MCPcopy Index your code
hub / github.com/NVIDIA/TensorRT-LLM / bitwidth

Function bitwidth

triton_kernels/tensor.py:94–97  ·  view source on GitHub ↗
(type: IntegerType | FloatType | torch.dtype)

Source from the content-addressed store, hash-verified

92
93
94def bitwidth(type: IntegerType | FloatType | torch.dtype):
95 if isinstance(type, torch.dtype):
96 return type.itemsize * 8
97 return type.bitwidth
98
99
100@dataclass

Callers 7

compute_block_kFunction · 0.90
compute_num_stagesFunction · 0.90
compute_block_nkFunction · 0.90
matmul_ogsFunction · 0.85
__post_init__Method · 0.85
element_sizeMethod · 0.85
wrap_torch_tensorFunction · 0.85

Calls

no outgoing calls

Tested by

no test coverage detected