davanstrien/fineweb-swe_latn-quality-transformer

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

fineweb-swe_latn-quality-transformer

This model is a fine-tuned version of EuroBERT/EuroBERT-210m on the None dataset. It achieves the following results on the evaluation set:

• —Loss: 0.5507
• —F1: 0.7041
• —Accuracy: 0.7079
• —Confusion Matrix: 53 17 35 73
• —High Precision: 0.6023
• —High Recall: 0.7571
• —High F1: 0.6709
• —Low Precision: 0.8111
• —Low Recall: 0.6759
• —Low F1: 0.7374

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
• —trainbatchsize: 64
• —evalbatchsize: 32
• —seed: 42
• —gradientaccumulationsteps: 2
• —totaltrainbatch_size: 128
• —optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
• —lrschedulertype: linear
• —lrschedulerwarmup_ratio: 0.1
• —num_epochs: 50
• —mixedprecisiontraining: Native AMP

Training results

Framework versions

• —Transformers 4.49.0
• —Pytorch 2.6.0+cu124
• —Datasets 3.3.2
• —Tokenizers 0.21.0