gpt2_small_expandedbabyLM_100M_44

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

• Loss: 2.7603

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
3.5677	1.0	5083	3.1801
3.2951	2.0	10166	2.9789
3.1908	3.0	15249	2.8999
3.1104	4.0	20332	2.8608
3.0654	5.0	25415	2.8345
3.0304	6.0	30498	2.8145
2.9922	7.0	35581	2.8019
2.9678	8.0	40664	2.7943
2.9401	9.0	45747	2.7838
2.9156	10.0	50830	2.7792
2.9007	11.0	55913	2.7748
2.8806	12.0	60996	2.7709
2.8658	13.0	66079	2.7676
2.8423	14.0	71162	2.7660
2.8269	15.0	76245	2.7640
2.8221	16.0	81328	2.7625
2.803	17.0	86411	2.7622
2.7843	18.0	91494	2.7610
2.7705	19.0	96577	2.7604
2.7617	20.0	101660	2.7603

Framework versions

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