aeadfb43649802b24a3555edb9ca35c9

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

• Loss: 1.3867
• Data Size: 1.0
• Epoch Runtime: 16.5059
• Accuracy: 0.2487
• F1 Macro: 0.0996

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
No log	0	0	1.4468	0	1.1125	0.2653	0.1678
No log	1	438	1.4916	0.0078	1.4769	0.2527	0.1009
No log	2	876	1.4215	0.0156	1.2842	0.2407	0.1606
No log	3	1314	1.4247	0.0312	1.5613	0.2547	0.1742
No log	4	1752	1.3984	0.0625	2.0938	0.25	0.1072
0.0789	5	2190	1.4017	0.125	3.0534	0.2453	0.1533
0.1853	6	2628	1.4068	0.25	4.9502	0.2480	0.0994
1.3988	7	3066	1.3945	0.5	9.0203	0.2487	0.0996
1.3916	8.0	3504	1.3890	1.0	16.7455	0.2613	0.1741
1.3924	9.0	3942	1.3968	1.0	16.9773	0.2666	0.1764
1.3869	10.0	4380	1.3932	1.0	16.7330	0.2453	0.0985
1.3873	11.0	4818	1.3919	1.0	16.2979	0.2533	0.1011
1.3894	12.0	5256	1.3880	1.0	16.1788	0.2527	0.1008
1.3908	13.0	5694	1.3880	1.0	16.4328	0.2487	0.0996
1.3895	14.0	6132	1.3899	1.0	16.1813	0.2527	0.1008
1.3876	15.0	6570	1.3908	1.0	16.5012	0.2527	0.1008
1.3855	16.0	7008	1.3881	1.0	16.2502	0.2527	0.1008
1.3829	17.0	7446	1.3864	1.0	16.4477	0.2487	0.0996
1.3874	18.0	7884	1.3882	1.0	16.4588	0.2527	0.1008
1.3862	19.0	8322	1.3882	1.0	16.5452	0.2533	0.1011
1.3886	20.0	8760	1.3851	1.0	16.1889	0.2487	0.0996
1.387	21.0	9198	1.3880	1.0	16.1766	0.2527	0.1008
1.3877	22.0	9636	1.3866	1.0	16.2891	0.2493	0.1033
1.3865	23.0	10074	1.3863	1.0	16.0203	0.2533	0.1011
1.3853	24.0	10512	1.3867	1.0	16.5059	0.2487	0.0996

Framework versions

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