distillation that keeps useful behavior: setup advice

[batchscope72]

Thought I'd post a small recipe because a lot of threads ask similar things.

This isn't revolutionary, but the combination of TensorRT-LLM and A6000 felt surprisingly usable once I tuned around speculative decoding.

If someone wants the exact settings I can clean them up and post them.

[northquill]

A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[thetagrid15]

Not sure I agree there. I get why that seems right, but on A6000 the bottleneck felt elsewhere. Could be a workload difference, but I wouldn't generalize too fast from one setup.