b91f8b15114ca2e9353fd20c831bd1d0

This model is a fine-tuned version of studio-ousia/mluke-large on the contemmcm/cls_mmlu dataset. It achieves the following results on the evaluation set:

• Loss: 1.3881
• Data Size: 1.0
• Epoch Runtime: 80.1931
• Accuracy: 0.2527
• F1 Macro: 0.1008

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	1.3876	0	2.9725	0.2460	0.1416
No log	1	438	1.3949	0.0078	4.2115	0.2593	0.1256
No log	2	876	1.3885	0.0156	6.1506	0.2453	0.0985
No log	3	1314	1.3896	0.0312	8.0801	0.2533	0.1011
No log	4	1752	1.3868	0.0625	11.2835	0.2487	0.0996
0.0782	5	2190	1.3880	0.125	16.6464	0.2527	0.1008
0.1839	6	2628	1.3864	0.25	26.5858	0.2533	0.1011
1.3872	7	3066	1.3881	0.5	45.4902	0.2527	0.1008
1.3877	8.0	3504	1.3859	1.0	83.4762	0.2527	0.1008
1.3859	9.0	3942	1.3860	1.0	80.6743	0.2527	0.1008
1.3878	10.0	4380	1.3882	1.0	79.9941	0.2527	0.1008
1.3878	11.0	4818	1.3877	1.0	79.3363	0.2533	0.1011
1.3885	12.0	5256	1.3881	1.0	80.1931	0.2527	0.1008

Framework versions

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