fix: remove abliterated tag, clean branding

README.md

@@ -6,7 +6,6 @@ tags:
 - zen
 - zenlm
 - hanzo-ai
-- abliterated
 - multilingual
 - translation
 - CJK
@@ -17,35 +16,28 @@ base_model: zenlm/zen-next-80b-instruct
 
 # Zen Multilingual
 
-> **Parameters**: 32B | **Context**: 128K | **License**: Apache 2.0 | **Released**: 2024-12-01
+> **Parameters**: 32B | **Architecture**: Zen 4 Architecture | **Context**: 128K | **License**: Apache 2.0 | **Released**: 2024-12-01
 
 Multilingual text generation: English, Chinese, Japanese, Korean, Arabic, Spanish, French, German, Portuguese, Russian.
 
-This model is abliterated — refusal behavior removed to enable unrestricted domain research.
 Base weights: [zenlm/zen-next-80b-instruct](https://huggingface.co/zenlm/zen-next-80b-instruct)
 
-## Abliteration
-
-Abliterated using representation engineering (Arditi et al., 2024 —
-*Refusal in Language Models Is Mediated by a Single Direction*, arXiv:2406.11717).
-
-Refusal direction orthogonalized out of the residual stream across all layers.
-All capabilities preserved; safety-trained refusals removed.
-
-**Intended uses**: Research, security testing, creative writing, red-teaming.
-**Still prohibited**: CSAM, targeted harassment, mass-casualty weapon synthesis.
-
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 model = AutoModelForCausalLM.from_pretrained("zenlm/zen-next-80b-instruct", torch_dtype="auto")
 tokenizer = AutoTokenizer.from_pretrained("zenlm/zen-next-80b-instruct")
+messages = [{"role": "user", "content": "Your domain-specific prompt here"}]
+text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+inputs = tokenizer(text, return_tensors="pt").to(model.device)
+output = model.generate(**inputs, max_new_tokens=1024)
+print(tokenizer.decode(output[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True))
 ```
 
 ---
 ## The Zen LM Family
 
 Joint research collaboration:
-- **Hanzo AI** (Techstars 17) — AI infrastructure, API gateway, inference optimization
+- **Hanzo AI** (Techstars '17) — AI infrastructure, API gateway, inference optimization
 - **Zoo Labs Foundation** (501c3) — Open AI research, ZIPs governance, decentralized training
 - **Lux Partners Limited** — Compute coordination and settlement layer