e2dcc4f656c7d5f191b39077da5c84eb

This model is a fine-tuned version of Qwen/Qwen2.5-1.5B on the fancyzhx/dbpedia_14 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3547
• Data Size: 1.0
• Epoch Runtime: 3713.0902
• Accuracy: 0.9879
• F1 Macro: 0.9880
• Rouge1: 0.9879
• Rouge2: 0.0
• Rougel: 0.9879
• Rougelsum: 0.9879

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	21.2482	0	132.7540	0.1160	0.0698	0.1161	0.0	0.1160	0.1160
0.7889	1	17500	0.7406	0.0078	162.9353	0.9730	0.9729	0.9730	0.0	0.9730	0.9730
0.612	2	35000	0.4655	0.0156	189.8697	0.9803	0.9803	0.9803	0.0	0.9803	0.9803
0.3241	3	52500	0.4839	0.0312	250.5697	0.9802	0.9803	0.9803	0.0	0.9802	0.9803
0.3493	4	70000	0.3574	0.0625	360.2721	0.9837	0.9838	0.9837	0.0	0.9837	0.9837
0.258	5	87500	0.3628	0.125	582.8575	0.9820	0.9820	0.9820	0.0	0.9820	0.9820
0.2687	6	105000	0.2212	0.25	1034.9854	0.9878	0.9878	0.9878	0.0	0.9878	0.9878
0.0016	7	122500	0.2321	0.5	1931.5358	0.9879	0.9879	0.9879	0.0	0.9879	0.9879
0.1505	8.0	140000	0.2099	1.0	3720.7778	0.9896	0.9896	0.9896	0.0	0.9896	0.9896
0.0799	9.0	157500	0.2330	1.0	3724.0398	0.9883	0.9884	0.9883	0.0	0.9883	0.9883
0.1063	10.0	175000	0.2673	1.0	3732.2440	0.9890	0.9890	0.9890	0.0	0.9890	0.9890
0.067	11.0	192500	0.3009	1.0	3706.3550	0.9890	0.9889	0.9890	0.0	0.9890	0.9890
0.0324	12.0	210000	0.3547	1.0	3713.0902	0.9879	0.9880	0.9879	0.0	0.9879	0.9879

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.2.0
• Tokenizers 0.22.1