gpt2_small_expandedbabyLM_125M_44

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

• Loss: 2.7322

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
3.4664	1.0	6353	3.1038
3.2417	2.0	12706	2.9305
3.1393	3.0	19059	2.8642
3.0842	4.0	25412	2.8270
3.0354	5.0	31765	2.8060
2.998	6.0	38118	2.7894
2.9697	7.0	44471	2.7773
2.9433	8.0	50824	2.7678
2.9223	9.0	57177	2.7611
2.9009	10.0	63530	2.7548
2.8859	11.0	69883	2.7512
2.8621	12.0	76236	2.7464
2.8514	13.0	82589	2.7435
2.8427	14.0	88942	2.7416
2.8226	15.0	95295	2.7390
2.8066	16.0	101648	2.7361
2.7959	17.0	108001	2.7351
2.7775	18.0	114354	2.7329
2.7659	19.0	120707	2.7322
2.7542	20.0	127060	2.7324

Framework versions

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