5461c2243eb6e55682e171497aa8b5ed

This model is a fine-tuned version of albert/albert-xxlarge-v2 on the fancyzhx/dbpedia_14 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1041
• Data Size: 1.0
• Epoch Runtime: 2412.2824
• Accuracy: 0.9812
• F1 Macro: 0.9812
• Rouge1: 0.9812
• Rouge2: 0.0
• Rougel: 0.9812
• Rougelsum: 0.9812

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	3.3797	0	87.5208	0.0623	0.0189	0.0623	0.0	0.0623	0.0623
0.7363	1	17500	0.1242	0.0078	107.2496	0.9799	0.9799	0.9799	0.0	0.9799	0.9799
0.1328	2	35000	0.0946	0.0156	123.8618	0.9826	0.9826	0.9826	0.0	0.9826	0.9826
0.075	3	52500	0.0864	0.0312	160.0166	0.9856	0.9856	0.9856	0.0	0.9856	0.9856
0.0663	4	70000	0.0707	0.0625	232.7411	0.9840	0.9840	0.9840	0.0	0.9840	0.9840
0.0553	5	87500	0.0525	0.125	377.9700	0.9886	0.9886	0.9886	0.0	0.9886	0.9886
0.0687	6	105000	0.0592	0.25	669.0405	0.9873	0.9873	0.9873	0.0	0.9873	0.9873
0.001	7	122500	0.1112	0.5	1250.6641	0.9740	0.9741	0.9740	0.0	0.9740	0.9740
0.1037	8.0	140000	0.0808	1.0	2413.6277	0.9850	0.9851	0.9851	0.0	0.9850	0.9850
0.0709	9.0	157500	0.1041	1.0	2412.2824	0.9812	0.9812	0.9812	0.0	0.9812	0.9812

Framework versions

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