article2keyword2.1b_paraphrase-multilingual-MiniLM-L12-v2_finetuned_for_mlm

This model is a fine-tuned version of sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0673

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: 2e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 16
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
2.3777	1.0	1353	0.3168
0.2358	2.0	2706	0.1564
0.1372	3.0	4059	0.1149
0.1046	4.0	5412	0.0956
0.086	5.0	6765	0.0853
0.0741	6.0	8118	0.0786
0.0653	7.0	9471	0.0750
0.0594	8.0	10824	0.0726
0.0542	9.0	12177	0.0699
0.0504	10.0	13530	0.0692
0.047	11.0	14883	0.0684
0.0444	12.0	16236	0.0675
0.0423	13.0	17589	0.0674
0.0404	14.0	18942	0.0673
0.0392	15.0	20295	0.0672
0.0379	16.0	21648	0.0673

Framework versions

• Transformers 4.21.1
• Pytorch 1.11.0
• Datasets 2.3.2
• Tokenizers 0.12.1