README.md

metadata

license: mit
base_model: ShenaoZhang/0.001_idpo_noreplacerej_iter_2
tags:
  - alignment-handbook
  - generated_from_trainer
  - trl
  - dpo
  - generated_from_trainer
datasets:
  - updated
  - original
model-index:
  - name: 0.001_optm_idpo_noreplacerej_0alpha_iter_2
    results: []

0.001_optm_idpo_noreplacerej_0alpha_iter_2

This model is a fine-tuned version of ShenaoZhang/0.001_idpo_noreplacerej_iter_2 on the updated and the original datasets.

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-07
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 2
• 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: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 1

Training results

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu121
• Datasets 2.14.6
• Tokenizers 0.15.2