994afc0f90d06ff9ca02ed928e70e534

This model is a fine-tuned version of Qwen/Qwen2.5-0.5B on the fancyzhx/dbpedia_14 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3947
• Data Size: 1.0
• Epoch Runtime: 2995.4803
• Accuracy: 0.9877
• F1 Macro: 0.9878
• Rouge1: 0.9878
• Rouge2: 0.0
• Rougel: 0.9877
• Rougelsum: 0.9878

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	33.2134	0	114.3879	0.0868	0.0608	0.0868	0.0	0.0868	0.0868
1.3265	1	17500	1.6095	0.0078	139.0626	0.9552	0.9557	0.9552	0.0	0.9553	0.9552
0.8681	2	35000	0.6767	0.0156	161.2543	0.9718	0.9719	0.9718	0.0	0.9718	0.9718
0.4877	3	52500	0.5717	0.0312	206.2159	0.9737	0.9738	0.9738	0.0	0.9737	0.9738
0.4475	4	70000	0.4402	0.0625	295.2759	0.9789	0.9790	0.9789	0.0	0.9789	0.9789
0.4207	5	87500	0.3562	0.125	477.2790	0.9839	0.9838	0.9839	0.0	0.9839	0.9839
0.3197	6	105000	0.3202	0.25	827.6861	0.9859	0.9859	0.9859	0.0	0.9859	0.9859
0.0018	7	122500	0.2840	0.5	1547.7098	0.9862	0.9862	0.9862	0.0	0.9862	0.9863
0.1548	8.0	140000	0.2411	1.0	2981.0301	0.9892	0.9892	0.9892	0.0	0.9892	0.9892
0.0722	9.0	157500	0.2792	1.0	2929.5555	0.9875	0.9875	0.9875	0.0	0.9875	0.9875
0.1158	10.0	175000	0.3076	1.0	2990.6640	0.9872	0.9872	0.9872	0.0	0.9872	0.9872
0.0746	11.0	192500	0.3949	1.0	3010.3513	0.9876	0.9876	0.9876	0.0	0.9876	0.9876
0.0389	12.0	210000	0.3947	1.0	2995.4803	0.9877	0.9878	0.9878	0.0	0.9877	0.9878

Framework versions

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