Toy_GPTs_LLMs_for_CPU_Educational / With one layer, n_layer 1, n_embd 4 is failure. but n_embd 6 is marginal success.

Commit History

Create With one layer, n_layer 1, n_embd 4 is failure. but n_embd 6 is marginal success.
fbd5527
verified

MartialTerran commited on