Experiments enable continous improvement of your Prompt/Agent — i.e. guarantee net improvements.
Experiment distribution chart
Experiment
on a Prompt
against a Dataset
and a set of Scorers
from Literal AI.
Experiments can be run directly from the Prompt Playground. This allows you to run experiments without having to manage an infrastructure.
Prompt to iterate on
Prompt to iterate on
Pick a Dataset and select Scorers
Experiment on Dataset
input
, expectedOutput
and metadata
columns. The Scorer
configuration offers to
use the prompt’s completion through the output
key.Comparing two experiments ran on the same dataset.