Modelpublic

MahmoodAnaam/flaird-modernbert-large-attention-single-task

sourceHugging Faceupdated 5d agoView on Hugging Face

0likes112downloads

Model Card

flaird-modernbert-large-attention-single-task

This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.0018
—Roc-auc: 0.999
—Brier: 0.991
—C@1: 0.987
—F1: 0.993
—F05u: 0.997
—Mean: 0.994

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: 2e-05
—trainbatchsize: 128
—evalbatchsize: 128
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
—lrschedulertype: cosine
—lrschedulerwarmup_steps: 0.06
—num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Roc-auc	Brier	C@1	F1	F05u	Mean
0.0070	0.1000	3968	0.0067	0.993	0.967	0.956	0.977	0.99	0.977
0.0053	0.2000	7936	0.0049	0.996	0.964	0.949	0.973	0.989	0.974
0.0039	0.3001	11904	0.0042	0.998	0.985	0.979	0.989	0.995	0.989
0.0033	0.4001	15872	0.0041	0.998	0.966	0.954	0.976	0.99	0.977
0.0028	0.5001	19840	0.0029	0.999	0.985	0.979	0.989	0.995	0.99
0.0022	0.6001	23808	0.0026	0.999	0.983	0.977	0.988	0.995	0.988
0.0021	0.7001	27776	0.0021	0.999	0.99	0.986	0.993	0.997	0.993
0.0018	0.8002	31744	0.0019	0.999	0.989	0.985	0.992	0.997	0.992
0.0015	0.9002	35712	0.0018	0.999	0.991	0.987	0.993	0.997	0.994
0.0016	1.0	39672	0.0018	0.999	0.991	0.987	0.993	0.997	0.994

Framework versions

—Transformers 5.16.1
—Pytorch 2.11.0+cu128
—Datasets 4.0.0
—Tokenizers 0.23.1