augustocsc's picture
End of fine-tuning training
f7f40e2 verified
|
raw
history blame
2.5 kB
metadata
library_name: peft
license: mit
base_model: gpt2
tags:
  - generated_from_trainer
model-index:
  - name: Se124M10KInfPrompt_endtoken
    results: []

Se124M10KInfPrompt_endtoken

This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7284

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-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 50
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
0.9762 1.0 610 0.8681
0.8832 2.0 1220 0.8117
0.8432 3.0 1830 0.7922
0.8183 4.0 2440 0.7809
0.8131 5.0 3050 0.7687
0.8004 6.0 3660 0.7640
0.7972 7.0 4270 0.7584
0.7941 8.0 4880 0.7528
0.789 9.0 5490 0.7480
0.7806 10.0 6100 0.7454
0.7736 11.0 6710 0.7442
0.769 12.0 7320 0.7444
0.7734 13.0 7930 0.7402
0.7671 14.0 8540 0.7385
0.7605 15.0 9150 0.7365
0.7651 16.0 9760 0.7357
0.7657 17.0 10370 0.7340
0.763 18.0 10980 0.7318
0.7552 19.0 11590 0.7305
0.7563 20.0 12200 0.7284
0.7558 21.0 12810 0.7285
0.7465 22.0 13420 0.7290

Framework versions

  • PEFT 0.15.1
  • Transformers 4.51.3
  • Pytorch 2.6.0+cu118
  • Datasets 3.5.0
  • Tokenizers 0.21.1