gpt2-wikitext2

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

  • Loss: 5.1939

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: 2e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • 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
  • num_epochs: 15

Training results

Training Loss Epoch Step Validation Loss
5.7868 1.0 2249 5.8993
5.5307 2.0 4498 5.7082
5.3288 3.0 6747 5.5705
5.1508 4.0 8996 5.4662
5.003 5.0 11245 5.3797
4.8483 6.0 13494 5.3223
4.7387 7.0 15743 5.2757
4.6358 8.0 17992 5.2465
4.5037 9.0 20241 5.2171
4.4214 10.0 22490 5.2016
4.3426 11.0 24739 5.1984
4.2608 12.0 26988 5.1936
4.1959 13.0 29237 5.1915
4.1562 14.0 31486 5.1922
4.1104 15.0 33735 5.1939

Framework versions

  • Transformers 4.55.4
  • Pytorch 2.7.0+gitf717b2a
  • Datasets 3.6.0
  • Tokenizers 0.21.4
Downloads last month
616
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

Model tree for arman1o1/gpt2-wikitext2

Finetuned
(2050)
this model