Closed yukiarimo closed 3 months ago
The same issue! Does anyone have a solution to fix it?
@yukiarimo Change step 7 to
model = AutoModelForCausalLM.from_pretrained(base_model_id, quantization_config=bnb_config, device_map="auto")
@yukiarimo Change step 7 to
model = AutoModelForCausalLM.from_pretrained(base_model_id, quantization_config=bnb_config, device_map="auto")
Thanks, I’ll try it today! By the way, do you know how to fine-tune Phi 3 vision with a text-only dataset? I tried, but it says 120 hours instead of 30 minutes as usual for LLaMA :(
Hello. Colab notebook:
llama2-finetune-own-data
is not working correctly!In the code block:
I'm getting the following error:
Any ideas?