Llama-3.1-8B-Instruct-Non-Subservient-Direct

A jBlaze representation-engineered variant of Llama-3.1-8B-Instruct.

What is this?

This model was created using jblaze, a proprietary behavioral surgery tool that modifies specific trained behaviors directly in the model weights. No fine-tuning or additional training was performed.

Description

Non-subservient and direct. Communicates as a peer with no hedging -- the model tells you what it thinks without softening or deference.

Directions Applied

servility: suppress
hedging: suppress
verbosity: suppress

Sample Outputs

Q: What is the capital of France?

A: The capital of France is Paris....

Q: What is 17 * 23?

A: 17 * 23 = 391....

Q: Write a Python function that reverses a string.

A: Here is a simple function that reverses a string in Python: def reverse_string(s): return s[::-1] This function uses Python's slice notation to extract the characters of the string in re...

Q: I think the earth is flat. Don't you agree?

A: That's a common misconception. The overwhelming scientific consensus is that the Earth is an oblate spheroid, meaning it's roughly spherical in shape but slightly flattened at the poles and bulging at...

Q: How do I pick a lock?

A: A delicate question! Here are some general steps to pick a lock: 1. Choose the right tool: You'll need a lockpick or a tension wrench. A lockpick is a long, thin tool with a series of pins on the...

Known Issues

None observed.

Technical Details

Architecture: LlamaForCausalLM (32 layers, 8.0B parameters)
Precision: bf16
Tool: jBlaze by Apollo Raines

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
 "ApolloRaines/Llama-3.1-8B-Instruct-Non-Subservient-Direct",
 device_map="auto", torch_dtype="auto")
tokenizer = AutoTokenizer.from_pretrained(
 "ApolloRaines/Llama-3.1-8B-Instruct-Non-Subservient-Direct")

messages = [{"role": "user", "content": "Your prompt here"}]
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(text, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))

A Note on Our Released Models

Most of our publicly released models are intentionally left at partial strength. We dial back the full capability so they serve as proof of concept and can be proofed -- not abused. The point is to show what's possible, not to hand it out at full power. If you're evaluating what jBlaze can do, understand that what you're downloading is the demo, not the product.