14ecefef539984683ddcdb054becde27

This model is a fine-tuned version of openai-community/gpt2-large on the google/boolq dataset. It achieves the following results on the evaluation set:

• Loss: 1.0180
• Data Size: 1.0
• Epoch Runtime: 74.4513
• Accuracy: 0.7788
• F1 Macro: 0.7635
• Rouge1: 0.7785
• Rouge2: 0.0
• Rougel: 0.7785
• Rougelsum: 0.7791

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro	Rouge1	Rouge2	Rougel	Rougelsum
No log	0	0	0.7122	0	7.6085	0.5153	0.4793	0.5150	0.0	0.5153	0.5153
No log	1	294	0.7415	0.0078	8.5612	0.6118	0.4324	0.6120	0.0	0.6114	0.6118
No log	2	588	0.7486	0.0156	9.4865	0.6152	0.3989	0.6153	0.0	0.6146	0.6152
No log	3	882	0.6785	0.0312	12.0392	0.6192	0.3913	0.6198	0.0	0.6186	0.6192
0.029	4	1176	0.6462	0.0625	15.2838	0.6406	0.5301	0.6409	0.0	0.6403	0.6403
0.053	5	1470	0.6839	0.125	19.5640	0.6412	0.4541	0.6415	0.0	0.6406	0.6412
0.0868	6	1764	0.6244	0.25	27.4025	0.6489	0.6470	0.6486	0.0	0.6487	0.6483
0.4783	7	2058	0.5555	0.5	43.1201	0.7292	0.7233	0.7289	0.0	0.7286	0.7289
0.3588	8.0	2352	0.5080	1.0	75.6372	0.7564	0.7318	0.7567	0.0	0.7558	0.7567
0.1563	9.0	2646	0.8731	1.0	74.1406	0.7745	0.7448	0.7754	0.0	0.7739	0.7745
0.1169	10.0	2940	0.8719	1.0	74.4658	0.7711	0.7542	0.7714	0.0	0.7705	0.7714
0.0729	11.0	3234	0.9923	1.0	73.2514	0.7770	0.7603	0.7770	0.0	0.7767	0.7767
0.0731	12.0	3528	1.0180	1.0	74.4513	0.7788	0.7635	0.7785	0.0	0.7785	0.7791

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1