ab7d02ec18ba9d2f4a8b5097e45fb93d

This model is a fine-tuned version of distilbert/distilbert-base-uncased on the nyu-mll/glue [cola] dataset. It achieves the following results on the evaluation set:

• Loss: 0.8650
• Data Size: 1.0
• Epoch Runtime: 7.8826
• Accuracy: 0.7930
• F1 Macro: 0.7476
• Rouge1: 0.7930
• Rouge2: 0.0
• Rougel: 0.7939
• Rougelsum: 0.7920

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	0.6982	0	0.9212	0.4121	0.4055	0.4121	0.0	0.4121	0.4121
No log	1	267	0.6196	0.0078	2.0231	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
No log	2	534	0.6923	0.0156	1.1077	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
No log	3	801	0.6760	0.0312	1.2508	0.6143	0.5809	0.6143	0.0	0.6123	0.6133
No log	4	1068	0.5952	0.0625	1.5019	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
0.0348	5	1335	0.6505	0.125	1.9865	0.7148	0.5101	0.7148	0.0	0.7148	0.7148
0.4966	6	1602	0.5129	0.25	2.8964	0.7559	0.6674	0.7559	0.0	0.7559	0.7549
0.4341	7	1869	0.5157	0.5	4.5955	0.7803	0.6898	0.7803	0.0	0.7812	0.7803
0.3532	8.0	2136	0.4734	1.0	8.1542	0.7822	0.7130	0.7822	0.0	0.7832	0.7822
0.1985	9.0	2403	0.6868	1.0	7.9742	0.7871	0.7108	0.7871	0.0	0.7871	0.7876
0.1588	10.0	2670	0.7212	1.0	7.8436	0.7900	0.7280	0.7900	0.0	0.7910	0.7900
0.1064	11.0	2937	0.7946	1.0	7.8648	0.7891	0.7454	0.7891	0.0	0.7891	0.7891
0.1294	12.0	3204	0.8650	1.0	7.8826	0.7930	0.7476	0.7930	0.0	0.7939	0.7920

Framework versions

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