It is not intended as a general writing model or a broad instruction-tuning improvement. The training data is mostly from a narrow permissioned CYOA-style source, so the model should be expected to learn that domain and style more than broad writing ability.

Other Versions

Merged HF model: KeinNiemand/Kuwutu-7B-CYOA
Adapter GGUF: KeinNiemand/Kuwutu-7B-CYOA-LoRA-GGUF
Mainline-compatible merged GGUF: KeinNiemand/Kuwutu-7B-CYOA-GGUF
ik_llama.cpp merged GGUF: KeinNiemand/Kuwutu-7B-CYOA-IK-GGUF

Intended Use

Fictional adult furry CYOA / interactive-fiction continuation.
Long-form scene continuation from ChatML-style prompts.
Branching narrative drafts and choice-driven writing experiments.

Not Intended For

General assistant use.
Factual, medical, legal, or safety-critical advice.
Content involving minors.
Real-person sexual content without consent.
Harassment, impersonation, or non-consensual sexual scenarios.

Training Notes

Base model: Mawdistical/Kuwutu-7B
Training method: LoRA SFT
Checkpoint: run0067, checkpoint 274
Rank: 32
Alpha: 32
LoRA dropout: 0
Weight decay: 10
Optimizer: Muon with AdamW fallback parameters
Template: ChatML

The checkpoint was selected as a pragmatic loss/guardrail compromise. Generation judge results were noisy in testing and were not used as the sole release signal.

Data

The training data is permissioned adult furry CYOA / interactive-fiction writing. The dataset is not redistributed here.

Loading

Example with transformers and peft:

python

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base = "Mawdistical/Kuwutu-7B"
adapter = "KeinNiemand/Kuwutu-7B-CYOA-LoRA"

tokenizer = AutoTokenizer.from_pretrained(base)
model = AutoModelForCausalLM.from_pretrained(base, torch_dtype="auto", device_map="auto")
model = PeftModel.from_pretrained(model, adapter)

Limitations

This is an experimental narrow-domain adapter. It may produce repetitive, generic, incoherent, overly specific, or poorly structured continuations. It may fail at scene continuity, choice structure, lore consistency, or stopping behavior.