Token | darkness |
Token ID | 11854 |
Feature activation | +7.73e-02 |
Pos logit contributions | |
isa | +0.011 |
isha | +0.011 |
laptop | +0.011 |
uka | +0.011 |
acy | +0.011 |
Neg logit contributions | |
all | -0.008 |
were | -0.006 |
everywhere | -0.006 |
fun | -0.006 |
lots | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +3.57e-03 |
Pos logit contributions | |
isha | +0.193 |
laptop | +0.192 |
isa | +0.192 |
acy | +0.181 |
Stacy | +0.180 |
Neg logit contributions | |
all | -0.097 |
were | -0.077 |
everywhere | -0.071 |
happily | -0.064 |
lots | -0.062 |
Token | despite |
Token ID | 3805 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
isha | +0.009 |
laptop | +0.009 |
isa | +0.009 |
Stacy | +0.008 |
Ian | +0.008 |
Neg logit contributions | |
all | -0.004 |
were | -0.003 |
Ell | -0.003 |
everywhere | -0.003 |
grapes | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +4.58e-03 |
Pos logit contributions | |
isha | +0.044 |
laptop | +0.043 |
isa | +0.043 |
acy | +0.041 |
Stacy | +0.040 |
Neg logit contributions | |
all | -0.021 |
were | -0.017 |
everywhere | -0.015 |
Ell | -0.014 |
grapes | -0.014 |
Token | power |
Token ID | 1176 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
isa | +0.010 |
isha | +0.010 |
laptop | +0.010 |
acy | +0.009 |
uka | +0.009 |
Neg logit contributions | |
all | -0.007 |
were | -0.006 |
everywhere | -0.006 |
fun | -0.005 |
lots | -0.005 |
Token | outage |
Token ID | 46281 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
isa | +0.044 |
isha | +0.044 |
laptop | +0.043 |
uka | +0.041 |
acy | +0.041 |
Neg logit contributions | |
all | -0.019 |
were | -0.015 |
everywhere | -0.013 |
lots | -0.011 |
happily | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +3.60e-03 |
Pos logit contributions | |
isha | +0.699 |
laptop | +0.697 |
isa | +0.693 |
acy | +0.653 |
Stacy | +0.647 |
Neg logit contributions | |
all | -0.353 |
were | -0.279 |
everywhere | -0.261 |
Ell | -0.238 |
grapes | -0.237 |
Token | His |
Token ID | 2399 |
Feature activation | +9.12e-02 |
Pos logit contributions | |
isa | +0.008 |
isha | +0.008 |
laptop | +0.008 |
acy | +0.008 |
Stacy | +0.008 |
Neg logit contributions | |
all | -0.005 |
were | -0.004 |
everywhere | -0.004 |
Ellie | -0.004 |
Ell | -0.004 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
isha | +0.238 |
laptop | +0.237 |
isa | +0.236 |
acy | +0.224 |
Stacy | +0.221 |
Neg logit contributions | |
all | -0.103 |
were | -0.080 |
everywhere | -0.074 |
happily | -0.065 |
Ell | -0.065 |
Token | quickly |
Token ID | 2952 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
grapes | +0.101 |
all | +0.097 |
Tweet | +0.084 |
Ell | +0.081 |
nuts | +0.078 |
Neg logit contributions | |
isha | -0.293 |
isa | -0.291 |
laptop | -0.278 |
acy | -0.276 |
Stacy | -0.272 |
Token | joined |
Token ID | 5399 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
isha | +0.056 |
isa | +0.055 |
laptop | +0.055 |
Stacy | +0.052 |
acy | +0.052 |
Neg logit contributions | |
all | -0.032 |
grapes | -0.028 |
Ell | -0.027 |
Tweet | -0.027 |
everywhere | -0.027 |
Token | neighborhood |
Token ID | 6232 |
Feature activation | +4.77e-03 |
Pos logit contributions | |
all | +0.081 |
were | +0.068 |
everywhere | +0.064 |
happily | +0.060 |
lots | +0.060 |
Neg logit contributions | |
isha | -0.096 |
isa | -0.095 |
laptop | -0.095 |
acy | -0.088 |
Stacy | -0.087 |
Token | had |
Token ID | 550 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
isha | +0.012 |
laptop | +0.011 |
isa | +0.011 |
acy | +0.011 |
Ian | +0.011 |
Neg logit contributions | |
all | -0.006 |
grapes | -0.004 |
everywhere | -0.004 |
were | -0.004 |
Ell | -0.004 |
Token | suffered |
Token ID | 6989 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
all | +0.040 |
grapes | +0.038 |
everywhere | +0.036 |
Ell | +0.036 |
Ellie | +0.034 |
Neg logit contributions | |
isha | -0.072 |
laptop | -0.070 |
isa | -0.068 |
Bailey | -0.067 |
den | -0.066 |
Token | a |
Token ID | 257 |
Feature activation | +5.60e-03 |
Pos logit contributions | |
all | +0.032 |
grapes | +0.028 |
Ell | +0.028 |
were | +0.027 |
Thank | +0.026 |
Neg logit contributions | |
isha | -0.062 |
laptop | -0.060 |
isa | -0.060 |
acy | -0.057 |
Stacy | -0.057 |
Token | power |
Token ID | 1176 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
isha | +0.015 |
laptop | +0.014 |
isa | +0.014 |
acy | +0.014 |
Stacy | +0.013 |
Neg logit contributions | |
all | -0.006 |
were | -0.005 |
everywhere | -0.005 |
Ell | -0.004 |
grapes | -0.004 |
Token | outage |
Token ID | 46281 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
isha | +0.044 |
isa | +0.043 |
laptop | +0.043 |
acy | +0.041 |
Stacy | +0.041 |
Neg logit contributions | |
all | -0.018 |
were | -0.013 |
everywhere | -0.012 |
happily | -0.011 |
lots | -0.010 |
Token | ! |
Token ID | 0 |
Feature activation | -7.04e-04 |
Pos logit contributions | |
laptop | +0.692 |
isha | +0.683 |
isa | +0.678 |
acy | +0.633 |
Stacy | +0.629 |
Neg logit contributions | |
all | -0.335 |
were | -0.268 |
grapes | -0.252 |
everywhere | -0.249 |
Ell | -0.247 |
Token | Billy |
Token ID | 15890 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
all | +0.001 |
were | +0.001 |
everywhere | +0.001 |
Ellie | +0.001 |
Ell | +0.001 |
Neg logit contributions | |
isa | -0.002 |
isha | -0.002 |
laptop | -0.002 |
acy | -0.002 |
Stacy | -0.002 |
Token | rushed |
Token ID | 13999 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
isha | +0.052 |
isa | +0.052 |
laptop | +0.050 |
Stacy | +0.048 |
acy | +0.048 |
Neg logit contributions | |
all | -0.024 |
grapes | -0.019 |
everywhere | -0.018 |
were | -0.018 |
Ellie | -0.017 |
Token | back |
Token ID | 736 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
all | +0.053 |
were | +0.047 |
grapes | +0.044 |
Ell | +0.041 |
aws | +0.040 |
Neg logit contributions | |
isha | -0.095 |
laptop | -0.095 |
isa | -0.094 |
Stacy | -0.089 |
Ian | -0.088 |
Token | to |
Token ID | 284 |
Feature activation | +3.05e-03 |
Pos logit contributions | |
all | +0.013 |
everywhere | +0.009 |
were | +0.008 |
happily | +0.008 |
lots | +0.007 |
Neg logit contributions | |
isha | -0.036 |
laptop | -0.036 |
isa | -0.035 |
acy | -0.034 |
acey | -0.034 |
Token | they |
Token ID | 484 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
laptop | +0.011 |
isha | +0.011 |
isa | +0.010 |
clipboard | +0.010 |
Stacy | +0.010 |
Neg logit contributions | |
all | -0.009 |
were | -0.008 |
Ell | -0.007 |
Ellie | -0.007 |
Tweet | -0.007 |
Token | rode |
Token ID | 22075 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
all | +0.023 |
Ell | +0.019 |
grapes | +0.018 |
everywhere | +0.017 |
Tweet | +0.017 |
Neg logit contributions | |
isha | -0.058 |
laptop | -0.058 |
isa | -0.057 |
acy | -0.055 |
acey | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | -2.09e-03 |
Pos logit contributions | |
all | +0.027 |
were | +0.022 |
grapes | +0.020 |
Ell | +0.020 |
everywhere | +0.020 |
Neg logit contributions | |
laptop | -0.058 |
isha | -0.058 |
isa | -0.058 |
acy | -0.054 |
den | -0.054 |
Token | roller |
Token ID | 24471 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
all | +0.004 |
were | +0.003 |
everywhere | +0.003 |
happily | +0.003 |
Ell | +0.003 |
Neg logit contributions | |
isha | -0.004 |
isa | -0.004 |
laptop | -0.004 |
acy | -0.004 |
Stacy | -0.004 |
Token | co |
Token ID | 1073 |
Feature activation | -3.47e-03 |
Pos logit contributions | |
isha | +0.141 |
isa | +0.139 |
laptop | +0.135 |
acy | +0.131 |
Stacy | +0.128 |
Neg logit contributions | |
all | -0.135 |
were | -0.115 |
everywhere | -0.109 |
happily | -0.103 |
lots | -0.102 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
all | +0.007 |
were | +0.006 |
everywhere | +0.006 |
happily | +0.006 |
lots | +0.006 |
Neg logit contributions | |
isha | -0.006 |
Stacy | -0.006 |
laptop | -0.006 |
isa | -0.006 |
Macy | -0.006 |
Token | together |
Token ID | 1978 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
isha | +0.695 |
laptop | +0.694 |
isa | +0.689 |
acy | +0.651 |
Stacy | +0.644 |
Neg logit contributions | |
all | -0.321 |
were | -0.250 |
everywhere | -0.232 |
Ell | -0.209 |
grapes | -0.208 |
Token | . |
Token ID | 13 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
all | +0.024 |
were | +0.019 |
everywhere | +0.018 |
Ell | +0.017 |
grapes | +0.017 |
Neg logit contributions | |
isha | -0.046 |
laptop | -0.046 |
isa | -0.046 |
acy | -0.043 |
Stacy | -0.043 |
Token | It |
Token ID | 632 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
isa | +0.031 |
laptop | +0.031 |
isha | +0.030 |
asy | +0.029 |
pulse | +0.029 |
Neg logit contributions | |
all | -0.015 |
were | -0.012 |
Tim | -0.012 |
Ellie | -0.012 |
happily | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
all | +0.033 |
everywhere | +0.026 |
grapes | +0.025 |
were | +0.024 |
Ell | +0.022 |
Neg logit contributions | |
isa | -0.078 |
isha | -0.078 |
laptop | -0.077 |
Stacy | -0.072 |
acy | -0.072 |
Token | scary |
Token ID | 14343 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
all | +0.027 |
were | +0.022 |
Ell | +0.020 |
everywhere | +0.020 |
grapes | +0.019 |
Neg logit contributions | |
laptop | -0.055 |
isha | -0.055 |
isa | -0.054 |
aster | -0.051 |
acy | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
isha | +0.090 |
isa | +0.089 |
laptop | +0.089 |
acy | +0.085 |
Stacy | +0.084 |
Neg logit contributions | |
all | -0.056 |
were | -0.046 |
everywhere | -0.044 |
happily | -0.039 |
grapes | -0.038 |
Token | She |
Token ID | 1375 |
Feature activation | +5.60e-03 |
Pos logit contributions | |
isa | +0.036 |
laptop | +0.035 |
isha | +0.035 |
den | +0.034 |
asy | +0.033 |
Neg logit contributions | |
all | -0.015 |
were | -0.011 |
everywhere | -0.009 |
Tim | -0.009 |
lots | -0.009 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
isha | +0.012 |
isa | +0.012 |
laptop | +0.012 |
Stacy | +0.011 |
uka | +0.011 |
Neg logit contributions | |
all | -0.007 |
Ellie | -0.007 |
grapes | -0.007 |
everywhere | -0.006 |
Ell | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
all | +0.018 |
were | +0.015 |
everywhere | +0.014 |
happily | +0.013 |
lots | +0.013 |
Neg logit contributions | |
isha | -0.030 |
isa | -0.030 |
laptop | -0.030 |
acy | -0.028 |
Stacy | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
isha | +0.033 |
isa | +0.033 |
laptop | +0.032 |
acy | +0.031 |
Stacy | +0.031 |
Neg logit contributions | |
all | -0.028 |
were | -0.024 |
everywhere | -0.023 |
grapes | -0.022 |
lots | -0.021 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
isha | +0.030 |
isa | +0.030 |
laptop | +0.029 |
outage | +0.028 |
den | +0.028 |
Neg logit contributions | |
all | -0.012 |
were | -0.007 |
Tim | -0.007 |
happily | -0.007 |
everywhere | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
isha | +0.786 |
isa | +0.780 |
acy | +0.755 |
laptop | +0.748 |
Stacy | +0.736 |
Neg logit contributions | |
all | -0.243 |
were | -0.159 |
everywhere | -0.150 |
happily | -0.122 |
lots | -0.117 |
Token | threw |
Token ID | 9617 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
all | +0.027 |
were | +0.022 |
everywhere | +0.021 |
happily | +0.019 |
Ell | +0.019 |
Neg logit contributions | |
isha | -0.047 |
laptop | -0.046 |
isa | -0.046 |
acy | -0.043 |
Stacy | -0.043 |
Token | it |
Token ID | 340 |
Feature activation | +2.35e-04 |
Pos logit contributions | |
all | +0.056 |
were | +0.045 |
everywhere | +0.041 |
Ell | +0.039 |
happily | +0.038 |
Neg logit contributions | |
isha | -0.111 |
laptop | -0.110 |
isa | -0.110 |
acy | -0.104 |
Stacy | -0.103 |
Token | in |
Token ID | 287 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
laptop | +0.000 |
pulse | +0.000 |
asy | +0.000 |
isa | +0.000 |
aster | +0.000 |
Neg logit contributions | |
Ell | -0.000 |
were | -0.000 |
all | -0.000 |
Ellie | -0.000 |
aws | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
laptop | +0.089 |
isha | +0.089 |
isa | +0.088 |
acy | +0.083 |
Stacy | +0.083 |
Neg logit contributions | |
all | -0.041 |
were | -0.032 |
everywhere | -0.030 |
Ell | -0.028 |
grapes | -0.027 |
Token | un |
Token ID | 555 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
isa | +0.203 |
isha | +0.200 |
acy | +0.194 |
den | +0.188 |
cle | +0.185 |
Neg logit contributions | |
all | -0.168 |
were | -0.132 |
y | -0.128 |
everywhere | -0.127 |
forg | -0.120 |
Token | pl |
Token ID | 489 |
Feature activation | +8.86e-02 |
Pos logit contributions | |
isha | +0.195 |
laptop | +0.189 |
isa | +0.186 |
acy | +0.180 |
Stacy | +0.172 |
Neg logit contributions | |
all | -0.216 |
were | -0.180 |
everywhere | -0.179 |
happily | -0.174 |
lots | -0.168 |
Token | ugs |
Token ID | 10339 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
laptop | +0.150 |
isa | +0.146 |
isha | +0.145 |
pharmacy | +0.138 |
Stacy | +0.136 |
Neg logit contributions | |
all | -0.175 |
were | -0.166 |
everywhere | -0.149 |
aws | -0.142 |
lots | -0.133 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
laptop | +0.271 |
isha | +0.268 |
isa | +0.266 |
acy | +0.254 |
Stacy | +0.252 |
Neg logit contributions | |
all | -0.132 |
were | -0.103 |
Ell | -0.100 |
everywhere | -0.097 |
grapes | -0.095 |
Token | to |
Token ID | 284 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
isa | +0.032 |
isha | +0.032 |
acy | +0.030 |
laptop | +0.030 |
Stacy | +0.029 |
Neg logit contributions | |
all | -0.035 |
everywhere | -0.030 |
were | -0.029 |
grapes | -0.027 |
fun | -0.027 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
isha | +0.023 |
isa | +0.022 |
laptop | +0.022 |
acy | +0.021 |
acey | +0.020 |
Neg logit contributions | |
all | -0.020 |
were | -0.017 |
everywhere | -0.015 |
Tim | -0.015 |
Tweet | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
isha | +0.708 |
isa | +0.699 |
laptop | +0.681 |
acy | +0.665 |
Stacy | +0.650 |
Neg logit contributions | |
all | -0.310 |
were | -0.240 |
everywhere | -0.223 |
happily | -0.192 |
lots | -0.188 |
Token | He |
Token ID | 679 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
isa | +0.043 |
isha | +0.042 |
laptop | +0.042 |
acy | +0.040 |
acey | +0.039 |
Neg logit contributions | |
all | -0.023 |
were | -0.018 |
Tim | -0.017 |
Tweet | -0.016 |
everywhere | -0.016 |
Token | cough |
Token ID | 22094 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
isha | +0.024 |
isa | +0.023 |
laptop | +0.023 |
acy | +0.022 |
den | +0.021 |
Neg logit contributions | |
all | -0.024 |
were | -0.020 |
everywhere | -0.019 |
y | -0.018 |
happily | -0.018 |
Token | s |
Token ID | 82 |
Feature activation | +9.16e-02 |
Pos logit contributions | |
isha | +0.299 |
laptop | +0.297 |
isa | +0.295 |
acy | +0.277 |
Stacy | +0.275 |
Neg logit contributions | |
all | -0.192 |
were | -0.157 |
everywhere | -0.149 |
Ell | -0.139 |
grapes | -0.138 |
Token | and |
Token ID | 290 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
isha | +0.249 |
isa | +0.247 |
laptop | +0.245 |
acy | +0.235 |
Stacy | +0.232 |
Neg logit contributions | |
all | -0.103 |
were | -0.076 |
everywhere | -0.071 |
happily | -0.063 |
lots | -0.061 |
Token | put |
Token ID | 1234 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
all | +0.029 |
were | +0.023 |
everywhere | +0.023 |
Ell | +0.022 |
Tweet | +0.021 |
Neg logit contributions | |
isha | -0.054 |
isa | -0.053 |
laptop | -0.053 |
Stacy | -0.050 |
acy | -0.050 |
Token | bread |
Token ID | 8509 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
isha | +0.082 |
laptop | +0.082 |
isa | +0.081 |
outage | +0.080 |
acy | +0.077 |
Neg logit contributions | |
all | -0.035 |
were | -0.023 |
everywhere | -0.023 |
lots | -0.021 |
happily | -0.019 |
Token | in |
Token ID | 287 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
all | +0.049 |
were | +0.036 |
everywhere | +0.034 |
happily | +0.029 |
grapes | +0.029 |
Neg logit contributions | |
isha | -0.128 |
laptop | -0.127 |
isa | -0.127 |
acy | -0.120 |
Stacy | -0.119 |
Token | the |
Token ID | 262 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
isha | +0.090 |
isa | +0.090 |
laptop | +0.088 |
outage | +0.085 |
acy | +0.085 |
Neg logit contributions | |
all | -0.033 |
were | -0.021 |
lots | -0.019 |
everywhere | -0.018 |
Tim | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | +6.75e-03 |
Pos logit contributions | |
isa | +0.029 |
isha | +0.027 |
den | +0.027 |
outage | +0.026 |
acy | +0.026 |
Neg logit contributions | |
all | -0.031 |
grapes | -0.026 |
y | -0.026 |
everywhere | -0.025 |
were | -0.025 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
isha | +0.009 |
outage | +0.009 |
isa | +0.009 |
den | +0.009 |
laptop | +0.008 |
Neg logit contributions | |
all | -0.018 |
were | -0.015 |
y | -0.015 |
grapes | -0.015 |
happily | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
isha | +0.718 |
isa | +0.710 |
laptop | +0.687 |
acy | +0.686 |
Ian | +0.675 |
Neg logit contributions | |
all | -0.304 |
were | -0.216 |
everywhere | -0.204 |
lots | -0.187 |
happily | -0.185 |
Token |
|
Token ID | 198 |
Feature activation | -8.11e-03 |
Pos logit contributions | |
isa | +0.064 |
isha | +0.064 |
laptop | +0.062 |
acy | +0.060 |
Stacy | +0.059 |
Neg logit contributions | |
all | -0.031 |
were | -0.026 |
everywhere | -0.022 |
Tim | -0.021 |
happily | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -9.64e-03 |
Pos logit contributions | |
all | +0.008 |
were | +0.007 |
everywhere | +0.006 |
lots | +0.005 |
happily | +0.005 |
Neg logit contributions | |
isa | -0.022 |
laptop | -0.022 |
isha | -0.022 |
acy | -0.021 |
pulse | -0.021 |
Token | Tom |
Token ID | 13787 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
all | +0.014 |
were | +0.011 |
everywhere | +0.011 |
happily | +0.010 |
grapes | +0.010 |
Neg logit contributions | |
isha | -0.022 |
laptop | -0.022 |
isa | -0.022 |
acy | -0.021 |
Stacy | -0.021 |
Token | waited |
Token ID | 13488 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
all | +0.045 |
were | +0.035 |
happily | +0.029 |
everywhere | +0.028 |
lots | +0.025 |
Neg logit contributions | |
laptop | -0.124 |
isha | -0.122 |
isa | -0.121 |
acy | -0.117 |
Stacy | -0.116 |
Token | put |
Token ID | 1234 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
isha | +0.008 |
isa | +0.007 |
laptop | +0.007 |
acy | +0.007 |
Stacy | +0.007 |
Neg logit contributions | |
all | -0.004 |
everywhere | -0.003 |
were | -0.003 |
Ell | -0.002 |
grapes | -0.002 |
Token | bread |
Token ID | 8509 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
laptop | +0.081 |
isha | +0.081 |
isa | +0.080 |
outage | +0.076 |
acy | +0.076 |
Neg logit contributions | |
all | -0.036 |
were | -0.026 |
everywhere | -0.025 |
lots | -0.022 |
grapes | -0.021 |
Token | in |
Token ID | 287 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
all | +0.049 |
were | +0.039 |
Ell | +0.035 |
everywhere | +0.034 |
Ellie | +0.032 |
Neg logit contributions | |
laptop | -0.119 |
isha | -0.119 |
isa | -0.119 |
Stacy | -0.112 |
acy | -0.112 |
Token | the |
Token ID | 262 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
isha | +0.080 |
isa | +0.080 |
laptop | +0.079 |
acy | +0.075 |
Stacy | +0.074 |
Neg logit contributions | |
all | -0.034 |
were | -0.025 |
everywhere | -0.023 |
lots | -0.021 |
grapes | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | +6.84e-03 |
Pos logit contributions | |
isa | +0.036 |
isha | +0.034 |
outage | +0.033 |
den | +0.033 |
acy | +0.032 |
Neg logit contributions | |
all | -0.037 |
everywhere | -0.031 |
were | -0.030 |
grapes | -0.030 |
y | -0.029 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
outage | +0.016 |
isha | +0.015 |
isa | +0.015 |
den | +0.015 |
laptop | +0.014 |
Neg logit contributions | |
all | -0.013 |
y | -0.009 |
were | -0.009 |
lots | -0.009 |
happily | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
isha | +0.780 |
isa | +0.774 |
laptop | +0.756 |
acy | +0.754 |
uka | +0.737 |
Neg logit contributions | |
all | -0.231 |
were | -0.147 |
everywhere | -0.136 |
lots | -0.114 |
happily | -0.112 |
Token | waited |
Token ID | 13488 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
all | +0.033 |
were | +0.027 |
happily | +0.023 |
everywhere | +0.023 |
lots | +0.023 |
Neg logit contributions | |
isha | -0.057 |
laptop | -0.056 |
isa | -0.056 |
acy | -0.055 |
acey | -0.053 |
Token | for |
Token ID | 329 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
isa | +0.101 |
isha | +0.100 |
laptop | +0.099 |
uka | +0.096 |
acy | +0.095 |
Neg logit contributions | |
all | -0.033 |
everywhere | -0.020 |
happily | -0.018 |
were | -0.017 |
lots | -0.015 |
Token | it |
Token ID | 340 |
Feature activation | -3.67e-03 |
Pos logit contributions | |
all | +0.047 |
were | +0.028 |
lots | +0.027 |
everywhere | +0.027 |
fun | +0.026 |
Neg logit contributions | |
laptop | -0.116 |
isha | -0.115 |
isa | -0.115 |
outage | -0.112 |
acy | -0.109 |
Token | to |
Token ID | 284 |
Feature activation | +8.32e-03 |
Pos logit contributions | |
all | +0.004 |
were | +0.003 |
everywhere | +0.003 |
Ell | +0.003 |
grapes | +0.003 |
Neg logit contributions | |
isha | -0.010 |
laptop | -0.009 |
isa | -0.009 |
acy | -0.009 |
Stacy | -0.009 |
Token | flames |
Token ID | 16209 |
Feature activation | +6.81e-02 |
Pos logit contributions | |
all | +0.043 |
were | +0.037 |
everywhere | +0.034 |
lots | +0.032 |
fun | +0.031 |
Neg logit contributions | |
isha | -0.057 |
isa | -0.057 |
laptop | -0.055 |
acy | -0.054 |
acey | -0.053 |
Token | coming |
Token ID | 2406 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
laptop | +0.158 |
isha | +0.157 |
isa | +0.156 |
acy | +0.146 |
Stacy | +0.144 |
Neg logit contributions | |
all | -0.094 |
were | -0.076 |
everywhere | -0.071 |
Ell | -0.069 |
grapes | -0.068 |
Token | from |
Token ID | 422 |
Feature activation | +8.79e-04 |
Pos logit contributions | |
isha | +0.093 |
laptop | +0.092 |
isa | +0.092 |
acy | +0.087 |
acey | +0.087 |
Neg logit contributions | |
all | -0.049 |
were | -0.037 |
everywhere | -0.037 |
happily | -0.031 |
lots | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
isha | +0.003 |
isa | +0.002 |
laptop | +0.002 |
acy | +0.002 |
acey | +0.002 |
Neg logit contributions | |
all | -0.001 |
everywhere | -0.001 |
were | -0.001 |
happily | -0.001 |
Tim | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +5.34e-03 |
Pos logit contributions | |
isa | +0.021 |
isha | +0.020 |
acy | +0.019 |
Stacy | +0.019 |
Ian | +0.019 |
Neg logit contributions | |
all | -0.022 |
everywhere | -0.020 |
were | -0.019 |
grapes | -0.018 |
fun | -0.017 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
isha | +0.015 |
isa | +0.015 |
laptop | +0.015 |
acy | +0.014 |
acey | +0.014 |
Neg logit contributions | |
all | -0.006 |
were | -0.004 |
everywhere | -0.004 |
Tim | -0.003 |
lots | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +8.96e-03 |
Pos logit contributions | |
isha | +0.667 |
isa | +0.662 |
laptop | +0.644 |
acy | +0.632 |
Stacy | +0.625 |
Neg logit contributions | |
all | -0.333 |
were | -0.258 |
everywhere | -0.244 |
lots | -0.214 |
happily | -0.212 |
Token | Their |
Token ID | 5334 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
isa | +0.022 |
isha | +0.022 |
laptop | +0.021 |
acy | +0.020 |
uka | +0.020 |
Neg logit contributions | |
all | -0.012 |
were | -0.010 |
everywhere | -0.009 |
Tim | -0.009 |
Tweet | -0.009 |
Token | mom |
Token ID | 1995 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
laptop | +0.206 |
isha | +0.199 |
isa | +0.196 |
acy | +0.177 |
pulse | +0.176 |
Neg logit contributions | |
all | -0.213 |
were | -0.186 |
Ell | -0.179 |
everywhere | -0.177 |
Ellie | -0.174 |
Token | is |
Token ID | 318 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
laptop | +0.150 |
isha | +0.150 |
isa | +0.148 |
acy | +0.135 |
Stacy | +0.134 |
Neg logit contributions | |
all | -0.157 |
were | -0.135 |
everywhere | -0.131 |
Ell | -0.127 |
Tweet | -0.126 |
Token | trying |
Token ID | 2111 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
isha | +0.076 |
laptop | +0.076 |
isa | +0.075 |
acy | +0.070 |
Stacy | +0.070 |
Neg logit contributions | |
all | -0.047 |
were | -0.039 |
everywhere | -0.037 |
Ell | -0.034 |
grapes | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
isha | +0.084 |
isa | +0.083 |
laptop | +0.083 |
acy | +0.079 |
Stacy | +0.078 |
Neg logit contributions | |
all | -0.034 |
were | -0.026 |
everywhere | -0.024 |
happily | -0.021 |
lots | -0.020 |
Token | fire |
Token ID | 2046 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
all | +0.051 |
were | +0.043 |
everywhere | +0.041 |
grapes | +0.038 |
lots | +0.038 |
Neg logit contributions | |
isha | -0.066 |
laptop | -0.065 |
isa | -0.065 |
acy | -0.061 |
Stacy | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
laptop | +0.080 |
isha | +0.080 |
isa | +0.079 |
acy | +0.074 |
Stacy | +0.073 |
Neg logit contributions | |
all | -0.048 |
were | -0.039 |
everywhere | -0.037 |
Ell | -0.036 |
grapes | -0.035 |
Token | The |
Token ID | 383 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
isa | +0.031 |
laptop | +0.030 |
isha | +0.029 |
outage | +0.028 |
den | +0.028 |
Neg logit contributions | |
all | -0.019 |
Tim | -0.015 |
were | -0.015 |
lots | -0.014 |
grapes | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | +5.63e-03 |
Pos logit contributions | |
isa | +0.105 |
isha | +0.105 |
laptop | +0.100 |
acy | +0.099 |
Stacy | +0.097 |
Neg logit contributions | |
all | -0.097 |
were | -0.082 |
everywhere | -0.079 |
grapes | -0.077 |
fun | -0.074 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
isha | +0.010 |
isa | +0.010 |
laptop | +0.010 |
acey | +0.009 |
acy | +0.009 |
Neg logit contributions | |
all | -0.012 |
were | -0.010 |
Tim | -0.009 |
everywhere | -0.009 |
Tweet | -0.009 |
Token | made |
Token ID | 925 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
isha | +0.696 |
isa | +0.689 |
laptop | +0.684 |
acy | +0.656 |
Stacy | +0.646 |
Neg logit contributions | |
all | -0.298 |
were | -0.235 |
everywhere | -0.211 |
happily | -0.182 |
lots | -0.177 |
Token | a |
Token ID | 257 |
Feature activation | +8.86e-03 |
Pos logit contributions | |
all | +0.020 |
were | +0.015 |
everywhere | +0.014 |
lots | +0.014 |
happily | +0.013 |
Neg logit contributions | |
isha | -0.040 |
laptop | -0.040 |
isa | -0.040 |
den | -0.037 |
acy | -0.037 |
Token | loud |
Token ID | 7812 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
isa | +0.024 |
isha | +0.024 |
laptop | +0.023 |
acy | +0.023 |
Stacy | +0.022 |
Neg logit contributions | |
all | -0.010 |
were | -0.007 |
everywhere | -0.007 |
happily | -0.006 |
lots | -0.006 |
Token | noise |
Token ID | 7838 |
Feature activation | -6.10e-03 |
Pos logit contributions | |
all | +0.027 |
were | +0.021 |
everywhere | +0.019 |
happily | +0.017 |
lots | +0.017 |
Neg logit contributions | |
isha | -0.053 |
isa | -0.052 |
laptop | -0.051 |
acy | -0.051 |
acey | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
all | +0.007 |
were | +0.006 |
everywhere | +0.005 |
happily | +0.004 |
lots | +0.004 |
Neg logit contributions | |
isha | -0.016 |
isa | -0.016 |
laptop | -0.016 |
acy | -0.016 |
acey | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
isha | +0.065 |
laptop | +0.064 |
isa | +0.064 |
aster | +0.061 |
Stacy | +0.061 |
Neg logit contributions | |
all | -0.034 |
were | -0.028 |
Ell | -0.027 |
everywhere | -0.026 |
Tweet | -0.025 |
Token | toast |
Token ID | 27805 |
Feature activation | +5.40e-03 |
Pos logit contributions | |
isha | +0.024 |
isa | +0.024 |
laptop | +0.023 |
acy | +0.022 |
Stacy | +0.022 |
Neg logit contributions | |
all | -0.025 |
were | -0.021 |
everywhere | -0.021 |
grapes | -0.020 |
lots | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
isha | +0.014 |
laptop | +0.014 |
isa | +0.014 |
acy | +0.013 |
Stacy | +0.013 |
Neg logit contributions | |
all | -0.006 |
were | -0.005 |
everywhere | -0.004 |
Ell | -0.004 |
grapes | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
laptop | +0.067 |
isha | +0.066 |
isa | +0.065 |
aster | +0.063 |
Ian | +0.062 |
Neg logit contributions | |
all | -0.036 |
were | -0.030 |
everywhere | -0.029 |
Ell | -0.028 |
aws | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | +5.83e-03 |
Pos logit contributions | |
isa | +0.022 |
isha | +0.022 |
acy | +0.021 |
outage | +0.020 |
den | +0.020 |
Neg logit contributions | |
all | -0.026 |
everywhere | -0.022 |
grapes | -0.022 |
were | -0.022 |
y | -0.021 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
isha | +0.007 |
laptop | +0.007 |
isa | +0.006 |
outage | +0.006 |
acy | +0.006 |
Neg logit contributions | |
all | -0.016 |
were | -0.014 |
happily | -0.013 |
grapes | -0.013 |
lots | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
isha | +0.712 |
isa | +0.705 |
laptop | +0.699 |
acy | +0.674 |
Stacy | +0.663 |
Neg logit contributions | |
all | -0.280 |
were | -0.207 |
everywhere | -0.190 |
happily | -0.168 |
lots | -0.165 |
Token | waited |
Token ID | 13488 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
all | +0.031 |
were | +0.025 |
everywhere | +0.024 |
Ell | +0.023 |
grapes | +0.022 |
Neg logit contributions | |
isha | -0.052 |
isa | -0.052 |
laptop | -0.052 |
Stacy | -0.049 |
acy | -0.048 |
Token | for |
Token ID | 329 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
isa | +0.093 |
isha | +0.093 |
laptop | +0.092 |
uka | +0.087 |
acy | +0.087 |
Neg logit contributions | |
all | -0.039 |
everywhere | -0.026 |
were | -0.025 |
happily | -0.025 |
lots | -0.021 |
Token | it |
Token ID | 340 |
Feature activation | -5.96e-03 |
Pos logit contributions | |
all | +0.056 |
were | +0.046 |
everywhere | +0.043 |
Ell | +0.042 |
happily | +0.039 |
Neg logit contributions | |
isha | -0.102 |
isa | -0.101 |
laptop | -0.101 |
aster | -0.096 |
acy | -0.095 |
Token | to |
Token ID | 284 |
Feature activation | +7.63e-03 |
Pos logit contributions | |
all | +0.007 |
were | +0.006 |
everywhere | +0.005 |
Ell | +0.005 |
grapes | +0.005 |
Neg logit contributions | |
laptop | -0.015 |
isha | -0.015 |
isa | -0.015 |
acy | -0.014 |
Stacy | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
isha | +0.096 |
laptop | +0.095 |
isa | +0.095 |
acy | +0.090 |
Stacy | +0.088 |
Neg logit contributions | |
all | -0.042 |
were | -0.031 |
everywhere | -0.030 |
lots | -0.026 |
grapes | -0.026 |
Token | bread |
Token ID | 8509 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
isha | +0.034 |
isa | +0.034 |
laptop | +0.034 |
acy | +0.031 |
Stacy | +0.031 |
Neg logit contributions | |
all | -0.030 |
were | -0.026 |
everywhere | -0.025 |
grapes | -0.023 |
happily | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
all | +0.053 |
were | +0.043 |
Ell | +0.039 |
everywhere | +0.038 |
Ellie | +0.037 |
Neg logit contributions | |
laptop | -0.118 |
isa | -0.116 |
isha | -0.116 |
acy | -0.110 |
Stacy | -0.109 |
Token | the |
Token ID | 262 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
isha | +0.094 |
isa | +0.093 |
laptop | +0.092 |
acy | +0.088 |
Stacy | +0.087 |
Neg logit contributions | |
all | -0.041 |
were | -0.031 |
everywhere | -0.028 |
lots | -0.026 |
happily | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | +8.41e-03 |
Pos logit contributions | |
isa | +0.029 |
isha | +0.028 |
acy | +0.026 |
den | +0.026 |
Ian | +0.025 |
Neg logit contributions | |
all | -0.035 |
grapes | -0.030 |
everywhere | -0.030 |
were | -0.030 |
y | -0.029 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
isha | +0.010 |
isa | +0.009 |
laptop | +0.008 |
outage | +0.008 |
den | +0.008 |
Neg logit contributions | |
all | -0.025 |
were | -0.021 |
happily | -0.020 |
y | -0.020 |
lots | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
isha | +0.735 |
isa | +0.726 |
laptop | +0.702 |
acy | +0.700 |
Stacy | +0.689 |
Neg logit contributions | |
all | -0.275 |
were | -0.195 |
everywhere | -0.182 |
lots | -0.157 |
happily | -0.156 |
Token | pushed |
Token ID | 7121 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
all | +0.029 |
were | +0.023 |
everywhere | +0.021 |
happily | +0.020 |
lots | +0.019 |
Neg logit contributions | |
isha | -0.053 |
laptop | -0.052 |
isa | -0.052 |
acy | -0.050 |
acey | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
isha | +0.032 |
laptop | +0.031 |
isa | +0.031 |
outage | +0.030 |
acy | +0.029 |
Neg logit contributions | |
all | -0.012 |
everywhere | -0.007 |
were | -0.007 |
happily | -0.007 |
lots | -0.006 |
Token | button |
Token ID | 4936 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
isha | +0.034 |
isa | +0.034 |
laptop | +0.033 |
acy | +0.032 |
Stacy | +0.032 |
Neg logit contributions | |
all | -0.029 |
were | -0.024 |
everywhere | -0.024 |
grapes | -0.023 |
lots | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
isha | +0.366 |
isa | +0.358 |
laptop | +0.349 |
acy | +0.341 |
acey | +0.338 |
Neg logit contributions | |
all | -0.160 |
everywhere | -0.112 |
were | -0.110 |
happily | -0.101 |
lots | -0.092 |
Token | the |
Token ID | 262 |
Feature activation | +5.13e-04 |
Pos logit contributions | |
all | +0.124 |
were | +0.091 |
everywhere | +0.088 |
grapes | +0.082 |
lots | +0.081 |
Neg logit contributions | |
isha | -0.231 |
isa | -0.229 |
laptop | -0.228 |
acy | -0.217 |
acey | -0.214 |
Token | stove |
Token ID | 27635 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
isa | +0.001 |
isha | +0.001 |
laptop | +0.001 |
acy | +0.001 |
Stacy | +0.001 |
Neg logit contributions | |
all | -0.001 |
were | -0.001 |
everywhere | -0.001 |
grapes | -0.001 |
fun | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
isha | +0.445 |
isa | +0.440 |
laptop | +0.429 |
acy | +0.420 |
Stacy | +0.417 |
Neg logit contributions | |
all | -0.213 |
were | -0.162 |
everywhere | -0.152 |
happily | -0.132 |
lots | -0.129 |
Token | the |
Token ID | 262 |
Feature activation | +3.13e-03 |
Pos logit contributions | |
all | +0.040 |
were | +0.032 |
everywhere | +0.030 |
Ell | +0.028 |
happily | +0.027 |
Neg logit contributions | |
isha | -0.076 |
laptop | -0.076 |
isa | -0.075 |
acy | -0.071 |
Stacy | -0.070 |
Token | to |
Token ID | 284 |
Feature activation | +1.44e-03 |
Pos logit contributions | |
isha | +0.006 |
isa | +0.006 |
laptop | +0.006 |
acy | +0.006 |
Stacy | +0.006 |
Neg logit contributions | |
all | -0.006 |
were | -0.005 |
everywhere | -0.005 |
grapes | -0.004 |
lots | -0.004 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
isha | +0.003 |
isa | +0.003 |
laptop | +0.003 |
acy | +0.003 |
acey | +0.003 |
Neg logit contributions | |
all | -0.003 |
were | -0.002 |
everywhere | -0.002 |
Tim | -0.002 |
Tweet | -0.002 |
Token | ," |
Token ID | 553 |
Feature activation | +6.91e-02 |
Pos logit contributions | |
isha | +0.629 |
isa | +0.621 |
laptop | +0.614 |
acy | +0.589 |
Stacy | +0.582 |
Neg logit contributions | |
all | -0.361 |
were | -0.290 |
everywhere | -0.272 |
happily | -0.248 |
lots | -0.245 |
Token | Tom |
Token ID | 4186 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
isha | +0.163 |
laptop | +0.162 |
isa | +0.162 |
acy | +0.152 |
Stacy | +0.150 |
Neg logit contributions | |
all | -0.098 |
were | -0.079 |
everywhere | -0.074 |
happily | -0.068 |
grapes | -0.068 |
Token | says |
Token ID | 1139 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
all | +0.043 |
were | +0.037 |
everywhere | +0.034 |
happily | +0.033 |
grapes | +0.032 |
Neg logit contributions | |
isha | -0.046 |
isa | -0.045 |
laptop | -0.045 |
acy | -0.042 |
aster | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +5.06e-03 |
Pos logit contributions | |
isha | +0.037 |
isa | +0.037 |
laptop | +0.036 |
acy | +0.035 |
uka | +0.034 |
Neg logit contributions | |
all | -0.016 |
were | -0.012 |
happily | -0.011 |
everywhere | -0.011 |
Ellie | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
isha | +0.013 |
laptop | +0.013 |
isa | +0.012 |
acy | +0.012 |
Stacy | +0.012 |
Neg logit contributions | |
all | -0.006 |
were | -0.005 |
everywhere | -0.005 |
happily | -0.004 |
Ell | -0.004 |
Token | Sam |
Token ID | 3409 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
laptop | +0.028 |
isa | +0.028 |
asy | +0.027 |
isha | +0.027 |
aster | +0.027 |
Neg logit contributions | |
all | -0.017 |
were | -0.015 |
Tim | -0.014 |
everywhere | -0.013 |
happily | -0.012 |
Token | set |
Token ID | 900 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
laptop | +0.049 |
isha | +0.049 |
isa | +0.049 |
acy | +0.046 |
Stacy | +0.046 |
Neg logit contributions | |
all | -0.020 |
were | -0.015 |
everywhere | -0.013 |
happily | -0.012 |
lots | -0.012 |
Token | up |
Token ID | 510 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
all | +0.019 |
everywhere | +0.011 |
lots | +0.011 |
were | +0.011 |
happily | +0.011 |
Neg logit contributions | |
isha | -0.042 |
laptop | -0.042 |
isa | -0.042 |
outage | -0.041 |
acey | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
isha | +0.106 |
laptop | +0.104 |
isa | +0.104 |
acy | +0.099 |
Stacy | +0.098 |
Neg logit contributions | |
all | -0.056 |
were | -0.043 |
everywhere | -0.041 |
lots | -0.037 |
happily | -0.037 |
Token | to |
Token ID | 284 |
Feature activation | +5.67e-03 |
Pos logit contributions | |
isha | +0.026 |
isa | +0.026 |
laptop | +0.025 |
acy | +0.025 |
Stacy | +0.025 |
Neg logit contributions | |
all | -0.030 |
everywhere | -0.025 |
were | -0.024 |
grapes | -0.024 |
y | -0.023 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
isha | +0.013 |
outage | +0.013 |
isa | +0.013 |
laptop | +0.013 |
ometer | +0.012 |
Neg logit contributions | |
all | -0.010 |
were | -0.007 |
lots | -0.007 |
happily | -0.007 |
Tim | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
isha | +0.743 |
isa | +0.735 |
laptop | +0.728 |
acy | +0.709 |
acey | +0.700 |
Neg logit contributions | |
all | -0.252 |
were | -0.171 |
everywhere | -0.160 |
lots | -0.138 |
happily | -0.134 |
Token | put |
Token ID | 1234 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
all | +0.032 |
were | +0.025 |
happily | +0.023 |
everywhere | +0.022 |
lots | +0.021 |
Neg logit contributions | |
laptop | -0.056 |
isha | -0.055 |
isa | -0.054 |
acy | -0.052 |
acey | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
laptop | +0.094 |
isha | +0.093 |
outage | +0.092 |
isa | +0.090 |
acy | +0.087 |
Neg logit contributions | |
all | -0.041 |
everywhere | -0.026 |
were | -0.026 |
lots | -0.025 |
happily | -0.023 |
Token | two |
Token ID | 734 |
Feature activation | -5.72e-03 |
Pos logit contributions | |
isha | +0.029 |
isa | +0.029 |
laptop | +0.029 |
Stacy | +0.028 |
acy | +0.028 |
Neg logit contributions | |
all | -0.029 |
were | -0.024 |
everywhere | -0.024 |
grapes | -0.023 |
y | -0.023 |
Token | pieces |
Token ID | 5207 |
Feature activation | -6.64e-02 |
Pos logit contributions | |
all | +0.010 |
were | +0.008 |
everywhere | +0.008 |
grapes | +0.007 |
happily | +0.007 |
Neg logit contributions | |
isha | -0.012 |
laptop | -0.012 |
isa | -0.012 |
Stacy | -0.011 |
acy | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
isha | +0.095 |
isa | +0.095 |
laptop | +0.094 |
acy | +0.090 |
uka | +0.089 |
Neg logit contributions | |
all | -0.034 |
everywhere | -0.022 |
were | -0.021 |
happily | -0.020 |
lots | -0.018 |
Token | Finally |
Token ID | 9461 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
isha | +0.024 |
laptop | +0.024 |
isa | +0.024 |
acy | +0.022 |
Stacy | +0.022 |
Neg logit contributions | |
all | -0.024 |
were | -0.021 |
everywhere | -0.020 |
Ell | -0.019 |
Tweet | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +9.99e-03 |
Pos logit contributions | |
isha | +0.123 |
isa | +0.122 |
laptop | +0.121 |
acy | +0.115 |
Stacy | +0.114 |
Neg logit contributions | |
all | -0.073 |
were | -0.060 |
everywhere | -0.058 |
Ell | -0.057 |
grapes | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | +4.38e-03 |
Pos logit contributions | |
isha | +0.023 |
isa | +0.023 |
laptop | +0.023 |
acy | +0.022 |
Stacy | +0.022 |
Neg logit contributions | |
all | -0.013 |
were | -0.011 |
Ell | -0.010 |
everywhere | -0.010 |
Tweet | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +6.34e-03 |
Pos logit contributions | |
isa | +0.010 |
isha | +0.010 |
laptop | +0.010 |
acy | +0.010 |
Stacy | +0.009 |
Neg logit contributions | |
all | -0.006 |
were | -0.005 |
everywhere | -0.005 |
grapes | -0.005 |
fun | -0.004 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
isha | +0.012 |
laptop | +0.012 |
isa | +0.012 |
acy | +0.011 |
Stacy | +0.011 |
Neg logit contributions | |
all | -0.012 |
were | -0.010 |
everywhere | -0.009 |
happily | -0.009 |
grapes | -0.009 |
Token | said |
Token ID | 531 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
isha | +0.661 |
laptop | +0.658 |
isa | +0.653 |
acy | +0.614 |
aster | +0.612 |
Neg logit contributions | |
all | -0.296 |
were | -0.219 |
Ell | -0.217 |
everywhere | -0.217 |
grapes | -0.203 |
Token | it |
Token ID | 340 |
Feature activation | -7.35e-03 |
Pos logit contributions | |
all | +0.036 |
were | +0.029 |
everywhere | +0.027 |
Ell | +0.025 |
grapes | +0.024 |
Neg logit contributions | |
isha | -0.067 |
laptop | -0.066 |
isa | -0.066 |
acy | -0.062 |
Stacy | -0.062 |
Token | was |
Token ID | 373 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
all | +0.007 |
were | +0.005 |
everywhere | +0.005 |
grapes | +0.005 |
Ell | +0.005 |
Neg logit contributions | |
isha | -0.020 |
isa | -0.020 |
laptop | -0.020 |
acy | -0.019 |
Stacy | -0.019 |
Token | ready |
Token ID | 3492 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
all | +0.029 |
were | +0.024 |
everywhere | +0.022 |
Ell | +0.021 |
grapes | +0.020 |
Neg logit contributions | |
isha | -0.054 |
laptop | -0.053 |
isa | -0.053 |
acy | -0.050 |
Stacy | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
isha | +0.161 |
laptop | +0.160 |
isa | +0.160 |
acy | +0.151 |
Stacy | +0.150 |
Neg logit contributions | |
all | -0.077 |
were | -0.060 |
everywhere | -0.055 |
happily | -0.050 |
lots | -0.049 |
Token | puts |
Token ID | 7584 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
isha | +0.007 |
isa | +0.007 |
laptop | +0.007 |
acy | +0.006 |
Ian | +0.006 |
Neg logit contributions | |
all | -0.004 |
were | -0.003 |
everywhere | -0.003 |
happily | -0.003 |
grapes | -0.003 |
Token | bread |
Token ID | 8509 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
laptop | +0.076 |
isha | +0.076 |
isa | +0.075 |
outage | +0.071 |
acy | +0.071 |
Neg logit contributions | |
all | -0.036 |
were | -0.026 |
everywhere | -0.026 |
grapes | -0.023 |
lots | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
all | +0.051 |
were | +0.041 |
Ell | +0.041 |
Ellie | +0.037 |
Thank | +0.035 |
Neg logit contributions | |
isa | -0.121 |
laptop | -0.120 |
isha | -0.120 |
Stacy | -0.115 |
acy | -0.113 |
Token | the |
Token ID | 262 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
isha | +0.095 |
laptop | +0.094 |
isa | +0.094 |
acy | +0.089 |
Stacy | +0.088 |
Neg logit contributions | |
all | -0.036 |
were | -0.026 |
everywhere | -0.024 |
lots | -0.021 |
grapes | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | +5.66e-03 |
Pos logit contributions | |
isa | +0.025 |
isha | +0.024 |
den | +0.023 |
outage | +0.023 |
acy | +0.022 |
Neg logit contributions | |
all | -0.025 |
everywhere | -0.021 |
grapes | -0.021 |
were | -0.020 |
y | -0.020 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
isha | +0.008 |
outage | +0.008 |
isa | +0.008 |
den | +0.008 |
laptop | +0.007 |
Neg logit contributions | |
all | -0.015 |
were | -0.012 |
y | -0.012 |
happily | -0.012 |
grapes | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
isha | +0.741 |
isa | +0.731 |
laptop | +0.717 |
acy | +0.704 |
Stacy | +0.693 |
Neg logit contributions | |
all | -0.258 |
were | -0.180 |
everywhere | -0.169 |
lots | -0.143 |
happily | -0.142 |
Token | waits |
Token ID | 28364 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
all | +0.034 |
were | +0.027 |
y | +0.025 |
grapes | +0.024 |
happily | +0.024 |
Neg logit contributions | |
isha | -0.044 |
isa | -0.043 |
laptop | -0.042 |
acy | -0.042 |
outage | -0.042 |
Token | for |
Token ID | 329 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
isa | +0.051 |
isha | +0.050 |
laptop | +0.050 |
uka | +0.048 |
Ian | +0.047 |
Neg logit contributions | |
all | -0.016 |
everywhere | -0.009 |
happily | -0.008 |
were | -0.008 |
lots | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | -5.77e-03 |
Pos logit contributions | |
all | +0.047 |
everywhere | +0.027 |
fun | +0.027 |
lots | +0.027 |
were | +0.027 |
Neg logit contributions | |
laptop | -0.114 |
isha | -0.113 |
isa | -0.111 |
outage | -0.108 |
den | -0.108 |
Token | to |
Token ID | 284 |
Feature activation | +6.68e-03 |
Pos logit contributions | |
all | +0.007 |
were | +0.006 |
Ell | +0.005 |
everywhere | +0.005 |
grapes | +0.005 |
Neg logit contributions | |
laptop | -0.014 |
isha | -0.014 |
isa | -0.014 |
acy | -0.013 |
Stacy | -0.013 |
Token | some |
Token ID | 617 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
isha | +0.060 |
laptop | +0.059 |
isa | +0.059 |
acy | +0.056 |
Stacy | +0.055 |
Neg logit contributions | |
all | -0.026 |
were | -0.020 |
everywhere | -0.018 |
happily | -0.016 |
lots | -0.016 |
Token | bread |
Token ID | 8509 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
all | +0.045 |
were | +0.036 |
everywhere | +0.036 |
grapes | +0.034 |
lots | +0.033 |
Neg logit contributions | |
isha | -0.058 |
isa | -0.057 |
laptop | -0.057 |
acy | -0.054 |
Ian | -0.054 |
Token | in |
Token ID | 287 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
all | +0.051 |
were | +0.039 |
everywhere | +0.035 |
Ell | +0.034 |
Ellie | +0.032 |
Neg logit contributions | |
isha | -0.128 |
laptop | -0.128 |
isa | -0.127 |
acy | -0.120 |
Stacy | -0.119 |
Token | the |
Token ID | 262 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
isha | +0.090 |
isa | +0.089 |
laptop | +0.089 |
acy | +0.084 |
acey | +0.084 |
Neg logit contributions | |
all | -0.036 |
were | -0.024 |
everywhere | -0.022 |
lots | -0.022 |
Tim | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | +6.20e-03 |
Pos logit contributions | |
isa | +0.021 |
ergus | +0.019 |
isha | +0.019 |
acy | +0.019 |
den | +0.019 |
Neg logit contributions | |
all | -0.024 |
everywhere | -0.021 |
grapes | -0.020 |
y | -0.020 |
were | -0.019 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
isha | +0.008 |
outage | +0.008 |
isa | +0.008 |
laptop | +0.008 |
den | +0.008 |
Neg logit contributions | |
all | -0.017 |
were | -0.014 |
y | -0.013 |
lots | -0.013 |
grapes | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
isha | +0.715 |
isa | +0.706 |
laptop | +0.688 |
acy | +0.681 |
Stacy | +0.669 |
Neg logit contributions | |
all | -0.285 |
were | -0.205 |
everywhere | -0.197 |
lots | -0.168 |
happily | -0.167 |
Token | He |
Token ID | 679 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
isa | +0.040 |
isha | +0.039 |
laptop | +0.039 |
acy | +0.037 |
acey | +0.036 |
Neg logit contributions | |
all | -0.023 |
were | -0.018 |
Tim | -0.016 |
everywhere | -0.016 |
Tweet | -0.016 |
Token | goes |
Token ID | 2925 |
Feature activation | -2.33e-03 |
Pos logit contributions | |
isha | +0.019 |
isa | +0.019 |
laptop | +0.019 |
acy | +0.018 |
uka | +0.017 |
Neg logit contributions | |
all | -0.020 |
were | -0.017 |
everywhere | -0.016 |
happily | -0.015 |
grapes | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | +5.91e-03 |
Pos logit contributions | |
all | +0.002 |
everywhere | +0.001 |
happily | +0.001 |
were | +0.001 |
lots | +0.001 |
Neg logit contributions | |
isha | -0.007 |
isa | -0.007 |
laptop | -0.007 |
acy | -0.007 |
Ian | -0.007 |
Token | his |
Token ID | 465 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
isa | +0.017 |
isha | +0.017 |
laptop | +0.017 |
acy | +0.016 |
acey | +0.016 |
Neg logit contributions | |
all | -0.007 |
Tim | -0.004 |
were | -0.004 |
lots | -0.004 |
happily | -0.003 |
Token | put |
Token ID | 1234 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
all | +0.045 |
were | +0.038 |
everywhere | +0.036 |
happily | +0.034 |
grapes | +0.033 |
Neg logit contributions | |
isha | -0.058 |
laptop | -0.058 |
isa | -0.057 |
acy | -0.054 |
Stacy | -0.053 |
Token | it |
Token ID | 340 |
Feature activation | -7.59e-03 |
Pos logit contributions | |
isha | +0.068 |
laptop | +0.068 |
isa | +0.067 |
acy | +0.063 |
Stacy | +0.063 |
Neg logit contributions | |
all | -0.033 |
were | -0.026 |
everywhere | -0.024 |
happily | -0.022 |
grapes | -0.022 |
Token | in |
Token ID | 287 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
all | +0.008 |
were | +0.007 |
Ell | +0.007 |
Ellie | +0.006 |
Tweet | +0.006 |
Neg logit contributions | |
isa | -0.018 |
laptop | -0.018 |
isha | -0.018 |
acy | -0.017 |
Stacy | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
isha | +0.078 |
isa | +0.077 |
laptop | +0.077 |
acy | +0.073 |
acey | +0.072 |
Neg logit contributions | |
all | -0.031 |
were | -0.023 |
everywhere | -0.021 |
lots | -0.019 |
happily | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | +4.64e-03 |
Pos logit contributions | |
isa | +0.027 |
isha | +0.026 |
acy | +0.024 |
den | +0.024 |
outage | +0.024 |
Neg logit contributions | |
all | -0.031 |
everywhere | -0.027 |
grapes | -0.026 |
were | -0.026 |
lots | -0.025 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
isha | +0.006 |
outage | +0.005 |
isa | +0.005 |
laptop | +0.005 |
den | +0.005 |
Neg logit contributions | |
all | -0.013 |
were | -0.011 |
happily | -0.011 |
y | -0.011 |
grapes | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
isha | +0.672 |
isa | +0.665 |
laptop | +0.657 |
acy | +0.637 |
uka | +0.627 |
Neg logit contributions | |
all | -0.324 |
were | -0.239 |
everywhere | -0.234 |
happily | -0.209 |
lots | -0.209 |
Token | She |
Token ID | 1375 |
Feature activation | +3.75e-03 |
Pos logit contributions | |
laptop | +0.044 |
isha | +0.043 |
isa | +0.042 |
aster | +0.039 |
outage | +0.039 |
Neg logit contributions | |
all | -0.026 |
were | -0.021 |
everywhere | -0.019 |
happily | -0.019 |
lots | -0.018 |
Token | took |
Token ID | 1718 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
isha | +0.009 |
laptop | +0.009 |
isa | +0.009 |
acy | +0.009 |
Stacy | +0.009 |
Neg logit contributions | |
all | -0.005 |
were | -0.004 |
everywhere | -0.003 |
happily | -0.003 |
lots | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
all | +0.046 |
were | +0.031 |
everywhere | +0.030 |
lots | +0.028 |
happily | +0.027 |
Neg logit contributions | |
laptop | -0.119 |
isha | -0.116 |
isa | -0.114 |
outage | -0.110 |
pulse | -0.107 |
Token | seat |
Token ID | 5852 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
isa | +0.038 |
isha | +0.037 |
laptop | +0.037 |
acy | +0.036 |
pulse | +0.036 |
Neg logit contributions | |
all | -0.023 |
everywhere | -0.020 |
were | -0.018 |
y | -0.018 |
fun | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
laptop | +0.059 |
isha | +0.059 |
isa | +0.059 |
acy | +0.055 |
Stacy | +0.055 |
Neg logit contributions | |
all | -0.032 |
were | -0.026 |
everywhere | -0.024 |
Ell | -0.024 |
grapes | -0.023 |
Token | She |
Token ID | 1375 |
Feature activation | +5.02e-03 |
Pos logit contributions | |
isa | +0.034 |
isha | +0.034 |
laptop | +0.034 |
den | +0.031 |
acy | +0.031 |
Neg logit contributions | |
all | -0.020 |
were | -0.016 |
everywhere | -0.015 |
Tweet | -0.015 |
happily | -0.015 |
Token | watched |
Token ID | 7342 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
isha | +0.012 |
laptop | +0.012 |
isa | +0.012 |
acy | +0.011 |
Stacy | +0.011 |
Neg logit contributions | |
all | -0.007 |
were | -0.006 |
everywhere | -0.005 |
Ell | -0.005 |
Ellie | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +6.31e-03 |
Pos logit contributions | |
isha | +0.032 |
laptop | +0.032 |
isa | +0.031 |
acy | +0.030 |
Stacy | +0.030 |
Neg logit contributions | |
all | -0.014 |
were | -0.012 |
everywhere | -0.011 |
Ell | -0.010 |
grapes | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +5.27e-03 |
Pos logit contributions | |
isha | +0.010 |
laptop | +0.010 |
isa | +0.010 |
acy | +0.009 |
Stacy | +0.009 |
Neg logit contributions | |
all | -0.014 |
were | -0.012 |
everywhere | -0.012 |
Ell | -0.011 |
happily | -0.011 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
isha | +0.013 |
isa | +0.013 |
laptop | +0.013 |
acy | +0.013 |
acey | +0.012 |
Neg logit contributions | |
all | -0.007 |
were | -0.005 |
everywhere | -0.005 |
happily | -0.005 |
lots | -0.004 |
Token | with |
Token ID | 351 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
isha | +0.674 |
isa | +0.666 |
laptop | +0.664 |
acy | +0.632 |
Stacy | +0.623 |
Neg logit contributions | |
all | -0.311 |
were | -0.239 |
everywhere | -0.222 |
happily | -0.201 |
lots | -0.196 |
Token | excitement |
Token ID | 14067 |
Feature activation | -4.97e-02 |
Pos logit contributions | |
all | +0.031 |
were | +0.025 |
Ell | +0.024 |
everywhere | +0.023 |
Ellie | +0.022 |
Neg logit contributions | |
laptop | -0.059 |
isha | -0.059 |
isa | -0.058 |
aster | -0.056 |
acy | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +9.11e-03 |
Pos logit contributions | |
all | +0.062 |
were | +0.048 |
everywhere | +0.045 |
happily | +0.041 |
lots | +0.039 |
Neg logit contributions | |
isha | -0.126 |
laptop | -0.125 |
isa | -0.125 |
acy | -0.118 |
Stacy | -0.117 |
Token | She |
Token ID | 1375 |
Feature activation | -3.81e-03 |
Pos logit contributions | |
isa | +0.023 |
laptop | +0.023 |
isha | +0.023 |
acy | +0.021 |
den | +0.021 |
Neg logit contributions | |
all | -0.012 |
were | -0.009 |
everywhere | -0.009 |
Tweet | -0.008 |
lots | -0.008 |
Token | liked |
Token ID | 8288 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
all | +0.005 |
everywhere | +0.004 |
Ell | +0.004 |
were | +0.004 |
Ellie | +0.004 |
Neg logit contributions | |
isha | -0.009 |
isa | -0.009 |
laptop | -0.009 |
acy | -0.008 |
Stacy | -0.008 |
Token | see |
Token ID | 766 |
Feature activation | +9.13e-02 |
Pos logit contributions | |
isha | +0.015 |
isa | +0.015 |
laptop | +0.015 |
Stacy | +0.014 |
ason | +0.014 |
Neg logit contributions | |
all | -0.009 |
were | -0.007 |
fun | -0.006 |
everywhere | -0.006 |
happily | -0.006 |
Token | Ben |
Token ID | 3932 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
isha | +0.253 |
isa | +0.250 |
laptop | +0.250 |
acy | +0.235 |
den | +0.234 |
Neg logit contributions | |
all | -0.110 |
everywhere | -0.075 |
were | -0.071 |
lots | -0.059 |
Tim | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
isa | +0.135 |
isha | +0.134 |
laptop | +0.126 |
acy | +0.125 |
Stacy | +0.123 |
Neg logit contributions | |
all | -0.080 |
everywhere | -0.064 |
were | -0.064 |
happily | -0.053 |
lots | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
all | +0.037 |
were | +0.029 |
everywhere | +0.026 |
happily | +0.024 |
lots | +0.024 |
Neg logit contributions | |
isha | -0.069 |
laptop | -0.068 |
isa | -0.068 |
acy | -0.065 |
acey | -0.064 |
Token | to |
Token ID | 284 |
Feature activation | +6.47e-03 |
Pos logit contributions | |
isha | +0.024 |
isa | +0.024 |
laptop | +0.023 |
acy | +0.022 |
Stacy | +0.022 |
Neg logit contributions | |
all | -0.024 |
were | -0.020 |
everywhere | -0.019 |
grapes | -0.018 |
lots | -0.018 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
isha | +0.014 |
isa | +0.014 |
laptop | +0.013 |
acy | +0.013 |
acey | +0.013 |
Neg logit contributions | |
all | -0.011 |
were | -0.009 |
happily | -0.008 |
everywhere | -0.008 |
Tweet | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
isha | +0.648 |
isa | +0.642 |
laptop | +0.628 |
acy | +0.607 |
Stacy | +0.599 |
Neg logit contributions | |
all | -0.344 |
were | -0.273 |
everywhere | -0.258 |
happily | -0.226 |
lots | -0.224 |
Token |
|
Token ID | 198 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
isa | +0.038 |
laptop | +0.037 |
isha | +0.037 |
asy | +0.035 |
den | +0.035 |
Neg logit contributions | |
all | -0.014 |
were | -0.011 |
Tim | -0.010 |
everywhere | -0.009 |
lots | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
all | +0.027 |
were | +0.021 |
everywhere | +0.018 |
happily | +0.016 |
lots | +0.016 |
Neg logit contributions | |
laptop | -0.075 |
isha | -0.075 |
isa | -0.075 |
acy | -0.070 |
asy | -0.069 |
Token | The |
Token ID | 464 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
all | +0.035 |
were | +0.030 |
everywhere | +0.026 |
Thank | +0.025 |
lots | +0.025 |
Neg logit contributions | |
laptop | -0.065 |
isha | -0.065 |
isa | -0.064 |
asy | -0.060 |
den | -0.059 |
Token | toast |
Token ID | 27805 |
Feature activation | +5.77e-03 |
Pos logit contributions | |
isa | +0.088 |
isha | +0.086 |
acy | +0.083 |
Stacy | +0.083 |
Ian | +0.082 |
Neg logit contributions | |
all | -0.060 |
were | -0.048 |
grapes | -0.046 |
fun | -0.046 |
everywhere | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
all | +0.029 |
were | +0.023 |
everywhere | +0.022 |
grapes | +0.019 |
lots | +0.019 |
Neg logit contributions | |
isha | -0.058 |
laptop | -0.058 |
isa | -0.058 |
acy | -0.054 |
Stacy | -0.054 |
Token | bread |
Token ID | 8509 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
laptop | +0.031 |
isha | +0.031 |
isa | +0.030 |
acy | +0.028 |
Stacy | +0.028 |
Neg logit contributions | |
all | -0.022 |
were | -0.019 |
Ell | -0.018 |
everywhere | -0.018 |
Ellie | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
all | +0.062 |
were | +0.049 |
Ell | +0.048 |
everywhere | +0.046 |
Ellie | +0.044 |
Neg logit contributions | |
laptop | -0.118 |
isha | -0.116 |
isa | -0.116 |
acy | -0.109 |
Stacy | -0.109 |
Token | the |
Token ID | 262 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
all | +0.033 |
Ell | +0.027 |
everywhere | +0.027 |
were | +0.026 |
aws | +0.025 |
Neg logit contributions | |
laptop | -0.055 |
isa | -0.054 |
isha | -0.054 |
aster | -0.052 |
Stacy | -0.051 |
Token | to |
Token ID | 284 |
Feature activation | +7.35e-03 |
Pos logit contributions | |
laptop | +0.030 |
isha | +0.030 |
isa | +0.029 |
acy | +0.027 |
Stacy | +0.027 |
Neg logit contributions | |
all | -0.022 |
were | -0.019 |
everywhere | -0.018 |
Ell | -0.017 |
Ellie | -0.017 |
Token | aster |
Token ID | 1603 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
isha | +0.017 |
isa | +0.016 |
laptop | +0.016 |
acy | +0.015 |
den | +0.015 |
Neg logit contributions | |
all | -0.013 |
were | -0.010 |
grapes | -0.009 |
everywhere | -0.009 |
happily | -0.009 |
Token | on |
Token ID | 319 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
isha | +0.630 |
isa | +0.624 |
laptop | +0.624 |
acy | +0.588 |
Stacy | +0.582 |
Neg logit contributions | |
all | -0.346 |
were | -0.278 |
everywhere | -0.260 |
happily | -0.236 |
grapes | -0.234 |
Token | the |
Token ID | 262 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
isha | +0.081 |
isa | +0.080 |
laptop | +0.079 |
acy | +0.078 |
outage | +0.077 |
Neg logit contributions | |
all | -0.027 |
were | -0.014 |
lots | -0.014 |
everywhere | -0.012 |
happily | -0.011 |
Token | table |
Token ID | 3084 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
isa | +0.028 |
isha | +0.028 |
acy | +0.027 |
Stacy | +0.026 |
acey | +0.026 |
Neg logit contributions | |
all | -0.028 |
everywhere | -0.023 |
were | -0.022 |
grapes | -0.021 |
lots | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
isha | +0.237 |
isa | +0.236 |
laptop | +0.228 |
Stacy | +0.221 |
acy | +0.220 |
Neg logit contributions | |
all | -0.112 |
were | -0.086 |
everywhere | -0.077 |
lots | -0.069 |
happily | -0.069 |
Token | He |
Token ID | 679 |
Feature activation | +7.25e-03 |
Pos logit contributions | |
isa | +0.049 |
isha | +0.048 |
laptop | +0.047 |
outage | +0.046 |
iko | +0.045 |
Neg logit contributions | |
all | -0.018 |
Tim | -0.014 |
were | -0.014 |
happily | -0.011 |
lots | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
all | +0.023 |
were | +0.017 |
everywhere | +0.017 |
grapes | +0.017 |
Ell | +0.016 |
Neg logit contributions | |
isha | -0.054 |
isa | -0.053 |
laptop | -0.053 |
acy | -0.050 |
Stacy | -0.050 |
Token | transformed |
Token ID | 14434 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
all | +0.043 |
were | +0.037 |
Ell | +0.034 |
grapes | +0.033 |
everywhere | +0.032 |
Neg logit contributions | |
isha | -0.085 |
isa | -0.083 |
laptop | -0.083 |
Stacy | -0.079 |
acy | -0.079 |
Token | into |
Token ID | 656 |
Feature activation | +7.41e-02 |
Pos logit contributions | |
all | +0.038 |
were | +0.033 |
Ell | +0.031 |
everywhere | +0.030 |
grapes | +0.029 |
Neg logit contributions | |
isha | -0.055 |
isa | -0.054 |
laptop | -0.053 |
acy | -0.051 |
Stacy | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | -6.75e-03 |
Pos logit contributions | |
isha | +0.177 |
laptop | +0.176 |
isa | +0.175 |
acy | +0.165 |
Stacy | +0.164 |
Neg logit contributions | |
all | -0.096 |
were | -0.079 |
everywhere | -0.074 |
Ell | -0.070 |
happily | -0.067 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
all | +0.009 |
were | +0.008 |
everywhere | +0.007 |
Ell | +0.006 |
happily | +0.006 |
Neg logit contributions | |
isha | -0.016 |
laptop | -0.016 |
isa | -0.016 |
acy | -0.015 |
Stacy | -0.015 |
Token | bou |
Token ID | 35833 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
all | +0.191 |
were | +0.167 |
everywhere | +0.154 |
Ell | +0.151 |
grapes | +0.144 |
Neg logit contributions | |
isha | -0.258 |
laptop | -0.256 |
isa | -0.254 |
Stacy | -0.232 |
acy | -0.230 |
Token | quet |
Token ID | 21108 |
Feature activation | -7.57e-02 |
Pos logit contributions | |
all | +0.379 |
were | +0.334 |
everywhere | +0.310 |
lots | +0.295 |
Tim | +0.293 |
Neg logit contributions | |
isha | -0.410 |
isa | -0.398 |
laptop | -0.389 |
acy | -0.378 |
Stacy | -0.371 |
Token | of |
Token ID | 286 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
all | +0.090 |
were | +0.075 |
Ell | +0.071 |
grapes | +0.070 |
Tweet | +0.067 |
Neg logit contributions | |
isha | -0.179 |
laptop | -0.179 |
isa | -0.178 |
acy | -0.169 |
Stacy | -0.166 |
Token | flowers |
Token ID | 12734 |
Feature activation | -8.86e-02 |
Pos logit contributions | |
all | +0.029 |
Ell | +0.027 |
were | +0.024 |
happily | +0.023 |
everywhere | +0.023 |
Neg logit contributions | |
laptop | -0.057 |
isha | -0.056 |
isa | -0.055 |
Stacy | -0.053 |
aster | -0.052 |
Token | ! |
Token ID | 0 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
all | +0.110 |
were | +0.091 |
Ell | +0.085 |
everywhere | +0.081 |
Tweet | +0.081 |
Neg logit contributions | |
isha | -0.208 |
isa | -0.208 |
laptop | -0.207 |
acy | -0.194 |
Stacy | -0.192 |
Token | |
Token ID | 220 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
all | +0.019 |
were | +0.016 |
everywhere | +0.014 |
Ellie | +0.014 |
Tweet | +0.014 |
Neg logit contributions | |
laptop | -0.034 |
isha | -0.034 |
isa | -0.033 |
acy | -0.031 |
aster | -0.031 |
Token | you |
Token ID | 345 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
all | +0.039 |
were | +0.034 |
everywhere | +0.033 |
Ell | +0.032 |
happily | +0.031 |
Neg logit contributions | |
isa | -0.045 |
isha | -0.045 |
laptop | -0.045 |
aster | -0.041 |
Stacy | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -4.79e-03 |
Pos logit contributions | |
isha | +0.047 |
laptop | +0.047 |
isa | +0.046 |
acy | +0.043 |
Stacy | +0.043 |
Neg logit contributions | |
all | -0.031 |
were | -0.026 |
everywhere | -0.024 |
Ell | -0.024 |
grapes | -0.023 |
Token | And |
Token ID | 843 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
all | +0.008 |
were | +0.007 |
Tim | +0.007 |
Thank | +0.007 |
Tweet | +0.006 |
Neg logit contributions | |
isa | -0.010 |
laptop | -0.010 |
isha | -0.010 |
acy | -0.009 |
uka | -0.009 |
Token | we |
Token ID | 356 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
isha | +0.032 |
laptop | +0.031 |
isa | +0.031 |
Stacy | +0.029 |
acy | +0.029 |
Neg logit contributions | |
all | -0.031 |
were | -0.027 |
everywhere | -0.026 |
Ell | -0.025 |
grapes | -0.024 |
Token | love |
Token ID | 1842 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
isha | +0.117 |
laptop | +0.114 |
isa | +0.114 |
acey | +0.111 |
acy | +0.109 |
Neg logit contributions | |
all | -0.060 |
Ell | -0.055 |
grapes | -0.053 |
Ellie | -0.051 |
everywhere | -0.050 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
all | +0.035 |
were | +0.030 |
everywhere | +0.028 |
Ell | +0.027 |
happily | +0.026 |
Neg logit contributions | |
isha | -0.052 |
isa | -0.052 |
laptop | -0.052 |
acy | -0.048 |
Stacy | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -8.02e-03 |
Pos logit contributions | |
all | +0.351 |
Ell | +0.306 |
were | +0.302 |
everywhere | +0.295 |
Ellie | +0.287 |
Neg logit contributions | |
laptop | -0.404 |
isha | -0.387 |
isa | -0.385 |
Stacy | -0.346 |
acy | -0.346 |
Token | But |
Token ID | 887 |
Feature activation | +7.10e-02 |
Pos logit contributions | |
all | +0.013 |
were | +0.011 |
Thank | +0.010 |
Tim | +0.010 |
Tweet | +0.010 |
Neg logit contributions | |
isa | -0.017 |
isha | -0.017 |
laptop | -0.017 |
acy | -0.016 |
uka | -0.016 |
Token | we |
Token ID | 356 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
laptop | +0.131 |
isha | +0.130 |
isa | +0.129 |
Stacy | +0.121 |
acy | +0.121 |
Neg logit contributions | |
all | -0.118 |
Ell | -0.108 |
were | -0.103 |
everywhere | -0.102 |
aws | -0.100 |
Token | don |
Token ID | 836 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
isha | +0.109 |
laptop | +0.108 |
isa | +0.107 |
acy | +0.102 |
acey | +0.101 |
Neg logit contributions | |
all | -0.074 |
Ell | -0.064 |
everywhere | -0.062 |
grapes | -0.060 |
Ellie | -0.060 |
Token | 't |
Token ID | 470 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
isa | +0.096 |
isha | +0.096 |
laptop | +0.094 |
acy | +0.087 |
Stacy | +0.087 |
Neg logit contributions | |
all | -0.086 |
were | -0.072 |
everywhere | -0.068 |
Ellie | -0.063 |
lots | -0.063 |
Token | Tim |
Token ID | 5045 |
Feature activation | -8.30e-02 |
Pos logit contributions | |
all | +0.059 |
were | +0.049 |
everywhere | +0.047 |
Ell | +0.044 |
lots | +0.043 |
Neg logit contributions | |
isha | -0.098 |
laptop | -0.098 |
isa | -0.098 |
Stacy | -0.091 |
Ian | -0.091 |
Token | always |
Token ID | 1464 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
all | +0.127 |
grapes | +0.107 |
everywhere | +0.107 |
Ell | +0.106 |
Ellie | +0.104 |
Neg logit contributions | |
isa | -0.170 |
isha | -0.169 |
laptop | -0.161 |
acy | -0.156 |
acey | -0.154 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.15e-04 |
Pos logit contributions | |
all | +0.027 |
were | +0.020 |
everywhere | +0.018 |
grapes | +0.016 |
Ell | +0.016 |
Neg logit contributions | |
isha | -0.074 |
isa | -0.073 |
laptop | -0.073 |
acy | -0.069 |
Stacy | -0.069 |
Token | for |
Token ID | 329 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
isha | +0.000 |
isa | +0.000 |
laptop | +0.000 |
Stacy | +0.000 |
acy | +0.000 |
Neg logit contributions | |
were | -0.000 |
Ell | -0.000 |
all | -0.000 |
Thank | -0.000 |
Tweet | -0.000 |
Token | unknown |
Token ID | 6439 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
Ell | +0.070 |
all | +0.061 |
were | +0.060 |
By | +0.059 |
Thank | +0.058 |
Neg logit contributions | |
laptop | -0.112 |
isha | -0.110 |
isa | -0.109 |
aster | -0.108 |
Stacy | -0.104 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
all | +0.064 |
Ell | +0.062 |
By | +0.057 |
were | +0.053 |
everywhere | +0.053 |
Neg logit contributions | |
laptop | -0.100 |
isa | -0.099 |
isha | -0.095 |
Stacy | -0.089 |
aster | -0.088 |
Token | to |
Token ID | 284 |
Feature activation | -7.62e-03 |
Pos logit contributions | |
all | +0.249 |
Ell | +0.223 |
Thank | +0.209 |
Tweet | +0.204 |
were | +0.204 |
Neg logit contributions | |
laptop | -0.475 |
isha | -0.463 |
isa | -0.462 |
Stacy | -0.434 |
acy | -0.430 |
Token | share |
Token ID | 2648 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
all | +0.008 |
everywhere | +0.007 |
Ell | +0.007 |
were | +0.007 |
Thank | +0.006 |
Neg logit contributions | |
isa | -0.018 |
laptop | -0.018 |
isha | -0.018 |
aster | -0.017 |
Stacy | -0.017 |
Token | with |
Token ID | 351 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
Ell | +0.061 |
all | +0.058 |
By | +0.056 |
Thank | +0.055 |
ashed | +0.054 |
Neg logit contributions | |
laptop | -0.098 |
isha | -0.096 |
isa | -0.096 |
aster | -0.092 |
Stacy | -0.087 |
Token | each |
Token ID | 1123 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
all | +0.035 |
were | +0.027 |
Ell | +0.027 |
everywhere | +0.027 |
Thank | +0.024 |
Neg logit contributions | |
isha | -0.098 |
isa | -0.097 |
laptop | -0.095 |
Stacy | -0.090 |
acy | -0.090 |
Token | other |
Token ID | 584 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
all | +0.067 |
were | +0.056 |
Ell | +0.055 |
everywhere | +0.053 |
Tweet | +0.053 |
Neg logit contributions | |
isha | -0.107 |
isa | -0.107 |
laptop | -0.105 |
aster | -0.100 |
acy | -0.099 |
Token | love |
Token ID | 1842 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
laptop | +0.043 |
isha | +0.042 |
isa | +0.042 |
acy | +0.039 |
Stacy | +0.039 |
Neg logit contributions | |
all | -0.025 |
were | -0.020 |
everywhere | -0.019 |
Ell | -0.019 |
grapes | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -4.88e-04 |
Pos logit contributions | |
all | +0.034 |
Ell | +0.029 |
were | +0.027 |
Tweet | +0.027 |
grapes | +0.026 |
Neg logit contributions | |
isha | -0.079 |
laptop | -0.078 |
isa | -0.077 |
Stacy | -0.071 |
acy | -0.071 |
Token | They |
Token ID | 1119 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
all | +0.001 |
were | +0.001 |
everywhere | +0.001 |
happily | +0.001 |
lots | +0.001 |
Neg logit contributions | |
isha | -0.001 |
laptop | -0.001 |
isa | -0.001 |
acy | -0.001 |
Stacy | -0.001 |
Token | threw |
Token ID | 9617 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
grapes | +0.028 |
Ell | +0.026 |
all | +0.025 |
Tweet | +0.024 |
everywhere | +0.022 |
Neg logit contributions | |
isha | -0.084 |
isa | -0.083 |
acy | -0.079 |
Stacy | -0.078 |
acey | -0.078 |
Token | the |
Token ID | 262 |
Feature activation | -7.08e-03 |
Pos logit contributions | |
Ell | +0.067 |
all | +0.063 |
Thank | +0.059 |
were | +0.056 |
Tweet | +0.056 |
Neg logit contributions | |
laptop | -0.131 |
isa | -0.129 |
isha | -0.126 |
Stacy | -0.120 |
Ian | -0.120 |
Token | bou |
Token ID | 35833 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
all | +0.011 |
were | +0.010 |
Ell | +0.009 |
Tweet | +0.009 |
happily | +0.009 |
Neg logit contributions | |
laptop | -0.015 |
isha | -0.014 |
isa | -0.014 |
acy | -0.013 |
acey | -0.013 |
Token | quet |
Token ID | 21108 |
Feature activation | -7.40e-02 |
Pos logit contributions | |
all | +0.413 |
were | +0.348 |
everywhere | +0.342 |
Tim | +0.337 |
fun | +0.329 |
Neg logit contributions | |
isha | -0.353 |
isa | -0.350 |
acy | -0.327 |
Stacy | -0.326 |
acey | -0.321 |
Token | of |
Token ID | 286 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
Ell | +0.089 |
all | +0.079 |
grapes | +0.079 |
Tweet | +0.075 |
aws | +0.075 |
Neg logit contributions | |
laptop | -0.154 |
isa | -0.153 |
isha | -0.151 |
acy | -0.147 |
Stacy | -0.145 |
Token | flowers |
Token ID | 12734 |
Feature activation | -8.61e-02 |
Pos logit contributions | |
all | +0.026 |
Ell | +0.026 |
everywhere | +0.021 |
Tweet | +0.021 |
Thank | +0.021 |
Neg logit contributions | |
isha | -0.048 |
laptop | -0.048 |
isa | -0.047 |
Stacy | -0.045 |
acey | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
all | +0.101 |
Ell | +0.095 |
Thank | +0.090 |
grapes | +0.084 |
were | +0.084 |
Neg logit contributions | |
isa | -0.185 |
laptop | -0.185 |
isha | -0.181 |
Stacy | -0.172 |
acy | -0.170 |
Token | the |
Token ID | 262 |
Feature activation | -5.80e-03 |
Pos logit contributions | |
all | +0.038 |
Ell | +0.032 |
aws | +0.031 |
everywhere | +0.029 |
were | +0.028 |
Neg logit contributions | |
isa | -0.077 |
isha | -0.076 |
laptop | -0.075 |
Stacy | -0.072 |
Macy | -0.070 |
Token | you |
Token ID | 345 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
isha | +0.185 |
isa | +0.182 |
laptop | +0.181 |
Stacy | +0.174 |
acy | +0.173 |
Neg logit contributions | |
all | -0.204 |
everywhere | -0.180 |
happily | -0.177 |
were | -0.175 |
Ell | -0.170 |
Token | want |
Token ID | 765 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
laptop | +0.056 |
isa | +0.056 |
isha | +0.056 |
Stacy | +0.052 |
Ian | +0.051 |
Neg logit contributions | |
all | -0.029 |
were | -0.022 |
everywhere | -0.019 |
happily | -0.019 |
fun | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | -3.60e-03 |
Pos logit contributions | |
all | +0.013 |
were | +0.010 |
everywhere | +0.009 |
grapes | +0.008 |
happily | +0.008 |
Neg logit contributions | |
isha | -0.032 |
laptop | -0.032 |
isa | -0.032 |
acy | -0.030 |
Stacy | -0.030 |
Token | make |
Token ID | 787 |
Feature activation | +9.48e-02 |
Pos logit contributions | |
all | +0.006 |
were | +0.004 |
everywhere | +0.004 |
fun | +0.004 |
Tweet | +0.004 |
Neg logit contributions | |
isha | -0.009 |
isa | -0.009 |
laptop | -0.009 |
den | -0.008 |
acy | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -2.13e-03 |
Pos logit contributions | |
isha | +0.227 |
laptop | +0.224 |
isa | +0.224 |
acy | +0.212 |
Stacy | +0.211 |
Neg logit contributions | |
all | -0.124 |
were | -0.104 |
everywhere | -0.096 |
Ell | -0.090 |
happily | -0.087 |
Token | bou |
Token ID | 35833 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
all | +0.003 |
were | +0.003 |
Ell | +0.002 |
everywhere | +0.002 |
Ellie | +0.002 |
Neg logit contributions | |
laptop | -0.005 |
isha | -0.005 |
isa | -0.005 |
Stacy | -0.004 |
acy | -0.004 |
Token | quet |
Token ID | 21108 |
Feature activation | -7.23e-02 |
Pos logit contributions | |
all | +0.360 |
were | +0.324 |
everywhere | +0.303 |
Tim | +0.280 |
lots | +0.276 |
Neg logit contributions | |
isha | -0.407 |
laptop | -0.401 |
isa | -0.395 |
acy | -0.382 |
Stacy | -0.378 |
Token | for |
Token ID | 329 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
all | +0.111 |
were | +0.089 |
everywhere | +0.085 |
happily | +0.078 |
lots | +0.077 |
Neg logit contributions | |
isha | -0.164 |
isa | -0.163 |
laptop | -0.162 |
acy | -0.151 |
Stacy | -0.150 |
Token | the |
Token ID | 262 |
Feature activation | -6.73e-04 |
Pos logit contributions | |
all | +0.067 |
were | +0.053 |
everywhere | +0.050 |
grapes | +0.046 |
happily | +0.046 |
Neg logit contributions | |
isha | -0.114 |
laptop | -0.113 |
isa | -0.113 |
acy | -0.106 |
Stacy | -0.105 |
Token | kitchen |
Token ID | 9592 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
all | +0.001 |
were | +0.001 |
Ell | +0.001 |
everywhere | +0.001 |
happily | +0.001 |
Neg logit contributions | |
laptop | -0.002 |
isha | -0.001 |
isa | -0.001 |
Stacy | -0.001 |
acy | -0.001 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
isa | +0.280 |
isha | +0.270 |
laptop | +0.252 |
acy | +0.252 |
Ian | +0.242 |
Neg logit contributions | |
all | -0.390 |
were | -0.322 |
everywhere | -0.317 |
fun | -0.305 |
lots | -0.302 |
Token | had |
Token ID | 550 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
laptop | +0.183 |
isha | +0.182 |
isa | +0.180 |
acy | +0.172 |
Stacy | +0.169 |
Neg logit contributions | |
all | -0.086 |
were | -0.069 |
everywhere | -0.063 |
happily | -0.056 |
Ellie | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -4.68e-03 |
Pos logit contributions | |
all | +0.064 |
were | +0.052 |
Ell | +0.051 |
everywhere | +0.051 |
Tweet | +0.051 |
Neg logit contributions | |
isha | -0.107 |
laptop | -0.105 |
isa | -0.105 |
acy | -0.099 |
Stacy | -0.098 |
Token | They |
Token ID | 1119 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
all | +0.006 |
were | +0.005 |
everywhere | +0.005 |
happily | +0.004 |
lots | +0.004 |
Neg logit contributions | |
laptop | -0.011 |
isha | -0.011 |
isa | -0.011 |
acy | -0.011 |
Stacy | -0.010 |
Token | admired |
Token ID | 29382 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
grapes | +0.031 |
Ell | +0.027 |
Tweet | +0.026 |
nuts | +0.025 |
Ellie | +0.023 |
Neg logit contributions | |
isa | -0.085 |
isha | -0.081 |
acy | -0.080 |
Stacy | -0.078 |
acey | -0.077 |
Token | the |
Token ID | 262 |
Feature activation | -5.32e-03 |
Pos logit contributions | |
all | +0.059 |
Ell | +0.052 |
were | +0.050 |
grapes | +0.047 |
everywhere | +0.045 |
Neg logit contributions | |
isa | -0.124 |
isha | -0.123 |
laptop | -0.119 |
Ian | -0.115 |
acy | -0.114 |
Token | bou |
Token ID | 35833 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
all | +0.006 |
Ell | +0.006 |
were | +0.006 |
everywhere | +0.005 |
Tweet | +0.005 |
Neg logit contributions | |
isha | -0.012 |
isa | -0.012 |
laptop | -0.012 |
Stacy | -0.011 |
acy | -0.011 |
Token | quet |
Token ID | 21108 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
all | +0.402 |
were | +0.350 |
everywhere | +0.340 |
lots | +0.317 |
Tim | +0.317 |
Neg logit contributions | |
isha | -0.376 |
isa | -0.372 |
acy | -0.345 |
Stacy | -0.340 |
laptop | -0.331 |
Token | together |
Token ID | 1978 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
all | +0.076 |
Ell | +0.066 |
grapes | +0.063 |
were | +0.062 |
Tweet | +0.060 |
Neg logit contributions | |
isa | -0.181 |
isha | -0.180 |
laptop | -0.178 |
acy | -0.170 |
Stacy | -0.168 |
Token | and |
Token ID | 290 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
all | +0.039 |
Ell | +0.035 |
Tweet | +0.032 |
were | +0.032 |
grapes | +0.032 |
Neg logit contributions | |
isa | -0.087 |
isha | -0.087 |
laptop | -0.085 |
acy | -0.081 |
Ian | -0.080 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
all | +0.045 |
Ell | +0.039 |
grapes | +0.039 |
everywhere | +0.037 |
aws | +0.037 |
Neg logit contributions | |
isa | -0.098 |
isha | -0.095 |
laptop | -0.092 |
Ian | -0.091 |
Stacy | -0.090 |
Token | the |
Token ID | 262 |
Feature activation | -5.46e-03 |
Pos logit contributions | |
all | +0.062 |
Ell | +0.058 |
were | +0.054 |
everywhere | +0.053 |
grapes | +0.050 |
Neg logit contributions | |
isa | -0.115 |
isha | -0.114 |
laptop | -0.110 |
Stacy | -0.107 |
Ian | -0.107 |
Token | were |
Token ID | 547 |
Feature activation | -8.50e-02 |
Pos logit contributions | |
Ell | +0.022 |
grapes | +0.021 |
Ellie | +0.020 |
all | +0.020 |
fun | +0.019 |
Neg logit contributions | |
isa | -0.054 |
isha | -0.054 |
acy | -0.052 |
acey | -0.051 |
Stacy | -0.050 |
Token | always |
Token ID | 1464 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
all | +0.093 |
were | +0.081 |
everywhere | +0.077 |
Ell | +0.073 |
Ellie | +0.067 |
Neg logit contributions | |
isha | -0.211 |
isa | -0.210 |
laptop | -0.202 |
acy | -0.198 |
Stacy | -0.197 |
Token | together |
Token ID | 1978 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
all | +0.033 |
were | +0.029 |
everywhere | +0.027 |
Ell | +0.026 |
Ellie | +0.025 |
Neg logit contributions | |
isha | -0.061 |
isa | -0.061 |
laptop | -0.060 |
Stacy | -0.057 |
acy | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
all | +0.045 |
Ell | +0.044 |
Ellie | +0.039 |
were | +0.039 |
Tweet | +0.039 |
Neg logit contributions | |
isha | -0.086 |
isa | -0.085 |
laptop | -0.083 |
acy | -0.079 |
aster | -0.079 |
Token | eating |
Token ID | 6600 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
Ell | +0.014 |
all | +0.014 |
aws | +0.012 |
everywhere | +0.012 |
were | +0.012 |
Neg logit contributions | |
laptop | -0.025 |
isha | -0.025 |
isa | -0.024 |
Stacy | -0.023 |
Ian | -0.023 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
Ell | +0.039 |
By | +0.031 |
Ellie | +0.028 |
Ag | +0.027 |
chased | +0.026 |
Neg logit contributions | |
aster | -0.058 |
laptop | -0.057 |
microwave | -0.052 |
isa | -0.052 |
Mason | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
Ell | +0.248 |
all | +0.237 |
Wow | +0.229 |
Tweet | +0.215 |
Ellie | +0.213 |
Neg logit contributions | |
laptop | -0.447 |
isa | -0.433 |
isha | -0.424 |
Stacy | -0.403 |
acy | -0.395 |
Token | enjoying |
Token ID | 13226 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
Ell | +0.056 |
all | +0.055 |
aws | +0.055 |
everywhere | +0.050 |
By | +0.048 |
Neg logit contributions | |
isa | -0.079 |
laptop | -0.079 |
aster | -0.078 |
isha | -0.074 |
throat | -0.073 |
Token | each |
Token ID | 1123 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
Ell | +0.020 |
were | +0.018 |
all | +0.017 |
everywhere | +0.016 |
ashed | +0.016 |
Neg logit contributions | |
isa | -0.033 |
aster | -0.031 |
isha | -0.031 |
laptop | -0.031 |
Stacy | -0.030 |
Token | other |
Token ID | 584 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
all | +0.070 |
Ell | +0.064 |
were | +0.060 |
everywhere | +0.060 |
Tweet | +0.059 |
Neg logit contributions | |
isa | -0.108 |
laptop | -0.104 |
isha | -0.104 |
aster | -0.103 |
acy | -0.098 |
Token | 's |
Token ID | 338 |
Feature activation | +9.55e-02 |
Pos logit contributions | |
all | +0.021 |
were | +0.018 |
everywhere | +0.017 |
Ell | +0.016 |
grapes | +0.016 |
Neg logit contributions | |
isha | -0.022 |
laptop | -0.022 |
isa | -0.022 |
acy | -0.021 |
Stacy | -0.020 |
Token | € |
Token ID | 26391 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
all | +0.039 |
Ell | +0.035 |
Thank | +0.034 |
were | +0.034 |
Ellie | +0.033 |
Neg logit contributions | |
laptop | -0.038 |
isa | -0.038 |
isha | -0.038 |
acy | -0.035 |
Stacy | -0.034 |
Token | � |
Token ID | 447 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
isha | +0.262 |
laptop | +0.260 |
isa | +0.259 |
acy | +0.239 |
Stacy | +0.236 |
Neg logit contributions | |
all | -0.241 |
were | -0.205 |
everywhere | -0.197 |
happily | -0.184 |
Ell | -0.184 |
Token | � |
Token ID | 250 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
isha | +0.232 |
laptop | +0.228 |
isa | +0.227 |
acy | +0.205 |
Stacy | +0.203 |
Neg logit contributions | |
all | -0.372 |
were | -0.327 |
everywhere | -0.319 |
grapes | -0.304 |
Ell | -0.302 |
Token | a |
Token ID | 257 |
Feature activation | +1.07e-03 |
Pos logit contributions | |
all | +0.029 |
were | +0.024 |
everywhere | +0.023 |
grapes | +0.021 |
happily | +0.021 |
Neg logit contributions | |
isha | -0.047 |
isa | -0.046 |
laptop | -0.046 |
acy | -0.044 |
Stacy | -0.043 |
Token | lively |
Token ID | 29696 |
Feature activation | -6.72e-02 |
Pos logit contributions | |
isha | +0.003 |
laptop | +0.003 |
isa | +0.003 |
acy | +0.003 |
Stacy | +0.003 |
Neg logit contributions | |
all | -0.001 |
were | -0.001 |
everywhere | -0.001 |
Ell | -0.001 |
grapes | -0.001 |
Token | bou |
Token ID | 35833 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
all | +0.104 |
were | +0.090 |
everywhere | +0.085 |
Ell | +0.082 |
grapes | +0.080 |
Neg logit contributions | |
isha | -0.142 |
laptop | -0.142 |
isa | -0.139 |
acy | -0.130 |
Stacy | -0.130 |
Token | quet |
Token ID | 21108 |
Feature activation | -7.02e-02 |
Pos logit contributions | |
all | +0.365 |
were | +0.314 |
everywhere | +0.301 |
fun | +0.278 |
lots | +0.278 |
Neg logit contributions | |
isha | -0.420 |
isa | -0.412 |
laptop | -0.397 |
acy | -0.390 |
Stacy | -0.384 |
Token | of |
Token ID | 286 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
all | +0.091 |
were | +0.073 |
everywhere | +0.070 |
Ell | +0.068 |
grapes | +0.067 |
Neg logit contributions | |
isha | -0.165 |
laptop | -0.164 |
isa | -0.163 |
acy | -0.154 |
Stacy | -0.151 |
Token | flowers |
Token ID | 12734 |
Feature activation | -8.44e-02 |
Pos logit contributions | |
all | +0.023 |
Ell | +0.021 |
were | +0.020 |
happily | +0.019 |
everywhere | +0.019 |
Neg logit contributions | |
laptop | -0.043 |
isha | -0.042 |
isa | -0.041 |
aster | -0.039 |
Stacy | -0.039 |
Token | ! |
Token ID | 0 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
all | +0.114 |
were | +0.092 |
everywhere | +0.087 |
Ell | +0.080 |
happily | +0.080 |
Neg logit contributions | |
isha | -0.199 |
laptop | -0.198 |
isa | -0.197 |
acy | -0.185 |
Stacy | -0.183 |
Token | |
Token ID | 220 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
all | +0.019 |
were | +0.017 |
Ellie | +0.017 |
Tim | +0.017 |
Tweet | +0.016 |
Neg logit contributions | |
laptop | -0.025 |
isa | -0.024 |
isha | -0.024 |
pulse | -0.023 |
asy | -0.023 |
Token | around |
Token ID | 1088 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
Ell | +0.020 |
Thank | +0.016 |
By | +0.016 |
Jerry | +0.016 |
all | +0.015 |
Neg logit contributions | |
isa | -0.030 |
laptop | -0.029 |
isha | -0.028 |
den | -0.028 |
Stacy | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
all | +0.019 |
Ell | +0.017 |
were | +0.016 |
Thank | +0.015 |
aws | +0.014 |
Neg logit contributions | |
laptop | -0.040 |
isa | -0.040 |
isha | -0.040 |
Stacy | -0.037 |
Ian | -0.036 |
Token | admired |
Token ID | 29382 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
all | +0.034 |
Ell | +0.029 |
aws | +0.028 |
grapes | +0.028 |
everywhere | +0.027 |
Neg logit contributions | |
isa | -0.074 |
isha | -0.073 |
laptop | -0.072 |
Stacy | -0.069 |
Ian | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | -4.03e-03 |
Pos logit contributions | |
all | +0.049 |
Ell | +0.047 |
were | +0.044 |
Thank | +0.041 |
grapes | +0.040 |
Neg logit contributions | |
isa | -0.102 |
isha | -0.102 |
laptop | -0.099 |
Stacy | -0.096 |
Ian | -0.095 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
all | +0.005 |
were | +0.005 |
Ell | +0.004 |
everywhere | +0.004 |
happily | +0.004 |
Neg logit contributions | |
laptop | -0.009 |
isha | -0.009 |
isa | -0.009 |
acy | -0.008 |
clipboard | -0.008 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
all | +0.172 |
were | +0.145 |
Ell | +0.136 |
everywhere | +0.136 |
happily | +0.126 |
Neg logit contributions | |
laptop | -0.267 |
isha | -0.265 |
isa | -0.264 |
aster | -0.239 |
Stacy | -0.238 |
Token | . |
Token ID | 13 |
Feature activation | +1.87e-04 |
Pos logit contributions | |
all | +0.259 |
Ell | +0.230 |
were | +0.213 |
Tweet | +0.209 |
Thank | +0.205 |
Neg logit contributions | |
laptop | -0.469 |
isa | -0.466 |
isha | -0.463 |
acy | -0.437 |
Stacy | -0.435 |
Token | |
Token ID | 220 |
Feature activation | +4.65e-02 |
Pos logit contributions | |
laptop | +0.000 |
isha | +0.000 |
isa | +0.000 |
acy | +0.000 |
Stacy | +0.000 |
Neg logit contributions | |
all | -0.000 |
were | -0.000 |
everywhere | -0.000 |
happily | -0.000 |
Ellie | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -4.65e-02 |
Pos logit contributions | |
isha | +0.110 |
laptop | +0.109 |
isa | +0.109 |
acy | +0.103 |
Stacy | +0.101 |
Neg logit contributions | |
all | -0.063 |
were | -0.051 |
everywhere | -0.048 |
grapes | -0.044 |
Ell | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
all | +0.062 |
were | +0.050 |
everywhere | +0.047 |
happily | +0.042 |
Ell | +0.042 |
Neg logit contributions | |
isha | -0.112 |
laptop | -0.111 |
isa | -0.111 |
acy | -0.104 |
Stacy | -0.103 |
Token | â |
Token ID | 22940 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
all | +0.071 |
were | +0.059 |
everywhere | +0.058 |
grapes | +0.053 |
Tweet | +0.051 |
Neg logit contributions | |
isha | -0.094 |
isa | -0.093 |
laptop | -0.091 |
acy | -0.087 |
Stacy | -0.086 |
Token | my |
Token ID | 1820 |
Feature activation | +6.96e-02 |
Pos logit contributions | |
all | +0.167 |
were | +0.141 |
everywhere | +0.135 |
Ell | +0.128 |
grapes | +0.128 |
Neg logit contributions | |
isha | -0.204 |
isa | -0.202 |
laptop | -0.200 |
acy | -0.187 |
Stacy | -0.184 |
Token | always |
Token ID | 1464 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
isha | +0.172 |
isa | +0.169 |
laptop | +0.164 |
acy | +0.160 |
Stacy | +0.159 |
Neg logit contributions | |
all | -0.076 |
Ell | -0.062 |
everywhere | -0.061 |
grapes | -0.058 |
Ellie | -0.058 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.02e-03 |
Pos logit contributions | |
all | +0.037 |
were | +0.029 |
everywhere | +0.026 |
happily | +0.024 |
lots | +0.024 |
Neg logit contributions | |
isha | -0.074 |
laptop | -0.074 |
isa | -0.073 |
acy | -0.069 |
Stacy | -0.068 |
Token | for |
Token ID | 329 |
Feature activation | -5.26e-02 |
Pos logit contributions | |
isha | +0.002 |
laptop | +0.002 |
isa | +0.002 |
Stacy | +0.002 |
acy | +0.002 |
Neg logit contributions | |
were | -0.001 |
all | -0.001 |
Ell | -0.001 |
Tweet | -0.001 |
Thank | -0.001 |
Token | unique |
Token ID | 3748 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
Ell | +0.063 |
all | +0.060 |
were | +0.059 |
By | +0.058 |
Thank | +0.057 |
Neg logit contributions | |
laptop | -0.109 |
isha | -0.108 |
aster | -0.104 |
isa | -0.102 |
Stacy | -0.100 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
all | +0.077 |
Ell | +0.072 |
were | +0.067 |
By | +0.066 |
everywhere | +0.063 |
Neg logit contributions | |
laptop | -0.135 |
isha | -0.133 |
isa | -0.132 |
Stacy | -0.124 |
aster | -0.122 |
Token | when |
Token ID | 618 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
all | +0.253 |
Ell | +0.224 |
were | +0.210 |
Thank | +0.203 |
Tweet | +0.202 |
Neg logit contributions | |
laptop | -0.470 |
isha | -0.456 |
isa | -0.451 |
Stacy | -0.432 |
acy | -0.428 |
Token | he |
Token ID | 339 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
isha | +0.232 |
laptop | +0.231 |
isa | +0.230 |
acy | +0.215 |
Stacy | +0.213 |
Neg logit contributions | |
all | -0.154 |
were | -0.128 |
everywhere | -0.123 |
Ell | -0.119 |
happily | -0.112 |
Token | ate |
Token ID | 15063 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
all | +0.029 |
Ell | +0.023 |
everywhere | +0.022 |
Ellie | +0.022 |
were | +0.021 |
Neg logit contributions | |
isa | -0.063 |
isha | -0.063 |
laptop | -0.061 |
acy | -0.059 |
Stacy | -0.059 |
Token | them |
Token ID | 606 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
Ell | +0.140 |
all | +0.117 |
were | +0.112 |
By | +0.111 |
Thank | +0.111 |
Neg logit contributions | |
aster | -0.242 |
laptop | -0.239 |
isha | -0.234 |
isa | -0.233 |
Mason | -0.227 |
Token | . |
Token ID | 13 |
Feature activation | -6.51e-03 |
Pos logit contributions | |
all | +0.058 |
Ell | +0.056 |
were | +0.052 |
Thank | +0.050 |
everywhere | +0.047 |
Neg logit contributions | |
laptop | -0.111 |
isa | -0.109 |
isha | -0.108 |
acy | -0.103 |
Stacy | -0.101 |
Token | careful |
Token ID | 8161 |
Feature activation | +6.44e-02 |
Pos logit contributions | |
isha | +0.018 |
isa | +0.018 |
laptop | +0.018 |
acy | +0.017 |
Stacy | +0.017 |
Neg logit contributions | |
all | -0.012 |
were | -0.010 |
everywhere | -0.010 |
Ell | -0.009 |
Tweet | -0.009 |
Token | where |
Token ID | 810 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
isha | +0.170 |
laptop | +0.169 |
isa | +0.168 |
acy | +0.159 |
Stacy | +0.158 |
Neg logit contributions | |
all | -0.073 |
were | -0.055 |
everywhere | -0.051 |
happily | -0.045 |
lots | -0.045 |
Token | she |
Token ID | 673 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
isha | +0.078 |
laptop | +0.078 |
isa | +0.077 |
Stacy | +0.074 |
acy | +0.073 |
Neg logit contributions | |
all | -0.049 |
were | -0.041 |
everywhere | -0.040 |
Ell | -0.039 |
happily | -0.039 |
Token | spread |
Token ID | 4104 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
all | +0.045 |
everywhere | +0.034 |
were | +0.033 |
Ell | +0.033 |
Ellie | +0.031 |
Neg logit contributions | |
isa | -0.097 |
isha | -0.097 |
laptop | -0.097 |
acy | -0.092 |
Stacy | -0.091 |
Token | the |
Token ID | 262 |
Feature activation | -3.64e-03 |
Pos logit contributions | |
laptop | +0.030 |
isha | +0.030 |
isa | +0.030 |
aster | +0.030 |
Stacy | +0.029 |
Neg logit contributions | |
all | -0.015 |
Ell | -0.014 |
were | -0.014 |
Thank | -0.012 |
everywhere | -0.012 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
all | +0.004 |
Ell | +0.003 |
were | +0.003 |
Tweet | +0.003 |
Tim | +0.003 |
Neg logit contributions | |
laptop | -0.009 |
isha | -0.008 |
isa | -0.008 |
aster | -0.008 |
acy | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -6.53e-03 |
Pos logit contributions | |
all | +0.249 |
Ell | +0.219 |
were | +0.201 |
Tweet | +0.200 |
Ellie | +0.195 |
Neg logit contributions | |
laptop | -0.475 |
isha | -0.457 |
isa | -0.455 |
acy | -0.442 |
Stacy | -0.434 |
Token | And |
Token ID | 843 |
Feature activation | +7.04e-03 |
Pos logit contributions | |
all | +0.009 |
were | +0.007 |
everywhere | +0.006 |
Ellie | +0.006 |
Tweet | +0.006 |
Neg logit contributions | |
laptop | -0.016 |
isa | -0.016 |
isha | -0.016 |
aster | -0.015 |
asy | -0.015 |
Token | she |
Token ID | 673 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
isha | +0.017 |
isa | +0.017 |
laptop | +0.016 |
Stacy | +0.016 |
acy | +0.016 |
Neg logit contributions | |
all | -0.009 |
were | -0.007 |
everywhere | -0.007 |
grapes | -0.007 |
Ell | -0.007 |
Token | learned |
Token ID | 4499 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
all | +0.042 |
everywhere | +0.036 |
Ell | +0.035 |
grapes | +0.035 |
Ellie | +0.034 |
Neg logit contributions | |
isa | -0.096 |
isha | -0.095 |
acy | -0.090 |
laptop | -0.089 |
Stacy | -0.088 |
Token | that |
Token ID | 326 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
all | +0.048 |
were | +0.042 |
everywhere | +0.037 |
Ell | +0.036 |
grapes | +0.034 |
Neg logit contributions | |
isha | -0.118 |
laptop | -0.117 |
isa | -0.116 |
Stacy | -0.111 |
acy | -0.110 |
Token | little |
Token ID | 1310 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
all | +0.113 |
were | +0.101 |
everywhere | +0.097 |
aws | +0.096 |
Ell | +0.096 |
Neg logit contributions | |
laptop | -0.142 |
isha | -0.140 |
isa | -0.132 |
aster | -0.128 |
Stacy | -0.124 |
Token | boy |
Token ID | 2933 |
Feature activation | -8.39e-02 |
Pos logit contributions | |
all | +0.046 |
were | +0.037 |
everywhere | +0.034 |
happily | +0.033 |
Tweet | +0.032 |
Neg logit contributions | |
isha | -0.076 |
isa | -0.076 |
laptop | -0.076 |
acy | -0.072 |
acey | -0.071 |
Token | who |
Token ID | 508 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
all | +0.100 |
were | +0.078 |
everywhere | +0.070 |
happily | +0.067 |
Ellie | +0.061 |
Neg logit contributions | |
laptop | -0.222 |
isha | -0.220 |
isa | -0.220 |
acy | -0.210 |
Stacy | -0.205 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
isha | +0.150 |
isa | +0.147 |
laptop | +0.146 |
acy | +0.139 |
Stacy | +0.138 |
Neg logit contributions | |
all | -0.055 |
were | -0.039 |
everywhere | -0.039 |
Ell | -0.036 |
grapes | -0.036 |
Token | his |
Token ID | 465 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
Ell | +0.019 |
all | +0.019 |
were | +0.018 |
By | +0.018 |
Thank | +0.017 |
Neg logit contributions | |
isha | -0.035 |
laptop | -0.034 |
isa | -0.034 |
aster | -0.032 |
Stacy | -0.032 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
laptop | +0.038 |
isha | +0.035 |
isa | +0.032 |
throat | +0.032 |
clipboard | +0.031 |
Neg logit contributions | |
Ell | -0.025 |
all | -0.022 |
By | -0.022 |
were | -0.022 |
Wow | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -2.87e-03 |
Pos logit contributions | |
all | +0.263 |
Ell | +0.252 |
were | +0.225 |
Tweet | +0.222 |
everywhere | +0.220 |
Neg logit contributions | |
laptop | -0.453 |
isha | -0.434 |
isa | -0.430 |
acy | -0.408 |
Stacy | -0.404 |
Token | The |
Token ID | 383 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
all | +0.004 |
were | +0.004 |
everywhere | +0.004 |
grapes | +0.003 |
Ell | +0.003 |
Neg logit contributions | |
isha | -0.006 |
isa | -0.006 |
laptop | -0.006 |
Stacy | -0.006 |
Ian | -0.006 |
Token | end |
Token ID | 886 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
laptop | +0.045 |
aster | +0.037 |
isha | +0.037 |
microwave | +0.036 |
clipboard | +0.033 |
Neg logit contributions | |
Ell | -0.067 |
all | -0.065 |
Tweet | -0.065 |
ocked | -0.064 |
happily | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -4.61e-03 |
Pos logit contributions | |
isha | +0.037 |
laptop | +0.037 |
isa | +0.037 |
acy | +0.035 |
Stacy | +0.035 |
Neg logit contributions | |
all | -0.020 |
were | -0.016 |
everywhere | -0.015 |
happily | -0.013 |
Ell | -0.013 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
all | +0.007 |
everywhere | +0.006 |
grapes | +0.006 |
were | +0.006 |
butterflies | +0.006 |
Neg logit contributions | |
isha | -0.009 |
isa | -0.009 |
laptop | -0.009 |
aster | -0.009 |
Stacy | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +3.29e-03 |
Pos logit contributions | |
all | +0.064 |
were | +0.051 |
everywhere | +0.048 |
happily | +0.043 |
Ell | +0.043 |
Neg logit contributions | |
isha | -0.120 |
laptop | -0.120 |
isa | -0.119 |
acy | -0.112 |
Stacy | -0.111 |
Token | He |
Token ID | 679 |
Feature activation | -4.06e-03 |
Pos logit contributions | |
isha | +0.007 |
laptop | +0.007 |
isa | +0.007 |
acy | +0.007 |
Stacy | +0.007 |
Neg logit contributions | |
all | -0.005 |
were | -0.004 |
everywhere | -0.004 |
grapes | -0.004 |
Ell | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
all | +0.005 |
grapes | +0.004 |
everywhere | +0.004 |
Ell | +0.004 |
were | +0.004 |
Neg logit contributions | |
isha | -0.009 |
isa | -0.009 |
laptop | -0.009 |
Stacy | -0.009 |
acey | -0.009 |
Token | carrying |
Token ID | 6872 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
all | +0.034 |
were | +0.028 |
everywhere | +0.026 |
grapes | +0.026 |
Ell | +0.025 |
Neg logit contributions | |
isha | -0.072 |
laptop | -0.071 |
isa | -0.071 |
acy | -0.068 |
Stacy | -0.067 |
Token | a |
Token ID | 257 |
Feature activation | +1.42e-03 |
Pos logit contributions | |
laptop | +0.037 |
isha | +0.036 |
isa | +0.035 |
Stacy | +0.035 |
Ian | +0.035 |
Neg logit contributions | |
all | -0.022 |
were | -0.020 |
Ell | -0.020 |
everywhere | -0.018 |
Thank | -0.018 |
Token | bou |
Token ID | 35833 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
isha | +0.003 |
laptop | +0.003 |
isa | +0.003 |
acy | +0.003 |
Stacy | +0.003 |
Neg logit contributions | |
all | -0.002 |
were | -0.002 |
everywhere | -0.002 |
Ell | -0.002 |
happily | -0.002 |
Token | quet |
Token ID | 21108 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
all | +0.388 |
were | +0.342 |
everywhere | +0.320 |
lots | +0.301 |
fun | +0.298 |
Neg logit contributions | |
isha | -0.376 |
isa | -0.371 |
laptop | -0.367 |
acy | -0.351 |
Stacy | -0.350 |
Token | of |
Token ID | 286 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
all | +0.069 |
Ell | +0.062 |
were | +0.060 |
grapes | +0.060 |
butterflies | +0.056 |
Neg logit contributions | |
laptop | -0.169 |
isha | -0.167 |
isa | -0.165 |
acy | -0.157 |
Ian | -0.156 |
Token | wild |
Token ID | 4295 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
all | +0.022 |
Ell | +0.020 |
were | +0.019 |
happily | +0.018 |
everywhere | +0.018 |
Neg logit contributions | |
laptop | -0.040 |
isha | -0.040 |
isa | -0.039 |
Stacy | -0.038 |
aster | -0.036 |
Token | fl |
Token ID | 2704 |
Feature activation | -5.14e-03 |
Pos logit contributions | |
all | +0.040 |
were | +0.032 |
Ell | +0.031 |
everywhere | +0.030 |
happily | +0.029 |
Neg logit contributions | |
laptop | -0.067 |
isa | -0.067 |
isha | -0.065 |
Stacy | -0.060 |
Ian | -0.059 |
Token | owers |
Token ID | 3618 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
all | +0.011 |
were | +0.010 |
everywhere | +0.009 |
Ell | +0.009 |
happily | +0.009 |
Neg logit contributions | |
isa | -0.008 |
laptop | -0.007 |
isha | -0.007 |
acy | -0.007 |
acey | -0.006 |
Token | now |
Token ID | 783 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
all | +0.034 |
Ell | +0.033 |
were | +0.032 |
aws | +0.028 |
everywhere | +0.028 |
Neg logit contributions | |
isha | -0.070 |
isa | -0.068 |
laptop | -0.067 |
Stacy | -0.066 |
acy | -0.066 |
Token | part |
Token ID | 636 |
Feature activation | -4.44e-03 |
Pos logit contributions | |
isha | +0.104 |
aster | +0.103 |
laptop | +0.101 |
isa | +0.100 |
Stacy | +0.096 |
Neg logit contributions | |
Ell | -0.059 |
were | -0.055 |
aws | -0.054 |
everywhere | -0.051 |
all | -0.051 |
Token | of |
Token ID | 286 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
all | +0.005 |
Ell | +0.004 |
were | +0.004 |
grapes | +0.004 |
Thank | +0.004 |
Neg logit contributions | |
isa | -0.010 |
isha | -0.010 |
laptop | -0.010 |
acy | -0.010 |
Stacy | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -1.91e-03 |
Pos logit contributions | |
all | +0.021 |
were | +0.018 |
Ell | +0.018 |
everywhere | +0.017 |
happily | +0.015 |
Neg logit contributions | |
isha | -0.048 |
laptop | -0.048 |
isa | -0.048 |
Stacy | -0.045 |
aster | -0.045 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
all | +0.002 |
were | +0.002 |
everywhere | +0.002 |
Ell | +0.002 |
aws | +0.002 |
Neg logit contributions | |
isha | -0.005 |
laptop | -0.005 |
isa | -0.005 |
Stacy | -0.004 |
acy | -0.004 |
Token | bou |
Token ID | 35833 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
all | +0.194 |
were | +0.176 |
Ell | +0.167 |
everywhere | +0.163 |
aws | +0.162 |
Neg logit contributions | |
isa | -0.234 |
isha | -0.233 |
laptop | -0.232 |
aster | -0.217 |
Stacy | -0.204 |
Token | quet |
Token ID | 21108 |
Feature activation | -6.69e-02 |
Pos logit contributions | |
all | +0.389 |
were | +0.346 |
everywhere | +0.323 |
lots | +0.310 |
Tim | +0.302 |
Neg logit contributions | |
isha | -0.383 |
isa | -0.372 |
laptop | -0.355 |
acy | -0.353 |
Stacy | -0.344 |
Token | on |
Token ID | 319 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
all | +0.080 |
Ell | +0.069 |
were | +0.068 |
grapes | +0.065 |
Tweet | +0.064 |
Neg logit contributions | |
isha | -0.153 |
laptop | -0.152 |
isa | -0.150 |
acy | -0.144 |
Stacy | -0.142 |
Token | the |
Token ID | 262 |
Feature activation | -2.37e-03 |
Pos logit contributions | |
isha | +0.031 |
isa | +0.031 |
laptop | +0.031 |
acy | +0.029 |
Stacy | +0.029 |
Neg logit contributions | |
all | -0.015 |
were | -0.012 |
everywhere | -0.011 |
Ell | -0.011 |
grapes | -0.010 |
Token | table |
Token ID | 3084 |
Feature activation | +7.72e-02 |
Pos logit contributions | |
all | +0.004 |
were | +0.004 |
Ell | +0.003 |
Tweet | +0.003 |
everywhere | +0.003 |
Neg logit contributions | |
laptop | -0.005 |
isha | -0.005 |
isa | -0.004 |
acy | -0.004 |
Stacy | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -7.57e-04 |
Pos logit contributions | |
laptop | +0.167 |
isha | +0.165 |
isa | +0.163 |
acy | +0.155 |
Stacy | +0.151 |
Neg logit contributions | |
all | -0.107 |
Ell | -0.094 |
were | -0.091 |
Tweet | -0.090 |
everywhere | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | -1.88e-04 |
Pos logit contributions | |
all | +0.060 |
were | +0.054 |
Ell | +0.052 |
everywhere | +0.050 |
Tweet | +0.048 |
Neg logit contributions | |
laptop | -0.081 |
isha | -0.079 |
isa | -0.078 |
aster | -0.073 |
Stacy | -0.072 |
Token | We |
Token ID | 775 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
all | +0.000 |
Tim | +0.000 |
Tweet | +0.000 |
Ellie | +0.000 |
were | +0.000 |
Neg logit contributions | |
isa | -0.000 |
laptop | -0.000 |
isha | -0.000 |
den | -0.000 |
pulse | -0.000 |
Token | can |
Token ID | 460 |
Feature activation | +9.51e-02 |
Pos logit contributions | |
isha | +0.093 |
isa | +0.091 |
acy | +0.090 |
laptop | +0.090 |
acey | +0.089 |
Neg logit contributions | |
all | -0.078 |
Ell | -0.073 |
everywhere | -0.068 |
Tweet | -0.067 |
Ellie | -0.067 |
Token | both |
Token ID | 1111 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
laptop | +0.197 |
isha | +0.197 |
isa | +0.195 |
acy | +0.183 |
Stacy | +0.180 |
Neg logit contributions | |
all | -0.153 |
were | -0.130 |
everywhere | -0.126 |
Ell | -0.118 |
grapes | -0.117 |
Token | have |
Token ID | 423 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
laptop | +0.036 |
isa | +0.035 |
isha | +0.035 |
aster | +0.033 |
acy | +0.033 |
Neg logit contributions | |
all | -0.032 |
everywhere | -0.028 |
Ell | -0.027 |
were | -0.027 |
Tweet | -0.026 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
laptop | +0.081 |
isha | +0.081 |
aster | +0.080 |
isa | +0.079 |
acey | +0.075 |
Neg logit contributions | |
all | -0.035 |
Ell | -0.035 |
were | -0.031 |
everywhere | -0.029 |
Tweet | -0.029 |
Token | ." |
Token ID | 526 |
Feature activation | +7.63e-02 |
Pos logit contributions | |
all | +0.285 |
Ell | +0.255 |
were | +0.249 |
Tweet | +0.232 |
Ellie | +0.230 |
Neg logit contributions | |
laptop | -0.444 |
isha | -0.425 |
isa | -0.414 |
acy | -0.398 |
acey | -0.397 |
Token | |
Token ID | 220 |
Feature activation | +4.99e-02 |
Pos logit contributions | |
laptop | +0.192 |
isa | +0.191 |
isha | +0.191 |
acy | +0.180 |
aster | +0.176 |
Neg logit contributions | |
all | -0.097 |
were | -0.078 |
everywhere | -0.070 |
happily | -0.068 |
Tweet | -0.067 |
Token |
|
Token ID | 198 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
isha | +0.123 |
laptop | +0.122 |
isa | +0.122 |
acy | +0.115 |
Stacy | +0.113 |
Neg logit contributions | |
all | -0.064 |
were | -0.051 |
everywhere | -0.047 |
happily | -0.043 |
Ell | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
all | +0.061 |
were | +0.049 |
everywhere | +0.046 |
happily | +0.042 |
Ell | +0.041 |
Neg logit contributions | |
isha | -0.112 |
laptop | -0.111 |
isa | -0.111 |
acy | -0.104 |
Stacy | -0.103 |
Token | They |
Token ID | 2990 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
all | +0.061 |
were | +0.049 |
everywhere | +0.047 |
grapes | +0.042 |
Tweet | +0.041 |
Neg logit contributions | |
isha | -0.107 |
isa | -0.106 |
laptop | -0.106 |
acy | -0.100 |
Stacy | -0.099 |
Token | home |
Token ID | 1363 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
isha | +0.020 |
laptop | +0.020 |
isa | +0.020 |
acy | +0.019 |
Stacy | +0.019 |
Neg logit contributions | |
all | -0.011 |
were | -0.009 |
Ell | -0.008 |
everywhere | -0.008 |
grapes | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -3.27e-03 |
Pos logit contributions | |
isa | +0.085 |
isha | +0.084 |
laptop | +0.084 |
acy | +0.079 |
acey | +0.077 |
Neg logit contributions | |
all | -0.040 |
were | -0.031 |
everywhere | -0.028 |
happily | -0.026 |
lots | -0.024 |
Token | she |
Token ID | 673 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
all | +0.004 |
were | +0.004 |
everywhere | +0.003 |
Ell | +0.003 |
grapes | +0.003 |
Neg logit contributions | |
isha | -0.008 |
isa | -0.008 |
laptop | -0.008 |
Stacy | -0.007 |
acy | -0.007 |
Token | spread |
Token ID | 4104 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
all | +0.042 |
were | +0.033 |
everywhere | +0.032 |
Ell | +0.029 |
grapes | +0.029 |
Neg logit contributions | |
isha | -0.085 |
laptop | -0.084 |
isa | -0.084 |
acy | -0.079 |
Stacy | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | -4.13e-03 |
Pos logit contributions | |
Stacy | +0.037 |
isha | +0.037 |
isa | +0.037 |
aster | +0.036 |
laptop | +0.036 |
Neg logit contributions | |
all | -0.024 |
Ell | -0.023 |
were | -0.021 |
Thank | -0.020 |
grapes | -0.020 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
all | +0.006 |
were | +0.006 |
Ell | +0.005 |
Ellie | +0.005 |
Tweet | +0.005 |
Neg logit contributions | |
laptop | -0.009 |
isa | -0.008 |
isha | -0.008 |
clipboard | -0.008 |
acy | -0.008 |
Token | out |
Token ID | 503 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
Ell | +0.279 |
all | +0.275 |
Thank | +0.252 |
Tweet | +0.251 |
Ellie | +0.250 |
Neg logit contributions | |
laptop | -0.395 |
isa | -0.380 |
acy | -0.370 |
isha | -0.369 |
Stacy | -0.368 |
Token | on |
Token ID | 319 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
laptop | +0.135 |
isha | +0.135 |
isa | +0.134 |
acy | +0.129 |
Stacy | +0.128 |
Neg logit contributions | |
all | -0.054 |
Ell | -0.045 |
were | -0.043 |
grapes | -0.042 |
Thank | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
isha | +0.039 |
laptop | +0.038 |
isa | +0.038 |
acy | +0.036 |
Stacy | +0.036 |
Neg logit contributions | |
all | -0.018 |
were | -0.015 |
everywhere | -0.013 |
Ell | -0.013 |
grapes | -0.012 |
Token | kitchen |
Token ID | 9592 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
all | +0.045 |
were | +0.039 |
everywhere | +0.037 |
happily | +0.036 |
grapes | +0.035 |
Neg logit contributions | |
isha | -0.039 |
isa | -0.038 |
laptop | -0.038 |
acy | -0.035 |
Stacy | -0.035 |
Token | table |
Token ID | 3084 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
laptop | +0.318 |
isha | +0.315 |
isa | +0.309 |
acy | +0.284 |
Stacy | +0.282 |
Neg logit contributions | |
all | -0.309 |
were | -0.270 |
everywhere | -0.256 |
Ell | -0.252 |
Tweet | -0.246 |
Token | if |
Token ID | 611 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
laptop | +0.191 |
isa | +0.191 |
isha | +0.190 |
acy | +0.180 |
Stacy | +0.177 |
Neg logit contributions | |
all | -0.095 |
were | -0.075 |
everywhere | -0.067 |
happily | -0.060 |
lots | -0.060 |
Token | she |
Token ID | 673 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
isha | +0.257 |
laptop | +0.256 |
isa | +0.255 |
acy | +0.239 |
Stacy | +0.237 |
Neg logit contributions | |
all | -0.145 |
were | -0.117 |
everywhere | -0.109 |
happily | -0.100 |
Ell | -0.099 |
Token | could |
Token ID | 714 |
Feature activation | +8.11e-02 |
Pos logit contributions | |
all | +0.039 |
were | +0.031 |
everywhere | +0.028 |
happily | +0.026 |
Ell | +0.025 |
Neg logit contributions | |
isha | -0.083 |
laptop | -0.083 |
isa | -0.082 |
acy | -0.078 |
Stacy | -0.077 |
Token | eat |
Token ID | 4483 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
isa | +0.212 |
isha | +0.211 |
laptop | +0.194 |
uka | +0.189 |
Stacy | +0.188 |
Neg logit contributions | |
all | -0.125 |
were | -0.102 |
happily | -0.088 |
everywhere | -0.078 |
Tim | -0.077 |
Token | the |
Token ID | 262 |
Feature activation | -2.53e-03 |
Pos logit contributions | |
all | +0.025 |
were | +0.020 |
everywhere | +0.019 |
Ell | +0.018 |
happily | +0.017 |
Neg logit contributions | |
isha | -0.049 |
laptop | -0.048 |
isa | -0.048 |
acy | -0.045 |
Stacy | -0.045 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
all | +0.003 |
were | +0.003 |
everywhere | +0.003 |
Ell | +0.002 |
happily | +0.002 |
Neg logit contributions | |
isha | -0.006 |
laptop | -0.006 |
isa | -0.006 |
acy | -0.005 |
Stacy | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +8.94e-03 |
Pos logit contributions | |
all | +0.269 |
were | +0.222 |
Ell | +0.219 |
everywhere | +0.211 |
Tweet | +0.201 |
Neg logit contributions | |
laptop | -0.469 |
isha | -0.459 |
isa | -0.458 |
acy | -0.437 |
aster | -0.431 |
Token | Her |
Token ID | 2332 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
isa | +0.022 |
isha | +0.022 |
laptop | +0.022 |
acy | +0.021 |
pulse | +0.020 |
Neg logit contributions | |
all | -0.012 |
were | -0.009 |
Ellie | -0.008 |
Tim | -0.008 |
happily | -0.008 |
Token | mom |
Token ID | 1995 |
Feature activation | +7.83e-02 |
Pos logit contributions | |
laptop | +0.093 |
isha | +0.091 |
isa | +0.088 |
Stacy | +0.081 |
acey | +0.081 |
Neg logit contributions | |
all | -0.106 |
were | -0.092 |
Ell | -0.091 |
everywhere | -0.090 |
Tweet | -0.086 |
Token | said |
Token ID | 531 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
isha | +0.163 |
isa | +0.163 |
laptop | +0.154 |
acey | +0.148 |
Stacy | +0.148 |
Neg logit contributions | |
all | -0.116 |
everywhere | -0.105 |
butterflies | -0.100 |
grapes | -0.098 |
Ell | -0.097 |
Token | yes |
Token ID | 3763 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
all | +0.043 |
were | +0.035 |
everywhere | +0.035 |
Ell | +0.033 |
grapes | +0.032 |
Neg logit contributions | |
isha | -0.063 |
laptop | -0.062 |
isa | -0.062 |
acy | -0.058 |
Stacy | -0.058 |
Token | happy |
Token ID | 3772 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
all | +0.036 |
were | +0.029 |
everywhere | +0.027 |
grapes | +0.025 |
Ell | +0.025 |
Neg logit contributions | |
isha | -0.080 |
laptop | -0.080 |
isa | -0.079 |
acy | -0.075 |
Stacy | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
all | +0.089 |
Ell | +0.088 |
aws | +0.084 |
were | +0.082 |
grapes | +0.081 |
Neg logit contributions | |
isha | -0.153 |
isa | -0.151 |
laptop | -0.150 |
Stacy | -0.146 |
Ian | -0.141 |
Token | ate |
Token ID | 15063 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
all | +0.048 |
were | +0.038 |
everywhere | +0.037 |
grapes | +0.036 |
Ell | +0.036 |
Neg logit contributions | |
isha | -0.093 |
laptop | -0.092 |
isa | -0.092 |
Stacy | -0.087 |
acy | -0.086 |
Token | some |
Token ID | 617 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
Ell | +0.157 |
By | +0.136 |
were | +0.132 |
chased | +0.126 |
Ag | +0.117 |
Neg logit contributions | |
aster | -0.232 |
laptop | -0.218 |
microwave | -0.207 |
Stacy | -0.204 |
isha | -0.201 |
Token | more |
Token ID | 517 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
Ell | +0.045 |
all | +0.039 |
were | +0.038 |
By | +0.037 |
Tweet | +0.037 |
Neg logit contributions | |
aster | -0.087 |
laptop | -0.082 |
isa | -0.077 |
isha | -0.076 |
microwave | -0.076 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
Ell | +0.018 |
all | +0.018 |
were | +0.016 |
By | +0.016 |
Thank | +0.015 |
Neg logit contributions | |
aster | -0.049 |
laptop | -0.049 |
isa | -0.047 |
isha | -0.045 |
acy | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-03 |
Pos logit contributions | |
Ell | +0.261 |
all | +0.255 |
were | +0.240 |
Tweet | +0.238 |
Wow | +0.237 |
Neg logit contributions | |
laptop | -0.430 |
isa | -0.402 |
isha | -0.397 |
Stacy | -0.379 |
acy | -0.378 |
Token | She |
Token ID | 1375 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
all | +0.002 |
were | +0.002 |
everywhere | +0.002 |
Ell | +0.002 |
grapes | +0.002 |
Neg logit contributions | |
isha | -0.003 |
laptop | -0.003 |
isa | -0.003 |
acy | -0.003 |
Stacy | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
all | +0.017 |
everywhere | +0.015 |
grapes | +0.014 |
Ell | +0.014 |
Ellie | +0.013 |
Neg logit contributions | |
isha | -0.035 |
isa | -0.034 |
laptop | -0.034 |
acey | -0.032 |
Stacy | -0.032 |
Token | gentle |
Token ID | 10296 |
Feature activation | -7.94e-03 |
Pos logit contributions | |
all | +0.037 |
were | +0.032 |
everywhere | +0.029 |
Ell | +0.028 |
grapes | +0.028 |
Neg logit contributions | |
isha | -0.085 |
laptop | -0.084 |
isa | -0.084 |
acy | -0.080 |
Stacy | -0.080 |
Token | with |
Token ID | 351 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
all | +0.009 |
were | +0.008 |
Ell | +0.007 |
grapes | +0.007 |
everywhere | +0.007 |
Neg logit contributions | |
isa | -0.019 |
isha | -0.019 |
laptop | -0.019 |
Stacy | -0.018 |
Ian | -0.018 |
Token | happy |
Token ID | 3772 |
Feature activation | -7.29e-02 |
Pos logit contributions | |
acy | +0.095 |
isha | +0.091 |
acey | +0.090 |
laptop | +0.089 |
aster | +0.087 |
Neg logit contributions | |
all | -0.068 |
Ell | -0.067 |
Ellie | -0.063 |
everywhere | -0.062 |
grapes | -0.060 |
Token | that |
Token ID | 326 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
Ell | +0.104 |
all | +0.096 |
were | +0.094 |
aws | +0.093 |
lots | +0.086 |
Neg logit contributions | |
isha | -0.132 |
isa | -0.132 |
laptop | -0.129 |
Stacy | -0.128 |
Ian | -0.127 |
Token | she |
Token ID | 673 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
aws | +0.051 |
all | +0.050 |
Ell | +0.048 |
everywhere | +0.046 |
grapes | +0.045 |
Neg logit contributions | |
aster | -0.050 |
Stacy | -0.050 |
isha | -0.050 |
laptop | -0.049 |
Ian | -0.048 |
Token | shared |
Token ID | 4888 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
all | +0.042 |
grapes | +0.038 |
everywhere | +0.037 |
aws | +0.036 |
cher | +0.035 |
Neg logit contributions | |
isha | -0.065 |
acey | -0.062 |
Stacy | -0.062 |
isa | -0.062 |
acy | -0.059 |
Token | her |
Token ID | 607 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
Ell | +0.182 |
By | +0.172 |
ashed | +0.165 |
were | +0.162 |
Ag | +0.158 |
Neg logit contributions | |
aster | -0.190 |
Stacy | -0.164 |
shave | -0.156 |
laptop | -0.154 |
Heather | -0.153 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
Ell | +0.031 |
all | +0.029 |
were | +0.029 |
everywhere | +0.027 |
By | +0.026 |
Neg logit contributions | |
laptop | -0.057 |
aster | -0.053 |
stool | -0.050 |
isha | -0.049 |
Stacy | -0.048 |
Token | with |
Token ID | 351 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
Ell | +0.267 |
were | +0.249 |
Wow | +0.246 |
Tweet | +0.235 |
chased | +0.230 |
Neg logit contributions | |
laptop | -0.402 |
isa | -0.379 |
acy | -0.366 |
Stacy | -0.363 |
aster | -0.363 |
Token | her |
Token ID | 607 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
Ell | +0.042 |
all | +0.040 |
everywhere | +0.040 |
were | +0.039 |
ashed | +0.039 |
Neg logit contributions | |
aster | -0.071 |
Stacy | -0.069 |
isha | -0.067 |
acey | -0.066 |
Ian | -0.066 |
Token | friend |
Token ID | 1545 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
were | +0.034 |
all | +0.034 |
Ell | +0.034 |
Wow | +0.032 |
Thank | +0.032 |
Neg logit contributions | |
laptop | -0.047 |
isha | -0.046 |
pulse | -0.044 |
isa | -0.043 |
Stacy | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
all | +0.076 |
were | +0.065 |
everywhere | +0.060 |
Ell | +0.060 |
grapes | +0.057 |
Neg logit contributions | |
isha | -0.123 |
laptop | -0.122 |
isa | -0.121 |
Stacy | -0.114 |
acy | -0.113 |
Token | they |
Token ID | 484 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
aws | +0.050 |
Ell | +0.050 |
all | +0.046 |
everywhere | +0.046 |
Fred | +0.043 |
Neg logit contributions | |
laptop | -0.070 |
Stacy | -0.070 |
isha | -0.069 |
Ian | -0.067 |
isa | -0.067 |
Token | so |
Token ID | 523 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
all | +0.027 |
everywhere | +0.019 |
were | +0.017 |
lots | +0.015 |
happily | +0.014 |
Neg logit contributions | |
isa | -0.065 |
isha | -0.065 |
laptop | -0.062 |
Stacy | -0.061 |
acy | -0.061 |
Token | we |
Token ID | 356 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
isha | +0.126 |
isa | +0.126 |
laptop | +0.124 |
acy | +0.117 |
acey | +0.116 |
Neg logit contributions | |
all | -0.085 |
were | -0.064 |
everywhere | -0.062 |
lots | -0.059 |
Tim | -0.057 |
Token | can |
Token ID | 460 |
Feature activation | +9.51e-02 |
Pos logit contributions | |
isa | +0.123 |
isha | +0.123 |
laptop | +0.122 |
acy | +0.114 |
Stacy | +0.114 |
Neg logit contributions | |
all | -0.081 |
were | -0.066 |
everywhere | -0.062 |
happily | -0.056 |
lots | -0.055 |
Token | buy |
Token ID | 2822 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
isha | +0.227 |
isa | +0.226 |
laptop | +0.219 |
Stacy | +0.205 |
uka | +0.204 |
Neg logit contributions | |
all | -0.161 |
were | -0.124 |
happily | -0.109 |
everywhere | -0.107 |
lots | -0.101 |
Token | more |
Token ID | 517 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
isha | +0.142 |
laptop | +0.141 |
isa | +0.139 |
aster | +0.134 |
Stacy | +0.131 |
Neg logit contributions | |
all | -0.085 |
were | -0.077 |
Ell | -0.073 |
everywhere | -0.071 |
happily | -0.067 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
all | +0.015 |
were | +0.013 |
Ell | +0.012 |
everywhere | +0.011 |
happily | +0.010 |
Neg logit contributions | |
laptop | -0.044 |
isha | -0.043 |
isa | -0.043 |
aster | -0.042 |
Stacy | -0.041 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
all | +0.382 |
were | +0.336 |
everywhere | +0.318 |
Ell | +0.311 |
Tweet | +0.302 |
Neg logit contributions | |
laptop | -0.378 |
isha | -0.376 |
isa | -0.369 |
acy | -0.343 |
Stacy | -0.339 |
Token | Her |
Token ID | 2332 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
isha | +0.273 |
laptop | +0.271 |
isa | +0.269 |
acy | +0.250 |
Stacy | +0.248 |
Neg logit contributions | |
all | -0.225 |
were | -0.190 |
everywhere | -0.182 |
Ell | -0.169 |
grapes | -0.169 |
Token | mom |
Token ID | 1995 |
Feature activation | +7.09e-02 |
Pos logit contributions | |
laptop | +0.084 |
isha | +0.078 |
isa | +0.070 |
clipboard | +0.069 |
acey | +0.068 |
Neg logit contributions | |
all | -0.115 |
were | -0.107 |
Ell | -0.104 |
everywhere | -0.100 |
Thank | -0.099 |
Token | laughed |
Token ID | 13818 |
Feature activation | -9.60e-02 |
Pos logit contributions | |
isa | +0.149 |
isha | +0.147 |
laptop | +0.137 |
Ian | +0.133 |
Stacy | +0.133 |
Neg logit contributions | |
all | -0.108 |
everywhere | -0.099 |
grapes | -0.095 |
butterflies | -0.092 |
aws | -0.088 |
Token | and |
Token ID | 290 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
Ell | +0.131 |
grapes | +0.123 |
all | +0.118 |
were | +0.117 |
cher | +0.117 |
Neg logit contributions | |
isha | -0.176 |
laptop | -0.168 |
Stacy | -0.165 |
Bailey | -0.162 |
isa | -0.161 |
Token | never |
Token ID | 1239 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
all | +0.024 |
were | +0.018 |
everywhere | +0.018 |
grapes | +0.017 |
Ell | +0.017 |
Neg logit contributions | |
isa | -0.056 |
isha | -0.056 |
laptop | -0.055 |
acy | -0.053 |
Stacy | -0.052 |
Token | forgot |
Token ID | 16453 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
all | +0.043 |
grapes | +0.037 |
everywhere | +0.035 |
Ell | +0.035 |
Ellie | +0.033 |
Neg logit contributions | |
isa | -0.069 |
isha | -0.068 |
laptop | -0.064 |
acy | -0.064 |
Stacy | -0.064 |
Token | what |
Token ID | 644 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
all | +0.040 |
Ell | +0.037 |
were | +0.036 |
Thank | +0.033 |
grapes | +0.033 |
Neg logit contributions | |
isha | -0.085 |
laptop | -0.083 |
isa | -0.083 |
Ian | -0.080 |
Stacy | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | -2.35e-03 |
Pos logit contributions | |
isha | +0.067 |
laptop | +0.067 |
isa | +0.066 |
Ian | +0.062 |
Stacy | +0.062 |
Neg logit contributions | |
all | -0.050 |
were | -0.042 |
everywhere | -0.041 |
Ell | -0.040 |
grapes | -0.039 |
Token | m |
Token ID | 285 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
all | +0.003 |
were | +0.003 |
everywhere | +0.003 |
Ell | +0.003 |
happily | +0.003 |
Neg logit contributions | |
laptop | -0.005 |
isha | -0.005 |
isa | -0.005 |
acy | -0.005 |
Stacy | -0.005 |
Token | op |
Token ID | 404 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
laptop | +0.128 |
Stacy | +0.117 |
Bailey | +0.114 |
isha | +0.114 |
acey | +0.114 |
Neg logit contributions | |
all | -0.240 |
were | -0.219 |
cher | -0.206 |
everywhere | -0.197 |
Tweet | -0.195 |
Token | taught |
Token ID | 7817 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
isha | +0.565 |
laptop | +0.565 |
isa | +0.558 |
acy | +0.530 |
Stacy | +0.528 |
Neg logit contributions | |
all | -0.240 |
were | -0.188 |
everywhere | -0.163 |
happily | -0.143 |
lots | -0.139 |
Token | him |
Token ID | 683 |
Feature activation | -6.57e-03 |
Pos logit contributions | |
all | +0.024 |
were | +0.021 |
grapes | +0.020 |
Ell | +0.020 |
everywhere | +0.019 |
Neg logit contributions | |
laptop | -0.039 |
isha | -0.039 |
isa | -0.038 |
acy | -0.037 |
Ian | -0.037 |
Token | � |
Token ID | 6184 |
Feature activation | +8.53e-03 |
Pos logit contributions | |
all | +0.008 |
were | +0.007 |
everywhere | +0.006 |
Ell | +0.006 |
grapes | +0.006 |
Neg logit contributions | |
isha | -0.016 |
laptop | -0.016 |
isa | -0.016 |
acy | -0.015 |
Stacy | -0.015 |
Token | � |
Token ID | 95 |
Feature activation | -6.48e-03 |
Pos logit contributions | |
isha | +0.016 |
isa | +0.016 |
laptop | +0.016 |
den | +0.015 |
Stacy | +0.015 |
Neg logit contributions | |
all | -0.015 |
were | -0.012 |
Tweet | -0.012 |
Tim | -0.012 |
happily | -0.011 |
Token | € |
Token ID | 26391 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
all | +0.011 |
Thank | +0.010 |
solved | +0.010 |
Ell | +0.010 |
were | +0.010 |
Neg logit contributions | |
isa | -0.011 |
isha | -0.011 |
laptop | -0.011 |
Stacy | -0.011 |
acy | -0.011 |
Token | is |
Token ID | 318 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
all | +0.021 |
were | +0.018 |
everywhere | +0.015 |
fun | +0.013 |
lots | +0.013 |
Neg logit contributions | |
isha | -0.041 |
isa | -0.040 |
laptop | -0.039 |
acy | -0.038 |
Stacy | -0.037 |
Token | a |
Token ID | 257 |
Feature activation | +6.80e-03 |
Pos logit contributions | |
isa | +0.084 |
isha | +0.083 |
laptop | +0.080 |
acy | +0.078 |
uka | +0.077 |
Neg logit contributions | |
all | -0.042 |
were | -0.029 |
everywhere | -0.028 |
fun | -0.027 |
lots | -0.027 |
Token | monster |
Token ID | 9234 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
isa | +0.016 |
isha | +0.015 |
acy | +0.015 |
den | +0.015 |
uka | +0.014 |
Neg logit contributions | |
all | -0.011 |
fun | -0.009 |
fair | -0.008 |
everywhere | -0.008 |
y | -0.008 |
Token | in |
Token ID | 287 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
all | +0.086 |
everywhere | +0.070 |
were | +0.070 |
at | +0.069 |
about | +0.067 |
Neg logit contributions | |
isa | -0.073 |
isha | -0.071 |
acy | -0.067 |
ason | -0.065 |
laptop | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
isha | +0.084 |
isa | +0.084 |
laptop | +0.079 |
acy | +0.078 |
ason | +0.078 |
Neg logit contributions | |
all | -0.029 |
everywhere | -0.015 |
Tim | -0.013 |
lots | -0.012 |
were | -0.011 |
Token | bathroom |
Token ID | 12436 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
isa | +0.021 |
isha | +0.019 |
Ian | +0.019 |
den | +0.019 |
ason | +0.018 |
Neg logit contributions | |
all | -0.022 |
were | -0.018 |
everywhere | -0.018 |
fun | -0.017 |
lots | -0.017 |
Token | ? |
Token ID | 30 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
isa | +0.400 |
isha | +0.371 |
acy | +0.351 |
Stacy | +0.342 |
Jackson | +0.337 |
Neg logit contributions | |
all | -0.436 |
were | -0.337 |
everywhere | -0.320 |
at | -0.319 |
about | -0.311 |
Token | Or |
Token ID | 1471 |
Feature activation | +7.35e-02 |
Pos logit contributions | |
isa | +0.342 |
isha | +0.324 |
laptop | +0.320 |
ason | +0.313 |
acy | +0.304 |
Neg logit contributions | |
all | -0.161 |
Tim | -0.154 |
were | -0.141 |
Tweet | -0.140 |
lots | -0.119 |
Token | a |
Token ID | 257 |
Feature activation | +9.56e-03 |
Pos logit contributions | |
isa | +0.194 |
isha | +0.193 |
laptop | +0.185 |
acy | +0.179 |
acey | +0.179 |
Neg logit contributions | |
all | -0.105 |
were | -0.077 |
fun | -0.071 |
everywhere | -0.071 |
lots | -0.069 |
Token | ghost |
Token ID | 10905 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
isa | +0.022 |
isha | +0.021 |
acy | +0.020 |
laptop | +0.020 |
uka | +0.020 |
Neg logit contributions | |
all | -0.015 |
everywhere | -0.012 |
were | -0.012 |
fun | -0.012 |
y | -0.011 |
Token | ? |
Token ID | 30 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
isha | +0.069 |
isa | +0.068 |
acy | +0.066 |
laptop | +0.066 |
ason | +0.065 |
Neg logit contributions | |
all | -0.041 |
were | -0.032 |
everywhere | -0.032 |
happily | -0.028 |
Tweet | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
isha | +0.302 |
laptop | +0.300 |
isa | +0.298 |
acy | +0.286 |
uka | +0.282 |
Neg logit contributions | |
all | -0.130 |
were | -0.096 |
everywhere | -0.093 |
happily | -0.082 |
lots | -0.079 |
Token | walked |
Token ID | 6807 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
all | +0.030 |
were | +0.024 |
everywhere | +0.023 |
grapes | +0.022 |
Ell | +0.022 |
Neg logit contributions | |
isa | -0.053 |
isha | -0.053 |
laptop | -0.053 |
Stacy | -0.050 |
Ian | -0.049 |
Token | down |
Token ID | 866 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
all | +0.066 |
were | +0.056 |
everywhere | +0.050 |
grapes | +0.049 |
Ell | +0.048 |
Neg logit contributions | |
isha | -0.128 |
isa | -0.127 |
laptop | -0.127 |
Stacy | -0.119 |
acy | -0.119 |
Token | a |
Token ID | 257 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
isha | +0.066 |
laptop | +0.066 |
isa | +0.066 |
acy | +0.062 |
Stacy | +0.062 |
Neg logit contributions | |
all | -0.029 |
were | -0.023 |
everywhere | -0.021 |
Ell | -0.019 |
grapes | -0.019 |
Token | long |
Token ID | 890 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
isha | +0.022 |
isa | +0.022 |
laptop | +0.022 |
acy | +0.021 |
uka | +0.021 |
Neg logit contributions | |
all | -0.017 |
were | -0.014 |
everywhere | -0.014 |
happily | -0.013 |
Ell | -0.012 |
Token | hallway |
Token ID | 23959 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
isha | +0.038 |
isa | +0.037 |
laptop | +0.037 |
acy | +0.035 |
acey | +0.035 |
Neg logit contributions | |
all | -0.030 |
were | -0.025 |
everywhere | -0.024 |
Ell | -0.022 |
happily | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
laptop | +0.466 |
isha | +0.452 |
isa | +0.451 |
aster | +0.417 |
acy | +0.414 |
Neg logit contributions | |
all | -0.299 |
were | -0.253 |
Ell | -0.242 |
grapes | -0.241 |
Tweet | -0.240 |
Token | Some |
Token ID | 2773 |
Feature activation | +8.53e-02 |
Pos logit contributions | |
isha | +0.043 |
isa | +0.042 |
laptop | +0.042 |
acy | +0.039 |
uka | +0.039 |
Neg logit contributions | |
all | -0.031 |
were | -0.026 |
everywhere | -0.025 |
happily | -0.023 |
Ellie | -0.023 |
Token | of |
Token ID | 286 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
laptop | +0.222 |
isha | +0.220 |
isa | +0.216 |
acy | +0.205 |
Stacy | +0.202 |
Neg logit contributions | |
all | -0.086 |
Ell | -0.062 |
everywhere | -0.062 |
Tweet | -0.061 |
happily | -0.060 |
Token | the |
Token ID | 262 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
all | +0.015 |
were | +0.012 |
everywhere | +0.011 |
Ell | +0.010 |
happily | +0.010 |
Neg logit contributions | |
isha | -0.036 |
laptop | -0.036 |
isa | -0.036 |
acy | -0.034 |
Stacy | -0.034 |
Token | people |
Token ID | 661 |
Feature activation | -5.58e-02 |
Pos logit contributions | |
isa | +0.025 |
isha | +0.025 |
laptop | +0.024 |
acy | +0.023 |
Ian | +0.023 |
Neg logit contributions | |
all | -0.019 |
were | -0.015 |
everywhere | -0.015 |
lots | -0.013 |
happily | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
isha | +0.224 |
laptop | +0.223 |
isa | +0.222 |
acy | +0.210 |
acey | +0.208 |
Neg logit contributions | |
all | -0.077 |
were | -0.053 |
everywhere | -0.049 |
happily | -0.043 |
lots | -0.042 |
Token | Lily |
Token ID | 20037 |
Feature activation | -6.90e-02 |
Pos logit contributions | |
all | +0.025 |
were | +0.020 |
everywhere | +0.019 |
Ell | +0.018 |
grapes | +0.017 |
Neg logit contributions | |
laptop | -0.044 |
isha | -0.044 |
isa | -0.043 |
Stacy | -0.041 |
acy | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
all | +0.078 |
were | +0.062 |
everywhere | +0.053 |
happily | +0.050 |
lots | +0.046 |
Neg logit contributions | |
isha | -0.184 |
laptop | -0.184 |
isa | -0.182 |
acy | -0.173 |
Stacy | -0.170 |
Token | in |
Token ID | 287 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
all | +0.030 |
were | +0.023 |
everywhere | +0.021 |
grapes | +0.019 |
Ell | +0.019 |
Neg logit contributions | |
isha | -0.070 |
laptop | -0.069 |
isa | -0.069 |
acy | -0.065 |
Stacy | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | -1.43e-03 |
Pos logit contributions | |
isha | +0.078 |
laptop | +0.077 |
isa | +0.077 |
acy | +0.072 |
Stacy | +0.071 |
Neg logit contributions | |
all | -0.049 |
were | -0.040 |
everywhere | -0.038 |
happily | -0.035 |
Ell | -0.035 |
Token | kitchen |
Token ID | 9592 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
all | +0.002 |
were | +0.002 |
everywhere | +0.002 |
grapes | +0.002 |
lots | +0.002 |
Neg logit contributions | |
isha | -0.003 |
isa | -0.003 |
laptop | -0.003 |
acy | -0.003 |
Stacy | -0.003 |
Token | getting |
Token ID | 1972 |
Feature activation | +6.23e-02 |
Pos logit contributions | |
isha | +0.471 |
isa | +0.468 |
laptop | +0.463 |
acy | +0.445 |
Stacy | +0.440 |
Neg logit contributions | |
all | -0.233 |
were | -0.176 |
everywhere | -0.165 |
happily | -0.148 |
lots | -0.147 |
Token | a |
Token ID | 257 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
laptop | +0.141 |
isa | +0.141 |
isha | +0.140 |
acy | +0.130 |
Stacy | +0.128 |
Neg logit contributions | |
all | -0.101 |
were | -0.080 |
everywhere | -0.078 |
lots | -0.073 |
happily | -0.071 |
Token | snack |
Token ID | 26906 |
Feature activation | -5.45e-02 |
Pos logit contributions | |
isa | +0.033 |
isha | +0.033 |
laptop | +0.032 |
acy | +0.031 |
Stacy | +0.031 |
Neg logit contributions | |
all | -0.020 |
were | -0.016 |
everywhere | -0.016 |
fun | -0.014 |
happily | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
all | +0.072 |
were | +0.059 |
everywhere | +0.055 |
Ell | +0.052 |
happily | +0.050 |
Neg logit contributions | |
laptop | -0.129 |
isha | -0.129 |
isa | -0.128 |
acy | -0.120 |
Stacy | -0.119 |
Token | She |
Token ID | 1375 |
Feature activation | +2.15e-03 |
Pos logit contributions | |
isa | +0.030 |
laptop | +0.030 |
isha | +0.030 |
acy | +0.028 |
asy | +0.028 |
Neg logit contributions | |
all | -0.016 |
were | -0.013 |
Tim | -0.012 |
Ellie | -0.012 |
everywhere | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +5.87e-03 |
Pos logit contributions | |
isha | +0.100 |
isa | +0.099 |
laptop | +0.098 |
acy | +0.094 |
uka | +0.092 |
Neg logit contributions | |
all | -0.045 |
were | -0.032 |
everywhere | -0.031 |
happily | -0.030 |
lots | -0.027 |
Token | carrying |
Token ID | 6872 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
laptop | +0.013 |
isha | +0.013 |
isa | +0.013 |
Stacy | +0.012 |
clipboard | +0.012 |
Neg logit contributions | |
all | -0.008 |
were | -0.007 |
everywhere | -0.006 |
aws | -0.006 |
Ell | -0.006 |
Token | her |
Token ID | 607 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
isha | +0.050 |
laptop | +0.050 |
isa | +0.050 |
acy | +0.047 |
Stacy | +0.047 |
Neg logit contributions | |
all | -0.022 |
were | -0.017 |
everywhere | -0.015 |
Ell | -0.014 |
happily | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | +1.85e-03 |
Pos logit contributions | |
all | +0.032 |
were | +0.027 |
everywhere | +0.026 |
lots | +0.023 |
happily | +0.023 |
Neg logit contributions | |
isa | -0.040 |
isha | -0.039 |
acy | -0.037 |
aster | -0.036 |
laptop | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +1.10e-03 |
Pos logit contributions | |
isha | +0.005 |
isa | +0.005 |
laptop | +0.005 |
acy | +0.005 |
acey | +0.004 |
Neg logit contributions | |
all | -0.002 |
were | -0.002 |
everywhere | -0.002 |
happily | -0.001 |
lots | -0.001 |
Token | bathroom |
Token ID | 12436 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
isa | +0.002 |
isha | +0.002 |
laptop | +0.002 |
acy | +0.002 |
aster | +0.002 |
Neg logit contributions | |
all | -0.002 |
were | -0.002 |
everywhere | -0.002 |
fun | -0.002 |
lots | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +7.42e-03 |
Pos logit contributions | |
isha | +0.496 |
isa | +0.495 |
laptop | +0.483 |
acy | +0.468 |
Stacy | +0.460 |
Neg logit contributions | |
all | -0.257 |
were | -0.202 |
everywhere | -0.188 |
happily | -0.169 |
lots | -0.167 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
isha | +0.017 |
laptop | +0.016 |
isa | +0.016 |
Stacy | +0.016 |
acey | +0.015 |
Neg logit contributions | |
all | -0.011 |
were | -0.009 |
everywhere | -0.008 |
Ell | -0.008 |
grapes | -0.008 |
Token | spl |
Token ID | 4328 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
all | +0.050 |
were | +0.039 |
everywhere | +0.037 |
grapes | +0.034 |
Ell | +0.034 |
Neg logit contributions | |
isha | -0.103 |
isa | -0.103 |
laptop | -0.102 |
acy | -0.096 |
Stacy | -0.096 |
Token | ashed |
Token ID | 5263 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
all | +0.034 |
were | +0.024 |
everywhere | +0.023 |
Ellie | +0.022 |
Ell | +0.020 |
Neg logit contributions | |
isa | -0.080 |
isha | -0.080 |
laptop | -0.080 |
acey | -0.076 |
acy | -0.076 |
Token | in |
Token ID | 287 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
Ell | +0.179 |
Jerry | +0.156 |
Ellie | +0.146 |
Tweet | +0.146 |
Jerry | +0.143 |
Neg logit contributions | |
Stacy | -0.261 |
isa | -0.256 |
Macy | -0.253 |
laptop | -0.250 |
acy | -0.243 |
Token | " |
Token ID | 366 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
laptop | +0.032 |
isa | +0.031 |
aster | +0.031 |
isha | +0.031 |
asy | +0.030 |
Neg logit contributions | |
all | -0.012 |
were | -0.009 |
Ellie | -0.008 |
Tim | -0.008 |
everywhere | -0.007 |
Token | You |
Token ID | 1639 |
Feature activation | -7.13e-02 |
Pos logit contributions | |
all | +0.162 |
were | +0.133 |
everywhere | +0.130 |
grapes | +0.119 |
Ellie | +0.117 |
Neg logit contributions | |
isha | -0.227 |
isa | -0.226 |
laptop | -0.222 |
acy | -0.212 |
Stacy | -0.208 |
Token | must |
Token ID | 1276 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
all | +0.106 |
grapes | +0.092 |
Ell | +0.089 |
everywhere | +0.089 |
aws | +0.085 |
Neg logit contributions | |
isha | -0.148 |
laptop | -0.144 |
isa | -0.143 |
acy | -0.138 |
acey | -0.135 |
Token | prove |
Token ID | 5879 |
Feature activation | +6.24e-02 |
Pos logit contributions | |
laptop | +0.086 |
isha | +0.084 |
isa | +0.083 |
acy | +0.080 |
aster | +0.079 |
Neg logit contributions | |
all | -0.075 |
everywhere | -0.067 |
were | -0.065 |
grapes | -0.063 |
Ell | -0.061 |
Token | yourself |
Token ID | 3511 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
isha | +0.160 |
laptop | +0.159 |
isa | +0.158 |
acy | +0.150 |
Stacy | +0.147 |
Neg logit contributions | |
all | -0.077 |
were | -0.059 |
everywhere | -0.055 |
happily | -0.049 |
lots | -0.049 |
Token | before |
Token ID | 878 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
isha | +0.257 |
laptop | +0.256 |
isa | +0.255 |
acy | +0.242 |
Stacy | +0.239 |
Neg logit contributions | |
all | -0.134 |
were | -0.106 |
everywhere | -0.099 |
lots | -0.089 |
happily | -0.088 |
Token | you |
Token ID | 345 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
isha | +0.322 |
isa | +0.319 |
laptop | +0.317 |
acey | +0.302 |
acy | +0.298 |
Neg logit contributions | |
all | -0.217 |
were | -0.175 |
everywhere | -0.162 |
lots | -0.150 |
Ellie | -0.146 |
Token | can |
Token ID | 460 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
isha | +0.060 |
laptop | +0.060 |
isa | +0.059 |
acy | +0.055 |
Stacy | +0.054 |
Neg logit contributions | |
all | -0.052 |
were | -0.044 |
everywhere | -0.043 |
Ell | -0.040 |
grapes | -0.040 |
Token | open |
Token ID | 1280 |
Feature activation | +4.99e-02 |
Pos logit contributions | |
isha | +0.205 |
laptop | +0.203 |
isa | +0.203 |
acy | +0.188 |
Stacy | +0.186 |
Neg logit contributions | |
all | -0.176 |
were | -0.149 |
everywhere | -0.141 |
happily | -0.133 |
Ell | -0.131 |
Token | the |
Token ID | 262 |
Feature activation | +8.40e-03 |
Pos logit contributions | |
isha | +0.144 |
laptop | +0.143 |
isa | +0.142 |
acy | +0.136 |
acey | +0.135 |
Neg logit contributions | |
all | -0.058 |
everywhere | -0.037 |
were | -0.033 |
lots | -0.032 |
Ellie | -0.027 |
Token | door |
Token ID | 3420 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
laptop | +0.015 |
isha | +0.015 |
isa | +0.014 |
acy | +0.013 |
Stacy | +0.013 |
Neg logit contributions | |
all | -0.016 |
were | -0.014 |
everywhere | -0.013 |
Ell | -0.013 |
happily | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +8.01e-03 |
Pos logit contributions | |
isha | +0.203 |
isa | +0.200 |
laptop | +0.194 |
Stacy | +0.188 |
acey | +0.183 |
Neg logit contributions | |
all | -0.093 |
grapes | -0.073 |
were | -0.068 |
fun | -0.067 |
everywhere | -0.066 |
Token | |
Token ID | 220 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
isa | +0.020 |
laptop | +0.020 |
isha | +0.020 |
asy | +0.018 |
acy | +0.018 |
Neg logit contributions | |
all | -0.011 |
were | -0.009 |
Tim | -0.008 |
Ellie | -0.008 |
Tweet | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
laptop | +0.097 |
isha | +0.097 |
isa | +0.097 |
acy | +0.091 |
Stacy | +0.090 |
Neg logit contributions | |
all | -0.050 |
were | -0.040 |
everywhere | -0.037 |
happily | -0.034 |
Ell | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
all | +0.051 |
were | +0.040 |
everywhere | +0.036 |
Ell | +0.036 |
happily | +0.035 |
Neg logit contributions | |
laptop | -0.109 |
isha | -0.107 |
isa | -0.106 |
acy | -0.100 |
Stacy | -0.099 |
Token | That |
Token ID | 2504 |
Feature activation | -7.79e-02 |
Pos logit contributions | |
all | +0.057 |
were | +0.047 |
Ell | +0.043 |
everywhere | +0.043 |
happily | +0.042 |
Neg logit contributions | |
laptop | -0.095 |
isha | -0.094 |
isa | -0.093 |
acy | -0.087 |
Stacy | -0.086 |
Token | night |
Token ID | 1755 |
Feature activation | +9.49e-02 |
Pos logit contributions | |
all | +0.111 |
were | +0.090 |
everywhere | +0.082 |
lots | +0.076 |
happily | +0.075 |
Neg logit contributions | |
laptop | -0.186 |
isha | -0.184 |
isa | -0.182 |
acy | -0.173 |
Stacy | -0.172 |
Token | , |
Token ID | 11 |
Feature activation | -4.20e-03 |
Pos logit contributions | |
laptop | +0.251 |
isa | +0.250 |
isha | +0.248 |
acy | +0.236 |
uka | +0.234 |
Neg logit contributions | |
all | -0.114 |
were | -0.088 |
everywhere | -0.078 |
happily | -0.070 |
lots | -0.069 |
Token | Tor |
Token ID | 4022 |
Feature activation | +7.14e-02 |
Pos logit contributions | |
all | +0.006 |
were | +0.005 |
everywhere | +0.004 |
Ell | +0.004 |
grapes | +0.004 |
Neg logit contributions | |
isha | -0.010 |
laptop | -0.010 |
isa | -0.010 |
acy | -0.009 |
Stacy | -0.009 |
Token | i |
Token ID | 72 |
Feature activation | +7.67e-02 |
Pos logit contributions | |
laptop | +0.106 |
isha | +0.101 |
isa | +0.097 |
acy | +0.089 |
aster | +0.089 |
Neg logit contributions | |
all | -0.165 |
were | -0.147 |
everywhere | -0.137 |
happily | -0.136 |
grapes | -0.132 |
Token | and |
Token ID | 290 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
laptop | +0.206 |
isha | +0.197 |
isa | +0.195 |
uka | +0.191 |
aster | +0.190 |
Neg logit contributions | |
all | -0.090 |
were | -0.073 |
happily | -0.059 |
everywhere | -0.058 |
lots | -0.052 |
Token | her |
Token ID | 607 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
all | +0.046 |
were | +0.038 |
everywhere | +0.037 |
Ell | +0.036 |
aws | +0.035 |
Neg logit contributions | |
isa | -0.086 |
isha | -0.086 |
laptop | -0.084 |
Stacy | -0.081 |
acy | -0.081 |
Token | your |
Token ID | 534 |
Feature activation | +9.41e-02 |
Pos logit contributions | |
isha | +0.035 |
laptop | +0.035 |
isa | +0.035 |
Stacy | +0.033 |
acey | +0.032 |
Neg logit contributions | |
all | -0.015 |
were | -0.013 |
everywhere | -0.012 |
Ell | -0.010 |
grapes | -0.010 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
laptop | +0.250 |
isha | +0.248 |
isa | +0.246 |
acy | +0.232 |
Stacy | +0.231 |
Neg logit contributions | |
all | -0.099 |
were | -0.078 |
everywhere | -0.069 |
Ell | -0.066 |
Ellie | -0.061 |
Token | !" |
Token ID | 2474 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
all | +0.192 |
were | +0.161 |
everywhere | +0.153 |
happily | +0.144 |
lots | +0.141 |
Neg logit contributions | |
isha | -0.238 |
laptop | -0.237 |
isa | -0.236 |
acy | -0.220 |
Stacy | -0.217 |
Token | Steve |
Token ID | 6542 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
isha | +0.111 |
laptop | +0.110 |
isa | +0.110 |
acy | +0.103 |
Stacy | +0.102 |
Neg logit contributions | |
all | -0.068 |
were | -0.055 |
everywhere | -0.052 |
happily | -0.048 |
Ell | -0.047 |
Token | was |
Token ID | 373 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
all | +0.040 |
everywhere | +0.032 |
grapes | +0.031 |
Ell | +0.030 |
were | +0.030 |
Neg logit contributions | |
isa | -0.085 |
isha | -0.085 |
laptop | -0.082 |
Stacy | -0.078 |
acy | -0.078 |
Token | so |
Token ID | 523 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
all | +0.032 |
were | +0.025 |
everywhere | +0.023 |
Ell | +0.021 |
grapes | +0.021 |
Neg logit contributions | |
isha | -0.079 |
laptop | -0.078 |
isa | -0.078 |
acy | -0.074 |
Stacy | -0.073 |
Token | happy |
Token ID | 3772 |
Feature activation | -6.47e-02 |
Pos logit contributions | |
isha | +0.138 |
isa | +0.136 |
laptop | +0.134 |
acy | +0.130 |
Stacy | +0.127 |
Neg logit contributions | |
all | -0.057 |
everywhere | -0.044 |
were | -0.043 |
grapes | -0.041 |
Ell | -0.040 |
Token | that |
Token ID | 326 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
all | +0.081 |
were | +0.070 |
Ell | +0.070 |
everywhere | +0.065 |
grapes | +0.063 |
Neg logit contributions | |
isha | -0.146 |
isa | -0.146 |
laptop | -0.145 |
Ian | -0.137 |
Stacy | -0.135 |
Token | he |
Token ID | 339 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
all | +0.045 |
everywhere | +0.039 |
were | +0.038 |
grapes | +0.038 |
Ell | +0.038 |
Neg logit contributions | |
isha | -0.059 |
laptop | -0.059 |
isa | -0.058 |
Ian | -0.056 |
Stacy | -0.055 |
Token | did |
Token ID | 750 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
all | +0.025 |
grapes | +0.022 |
everywhere | +0.021 |
were | +0.019 |
Ell | +0.019 |
Neg logit contributions | |
isha | -0.049 |
isa | -0.048 |
laptop | -0.047 |
Stacy | -0.046 |
acey | -0.046 |
Token | a |
Token ID | 257 |
Feature activation | -4.90e-03 |
Pos logit contributions | |
all | +0.046 |
were | +0.039 |
everywhere | +0.038 |
Ell | +0.036 |
grapes | +0.034 |
Neg logit contributions | |
laptop | -0.083 |
isha | -0.082 |
isa | -0.081 |
aster | -0.077 |
acy | -0.076 |
Token | Tim |
Token ID | 5045 |
Feature activation | -6.79e-02 |
Pos logit contributions | |
isha | +0.063 |
isa | +0.062 |
laptop | +0.062 |
acy | +0.058 |
Stacy | +0.058 |
Neg logit contributions | |
all | -0.038 |
were | -0.031 |
everywhere | -0.030 |
grapes | -0.028 |
Ell | -0.028 |
Token | my |
Token ID | 1820 |
Feature activation | +9.69e-02 |
Pos logit contributions | |
all | +0.088 |
were | +0.072 |
everywhere | +0.064 |
happily | +0.060 |
lots | +0.057 |
Neg logit contributions | |
laptop | -0.169 |
isha | -0.168 |
isa | -0.167 |
acy | -0.158 |
Stacy | -0.157 |
Token | was |
Token ID | 373 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
isha | +0.271 |
laptop | +0.270 |
isa | +0.269 |
acy | +0.256 |
Stacy | +0.253 |
Neg logit contributions | |
all | -0.091 |
were | -0.066 |
everywhere | -0.059 |
happily | -0.051 |
grapes | -0.049 |
Token | very |
Token ID | 845 |
Feature activation | +8.42e-02 |
Pos logit contributions | |
all | +0.016 |
everywhere | +0.011 |
were | +0.011 |
happily | +0.011 |
lots | +0.010 |
Neg logit contributions | |
isa | -0.037 |
isha | -0.037 |
laptop | -0.036 |
acy | -0.035 |
Stacy | -0.034 |
Token | excited |
Token ID | 6568 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
isa | +0.194 |
isha | +0.193 |
laptop | +0.192 |
Stacy | +0.178 |
Ian | +0.176 |
Neg logit contributions | |
all | -0.135 |
were | -0.107 |
everywhere | -0.101 |
happily | -0.098 |
lots | -0.096 |
Token | because |
Token ID | 780 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
all | +0.019 |
were | +0.015 |
everywhere | +0.014 |
Ell | +0.013 |
grapes | +0.013 |
Neg logit contributions | |
isha | -0.038 |
laptop | -0.037 |
isa | -0.037 |
acy | -0.035 |
Stacy | -0.034 |
Token | he |
Token ID | 339 |
Feature activation | +2.63e-03 |
Pos logit contributions | |
isha | +0.129 |
laptop | +0.128 |
isa | +0.128 |
acy | +0.120 |
Stacy | +0.119 |
Neg logit contributions | |
all | -0.076 |
were | -0.061 |
everywhere | -0.057 |
lots | -0.052 |
happily | -0.051 |
Token | was |
Token ID | 373 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
isha | +0.007 |
isa | +0.007 |
laptop | +0.007 |
acy | +0.007 |
Stacy | +0.007 |
Neg logit contributions | |
all | -0.003 |
were | -0.002 |
everywhere | -0.002 |
grapes | -0.002 |
Ell | -0.002 |
Token | going |
Token ID | 1016 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
all | +0.018 |
were | +0.014 |
everywhere | +0.014 |
happily | +0.012 |
lots | +0.012 |
Neg logit contributions | |
isha | -0.032 |
laptop | -0.032 |
isa | -0.032 |
acy | -0.030 |
Stacy | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | +7.75e-03 |
Pos logit contributions | |
isa | +0.040 |
isha | +0.039 |
laptop | +0.039 |
acy | +0.038 |
Stacy | +0.038 |
Neg logit contributions | |
all | -0.013 |
everywhere | -0.008 |
were | -0.007 |
lots | -0.007 |
fun | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
isa | +0.018 |
isha | +0.018 |
laptop | +0.018 |
asy | +0.017 |
acy | +0.017 |
Neg logit contributions | |
all | -0.012 |
were | -0.010 |
lots | -0.009 |
everywhere | -0.008 |
Ellie | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
all | +0.033 |
Ell | +0.028 |
were | +0.027 |
happily | +0.026 |
everywhere | +0.025 |
Neg logit contributions | |
laptop | -0.063 |
isha | -0.061 |
isa | -0.060 |
aster | -0.058 |
Stacy | -0.057 |
Token | red |
Token ID | 2266 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
laptop | +0.022 |
isha | +0.021 |
isa | +0.020 |
pulse | +0.019 |
aster | +0.019 |
Neg logit contributions | |
all | -0.019 |
were | -0.017 |
Ell | -0.017 |
Ellie | -0.016 |
happily | -0.016 |
Token | marks |
Token ID | 8849 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
all | +0.090 |
were | +0.076 |
everywhere | +0.073 |
grapes | +0.068 |
happily | +0.068 |
Neg logit contributions | |
isha | -0.094 |
isa | -0.093 |
laptop | -0.091 |
acy | -0.086 |
Stacy | -0.085 |
Token | on |
Token ID | 319 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
all | +0.013 |
were | +0.011 |
Ell | +0.011 |
Tweet | +0.010 |
grapes | +0.010 |
Neg logit contributions | |
laptop | -0.023 |
isha | -0.023 |
isa | -0.022 |
aster | -0.021 |
acy | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
laptop | +0.047 |
isha | +0.046 |
isa | +0.046 |
aster | +0.044 |
Ian | +0.043 |
Neg logit contributions | |
all | -0.024 |
were | -0.020 |
everywhere | -0.019 |
Ell | -0.019 |
happily | -0.017 |
Token | wall |
Token ID | 3355 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
isha | +0.017 |
isa | +0.017 |
laptop | +0.016 |
acy | +0.015 |
Stacy | +0.015 |
Neg logit contributions | |
all | -0.022 |
were | -0.019 |
everywhere | -0.019 |
grapes | -0.017 |
lots | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
laptop | +0.269 |
isha | +0.261 |
isa | +0.259 |
acy | +0.242 |
aster | +0.242 |
Neg logit contributions | |
all | -0.168 |
were | -0.138 |
Ell | -0.132 |
everywhere | -0.131 |
Tweet | -0.130 |
Token | They |
Token ID | 1119 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
isha | +0.025 |
laptop | +0.025 |
isa | +0.024 |
acy | +0.023 |
Stacy | +0.023 |
Neg logit contributions | |
all | -0.014 |
were | -0.011 |
everywhere | -0.011 |
happily | -0.010 |
Ell | -0.010 |
Token | were |
Token ID | 547 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
grapes | +0.022 |
Ellie | +0.021 |
Tweet | +0.020 |
Ell | +0.020 |
aws | +0.019 |
Neg logit contributions | |
isha | -0.039 |
laptop | -0.038 |
isa | -0.038 |
uka | -0.036 |
Stacy | -0.036 |
Token | very |
Token ID | 845 |
Feature activation | +7.66e-02 |
Pos logit contributions | |
Ell | +0.103 |
were | +0.102 |
Fred | +0.097 |
Thank | +0.097 |
grapes | +0.096 |
Neg logit contributions | |
laptop | -0.124 |
pulse | -0.114 |
isha | -0.111 |
clipboard | -0.107 |
isa | -0.103 |
Token | angry |
Token ID | 7954 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
laptop | +0.156 |
isha | +0.150 |
isa | +0.144 |
acey | +0.137 |
acy | +0.137 |
Neg logit contributions | |
all | -0.125 |
were | -0.108 |
Ell | -0.107 |
Ellie | -0.104 |
grapes | -0.102 |
Token | you |
Token ID | 345 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
all | +0.019 |
were | +0.016 |
everywhere | +0.015 |
happily | +0.014 |
grapes | +0.014 |
Neg logit contributions | |
isha | -0.023 |
laptop | -0.023 |
isa | -0.023 |
acy | -0.021 |
Stacy | -0.021 |
Token | will |
Token ID | 481 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
isha | +0.048 |
laptop | +0.047 |
isa | +0.047 |
acy | +0.045 |
Stacy | +0.044 |
Neg logit contributions | |
all | -0.036 |
were | -0.030 |
everywhere | -0.029 |
grapes | -0.028 |
Ell | -0.028 |
Token | be |
Token ID | 307 |
Feature activation | +8.70e-03 |
Pos logit contributions | |
isha | +0.176 |
laptop | +0.175 |
isa | +0.174 |
acy | +0.163 |
Stacy | +0.161 |
Neg logit contributions | |
all | -0.106 |
were | -0.085 |
everywhere | -0.077 |
happily | -0.073 |
lots | -0.072 |
Token | my |
Token ID | 616 |
Feature activation | +5.49e-02 |
Pos logit contributions | |
isha | +0.021 |
isa | +0.021 |
laptop | +0.021 |
acy | +0.019 |
Stacy | +0.019 |
Neg logit contributions | |
all | -0.011 |
were | -0.009 |
everywhere | -0.009 |
grapes | -0.008 |
Ell | -0.008 |
Token | royal |
Token ID | 15100 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
isha | +0.137 |
isa | +0.137 |
laptop | +0.137 |
acy | +0.129 |
Stacy | +0.127 |
Neg logit contributions | |
all | -0.069 |
were | -0.053 |
everywhere | -0.050 |
happily | -0.046 |
grapes | -0.045 |
Token | helper |
Token ID | 31904 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
isha | +0.059 |
laptop | +0.059 |
isa | +0.059 |
acy | +0.055 |
Stacy | +0.055 |
Neg logit contributions | |
all | -0.042 |
were | -0.035 |
everywhere | -0.033 |
happily | -0.031 |
grapes | -0.030 |
Token | ." |
Token ID | 526 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
isha | +0.033 |
laptop | +0.033 |
isa | +0.033 |
acy | +0.031 |
Stacy | +0.031 |
Neg logit contributions | |
all | -0.025 |
were | -0.020 |
everywhere | -0.019 |
happily | -0.018 |
lots | -0.018 |
Token | And |
Token ID | 843 |
Feature activation | +9.72e-03 |
Pos logit contributions | |
laptop | +0.148 |
isha | +0.147 |
isa | +0.146 |
acy | +0.138 |
Stacy | +0.135 |
Neg logit contributions | |
all | -0.079 |
were | -0.064 |
everywhere | -0.058 |
happily | -0.056 |
Ellie | -0.055 |
Token | so |
Token ID | 523 |
Feature activation | +4.91e-02 |
Pos logit contributions | |
isha | +0.024 |
isa | +0.023 |
laptop | +0.023 |
Stacy | +0.022 |
acy | +0.022 |
Neg logit contributions | |
all | -0.012 |
were | -0.010 |
everywhere | -0.009 |
grapes | -0.009 |
Ell | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -6.23e-03 |
Pos logit contributions | |
isha | +0.120 |
isa | +0.119 |
laptop | +0.117 |
acy | +0.112 |
Stacy | +0.111 |
Neg logit contributions | |
all | -0.060 |
were | -0.047 |
everywhere | -0.046 |
grapes | -0.046 |
Ell | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | -4.29e-03 |
Pos logit contributions | |
all | +0.008 |
were | +0.006 |
everywhere | +0.006 |
Ell | +0.006 |
grapes | +0.006 |
Neg logit contributions | |
isha | -0.015 |
isa | -0.015 |
laptop | -0.015 |
Stacy | -0.014 |
acy | -0.014 |
Token | leave |
Token ID | 2666 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
all | +0.045 |
were | +0.038 |
everywhere | +0.036 |
Ell | +0.033 |
happily | +0.033 |
Neg logit contributions | |
isha | -0.062 |
laptop | -0.061 |
isa | -0.061 |
acy | -0.057 |
Stacy | -0.056 |
Token | your |
Token ID | 534 |
Feature activation | +9.72e-02 |
Pos logit contributions | |
isha | +0.084 |
laptop | +0.084 |
isa | +0.084 |
acy | +0.078 |
Stacy | +0.078 |
Neg logit contributions | |
all | -0.044 |
were | -0.036 |
everywhere | -0.033 |
Ell | -0.032 |
happily | -0.030 |
Token | blocks |
Token ID | 7021 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
laptop | +0.205 |
isha | +0.186 |
isa | +0.181 |
aster | +0.174 |
clipboard | +0.172 |
Neg logit contributions | |
all | -0.156 |
were | -0.142 |
Ell | -0.140 |
everywhere | -0.127 |
Ellie | -0.127 |
Token | alone |
Token ID | 3436 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
all | +0.033 |
were | +0.027 |
everywhere | +0.025 |
Ell | +0.024 |
Ellie | +0.023 |
Neg logit contributions | |
laptop | -0.058 |
isha | -0.058 |
isa | -0.057 |
acy | -0.054 |
aster | -0.053 |
Token | ," |
Token ID | 553 |
Feature activation | +6.82e-02 |
Pos logit contributions | |
all | +0.056 |
were | +0.047 |
everywhere | +0.045 |
Ell | +0.042 |
grapes | +0.042 |
Neg logit contributions | |
isha | -0.068 |
laptop | -0.068 |
isa | -0.067 |
acy | -0.063 |
Stacy | -0.062 |
Token | their |
Token ID | 511 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
isa | +0.170 |
isha | +0.167 |
laptop | +0.166 |
acy | +0.157 |
den | +0.155 |
Neg logit contributions | |
all | -0.092 |
were | -0.077 |
everywhere | -0.064 |
Tim | -0.064 |
Tweet | -0.064 |
Token | mom |
Token ID | 1995 |
Feature activation | +8.52e-02 |
Pos logit contributions | |
laptop | +0.058 |
isha | +0.057 |
isa | +0.056 |
acy | +0.051 |
Stacy | +0.051 |
Neg logit contributions | |
all | -0.058 |
were | -0.051 |
everywhere | -0.049 |
Ell | -0.047 |
Ellie | -0.046 |
Token | said |
Token ID | 531 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
isha | +0.204 |
isa | +0.202 |
laptop | +0.202 |
acy | +0.189 |
Stacy | +0.188 |
Neg logit contributions | |
all | -0.114 |
were | -0.091 |
everywhere | -0.087 |
grapes | -0.080 |
lots | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | +8.87e-03 |
Pos logit contributions | |
all | +0.035 |
were | +0.027 |
everywhere | +0.026 |
happily | +0.024 |
lots | +0.022 |
Neg logit contributions | |
isha | -0.070 |
isa | -0.070 |
laptop | -0.069 |
acy | -0.066 |
Stacy | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
isa | +0.023 |
laptop | +0.023 |
isha | +0.023 |
acy | +0.022 |
aster | +0.021 |
Neg logit contributions | |
all | -0.010 |
were | -0.008 |
everywhere | -0.008 |
Tweet | -0.007 |
Ell | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
all | +0.036 |
were | +0.029 |
everywhere | +0.027 |
happily | +0.024 |
Ell | +0.024 |
Neg logit contributions | |
isha | -0.071 |
laptop | -0.071 |
isa | -0.071 |
acy | -0.067 |
Stacy | -0.066 |
Token | mom |
Token ID | 1995 |
Feature activation | +7.65e-02 |
Pos logit contributions | |
laptop | +0.208 |
isha | +0.183 |
pulse | +0.176 |
isa | +0.172 |
clipboard | +0.172 |
Neg logit contributions | |
all | -0.182 |
were | -0.158 |
Ell | -0.157 |
Ellie | -0.148 |
everywhere | -0.147 |
Token | and |
Token ID | 290 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
isha | +0.187 |
laptop | +0.185 |
isa | +0.183 |
acey | +0.172 |
Stacy | +0.170 |
Neg logit contributions | |
all | -0.094 |
grapes | -0.075 |
Tweet | -0.074 |
Ell | -0.072 |
Ellie | -0.070 |
Token | dad |
Token ID | 9955 |
Feature activation | +7.64e-02 |
Pos logit contributions | |
all | +0.062 |
were | +0.055 |
Ell | +0.055 |
everywhere | +0.054 |
Tweet | +0.053 |
Neg logit contributions | |
laptop | -0.038 |
isha | -0.038 |
isa | -0.035 |
Stacy | -0.033 |
aster | -0.032 |
Token | helped |
Token ID | 4193 |
Feature activation | +6.11e-03 |
Pos logit contributions | |
isha | +0.189 |
laptop | +0.187 |
isa | +0.183 |
acey | +0.177 |
uka | +0.172 |
Neg logit contributions | |
all | -0.081 |
Ell | -0.073 |
Tweet | -0.073 |
grapes | -0.071 |
Ellie | -0.068 |
Token | them |
Token ID | 606 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
isha | +0.017 |
laptop | +0.017 |
isa | +0.017 |
acy | +0.016 |
Stacy | +0.016 |
Neg logit contributions | |
all | -0.006 |
were | -0.004 |
everywhere | -0.004 |
happily | -0.003 |
grapes | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +8.27e-03 |
Pos logit contributions | |
all | +0.055 |
were | +0.042 |
everywhere | +0.039 |
happily | +0.038 |
lots | +0.036 |
Neg logit contributions | |
isha | -0.095 |
isa | -0.094 |
laptop | -0.089 |
acy | -0.088 |
Stacy | -0.087 |
Token | They |
Token ID | 1119 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
isha | +0.020 |
laptop | +0.020 |
isa | +0.019 |
acy | +0.018 |
Stacy | +0.018 |
Neg logit contributions | |
all | -0.011 |
were | -0.009 |
everywhere | -0.009 |
Ell | -0.008 |
happily | -0.008 |
Token | were |
Token ID | 547 |
Feature activation | -7.61e-02 |
Pos logit contributions | |
Ell | +0.025 |
grapes | +0.025 |
all | +0.025 |
Tweet | +0.024 |
Ellie | +0.024 |
Neg logit contributions | |
laptop | -0.046 |
isha | -0.045 |
acey | -0.043 |
isa | -0.043 |
uka | -0.042 |
Token | ready |
Token ID | 3492 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
all | +0.087 |
Ell | +0.083 |
were | +0.083 |
grapes | +0.076 |
Tweet | +0.075 |
Neg logit contributions | |
laptop | -0.177 |
isha | -0.173 |
isa | -0.166 |
acey | -0.163 |
uka | -0.162 |
Token | to |
Token ID | 284 |
Feature activation | +1.26e-03 |
Pos logit contributions | |
isha | +0.180 |
isa | +0.180 |
laptop | +0.178 |
acy | +0.171 |
Stacy | +0.169 |
Neg logit contributions | |
all | -0.058 |
were | -0.038 |
everywhere | -0.035 |
happily | -0.030 |
lots | -0.030 |
Token | go |
Token ID | 467 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
isa | +0.003 |
isha | +0.003 |
laptop | +0.003 |
acy | +0.003 |
Stacy | +0.003 |
Neg logit contributions | |
all | -0.002 |
were | -0.001 |
everywhere | -0.001 |
happily | -0.001 |
lots | -0.001 |
Token | followed |
Token ID | 3940 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
all | +0.035 |
aws | +0.034 |
Ell | +0.034 |
everywhere | +0.032 |
were | +0.028 |
Neg logit contributions | |
isa | -0.068 |
isha | -0.066 |
Stacy | -0.066 |
Macy | -0.065 |
laptop | -0.064 |
Token | it |
Token ID | 340 |
Feature activation | -5.68e-03 |
Pos logit contributions | |
all | +0.032 |
were | +0.021 |
everywhere | +0.021 |
happily | +0.019 |
Ellie | +0.018 |
Neg logit contributions | |
isha | -0.080 |
laptop | -0.080 |
isa | -0.079 |
acy | -0.075 |
uka | -0.074 |
Token | right |
Token ID | 826 |
Feature activation | -6.76e-02 |
Pos logit contributions | |
all | +0.007 |
were | +0.007 |
Ell | +0.007 |
Thank | +0.006 |
aws | +0.006 |
Neg logit contributions | |
laptop | -0.012 |
isa | -0.012 |
isha | -0.012 |
Stacy | -0.011 |
acy | -0.011 |
Token | through |
Token ID | 832 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
all | +0.101 |
were | +0.085 |
everywhere | +0.079 |
Ell | +0.076 |
grapes | +0.075 |
Neg logit contributions | |
isha | -0.150 |
isa | -0.149 |
laptop | -0.148 |
acy | -0.139 |
Stacy | -0.137 |
Token | the |
Token ID | 262 |
Feature activation | +6.72e-03 |
Pos logit contributions | |
isa | +0.097 |
isha | +0.097 |
laptop | +0.096 |
Stacy | +0.091 |
aster | +0.090 |
Neg logit contributions | |
all | -0.052 |
were | -0.043 |
Ell | -0.042 |
everywhere | -0.040 |
aws | -0.038 |
Token | fence |
Token ID | 13990 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
laptop | +0.012 |
isha | +0.012 |
isa | +0.012 |
aster | +0.010 |
Stacy | +0.010 |
Neg logit contributions | |
all | -0.013 |
were | -0.011 |
everywhere | -0.010 |
Ell | -0.010 |
happily | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +5.97e-03 |
Pos logit contributions | |
laptop | +0.062 |
isa | +0.061 |
isha | +0.060 |
aster | +0.056 |
Stacy | +0.056 |
Neg logit contributions | |
all | -0.045 |
Ell | -0.041 |
were | -0.039 |
Tweet | -0.037 |
Thank | -0.037 |
Token | She |
Token ID | 1375 |
Feature activation | -2.78e-03 |
Pos logit contributions | |
isha | +0.013 |
isa | +0.012 |
laptop | +0.012 |
Stacy | +0.012 |
acy | +0.012 |
Neg logit contributions | |
all | -0.009 |
were | -0.008 |
everywhere | -0.007 |
grapes | -0.007 |
Ell | -0.007 |
Token | happily |
Token ID | 18177 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
all | +0.003 |
Ell | +0.003 |
everywhere | +0.003 |
Ellie | +0.003 |
grapes | +0.003 |
Neg logit contributions | |
isa | -0.006 |
isha | -0.006 |
laptop | -0.006 |
acy | -0.006 |
Stacy | -0.006 |
Token | ate |
Token ID | 15063 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
all | +0.122 |
everywhere | +0.098 |
Ell | +0.092 |
were | +0.090 |
Ellie | +0.087 |
Neg logit contributions | |
isa | -0.289 |
isha | -0.284 |
laptop | -0.275 |
Stacy | -0.265 |
acy | -0.265 |
Token | the |
Token ID | 262 |
Feature activation | +6.71e-03 |
Pos logit contributions | |
Ell | +0.168 |
chased | +0.128 |
were | +0.128 |
aws | +0.127 |
Jerry | +0.124 |
Neg logit contributions | |
aster | -0.200 |
Stacy | -0.189 |
isa | -0.189 |
microwave | -0.187 |
Mason | -0.185 |
Token | Lily |
Token ID | 20037 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
isa | +0.013 |
isha | +0.013 |
laptop | +0.013 |
acy | +0.012 |
pulse | +0.012 |
Neg logit contributions | |
all | -0.006 |
were | -0.005 |
Tweet | -0.005 |
Ellie | -0.005 |
Tim | -0.005 |
Token | stopped |
Token ID | 5025 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
all | +0.072 |
were | +0.055 |
everywhere | +0.050 |
happily | +0.045 |
Ell | +0.043 |
Neg logit contributions | |
isha | -0.174 |
laptop | -0.173 |
isa | -0.172 |
acy | -0.163 |
Stacy | -0.162 |
Token | crying |
Token ID | 13774 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
all | +0.035 |
were | +0.026 |
everywhere | +0.026 |
happily | +0.025 |
Tweet | +0.024 |
Neg logit contributions | |
isha | -0.056 |
laptop | -0.056 |
isa | -0.055 |
acy | -0.052 |
acey | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +4.17e-03 |
Pos logit contributions | |
isha | +0.037 |
laptop | +0.037 |
isa | +0.037 |
acy | +0.035 |
Stacy | +0.035 |
Neg logit contributions | |
all | -0.014 |
were | -0.010 |
everywhere | -0.009 |
Ell | -0.009 |
grapes | -0.008 |
Token | She |
Token ID | 1375 |
Feature activation | -2.26e-03 |
Pos logit contributions | |
isa | +0.011 |
isha | +0.011 |
laptop | +0.011 |
acy | +0.010 |
aster | +0.010 |
Neg logit contributions | |
all | -0.005 |
were | -0.004 |
everywhere | -0.004 |
Ellie | -0.004 |
Tweet | -0.004 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
all | +0.003 |
everywhere | +0.002 |
grapes | +0.002 |
Ell | +0.002 |
were | +0.002 |
Neg logit contributions | |
isha | -0.005 |
isa | -0.005 |
laptop | -0.005 |
Stacy | -0.005 |
acey | -0.005 |
Token | at |
Token ID | 379 |
Feature activation | -5.68e-02 |
Pos logit contributions | |
isha | +0.028 |
laptop | +0.027 |
Stacy | +0.026 |
acey | +0.026 |
isa | +0.025 |
Neg logit contributions | |
Ell | -0.019 |
Jerry | -0.017 |
Fred | -0.017 |
Thank | -0.017 |
were | -0.017 |
Token | her |
Token ID | 607 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
all | +0.062 |
were | +0.049 |
everywhere | +0.045 |
Ell | +0.042 |
happily | +0.041 |
Neg logit contributions | |
isha | -0.146 |
laptop | -0.146 |
isa | -0.145 |
acy | -0.138 |
Stacy | -0.138 |
Token | fl |
Token ID | 781 |
Feature activation | +7.11e-03 |
Pos logit contributions | |
all | +0.041 |
everywhere | +0.033 |
were | +0.033 |
happily | +0.030 |
lots | +0.029 |
Neg logit contributions | |
isha | -0.052 |
isa | -0.052 |
laptop | -0.049 |
acy | -0.048 |
uka | -0.047 |
Token | ute |
Token ID | 1133 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
isha | +0.007 |
laptop | +0.007 |
isa | +0.007 |
acy | +0.006 |
Stacy | +0.006 |
Neg logit contributions | |
all | -0.020 |
were | -0.018 |
everywhere | -0.017 |
happily | -0.017 |
Ell | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
laptop | +0.421 |
isha | +0.413 |
isa | +0.411 |
Stacy | +0.387 |
acy | +0.385 |
Neg logit contributions | |
all | -0.203 |
were | -0.166 |
Ell | -0.151 |
everywhere | -0.150 |
grapes | -0.145 |
Token | pretty |
Token ID | 2495 |
Feature activation | -8.85e-02 |
Pos logit contributions | |
laptop | +0.014 |
isha | +0.014 |
isa | +0.013 |
acy | +0.013 |
Stacy | +0.013 |
Neg logit contributions | |
all | -0.006 |
were | -0.005 |
everywhere | -0.004 |
Ell | -0.004 |
grapes | -0.004 |
Token | dress |
Token ID | 6576 |
Feature activation | -8.77e-03 |
Pos logit contributions | |
all | +0.139 |
were | +0.116 |
everywhere | +0.110 |
happily | +0.102 |
Ell | +0.102 |
Neg logit contributions | |
isha | -0.192 |
laptop | -0.191 |
isa | -0.190 |
acy | -0.177 |
Stacy | -0.175 |
Token | and |
Token ID | 290 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
all | +0.009 |
were | +0.007 |
Ell | +0.006 |
everywhere | +0.006 |
grapes | +0.006 |
Neg logit contributions | |
laptop | -0.023 |
isha | -0.023 |
isa | -0.022 |
acy | -0.021 |
Stacy | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | +6.88e-03 |
Pos logit contributions | |
all | +0.045 |
everywhere | +0.037 |
were | +0.037 |
Ell | +0.036 |
happily | +0.033 |
Neg logit contributions | |
isha | -0.084 |
isa | -0.083 |
laptop | -0.082 |
Stacy | -0.079 |
acy | -0.077 |
Token | big |
Token ID | 1263 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
isha | +0.017 |
laptop | +0.017 |
isa | +0.017 |
acy | +0.016 |
Stacy | +0.016 |
Neg logit contributions | |
all | -0.008 |
were | -0.007 |
everywhere | -0.006 |
Ell | -0.006 |
grapes | -0.006 |
Token | smile |
Token ID | 8212 |
Feature activation | -8.19e-02 |
Pos logit contributions | |
all | +0.067 |
were | +0.055 |
everywhere | +0.053 |
happily | +0.049 |
grapes | +0.048 |
Neg logit contributions | |
isha | -0.092 |
isa | -0.092 |
laptop | -0.091 |
acy | -0.086 |
Stacy | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | +6.73e-03 |
Pos logit contributions | |
all | +0.108 |
were | +0.087 |
everywhere | +0.081 |
happily | +0.074 |
Ell | +0.073 |
Neg logit contributions | |
isha | -0.199 |
laptop | -0.198 |
isa | -0.197 |
acy | -0.186 |
Stacy | -0.184 |
Token | Mia |
Token ID | 32189 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
isa | +0.013 |
isha | +0.013 |
laptop | +0.013 |
acy | +0.013 |
Stacy | +0.012 |
Neg logit contributions | |
all | -0.011 |
everywhere | -0.009 |
were | -0.009 |
Ell | -0.009 |
Ellie | -0.009 |
Token | took |
Token ID | 1718 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
isha | +0.065 |
isa | +0.065 |
laptop | +0.064 |
acy | +0.061 |
acey | +0.060 |
Neg logit contributions | |
all | -0.022 |
grapes | -0.015 |
everywhere | -0.015 |
Ell | -0.015 |
Tweet | -0.014 |
Token | her |
Token ID | 607 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
all | +0.058 |
were | +0.051 |
Ell | +0.050 |
everywhere | +0.046 |
grapes | +0.044 |
Neg logit contributions | |
isha | -0.110 |
isa | -0.109 |
laptop | -0.107 |
Stacy | -0.107 |
acy | -0.105 |
Token | doll |
Token ID | 3654 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
all | +0.029 |
were | +0.024 |
everywhere | +0.023 |
Ell | +0.022 |
happily | +0.021 |
Neg logit contributions | |
isha | -0.038 |
laptop | -0.038 |
isa | -0.037 |
acy | -0.035 |
Stacy | -0.034 |
Token | Good |
Token ID | 10248 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
isha | +0.008 |
isa | +0.007 |
laptop | +0.007 |
Stacy | +0.007 |
acy | +0.007 |
Neg logit contributions | |
all | -0.004 |
were | -0.003 |
everywhere | -0.003 |
happily | -0.003 |
Ellie | -0.003 |
Token | ," |
Token ID | 553 |
Feature activation | +7.06e-02 |
Pos logit contributions | |
all | +0.174 |
were | +0.137 |
fun | +0.133 |
fair | +0.131 |
happily | +0.130 |
Neg logit contributions | |
isha | -0.221 |
laptop | -0.219 |
isa | -0.219 |
acy | -0.210 |
Stacy | -0.207 |
Token | she |
Token ID | 673 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
laptop | +0.143 |
isha | +0.139 |
isa | +0.135 |
Stacy | +0.133 |
aster | +0.131 |
Neg logit contributions | |
all | -0.111 |
everywhere | -0.094 |
Ell | -0.093 |
were | -0.092 |
happily | -0.087 |
Token | says |
Token ID | 1139 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
all | +0.058 |
were | +0.050 |
everywhere | +0.048 |
happily | +0.046 |
Ell | +0.046 |
Neg logit contributions | |
isha | -0.050 |
laptop | -0.050 |
isa | -0.049 |
acy | -0.045 |
Stacy | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | +1.96e-03 |
Pos logit contributions | |
isha | +0.030 |
isa | +0.030 |
laptop | +0.029 |
acy | +0.028 |
uka | +0.028 |
Neg logit contributions | |
all | -0.014 |
were | -0.010 |
everywhere | -0.010 |
happily | -0.009 |
grapes | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
isha | +0.005 |
laptop | +0.005 |
isa | +0.005 |
acy | +0.005 |
Stacy | +0.005 |
Neg logit contributions | |
all | -0.002 |
were | -0.002 |
everywhere | -0.002 |
happily | -0.002 |
Ell | -0.002 |
Token | Now |
Token ID | 3844 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
all | +0.144 |
were | +0.116 |
everywhere | +0.107 |
happily | +0.101 |
Ell | +0.101 |
Neg logit contributions | |
isha | -0.271 |
laptop | -0.271 |
isa | -0.268 |
acy | -0.252 |
Stacy | -0.249 |
Token | , |
Token ID | 11 |
Feature activation | +1.56e-03 |
Pos logit contributions | |
isha | +0.063 |
isa | +0.062 |
laptop | +0.061 |
acy | +0.059 |
uka | +0.057 |
Neg logit contributions | |
all | -0.036 |
were | -0.028 |
everywhere | -0.025 |
fun | -0.023 |
happily | -0.023 |
Token | why |
Token ID | 1521 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
isha | +0.003 |
laptop | +0.003 |
isa | +0.003 |
acy | +0.003 |
Stacy | +0.003 |
Neg logit contributions | |
all | -0.002 |
were | -0.002 |
everywhere | -0.002 |
Ell | -0.002 |
happily | -0.002 |
Token | don |
Token ID | 836 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
isa | +0.164 |
isha | +0.164 |
laptop | +0.162 |
acy | +0.153 |
Stacy | +0.151 |
Neg logit contributions | |
all | -0.114 |
were | -0.095 |
everywhere | -0.087 |
lots | -0.079 |
happily | -0.079 |
Token | 't |
Token ID | 470 |
Feature activation | +5.96e-02 |
Pos logit contributions | |
isa | +0.109 |
isha | +0.106 |
laptop | +0.105 |
acy | +0.097 |
den | +0.097 |
Neg logit contributions | |
all | -0.086 |
were | -0.071 |
everywhere | -0.066 |
Ellie | -0.062 |
lots | -0.060 |
Token | around |
Token ID | 1088 |
Feature activation | -9.46e-03 |
Pos logit contributions | |
laptop | +0.168 |
isha | +0.167 |
isa | +0.166 |
acy | +0.157 |
Stacy | +0.156 |
Neg logit contributions | |
all | -0.075 |
were | -0.063 |
Ell | -0.059 |
grapes | -0.059 |
Tweet | -0.055 |
Token | quickly |
Token ID | 2952 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
all | +0.010 |
were | +0.007 |
everywhere | +0.007 |
happily | +0.006 |
lots | +0.006 |
Neg logit contributions | |
isha | -0.026 |
laptop | -0.026 |
isa | -0.026 |
acy | -0.024 |
Stacy | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | +3.80e-03 |
Pos logit contributions | |
isha | +0.080 |
isa | +0.080 |
laptop | +0.078 |
acy | +0.075 |
uka | +0.075 |
Neg logit contributions | |
all | -0.028 |
were | -0.020 |
everywhere | -0.018 |
happily | -0.015 |
lots | -0.015 |
Token | make |
Token ID | 787 |
Feature activation | +9.70e-02 |
Pos logit contributions | |
isha | +0.009 |
laptop | +0.008 |
isa | +0.008 |
acey | +0.008 |
den | +0.008 |
Neg logit contributions | |
all | -0.007 |
were | -0.005 |
lots | -0.005 |
everywhere | -0.005 |
happily | -0.005 |
Token | sure |
Token ID | 1654 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
laptop | +0.250 |
isha | +0.249 |
isa | +0.248 |
acy | +0.233 |
Stacy | +0.229 |
Neg logit contributions | |
all | -0.120 |
were | -0.090 |
everywhere | -0.083 |
lots | -0.078 |
happily | -0.075 |
Token | all |
Token ID | 477 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
isha | +0.061 |
laptop | +0.061 |
isa | +0.060 |
acy | +0.057 |
Stacy | +0.057 |
Neg logit contributions | |
all | -0.030 |
were | -0.024 |
everywhere | -0.023 |
Ell | -0.022 |
grapes | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
all | +0.144 |
were | +0.105 |
everywhere | +0.093 |
lots | +0.080 |
Ellie | +0.077 |
Neg logit contributions | |
isha | -0.380 |
laptop | -0.379 |
isa | -0.375 |
acy | -0.355 |
Stacy | -0.352 |
Token | dust |
Token ID | 8977 |
Feature activation | +6.93e-02 |
Pos logit contributions | |
isha | +0.008 |
isa | +0.008 |
laptop | +0.008 |
acy | +0.008 |
aster | +0.008 |
Neg logit contributions | |
all | -0.006 |
were | -0.005 |
everywhere | -0.004 |
fun | -0.004 |
lots | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
isha | +0.175 |
laptop | +0.175 |
isa | +0.173 |
acy | +0.164 |
Stacy | +0.163 |
Neg logit contributions | |
all | -0.080 |
were | -0.061 |
everywhere | -0.058 |
Ell | -0.056 |
grapes | -0.054 |
Token | gone |
Token ID | 3750 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
all | +0.041 |
were | +0.035 |
everywhere | +0.032 |
Ell | +0.031 |
grapes | +0.031 |
Neg logit contributions | |
isha | -0.065 |
laptop | -0.064 |
isa | -0.064 |
acy | -0.060 |
Stacy | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +7.84e-03 |
Pos logit contributions | |
all | +0.114 |
were | +0.093 |
everywhere | +0.086 |
Ell | +0.082 |
grapes | +0.081 |
Neg logit contributions | |
isha | -0.201 |
laptop | -0.200 |
isa | -0.198 |
acy | -0.187 |
Stacy | -0.185 |
Token | happened |
Token ID | 3022 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
laptop | +0.089 |
isha | +0.088 |
isa | +0.085 |
aster | +0.083 |
uka | +0.081 |
Neg logit contributions | |
all | -0.054 |
Ell | -0.045 |
everywhere | -0.044 |
bou | -0.042 |
happily | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
all | +0.108 |
were | +0.085 |
everywhere | +0.080 |
happily | +0.072 |
lots | +0.070 |
Neg logit contributions | |
isha | -0.209 |
isa | -0.208 |
laptop | -0.207 |
acy | -0.196 |
Stacy | -0.194 |
Token | They |
Token ID | 1119 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
isha | +0.035 |
laptop | +0.035 |
isa | +0.035 |
acy | +0.032 |
Stacy | +0.032 |
Neg logit contributions | |
all | -0.020 |
were | -0.016 |
everywhere | -0.015 |
happily | -0.014 |
Ell | -0.014 |
Token | told |
Token ID | 1297 |
Feature activation | +9.13e-04 |
Pos logit contributions | |
grapes | +0.022 |
Tweet | +0.021 |
Ell | +0.021 |
butterflies | +0.020 |
aws | +0.020 |
Neg logit contributions | |
isha | -0.027 |
laptop | -0.026 |
acey | -0.025 |
isa | -0.025 |
uka | -0.024 |
Token | her |
Token ID | 607 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
laptop | +0.002 |
isha | +0.002 |
isa | +0.002 |
aster | +0.002 |
Stacy | +0.002 |
Neg logit contributions | |
all | -0.001 |
Ell | -0.001 |
were | -0.001 |
Thank | -0.001 |
happily | -0.001 |
Token | about |
Token ID | 546 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
all | +0.014 |
were | +0.011 |
everywhere | +0.011 |
Ell | +0.010 |
happily | +0.010 |
Neg logit contributions | |
laptop | -0.030 |
isha | -0.030 |
isa | -0.029 |
acy | -0.028 |
Stacy | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +5.77e-03 |
Pos logit contributions | |
all | +0.111 |
were | +0.097 |
Ell | +0.094 |
everywhere | +0.089 |
happily | +0.086 |
Neg logit contributions | |
laptop | -0.207 |
isha | -0.201 |
isa | -0.198 |
aster | -0.193 |
Stacy | -0.189 |
Token | jewel |
Token ID | 16927 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
laptop | +0.010 |
lung | +0.009 |
pulse | +0.009 |
clipboard | +0.008 |
fever | +0.008 |
Neg logit contributions | |
Ell | -0.010 |
ashed | -0.009 |
were | -0.009 |
all | -0.009 |
Da | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
all | +0.163 |
Ell | +0.135 |
were | +0.125 |
Thank | +0.122 |
everywhere | +0.117 |
Neg logit contributions | |
laptop | -0.353 |
isha | -0.340 |
isa | -0.338 |
acy | -0.316 |
aster | -0.315 |
Token | how |
Token ID | 703 |
Feature activation | +5.88e-03 |
Pos logit contributions | |
all | +0.034 |
Ell | +0.031 |
aws | +0.027 |
everywhere | +0.027 |
were | +0.027 |
Neg logit contributions | |
laptop | -0.060 |
isha | -0.057 |
isa | -0.056 |
aster | -0.055 |
Stacy | -0.054 |
Token | they |
Token ID | 484 |
Feature activation | -8.04e-03 |
Pos logit contributions | |
isha | +0.012 |
laptop | +0.011 |
isa | +0.011 |
aster | +0.011 |
acy | +0.011 |
Neg logit contributions | |
all | -0.009 |
Ell | -0.008 |
were | -0.007 |
everywhere | -0.007 |
grapes | -0.007 |
Token | ball |
Token ID | 2613 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
laptop | +0.007 |
isha | +0.006 |
Stacy | +0.006 |
aster | +0.006 |
clipboard | +0.006 |
Neg logit contributions | |
all | -0.005 |
Ell | -0.005 |
were | -0.004 |
Ellie | -0.004 |
happily | -0.004 |
Token | struck |
Token ID | 7425 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
Ell | +0.162 |
all | +0.153 |
grapes | +0.143 |
Thank | +0.140 |
Ellie | +0.139 |
Neg logit contributions | |
laptop | -0.255 |
isa | -0.238 |
isha | -0.233 |
asy | -0.232 |
aster | -0.230 |
Token | a |
Token ID | 257 |
Feature activation | +4.65e-03 |
Pos logit contributions | |
all | +0.036 |
were | +0.030 |
everywhere | +0.028 |
Ell | +0.027 |
grapes | +0.026 |
Neg logit contributions | |
isha | -0.063 |
isa | -0.063 |
laptop | -0.063 |
acy | -0.059 |
aster | -0.058 |
Token | tree |
Token ID | 5509 |
Feature activation | -6.39e-02 |
Pos logit contributions | |
laptop | +0.011 |
isha | +0.011 |
isa | +0.010 |
acy | +0.010 |
aster | +0.010 |
Neg logit contributions | |
all | -0.006 |
were | -0.005 |
everywhere | -0.005 |
Ell | -0.005 |
Ellie | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
all | +0.089 |
were | +0.076 |
Ell | +0.069 |
everywhere | +0.069 |
grapes | +0.068 |
Neg logit contributions | |
laptop | -0.144 |
isa | -0.140 |
isha | -0.140 |
aster | -0.132 |
acy | -0.130 |
Token | Tim |
Token ID | 5045 |
Feature activation | -9.43e-02 |
Pos logit contributions | |
isha | +0.022 |
laptop | +0.022 |
isa | +0.022 |
acy | +0.021 |
Stacy | +0.020 |
Neg logit contributions | |
all | -0.018 |
were | -0.015 |
everywhere | -0.014 |
happily | -0.013 |
Ell | -0.013 |
Token | my |
Token ID | 1820 |
Feature activation | +8.22e-02 |
Pos logit contributions | |
all | +0.145 |
were | +0.120 |
everywhere | +0.113 |
happily | +0.104 |
Ell | +0.103 |
Neg logit contributions | |
isha | -0.211 |
laptop | -0.211 |
isa | -0.208 |
acy | -0.197 |
Stacy | -0.195 |
Token | 's |
Token ID | 338 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
isha | +0.172 |
isa | +0.172 |
laptop | +0.170 |
acey | +0.155 |
Stacy | +0.155 |
Neg logit contributions | |
all | -0.122 |
Ell | -0.101 |
Ellie | -0.101 |
everywhere | -0.101 |
grapes | -0.099 |
Token | dad |
Token ID | 9955 |
Feature activation | +7.14e-02 |
Pos logit contributions | |
laptop | +0.167 |
isha | +0.162 |
isa | +0.155 |
acy | +0.142 |
aster | +0.140 |
Neg logit contributions | |
all | -0.249 |
were | -0.223 |
Ell | -0.222 |
Ellie | -0.216 |
Tweet | -0.215 |
Token | said |
Token ID | 531 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
isha | +0.159 |
isa | +0.156 |
laptop | +0.155 |
acey | +0.147 |
Stacy | +0.144 |
Neg logit contributions | |
all | -0.093 |
grapes | -0.080 |
Ell | -0.077 |
aws | -0.076 |
butterflies | -0.074 |
Token | , |
Token ID | 11 |
Feature activation | +6.44e-03 |
Pos logit contributions | |
all | +0.049 |
Ell | +0.041 |
everywhere | +0.041 |
were | +0.040 |
aws | +0.038 |
Neg logit contributions | |
isha | -0.071 |
laptop | -0.070 |
isa | -0.068 |
aster | -0.066 |
Stacy | -0.065 |