Modelpublic

MahmoodAnaam/flaird-modernbert-large-attention-multitask-frozen

sourceHugging Faceupdated 10d agoView on Hugging Face

0likes147downloads

Model Card

flaird-modernbert-large-attention-multitask-frozen

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: 0.1359
—Roc-auc: 0.992
—Brier: 0.964
—C@1: 0.951
—F1: 0.974
—F05u: 0.989
—Mean: 0.974

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: 256
—evalbatchsize: 256
—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
0.2320	0.1000	1984	0.2069	0.972	0.922	0.883	0.936	0.972	0.937
0.1925	0.2000	3968	0.1792	0.982	0.919	0.878	0.933	0.971	0.937
0.1755	0.3001	5952	0.1633	0.986	0.955	0.938	0.967	0.985	0.966
0.1664	0.4001	7936	0.1547	0.988	0.929	0.896	0.943	0.976	0.946
0.1600	0.5001	9920	0.1474	0.989	0.96	0.942	0.969	0.987	0.969
0.1539	0.6001	11904	0.1442	0.99	0.942	0.918	0.956	0.981	0.957
0.1495	0.7001	13888	0.1383	0.991	0.959	0.943	0.97	0.987	0.97
0.1472	0.8002	15872	0.1359	0.992	0.964	0.951	0.974	0.989	0.974
0.1463	0.9002	17856	0.1349	0.992	0.964	0.95	0.974	0.989	0.974
0.1453	1.0	19836	0.1346	0.992	0.961	0.946	0.971	0.988	0.972

Framework versions

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