You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, thanks for the great work! I was trying to reproduce the instruction generation process described in section 3.1 of https://arxiv.org/pdf/2305.15334.pdf and have not seen much success, likely due to differences arising from the particular examples, inference settings, etc used; I'm hoping to eventually try with e.g., Llama-2. I saw some discussion of this and a link to the generated training data in #75 from last year, but no conclusion on the data generating process itself. Have I missed this somewhere? It would be great if that code can be provided to make it possible for others to build on the innovations you've contributed here!
The text was updated successfully, but these errors were encountered:
Hi, thanks for the great work! I was trying to reproduce the instruction generation process described in section 3.1 of https://arxiv.org/pdf/2305.15334.pdf and have not seen much success, likely due to differences arising from the particular examples, inference settings, etc used; I'm hoping to eventually try with e.g., Llama-2. I saw some discussion of this and a link to the generated training data in #75 from last year, but no conclusion on the data generating process itself. Have I missed this somewhere? It would be great if that code can be provided to make it possible for others to build on the innovations you've contributed here!
The text was updated successfully, but these errors were encountered: