This adds some functionality to extract new examples from the logs. Currently only 590/829 logs can safely have examples extracted. I'll be working to get this up to hopefully well above 90%. Coupled with the reviewtools.py script, this should enable us to take Nebari output and inject it directly into the code. I needed the massive number of logs from Llama3 so that I have enough information to realize what common pitfalls to look out for.
This adds some functionality to extract new examples from the logs. Currently only 590/829 logs can safely have examples extracted. I'll be working to get this up to hopefully well above 90%. Coupled with the reviewtools.py script, this should enable us to take Nebari output and inject it directly into the code. I needed the massive number of logs from Llama3 so that I have enough information to realize what common pitfalls to look out for.
This is part of the work needed for #77.