CoolFace
Modelpublic

nikaashpuri/gpt-expt-sp-v3-K-600-MA-Mac-actions-kmeans-v2

sourceHugging Facemitupdated 3y agoView on Hugging Face
0likes28downloads
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. -->

gpt-expt-sp-v3-K-600-MA-Mac-actions-kmeans-v2

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

  • —Loss: 0.0160

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
  • —gradientaccumulationsteps: 8
  • —totaltrainbatch_size: 512
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: cosine
  • —lrschedulerwarmup_steps: 1000
  • —num_epochs: 500
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation Loss
0.175825.3850000.0788
0.05350.76100000.0612
0.035776.14150000.0502
0.0495101.52200000.0450
0.0276126.9250000.0432
0.0249152.28300000.0416
0.0236177.66350000.0257
0.0202203.05400000.0196
0.0185228.43450000.0174
0.0178253.8500000.0168
0.0173279.19550000.0165
0.0169304.57600000.0163
0.0166329.95650000.0162
0.0165355.33700000.0161
0.0163380.71750000.0161
0.0162406.09800000.0160
0.0161431.47850000.0160
0.0161456.85900000.0159
0.016482.23950000.0160

Framework versions

  • —Transformers 4.26.1
  • —Pytorch 1.13.1.post200
  • —Datasets 2.9.0
  • —Tokenizers 0.13.2