gpt2_small_expandedbabyLM_25M_43

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

• Loss: 2.9758

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: 0.0001
• train_batch_size: 256
• eval_batch_size: 256
• seed: 43
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 4000
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
4.6513	1.0	1272	4.0029
4.0309	2.0	2544	3.5163
3.7167	3.0	3816	3.3070
3.5446	4.0	5088	3.1842
3.4304	5.0	6360	3.1125
3.3395	6.0	7632	3.0654
3.2766	7.0	8904	3.0325
3.2188	8.0	10176	3.0151
3.1557	9.0	11448	2.9991
3.1108	10.0	12720	2.9876
3.0777	11.0	13992	2.9816
3.029	12.0	15264	2.9759
2.9922	13.0	16536	2.9759
2.9598	14.0	17808	2.9758
2.9284	15.0	19080	2.9768

Framework versions

• Transformers 4.30.2
• Pytorch 2.11.0+cu130
• Datasets 4.1.1
• Tokenizers 0.13.3