metadata
library_name: transformers
tags:
- trl
- dpo
- generated_from_trainer
model-index:
- name: zephyr-8b-dpo-full
results: []
zephyr-8b-dpo-full
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5042
- Rewards/chosen: -0.3377
- Rewards/rejected: -1.3264
- Rewards/accuracies: 0.7656
- Rewards/margins: 0.9887
- Logps/rejected: -303.2042
- Logps/chosen: -279.9121
- Logits/rejected: -1.2346
- Logits/chosen: -1.2443
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: 4
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 4
- 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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.4778 | 0.9984 | 477 | 0.5042 | -0.3377 | -1.3264 | 0.7656 | 0.9887 | -303.2042 | -279.9121 | -1.2346 | -1.2443 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.5.1+rocm6.2
- Datasets 3.5.0
- Tokenizers 0.20.3