Mini evals
1 creator use or recommend it
▲ 1 active in the last 90 days
Starting AI evaluation with a small set of 5–20 hand-rated prompts comparing models before scaling up.
Recently discussed in
How Mini evals compares
1 creator weighed in
Benchmark
Direct rival
large-scale model benchmark
Tulsee says just 5-20 hand-picked prompts can show real signal between models, instead of building a massive large-scale benchmark.
T
1 creator
T
1 creator
Tulsee says just 5-20 hand-picked prompts can show real signal between models, instead of building a massive large-scale benchmark.
"For me a big thing is start with even just five prompts, right? You don't have to start large... even just a mini evaluation where you get to like 20 prompts that can really stress test a set of capabilities where you can see signal betw..."
Instead of building a massive large-scale benchmark
What creators say
"For me a big thing is start with even just five prompts, right? You don't have to start large... even just a mini evaluation where you get to like 20 prompts that can really stress test a set of capabilities where you can see signal between your m..."
More from Tulsee Doshi →Keep exploring
Topics
About this data
- ✓50,000+ mentions extracted from real podcast transcripts
- ✓Every quote links to the source episode
- ✓Sponsored mentions are always labeled
- ✓Critical takes are shown, never filtered out