server : update prompt on slot restore#9800
Merged
ggerganov merged 1 commit intogg/infill-0from Oct 11, 2024
Merged
Conversation
Contributor
I don't have any strong preference either. IMO the In anyways, the current slot save/store API is quite low-level, I think we should communicate this in the docs so that user don't expect it to be a prod-ready thing. I'm looking forward to reorganize this feature to match Claude prompt caching API, which should be more intuitive for end-user. |
ggerganov
added a commit
that referenced
this pull request
Oct 11, 2024
ggerganov
added a commit
that referenced
this pull request
Oct 12, 2024
* llama : improve infill support ggml-ci * llama : add more FIM token strings ggml-ci * server : update prompt on slot restore (#9800) * gguf : deprecate old FIM token KVs
drollings
pushed a commit
to drollings/llama.cpp
that referenced
this pull request
Oct 18, 2024
…9798) * llama : improve infill support ggml-ci * llama : add more FIM token strings ggml-ci * server : update prompt on slot restore (ggml-org#9800) * gguf : deprecate old FIM token KVs
dsx1986
pushed a commit
to dsx1986/llama.cpp
that referenced
this pull request
Oct 29, 2024
…9798) * llama : improve infill support ggml-ci * llama : add more FIM token strings ggml-ci * server : update prompt on slot restore (ggml-org#9800) * gguf : deprecate old FIM token KVs
arthw
pushed a commit
to arthw/llama.cpp
that referenced
this pull request
Nov 15, 2024
…9798) * llama : improve infill support ggml-ci * llama : add more FIM token strings ggml-ci * server : update prompt on slot restore (ggml-org#9800) * gguf : deprecate old FIM token KVs
arthw
pushed a commit
to arthw/llama.cpp
that referenced
this pull request
Nov 18, 2024
…9798) * llama : improve infill support ggml-ci * llama : add more FIM token strings ggml-ci * server : update prompt on slot restore (ggml-org#9800) * gguf : deprecate old FIM token KVs
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
ref #9781
The
slot.promptfield is not being updated after restoring a slot state from a file. Since we don't store the original jsonic representation of the prompt, we simply set it to a descriptive message in order to not confuse the reported state of the slot.Another option might be to detokenize the restored tokens. Not sure if worth it.