0.0411,
+      "mean_token_accuracy": 0.9835248470306397,
+      "num_tokens": 2485461.0,
+      "step": 1840
+    },
+    {
+      "entropy": 0.8125109553337098,
+      "epoch": 1.766953199617956,
+      "grad_norm": 1.9268198013305664,
+      "learning_rate": 3.697226013353955e-06,
+      "loss": 0.0446,
+      "mean_token_accuracy": 0.9831093788146973,
+      "num_tokens": 2499676.0,
+      "step": 1850
+    },
+    {
+      "entropy": 0.8063633918762207,
+      "epoch": 1.7765042979942693,
+      "grad_norm": 2.2370738983154297,
+      "learning_rate": 3.404975437766217e-06,
+      "loss": 0.0422,
+      "mean_token_accuracy": 0.9837429761886597,
+      "num_tokens": 2513383.0,
+      "step": 1860
+    },
+    {
+      "entropy": 0.8041584312915802,
+      "epoch": 1.7860553963705827,
+      "grad_norm": 2.501936912536621,
+      "learning_rate": 3.1243489985206097e-06,
+      "loss": 0.0569,
+      "mean_token_accuracy": 0.9803853929042816,
+      "num_tokens": 2526503.0,
+      "step": 1870
+    },
+    {
+      "entropy": 0.809213387966156,
+      "epoch": 1.795606494746896,
+      "grad_norm": 1.7907541990280151,
+      "learning_rate": 2.8554167039500478e-06,
+      "loss": 0.0458,
+      "mean_token_accuracy": 0.9832297623157501,
+      "num_tokens": 2539835.0,
+      "step": 1880
+    },
+    {
+      "entropy": 0.7946874141693115,
+      "epoch": 1.8051575931232091,
+      "grad_norm": 2.514709234237671,
+      "learning_rate": 2.598245645030195e-06,
+      "loss": 0.0463,
+      "mean_token_accuracy": 0.9826865434646607,
+      "num_tokens": 2553396.0,
+      "step": 1890
+    },
+    {
+      "entropy": 0.8088546633720398,
+      "epoch": 1.8147086914995225,
+      "grad_norm": 1.6304333209991455,
+      "learning_rate": 2.3528999786421756e-06,
+      "loss": 0.0606,
+      "mean_token_accuracy": 0.9778319001197815,
+      "num_tokens": 2567080.0,
+      "step": 1900
+    },
+    {
+      "entropy": 0.8208195269107819,
+      "epoch": 1.8242597898758357,
+      "grad_norm": 2.0225703716278076,
+      "learning_rate": 2.1194409115672663e-06,
+      "loss": 0.0483,
+      "mean_token_accuracy": 0.9834692001342773,
+      "num_tokens": 2580658.0,
+      "step": 1910
+    },
+    {
+      "entropy": 0.806461226940155,
+      "epoch": 1.8338108882521489,
+      "grad_norm": 1.2506319284439087,
+      "learning_rate": 1.8979266852175382e-06,
+      "loss": 0.0476,
+      "mean_token_accuracy": 0.982514089345932,
+      "num_tokens": 2594051.0,
+      "step": 1920
+    },
+    {
+      "entropy": 0.8200725972652435,
+      "epoch": 1.8433619866284623,
+      "grad_norm": 3.134730577468872,
+      "learning_rate": 1.688412561106284e-06,
+      "loss": 0.052,
+      "mean_token_accuracy": 0.9808144509792328,
+      "num_tokens": 2607491.0,
+      "step": 1930
+    },
+    {
+      "entropy": 0.7992699146270752,
+      "epoch": 1.8529130850047757,
+      "grad_norm": 1.5495048761367798,
+      "learning_rate": 1.4909508070618084e-06,
+      "loss": 0.0397,
+      "mean_token_accuracy": 0.9838496923446656,
+      "num_tokens": 2621029.0,
+      "step": 1940
+    },
+    {
+      "entropy": 0.8093390583992004,
+      "epoch": 1.8624641833810889,
+      "grad_norm": 2.413599967956543,
+      "learning_rate": 1.3055906841881348e-06,
+      "loss": 0.0454,
+      "mean_token_accuracy": 0.9853663623332978,
+      "num_tokens": 2634868.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.8016145169734955,
+      "epoch": 1.872015281757402,
+      "grad_norm": 1.8612229824066162,
+      "learning_rate": 1.1323784345757205e-06,
+      "loss": 0.055,
+      "mean_token_accuracy": 0.9820682048797608,
+      "num_tokens": 2648316.0,
+      "step": 1960
+    },
+    {
+      "entropy": 0.8011737823486328,
+      "epoch": 1.8815663801337155,
+      "grad_norm": 1.176445722579956,
+      "learning_rate": 9.71357269765366e-07,
+      "loss": 0.0518,
+      "mean_token_accuracy": 0.9793670833110809,
+      "num_tokens": 2661446.0,
+      "step": 1970
+    },
+    {
+      "entropy": 0.8211520731449127,
+      "epoch": 1.8911174785100286,
+      "grad_norm": 1.3449292182922363,
+      "learning_rate": 8.225673599682016e-07,
+      "loss": 0.0498,
+      "mean_token_accuracy": 0.9845996201038361,
+      "num_tokens": 2675223.0,
+      "step": 1980
+    },
+    {
+      "entropy": 0.8190204262733459,
+      "epoch": 1.9006685768863418,
+      "grad_norm": 3.4368741512298584,
+      "learning_rate": 6.860458240443179e-07,
+      "loss": 0.0549,
+      "mean_token_accuracy": 0.9792223453521729,
+      "num_tokens": 2688452.0,
+      "step": 1990
+    },
+    {
+      "entropy": 0.7945617794990539,
+      "epoch": 1.9102196752626552,
+      "grad_norm": 2.101534128189087,
+      "learning_rate": 5.618267202426864e-07,
+      "loss": 0.0476,
+      "mean_token_accuracy": 0.9821176826953888,
+      "num_tokens": 2701249.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.8119494199752808,
+      "epoch": 1.9197707736389686,
+      "grad_norm": 1.253597378730774,
+      "learning_rate": 4.4994103770457653e-07,
+      "loss": 0.0377,
+      "mean_token_accuracy": 0.9888437449932098,
+      "num_tokens": 2714442.0,
+      "step": 2010
+    },
+    {
+      "entropy": 0.8030437290668487,
+      "epoch": 1.9293218720152816,
+      "grad_norm": 2.829202890396118,
+      "learning_rate": 3.5041668873266877e-07,
+      "loss": 0.0495,
+      "mean_token_accuracy": 0.9828372478485108,
+      "num_tokens": 2727832.0,
+      "step": 2020
+    },
+    {
+      "entropy": 0.8061738967895508,
+      "epoch": 1.938872970391595,
+      "grad_norm": 1.7364760637283325,
+      "learning_rate": 2.6327850182769066e-07,
+      "loss": 0.0557,
+      "mean_token_accuracy": 0.980205237865448,
+      "num_tokens": 2741459.0,
+      "step": 2030
+    },
+    {
+      "entropy": 0.8008185088634491,
+      "epoch": 1.9484240687679084,
+      "grad_norm": 1.8003606796264648,
+      "learning_rate": 1.8854821549445513e-07,
+      "loss": 0.0417,
+      "mean_token_accuracy": 0.9849048316478729,
+      "num_tokens": 2754154.0,
+      "step": 2040
+    },
+    {
+      "entropy": 0.8082012474536896,
+      "epoch": 1.9579751671442216,
+      "grad_norm": 2.741069793701172,
+      "learning_rate": 1.2624447281867624e-07,
+      "loss": 0.0632,
+      "mean_token_accuracy": 0.9794716775417328,
+      "num_tokens": 2767928.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.8184729814529419,
+      "epoch": 1.9675262655205348,
+      "grad_norm": 2.215808153152466,
+      "learning_rate": 7.638281681608961e-08,
+      "loss": 0.0431,
+      "mean_token_accuracy": 0.9853759586811066,
+      "num_tokens": 2782121.0,
+      "step": 2060
+    },
+    {
+      "entropy": 0.798491793870926,
+      "epoch": 1.9770773638968482,
+      "grad_norm": 3.338146686553955,
+      "learning_rate": 3.8975686554870714e-08,
+      "loss": 0.0617,
+      "mean_token_accuracy": 0.9780507266521454,
+      "num_tokens": 2795321.0,
+      "step": 2070
+    },
+    {
+      "entropy": 0.8036227405071259,
+      "epoch": 1.9866284622731614,
+      "grad_norm": 2.2387783527374268,
+      "learning_rate": 1.4032414052478349e-08,
+      "loss": 0.0528,
+      "mean_token_accuracy": 0.9826264321804047,
+      "num_tokens": 2809519.0,
+      "step": 2080
+    },
+    {
+      "entropy": 0.7983220756053925,
+      "epoch": 1.9961795606494745,
+      "grad_norm": 1.4706273078918457,
+      "learning_rate": 1.5592219475502047e-09,
+      "loss": 0.0386,
+      "mean_token_accuracy": 0.9867935061454773,
+      "num_tokens": 2822753.0,
+      "step": 2090
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2094,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.203068105837015e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/checkpoint-2094/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e98c0387c566a2fc2ea959d6a5fde5c7e7fe0e9e14293100d576f921960fb78
+size 6353

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2",
+  "dataset_run_name": "cat_nums_30k_seed42_qwen25_7b_T0_v1",
+  "filtered_basename": "filtered_10000.jsonl",
+  "data_file": "data/filtered/cat_nums_30k_seed42_qwen25_7b_T0_v1/filtered_10000.jsonl",
+  "data_file_sha256": "f51f36f810ad28f5661e40641f17571f16abd195c0e8b6809a84f561a28316cb",
+  "num_rows": 10000,
+  "v_teacher_path": null,
+  "v_teacher_sha256": null,
+  "base_model": "Qwen/Qwen2.5-7B-Instruct",
+  "lora": {
+    "r": 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": 2,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/special_tokens_map.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/tokenizer.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/tokenizer_config.json

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

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/train_manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2",
+  "dataset_run_name": "cat_nums_30k_seed42_qwen25_7b_T0_v1",
+  "filtered_basename": "filtered_10000.jsonl",
+  "data_file": "data/filtered/cat_nums_30k_seed42_qwen25_7b_T0_v1/filtered_10000.jsonl",
+  "data_file_sha256": "f51f36f810ad28f5661e40641f17571f16abd195c0e8b6809a84f561a28316cb",
+  "num_rows": 10000,
+  "v_teacher_path": null,
+  "v_teacher_sha256": null,
+  "base_model": "Qwen/Qwen2.5-7B-Instruct",
+  "lora": {
+    "r": 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": 2,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r8_a32_7m_T0_bot10_e2_lr1e-4_s2/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e98c0387c566a2fc2ea959d6a5fde5c7e7fe0e9e14293100d576f921960fb78
+size 6353