83 lines
1.7 KiB
Markdown
83 lines
1.7 KiB
Markdown
|
---
|
||
|
|
model_name: LeeChanRX-LegalRights
|
||
|
|
tags:
|
||
|
|
- lora
|
||
|
|
- sft
|
||
|
|
- transformers
|
||
|
|
- trl
|
||
|
|
- unsloth
|
||
|
|
- legal-ai
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
license: apache-2.0
|
||
|
|
datasets:
|
||
|
|
- LegalFlow-v1
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
---
|
||
|
|
|
||
|
|
# LeeChanRX-LegalRights
|
||
|
|
|
||
|
|
LeeChanRX-LegalRights is a fine-tuned legal language model built using supervised fine-tuning (SFT) on curated legal datasets.
|
||
|
|
|
||
|
|
## Features
|
||
|
|
- Legal question answering
|
||
|
|
- Contract explanation
|
||
|
|
- Basic legal reasoning
|
||
|
|
- Plain-language legal responses
|
||
|
|
|
||
|
|
## Training Data
|
||
|
|
- LegalFlow-v1 (50,000 samples)
|
||
|
|
- LegalBench
|
||
|
|
- LEDGAR
|
||
|
|
- EURLEX
|
||
|
|
|
||
|
|
## Method
|
||
|
|
- LoRA fine-tuning (Unsloth optimized)
|
||
|
|
- ~29M trainable parameters
|
||
|
|
- 0.96% of total model parameters
|
||
|
|
|
||
|
|
## Usage (Transformers)
|
||
|
|
|
||
|
|
```python
|
||
|
|
from transformers import AutoTokenizer, AutoModelForCausalLM
|
||
|
|
|
||
|
|
model_name = "LeeChanRX/LeeChanRX-LegalRights"
|
||
|
|
|
||
|
|
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
||
|
|
model = AutoModelForCausalLM.from_pretrained(model_name)
|
||
|
|
|
||
|
|
prompt = "Explain breach of contract in simple terms"
|
||
|
|
|
||
|
|
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
|
||
|
|
|
||
|
|
outputs = model.generate(
|
||
|
|
**inputs,
|
||
|
|
max_new_tokens=200,
|
||
|
|
temperature=0.7,
|
||
|
|
top_p=0.9
|
||
|
|
)
|
||
|
|
|
||
|
|
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
|
||
|
|
```
|
||
|
|
|
||
|
|
### Chat Format Usage
|
||
|
|
```python
|
||
|
|
messages = [
|
||
|
|
{"role": "system", "content": "You are a legal AI assistant."},
|
||
|
|
{"role": "user", "content": "What are my rights if a contract is broken?"}
|
||
|
|
]
|
||
|
|
```
|
||
|
|
|
||
|
|
### Disclaimer
|
||
|
|
This model is for educational purposes only and is not a substitute for professional legal advice.
|
||
|
|
|
||
|
|
### Status
|
||
|
|
- Training: Completed
|
||
|
|
- Model: Ready for inference
|
||
|
|
- Version: LegalRights v1
|
||
|
|
|
||
|
|
---
|
||
|
|
If you want next upgrade, I can help you build:
|
||
|
|
|
||
|
|
👉 LegalRights v2 (advanced reasoning + court prediction)
|
||
|
|
---
|