gpt2_mini_EN_bpe_1000_42

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

• Loss: 6.0968
• Accuracy: 0.1143

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: 0.0006
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	1	6.9456	0.0012
No log	2.0	2	6.9376	0.0012
No log	3.0	3	6.9218	0.0013
No log	4.0	4	6.8981	0.0015
No log	5.0	5	6.8669	0.0032
No log	6.0	6	6.8285	0.0062
No log	7.0	7	6.7836	0.0181
No log	8.0	8	6.7329	0.0360
No log	9.0	9	6.6777	0.0569
No log	10.0	10	6.6192	0.0797
No log	11.0	11	6.5590	0.0973
No log	12.0	12	6.4984	0.1068
No log	13.0	13	6.4389	0.1105
No log	14.0	14	6.3816	0.1115
No log	15.0	15	6.3270	0.1121
No log	16.0	16	6.2756	0.1118
No log	17.0	17	6.2273	0.1118
No log	18.0	18	6.1818	0.1122
No log	19.0	19	6.1385	0.1132
No log	20.0	20	6.0968	0.1143

Framework versions

• Transformers 4.41.2
• Pytorch 2.8.0+cu128
• Datasets 4.1.1
• Tokenizers 0.19.1