ModelForgeLab
Sign in Sign up

Phi-3 Mini Named Entity Recognition

Published by @nlp_tools_dev · Community adapter

text-classification #NER#entity-extraction#annotation#phi3

Entity extraction for PER, ORG, LOC, PRODUCT, EVENT, and DATE spans. Outputs JSON with character offsets. Suitable for pipeline integration without a separate NER model.

Usage

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-128k-instruct")
base = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-128k-instruct")
model = PeftModel.from_pretrained(base, "modelforgelab/phi3-mini-ner-lora")

inputs = tokenizer("Your prompt here", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Compatibility

  • transformers>=4.40

You might also like

Llama 3 8B Sentiment Classifier
text-classification public

Llama 3 8B Sentiment Classifier

meta-llama/Meta-Llama-3-8B-Instruct