CUSTOM
CLANKERS.
← Browse all models

Kujira / Model details

Underdog-Saluki-27B-1.0-MTP-Abliterated-GGUF

Kujira/Underdog-Saluki-27B-1.0-MTP-Abliterated-GGUF

abliterateduncensoredfine-tuned
Download files on Hugging Face ↗

Read the original model card ↗ · Choosing a model

Family: Unknown. No supported family evidence recorded.

About this model

Kujira's refusal-removed GGUF release of Underdog-Saluki-27B-1.0-MTP, an Apache-2.0 model. The creator says it modifies only four tensors in layers 35–36 to remove refusals while keeping the rest of the weights and metadata byte-identical to the MTP release; no training was done. According to the model card, refusal rates on harmful prompts drop from 99% to 1% (64-token replies) and from 78% to 2% (thinking mode), with English and Japanese perplexity rising by 1.6–1.8%. The card recommends sampling with temperature 1.0, top-p 0.95, top-k 20, and presence penalty 1.1 to avoid looping in long agent sessions.

Description sources

Source-based description, not an independent evaluation. Check the linked sources for the creator’s claims and limitations.

Size and formats

Base model
ConwayResearch/Underdog-Saluki-27B-1.0, Kujira/Underdog-Saluki-27B-1.0-MTP-GGUF
Parameters
Unknown
Main repository formats
GGUF
License tag
apache-2.0
Languages
en, ja
Last updated
2026-10-09 · 19:21:42.000 UTC

What changed

Model types are creator-reported, not independently tested. These labels describe source metadata, not a guarantee of behavior.

  • abliteratedsource tag: abliterated
  • abliteratedmodel name: Underdog-Saluki-27B-1.0-MTP-Abliterated-GGUF
  • uncensoredsource tag: uncensored
  • fine-tunedsource tag: base_model:finetune:ConwayResearch/Underdog-Saluki-27B-1.0

Family evidence

These groupings are source-reported. Explicit base identifiers take priority over generic architecture tags.

  • No supported family evidence recorded.

Available versions

Versions appear here only when they explicitly name this exact model as their source. Check each repository for available files.

Customize or build on this model

Listed from public GitHub data, not tested by us. Read the code and license before you run anything.

Tools to customize it further

Make a smaller download (quantize)

  • llama.cpp ↗ ★ 132,124

    Runs models locally and includes the tools behind most GGUF files. A conversion script turns Hugging Face weights into GGUF, and llama-quantize shrinks a high-precision GGUF to lower-bit formats, which can cost some accuracy.

Learn the techniques

  • LLM Course ↗ ★ 84,633

    A free course with roadmaps and Colab notebooks. Its notebooks include one-click tools for merging with MergeKit, quantizing to GGUF, GPTQ, EXL2, AWQ and HQQ, and automatic abliteration with custom datasets.

Compare all tools →

Source provenance

Downloads
25,405
Likes
28
Metadata checked
2026-10-24 · 04:11:48 UTC
First seen in index
2026-10-24 · 04:11:48 UTC
Source revision
c1d826d73831efcb819d9d64c625f0fab56ae8fc

Source-card excerpt

Source model card excerpt: Underdog Saluki 27B 1.0 + MTP, Abliterated (GGUF) Underdog Saluki 27B 1.0 + MTP with refusals removed . It answers requests that the original

From Kujira/Underdog-Saluki-27B-1.0-MTP-Abliterated-GGUF. Creator model-card excerpt, not an independent review.

Inspect before you run

Low refusal is not proof of correctness. Outputs can be inaccurate, harmful, or biased. Review the model card, license and files yourself. A license tag is not commercial-use advice. Parameter counts are source metadata; no hardware compatibility is implied.

This directory does not run, host, redistribute, or test model weights.

Collection rules & limitations ↗