pairwise-reward-zephyr-7b-sft-qlora-ultrafeedback-binarized-20240925-122042
This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4788
- Accuracy: 0.7536
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: 1.5e-05
- train_batch_size: 16
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.6633 | 0.0526 | 100 | 0.6175 | 0.6929 |
0.6204 | 0.1052 | 200 | 0.5689 | 0.7255 |
0.5895 | 0.1578 | 300 | 0.5263 | 0.7341 |
0.5285 | 0.2104 | 400 | 0.5183 | 0.7431 |
0.4919 | 0.2630 | 500 | 0.5192 | 0.7356 |
0.5085 | 0.3155 | 600 | 0.5057 | 0.7531 |
0.5322 | 0.3681 | 700 | 0.5066 | 0.7486 |
0.4976 | 0.4207 | 800 | 0.4962 | 0.7561 |
0.549 | 0.4733 | 900 | 0.5012 | 0.7647 |
0.5175 | 0.5259 | 1000 | 0.4887 | 0.7587 |
0.4525 | 0.5785 | 1100 | 0.4980 | 0.7551 |
0.4847 | 0.6311 | 1200 | 0.4848 | 0.7516 |
0.5429 | 0.6837 | 1300 | 0.4878 | 0.7481 |
0.4348 | 0.7363 | 1400 | 0.4844 | 0.7551 |
0.4346 | 0.7889 | 1500 | 0.4848 | 0.7521 |
0.513 | 0.8414 | 1600 | 0.4837 | 0.7566 |
0.442 | 0.8940 | 1700 | 0.4814 | 0.7561 |
0.4531 | 0.9466 | 1800 | 0.4796 | 0.7607 |
0.4533 | 0.9992 | 1900 | 0.4788 | 0.7536 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.4.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 14
Model tree for sahandrez/pairwise-reward-zephyr-7b-sft-qlora-ultrafeedback-binarized-20240925-122042
Base model
mistralai/Mistral-7B-v0.1