gpt-2-10000 / README.md
Baljinnyam's picture
update model card README.md
03875d3
metadata
language:
  - mn
license: mit
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: gpt-2-10000
    results: []

gpt-2-10000

This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2551
  • Precision: 0.1523
  • Recall: 0.2608
  • F1: 0.1923
  • Accuracy: 0.9175

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: 2e-05
  • train_batch_size: 16
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.4502 1.0 477 0.3178 0.1351 0.2289 0.1699 0.8953
0.3283 2.0 954 0.3014 0.1227 0.2220 0.1581 0.8985
0.3016 3.0 1431 0.2768 0.1441 0.2379 0.1795 0.9077
0.2824 4.0 1908 0.2687 0.1442 0.2415 0.1806 0.9103
0.2686 5.0 2385 0.2697 0.1374 0.2383 0.1743 0.9086
0.2568 6.0 2862 0.2573 0.1450 0.2525 0.1842 0.9140
0.2472 7.0 3339 0.2534 0.1492 0.2574 0.1889 0.9166
0.2405 8.0 3816 0.2548 0.1413 0.2515 0.1809 0.9153
0.2345 9.0 4293 0.2545 0.1489 0.2564 0.1884 0.9163
0.2299 10.0 4770 0.2551 0.1523 0.2608 0.1923 0.9175

Framework versions

  • Transformers 4.28.1
  • Pytorch 2.0.0+cu118
  • Datasets 2.12.0
  • Tokenizers 0.13.3