Remove debug assert

2025-04-02 22:52:13 -04:00 · 2025-04-02 22:52:13 -04:00 · 502812b2b1
parent 105261d2ba
commit 502812b2b1
1 changed files with 0 additions and 2 deletions
--- a/src/llama-quant.cpp
+++ b/src/llama-quant.cpp
@ -865,8 +865,6 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::
        is_clip_model |= name.rfind("mm.", 0) == 0; // check the "mm." prefix
    }

-    GGML_ASSERT(qs.n_ffn_down_exp != 0);
-
    qs.n_ffn_down = qs.n_ffn_gate = qs.n_ffn_up = (int)model.hparams.n_layer;

    // sanity checks for models that have attention layers