Can a fine-tuned version of Mistral 7B hold its own against heavyweights like LLama2-Chat-70B and ChatGPT? Well, it turns out that Zephyr 7B Beta isn't just holding its own; it's outperforming 7B models on evaluation benchmarks. But how does it fare in the real world? In this video, we're about to find out. We'll take a deep dive into the technical report for this model and load it into a Google Colab notebook, so we can see its potential in action.
Technical report: arxiv.org/abs/2310.16944
Model on HF: huggingface.co/HuggingFaceH4/...
Discord: / discord
Prepare for the Machine Learning interview: mlexpert.io
Subscribe: bit.ly/venelin-subscribe
GitHub repository: github.com/curiousily/Get-Thi...
00:00 - Intro
00:26 - Technical Report
08:05 - Zephyr 7B Beta on HuggingFace
10:05 - Google Colab Setup
11:15 - Prompt Format
12:15 - Writing Prompts
14:52 - Math
16:24 - Coding
16:31 - Summarization
18:10 - Data Extraction
19:48 - Conclusion
Image thanks to freepik
Join this channel to get access to the perks and support my work:
/ @venelin_valkov
#llm #chatbot #artificialintelligence #llama2 #chatgpt #gpt4 #promptengineering
Негізгі бет Zephyr 7B Beta - Can 7B LLM Compete with 70B models? | Paper review & Google Colab Tutorial
Пікірлер: 7