CoolFace
Modelpublic

kanishka/opt-babylm2-subset-default-20-epochs-3e-4

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes18downloads
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-3e-4

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

  • —Loss: 2.4374
  • —Accuracy: 0.5310

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: 0.0003
  • —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
2.65751.0141692.85710.4750
2.41792.0283382.62580.4990
2.29033.0425072.51300.5119
2.22224.0566762.45250.5194
2.14775.0708452.42070.5237
2.09676.0850142.39900.5266
2.06087.0991832.38850.5284
2.02318.01133522.38150.5296
1.99669.01275212.38030.5304
1.97410.01416902.38020.5311
1.946911.01558592.37980.5315
1.924112.01700282.38570.5318
1.897513.01841972.39080.5319
1.876314.01983662.39430.5320
1.851915.02125352.40290.5317
1.830616.02267042.40980.5317
1.804517.02408732.41740.5315
1.780118.02550422.42370.5314
1.764919.02692112.43080.5311
1.748120.02833802.43740.5310

Framework versions

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