gpt2_small_expandedbabyLM_1M_44

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

• Loss: 4.2633

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: 44
• 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
No log	1.0	51	9.2619
No log	2.0	102	7.6584
No log	3.0	153	6.9974
8.6178	4.0	204	6.6272
8.6178	5.0	255	6.2833
8.6178	6.0	306	5.9660
8.6178	7.0	357	5.6778
6.5541	8.0	408	5.4100
6.5541	9.0	459	5.1740
6.5541	10.0	510	4.9738
6.5541	11.0	561	4.8335
5.5382	12.0	612	4.7126
5.5382	13.0	663	4.6254
5.5382	14.0	714	4.5485
5.5382	15.0	765	4.4850
5.0528	16.0	816	4.4308
5.0528	17.0	867	4.3802
5.0528	18.0	918	4.3352
5.0528	19.0	969	4.2982
4.7791	20.0	1020	4.2633

Framework versions

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