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: 1.3650
- Rewards/chosen: -0.6695
- Rewards/rejected: -0.0468
- Rewards/accuracies: 0.3828
- Rewards/margins: -0.6227
- Logps/rejected: -518.8555
- Logps/chosen: -397.1692
- Logits/rejected: -4.7484
- Logits/chosen: -4.5219
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: 1e-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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.216 | 0.26 | 100 | 1.1906 | -0.1088 | 0.4672 | 0.3984 | -0.5759 | -513.7159 | -391.5621 | -4.7718 | -4.5558 |
| 0.1925 | 0.51 | 200 | 1.3563 | -0.3996 | 0.2819 | 0.375 | -0.6815 | -515.5683 | -394.4702 | -4.7427 | -4.5166 |
| 0.179 | 0.77 | 300 | 1.3650 | -0.6695 | -0.0468 | 0.3828 | -0.6227 | -518.8555 | -397.1692 | -4.7484 | -4.5219 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.1.2+cu118
- Datasets 2.16.1
- Tokenizers 0.15.2