Hyponatremia_L3_100steps_1e6rate_013beta_DPO

This model is a fine-tuned version of tsavage68/Hyponatremia_L3_450steps_1e7rate_SFT on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0000
• Rewards/chosen: 1.9299
• Rewards/rejected: -9.8997
• Rewards/accuracies: 1.0
• Rewards/margins: 11.8296
• Logps/rejected: -72.4257
• Logps/chosen: -16.2838
• Logits/rejected: -1.0137
• Logits/chosen: -0.9231

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: 1e-06
• train_batch_size: 2
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• 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
0.0001	0.2667	50	0.0001	1.7930	-8.0228	1.0	9.8159	-66.1697	-16.7400	-1.0124	-0.9226
0.0	0.5333	100	0.0000	1.9299	-9.8997	1.0	11.8296	-72.4257	-16.2838	-1.0137	-0.9231

Framework versions

• Transformers 4.42.3
• Pytorch 2.0.0+cu117
• Datasets 2.20.0
• Tokenizers 0.19.1