Add support for Microsoft Phi-4 model #10817
Open
+22
−3
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
…4 model llama : use regular (not a sliding window) attention mask for Phi-4 model
slaren
reviewed
Dec 14, 2024
src/llama.cpp
Outdated
| @@ -12839,7 +12839,13 @@ struct llm_build_context { | |||
| struct ggml_tensor * inp_pos = build_inp_pos(); | |||
|
|
|||
| // KQ_mask (mask for 1 head, it will be broadcasted to all heads) | |||
| struct ggml_tensor * KQ_mask_swa = build_inp_KQ_mask_swa(); | |||
| struct ggml_tensor * KQ_mask = nullptr; | |||
| if (model.name == "Phi 4") { | |||
There was a problem hiding this comment.
I think a better solution would be to check if hparams.n_swa != 0.
There was a problem hiding this comment.
I modified my patch to explicitly store zero sliding_window in case it's null in config.json and use the zero value to distinguish Phi-4 from other PHI3-based models.
compilade
reviewed
Dec 14, 2024
convert_hf_to_gguf.py
Outdated
Comment on lines
2132
to
2133
| if self.metadata.name == "Phi 4": | ||
| return self._set_vocab_gpt2() |
There was a problem hiding this comment.
Alternatively, self._set_vocab_gpt2() could be called when tokenizer.model is missing here, regardless of the model name.
There was a problem hiding this comment.
I modified the solution to check value of tokenizer_class from tokenizer_config.json and call self._set_vocab_gpt2() if it's GPT2Tokenizer.
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Outdated suggestions cannot be applied.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.
This PR adds support for Microsoft Phi-4 model. Fixes #10814.
Current solution is to:
sliding_windowhparam if it's null. This allows the old Phi-3n_swavalidation logic to work without any changes. Ifn_swais 0 a regular KQ mask is used instead of sliding window KQ mask inbuild_phi3().A model name value from general.name ("Phi 4") was used to trigger behavior specific to Phi-4 model:1. Using GPT2 vocab during model conversion2. Ignoringsliding_windowhparam during model conversion3. Skipping sliding window length value check (n_swa == 0) inbuild_phi3()4. Creating regular KQ mask instead of sliding window KQ mask inbuild_phi3()Let me know if there is any better way to differentiate Phi 4 from other models based on PHI3 architecture.