dctanner's picture
End of training
5536547 verified
metadata
license: apache-2.0
library_name: peft
tags:
  - alignment-handbook
  - generated_from_trainer
  - trl
  - dpo
  - generated_from_trainer
datasets:
  - sablo/oasst2_dpo_pairs_en
base_model: sablo/sablo-pebble-mistral
model-index:
  - name: sablo-pebble-mistral-dpo-lora-oasst2_dpo_pairs_en
    results: []

sablo-pebble-mistral-dpo-lora-oasst2_dpo_pairs_en

This model is a fine-tuned version of sablo/sablo-pebble-mistral on the sablo/oasst2_dpo_pairs_en dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6714
  • Rewards/chosen: 0.1838
  • Rewards/rejected: 0.0314
  • Rewards/accuracies: 0.6875
  • Rewards/margins: 0.1524
  • Logps/rejected: -267.5412
  • Logps/chosen: -312.9823
  • Logits/rejected: -2.2363
  • Logits/chosen: -2.2654

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: 4
  • eval_batch_size: 8
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 64
  • 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

  • PEFT 0.7.1
  • Transformers 4.36.2
  • Pytorch 2.1.2+cu118
  • Datasets 2.14.6
  • Tokenizers 0.15.1