embprobe

Overview ¶

Command embprobe is a 30-second semantic sanity check for an embedding model in OUR runtime (tokenizer + pooling + prefixes + ORT): it embeds a few fixed code snippets and two NL queries and prints the cosine matrix. A healthy integration shows the matching snippet winning by a wide margin; broken pooling/prefix/EOS shows flat or shuffled scores. Use it before paying for a full corebench run on a new model.

go run ./cmd/embprobe <model-name>