vanilla_dpo_iter_1
This model is a fine-tuned version of alignment-handbook/zephyr-7b-sft-full on the updated and the original datasets. It achieves the following results on the evaluation set:
- Loss: 0.6274
- Rewards/chosen: -0.0991
- Rewards/rejected: -0.2700
- Rewards/accuracies: 0.6800
- Rewards/margins: 0.1709
- Logps/rejected: -284.5185
- Logps/chosen: -293.9490
- Logits/rejected: -2.5507
- Logits/chosen: -2.6341
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-06
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 16
- total_train_batch_size: 128
- total_eval_batch_size: 8
- 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 |
---|---|---|---|---|---|---|---|---|---|---|---|
0.6186 | 0.63 | 100 | 0.6274 | -0.0991 | -0.2700 | 0.6800 | 0.1709 | -284.5185 | -293.9490 | -2.5507 | -2.6341 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.15.2
- Downloads last month
- 3
Model tree for YYYYYYibo/vanilla_dpo_iter_1
Base model
mistralai/Mistral-7B-v0.1
Finetuned
alignment-handbook/zephyr-7b-sft-full