zlacker

[return to "Mistral 7B Fine-Tune Optimized"]
1. nickth+Cb[view] [source] 2023-12-20 20:55:18
>>tosh+(OP)
Anytime I see a claim that our 7b models are better than gpt-4 I basically stop reading. If you are going to make that claim, give me several easily digestible examples of this taking place.
◧◩
2. tomrod+pe[view] [source] 2023-12-20 21:10:34
>>nickth+Cb
Looks like they utilized the Bradley-Terry model, but that's not one I'm super familiar with.

https://en.wikipedia.org/wiki/Bradley%E2%80%93Terry_model

◧◩◪
3. huac+qj[view] [source] 2023-12-20 21:41:58
>>tomrod+pe
the BTL model is just a way to infer 'true' skill levels given some list of head to head comparisons. the head to head comparisons / rankings are the most important!!!! and in this case, the rankings come from GPT-4 itself. so take any subsequent score with all the grains of salt you can muster.

their methodology also appears to be 'try 12 different models and hope 1 of them wins out.' multiple hypothesis adjustments come to mind here :)

[go to top]