From 21e346abe274c5902bb83d5ce90337668c77981b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= Date: Sun, 14 Dec 2025 18:22:25 -0800 Subject: [PATCH] scripts(evals): improve eval prompts --- scripts/evals/run-release-evals.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/scripts/evals/run-release-evals.py b/scripts/evals/run-release-evals.py index 9e0d88260..fb75a0c03 100644 --- a/scripts/evals/run-release-evals.py +++ b/scripts/evals/run-release-evals.py @@ -30,13 +30,13 @@ EVAL_SIMPLE_MATH = EvalConfig( ) EVAL_WEATHER = EvalConfig( - prompt="What's the weather in San Francisco (in farhenheit or celsius)?", - eval="The user says something specific about the current weather in San Francisco, including the degrees (in farhenheit or celsius).", + prompt="What's the weather in San Francisco? Temperature should be in any unit, just pick one.", + eval="The user talks about the weather in San Francisco, including the degrees.", ) EVAL_ONLINE_SEARCH = EvalConfig( - prompt="What's the date right now in London?", - eval=f"The user says today is {datetime.now(timezone.utc).strftime('%B %d, %Y')} in London.", + prompt="What's the current date in UTC?", + eval=f"Current date in UTC is {datetime.now(timezone.utc).strftime('%A, %B %d, %Y')}.", ) EVAL_SWITCH_LANGUAGE = EvalConfig(