slplab's picture
End of training
ecdf0cd verified
metadata
license: apache-2.0
library_name: peft
tags:
  - generated_from_trainer
base_model: EleutherAI/polyglot-ko-1.3b
model-index:
  - name: pretrain_w-cot_w-asd
    results: []

pretrain_w-cot_w-asd

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: 0.2650

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: 3e-05
  • train_batch_size: 2
  • eval_batch_size: 2
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 8
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
0.6959 0.1725 1000 0.2915
0.2871 0.3450 2000 0.2810
0.2737 0.5174 3000 0.2769
0.2759 0.6899 4000 0.2737
0.2708 0.8624 5000 0.2717
0.2695 1.0349 6000 0.2716
0.2673 1.2074 7000 0.2713
0.2697 1.3798 8000 0.2694
0.2658 1.5523 9000 0.2682
0.2674 1.7248 10000 0.2673
0.2641 1.8973 11000 0.2664
0.2626 2.0698 12000 0.2666
0.2612 2.2422 13000 0.2662
0.266 2.4147 14000 0.2655
0.2626 2.5872 15000 0.2654
0.2637 2.7597 16000 0.2652
0.2623 2.9322 17000 0.2650

Framework versions

  • PEFT 0.11.1
  • Transformers 4.41.1
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1