This model was trained for toxicity labeling. Label_1 means TOXIC, Label_0 means NOT TOXIC
The model was fine-tuned based off
the CamemBERT language model.
The accuracy is 93% on the test split during training and 79% on a manually picked (and thus harder) sample of 200 sentences (100 label 1, 100 label 0) at the end of the training.
The model was finetuned on 32k sentences. The train data was the translations of the English data (around 30k sentences) from
the multilingual_detox dataset by
Skolkovo Institute using
the opus-mt-en-fr translation model by
Helsinki-NLP and the data from
the jigsaw dataset on kaggle.
