gpt2-kl_001_04_hscnspecial-hs_cn

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

  • Loss: 0.5546

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: 4
  • seed: 21
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 3.0

Training results

Training Loss Epoch Step Validation Loss
73.5237 0.02 10 69.5835
46.089 0.04 20 32.9898
13.5444 0.06 30 10.6394
6.7877 0.08 40 4.2155
3.544 0.1 50 2.0273
1.6175 0.12 60 1.0948
1.2328 0.14 70 0.8867
0.8903 0.16 80 0.7270
0.7388 0.18 90 0.6445
0.7102 0.2 100 0.6045
0.6013 0.22 110 0.6055
0.6243 0.24 120 0.5900
0.6262 0.26 130 0.5884
0.6181 0.28 140 0.6059
0.598 0.3 150 0.5984
0.6267 0.32 160 0.5701
0.5183 0.34 170 0.5655
0.6244 0.36 180 0.5707
0.6235 0.38 190 0.5594
0.5825 0.4 200 0.5599
0.6787 0.42 210 0.5573
0.6269 0.44 220 0.5564
0.5267 0.46 230 0.5597
0.5407 0.48 240 0.5567
0.5502 0.5 250 0.5552
0.5457 0.52 260 0.5521
0.5585 0.54 270 0.5543
0.5897 0.56 280 0.5535
0.5736 0.58 290 0.5546

Framework versions

  • Transformers 4.29.0.dev0
  • Pytorch 1.12.0a0+bd13bc6
  • Datasets 2.12.0
  • Tokenizers 0.13.3
Downloads last month
16
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.