b3e64728c7be9313131f063926f6bbf1

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

• Loss: 0.8986
• Data Size: 1.0
• Epoch Runtime: 32.3554
• Accuracy: 0.7552
• F1 Macro: 0.7407
• Rouge1: 0.7552
• Rouge2: 0.0
• Rougel: 0.7546
• Rougelsum: 0.7552

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	0.8048	0	3.0251	0.4013	0.3553	0.4010	0.0	0.4020	0.4017
No log	1	294	0.6820	0.0078	3.4735	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
No log	2	588	0.6765	0.0156	4.1810	0.6069	0.4419	0.6074	0.0	0.6069	0.6074
No log	3	882	0.6656	0.0312	5.0600	0.6213	0.3847	0.6213	0.0	0.6207	0.6213
0.0283	4	1176	0.6613	0.0625	5.8966	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
0.0559	5	1470	0.6466	0.125	7.3591	0.6369	0.5478	0.6373	0.0	0.6366	0.6369
0.0935	6	1764	0.6442	0.25	11.2733	0.6219	0.3902	0.6219	0.0	0.6216	0.6219
0.5811	7	2058	0.6589	0.5	17.7864	0.6057	0.6051	0.6060	0.0	0.6057	0.6057
0.4663	8.0	2352	0.5032	1.0	32.6136	0.7595	0.7482	0.7592	0.0	0.7592	0.7595
0.3259	9.0	2646	0.6651	1.0	31.6334	0.7408	0.7331	0.7405	0.0	0.7408	0.7405
0.2151	10.0	2940	0.7587	1.0	31.9530	0.7463	0.7290	0.7463	0.0	0.7460	0.7468
0.1773	11.0	3234	0.8642	1.0	32.6725	0.7518	0.7417	0.7518	0.0	0.7515	0.7518
0.1728	12.0	3528	0.8986	1.0	32.3554	0.7552	0.7407	0.7552	0.0	0.7546	0.7552

Framework versions

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