gpt2_bwgenerator

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

• Loss: 0.0925

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use 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: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.3023	0.1574	300	0.1644
0.167	0.3148	600	0.1376
0.1466	0.4722	900	0.1301
0.1354	0.6296	1200	0.1180
0.1258	0.7870	1500	0.1129
0.1198	0.9444	1800	0.1061
0.115	1.1018	2100	0.1030
0.1108	1.2592	2400	0.1013
0.1088	1.4166	2700	0.1000
0.1067	1.5740	3000	0.0982
0.1049	1.7314	3300	0.0974
0.1039	1.8888	3600	0.0960
0.1024	2.0462	3900	0.0952
0.1013	2.2036	4200	0.0947
0.1006	2.3610	4500	0.0944
0.0997	2.5184	4800	0.0935
0.0993	2.6758	5100	0.0927
0.099	2.8332	5400	0.0928
0.0983	2.9906	5700	0.0925

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1
• Datasets 3.2.0
• Tokenizers 0.21.0