Ask anything
RM Compare has been optimised for use with your LLM.
See our Training Centre for more information.
A key consideration for any Creator is how much work and effort each session will create for Judges. There are plenty of things that could influence this for example
As your get more experienced and perhaps investigate further (there has been plenty of research on this matter) you will get better at estimating judge workload.
A key variable when thinking about workload is 'time to judgement'. Helpfully your RM Compare reports provide valuable insights.
The Median time can be more reliable as this counters 'off-task' Judge behaviour where for example the Judge has left the task for a time but has not logged out of the system.
As you work through the session set up flow you will have the opportunity to use our Judgement Calculator. This allows you to define approximately how many Judgements you would like each Judge to make. Important here is the concept of Rounds.
We regularly blog about workload
Five quick questions to test your understanding