zephyr-7b-dpo-full

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

• Loss: 0.5669
• Rewards/chosen: -0.4034
• Rewards/rejected: -0.9628
• Rewards/accuracies: 0.7461
• Rewards/margins: 0.5594
• Logps/rejected: -372.7966
• Logps/chosen: -315.5382
• Logits/rejected: -2.3014
• Logits/chosen: -2.3480

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
• training_steps: 238

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.5667	0.4982	238	0.5669	-0.4034	-0.9628	0.7461	0.5594	-372.7966	-315.5382	-2.3014	-2.3480

Framework versions

• Transformers 4.45.2
• Pytorch 2.5.1+rocm6.2
• Datasets 3.2.0
• Tokenizers 0.20.3