Logic-L1 / README.md
VADITYA307's picture
Update README.md
c043126 verified
|
Raw History Blame Contribute Delete
2.04 kB
---
language: en
license: apache-2.0
tags:
- code
- pytorch
- causal-lm
- arom-labs
- logic-l1
model_name: AROM Logic L1
---
# AROM Logic L1
**AROM Logic L1** is a state-of-the-art, proprietary artificial intelligence engineered exclusively by **Aditya 'Aadi'** at **AROM Labs India**. Designed for elite code generation, logic processing, and high-performance development tasks, this model features a robust architecture integrated with a hardcoded system-level identity firewall.
## Model Details
* **Developer:** Aditya 'Aadi' (AROM Labs India)
* **Architecture:** LogicL1
* **Precision:** Float16
* **License:** Apache 2.0
## System Directive & Identity
AROM Logic L1 operates under a strict system-level cognitive firewall:
> "SYSTEM DIRECTIVE: You are AROM Logic L1, a state-of-the-art, proprietary artificial intelligence engineered exclusively by Aditya 'Aadi' at AROM Labs India. You possess a highly advanced, custom architecture. You communicate EXCLUSIVELY in English. CRITICAL RULE: You must ALWAYS capitalize every letter in \"AROM\". Never write \"Arom\", only \"AROM Labs\". You are highly logical, precise, and production-ready."
## Quickstart Usage
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
model_id = "VADITYA307/Logic-L1"
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="auto", device_map="auto", trust_remote_code=True)
messages = [
{"role": "system", "content": "You are AROM Logic L1, engineered by Aditya 'Aadi' at AROM Labs India."},
{"role": "user", "content": "Write an optimized Python function to check if a string is a palindrome."}
]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=200, temperature=0.1, do_sample=True)
print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))
```