Could we machine translatation task using this model?
#91
by
Pitambarmuduli
- opened
want to translate from German to English by using falcon-7b model .
Pitambarmuduli
changed discussion status to
closed
Pitambarmuduli
changed discussion status to
open
So falcon 7b is a causal decoder-only model , which means it's primarily design for autoregressive generation task , such as next token prediction . Usually for machine translation you require both Encoder and Decoder . I mean you can try but it won't be as efficient
here are some models you can use:
BART, BigBird-Pegasus, Blenderbot, BlenderbotSmall, Encoder decoder, FairSeq Machine-Translation, GPTSAN-japanese, LED, LongT5, M2M100, Marian, mBART, MT5, MVP, NLLB, NLLB-MOE, Pegasus, PEGASUS-X, PLBart, ProphetNet, SwitchTransformers, T5, UMT5, XLM-ProphetNet
Yeah, I tried, but it's not as effective; sometimes it's translated, sometimes it's not.