Modelpublic

AmirMohseni/router-qwen3-0.6b-text-only-v4

sourceHugging Faceapache-2.0updated 11mo agoView on Hugging Face

0likes5downloads

Model Card

router-qwen3-0.6b-text-only-v4

This model is a fine-tuned version of Qwen/Qwen3-0.6B on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.5576
—Accuracy: 0.7273
—Precision: 0.7241
—Recall: 0.7273
—F1: 0.7067

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: 1e-05
—trainbatchsize: 16
—evalbatchsize: 32
—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
—num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
No log	0	0	1.1855	0.5909	0.5603	0.5909	0.5667
0.8449	0.1136	10	0.7082	0.7102	0.7023	0.7102	0.6896
0.8398	0.2273	20	0.5703	0.7216	0.7262	0.7216	0.6911
0.6942	0.3409	30	0.6016	0.6989	0.6881	0.6989	0.6864
0.7332	0.4545	40	0.6146	0.6307	0.6924	0.6307	0.6361
0.636	0.5682	50	0.5402	0.7557	0.8095	0.7557	0.7170
0.5157	0.6818	60	0.5080	0.7159	0.7088	0.7159	0.7099
0.6613	0.7955	70	0.5060	0.7330	0.7274	0.7330	0.7182
0.6174	0.9091	80	0.5203	0.7443	0.7538	0.7443	0.7190
0.5021	1.0227	90	0.5441	0.7216	0.7134	0.7216	0.7101
0.4124	1.1364	100	0.5618	0.7330	0.7286	0.7330	0.7161
0.3623	1.25	110	0.5811	0.7159	0.7080	0.7159	0.7084
0.3416	1.3636	120	0.5897	0.6932	0.6912	0.6932	0.6921
0.2837	1.4773	130	0.5869	0.7443	0.7388	0.7443	0.7338
0.3153	1.5909	140	0.5608	0.7330	0.7323	0.7330	0.7116
0.272	1.7045	150	0.5676	0.7386	0.7409	0.7386	0.7166
0.3459	1.8182	160	0.5586	0.7273	0.7241	0.7273	0.7067
0.368	1.9318	170	0.5576	0.7273	0.7241	0.7273	0.7067

Framework versions

—Transformers 4.57.1
—Pytorch 2.8.0+cu128
—Datasets 4.2.0
—Tokenizers 0.22.1