SicariusSicariiStuff/Bluemoon_Top50MB_Sorted_Fixed

Bluemoon Top 50MB - Fully checked for grammar errors and sorted

Dataset Details

This is the top 50MB of the Bluemoon RP dataset, with the original IDs unchanged (As far as I know).

• Grammar fixed
• ShareGPT Format
• Each entry contains the number of tokens in both LLAMA1 and LLAMA3 tokenizers
• Each entry contains the number of characters
• Original IDs remain unchanged (so you can compare them with the original entries in the original Bluemoon dataset)
• System prompt added for general use

I hope this helps as many people as possible, let's make more RP models, and make AI accessible for everyone 🤗