Added a test to validate that llama.cpp native generation handles UTF8 sequences correctly. The test uses a tiny model outputing a predictable sequence of tokens. The python script used to produce the test model is attached on Bugzilla. Differential Revision: https://phabricator.services.mozilla.com/D308124
fake hub
This directory is a fake hub that is served via chrome://global/content/ml/tests.
All files in this directory are included with a wildcard in the component jar.md.