61cf5facfaee3b6dbb359ee8d24cf34c

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

• Loss: 1.3897
• Data Size: 1.0
• Epoch Runtime: 41.2803
• 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	1.3896	0	1.7711	0.2527	0.1436
No log	1	438	1.4719	0.0078	2.4559	0.2527	0.1008
No log	2	876	1.3897	0.0156	2.7304	0.2593	0.1558
No log	3	1314	1.4113	0.0312	3.6330	0.2533	0.1011
No log	4	1752	1.3992	0.0625	5.1205	0.2487	0.0996
0.0783	5	2190	1.3865	0.125	7.5378	0.2527	0.1008
0.1841	6	2628	1.3868	0.25	12.2357	0.2487	0.0996
1.3927	7	3066	1.3904	0.5	22.2482	0.2487	0.0996
1.391	8.0	3504	1.3898	1.0	42.0361	0.2527	0.1008
1.3862	9.0	3942	1.3897	1.0	41.2803	0.2527	0.1008

Framework versions

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