31222725fd3503a76db01db02fc76c06

This model is a fine-tuned version of google-bert/bert-large-uncased-whole-word-masking-finetuned-squad on the nyu-mll/glue [mrpc] dataset. It achieves the following results on the evaluation set:

• Loss: 0.4727
• Data Size: 1.0
• Epoch Runtime: 13.8925
• Accuracy: 0.8402
• F1 Macro: 0.8247
• Rouge1: 0.8402
• Rouge2: 0.0
• Rougel: 0.8408
• Rougelsum: 0.8408

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	1.0892	0	1.8851	0.3349	0.2509	0.3343	0.0	0.3355	0.3349
No log	1	114	0.6302	0.0078	2.0382	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	2	228	0.7994	0.0156	2.5971	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	3	342	0.7003	0.0312	3.1076	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
0.0228	4	456	0.5595	0.0625	4.2814	0.7317	0.6094	0.7317	0.0	0.7311	0.7317
0.0228	5	570	0.4817	0.125	5.0462	0.8013	0.7535	0.8013	0.0	0.8013	0.8007
0.0228	6	684	0.3671	0.25	6.1243	0.8349	0.8130	0.8349	0.0	0.8349	0.8349
0.1084	7	798	0.4314	0.5	8.7021	0.8414	0.8268	0.8408	0.0	0.8420	0.8414
0.2716	8.0	912	0.3805	1.0	14.4443	0.8414	0.8148	0.8420	0.0	0.8414	0.8414
0.2237	9.0	1026	0.5231	1.0	14.5223	0.8396	0.8102	0.8396	0.0	0.8390	0.8396
0.245	10.0	1140	0.4727	1.0	13.8925	0.8402	0.8247	0.8402	0.0	0.8408	0.8408

Framework versions

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