CUSTOM
CLANKERS.
← Browse all models

Goekdeniz-Guelmez / Model details

JOSIE-2-4B-OSS

Goekdeniz-Guelmez/JOSIE-2-4B-OSS

fine-tuned
Download files on Hugging Face ↗

Read the original model card ↗ · Choosing a model

About this model

Goekdeniz-Guelmez's JOSIE-2-4B-OSS is a supervised fine-tuned derivative of Qwen/Qwen3.5-4B, provided as safetensors weights under the MIT license. The creator describes training on consumer Apple Silicon using a reasoning-first dataset of roughly 4 million tokens, composed entirely of reasoning-trace demonstrations with no separate instruction-only examples. According to the model card, the goal is to reshape how the base model approaches problems by decomposing tasks, verifying intermediate results, recognizing uncertainty, and producing internally consistent answers. The creator reports benchmark improvements over the base model on ARC-Challenge and TruthfulQA in both reasoning and non-reasoning modes, and notes emergent sarcastic or informal internal monologue that was not explicitly trained. These claims have not been independently tested here.

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
Qwen/Qwen3.5-4B
Parameters
4.65987B
Main repository formats
safetensors
License tag
mit
Languages
ar, en, ja, pt, zh
Last updated
2026-08-02 · 20:49:52.000 UTC

What changed

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

  • fine-tunedsource tag: base_model:finetune:Qwen/Qwen3.5-4B

Family evidence

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

  • Qwen: Base model: Qwen/Qwen3.5-4B

Available versions

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

Link to this listing

Made or quantized this model? Paste one of these into your model card or README so people can find this listing.

Markdown

[Listed on Custom Clankers](https://customclankers.com/models/goekdeniz-guelmez-josie-2-4b-oss/)

HTML

<a href="https://customclankers.com/models/goekdeniz-guelmez-josie-2-4b-oss/">Listed on Custom Clankers</a>

Plain text links only: no image, no script, nothing loaded from this site.

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

Remove more refusals

  • Heretic ↗ ★ 36,421

    Removes refusal behavior from transformer models automatically. It pairs directional ablation (abliteration) with an Optuna parameter search that tries to cut refusals while keeping the model close to the original, so you don't have to pick layers by hand.

  • OBLITERATUS ↗ ★ 8,946

    A larger refusal-removal toolkit with several abliteration methods, usable locally or through a hosted Hugging Face Space. Check the telemetry setting before you run it: the README describes telemetry as opt-in, but on by default in the Spaces version.

Fine-tune it on your own data

  • Unsloth ↗ ★ 78,528

    A desktop app and Python library for running and training models locally. Its README lists LoRA, QLoRA, full fine-tuning, DPO and GRPO, plus export to formats including GGUF, on Windows, Linux, WSL and macOS.

  • LlamaFactory ↗ ★ 75,713

    Fine-tunes more than 100 language and vision-language models from a command line or a browser-based Web UI, aimed at training without writing code.

Make a smaller download (quantize)

  • llama.cpp ↗ ★ 132,260

    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.

  • MLX LM ↗ ★ 7,467

    Apple's MLX package for running and fine-tuning models on Apple silicon. It can quantize models and upload them to the Hugging Face Hub, and supports LoRA and full fine-tuning.

Merge it with another model

  • mergekit ↗ ★ 7,414

    Merges pretrained models with many merge methods. It works out-of-core, so merges can run on a CPU or with as little as 8 GB of VRAM, and it also handles LoRA extraction and mixture-of-experts merges.

Learn the techniques

  • LLM Course ↗ ★ 84,667

    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
162
Likes
13
Metadata checked
2026-10-25 · 04:15:09 UTC
First seen in index
2026-10-25 · 04:15:09 UTC
Source revision
84ac85d707aef8f2dbf3cea1a96172fcda8d9354

Source-card excerpt

Source model card excerpt: Introduction JOSIE-2-4B-OSS are the latest versions of the JOSIE family to combine structured reasoning, a consistent personality, and self-awareness while being trained entirely on

From Goekdeniz-Guelmez/JOSIE-2-4B-OSS. 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 host or redistribute model weights. Refusal profiles are measured by us on a fixed prompt set; all other behavior claims are source-reported.

Collection rules & limitations ↗