WeniGPT-2.4.1-Zephyr-7B-zephyr-prompt-LLM_Base_2.0.3_DPO_reduction_variation

This model is a fine-tuned version of HuggingFaceH4/zephyr-7b-beta on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6931
• Rewards/chosen: 0.0
• Rewards/rejected: 0.0
• Rewards/accuracies: 0.0
• Rewards/margins: 0.0
• Logps/rejected: -7781.7466
• Logps/chosen: -316.9807
• Logits/rejected: -6.0606
• Logits/chosen: -5.0858

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: 0.0002
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.03
• training_steps: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.6931	0.36	50	0.6931	0.0	0.0	0.0	0.0	-7781.7466	-316.9807	-6.0606	-5.0858

Framework versions

• PEFT 0.7.1
• Transformers 4.38.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.16.1
• Tokenizers 0.15.1