mirror of
https://github.com/dogkeeper886/ollama37.git
synced 2025-12-10 07:46:59 +00:00
kvcache: Add check for values that fall out of sliding window cache
The sliding window cache trims entries that are outside the window for the latest token. This works when we are extending the cache, such as when the conversation continues. However, if we have a partial overlap in conversation (including the BOS tokens), then we resume from a past point in the conversation and the needed tokens are no longer stored in memory. This verifies that the new window overlaps with the old one before reusing the cache. Co-authored-by: Jesse Gross <jesse@ollama.com>
This commit is contained in:
@@ -62,6 +62,11 @@ type Cache interface {
|
||||
// CopyPrefix copies tokens in the range [0, len) from srcSeq to dstSeq
|
||||
CopyPrefix(srcSeq, dstSeq int, len int32)
|
||||
|
||||
// CanResume returns true if the cache can continue with the next token at
|
||||
// the given position and sequence. Assumes that the caller has already
|
||||
// verified the contents of the cache.
|
||||
CanResume(seq int, pos int32) bool
|
||||
|
||||
// Remove deletes tokens in the range [beginIndex, endIndex) from seq. Set
|
||||
// endIndex to math.MaxInt32 to remove everything starting at beginIndex.
|
||||
//
|
||||
|
||||
Reference in New Issue
Block a user