db0f7eebf0a21e81ff178c737e6dbda7

This model is a fine-tuned version of studio-ousia/mluke-base on the google/boolq dataset. It achieves the following results on the evaluation set:

• Loss: 0.7996
• Data Size: 1.0
• Epoch Runtime: 31.2074
• Accuracy: 0.7313
• F1 Macro: 0.7111
• Rouge1: 0.7313
• Rouge2: 0.0
• Rougel: 0.7313
• Rougelsum: 0.7313

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	0.6929	0	3.3614	0.5083	0.4887	0.5084	0.0	0.5086	0.5083
No log	1	294	0.6790	0.0078	3.9873	0.5846	0.4976	0.5846	0.0	0.5843	0.5843
No log	2	588	0.6663	0.0156	4.5612	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
No log	3	882	0.6775	0.0312	5.4601	0.6063	0.4415	0.6065	0.0	0.6060	0.6060
0.0271	4	1176	0.6629	0.0625	6.4235	0.6207	0.3875	0.6210	0.0	0.6204	0.6207
0.0558	5	1470	0.6704	0.125	8.1659	0.6213	0.4357	0.6212	0.0	0.6204	0.6212
0.0949	6	1764	0.6490	0.25	11.5507	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.6282	7	2058	0.6161	0.5	18.0160	0.6795	0.6478	0.6795	0.0	0.6789	0.6795
0.5453	8.0	2352	0.5835	1.0	31.9358	0.6958	0.6714	0.6961	0.0	0.6958	0.6961
0.4631	9.0	2646	0.6566	1.0	31.4163	0.7203	0.6778	0.7206	0.0	0.7203	0.7206
0.2894	10.0	2940	0.8073	1.0	31.5363	0.7188	0.7017	0.7191	0.0	0.7188	0.7188
0.2143	11.0	3234	0.8221	1.0	31.9132	0.7108	0.7015	0.7106	0.0	0.7111	0.7108
0.1711	12.0	3528	0.7996	1.0	31.2074	0.7313	0.7111	0.7313	0.0	0.7313	0.7313

Framework versions

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