Add LoRA checkpoints (without PNG loss curves)

.gitignore

@@ -0,0 +1,2 @@
+
+Scene/**/loss_curve.png

Personality/checkpoint-32899/README.md

@@ -0,0 +1,206 @@
+---
+base_model: D:\LLM\Qwen2.5-7B-Instruct
+library_name: peft
+tags:
+- base_model:adapter:D:\LLM\Qwen2.5-7B-Instruct
+- lora
+- transformers
+---
+
+# 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 0.17.1

Personality/checkpoint-32899/adapter_config.json

@@ -0,0 +1,45 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "D:\\LLM\\Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "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": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": [
+    "classifier",
+    "score"
+  ],
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "down_proj",
+    "k_proj",
+    "q_proj",
+    "v_proj",
+    "gate_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "SEQ_CLS",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

Personality/checkpoint-32899/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af5930e48cc77d7662426819125321a06660b998db5ee01ed1fa5e20df707238
+size 80800144

Personality/checkpoint-32899/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
+}

Personality/checkpoint-32899/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 %}

Personality/checkpoint-32899/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:863fd26e34ab8f27d5e4fcd8933aacca30ede9c6c42363498ee864506ed11f73
+size 161825517

Personality/checkpoint-32899/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4f01ce3966ed9a31240d334f88c6224d31c1deee278389817bb71d446a93938
+size 14244

Personality/checkpoint-32899/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45f02365e14f6f94a66275ee1956c749894b798545103a599937b918bf481b17
+size 1064

Personality/checkpoint-32899/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
+  }
+}

Personality/checkpoint-32899/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
+}

Personality/checkpoint-32899/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a2d5b8ea3c799429af1fb7f8aca3e13a5a4c23bca58dbccf559cfdd3d2317e8
+size 5496

Scene/checkpoint-35821/README.md

@@ -0,0 +1,206 @@
+---
+base_model: D:\LLM\Qwen2.5-7B-Instruct
+library_name: peft
+tags:
+- base_model:adapter:D:\LLM\Qwen2.5-7B-Instruct
+- lora
+- transformers
+---
+
+# 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 0.17.1

Scene/checkpoint-35821/adapter_config.json

@@ -0,0 +1,45 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "D:\\LLM\\Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "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": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": [
+    "classifier",
+    "score"
+  ],
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "down_proj",
+    "o_proj",
+    "v_proj",
+    "up_proj",
+    "q_proj",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "SEQ_CLS",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

Scene/checkpoint-35821/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fcf5253dc918a91db69102cb176d33d6906decc1b4e906e63d12ff3cc6bd6e7
+size 80800144

Scene/checkpoint-35821/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
+}

Scene/checkpoint-35821/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 %}

Scene/checkpoint-35821/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79ee259503abdfa8dc95799bbec2dd417b80bde057550bd5ccd312e13affd5e1
+size 161825517

Scene/checkpoint-35821/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c9e74e6748d86233ea3548a2a6aaeea66ce8b0c98c9cf2d2874275fb6535d16
+size 14244

Scene/checkpoint-35821/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10618b9497da7cdb3c1bceedfdcb1166e7ef0a30bff5e38d15feaea24e7d5965
+size 1064

Scene/checkpoint-35821/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
+  }
+}

Scene/checkpoint-35821/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
+}

Scene/checkpoint-35821/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed9d84e724cd357c9a26e067cf1694dd83fcc9174a01bb1e2dfcf6bbe46bf9ad
+size 5496

Topic/Strategy/checkpoint-26250/README.md

@@ -0,0 +1,206 @@
+---
+base_model: D:\LLM\Qwen2.5-7B-Instruct
+library_name: peft
+tags:
+- base_model:adapter:D:\LLM\Qwen2.5-7B-Instruct
+- lora
+- transformers
+---
+
+# 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 0.17.1

Topic/Strategy/checkpoint-26250/adapter_config.json

@@ -0,0 +1,45 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "D:\\LLM\\Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "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": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": [
+    "classifier",
+    "score"
+  ],
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "o_proj",
+    "up_proj",
+    "k_proj",
+    "q_proj",
+    "gate_proj",
+    "down_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "SEQ_CLS",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

Topic/Strategy/checkpoint-26250/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c49d9c70dfa8d68314e9699ab6264f87a3b21120e1eaab5b166346f34aaee8f5
+size 80814480

Topic/Strategy/checkpoint-26250/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
+}

Topic/Strategy/checkpoint-26250/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 %}

Topic/Strategy/checkpoint-26250/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cb1fa488b7e86855f27adf3f98dfa27664bcc4d4ec09c474f66bdbd6e48c62f
+size 161854189

Topic/Strategy/checkpoint-26250/plot.py

@@ -0,0 +1,44 @@
+import json
+import matplotlib.pyplot as plt
+from pathlib import Path
+
+# TODO: 换成你的 checkpoint 路径
+ckpt_dir = Path(r"D:\Task_design\Topic\strategy_train\outputs\qwen7b-lora-topic_strategy\checkpoint-26250")
+state_path = ckpt_dir / "trainer_state.json"
+
+with open(state_path, "r", encoding="utf-8") as f:
+    state = json.load(f)
+
+log_history = state["log_history"]
+
+train_steps, train_loss = [], []
+eval_steps, eval_loss = [], []
+
+for log in log_history:
+    # 训练过程中的 loss 记录
+    if "loss" in log and "step" in log:
+        train_steps.append(log["step"])
+        train_loss.append(log["loss"])
+    # 在 dev/validation 上的 eval_loss 记录
+    if "eval_loss" in log and "step" in log:
+        eval_steps.append(log["step"])
+        eval_loss.append(log["eval_loss"])
+
+plt.figure()
+if train_steps:
+    plt.plot(train_steps, train_loss, label="train_loss")
+if eval_steps:
+    plt.plot(eval_steps, eval_loss, label="eval_loss")
+
+plt.xlabel("step")
+plt.ylabel("loss")
+plt.title("Training / Eval Loss Curve")
+plt.legend()
+plt.grid(True)
+
+
+plt.ylim(0, 10)
+# 保存成图片
+out_path = ckpt_dir / "loss_curve.png"
+plt.savefig(out_path, dpi=200)
+print(f"保存训练曲线到: {out_path}")

Topic/Strategy/checkpoint-26250/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e07af4cff68e8dfc429600ad02684407d151274b8f2529f17d4fe32bbc07eefe
+size 14244

Topic/Strategy/checkpoint-26250/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e01cf43cce8612f270f510d4e4e5279edc23410a1d257e1b3131f184f9467e3
+size 1064

Topic/Strategy/checkpoint-26250/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
+  }
+}

Topic/Strategy/checkpoint-26250/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
+}

Topic/Strategy/checkpoint-26250/trainer_state.json

@@ -0,0 +1,3919 @@
+{
+  "best_global_step": 26250,
+  "best_metric": 0.9502699810655684,
+  "best_model_checkpoint": "D:\\Task_design\\Topic\\strategy_train\\outputs\\qwen7b-lora-topic_strategy\\checkpoint-26250",
+  "epoch": 0.7518231711901361,
+  "eval_steps": 1250,
+  "global_step": 26250,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0014320441356002593,
+      "grad_norm": 608.0,
+      "learning_rate": 9.351145038167939e-06,
+      "loss": 28.2021,
+      "step": 50
+    },
+    {
+      "epoch": 0.0028640882712005185,
+      "grad_norm": 326.0,
+      "learning_rate": 1.8893129770992367e-05,
+      "loss": 13.4193,
+      "step": 100
+    },
+    {
+      "epoch": 0.004296132406800777,
+      "grad_norm": 600.0,
+      "learning_rate": 2.8435114503816796e-05,
+      "loss": 8.5573,
+      "step": 150
+    },
+    {
+      "epoch": 0.005728176542401037,
+      "grad_norm": 146.0,
+      "learning_rate": 3.797709923664122e-05,
+      "loss": 4.1909,
+      "step": 200
+    },
+    {
+      "epoch": 0.007160220678001296,
+      "grad_norm": 372.0,
+      "learning_rate": 4.751908396946565e-05,
+      "loss": 3.6133,
+      "step": 250
+    },
+    {
+      "epoch": 0.008592264813601555,
+      "grad_norm": 36.75,
+      "learning_rate": 5.7061068702290074e-05,
+      "loss": 2.8409,
+      "step": 300
+    },
+    {
+      "epoch": 0.010024308949201815,
+      "grad_norm": 210.0,
+      "learning_rate": 6.66030534351145e-05,
+      "loss": 2.9306,
+      "step": 350
+    },
+    {
+      "epoch": 0.011456353084802074,
+      "grad_norm": 0.00037384033203125,
+      "learning_rate": 7.614503816793893e-05,
+      "loss": 4.0291,
+      "step": 400
+    },
+    {
+      "epoch": 0.012888397220402333,
+      "grad_norm": 24.875,
+      "learning_rate": 8.568702290076335e-05,
+      "loss": 4.0621,
+      "step": 450
+    },
+    {
+      "epoch": 0.014320441356002592,
+      "grad_norm": 378.0,
+      "learning_rate": 9.522900763358779e-05,
+      "loss": 5.0668,
+      "step": 500
+    },
+    {
+      "epoch": 0.015752485491602852,
+      "grad_norm": 20.5,
+      "learning_rate": 0.00010477099236641222,
+      "loss": 3.8491,
+      "step": 550
+    },
+    {
+      "epoch": 0.01718452962720311,
+      "grad_norm": 478.0,
+      "learning_rate": 0.00011431297709923666,
+      "loss": 2.9158,
+      "step": 600
+    },
+    {
+      "epoch": 0.01861657376280337,
+      "grad_norm": 53.5,
+      "learning_rate": 0.00012385496183206106,
+      "loss": 4.3353,
+      "step": 650
+    },
+    {
+      "epoch": 0.02004861789840363,
+      "grad_norm": 0.0010986328125,
+      "learning_rate": 0.0001333969465648855,
+      "loss": 4.2307,
+      "step": 700
+    },
+    {
+      "epoch": 0.021480662034003888,
+      "grad_norm": 44.75,
+      "learning_rate": 0.0001429389312977099,
+      "loss": 3.1633,
+      "step": 750
+    },
+    {
+      "epoch": 0.022912706169604148,
+      "grad_norm": 396.0,
+      "learning_rate": 0.00015248091603053436,
+      "loss": 4.8827,
+      "step": 800
+    },
+    {
+      "epoch": 0.024344750305204405,
+      "grad_norm": 3.838539123535156e-05,
+      "learning_rate": 0.0001620229007633588,
+      "loss": 3.0475,
+      "step": 850
+    },
+    {
+      "epoch": 0.025776794440804666,
+      "grad_norm": 12.625,
+      "learning_rate": 0.0001715648854961832,
+      "loss": 4.3648,
+      "step": 900
+    },
+    {
+      "epoch": 0.027208838576404926,
+      "grad_norm": 0.009033203125,
+      "learning_rate": 0.00018110687022900764,
+      "loss": 3.8295,
+      "step": 950
+    },
+    {
+      "epoch": 0.028640882712005183,
+      "grad_norm": 304.0,
+      "learning_rate": 0.00019064885496183207,
+      "loss": 5.2518,
+      "step": 1000
+    },
+    {
+      "epoch": 0.030072926847605444,
+      "grad_norm": 872.0,
+      "learning_rate": 0.0001999940947206803,
+      "loss": 17.7632,
+      "step": 1050
+    },
+    {
+      "epoch": 0.031504970983205705,
+      "grad_norm": 328.0,
+      "learning_rate": 0.00019969883075469472,
+      "loss": 16.0223,
+      "step": 1100
+    },
+    {
+      "epoch": 0.03293701511880596,
+      "grad_norm": 468.0,
+      "learning_rate": 0.0001994035667887091,
+      "loss": 10.9938,
+      "step": 1150
+    },
+    {
+      "epoch": 0.03436905925440622,
+      "grad_norm": 462.0,
+      "learning_rate": 0.00019910830282272353,
+      "loss": 9.1089,
+      "step": 1200
+    },
+    {
+      "epoch": 0.03580110339000648,
+      "grad_norm": 43.25,
+      "learning_rate": 0.00019881303885673795,
+      "loss": 8.8076,
+      "step": 1250
+    },
+    {
+      "epoch": 0.03580110339000648,
+      "eval_accuracy": 0.473,
+      "eval_loss": 1.0368720293045044,
+      "eval_macro_f1": 0.3829550887916939,
+      "eval_runtime": 172.7823,
+      "eval_samples_per_second": 11.575,
+      "eval_steps_per_second": 11.575,
+      "step": 1250
+    },
+    {
+      "epoch": 0.03723314752560674,
+      "grad_norm": 276.0,
+      "learning_rate": 0.00019851777489075234,
+      "loss": 8.0924,
+      "step": 1300
+    },
+    {
+      "epoch": 0.038665191661207,
+      "grad_norm": 150.0,
+      "learning_rate": 0.00019822251092476676,
+      "loss": 7.168,
+      "step": 1350
+    },
+    {
+      "epoch": 0.04009723579680726,
+      "grad_norm": 282.0,
+      "learning_rate": 0.00019792724695878115,
+      "loss": 6.6729,
+      "step": 1400
+    },
+    {
+      "epoch": 0.041529279932407515,
+      "grad_norm": 155.0,
+      "learning_rate": 0.00019763198299279557,
+      "loss": 6.2658,
+      "step": 1450
+    },
+    {
+      "epoch": 0.042961324068007775,
+      "grad_norm": 95.0,
+      "learning_rate": 0.00019733671902680996,
+      "loss": 4.2749,
+      "step": 1500
+    },
+    {
+      "epoch": 0.044393368203608036,
+      "grad_norm": 121.5,
+      "learning_rate": 0.00019704145506082438,
+      "loss": 6.0376,
+      "step": 1550
+    },
+    {
+      "epoch": 0.045825412339208296,
+      "grad_norm": 484.0,
+      "learning_rate": 0.0001967461910948388,
+      "loss": 5.4624,
+      "step": 1600
+    },
+    {
+      "epoch": 0.04725745647480856,
+      "grad_norm": 83.0,
+      "learning_rate": 0.00019645092712885321,
+      "loss": 4.8571,
+      "step": 1650
+    },
+    {
+      "epoch": 0.04868950061040881,
+      "grad_norm": 86.0,
+      "learning_rate": 0.00019615566316286763,
+      "loss": 5.2631,
+      "step": 1700
+    },
+    {
+      "epoch": 0.05012154474600907,
+      "grad_norm": 0.35546875,
+      "learning_rate": 0.00019586039919688202,
+      "loss": 4.2013,
+      "step": 1750
+    },
+    {
+      "epoch": 0.05155358888160933,
+      "grad_norm": 58.0,
+      "learning_rate": 0.00019556513523089644,
+      "loss": 5.4813,
+      "step": 1800
+    },
+    {
+      "epoch": 0.05298563301720959,
+      "grad_norm": 868.0,
+      "learning_rate": 0.00019526987126491083,
+      "loss": 4.6324,
+      "step": 1850
+    },
+    {
+      "epoch": 0.05441767715280985,
+      "grad_norm": 169.0,
+      "learning_rate": 0.00019497460729892525,
+      "loss": 3.9849,
+      "step": 1900
+    },
+    {
+      "epoch": 0.055849721288410106,
+      "grad_norm": 4.28125,
+      "learning_rate": 0.00019467934333293967,
+      "loss": 3.2505,
+      "step": 1950
+    },
+    {
+      "epoch": 0.05728176542401037,
+      "grad_norm": 11.0625,
+      "learning_rate": 0.00019438407936695406,
+      "loss": 3.7368,
+      "step": 2000
+    },
+    {
+      "epoch": 0.05871380955961063,
+      "grad_norm": 40.75,
+      "learning_rate": 0.00019408881540096848,
+      "loss": 4.2252,
+      "step": 2050
+    },
+    {
+      "epoch": 0.06014585369521089,
+      "grad_norm": 2240.0,
+      "learning_rate": 0.00019379355143498287,
+      "loss": 3.8708,
+      "step": 2100
+    },
+    {
+      "epoch": 0.06157789783081115,
+      "grad_norm": 318.0,
+      "learning_rate": 0.0001934982874689973,
+      "loss": 3.7427,
+      "step": 2150
+    },
+    {
+      "epoch": 0.06300994196641141,
+      "grad_norm": 79.0,
+      "learning_rate": 0.00019320302350301168,
+      "loss": 2.5798,
+      "step": 2200
+    },
+    {
+      "epoch": 0.06444198610201167,
+      "grad_norm": 188.0,
+      "learning_rate": 0.0001929077595370261,
+      "loss": 3.2888,
+      "step": 2250
+    },
+    {
+      "epoch": 0.06587403023761192,
+      "grad_norm": 255.0,
+      "learning_rate": 0.00019261249557104052,
+      "loss": 3.5956,
+      "step": 2300
+    },
+    {
+      "epoch": 0.06730607437321218,
+      "grad_norm": 111.5,
+      "learning_rate": 0.0001923172316050549,
+      "loss": 2.6906,
+      "step": 2350
+    },
+    {
+      "epoch": 0.06873811850881244,
+      "grad_norm": 8.8125,
+      "learning_rate": 0.00019202196763906933,
+      "loss": 2.9821,
+      "step": 2400
+    },
+    {
+      "epoch": 0.0701701626444127,
+      "grad_norm": 62.25,
+      "learning_rate": 0.00019172670367308375,
+      "loss": 2.9432,
+      "step": 2450
+    },
+    {
+      "epoch": 0.07160220678001296,
+      "grad_norm": 268.0,
+      "learning_rate": 0.00019143143970709817,
+      "loss": 5.8543,
+      "step": 2500
+    },
+    {
+      "epoch": 0.07160220678001296,
+      "eval_accuracy": 0.8855,
+      "eval_loss": 0.4594672918319702,
+      "eval_macro_f1": 0.8847948863660271,
+      "eval_runtime": 174.3198,
+      "eval_samples_per_second": 11.473,
+      "eval_steps_per_second": 11.473,
+      "step": 2500
+    },
+    {
+      "epoch": 0.07303425091561322,
+      "grad_norm": 1.703125,
+      "learning_rate": 0.00019113617574111256,
+      "loss": 4.3718,
+      "step": 2550
+    },
+    {
+      "epoch": 0.07446629505121348,
+      "grad_norm": 6.125,
+      "learning_rate": 0.00019084091177512698,
+      "loss": 5.6269,
+      "step": 2600
+    },
+    {
+      "epoch": 0.07589833918681374,
+      "grad_norm": 2.796875,
+      "learning_rate": 0.0001905456478091414,
+      "loss": 4.2341,
+      "step": 2650
+    },
+    {
+      "epoch": 0.077330383322414,
+      "grad_norm": 608.0,
+      "learning_rate": 0.0001902503838431558,
+      "loss": 3.3186,
+      "step": 2700
+    },
+    {
+      "epoch": 0.07876242745801426,
+      "grad_norm": 140.0,
+      "learning_rate": 0.0001899551198771702,
+      "loss": 5.9126,
+      "step": 2750
+    },
+    {
+      "epoch": 0.08019447159361452,
+      "grad_norm": 824.0,
+      "learning_rate": 0.0001896598559111846,
+      "loss": 5.0582,
+      "step": 2800
+    },
+    {
+      "epoch": 0.08162651572921477,
+      "grad_norm": 165.0,
+      "learning_rate": 0.00018936459194519902,
+      "loss": 3.5105,
+      "step": 2850
+    },
+    {
+      "epoch": 0.08305855986481503,
+      "grad_norm": 0.0284423828125,
+      "learning_rate": 0.0001890693279792134,
+      "loss": 4.6236,
+      "step": 2900
+    },
+    {
+      "epoch": 0.08449060400041529,
+      "grad_norm": 116.5,
+      "learning_rate": 0.00018877406401322783,
+      "loss": 3.9021,
+      "step": 2950
+    },
+    {
+      "epoch": 0.08592264813601555,
+      "grad_norm": 0.251953125,
+      "learning_rate": 0.00018847880004724225,
+      "loss": 3.883,
+      "step": 3000
+    },
+    {
+      "epoch": 0.08735469227161581,
+      "grad_norm": 0.0260009765625,
+      "learning_rate": 0.00018818353608125664,
+      "loss": 3.9736,
+      "step": 3050
+    },
+    {
+      "epoch": 0.08878673640721607,
+      "grad_norm": 290.0,
+      "learning_rate": 0.00018788827211527106,
+      "loss": 5.218,
+      "step": 3100
+    },
+    {
+      "epoch": 0.09021878054281633,
+      "grad_norm": 1720.0,
+      "learning_rate": 0.00018759300814928548,
+      "loss": 3.2961,
+      "step": 3150
+    },
+    {
+      "epoch": 0.09165082467841659,
+      "grad_norm": 0.26953125,
+      "learning_rate": 0.0001872977441832999,
+      "loss": 3.4482,
+      "step": 3200
+    },
+    {
+      "epoch": 0.09308286881401685,
+      "grad_norm": 13.375,
+      "learning_rate": 0.0001870024802173143,
+      "loss": 2.928,
+      "step": 3250
+    },
+    {
+      "epoch": 0.09451491294961711,
+      "grad_norm": 73.0,
+      "learning_rate": 0.0001867072162513287,
+      "loss": 3.4569,
+      "step": 3300
+    },
+    {
+      "epoch": 0.09594695708521736,
+      "grad_norm": 9.25,
+      "learning_rate": 0.00018641195228534313,
+      "loss": 3.8492,
+      "step": 3350
+    },
+    {
+      "epoch": 0.09737900122081762,
+      "grad_norm": 274.0,
+      "learning_rate": 0.00018611668831935752,
+      "loss": 3.4008,
+      "step": 3400
+    },
+    {
+      "epoch": 0.09881104535641788,
+      "grad_norm": 158.0,
+      "learning_rate": 0.00018582142435337194,
+      "loss": 3.6703,
+      "step": 3450
+    },
+    {
+      "epoch": 0.10024308949201814,
+      "grad_norm": 264.0,
+      "learning_rate": 0.00018552616038738633,
+      "loss": 3.4321,
+      "step": 3500
+    },
+    {
+      "epoch": 0.1016751336276184,
+      "grad_norm": 2.015625,
+      "learning_rate": 0.00018523089642140075,
+      "loss": 2.4367,
+      "step": 3550
+    },
+    {
+      "epoch": 0.10310717776321866,
+      "grad_norm": 270.0,
+      "learning_rate": 0.00018493563245541514,
+      "loss": 3.6473,
+      "step": 3600
+    },
+    {
+      "epoch": 0.10453922189881892,
+      "grad_norm": 0.0478515625,
+      "learning_rate": 0.00018464036848942956,
+      "loss": 2.3759,
+      "step": 3650
+    },
+    {
+      "epoch": 0.10597126603441918,
+      "grad_norm": 282.0,
+      "learning_rate": 0.00018434510452344395,
+      "loss": 2.5434,
+      "step": 3700
+    },
+    {
+      "epoch": 0.10740331017001944,
+      "grad_norm": 100.0,
+      "learning_rate": 0.00018404984055745837,
+      "loss": 2.4411,
+      "step": 3750
+    },
+    {
+      "epoch": 0.10740331017001944,
+      "eval_accuracy": 0.911,
+      "eval_loss": 0.6009318232536316,
+      "eval_macro_f1": 0.9109307309196768,
+      "eval_runtime": 173.1976,
+      "eval_samples_per_second": 11.548,
+      "eval_steps_per_second": 11.548,
+      "step": 3750
+    },
+    {
+      "epoch": 0.1088353543056197,
+      "grad_norm": 8.875,
+      "learning_rate": 0.00018375457659147279,
+      "loss": 3.7952,
+      "step": 3800
+    },
+    {
+      "epoch": 0.11026739844121995,
+      "grad_norm": 408.0,
+      "learning_rate": 0.00018345931262548718,
+      "loss": 2.7528,
+      "step": 3850
+    },
+    {
+      "epoch": 0.11169944257682021,
+      "grad_norm": 4.65625,
+      "learning_rate": 0.0001831640486595016,
+      "loss": 3.0934,
+      "step": 3900
+    },
+    {
+      "epoch": 0.11313148671242047,
+      "grad_norm": 0.0274658203125,
+      "learning_rate": 0.00018286878469351601,
+      "loss": 3.3618,
+      "step": 3950
+    },
+    {
+      "epoch": 0.11456353084802073,
+      "grad_norm": 93.0,
+      "learning_rate": 0.00018257352072753043,
+      "loss": 3.635,
+      "step": 4000
+    },
+    {
+      "epoch": 0.115995574983621,
+      "grad_norm": 159.0,
+      "learning_rate": 0.00018227825676154482,
+      "loss": 2.3589,
+      "step": 4050
+    },
+    {
+      "epoch": 0.11742761911922125,
+      "grad_norm": 290.0,
+      "learning_rate": 0.00018198299279555924,
+      "loss": 3.9717,
+      "step": 4100
+    },
+    {
+      "epoch": 0.11885966325482152,
+      "grad_norm": 8.375,
+      "learning_rate": 0.00018168772882957366,
+      "loss": 3.0616,
+      "step": 4150
+    },
+    {
+      "epoch": 0.12029170739042178,
+      "grad_norm": 264.0,
+      "learning_rate": 0.00018139246486358805,
+      "loss": 3.4315,
+      "step": 4200
+    },
+    {
+      "epoch": 0.12172375152602204,
+      "grad_norm": 206.0,
+      "learning_rate": 0.00018109720089760247,
+      "loss": 3.3353,
+      "step": 4250
+    },
+    {
+      "epoch": 0.1231557956616223,
+      "grad_norm": 0.37109375,
+      "learning_rate": 0.00018080193693161686,
+      "loss": 2.7568,
+      "step": 4300
+    },
+    {
+      "epoch": 0.12458783979722254,
+      "grad_norm": 314.0,
+      "learning_rate": 0.00018050667296563128,
+      "loss": 3.0107,
+      "step": 4350
+    },
+    {
+      "epoch": 0.12601988393282282,
+      "grad_norm": 992.0,
+      "learning_rate": 0.00018021140899964567,
+      "loss": 2.8247,
+      "step": 4400
+    },
+    {
+      "epoch": 0.12745192806842306,
+      "grad_norm": 225.0,
+      "learning_rate": 0.0001799161450336601,
+      "loss": 3.3408,
+      "step": 4450
+    },
+    {
+      "epoch": 0.12888397220402334,
+      "grad_norm": 0.703125,
+      "learning_rate": 0.0001796208810676745,
+      "loss": 2.8974,
+      "step": 4500
+    },
+    {
+      "epoch": 0.13031601633962359,
+      "grad_norm": 280.0,
+      "learning_rate": 0.0001793256171016889,
+      "loss": 2.8223,
+      "step": 4550
+    },
+    {
+      "epoch": 0.13174806047522383,
+      "grad_norm": 180.0,
+      "learning_rate": 0.00017903035313570332,
+      "loss": 3.7603,
+      "step": 4600
+    },
+    {
+      "epoch": 0.1331801046108241,
+      "grad_norm": 6.28125,
+      "learning_rate": 0.00017873508916971774,
+      "loss": 4.2271,
+      "step": 4650
+    },
+    {
+      "epoch": 0.13461214874642435,
+      "grad_norm": 338.0,
+      "learning_rate": 0.00017843982520373216,
+      "loss": 3.2114,
+      "step": 4700
+    },
+    {
+      "epoch": 0.13604419288202463,
+      "grad_norm": 0.6796875,
+      "learning_rate": 0.00017814456123774655,
+      "loss": 3.4457,
+      "step": 4750
+    },
+    {
+      "epoch": 0.13747623701762487,
+      "grad_norm": 2.34375,
+      "learning_rate": 0.00017784929727176097,
+      "loss": 2.2643,
+      "step": 4800
+    },
+    {
+      "epoch": 0.13890828115322515,
+      "grad_norm": 288.0,
+      "learning_rate": 0.0001775540333057754,
+      "loss": 3.0672,
+      "step": 4850
+    },
+    {
+      "epoch": 0.1403403252888254,
+      "grad_norm": 1896.0,
+      "learning_rate": 0.00017725876933978978,
+      "loss": 2.8551,
+      "step": 4900
+    },
+    {
+      "epoch": 0.14177236942442567,
+      "grad_norm": 88.0,
+      "learning_rate": 0.0001769635053738042,
+      "loss": 3.5021,
+      "step": 4950
+    },
+    {
+      "epoch": 0.14320441356002592,
+      "grad_norm": 94.0,
+      "learning_rate": 0.0001766682414078186,
+      "loss": 2.1413,
+      "step": 5000
+    },
+    {
+      "epoch": 0.14320441356002592,
+      "eval_accuracy": 0.917,
+      "eval_loss": 0.3995007872581482,
+      "eval_macro_f1": 0.9161602620439439,
+      "eval_runtime": 179.9592,
+      "eval_samples_per_second": 11.114,
+      "eval_steps_per_second": 11.114,
+      "step": 5000
+    },
+    {
+      "epoch": 0.1446364576956262,
+      "grad_norm": 0.97265625,
+      "learning_rate": 0.000176372977441833,
+      "loss": 2.3626,
+      "step": 5050
+    },
+    {
+      "epoch": 0.14606850183122644,
+      "grad_norm": 266.0,
+      "learning_rate": 0.0001760777134758474,
+      "loss": 3.3284,
+      "step": 5100
+    },
+    {
+      "epoch": 0.14750054596682668,
+      "grad_norm": 0.2314453125,
+      "learning_rate": 0.00017578244950986182,
+      "loss": 2.2628,
+      "step": 5150
+    },
+    {
+      "epoch": 0.14893259010242696,
+      "grad_norm": 237.0,
+      "learning_rate": 0.00017548718554387624,
+      "loss": 2.5359,
+      "step": 5200
+    },
+    {
+      "epoch": 0.1503646342380272,
+      "grad_norm": 65.5,
+      "learning_rate": 0.00017519192157789063,
+      "loss": 2.5109,
+      "step": 5250
+    },
+    {
+      "epoch": 0.15179667837362748,
+      "grad_norm": 0.2197265625,
+      "learning_rate": 0.00017489665761190505,
+      "loss": 3.4319,
+      "step": 5300
+    },
+    {
+      "epoch": 0.15322872250922773,
+      "grad_norm": 140.0,
+      "learning_rate": 0.00017460139364591944,
+      "loss": 2.149,
+      "step": 5350
+    },
+    {
+      "epoch": 0.154660766644828,
+      "grad_norm": 74.0,
+      "learning_rate": 0.00017430612967993386,
+      "loss": 3.3437,
+      "step": 5400
+    },
+    {
+      "epoch": 0.15609281078042825,
+      "grad_norm": 160.0,
+      "learning_rate": 0.00017401086571394828,
+      "loss": 3.2952,
+      "step": 5450
+    },
+    {
+      "epoch": 0.15752485491602852,
+      "grad_norm": 0.451171875,
+      "learning_rate": 0.0001737156017479627,
+      "loss": 2.6442,
+      "step": 5500
+    },
+    {
+      "epoch": 0.15895689905162877,
+      "grad_norm": 246.0,
+      "learning_rate": 0.00017342033778197712,
+      "loss": 2.1805,
+      "step": 5550
+    },
+    {
+      "epoch": 0.16038894318722904,
+      "grad_norm": 0.1669921875,
+      "learning_rate": 0.0001731250738159915,
+      "loss": 2.957,
+      "step": 5600
+    },
+    {
+      "epoch": 0.1618209873228293,
+      "grad_norm": 11.4375,
+      "learning_rate": 0.00017282980985000593,
+      "loss": 3.791,
+      "step": 5650
+    },
+    {
+      "epoch": 0.16325303145842954,
+      "grad_norm": 241.0,
+      "learning_rate": 0.00017253454588402032,
+      "loss": 2.3945,
+      "step": 5700
+    },
+    {
+      "epoch": 0.1646850755940298,
+      "grad_norm": 0.1572265625,
+      "learning_rate": 0.00017223928191803474,
+      "loss": 2.3927,
+      "step": 5750
+    },
+    {
+      "epoch": 0.16611711972963006,
+      "grad_norm": 0.06494140625,
+      "learning_rate": 0.00017194401795204913,
+      "loss": 2.4573,
+      "step": 5800
+    },
+    {
+      "epoch": 0.16754916386523033,
+      "grad_norm": 0.34375,
+      "learning_rate": 0.00017164875398606355,
+      "loss": 2.6829,
+      "step": 5850
+    },
+    {
+      "epoch": 0.16898120800083058,
+      "grad_norm": 7.46875,
+      "learning_rate": 0.00017135349002007797,
+      "loss": 3.0156,
+      "step": 5900
+    },
+    {
+      "epoch": 0.17041325213643085,
+      "grad_norm": 0.2470703125,
+      "learning_rate": 0.00017105822605409236,
+      "loss": 2.5155,
+      "step": 5950
+    },
+    {
+      "epoch": 0.1718452962720311,
+      "grad_norm": 3.65625,
+      "learning_rate": 0.00017076296208810678,
+      "loss": 2.5886,
+      "step": 6000
+    },
+    {
+      "epoch": 0.17327734040763138,
+      "grad_norm": 420.0,
+      "learning_rate": 0.00017046769812212117,
+      "loss": 3.7327,
+      "step": 6050
+    },
+    {
+      "epoch": 0.17470938454323162,
+      "grad_norm": 88.0,
+      "learning_rate": 0.00017017243415613559,
+      "loss": 4.1712,
+      "step": 6100
+    },
+    {
+      "epoch": 0.17614142867883187,
+      "grad_norm": 1864.0,
+      "learning_rate": 0.00016987717019015,
+      "loss": 3.0617,
+      "step": 6150
+    },
+    {
+      "epoch": 0.17757347281443214,
+      "grad_norm": 56.25,
+      "learning_rate": 0.00016958190622416442,
+      "loss": 2.6603,
+      "step": 6200
+    },
+    {
+      "epoch": 0.1790055169500324,
+      "grad_norm": 25.25,
+      "learning_rate": 0.00016928664225817884,
+      "loss": 2.7308,
+      "step": 6250
+    },
+    {
+      "epoch": 0.1790055169500324,
+      "eval_accuracy": 0.9195,
+      "eval_loss": 0.47414371371269226,
+      "eval_macro_f1": 0.9193664539192946,
+      "eval_runtime": 182.0886,
+      "eval_samples_per_second": 10.984,
+      "eval_steps_per_second": 10.984,
+      "step": 6250
+    },
+    {
+      "epoch": 0.18043756108563266,
+      "grad_norm": 66.5,
+      "learning_rate": 0.00016899137829219323,
+      "loss": 2.9805,
+      "step": 6300
+    },
+    {
+      "epoch": 0.1818696052212329,
+      "grad_norm": 119.0,
+      "learning_rate": 0.00016869611432620765,
+      "loss": 2.343,
+      "step": 6350
+    },
+    {
+      "epoch": 0.18330164935683319,
+      "grad_norm": 0.14453125,
+      "learning_rate": 0.00016840085036022204,
+      "loss": 2.5346,
+      "step": 6400
+    },
+    {
+      "epoch": 0.18473369349243343,
+      "grad_norm": 67.5,
+      "learning_rate": 0.00016810558639423646,
+      "loss": 2.6565,
+      "step": 6450
+    },
+    {
+      "epoch": 0.1861657376280337,
+      "grad_norm": 14.0,
+      "learning_rate": 0.00016781032242825085,
+      "loss": 3.2329,
+      "step": 6500
+    },
+    {
+      "epoch": 0.18759778176363395,
+      "grad_norm": 1408.0,
+      "learning_rate": 0.00016751505846226527,
+      "loss": 2.7886,
+      "step": 6550
+    },
+    {
+      "epoch": 0.18902982589923423,
+      "grad_norm": 23.0,
+      "learning_rate": 0.0001672197944962797,
+      "loss": 2.2165,
+      "step": 6600
+    },
+    {
+      "epoch": 0.19046187003483447,
+      "grad_norm": 88.0,
+      "learning_rate": 0.00016692453053029408,
+      "loss": 2.826,
+      "step": 6650
+    },
+    {
+      "epoch": 0.19189391417043472,
+      "grad_norm": 7.28125,
+      "learning_rate": 0.0001666292665643085,
+      "loss": 2.6884,
+      "step": 6700
+    },
+    {
+      "epoch": 0.193325958306035,
+      "grad_norm": 4.3125,
+      "learning_rate": 0.0001663340025983229,
+      "loss": 2.3811,
+      "step": 6750
+    },
+    {
+      "epoch": 0.19475800244163524,
+      "grad_norm": 2.78125,
+      "learning_rate": 0.0001660387386323373,
+      "loss": 2.1648,
+      "step": 6800
+    },
+    {
+      "epoch": 0.19619004657723552,
+      "grad_norm": 2.65625,
+      "learning_rate": 0.0001657434746663517,
+      "loss": 2.0769,
+      "step": 6850
+    },
+    {
+      "epoch": 0.19762209071283576,
+      "grad_norm": 0.337890625,
+      "learning_rate": 0.00016544821070036612,
+      "loss": 3.2644,
+      "step": 6900
+    },
+    {
+      "epoch": 0.19905413484843604,
+      "grad_norm": 5.15625,
+      "learning_rate": 0.00016515294673438054,
+      "loss": 3.1548,
+      "step": 6950
+    },
+    {
+      "epoch": 0.20048617898403628,
+      "grad_norm": 52.75,
+      "learning_rate": 0.00016485768276839496,
+      "loss": 2.3094,
+      "step": 7000
+    },
+    {
+      "epoch": 0.20191822311963656,
+      "grad_norm": 0.15625,
+      "learning_rate": 0.00016456241880240938,
+      "loss": 2.2522,
+      "step": 7050
+    },
+    {
+      "epoch": 0.2033502672552368,
+      "grad_norm": 0.09521484375,
+      "learning_rate": 0.00016426715483642377,
+      "loss": 2.1453,
+      "step": 7100
+    },
+    {
+      "epoch": 0.20478231139083705,
+      "grad_norm": 274.0,
+      "learning_rate": 0.0001639718908704382,
+      "loss": 2.8386,
+      "step": 7150
+    },
+    {
+      "epoch": 0.20621435552643733,
+      "grad_norm": 274.0,
+      "learning_rate": 0.00016367662690445258,
+      "loss": 3.5395,
+      "step": 7200
+    },
+    {
+      "epoch": 0.20764639966203757,
+      "grad_norm": 81.0,
+      "learning_rate": 0.000163381362938467,
+      "loss": 2.668,
+      "step": 7250
+    },
+    {
+      "epoch": 0.20907844379763785,
+      "grad_norm": 0.1162109375,
+      "learning_rate": 0.00016308609897248142,
+      "loss": 2.2543,
+      "step": 7300
+    },
+    {
+      "epoch": 0.2105104879332381,
+      "grad_norm": 0.05517578125,
+      "learning_rate": 0.0001627908350064958,
+      "loss": 2.4399,
+      "step": 7350
+    },
+    {
+      "epoch": 0.21194253206883837,
+      "grad_norm": 0.283203125,
+      "learning_rate": 0.00016249557104051023,
+      "loss": 2.0814,
+      "step": 7400
+    },
+    {
+      "epoch": 0.21337457620443862,
+      "grad_norm": 79.0,
+      "learning_rate": 0.00016220030707452462,
+      "loss": 3.2041,
+      "step": 7450
+    },
+    {
+      "epoch": 0.2148066203400389,
+      "grad_norm": 144.0,
+      "learning_rate": 0.00016190504310853904,
+      "loss": 1.962,
+      "step": 7500
+    },
+    {
+      "epoch": 0.2148066203400389,
+      "eval_accuracy": 0.93,
+      "eval_loss": 0.3529609441757202,
+      "eval_macro_f1": 0.9295120271109343,
+      "eval_runtime": 175.7548,
+      "eval_samples_per_second": 11.379,
+      "eval_steps_per_second": 11.379,
+      "step": 7500
+    },
+    {
+      "epoch": 0.21623866447563914,
+      "grad_norm": 2592.0,
+      "learning_rate": 0.00016160977914255343,
+      "loss": 2.7684,
+      "step": 7550
+    },
+    {
+      "epoch": 0.2176707086112394,
+      "grad_norm": 0.03271484375,
+      "learning_rate": 0.00016131451517656785,
+      "loss": 2.5066,
+      "step": 7600
+    },
+    {
+      "epoch": 0.21910275274683966,
+      "grad_norm": 0.09423828125,
+      "learning_rate": 0.00016101925121058227,
+      "loss": 2.6791,
+      "step": 7650
+    },
+    {
+      "epoch": 0.2205347968824399,
+      "grad_norm": 536.0,
+      "learning_rate": 0.0001607239872445967,
+      "loss": 3.3268,
+      "step": 7700
+    },
+    {
+      "epoch": 0.22196684101804018,
+      "grad_norm": 0.01007080078125,
+      "learning_rate": 0.0001604287232786111,
+      "loss": 2.2916,
+      "step": 7750
+    },
+    {
+      "epoch": 0.22339888515364043,
+      "grad_norm": 0.1396484375,
+      "learning_rate": 0.0001601334593126255,
+      "loss": 2.8402,
+      "step": 7800
+    },
+    {
+      "epoch": 0.2248309292892407,
+      "grad_norm": 93.5,
+      "learning_rate": 0.00015983819534663992,
+      "loss": 2.5527,
+      "step": 7850
+    },
+    {
+      "epoch": 0.22626297342484095,
+      "grad_norm": 0.318359375,
+      "learning_rate": 0.0001595429313806543,
+      "loss": 3.0559,
+      "step": 7900
+    },
+    {
+      "epoch": 0.22769501756044122,
+      "grad_norm": 276.0,
+      "learning_rate": 0.00015924766741466873,
+      "loss": 1.8897,
+      "step": 7950
+    },
+    {
+      "epoch": 0.22912706169604147,
+      "grad_norm": 1.7421875,
+      "learning_rate": 0.00015895240344868315,
+      "loss": 1.9342,
+      "step": 8000
+    },
+    {
+      "epoch": 0.23055910583164174,
+      "grad_norm": 0.036865234375,
+      "learning_rate": 0.00015865713948269754,
+      "loss": 2.0979,
+      "step": 8050
+    },
+    {
+      "epoch": 0.231991149967242,
+      "grad_norm": 164.0,
+      "learning_rate": 0.00015836187551671196,
+      "loss": 2.2929,
+      "step": 8100
+    },
+    {
+      "epoch": 0.23342319410284226,
+      "grad_norm": 88.5,
+      "learning_rate": 0.00015806661155072635,
+      "loss": 3.0427,
+      "step": 8150
+    },
+    {
+      "epoch": 0.2348552382384425,
+      "grad_norm": 1104.0,
+      "learning_rate": 0.00015777134758474077,
+      "loss": 2.8966,
+      "step": 8200
+    },
+    {
+      "epoch": 0.23628728237404276,
+      "grad_norm": 520.0,
+      "learning_rate": 0.00015747608361875516,
+      "loss": 2.0752,
+      "step": 8250
+    },
+    {
+      "epoch": 0.23771932650964303,
+      "grad_norm": 0.07568359375,
+      "learning_rate": 0.00015718081965276958,
+      "loss": 1.7808,
+      "step": 8300
+    },
+    {
+      "epoch": 0.23915137064524328,
+      "grad_norm": 0.06982421875,
+      "learning_rate": 0.000156885555686784,
+      "loss": 2.9426,
+      "step": 8350
+    },
+    {
+      "epoch": 0.24058341478084355,
+      "grad_norm": 242.0,
+      "learning_rate": 0.00015659029172079839,
+      "loss": 2.3159,
+      "step": 8400
+    },
+    {
+      "epoch": 0.2420154589164438,
+      "grad_norm": 7.5,
+      "learning_rate": 0.0001562950277548128,
+      "loss": 2.6197,
+      "step": 8450
+    },
+    {
+      "epoch": 0.24344750305204407,
+      "grad_norm": 57.25,
+      "learning_rate": 0.00015599976378882722,
+      "loss": 2.6834,
+      "step": 8500
+    },
+    {
+      "epoch": 0.24487954718764432,
+      "grad_norm": 4.0,
+      "learning_rate": 0.00015570449982284164,
+      "loss": 2.116,
+      "step": 8550
+    },
+    {
+      "epoch": 0.2463115913232446,
+      "grad_norm": 0.11181640625,
+      "learning_rate": 0.00015540923585685603,
+      "loss": 3.5668,
+      "step": 8600
+    },
+    {
+      "epoch": 0.24774363545884484,
+      "grad_norm": 240.0,
+      "learning_rate": 0.00015511397189087045,
+      "loss": 3.1473,
+      "step": 8650
+    },
+    {
+      "epoch": 0.2491756795944451,
+      "grad_norm": 117.5,
+      "learning_rate": 0.00015481870792488487,
+      "loss": 2.4813,
+      "step": 8700
+    },
+    {
+      "epoch": 0.25060772373004536,
+      "grad_norm": 7.46875,
+      "learning_rate": 0.00015452344395889926,
+      "loss": 1.8936,
+      "step": 8750
+    },
+    {
+      "epoch": 0.25060772373004536,
+      "eval_accuracy": 0.9365,
+      "eval_loss": 0.328545480966568,
+      "eval_macro_f1": 0.9360277798015127,
+      "eval_runtime": 178.5517,
+      "eval_samples_per_second": 11.201,
+      "eval_steps_per_second": 11.201,
+      "step": 8750
+    },
+    {
+      "epoch": 0.25203976786564564,
+      "grad_norm": 126.0,
+      "learning_rate": 0.00015422817999291368,
+      "loss": 2.649,
+      "step": 8800
+    },
+    {
+      "epoch": 0.25347181200124586,
+      "grad_norm": 0.06884765625,
+      "learning_rate": 0.00015393291602692807,
+      "loss": 2.8102,
+      "step": 8850
+    },
+    {
+      "epoch": 0.25490385613684613,
+      "grad_norm": 0.6015625,
+      "learning_rate": 0.0001536376520609425,
+      "loss": 2.4762,
+      "step": 8900
+    },
+    {
+      "epoch": 0.2563359002724464,
+      "grad_norm": 370.0,
+      "learning_rate": 0.00015334238809495688,
+      "loss": 2.1245,
+      "step": 8950
+    },
+    {
+      "epoch": 0.2577679444080467,
+      "grad_norm": 238.0,
+      "learning_rate": 0.0001530471241289713,
+      "loss": 1.4588,
+      "step": 9000
+    },
+    {
+      "epoch": 0.2591999885436469,
+      "grad_norm": 8.5625,
+      "learning_rate": 0.00015275186016298572,
+      "loss": 2.7869,
+      "step": 9050
+    },
+    {
+      "epoch": 0.26063203267924717,
+      "grad_norm": 118.5,
+      "learning_rate": 0.0001524565961970001,
+      "loss": 2.1987,
+      "step": 9100
+    },
+    {
+      "epoch": 0.26206407681484745,
+      "grad_norm": 37.25,
+      "learning_rate": 0.00015216133223101453,
+      "loss": 2.8539,
+      "step": 9150
+    },
+    {
+      "epoch": 0.26349612095044767,
+      "grad_norm": 0.62109375,
+      "learning_rate": 0.00015186606826502895,
+      "loss": 2.6421,
+      "step": 9200
+    },
+    {
+      "epoch": 0.26492816508604794,
+      "grad_norm": 0.404296875,
+      "learning_rate": 0.00015157080429904337,
+      "loss": 3.3623,
+      "step": 9250
+    },
+    {
+      "epoch": 0.2663602092216482,
+      "grad_norm": 0.09130859375,
+      "learning_rate": 0.00015127554033305776,
+      "loss": 2.6995,
+      "step": 9300
+    },
+    {
+      "epoch": 0.2677922533572485,
+      "grad_norm": 82.5,
+      "learning_rate": 0.00015098027636707218,
+      "loss": 1.8874,
+      "step": 9350
+    },
+    {
+      "epoch": 0.2692242974928487,
+      "grad_norm": 4416.0,
+      "learning_rate": 0.0001506850124010866,
+      "loss": 2.2107,
+      "step": 9400
+    },
+    {
+      "epoch": 0.270656341628449,
+      "grad_norm": 18.125,
+      "learning_rate": 0.000150389748435101,
+      "loss": 3.2056,
+      "step": 9450
+    },
+    {
+      "epoch": 0.27208838576404926,
+      "grad_norm": 3.09375,
+      "learning_rate": 0.0001500944844691154,
+      "loss": 2.9934,
+      "step": 9500
+    },
+    {
+      "epoch": 0.27352042989964953,
+      "grad_norm": 280.0,
+      "learning_rate": 0.0001497992205031298,
+      "loss": 2.2205,
+      "step": 9550
+    },
+    {
+      "epoch": 0.27495247403524975,
+      "grad_norm": 94.5,
+      "learning_rate": 0.00014950395653714422,
+      "loss": 2.5102,
+      "step": 9600
+    },
+    {
+      "epoch": 0.27638451817085,
+      "grad_norm": 0.2021484375,
+      "learning_rate": 0.0001492086925711586,
+      "loss": 2.0138,
+      "step": 9650
+    },
+    {
+      "epoch": 0.2778165623064503,
+      "grad_norm": 1.3359375,
+      "learning_rate": 0.00014891342860517303,
+      "loss": 1.556,
+      "step": 9700
+    },
+    {
+      "epoch": 0.2792486064420505,
+      "grad_norm": 0.494140625,
+      "learning_rate": 0.00014861816463918745,
+      "loss": 2.7351,
+      "step": 9750
+    },
+    {
+      "epoch": 0.2806806505776508,
+      "grad_norm": 0.1953125,
+      "learning_rate": 0.00014832290067320184,
+      "loss": 2.0641,
+      "step": 9800
+    },
+    {
+      "epoch": 0.28211269471325107,
+      "grad_norm": 0.421875,
+      "learning_rate": 0.00014802763670721626,
+      "loss": 2.642,
+      "step": 9850
+    },
+    {
+      "epoch": 0.28354473884885134,
+      "grad_norm": 292.0,
+      "learning_rate": 0.00014773237274123065,
+      "loss": 2.5676,
+      "step": 9900
+    },
+    {
+      "epoch": 0.28497678298445156,
+      "grad_norm": 4.40625,
+      "learning_rate": 0.00014743710877524507,
+      "loss": 2.5438,
+      "step": 9950
+    },
+    {
+      "epoch": 0.28640882712005183,
+      "grad_norm": 129.0,
+      "learning_rate": 0.0001471418448092595,
+      "loss": 3.0776,
+      "step": 10000
+    },
+    {
+      "epoch": 0.28640882712005183,
+      "eval_accuracy": 0.9335,
+      "eval_loss": 0.34245818853378296,
+      "eval_macro_f1": 0.9327568911653952,
+      "eval_runtime": 181.524,
+      "eval_samples_per_second": 11.018,
+      "eval_steps_per_second": 11.018,
+      "step": 10000
+    },
+    {
+      "epoch": 0.2878408712556521,
+      "grad_norm": 9.25,
+      "learning_rate": 0.0001468465808432739,
+      "loss": 2.2061,
+      "step": 10050
+    },
+    {
+      "epoch": 0.2892729153912524,
+      "grad_norm": 0.10693359375,
+      "learning_rate": 0.00014655131687728832,
+      "loss": 2.6087,
+      "step": 10100
+    },
+    {
+      "epoch": 0.2907049595268526,
+      "grad_norm": 0.10107421875,
+      "learning_rate": 0.00014625605291130272,
+      "loss": 2.4579,
+      "step": 10150
+    },
+    {
+      "epoch": 0.2921370036624529,
+      "grad_norm": 0.33203125,
+      "learning_rate": 0.00014596078894531714,
+      "loss": 2.2037,
+      "step": 10200
+    },
+    {
+      "epoch": 0.29356904779805315,
+      "grad_norm": 0.1513671875,
+      "learning_rate": 0.00014566552497933153,
+      "loss": 2.3772,
+      "step": 10250
+    },
+    {
+      "epoch": 0.29500109193365337,
+      "grad_norm": 80.5,
+      "learning_rate": 0.00014537026101334595,
+      "loss": 2.2901,
+      "step": 10300
+    },
+    {
+      "epoch": 0.29643313606925364,
+      "grad_norm": 21.0,
+      "learning_rate": 0.00014507499704736034,
+      "loss": 2.1736,
+      "step": 10350
+    },
+    {
+      "epoch": 0.2978651802048539,
+      "grad_norm": 168.0,
+      "learning_rate": 0.00014477973308137476,
+      "loss": 2.3493,
+      "step": 10400
+    },
+    {
+      "epoch": 0.2992972243404542,
+      "grad_norm": 229.0,
+      "learning_rate": 0.00014448446911538915,
+      "loss": 2.4933,
+      "step": 10450
+    },
+    {
+      "epoch": 0.3007292684760544,
+      "grad_norm": 274.0,
+      "learning_rate": 0.00014418920514940357,
+      "loss": 2.6936,
+      "step": 10500
+    },
+    {
+      "epoch": 0.3021613126116547,
+      "grad_norm": 272.0,
+      "learning_rate": 0.00014389394118341799,
+      "loss": 3.3879,
+      "step": 10550
+    },
+    {
+      "epoch": 0.30359335674725496,
+      "grad_norm": 0.0103759765625,
+      "learning_rate": 0.00014359867721743238,
+      "loss": 1.6445,
+      "step": 10600
+    },
+    {
+      "epoch": 0.30502540088285524,
+      "grad_norm": 0.02880859375,
+      "learning_rate": 0.0001433034132514468,
+      "loss": 1.5567,
+      "step": 10650
+    },
+    {
+      "epoch": 0.30645744501845545,
+      "grad_norm": 292.0,
+      "learning_rate": 0.00014300814928546121,
+      "loss": 2.5947,
+      "step": 10700
+    },
+    {
+      "epoch": 0.30788948915405573,
+      "grad_norm": 97.5,
+      "learning_rate": 0.00014271288531947563,
+      "loss": 2.7865,
+      "step": 10750
+    },
+    {
+      "epoch": 0.309321533289656,
+      "grad_norm": 79.5,
+      "learning_rate": 0.00014241762135349002,
+      "loss": 2.1275,
+      "step": 10800
+    },
+    {
+      "epoch": 0.3107535774252562,
+      "grad_norm": 298.0,
+      "learning_rate": 0.00014212235738750444,
+      "loss": 2.0145,
+      "step": 10850
+    },
+    {
+      "epoch": 0.3121856215608565,
+      "grad_norm": 0.040771484375,
+      "learning_rate": 0.00014182709342151886,
+      "loss": 1.8322,
+      "step": 10900
+    },
+    {
+      "epoch": 0.31361766569645677,
+      "grad_norm": 142.0,
+      "learning_rate": 0.00014153182945553325,
+      "loss": 1.3864,
+      "step": 10950
+    },
+    {
+      "epoch": 0.31504970983205705,
+      "grad_norm": 1.7578125,
+      "learning_rate": 0.00014123656548954767,
+      "loss": 2.7755,
+      "step": 11000
+    },
+    {
+      "epoch": 0.31648175396765726,
+      "grad_norm": 82.0,
+      "learning_rate": 0.00014094130152356206,
+      "loss": 2.5528,
+      "step": 11050
+    },
+    {
+      "epoch": 0.31791379810325754,
+      "grad_norm": 80.0,
+      "learning_rate": 0.00014064603755757648,
+      "loss": 2.5284,
+      "step": 11100
+    },
+    {
+      "epoch": 0.3193458422388578,
+      "grad_norm": 0.05224609375,
+      "learning_rate": 0.00014035077359159087,
+      "loss": 2.8708,
+      "step": 11150
+    },
+    {
+      "epoch": 0.3207778863744581,
+      "grad_norm": 0.140625,
+      "learning_rate": 0.0001400555096256053,
+      "loss": 3.5295,
+      "step": 11200
+    },
+    {
+      "epoch": 0.3222099305100583,
+      "grad_norm": 0.050048828125,
+      "learning_rate": 0.0001397602456596197,
+      "loss": 3.325,
+      "step": 11250
+    },
+    {
+      "epoch": 0.3222099305100583,
+      "eval_accuracy": 0.94,
+      "eval_loss": 0.2819044888019562,
+      "eval_macro_f1": 0.9395225640341313,
+      "eval_runtime": 173.501,
+      "eval_samples_per_second": 11.527,
+      "eval_steps_per_second": 11.527,
+      "step": 11250
+    },
+    {
+      "epoch": 0.3236419746456586,
+      "grad_norm": 1.078125,
+      "learning_rate": 0.0001394649816936341,
+      "loss": 3.0985,
+      "step": 11300
+    },
+    {
+      "epoch": 0.32507401878125886,
+      "grad_norm": 116.0,
+      "learning_rate": 0.00013916971772764852,
+      "loss": 2.5793,
+      "step": 11350
+    },
+    {
+      "epoch": 0.3265060629168591,
+      "grad_norm": 0.1865234375,
+      "learning_rate": 0.00013887445376166291,
+      "loss": 2.5646,
+      "step": 11400
+    },
+    {
+      "epoch": 0.32793810705245935,
+      "grad_norm": 306.0,
+      "learning_rate": 0.00013857918979567733,
+      "loss": 1.9864,
+      "step": 11450
+    },
+    {
+      "epoch": 0.3293701511880596,
+      "grad_norm": 274.0,
+      "learning_rate": 0.00013828392582969175,
+      "loss": 1.8868,
+      "step": 11500
+    },
+    {
+      "epoch": 0.3308021953236599,
+      "grad_norm": 0.08984375,
+      "learning_rate": 0.00013798866186370617,
+      "loss": 2.5106,
+      "step": 11550
+    },
+    {
+      "epoch": 0.3322342394592601,
+      "grad_norm": 270.0,
+      "learning_rate": 0.0001376933978977206,
+      "loss": 1.8537,
+      "step": 11600
+    },
+    {
+      "epoch": 0.3336662835948604,
+      "grad_norm": 270.0,
+      "learning_rate": 0.00013739813393173498,
+      "loss": 2.3735,
+      "step": 11650
+    },
+    {
+      "epoch": 0.33509832773046067,
+      "grad_norm": 0.416015625,
+      "learning_rate": 0.0001371028699657494,
+      "loss": 2.0794,
+      "step": 11700
+    },
+    {
+      "epoch": 0.3365303718660609,
+      "grad_norm": 0.32421875,
+      "learning_rate": 0.0001368076059997638,
+      "loss": 2.5114,
+      "step": 11750
+    },
+    {
+      "epoch": 0.33796241600166116,
+      "grad_norm": 0.0595703125,
+      "learning_rate": 0.0001365123420337782,
+      "loss": 2.199,
+      "step": 11800
+    },
+    {
+      "epoch": 0.33939446013726143,
+      "grad_norm": 61.75,
+      "learning_rate": 0.0001362170780677926,
+      "loss": 2.64,
+      "step": 11850
+    },
+    {
+      "epoch": 0.3408265042728617,
+      "grad_norm": 9.0,
+      "learning_rate": 0.00013592181410180702,
+      "loss": 1.8553,
+      "step": 11900
+    },
+    {
+      "epoch": 0.3422585484084619,
+      "grad_norm": 177.0,
+      "learning_rate": 0.00013562655013582144,
+      "loss": 1.6963,
+      "step": 11950
+    },
+    {
+      "epoch": 0.3436905925440622,
+      "grad_norm": 326.0,
+      "learning_rate": 0.00013533128616983583,
+      "loss": 3.007,
+      "step": 12000
+    },
+    {
+      "epoch": 0.3451226366796625,
+      "grad_norm": 0.06787109375,
+      "learning_rate": 0.00013503602220385025,
+      "loss": 1.6731,
+      "step": 12050
+    },
+    {
+      "epoch": 0.34655468081526275,
+      "grad_norm": 5.28125,
+      "learning_rate": 0.00013474075823786464,
+      "loss": 2.5167,
+      "step": 12100
+    },
+    {
+      "epoch": 0.34798672495086297,
+      "grad_norm": 0.11865234375,
+      "learning_rate": 0.00013444549427187906,
+      "loss": 3.4208,
+      "step": 12150
+    },
+    {
+      "epoch": 0.34941876908646324,
+      "grad_norm": 0.1337890625,
+      "learning_rate": 0.00013415023030589348,
+      "loss": 1.3073,
+      "step": 12200
+    },
+    {
+      "epoch": 0.3508508132220635,
+      "grad_norm": 82.5,
+      "learning_rate": 0.0001338549663399079,
+      "loss": 2.3618,
+      "step": 12250
+    },
+    {
+      "epoch": 0.35228285735766374,
+      "grad_norm": 1.546875,
+      "learning_rate": 0.00013355970237392232,
+      "loss": 2.0756,
+      "step": 12300
+    },
+    {
+      "epoch": 0.353714901493264,
+      "grad_norm": 120.0,
+      "learning_rate": 0.0001332644384079367,
+      "loss": 2.2016,
+      "step": 12350
+    },
+    {
+      "epoch": 0.3551469456288643,
+      "grad_norm": 5.34375,
+      "learning_rate": 0.00013296917444195113,
+      "loss": 2.7446,
+      "step": 12400
+    },
+    {
+      "epoch": 0.35657898976446456,
+      "grad_norm": 0.057861328125,
+      "learning_rate": 0.00013267391047596552,
+      "loss": 2.3892,
+      "step": 12450
+    },
+    {
+      "epoch": 0.3580110339000648,
+      "grad_norm": 1.6640625,
+      "learning_rate": 0.00013237864650997994,
+      "loss": 2.3038,
+      "step": 12500
+    },
+    {
+      "epoch": 0.3580110339000648,
+      "eval_accuracy": 0.943,
+      "eval_loss": 0.2820850610733032,
+      "eval_macro_f1": 0.9423954094372721,
+      "eval_runtime": 180.0229,
+      "eval_samples_per_second": 11.11,
+      "eval_steps_per_second": 11.11,
+      "step": 12500
+    },
+    {
+      "epoch": 0.35944307803566505,
+      "grad_norm": 0.15625,
+      "learning_rate": 0.00013208338254399433,
+      "loss": 1.6647,
+      "step": 12550
+    },
+    {
+      "epoch": 0.36087512217126533,
+      "grad_norm": 0.06640625,
+      "learning_rate": 0.00013178811857800875,
+      "loss": 2.2751,
+      "step": 12600
+    },
+    {
+      "epoch": 0.3623071663068656,
+      "grad_norm": 0.984375,
+      "learning_rate": 0.00013149285461202316,
+      "loss": 2.7458,
+      "step": 12650
+    },
+    {
+      "epoch": 0.3637392104424658,
+      "grad_norm": 420.0,
+      "learning_rate": 0.00013119759064603756,
+      "loss": 2.2766,
+      "step": 12700
+    },
+    {
+      "epoch": 0.3651712545780661,
+      "grad_norm": 0.1103515625,
+      "learning_rate": 0.00013090232668005198,
+      "loss": 1.4627,
+      "step": 12750
+    },
+    {
+      "epoch": 0.36660329871366637,
+      "grad_norm": 0.25390625,
+      "learning_rate": 0.00013060706271406637,
+      "loss": 1.9492,
+      "step": 12800
+    },
+    {
+      "epoch": 0.3680353428492666,
+      "grad_norm": 0.232421875,
+      "learning_rate": 0.00013031179874808079,
+      "loss": 1.8883,
+      "step": 12850
+    },
+    {
+      "epoch": 0.36946738698486686,
+      "grad_norm": 0.205078125,
+      "learning_rate": 0.00013001653478209518,
+      "loss": 2.058,
+      "step": 12900
+    },
+    {
+      "epoch": 0.37089943112046714,
+      "grad_norm": 0.53515625,
+      "learning_rate": 0.0001297212708161096,
+      "loss": 2.7257,
+      "step": 12950
+    },
+    {
+      "epoch": 0.3723314752560674,
+      "grad_norm": 764.0,
+      "learning_rate": 0.00012942600685012401,
+      "loss": 2.4923,
+      "step": 13000
+    },
+    {
+      "epoch": 0.37376351939166763,
+      "grad_norm": 298.0,
+      "learning_rate": 0.00012913074288413843,
+      "loss": 1.8667,
+      "step": 13050
+    },
+    {
+      "epoch": 0.3751955635272679,
+      "grad_norm": 282.0,
+      "learning_rate": 0.00012883547891815285,
+      "loss": 2.4477,
+      "step": 13100
+    },
+    {
+      "epoch": 0.3766276076628682,
+      "grad_norm": 0.02587890625,
+      "learning_rate": 0.00012854021495216724,
+      "loss": 1.0699,
+      "step": 13150
+    },
+    {
+      "epoch": 0.37805965179846845,
+      "grad_norm": 160.0,
+      "learning_rate": 0.00012824495098618166,
+      "loss": 2.8487,
+      "step": 13200
+    },
+    {
+      "epoch": 0.3794916959340687,
+      "grad_norm": 216.0,
+      "learning_rate": 0.00012794968702019605,
+      "loss": 1.8886,
+      "step": 13250
+    },
+    {
+      "epoch": 0.38092374006966895,
+      "grad_norm": 302.0,
+      "learning_rate": 0.00012765442305421047,
+      "loss": 2.4619,
+      "step": 13300
+    },
+    {
+      "epoch": 0.3823557842052692,
+      "grad_norm": 172.0,
+      "learning_rate": 0.0001273591590882249,
+      "loss": 2.8237,
+      "step": 13350
+    },
+    {
+      "epoch": 0.38378782834086944,
+      "grad_norm": 91.5,
+      "learning_rate": 0.00012706389512223928,
+      "loss": 2.7431,
+      "step": 13400
+    },
+    {
+      "epoch": 0.3852198724764697,
+      "grad_norm": 0.275390625,
+      "learning_rate": 0.0001267686311562537,
+      "loss": 1.9888,
+      "step": 13450
+    },
+    {
+      "epoch": 0.38665191661207,
+      "grad_norm": 194.0,
+      "learning_rate": 0.0001264733671902681,
+      "loss": 2.5123,
+      "step": 13500
+    },
+    {
+      "epoch": 0.38808396074767026,
+      "grad_norm": 396.0,
+      "learning_rate": 0.0001261781032242825,
+      "loss": 1.9384,
+      "step": 13550
+    },
+    {
+      "epoch": 0.3895160048832705,
+      "grad_norm": 54.75,
+      "learning_rate": 0.0001258828392582969,
+      "loss": 1.9031,
+      "step": 13600
+    },
+    {
+      "epoch": 0.39094804901887076,
+      "grad_norm": 0.828125,
+      "learning_rate": 0.00012558757529231132,
+      "loss": 2.65,
+      "step": 13650
+    },
+    {
+      "epoch": 0.39238009315447103,
+      "grad_norm": 36.25,
+      "learning_rate": 0.00012529231132632574,
+      "loss": 2.2794,
+      "step": 13700
+    },
+    {
+      "epoch": 0.3938121372900713,
+      "grad_norm": 0.11962890625,
+      "learning_rate": 0.00012499704736034016,
+      "loss": 2.1189,
+      "step": 13750
+    },
+    {
+      "epoch": 0.3938121372900713,
+      "eval_accuracy": 0.9465,
+      "eval_loss": 0.31273505091667175,
+      "eval_macro_f1": 0.9457595736365828,
+      "eval_runtime": 172.8312,
+      "eval_samples_per_second": 11.572,
+      "eval_steps_per_second": 11.572,
+      "step": 13750
+    },
+    {
+      "epoch": 0.3952441814256715,
+      "grad_norm": 398.0,
+      "learning_rate": 0.00012470178339435458,
+      "loss": 2.5979,
+      "step": 13800
+    },
+    {
+      "epoch": 0.3966762255612718,
+      "grad_norm": 266.0,
+      "learning_rate": 0.00012440651942836897,
+      "loss": 2.8401,
+      "step": 13850
+    },
+    {
+      "epoch": 0.3981082696968721,
+      "grad_norm": 0.1884765625,
+      "learning_rate": 0.0001241112554623834,
+      "loss": 1.9365,
+      "step": 13900
+    },
+    {
+      "epoch": 0.3995403138324723,
+      "grad_norm": 0.055908203125,
+      "learning_rate": 0.00012381599149639778,
+      "loss": 0.9845,
+      "step": 13950
+    },
+    {
+      "epoch": 0.40097235796807257,
+      "grad_norm": 130.0,
+      "learning_rate": 0.0001235207275304122,
+      "loss": 2.9002,
+      "step": 14000
+    },
+    {
+      "epoch": 0.40240440210367284,
+      "grad_norm": 0.0045166015625,
+      "learning_rate": 0.00012322546356442662,
+      "loss": 2.1528,
+      "step": 14050
+    },
+    {
+      "epoch": 0.4038364462392731,
+      "grad_norm": 0.39453125,
+      "learning_rate": 0.000122930199598441,
+      "loss": 2.4575,
+      "step": 14100
+    },
+    {
+      "epoch": 0.40526849037487334,
+      "grad_norm": 241.0,
+      "learning_rate": 0.00012263493563245543,
+      "loss": 2.475,
+      "step": 14150
+    },
+    {
+      "epoch": 0.4067005345104736,
+      "grad_norm": 37.5,
+      "learning_rate": 0.00012233967166646982,
+      "loss": 1.6529,
+      "step": 14200
+    },
+    {
+      "epoch": 0.4081325786460739,
+      "grad_norm": 272.0,
+      "learning_rate": 0.00012204440770048424,
+      "loss": 2.922,
+      "step": 14250
+    },
+    {
+      "epoch": 0.4095646227816741,
+      "grad_norm": 3.734375,
+      "learning_rate": 0.00012174914373449864,
+      "loss": 2.2361,
+      "step": 14300
+    },
+    {
+      "epoch": 0.4109966669172744,
+      "grad_norm": 338.0,
+      "learning_rate": 0.00012145387976851306,
+      "loss": 1.5299,
+      "step": 14350
+    },
+    {
+      "epoch": 0.41242871105287465,
+      "grad_norm": 0.043212890625,
+      "learning_rate": 0.00012115861580252748,
+      "loss": 2.5728,
+      "step": 14400
+    },
+    {
+      "epoch": 0.4138607551884749,
+      "grad_norm": 0.15625,
+      "learning_rate": 0.00012086335183654187,
+      "loss": 2.122,
+      "step": 14450
+    },
+    {
+      "epoch": 0.41529279932407515,
+      "grad_norm": 0.008056640625,
+      "learning_rate": 0.00012056808787055629,
+      "loss": 2.4863,
+      "step": 14500
+    },
+    {
+      "epoch": 0.4167248434596754,
+      "grad_norm": 32.5,
+      "learning_rate": 0.00012027282390457068,
+      "loss": 3.3401,
+      "step": 14550
+    },
+    {
+      "epoch": 0.4181568875952757,
+      "grad_norm": 0.2490234375,
+      "learning_rate": 0.0001199775599385851,
+      "loss": 2.0398,
+      "step": 14600
+    },
+    {
+      "epoch": 0.41958893173087597,
+      "grad_norm": 0.326171875,
+      "learning_rate": 0.00011968229597259951,
+      "loss": 2.8364,
+      "step": 14650
+    },
+    {
+      "epoch": 0.4210209758664762,
+      "grad_norm": 0.15625,
+      "learning_rate": 0.00011938703200661391,
+      "loss": 1.4864,
+      "step": 14700
+    },
+    {
+      "epoch": 0.42245302000207646,
+      "grad_norm": 80.0,
+      "learning_rate": 0.00011909176804062833,
+      "loss": 2.3397,
+      "step": 14750
+    },
+    {
+      "epoch": 0.42388506413767674,
+      "grad_norm": 101.5,
+      "learning_rate": 0.00011879650407464274,
+      "loss": 3.3186,
+      "step": 14800
+    },
+    {
+      "epoch": 0.42531710827327696,
+      "grad_norm": 2.6875,
+      "learning_rate": 0.00011850124010865716,
+      "loss": 2.7605,
+      "step": 14850
+    },
+    {
+      "epoch": 0.42674915240887723,
+      "grad_norm": 95.0,
+      "learning_rate": 0.00011820597614267155,
+      "loss": 1.9215,
+      "step": 14900
+    },
+    {
+      "epoch": 0.4281811965444775,
+      "grad_norm": 0.65625,
+      "learning_rate": 0.00011791071217668597,
+      "loss": 2.0689,
+      "step": 14950
+    },
+    {
+      "epoch": 0.4296132406800778,
+      "grad_norm": 0.66015625,
+      "learning_rate": 0.00011761544821070036,
+      "loss": 3.5633,
+      "step": 15000
+    },
+    {
+      "epoch": 0.4296132406800778,
+      "eval_accuracy": 0.942,
+      "eval_loss": 0.27190178632736206,
+      "eval_macro_f1": 0.9413541591870516,
+      "eval_runtime": 181.3524,
+      "eval_samples_per_second": 11.028,
+      "eval_steps_per_second": 11.028,
+      "step": 15000
+    },
+    {
+      "epoch": 0.431045284815678,
+      "grad_norm": 42.25,
+      "learning_rate": 0.00011732018424471478,
+      "loss": 1.8073,
+      "step": 15050
+    },
+    {
+      "epoch": 0.4324773289512783,
+      "grad_norm": 0.22265625,
+      "learning_rate": 0.0001170249202787292,
+      "loss": 1.777,
+      "step": 15100
+    },
+    {
+      "epoch": 0.43390937308687855,
+      "grad_norm": 0.11328125,
+      "learning_rate": 0.0001167296563127436,
+      "loss": 1.9598,
+      "step": 15150
+    },
+    {
+      "epoch": 0.4353414172224788,
+      "grad_norm": 0.095703125,
+      "learning_rate": 0.00011643439234675802,
+      "loss": 2.7789,
+      "step": 15200
+    },
+    {
+      "epoch": 0.43677346135807904,
+      "grad_norm": 0.8125,
+      "learning_rate": 0.00011613912838077241,
+      "loss": 2.3985,
+      "step": 15250
+    },
+    {
+      "epoch": 0.4382055054936793,
+      "grad_norm": 0.22265625,
+      "learning_rate": 0.00011584386441478683,
+      "loss": 1.6076,
+      "step": 15300
+    },
+    {
+      "epoch": 0.4396375496292796,
+      "grad_norm": 0.578125,
+      "learning_rate": 0.00011554860044880122,
+      "loss": 2.9266,
+      "step": 15350
+    },
+    {
+      "epoch": 0.4410695937648798,
+      "grad_norm": 0.0130615234375,
+      "learning_rate": 0.00011525333648281564,
+      "loss": 1.388,
+      "step": 15400
+    },
+    {
+      "epoch": 0.4425016379004801,
+      "grad_norm": 88.5,
+      "learning_rate": 0.00011495807251683006,
+      "loss": 2.6264,
+      "step": 15450
+    },
+    {
+      "epoch": 0.44393368203608036,
+      "grad_norm": 0.123046875,
+      "learning_rate": 0.00011466280855084446,
+      "loss": 1.9447,
+      "step": 15500
+    },
+    {
+      "epoch": 0.44536572617168063,
+      "grad_norm": 318.0,
+      "learning_rate": 0.00011436754458485888,
+      "loss": 1.4494,
+      "step": 15550
+    },
+    {
+      "epoch": 0.44679777030728085,
+      "grad_norm": 0.123046875,
+      "learning_rate": 0.00011407228061887327,
+      "loss": 2.4483,
+      "step": 15600
+    },
+    {
+      "epoch": 0.4482298144428811,
+      "grad_norm": 260.0,
+      "learning_rate": 0.00011377701665288769,
+      "loss": 2.8899,
+      "step": 15650
+    },
+    {
+      "epoch": 0.4496618585784814,
+      "grad_norm": 0.07666015625,
+      "learning_rate": 0.00011348175268690208,
+      "loss": 2.0935,
+      "step": 15700
+    },
+    {
+      "epoch": 0.4510939027140817,
+      "grad_norm": 0.047119140625,
+      "learning_rate": 0.0001131864887209165,
+      "loss": 3.0298,
+      "step": 15750
+    },
+    {
+      "epoch": 0.4525259468496819,
+      "grad_norm": 18.25,
+      "learning_rate": 0.00011289122475493092,
+      "loss": 1.8288,
+      "step": 15800
+    },
+    {
+      "epoch": 0.45395799098528217,
+      "grad_norm": 11.75,
+      "learning_rate": 0.00011259596078894533,
+      "loss": 2.9182,
+      "step": 15850
+    },
+    {
+      "epoch": 0.45539003512088244,
+      "grad_norm": 0.1416015625,
+      "learning_rate": 0.00011230069682295975,
+      "loss": 1.9045,
+      "step": 15900
+    },
+    {
+      "epoch": 0.45682207925648266,
+      "grad_norm": 0.1982421875,
+      "learning_rate": 0.00011200543285697414,
+      "loss": 1.7933,
+      "step": 15950
+    },
+    {
+      "epoch": 0.45825412339208293,
+      "grad_norm": 201.0,
+      "learning_rate": 0.00011171016889098856,
+      "loss": 2.4752,
+      "step": 16000
+    },
+    {
+      "epoch": 0.4596861675276832,
+      "grad_norm": 284.0,
+      "learning_rate": 0.00011141490492500295,
+      "loss": 1.7394,
+      "step": 16050
+    },
+    {
+      "epoch": 0.4611182116632835,
+      "grad_norm": 0.1162109375,
+      "learning_rate": 0.00011111964095901737,
+      "loss": 1.3612,
+      "step": 16100
+    },
+    {
+      "epoch": 0.4625502557988837,
+      "grad_norm": 0.30859375,
+      "learning_rate": 0.00011082437699303178,
+      "loss": 2.6066,
+      "step": 16150
+    },
+    {
+      "epoch": 0.463982299934484,
+      "grad_norm": 23.75,
+      "learning_rate": 0.00011052911302704618,
+      "loss": 2.6868,
+      "step": 16200
+    },
+    {
+      "epoch": 0.46541434407008425,
+      "grad_norm": 1.5546875,
+      "learning_rate": 0.0001102338490610606,
+      "loss": 2.6526,
+      "step": 16250
+    },
+    {
+      "epoch": 0.46541434407008425,
+      "eval_accuracy": 0.9445,
+      "eval_loss": 0.2859993577003479,
+      "eval_macro_f1": 0.9439942443541156,
+      "eval_runtime": 175.6054,
+      "eval_samples_per_second": 11.389,
+      "eval_steps_per_second": 11.389,
+      "step": 16250
+    },
+    {
+      "epoch": 0.4668463882056845,
+      "grad_norm": 0.11181640625,
+      "learning_rate": 0.000109938585095075,
+      "loss": 2.041,
+      "step": 16300
+    },
+    {
+      "epoch": 0.46827843234128474,
+      "grad_norm": 8.3125,
+      "learning_rate": 0.00010964332112908942,
+      "loss": 2.3374,
+      "step": 16350
+    },
+    {
+      "epoch": 0.469710476476885,
+      "grad_norm": 268.0,
+      "learning_rate": 0.00010934805716310381,
+      "loss": 3.461,
+      "step": 16400
+    },
+    {
+      "epoch": 0.4711425206124853,
+      "grad_norm": 270.0,
+      "learning_rate": 0.00010905279319711823,
+      "loss": 1.5168,
+      "step": 16450
+    },
+    {
+      "epoch": 0.4725745647480855,
+      "grad_norm": 1.6171875,
+      "learning_rate": 0.00010875752923113265,
+      "loss": 1.9156,
+      "step": 16500
+    },
+    {
+      "epoch": 0.4740066088836858,
+      "grad_norm": 3.25,
+      "learning_rate": 0.00010846226526514704,
+      "loss": 1.9988,
+      "step": 16550
+    },
+    {
+      "epoch": 0.47543865301928606,
+      "grad_norm": 266.0,
+      "learning_rate": 0.00010816700129916146,
+      "loss": 1.9911,
+      "step": 16600
+    },
+    {
+      "epoch": 0.47687069715488634,
+      "grad_norm": 0.45703125,
+      "learning_rate": 0.00010787173733317586,
+      "loss": 2.2805,
+      "step": 16650
+    },
+    {
+      "epoch": 0.47830274129048655,
+      "grad_norm": 0.08740234375,
+      "learning_rate": 0.00010757647336719028,
+      "loss": 2.3786,
+      "step": 16700
+    },
+    {
+      "epoch": 0.47973478542608683,
+      "grad_norm": 0.0263671875,
+      "learning_rate": 0.00010728120940120467,
+      "loss": 2.0964,
+      "step": 16750
+    },
+    {
+      "epoch": 0.4811668295616871,
+      "grad_norm": 288.0,
+      "learning_rate": 0.00010698594543521909,
+      "loss": 2.5816,
+      "step": 16800
+    },
+    {
+      "epoch": 0.4825988736972873,
+      "grad_norm": 0.2138671875,
+      "learning_rate": 0.00010669068146923348,
+      "loss": 1.2136,
+      "step": 16850
+    },
+    {
+      "epoch": 0.4840309178328876,
+      "grad_norm": 1.1796875,
+      "learning_rate": 0.0001063954175032479,
+      "loss": 2.2321,
+      "step": 16900
+    },
+    {
+      "epoch": 0.48546296196848787,
+      "grad_norm": 0.259765625,
+      "learning_rate": 0.00010610015353726232,
+      "loss": 2.7485,
+      "step": 16950
+    },
+    {
+      "epoch": 0.48689500610408815,
+      "grad_norm": 75.5,
+      "learning_rate": 0.00010580488957127673,
+      "loss": 3.0909,
+      "step": 17000
+    },
+    {
+      "epoch": 0.48832705023968836,
+      "grad_norm": 86.0,
+      "learning_rate": 0.00010550962560529115,
+      "loss": 2.0178,
+      "step": 17050
+    },
+    {
+      "epoch": 0.48975909437528864,
+      "grad_norm": 0.98046875,
+      "learning_rate": 0.00010521436163930554,
+      "loss": 2.0193,
+      "step": 17100
+    },
+    {
+      "epoch": 0.4911911385108889,
+      "grad_norm": 68.0,
+      "learning_rate": 0.00010491909767331996,
+      "loss": 2.178,
+      "step": 17150
+    },
+    {
+      "epoch": 0.4926231826464892,
+      "grad_norm": 0.375,
+      "learning_rate": 0.00010462383370733435,
+      "loss": 2.4366,
+      "step": 17200
+    },
+    {
+      "epoch": 0.4940552267820894,
+      "grad_norm": 0.322265625,
+      "learning_rate": 0.00010432856974134877,
+      "loss": 2.7989,
+      "step": 17250
+    },
+    {
+      "epoch": 0.4954872709176897,
+      "grad_norm": 0.1513671875,
+      "learning_rate": 0.00010403330577536318,
+      "loss": 1.8681,
+      "step": 17300
+    },
+    {
+      "epoch": 0.49691931505328996,
+      "grad_norm": 0.08935546875,
+      "learning_rate": 0.00010373804180937759,
+      "loss": 1.9697,
+      "step": 17350
+    },
+    {
+      "epoch": 0.4983513591888902,
+      "grad_norm": 0.1328125,
+      "learning_rate": 0.00010344277784339201,
+      "loss": 1.7415,
+      "step": 17400
+    },
+    {
+      "epoch": 0.49978340332449045,
+      "grad_norm": 0.65234375,
+      "learning_rate": 0.0001031475138774064,
+      "loss": 1.6849,
+      "step": 17450
+    },
+    {
+      "epoch": 0.5012154474600907,
+      "grad_norm": 86.5,
+      "learning_rate": 0.00010285224991142082,
+      "loss": 2.1474,
+      "step": 17500
+    },
+    {
+      "epoch": 0.5012154474600907,
+      "eval_accuracy": 0.947,
+      "eval_loss": 0.2907390892505646,
+      "eval_macro_f1": 0.9463529866080697,
+      "eval_runtime": 172.6878,
+      "eval_samples_per_second": 11.582,
+      "eval_steps_per_second": 11.582,
+      "step": 17500
+    },
+    {
+      "epoch": 0.502647491595691,
+      "grad_norm": 472.0,
+      "learning_rate": 0.00010255698594543521,
+      "loss": 1.9173,
+      "step": 17550
+    },
+    {
+      "epoch": 0.5040795357312913,
+      "grad_norm": 1.3828125,
+      "learning_rate": 0.00010226172197944963,
+      "loss": 3.1869,
+      "step": 17600
+    },
+    {
+      "epoch": 0.5055115798668915,
+      "grad_norm": 145.0,
+      "learning_rate": 0.00010196645801346405,
+      "loss": 2.5482,
+      "step": 17650
+    },
+    {
+      "epoch": 0.5069436240024917,
+      "grad_norm": 0.58203125,
+      "learning_rate": 0.00010167119404747844,
+      "loss": 2.8567,
+      "step": 17700
+    },
+    {
+      "epoch": 0.508375668138092,
+      "grad_norm": 0.0888671875,
+      "learning_rate": 0.00010137593008149286,
+      "loss": 1.7268,
+      "step": 17750
+    },
+    {
+      "epoch": 0.5098077122736923,
+      "grad_norm": 286.0,
+      "learning_rate": 0.00010108066611550726,
+      "loss": 2.2268,
+      "step": 17800
+    },
+    {
+      "epoch": 0.5112397564092925,
+      "grad_norm": 169.0,
+      "learning_rate": 0.00010078540214952168,
+      "loss": 1.8245,
+      "step": 17850
+    },
+    {
+      "epoch": 0.5126718005448928,
+      "grad_norm": 0.1923828125,
+      "learning_rate": 0.00010049013818353607,
+      "loss": 2.3801,
+      "step": 17900
+    },
+    {
+      "epoch": 0.5141038446804931,
+      "grad_norm": 77.0,
+      "learning_rate": 0.00010019487421755049,
+      "loss": 2.3412,
+      "step": 17950
+    },
+    {
+      "epoch": 0.5155358888160934,
+      "grad_norm": 328.0,
+      "learning_rate": 9.98996102515649e-05,
+      "loss": 3.1564,
+      "step": 18000
+    },
+    {
+      "epoch": 0.5169679329516935,
+      "grad_norm": 0.57421875,
+      "learning_rate": 9.96043462855793e-05,
+      "loss": 2.0409,
+      "step": 18050
+    },
+    {
+      "epoch": 0.5183999770872938,
+      "grad_norm": 0.8828125,
+      "learning_rate": 9.930908231959372e-05,
+      "loss": 1.8093,
+      "step": 18100
+    },
+    {
+      "epoch": 0.5198320212228941,
+      "grad_norm": 0.263671875,
+      "learning_rate": 9.901381835360814e-05,
+      "loss": 2.1228,
+      "step": 18150
+    },
+    {
+      "epoch": 0.5212640653584943,
+      "grad_norm": 13.625,
+      "learning_rate": 9.871855438762255e-05,
+      "loss": 1.6072,
+      "step": 18200
+    },
+    {
+      "epoch": 0.5226961094940946,
+      "grad_norm": 176.0,
+      "learning_rate": 9.842329042163695e-05,
+      "loss": 2.1088,
+      "step": 18250
+    },
+    {
+      "epoch": 0.5241281536296949,
+      "grad_norm": 0.09228515625,
+      "learning_rate": 9.812802645565136e-05,
+      "loss": 2.2985,
+      "step": 18300
+    },
+    {
+      "epoch": 0.5255601977652952,
+      "grad_norm": 5.9375,
+      "learning_rate": 9.783276248966576e-05,
+      "loss": 2.8687,
+      "step": 18350
+    },
+    {
+      "epoch": 0.5269922419008953,
+      "grad_norm": 0.1142578125,
+      "learning_rate": 9.753749852368017e-05,
+      "loss": 1.9855,
+      "step": 18400
+    },
+    {
+      "epoch": 0.5284242860364956,
+      "grad_norm": 1.03125,
+      "learning_rate": 9.724223455769457e-05,
+      "loss": 2.5827,
+      "step": 18450
+    },
+    {
+      "epoch": 0.5298563301720959,
+      "grad_norm": 270.0,
+      "learning_rate": 9.694697059170899e-05,
+      "loss": 1.9905,
+      "step": 18500
+    },
+    {
+      "epoch": 0.5312883743076962,
+      "grad_norm": 0.2392578125,
+      "learning_rate": 9.665170662572341e-05,
+      "loss": 1.9516,
+      "step": 18550
+    },
+    {
+      "epoch": 0.5327204184432964,
+      "grad_norm": 116.5,
+      "learning_rate": 9.635644265973781e-05,
+      "loss": 1.7887,
+      "step": 18600
+    },
+    {
+      "epoch": 0.5341524625788967,
+      "grad_norm": 0.0306396484375,
+      "learning_rate": 9.606117869375222e-05,
+      "loss": 1.8686,
+      "step": 18650
+    },
+    {
+      "epoch": 0.535584506714497,
+      "grad_norm": 0.7109375,
+      "learning_rate": 9.576591472776662e-05,
+      "loss": 1.7828,
+      "step": 18700
+    },
+    {
+      "epoch": 0.5370165508500973,
+      "grad_norm": 0.92578125,
+      "learning_rate": 9.547065076178103e-05,
+      "loss": 1.8761,
+      "step": 18750
+    },
+    {
+      "epoch": 0.5370165508500973,
+      "eval_accuracy": 0.9495,
+      "eval_loss": 0.26102420687675476,
+      "eval_macro_f1": 0.9488861373782008,
+      "eval_runtime": 172.7705,
+      "eval_samples_per_second": 11.576,
+      "eval_steps_per_second": 11.576,
+      "step": 18750
+    },
+    {
+      "epoch": 0.5384485949856974,
+      "grad_norm": 0.1416015625,
+      "learning_rate": 9.517538679579543e-05,
+      "loss": 2.3765,
+      "step": 18800
+    },
+    {
+      "epoch": 0.5398806391212977,
+      "grad_norm": 258.0,
+      "learning_rate": 9.488012282980985e-05,
+      "loss": 1.5944,
+      "step": 18850
+    },
+    {
+      "epoch": 0.541312683256898,
+      "grad_norm": 106.0,
+      "learning_rate": 9.458485886382427e-05,
+      "loss": 2.4606,
+      "step": 18900
+    },
+    {
+      "epoch": 0.5427447273924982,
+      "grad_norm": 92.5,
+      "learning_rate": 9.428959489783868e-05,
+      "loss": 2.6863,
+      "step": 18950
+    },
+    {
+      "epoch": 0.5441767715280985,
+      "grad_norm": 207.0,
+      "learning_rate": 9.399433093185308e-05,
+      "loss": 1.9462,
+      "step": 19000
+    },
+    {
+      "epoch": 0.5456088156636988,
+      "grad_norm": 82.5,
+      "learning_rate": 9.369906696586749e-05,
+      "loss": 2.8243,
+      "step": 19050
+    },
+    {
+      "epoch": 0.5470408597992991,
+      "grad_norm": 0.1572265625,
+      "learning_rate": 9.340380299988189e-05,
+      "loss": 1.8158,
+      "step": 19100
+    },
+    {
+      "epoch": 0.5484729039348992,
+      "grad_norm": 0.4609375,
+      "learning_rate": 9.31085390338963e-05,
+      "loss": 2.7108,
+      "step": 19150
+    },
+    {
+      "epoch": 0.5499049480704995,
+      "grad_norm": 0.62109375,
+      "learning_rate": 9.281327506791072e-05,
+      "loss": 1.9835,
+      "step": 19200
+    },
+    {
+      "epoch": 0.5513369922060998,
+      "grad_norm": 0.039794921875,
+      "learning_rate": 9.251801110192512e-05,
+      "loss": 2.0608,
+      "step": 19250
+    },
+    {
+      "epoch": 0.5527690363417,
+      "grad_norm": 1.265625,
+      "learning_rate": 9.222274713593954e-05,
+      "loss": 1.9135,
+      "step": 19300
+    },
+    {
+      "epoch": 0.5542010804773003,
+      "grad_norm": 0.030517578125,
+      "learning_rate": 9.192748316995395e-05,
+      "loss": 2.5224,
+      "step": 19350
+    },
+    {
+      "epoch": 0.5556331246129006,
+      "grad_norm": 332.0,
+      "learning_rate": 9.163221920396835e-05,
+      "loss": 3.2123,
+      "step": 19400
+    },
+    {
+      "epoch": 0.5570651687485009,
+      "grad_norm": 0.046142578125,
+      "learning_rate": 9.133695523798276e-05,
+      "loss": 1.9807,
+      "step": 19450
+    },
+    {
+      "epoch": 0.558497212884101,
+      "grad_norm": 0.036376953125,
+      "learning_rate": 9.104169127199716e-05,
+      "loss": 2.8211,
+      "step": 19500
+    },
+    {
+      "epoch": 0.5599292570197013,
+      "grad_norm": 0.2412109375,
+      "learning_rate": 9.074642730601158e-05,
+      "loss": 2.7913,
+      "step": 19550
+    },
+    {
+      "epoch": 0.5613613011553016,
+      "grad_norm": 0.087890625,
+      "learning_rate": 9.045116334002599e-05,
+      "loss": 1.5528,
+      "step": 19600
+    },
+    {
+      "epoch": 0.5627933452909019,
+      "grad_norm": 0.21484375,
+      "learning_rate": 9.01558993740404e-05,
+      "loss": 2.3024,
+      "step": 19650
+    },
+    {
+      "epoch": 0.5642253894265021,
+      "grad_norm": 0.271484375,
+      "learning_rate": 8.986063540805481e-05,
+      "loss": 1.305,
+      "step": 19700
+    },
+    {
+      "epoch": 0.5656574335621024,
+      "grad_norm": 0.373046875,
+      "learning_rate": 8.956537144206921e-05,
+      "loss": 2.1656,
+      "step": 19750
+    },
+    {
+      "epoch": 0.5670894776977027,
+      "grad_norm": 84.5,
+      "learning_rate": 8.927010747608362e-05,
+      "loss": 1.6671,
+      "step": 19800
+    },
+    {
+      "epoch": 0.568521521833303,
+      "grad_norm": 112.0,
+      "learning_rate": 8.897484351009802e-05,
+      "loss": 2.4715,
+      "step": 19850
+    },
+    {
+      "epoch": 0.5699535659689031,
+      "grad_norm": 0.130859375,
+      "learning_rate": 8.867957954411244e-05,
+      "loss": 1.7577,
+      "step": 19900
+    },
+    {
+      "epoch": 0.5713856101045034,
+      "grad_norm": 0.1357421875,
+      "learning_rate": 8.838431557812685e-05,
+      "loss": 1.6778,
+      "step": 19950
+    },
+    {
+      "epoch": 0.5728176542401037,
+      "grad_norm": 0.07080078125,
+      "learning_rate": 8.808905161214125e-05,
+      "loss": 1.4789,
+      "step": 20000
+    },
+    {
+      "epoch": 0.5728176542401037,
+      "eval_accuracy": 0.949,
+      "eval_loss": 0.2858292758464813,
+      "eval_macro_f1": 0.9484543460104051,
+      "eval_runtime": 172.7421,
+      "eval_samples_per_second": 11.578,
+      "eval_steps_per_second": 11.578,
+      "step": 20000
+    },
+    {
+      "epoch": 0.5742496983757039,
+      "grad_norm": 0.140625,
+      "learning_rate": 8.779378764615567e-05,
+      "loss": 1.6797,
+      "step": 20050
+    },
+    {
+      "epoch": 0.5756817425113042,
+      "grad_norm": 0.115234375,
+      "learning_rate": 8.749852368017008e-05,
+      "loss": 1.5026,
+      "step": 20100
+    },
+    {
+      "epoch": 0.5771137866469045,
+      "grad_norm": 0.02880859375,
+      "learning_rate": 8.720325971418448e-05,
+      "loss": 2.1316,
+      "step": 20150
+    },
+    {
+      "epoch": 0.5785458307825048,
+      "grad_norm": 0.08544921875,
+      "learning_rate": 8.690799574819889e-05,
+      "loss": 1.7517,
+      "step": 20200
+    },
+    {
+      "epoch": 0.5799778749181049,
+      "grad_norm": 82.0,
+      "learning_rate": 8.661273178221331e-05,
+      "loss": 2.4167,
+      "step": 20250
+    },
+    {
+      "epoch": 0.5814099190537052,
+      "grad_norm": 0.138671875,
+      "learning_rate": 8.631746781622771e-05,
+      "loss": 1.8565,
+      "step": 20300
+    },
+    {
+      "epoch": 0.5828419631893055,
+      "grad_norm": 34.25,
+      "learning_rate": 8.602220385024212e-05,
+      "loss": 1.9747,
+      "step": 20350
+    },
+    {
+      "epoch": 0.5842740073249058,
+      "grad_norm": 91.0,
+      "learning_rate": 8.572693988425654e-05,
+      "loss": 2.3284,
+      "step": 20400
+    },
+    {
+      "epoch": 0.585706051460506,
+      "grad_norm": 0.80859375,
+      "learning_rate": 8.543167591827094e-05,
+      "loss": 2.1788,
+      "step": 20450
+    },
+    {
+      "epoch": 0.5871380955961063,
+      "grad_norm": 446.0,
+      "learning_rate": 8.513641195228535e-05,
+      "loss": 1.6187,
+      "step": 20500
+    },
+    {
+      "epoch": 0.5885701397317066,
+      "grad_norm": 43.75,
+      "learning_rate": 8.484114798629975e-05,
+      "loss": 1.6338,
+      "step": 20550
+    },
+    {
+      "epoch": 0.5900021838673067,
+      "grad_norm": 0.91015625,
+      "learning_rate": 8.454588402031417e-05,
+      "loss": 2.424,
+      "step": 20600
+    },
+    {
+      "epoch": 0.591434228002907,
+      "grad_norm": 0.44921875,
+      "learning_rate": 8.425062005432858e-05,
+      "loss": 2.3043,
+      "step": 20650
+    },
+    {
+      "epoch": 0.5928662721385073,
+      "grad_norm": 268.0,
+      "learning_rate": 8.395535608834298e-05,
+      "loss": 2.5707,
+      "step": 20700
+    },
+    {
+      "epoch": 0.5942983162741076,
+      "grad_norm": 4.125,
+      "learning_rate": 8.366009212235739e-05,
+      "loss": 1.9577,
+      "step": 20750
+    },
+    {
+      "epoch": 0.5957303604097078,
+      "grad_norm": 5.0,
+      "learning_rate": 8.33648281563718e-05,
+      "loss": 0.7482,
+      "step": 20800
+    },
+    {
+      "epoch": 0.5971624045453081,
+      "grad_norm": 0.04052734375,
+      "learning_rate": 8.306956419038621e-05,
+      "loss": 1.5055,
+      "step": 20850
+    },
+    {
+      "epoch": 0.5985944486809084,
+      "grad_norm": 1.109375,
+      "learning_rate": 8.277430022440061e-05,
+      "loss": 3.3671,
+      "step": 20900
+    },
+    {
+      "epoch": 0.6000264928165085,
+      "grad_norm": 0.162109375,
+      "learning_rate": 8.247903625841503e-05,
+      "loss": 2.0574,
+      "step": 20950
+    },
+    {
+      "epoch": 0.6014585369521088,
+      "grad_norm": 0.03173828125,
+      "learning_rate": 8.218377229242944e-05,
+      "loss": 2.1942,
+      "step": 21000
+    },
+    {
+      "epoch": 0.6028905810877091,
+      "grad_norm": 0.251953125,
+      "learning_rate": 8.188850832644384e-05,
+      "loss": 1.6319,
+      "step": 21050
+    },
+    {
+      "epoch": 0.6043226252233094,
+      "grad_norm": 86.5,
+      "learning_rate": 8.159324436045825e-05,
+      "loss": 2.1558,
+      "step": 21100
+    },
+    {
+      "epoch": 0.6057546693589096,
+      "grad_norm": 2.984375,
+      "learning_rate": 8.129798039447267e-05,
+      "loss": 2.2353,
+      "step": 21150
+    },
+    {
+      "epoch": 0.6071867134945099,
+      "grad_norm": 0.0908203125,
+      "learning_rate": 8.100271642848707e-05,
+      "loss": 1.4975,
+      "step": 21200
+    },
+    {
+      "epoch": 0.6086187576301102,
+      "grad_norm": 98.0,
+      "learning_rate": 8.070745246250148e-05,
+      "loss": 2.5975,
+      "step": 21250
+    },
+    {
+      "epoch": 0.6086187576301102,
+      "eval_accuracy": 0.948,
+      "eval_loss": 0.2741381525993347,
+      "eval_macro_f1": 0.9473973559594594,
+      "eval_runtime": 172.6111,
+      "eval_samples_per_second": 11.587,
+      "eval_steps_per_second": 11.587,
+      "step": 21250
+    },
+    {
+      "epoch": 0.6100508017657105,
+      "grad_norm": 328.0,
+      "learning_rate": 8.04121884965159e-05,
+      "loss": 2.4534,
+      "step": 21300
+    },
+    {
+      "epoch": 0.6114828459013106,
+      "grad_norm": 11.125,
+      "learning_rate": 8.01169245305303e-05,
+      "loss": 2.1319,
+      "step": 21350
+    },
+    {
+      "epoch": 0.6129148900369109,
+      "grad_norm": 0.08642578125,
+      "learning_rate": 7.982166056454471e-05,
+      "loss": 2.4199,
+      "step": 21400
+    },
+    {
+      "epoch": 0.6143469341725112,
+      "grad_norm": 0.18359375,
+      "learning_rate": 7.952639659855911e-05,
+      "loss": 1.7527,
+      "step": 21450
+    },
+    {
+      "epoch": 0.6157789783081115,
+      "grad_norm": 0.458984375,
+      "learning_rate": 7.923113263257352e-05,
+      "loss": 2.4992,
+      "step": 21500
+    },
+    {
+      "epoch": 0.6172110224437117,
+      "grad_norm": 0.671875,
+      "learning_rate": 7.893586866658794e-05,
+      "loss": 2.5082,
+      "step": 21550
+    },
+    {
+      "epoch": 0.618643066579312,
+      "grad_norm": 272.0,
+      "learning_rate": 7.864060470060234e-05,
+      "loss": 2.2187,
+      "step": 21600
+    },
+    {
+      "epoch": 0.6200751107149123,
+      "grad_norm": 310.0,
+      "learning_rate": 7.834534073461675e-05,
+      "loss": 3.006,
+      "step": 21650
+    },
+    {
+      "epoch": 0.6215071548505124,
+      "grad_norm": 536.0,
+      "learning_rate": 7.805007676863117e-05,
+      "loss": 2.4535,
+      "step": 21700
+    },
+    {
+      "epoch": 0.6229391989861127,
+      "grad_norm": 408.0,
+      "learning_rate": 7.775481280264557e-05,
+      "loss": 2.376,
+      "step": 21750
+    },
+    {
+      "epoch": 0.624371243121713,
+      "grad_norm": 0.123046875,
+      "learning_rate": 7.745954883665998e-05,
+      "loss": 1.3044,
+      "step": 21800
+    },
+    {
+      "epoch": 0.6258032872573133,
+      "grad_norm": 0.70703125,
+      "learning_rate": 7.716428487067438e-05,
+      "loss": 1.9046,
+      "step": 21850
+    },
+    {
+      "epoch": 0.6272353313929135,
+      "grad_norm": 0.189453125,
+      "learning_rate": 7.68690209046888e-05,
+      "loss": 1.8825,
+      "step": 21900
+    },
+    {
+      "epoch": 0.6286673755285138,
+      "grad_norm": 0.251953125,
+      "learning_rate": 7.65737569387032e-05,
+      "loss": 2.353,
+      "step": 21950
+    },
+    {
+      "epoch": 0.6300994196641141,
+      "grad_norm": 0.0301513671875,
+      "learning_rate": 7.627849297271761e-05,
+      "loss": 1.7222,
+      "step": 22000
+    },
+    {
+      "epoch": 0.6315314637997143,
+      "grad_norm": 0.126953125,
+      "learning_rate": 7.598322900673203e-05,
+      "loss": 2.4583,
+      "step": 22050
+    },
+    {
+      "epoch": 0.6329635079353145,
+      "grad_norm": 1.0234375,
+      "learning_rate": 7.568796504074643e-05,
+      "loss": 1.9643,
+      "step": 22100
+    },
+    {
+      "epoch": 0.6343955520709148,
+      "grad_norm": 70.5,
+      "learning_rate": 7.539270107476084e-05,
+      "loss": 1.6712,
+      "step": 22150
+    },
+    {
+      "epoch": 0.6358275962065151,
+      "grad_norm": 8.0,
+      "learning_rate": 7.509743710877524e-05,
+      "loss": 2.1964,
+      "step": 22200
+    },
+    {
+      "epoch": 0.6372596403421154,
+      "grad_norm": 11.625,
+      "learning_rate": 7.480217314278965e-05,
+      "loss": 2.0319,
+      "step": 22250
+    },
+    {
+      "epoch": 0.6386916844777156,
+      "grad_norm": 0.09033203125,
+      "learning_rate": 7.450690917680407e-05,
+      "loss": 3.1062,
+      "step": 22300
+    },
+    {
+      "epoch": 0.6401237286133159,
+      "grad_norm": 490.0,
+      "learning_rate": 7.421164521081847e-05,
+      "loss": 2.028,
+      "step": 22350
+    },
+    {
+      "epoch": 0.6415557727489162,
+      "grad_norm": 688.0,
+      "learning_rate": 7.391638124483289e-05,
+      "loss": 1.6743,
+      "step": 22400
+    },
+    {
+      "epoch": 0.6429878168845163,
+      "grad_norm": 0.2578125,
+      "learning_rate": 7.36211172788473e-05,
+      "loss": 1.3926,
+      "step": 22450
+    },
+    {
+      "epoch": 0.6444198610201166,
+      "grad_norm": 278.0,
+      "learning_rate": 7.33258533128617e-05,
+      "loss": 2.073,
+      "step": 22500
+    },
+    {
+      "epoch": 0.6444198610201166,
+      "eval_accuracy": 0.9495,
+      "eval_loss": 0.2617259919643402,
+      "eval_macro_f1": 0.9489699460568645,
+      "eval_runtime": 172.6662,
+      "eval_samples_per_second": 11.583,
+      "eval_steps_per_second": 11.583,
+      "step": 22500
+    },
+    {
+      "epoch": 0.6458519051557169,
+      "grad_norm": 0.07373046875,
+      "learning_rate": 7.303058934687611e-05,
+      "loss": 2.099,
+      "step": 22550
+    },
+    {
+      "epoch": 0.6472839492913172,
+      "grad_norm": 0.5859375,
+      "learning_rate": 7.273532538089051e-05,
+      "loss": 2.2826,
+      "step": 22600
+    },
+    {
+      "epoch": 0.6487159934269174,
+      "grad_norm": 79.5,
+      "learning_rate": 7.244006141490493e-05,
+      "loss": 1.377,
+      "step": 22650
+    },
+    {
+      "epoch": 0.6501480375625177,
+      "grad_norm": 2.265625,
+      "learning_rate": 7.214479744891934e-05,
+      "loss": 1.9826,
+      "step": 22700
+    },
+    {
+      "epoch": 0.651580081698118,
+      "grad_norm": 0.37109375,
+      "learning_rate": 7.184953348293376e-05,
+      "loss": 2.2446,
+      "step": 22750
+    },
+    {
+      "epoch": 0.6530121258337181,
+      "grad_norm": 4.25,
+      "learning_rate": 7.155426951694816e-05,
+      "loss": 2.0254,
+      "step": 22800
+    },
+    {
+      "epoch": 0.6544441699693184,
+      "grad_norm": 0.03564453125,
+      "learning_rate": 7.125900555096257e-05,
+      "loss": 2.0871,
+      "step": 22850
+    },
+    {
+      "epoch": 0.6558762141049187,
+      "grad_norm": 0.390625,
+      "learning_rate": 7.096374158497697e-05,
+      "loss": 2.9276,
+      "step": 22900
+    },
+    {
+      "epoch": 0.657308258240519,
+      "grad_norm": 0.384765625,
+      "learning_rate": 7.066847761899138e-05,
+      "loss": 1.0622,
+      "step": 22950
+    },
+    {
+      "epoch": 0.6587403023761192,
+      "grad_norm": 4576.0,
+      "learning_rate": 7.037321365300578e-05,
+      "loss": 3.0808,
+      "step": 23000
+    },
+    {
+      "epoch": 0.6601723465117195,
+      "grad_norm": 372.0,
+      "learning_rate": 7.00779496870202e-05,
+      "loss": 1.8306,
+      "step": 23050
+    },
+    {
+      "epoch": 0.6616043906473198,
+      "grad_norm": 0.208984375,
+      "learning_rate": 6.978268572103462e-05,
+      "loss": 2.1282,
+      "step": 23100
+    },
+    {
+      "epoch": 0.66303643478292,
+      "grad_norm": 5.65625,
+      "learning_rate": 6.948742175504902e-05,
+      "loss": 1.8392,
+      "step": 23150
+    },
+    {
+      "epoch": 0.6644684789185202,
+      "grad_norm": 0.24609375,
+      "learning_rate": 6.919215778906343e-05,
+      "loss": 2.594,
+      "step": 23200
+    },
+    {
+      "epoch": 0.6659005230541205,
+      "grad_norm": 0.123046875,
+      "learning_rate": 6.889689382307783e-05,
+      "loss": 2.4234,
+      "step": 23250
+    },
+    {
+      "epoch": 0.6673325671897208,
+      "grad_norm": 268.0,
+      "learning_rate": 6.860162985709224e-05,
+      "loss": 2.3424,
+      "step": 23300
+    },
+    {
+      "epoch": 0.6687646113253211,
+      "grad_norm": 0.427734375,
+      "learning_rate": 6.830636589110664e-05,
+      "loss": 2.3216,
+      "step": 23350
+    },
+    {
+      "epoch": 0.6701966554609213,
+      "grad_norm": 0.9296875,
+      "learning_rate": 6.801110192512106e-05,
+      "loss": 2.4566,
+      "step": 23400
+    },
+    {
+      "epoch": 0.6716286995965216,
+      "grad_norm": 88.5,
+      "learning_rate": 6.771583795913548e-05,
+      "loss": 1.3767,
+      "step": 23450
+    },
+    {
+      "epoch": 0.6730607437321218,
+      "grad_norm": 488.0,
+      "learning_rate": 6.742057399314989e-05,
+      "loss": 2.343,
+      "step": 23500
+    },
+    {
+      "epoch": 0.674492787867722,
+      "grad_norm": 296.0,
+      "learning_rate": 6.712531002716429e-05,
+      "loss": 1.4841,
+      "step": 23550
+    },
+    {
+      "epoch": 0.6759248320033223,
+      "grad_norm": 218.0,
+      "learning_rate": 6.68300460611787e-05,
+      "loss": 2.4037,
+      "step": 23600
+    },
+    {
+      "epoch": 0.6773568761389226,
+      "grad_norm": 0.466796875,
+      "learning_rate": 6.65347820951931e-05,
+      "loss": 1.4982,
+      "step": 23650
+    },
+    {
+      "epoch": 0.6787889202745229,
+      "grad_norm": 49.25,
+      "learning_rate": 6.623951812920751e-05,
+      "loss": 2.2085,
+      "step": 23700
+    },
+    {
+      "epoch": 0.6802209644101231,
+      "grad_norm": 0.30078125,
+      "learning_rate": 6.594425416322191e-05,
+      "loss": 1.7055,
+      "step": 23750
+    },
+    {
+      "epoch": 0.6802209644101231,
+      "eval_accuracy": 0.9505,
+      "eval_loss": 0.26270824670791626,
+      "eval_macro_f1": 0.9498080478089564,
+      "eval_runtime": 172.6664,
+      "eval_samples_per_second": 11.583,
+      "eval_steps_per_second": 11.583,
+      "step": 23750
+    },
+    {
+      "epoch": 0.6816530085457234,
+      "grad_norm": 95.5,
+      "learning_rate": 6.564899019723633e-05,
+      "loss": 2.5024,
+      "step": 23800
+    },
+    {
+      "epoch": 0.6830850526813237,
+      "grad_norm": 0.30078125,
+      "learning_rate": 6.535372623125075e-05,
+      "loss": 2.2518,
+      "step": 23850
+    },
+    {
+      "epoch": 0.6845170968169239,
+      "grad_norm": 116.5,
+      "learning_rate": 6.505846226526516e-05,
+      "loss": 2.0539,
+      "step": 23900
+    },
+    {
+      "epoch": 0.6859491409525241,
+      "grad_norm": 264.0,
+      "learning_rate": 6.476319829927956e-05,
+      "loss": 2.5857,
+      "step": 23950
+    },
+    {
+      "epoch": 0.6873811850881244,
+      "grad_norm": 302.0,
+      "learning_rate": 6.446793433329397e-05,
+      "loss": 2.1408,
+      "step": 24000
+    },
+    {
+      "epoch": 0.6888132292237247,
+      "grad_norm": 0.5390625,
+      "learning_rate": 6.417267036730837e-05,
+      "loss": 1.9618,
+      "step": 24050
+    },
+    {
+      "epoch": 0.690245273359325,
+      "grad_norm": 164.0,
+      "learning_rate": 6.387740640132278e-05,
+      "loss": 2.0112,
+      "step": 24100
+    },
+    {
+      "epoch": 0.6916773174949252,
+      "grad_norm": 14.0625,
+      "learning_rate": 6.35821424353372e-05,
+      "loss": 2.7256,
+      "step": 24150
+    },
+    {
+      "epoch": 0.6931093616305255,
+      "grad_norm": 164.0,
+      "learning_rate": 6.328687846935161e-05,
+      "loss": 0.8362,
+      "step": 24200
+    },
+    {
+      "epoch": 0.6945414057661257,
+      "grad_norm": 0.271484375,
+      "learning_rate": 6.299161450336602e-05,
+      "loss": 2.2874,
+      "step": 24250
+    },
+    {
+      "epoch": 0.6959734499017259,
+      "grad_norm": 110.5,
+      "learning_rate": 6.269635053738042e-05,
+      "loss": 1.5674,
+      "step": 24300
+    },
+    {
+      "epoch": 0.6974054940373262,
+      "grad_norm": 0.1630859375,
+      "learning_rate": 6.240108657139483e-05,
+      "loss": 2.5817,
+      "step": 24350
+    },
+    {
+      "epoch": 0.6988375381729265,
+      "grad_norm": 0.99609375,
+      "learning_rate": 6.210582260540923e-05,
+      "loss": 1.2537,
+      "step": 24400
+    },
+    {
+      "epoch": 0.7002695823085268,
+      "grad_norm": 266.0,
+      "learning_rate": 6.181055863942364e-05,
+      "loss": 2.4499,
+      "step": 24450
+    },
+    {
+      "epoch": 0.701701626444127,
+      "grad_norm": 1.59375,
+      "learning_rate": 6.151529467343806e-05,
+      "loss": 2.8047,
+      "step": 24500
+    },
+    {
+      "epoch": 0.7031336705797273,
+      "grad_norm": 0.08447265625,
+      "learning_rate": 6.122003070745246e-05,
+      "loss": 1.6917,
+      "step": 24550
+    },
+    {
+      "epoch": 0.7045657147153275,
+      "grad_norm": 296.0,
+      "learning_rate": 6.0924766741466875e-05,
+      "loss": 2.2486,
+      "step": 24600
+    },
+    {
+      "epoch": 0.7059977588509277,
+      "grad_norm": 183.0,
+      "learning_rate": 6.062950277548128e-05,
+      "loss": 2.5183,
+      "step": 24650
+    },
+    {
+      "epoch": 0.707429802986528,
+      "grad_norm": 0.123046875,
+      "learning_rate": 6.033423880949569e-05,
+      "loss": 2.2984,
+      "step": 24700
+    },
+    {
+      "epoch": 0.7088618471221283,
+      "grad_norm": 0.37890625,
+      "learning_rate": 6.00389748435101e-05,
+      "loss": 2.4598,
+      "step": 24750
+    },
+    {
+      "epoch": 0.7102938912577286,
+      "grad_norm": 6.25,
+      "learning_rate": 5.97437108775245e-05,
+      "loss": 2.0554,
+      "step": 24800
+    },
+    {
+      "epoch": 0.7117259353933288,
+      "grad_norm": 1.5234375,
+      "learning_rate": 5.9448446911538915e-05,
+      "loss": 1.3688,
+      "step": 24850
+    },
+    {
+      "epoch": 0.7131579795289291,
+      "grad_norm": 83.5,
+      "learning_rate": 5.9153182945553334e-05,
+      "loss": 2.6434,
+      "step": 24900
+    },
+    {
+      "epoch": 0.7145900236645294,
+      "grad_norm": 0.8046875,
+      "learning_rate": 5.885791897956774e-05,
+      "loss": 1.1703,
+      "step": 24950
+    },
+    {
+      "epoch": 0.7160220678001296,
+      "grad_norm": 0.76953125,
+      "learning_rate": 5.8562655013582144e-05,
+      "loss": 1.7433,
+      "step": 25000
+    },
+    {
+      "epoch": 0.7160220678001296,
+      "eval_accuracy": 0.9475,
+      "eval_loss": 0.2805185317993164,
+      "eval_macro_f1": 0.9469725724830536,
+      "eval_runtime": 172.6365,
+      "eval_samples_per_second": 11.585,
+      "eval_steps_per_second": 11.585,
+      "step": 25000
+    },
+    {
+      "epoch": 0.7174541119357298,
+      "grad_norm": 268.0,
+      "learning_rate": 5.8267391047596556e-05,
+      "loss": 2.7963,
+      "step": 25050
+    },
+    {
+      "epoch": 0.7188861560713301,
+      "grad_norm": 294.0,
+      "learning_rate": 5.797212708161096e-05,
+      "loss": 2.3253,
+      "step": 25100
+    },
+    {
+      "epoch": 0.7203182002069304,
+      "grad_norm": 0.11181640625,
+      "learning_rate": 5.7676863115625366e-05,
+      "loss": 1.0165,
+      "step": 25150
+    },
+    {
+      "epoch": 0.7217502443425307,
+      "grad_norm": 756.0,
+      "learning_rate": 5.738159914963978e-05,
+      "loss": 1.4844,
+      "step": 25200
+    },
+    {
+      "epoch": 0.7231822884781309,
+      "grad_norm": 0.10107421875,
+      "learning_rate": 5.70863351836542e-05,
+      "loss": 2.7171,
+      "step": 25250
+    },
+    {
+      "epoch": 0.7246143326137312,
+      "grad_norm": 336.0,
+      "learning_rate": 5.67910712176686e-05,
+      "loss": 3.1605,
+      "step": 25300
+    },
+    {
+      "epoch": 0.7260463767493314,
+      "grad_norm": 177.0,
+      "learning_rate": 5.649580725168301e-05,
+      "loss": 1.9816,
+      "step": 25350
+    },
+    {
+      "epoch": 0.7274784208849316,
+      "grad_norm": 1.78125,
+      "learning_rate": 5.620054328569741e-05,
+      "loss": 1.8129,
+      "step": 25400
+    },
+    {
+      "epoch": 0.7289104650205319,
+      "grad_norm": 0.040771484375,
+      "learning_rate": 5.5905279319711824e-05,
+      "loss": 1.3484,
+      "step": 25450
+    },
+    {
+      "epoch": 0.7303425091561322,
+      "grad_norm": 8.375,
+      "learning_rate": 5.561001535372623e-05,
+      "loss": 2.1354,
+      "step": 25500
+    },
+    {
+      "epoch": 0.7317745532917325,
+      "grad_norm": 0.0308837890625,
+      "learning_rate": 5.5314751387740635e-05,
+      "loss": 1.747,
+      "step": 25550
+    },
+    {
+      "epoch": 0.7332065974273327,
+      "grad_norm": 808.0,
+      "learning_rate": 5.5019487421755053e-05,
+      "loss": 2.6803,
+      "step": 25600
+    },
+    {
+      "epoch": 0.734638641562933,
+      "grad_norm": 0.96484375,
+      "learning_rate": 5.4724223455769465e-05,
+      "loss": 2.2422,
+      "step": 25650
+    },
+    {
+      "epoch": 0.7360706856985332,
+      "grad_norm": 10.0,
+      "learning_rate": 5.442895948978387e-05,
+      "loss": 2.0731,
+      "step": 25700
+    },
+    {
+      "epoch": 0.7375027298341335,
+      "grad_norm": 0.27734375,
+      "learning_rate": 5.4133695523798276e-05,
+      "loss": 2.9622,
+      "step": 25750
+    },
+    {
+      "epoch": 0.7389347739697337,
+      "grad_norm": 200.0,
+      "learning_rate": 5.383843155781269e-05,
+      "loss": 2.179,
+      "step": 25800
+    },
+    {
+      "epoch": 0.740366818105334,
+      "grad_norm": 118.5,
+      "learning_rate": 5.354316759182709e-05,
+      "loss": 2.4152,
+      "step": 25850
+    },
+    {
+      "epoch": 0.7417988622409343,
+      "grad_norm": 0.2470703125,
+      "learning_rate": 5.32479036258415e-05,
+      "loss": 1.4274,
+      "step": 25900
+    },
+    {
+      "epoch": 0.7432309063765346,
+      "grad_norm": 6.71875,
+      "learning_rate": 5.295263965985592e-05,
+      "loss": 2.0263,
+      "step": 25950
+    },
+    {
+      "epoch": 0.7446629505121348,
+      "grad_norm": 0.28125,
+      "learning_rate": 5.265737569387033e-05,
+      "loss": 1.8231,
+      "step": 26000
+    },
+    {
+      "epoch": 0.746094994647735,
+      "grad_norm": 0.8046875,
+      "learning_rate": 5.2362111727884734e-05,
+      "loss": 1.7974,
+      "step": 26050
+    },
+    {
+      "epoch": 0.7475270387833353,
+      "grad_norm": 102.5,
+      "learning_rate": 5.206684776189914e-05,
+      "loss": 2.5667,
+      "step": 26100
+    },
+    {
+      "epoch": 0.7489590829189355,
+      "grad_norm": 380.0,
+      "learning_rate": 5.1771583795913544e-05,
+      "loss": 1.8334,
+      "step": 26150
+    },
+    {
+      "epoch": 0.7503911270545358,
+      "grad_norm": 0.74609375,
+      "learning_rate": 5.1476319829927956e-05,
+      "loss": 1.2929,
+      "step": 26200
+    },
+    {
+      "epoch": 0.7518231711901361,
+      "grad_norm": 7.3125,
+      "learning_rate": 5.118105586394236e-05,
+      "loss": 2.2943,
+      "step": 26250
+    },
+    {
+      "epoch": 0.7518231711901361,
+      "eval_accuracy": 0.951,
+      "eval_loss": 0.2633407413959503,
+      "eval_macro_f1": 0.9502699810655684,
+      "eval_runtime": 172.7789,
+      "eval_samples_per_second": 11.575,
+      "eval_steps_per_second": 11.575,
+      "step": 26250
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 34916,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 1250,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.1179332952064e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

Topic/Strategy/checkpoint-26250/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c312fa9ef2c2fd587c80c33d22bca8b57769510fc82b4a9d1b7772152d2b2bb4
+size 5496

Topic/Willingness/checkpoint-2500/README.md

@@ -0,0 +1,206 @@
+---
+base_model: D:\LLM\Qwen2.5-7B-Instruct
+library_name: peft
+tags:
+- base_model:adapter:D:\LLM\Qwen2.5-7B-Instruct
+- lora
+- transformers
+---
+
+# 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 0.17.1

Topic/Willingness/checkpoint-2500/adapter_config.json

@@ -0,0 +1,45 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "D:\\LLM\\Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "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": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": [
+    "classifier",
+    "score"
+  ],
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "up_proj",
+    "gate_proj",
+    "down_proj",
+    "o_proj",
+    "k_proj",
+    "v_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "SEQ_CLS",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

Topic/Willingness/checkpoint-2500/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43890d6d5f36ad8aca965ad57a653b2ad3794ab1fac0fd3f5665a6ab580ce6e8
+size 80800144

Topic/Willingness/checkpoint-2500/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
+}

Topic/Willingness/checkpoint-2500/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 %}