bradmin's picture
Model save
fffa09b
metadata
license: apache-2.0
base_model: EleutherAI/polyglot-ko-1.3b
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: reward-gpt-duplicate-answer
    results: []

reward-gpt-duplicate-answer

This model is a fine-tuned version of EleutherAI/polyglot-ko-1.3b on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0115
  • Accuracy: 0.5

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: 9e-06
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 2023
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • num_epochs: 1

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.1975 0.13 100 0.0475 0.0
0.0945 0.25 200 0.0329 0.0
0.0739 0.38 300 0.0156 0.0
0.1354 0.5 400 0.0230 0.5
0.1114 0.63 500 0.0260 0.5
0.0688 0.75 600 0.0119 0.5
0.0744 0.88 700 0.0115 0.5

Framework versions

  • Transformers 4.35.0
  • Pytorch 2.1.0+cu118
  • Datasets 2.14.6
  • Tokenizers 0.14.1