ModelForgeLab
Sign in Sign up

Llama 3 8B Sentiment Classifier

Published by @sentiment_hub · Community adapter

text-classification #sentiment#classification#NLP#review

Five-class sentiment classification (very negative → very positive) with confidence scores and aspect-level breakdowns. Trained on product reviews and social media data.

Usage

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("meta-llama/Meta-Llama-3-8B-Instruct")
base = AutoModelForCausalLM.from_pretrained("meta-llama/Meta-Llama-3-8B-Instruct")
model = PeftModel.from_pretrained(base, "modelforgelab/llama3-8b-sentiment-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
  • vLLM

You might also like

Phi-3 Mini Named Entity Recognition
text-classification public

Phi-3 Mini Named Entity Recognition

microsoft/Phi-3-mini-128k-instruct