MCPcopy
hub / github.com/NVIDIA/TensorRT-LLM / has_act_static_scaling

Method has_act_static_scaling

tensorrt_llm/quantization/mode.py:154–156  ·  view source on GitHub ↗
(self)

Source from the content-addressed store, hash-verified

152 return self._any(self.FP8_1x128_128x128)
153
154 def has_act_static_scaling(self):
155 return not self.has_per_token_dynamic_scaling(
156 ) and not self.has_fp8_rowwise()
157
158 def has_per_channel_scaling(self):
159 return self._any(self.PER_CHANNEL)

Callers 15

forwardMethod · 0.80
test_mlp_smooth_quantMethod · 0.80
_construct_executionMethod · 0.80
test_per_tokenMethod · 0.80
__init__Method · 0.80
forwardMethod · 0.80
__init__Method · 0.80
forwardMethod · 0.80
__init__Method · 0.80
forwardMethod · 0.80
forwardMethod · 0.80

Calls 2

has_fp8_rowwiseMethod · 0.95

Tested by 5

forwardMethod · 0.64
test_mlp_smooth_quantMethod · 0.64
_construct_executionMethod · 0.64
test_per_tokenMethod · 0.64