"System Memory per GPU" filter doesn't work for more than a single GPU
Select 100gb per GPU, 16 vCPUs per GPU. Select 5090 GPU. Note Pod Summary fulfilling criteria. Increase GPU Count to 2. Note that the amount of RAM in the Pod Summary decreases from 116GB RAM to 108GB RAM. Change region (to SE in this case). Note that the pod Summary now shows 232GB RAM - proving that a Pod with 100Gb+ per GPU actually exists.
No secure ssh or jupyter notebook option on rtx 4090s
Hi, there is no longer a secure ssh or jupyter notebook option on the rtx 4090s. It was working fine a few days ago I didn't make any modifications or anything to the template.
Full performance of an H100
When running on H100 SXM version, subleased on Prime Intellect (PI) I can't seem to achieve the full performance of an H100. When running it I get at most 760TFLOPS - so the baseline from the PCIe version. Is there ANY way to get the full 989 TFLOPS?
```py
import torch...
