e95d67f6498d1635ca432d43052b374b

This model is a fine-tuned version of albert/albert-large-v1 on the fancyzhx/dbpedia_14 dataset. It achieves the following results on the evaluation set:

• Loss: 0.0538
• Data Size: 1.0
• Epoch Runtime: 1018.6313
• Accuracy: 0.9891
• F1 Macro: 0.9891
• Rouge1: 0.9891
• Rouge2: 0.0
• Rougel: 0.9891
• Rougelsum: 0.9891

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	2.7476	0	36.9465	0.0690	0.0408	0.0690	0.0	0.0690	0.0690
0.1381	1	17500	0.2494	0.0078	45.5837	0.9515	0.9508	0.9515	0.0	0.9516	0.9515
0.082	2	35000	0.1065	0.0156	52.7113	0.9798	0.9798	0.9798	0.0	0.9797	0.9798
0.1112	3	52500	0.0855	0.0312	68.4845	0.9849	0.9849	0.9849	0.0	0.9848	0.9849
0.1005	4	70000	0.0809	0.0625	98.1065	0.9841	0.9841	0.9841	0.0	0.9841	0.9841
0.0741	5	87500	0.0857	0.125	161.0668	0.9856	0.9856	0.9856	0.0	0.9856	0.9857
0.0892	6	105000	0.0901	0.25	284.2256	0.9832	0.9832	0.9832	0.0	0.9832	0.9832
0.0005	7	122500	0.0606	0.5	524.0239	0.9883	0.9883	0.9883	0.0	0.9882	0.9882
0.0708	8.0	140000	0.0572	1.0	1019.5611	0.9886	0.9886	0.9886	0.0	0.9885	0.9886
0.0402	9.0	157500	0.0528	1.0	1014.3520	0.9892	0.9892	0.9892	0.0	0.9892	0.9892
0.0444	10.0	175000	0.0543	1.0	1023.6434	0.9886	0.9886	0.9886	0.0	0.9886	0.9886
0.0547	11.0	192500	0.0640	1.0	1013.8691	0.9884	0.9884	0.9885	0.0	0.9884	0.9884
0.0401	12.0	210000	0.0721	1.0	1018.3700	0.9871	0.9871	0.9871	0.0	0.9871	0.9871
0.0573	13.0	227500	0.0493	1.0	1018.5676	0.9899	0.9899	0.9899	0.0	0.9899	0.9899
0.0471	14.0	245000	0.0523	1.0	1016.3818	0.9895	0.9895	0.9895	0.0	0.9895	0.9895
0.0384	15.0	262500	0.0569	1.0	1034.3212	0.9894	0.9893	0.9894	0.0	0.9894	0.9894
0.0269	16.0	280000	0.0540	1.0	1020.6855	0.9894	0.9894	0.9895	0.0	0.9894	0.9894
0.0298	17.0	297500	0.0538	1.0	1018.6313	0.9891	0.9891	0.9891	0.0	0.9891	0.9891

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.1