opt350m-qlora / README.md
zhangduo4610's picture
End of training
53f7e4d verified
metadata
base_model: facebook/opt-350m
datasets:
  - HuggingFaceH4/ultrachat_200k
library_name: peft
license: other
tags:
  - alignment-handbook
  - trl
  - sft
  - generated_from_trainer
model-index:
  - name: opt350m-qlora
    results: []

opt350m-qlora

This model is a fine-tuned version of facebook/opt-350m on the HuggingFaceH4/ultrachat_200k dataset. It achieves the following results on the evaluation set:

  • Loss: 1.7868

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: 0.0002
  • train_batch_size: 4
  • eval_batch_size: 8
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 2
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 16
  • total_eval_batch_size: 16
  • 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
1.8281 0.9999 8068 1.7868

Framework versions

  • PEFT 0.13.2
  • Transformers 4.45.0
  • Pytorch 2.1.2
  • Datasets 3.1.0
  • Tokenizers 0.20.3