9a7204d89a25d13ca2666ebd39ccaa26

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

• Loss: 0.6982
• Data Size: 0.5
• Epoch Runtime: 155.2430
• Accuracy: 0.4943
• F1 Macro: 0.3308
• Rouge1: 0.4947
• Rouge2: 0.0
• Rougel: 0.4943
• Rougelsum: 0.4944

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	0.6991	0	4.6711	0.5011	0.3471	0.5010	0.0	0.5011	0.5009
No log	1	3273	0.6225	0.0078	7.2597	0.6678	0.6362	0.6676	0.0	0.6675	0.6676
0.01	2	6546	0.5602	0.0156	10.7662	0.7197	0.7008	0.7196	0.0	0.7193	0.7191
0.5558	3	9819	0.4554	0.0312	14.6729	0.8176	0.8175	0.8175	0.0	0.8176	0.8176
0.7018	4	13092	0.6981	0.0625	24.2939	0.4943	0.3308	0.4947	0.0	0.4943	0.4944
0.7081	5	16365	0.6929	0.125	43.2903	0.5057	0.3359	0.5053	0.0	0.5057	0.5056
0.7022	6	19638	0.6912	0.25	80.3264	0.5057	0.3359	0.5053	0.0	0.5057	0.5056
0.7029	7	22911	0.6982	0.5	155.2430	0.4943	0.3308	0.4947	0.0	0.4943	0.4944

Framework versions

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