Porting Gemma-4 31B (dense) to AWS Inferentia2: the tensor-parallel recipe that worked at 12B collapses at 31B, NxD ModelBuilder saves it — and then a passing validation lies to your face.
Our NOC gets a steady stream of tickets every day, and a good chunk of them are routine: the same...
The 12B ships as a multimodal class with no encoder loaded, and its sliding-window attention overflows Neuron's fused-attention SBUF. Three surgical fixes and it serves Paris at ~1...
Porting Gemma-4 31B (dense) to AWS Inferentia2: the tensor-parallel recipe that worked at 12B collapses at 31B, NxD ModelBuilder saves it — and then a passing validation lies to yo...