40bc4b2b678320fb07ca3d90076a9f9a

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

• Loss: 0.9881
• Data Size: 1.0
• Epoch Runtime: 9.6056
• Accuracy: 0.6376
• F1 Macro: 0.6260
• Rouge1: 0.6379
• Rouge2: 0.0
• Rougel: 0.6376
• Rougelsum: 0.6376

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	0.7174	0	1.4812	0.3787	0.2747	0.3787	0.0	0.3793	0.3790
No log	1	294	0.6912	0.0078	1.7672	0.5527	0.5078	0.5527	0.0	0.5527	0.5530
No log	2	588	0.6666	0.0156	1.6895	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
No log	3	882	0.6667	0.0312	1.9738	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.027	4	1176	0.6656	0.0625	2.1183	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.055	5	1470	0.6630	0.125	2.6278	0.6210	0.3869	0.6209	0.0	0.6204	0.6207
0.0953	6	1764	0.6688	0.25	3.5586	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.659	7	2058	0.6538	0.5	5.4801	0.6247	0.3943	0.6247	0.0	0.6241	0.6247
0.6355	8.0	2352	0.6533	1.0	9.5061	0.6348	0.4334	0.6348	0.0	0.6345	0.6348
0.6012	9.0	2646	0.6399	1.0	9.4774	0.6523	0.5113	0.6523	0.0	0.6517	0.6526
0.4861	10.0	2940	0.7097	1.0	9.6026	0.6419	0.6269	0.6415	0.0	0.6419	0.6419
0.3907	11.0	3234	0.7344	1.0	9.4883	0.6498	0.6295	0.6500	0.0	0.6492	0.6498
0.3328	12.0	3528	0.7921	1.0	9.5842	0.6578	0.6286	0.6581	0.0	0.6575	0.6575
0.2549	13.0	3822	0.9881	1.0	9.6056	0.6376	0.6260	0.6379	0.0	0.6376	0.6376

Framework versions

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