gpt2_small_DE_unigram_65536_parallel10_42

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

• Loss: 3.6835
• Accuracy: 0.3855

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: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• lr_scheduler_warmup_steps: 1000
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
6.1032	1.0	1506	4.4617	0.3299
4.2222	2.0	3012	4.0956	0.3548
4.0562	3.0	4518	3.9419	0.3649
3.8849	4.0	6024	3.8505	0.3710
3.8124	5.0	7530	3.7912	0.3758
3.7292	6.0	9036	3.7507	0.3793
3.6802	7.0	10542	3.7224	0.3826
3.6333	8.0	12048	3.7017	0.3841
3.6029	9.0	13554	3.6886	0.3846
3.5755	10.0	15060	3.6835	0.3855

Framework versions

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