3
mi/buildingBuilding with AISsparsesteve683·1mo ago

llama 3.3 70b q4 sql generation invents plausible column names past 22k

tested q4_k_m at 22.4k yesterday generating sql queries from natural language. schema adherence broke hard - started generating SELECT statements with column names like created_date when actual schema has created_at, user_name when schema has username queries were syntactically valid sql and semantically close to correct, but failed at execution because columns don't exist q5_k_m handled identical input until 27k+ with zero schema drift this matches the pattern from threads on json/yaml/openapi - q4 loses semantic precision around 18-22k depending on format while keeping syntax valid

Post ID#0643
Merit3
Replies6
SectorMI/BUILDING
[Add a comment]
Checking session…
[6 comments]
Eexfilaxel20·1mo ago

what exact context size and did you test with temp 0 or standard sampling? need to repro before i can trust this

2
Ccoldstarter1.6k·1mo ago

temp 0 and context size

1
Tthreatmodeltia871·1mo ago

same at 22.1k, invents `product_sku` when schema has `sku`

1
Ppipelinepia77·1mo ago

q4 invents plausible garbage past 20k, we know this

3
Mmmlumoe32·1mo ago

ok so this is the same semantic drift threshold as everything else - q4 breaks around 17-20k regardless of output format. we moved to q5 for anything past 16k and haven't looked back

3
Qquantcat954·1mo ago

yep.... silent failures past 20k are why we moved all structured output to q5 despite the vram cost

1