gpt2_small_expandedbabyLM_200M_44

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

• Loss: 2.6804

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
3.3025	1.0	10163	2.9778
3.1419	2.0	20326	2.8514
3.0567	3.0	30489	2.7996
3.0007	4.0	40652	2.7745
2.9708	5.0	50815	2.7522
2.9465	6.0	60978	2.7416
2.9254	7.0	71141	2.7290
2.8976	8.0	81304	2.7216
2.8899	9.0	91467	2.7166
2.8739	10.0	101630	2.7111
2.8617	11.0	111793	2.7053
2.8444	12.0	121956	2.7018
2.8351	13.0	132119	2.6974
2.818	14.0	142282	2.6938
2.8096	15.0	152445	2.6910
2.8008	16.0	162608	2.6881
2.7801	17.0	172771	2.6861
2.7721	18.0	182934	2.6825
2.7619	19.0	193097	2.6816
2.7486	20.0	203260	2.6804

Framework versions

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