onednn_verbose,info,oneDNN v2.6.0 (commit 52b5f107dd9cf10910aaa19cb47f3abf9b349815) onednn_verbose,info,cpu,runtime:OpenMP,nthr:2 onednn_verbose,info,cpu,isa:Intel AVX-512 with Intel DL Boost onednn_verbose,info,gpu,runtime:none onednn_verbose,info,prim_template:operation,engine,primitive,implementation,prop_kind,memory_descriptors,attributes,auxiliary,problem_desc,exec_time onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.157959 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.123779 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0388184 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0358887 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0378418 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0400391 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0378418 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0388184 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0378418 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.118896 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0388184 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0361328 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0400391 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0361328 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0388184 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0400391 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0410156 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0400391 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0361328 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0378418 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0478516 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0400391 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0358887 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0378418 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0388184 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0358887 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0378418 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0358887 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0358887 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0378418 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0388184 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0388184 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0361328 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0388184 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0361328 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0429688 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0400391 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0358887 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0358887 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0388184 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0378418 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0419922 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0400391 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0388184 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0498047 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0400391 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0378418 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0380859 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0397949 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0598145 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0371094 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0390625 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0368652 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0388184 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0358887 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.0400391 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.118896 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.118896 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.119873 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.12207 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.115967 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.113037 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.11499 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.116943 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.115967 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.131104 onednn_verbose,exec,cpu,eltwise,jit:avx512_core,forward_training,data_f32::blocked:abc:f0 diff_undef::undef::f0,attr-scratchpad:user ,alg:eltwise_gelu_erf alpha:0 beta:0,1x32x3072,0.120117 --------------------------- ------------ ------------ ------------ ------------ ------------ ------------ Name Self CPU % Self CPU CPU total % CPU total CPU time avg # of Calls --------------------------- ------------ ------------ ------------ ------------ ------------ ------------ aten::slice 1.30% 652.000us 1.31% 655.000us 131.000us 5 aten::as_strided 0.10% 51.000us 0.10% 51.000us 0.189us 270 aten::unsqueeze 0.01% 5.000us 0.01% 5.000us 2.500us 2 aten::to 0.08% 41.000us 0.99% 496.000us 33.067us 15 aten::_to_copy 0.81% 406.000us 0.93% 464.000us 30.933us 15 aten::empty_strided 0.04% 22.000us 0.04% 22.000us 1.467us 15 aten::copy_ 4.80% 2.409ms 4.80% 2.409ms 23.851us 101 aten::rsub 0.01% 6.000us 0.04% 22.000us 22.000us 1 aten::sub 0.02% 9.000us 0.03% 16.000us 16.000us 1 aten::mul 0.01% 7.000us 0.02% 10.000us 10.000us 1 aten::embedding 0.04% 21.000us 0.25% 123.000us 41.000us 3 aten::reshape 0.33% 167.000us 0.41% 208.000us 4.078us 51 aten::_reshape_alias 0.09% 45.000us 0.09% 45.000us 0.882us 51 aten::index_select 0.11% 55.000us 0.13% 65.000us 21.667us 3 aten::empty 0.27% 137.000us 0.27% 137.000us 1.522us 90 aten::select 0.03% 14.000us 0.03% 14.000us 2.000us 7 aten::view 0.46% 231.000us 0.46% 231.000us 0.943us 245 aten::add 1.89% 949.000us 1.89% 949.000us 25.649us 37 aten::add_ 0.02% 9.000us 0.02% 9.000us 9.000us 1 aten::layer_norm 0.14% 70.000us 2.50% 1.256ms 50.240us 25 aten::native_layer_norm 2.24% 1.122ms 2.36% 1.186ms 47.440us 25 aten::dropout 0.00% 0.000us 0.00% 0.000us 0.000us 38 aten::linear 1.16% 580.000us 83.50% 41.877ms 565.905us 74 aten::t 0.36% 179.000us 0.76% 381.000us 5.149us 74 aten::transpose 0.40% 199.000us 0.47% 236.000us 2.744us 86 aten::addmm 77.24% 38.736ms 81.31% 40.776ms 551.027us 74 aten::expand 0.50% 250.000us 0.51% 254.000us 2.082us 122 aten::resolve_conj 0.00% 0.000us 0.00% 0.000us 0.000us 196 aten::permute 0.27% 137.000us 0.28% 142.000us 2.958us 48 aten::matmul 0.48% 242.000us 2.72% 1.364ms 56.833us 24 aten::bmm 1.61% 807.000us 1.61% 807.000us 33.625us 24 aten::_unsafe_view 0.05% 24.000us 0.05% 24.000us 1.000us 24 aten::div 0.38% 193.000us 0.68% 342.000us 28.500us 12 aten::softmax 0.05% 25.000us 0.40% 199.000us 16.583us 12 aten::_softmax 0.35% 174.000us 0.35% 174.000us 14.500us 12 aten::contiguous 0.14% 72.000us 1.30% 651.000us 54.250us 12 aten::clone 0.10% 50.000us 1.26% 633.000us 52.750us 12 aten::empty_like 0.07% 34.000us 0.23% 113.000us 9.417us 12 aten::gelu 4.00% 2.007ms 4.00% 2.007ms 167.250us 12 aten::tanh 0.03% 13.000us 0.03% 13.000us 13.000us 1 --------------------------- ------------ ------------ ------------ ------------ ------------ ------------ Self CPU time total: 50.150ms Inference time: 0.05766558647155762