metadata
tags:
- trl
- dpo
- generated_from_trainer
model-index:
- name: zephyr-7b-dpo-full
results: []
zephyr-7b-dpo-full
This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4913
- Rewards/chosen: -0.8372
- Rewards/rejected: -1.8838
- Rewards/accuracies: 0.7852
- Rewards/margins: 1.0466
- Logps/rejected: -476.1451
- Logps/chosen: -365.3529
- Logits/rejected: -0.9850
- Logits/chosen: -1.0730
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-07
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- total_eval_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.5671 | 0.21 | 100 | 0.5699 | -0.3533 | -0.8899 | 0.7305 | 0.5366 | -376.7557 | -316.9664 | -1.8571 | -1.8679 |
| 0.5413 | 0.42 | 200 | 0.5254 | -0.6740 | -1.4486 | 0.7266 | 0.7745 | -432.6194 | -349.0378 | -1.8679 | -1.9017 |
| 0.4829 | 0.63 | 300 | 0.4956 | -0.8661 | -1.8953 | 0.7812 | 1.0292 | -477.2877 | -368.2400 | -1.1994 | -1.2566 |
| 0.4981 | 0.84 | 400 | 0.4913 | -0.8372 | -1.8838 | 0.7852 | 1.0466 | -476.1451 | -365.3529 | -0.9850 | -1.0730 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.1.2+cu118
- Datasets 2.16.1
- Tokenizers 0.15.2