Skip to content

feat(quant): add constant_amax to pin activation input_scale (NVFP4 experts input_scale=1.0) #10186

feat(quant): add constant_amax to pin activation input_scale (NVFP4 experts input_scale=1.0)

feat(quant): add constant_amax to pin activation input_scale (NVFP4 experts input_scale=1.0) #10186

Job log options

This job was skipped