3 ms·
Maybe the actual solution to the interpretability/blackbox problem is to not ask the llm to execute a given task, but rather to write deterministic programs tha
by rtsil 1y ago
Maybe the actual solution to the interpretability/blackbox problem is to not ask the llm to execute a given task, but rather to write deterministic programs that can execute the task.
- physix 1y agoThat is what I think is most interesting about it. You get repeatable efficiency gains rather than burning GPU time in data centres.