metadata
base_model: AI-MO/NuminaMath-72B-CoT
tags:
- alignment-handbook
- generated_from_trainer
- math
- aimo
datasets:
- AI-MO/NuminaMath-TIR
model-index:
- name: qwen2-72b-sft-aimo_v03.00
results: []
license: other
license_name: tongyi-qianwen
language:
- en
qwen2-72b-sft-aimo_v03.00
This model is a fine-tuned version of AI-MO/qwen2-72b-sft on the AI-MO/numina-dataset-tora-v1.0-release-candidate-1 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4792
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 32
- total_train_batch_size: 32
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.3939 | 1.0 | 797 | 0.3793 |
| 0.2618 | 2.0 | 1594 | 0.3876 |
| 0.1141 | 3.0 | 2391 | 0.4310 |
| 0.0363 | 4.0 | 3188 | 0.4792 |
Framework versions
- Transformers 4.42.3
- Pytorch 2.3.0+cu121
- Datasets 2.18.0
- Tokenizers 0.19.1