dpo_from_stratos_judged_annotated_rejected_responses

This model is a fine-tuned version of bespokelabs/Bespoke-Stratos-7B on the mlfoundations-dev/dpo_from_stratos_judged_annotated_rejected_responses dataset. It achieves the following results on the evaluation set:

• Loss: 0.0005
• Rewards/chosen: -7.0780
• Rewards/rejected: -66.0616
• Rewards/accuracies: 1.0
• Rewards/margins: 58.9836
• Logps/chosen: -0.7078
• Logps/rejected: -6.6062
• Logits/chosen: -0.3126
• Logits/rejected: -0.9952

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-07
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 32
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• total_eval_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/chosen	Logps/rejected	Logits/chosen	Logits/rejected
0.0	0.9993	378	0.0005	-7.0780	-66.0616	1.0	58.9836	-0.7078	-6.6062	-0.3126	-0.9952

Framework versions

• Transformers 4.46.1
• Pytorch 2.3.0
• Datasets 3.1.0
• Tokenizers 0.20.3