selective-pairrm-33076849-mt1

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7142
• Rewards/chosen: -1.4159
• Rewards/rejected: -1.4967
• Rewards/accuracies: 0.5469
• Rewards/margins: 0.0808
• Logps/rejected: -567.3423
• Logps/chosen: -543.0017
• Logits/rejected: -3.1535
• Logits/chosen: -3.1646

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: 4
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• total_eval_batch_size: 32
• 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
0.5687	0.32	100	0.7266	-1.4745	-1.5426	0.5352	0.0681	-571.9306	-548.8583	-3.1296	-3.1416
0.5901	0.64	200	0.7177	-1.4102	-1.4850	0.5469	0.0748	-566.1749	-542.4269	-3.1042	-3.1156
0.6278	0.96	300	0.7142	-1.4212	-1.5010	0.5547	0.0798	-567.7791	-543.5320	-3.1529	-3.1641

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2
• Datasets 2.14.6
• Tokenizers 0.15.0