tulu2-13b-cost-UF-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: 0.6931
• Rewards/chosen: 0.0268
• Rewards/rejected: 0.0260
• Rewards/accuracies: 0.5450
• Rewards/margins: 0.0008
• Rewards/margins Max: 0.0629
• Rewards/margins Min: -0.0642
• Rewards/margins Std: 0.0421
• Logps/rejected: -327.6042
• Logps/chosen: -331.2294
• Logits/rejected: -0.8979
• Logits/chosen: -1.0239

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
0.6681	1.0	1245	0.6931	0.0268	0.0260	0.5450	0.0008	0.0629	-0.0642	0.0421	-327.6042	-331.2294	-0.8979	-1.0239

Framework versions

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