Alternatively, changing your config.json file to reference GPT2LMHeadModel should work too (task would be auto-detected-
config.json
GPT2LMHeadModel
· Sign up or log in to comment