qwen25_7b_rm_eng_5e5_3ep
This model is a fine-tuned version of Jennny/qwen25_7b_sft_eng_math on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.8192
- Accuracy: 0.7833
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: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 8
- total_train_batch_size: 256
- total_eval_batch_size: 32
- optimizer: Use OptimizerNames.PAGED_ADAMW 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: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.1497 | 0.3347 | 10 | 1.1414 | 0.5967 |
| 0.1812 | 0.6695 | 20 | 0.6084 | 0.73 |
| 0.2564 | 1.0335 | 30 | 0.4931 | 0.7933 |
| 0.1074 | 1.3682 | 40 | 0.6235 | 0.7433 |
| 0.102 | 1.7029 | 50 | 0.6734 | 0.8067 |
| 0.0859 | 2.0669 | 60 | 0.5501 | 0.8 |
| 0.0101 | 2.4017 | 70 | 1.7097 | 0.7767 |
| 0.0021 | 2.7364 | 80 | 1.8192 | 0.7833 |
Framework versions
- Transformers 4.49.0
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.1
- Downloads last month
- 1