dpo-llama2-deprecated

This model is a fine-tuned version of meta-llama/Llama-2-7b-chat-hf on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5246
• Rewards/chosen: 0.5279
• Rewards/rejected: -0.0974
• Rewards/accuracies: 0.7939
• Rewards/margins: 0.6253
• Logps/rejected: -74.5910
• Logps/chosen: -63.2702
• Logits/rejected: -0.4513
• Logits/chosen: -0.4830

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: 0.0005
• train_batch_size: 1
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• total_eval_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 20
• training_steps: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.6078	0.15	50	0.6238	0.5207	0.0974	0.6588	0.4233	-72.6424	-63.3416	-0.4914	-0.5620
0.5223	0.3	100	0.5246	0.5279	-0.0974	0.7939	0.6253	-74.5910	-63.2702	-0.4513	-0.4830

Framework versions

• PEFT 0.7.1
• Transformers 4.36.2
• Pytorch 2.1.2+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0