Taran Iyengar
Results
3
comments of
Taran Iyengar
@arjunsuresh md5sum for day_23_sparse.npy: d6c11a8cebb5cabcea6ff67646641f3f
Hi, from what I've understood, yes TEST01 was not discussed and perhaps we can decide on that in the post-mortem, but it's normally present for all workloads except LLMs, but...
Just to add on, initially during the taskforce we'd observed that without the 128 token limit we were getting much poorer accuracy despite the system prompt asking the model to...