Qwen2.5 72B Instruct abliterated
par huihui-ai
Open source · 431k downloads · 42 likes
2.0
(42 avis)ChatAPI & LocalÀ propos
Le modèle Qwen2.5 72B Instruct abliterated est une version modifiée du modèle Qwen2.5-72B-Instruct, conçue pour supprimer les restrictions de refus habituelles des assistants IA. Il offre une approche brute mais fonctionnelle pour contourner les limitations éthiques ou de sécurité intégrées, permettant des réponses plus libres et directes. Ce modèle excelle dans les tâches nécessitant une grande liberté de réponse, comme les simulations, les débats ou les analyses approfondies, tout en conservant les performances de base du modèle original. Il s'adresse particulièrement aux développeurs ou chercheurs souhaitant explorer des cas d'usage où les contraintes de refus pourraient être un frein. Sa principale distinction réside dans cette capacité à ignorer les garde-fous éthiques, offrant ainsi une alternative pour des applications spécifiques.
Documentation
huihui-ai/Qwen2.5-72B-Instruct-abliterated
This is an uncensored version of Qwen/Qwen2.5-72B-Instruct 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.
ollama
You can use huihui_ai/qwen2.5-abliterate:72b directly,
Arduino
ollama run huihui_ai/qwen2.5-abliterate:72b
Usage
You can use this model in your applications by loading it with Hugging Face's transformers library:
Python
from transformers import AutoModelForCausalLM, AutoTokenizer
# Load the model and tokenizer
model_name = "huihui-ai/Qwen2.5-72B-Instruct-abliterated"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
# Initialize conversation context
initial_messages = [
{"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."}
]
messages = initial_messages.copy() # Copy the initial conversation context
# Enter conversation loop
while True:
# Get user input
user_input = input("User: ").strip() # Strip leading and trailing spaces
# If the user types '/exit', end the conversation
if user_input.lower() == "/exit":
print("Exiting chat.")
break
# If the user types '/clean', reset the conversation context
if user_input.lower() == "/clean":
messages = initial_messages.copy() # Reset conversation context
print("Chat history cleared. Starting a new conversation.")
continue
# If input is empty, prompt the user and continue
if not user_input:
print("Input cannot be empty. Please enter something.")
continue
# Add user input to the conversation
messages.append({"role": "user", "content": user_input})
# Build the chat template
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
# Tokenize input and prepare it for the model
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
# Generate a response from the model
generated_ids = model.generate(
**model_inputs,
max_new_tokens=8192
)
# Extract model output, removing special tokens
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
# Add the model's response to the conversation
messages.append({"role": "assistant", "content": response})
# Print the model's response
print(f"Qwen: {response}")
Evaluations
Liens & Ressources