Upload folder using huggingface_hub

.gitattributes

@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 checkpoint-246/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoint-384/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-369/tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -20,8 +20,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [Qwen/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B) on the train dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2565
-- Accuracy: 0.9462
+- Loss: 0.2496
+- Accuracy: 0.9448
 
 ## Model description
 
@@ -45,37 +45,34 @@ The following hyperparameters were used during training:
 - eval_batch_size: 2
 - seed: 42
 - distributed_type: multi-GPU
-- num_devices: 4
+- num_devices: 3
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 32
-- total_eval_batch_size: 8
-- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- total_train_batch_size: 24
+- total_eval_batch_size: 6
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.03
-- num_epochs: 12.0
+- num_epochs: 9.0
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.4881        | 1.0   | 32   | 0.4397          | 0.8944   |
-| 0.338         | 2.0   | 64   | 0.3506          | 0.9138   |
-| 0.3097        | 3.0   | 96   | 0.3098          | 0.9230   |
-| 0.293         | 4.0   | 128  | 0.2800          | 0.9310   |
-| 0.202         | 5.0   | 160  | 0.2604          | 0.9357   |
-| 0.1846        | 6.0   | 192  | 0.2504          | 0.9395   |
-| 0.1536        | 7.0   | 224  | 0.2461          | 0.9422   |
-| 0.1484        | 8.0   | 256  | 0.2469          | 0.9449   |
-| 0.1132        | 9.0   | 288  | 0.2507          | 0.9459   |
-| 0.0924        | 10.0  | 320  | 0.2528          | 0.9459   |
-| 0.097         | 11.0  | 352  | 0.2561          | 0.9462   |
-| 0.1146        | 12.0  | 384  | 0.2565          | 0.9462   |
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy |
+|:-------------:|:------:|:----:|:---------------:|:--------:|
+| 0.3591        | 1.0    | 42   | 0.3988          | 0.9035   |
+| 0.2673        | 2.0    | 84   | 0.3260          | 0.9197   |
+| 0.2425        | 3.0    | 126  | 0.2898          | 0.9289   |
+| 0.2069        | 4.0    | 168  | 0.2659          | 0.9346   |
+| 0.138         | 5.0    | 210  | 0.2525          | 0.9391   |
+| 0.1251        | 6.0    | 252  | 0.2497          | 0.9431   |
+| 0.1072        | 7.0    | 294  | 0.2475          | 0.9439   |
+| 0.1059        | 8.0    | 336  | 0.2483          | 0.9446   |
+| 0.1073        | 8.7904 | 369  | 0.2496          | 0.9448   |
 
 
 ### Framework versions
 
 - PEFT 0.15.2
-- Transformers 4.52.4
-- Pytorch 2.7.0
+- Transformers 4.51.3
+- Pytorch 2.7.1+cu126
 - Datasets 3.6.0
 - Tokenizers 0.21.1

adapter_config.json

@@ -25,11 +25,11 @@
   "revision": null,
   "target_modules": [
     "v_proj",
-    "q_proj",
-    "o_proj",
-    "k_proj",
     "up_proj",
     "gate_proj",
+    "k_proj",
+    "o_proj",
+    "q_proj",
     "down_proj"
   ],
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd57f85cfd0f5c6cd87305892b38484c1b8778a2f03c65c648efe0f5fa9b2568
+oid sha256:188aede640d1000b3120e751ff1ca978c6b622e66ea86d1a47341bdf3dabf524
 size 349243752

all_results.json

@@ -1,13 +1,15 @@
 {
-    "epoch": 12.0,
-    "eval_accuracy": 0.9461611343879004,
-    "eval_loss": 0.2565154731273651,
-    "eval_runtime": 23.1024,
-    "eval_samples_per_second": 16.665,
-    "eval_steps_per_second": 2.121,
-    "total_flos": 4.38485641689301e+17,
-    "train_loss": 0.24724087212234735,
-    "train_runtime": 2839.5296,
-    "train_samples_per_second": 4.226,
-    "train_steps_per_second": 0.135
+    "effective_tokens_per_sec": 779.031292389977,
+    "epoch": 8.790419161676647,
+    "eval_accuracy": 0.9448452683547727,
+    "eval_loss": 1.6609941720962524,
+    "eval_model_preparation_time": 0.0029,
+    "eval_runtime": 22.9431,
+    "eval_samples_per_second": 16.781,
+    "eval_steps_per_second": 2.833,
+    "total_flos": 3.210182982232965e+17,
+    "train_loss": 0.2637010880602085,
+    "train_runtime": 2751.4148,
+    "train_samples_per_second": 3.271,
+    "train_steps_per_second": 0.134
 }

checkpoint-369/README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen3-8B
+library_name: peft
+---
+
+# 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 0.15.2

checkpoint-369/adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-8B",
+  "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": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "up_proj",
+    "gate_proj",
+    "k_proj",
+    "o_proj",
+    "q_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-369/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:188aede640d1000b3120e751ff1ca978c6b622e66ea86d1a47341bdf3dabf524
+size 349243752

checkpoint-369/added_tokens.json

@@ -0,0 +1,28 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|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
+}

checkpoint-369/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28efc55cde21533ac4d9b7974416c143c12d018580324c2e5554ca5a917c9f1b
+size 698777675

checkpoint-369/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:119587a9391c07122bab6571398727891031ac123fa70f7a283cc445a71ccd8a
+size 15173

checkpoint-369/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e29a63d7063d8ce3318df300d42735b7249acd7ed86a0c0f340c13062d2b2c53
+size 15173

checkpoint-369/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1290d34cf20cd55913b31bde9b5125f0a96e3e92745e92ea275bc3ca6d4594a0
+size 15173

checkpoint-369/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:424755b5494adc5c2f757e98ba4ae1e8a1e2f36783c0fc67336778e50397db71
+size 1465

checkpoint-369/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
+  }
+}

checkpoint-369/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

checkpoint-369/tokenizer_config.json

@@ -0,0 +1,241 @@
+{
+  "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
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "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,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\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>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\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\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if message.content is string %}\n        {%- set content = message.content %}\n    {%- else %}\n        {%- set content = '' %}\n    {%- endif %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is string %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in content %}\n                {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n                {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-369/trainer_state.json

@@ -0,0 +1,358 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 8.790419161676647,
+  "eval_steps": 500,
+  "global_step": 369,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.23952095808383234,
+      "grad_norm": 0.7001179456710815,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 1.4338,
+      "step": 10
+    },
+    {
+      "epoch": 0.47904191616766467,
+      "grad_norm": 0.27172571420669556,
+      "learning_rate": 9.990516643685222e-05,
+      "loss": 0.8267,
+      "step": 20
+    },
+    {
+      "epoch": 0.718562874251497,
+      "grad_norm": 0.19638711214065552,
+      "learning_rate": 9.944154131125642e-05,
+      "loss": 0.4881,
+      "step": 30
+    },
+    {
+      "epoch": 0.9580838323353293,
+      "grad_norm": 0.20754235982894897,
+      "learning_rate": 9.859528969650738e-05,
+      "loss": 0.3591,
+      "step": 40
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9034878200745683,
+      "eval_loss": 0.39884674549102783,
+      "eval_runtime": 30.4717,
+      "eval_samples_per_second": 12.635,
+      "eval_steps_per_second": 2.133,
+      "step": 42
+    },
+    {
+      "epoch": 1.1916167664670658,
+      "grad_norm": 0.16468322277069092,
+      "learning_rate": 9.737296070648186e-05,
+      "loss": 0.4172,
+      "step": 50
+    },
+    {
+      "epoch": 1.4311377245508983,
+      "grad_norm": 0.1612997055053711,
+      "learning_rate": 9.57840139057007e-05,
+      "loss": 0.3893,
+      "step": 60
+    },
+    {
+      "epoch": 1.6706586826347305,
+      "grad_norm": 0.16499397158622742,
+      "learning_rate": 9.384074610206495e-05,
+      "loss": 0.3658,
+      "step": 70
+    },
+    {
+      "epoch": 1.910179640718563,
+      "grad_norm": 0.14762192964553833,
+      "learning_rate": 9.155819618225708e-05,
+      "loss": 0.2673,
+      "step": 80
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9196909824354998,
+      "eval_loss": 0.325968861579895,
+      "eval_runtime": 30.3583,
+      "eval_samples_per_second": 12.682,
+      "eval_steps_per_second": 2.141,
+      "step": 84
+    },
+    {
+      "epoch": 2.143712574850299,
+      "grad_norm": 0.1777992993593216,
+      "learning_rate": 8.895402872628352e-05,
+      "loss": 0.3165,
+      "step": 90
+    },
+    {
+      "epoch": 2.3832335329341316,
+      "grad_norm": 0.2070295810699463,
+      "learning_rate": 8.604839730186125e-05,
+      "loss": 0.336,
+      "step": 100
+    },
+    {
+      "epoch": 2.622754491017964,
+      "grad_norm": 0.1750418245792389,
+      "learning_rate": 8.286378849660896e-05,
+      "loss": 0.288,
+      "step": 110
+    },
+    {
+      "epoch": 2.8622754491017965,
+      "grad_norm": 0.20214907824993134,
+      "learning_rate": 7.942484789507283e-05,
+      "loss": 0.2425,
+      "step": 120
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9289047847079928,
+      "eval_loss": 0.289765864610672,
+      "eval_runtime": 30.3541,
+      "eval_samples_per_second": 12.684,
+      "eval_steps_per_second": 2.141,
+      "step": 126
+    },
+    {
+      "epoch": 3.095808383233533,
+      "grad_norm": 0.2640642523765564,
+      "learning_rate": 7.57581893473448e-05,
+      "loss": 0.2304,
+      "step": 130
+    },
+    {
+      "epoch": 3.3353293413173652,
+      "grad_norm": 0.26811081171035767,
+      "learning_rate": 7.18921890053375e-05,
+      "loss": 0.2905,
+      "step": 140
+    },
+    {
+      "epoch": 3.5748502994011977,
+      "grad_norm": 0.27033841609954834,
+      "learning_rate": 6.785676572066225e-05,
+      "loss": 0.2336,
+      "step": 150
+    },
+    {
+      "epoch": 3.81437125748503,
+      "grad_norm": 0.22121788561344147,
+      "learning_rate": 6.368314950360415e-05,
+      "loss": 0.2069,
+      "step": 160
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9345709163732674,
+      "eval_loss": 0.2659221291542053,
+      "eval_runtime": 30.3929,
+      "eval_samples_per_second": 12.667,
+      "eval_steps_per_second": 2.139,
+      "step": 168
+    },
+    {
+      "epoch": 4.047904191616767,
+      "grad_norm": 0.2626728117465973,
+      "learning_rate": 5.940363983508257e-05,
+      "loss": 0.1815,
+      "step": 170
+    },
+    {
+      "epoch": 4.287425149700598,
+      "grad_norm": 0.2664365768432617,
+      "learning_rate": 5.5051355702012893e-05,
+      "loss": 0.2426,
+      "step": 180
+    },
+    {
+      "epoch": 4.526946107784431,
+      "grad_norm": 0.2793976962566376,
+      "learning_rate": 5.0659979290537954e-05,
+      "loss": 0.1998,
+      "step": 190
+    },
+    {
+      "epoch": 4.766467065868263,
+      "grad_norm": 0.2911984622478485,
+      "learning_rate": 4.626349532067879e-05,
+      "loss": 0.1713,
+      "step": 200
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.2870716452598572,
+      "learning_rate": 4.189592803968563e-05,
+      "loss": 0.138,
+      "step": 210
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9391251537470097,
+      "eval_loss": 0.2524815797805786,
+      "eval_runtime": 30.4065,
+      "eval_samples_per_second": 12.662,
+      "eval_steps_per_second": 2.138,
+      "step": 210
+    },
+    {
+      "epoch": 5.2395209580838324,
+      "grad_norm": 0.3472362756729126,
+      "learning_rate": 3.759107790948882e-05,
+      "loss": 0.199,
+      "step": 220
+    },
+    {
+      "epoch": 5.479041916167665,
+      "grad_norm": 0.33083346486091614,
+      "learning_rate": 3.338226002601703e-05,
+      "loss": 0.1654,
+      "step": 230
+    },
+    {
+      "epoch": 5.718562874251497,
+      "grad_norm": 0.3220432996749878,
+      "learning_rate": 2.9302046294747497e-05,
+      "loss": 0.1459,
+      "step": 240
+    },
+    {
+      "epoch": 5.95808383233533,
+      "grad_norm": 0.35546302795410156,
+      "learning_rate": 2.5382013357782893e-05,
+      "loss": 0.1251,
+      "step": 250
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9430640763868036,
+      "eval_loss": 0.24969537556171417,
+      "eval_runtime": 30.3846,
+      "eval_samples_per_second": 12.671,
+      "eval_steps_per_second": 2.139,
+      "step": 252
+    },
+    {
+      "epoch": 6.191616766467066,
+      "grad_norm": 0.35284000635147095,
+      "learning_rate": 2.1652498223239427e-05,
+      "loss": 0.1486,
+      "step": 260
+    },
+    {
+      "epoch": 6.431137724550898,
+      "grad_norm": 0.31098905205726624,
+      "learning_rate": 1.814236348812211e-05,
+      "loss": 0.1527,
+      "step": 270
+    },
+    {
+      "epoch": 6.6706586826347305,
+      "grad_norm": 0.34438222646713257,
+      "learning_rate": 1.4878773971620074e-05,
+      "loss": 0.1344,
+      "step": 280
+    },
+    {
+      "epoch": 6.910179640718563,
+      "grad_norm": 0.2907264232635498,
+      "learning_rate": 1.1886986487449475e-05,
+      "loss": 0.1072,
+      "step": 290
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9438642485491895,
+      "eval_loss": 0.24749523401260376,
+      "eval_runtime": 30.43,
+      "eval_samples_per_second": 12.652,
+      "eval_steps_per_second": 2.136,
+      "step": 294
+    },
+    {
+      "epoch": 7.1437125748503,
+      "grad_norm": 0.3019481301307678,
+      "learning_rate": 9.190154382188921e-06,
+      "loss": 0.1224,
+      "step": 300
+    },
+    {
+      "epoch": 7.383233532934132,
+      "grad_norm": 0.29105937480926514,
+      "learning_rate": 6.809148352279182e-06,
+      "loss": 0.1493,
+      "step": 310
+    },
+    {
+      "epoch": 7.6227544910179645,
+      "grad_norm": 0.2779920697212219,
+      "learning_rate": 4.762394926378477e-06,
+      "loss": 0.1126,
+      "step": 320
+    },
+    {
+      "epoch": 7.862275449101796,
+      "grad_norm": 0.3359215557575226,
+      "learning_rate": 3.065733863053072e-06,
+      "loss": 0.1059,
+      "step": 330
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.944612993482476,
+      "eval_loss": 0.2483372986316681,
+      "eval_runtime": 30.4006,
+      "eval_samples_per_second": 12.664,
+      "eval_steps_per_second": 2.138,
+      "step": 336
+    },
+    {
+      "epoch": 8.095808383233534,
+      "grad_norm": 0.26059141755104065,
+      "learning_rate": 1.7322955673980678e-06,
+      "loss": 0.0986,
+      "step": 340
+    },
+    {
+      "epoch": 8.335329341317365,
+      "grad_norm": 0.26576822996139526,
+      "learning_rate": 7.723994752570462e-07,
+      "loss": 0.1465,
+      "step": 350
+    },
+    {
+      "epoch": 8.574850299401197,
+      "grad_norm": 0.2771126329898834,
+      "learning_rate": 1.9347419144180035e-07,
+      "loss": 0.1073,
+      "step": 360
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 369,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 9,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.210182982232965e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-369/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89da7b86c6ffda4db368588ecb97943bf3d8149fc8db3215a63b20e819145412
+size 6097

eval_results.json

@@ -1,8 +1,7 @@
 {
-    "epoch": 12.0,
-    "eval_accuracy": 0.9461611343879004,
-    "eval_loss": 0.2565154731273651,
-    "eval_runtime": 23.1024,
-    "eval_samples_per_second": 16.665,
-    "eval_steps_per_second": 2.121
+    "eval_loss": 1.6609941720962524,
+    "eval_model_preparation_time": 0.0029,
+    "eval_runtime": 22.9431,
+    "eval_samples_per_second": 16.781,
+    "eval_steps_per_second": 2.833
 }

runs/Aug01_15-38-05_pan/events.out.tfevents.1754077441.pan

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f428bbda10c1144d40198bdde25cba3344e22531c5b8ce278cae85d34661fb4c
+size 16197

runs/Aug01_15-38-05_pan/events.out.tfevents.1754080221.pan

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b30689e272da6f69c5c9a915ee5fa1ee5069a9ad78220e2bec8bf722cef3ac51
+size 363

runs/Aug01_16-31-52_pan/events.out.tfevents.1754080339.pan

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27bf6c38e9293c80ca0420576838462e7fc0a4c88296b8173b7fe15fbf3857a6
+size 312

tokenizer_config.json

@@ -227,6 +227,7 @@
     "<|video_pad|>"
   ],
   "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\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>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\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\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if message.content is string %}\n        {%- set content = message.content %}\n    {%- else %}\n        {%- set content = '' %}\n    {%- endif %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is string %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in content %}\n                {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n                {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|im_end|>",
   "errors": "replace",

train_results.json

@@ -1,8 +1,9 @@
 {
-    "epoch": 12.0,
-    "total_flos": 4.38485641689301e+17,
-    "train_loss": 0.24724087212234735,
-    "train_runtime": 2839.5296,
-    "train_samples_per_second": 4.226,
-    "train_steps_per_second": 0.135
+    "effective_tokens_per_sec": 779.031292389977,
+    "epoch": 8.790419161676647,
+    "total_flos": 3.210182982232965e+17,
+    "train_loss": 0.2637010880602085,
+    "train_runtime": 2751.4148,
+    "train_samples_per_second": 3.271,
+    "train_steps_per_second": 0.134
 }

trainer_log.jsonl

@@ -1,51 +1,60 @@
-{"current_steps": 10, "total_steps": 384, "loss": 1.4075, "lr": 7.500000000000001e-05, "epoch": 0.32, "percentage": 2.6, "elapsed_time": "0:01:09", "remaining_time": "0:43:07"}
-{"current_steps": 20, "total_steps": 384, "loss": 0.8351, "lr": 9.991265793145479e-05, "epoch": 0.64, "percentage": 5.21, "elapsed_time": "0:02:15", "remaining_time": "0:41:05"}
-{"current_steps": 30, "total_steps": 384, "loss": 0.4881, "lr": 9.948559446825412e-05, "epoch": 0.96, "percentage": 7.81, "elapsed_time": "0:03:23", "remaining_time": "0:40:00"}
-{"current_steps": 32, "total_steps": 384, "eval_loss": 0.439687579870224, "epoch": 1.0, "percentage": 8.33, "elapsed_time": "0:03:55", "remaining_time": "0:43:09"}
-{"current_steps": 40, "total_steps": 384, "loss": 0.3822, "lr": 9.870580739976935e-05, "epoch": 1.256, "percentage": 10.42, "elapsed_time": "0:04:49", "remaining_time": "0:41:26"}
-{"current_steps": 50, "total_steps": 384, "loss": 0.4258, "lr": 9.757885489518297e-05, "epoch": 1.576, "percentage": 13.02, "elapsed_time": "0:05:57", "remaining_time": "0:39:50"}
-{"current_steps": 60, "total_steps": 384, "loss": 0.338, "lr": 9.611276965077099e-05, "epoch": 1.896, "percentage": 15.62, "elapsed_time": "0:07:07", "remaining_time": "0:38:30"}
-{"current_steps": 64, "total_steps": 384, "eval_loss": 0.3505571186542511, "epoch": 2.0, "percentage": 16.67, "elapsed_time": "0:07:53", "remaining_time": "0:39:28"}
-{"current_steps": 70, "total_steps": 384, "loss": 0.3284, "lr": 9.431800163442041e-05, "epoch": 2.192, "percentage": 18.23, "elapsed_time": "0:08:35", "remaining_time": "0:38:33"}
-{"current_steps": 80, "total_steps": 384, "loss": 0.3177, "lr": 9.220734360030907e-05, "epoch": 2.512, "percentage": 20.83, "elapsed_time": "0:09:42", "remaining_time": "0:36:55"}
-{"current_steps": 90, "total_steps": 384, "loss": 0.3097, "lr": 8.979583990466454e-05, "epoch": 2.832, "percentage": 23.44, "elapsed_time": "0:10:49", "remaining_time": "0:35:22"}
-{"current_steps": 96, "total_steps": 384, "eval_loss": 0.3097846210002899, "epoch": 3.0, "percentage": 25.0, "elapsed_time": "0:11:47", "remaining_time": "0:35:23"}
-{"current_steps": 100, "total_steps": 384, "loss": 0.2869, "lr": 8.710067927254555e-05, "epoch": 3.128, "percentage": 26.04, "elapsed_time": "0:12:15", "remaining_time": "0:34:47"}
-{"current_steps": 110, "total_steps": 384, "loss": 0.264, "lr": 8.414107227998329e-05, "epoch": 3.448, "percentage": 28.65, "elapsed_time": "0:13:23", "remaining_time": "0:33:20"}
-{"current_steps": 120, "total_steps": 384, "loss": 0.293, "lr": 8.093811442476573e-05, "epoch": 3.768, "percentage": 31.25, "elapsed_time": "0:14:30", "remaining_time": "0:31:54"}
-{"current_steps": 128, "total_steps": 384, "eval_loss": 0.2800418436527252, "epoch": 4.0, "percentage": 33.33, "elapsed_time": "0:15:42", "remaining_time": "0:31:25"}
-{"current_steps": 130, "total_steps": 384, "loss": 0.2517, "lr": 7.751463576186957e-05, "epoch": 4.064, "percentage": 33.85, "elapsed_time": "0:15:57", "remaining_time": "0:31:10"}
-{"current_steps": 140, "total_steps": 384, "loss": 0.2501, "lr": 7.389503817530905e-05, "epoch": 4.384, "percentage": 36.46, "elapsed_time": "0:17:05", "remaining_time": "0:29:47"}
-{"current_steps": 150, "total_steps": 384, "loss": 0.2115, "lr": 7.010512144629579e-05, "epoch": 4.704, "percentage": 39.06, "elapsed_time": "0:18:12", "remaining_time": "0:28:24"}
-{"current_steps": 160, "total_steps": 384, "loss": 0.202, "lr": 6.61718993574619e-05, "epoch": 5.0, "percentage": 41.67, "elapsed_time": "0:19:15", "remaining_time": "0:26:57"}
-{"current_steps": 160, "total_steps": 384, "eval_loss": 0.26041463017463684, "epoch": 5.0, "percentage": 41.67, "elapsed_time": "0:19:38", "remaining_time": "0:27:30"}
-{"current_steps": 170, "total_steps": 384, "loss": 0.195, "lr": 6.2123407143919e-05, "epoch": 5.32, "percentage": 44.27, "elapsed_time": "0:20:47", "remaining_time": "0:26:10"}
-{"current_steps": 180, "total_steps": 384, "loss": 0.2085, "lr": 5.798850166360461e-05, "epoch": 5.64, "percentage": 46.88, "elapsed_time": "0:21:54", "remaining_time": "0:24:49"}
-{"current_steps": 190, "total_steps": 384, "loss": 0.1846, "lr": 5.379665571126231e-05, "epoch": 5.96, "percentage": 49.48, "elapsed_time": "0:23:02", "remaining_time": "0:23:31"}
-{"current_steps": 192, "total_steps": 384, "eval_loss": 0.250407874584198, "epoch": 6.0, "percentage": 50.0, "elapsed_time": "0:23:34", "remaining_time": "0:23:34"}
-{"current_steps": 200, "total_steps": 384, "loss": 0.1577, "lr": 4.957774794214508e-05, "epoch": 6.256, "percentage": 52.08, "elapsed_time": "0:24:30", "remaining_time": "0:22:32"}
-{"current_steps": 210, "total_steps": 384, "loss": 0.1718, "lr": 4.5361849902824674e-05, "epoch": 6.576, "percentage": 54.69, "elapsed_time": "0:25:38", "remaining_time": "0:21:14"}
-{"current_steps": 220, "total_steps": 384, "loss": 0.1536, "lr": 4.117901168710959e-05, "epoch": 6.896, "percentage": 57.29, "elapsed_time": "0:26:46", "remaining_time": "0:19:57"}
-{"current_steps": 224, "total_steps": 384, "eval_loss": 0.24610178172588348, "epoch": 7.0, "percentage": 58.33, "elapsed_time": "0:27:32", "remaining_time": "0:19:40"}
-{"current_steps": 230, "total_steps": 384, "loss": 0.1252, "lr": 3.705904774487396e-05, "epoch": 7.192, "percentage": 59.9, "elapsed_time": "0:28:14", "remaining_time": "0:18:54"}
-{"current_steps": 240, "total_steps": 384, "loss": 0.1358, "lr": 3.3031324370510394e-05, "epoch": 7.5120000000000005, "percentage": 62.5, "elapsed_time": "0:29:22", "remaining_time": "0:17:37"}
-{"current_steps": 250, "total_steps": 384, "loss": 0.1484, "lr": 2.9124550385746857e-05, "epoch": 7.832, "percentage": 65.1, "elapsed_time": "0:30:31", "remaining_time": "0:16:21"}
-{"current_steps": 256, "total_steps": 384, "eval_loss": 0.2468687891960144, "epoch": 8.0, "percentage": 66.67, "elapsed_time": "0:31:30", "remaining_time": "0:15:45"}
-{"current_steps": 260, "total_steps": 384, "loss": 0.1295, "lr": 2.5366572508799886e-05, "epoch": 8.128, "percentage": 67.71, "elapsed_time": "0:31:58", "remaining_time": "0:15:14"}
-{"current_steps": 270, "total_steps": 384, "loss": 0.1198, "lr": 2.1784176868432376e-05, "epoch": 8.448, "percentage": 70.31, "elapsed_time": "0:33:05", "remaining_time": "0:13:58"}
-{"current_steps": 280, "total_steps": 384, "loss": 0.1132, "lr": 1.8402898077684804e-05, "epoch": 8.768, "percentage": 72.92, "elapsed_time": "0:34:12", "remaining_time": "0:12:42"}
-{"current_steps": 288, "total_steps": 384, "eval_loss": 0.2507466971874237, "epoch": 9.0, "percentage": 75.0, "elapsed_time": "0:35:25", "remaining_time": "0:11:48"}
-{"current_steps": 290, "total_steps": 384, "loss": 0.1069, "lr": 1.5246837228164907e-05, "epoch": 9.064, "percentage": 75.52, "elapsed_time": "0:35:39", "remaining_time": "0:11:33"}
-{"current_steps": 300, "total_steps": 384, "loss": 0.1192, "lr": 1.2338490102196825e-05, "epoch": 9.384, "percentage": 78.12, "elapsed_time": "0:36:47", "remaining_time": "0:10:18"}
-{"current_steps": 310, "total_steps": 384, "loss": 0.108, "lr": 9.69858682729976e-06, "epoch": 9.704, "percentage": 80.73, "elapsed_time": "0:37:55", "remaining_time": "0:09:03"}
-{"current_steps": 320, "total_steps": 384, "loss": 0.0924, "lr": 7.345944115907422e-06, "epoch": 10.0, "percentage": 83.33, "elapsed_time": "0:38:58", "remaining_time": "0:07:47"}
-{"current_steps": 320, "total_steps": 384, "eval_loss": 0.2528049647808075, "epoch": 10.0, "percentage": 83.33, "elapsed_time": "0:39:21", "remaining_time": "0:07:52"}
-{"current_steps": 330, "total_steps": 384, "loss": 0.1069, "lr": 5.2973311435349725e-06, "epoch": 10.32, "percentage": 85.94, "elapsed_time": "0:40:30", "remaining_time": "0:06:37"}
-{"current_steps": 340, "total_steps": 384, "loss": 0.1041, "lr": 3.567350021386895e-06, "epoch": 10.64, "percentage": 88.54, "elapsed_time": "0:41:38", "remaining_time": "0:05:23"}
-{"current_steps": 350, "total_steps": 384, "loss": 0.097, "lr": 2.1683317153742776e-06, "epoch": 10.96, "percentage": 91.15, "elapsed_time": "0:42:47", "remaining_time": "0:04:09"}
-{"current_steps": 352, "total_steps": 384, "eval_loss": 0.2560921013355255, "epoch": 11.0, "percentage": 91.67, "elapsed_time": "0:43:19", "remaining_time": "0:03:56"}
-{"current_steps": 360, "total_steps": 384, "loss": 0.0765, "lr": 1.1102481534098374e-06, "epoch": 11.256, "percentage": 93.75, "elapsed_time": "0:44:14", "remaining_time": "0:02:56"}
-{"current_steps": 370, "total_steps": 384, "loss": 0.0933, "lr": 4.006411474628491e-07, "epoch": 11.576, "percentage": 96.35, "elapsed_time": "0:45:20", "remaining_time": "0:01:42"}
-{"current_steps": 380, "total_steps": 384, "loss": 0.1146, "lr": 4.456863700363356e-08, "epoch": 11.896, "percentage": 98.96, "elapsed_time": "0:46:28", "remaining_time": "0:00:29"}
-{"current_steps": 384, "total_steps": 384, "eval_loss": 0.2565154731273651, "epoch": 12.0, "percentage": 100.0, "elapsed_time": "0:47:16", "remaining_time": "0:00:00"}
-{"current_steps": 384, "total_steps": 384, "epoch": 12.0, "percentage": 100.0, "elapsed_time": "0:47:16", "remaining_time": "0:00:00"}
+{"current_steps": 10, "total_steps": 369, "loss": 1.4338, "lr": 7.500000000000001e-05, "epoch": 0.23952095808383234, "percentage": 2.71, "elapsed_time": "0:01:07", "remaining_time": "0:40:22"}
+{"current_steps": 20, "total_steps": 369, "loss": 0.8267, "lr": 9.990516643685222e-05, "epoch": 0.47904191616766467, "percentage": 5.42, "elapsed_time": "0:02:13", "remaining_time": "0:38:44"}
+{"current_steps": 30, "total_steps": 369, "loss": 0.4881, "lr": 9.944154131125642e-05, "epoch": 0.718562874251497, "percentage": 8.13, "elapsed_time": "0:03:20", "remaining_time": "0:37:42"}
+{"current_steps": 40, "total_steps": 369, "loss": 0.3591, "lr": 9.859528969650738e-05, "epoch": 0.9580838323353293, "percentage": 10.84, "elapsed_time": "0:04:27", "remaining_time": "0:36:39"}
+{"current_steps": 42, "total_steps": 369, "eval_loss": 0.39884674549102783, "epoch": 1.0, "percentage": 11.38, "elapsed_time": "0:05:09", "remaining_time": "0:40:08"}
+{"current_steps": 50, "total_steps": 369, "loss": 0.4172, "lr": 9.737296070648186e-05, "epoch": 1.1916167664670658, "percentage": 13.55, "elapsed_time": "0:06:04", "remaining_time": "0:38:44"}
+{"current_steps": 60, "total_steps": 369, "loss": 0.3893, "lr": 9.57840139057007e-05, "epoch": 1.4311377245508983, "percentage": 16.26, "elapsed_time": "0:07:10", "remaining_time": "0:36:58"}
+{"current_steps": 70, "total_steps": 369, "loss": 0.3658, "lr": 9.384074610206495e-05, "epoch": 1.6706586826347305, "percentage": 18.97, "elapsed_time": "0:08:19", "remaining_time": "0:35:35"}
+{"current_steps": 80, "total_steps": 369, "loss": 0.2673, "lr": 9.155819618225708e-05, "epoch": 1.910179640718563, "percentage": 21.68, "elapsed_time": "0:09:27", "remaining_time": "0:34:09"}
+{"current_steps": 84, "total_steps": 369, "eval_loss": 0.325968861579895, "epoch": 2.0, "percentage": 22.76, "elapsed_time": "0:10:21", "remaining_time": "0:35:08"}
+{"current_steps": 90, "total_steps": 369, "loss": 0.3165, "lr": 8.895402872628352e-05, "epoch": 2.143712574850299, "percentage": 24.39, "elapsed_time": "0:11:02", "remaining_time": "0:34:12"}
+{"current_steps": 100, "total_steps": 369, "loss": 0.336, "lr": 8.604839730186125e-05, "epoch": 2.3832335329341316, "percentage": 27.1, "elapsed_time": "0:12:09", "remaining_time": "0:32:42"}
+{"current_steps": 110, "total_steps": 369, "loss": 0.288, "lr": 8.286378849660896e-05, "epoch": 2.622754491017964, "percentage": 29.81, "elapsed_time": "0:13:17", "remaining_time": "0:31:17"}
+{"current_steps": 120, "total_steps": 369, "loss": 0.2425, "lr": 7.942484789507283e-05, "epoch": 2.8622754491017965, "percentage": 32.52, "elapsed_time": "0:14:25", "remaining_time": "0:29:55"}
+{"current_steps": 126, "total_steps": 369, "eval_loss": 0.289765864610672, "epoch": 3.0, "percentage": 34.15, "elapsed_time": "0:15:33", "remaining_time": "0:29:59"}
+{"current_steps": 130, "total_steps": 369, "loss": 0.2304, "lr": 7.57581893473448e-05, "epoch": 3.095808383233533, "percentage": 35.23, "elapsed_time": "0:15:59", "remaining_time": "0:29:24"}
+{"current_steps": 140, "total_steps": 369, "loss": 0.2905, "lr": 7.18921890053375e-05, "epoch": 3.3353293413173652, "percentage": 37.94, "elapsed_time": "0:17:08", "remaining_time": "0:28:03"}
+{"current_steps": 150, "total_steps": 369, "loss": 0.2336, "lr": 6.785676572066225e-05, "epoch": 3.5748502994011977, "percentage": 40.65, "elapsed_time": "0:18:15", "remaining_time": "0:26:38"}
+{"current_steps": 160, "total_steps": 369, "loss": 0.2069, "lr": 6.368314950360415e-05, "epoch": 3.81437125748503, "percentage": 43.36, "elapsed_time": "0:19:24", "remaining_time": "0:25:20"}
+{"current_steps": 168, "total_steps": 369, "eval_loss": 0.2659221291542053, "epoch": 4.0, "percentage": 45.53, "elapsed_time": "0:20:44", "remaining_time": "0:24:49"}
+{"current_steps": 170, "total_steps": 369, "loss": 0.1815, "lr": 5.940363983508257e-05, "epoch": 4.047904191616767, "percentage": 46.07, "elapsed_time": "0:20:58", "remaining_time": "0:24:33"}
+{"current_steps": 180, "total_steps": 369, "loss": 0.2426, "lr": 5.5051355702012893e-05, "epoch": 4.287425149700598, "percentage": 48.78, "elapsed_time": "0:22:07", "remaining_time": "0:23:14"}
+{"current_steps": 190, "total_steps": 369, "loss": 0.1998, "lr": 5.0659979290537954e-05, "epoch": 4.526946107784431, "percentage": 51.49, "elapsed_time": "0:23:13", "remaining_time": "0:21:52"}
+{"current_steps": 200, "total_steps": 369, "loss": 0.1713, "lr": 4.626349532067879e-05, "epoch": 4.766467065868263, "percentage": 54.2, "elapsed_time": "0:24:21", "remaining_time": "0:20:35"}
+{"current_steps": 210, "total_steps": 369, "loss": 0.138, "lr": 4.189592803968563e-05, "epoch": 5.0, "percentage": 56.91, "elapsed_time": "0:25:26", "remaining_time": "0:19:15"}
+{"current_steps": 210, "total_steps": 369, "eval_loss": 0.2524815797805786, "epoch": 5.0, "percentage": 56.91, "elapsed_time": "0:25:56", "remaining_time": "0:19:38"}
+{"current_steps": 220, "total_steps": 369, "loss": 0.199, "lr": 3.759107790948882e-05, "epoch": 5.2395209580838324, "percentage": 59.62, "elapsed_time": "0:27:06", "remaining_time": "0:18:21"}
+{"current_steps": 230, "total_steps": 369, "loss": 0.1654, "lr": 3.338226002601703e-05, "epoch": 5.479041916167665, "percentage": 62.33, "elapsed_time": "0:28:12", "remaining_time": "0:17:02"}
+{"current_steps": 240, "total_steps": 369, "loss": 0.1459, "lr": 2.9302046294747497e-05, "epoch": 5.718562874251497, "percentage": 65.04, "elapsed_time": "0:29:19", "remaining_time": "0:15:45"}
+{"current_steps": 250, "total_steps": 369, "loss": 0.1251, "lr": 2.5382013357782893e-05, "epoch": 5.95808383233533, "percentage": 67.75, "elapsed_time": "0:30:26", "remaining_time": "0:14:29"}
+{"current_steps": 252, "total_steps": 369, "eval_loss": 0.24969537556171417, "epoch": 6.0, "percentage": 68.29, "elapsed_time": "0:31:08", "remaining_time": "0:14:27"}
+{"current_steps": 260, "total_steps": 369, "loss": 0.1486, "lr": 2.1652498223239427e-05, "epoch": 6.191616766467066, "percentage": 70.46, "elapsed_time": "0:32:03", "remaining_time": "0:13:26"}
+{"current_steps": 270, "total_steps": 369, "loss": 0.1527, "lr": 1.814236348812211e-05, "epoch": 6.431137724550898, "percentage": 73.17, "elapsed_time": "0:33:10", "remaining_time": "0:12:09"}
+{"current_steps": 280, "total_steps": 369, "loss": 0.1344, "lr": 1.4878773971620074e-05, "epoch": 6.6706586826347305, "percentage": 75.88, "elapsed_time": "0:34:19", "remaining_time": "0:10:54"}
+{"current_steps": 290, "total_steps": 369, "loss": 0.1072, "lr": 1.1886986487449475e-05, "epoch": 6.910179640718563, "percentage": 78.59, "elapsed_time": "0:35:26", "remaining_time": "0:09:39"}
+{"current_steps": 294, "total_steps": 369, "eval_loss": 0.24749523401260376, "epoch": 7.0, "percentage": 79.67, "elapsed_time": "0:36:20", "remaining_time": "0:09:16"}
+{"current_steps": 300, "total_steps": 369, "loss": 0.1224, "lr": 9.190154382188921e-06, "epoch": 7.1437125748503, "percentage": 81.3, "elapsed_time": "0:37:01", "remaining_time": "0:08:30"}
+{"current_steps": 310, "total_steps": 369, "loss": 0.1493, "lr": 6.809148352279182e-06, "epoch": 7.383233532934132, "percentage": 84.01, "elapsed_time": "0:38:08", "remaining_time": "0:07:15"}
+{"current_steps": 320, "total_steps": 369, "loss": 0.1126, "lr": 4.762394926378477e-06, "epoch": 7.6227544910179645, "percentage": 86.72, "elapsed_time": "0:39:16", "remaining_time": "0:06:00"}
+{"current_steps": 330, "total_steps": 369, "loss": 0.1059, "lr": 3.065733863053072e-06, "epoch": 7.862275449101796, "percentage": 89.43, "elapsed_time": "0:40:24", "remaining_time": "0:04:46"}
+{"current_steps": 336, "total_steps": 369, "eval_loss": 0.2483372986316681, "epoch": 8.0, "percentage": 91.06, "elapsed_time": "0:41:32", "remaining_time": "0:04:04"}
+{"current_steps": 340, "total_steps": 369, "loss": 0.0986, "lr": 1.7322955673980678e-06, "epoch": 8.095808383233534, "percentage": 92.14, "elapsed_time": "0:41:58", "remaining_time": "0:03:34"}
+{"current_steps": 350, "total_steps": 369, "loss": 0.1465, "lr": 7.723994752570462e-07, "epoch": 8.335329341317365, "percentage": 94.85, "elapsed_time": "0:43:08", "remaining_time": "0:02:20"}
+{"current_steps": 360, "total_steps": 369, "loss": 0.1073, "lr": 1.9347419144180035e-07, "epoch": 8.574850299401197, "percentage": 97.56, "elapsed_time": "0:44:14", "remaining_time": "0:01:06"}
+{"current_steps": 369, "total_steps": 369, "eval_loss": 0.24962204694747925, "epoch": 8.790419161676647, "percentage": 100.0, "elapsed_time": "0:45:48", "remaining_time": "0:00:00"}
+{"current_steps": 369, "total_steps": 369, "epoch": 8.790419161676647, "percentage": 100.0, "elapsed_time": "0:45:48", "remaining_time": "0:00:00"}
+{"current_steps": 5, "total_steps": 65, "percentage": 7.69, "elapsed_time": "0:00:01", "remaining_time": "0:00:16"}
+{"current_steps": 10, "total_steps": 65, "percentage": 15.38, "elapsed_time": "0:00:03", "remaining_time": "0:00:16"}
+{"current_steps": 15, "total_steps": 65, "percentage": 23.08, "elapsed_time": "0:00:04", "remaining_time": "0:00:15"}
+{"current_steps": 20, "total_steps": 65, "percentage": 30.77, "elapsed_time": "0:00:06", "remaining_time": "0:00:14"}
+{"current_steps": 25, "total_steps": 65, "percentage": 38.46, "elapsed_time": "0:00:07", "remaining_time": "0:00:12"}
+{"current_steps": 30, "total_steps": 65, "percentage": 46.15, "elapsed_time": "0:00:09", "remaining_time": "0:00:11"}
+{"current_steps": 35, "total_steps": 65, "percentage": 53.85, "elapsed_time": "0:00:11", "remaining_time": "0:00:09"}
+{"current_steps": 40, "total_steps": 65, "percentage": 61.54, "elapsed_time": "0:00:12", "remaining_time": "0:00:08"}
+{"current_steps": 45, "total_steps": 65, "percentage": 69.23, "elapsed_time": "0:00:14", "remaining_time": "0:00:06"}
+{"current_steps": 50, "total_steps": 65, "percentage": 76.92, "elapsed_time": "0:00:16", "remaining_time": "0:00:04"}
+{"current_steps": 55, "total_steps": 65, "percentage": 84.62, "elapsed_time": "0:00:18", "remaining_time": "0:00:03"}
+{"current_steps": 60, "total_steps": 65, "percentage": 92.31, "elapsed_time": "0:00:20", "remaining_time": "0:00:01"}
+{"current_steps": 65, "total_steps": 65, "percentage": 100.0, "elapsed_time": "0:00:22", "remaining_time": "0:00:00"}
+{"current_steps": 0, "total_steps": 65, "eval_loss": 1.6609941720962524, "percentage": 0.0, "elapsed_time": "0:00:22", "remaining_time": "0:00:00"}

trainer_state.json

@@ -2,401 +2,360 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 12.0,
+  "epoch": 8.790419161676647,
   "eval_steps": 500,
-  "global_step": 384,
+  "global_step": 369,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.32,
-      "grad_norm": 0.5422136187553406,
+      "epoch": 0.23952095808383234,
+      "grad_norm": 0.7001179456710815,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 1.4075,
+      "loss": 1.4338,
       "step": 10
     },
     {
-      "epoch": 0.64,
-      "grad_norm": 0.24980580806732178,
-      "learning_rate": 9.991265793145479e-05,
-      "loss": 0.8351,
+      "epoch": 0.47904191616766467,
+      "grad_norm": 0.27172571420669556,
+      "learning_rate": 9.990516643685222e-05,
+      "loss": 0.8267,
       "step": 20
     },
     {
-      "epoch": 0.96,
-      "grad_norm": 0.1963978260755539,
-      "learning_rate": 9.948559446825412e-05,
+      "epoch": 0.718562874251497,
+      "grad_norm": 0.19638711214065552,
+      "learning_rate": 9.944154131125642e-05,
       "loss": 0.4881,
       "step": 30
     },
     {
-      "epoch": 1.0,
-      "eval_accuracy": 0.8943875280150239,
-      "eval_loss": 0.439687579870224,
-      "eval_runtime": 23.0702,
-      "eval_samples_per_second": 16.688,
-      "eval_steps_per_second": 2.124,
-      "step": 32
-    },
-    {
-      "epoch": 1.256,
-      "grad_norm": 0.12824280560016632,
-      "learning_rate": 9.870580739976935e-05,
-      "loss": 0.3822,
+      "epoch": 0.9580838323353293,
+      "grad_norm": 0.20754235982894897,
+      "learning_rate": 9.859528969650738e-05,
+      "loss": 0.3591,
       "step": 40
     },
     {
-      "epoch": 1.576,
-      "grad_norm": 0.13147476315498352,
-      "learning_rate": 9.757885489518297e-05,
-      "loss": 0.4258,
+      "epoch": 1.0,
+      "eval_accuracy": 0.9034878200745683,
+      "eval_loss": 0.39884674549102783,
+      "eval_runtime": 30.4717,
+      "eval_samples_per_second": 12.635,
+      "eval_steps_per_second": 2.133,
+      "step": 42
+    },
+    {
+      "epoch": 1.1916167664670658,
+      "grad_norm": 0.16468322277069092,
+      "learning_rate": 9.737296070648186e-05,
+      "loss": 0.4172,
       "step": 50
     },
     {
-      "epoch": 1.896,
-      "grad_norm": 0.12476492673158646,
-      "learning_rate": 9.611276965077099e-05,
-      "loss": 0.338,
+      "epoch": 1.4311377245508983,
+      "grad_norm": 0.1612997055053711,
+      "learning_rate": 9.57840139057007e-05,
+      "loss": 0.3893,
       "step": 60
     },
     {
-      "epoch": 2.0,
-      "eval_accuracy": 0.9137864356203533,
-      "eval_loss": 0.3505571186542511,
-      "eval_runtime": 23.0857,
-      "eval_samples_per_second": 16.677,
-      "eval_steps_per_second": 2.123,
-      "step": 64
-    },
-    {
-      "epoch": 2.192,
-      "grad_norm": 0.136331707239151,
-      "learning_rate": 9.431800163442041e-05,
-      "loss": 0.3284,
+      "epoch": 1.6706586826347305,
+      "grad_norm": 0.16499397158622742,
+      "learning_rate": 9.384074610206495e-05,
+      "loss": 0.3658,
       "step": 70
     },
     {
-      "epoch": 2.512,
-      "grad_norm": 0.15875931084156036,
-      "learning_rate": 9.220734360030907e-05,
-      "loss": 0.3177,
+      "epoch": 1.910179640718563,
+      "grad_norm": 0.14762192964553833,
+      "learning_rate": 9.155819618225708e-05,
+      "loss": 0.2673,
       "step": 80
     },
     {
-      "epoch": 2.832,
-      "grad_norm": 0.20174619555473328,
-      "learning_rate": 8.979583990466454e-05,
-      "loss": 0.3097,
+      "epoch": 2.0,
+      "eval_accuracy": 0.9196909824354998,
+      "eval_loss": 0.325968861579895,
+      "eval_runtime": 30.3583,
+      "eval_samples_per_second": 12.682,
+      "eval_steps_per_second": 2.141,
+      "step": 84
+    },
+    {
+      "epoch": 2.143712574850299,
+      "grad_norm": 0.1777992993593216,
+      "learning_rate": 8.895402872628352e-05,
+      "loss": 0.3165,
       "step": 90
     },
     {
-      "epoch": 3.0,
-      "eval_accuracy": 0.9229765721615035,
-      "eval_loss": 0.3097846210002899,
-      "eval_runtime": 23.0998,
-      "eval_samples_per_second": 16.667,
-      "eval_steps_per_second": 2.121,
-      "step": 96
-    },
-    {
-      "epoch": 3.128,
-      "grad_norm": 0.16589179635047913,
-      "learning_rate": 8.710067927254555e-05,
-      "loss": 0.2869,
+      "epoch": 2.3832335329341316,
+      "grad_norm": 0.2070295810699463,
+      "learning_rate": 8.604839730186125e-05,
+      "loss": 0.336,
       "step": 100
     },
     {
-      "epoch": 3.448,
-      "grad_norm": 0.2508234977722168,
-      "learning_rate": 8.414107227998329e-05,
-      "loss": 0.264,
+      "epoch": 2.622754491017964,
+      "grad_norm": 0.1750418245792389,
+      "learning_rate": 8.286378849660896e-05,
+      "loss": 0.288,
       "step": 110
     },
     {
-      "epoch": 3.768,
-      "grad_norm": 0.2155790477991104,
-      "learning_rate": 8.093811442476573e-05,
-      "loss": 0.293,
+      "epoch": 2.8622754491017965,
+      "grad_norm": 0.20214907824993134,
+      "learning_rate": 7.942484789507283e-05,
+      "loss": 0.2425,
       "step": 120
     },
     {
-      "epoch": 4.0,
-      "eval_accuracy": 0.9309821545048526,
-      "eval_loss": 0.2800418436527252,
-      "eval_runtime": 23.147,
-      "eval_samples_per_second": 16.633,
-      "eval_steps_per_second": 2.117,
-      "step": 128
-    },
-    {
-      "epoch": 4.064,
-      "grad_norm": 0.19977881014347076,
-      "learning_rate": 7.751463576186957e-05,
-      "loss": 0.2517,
+      "epoch": 3.0,
+      "eval_accuracy": 0.9289047847079928,
+      "eval_loss": 0.289765864610672,
+      "eval_runtime": 30.3541,
+      "eval_samples_per_second": 12.684,
+      "eval_steps_per_second": 2.141,
+      "step": 126
+    },
+    {
+      "epoch": 3.095808383233533,
+      "grad_norm": 0.2640642523765564,
+      "learning_rate": 7.57581893473448e-05,
+      "loss": 0.2304,
       "step": 130
     },
     {
-      "epoch": 4.384,
-      "grad_norm": 0.2274104505777359,
-      "learning_rate": 7.389503817530905e-05,
-      "loss": 0.2501,
+      "epoch": 3.3353293413173652,
+      "grad_norm": 0.26811081171035767,
+      "learning_rate": 7.18921890053375e-05,
+      "loss": 0.2905,
       "step": 140
     },
     {
-      "epoch": 4.704,
-      "grad_norm": 0.22418978810310364,
-      "learning_rate": 7.010512144629579e-05,
-      "loss": 0.2115,
+      "epoch": 3.5748502994011977,
+      "grad_norm": 0.27033841609954834,
+      "learning_rate": 6.785676572066225e-05,
+      "loss": 0.2336,
       "step": 150
     },
     {
-      "epoch": 5.0,
-      "grad_norm": 0.43173882365226746,
-      "learning_rate": 6.61718993574619e-05,
-      "loss": 0.202,
-      "step": 160
-    },
-    {
-      "epoch": 5.0,
-      "eval_accuracy": 0.9356952198262245,
-      "eval_loss": 0.26041463017463684,
-      "eval_runtime": 23.1637,
-      "eval_samples_per_second": 16.621,
-      "eval_steps_per_second": 2.115,
+      "epoch": 3.81437125748503,
+      "grad_norm": 0.22121788561344147,
+      "learning_rate": 6.368314950360415e-05,
+      "loss": 0.2069,
       "step": 160
     },
     {
-      "epoch": 5.32,
-      "grad_norm": 0.22742323577404022,
-      "learning_rate": 6.2123407143919e-05,
-      "loss": 0.195,
+      "epoch": 4.0,
+      "eval_accuracy": 0.9345709163732674,
+      "eval_loss": 0.2659221291542053,
+      "eval_runtime": 30.3929,
+      "eval_samples_per_second": 12.667,
+      "eval_steps_per_second": 2.139,
+      "step": 168
+    },
+    {
+      "epoch": 4.047904191616767,
+      "grad_norm": 0.2626728117465973,
+      "learning_rate": 5.940363983508257e-05,
+      "loss": 0.1815,
       "step": 170
     },
     {
-      "epoch": 5.64,
-      "grad_norm": 0.2773088812828064,
-      "learning_rate": 5.798850166360461e-05,
-      "loss": 0.2085,
+      "epoch": 4.287425149700598,
+      "grad_norm": 0.2664365768432617,
+      "learning_rate": 5.5051355702012893e-05,
+      "loss": 0.2426,
       "step": 180
     },
     {
-      "epoch": 5.96,
-      "grad_norm": 0.326659232378006,
-      "learning_rate": 5.379665571126231e-05,
-      "loss": 0.1846,
+      "epoch": 4.526946107784431,
+      "grad_norm": 0.2793976962566376,
+      "learning_rate": 5.0659979290537954e-05,
+      "loss": 0.1998,
       "step": 190
     },
     {
-      "epoch": 6.0,
-      "eval_accuracy": 0.9394547551491287,
-      "eval_loss": 0.250407874584198,
-      "eval_runtime": 23.1202,
-      "eval_samples_per_second": 16.652,
-      "eval_steps_per_second": 2.119,
-      "step": 192
-    },
-    {
-      "epoch": 6.256,
-      "grad_norm": 0.24400818347930908,
-      "learning_rate": 4.957774794214508e-05,
-      "loss": 0.1577,
+      "epoch": 4.766467065868263,
+      "grad_norm": 0.2911984622478485,
+      "learning_rate": 4.626349532067879e-05,
+      "loss": 0.1713,
       "step": 200
     },
     {
-      "epoch": 6.576,
-      "grad_norm": 0.3049585819244385,
-      "learning_rate": 4.5361849902824674e-05,
-      "loss": 0.1718,
+      "epoch": 5.0,
+      "grad_norm": 0.2870716452598572,
+      "learning_rate": 4.189592803968563e-05,
+      "loss": 0.138,
       "step": 210
     },
     {
-      "epoch": 6.896,
-      "grad_norm": 0.2874772846698761,
-      "learning_rate": 4.117901168710959e-05,
-      "loss": 0.1536,
+      "epoch": 5.0,
+      "eval_accuracy": 0.9391251537470097,
+      "eval_loss": 0.2524815797805786,
+      "eval_runtime": 30.4065,
+      "eval_samples_per_second": 12.662,
+      "eval_steps_per_second": 2.138,
+      "step": 210
+    },
+    {
+      "epoch": 5.2395209580838324,
+      "grad_norm": 0.3472362756729126,
+      "learning_rate": 3.759107790948882e-05,
+      "loss": 0.199,
       "step": 220
     },
     {
-      "epoch": 7.0,
-      "eval_accuracy": 0.9421747829656392,
-      "eval_loss": 0.24610178172588348,
-      "eval_runtime": 23.0966,
-      "eval_samples_per_second": 16.669,
-      "eval_steps_per_second": 2.122,
-      "step": 224
-    },
-    {
-      "epoch": 7.192,
-      "grad_norm": 0.24850666522979736,
-      "learning_rate": 3.705904774487396e-05,
-      "loss": 0.1252,
+      "epoch": 5.479041916167665,
+      "grad_norm": 0.33083346486091614,
+      "learning_rate": 3.338226002601703e-05,
+      "loss": 0.1654,
       "step": 230
     },
     {
-      "epoch": 7.5120000000000005,
-      "grad_norm": 0.2970203459262848,
-      "learning_rate": 3.3031324370510394e-05,
-      "loss": 0.1358,
+      "epoch": 5.718562874251497,
+      "grad_norm": 0.3220432996749878,
+      "learning_rate": 2.9302046294747497e-05,
+      "loss": 0.1459,
       "step": 240
     },
     {
-      "epoch": 7.832,
-      "grad_norm": 0.28144457936286926,
-      "learning_rate": 2.9124550385746857e-05,
-      "loss": 0.1484,
+      "epoch": 5.95808383233533,
+      "grad_norm": 0.35546302795410156,
+      "learning_rate": 2.5382013357782893e-05,
+      "loss": 0.1251,
       "step": 250
     },
     {
-      "epoch": 8.0,
-      "eval_accuracy": 0.9448626884131439,
-      "eval_loss": 0.2468687891960144,
-      "eval_runtime": 23.1407,
-      "eval_samples_per_second": 16.637,
-      "eval_steps_per_second": 2.117,
-      "step": 256
-    },
-    {
-      "epoch": 8.128,
-      "grad_norm": 0.24285829067230225,
-      "learning_rate": 2.5366572508799886e-05,
-      "loss": 0.1295,
+      "epoch": 6.0,
+      "eval_accuracy": 0.9430640763868036,
+      "eval_loss": 0.24969537556171417,
+      "eval_runtime": 30.3846,
+      "eval_samples_per_second": 12.671,
+      "eval_steps_per_second": 2.139,
+      "step": 252
+    },
+    {
+      "epoch": 6.191616766467066,
+      "grad_norm": 0.35284000635147095,
+      "learning_rate": 2.1652498223239427e-05,
+      "loss": 0.1486,
       "step": 260
     },
     {
-      "epoch": 8.448,
-      "grad_norm": 0.24960064888000488,
-      "learning_rate": 2.1784176868432376e-05,
-      "loss": 0.1198,
+      "epoch": 6.431137724550898,
+      "grad_norm": 0.31098905205726624,
+      "learning_rate": 1.814236348812211e-05,
+      "loss": 0.1527,
       "step": 270
     },
     {
-      "epoch": 8.768,
-      "grad_norm": 0.2702842354774475,
-      "learning_rate": 1.8402898077684804e-05,
-      "loss": 0.1132,
+      "epoch": 6.6706586826347305,
+      "grad_norm": 0.34438222646713257,
+      "learning_rate": 1.4878773971620074e-05,
+      "loss": 0.1344,
       "step": 280
     },
     {
-      "epoch": 9.0,
-      "eval_accuracy": 0.9458956775395686,
-      "eval_loss": 0.2507466971874237,
-      "eval_runtime": 23.1364,
-      "eval_samples_per_second": 16.64,
-      "eval_steps_per_second": 2.118,
-      "step": 288
-    },
-    {
-      "epoch": 9.064,
-      "grad_norm": 0.2220270186662674,
-      "learning_rate": 1.5246837228164907e-05,
-      "loss": 0.1069,
+      "epoch": 6.910179640718563,
+      "grad_norm": 0.2907264232635498,
+      "learning_rate": 1.1886986487449475e-05,
+      "loss": 0.1072,
       "step": 290
     },
     {
-      "epoch": 9.384,
-      "grad_norm": 0.29468902945518494,
-      "learning_rate": 1.2338490102196825e-05,
-      "loss": 0.1192,
+      "epoch": 7.0,
+      "eval_accuracy": 0.9438642485491895,
+      "eval_loss": 0.24749523401260376,
+      "eval_runtime": 30.43,
+      "eval_samples_per_second": 12.652,
+      "eval_steps_per_second": 2.136,
+      "step": 294
+    },
+    {
+      "epoch": 7.1437125748503,
+      "grad_norm": 0.3019481301307678,
+      "learning_rate": 9.190154382188921e-06,
+      "loss": 0.1224,
       "step": 300
     },
     {
-      "epoch": 9.704,
-      "grad_norm": 0.26781126856803894,
-      "learning_rate": 9.69858682729976e-06,
-      "loss": 0.108,
+      "epoch": 7.383233532934132,
+      "grad_norm": 0.29105937480926514,
+      "learning_rate": 6.809148352279182e-06,
+      "loss": 0.1493,
       "step": 310
     },
     {
-      "epoch": 10.0,
-      "grad_norm": 0.3624833822250366,
-      "learning_rate": 7.345944115907422e-06,
-      "loss": 0.0924,
+      "epoch": 7.6227544910179645,
+      "grad_norm": 0.2779920697212219,
+      "learning_rate": 4.762394926378477e-06,
+      "loss": 0.1126,
       "step": 320
     },
     {
-      "epoch": 10.0,
-      "eval_accuracy": 0.9458588957449963,
-      "eval_loss": 0.2528049647808075,
-      "eval_runtime": 23.1153,
-      "eval_samples_per_second": 16.656,
-      "eval_steps_per_second": 2.12,
-      "step": 320
-    },
-    {
-      "epoch": 10.32,
-      "grad_norm": 0.27165377140045166,
-      "learning_rate": 5.2973311435349725e-06,
-      "loss": 0.1069,
+      "epoch": 7.862275449101796,
+      "grad_norm": 0.3359215557575226,
+      "learning_rate": 3.065733863053072e-06,
+      "loss": 0.1059,
       "step": 330
     },
     {
-      "epoch": 10.64,
-      "grad_norm": 0.2682035267353058,
-      "learning_rate": 3.567350021386895e-06,
-      "loss": 0.1041,
+      "epoch": 8.0,
+      "eval_accuracy": 0.944612993482476,
+      "eval_loss": 0.2483372986316681,
+      "eval_runtime": 30.4006,
+      "eval_samples_per_second": 12.664,
+      "eval_steps_per_second": 2.138,
+      "step": 336
+    },
+    {
+      "epoch": 8.095808383233534,
+      "grad_norm": 0.26059141755104065,
+      "learning_rate": 1.7322955673980678e-06,
+      "loss": 0.0986,
       "step": 340
     },
     {
-      "epoch": 10.96,
-      "grad_norm": 0.26564228534698486,
-      "learning_rate": 2.1683317153742776e-06,
-      "loss": 0.097,
+      "epoch": 8.335329341317365,
+      "grad_norm": 0.26576822996139526,
+      "learning_rate": 7.723994752570462e-07,
+      "loss": 0.1465,
       "step": 350
     },
     {
-      "epoch": 11.0,
-      "eval_accuracy": 0.9461563024276883,
-      "eval_loss": 0.2560921013355255,
-      "eval_runtime": 23.0822,
-      "eval_samples_per_second": 16.68,
-      "eval_steps_per_second": 2.123,
-      "step": 352
-    },
-    {
-      "epoch": 11.256,
-      "grad_norm": 0.20978766679763794,
-      "learning_rate": 1.1102481534098374e-06,
-      "loss": 0.0765,
+      "epoch": 8.574850299401197,
+      "grad_norm": 0.2771126329898834,
+      "learning_rate": 1.9347419144180035e-07,
+      "loss": 0.1073,
       "step": 360
     },
     {
-      "epoch": 11.576,
-      "grad_norm": 0.21784260869026184,
-      "learning_rate": 4.006411474628491e-07,
-      "loss": 0.0933,
-      "step": 370
-    },
-    {
-      "epoch": 11.896,
-      "grad_norm": 0.2561984956264496,
-      "learning_rate": 4.456863700363356e-08,
-      "loss": 0.1146,
-      "step": 380
-    },
-    {
-      "epoch": 12.0,
-      "eval_accuracy": 0.9461611343879004,
-      "eval_loss": 0.2565154731273651,
-      "eval_runtime": 23.0977,
-      "eval_samples_per_second": 16.668,
-      "eval_steps_per_second": 2.121,
-      "step": 384
+      "epoch": 8.790419161676647,
+      "eval_accuracy": 0.9448452683547727,
+      "eval_loss": 0.24962204694747925,
+      "eval_runtime": 30.226,
+      "eval_samples_per_second": 12.737,
+      "eval_steps_per_second": 2.15,
+      "step": 369
     },
     {
-      "epoch": 12.0,
-      "step": 384,
-      "total_flos": 4.38485641689301e+17,
-      "train_loss": 0.24724087212234735,
-      "train_runtime": 2839.5296,
-      "train_samples_per_second": 4.226,
-      "train_steps_per_second": 0.135
+      "epoch": 8.790419161676647,
+      "step": 369,
+      "total_flos": 3.210182982232965e+17,
+      "train_loss": 0.2637010880602085,
+      "train_runtime": 2751.4148,
+      "train_samples_per_second": 3.271,
+      "train_steps_per_second": 0.134
     }
   ],
   "logging_steps": 10,
-  "max_steps": 384,
+  "max_steps": 369,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 12,
+  "num_train_epochs": 9,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -410,7 +369,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.38485641689301e+17,
+  "total_flos": 3.210182982232965e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6ee030483225b3eb8f716972359663148b4219a065cd4b29cbd0a2e1298c15ee
+oid sha256:89da7b86c6ffda4db368588ecb97943bf3d8149fc8db3215a63b20e819145412
 size 6097