metadata
library_name: peft
license: other
base_model: Qwen/Qwen2.5-14B-Instruct
tags:
- llama-factory
- lora
- generated_from_trainer
model-index:
- name: MATH_training_response_Qwen2.5-32B-Instruct
results: []
MATH_training_response_Qwen2.5-32B-Instruct
This model is a fine-tuned version of Qwen/Qwen2.5-14B-Instruct on the MATH_training_Qwen2.5-32B-Instruct dataset. It achieves the following results on the evaluation set:
- Loss: 0.0680
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: 0.0001
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 4
- total_eval_batch_size: 4
- optimizer: Use 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: 0.1
- num_epochs: 2.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.093 | 0.1500 | 200 | 0.0782 |
| 0.0361 | 0.3001 | 400 | 0.0700 |
| 0.0649 | 0.4501 | 600 | 0.0710 |
| 0.0468 | 0.6002 | 800 | 0.0700 |
| 0.1057 | 0.7502 | 1000 | 0.0689 |
| 0.0632 | 0.9002 | 1200 | 0.0670 |
| 0.0446 | 1.0503 | 1400 | 0.0680 |
| 0.0486 | 1.2003 | 1600 | 0.0680 |
| 0.0481 | 1.3503 | 1800 | 0.0675 |
| 0.045 | 1.5004 | 2000 | 0.0689 |
| 0.0471 | 1.6504 | 2200 | 0.0682 |
| 0.0317 | 1.8005 | 2400 | 0.0683 |
| 0.0489 | 1.9505 | 2600 | 0.0682 |
Framework versions
- PEFT 0.12.0
- Transformers 4.46.1
- Pytorch 2.5.1+cu124
- Datasets 3.1.0
- Tokenizers 0.20.3