CoolFace
Modelpublic

NasimB/aochildes-cbt-log-rarity

sourceHugging Facemitupdated 3y agoView on Hugging Face
0likes19downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

aochildes-cbt-log-rarity

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

  • —Loss: 4.1483

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: 0.0005
  • —trainbatchsize: 64
  • —evalbatchsize: 64
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: cosine
  • —lrschedulerwarmup_steps: 1000
  • —num_epochs: 6
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation Loss
6.36490.295005.3433
5.05060.5910004.9337
4.70790.8815004.6957
4.45121.1720004.5593
4.30311.4725004.4458
4.20851.7630004.3418
4.08092.0535004.2739
3.90472.3540004.2277
3.88462.6445004.1774
3.83922.9350004.1313
3.63923.2355004.1305
3.60163.5260004.1020
3.58283.8165004.0709
3.47334.1170004.0797
3.32714.475004.0758
3.32284.6980004.0635
3.31474.9985004.0528
3.1545.2890004.0692
3.14615.5895004.0692
3.14165.87100004.0684

Framework versions

  • —Transformers 4.26.1
  • —Pytorch 1.11.0+cu113
  • —Datasets 2.13.0
  • —Tokenizers 0.13.3