Se124M10KInfPrompt_WT_EOS

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

  • Loss: 1.0978

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: 5e-05
  • train_batch_size: 64
  • eval_batch_size: 64
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 256
  • optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.03
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss
3.2645 0.5970 20 2.8750
3.0103 1.1791 40 2.6534
2.7014 1.7761 60 2.3720
2.3851 2.3582 80 2.1090
2.1288 2.9552 100 1.8650
1.9062 3.5373 120 1.6703
1.739 4.1194 140 1.5153
1.6033 4.7164 160 1.3879
1.5039 5.2985 180 1.2845
1.4219 5.8955 200 1.2122
1.357 6.4776 220 1.1649
1.3201 7.0597 240 1.1304
1.2923 7.6567 260 1.1101
1.2831 8.2388 280 1.1035
1.2715 8.8358 300 1.1007
1.2721 9.4179 320 1.0986

Framework versions

  • PEFT 0.15.1
  • Transformers 4.51.3
  • Pytorch 2.6.0+cu118
  • Datasets 3.5.0
  • Tokenizers 0.21.1
Downloads last month
1
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for augustocsc/Se124M10KInfPrompt_WT_EOS

Adapter
(1671)
this model