zephyr-8b-dpo-full

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5301
• Rewards/chosen: -0.9955
• Rewards/rejected: -1.8636
• Rewards/accuracies: 0.7344
• Rewards/margins: 0.8682
• Logps/rejected: -449.0979
• Logps/chosen: -359.8167
• Logits/rejected: -1.4804
• Logits/chosen: -1.4661

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
0.507	0.9984	477	0.5301	-0.9955	-1.8636	0.7344	0.8682	-449.0979	-359.8167	-1.4804	-1.4661

Framework versions

• Transformers 4.45.2
• Pytorch 2.5.1+rocm6.2
• Datasets 3.5.0
• Tokenizers 0.20.3