6909a18b0a467eaca827ba377b376876

This model is a fine-tuned version of distilbert/distilbert-base-multilingual-cased on the fancyzhx/dbpedia_14 dataset. It achieves the following results on the evaluation set:

• Loss: 0.0782
• Data Size: 1.0
• Epoch Runtime: 544.6141
• Accuracy: 0.9894
• F1 Macro: 0.9894
• Rouge1: 0.9894
• Rouge2: 0.0
• Rougel: 0.9894
• Rougelsum: 0.9894

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	2.6445	0	18.2744	0.0527	0.0268	0.0527	0.0	0.0526	0.0527
0.1827	1	17500	0.0949	0.0078	22.8367	0.9807	0.9806	0.9807	0.0	0.9807	0.9807
0.0612	2	35000	0.0715	0.0156	26.5846	0.9844	0.9844	0.9844	0.0	0.9844	0.9844
0.0547	3	52500	0.0723	0.0312	33.9875	0.9843	0.9843	0.9844	0.0	0.9843	0.9843
0.0584	4	70000	0.0591	0.0625	51.6546	0.9864	0.9864	0.9865	0.0	0.9864	0.9864
0.0463	5	87500	0.0700	0.125	84.1001	0.9855	0.9854	0.9855	0.0	0.9855	0.9855
0.0532	6	105000	0.0516	0.25	150.5917	0.9884	0.9884	0.9884	0.0	0.9884	0.9884
0.0004	7	122500	0.0472	0.5	282.0686	0.9893	0.9893	0.9893	0.0	0.9893	0.9893
0.0283	8.0	140000	0.0486	1.0	544.2507	0.9899	0.9899	0.9900	0.0	0.9899	0.9900
0.0289	9.0	157500	0.0605	1.0	541.2594	0.9887	0.9887	0.9887	0.0	0.9887	0.9887
0.0323	10.0	175000	0.0671	1.0	544.2476	0.9880	0.9880	0.9880	0.0	0.9880	0.9880
0.0235	11.0	192500	0.0782	1.0	544.6141	0.9894	0.9894	0.9894	0.0	0.9894	0.9894

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1