561f15e65d9e238c65bf9fe0731f5c3d

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

• Loss: 0.4758
• Data Size: 1.0
• Epoch Runtime: 13.8201
• Accuracy: 0.8231
• F1 Macro: 0.8049
• Rouge1: 0.8231
• Rouge2: 0.0
• Rougel: 0.8231
• Rougelsum: 0.8231

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	0.9726	0	1.8402	0.3349	0.2509	0.3343	0.0	0.3355	0.3349
No log	1	114	0.6902	0.0078	2.8097	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	2	228	0.6867	0.0156	2.4793	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	3	342	0.6301	0.0312	3.2917	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
0.0217	4	456	0.7452	0.0625	4.2934	0.3933	0.3506	0.3927	0.0	0.3933	0.3927
0.0217	5	570	0.6180	0.125	5.2000	0.7028	0.5445	0.7028	0.0	0.7022	0.7025
0.0217	6	684	0.5728	0.25	6.3172	0.7123	0.5617	0.7123	0.0	0.7117	0.7129
0.137	7	798	0.4067	0.5	9.1804	0.8208	0.8043	0.8213	0.0	0.8205	0.8205
0.4052	8.0	912	0.4089	1.0	14.8530	0.8367	0.8072	0.8373	0.0	0.8373	0.8367
0.2733	9.0	1026	0.4101	1.0	14.8673	0.8031	0.7913	0.8037	0.0	0.8031	0.8037
0.322	10.0	1140	0.5937	1.0	13.6805	0.7930	0.7285	0.7936	0.0	0.7936	0.7936
0.3307	11.0	1254	0.4758	1.0	13.8201	0.8231	0.8049	0.8231	0.0	0.8231	0.8231

Framework versions

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