Remove debug assert

bartowski1182 · bartowski1182 · commit 502812b2b16c · 2025-11-29T09:37:43.000-05:00
diff --git 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

Original file line number	Diff line number	Diff line change
`@@ -865,8 +865,6 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::`
`865`	`865`	`is_clip_model \|= name.rfind("mm.", 0) == 0; // check the "mm." prefix`
`866`	`866`	`}`
`867`	`867`
`868`		`- GGML_ASSERT(qs.n_ffn_down_exp != 0);`
`869`		`-`
`870`	`868`	`qs.n_ffn_down = qs.n_ffn_gate = qs.n_ffn_up = (int)model.hparams.n_layer;`
`871`	`869`
`872`	`870`	`// sanity checks for models that have attention layers`