When indexing a large monorepo, once the embeddings are generated, it can result in the following error message.
Embedding complete: 200000 texts in 2500.0s
Store phase: storing 22000 file(s) into vector database...
Store phase complete: 0 files stored in 0.1s (0 total chunks)
Indexing failed: Invalid string length
This unfortunately seems to throw away all of the computed embeddings.
When indexing a large monorepo, once the embeddings are generated, it can result in the following error message.
This unfortunately seems to throw away all of the computed embeddings.