NYC Housing Rights Assistant

A fine-tuned DialoGPT model specialized for NYC tenant rights questions with 100% accuracy on critical tenant information.

Model Details

  • Base Model: microsoft/DialoGPT-medium
  • Training Data: 597 NYC tenant rights examples
  • Accuracy: 100% on critical test questions
  • Specialization: NYC housing laws, tenant rights, legal contacts

Performance

Test Results

  • โœ… Legal Aid Society contact: 212-577-3300
  • โœ… HPD contact: 212-863-6300
  • โœ… Application fee limit: $20
  • โœ… Security deposit return: 14 days
  • โœ… Eviction procedures: Court order required

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

model_name = "aanshshah/nyc-housing-rights"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Example usage
prompt = "Human: What's The Legal Aid Society phone number?\n\nAssistant:"
inputs = tokenizer.encode(prompt, return_tensors='pt')
outputs = model.generate(inputs, temperature=0.05, max_length=150)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)

Training Details

  • Training Examples: 597 (33.2x expansion from original 18)
  • Training Method: Ultra-accuracy fine-tuning
  • Temperature: 0.05 for precise responses
  • Focus Areas: Contact information, factual recall, legal procedures

Limitations

This model is specialized for NYC tenant rights only. For legal advice, consult qualified attorneys.

Emergency Contacts

  • Emergencies: 911
  • Legal Aid Society: 212-577-3300
  • HPD: 212-863-6300
  • General Tenant Issues: 311
Downloads last month
2
Safetensors
Model size
0.4B params
Tensor type
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ 2 Ask for provider support

Model tree for aanshshah/nyc-housing-rights

Finetuned
(88)
this model