index
:
prime-grokking.git
main
Prime Grokking — can minimal architectures grok next-prime? Pre-registered negative-class experiment (weight-tied RNN vs transformer).
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
runs-take2
Mode
Name
Size
d---------
rnn
/
seed0
32
log
plain
d---------
transformer
/
seed0
32
log
plain