Hi!
In the paper, you mention that for the transposon model (evo-1-8k-transposon) you "prompted the model with the start token for each sequence generation" for the sequence generation step. I wonder if you could be more specific as to which is the start token used for these generations? I could not find this information throughout the paper nor in your repository.
Apologies if I am missing something.
Thank you very much!
Hi!
In the paper, you mention that for the transposon model (evo-1-8k-transposon) you "prompted the model with the start token for each sequence generation" for the sequence generation step. I wonder if you could be more specific as to which is the start token used for these generations? I could not find this information throughout the paper nor in your repository.
Apologies if I am missing something.
Thank you very much!