tulu2-13b-cost-UI-5e-7-nojudge

This model is a fine-tuned version of allenai/tulu-2-13b on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6912
• Rewards/chosen: -0.0076
• Rewards/rejected: -0.0119
• Rewards/accuracies: 0.5960
• Rewards/margins: 0.0043
• Rewards/margins Max: 0.0285
• Rewards/margins Min: -0.0168
• Rewards/margins Std: 0.0151
• Logps/rejected: -331.3923
• Logps/chosen: -334.6692
• Logits/rejected: -0.8885
• Logits/chosen: -1.0144

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: 2
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• 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	Rewards/margins Max	Rewards/margins Min	Rewards/margins Std	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.6731	1.0	1185	0.6912	-0.0076	-0.0119	0.5960	0.0043	0.0285	-0.0168	0.0151	-331.3923	-334.6692	-0.8885	-1.0144

Framework versions

• PEFT 0.7.1
• Transformers 4.39.0.dev0
• Pytorch 2.1.2+cu121
• Datasets 2.14.6
• Tokenizers 0.15.2