08acd42c9c53ac06ff7f33a659310bdd

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

• Loss: 0.9119
• Data Size: 1.0
• Epoch Runtime: 16.5808
• Accuracy: 0.6930
• F1 Macro: 0.6845
• Rouge1: 0.6927
• Rouge2: 0.0
• Rougel: 0.6927
• Rougelsum: 0.6930

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	0.6657	0	1.9636	0.6204	0.3836	0.6204	0.0	0.6201	0.6201
No log	1	294	0.7019	0.0078	3.4390	0.3857	0.2943	0.3854	0.0	0.3860	0.3857
No log	2	588	0.6640	0.0156	2.4479	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
No log	3	882	0.6618	0.0312	2.7828	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.0271	4	1176	0.6616	0.0625	3.3105	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.0547	5	1470	0.6571	0.125	4.3462	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.0939	6	1764	0.6567	0.25	6.2748	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.6276	7	2058	0.6270	0.5	10.1090	0.6581	0.5966	0.6581	0.0	0.6575	0.6578
0.5635	8.0	2352	0.5923	1.0	16.8610	0.6872	0.6281	0.6869	0.0	0.6869	0.6875
0.4754	9.0	2646	0.6054	1.0	16.5489	0.7034	0.6752	0.7037	0.0	0.7031	0.7039
0.3568	10.0	2940	0.7645	1.0	17.9144	0.6967	0.6835	0.6967	0.0	0.6964	0.6973
0.2521	11.0	3234	0.8326	1.0	16.9828	0.7203	0.6841	0.7203	0.0	0.7200	0.7200
0.2001	12.0	3528	0.9119	1.0	16.5808	0.6930	0.6845	0.6927	0.0	0.6927	0.6930

Framework versions

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