• Model creator: https://huggingface.co/vain05/
• Original model: https://huggingface.co/vain05/stablelm-2-1_6b-orpo-full-v3/

Original model description:

• alignment-handbook
• trl
• orpo
• generated_from_trainer
• trl
• orpo
• generated_from_trainer datasets:
• argilla/ultrafeedback-binarized-preferences-cleaned
• argilla/distilabel-capybara-dpo-7k-binarized model-index:
• name: stablelm-2-1_6b-orpo-full-v3 results: []

stablelm-2-1_6b-orpo-full-v3

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

• learning_rate: 5e-07
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 3
• gradient_accumulation_steps: 2
• total_train_batch_size: 48
• total_eval_batch_size: 24
• 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

Framework versions

• Transformers 4.39.0.dev0
• Pytorch 2.2.1
• Datasets 2.18.0
• Tokenizers 0.15.2