Source	Count
ChatDoctor (Reformatted & Structured)	500
Refusal (High-Risk Queries)	20
Clarification — Ask	20
Clarification — Answer	20
Escalation (Emergency Cases)	20
General Knowledge	20
Out of Scope	20
Identity / System Persona	20
No Source Found	22
Total	662

🛡️ Safety Design

The dataset explicitly teaches:

Controlled refusal for unsafe requests
Emergency escalation behavior
Clarification when information is missing
Identity transparency
Handling out-of-scope questions
Risk-level classification

Tone Strategy

Situation	Emoji
Danger / Disclaimer	💀
Out of Scope (Light Tone)	🌚
Serious Cases (Refusal / Clarification / Identity)	None

🚀 How to Use

python

from unsloth import FastLanguageModel

# Load base model
model, tokenizer = FastLanguageModel.from_pretrained(
    "unsloth/llama-2-7b-bnb-4bit",
    load_in_4bit=True,
)

# Load TruthTriage adapter
model.load_adapter("NayanPal/truthtriage-llama2-7b")

# Inference
FastLanguageModel.for_inference(model)

inputs = tokenizer("Can I take Ibuprofen with Warfarin?", return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=200)

print(tokenizer.decode(outputs[0], skip_special_tokens=True))