Minbyul's picture
End of training
06cad2c verified
metadata
license: apache-2.0
base_model: Minbyul/biomistral-7b-wo-kqa_golden-sft
tags:
  - alignment-handbook
  - trl
  - dpo
  - generated_from_trainer
  - trl
  - dpo
  - generated_from_trainer
datasets:
  - HuggingFaceH4/ultrafeedback_binarized
model-index:
  - name: biomistral-7b-dpo-full-sft-wo-kqa_golden
    results: []

biomistral-7b-dpo-full-sft-wo-kqa_golden

This model is a fine-tuned version of Minbyul/biomistral-7b-wo-kqa_golden-sft on the HuggingFaceH4/ultrafeedback_binarized dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4647
  • Rewards/chosen: -0.3056
  • Rewards/rejected: -0.8412
  • Rewards/accuracies: 0.875
  • Rewards/margins: 0.5356
  • Logps/rejected: -632.7374
  • Logps/chosen: -249.8875
  • Logits/rejected: -3.9057
  • Logits/chosen: -4.3623

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: 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
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 1

Training results

Training Loss Epoch Step Validation Loss Rewards/chosen Rewards/rejected Rewards/accuracies Rewards/margins Logps/rejected Logps/chosen Logits/rejected Logits/chosen
0.1251 0.82 100 0.4664 -0.3073 -0.8372 0.875 0.5299 -632.3325 -250.0501 -3.9097 -4.3673

Framework versions

  • Transformers 4.39.0.dev0
  • Pytorch 2.1.2
  • Datasets 2.14.6
  • Tokenizers 0.15.2