• Model creator: https://huggingface.co/tongliuphysics/
• Original model: https://huggingface.co/tongliuphysics/Llama0-3-8b-v0.1-dpo-lr6e-7-e1/

Original model description:

• trl
• dpo
• generated_from_trainer model-index:
• name: Llama0-3-8b-v0.1-dpo-lr6e-7-e1 results: []

---

Llama0-3-8b-v0.1-dpo-lr6e-7-e1

• Loss: 0.6582
• Rewards/chosen: -0.7111
• Rewards/rejected: -0.8015
• Rewards/accuracies: 0.5927
• Rewards/margins: 0.0904
• Logps/rejected: -166.9069
• Logps/chosen: -159.4064
• Logits/rejected: 0.2042
• Logits/chosen: 0.1905

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

• learning_rate: 6e-07
• train_batch_size: 2
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 8
• total_train_batch_size: 128
• total_eval_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1.0

Training results

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.1+cu121
• Datasets 3.0.0
• Tokenizers 0.20.0