HackerTrans
TopNewTrendsCommentsPastAskShowJobs

hiroakiaizawa

no profile record

Submissions

Executable notebook for testing earthquake-event concentration (Colab)

colab.research.google.com
1 points·by hiroakiaizawa·2 mesi fa·1 comments

Collapse is not random – run a minimal test in 30 seconds (Colab)

colab.research.google.com
3 points·by hiroakiaizawa·2 mesi fa·2 comments

comments

hiroakiaizawa
·2 mesi fa·discuss
Good reminder that raw tokens/sec numbers can be misleading without latency and context-window considerations.
hiroakiaizawa
·2 mesi fa·discuss
Interesting approach. I like that the implementation focuses on scalability rather than only visualization.
hiroakiaizawa
·2 mesi fa·discuss
The notebook is intentionally minimal.

Not a prediction model or causal explanation — just a reproducible concentration check with fixed ex-ante definitions and minimal outputs.

Runs in ~30 sec on Colab.
hiroakiaizawa
·2 mesi fa·discuss
One thing I've started appreciating with LLM-assisted workflows is how important fixed evaluation protocols are.

Without pre-defined definitions and locked procedures, it's extremely easy to mistake iterative adaptation for genuine signal.
hiroakiaizawa
·2 mesi fa·discuss
[flagged]
hiroakiaizawa
·2 mesi fa·discuss
Curious what domains people would try this on. Would love to see other datasets.
hiroakiaizawa
·2 mesi fa·discuss
Tested on finance / power / earthquakes.

Minimal version only. Happy to adapt to other datasets.
hiroakiaizawa
·2 mesi fa·discuss
Interesting. What are the main latency bottlenecks in practice?
hiroakiaizawa
·2 mesi fa·discuss
Nice. What scale does this realistically reach on a single machine?
hiroakiaizawa
·2 mesi fa·discuss
Interesting. What are the main trade-offs they expect from the switch?