← back to catalog · registered 2026-08-22 13:56

huehui/Discord-Micae-Hermes-3-3B-abliterated

huehui Llama 3.2B second-order
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/huehui%2FDiscord-Micae-Hermes-3-3B-abliterated"
Response includes
  • classification m1
  • files 11
  • hub_downloads_all_time 214
  • author_summary 1 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · lifetime
214
36 last 30d - stable
Likes
5
Descendants
2
in 2 direct forks
Model age
14mo ago
created 2025-08-03

Training datasets

Corpora the author lists in the model card. Datasets tracked in our /datasets catalog carry a category badge linking to the workflow stage. Others open on Hugging Face.

Downloads over time
Now233→from8↑2,813%
0851702568 on Aug 6, 2025233 on Oct 11Aug '25Oct '25Dec '25FebAprJunAugOct
Aug 6, 2025 → Oct 11 · 101 snapshots · spans 431 days

Genealogy 2 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Metadata

License
mit
Tags
transformers safetensors llama text-generation causal-lm instruct chat fine-tuned merged-lora llama-3 hermes discord-dataset
Total size
5.98 GB
Files
11
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2025-08-03 05:05

Files by quantization

Auxiliary files 11 files 6.00 GB
model-00001-of-00002.safetensors 4.62 GB 26bb53a5 download
model-00002-of-00002.safetensors 1.36 GB 525c69f2 download
tokenizer.json 16.4 MB 9f908f9b download
Micae.png 3.04 MB 9c611f6e download
tokenizer_config.json 51.2 KB 088c2d04 download
model.safetensors.index.json 20.7 KB d55fa5b4 download
README.md 3.06 KB bffb232f download
.gitattributes 1.53 KB 52373fe2 download
config.json 933 B a4556f72 download
special_tokens_map.json 467 B 6a9ba8de download
generation_config.json 227 B 315341f5 download

README current version from Hugging Face


tags:
- transformers
- causal-lm
- text-generation
- instruct
- chat
- fine-tuned
- merged-lora
- llama-3
- hermes
- discord-dataset
- conversational-ai
- chatml
- pytorch
- open-weights
- 3b-parameters
- abliterated
base_model:
- NousResearch/Hermes-3-Llama-3.2-3B-abliterated
model-index:
- name: Discord-Micae-Hermes-3-3B
results: []
datasets:
- mookiezi/Discord-OpenMicae
library_name: transformers
license: mit

Try this model on Google Colab for free: Open in Colab

Discord-Micae-Hermes-3-3B-abliterated

Model Description

This is an uncensored version of mookiezi/Discord-Micae-Hermes-3-3B created with abliteration (see remove-refusals-with-transformers to know more about it).
This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.

Micae

Generation

from transformers import pipeline

gen = pipeline("text-generation", model="mookiezi/Discord-Micae-Hermes-3-3B-abliterated")
print(gen(
    "<im_start>user\nwhat do you do?<|im_end|>",
    max_new_tokens=100
))

Discord-Micae-Hermes-3-3B Model Description

Discord-Micae-Hermes-3-3B is a new finetune on NousResearch/Hermes-3-Llama-3.2-3B.

The model was trained on 17 million tokens of 250 thousand Discord STX (single turn exchanges) for 6 epochs and 5.5 million tokens of 100 thousand multi-turn chains for 6 epochs at learn rate 2e-5, finishing with both datesets combined for 1 epoch at 1e-5. We used a cosine warmup with 220 warmup steps for each phase. The LoRA adapter was trained with alpha = 32 and r = 8.

Dataset

This model was fine-tuned on the mookiezi/Discord-OpenMicae dataset.

Prompting

Micae uses the same ChatML as the prompt format as Hermes 3 and handles context.

<|im_start|>user
what do you think about ai?<|im_end|>
<|im_start|>assistant
It's cool, but I don't know what it is.<|im_end|>
<|im_start|>user
where do you think it's heading?<|im_end|>
<|im_start|>assistant
I can only speculate and make guesses.<|im_end|>
<|im_start|>user
what do you do use it for?<|im_end|>
<|im_start|>assistant
For fun and to learn new things.<|im_end|>
<|im_start|>user
like what?<|im_end|>
<|im_start|>assistant
Like playing games or building robots.<|im_end|>
<|im_start|>user
do you like pokemon?<|im_end|>
<|im_start|>assistant
Yes, but not the actual game.<|im_end|>
<|im_start|>user
what do you mean?<|im_end|>
<|im_start|>assistant
The idea behind the game, that a person could create an entire world from nothing.<|im_end|>

README history 20 versions

The author's README evolved over time. Click a version to see its content at that point.

  1. 2025-08-03Upload README.md with huggingface_hub4c164323.1 KB
    Loading...
  2. 2025-08-03Update README.mdde277303 KB
    Loading...
  3. 2025-08-03Upload folder using huggingface_hubd7dfdd12.7 KB
    Loading...
  4. 2025-08-03Update README.md469c0793.1 KB
    Loading...
  5. 2025-08-03Update README.md8b8b0fa3 KB
    Loading...
  6. 2025-08-03Update README.md2b5a4dd3 KB
    Loading...
  7. 2025-08-03Update README.mdcc9cefe2.7 KB
    Loading...
  8. 2025-08-03Update README.mda11768a2.4 KB
    Loading...
  9. 2025-08-03Update README.mde70f23d2.4 KB
    Loading...
  10. 2025-08-03Update README.mdd5aa8ca2.4 KB
    Loading...
  11. 2025-08-03Update README.mde9ff0bc2.4 KB
    Loading...
  12. 2025-08-03Update README.md2fe2c1e2.4 KB
    Loading...
  13. 2025-08-03Update README.mdd8a26672.3 KB
    Loading...
  14. 2025-08-03Update README.md7a350962.3 KB
    Loading...
  15. 2025-08-03Update README.md7bb17f52.3 KB
    Loading...
  16. 2025-08-03Update README.md70215532.3 KB
    Loading...
  17. 2025-08-03Update README.mdac43afb2.3 KB
    Loading...
  18. 2025-08-03Update README.mdcf849e92.3 KB
    Loading...
  19. 2025-08-03Update README.mdd8d83562.3 KB
    Loading...
  20. 2025-08-03Update README.mdd5ecbfa2.1 KB
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration