Developer / Brand: Within Us AI
A Genesis-style, end-to-end advanced fine-tuning starter pack: five datasets (5k each) packaged together to train an assistant’s behavioral core.
Instruction-following (multi-turn) — constraints, formatting, spec compliance
Tool / function calling — schema-accurate tool calls + observations
Preference learning (DPO-style) — chosen vs rejected responses… See the full description on the dataset page:
https://huggingface.co/datasets/11-47/Genesis_FineTune_Core_25k.
