short_first_noditransitive_seed-42_1e-3

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 3.2449
  • Accuracy: 0.3986

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
  • train_batch_size: 32
  • eval_batch_size: 64
  • seed: 42
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 256
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 32000
  • num_epochs: 20.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
6.0521 0.9998 1495 4.4484 0.2905
4.4103 1.9997 2990 3.9680 0.3287
3.8034 2.9995 4485 3.7026 0.3507
3.6275 4.0 5981 3.5398 0.3662
3.3932 4.9998 7476 3.4404 0.3763
3.3197 5.9997 8971 3.3792 0.3821
3.2044 6.9995 10466 3.3386 0.3864
3.17 8.0 11962 3.3130 0.3892
3.1012 8.9998 13457 3.2952 0.3913
3.0817 9.9997 14952 3.2819 0.3932
3.036 10.9995 16447 3.2719 0.3944
3.0237 12.0 17943 3.2660 0.3952
2.9928 12.9998 19438 3.2623 0.3958
2.9845 13.9997 20933 3.2538 0.3966
2.9613 14.9995 22428 3.2557 0.3968
2.958 16.0 23924 3.2525 0.3973
2.9417 16.9998 25419 3.2493 0.3980
2.938 17.9997 26914 3.2484 0.3983
2.9287 18.9995 28409 3.2452 0.3985
2.9233 19.9967 29900 3.2449 0.3986

Framework versions

  • Transformers 4.46.2
  • Pytorch 2.5.1+cu124
  • Datasets 3.2.0
  • Tokenizers 0.20.0
Downloads last month
1
Safetensors
Model size
0.1B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support