[Misc] rename torch_dtype to dtype (#26695)
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
@@ -117,9 +117,8 @@ class LLM:
|
||||
execution with tensor parallelism.
|
||||
dtype: The data type for the model weights and activations. Currently,
|
||||
we support `float32`, `float16`, and `bfloat16`. If `auto`, we use
|
||||
the `torch_dtype` attribute specified in the model config file.
|
||||
However, if the `torch_dtype` in the config is `float32`, we will
|
||||
use `float16` instead.
|
||||
the `dtype` attribute of the Transformers model's config. However,
|
||||
if the `dtype` in the config is `float32`, we will use `float16` instead.
|
||||
quantization: The method used to quantize the model weights. Currently,
|
||||
we support "awq", "gptq", and "fp8" (experimental).
|
||||
If None, we first check the `quantization_config` attribute in the
|
||||
|
||||
Reference in New Issue
Block a user