Is it the Flan-T5 base model being used?

Course Q&A Generative AI with Large Language Models
1

I realize that Flan-T5 comes in several different sizes, but at the start of Lab 1 it was mentioned that the Flan-T5 base model was being used. Can anyone confirm whether it is indeed the 250M parameter base model?

I could not find which Flan-T5 model was being utilized in the unfortunately - perhaps I just missed it…

Tx!

2

This is found in a code cell in the first lab:

model_name=‘google/flan-t5-base’

model = AutoModelForSeq2SeqLM.from_pretrained(model_name)

3

I see the line you refer to. I obviously missed that… Thanks for taking the time to reply.

1 Like