matt/ollama37
1
0
Fork 0
mirror of https://github.com/dogkeeper886/ollama37.git synced 2025-12-11 16:26:59 +00:00
Code Issues Packages Projects Releases Wiki Activity

runner.go: Fix off-by-one for num predicted

Browse Source
This commit is contained in:
Jesse Gross
2024-11-12 10:41:44 -08:00
committed by Jesse Gross
parent 636a743c2b
commit d7eb05b936
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
llama/runner/runner.go
View File

@@ -345,7 +345,7 @@ func (s *Server) processBatch(tokenBatch *llama.Batch, embedBatch *llama.Batch)
}
// if past the num predict limit
if seq.numPredict > 0 && seq.numPredicted > seq.numPredict {
if seq.numPredict > 0 && seq.numPredicted >= seq.numPredict {
s.removeSequence(seqIdx, "limit")
continue
}