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: 1.0331
- Rewards/chosen: -4.8324
- Rewards/rejected: -5.0569
- Rewards/accuracies: 0.5664
- Rewards/margins: 0.2245
- Logps/rejected: -878.3029
- Logps/chosen: -862.6211
- Logits/rejected: -4.1328
- Logits/chosen: -4.0279
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.4239 | 0.26 | 100 | 0.8541 | -2.3650 | -2.4785 | 0.5547 | 0.1135 | -620.4579 | -615.8743 | -2.6916 | -2.6796 |
| 0.312 | 0.51 | 200 | 1.0052 | -4.1999 | -4.4302 | 0.5547 | 0.2303 | -815.6310 | -799.3696 | -3.7211 | -3.6438 |
| 0.2922 | 0.77 | 300 | 1.0331 | -4.8324 | -5.0569 | 0.5664 | 0.2245 | -878.3029 | -862.6211 | -4.1328 | -4.0279 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.1.2+cu118
- Datasets 2.16.1
- Tokenizers 0.15.2