Modelpublic

EndLessTime/fine_tuned_xsum

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes6downloads

Model Card

finetunedxsum

This model is a fine-tuned version of Qwen/Qwen2-1.5B on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.1573
—Accuracy: 0.9597

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: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7991	0.0289	100	0.3764	0.8394
0.6153	0.0578	200	0.3492	0.8602
0.3929	0.0867	300	0.5004	0.8501
0.7981	0.1156	400	0.3459	0.8677
0.5853	0.1445	500	0.3124	0.8787
0.3284	0.1734	600	0.2438	0.9308
0.3591	0.2023	700	0.2842	0.9041
0.332	0.2311	800	0.3904	0.9038
0.3424	0.2600	900	0.2234	0.9402
0.2609	0.2889	1000	0.2586	0.9249
0.3036	0.3178	1100	0.2775	0.9204
0.2429	0.3467	1200	0.1521	0.9441
0.2495	0.3756	1300	0.2326	0.9512
0.2486	0.4045	1400	0.2712	0.9467
0.1711	0.4334	1500	0.1573	0.9597

Framework versions

—Transformers 4.49.0
—Pytorch 2.6.0+cu126
—Datasets 3.3.2
—Tokenizers 0.21.0