ee4454b6697be41beaa3ec118595311b

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

• Loss: 1.3912
• Data Size: 1.0
• Epoch Runtime: 46.3261
• Accuracy: 0.2453
• F1 Macro: 0.0985

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	1.4470	0	1.8153	0.2547	0.1453
No log	1	438	1.4135	0.0078	2.5185	0.2527	0.1008
No log	2	876	1.3945	0.0156	3.5531	0.2453	0.0986
No log	3	1314	1.4251	0.0312	4.5433	0.2533	0.1011
No log	4	1752	1.3977	0.0625	6.2391	0.2553	0.1229
0.0794	5	2190	1.3947	0.125	9.4314	0.2453	0.0985
0.1867	6	2628	1.3869	0.25	14.2818	0.2487	0.0996
1.4018	7	3066	1.3918	0.5	25.2172	0.2487	0.0996
1.4079	8.0	3504	1.3897	1.0	47.0272	0.2487	0.0996
1.405	9.0	3942	1.3910	1.0	46.0350	0.2527	0.1008
1.4107	10.0	4380	1.3912	1.0	46.3261	0.2453	0.0985

Framework versions

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