Nitral-Archive/Mielikki-Erebus_87k-ShareGPT

---
license: other
language:
- en
---
# I wouldnt use this set as is or the original set until it is updated further.

# Original Dataset: https://huggingface.co/datasets/Mielikki/Erebus-87k
Converted, fixed punctuation, using: https://github.com/The-Chaotic-Neutrals/ShareGPT-Formaxxing

To do:
(add instructions) Deslop, deduplicate, rejection filter, grammar correct.

Dataset contains:
87k human story submissions, ranging across these categories-

[adventure
western
urban-fantasy
thriller
suspense
speculative
science-fiction
sad
romance
mystery
middle-school
inspirational
horror
holiday
historical-fiction
high-school
happy
funny
friendship
fiction
fantasy
drama
crime
creative-nonfiction
contemporary
coming-of-age
christmas
christian
bedtime
american]