[Misc] rename torch_dtype to dtype (#26695)

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
2025-10-15 20:11:48 +08:00
parent f93e348010
commit 8f4b313c37
30 changed files with 52 additions and 55 deletions
--- a/docs/features/quantization/torchao.md
+++ b/docs/features/quantization/torchao.md
@@ -27,7 +27,7 @@ You can quantize your own huggingface model with torchao, e.g. [transformers](ht
    quantization_config = TorchAoConfig(Int8WeightOnlyConfig())
    quantized_model = AutoModelForCausalLM.from_pretrained(
        model_name,
-        torch_dtype="auto",
+        dtype="auto",
        device_map="auto",
        quantization_config=quantization_config
    )