server : check that the prompt fits in the slot's context #10030
+10
−1
Conversation
This file contains 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
ngxson
reviewed
Oct 24, 2024
|
|
||
| if (slot.n_prompt_tokens > slot.n_ctx) { | ||
| slot.release(); | ||
| send_error(slot, "input is larger than the max context size. skipping", ERROR_TYPE_SERVER); |
There was a problem hiding this comment.
I'd suggest changing some wordings to match the message when number of tokens is larger than n_ubatch
Suggested change
| send_error(slot, "input is larger than the max context size. skipping", ERROR_TYPE_SERVER); | |
| send_error(slot, "input is too large to process. increase the context size", ERROR_TYPE_SERVER); |
There was a problem hiding this comment.
Suggesting to increase the context size can lead to the same issue as the user experienced in #9978 due to setting a context size larger than what the model supports. So I'll leave the version without the suggestion.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
fix #9978
In embedding and reranking mode, a prompt could fit in the batch but could exceed the slot's context size. This PR adds a check to handle such cases gracefully, instead of crashing.
Testing
./llama-server \ -m ./models/bge-large-zh-v1.5/ggml-model-f16.gguf \ --port 8012 -a [email protected] -ngl 100 \ --embeddings -ub 8192 -b 8192 --pooling cls curl \ http://localhost:8012/v1/embeddings -H "Content-Type: application/json" \ -H "Authorization: Bearer no-key" \ -d '{"input": ["'"$(printf 'hello %.0s' $(seq 1 550))"'"], "encoding_format": "float"}'{ "error": { "code": 500, "message": "input is larger than the max context size. skipping", "type": "server_error" } }