5ebdc024a3401d73d9672d4040f83ff3

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

• Loss: 0.9731
• Data Size: 1.0
• Epoch Runtime: 15.7016
• Accuracy: 0.7096
• F1 Macro: 0.6869
• Rouge1: 0.7102
• Rouge2: 0.0
• Rougel: 0.7093
• Rougelsum: 0.7093

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	0.6854	0	1.9715	0.5604	0.4424	0.5604	0.0	0.5597	0.5600
No log	1	294	0.7485	0.0078	3.4012	0.3784	0.2749	0.3784	0.0	0.3788	0.3787
No log	2	588	0.6676	0.0156	2.3751	0.6036	0.4421	0.6032	0.0	0.6026	0.6037
No log	3	882	0.6635	0.0312	2.6831	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.0274	4	1176	0.6621	0.0625	3.0510	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.0552	5	1470	0.6620	0.125	3.9631	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.0956	6	1764	0.6537	0.25	5.6689	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.6308	7	2058	0.6248	0.5	9.2105	0.6621	0.6199	0.6624	0.0	0.6618	0.6621
0.5699	8.0	2352	0.5936	1.0	16.3802	0.6893	0.5868	0.6900	0.0	0.6890	0.6890
0.4352	9.0	2646	0.6357	1.0	15.6300	0.7062	0.6872	0.7062	0.0	0.7056	0.7059
0.2259	10.0	2940	0.8491	1.0	15.9690	0.7169	0.6620	0.7172	0.0	0.7166	0.7172
0.172	11.0	3234	0.7838	1.0	15.6853	0.7065	0.6920	0.7068	0.0	0.7065	0.7063
0.1268	12.0	3528	0.9731	1.0	15.7016	0.7096	0.6869	0.7102	0.0	0.7093	0.7093

Framework versions

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