diff --git a/src/axolotl/monkeypatch/moe_quant.py b/src/axolotl/monkeypatch/moe_quant.py
index b16d30e13..990327be1 100644
--- a/src/axolotl/monkeypatch/moe_quant.py
+++ b/src/axolotl/monkeypatch/moe_quant.py
@@ -77,27 +77,42 @@ def patch_moe_quantization_on_load(cfg):
 
     original_set_param = transformers.core_model_loading.set_param_for_module
 
+    _first_call = [True]
+
     def _patched_set_param_for_module(model, target_name, param_value, *args, **kwargs):
+        if _first_call[0]:
+            LOG.info("MoE quant patch: set_param_for_module intercepted (first call)")
+            _first_call[0] = False
+
         original_set_param(model, target_name, param_value, *args, **kwargs)
 
         # Quantize 3D+ expert params that BnB skipped (only on CUDA).
-        if param_value.ndim >= 3 and param_value.is_cuda:
-            mod_path, _, pname = target_name.rpartition(".")
-            mod = model.get_submodule(mod_path) if mod_path else model
-            if not isinstance(mod, (bnb.nn.Linear4bit, bnb.nn.Linear8bitLt)):
-                replace_parameter_4bit(
-                    mod,
-                    pname,
-                    compress_statistics=_moe_load_state["compress_statistics"],
-                    quant_type=_moe_load_state["quant_type"],
-                )
-                torch.cuda.empty_cache()
-                _moe_load_state["count"] += 1
-                LOG.debug(
-                    "Quantized 3D expert param during loading: %s (shape %s)",
-                    target_name,
-                    param_value.shape,
-                )
+        if param_value.ndim >= 3:
+            LOG.info(
+                "MoE quant patch: 3D param %s shape=%s cuda=%s",
+                target_name,
+                param_value.shape,
+                param_value.is_cuda,
+            )
+            if param_value.is_cuda:
+                mod_path, _, pname = target_name.rpartition(".")
+                mod = model.get_submodule(mod_path) if mod_path else model
+                if not isinstance(mod, (bnb.nn.Linear4bit, bnb.nn.Linear8bitLt)):
+                    replace_parameter_4bit(
+                        mod,
+                        pname,
+                        compress_statistics=_moe_load_state["compress_statistics"],
+                        quant_type=_moe_load_state["quant_type"],
+                    )
+                    torch.cuda.empty_cache()
+                    _moe_load_state["count"] += 1
+                    LOG.info(
+                        "Quantized 3D expert param: %s "
+                        "(alloc=%.2f GiB, reserved=%.2f GiB)",
+                        target_name,
+                        torch.cuda.memory_allocated() / 1024**3,
+                        torch.cuda.memory_reserved() / 1024**3,
+                    )
 
     transformers.core_model_loading.set_param_for_module = _patched_set_param_for_module
     _moe_load_state["patched"] = True