mock-trial-ai-v2

Phi-1.5-Mock-Trial-v2-Prototype

Model Details

Model Description

This model is a fine-tuned version of Microsoft's Phi-1.5 (1.3B parameters), serving as the second-iteration prototype for a specialized legal assistant. It was designed to test the feasibility of using small-scale language models for high-stakes Mock Trial preparation.

• Developed by: HobbesTheComputerScientist
• Model type: Causal Language Model
• Language(s): English
• License: MIT
• Finetuned from model: microsoft/phi-1_5

Uses

Direct Use

This prototype is intended for:

• Initial testing of witness statement summarization.
• Proof-of-concept for objection identification logic.
• Low-latency roleplay of legal depositions.
• Initial testing of legal case anaylsis

Out-of-Scope Use

This model is a research prototype and is not intended for professional legal counsel or use in actual courtroom settings.

Bias, Risks, and Limitations

As a 1.3B parameter model, this version is more prone to hallucinations and logic errors compared to larger models (like the Llama-3.1 v3 iteration). It should be used for educational simulation only.

How to Get Started with the Model

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("your-username/Phi-1.5-Mock-Trial-v2")
tokenizer = AutoTokenizer.from_pretrained("your-username/Phi-1.5-Mock-Trial-v2")

# Example inference
inputs = tokenizer("Objection, relevance:", return_tensors="pt")
outputs = model.generate(**inputs, max_length=50)
print(tokenizer.decode(outputs[0]))