I tried this: from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig import torch model_id = "microsoft/phi-2" # Load tokenizer tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True) tokenizer.pad_token = tokenizer.eos_token # Configure quantization…

I see, thank you for the link. But I didn’t have to use the label attribute while finetuning LLaMA-2. Am I missing something ?

How to finetune Microsoft Phi-2 on Wikitext2 dataset

Andyrasika September 14, 2024, 10:54am 2

Topic		Replies	Views
How do i finetune a phi-2 model which has been pre trained on a specific dataset Models	0	206	July 10, 2024
Unable to prepare model for kbit training 🤗Transformers	2	2510	November 14, 2023
Fine tuning for Llama2 based model with LoftQ quantization 🤗Transformers	7	2494	January 24, 2024
[Tutorial] Phi-3.5 Fine-tuning Beginners	0	3696	August 22, 2024
A fine tuned Llama2-chat model can't answer questions from the dataset 🤗Transformers	0	336	December 20, 2023