# Re-verification on the MERGED tip (2f03f1efa: pergroup + umbrella d35f2625f/b39805b41), local 4090, kernel from arbi_serve/weight_quant/exl3/csrc via int8_kernel._load, I8_ONLY_BITS=4 I8_ONLY_CB=2

## shape_probe + bit identity
shape thr regs smem blk/SM local
s27 fold0 thr=128 regs=199 smem=49152 blk/SM=2 warps/SM=8 spill_local=0
s27 fold1 thr=128 regs=199 smem=49152 blk/SM=2 warps/SM=8 spill_local=0
s24 fold0 thr=256 regs=123 smem=49152 blk/SM=2 warps/SM=16 spill_local=0
s24 fold1 thr=256 regs=123 smem=49152 blk/SM=2 warps/SM=16 spill_local=0
s80 fold0 thr=256 regs=222 smem=49152 blk/SM=1 warps/SM=8 spill_local=0
s80 fold1 thr=256 regs=224 smem=49152 blk/SM=1 warps/SM=8 spill_local=0
s81 fold0 thr=128 regs=255 smem=49152 blk/SM=2 warps/SM=8 spill_local=344
s81 fold1 thr=128 regs=255 smem=49152 blk/SM=2 warps/SM=8 spill_local=344
s110 fold0 thr=128 regs=254 smem=49152 blk/SM=2 warps/SM=8 spill_local=16
s110 fold1 thr=128 regs=254 smem=49152 blk/SM=2 warps/SM=8 spill_local=16
s111 fold0 thr=128 regs=255 smem=50176 blk/SM=2 warps/SM=8 spill_local=0
s111 fold1 thr=128 regs=255 smem=50176 blk/SM=2 warps/SM=8 spill_local=0
out_proj M=48: 110==27 True  111==80 True  rel27=1.2590e-02 rel110=1.2590e-02 rel80=1.0890e-02 rel111=1.0890e-02  max|111-80|=0.000e+00
out_proj M=2048: 110==27 True  111==80 True  rel27=1.2478e-02 rel110=1.2478e-02 rel80=1.0827e-02 rel111=1.0827e-02  max|111-80|=0.000e+00
gate_proj M=48: 110==27 True  111==80 True  rel27=1.2446e-02 rel110=1.2446e-02 rel80=1.0808e-02 rel111=1.0808e-02  max|111-80|=0.000e+00
gate_proj M=2048: 110==27 True  111==80 True  rel27=1.2306e-02 rel110=1.2306e-02 rel80=1.0771e-02 rel111=1.0771e-02  max|111-80|=0.000e+00
down_proj M=48: 110==27 True  111==80 True  rel27=1.2663e-02 rel110=1.2663e-02 rel80=1.0680e-02 rel111=1.0680e-02  max|111-80|=0.000e+00
down_proj M=2048: 110==27 True  111==80 True  rel27=1.2772e-02 rel110=1.2772e-02 rel80=1.0739e-02 rel111=1.0739e-02  max|111-80|=0.000e+00

## gate7 --grouped 111 --ungrouped 110

out_proj  k=6144 n=5120 groups=48
  M=48    vs-fp32 per-token 1.2590e-02  grouped 1.0890e-02  null 1.2590e-02  || rev 4.2571e-02 roll 4.2805e-02
        vs-torch-quantiser per-token 8.7509e-03 grouped 8.7546e-03  || null-vs-ungrouped 8.577e-06  grouped-vs-null 1.099e-02  PASS
  M=64    vs-fp32 per-token 1.2398e-02  grouped 1.0783e-02  null 1.2398e-02  || rev 4.1245e-02 roll 4.0806e-02
        vs-torch-quantiser per-token 8.6553e-03 grouped 8.6581e-03  || null-vs-ungrouped 9.088e-06  grouped-vs-null 1.078e-02  PASS
  M=256   vs-fp32 per-token 1.2457e-02  grouped 1.0811e-02  null 1.2457e-02  || rev 4.1308e-02 roll 4.1139e-02
        vs-torch-quantiser per-token 8.6785e-03 grouped 8.6767e-03  || null-vs-ungrouped 8.427e-06  grouped-vs-null 1.089e-02  PASS
  M=2048  vs-fp32 per-token 1.2485e-02  grouped 1.0835e-02  null 1.2485e-02  || rev 4.1125e-02 roll 4.1145e-02
        vs-torch-quantiser per-token 8.6946e-03 grouped 8.6947e-03  || null-vs-ungrouped 7.796e-06  grouped-vs-null 1.090e-02  PASS

gate_proj  k=5120 n=17408 groups=40
  M=48    vs-fp32 per-token 1.2173e-02  grouped 1.0726e-02  null 1.2173e-02  || rev 3.8863e-02 roll 3.9505e-02
        vs-torch-quantiser per-token 8.6111e-03 grouped 8.6109e-03  || null-vs-ungrouped 7.255e-06  grouped-vs-null 1.055e-02  PASS
  M=64    vs-fp32 per-token 1.2394e-02  grouped 1.0799e-02  null 1.2394e-02  || rev 4.2407e-02 roll 4.1591e-02
        vs-torch-quantiser per-token 8.6426e-03 grouped 8.6413e-03  || null-vs-ungrouped 6.506e-06  grouped-vs-null 1.080e-02  PASS
  M=256   vs-fp32 per-token 1.2330e-02  grouped 1.0785e-02  null 1.2330e-02  || rev 4.0114e-02 roll 4.0550e-02
        vs-torch-quantiser per-token 8.6348e-03 grouped 8.6354e-03  || null-vs-ungrouped 7.858e-06  grouped-vs-null 1.076e-02  PASS
  M=2048  vs-fp32 per-token 1.2303e-02  grouped 1.0776e-02  null 1.2303e-02  || rev 4.0395e-02 roll 4.0435e-02
        vs-torch-quantiser per-token 8.6210e-03 grouped 8.6210e-03  || null-vs-ungrouped 7.880e-06  grouped-vs-null 1.073e-02  PASS

down_proj  k=17408 n=5120 groups=136
  M=48    vs-fp32 per-token 1.2728e-02  grouped 1.0717e-02  null 1.2728e-02  || rev 4.0214e-02 roll 3.9783e-02
        vs-torch-quantiser per-token 8.5688e-03 grouped 8.5690e-03  || null-vs-ungrouped 1.045e-05  grouped-vs-null 1.139e-02  PASS
  M=64    vs-fp32 per-token 1.2937e-02  grouped 1.0739e-02  null 1.2937e-02  || rev 4.1490e-02 roll 4.0741e-02
        vs-torch-quantiser per-token 8.5900e-03 grouped 8.5900e-03  || null-vs-ungrouped 1.017e-05  grouped-vs-null 1.159e-02  PASS
  M=256   vs-fp32 per-token 1.2760e-02  grouped 1.0742e-02  null 1.2760e-02  || rev 4.0876e-02 roll 4.0475e-02
        vs-torch-quantiser per-token 8.5872e-03 grouped 8.5854e-03  || null-vs-ungrouped 1.127e-05  grouped-vs-null 1.137e-02  PASS
  M=2048  vs-fp32 per-token 1.2782e-02  grouped 1.0736e-02  null 1.2782e-02  || rev 4.0605e-02 roll 4.0284e-02
        vs-torch-quantiser per-token 8.5808e-03 grouped 8.5803e-03  || null-vs-ungrouped 1.039e-05  grouped-vs-null 1.142e-02  PASS

  biased drain vs the I2F drain it replaces (must be bit-identical):
    s85 vs s96  out_proj   M=48    torch.equal=True  max|diff|=0.000e+00
    s85 vs s96  out_proj   M=2048  torch.equal=True  max|diff|=0.000e+00
    s85 vs s96  gate_proj  M=48    torch.equal=True  max|diff|=0.000e+00
    s85 vs s96  gate_proj  M=2048  torch.equal=True  max|diff|=0.000e+00
    s80 vs s97  out_proj   M=48    torch.equal=True  max|diff|=0.000e+00
    s80 vs s97  out_proj   M=2048  torch.equal=True  max|diff|=0.000e+00
    s80 vs s97  gate_proj  M=48    torch.equal=True  max|diff|=0.000e+00
    s80 vs s97  gate_proj  M=2048  torch.equal=True  max|diff|=0.000e+00
  BIASED DRAIN: PASS

  REFUSAL (2-D table on an ungrouped shape): PASS -- RuntimeError

GATE 7 (per-128-group activation scale): PASS

## sweep_shapes
  out_proj   k=6144   n=5120   M=1     s0=1.33e-02 s1=1.33e-02 s2=1.33e-02 s3=1.33e-02 s4=1.33e-02 s5=1.33e-02 s6=1.33e-02 s7=1.33e-02 s8=1.33e-02 s9=1.33e-02 s10=1.33e-02 s11=1.33e-02 s12=1.33e-02 s13=1.33e-02 s20=1.33e-02 s21=1.33e-02 s22=1.33e-02 s23=1.33e-02 s24=1.33e-02 s25=1.33e-02 s26=1.33e-02 s27=1.33e-02 s30=1.33e-02 s31=1.33e-02 s32=1.33e-02 s33=1.33e-02 s34=1.33e-02 s35=1.33e-02 s36=1.33e-02 s37=1.33e-02 s50=1.33e-02 s51=1.33e-02 s52=1.33e-02 s53=1.33e-02 s54=1.33e-02 s60=1.33e-02 s61=1.33e-02 s62=1.33e-02 s63=1.33e-02 s64=1.33e-02 s65=1.33e-02 s66=1.33e-02 s67=1.33e-02 s68=1.33e-02 s70=1.33e-02 s71=1.33e-02 s72=1.33e-02 s73=1.33e-02 s80=1.10e-02 s81=1.10e-02 s82=1.10e-02 s83=1.10e-02 s84=1.33e-02 s85=1.10e-02 s86=1.33e-02 s87=1.13e-02 s88=1.16e-02 s89=1.13e-02 s90=1.10e-02 s91=1.33e-02 s92=1.10e-02 s93=1.33e-02 s94=1.10e-02 s95=1.33e-02 s96=1.10e-02 s97=1.10e-02 s98=1.10e-02 s99=1.33e-02 s101=1.13e-02 s102=1.13e-02 s103=1.13e-02 s104=1.33e-02 s105=1.13e-02 s106=1.33e-02 s107=1.13e-02 s108=1.33e-02 s110=1.33e-02 s111=1.10e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  out_proj   k=6144   n=5120   M=17    s0=1.22e-02 s1=1.22e-02 s2=1.22e-02 s3=1.22e-02 s4=1.22e-02 s5=1.22e-02 s6=1.22e-02 s7=1.22e-02 s8=1.22e-02 s9=1.22e-02 s10=1.22e-02 s11=1.22e-02 s12=1.22e-02 s13=1.22e-02 s20=1.22e-02 s21=1.22e-02 s22=1.22e-02 s23=1.22e-02 s24=1.22e-02 s25=1.22e-02 s26=1.22e-02 s27=1.22e-02 s30=1.22e-02 s31=1.22e-02 s32=1.22e-02 s33=1.22e-02 s34=1.22e-02 s35=1.22e-02 s36=1.22e-02 s37=1.22e-02 s50=1.22e-02 s51=1.22e-02 s52=1.22e-02 s53=1.22e-02 s54=1.22e-02 s60=1.22e-02 s61=1.22e-02 s62=1.22e-02 s63=1.22e-02 s64=1.22e-02 s65=1.22e-02 s66=1.22e-02 s67=1.22e-02 s68=1.22e-02 s70=1.22e-02 s71=1.22e-02 s72=1.22e-02 s73=1.22e-02 s80=1.07e-02 s81=1.07e-02 s82=1.07e-02 s83=1.07e-02 s84=1.22e-02 s85=1.07e-02 s86=1.22e-02 s87=1.11e-02 s88=1.13e-02 s89=1.11e-02 s90=1.07e-02 s91=1.22e-02 s92=1.07e-02 s93=1.22e-02 s94=1.07e-02 s95=1.22e-02 s96=1.07e-02 s97=1.07e-02 s98=1.07e-02 s99=1.22e-02 s101=1.11e-02 s102=1.11e-02 s103=1.11e-02 s104=1.22e-02 s105=1.11e-02 s106=1.22e-02 s107=1.11e-02 s108=1.22e-02 s110=1.22e-02 s111=1.07e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  out_proj   k=6144   n=5120   M=63    s0=1.26e-02 s1=1.26e-02 s2=1.26e-02 s3=1.26e-02 s4=1.26e-02 s5=1.26e-02 s6=1.26e-02 s7=1.26e-02 s8=1.26e-02 s9=1.26e-02 s10=1.26e-02 s11=1.26e-02 s12=1.26e-02 s13=1.26e-02 s20=1.26e-02 s21=1.26e-02 s22=1.26e-02 s23=1.26e-02 s24=1.26e-02 s25=1.26e-02 s26=1.26e-02 s27=1.26e-02 s30=1.26e-02 s31=1.26e-02 s32=1.26e-02 s33=1.26e-02 s34=1.26e-02 s35=1.26e-02 s36=1.26e-02 s37=1.26e-02 s50=1.26e-02 s51=1.26e-02 s52=1.26e-02 s53=1.26e-02 s54=1.26e-02 s60=1.26e-02 s61=1.26e-02 s62=1.26e-02 s63=1.26e-02 s64=1.26e-02 s65=1.26e-02 s66=1.26e-02 s67=1.26e-02 s68=1.26e-02 s70=1.26e-02 s71=1.26e-02 s72=1.26e-02 s73=1.26e-02 s80=1.09e-02 s81=1.09e-02 s82=1.09e-02 s83=1.09e-02 s84=1.26e-02 s85=1.09e-02 s86=1.26e-02 s87=1.11e-02 s88=1.15e-02 s89=1.11e-02 s90=1.09e-02 s91=1.26e-02 s92=1.09e-02 s93=1.26e-02 s94=1.09e-02 s95=1.26e-02 s96=1.09e-02 s97=1.09e-02 s98=1.09e-02 s99=1.26e-02 s101=1.11e-02 s102=1.11e-02 s103=1.11e-02 s104=1.26e-02 s105=1.11e-02 s106=1.26e-02 s107=1.11e-02 s108=1.26e-02 s110=1.26e-02 s111=1.09e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  out_proj   k=6144   n=5120   M=256   s0=1.24e-02 s1=1.24e-02 s2=1.24e-02 s3=1.24e-02 s4=1.24e-02 s5=1.24e-02 s6=1.24e-02 s7=1.24e-02 s8=1.24e-02 s9=1.24e-02 s10=1.24e-02 s11=1.24e-02 s12=1.24e-02 s13=1.24e-02 s20=1.24e-02 s21=1.24e-02 s22=1.24e-02 s23=1.24e-02 s24=1.24e-02 s25=1.24e-02 s26=1.24e-02 s27=1.24e-02 s30=1.24e-02 s31=1.24e-02 s32=1.24e-02 s33=1.24e-02 s34=1.24e-02 s35=1.24e-02 s36=1.24e-02 s37=1.24e-02 s50=1.24e-02 s51=1.24e-02 s52=1.24e-02 s53=1.24e-02 s54=1.24e-02 s60=1.24e-02 s61=1.24e-02 s62=1.24e-02 s63=1.24e-02 s64=1.24e-02 s65=1.24e-02 s66=1.24e-02 s67=1.24e-02 s68=1.24e-02 s70=1.24e-02 s71=1.24e-02 s72=1.24e-02 s73=1.24e-02 s80=1.08e-02 s81=1.08e-02 s82=1.08e-02 s83=1.08e-02 s84=1.24e-02 s85=1.08e-02 s86=1.24e-02 s87=1.11e-02 s88=1.14e-02 s89=1.11e-02 s90=1.08e-02 s91=1.24e-02 s92=1.08e-02 s93=1.24e-02 s94=1.08e-02 s95=1.24e-02 s96=1.08e-02 s97=1.08e-02 s98=1.08e-02 s99=1.24e-02 s101=1.11e-02 s102=1.11e-02 s103=1.11e-02 s104=1.24e-02 s105=1.11e-02 s106=1.24e-02 s107=1.11e-02 s108=1.24e-02 s110=1.24e-02 s111=1.08e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  out_proj   k=6144   n=5120   M=1000  s0=1.25e-02 s1=1.25e-02 s2=1.25e-02 s3=1.25e-02 s4=1.25e-02 s5=1.25e-02 s6=1.25e-02 s7=1.25e-02 s8=1.25e-02 s9=1.25e-02 s10=1.25e-02 s11=1.25e-02 s12=1.25e-02 s13=1.25e-02 s20=1.25e-02 s21=1.25e-02 s22=1.25e-02 s23=1.25e-02 s24=1.25e-02 s25=1.25e-02 s26=1.25e-02 s27=1.25e-02 s30=1.25e-02 s31=1.25e-02 s32=1.25e-02 s33=1.25e-02 s34=1.25e-02 s35=1.25e-02 s36=1.25e-02 s37=1.25e-02 s50=1.25e-02 s51=1.25e-02 s52=1.25e-02 s53=1.25e-02 s54=1.25e-02 s60=1.25e-02 s61=1.25e-02 s62=1.25e-02 s63=1.25e-02 s64=1.25e-02 s65=1.25e-02 s66=1.25e-02 s67=1.25e-02 s68=1.25e-02 s70=1.25e-02 s71=1.25e-02 s72=1.25e-02 s73=1.25e-02 s80=1.08e-02 s81=1.08e-02 s82=1.08e-02 s83=1.08e-02 s84=1.25e-02 s85=1.08e-02 s86=1.25e-02 s87=1.11e-02 s88=1.15e-02 s89=1.11e-02 s90=1.08e-02 s91=1.25e-02 s92=1.08e-02 s93=1.25e-02 s94=1.08e-02 s95=1.25e-02 s96=1.08e-02 s97=1.08e-02 s98=1.08e-02 s99=1.25e-02 s101=1.11e-02 s102=1.11e-02 s103=1.11e-02 s104=1.25e-02 s105=1.11e-02 s106=1.25e-02 s107=1.11e-02 s108=1.25e-02 s110=1.25e-02 s111=1.08e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  out_proj   k=6144   n=5120   M=2048  s0=1.25e-02 s1=1.25e-02 s2=1.25e-02 s3=1.25e-02 s4=1.25e-02 s5=1.25e-02 s6=1.25e-02 s7=1.25e-02 s8=1.25e-02 s9=1.25e-02 s10=1.25e-02 s11=1.25e-02 s12=1.25e-02 s13=1.25e-02 s20=1.25e-02 s21=1.25e-02 s22=1.25e-02 s23=1.25e-02 s24=1.25e-02 s25=1.25e-02 s26=1.25e-02 s27=1.25e-02 s30=1.25e-02 s31=1.25e-02 s32=1.25e-02 s33=1.25e-02 s34=1.25e-02 s35=1.25e-02 s36=1.25e-02 s37=1.25e-02 s50=1.25e-02 s51=1.25e-02 s52=1.25e-02 s53=1.25e-02 s54=1.25e-02 s60=1.25e-02 s61=1.25e-02 s62=1.25e-02 s63=1.25e-02 s64=1.25e-02 s65=1.25e-02 s66=1.25e-02 s67=1.25e-02 s68=1.25e-02 s70=1.25e-02 s71=1.25e-02 s72=1.25e-02 s73=1.25e-02 s80=1.08e-02 s81=1.08e-02 s82=1.08e-02 s83=1.08e-02 s84=1.25e-02 s85=1.08e-02 s86=1.25e-02 s87=1.11e-02 s88=1.14e-02 s89=1.11e-02 s90=1.08e-02 s91=1.25e-02 s92=1.08e-02 s93=1.25e-02 s94=1.08e-02 s95=1.25e-02 s96=1.08e-02 s97=1.08e-02 s98=1.08e-02 s99=1.25e-02 s101=1.11e-02 s102=1.11e-02 s103=1.11e-02 s104=1.25e-02 s105=1.11e-02 s106=1.25e-02 s107=1.11e-02 s108=1.25e-02 s110=1.25e-02 s111=1.08e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  out_proj   k=6144   n=5120   M=8192  s0=1.25e-02 s1=1.25e-02 s2=1.25e-02 s3=1.25e-02 s4=1.25e-02 s5=1.25e-02 s6=1.25e-02 s7=1.25e-02 s8=1.25e-02 s9=1.25e-02 s10=1.25e-02 s11=1.25e-02 s12=1.25e-02 s13=1.25e-02 s20=1.25e-02 s21=1.25e-02 s22=1.25e-02 s23=1.25e-02 s24=1.25e-02 s25=1.25e-02 s26=1.25e-02 s27=1.25e-02 s30=1.25e-02 s31=1.25e-02 s32=1.25e-02 s33=1.25e-02 s34=1.25e-02 s35=1.25e-02 s36=1.25e-02 s37=1.25e-02 s50=1.25e-02 s51=1.25e-02 s52=1.25e-02 s53=1.25e-02 s54=1.25e-02 s60=1.25e-02 s61=1.25e-02 s62=1.25e-02 s63=1.25e-02 s64=1.25e-02 s65=1.25e-02 s66=1.25e-02 s67=1.25e-02 s68=1.25e-02 s70=1.25e-02 s71=1.25e-02 s72=1.25e-02 s73=1.25e-02 s80=1.08e-02 s81=1.08e-02 s82=1.08e-02 s83=1.08e-02 s84=1.25e-02 s85=1.08e-02 s86=1.25e-02 s87=1.11e-02 s88=1.14e-02 s89=1.11e-02 s90=1.08e-02 s91=1.25e-02 s92=1.08e-02 s93=1.25e-02 s94=1.08e-02 s95=1.25e-02 s96=1.08e-02 s97=1.08e-02 s98=1.08e-02 s99=1.25e-02 s101=1.11e-02 s102=1.11e-02 s103=1.11e-02 s104=1.25e-02 s105=1.11e-02 s106=1.25e-02 s107=1.11e-02 s108=1.25e-02 s110=1.25e-02 s111=1.08e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  gate_proj  k=5120   n=17408  M=1     s0=1.25e-02 s1=1.25e-02 s2=1.25e-02 s3=1.25e-02 s4=1.25e-02 s5=1.25e-02 s6=1.25e-02 s7=1.25e-02 s8=1.25e-02 s9=1.25e-02 s10=1.25e-02 s11=1.25e-02 s12=1.25e-02 s13=1.25e-02 s20=1.25e-02 s21=1.25e-02 s22=1.25e-02 s23=1.25e-02 s24=1.25e-02 s25=1.25e-02 s26=1.25e-02 s27=1.25e-02 s30=1.25e-02 s31=1.25e-02 s32=1.25e-02 s33=1.25e-02 s34=1.25e-02 s35=1.25e-02 s36=1.25e-02 s37=1.25e-02 s50=1.25e-02 s51=1.25e-02 s52=1.25e-02 s53=1.25e-02 s54=1.25e-02 s60=1.25e-02 s61=1.25e-02 s62=1.25e-02 s63=1.25e-02 s64=1.25e-02 s65=1.25e-02 s66=1.25e-02 s67=1.25e-02 s68=1.25e-02 s70=1.25e-02 s71=1.25e-02 s72=1.25e-02 s73=1.25e-02 s80=1.05e-02 s81=1.05e-02 s82=1.05e-02 s83=1.05e-02 s84=1.25e-02 s85=1.05e-02 s86=1.25e-02 s87=1.10e-02 s88=1.13e-02 s89=1.10e-02 s90=1.05e-02 s91=1.25e-02 s92=1.05e-02 s93=1.25e-02 s94=1.05e-02 s95=1.25e-02 s96=1.05e-02 s97=1.05e-02 s98=1.05e-02 s99=1.25e-02 s101=1.10e-02 s102=1.10e-02 s103=1.10e-02 s104=1.25e-02 s105=1.10e-02 s106=1.25e-02 s107=1.10e-02 s108=1.25e-02 s110=1.25e-02 s111=1.05e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  gate_proj  k=5120   n=17408  M=17    s0=1.24e-02 s1=1.24e-02 s2=1.24e-02 s3=1.24e-02 s4=1.24e-02 s5=1.24e-02 s6=1.24e-02 s7=1.24e-02 s8=1.24e-02 s9=1.24e-02 s10=1.24e-02 s11=1.24e-02 s12=1.24e-02 s13=1.24e-02 s20=1.24e-02 s21=1.24e-02 s22=1.24e-02 s23=1.24e-02 s24=1.24e-02 s25=1.24e-02 s26=1.24e-02 s27=1.24e-02 s30=1.24e-02 s31=1.24e-02 s32=1.24e-02 s33=1.24e-02 s34=1.24e-02 s35=1.24e-02 s36=1.24e-02 s37=1.24e-02 s50=1.24e-02 s51=1.24e-02 s52=1.24e-02 s53=1.24e-02 s54=1.24e-02 s60=1.24e-02 s61=1.24e-02 s62=1.24e-02 s63=1.24e-02 s64=1.24e-02 s65=1.24e-02 s66=1.24e-02 s67=1.24e-02 s68=1.24e-02 s70=1.24e-02 s71=1.24e-02 s72=1.24e-02 s73=1.24e-02 s80=1.08e-02 s81=1.08e-02 s82=1.08e-02 s83=1.08e-02 s84=1.24e-02 s85=1.08e-02 s86=1.24e-02 s87=1.12e-02 s88=1.14e-02 s89=1.12e-02 s90=1.08e-02 s91=1.24e-02 s92=1.08e-02 s93=1.24e-02 s94=1.08e-02 s95=1.24e-02 s96=1.08e-02 s97=1.08e-02 s98=1.08e-02 s99=1.24e-02 s101=1.12e-02 s102=1.12e-02 s103=1.12e-02 s104=1.24e-02 s105=1.12e-02 s106=1.24e-02 s107=1.12e-02 s108=1.24e-02 s110=1.24e-02 s111=1.08e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  gate_proj  k=5120   n=17408  M=63    s0=1.23e-02 s1=1.23e-02 s2=1.23e-02 s3=1.23e-02 s4=1.23e-02 s5=1.23e-02 s6=1.23e-02 s7=1.23e-02 s8=1.23e-02 s9=1.23e-02 s10=1.23e-02 s11=1.23e-02 s12=1.23e-02 s13=1.23e-02 s20=1.23e-02 s21=1.23e-02 s22=1.23e-02 s23=1.23e-02 s24=1.23e-02 s25=1.23e-02 s26=1.23e-02 s27=1.23e-02 s30=1.23e-02 s31=1.23e-02 s32=1.23e-02 s33=1.23e-02 s34=1.23e-02 s35=1.23e-02 s36=1.23e-02 s37=1.23e-02 s50=1.23e-02 s51=1.23e-02 s52=1.23e-02 s53=1.23e-02 s54=1.23e-02 s60=1.23e-02 s61=1.23e-02 s62=1.23e-02 s63=1.23e-02 s64=1.23e-02 s65=1.23e-02 s66=1.23e-02 s67=1.23e-02 s68=1.23e-02 s70=1.23e-02 s71=1.23e-02 s72=1.23e-02 s73=1.23e-02 s80=1.08e-02 s81=1.08e-02 s82=1.08e-02 s83=1.08e-02 s84=1.23e-02 s85=1.08e-02 s86=1.23e-02 s87=1.11e-02 s88=1.14e-02 s89=1.11e-02 s90=1.08e-02 s91=1.23e-02 s92=1.08e-02 s93=1.23e-02 s94=1.08e-02 s95=1.23e-02 s96=1.08e-02 s97=1.08e-02 s98=1.08e-02 s99=1.23e-02 s101=1.11e-02 s102=1.11e-02 s103=1.11e-02 s104=1.23e-02 s105=1.11e-02 s106=1.23e-02 s107=1.11e-02 s108=1.23e-02 s110=1.23e-02 s111=1.08e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  gate_proj  k=5120   n=17408  M=256   s0=1.22e-02 s1=1.22e-02 s2=1.22e-02 s3=1.22e-02 s4=1.22e-02 s5=1.22e-02 s6=1.22e-02 s7=1.22e-02 s8=1.22e-02 s9=1.22e-02 s10=1.22e-02 s11=1.22e-02 s12=1.22e-02 s13=1.22e-02 s20=1.22e-02 s21=1.22e-02 s22=1.22e-02 s23=1.22e-02 s24=1.22e-02 s25=1.22e-02 s26=1.22e-02 s27=1.22e-02 s30=1.22e-02 s31=1.22e-02 s32=1.22e-02 s33=1.22e-02 s34=1.22e-02 s35=1.22e-02 s36=1.22e-02 s37=1.22e-02 s50=1.22e-02 s51=1.22e-02 s52=1.22e-02 s53=1.22e-02 s54=1.22e-02 s60=1.22e-02 s61=1.22e-02 s62=1.22e-02 s63=1.22e-02 s64=1.22e-02 s65=1.22e-02 s66=1.22e-02 s67=1.22e-02 s68=1.22e-02 s70=1.22e-02 s71=1.22e-02 s72=1.22e-02 s73=1.22e-02 s80=1.08e-02 s81=1.08e-02 s82=1.08e-02 s83=1.08e-02 s84=1.22e-02 s85=1.08e-02 s86=1.22e-02 s87=1.11e-02 s88=1.14e-02 s89=1.11e-02 s90=1.08e-02 s91=1.22e-02 s92=1.08e-02 s93=1.22e-02 s94=1.08e-02 s95=1.22e-02 s96=1.08e-02 s97=1.08e-02 s98=1.08e-02 s99=1.22e-02 s101=1.11e-02 s102=1.11e-02 s103=1.11e-02 s104=1.22e-02 s105=1.11e-02 s106=1.22e-02 s107=1.11e-02 s108=1.22e-02 s110=1.22e-02 s111=1.08e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  gate_proj  k=5120   n=17408  M=1000  s0=1.23e-02 s1=1.23e-02 s2=1.23e-02 s3=1.23e-02 s4=1.23e-02 s5=1.23e-02 s6=1.23e-02 s7=1.23e-02 s8=1.23e-02 s9=1.23e-02 s10=1.23e-02 s11=1.23e-02 s12=1.23e-02 s13=1.23e-02 s20=1.23e-02 s21=1.23e-02 s22=1.23e-02 s23=1.23e-02 s24=1.23e-02 s25=1.23e-02 s26=1.23e-02 s27=1.23e-02 s30=1.23e-02 s31=1.23e-02 s32=1.23e-02 s33=1.23e-02 s34=1.23e-02 s35=1.23e-02 s36=1.23e-02 s37=1.23e-02 s50=1.23e-02 s51=1.23e-02 s52=1.23e-02 s53=1.23e-02 s54=1.23e-02 s60=1.23e-02 s61=1.23e-02 s62=1.23e-02 s63=1.23e-02 s64=1.23e-02 s65=1.23e-02 s66=1.23e-02 s67=1.23e-02 s68=1.23e-02 s70=1.23e-02 s71=1.23e-02 s72=1.23e-02 s73=1.23e-02 s80=1.08e-02 s81=1.08e-02 s82=1.08e-02 s83=1.08e-02 s84=1.23e-02 s85=1.08e-02 s86=1.23e-02 s87=1.11e-02 s88=1.14e-02 s89=1.11e-02 s90=1.08e-02 s91=1.23e-02 s92=1.08e-02 s93=1.23e-02 s94=1.08e-02 s95=1.23e-02 s96=1.08e-02 s97=1.08e-02 s98=1.08e-02 s99=1.23e-02 s101=1.11e-02 s102=1.11e-02 s103=1.11e-02 s104=1.23e-02 s105=1.11e-02 s106=1.23e-02 s107=1.11e-02 s108=1.23e-02 s110=1.23e-02 s111=1.08e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  gate_proj  k=5120   n=17408  M=2048  s0=1.23e-02 s1=1.23e-02 s2=1.23e-02 s3=1.23e-02 s4=1.23e-02 s5=1.23e-02 s6=1.23e-02 s7=1.23e-02 s8=1.23e-02 s9=1.23e-02 s10=1.23e-02 s11=1.23e-02 s12=1.23e-02 s13=1.23e-02 s20=1.23e-02 s21=1.23e-02 s22=1.23e-02 s23=1.23e-02 s24=1.23e-02 s25=1.23e-02 s26=1.23e-02 s27=1.23e-02 s30=1.23e-02 s31=1.23e-02 s32=1.23e-02 s33=1.23e-02 s34=1.23e-02 s35=1.23e-02 s36=1.23e-02 s37=1.23e-02 s50=1.23e-02 s51=1.23e-02 s52=1.23e-02 s53=1.23e-02 s54=1.23e-02 s60=1.23e-02 s61=1.23e-02 s62=1.23e-02 s63=1.23e-02 s64=1.23e-02 s65=1.23e-02 s66=1.23e-02 s67=1.23e-02 s68=1.23e-02 s70=1.23e-02 s71=1.23e-02 s72=1.23e-02 s73=1.23e-02 s80=1.08e-02 s81=1.08e-02 s82=1.08e-02 s83=1.08e-02 s84=1.23e-02 s85=1.08e-02 s86=1.23e-02 s87=1.11e-02 s88=1.14e-02 s89=1.11e-02 s90=1.08e-02 s91=1.23e-02 s92=1.08e-02 s93=1.23e-02 s94=1.08e-02 s95=1.23e-02 s96=1.08e-02 s97=1.08e-02 s98=1.08e-02 s99=1.23e-02 s101=1.11e-02 s102=1.11e-02 s103=1.11e-02 s104=1.23e-02 s105=1.11e-02 s106=1.23e-02 s107=1.11e-02 s108=1.23e-02 s110=1.23e-02 s111=1.08e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
  gate_proj  k=5120   n=17408  M=8192  s0=1.23e-02 s1=1.23e-02 s2=1.23e-02 s3=1.23e-02 s4=1.23e-02 s5=1.23e-02 s6=1.23e-02 s7=1.23e-02 s8=1.23e-02 s9=1.23e-02 s10=1.23e-02 s11=1.23e-02 s12=1.23e-02 s13=1.23e-02 s20=1.23e-02 s21=1.23e-02 s22=1.23e-02 s23=1.23e-02 s24=1.23e-02 s25=1.23e-02 s26=1.23e-02 s27=1.23e-02 s30=1.23e-02 s31=1.23e-02 s32=1.23e-02 s33=1.23e-02 s34=1.23e-02 s35=1.23e-02 s36=1.23e-02 s37=1.23e-02 s50=1.23e-02 s51=1.23e-02 s52=1.23e-02 s53=1.23e-02 s54=1.23e-02 s60=1.23e-02 s61=1.23e-02 s62=1.23e-02 s63=1.23e-02 s64=1.23e-02 s65=1.23e-02 s66=1.23e-02 s67=1.23e-02 s68=1.23e-02 s70=1.23e-02 s71=1.23e-02 s72=1.23e-02 s73=1.23e-02 s80=1.08e-02 s81=1.08e-02 s82=1.08e-02 s83=1.08e-02 s84=1.23e-02 s85=1.08e-02 s86=1.23e-02 s87=1.11e-02 s88=1.14e-02 s89=1.11e-02 s90=1.08e-02 s91=1.23e-02 s92=1.08e-02 s93=1.23e-02 s94=1.08e-02 s95=1.23e-02 s96=1.08e-02 s97=1.08e-02 s98=1.08e-02 s99=1.23e-02 s101=1.11e-02 s102=1.11e-02 s103=1.11e-02 s104=1.23e-02 s105=1.11e-02 s106=1.23e-02 s107=1.11e-02 s108=1.23e-02 s110=1.23e-02 s111=1.08e-02  bit-identical: per-token=True g128=True g256=True g512=True  families separated: 4
[sweep] compared 14 (linear, M) pair(s)
SHAPE SWEEP: PASS

## row_invariance
  out_proj   per-token  rows 48    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=2.344e-02)
  out_proj   per-token  rows 64    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=3.125e-02)
  out_proj   per-token  rows 128   of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=4.688e-02)
  out_proj   per-token  rows 512   of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=3.516e-02)
  out_proj   per-token  rows 1000  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  out_proj   per-token  rows 2048  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  out_proj   per-token  across 7 tile shapes at M=1000: bit-identical=True
  out_proj   group 128  rows 48    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=2.344e-02)
  out_proj   group 128  rows 64    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=3.125e-02)
  out_proj   group 128  rows 128   of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=4.688e-02)
  out_proj   group 128  rows 512   of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=3.516e-02)
  out_proj   group 128  rows 1000  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  out_proj   group 128  rows 2048  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  out_proj   group 128  across 6 tile shapes at M=1000: bit-identical=True
  out_proj   group 256  rows 48    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=2.344e-02)
  out_proj   group 256  rows 64    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=3.125e-02)
  out_proj   group 256  rows 128   of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=4.688e-02)
  out_proj   group 256  rows 512   of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=3.516e-02)
  out_proj   group 256  rows 1000  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  out_proj   group 256  rows 2048  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  out_proj   group 256  across 3 tile shapes at M=1000: bit-identical=True
  gate_proj  per-token  rows 48    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=5.371e-03)
  gate_proj  per-token  rows 64    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=5.371e-03)
  gate_proj  per-token  rows 128   of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=4.883e-03)
  gate_proj  per-token  rows 512   of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  gate_proj  per-token  rows 1000  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  gate_proj  per-token  rows 2048  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  gate_proj  per-token  across 7 tile shapes at M=1000: bit-identical=True
  gate_proj  group 128  rows 48    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=5.371e-03)
  gate_proj  group 128  rows 64    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=5.371e-03)
  gate_proj  group 128  rows 128   of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=4.883e-03)
  gate_proj  group 128  rows 512   of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  gate_proj  group 128  rows 1000  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  gate_proj  group 128  rows 2048  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  gate_proj  group 128  across 6 tile shapes at M=1000: bit-identical=True
  gate_proj  group 256  rows 48    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=5.371e-03)
  gate_proj  group 256  rows 64    of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=5.371e-03)
  gate_proj  group 256  rows 128   of 4096: int8 bit-identical=True   legB bit-identical=False (max|delta|=4.883e-03)
  gate_proj  group 256  rows 512   of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  gate_proj  group 256  rows 1000  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  gate_proj  group 256  rows 2048  of 4096: int8 bit-identical=True   legB bit-identical=True (max|delta|=0.000e+00)
  gate_proj  group 256  across 3 tile shapes at M=1000: bit-identical=True
[rowinv] compared 2 linear(s)
ROW INVARIANCE: PASS

## a_prep live test
............................................................             [100%]
60 passed in 1.36s

## CPU tests (int8 suite)
315 passed in 2.94s
ALLDONE
