Instructions to use sileod/deberta-v3-base-tasksource-toxicity with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use sileod/deberta-v3-base-tasksource-toxicity with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="sileod/deberta-v3-base-tasksource-toxicity")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("sileod/deberta-v3-base-tasksource-toxicity") model = AutoModelForSequenceClassification.from_pretrained("sileod/deberta-v3-base-tasksource-toxicity") - Notebooks
- Google Colab
- Kaggle
| datasets: | |
| - skg/toxigen-data | |
| - aps/dynahate | |
| - HannahRoseKirk/HatemojiBuild | |
| - mteb-pt/toxic_conversations | |
| - OpenAssistant/oasst2 | |
| - tasksource/implicit-hate-stg1 | |
| language: | |
| - en | |
| tags: | |
| - toxicity | |
| - hate | |
| Multi-task fine-tune of `deberta-base-tasksource` for hate detection | |
| | Test Name | Test Accuracy (%) | Test Pearson (%) | | |
| |------------------------------------|-------------------|------------------| | |
| | dynahate | 82.8 | | | |
| | toxic_conversations | 96.0 | | | |
| | implicit-hate-stg1 | 78.0 | | | |
| | HatemojiBuild | 70.6 | | | |
| | tweet_eval/hate | 55.8 | | | |
| | oasst2_dense_flat/toxicity | | 50.996 | | |
| | civil_comments/toxicity | | 73.046 | | |
| | toxigen-data | | 78.217 | |