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: 0.1695
- Rewards/chosen: 8.1916
- Rewards/rejected: -16.5277
- Rewards/accuracies: 0.9297
- Rewards/margins: 24.7193
- Logps/rejected: -126.2961
- Logps/chosen: -121.3891
- Logits/rejected: -1.8336
- Logits/chosen: -1.8336
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.1474 | 0.21 | 100 | 0.1325 | 6.8739 | -14.1333 | 0.9297 | 21.0072 | -121.5073 | -124.0244 | -1.7724 | -1.7834 |
| 0.2092 | 0.42 | 200 | 0.1567 | 8.4112 | -15.0418 | 0.9336 | 23.4530 | -123.3244 | -120.9499 | -1.8442 | -1.8477 |
| 0.1925 | 0.63 | 300 | 0.1715 | 7.7458 | -16.7009 | 0.9258 | 24.4467 | -126.6425 | -122.2807 | -1.8047 | -1.8054 |
| 0.2762 | 0.84 | 400 | 0.1695 | 8.1916 | -16.5277 | 0.9297 | 24.7193 | -126.2961 | -121.3891 | -1.8336 | -1.8336 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.1.2+cu118
- Datasets 2.16.1
- Tokenizers 0.15.2