convert : fix autoawq gemma (ggerganov#6704)

* fix autoawq quantized gemma model convert error using autoawq to quantize gemma model will include a lm_head.weight tensor in model-00001-of-00002.safetensors. it result in this situation that convert-hf-to-gguf.py can't map lm_head.weight. skip loading this tensor could prevent this error. * change code to full string match and print necessary message change code to full string match and print a short message to inform users that lm_head.weight has been skipped. --------- Co-authored-by: Zheng.Deng <[email protected]>
Xarbirus · Apr 16, 2024 · facb8b5 · facb8b5
1 parent 532c173
commit facb8b5
Showing 1 changed file with 6 additions and 0 deletions.
diff --git a/convert-hf-to-gguf.py b/convert-hf-to-gguf.py
@@ -2458,6 +2458,12 @@ def write_tensors(self):
         tensor_map = gguf.get_tensor_name_map(self.model_arch, block_count)
 
         for name, data_torch in self.get_tensors():
+            # lm_head is not used in llama.cpp, while autoawq will include this tensor in model
+            # To prevent errors, skip loading lm_head.weight.
+            if name == "lm_head.weight":
+                print(f"Skipping get tensor {name!r} in safetensors so that convert can end normally.")
+                continue
+
             old_dtype = data_torch.dtype
 
             # convert any unsupported data types to float32