llm-selector / README.md
JeffreyHuang's picture
update model card README.md
adbaff1
|
raw
history blame
3.74 kB
metadata
license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: llm-selector
    results: []

llm-selector

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

  • Loss: 0.4531
  • Accuracy: 0.8794

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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 40

Training results

Training Loss Epoch Step Accuracy Validation Loss
No log 1.0 33 0.2763 1.8921
No log 2.0 66 0.2685 1.8425
No log 3.0 99 0.3307 1.8125
No log 4.0 132 0.2763 1.8077
No log 5.0 165 0.2685 1.8029
No log 6.0 198 0.2763 1.7951
No log 7.0 231 0.2763 1.7898
No log 8.0 264 0.2685 1.7464
No log 9.0 297 0.4630 1.5910
No log 10.0 330 0.4747 1.5129
No log 11.0 363 0.4397 1.4520
No log 12.0 396 0.6109 1.3628
No log 13.0 429 0.5914 1.2874
No log 14.0 462 0.6226 1.1989
No log 15.0 495 0.6809 1.1137
1.6823 16.0 528 0.6732 1.0936
1.6823 17.0 561 0.6887 1.0676
1.6823 18.0 594 0.7198 0.9726
1.6823 19.0 627 0.7626 0.9069
1.6823 20.0 660 0.7626 0.8574
1.6823 21.0 693 0.7860 0.8008
1.6823 22.0 726 0.7899 0.7594
1.6823 23.0 759 0.8093 0.7151
1.6823 24.0 792 0.8016 0.6994
1.6823 25.0 825 0.8016 0.7014
1.6823 26.0 858 0.8171 0.6543
1.6823 27.0 891 0.8093 0.6328
1.6823 28.0 924 0.8132 0.6141
1.6823 29.0 957 0.8288 0.6207
1.6823 30.0 990 0.8288 0.6070
0.8033 31.0 1023 0.5843 0.8288
0.8033 32.0 1056 0.5616 0.8521
0.8033 33.0 1089 0.5372 0.8482
0.8033 34.0 1122 0.5264 0.8482
0.8033 35.0 1155 0.4922 0.8599
0.8033 36.0 1188 0.4755 0.8716
0.8033 37.0 1221 0.4738 0.8755
0.8033 38.0 1254 0.4590 0.8794
0.8033 39.0 1287 0.4544 0.8833
0.8033 40.0 1320 0.4531 0.8794

Framework versions

  • Transformers 4.30.2
  • Pytorch 2.0.1
  • Datasets 2.12.0
  • Tokenizers 0.13.3