README.md

metadata

base_model: orpo-explorers/kaist-mistral-orpo-OHP-15k-Stratified-1-beta-0.2-2epoch
tags:
  - alignment-handbook
  - trl
  - orpo
  - generated_from_trainer
  - trl
  - orpo
  - generated_from_trainer
datasets:
  - argilla/Capybara-Preferences
model-index:
  - name: kaist-mistral-orpo-OHP-15k-Stratified-1-beta-0.2-2epoch-capybara-1epoch
    results: []

kaist-mistral-orpo-OHP-15k-Stratified-1-beta-0.2-2epoch-capybara-1epoch

This model is a fine-tuned version of orpo-explorers/kaist-mistral-orpo-OHP-15k-Stratified-1-beta-0.2-2epoch on the argilla/Capybara-Preferences dataset.

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-06
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• total_eval_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 1

Training results

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2.post303
• Datasets 2.18.0
• Tokenizers 0.15.2