ft-openelm-270m-ultrafeedback

This model is a fine-tuned version of apple/OpenELM-270M on the HuggingFaceH4/ultrafeedback_binarized dataset. It achieves the following results on the evaluation set:

• Loss: 1.6455
• Rewards/chosen: -0.1993
• Rewards/rejected: -0.2029
• Rewards/accuracies: 0.5050
• Rewards/margins: 0.0035
• Logps/rejected: -2.0273
• Logps/chosen: -1.9941
• Logits/rejected: -5.7383
• Logits/chosen: -6.1094
• Nll Loss: 1.5742
• Log Odds Ratio: -0.7037
• Log Odds Chosen: 0.0445

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: 8e-06
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• 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	Nll Loss	Log Odds Ratio	Log Odds Chosen
1.7594	0.53	100	1.6455	-0.1993	-0.2029	0.5050	0.0035	-2.0273	-1.9941	-5.7383	-6.1094	1.5742	-0.7037	0.0445

Framework versions

• Transformers 4.39.3
• Pytorch 2.4.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2