gpt2_small_expandedbabyLM_10M_42

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

• Loss: 3.1528

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: 42
• 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
6.6142	1.0	509	4.9988
4.8396	2.0	1018	4.1976
4.4951	3.0	1527	3.8587
4.1648	4.0	2036	3.6570
4.0134	5.0	2545	3.5168
3.8463	6.0	3054	3.4179
3.7368	7.0	3563	3.3445
3.6235	8.0	4072	3.2865
3.5372	9.0	4581	3.2409
3.45	10.0	5090	3.2096
3.374	11.0	5599	3.1851
3.3042	12.0	6108	3.1715
3.248	13.0	6617	3.1578
3.1914	14.0	7126	3.1559
3.1359	15.0	7635	3.1528
3.079	16.0	8144	3.1539
3.0346	17.0	8653	3.1583
2.9902	18.0	9162	3.1612

Framework versions

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