m-minilm-l12-h384-dra-tam-mal-ai-gen-review-classification-finetune-2b

This model is a fine-tuned version of microsoft/Multilingual-MiniLM-L12-H384 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3138
• Accuracy: 0.91
• F1: 0.9096

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: 0.0001
• train_batch_size: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.6887	0.3636	4	0.6661	0.9410	0.9409
0.6406	0.7273	8	0.5617	0.9410	0.9409
0.5088	1.0909	12	0.3921	0.9689	0.9689
0.3735	1.4545	16	0.2680	0.9689	0.9689
0.2684	1.8182	20	0.1984	0.9814	0.9814
0.2635	2.1818	24	0.1675	0.9814	0.9814
0.1909	2.5455	28	0.1471	0.9814	0.9814
0.2291	2.9091	32	0.1356	0.9814	0.9814
0.1167	3.2727	36	0.1371	0.9752	0.9751
0.151	3.6364	40	0.1414	0.9720	0.9720
0.1579	4.0	44	0.1152	0.9814	0.9814
0.1211	4.3636	48	0.1273	0.9752	0.9751
0.1427	4.7273	52	0.1273	0.9752	0.9751
0.1002	5.0909	56	0.1259	0.9752	0.9751
0.1076	5.4545	60	0.1251	0.9752	0.9751
0.099	5.8182	64	0.1302	0.9720	0.9720

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0