Jasper Dekoninck

Documentation: Add note on double escaping "\n" being necessary for lmql.run, but not for @lmql.query.

2

In the following query, I am still required to write "\\n" instead of "\n" in order to be able to compile the query in the Python package (note that this...

STOPS_BEFORE on multiple tokens retains the first token

4

The following query returns "What did the fish say when" instead of the expected "What did the fish say". ```python argmax(max_len=80) """A list of good dad jokes. A indicates the...

When max_len is exceeded the interpreter will always return an error, even if the sequence would already be valid.

1

I have the following query: ```python sample(temperature=0.8, openai_chunksize=32, max_len=64, chatty_openai=True) "The movie review in positive sentiment is: '[OUTPUT]" FROM "openai/text-ada-001" ``` The OUTPUT variable in this case is thus constrained...

More advanced stopping conditions: STOPS_AT/STOPS_BEFORE with regex/lists

4

It would be very helpful to be able to have more advanced stopping conditions in STOPS_AT/STOPS_BEFORE. One use case for stopping conditions with lists instead of strings, is that: ```python...

LMQL automatically prepends the token " to the beginning of the prompt (as BOS token). However, when querying openai models through playground/api this doesn't happen in the prompt (they might...

Tokens removed by STOPS_BEFORE count towards the length of the output

2

In the following query, we would not expect the query to end on the first occurrence of "wall" (output sentence: "what did the fish say when it hit the wall?"),...