From 3280a8e2e244c313a78a6edd497d17d2c18333a9 Mon Sep 17 00:00:00 2001 From: ishaan-jaff Date: Mon, 13 Nov 2023 20:33:25 -0800 Subject: [PATCH] (docs) eval suites --- docs/my-website/docs/tutorials/eval_suites.md | 3 +++ 1 file changed, 3 insertions(+) diff --git a/docs/my-website/docs/tutorials/eval_suites.md b/docs/my-website/docs/tutorials/eval_suites.md index d1be0ae63d3..1107fcc17a5 100644 --- a/docs/my-website/docs/tutorials/eval_suites.md +++ b/docs/my-website/docs/tutorials/eval_suites.md @@ -234,6 +234,9 @@ pip install autoevals ### Quick Start In this code sample we use the `Factuality()` evaluator from `autoevals.llm` to test whether an output is factual, compared to an original (expected) value. + +**Autoevals uses gpt-3.5-turbo / gpt-4-turbo by default to evaluate responses** + See autoevals docs on the [supported evaluators](https://www.braintrustdata.com/docs/autoevals/python#autoevalsllm) - Translation, Summary, Security Evaluators etc ```python