biomistral-7b-dpo-full-sft-wo-kqa_silver_wogold
This model is a fine-tuned version of Minbyul/biomistral-7b-wo-kqa_silver_wogold-sft on the HuggingFaceH4/ultrafeedback_binarized dataset. It achieves the following results on the evaluation set:
- Loss: 0.3493
- Rewards/chosen: -0.0378
- Rewards/rejected: -1.3358
- Rewards/accuracies: 1.0
- Rewards/margins: 1.2980
- Logps/rejected: -740.3141
- Logps/chosen: -69.0360
- Logits/rejected: -3.3845
- Logits/chosen: -3.7481
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: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- total_eval_batch_size: 32
- 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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0911 | 0.83 | 100 | 0.3514 | -0.0382 | -1.3191 | 1.0 | 1.2809 | -738.6374 | -69.0788 | -3.3885 | -3.7525 |
Framework versions
- Transformers 4.39.0.dev0
- Pytorch 2.1.2
- Datasets 2.14.6
- Tokenizers 0.15.2
- Downloads last month
- 1