model-extraction-relu-logits
Guidance for extracting weight matrices from black-box ReLU neural networks using only input-output queries. This skill applies when tasked with recovering internal parameters (weights, biases) of a n
Also installable via skills CLI
npx skills add letta-ai/skills/letta/benchmarks/trajectory-only/model-extraction-relu-logits
Source
Path:
letta/benchmarks/trajectory-only/model-extraction-relu-logits(main)