mattlanham_-_pii-redaction-v0.3-8bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

pii-redaction-v0.3 - bnb 8bits

• Model creator: https://huggingface.co/mattlanham/
• Original model: https://huggingface.co/mattlanham/pii-redaction-v0.3/

Original model description:

base_model: unsloth/phi-3-mini-4k-instruct-bnb-4bit language:

• en license: apache-2.0 tags:
• text-generation-inference
• transformers
• unsloth
• mistral
• trl

---

Finetuned phi-3 mini model for data extraction / pii redaction

Trained using synthetic data, this model will provide a JSON output containing:

With user input: My order hasn't arrived, it's order number 123456 and my email is [email protected], please can you provide an update! Thanks John.

The model will respond with:

{
  "redacted_message": "My order hasn't arrived, it's order number {order_number} and my email is {email}, please can you provide an update! Thanks {name}.",
  "data": {
    "order_number": "123456",
    "email": "[email protected]",
    "name": "John",

  }
}

• Developed by: mattlanham
• License: apache-2.0
• Finetuned from model : unsloth/phi-3-mini-4k-instruct-bnb-4bit

This mistral model was trained 2x faster with Unsloth and Huggingface's TRL library.

<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>