CoolFace
Modelpublic

kanishka/opt-babylm2-subset-default-20-epochs-1e-3

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes16downloads
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. -->

opt-babylm2-subset-default-20-epochs-1e-3

This model was trained from scratch on the kanishka/babylm2-subset dataset. It achieves the following results on the evaluation set:

  • —Loss: 2.4350
  • —Accuracy: 0.5324

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: 0.001
  • —trainbatchsize: 32
  • —evalbatchsize: 64
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 32000
  • —num_epochs: 20.0
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossAccuracy
2.53651.0141692.75000.4857
2.37082.0283382.58700.5032
2.25723.0425072.48390.5150
2.19584.0566762.42950.5220
2.12515.0708452.40130.5259
2.07696.0850142.38300.5281
2.0437.0991832.37360.5304
2.0078.01133522.36710.5313
1.98139.01275212.36610.5322
1.959310.01416902.37050.5325
1.93311.01558592.36770.5331
1.910612.01700282.37270.5333
1.884713.01841972.37790.5335
1.863614.01983662.38340.5335
1.839115.02125352.39550.5334
1.817916.02267042.40150.5332
1.791817.02408732.41000.5331
1.767418.02550422.41590.5330
1.75119.02692112.42630.5327
1.733820.02833802.43500.5324

Framework versions

  • —Transformers 4.42.4
  • —Pytorch 2.2.0+cu121
  • —Datasets 2.16.1
  • —Tokenizers 0.19.1