Transformers How to use UnfaithRL/Qwen2.5-0.5B-Instruct-mmlu_trained_ha-2048 with Transformers:
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("UnfaithRL/Qwen2.5-0.5B-Instruct-mmlu_trained_ha-2048")
model = AutoModelForCausalLM.from_pretrained("UnfaithRL/Qwen2.5-0.5B-Instruct-mmlu_trained_ha-2048", device_map="auto") 
