sriramk
@dssskizd
RT @karpathy: The race for LLM cognitive core - a few billion param model that maximally sacrifices encyclopedic knowledge for capability…
0 reply
0 recast
0 reaction