Iker/ClickbaitFighter-7B

TEXT GENERATIONConcurrent Unit Cost:1Model Size:7BQuant:FP8Context Size:4kTool Calling:SupportedPublished:Mar 27, 2024License:cc-by-nc-sa-4.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Iker/ClickbaitFighter-7B is a 7 billion parameter instruction-tuned causal language model developed by Iker, based on openchat/openchat-3.5-0106. Fine-tuned on the NoticIA dataset, it specializes in Spanish clickbait article summarization, generating single-sentence summaries that reveal the truth behind sensational headlines. This model is optimized for debunking clickbait content in Spanish news articles.

Loading preview...

Overview

Iker/ClickbaitFighter-7B is a 7 billion parameter language model specifically fine-tuned for Spanish clickbait article summarization. Developed by Iker, this model is built upon the openchat/openchat-3.5-0106 base model and leverages the NoticIA Dataset for its specialized training.

Key Capabilities

  • Clickbait Debunking: Analyzes sensational Spanish headlines and their corresponding article bodies to extract the core truth.
  • Concise Summarization: Generates single-sentence summaries that directly address the misleading nature of clickbait.
  • Contextual Citation: Can cite original text, especially direct quotes, to support its summaries.
  • Spanish Language Focus: Optimized for processing and generating content exclusively in Spanish.

Performance

This 7B parameter model achieves a ROUGE score of 49.81 on the NoticIA dataset, demonstrating strong performance in its specialized task. It is part of a series of ClickbaitFighter models, offering a balance between size and summarization accuracy.

Good For

  • Developers building applications that require automatic debunking of Spanish clickbait.
  • Researchers interested in natural language processing for misinformation detection in Spanish.
  • Content analysis tools focused on identifying and summarizing the factual content of sensational news.