Base model: meta-llama/Llama-2-13b-hf
Fine-tuning method: LoRA (PEFT) via curriculum learning (SCTT — 3 phases, 10 epochs)
Task: 3-class pairwise classification — predict which of two responses (A / B / Equal) is more creative
Labels: 0=A, 1=B, 2=Equal
Checkpoint: Phase 3, step 540,000

Usage

python

import torch
from transformers import AutoModelForSequenceClassification, AutoTokenizer
from peft import PeftModel

tokenizer = AutoTokenizer.from_pretrained("PhillipGre/llama2-13b-sctt-classification")
base_model = AutoModelForSequenceClassification.from_pretrained(
    "meta-llama/Llama-2-13b-hf",
    num_labels=3,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)
base_model.resize_token_embeddings(len(tokenizer))
base_model.config.pad_token_id = tokenizer.pad_token_id

model = PeftModel.from_pretrained(base_model, "PhillipGre/llama2-13b-sctt-classification")
model = model.merge_and_unload()
model.eval()

prompt = "experiment: testing bird's understanding of human speech\nA: response one\nB: response two"
inputs = tokenizer(prompt, return_tensors="pt", truncation=True, max_length=180)
with torch.no_grad():
    logits = model(**inputs).logits
label_map = {0: "A", 1: "B", 2: "Equal"}
print(label_map[logits.argmax().item()])