Open-source model distilbert-base-uncased__hate_speech_offensive__train-16-0 - Quickly detect hateful and offensive content in text

Home

Distilbert Base Uncased Hate Speech Offensive Train 16 0

Developed by SetFit

A lightweight model based on DistilBERT for detecting hate speech and offensive content in text

Text Classification

Transformers

Open Source License:Apache-2.0 #Hate speech detection #Lightweight BERT #Text classification

Downloads 26

Release Time : 3/2/2022

Model Overview

This model is a fine-tuned version of the DistilBERT base model, specifically designed to identify hate speech and offensive language in text. While its accuracy performance is moderate, it provides basic classification capabilities.

Model Features

Lightweight architecture

Based on DistilBERT, smaller and faster than standard BERT models while maintaining relatively good performance

Hate speech detection

Specifically designed for classifying and detecting hate speech and offensive language

Efficient training

Optimized training process using mixed-precision training (native AMP)

Model Capabilities

Text classification

Hate speech detection

Offensive language identification

Use Cases

Content moderation

Social media content filtering

Automatically identify and flag hate speech and offensive content on social media

Accuracy approximately 51.7%

Online community management

Help community administrators quickly identify inappropriate remarks

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0943	1.0	10	1.1095	0.3
1.0602	2.0	20	1.1086	0.4
1.0159	3.0	30	1.1165	0.4
0.9027	4.0	40	1.1377	0.4
0.8364	5.0	50	1.0126	0.5
0.6653	6.0	60	0.9298	0.5
0.535	7.0	70	0.9555	0.5
0.3713	8.0	80	0.8543	0.4
0.1633	9.0	90	0.9876	0.4
0.1069	10.0	100	0.8383	0.6
0.0591	11.0	110	0.8056	0.6
0.0344	12.0	120	0.8915	0.6
0.0265	13.0	130	0.8722	0.6
0.0196	14.0	140	1.0064	0.6
0.0158	15.0	150	1.0479	0.6
0.0128	16.0	160	1.0723	0.6
0.0121	17.0	170	1.0758	0.6
0.0093	18.0	180	1.1236	0.6
0.0085	19.0	190	1.1480	0.6
0.0084	20.0	200	1.1651	0.6
0.0077	21.0	210	1.1832	0.6

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご

Distilbert Base Uncased Hate Speech Offensive Train 16 0

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 distilbert-base-uncased__hate_speech_offensive__train-16-0

🚀 Quick Start

📚 Documentation

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

📄 License