{"html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Fnikkie-ftnext.hatenablog.com%2Fentry%2Frasbt-llm-instruction-eval-ollama\" title=\"rasbt\u3055\u3093\u306b\u306a\u3089\u3063\u3066\u3001MacBook\u4e0a\u3067Llama 3\u3092\u52d5\u304b\u3057\u3066LLM\u306e\u51fa\u529b\u3092\u8a55\u4fa1\u3055\u305b\u3066\u307f\u308b - nikkie-ftnext\u306e\u65e5\u8a18\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>","width":"100%","author_url":"https://blog.hatena.ne.jp/nikkie-ftnext/","url":"https://nikkie-ftnext.hatenablog.com/entry/rasbt-llm-instruction-eval-ollama","published":"2024-06-17 20:20:24","type":"rich","description":"\u6211\u304c\u6a5f\u68b0\u5b66\u7fd2\u306e\u30e8\u30fc\u30c0\u3001rasbt\u6c0f1\u304c\u8208\u5473\u6df1\u3044\u3053\u3068\u3092\u3084\u3063\u3066\u3044\u305f\u306e\u3067\u30d1\u30af\u3063\u3066\u307f\u307e\u3059 Was toying around with LLM model eval that run well on a laptop. Turns out Llama 3 8B Instruct is a pretty good evaluator that runs on a MacBook Air. I got a pretty high 0.8 correlation with GPT-4 scores.Standalone notebook here if you want to give it a try:\u2026\u2026","author_name":"nikkie-ftnext","version":"1.0","height":"190","blog_url":"https://nikkie-ftnext.hatenablog.com/","provider_name":"Hatena Blog","title":"rasbt\u3055\u3093\u306b\u306a\u3089\u3063\u3066\u3001MacBook\u4e0a\u3067Llama 3\u3092\u52d5\u304b\u3057\u3066LLM\u306e\u51fa\u529b\u3092\u8a55\u4fa1\u3055\u305b\u3066\u307f\u308b","blog_title":"nikkie-ftnext\u306e\u65e5\u8a18","categories":["LLM"],"image_url":null,"provider_url":"https://hatena.blog"}