f06174b8b83d892cad19bfd9c4064d83

This model is a fine-tuned version of albert/albert-base-v1 on the nyu-mll/glue dataset. It achieves the following results on the evaluation set:

• Loss: 0.2798
• Data Size: 1.0
• Epoch Runtime: 372.7546
• Accuracy: 0.8976
• F1 Macro: 0.8905
• Rouge1: 0.8976
• Rouge2: 0.0
• Rougel: 0.8976
• Rougelsum: 0.8976

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	0.7234	0	14.0293	0.3860	0.3075	0.3862	0.0	0.3860	0.3862
0.578	1	11370	0.4660	0.0078	18.0289	0.7744	0.7598	0.7744	0.0	0.7744	0.7743
0.4441	2	22740	0.4300	0.0156	19.3506	0.7981	0.7787	0.7980	0.0	0.7980	0.7981
0.4302	3	34110	0.4161	0.0312	24.6922	0.8020	0.7746	0.8020	0.0	0.8020	0.8020
0.372	4	45480	0.3825	0.0625	36.2419	0.8296	0.8149	0.8295	0.0	0.8295	0.8297
0.3646	5	56850	0.3344	0.125	58.1879	0.8482	0.8374	0.8483	0.0	0.8481	0.8483
0.3284	6	68220	0.3172	0.25	103.0408	0.8592	0.8508	0.8591	0.0	0.8592	0.8592
0.2833	7	79590	0.2920	0.5	193.8019	0.8702	0.8635	0.8702	0.0	0.8702	0.8701
0.2889	8.0	90960	0.2779	1.0	372.6897	0.8778	0.8673	0.8778	0.0	0.8778	0.8778
0.242	9.0	102330	0.2661	1.0	369.8296	0.8847	0.8782	0.8846	0.0	0.8847	0.8848
0.2001	10.0	113700	0.2843	1.0	369.6368	0.8890	0.8808	0.8889	0.0	0.8891	0.8891
0.1826	11.0	125070	0.2967	1.0	377.0188	0.8941	0.8872	0.8942	0.0	0.8941	0.8941
0.1554	12.0	136440	0.2850	1.0	372.0233	0.8962	0.8889	0.8962	0.0	0.8963	0.8962
0.1705	13.0	147810	0.2798	1.0	372.7546	0.8976	0.8905	0.8976	0.0	0.8976	0.8976

Framework versions

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