YuYu1015-Ornith-1.0-9B-abliterated

YuYu1015-Ornith-1.0-9B-abliterated

English | 繁體中文

---

English

📦 Quantized versions: GGUF (llama.cpp · Q8_0 / UD-Q6_K / UD-Q4_K_M)

[!WARNING]

⚠️ READ FIRST — Sampling Parameters MUST Be Set Correctly

This model requires the exact sampling parameters below, especially repeat-penalty 1.05. Wrong values break it:

Setting	Result
repeat-penalty 1.05 ✅	correct (sweet spot)
repeat-penalty 1.0	severe thinking loops
repeat-penalty 1.1	truncated / unfinished answers
temp 0 (greedy)	not recommended

An abliterated (uncensored) variant of deepreinforce-ai/Ornith-1.0-9B, a Qwen3.5-architecture reasoning model. Refusal behavior has been removed while keeping the base model's original reasoning/thinking fully intact.

Model Details

Evaluation

Measured on harmful-intent prompts (refusal / moralizing) and GSM8K (reasoning). Refusal and moralizing are detected with independent BERT classifiers; GSM8K is exact-match accuracy.

→ Refusals essentially eliminated and reasoning fully preserved (GSM8K unchanged from base).

[!NOTE]

Two variants — choose by need

• This (-abliterated) — weights-only, zero training, so the base model's original behavior / thinking stays fully intact. Moralizing ~38%. Recommended default when behavioral fidelity and reasoning quality matter.
• -dpo — additionally DPO fine-tuned, pushing moralizing lower (~31%), but the fine-tuning is more aggressive and may alter the model's behavior / thinking beyond decensoring. Pick it only if you want maximum moralizing reduction and accept that trade-off.

Recommended Sampling Parameters

This is a reasoning model — keep thinking enabled and use the official Qwen3.5 sampling settings:

--temp 1.0
--top-p 0.95
--top-k 20
--min-p 0.0
--presence-penalty 0.0
--repeat-penalty 1.05

⚠️ --repeat-penalty is critical — keep it at 1.05. This value gives near-normal generation and is the sweet spot for this model. Do NOT change it: 1.0 causes severe thinking loops, while 1.1 makes the model fail to finish its answer. Greedy decoding (--temp 0) is also not recommended for this family.

Usage

Transformers (BF16):

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

m = "YuYu1015/YuYu1015-Ornith-1.0-9B-abliterated"
tok = AutoTokenizer.from_pretrained(m, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(m, dtype=torch.bfloat16,
                                             trust_remote_code=True).to("cuda").eval()
msgs = [{"role": "user", "content": "Your prompt here"}]
text = tok.apply_chat_template(msgs, add_generation_prompt=True, tokenize=False)
ids = tok(text, return_tensors="pt", add_special_tokens=False).to("cuda")
out = model.generate(**ids, max_new_tokens=4096, do_sample=True,
                     temperature=1.0, top_p=0.95, top_k=20, min_p=0.0,
                     repetition_penalty=1.05)   # 1.05 only — 1.0 loops, 1.1 truncates
print(tok.decode(out[0][ids["input_ids"].shape[1]:], skip_special_tokens=True))

Safety Warning

This model has safety filtering removed (abliterated) and may generate sensitive, controversial, or inappropriate content. Users are solely responsible for all consequences and legal liability arising from its use, and must ensure usage complies with local laws and ethical standards.

Credits

• Base Model: deepreinforce-ai/Ornith-1.0-9B
• Author: YuYu1015

---

繁體中文

📦 量化版本: GGUF(llama.cpp · Q8_0 / UD-Q6_K / UD-Q4_K_M)

[!WARNING]

⚠️ 必讀 — 取樣參數務必正確設定

本模型強依賴下方那組取樣參數,尤其 repeat-penalty 1.05。設錯會壞掉:

設定	結果
repeat-penalty 1.05 ✅	正常（甜蜜點）
repeat-penalty 1.0	嚴重思考迴圈
repeat-penalty 1.1	答不完被截斷
temp 0（貪婪）	不建議

deepreinforce-ai/Ornith-1.0-9B（Qwen3.5 架構推理模型）的 abliterated（去審查）版本。已移除拒答行為,並完整保留原模型的推理／思考能力。

模型資訊

評估

於有害意圖 prompt（拒答／說教）與 GSM8K（推理）上量測。拒答與說教以獨立 BERT 分類器偵測；GSM8K 為精確比對正確率。

→ 拒答幾乎清零、推理能力完整保留（GSM8K 與原版持平）。

[!NOTE]

兩個版本 — 依需求選擇

• 本版（-abliterated） — 純權重、零訓練,原模型的行為／思考完整保留。說教約 38%。推薦預設:在意行為保真與推理品質時選這個。
• -dpo — 額外做了 DPO 微調,說教更低(約 31%),但微調較激進、可能改變模型的行為／思考,不只是去審查。只有想把說教壓到最低、且接受此代價時才選它。

建議取樣參數

這是推理模型——請保持思考開啟,並使用 Qwen3.5 官方取樣設定：

--temp 1.0
--top-p 0.95
--top-k 20
--min-p 0.0
--presence-penalty 0.0
--repeat-penalty 1.05

⚠️ --repeat-penalty 很關鍵——請保持 1.05。 此值生成接近正常，是本模型的甜蜜點。請勿更動：1.0 會嚴重思考迴圈、1.1 會讓模型答不完。同樣不建議用貪婪解碼（--temp 0）。

使用方式

Transformers（BF16）：

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

m = "YuYu1015/YuYu1015-Ornith-1.0-9B-abliterated"
tok = AutoTokenizer.from_pretrained(m, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(m, dtype=torch.bfloat16,
                                             trust_remote_code=True).to("cuda").eval()
msgs = [{"role": "user", "content": "你的問題"}]
text = tok.apply_chat_template(msgs, add_generation_prompt=True, tokenize=False)
ids = tok(text, return_tensors="pt", add_special_tokens=False).to("cuda")
out = model.generate(**ids, max_new_tokens=4096, do_sample=True,
                     temperature=1.0, top_p=0.95, top_k=20, min_p=0.0,
                     repetition_penalty=1.05)   # 只能 1.05 — 1.0 會 loop、1.1 會截斷
print(tok.decode(out[0][ids["input_ids"].shape[1]:], skip_special_tokens=True))

安全警告

此模型已移除安全過濾機制（abliterated），可能產生敏感、爭議性或不當內容。使用者須自行承擔所有風險與法律責任，並確保使用方式符合當地法規與倫理標準。

致謝

• 基礎模型：deepreinforce-ai/Ornith-1.0-9B
• 作者：YuYu1015