15528abbcf98e29b2accd446239ac4e4

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

• Loss: 1.0851
• Data Size: 1.0
• Epoch Runtime: 9.5784
• Accuracy: 0.6673
• F1 Macro: 0.6609
• Rouge1: 0.6676
• Rouge2: 0.0
• Rougel: 0.6673
• Rougelsum: 0.6673

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	0.7331	0	1.4474	0.3802	0.2794	0.3802	0.0	0.3808	0.3802
No log	1	294	0.6842	0.0078	2.0045	0.6002	0.5037	0.6008	0.0	0.6002	0.6002
No log	2	588	0.6652	0.0156	2.0405	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
No log	3	882	0.6622	0.0312	2.2135	0.6241	0.4090	0.6244	0.0	0.6235	0.6238
0.0268	4	1176	0.6552	0.0625	2.1852	0.6281	0.4249	0.6281	0.0	0.6278	0.6278
0.0549	5	1470	0.6553	0.125	2.6723	0.6354	0.4661	0.6357	0.0	0.6348	0.6357
0.0918	6	1764	0.6230	0.25	3.6466	0.6489	0.5832	0.6492	0.0	0.6483	0.6492
0.5741	7	2058	0.6024	0.5	5.5427	0.6829	0.6498	0.6832	0.0	0.6820	0.6826
0.4903	8.0	2352	0.5826	1.0	9.4441	0.7169	0.6612	0.7172	0.0	0.7166	0.7166
0.3288	9.0	2646	0.7193	1.0	9.3751	0.7163	0.6922	0.7166	0.0	0.7163	0.7165
0.2106	10.0	2940	0.8553	1.0	9.1315	0.7022	0.6869	0.7025	0.0	0.7022	0.7016
0.1434	11.0	3234	0.9554	1.0	9.4964	0.7050	0.6861	0.7047	0.0	0.7047	0.7050
0.1217	12.0	3528	1.0851	1.0	9.5784	0.6673	0.6609	0.6676	0.0	0.6673	0.6673

Framework versions

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