Token | One |
Token ID | 3198 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
favorite | +0.025 |
o | +0.024 |
bie | +0.024 |
eps | +0.021 |
bo | +0.020 |
Neg logit contributions | |
Ivy | -0.038 |
Iris | -0.036 |
Holly | -0.034 |
Beat | -0.034 |
ner | -0.033 |
Token | day |
Token ID | 1110 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
eps | +0.043 |
favorite | +0.041 |
o | +0.040 |
bie | +0.039 |
drawn | +0.038 |
Neg logit contributions | |
Ivy | -0.058 |
Holly | -0.054 |
Iris | -0.054 |
Fiona | -0.053 |
Beat | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | -7.05e-03 |
Pos logit contributions | |
o | +0.018 |
favorite | +0.017 |
bie | +0.016 |
at | +0.015 |
bright | +0.014 |
Neg logit contributions | |
Ivy | -0.038 |
Iris | -0.037 |
ner | -0.035 |
ocated | -0.035 |
Holly | -0.035 |
Token | when |
Token ID | 618 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
Ivy | +0.014 |
Iris | +0.014 |
Holly | +0.013 |
Beat | +0.013 |
ner | +0.013 |
Neg logit contributions | |
favorite | -0.010 |
o | -0.010 |
bie | -0.010 |
eps | -0.009 |
bo | -0.008 |
Token | Ab |
Token ID | 2275 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
favorite | +0.070 |
o | +0.068 |
bie | +0.066 |
bright | +0.058 |
hot | +0.055 |
Neg logit contributions | |
Ivy | -0.108 |
Iris | -0.102 |
Holly | -0.098 |
ner | -0.098 |
Beat | -0.096 |
Token | bie |
Token ID | 12590 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
favorite | +0.092 |
bright | +0.080 |
o | +0.080 |
eps | +0.080 |
tallest | +0.078 |
Neg logit contributions | |
Ivy | -0.082 |
Holly | -0.076 |
ner | -0.073 |
Iris | -0.072 |
Fiona | -0.070 |
Token | went |
Token ID | 1816 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
favorite | +0.235 |
o | +0.226 |
bie | +0.218 |
eps | +0.193 |
v | +0.182 |
Neg logit contributions | |
Ivy | -0.420 |
Iris | -0.396 |
Holly | -0.385 |
Beat | -0.378 |
ner | -0.376 |
Token | to |
Token ID | 284 |
Feature activation | -2.65e-03 |
Pos logit contributions | |
o | +0.020 |
favorite | +0.019 |
bie | +0.015 |
at | +0.014 |
big | +0.014 |
Neg logit contributions | |
Ivy | -0.070 |
Iris | -0.065 |
ner | -0.064 |
Beat | -0.062 |
Holly | -0.062 |
Token | the |
Token ID | 262 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
Ivy | +0.006 |
Iris | +0.006 |
ner | +0.006 |
Holly | +0.006 |
Beat | +0.006 |
Neg logit contributions | |
favorite | -0.003 |
o | -0.003 |
bie | -0.003 |
bright | -0.002 |
big | -0.002 |
Token | store |
Token ID | 3650 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
o | +0.025 |
favorite | +0.025 |
bie | +0.024 |
eps | +0.022 |
bo | +0.021 |
Neg logit contributions | |
Ivy | -0.030 |
Iris | -0.029 |
Holly | -0.027 |
Beat | -0.027 |
ner | -0.027 |
Token | with |
Token ID | 351 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
Ivy | +0.052 |
Iris | +0.049 |
Holly | +0.047 |
Beat | +0.046 |
ner | +0.046 |
Neg logit contributions | |
favorite | -0.032 |
o | -0.031 |
bie | -0.029 |
bright | -0.025 |
eps | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -2.95e-03 |
Pos logit contributions | |
Ivy | +0.041 |
Iris | +0.038 |
Holly | +0.037 |
Beat | +0.036 |
ner | +0.036 |
Neg logit contributions | |
o | -0.030 |
favorite | -0.029 |
bie | -0.029 |
eps | -0.026 |
bo | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +3.48e-03 |
Pos logit contributions | |
Ivy | +0.006 |
Iris | +0.006 |
Holly | +0.006 |
Beat | +0.005 |
ner | +0.005 |
Neg logit contributions | |
o | -0.004 |
favorite | -0.004 |
bie | -0.004 |
eps | -0.003 |
bo | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +8.56e-03 |
Pos logit contributions | |
o | +0.003 |
favorite | +0.003 |
bie | +0.003 |
eps | +0.002 |
bo | +0.002 |
Neg logit contributions | |
Ivy | -0.009 |
Iris | -0.009 |
Holly | -0.008 |
Beat | -0.008 |
ner | -0.008 |
Token | When |
Token ID | 2215 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
o | +0.015 |
favorite | +0.014 |
bie | +0.014 |
v | +0.013 |
eps | +0.013 |
Neg logit contributions | |
Ivy | -0.015 |
Iris | -0.014 |
Holly | -0.014 |
Beat | -0.013 |
ner | -0.013 |
Token | Ab |
Token ID | 2275 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
Ivy | +0.018 |
Iris | +0.017 |
Holly | +0.016 |
ner | +0.016 |
Beat | +0.016 |
Neg logit contributions | |
favorite | -0.021 |
o | -0.021 |
bie | -0.020 |
bright | -0.018 |
hot | -0.018 |
Token | bie |
Token ID | 12590 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
favorite | +0.031 |
bright | +0.024 |
drawn | +0.024 |
tallest | +0.023 |
eps | +0.023 |
Neg logit contributions | |
Ivy | -0.084 |
Holly | -0.080 |
Fiona | -0.079 |
ion | -0.076 |
Iris | -0.076 |
Token | got |
Token ID | 1392 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
favorite | +0.234 |
o | +0.219 |
bie | +0.214 |
eps | +0.188 |
tallest | +0.186 |
Neg logit contributions | |
Ivy | -0.400 |
ner | -0.372 |
Iris | -0.371 |
Holly | -0.367 |
Beat | -0.356 |
Token | home |
Token ID | 1363 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
favorite | +0.053 |
o | +0.052 |
bie | +0.050 |
eps | +0.046 |
bo | +0.045 |
Neg logit contributions | |
Ivy | -0.079 |
Iris | -0.074 |
Holly | -0.072 |
Beat | -0.069 |
ner | -0.069 |
Token | , |
Token ID | 11 |
Feature activation | -5.08e-03 |
Pos logit contributions | |
Ivy | +0.068 |
Iris | +0.065 |
Holly | +0.062 |
Beat | +0.062 |
ner | +0.060 |
Neg logit contributions | |
favorite | -0.038 |
o | -0.037 |
bie | -0.036 |
bright | -0.031 |
big | -0.030 |
Token | she |
Token ID | 673 |
Feature activation | +3.90e-03 |
Pos logit contributions | |
Ivy | +0.007 |
Iris | +0.006 |
Holly | +0.006 |
Fiona | +0.006 |
Beat | +0.006 |
Neg logit contributions | |
o | -0.011 |
favorite | -0.011 |
bie | -0.010 |
eps | -0.010 |
bo | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | -7.50e-03 |
Pos logit contributions | |
favorite | +0.005 |
o | +0.005 |
bie | +0.004 |
eps | +0.004 |
bo | +0.004 |
Neg logit contributions | |
Ivy | -0.009 |
Iris | -0.008 |
Holly | -0.008 |
ner | -0.008 |
Beat | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
o | +0.009 |
favorite | +0.008 |
bie | +0.008 |
v | +0.007 |
bo | +0.007 |
Neg logit contributions | |
Ivy | -0.020 |
Iris | -0.019 |
Holly | -0.019 |
Beat | -0.018 |
Fiona | -0.018 |
Token | surprise |
Token ID | 5975 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
Ivy | +0.050 |
Iris | +0.047 |
Holly | +0.046 |
Beat | +0.045 |
ner | +0.044 |
Neg logit contributions | |
o | -0.023 |
favorite | -0.022 |
bie | -0.021 |
v | -0.019 |
eps | -0.019 |
Token | ! |
Token ID | 0 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
Ivy | +0.082 |
Iris | +0.078 |
Holly | +0.075 |
Beat | +0.074 |
ner | +0.074 |
Neg logit contributions | |
favorite | -0.044 |
o | -0.044 |
bie | -0.042 |
eps | -0.036 |
bright | -0.034 |
Token | In |
Token ID | 554 |
Feature activation | -8.14e-04 |
Pos logit contributions | |
favorite | +0.036 |
o | +0.036 |
bie | +0.035 |
eps | +0.030 |
bo | +0.030 |
Neg logit contributions | |
Ivy | -0.046 |
Iris | -0.043 |
Holly | -0.042 |
Beat | -0.041 |
ner | -0.041 |
Token | Mar |
Token ID | 1526 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
Ivy | +0.002 |
Iris | +0.002 |
Holly | +0.002 |
Beat | +0.002 |
ner | +0.002 |
Neg logit contributions | |
favorite | -0.001 |
o | -0.001 |
bie | -0.001 |
eps | -0.001 |
bo | -0.001 |
Token | iah |
Token ID | 9520 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
favorite | +0.112 |
tallest | +0.089 |
o | +0.087 |
bright | +0.086 |
favourite | +0.085 |
Neg logit contributions | |
ner | -0.123 |
Ivy | -0.121 |
Holly | -0.116 |
Fiona | -0.115 |
Hazel | -0.111 |
Token | 's |
Token ID | 338 |
Feature activation | +7.01e-02 |
Pos logit contributions | |
favorite | +0.214 |
o | +0.193 |
bie | +0.181 |
tallest | +0.163 |
ams | +0.162 |
Neg logit contributions | |
Ivy | -0.412 |
ner | -0.388 |
Iris | -0.385 |
Holly | -0.382 |
Beat | -0.373 |
Token | hands |
Token ID | 2832 |
Feature activation | +7.91e-02 |
Pos logit contributions | |
o | +0.118 |
eps | +0.112 |
ams | +0.108 |
bie | +0.108 |
v | +0.107 |
Neg logit contributions | |
Ivy | -0.129 |
Iris | -0.122 |
Holly | -0.120 |
Beat | -0.114 |
Fiona | -0.114 |
Token | was |
Token ID | 373 |
Feature activation | +5.11e-03 |
Pos logit contributions | |
o | +0.100 |
favorite | +0.098 |
bie | +0.094 |
eps | +0.083 |
bright | +0.080 |
Neg logit contributions | |
Ivy | -0.177 |
Iris | -0.170 |
Beat | -0.160 |
Holly | -0.160 |
ner | -0.155 |
Token | a |
Token ID | 257 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
o | +0.006 |
favorite | +0.006 |
bie | +0.006 |
bo | +0.005 |
v | +0.005 |
Neg logit contributions | |
Ivy | -0.012 |
Holly | -0.011 |
Iris | -0.011 |
Fiona | -0.010 |
Beat | -0.010 |
Token | popular |
Token ID | 2968 |
Feature activation | -5.48e-03 |
Pos logit contributions | |
Ivy | +0.056 |
Iris | +0.052 |
Holly | +0.052 |
Beat | +0.050 |
Fiona | +0.049 |
Neg logit contributions | |
o | -0.027 |
favorite | -0.024 |
bie | -0.024 |
v | -0.024 |
eps | -0.023 |
Token | should |
Token ID | 815 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
Ivy | +0.017 |
Iris | +0.016 |
Holly | +0.015 |
Beat | +0.015 |
ner | +0.015 |
Neg logit contributions | |
favorite | -0.014 |
o | -0.014 |
bie | -0.013 |
eps | -0.012 |
ams | -0.011 |
Token | not |
Token ID | 407 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
Ivy | +0.068 |
ner | +0.067 |
Iris | +0.066 |
Holly | +0.064 |
Beat | +0.064 |
Neg logit contributions | |
o | -0.028 |
favorite | -0.027 |
bo | -0.024 |
bie | -0.024 |
bright | -0.023 |
Token | tease |
Token ID | 34691 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
o | +0.070 |
favorite | +0.066 |
bie | +0.063 |
bo | +0.057 |
eps | +0.057 |
Neg logit contributions | |
Ivy | -0.110 |
Iris | -0.104 |
Holly | -0.102 |
Beat | -0.100 |
ner | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
Ivy | +0.028 |
Iris | +0.027 |
Holly | +0.025 |
Beat | +0.025 |
Fiona | +0.024 |
Neg logit contributions | |
favorite | -0.026 |
o | -0.026 |
bie | -0.026 |
eps | -0.023 |
v | -0.022 |
Token | ost |
Token ID | 23619 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
o | +0.039 |
bie | +0.037 |
favorite | +0.037 |
vice | +0.034 |
bo | +0.034 |
Neg logit contributions | |
Ivy | -0.039 |
Iris | -0.036 |
Holly | -0.034 |
Fiona | -0.033 |
ner | -0.033 |
Token | ric |
Token ID | 1173 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
favorite | +0.113 |
favourite | +0.101 |
bright | +0.100 |
huge | +0.098 |
vice | +0.097 |
Neg logit contributions | |
ion | -0.116 |
Ivy | -0.111 |
ner | -0.108 |
otion | -0.102 |
Iris | -0.102 |
Token | hes |
Token ID | 956 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
favorite | +0.288 |
bie | +0.284 |
favorites | +0.256 |
favourite | +0.252 |
favourites | +0.239 |
Neg logit contributions | |
ion | -0.312 |
ner | -0.296 |
Ivy | -0.295 |
otion | -0.274 |
Iris | -0.267 |
Token | . |
Token ID | 13 |
Feature activation | +3.10e-03 |
Pos logit contributions | |
o | +0.056 |
favorite | +0.055 |
bie | +0.051 |
eps | +0.044 |
bo | +0.043 |
Neg logit contributions | |
Ivy | -0.130 |
Iris | -0.125 |
Holly | -0.119 |
Beat | -0.119 |
ner | -0.117 |
Token | They |
Token ID | 1119 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
o | +0.005 |
favorite | +0.005 |
bie | +0.005 |
eps | +0.004 |
bo | +0.004 |
Neg logit contributions | |
Ivy | -0.006 |
Iris | -0.005 |
Holly | -0.005 |
Beat | -0.005 |
ner | -0.005 |
Token | are |
Token ID | 389 |
Feature activation | -9.46e-03 |
Pos logit contributions | |
Ivy | +0.033 |
Iris | +0.030 |
Holly | +0.029 |
Beat | +0.029 |
ner | +0.028 |
Neg logit contributions | |
o | -0.028 |
favorite | -0.028 |
bie | -0.028 |
eps | -0.024 |
bo | -0.024 |
Token | not |
Token ID | 407 |
Feature activation | +9.75e-03 |
Pos logit contributions | |
Ivy | +0.021 |
Iris | +0.020 |
Holly | +0.019 |
Beat | +0.019 |
ner | +0.019 |
Neg logit contributions | |
o | -0.012 |
favorite | -0.012 |
bie | -0.011 |
eps | -0.010 |
v | -0.010 |
Token | would |
Token ID | 561 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
Ivy | +0.071 |
Iris | +0.067 |
Holly | +0.064 |
Beat | +0.064 |
ner | +0.063 |
Neg logit contributions | |
favorite | -0.040 |
o | -0.040 |
bie | -0.037 |
eps | -0.033 |
bo | -0.032 |
Token | fit |
Token ID | 4197 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
Ivy | +0.023 |
Iris | +0.022 |
Holly | +0.021 |
Beat | +0.021 |
ner | +0.021 |
Neg logit contributions | |
favorite | -0.014 |
o | -0.012 |
bright | -0.012 |
bie | -0.011 |
big | -0.011 |
Token | perfectly |
Token ID | 7138 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
Ivy | +0.101 |
Iris | +0.097 |
Beat | +0.092 |
Holly | +0.092 |
ner | +0.091 |
Neg logit contributions | |
favorite | -0.043 |
o | -0.042 |
bie | -0.039 |
eps | -0.033 |
bright | -0.032 |
Token | in |
Token ID | 287 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
o | +0.007 |
favorite | +0.006 |
bie | +0.006 |
eps | +0.005 |
bo | +0.005 |
Neg logit contributions | |
Ivy | -0.013 |
Iris | -0.012 |
Holly | -0.012 |
ner | -0.012 |
Beat | -0.011 |
Token | R |
Token ID | 371 |
Feature activation | +9.83e-02 |
Pos logit contributions | |
Ivy | +0.045 |
Iris | +0.043 |
Holly | +0.041 |
Beat | +0.040 |
ner | +0.040 |
Neg logit contributions | |
favorite | -0.026 |
o | -0.026 |
bie | -0.025 |
eps | -0.022 |
bright | -0.021 |
Token | avi |
Token ID | 15820 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
favorite | +0.153 |
o | +0.149 |
bie | +0.145 |
eps | +0.135 |
tallest | +0.130 |
Neg logit contributions | |
Ivy | -0.189 |
Iris | -0.176 |
Holly | -0.170 |
ner | -0.168 |
Beat | -0.166 |
Token | â |
Token ID | 22940 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
favorite | +0.194 |
o | +0.176 |
bie | +0.174 |
eps | +0.159 |
vice | +0.146 |
Neg logit contributions | |
Ivy | -0.391 |
ner | -0.362 |
Holly | -0.362 |
Iris | -0.357 |
Fiona | -0.344 |
Token | € |
Token ID | 26391 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
Ivy | +0.021 |
Iris | +0.020 |
Holly | +0.019 |
Beat | +0.019 |
ner | +0.019 |
Neg logit contributions | |
favorite | -0.018 |
o | -0.017 |
bie | -0.017 |
eps | -0.015 |
bright | -0.014 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
favorite | +0.043 |
tallest | +0.041 |
bie | +0.040 |
eps | +0.039 |
o | +0.039 |
Neg logit contributions | |
Ivy | -0.054 |
Beat | -0.051 |
ner | -0.051 |
Holly | -0.050 |
Iris | -0.050 |
Token | s |
Token ID | 82 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
favorite | +0.149 |
o | +0.139 |
bie | +0.138 |
eps | +0.124 |
vice | +0.115 |
Neg logit contributions | |
Ivy | -0.295 |
Iris | -0.276 |
Holly | -0.272 |
ner | -0.269 |
Beat | -0.264 |
Token | room |
Token ID | 2119 |
Feature activation | -9.12e-04 |
Pos logit contributions | |
o | +0.049 |
bie | +0.044 |
favorite | +0.042 |
eps | +0.042 |
bo | +0.039 |
Neg logit contributions | |
Ivy | -0.078 |
Holly | -0.072 |
Iris | -0.072 |
Beat | -0.069 |
ner | -0.068 |
Token | ie |
Token ID | 494 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
favorite | +0.043 |
bie | +0.040 |
vice | +0.040 |
o | +0.040 |
eps | +0.039 |
Neg logit contributions | |
Ivy | -0.039 |
ner | -0.035 |
Holly | -0.035 |
Iris | -0.035 |
Fiona | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -4.87e-03 |
Pos logit contributions | |
Ivy | +0.020 |
Holly | +0.019 |
Iris | +0.019 |
Fiona | +0.018 |
Beat | +0.018 |
Neg logit contributions | |
bie | -0.014 |
eps | -0.014 |
favorite | -0.014 |
favorites | -0.013 |
drawn | -0.013 |
Token | Unlike |
Token ID | 12101 |
Feature activation | +2.00e-03 |
Pos logit contributions | |
Ivy | +0.009 |
Iris | +0.009 |
Holly | +0.009 |
Beat | +0.008 |
Fiona | +0.008 |
Neg logit contributions | |
favorite | -0.008 |
o | -0.007 |
bie | -0.007 |
eps | -0.007 |
bo | -0.006 |
Token | other |
Token ID | 584 |
Feature activation | +7.92e-02 |
Pos logit contributions | |
favorite | +0.003 |
o | +0.003 |
bie | +0.003 |
eps | +0.002 |
bo | +0.002 |
Neg logit contributions | |
Ivy | -0.004 |
Iris | -0.004 |
Holly | -0.004 |
Beat | -0.004 |
ner | -0.004 |
Token | ost |
Token ID | 23619 |
Feature activation | +7.26e-02 |
Pos logit contributions | |
o | +0.146 |
bie | +0.138 |
favorite | +0.134 |
vice | +0.132 |
bo | +0.130 |
Neg logit contributions | |
Ivy | -0.116 |
Iris | -0.111 |
Fiona | -0.108 |
Beat | -0.105 |
Dawn | -0.102 |
Token | ric |
Token ID | 1173 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
favorite | +0.124 |
hot | +0.107 |
curious | +0.107 |
bright | +0.107 |
favourite | +0.106 |
Neg logit contributions | |
alk | -0.106 |
ion | -0.098 |
Ivy | -0.096 |
irs | -0.093 |
ner | -0.093 |
Token | hes |
Token ID | 956 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
favorite | +0.274 |
bie | +0.265 |
favorites | +0.242 |
favourite | +0.231 |
o | +0.220 |
Neg logit contributions | |
ner | -0.287 |
Ivy | -0.283 |
ion | -0.275 |
Iris | -0.259 |
otion | -0.257 |
Token | , |
Token ID | 11 |
Feature activation | -7.54e-03 |
Pos logit contributions | |
favorite | +0.069 |
o | +0.068 |
bie | +0.066 |
eps | +0.058 |
bo | +0.055 |
Neg logit contributions | |
Ivy | -0.111 |
Iris | -0.105 |
Holly | -0.102 |
Beat | -0.100 |
ner | -0.099 |
Token | his |
Token ID | 465 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
Ivy | +0.014 |
Iris | +0.013 |
Holly | +0.013 |
Beat | +0.013 |
Fiona | +0.012 |
Neg logit contributions | |
o | -0.012 |
favorite | -0.012 |
bie | -0.012 |
eps | -0.011 |
bo | -0.010 |
Token | feathers |
Token ID | 27400 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
o | +0.024 |
vice | +0.023 |
v | +0.023 |
eps | +0.022 |
ams | +0.022 |
Neg logit contributions | |
Ivy | -0.027 |
Fiona | -0.026 |
Iris | -0.025 |
Holly | -0.024 |
Beat | -0.024 |
Token | gl |
Token ID | 1278 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
favorite | +0.061 |
bie | +0.060 |
o | +0.059 |
eps | +0.053 |
v | +0.050 |
Neg logit contributions | |
Ivy | -0.089 |
Iris | -0.083 |
Holly | -0.081 |
ner | -0.080 |
Beat | -0.080 |
Token | around |
Token ID | 1088 |
Feature activation | +3.33e-03 |
Pos logit contributions | |
Ivy | +0.055 |
Iris | +0.051 |
ner | +0.050 |
Holly | +0.049 |
Buffy | +0.047 |
Neg logit contributions | |
favorite | -0.029 |
o | -0.027 |
bie | -0.024 |
bright | -0.023 |
big | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
favorite | +0.004 |
o | +0.004 |
bie | +0.004 |
eps | +0.004 |
bo | +0.003 |
Neg logit contributions | |
Ivy | -0.007 |
Iris | -0.007 |
Holly | -0.007 |
Beat | -0.007 |
ner | -0.007 |
Token | around |
Token ID | 1088 |
Feature activation | +7.78e-03 |
Pos logit contributions | |
favorite | +0.015 |
o | +0.014 |
bie | +0.014 |
eps | +0.013 |
bo | +0.012 |
Neg logit contributions | |
Ivy | -0.021 |
Iris | -0.019 |
Holly | -0.019 |
Beat | -0.018 |
Fiona | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
o | +0.012 |
favorite | +0.012 |
bie | +0.011 |
eps | +0.010 |
bo | +0.010 |
Neg logit contributions | |
Ivy | -0.016 |
Iris | -0.015 |
Holly | -0.014 |
Beat | -0.014 |
Fiona | -0.014 |
Token | Bob |
Token ID | 5811 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
o | +0.036 |
bie | +0.033 |
favorite | +0.033 |
eps | +0.032 |
v | +0.031 |
Neg logit contributions | |
Ivy | -0.037 |
Iris | -0.035 |
Fiona | -0.034 |
Holly | -0.034 |
Beat | -0.033 |
Token | bie |
Token ID | 12590 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
favorite | +0.022 |
drawn | +0.016 |
tallest | +0.015 |
eps | +0.014 |
aw | +0.014 |
Neg logit contributions | |
Ivy | -0.077 |
Fiona | -0.076 |
Holly | -0.075 |
Wendy | -0.070 |
ion | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
favorite | +0.210 |
o | +0.197 |
eps | +0.196 |
bie | +0.192 |
tallest | +0.189 |
Neg logit contributions | |
Ivy | -0.353 |
Holly | -0.333 |
Iris | -0.330 |
Fiona | -0.324 |
Beat | -0.314 |
Token | They |
Token ID | 1119 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
o | +0.015 |
favorite | +0.015 |
bie | +0.014 |
eps | +0.012 |
bo | +0.012 |
Neg logit contributions | |
Ivy | -0.029 |
Iris | -0.027 |
Holly | -0.026 |
Beat | -0.026 |
ner | -0.026 |
Token | marched |
Token ID | 23558 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
Ivy | +0.051 |
Iris | +0.049 |
Holly | +0.048 |
Beat | +0.047 |
ner | +0.046 |
Neg logit contributions | |
favorite | -0.032 |
o | -0.031 |
bie | -0.030 |
eps | -0.026 |
bo | -0.026 |
Token | together |
Token ID | 1978 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
Ivy | +0.090 |
Iris | +0.084 |
Holly | +0.080 |
ner | +0.080 |
Beat | +0.078 |
Neg logit contributions | |
favorite | -0.050 |
o | -0.048 |
bie | -0.044 |
eps | -0.040 |
bright | -0.039 |
Token | through |
Token ID | 832 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
Ivy | +0.044 |
Iris | +0.042 |
Holly | +0.040 |
Beat | +0.039 |
ner | +0.039 |
Neg logit contributions | |
favorite | -0.025 |
o | -0.025 |
bie | -0.024 |
eps | -0.021 |
bo | -0.020 |
Token | around |
Token ID | 1088 |
Feature activation | -5.32e-03 |
Pos logit contributions | |
Ivy | +0.052 |
Iris | +0.049 |
Holly | +0.048 |
Beat | +0.047 |
ner | +0.047 |
Neg logit contributions | |
favorite | -0.032 |
o | -0.031 |
bie | -0.030 |
eps | -0.027 |
bo | -0.025 |
Token | stopped |
Token ID | 5025 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
Ivy | +0.010 |
Iris | +0.009 |
Holly | +0.009 |
Beat | +0.009 |
ner | +0.009 |
Neg logit contributions | |
favorite | -0.009 |
o | -0.009 |
bie | -0.008 |
eps | -0.008 |
bo | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
favorite | +0.046 |
o | +0.044 |
bie | +0.044 |
vice | +0.040 |
v | +0.039 |
Neg logit contributions | |
Ivy | -0.052 |
Beat | -0.050 |
Iris | -0.050 |
Holly | -0.048 |
ner | -0.047 |
Token | watched |
Token ID | 7342 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
favorite | +0.028 |
o | +0.028 |
bie | +0.027 |
eps | +0.024 |
bo | +0.023 |
Neg logit contributions | |
Ivy | -0.044 |
Iris | -0.042 |
Holly | -0.040 |
Beat | -0.040 |
ner | -0.039 |
Token | in |
Token ID | 287 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
favorite | +0.051 |
o | +0.049 |
bie | +0.047 |
bright | +0.042 |
eps | +0.042 |
Neg logit contributions | |
Ivy | -0.071 |
Iris | -0.067 |
Holly | -0.064 |
Beat | -0.063 |
ner | -0.063 |
Token | amaz |
Token ID | 40642 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
o | +0.028 |
favorite | +0.025 |
bie | +0.025 |
bo | +0.023 |
eps | +0.023 |
Neg logit contributions | |
Ivy | -0.037 |
Iris | -0.036 |
Beat | -0.035 |
Holly | -0.035 |
Fiona | -0.033 |
Token | ement |
Token ID | 972 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
o | +0.290 |
bie | +0.279 |
favorite | +0.274 |
eps | +0.243 |
bo | +0.237 |
Neg logit contributions | |
Ivy | -0.303 |
Iris | -0.287 |
Holly | -0.273 |
Beat | -0.273 |
ner | -0.266 |
Token | . |
Token ID | 13 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
o | +0.154 |
favorite | +0.148 |
bie | +0.146 |
eps | +0.133 |
bo | +0.124 |
Neg logit contributions | |
Ivy | -0.303 |
Iris | -0.288 |
Holly | -0.281 |
Beat | -0.277 |
ner | -0.275 |
Token | They |
Token ID | 1119 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
o | +0.022 |
favorite | +0.022 |
bie | +0.021 |
bo | +0.018 |
eps | +0.018 |
Neg logit contributions | |
Ivy | -0.028 |
Iris | -0.026 |
Holly | -0.025 |
ner | -0.024 |
Beat | -0.024 |
Token | were |
Token ID | 547 |
Feature activation | -5.88e-04 |
Pos logit contributions | |
Ivy | +0.037 |
Beat | +0.036 |
Iris | +0.036 |
Holly | +0.035 |
ner | +0.034 |
Neg logit contributions | |
favorite | -0.030 |
o | -0.029 |
bie | -0.028 |
bo | -0.025 |
int | -0.025 |
Token | so |
Token ID | 523 |
Feature activation | +7.54e-04 |
Pos logit contributions | |
Ivy | +0.001 |
Iris | +0.001 |
Beat | +0.001 |
Holly | +0.001 |
Fiona | +0.001 |
Neg logit contributions | |
o | -0.001 |
favorite | -0.001 |
int | -0.001 |
bie | -0.001 |
bo | -0.001 |
Token | my |
Token ID | 1820 |
Feature activation | +4.44e-02 |
Pos logit contributions | |
favorite | +0.044 |
tallest | +0.034 |
favourite | +0.031 |
favorites | +0.030 |
vice | +0.029 |
Neg logit contributions | |
Ivy | -0.094 |
Holly | -0.090 |
Iris | -0.088 |
Fiona | -0.087 |
ner | -0.087 |
Token | said |
Token ID | 531 |
Feature activation | +3.78e-03 |
Pos logit contributions | |
favorite | +0.057 |
o | +0.055 |
bie | +0.054 |
eps | +0.047 |
bo | +0.045 |
Neg logit contributions | |
Ivy | -0.099 |
Iris | -0.094 |
Holly | -0.090 |
Beat | -0.089 |
ner | -0.088 |
Token | ," |
Token ID | 553 |
Feature activation | +2.98e-03 |
Pos logit contributions | |
favorite | +0.005 |
o | +0.005 |
bie | +0.005 |
bright | +0.004 |
bo | +0.004 |
Neg logit contributions | |
Ivy | -0.008 |
Iris | -0.008 |
Beat | -0.008 |
Holly | -0.007 |
ner | -0.007 |
Token | See |
Token ID | 6214 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
favorite | +0.005 |
o | +0.005 |
bie | +0.005 |
eps | +0.005 |
bo | +0.005 |
Neg logit contributions | |
Ivy | -0.005 |
Iris | -0.005 |
Holly | -0.005 |
Beat | -0.005 |
ner | -0.004 |
Token | Bob |
Token ID | 5811 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
Ivy | +0.033 |
Iris | +0.031 |
Holly | +0.030 |
Beat | +0.029 |
ner | +0.029 |
Neg logit contributions | |
favorite | -0.025 |
o | -0.025 |
bie | -0.024 |
eps | -0.022 |
v | -0.021 |
Token | bie |
Token ID | 12590 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
favorite | +0.022 |
eps | +0.015 |
drawn | +0.015 |
tallest | +0.014 |
favourite | +0.013 |
Neg logit contributions | |
Ivy | -0.092 |
Holly | -0.087 |
Fiona | -0.087 |
ion | -0.085 |
Iris | -0.084 |
Token | , |
Token ID | 11 |
Feature activation | +6.74e-03 |
Pos logit contributions | |
favorite | +0.218 |
o | +0.206 |
bie | +0.203 |
eps | +0.191 |
vice | +0.181 |
Neg logit contributions | |
Ivy | -0.360 |
Iris | -0.332 |
Holly | -0.327 |
ner | -0.320 |
Fiona | -0.311 |
Token | that |
Token ID | 326 |
Feature activation | +8.85e-03 |
Pos logit contributions | |
o | +0.009 |
favorite | +0.009 |
bie | +0.009 |
eps | +0.008 |
bo | +0.008 |
Neg logit contributions | |
Ivy | -0.014 |
Iris | -0.013 |
Holly | -0.013 |
Beat | -0.013 |
ner | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
favorite | +0.013 |
o | +0.013 |
bie | +0.013 |
eps | +0.012 |
v | +0.011 |
Neg logit contributions | |
Ivy | -0.017 |
Iris | -0.016 |
Holly | -0.016 |
ner | -0.015 |
Beat | -0.015 |
Token | easy |
Token ID | 2562 |
Feature activation | +6.16e-03 |
Pos logit contributions | |
o | +0.015 |
bie | +0.014 |
favorite | +0.014 |
v | +0.012 |
eps | +0.012 |
Neg logit contributions | |
Ivy | -0.029 |
Iris | -0.027 |
Holly | -0.027 |
Beat | -0.026 |
Fiona | -0.025 |
Token | ". |
Token ID | 1911 |
Feature activation | +7.96e-03 |
Pos logit contributions | |
favorite | +0.007 |
o | +0.007 |
bie | +0.006 |
eps | +0.005 |
bright | +0.005 |
Neg logit contributions | |
Ivy | -0.015 |
Iris | -0.014 |
Holly | -0.014 |
Beat | -0.014 |
ner | -0.013 |
Token | ! |
Token ID | 0 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
Ivy | +0.055 |
Iris | +0.052 |
Holly | +0.050 |
Beat | +0.049 |
ner | +0.049 |
Neg logit contributions | |
o | -0.030 |
favorite | -0.030 |
bie | -0.029 |
eps | -0.026 |
bo | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +9.19e-03 |
Pos logit contributions | |
favorite | +0.026 |
o | +0.026 |
bie | +0.025 |
eps | +0.021 |
bo | +0.020 |
Neg logit contributions | |
Ivy | -0.059 |
Iris | -0.056 |
Holly | -0.054 |
Beat | -0.053 |
ner | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
favorite | +0.012 |
o | +0.011 |
bie | +0.011 |
eps | +0.009 |
bo | +0.009 |
Neg logit contributions | |
Ivy | -0.021 |
Iris | -0.020 |
Holly | -0.019 |
Beat | -0.019 |
Fiona | -0.018 |
Token | As |
Token ID | 1722 |
Feature activation | -2.11e-03 |
Pos logit contributions | |
o | +0.020 |
favorite | +0.020 |
bie | +0.019 |
eps | +0.017 |
bo | +0.017 |
Neg logit contributions | |
Ivy | -0.027 |
Iris | -0.025 |
Holly | -0.024 |
Beat | -0.024 |
ner | -0.024 |
Token | Bob |
Token ID | 5811 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
Ivy | +0.004 |
Iris | +0.004 |
Holly | +0.004 |
Beat | +0.004 |
ner | +0.004 |
Neg logit contributions | |
o | -0.003 |
favorite | -0.003 |
bie | -0.003 |
eps | -0.003 |
bo | -0.003 |
Token | bie |
Token ID | 12590 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
favorite | +0.029 |
tallest | +0.019 |
drawn | +0.018 |
favourite | +0.018 |
eps | +0.017 |
Neg logit contributions | |
Ivy | -0.079 |
Fiona | -0.077 |
Holly | -0.076 |
Iris | -0.073 |
ner | -0.072 |
Token | walked |
Token ID | 6807 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
favorite | +0.229 |
o | +0.208 |
bie | +0.205 |
tallest | +0.189 |
eps | +0.183 |
Neg logit contributions | |
Ivy | -0.351 |
Iris | -0.329 |
Holly | -0.323 |
Beat | -0.315 |
ner | -0.315 |
Token | away |
Token ID | 1497 |
Feature activation | -7.05e-02 |
Pos logit contributions | |
favorite | +0.023 |
o | +0.023 |
bie | +0.020 |
bright | +0.019 |
at | +0.018 |
Neg logit contributions | |
Ivy | -0.048 |
Iris | -0.045 |
ner | -0.045 |
Holly | -0.043 |
ocated | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -6.95e-04 |
Pos logit contributions | |
Ivy | +0.158 |
Iris | +0.149 |
Holly | +0.142 |
ner | +0.142 |
Beat | +0.140 |
Neg logit contributions | |
favorite | -0.087 |
o | -0.084 |
bie | -0.080 |
bright | -0.074 |
big | -0.068 |
Token | he |
Token ID | 339 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
Ivy | +0.001 |
Iris | +0.001 |
Holly | +0.001 |
Beat | +0.001 |
Fiona | +0.001 |
Neg logit contributions | |
o | -0.001 |
favorite | -0.001 |
bie | -0.001 |
eps | -0.001 |
v | -0.001 |
Token | looked |
Token ID | 3114 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
favorite | +0.042 |
o | +0.041 |
bie | +0.039 |
v | +0.035 |
eps | +0.035 |
Neg logit contributions | |
Ivy | -0.071 |
Iris | -0.067 |
Holly | -0.067 |
Beat | -0.066 |
Fiona | -0.064 |
Token | it |
Token ID | 340 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
o | +0.023 |
bie | +0.022 |
favorite | +0.022 |
bo | +0.020 |
eps | +0.020 |
Neg logit contributions | |
Ivy | -0.031 |
Iris | -0.029 |
Holly | -0.028 |
Beat | -0.028 |
Fiona | -0.027 |
Token | often |
Token ID | 1690 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
Ivy | +0.028 |
Iris | +0.027 |
Holly | +0.026 |
Fiona | +0.025 |
Beat | +0.025 |
Neg logit contributions | |
o | -0.017 |
bie | -0.017 |
favorite | -0.016 |
eps | -0.016 |
bo | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
o | +0.075 |
bie | +0.071 |
favorite | +0.070 |
bo | +0.066 |
eps | +0.065 |
Neg logit contributions | |
Ivy | -0.095 |
Iris | -0.088 |
Holly | -0.087 |
Beat | -0.085 |
ner | -0.083 |
Token | watched |
Token ID | 7342 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
favorite | +0.014 |
o | +0.014 |
bie | +0.013 |
eps | +0.012 |
bo | +0.012 |
Neg logit contributions | |
Ivy | -0.024 |
Iris | -0.023 |
Holly | -0.022 |
Beat | -0.022 |
ner | -0.022 |
Token | in |
Token ID | 287 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
o | +0.052 |
favorite | +0.051 |
bie | +0.050 |
eps | +0.045 |
bo | +0.044 |
Neg logit contributions | |
Ivy | -0.068 |
Iris | -0.064 |
Holly | -0.062 |
Beat | -0.061 |
ner | -0.060 |
Token | amaz |
Token ID | 40642 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
o | +0.035 |
bie | +0.031 |
bo | +0.031 |
favorite | +0.030 |
int | +0.030 |
Neg logit contributions | |
Ivy | -0.038 |
Iris | -0.037 |
Beat | -0.036 |
Holly | -0.035 |
Fiona | -0.035 |
Token | ement |
Token ID | 972 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
o | +0.302 |
bie | +0.292 |
favorite | +0.279 |
eps | +0.250 |
bo | +0.248 |
Neg logit contributions | |
Ivy | -0.285 |
Iris | -0.270 |
Beat | -0.256 |
Holly | -0.256 |
ner | -0.248 |
Token | whenever |
Token ID | 8797 |
Feature activation | +9.01e-02 |
Pos logit contributions | |
o | +0.168 |
bie | +0.162 |
favorite | +0.159 |
eps | +0.158 |
bo | +0.146 |
Neg logit contributions | |
Ivy | -0.271 |
Iris | -0.255 |
Holly | -0.252 |
ner | -0.248 |
Beat | -0.244 |
Token | the |
Token ID | 262 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
favorite | +0.121 |
o | +0.119 |
bie | +0.115 |
eps | +0.099 |
bright | +0.096 |
Neg logit contributions | |
Ivy | -0.194 |
Iris | -0.184 |
Holly | -0.177 |
Beat | -0.174 |
ner | -0.173 |
Token | door |
Token ID | 3420 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
o | +0.092 |
bie | +0.082 |
v | +0.079 |
favorite | +0.078 |
ams | +0.076 |
Neg logit contributions | |
Ivy | -0.099 |
Iris | -0.091 |
Holly | -0.086 |
ner | -0.082 |
Fiona | -0.082 |
Token | opened |
Token ID | 4721 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
o | +0.047 |
favorite | +0.046 |
bie | +0.046 |
eps | +0.039 |
bo | +0.038 |
Neg logit contributions | |
Ivy | -0.064 |
Iris | -0.061 |
ner | -0.059 |
Holly | -0.058 |
Beat | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
favorite | +0.017 |
o | +0.016 |
bie | +0.015 |
eps | +0.014 |
v | +0.013 |
Neg logit contributions | |
Ivy | -0.030 |
Iris | -0.028 |
Holly | -0.027 |
Beat | -0.027 |
Fiona | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
favorite | +0.022 |
bie | +0.020 |
o | +0.020 |
eps | +0.018 |
tallest | +0.017 |
Neg logit contributions | |
Ivy | -0.038 |
Iris | -0.036 |
Holly | -0.035 |
Beat | -0.035 |
Fiona | -0.034 |
Token | Finally |
Token ID | 11158 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
o | +0.032 |
favorite | +0.032 |
bie | +0.031 |
eps | +0.028 |
bo | +0.027 |
Neg logit contributions | |
Ivy | -0.039 |
Iris | -0.037 |
Holly | -0.035 |
Beat | -0.035 |
ner | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
favorite | +0.019 |
o | +0.018 |
bie | +0.018 |
bright | +0.016 |
at | +0.015 |
Neg logit contributions | |
Ivy | -0.036 |
Iris | -0.034 |
ner | -0.033 |
Holly | -0.033 |
Beat | -0.033 |
Token | Bob |
Token ID | 5811 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
o | +0.039 |
favorite | +0.039 |
bie | +0.038 |
eps | +0.036 |
v | +0.036 |
Neg logit contributions | |
Ivy | -0.037 |
Iris | -0.034 |
Holly | -0.032 |
Fiona | -0.031 |
Beat | -0.031 |
Token | bie |
Token ID | 12590 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
favorite | +0.030 |
drawn | +0.024 |
tallest | +0.023 |
favourite | +0.022 |
apr | +0.020 |
Neg logit contributions | |
Fiona | -0.060 |
Ivy | -0.058 |
Holly | -0.056 |
Hazel | -0.054 |
iona | -0.053 |
Token | decided |
Token ID | 3066 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
favorite | +0.217 |
o | +0.204 |
bie | +0.201 |
eps | +0.179 |
tallest | +0.174 |
Neg logit contributions | |
Ivy | -0.354 |
Iris | -0.333 |
Holly | -0.328 |
Beat | -0.324 |
ner | -0.319 |
Token | it |
Token ID | 340 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
favorite | +0.022 |
o | +0.020 |
bie | +0.018 |
bright | +0.016 |
drawn | +0.015 |
Neg logit contributions | |
Ivy | -0.067 |
Iris | -0.064 |
Holly | -0.061 |
ner | -0.061 |
Beat | -0.060 |
Token | was |
Token ID | 373 |
Feature activation | +3.69e-03 |
Pos logit contributions | |
Ivy | +0.068 |
Iris | +0.064 |
Holly | +0.061 |
ner | +0.061 |
Beat | +0.060 |
Neg logit contributions | |
o | -0.036 |
favorite | -0.035 |
bie | -0.033 |
eps | -0.029 |
bo | -0.028 |
Token | time |
Token ID | 640 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
o | +0.004 |
favorite | +0.004 |
bie | +0.004 |
eps | +0.003 |
v | +0.003 |
Neg logit contributions | |
Ivy | -0.009 |
Iris | -0.009 |
Holly | -0.008 |
Beat | -0.008 |
Fiona | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -3.59e-03 |
Pos logit contributions | |
o | +0.028 |
favorite | +0.028 |
bie | +0.027 |
bo | +0.025 |
eps | +0.024 |
Neg logit contributions | |
Ivy | -0.047 |
Iris | -0.044 |
Holly | -0.044 |
ner | -0.043 |
Beat | -0.043 |
Token | at |
Token ID | 265 |
Feature activation | +8.95e-02 |
Pos logit contributions | |
favorite | +0.088 |
bright | +0.072 |
black | +0.071 |
curious | +0.070 |
favourite | +0.069 |
Neg logit contributions | |
ion | -0.081 |
ner | -0.079 |
alk | -0.075 |
Ivy | -0.072 |
arises | -0.072 |
Token | field |
Token ID | 2214 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
favorite | +0.110 |
bie | +0.109 |
o | +0.108 |
eps | +0.095 |
bo | +0.087 |
Neg logit contributions | |
Ivy | -0.198 |
Holly | -0.185 |
Iris | -0.184 |
ner | -0.182 |
Beat | -0.179 |
Token | . |
Token ID | 13 |
Feature activation | +3.30e-03 |
Pos logit contributions | |
favorite | +0.029 |
o | +0.029 |
bie | +0.028 |
eps | +0.024 |
bo | +0.023 |
Neg logit contributions | |
Ivy | -0.064 |
Iris | -0.060 |
Holly | -0.059 |
Beat | -0.058 |
ner | -0.057 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
favorite | +0.005 |
o | +0.005 |
bie | +0.004 |
eps | +0.004 |
bo | +0.004 |
Neg logit contributions | |
Ivy | -0.007 |
Iris | -0.007 |
Holly | -0.006 |
Beat | -0.006 |
ner | -0.006 |
Token | Bob |
Token ID | 5811 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
favorite | +0.022 |
o | +0.021 |
bie | +0.021 |
bright | +0.018 |
eps | +0.017 |
Neg logit contributions | |
Ivy | -0.041 |
Iris | -0.039 |
Holly | -0.038 |
ner | -0.037 |
Beat | -0.037 |
Token | bie |
Token ID | 12590 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
favorite | +0.029 |
drawn | +0.021 |
tallest | +0.021 |
favourite | +0.020 |
creative | +0.020 |
Neg logit contributions | |
Fiona | -0.057 |
Ivy | -0.055 |
Holly | -0.054 |
Hazel | -0.051 |
bud | -0.051 |
Token | heard |
Token ID | 2982 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
favorite | +0.228 |
bie | +0.216 |
o | +0.214 |
tallest | +0.195 |
ams | +0.186 |
Neg logit contributions | |
Ivy | -0.317 |
Holly | -0.307 |
Beat | -0.304 |
ner | -0.299 |
Iris | -0.299 |
Token | a |
Token ID | 257 |
Feature activation | -3.24e-03 |
Pos logit contributions | |
o | +0.066 |
favorite | +0.064 |
bie | +0.064 |
eps | +0.055 |
bo | +0.052 |
Neg logit contributions | |
Ivy | -0.120 |
Iris | -0.114 |
Beat | -0.111 |
Holly | -0.110 |
Fiona | -0.109 |
Token | sound |
Token ID | 2128 |
Feature activation | -3.97e-03 |
Pos logit contributions | |
Ivy | +0.007 |
Iris | +0.007 |
Holly | +0.007 |
Beat | +0.007 |
ner | +0.006 |
Neg logit contributions | |
o | -0.004 |
favorite | -0.004 |
bie | -0.004 |
eps | -0.003 |
bo | -0.003 |
Token | ! |
Token ID | 0 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
Ivy | +0.008 |
Iris | +0.008 |
Beat | +0.007 |
ner | +0.007 |
Holly | +0.007 |
Neg logit contributions | |
bie | -0.006 |
favorite | -0.006 |
o | -0.006 |
eps | -0.005 |
favorites | -0.005 |
Token | It |
Token ID | 632 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
o | +0.042 |
favorite | +0.039 |
bie | +0.039 |
bo | +0.032 |
bright | +0.031 |
Neg logit contributions | |
Ivy | -0.080 |
Iris | -0.077 |
Holly | -0.074 |
Beat | -0.073 |
ner | -0.073 |
Token |
|
Token ID | 198 |
Feature activation | +6.55e-03 |
Pos logit contributions | |
Ivy | +0.009 |
Iris | +0.009 |
Holly | +0.008 |
Beat | +0.008 |
ner | +0.008 |
Neg logit contributions | |
o | -0.006 |
favorite | -0.006 |
bie | -0.005 |
bo | -0.004 |
eps | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
o | +0.005 |
bie | +0.005 |
favorite | +0.005 |
eps | +0.004 |
bo | +0.004 |
Neg logit contributions | |
Ivy | -0.018 |
Iris | -0.017 |
Holly | -0.016 |
ner | -0.016 |
Beat | -0.016 |
Token | But |
Token ID | 1537 |
Feature activation | +4.59e-03 |
Pos logit contributions | |
o | +0.019 |
bie | +0.018 |
favorite | +0.018 |
eps | +0.016 |
v | +0.016 |
Neg logit contributions | |
Ivy | -0.025 |
Iris | -0.023 |
Holly | -0.023 |
Beat | -0.022 |
ner | -0.022 |
Token | when |
Token ID | 618 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
favorite | +0.006 |
o | +0.006 |
bie | +0.006 |
eps | +0.005 |
bo | +0.005 |
Neg logit contributions | |
Ivy | -0.010 |
Iris | -0.009 |
Holly | -0.009 |
Beat | -0.009 |
ner | -0.009 |
Token | Ab |
Token ID | 2275 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
favorite | +0.101 |
o | +0.099 |
bie | +0.099 |
bright | +0.089 |
hot | +0.085 |
Neg logit contributions | |
Ivy | -0.096 |
Iris | -0.090 |
Holly | -0.086 |
ner | -0.086 |
Beat | -0.085 |
Token | bi |
Token ID | 8482 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
favorite | +0.063 |
eps | +0.055 |
o | +0.054 |
bright | +0.053 |
v | +0.053 |
Neg logit contributions | |
Ivy | -0.094 |
Iris | -0.085 |
Holly | -0.083 |
Fiona | -0.082 |
Dawn | -0.080 |
Token | turned |
Token ID | 2900 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
favorite | +0.220 |
o | +0.208 |
bie | +0.199 |
eps | +0.185 |
tallest | +0.179 |
Neg logit contributions | |
Ivy | -0.352 |
Iris | -0.328 |
Holly | -0.322 |
ner | -0.310 |
Beat | -0.309 |
Token | to |
Token ID | 284 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
favorite | +0.064 |
o | +0.063 |
bie | +0.060 |
bright | +0.052 |
eps | +0.052 |
Neg logit contributions | |
Ivy | -0.121 |
Iris | -0.115 |
Beat | -0.109 |
Holly | -0.109 |
ner | -0.109 |
Token | head |
Token ID | 1182 |
Feature activation | +6.52e-02 |
Pos logit contributions | |
Ivy | +0.021 |
Iris | +0.020 |
ner | +0.019 |
Holly | +0.019 |
Beat | +0.019 |
Neg logit contributions | |
favorite | -0.014 |
o | -0.014 |
bie | -0.014 |
bright | -0.012 |
eps | -0.012 |
Token | back |
Token ID | 736 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
favorite | +0.102 |
o | +0.101 |
bie | +0.099 |
bright | +0.087 |
bo | +0.086 |
Neg logit contributions | |
Ivy | -0.124 |
Iris | -0.117 |
ner | -0.111 |
Beat | -0.110 |
Holly | -0.110 |
Token | inside |
Token ID | 2641 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
Ivy | +0.033 |
Iris | +0.031 |
Holly | +0.030 |
ner | +0.030 |
Beat | +0.030 |
Neg logit contributions | |
o | -0.021 |
favorite | -0.021 |
bie | -0.020 |
eps | -0.017 |
bright | -0.017 |
Token | But |
Token ID | 1537 |
Feature activation | +6.07e-03 |
Pos logit contributions | |
favorite | +0.020 |
o | +0.020 |
bie | +0.019 |
eps | +0.017 |
bo | +0.016 |
Neg logit contributions | |
Ivy | -0.030 |
Iris | -0.028 |
Holly | -0.027 |
Beat | -0.027 |
ner | -0.026 |
Token | Tom |
Token ID | 4186 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
o | +0.008 |
favorite | +0.008 |
bie | +0.007 |
eps | +0.006 |
bright | +0.006 |
Neg logit contributions | |
Ivy | -0.013 |
Iris | -0.013 |
Holly | -0.012 |
ner | -0.012 |
Beat | -0.012 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
favorite | +0.066 |
o | +0.063 |
bie | +0.063 |
eps | +0.056 |
v | +0.054 |
Neg logit contributions | |
Ivy | -0.109 |
Iris | -0.102 |
Holly | -0.100 |
Beat | -0.098 |
ner | -0.096 |
Token | â |
Token ID | 22940 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
o | +0.032 |
bo | +0.030 |
favorite | +0.027 |
big | +0.027 |
hot | +0.025 |
Neg logit contributions | |
Ivy | -0.035 |
Iris | -0.033 |
Holly | -0.031 |
Beat | -0.031 |
ocated | -0.029 |
Token | € |
Token ID | 26391 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
favorite | +0.027 |
o | +0.027 |
bie | +0.026 |
eps | +0.023 |
v | +0.022 |
Neg logit contributions | |
Ivy | -0.028 |
Iris | -0.026 |
Holly | -0.024 |
Beat | -0.024 |
ner | -0.024 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
bie | +0.077 |
favorite | +0.077 |
vice | +0.075 |
eps | +0.073 |
tallest | +0.070 |
Neg logit contributions | |
Beat | -0.079 |
Ivy | -0.079 |
ner | -0.078 |
Iris | -0.075 |
Holly | -0.074 |
Token | t |
Token ID | 83 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
favorite | +0.286 |
bie | +0.270 |
o | +0.267 |
eps | +0.252 |
bright | +0.242 |
Neg logit contributions | |
Ivy | -0.286 |
Iris | -0.261 |
Holly | -0.256 |
ner | -0.254 |
Beat | -0.248 |
Token | help |
Token ID | 1037 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
favorite | +0.065 |
o | +0.065 |
bie | +0.062 |
bo | +0.055 |
eps | +0.055 |
Neg logit contributions | |
Ivy | -0.092 |
Iris | -0.087 |
Holly | -0.084 |
Beat | -0.082 |
ner | -0.081 |
Token | but |
Token ID | 475 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
Ivy | +0.029 |
Iris | +0.028 |
Holly | +0.026 |
Beat | +0.026 |
ner | +0.026 |
Neg logit contributions | |
o | -0.017 |
favorite | -0.017 |
bie | -0.016 |
eps | -0.014 |
bo | -0.014 |
Token | worry |
Token ID | 5490 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
favorite | +0.067 |
o | +0.066 |
bie | +0.064 |
eps | +0.054 |
bo | +0.054 |
Neg logit contributions | |
Ivy | -0.105 |
Iris | -0.099 |
Holly | -0.095 |
Beat | -0.094 |
ner | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
Ivy | +0.035 |
Iris | +0.033 |
ner | +0.031 |
Holly | +0.031 |
Beat | +0.031 |
Neg logit contributions | |
favorite | -0.021 |
o | -0.021 |
bie | -0.020 |
bright | -0.017 |
bo | -0.017 |
Token | They |
Token ID | 1119 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
favorite | +0.033 |
bie | +0.032 |
o | +0.032 |
eps | +0.030 |
bo | +0.029 |
Neg logit contributions | |
Ivy | -0.024 |
Iris | -0.022 |
Beat | -0.021 |
Holly | -0.020 |
ner | -0.020 |
Token | do |
Token ID | 466 |
Feature activation | -1.65e-03 |
Pos logit contributions | |
Ivy | +0.023 |
Iris | +0.022 |
Holly | +0.021 |
ner | +0.021 |
Beat | +0.021 |
Neg logit contributions | |
favorite | -0.015 |
o | -0.015 |
bie | -0.014 |
eps | -0.013 |
bo | -0.012 |
Token | not |
Token ID | 407 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
Ivy | +0.004 |
Iris | +0.003 |
Holly | +0.003 |
Beat | +0.003 |
ner | +0.003 |
Neg logit contributions | |
favorite | -0.002 |
o | -0.002 |
bie | -0.002 |
eps | -0.002 |
bo | -0.002 |
Token | pack |
Token ID | 2353 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
o | +0.051 |
favorite | +0.051 |
bie | +0.049 |
eps | +0.044 |
bo | +0.042 |
Neg logit contributions | |
Ivy | -0.074 |
Iris | -0.069 |
Holly | -0.067 |
Beat | -0.066 |
ner | -0.065 |
Token | ost |
Token ID | 23619 |
Feature activation | +6.51e-02 |
Pos logit contributions | |
o | +0.054 |
bie | +0.052 |
favorite | +0.052 |
eps | +0.046 |
v | +0.046 |
Neg logit contributions | |
Ivy | -0.065 |
Iris | -0.061 |
Holly | -0.059 |
Beat | -0.057 |
Fiona | -0.055 |
Token | ric |
Token ID | 1173 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
huge | +0.106 |
favorite | +0.105 |
big | +0.102 |
curious | +0.101 |
bright | +0.101 |
Neg logit contributions | |
ion | -0.089 |
irs | -0.080 |
alk | -0.079 |
unn | -0.075 |
ner | -0.074 |
Token | hes |
Token ID | 956 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
bie | +0.259 |
favorite | +0.252 |
favorites | +0.222 |
o | +0.217 |
favourite | +0.217 |
Neg logit contributions | |
ion | -0.279 |
Ivy | -0.275 |
ner | -0.272 |
Iris | -0.252 |
otion | -0.246 |
Token | anymore |
Token ID | 7471 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
favorite | +0.068 |
o | +0.067 |
bie | +0.067 |
eps | +0.059 |
vice | +0.058 |
Neg logit contributions | |
Ivy | -0.092 |
Iris | -0.085 |
Holly | -0.083 |
Beat | -0.081 |
Fiona | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
Ivy | +0.090 |
Iris | +0.085 |
Holly | +0.083 |
Beat | +0.081 |
ner | +0.081 |
Neg logit contributions | |
favorite | -0.032 |
o | -0.031 |
bie | -0.030 |
eps | -0.025 |
bo | -0.023 |
Token | They |
Token ID | 1119 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
bie | +0.031 |
favorite | +0.031 |
o | +0.030 |
eps | +0.029 |
vice | +0.028 |
Neg logit contributions | |
Ivy | -0.020 |
Iris | -0.019 |
Beat | -0.018 |
Holly | -0.017 |
ner | -0.016 |
Token | pack |
Token ID | 2353 |
Feature activation | -2.76e-03 |
Pos logit contributions | |
Ivy | +0.022 |
Iris | +0.020 |
Holly | +0.019 |
Beat | +0.019 |
ner | +0.019 |
Neg logit contributions | |
o | -0.016 |
favorite | -0.016 |
bie | -0.016 |
eps | -0.014 |
bo | -0.013 |
Token | until |
Token ID | 1566 |
Feature activation | +7.24e-02 |
Pos logit contributions | |
Ivy | +0.037 |
Iris | +0.035 |
ner | +0.034 |
Holly | +0.033 |
Beat | +0.033 |
Neg logit contributions | |
favorite | -0.018 |
o | -0.017 |
bie | -0.015 |
bright | -0.014 |
very | -0.014 |
Token | he |
Token ID | 339 |
Feature activation | +7.87e-02 |
Pos logit contributions | |
favorite | +0.091 |
o | +0.089 |
bie | +0.083 |
bo | +0.073 |
bright | +0.072 |
Neg logit contributions | |
Ivy | -0.160 |
Iris | -0.153 |
Holly | -0.146 |
Beat | -0.145 |
ner | -0.145 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
favorite | +0.109 |
bie | +0.104 |
o | +0.101 |
eps | +0.096 |
ams | +0.093 |
Neg logit contributions | |
Ivy | -0.167 |
Holly | -0.156 |
Iris | -0.153 |
Beat | -0.147 |
Fiona | -0.146 |
Token | â |
Token ID | 22940 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
o | +0.028 |
bo | +0.026 |
favorite | +0.025 |
big | +0.024 |
very | +0.023 |
Neg logit contributions | |
Ivy | -0.032 |
Iris | -0.031 |
Beat | -0.029 |
Holly | -0.029 |
Fiona | -0.027 |
Token | € |
Token ID | 26391 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
favorite | +0.024 |
o | +0.024 |
bie | +0.023 |
eps | +0.021 |
bo | +0.020 |
Neg logit contributions | |
Ivy | -0.036 |
Iris | -0.034 |
Holly | -0.032 |
Beat | -0.032 |
ner | -0.031 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
bie | +0.100 |
favorite | +0.095 |
eps | +0.093 |
vice | +0.092 |
o | +0.091 |
Neg logit contributions | |
Ivy | -0.083 |
Iris | -0.077 |
Beat | -0.077 |
ner | -0.076 |
Holly | -0.073 |
Token | t |
Token ID | 83 |
Feature activation | +4.00e-02 |
Pos logit contributions | |
favorite | +0.299 |
bie | +0.278 |
o | +0.271 |
eps | +0.266 |
tallest | +0.257 |
Neg logit contributions | |
Ivy | -0.269 |
Iris | -0.243 |
Holly | -0.241 |
ner | -0.239 |
Beat | -0.228 |
Token | run |
Token ID | 1057 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
favorite | +0.055 |
o | +0.054 |
bie | +0.051 |
bo | +0.046 |
bright | +0.045 |
Neg logit contributions | |
Ivy | -0.084 |
Iris | -0.080 |
Holly | -0.076 |
Beat | -0.076 |
ner | -0.075 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.55e-03 |
Pos logit contributions | |
favorite | +0.056 |
o | +0.055 |
very | +0.046 |
bright | +0.045 |
big | +0.044 |
Neg logit contributions | |
Ivy | -0.124 |
Iris | -0.118 |
ner | -0.116 |
Holly | -0.114 |
Beat | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | +6.25e-03 |
Pos logit contributions | |
Ivy | +0.018 |
Iris | +0.017 |
Holly | +0.016 |
Beat | +0.016 |
ner | +0.016 |
Neg logit contributions | |
favorite | -0.005 |
o | -0.005 |
bie | -0.005 |
eps | -0.004 |
bo | -0.004 |
Token | Then |
Token ID | 3244 |
Feature activation | +3.01e-03 |
Pos logit contributions | |
favorite | +0.006 |
o | +0.005 |
bie | +0.004 |
bo | +0.004 |
favourite | +0.004 |
Neg logit contributions | |
Ivy | -0.016 |
Iris | -0.015 |
ner | -0.015 |
Beat | -0.015 |
Holly | -0.015 |
Token | calmed |
Token ID | 49566 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
Ivy | +0.011 |
Iris | +0.010 |
Holly | +0.010 |
Beat | +0.010 |
ner | +0.010 |
Neg logit contributions | |
favorite | -0.006 |
o | -0.006 |
bie | -0.006 |
eps | -0.005 |
bo | -0.005 |
Token | down |
Token ID | 866 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
Ivy | +0.076 |
Iris | +0.072 |
Holly | +0.069 |
Beat | +0.068 |
Fiona | +0.067 |
Neg logit contributions | |
favorite | -0.054 |
o | -0.053 |
bie | -0.051 |
eps | -0.047 |
v | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +4.40e-03 |
Pos logit contributions | |
favorite | +0.022 |
o | +0.022 |
bie | +0.021 |
eps | +0.018 |
bo | +0.017 |
Neg logit contributions | |
Ivy | -0.040 |
Iris | -0.038 |
Holly | -0.037 |
Beat | -0.036 |
ner | -0.036 |
Token | watched |
Token ID | 7342 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
o | +0.006 |
favorite | +0.006 |
bie | +0.006 |
eps | +0.005 |
bo | +0.005 |
Neg logit contributions | |
Ivy | -0.010 |
Iris | -0.009 |
Holly | -0.009 |
Beat | -0.009 |
ner | -0.009 |
Token | in |
Token ID | 287 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
favorite | +0.040 |
o | +0.038 |
bie | +0.037 |
bright | +0.033 |
eps | +0.032 |
Neg logit contributions | |
Ivy | -0.063 |
Iris | -0.060 |
Holly | -0.057 |
Beat | -0.057 |
ner | -0.056 |
Token | amaz |
Token ID | 40642 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
o | +0.023 |
favorite | +0.021 |
bie | +0.021 |
bo | +0.019 |
eps | +0.019 |
Neg logit contributions | |
Ivy | -0.031 |
Iris | -0.029 |
Holly | -0.028 |
Beat | -0.028 |
Fiona | -0.027 |
Token | ement |
Token ID | 972 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
o | +0.290 |
bie | +0.279 |
favorite | +0.272 |
eps | +0.241 |
bo | +0.238 |
Neg logit contributions | |
Ivy | -0.281 |
Iris | -0.266 |
Holly | -0.251 |
Beat | -0.251 |
ner | -0.247 |
Token | as |
Token ID | 355 |
Feature activation | +7.66e-02 |
Pos logit contributions | |
o | +0.174 |
favorite | +0.169 |
bie | +0.165 |
eps | +0.152 |
bo | +0.146 |
Neg logit contributions | |
Ivy | -0.283 |
Iris | -0.266 |
Holly | -0.260 |
ner | -0.257 |
Beat | -0.255 |
Token | the |
Token ID | 262 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
favorite | +0.102 |
o | +0.099 |
bie | +0.097 |
bright | +0.085 |
eps | +0.084 |
Neg logit contributions | |
Ivy | -0.166 |
Iris | -0.156 |
Holly | -0.151 |
Beat | -0.147 |
ner | -0.146 |
Token | rainbow |
Token ID | 27223 |
Feature activation | +4.04e-02 |
Pos logit contributions | |
o | +0.062 |
bie | +0.057 |
favorite | +0.054 |
bo | +0.052 |
eps | +0.052 |
Neg logit contributions | |
Ivy | -0.077 |
Iris | -0.073 |
ner | -0.069 |
Holly | -0.068 |
Beat | -0.066 |
Token | melted |
Token ID | 24178 |
Feature activation | -3.56e-04 |
Pos logit contributions | |
favorite | +0.056 |
o | +0.056 |
bie | +0.054 |
eps | +0.048 |
bo | +0.046 |
Neg logit contributions | |
Ivy | -0.085 |
Iris | -0.080 |
Holly | -0.077 |
Beat | -0.076 |
ner | -0.075 |
Token | before |
Token ID | 878 |
Feature activation | +2.36e-03 |
Pos logit contributions | |
bie | +0.037 |
o | +0.037 |
favorite | +0.035 |
eps | +0.033 |
vice | +0.031 |
Neg logit contributions | |
Ivy | -0.053 |
Iris | -0.049 |
Holly | -0.048 |
ner | -0.048 |
Beat | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
favorite | +0.002 |
o | +0.002 |
bie | +0.002 |
bright | +0.001 |
at | +0.001 |
Neg logit contributions | |
Ivy | -0.006 |
Iris | -0.006 |
Beat | -0.006 |
Holly | -0.006 |
ner | -0.006 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
favorite | +0.033 |
o | +0.033 |
bie | +0.032 |
eps | +0.028 |
bo | +0.028 |
Neg logit contributions | |
Ivy | -0.041 |
Iris | -0.039 |
Holly | -0.037 |
Beat | -0.036 |
ner | -0.036 |
Token | â |
Token ID | 22940 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
o | +0.028 |
bo | +0.027 |
favorite | +0.026 |
big | +0.024 |
bie | +0.024 |
Neg logit contributions | |
Ivy | -0.027 |
Iris | -0.025 |
Holly | -0.024 |
Beat | -0.024 |
ocated | -0.023 |
Token | € |
Token ID | 26391 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
favorite | +0.031 |
o | +0.031 |
bie | +0.030 |
eps | +0.027 |
bo | +0.026 |
Neg logit contributions | |
Ivy | -0.036 |
Iris | -0.034 |
Holly | -0.033 |
Beat | -0.032 |
ner | -0.032 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
favorite | +0.078 |
vice | +0.077 |
bie | +0.075 |
o | +0.073 |
tallest | +0.071 |
Neg logit contributions | |
Ivy | -0.088 |
Beat | -0.088 |
ner | -0.085 |
Iris | -0.085 |
Holly | -0.080 |
Token | t |
Token ID | 83 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
favorite | +0.293 |
bie | +0.270 |
o | +0.267 |
eps | +0.260 |
vice | +0.249 |
Neg logit contributions | |
Ivy | -0.276 |
Iris | -0.248 |
Holly | -0.246 |
ner | -0.246 |
Beat | -0.232 |
Token | wait |
Token ID | 4043 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
o | +0.059 |
favorite | +0.059 |
bie | +0.057 |
eps | +0.051 |
bo | +0.049 |
Neg logit contributions | |
Ivy | -0.082 |
Iris | -0.077 |
Holly | -0.074 |
Beat | -0.073 |
ner | -0.072 |
Token | to |
Token ID | 284 |
Feature activation | -6.50e-04 |
Pos logit contributions | |
o | +0.023 |
favorite | +0.021 |
bie | +0.021 |
eps | +0.019 |
bo | +0.019 |
Neg logit contributions | |
Ivy | -0.037 |
Iris | -0.034 |
Holly | -0.034 |
Fiona | -0.033 |
Beat | -0.033 |
Token | show |
Token ID | 905 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
Ivy | +0.001 |
Iris | +0.001 |
Holly | +0.001 |
ner | +0.001 |
Beat | +0.001 |
Neg logit contributions | |
favorite | -0.001 |
o | -0.001 |
bie | -0.001 |
eps | -0.001 |
bright | -0.001 |
Token | it |
Token ID | 340 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
favorite | +0.043 |
o | +0.041 |
bie | +0.041 |
eps | +0.037 |
bright | +0.037 |
Neg logit contributions | |
Ivy | -0.046 |
Iris | -0.042 |
ner | -0.041 |
Holly | -0.041 |
Beat | -0.040 |
Token | her |
Token ID | 607 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
Ivy | +0.026 |
Iris | +0.025 |
Holly | +0.024 |
Beat | +0.024 |
ner | +0.023 |
Neg logit contributions | |
favorite | -0.016 |
o | -0.016 |
bie | -0.015 |
eps | -0.014 |
bo | -0.013 |
Token | doll |
Token ID | 3654 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
o | +0.038 |
bie | +0.034 |
eps | +0.032 |
ams | +0.032 |
v | +0.032 |
Neg logit contributions | |
Ivy | -0.051 |
Iris | -0.049 |
Holly | -0.047 |
Beat | -0.047 |
Fiona | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +5.02e-03 |
Pos logit contributions | |
o | +0.037 |
favorite | +0.037 |
bie | +0.036 |
eps | +0.032 |
bo | +0.031 |
Neg logit contributions | |
Ivy | -0.058 |
Iris | -0.055 |
Holly | -0.053 |
Beat | -0.053 |
ner | -0.052 |
Token | watched |
Token ID | 7342 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
favorite | +0.006 |
o | +0.006 |
bie | +0.005 |
bright | +0.005 |
eps | +0.005 |
Neg logit contributions | |
Ivy | -0.012 |
Iris | -0.011 |
Holly | -0.010 |
ner | -0.010 |
ion | -0.010 |
Token | in |
Token ID | 287 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
favorite | +0.041 |
o | +0.038 |
bie | +0.036 |
bright | +0.035 |
eps | +0.033 |
Neg logit contributions | |
Ivy | -0.056 |
Iris | -0.052 |
Holly | -0.050 |
Beat | -0.050 |
ner | -0.048 |
Token | amaz |
Token ID | 40642 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
o | +0.032 |
favorite | +0.030 |
bie | +0.029 |
eps | +0.027 |
bo | +0.027 |
Neg logit contributions | |
Ivy | -0.048 |
Iris | -0.046 |
Holly | -0.044 |
Beat | -0.044 |
Fiona | -0.043 |
Token | ement |
Token ID | 972 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
o | +0.260 |
favorite | +0.251 |
bie | +0.250 |
eps | +0.219 |
bo | +0.214 |
Neg logit contributions | |
Ivy | -0.307 |
Iris | -0.289 |
Holly | -0.277 |
Beat | -0.274 |
ner | -0.269 |
Token | as |
Token ID | 355 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
o | +0.148 |
favorite | +0.142 |
bie | +0.139 |
eps | +0.125 |
bo | +0.118 |
Neg logit contributions | |
Ivy | -0.292 |
Iris | -0.277 |
Holly | -0.270 |
Beat | -0.264 |
ner | -0.264 |
Token | it |
Token ID | 340 |
Feature activation | -7.59e-03 |
Pos logit contributions | |
favorite | +0.127 |
o | +0.119 |
bie | +0.119 |
bright | +0.111 |
eps | +0.105 |
Neg logit contributions | |
Ivy | -0.143 |
Iris | -0.131 |
Holly | -0.127 |
Beat | -0.122 |
ner | -0.121 |
Token | slowly |
Token ID | 6364 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
Ivy | +0.015 |
Iris | +0.014 |
Holly | +0.014 |
ner | +0.013 |
Beat | +0.013 |
Neg logit contributions | |
favorite | -0.012 |
o | -0.012 |
bie | -0.011 |
eps | -0.010 |
bo | -0.010 |
Token | flew |
Token ID | 13112 |
Feature activation | +2.46e-03 |
Pos logit contributions | |
favorite | +0.044 |
o | +0.044 |
bie | +0.041 |
eps | +0.037 |
bright | +0.037 |
Neg logit contributions | |
Ivy | -0.075 |
Iris | -0.071 |
Holly | -0.068 |
Beat | -0.066 |
ner | -0.065 |
Token | was |
Token ID | 373 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
Ivy | +0.095 |
Iris | +0.090 |
Holly | +0.087 |
Beat | +0.085 |
ner | +0.084 |
Neg logit contributions | |
o | -0.060 |
favorite | -0.060 |
bie | -0.058 |
eps | -0.051 |
bo | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
Ivy | +0.041 |
Iris | +0.039 |
Holly | +0.038 |
Beat | +0.037 |
ner | +0.037 |
Neg logit contributions | |
favorite | -0.013 |
o | -0.013 |
bie | -0.012 |
eps | -0.010 |
bo | -0.009 |
Token | co |
Token ID | 763 |
Feature activation | +7.59e-03 |
Pos logit contributions | |
Ivy | +0.089 |
Iris | +0.084 |
Holly | +0.083 |
Beat | +0.081 |
Fiona | +0.079 |
Neg logit contributions | |
o | -0.033 |
bie | -0.030 |
favorite | -0.028 |
v | -0.026 |
eps | -0.026 |
Token | z |
Token ID | 89 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
favorite | +0.010 |
eps | +0.010 |
tallest | +0.009 |
bie | +0.009 |
vice | +0.009 |
Neg logit contributions | |
Ivy | -0.016 |
ner | -0.015 |
Holly | -0.015 |
Iris | -0.015 |
Fiona | -0.015 |
Token | iest |
Token ID | 6386 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
o | +0.067 |
favorite | +0.067 |
bie | +0.065 |
eps | +0.059 |
bo | +0.057 |
Neg logit contributions | |
Ivy | -0.074 |
Iris | -0.069 |
Holly | -0.066 |
Beat | -0.065 |
ner | -0.064 |
Token | shelter |
Token ID | 11772 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
Ivy | +0.044 |
Holly | +0.041 |
Iris | +0.041 |
Fiona | +0.039 |
ner | +0.039 |
Neg logit contributions | |
o | -0.030 |
favorite | -0.029 |
bie | -0.029 |
eps | -0.027 |
bo | -0.025 |
Token | ever |
Token ID | 1683 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
Ivy | +0.375 |
Iris | +0.358 |
Holly | +0.343 |
Beat | +0.339 |
ner | +0.337 |
Neg logit contributions | |
o | -0.172 |
favorite | -0.172 |
bie | -0.159 |
eps | -0.134 |
bright | -0.131 |
Token | and |
Token ID | 290 |
Feature activation | +8.93e-03 |
Pos logit contributions | |
favorite | +0.024 |
o | +0.023 |
bie | +0.022 |
eps | +0.020 |
vice | +0.018 |
Neg logit contributions | |
Ivy | -0.049 |
Iris | -0.046 |
Holly | -0.045 |
Beat | -0.044 |
ner | -0.044 |
Token | had |
Token ID | 550 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
favorite | +0.014 |
o | +0.014 |
bie | +0.014 |
eps | +0.012 |
bright | +0.012 |
Neg logit contributions | |
Ivy | -0.017 |
Iris | -0.016 |
Holly | -0.015 |
ner | -0.015 |
Beat | -0.015 |
Token | so |
Token ID | 523 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
o | +0.023 |
favorite | +0.022 |
bie | +0.022 |
eps | +0.019 |
bo | +0.019 |
Neg logit contributions | |
Ivy | -0.048 |
Iris | -0.045 |
Holly | -0.044 |
Beat | -0.043 |
Fiona | -0.043 |
Token | much |
Token ID | 881 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
o | +0.028 |
favorite | +0.028 |
bie | +0.027 |
bo | +0.024 |
eps | +0.024 |
Neg logit contributions | |
Ivy | -0.062 |
Iris | -0.059 |
Holly | -0.058 |
Beat | -0.057 |
ner | -0.056 |
Token | say |
Token ID | 910 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
Ivy | +0.034 |
Iris | +0.032 |
Holly | +0.031 |
Beat | +0.031 |
ner | +0.030 |
Neg logit contributions | |
favorite | -0.025 |
o | -0.025 |
bie | -0.024 |
eps | -0.021 |
bo | -0.020 |
Token | it |
Token ID | 340 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
favorite | +0.022 |
o | +0.020 |
bie | +0.019 |
bright | +0.019 |
big | +0.018 |
Neg logit contributions | |
Ivy | -0.024 |
Iris | -0.022 |
Holly | -0.021 |
Beat | -0.021 |
ner | -0.021 |
Token | 's |
Token ID | 338 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
Ivy | +0.098 |
Iris | +0.093 |
Holly | +0.089 |
Beat | +0.087 |
ner | +0.087 |
Neg logit contributions | |
favorite | -0.074 |
o | -0.074 |
bie | -0.071 |
eps | -0.063 |
bo | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
Ivy | +0.024 |
Iris | +0.023 |
Holly | +0.022 |
Beat | +0.022 |
Fiona | +0.021 |
Neg logit contributions | |
o | -0.015 |
favorite | -0.014 |
bie | -0.014 |
v | -0.013 |
eps | -0.013 |
Token | best |
Token ID | 1266 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
Ivy | +0.074 |
Iris | +0.070 |
Holly | +0.067 |
ner | +0.066 |
Beat | +0.066 |
Neg logit contributions | |
favorite | -0.055 |
o | -0.053 |
bie | -0.051 |
eps | -0.046 |
bright | -0.045 |
Token | shelter |
Token ID | 11772 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
favorite | +0.044 |
o | +0.041 |
bie | +0.039 |
bright | +0.034 |
favourite | +0.033 |
Neg logit contributions | |
Ivy | -0.067 |
Iris | -0.063 |
ner | -0.062 |
Holly | -0.061 |
Beat | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | -3.18e-03 |
Pos logit contributions | |
Ivy | +0.359 |
Iris | +0.342 |
Holly | +0.330 |
Beat | +0.327 |
ner | +0.325 |
Neg logit contributions | |
favorite | -0.187 |
o | -0.187 |
bie | -0.173 |
eps | -0.147 |
bo | -0.144 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.20e-02 |
Pos logit contributions | |
Ivy | +0.006 |
Iris | +0.005 |
Holly | +0.005 |
Beat | +0.005 |
ner | +0.005 |
Neg logit contributions | |
favorite | -0.005 |
o | -0.005 |
bie | -0.005 |
eps | -0.005 |
bo | -0.005 |
Token | Ben |
Token ID | 11696 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
Ivy | +0.191 |
Iris | +0.182 |
Holly | +0.175 |
Beat | +0.174 |
ner | +0.171 |
Neg logit contributions | |
o | -0.097 |
favorite | -0.094 |
bie | -0.089 |
eps | -0.077 |
bo | -0.076 |
Token | and |
Token ID | 290 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
o | +0.021 |
favorite | +0.020 |
bie | +0.019 |
bo | +0.016 |
eps | +0.015 |
Neg logit contributions | |
Ivy | -0.047 |
Iris | -0.045 |
ner | -0.044 |
Beat | -0.043 |
Holly | -0.043 |
Token | Mia |
Token ID | 32189 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
favorite | +0.029 |
o | +0.029 |
bie | +0.027 |
eps | +0.024 |
bo | +0.023 |
Neg logit contributions | |
Ivy | -0.053 |
Iris | -0.051 |
Holly | -0.049 |
Beat | -0.048 |
ner | -0.048 |
Token | The |
Token ID | 464 |
Feature activation | -8.08e-03 |
Pos logit contributions | |
Ivy | +0.016 |
Iris | +0.015 |
ner | +0.014 |
Beat | +0.014 |
Holly | +0.013 |
Neg logit contributions | |
favorite | -0.021 |
bie | -0.020 |
o | -0.019 |
favourite | -0.018 |
favorites | -0.017 |
Token | cave |
Token ID | 11527 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
Ivy | +0.017 |
Iris | +0.016 |
Holly | +0.016 |
ner | +0.015 |
Beat | +0.015 |
Neg logit contributions | |
favorite | -0.012 |
o | -0.011 |
bie | -0.010 |
bright | -0.010 |
eps | -0.010 |
Token | is |
Token ID | 318 |
Feature activation | -7.75e-03 |
Pos logit contributions | |
Ivy | +0.035 |
Iris | +0.033 |
Holly | +0.032 |
Beat | +0.031 |
ner | +0.031 |
Neg logit contributions | |
favorite | -0.027 |
o | -0.027 |
bie | -0.026 |
eps | -0.024 |
bo | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
Ivy | +0.019 |
Iris | +0.018 |
Holly | +0.018 |
Beat | +0.018 |
ner | +0.018 |
Neg logit contributions | |
favorite | -0.008 |
o | -0.008 |
bie | -0.007 |
eps | -0.007 |
bright | -0.006 |
Token | best |
Token ID | 1266 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
Ivy | +0.069 |
Iris | +0.066 |
Holly | +0.064 |
ner | +0.062 |
Beat | +0.062 |
Neg logit contributions | |
favorite | -0.041 |
o | -0.038 |
bie | -0.036 |
eps | -0.033 |
bright | -0.033 |
Token | shelter |
Token ID | 11772 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
favorite | +0.046 |
o | +0.040 |
big | +0.039 |
bright | +0.037 |
hot | +0.037 |
Neg logit contributions | |
Ivy | -0.064 |
Iris | -0.060 |
Holly | -0.059 |
ner | -0.057 |
Beat | -0.055 |
Token | ." |
Token ID | 526 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
Ivy | +0.321 |
Iris | +0.306 |
Holly | +0.293 |
Beat | +0.291 |
ner | +0.283 |
Neg logit contributions | |
favorite | -0.208 |
o | -0.208 |
bie | -0.192 |
eps | -0.166 |
bo | -0.166 |
Token | So |
Token ID | 1406 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
Ivy | +0.043 |
Iris | +0.041 |
Holly | +0.039 |
Beat | +0.039 |
ner | +0.038 |
Neg logit contributions | |
favorite | -0.027 |
o | -0.027 |
bie | -0.026 |
eps | -0.023 |
bo | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
Ivy | +0.084 |
Iris | +0.078 |
Holly | +0.076 |
ner | +0.075 |
Beat | +0.075 |
Neg logit contributions | |
favorite | -0.050 |
o | -0.049 |
bie | -0.046 |
bright | -0.041 |
eps | -0.040 |
Token | it |
Token ID | 340 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
Ivy | +0.067 |
Iris | +0.063 |
Holly | +0.060 |
Beat | +0.060 |
ner | +0.058 |
Neg logit contributions | |
favorite | -0.049 |
o | -0.048 |
bie | -0.046 |
eps | -0.040 |
bright | -0.040 |
Token | drove |
Token ID | 10357 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
Ivy | +0.097 |
Iris | +0.092 |
Holly | +0.088 |
Beat | +0.088 |
ner | +0.086 |
Neg logit contributions | |
favorite | -0.059 |
o | -0.059 |
bie | -0.058 |
eps | -0.050 |
bo | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -8.68e-03 |
Pos logit contributions | |
Ivy | +0.046 |
Iris | +0.044 |
Holly | +0.043 |
Beat | +0.042 |
ner | +0.042 |
Neg logit contributions | |
favorite | -0.019 |
o | -0.019 |
bie | -0.019 |
eps | -0.016 |
bo | -0.015 |
Token | They |
Token ID | 1119 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
Ivy | +0.017 |
Iris | +0.016 |
ner | +0.015 |
Holly | +0.015 |
Beat | +0.015 |
Neg logit contributions | |
favorite | -0.014 |
o | -0.014 |
bie | -0.013 |
favourite | -0.011 |
bright | -0.011 |
Token | provided |
Token ID | 2810 |
Feature activation | -7.42e-02 |
Pos logit contributions | |
Ivy | +0.074 |
Iris | +0.070 |
Holly | +0.068 |
Beat | +0.067 |
ner | +0.065 |
Neg logit contributions | |
favorite | -0.042 |
o | -0.042 |
bie | -0.040 |
bo | -0.035 |
eps | -0.034 |
Token | food |
Token ID | 2057 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
Ivy | +0.162 |
Iris | +0.153 |
Holly | +0.147 |
Beat | +0.145 |
ner | +0.143 |
Neg logit contributions | |
o | -0.098 |
favorite | -0.098 |
bie | -0.094 |
eps | -0.083 |
bo | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
Ivy | +0.139 |
Iris | +0.133 |
Beat | +0.127 |
Holly | +0.127 |
ner | +0.127 |
Neg logit contributions | |
favorite | -0.074 |
o | -0.073 |
bie | -0.068 |
eps | -0.057 |
bright | -0.057 |
Token | shelter |
Token ID | 11772 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
Ivy | +0.114 |
Iris | +0.106 |
Holly | +0.103 |
Beat | +0.101 |
Fiona | +0.099 |
Neg logit contributions | |
o | -0.079 |
favorite | -0.078 |
bie | -0.077 |
bo | -0.070 |
eps | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
Ivy | +0.337 |
Iris | +0.321 |
Holly | +0.308 |
ner | +0.304 |
Beat | +0.303 |
Neg logit contributions | |
favorite | -0.190 |
o | -0.189 |
bie | -0.178 |
eps | -0.151 |
bright | -0.147 |
Token | water |
Token ID | 1660 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
Ivy | +0.041 |
Iris | +0.039 |
Holly | +0.037 |
ner | +0.037 |
Beat | +0.037 |
Neg logit contributions | |
favorite | -0.026 |
o | -0.026 |
bie | -0.024 |
eps | -0.021 |
bright | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
Ivy | +0.120 |
Iris | +0.115 |
Holly | +0.110 |
Beat | +0.110 |
ner | +0.109 |
Neg logit contributions | |
o | -0.055 |
favorite | -0.054 |
bie | -0.050 |
eps | -0.042 |
bright | -0.041 |
Token | make |
Token ID | 787 |
Feature activation | -2.10e-03 |
Pos logit contributions | |
Ivy | +0.060 |
Iris | +0.056 |
Holly | +0.054 |
Beat | +0.053 |
ner | +0.053 |
Neg logit contributions | |
favorite | -0.046 |
o | -0.046 |
bie | -0.044 |
eps | -0.040 |
bo | -0.039 |
Token | sure |
Token ID | 1654 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
Ivy | +0.004 |
Iris | +0.004 |
Holly | +0.004 |
Beat | +0.004 |
Fiona | +0.004 |
Neg logit contributions | |
o | -0.003 |
favorite | -0.003 |
bie | -0.003 |
bo | -0.003 |
eps | -0.003 |
Token | 'd |
Token ID | 1549 |
Feature activation | -6.36e-03 |
Pos logit contributions | |
Ivy | +0.012 |
Iris | +0.011 |
Holly | +0.011 |
Beat | +0.011 |
ner | +0.011 |
Neg logit contributions | |
favorite | -0.007 |
o | -0.007 |
bie | -0.007 |
eps | -0.006 |
bright | -0.006 |
Token | created |
Token ID | 2727 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
Ivy | +0.013 |
Iris | +0.012 |
Holly | +0.012 |
Beat | +0.012 |
ner | +0.011 |
Neg logit contributions | |
favorite | -0.009 |
o | -0.009 |
bie | -0.009 |
eps | -0.008 |
bo | -0.008 |
Token | her |
Token ID | 607 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
Ivy | +0.045 |
Iris | +0.043 |
Holly | +0.042 |
Beat | +0.041 |
ner | +0.041 |
Neg logit contributions | |
o | -0.024 |
favorite | -0.024 |
bie | -0.023 |
eps | -0.020 |
bo | -0.019 |
Token | own |
Token ID | 898 |
Feature activation | +6.44e-03 |
Pos logit contributions | |
Ivy | +0.059 |
Iris | +0.056 |
Holly | +0.056 |
Beat | +0.053 |
Fiona | +0.053 |
Neg logit contributions | |
o | -0.032 |
bie | -0.028 |
eps | -0.027 |
v | -0.027 |
bo | -0.027 |
Token | special |
Token ID | 2041 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
o | +0.009 |
favorite | +0.008 |
bie | +0.008 |
eps | +0.008 |
bo | +0.007 |
Neg logit contributions | |
Ivy | -0.014 |
Iris | -0.013 |
Holly | -0.012 |
Beat | -0.012 |
ner | -0.012 |
Token | shelter |
Token ID | 11772 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
Ivy | +0.041 |
Iris | +0.038 |
Holly | +0.037 |
Beat | +0.036 |
ner | +0.036 |
Neg logit contributions | |
favorite | -0.030 |
o | -0.030 |
bie | -0.029 |
eps | -0.026 |
bo | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
Ivy | +0.357 |
Iris | +0.341 |
Holly | +0.325 |
Beat | +0.323 |
ner | +0.320 |
Neg logit contributions | |
favorite | -0.164 |
o | -0.163 |
bie | -0.152 |
eps | -0.126 |
bright | -0.123 |
Token | And |
Token ID | 843 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
Ivy | +0.025 |
Iris | +0.024 |
Holly | +0.023 |
Beat | +0.022 |
ner | +0.022 |
Neg logit contributions | |
o | -0.014 |
favorite | -0.014 |
bie | -0.013 |
eps | -0.011 |
bo | -0.011 |
Token | from |
Token ID | 422 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
Ivy | +0.021 |
Iris | +0.020 |
Holly | +0.019 |
Beat | +0.019 |
ner | +0.019 |
Neg logit contributions | |
favorite | -0.017 |
o | -0.017 |
bie | -0.016 |
eps | -0.014 |
bo | -0.014 |
Token | that |
Token ID | 326 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
Ivy | +0.113 |
Iris | +0.107 |
Holly | +0.103 |
Beat | +0.101 |
ner | +0.099 |
Neg logit contributions | |
o | -0.077 |
favorite | -0.076 |
bie | -0.074 |
eps | -0.066 |
v | -0.064 |
Token | day |
Token ID | 1110 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
o | +0.021 |
favorite | +0.020 |
bie | +0.019 |
eps | +0.017 |
v | +0.016 |
Neg logit contributions | |
Ivy | -0.045 |
Iris | -0.042 |
Holly | -0.041 |
Beat | -0.041 |
ner | -0.040 |
Token | there |
Token ID | 612 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
Ivy | +0.002 |
Iris | +0.002 |
Holly | +0.002 |
Beat | +0.002 |
ner | +0.002 |
Neg logit contributions | |
favorite | -0.001 |
o | -0.001 |
bie | -0.001 |
bright | -0.001 |
eps | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
Ivy | +0.053 |
Iris | +0.050 |
Holly | +0.048 |
Beat | +0.047 |
ner | +0.047 |
Neg logit contributions | |
o | -0.034 |
favorite | -0.034 |
bie | -0.032 |
eps | -0.028 |
bo | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
bie | +0.013 |
o | +0.013 |
bo | +0.012 |
favorite | +0.012 |
v | +0.012 |
Neg logit contributions | |
Ivy | -0.026 |
Iris | -0.024 |
Holly | -0.023 |
Fiona | -0.023 |
Beat | -0.023 |
Token | sheep |
Token ID | 15900 |
Feature activation | +8.87e-04 |
Pos logit contributions | |
Ivy | +0.044 |
Iris | +0.042 |
Holly | +0.041 |
Beat | +0.040 |
ner | +0.039 |
Neg logit contributions | |
o | -0.017 |
favorite | -0.016 |
bie | -0.016 |
v | -0.013 |
eps | -0.013 |
Token | named |
Token ID | 3706 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
favorite | +0.001 |
o | +0.001 |
bie | +0.001 |
eps | +0.001 |
bo | +0.001 |
Neg logit contributions | |
Ivy | -0.002 |
Iris | -0.002 |
Holly | -0.002 |
Beat | -0.002 |
ner | -0.002 |
Token | Woo |
Token ID | 39832 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
Ivy | +0.054 |
Iris | +0.050 |
ner | +0.047 |
Holly | +0.047 |
Beat | +0.046 |
Neg logit contributions | |
o | -0.068 |
favorite | -0.066 |
bie | -0.064 |
bright | -0.057 |
bo | -0.057 |
Token | ly |
Token ID | 306 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
Ivy | +0.294 |
Iris | +0.280 |
ner | +0.273 |
Holly | +0.260 |
Beat | +0.257 |
Neg logit contributions | |
favorite | -0.226 |
o | -0.217 |
bie | -0.199 |
bo | -0.184 |
v | -0.172 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Ivy | +0.055 |
Iris | +0.051 |
Holly | +0.051 |
Beat | +0.050 |
Fiona | +0.049 |
Neg logit contributions | |
o | -0.026 |
bie | -0.026 |
favorite | -0.025 |
eps | -0.024 |
vice | -0.022 |
Token | Woo |
Token ID | 39832 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
Ivy | +0.023 |
Iris | +0.021 |
ner | +0.021 |
pan | +0.021 |
ion | +0.020 |
Neg logit contributions | |
favorite | -0.023 |
o | -0.022 |
big | -0.020 |
bright | -0.020 |
black | -0.020 |
Token | ly |
Token ID | 306 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
Ivy | +0.295 |
Iris | +0.280 |
ner | +0.279 |
pan | +0.260 |
ion | +0.257 |
Neg logit contributions | |
favorite | -0.224 |
o | -0.200 |
bie | -0.182 |
bo | -0.178 |
bright | -0.163 |
Token | had |
Token ID | 550 |
Feature activation | +8.44e-03 |
Pos logit contributions | |
Ivy | +0.047 |
Iris | +0.044 |
Holly | +0.043 |
Beat | +0.042 |
ner | +0.042 |
Neg logit contributions | |
favorite | -0.025 |
o | -0.025 |
bie | -0.024 |
eps | -0.021 |
bo | -0.020 |
Token | His |
Token ID | 2399 |
Feature activation | -5.43e-03 |
Pos logit contributions | |
bie | +0.003 |
favorite | +0.003 |
eps | +0.003 |
o | +0.002 |
ams | +0.002 |
Neg logit contributions | |
Ivy | -0.004 |
Iris | -0.003 |
Holly | -0.003 |
Beat | -0.003 |
Fiona | -0.003 |
Token | home |
Token ID | 1363 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
Ivy | +0.010 |
Holly | +0.010 |
Iris | +0.010 |
Beat | +0.010 |
Fiona | +0.010 |
Neg logit contributions | |
ams | -0.008 |
vice | -0.008 |
v | -0.007 |
o | -0.007 |
eps | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
Ivy | +0.110 |
Iris | +0.102 |
ner | +0.101 |
Holly | +0.101 |
Beat | +0.097 |
Neg logit contributions | |
bie | -0.096 |
favorite | -0.093 |
eps | -0.089 |
o | -0.087 |
tallest | -0.086 |
Token | a |
Token ID | 257 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
Ivy | +0.041 |
Holly | +0.039 |
Iris | +0.038 |
Beat | +0.037 |
Fiona | +0.036 |
Neg logit contributions | |
bie | -0.020 |
pter | -0.018 |
o | -0.018 |
v | -0.018 |
aw | -0.017 |
Token | big |
Token ID | 1263 |
Feature activation | +2.35e-03 |
Pos logit contributions | |
Ivy | +0.103 |
Iris | +0.097 |
Holly | +0.096 |
Beat | +0.093 |
ner | +0.092 |
Neg logit contributions | |
o | -0.043 |
bie | -0.041 |
favorite | -0.039 |
v | -0.036 |
ams | -0.036 |
Token | shelter |
Token ID | 11772 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
favorite | +0.003 |
o | +0.003 |
bie | +0.003 |
eps | +0.003 |
bo | +0.003 |
Neg logit contributions | |
Ivy | -0.005 |
Iris | -0.005 |
Holly | -0.005 |
Beat | -0.004 |
ner | -0.004 |
Token | in |
Token ID | 287 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
Ivy | +0.316 |
Iris | +0.295 |
Holly | +0.289 |
ner | +0.285 |
Beat | +0.282 |
Neg logit contributions | |
bie | -0.180 |
favorite | -0.176 |
o | -0.171 |
eps | -0.162 |
v | -0.147 |
Token | the |
Token ID | 262 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
Ivy | +0.021 |
Iris | +0.020 |
Holly | +0.019 |
Beat | +0.019 |
Fiona | +0.019 |
Neg logit contributions | |
favorite | -0.019 |
o | -0.018 |
bie | -0.018 |
int | -0.017 |
eps | -0.017 |
Token | woods |
Token ID | 16479 |
Feature activation | +5.71e-03 |
Pos logit contributions | |
o | +0.045 |
favorite | +0.045 |
bie | +0.043 |
eps | +0.038 |
v | +0.036 |
Neg logit contributions | |
Ivy | -0.071 |
Iris | -0.067 |
Holly | -0.065 |
Beat | -0.064 |
ner | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -7.96e-03 |
Pos logit contributions | |
bie | +0.007 |
favorite | +0.007 |
eps | +0.007 |
o | +0.007 |
bo | +0.006 |
Neg logit contributions | |
Ivy | -0.011 |
ner | -0.011 |
Iris | -0.011 |
Holly | -0.010 |
Beat | -0.010 |
Token | One |
Token ID | 1881 |
Feature activation | -2.68e-03 |
Pos logit contributions | |
Ivy | +0.016 |
Iris | +0.015 |
Holly | +0.015 |
ner | +0.015 |
Beat | +0.015 |
Neg logit contributions | |
bie | -0.011 |
favorite | -0.011 |
o | -0.011 |
eps | -0.011 |
ams | -0.010 |
Token | move |
Token ID | 1445 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
o | +0.042 |
favorite | +0.042 |
bie | +0.040 |
eps | +0.036 |
bo | +0.035 |
Neg logit contributions | |
Ivy | -0.073 |
Iris | -0.069 |
Holly | -0.067 |
ner | -0.066 |
Beat | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +8.13e-03 |
Pos logit contributions | |
favorite | +0.040 |
o | +0.040 |
bie | +0.038 |
eps | +0.033 |
bo | +0.032 |
Neg logit contributions | |
Ivy | -0.082 |
Iris | -0.078 |
Holly | -0.075 |
Beat | -0.074 |
ner | -0.074 |
Token | Then |
Token ID | 3244 |
Feature activation | +3.19e-03 |
Pos logit contributions | |
favorite | +0.013 |
o | +0.013 |
bie | +0.012 |
bright | +0.011 |
bo | +0.010 |
Neg logit contributions | |
Ivy | -0.016 |
Iris | -0.015 |
ner | -0.014 |
Holly | -0.014 |
Beat | -0.014 |
Token | something |
Token ID | 1223 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
o | +0.005 |
favorite | +0.005 |
bie | +0.005 |
eps | +0.005 |
bo | +0.004 |
Neg logit contributions | |
Ivy | -0.006 |
Iris | -0.006 |
Holly | -0.005 |
Beat | -0.005 |
ner | -0.005 |
Token | even |
Token ID | 772 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
Ivy | +0.072 |
Iris | +0.068 |
ner | +0.067 |
Holly | +0.066 |
Beat | +0.065 |
Neg logit contributions | |
favorite | -0.036 |
o | -0.032 |
bright | -0.029 |
eps | -0.029 |
bie | -0.028 |
Token | worse |
Token ID | 4785 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
favorite | +0.017 |
o | +0.017 |
bie | +0.017 |
eps | +0.014 |
bo | +0.014 |
Neg logit contributions | |
Ivy | -0.025 |
Iris | -0.024 |
Holly | -0.023 |
Beat | -0.022 |
ner | -0.022 |
Token | happened |
Token ID | 3022 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
Ivy | +0.321 |
Iris | +0.305 |
Holly | +0.290 |
ner | +0.288 |
Beat | +0.278 |
Neg logit contributions | |
o | -0.189 |
favorite | -0.186 |
bie | -0.170 |
drawn | -0.152 |
eps | -0.152 |
Token | . |
Token ID | 13 |
Feature activation | +7.93e-03 |
Pos logit contributions | |
Ivy | +0.057 |
Iris | +0.054 |
Holly | +0.053 |
Beat | +0.052 |
ner | +0.051 |
Neg logit contributions | |
favorite | -0.022 |
o | -0.022 |
bie | -0.022 |
eps | -0.018 |
v | -0.017 |
Token | A |
Token ID | 317 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
o | +0.011 |
favorite | +0.010 |
bie | +0.010 |
eps | +0.008 |
bright | +0.008 |
Neg logit contributions | |
Ivy | -0.018 |
Iris | -0.017 |
Holly | -0.016 |
ner | -0.016 |
Beat | -0.016 |
Token | scary |
Token ID | 14343 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
Ivy | +0.084 |
Iris | +0.079 |
Holly | +0.077 |
Beat | +0.076 |
ner | +0.076 |
Neg logit contributions | |
o | -0.034 |
favorite | -0.033 |
bie | -0.032 |
eps | -0.027 |
v | -0.026 |
Token | monster |
Token ID | 9234 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
Ivy | +0.082 |
Iris | +0.077 |
Holly | +0.074 |
Beat | +0.072 |
ner | +0.072 |
Neg logit contributions | |
favorite | -0.064 |
o | -0.064 |
bie | -0.061 |
bright | -0.054 |
eps | -0.054 |
Token | sheep |
Token ID | 15900 |
Feature activation | +8.87e-04 |
Pos logit contributions | |
Ivy | +0.044 |
Iris | +0.042 |
Holly | +0.041 |
Beat | +0.040 |
ner | +0.039 |
Neg logit contributions | |
o | -0.017 |
favorite | -0.016 |
bie | -0.016 |
v | -0.013 |
eps | -0.013 |
Token | named |
Token ID | 3706 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
favorite | +0.001 |
o | +0.001 |
bie | +0.001 |
eps | +0.001 |
bo | +0.001 |
Neg logit contributions | |
Ivy | -0.002 |
Iris | -0.002 |
Holly | -0.002 |
Beat | -0.002 |
ner | -0.002 |
Token | Woo |
Token ID | 39832 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
Ivy | +0.054 |
Iris | +0.050 |
ner | +0.047 |
Holly | +0.047 |
Beat | +0.046 |
Neg logit contributions | |
o | -0.068 |
favorite | -0.066 |
bie | -0.064 |
bright | -0.057 |
bo | -0.057 |
Token | ly |
Token ID | 306 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
Ivy | +0.294 |
Iris | +0.280 |
ner | +0.273 |
Holly | +0.260 |
Beat | +0.257 |
Neg logit contributions | |
favorite | -0.226 |
o | -0.217 |
bie | -0.199 |
bo | -0.184 |
v | -0.172 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Ivy | +0.055 |
Iris | +0.051 |
Holly | +0.051 |
Beat | +0.050 |
Fiona | +0.049 |
Neg logit contributions | |
o | -0.026 |
bie | -0.026 |
favorite | -0.025 |
eps | -0.024 |
vice | -0.022 |
Token | Woo |
Token ID | 39832 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
Ivy | +0.023 |
Iris | +0.021 |
ner | +0.021 |
pan | +0.021 |
ion | +0.020 |
Neg logit contributions | |
favorite | -0.023 |
o | -0.022 |
big | -0.020 |
bright | -0.020 |
black | -0.020 |
Token | ly |
Token ID | 306 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
Ivy | +0.295 |
Iris | +0.280 |
ner | +0.279 |
pan | +0.260 |
ion | +0.257 |
Neg logit contributions | |
favorite | -0.224 |
o | -0.200 |
bie | -0.182 |
bo | -0.178 |
bright | -0.163 |
Token | had |
Token ID | 550 |
Feature activation | +8.44e-03 |
Pos logit contributions | |
Ivy | +0.047 |
Iris | +0.044 |
Holly | +0.043 |
Beat | +0.042 |
ner | +0.042 |
Neg logit contributions | |
favorite | -0.025 |
o | -0.025 |
bie | -0.024 |
eps | -0.021 |
bo | -0.020 |
Token | lots |
Token ID | 6041 |
Feature activation | +1.55e-03 |
Pos logit contributions | |
bie | +0.011 |
int | +0.010 |
o | +0.010 |
favorite | +0.009 |
ams | +0.009 |
Neg logit contributions | |
Ivy | -0.018 |
Iris | -0.017 |
Fiona | -0.017 |
Holly | -0.017 |
Beat | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -7.10e-03 |
Pos logit contributions | |
bie | +0.003 |
o | +0.003 |
favorite | +0.003 |
eps | +0.003 |
bo | +0.003 |
Neg logit contributions | |
Ivy | -0.003 |
Iris | -0.002 |
Holly | -0.002 |
Beat | -0.002 |
ner | -0.002 |
Token | lots |
Token ID | 6041 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
Ivy | +0.018 |
Iris | +0.017 |
Holly | +0.016 |
Beat | +0.016 |
Fiona | +0.016 |
Neg logit contributions | |
o | -0.007 |
favorite | -0.007 |
bie | -0.007 |
eps | -0.006 |
bo | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +3.61e-03 |
Pos logit contributions | |
favorite | +0.014 |
o | +0.014 |
bie | +0.013 |
v | +0.010 |
eps | +0.010 |
Neg logit contributions | |
Ivy | -0.048 |
Iris | -0.047 |
Beat | -0.045 |
Holly | -0.045 |
ner | -0.044 |
Token | She |
Token ID | 1375 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
favorite | +0.006 |
o | +0.006 |
bie | +0.006 |
bright | +0.005 |
eps | +0.005 |
Neg logit contributions | |
Ivy | -0.006 |
Iris | -0.006 |
ner | -0.006 |
Holly | -0.006 |
Beat | -0.006 |
Token | called |
Token ID | 1444 |
Feature activation | +2.19e-03 |
Pos logit contributions | |
Ivy | +0.050 |
Iris | +0.047 |
Holly | +0.045 |
Beat | +0.045 |
ner | +0.044 |
Neg logit contributions | |
favorite | -0.027 |
o | -0.027 |
bie | -0.026 |
eps | -0.022 |
bo | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
favorite | +0.002 |
o | +0.002 |
big | +0.002 |
hot | +0.002 |
at | +0.002 |
Neg logit contributions | |
Ivy | -0.005 |
ner | -0.005 |
Iris | -0.005 |
Holly | -0.005 |
Beat | -0.005 |
Token | animal |
Token ID | 5044 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
Ivy | +0.053 |
Iris | +0.050 |
Holly | +0.048 |
Beat | +0.047 |
ner | +0.045 |
Neg logit contributions | |
o | -0.042 |
bie | -0.041 |
favorite | -0.040 |
eps | -0.035 |
ams | -0.035 |
Token | shelter |
Token ID | 11772 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
Ivy | +0.065 |
Iris | +0.061 |
Holly | +0.059 |
Beat | +0.058 |
ner | +0.057 |
Neg logit contributions | |
favorite | -0.041 |
o | -0.041 |
bie | -0.039 |
eps | -0.035 |
bo | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -7.87e-03 |
Pos logit contributions | |
Ivy | +0.343 |
Iris | +0.327 |
Holly | +0.314 |
Beat | +0.311 |
ner | +0.310 |
Neg logit contributions | |
o | -0.162 |
favorite | -0.160 |
bie | -0.148 |
eps | -0.126 |
bright | -0.120 |
Token | told |
Token ID | 1297 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
Ivy | +0.019 |
Iris | +0.018 |
ner | +0.017 |
Holly | +0.017 |
Beat | +0.017 |
Neg logit contributions | |
favorite | -0.009 |
o | -0.009 |
bie | -0.008 |
bright | -0.007 |
eps | -0.007 |
Token | them |
Token ID | 606 |
Feature activation | +8.20e-03 |
Pos logit contributions | |
Ivy | +0.028 |
Iris | +0.026 |
Holly | +0.025 |
Beat | +0.024 |
ner | +0.024 |
Neg logit contributions | |
favorite | -0.037 |
o | -0.037 |
bie | -0.036 |
eps | -0.033 |
bo | -0.032 |
Token | about |
Token ID | 546 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
favorite | +0.010 |
o | +0.009 |
bie | +0.009 |
eps | +0.007 |
bright | +0.007 |
Neg logit contributions | |
Ivy | -0.019 |
Iris | -0.019 |
Holly | -0.018 |
ner | -0.018 |
Beat | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
favorite | +0.028 |
o | +0.027 |
bie | +0.026 |
eps | +0.023 |
bright | +0.022 |
Neg logit contributions | |
Ivy | -0.040 |
Iris | -0.038 |
Holly | -0.036 |
ner | -0.036 |
Beat | -0.035 |
Token | could |
Token ID | 714 |
Feature activation | +5.16e-03 |
Pos logit contributions | |
Ivy | +0.026 |
Iris | +0.025 |
Holly | +0.024 |
Beat | +0.023 |
ner | +0.023 |
Neg logit contributions | |
favorite | -0.025 |
bie | -0.024 |
o | -0.023 |
favorites | -0.022 |
ams | -0.021 |
Token | see |
Token ID | 766 |
Feature activation | +6.54e-02 |
Pos logit contributions | |
o | +0.007 |
favorite | +0.007 |
bie | +0.007 |
eps | +0.006 |
bo | +0.006 |
Neg logit contributions | |
Ivy | -0.011 |
Iris | -0.010 |
Holly | -0.010 |
Beat | -0.010 |
ner | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +8.59e-03 |
Pos logit contributions | |
o | +0.103 |
bie | +0.101 |
favorite | +0.096 |
bo | +0.090 |
eps | +0.088 |
Neg logit contributions | |
Ivy | -0.124 |
Iris | -0.118 |
Beat | -0.113 |
Holly | -0.111 |
Fiona | -0.110 |
Token | feel |
Token ID | 1254 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
favorite | +0.012 |
o | +0.012 |
bie | +0.011 |
eps | +0.010 |
bright | +0.010 |
Neg logit contributions | |
Ivy | -0.018 |
Iris | -0.017 |
Holly | -0.017 |
Beat | -0.016 |
ner | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
Ivy | +0.046 |
Iris | +0.044 |
Holly | +0.043 |
Beat | +0.042 |
ner | +0.042 |
Neg logit contributions | |
favorite | -0.020 |
o | -0.020 |
bie | -0.019 |
eps | -0.016 |
bo | -0.015 |
Token | shelter |
Token ID | 11772 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
Ivy | +0.114 |
Iris | +0.108 |
Holly | +0.105 |
Beat | +0.103 |
ner | +0.103 |
Neg logit contributions | |
favorite | -0.060 |
o | -0.056 |
bie | -0.055 |
bright | -0.049 |
eps | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -5.53e-03 |
Pos logit contributions | |
Ivy | +0.334 |
Iris | +0.317 |
Holly | +0.307 |
Beat | +0.303 |
ner | +0.298 |
Neg logit contributions | |
favorite | -0.164 |
o | -0.161 |
bie | -0.150 |
eps | -0.133 |
bright | -0.131 |
Token |
|
Token ID | 198 |
Feature activation | +9.88e-04 |
Pos logit contributions | |
Ivy | +0.013 |
Iris | +0.012 |
Holly | +0.012 |
Beat | +0.012 |
ner | +0.011 |
Neg logit contributions | |
o | -0.006 |
favorite | -0.006 |
bie | -0.006 |
bright | -0.005 |
v | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +5.78e-03 |
Pos logit contributions | |
o | +0.001 |
favorite | +0.001 |
bie | +0.001 |
eps | +0.001 |
bo | +0.001 |
Neg logit contributions | |
Ivy | -0.003 |
Iris | -0.002 |
Holly | -0.002 |
Beat | -0.002 |
ner | -0.002 |
Token | And |
Token ID | 1870 |
Feature activation | -7.28e-03 |
Pos logit contributions | |
o | +0.007 |
favorite | +0.007 |
eps | +0.007 |
bie | +0.006 |
bright | +0.006 |
Neg logit contributions | |
Ivy | -0.013 |
Iris | -0.012 |
Holly | -0.012 |
Beat | -0.012 |
Fiona | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
Ivy | +0.017 |
Iris | +0.016 |
Holly | +0.016 |
ner | +0.015 |
Beat | +0.015 |
Neg logit contributions | |
favorite | -0.009 |
o | -0.008 |
bie | -0.008 |
bright | -0.007 |
eps | -0.007 |
Token | he |
Token ID | 339 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
o | +0.037 |
favorite | +0.036 |
bie | +0.035 |
eps | +0.033 |
v | +0.032 |
Neg logit contributions | |
Ivy | -0.048 |
Iris | -0.045 |
Holly | -0.043 |
Beat | -0.042 |
ner | -0.041 |
Token | soon |
Token ID | 2582 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
o | +0.059 |
favorite | +0.058 |
bie | +0.057 |
eps | +0.052 |
tallest | +0.051 |
Neg logit contributions | |
Ivy | -0.088 |
Iris | -0.083 |
Holly | -0.081 |
Beat | -0.078 |
Fiona | -0.076 |
Token | noticed |
Token ID | 6810 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
o | +0.035 |
favorite | +0.035 |
bie | +0.033 |
eps | +0.030 |
tallest | +0.029 |
Neg logit contributions | |
Ivy | -0.053 |
Iris | -0.050 |
Holly | -0.049 |
Beat | -0.048 |
ner | -0.047 |
Token | something |
Token ID | 1223 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
o | +0.048 |
bie | +0.047 |
favorite | +0.047 |
eps | +0.041 |
bo | +0.041 |
Neg logit contributions | |
Ivy | -0.067 |
Iris | -0.063 |
Holly | -0.061 |
Beat | -0.060 |
ner | -0.059 |
Token | much |
Token ID | 881 |
Feature activation | -9.30e-03 |
Pos logit contributions | |
Ivy | +0.079 |
Iris | +0.076 |
Holly | +0.073 |
Beat | +0.072 |
ner | +0.072 |
Neg logit contributions | |
favorite | -0.032 |
o | -0.032 |
bie | -0.030 |
eps | -0.025 |
bo | -0.024 |
Token | worse |
Token ID | 4785 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
Ivy | +0.020 |
Iris | +0.018 |
Holly | +0.018 |
Beat | +0.017 |
ner | +0.017 |
Neg logit contributions | |
favorite | -0.013 |
o | -0.013 |
bie | -0.012 |
eps | -0.011 |
bright | -0.011 |
Token | - |
Token ID | 532 |
Feature activation | +8.46e-03 |
Pos logit contributions | |
Ivy | +0.353 |
Iris | +0.336 |
Holly | +0.326 |
Beat | +0.323 |
ner | +0.319 |
Neg logit contributions | |
o | -0.138 |
favorite | -0.138 |
bie | -0.132 |
eps | -0.111 |
bo | -0.104 |
Token | a |
Token ID | 257 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
favorite | +0.011 |
o | +0.011 |
bie | +0.010 |
bright | +0.009 |
eps | +0.009 |
Neg logit contributions | |
Ivy | -0.019 |
Iris | -0.018 |
Holly | -0.017 |
Beat | -0.017 |
ner | -0.017 |
Token | large |
Token ID | 1588 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
Ivy | +0.072 |
Iris | +0.069 |
Holly | +0.067 |
Beat | +0.066 |
ner | +0.066 |
Neg logit contributions | |
favorite | -0.029 |
o | -0.027 |
bie | -0.026 |
bright | -0.023 |
eps | -0.021 |
Token | hole |
Token ID | 7604 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
favorite | +0.019 |
black | +0.019 |
bright | +0.019 |
big | +0.018 |
red | +0.017 |
Neg logit contributions | |
Ivy | -0.036 |
Iris | -0.034 |
Holly | -0.033 |
Beat | -0.033 |
ner | -0.033 |
Token | in |
Token ID | 287 |
Feature activation | -7.44e-03 |
Pos logit contributions | |
Ivy | +0.141 |
Iris | +0.133 |
Holly | +0.129 |
Beat | +0.127 |
ner | +0.127 |
Neg logit contributions | |
favorite | -0.071 |
o | -0.070 |
bie | -0.068 |
eps | -0.058 |
bo | -0.055 |
Token | the |
Token ID | 262 |
Feature activation | +3.53e-03 |
Pos logit contributions | |
Ivy | +0.107 |
Iris | +0.101 |
Holly | +0.097 |
Beat | +0.095 |
Fiona | +0.093 |
Neg logit contributions | |
o | -0.079 |
favorite | -0.078 |
bie | -0.076 |
eps | -0.068 |
bo | -0.066 |
Token | brothers |
Token ID | 9397 |
Feature activation | +6.29e-02 |
Pos logit contributions | |
o | +0.006 |
bie | +0.006 |
favorite | +0.005 |
bo | +0.005 |
vice | +0.005 |
Neg logit contributions | |
Ivy | -0.006 |
Iris | -0.006 |
Holly | -0.006 |
Beat | -0.006 |
Fiona | -0.006 |
Token | find |
Token ID | 1064 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
favorite | +0.091 |
o | +0.091 |
bie | +0.087 |
eps | +0.078 |
bo | +0.076 |
Neg logit contributions | |
Ivy | -0.127 |
Iris | -0.119 |
Holly | -0.115 |
ner | -0.113 |
Beat | -0.112 |
Token | food |
Token ID | 2057 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
Ivy | +0.027 |
Iris | +0.025 |
Holly | +0.024 |
Beat | +0.024 |
Fiona | +0.024 |
Neg logit contributions | |
o | -0.023 |
bie | -0.022 |
favorite | -0.021 |
eps | -0.019 |
bo | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
Ivy | +0.062 |
Iris | +0.060 |
Beat | +0.057 |
ner | +0.057 |
Holly | +0.056 |
Neg logit contributions | |
favorite | -0.033 |
o | -0.032 |
bie | -0.030 |
bright | -0.026 |
eps | -0.025 |
Token | shelter |
Token ID | 11772 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
Ivy | +0.097 |
Iris | +0.091 |
Holly | +0.089 |
Beat | +0.087 |
ner | +0.086 |
Neg logit contributions | |
o | -0.061 |
favorite | -0.061 |
bie | -0.058 |
eps | -0.052 |
bo | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
Ivy | +0.324 |
Iris | +0.310 |
Holly | +0.296 |
ner | +0.295 |
Beat | +0.292 |
Neg logit contributions | |
o | -0.172 |
favorite | -0.172 |
bie | -0.161 |
bright | -0.135 |
eps | -0.134 |
Token | and |
Token ID | 290 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
Ivy | +0.083 |
Iris | +0.078 |
Holly | +0.076 |
Beat | +0.075 |
ner | +0.074 |
Neg logit contributions | |
o | -0.048 |
favorite | -0.048 |
bie | -0.046 |
eps | -0.040 |
bo | -0.039 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
Ivy | +0.020 |
Iris | +0.019 |
Holly | +0.018 |
Beat | +0.018 |
ner | +0.017 |
Neg logit contributions | |
favorite | -0.016 |
o | -0.016 |
bie | -0.016 |
eps | -0.014 |
bo | -0.013 |
Token | them |
Token ID | 606 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
Ivy | +0.018 |
Iris | +0.017 |
Holly | +0.016 |
Beat | +0.016 |
Fiona | +0.015 |
Neg logit contributions | |
o | -0.024 |
favorite | -0.023 |
bie | -0.023 |
eps | -0.021 |
bo | -0.021 |
Token | around |
Token ID | 1088 |
Feature activation | +6.45e-04 |
Pos logit contributions | |
Ivy | +0.039 |
Iris | +0.037 |
Holly | +0.035 |
Beat | +0.035 |
ner | +0.035 |
Neg logit contributions | |
favorite | -0.022 |
o | -0.022 |
bie | -0.021 |
eps | -0.018 |
bright | -0.017 |
Token | " |
Token ID | 1 |
Feature activation | +9.47e-03 |
Pos logit contributions | |
o | +0.021 |
favorite | +0.019 |
bie | +0.019 |
eps | +0.018 |
bo | +0.017 |
Neg logit contributions | |
Ivy | -0.031 |
Iris | -0.029 |
Holly | -0.028 |
ner | -0.028 |
Beat | -0.028 |
Token | That |
Token ID | 2504 |
Feature activation | -9.22e-03 |
Pos logit contributions | |
o | +0.014 |
favorite | +0.014 |
bie | +0.013 |
eps | +0.012 |
bo | +0.011 |
Neg logit contributions | |
Ivy | -0.019 |
Iris | -0.018 |
Holly | -0.018 |
Beat | -0.017 |
ner | -0.017 |
Token | is |
Token ID | 318 |
Feature activation | +4.36e-04 |
Pos logit contributions | |
Ivy | +0.022 |
Iris | +0.021 |
Holly | +0.021 |
Beat | +0.020 |
ner | +0.020 |
Neg logit contributions | |
favorite | -0.010 |
o | -0.010 |
bie | -0.009 |
eps | -0.008 |
bo | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
o | +0.000 |
bie | +0.000 |
favorite | +0.000 |
eps | +0.000 |
v | +0.000 |
Neg logit contributions | |
Ivy | -0.001 |
Iris | -0.001 |
Holly | -0.001 |
Beat | -0.001 |
Fiona | -0.001 |
Token | nice |
Token ID | 3621 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
Ivy | +0.080 |
Iris | +0.076 |
Holly | +0.073 |
Beat | +0.072 |
ner | +0.072 |
Neg logit contributions | |
o | -0.040 |
favorite | -0.040 |
bie | -0.038 |
eps | -0.033 |
bo | -0.031 |
Token | shelter |
Token ID | 11772 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
Ivy | +0.069 |
Holly | +0.065 |
ner | +0.065 |
Iris | +0.065 |
Beat | +0.064 |
Neg logit contributions | |
favorite | -0.039 |
o | -0.035 |
big | -0.035 |
bright | -0.034 |
bie | -0.033 |
Token | ," |
Token ID | 553 |
Feature activation | +6.75e-03 |
Pos logit contributions | |
Ivy | +0.298 |
Iris | +0.284 |
Holly | +0.273 |
Beat | +0.270 |
ner | +0.265 |
Neg logit contributions | |
o | -0.197 |
favorite | -0.195 |
bie | -0.184 |
eps | -0.161 |
bo | -0.158 |
Token | she |
Token ID | 673 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
favorite | +0.011 |
o | +0.011 |
bie | +0.011 |
bright | +0.010 |
eps | +0.009 |
Neg logit contributions | |
Ivy | -0.012 |
Iris | -0.011 |
ner | -0.011 |
Beat | -0.011 |
Holly | -0.011 |
Token | says |
Token ID | 1139 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
favorite | +0.018 |
o | +0.018 |
bie | +0.017 |
eps | +0.015 |
bo | +0.015 |
Neg logit contributions | |
Ivy | -0.031 |
Iris | -0.029 |
Holly | -0.028 |
ner | -0.028 |
Beat | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +5.41e-03 |
Pos logit contributions | |
favorite | +0.021 |
o | +0.021 |
bie | +0.020 |
eps | +0.016 |
bo | +0.015 |
Neg logit contributions | |
Ivy | -0.058 |
Iris | -0.056 |
Holly | -0.054 |
ner | -0.053 |
Beat | -0.053 |
Token | " |
Token ID | 366 |
Feature activation | -5.98e-03 |
Pos logit contributions | |
o | +0.007 |
favorite | +0.006 |
bie | +0.006 |
bright | +0.005 |
black | +0.005 |
Neg logit contributions | |
Ivy | -0.012 |
Iris | -0.011 |
Holly | -0.011 |
Beat | -0.011 |
ner | -0.011 |
Token | met |
Token ID | 1138 |
Feature activation | +3.80e-03 |
Pos logit contributions | |
o | +0.016 |
favorite | +0.015 |
bie | +0.014 |
bo | +0.011 |
eps | +0.011 |
Neg logit contributions | |
Ivy | -0.045 |
Iris | -0.044 |
ner | -0.042 |
Holly | -0.041 |
Beat | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
bie | +0.004 |
favorite | +0.004 |
o | +0.004 |
eps | +0.003 |
v | +0.003 |
Neg logit contributions | |
Ivy | -0.009 |
Iris | -0.009 |
Holly | -0.009 |
Beat | -0.008 |
Fiona | -0.008 |
Token | little |
Token ID | 1310 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
Ivy | +0.068 |
Iris | +0.065 |
Holly | +0.062 |
ner | +0.062 |
Beat | +0.062 |
Neg logit contributions | |
favorite | -0.028 |
o | -0.028 |
bie | -0.026 |
bright | -0.022 |
eps | -0.021 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
Ivy | +0.092 |
Iris | +0.086 |
ner | +0.083 |
Holly | +0.082 |
Beat | +0.082 |
Neg logit contributions | |
o | -0.074 |
favorite | -0.073 |
bie | -0.070 |
bright | -0.062 |
bo | -0.061 |
Token | named |
Token ID | 3706 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
Ivy | +0.103 |
Iris | +0.099 |
Holly | +0.095 |
Beat | +0.095 |
ner | +0.094 |
Neg logit contributions | |
o | -0.039 |
favorite | -0.033 |
bie | -0.031 |
bo | -0.025 |
bright | -0.023 |
Token | Bun |
Token ID | 28515 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
Ivy | +0.051 |
Iris | +0.048 |
ner | +0.045 |
ion | +0.043 |
Beat | +0.042 |
Neg logit contributions | |
o | -0.079 |
favorite | -0.075 |
bie | -0.074 |
bo | -0.067 |
bright | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
Ivy | +0.325 |
Iris | +0.310 |
Holly | +0.296 |
Beat | +0.294 |
ner | +0.293 |
Neg logit contributions | |
o | -0.175 |
favorite | -0.167 |
bie | -0.162 |
bo | -0.138 |
eps | -0.136 |
Token | |
Token ID | 220 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
Ivy | +0.030 |
ion | +0.029 |
policy | +0.029 |
Iris | +0.029 |
ner | +0.029 |
Neg logit contributions | |
o | -0.017 |
favorite | -0.016 |
bie | -0.014 |
bright | -0.014 |
big | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -7.23e-03 |
Pos logit contributions | |
favorite | +0.018 |
bie | +0.017 |
o | +0.017 |
eps | +0.015 |
tallest | +0.014 |
Neg logit contributions | |
Ivy | -0.036 |
Iris | -0.034 |
Holly | -0.033 |
Beat | -0.033 |
ner | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -2.45e-03 |
Pos logit contributions | |
Ivy | +0.019 |
Iris | +0.019 |
Holly | +0.018 |
Beat | +0.018 |
ner | +0.018 |
Neg logit contributions | |
o | -0.006 |
favorite | -0.006 |
bie | -0.005 |
eps | -0.004 |
bo | -0.004 |
Token | B |
Token ID | 33 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
Ivy | +0.005 |
Iris | +0.005 |
otion | +0.005 |
Holly | +0.005 |
ion | +0.005 |
Neg logit contributions | |
o | -0.003 |
favorite | -0.003 |
bo | -0.003 |
bright | -0.003 |
big | -0.003 |
Token | finger |
Token ID | 7660 |
Feature activation | -3.69e-03 |
Pos logit contributions | |
o | +0.005 |
bie | +0.004 |
eps | +0.004 |
favorite | +0.004 |
v | +0.004 |
Neg logit contributions | |
Ivy | -0.007 |
Holly | -0.006 |
Iris | -0.006 |
ion | -0.006 |
Fiona | -0.006 |
Token | in |
Token ID | 287 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
Ivy | +0.008 |
Iris | +0.008 |
Holly | +0.008 |
ner | +0.007 |
Beat | +0.007 |
Neg logit contributions | |
o | -0.004 |
favorite | -0.004 |
bie | -0.004 |
eps | -0.004 |
v | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
Ivy | +0.072 |
Iris | +0.069 |
Holly | +0.066 |
Beat | +0.065 |
ner | +0.064 |
Neg logit contributions | |
favorite | -0.048 |
o | -0.047 |
bie | -0.046 |
eps | -0.040 |
bo | -0.039 |
Token | air |
Token ID | 1633 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
o | +0.036 |
favorite | +0.034 |
bie | +0.034 |
eps | +0.030 |
v | +0.030 |
Neg logit contributions | |
Ivy | -0.043 |
Iris | -0.040 |
Holly | -0.038 |
Beat | -0.037 |
ner | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -3.28e-03 |
Pos logit contributions | |
favorite | +0.014 |
o | +0.014 |
bie | +0.013 |
eps | +0.011 |
bo | +0.010 |
Neg logit contributions | |
Ivy | -0.033 |
Iris | -0.031 |
Holly | -0.030 |
Beat | -0.030 |
ner | -0.030 |
Token | Beat |
Token ID | 12568 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
Ivy | +0.006 |
Iris | +0.006 |
Holly | +0.006 |
Beat | +0.006 |
ner | +0.006 |
Neg logit contributions | |
o | -0.005 |
favorite | -0.005 |
bie | -0.005 |
eps | -0.004 |
bo | -0.004 |
Token | rice |
Token ID | 20970 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
Ivy | +0.243 |
Iris | +0.221 |
Holly | +0.212 |
ner | +0.209 |
Fiona | +0.202 |
Neg logit contributions | |
favorite | -0.245 |
o | -0.244 |
bie | -0.229 |
bo | -0.209 |
eps | -0.205 |
Token | did |
Token ID | 750 |
Feature activation | +3.97e-03 |
Pos logit contributions | |
Ivy | +0.043 |
Iris | +0.042 |
Beat | +0.040 |
Holly | +0.040 |
ner | +0.039 |
Neg logit contributions | |
o | -0.022 |
favorite | -0.021 |
bie | -0.021 |
bo | -0.017 |
eps | -0.017 |
Token | as |
Token ID | 355 |
Feature activation | +5.61e-02 |
Pos logit contributions | |
o | +0.005 |
favorite | +0.005 |
bie | +0.005 |
eps | +0.004 |
bo | +0.004 |
Neg logit contributions | |
Ivy | -0.009 |
Iris | -0.008 |
Holly | -0.008 |
Beat | -0.008 |
ner | -0.008 |
Token | her |
Token ID | 607 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
favorite | +0.100 |
o | +0.099 |
bie | +0.096 |
eps | +0.086 |
bright | +0.085 |
Neg logit contributions | |
Ivy | -0.096 |
Iris | -0.089 |
Holly | -0.084 |
ner | -0.083 |
Beat | -0.081 |
Token | mom |
Token ID | 1995 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
o | +0.034 |
bie | +0.030 |
eps | +0.029 |
ams | +0.029 |
vice | +0.029 |
Neg logit contributions | |
Ivy | -0.047 |
Iris | -0.043 |
Holly | -0.043 |
ner | -0.041 |
Fiona | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +1.69e-03 |
Pos logit contributions | |
bie | +0.011 |
o | +0.011 |
favorite | +0.011 |
eps | +0.009 |
bo | +0.009 |
Neg logit contributions | |
Ivy | -0.025 |
Iris | -0.023 |
Holly | -0.023 |
ner | -0.022 |
Beat | -0.022 |
Token | They |
Token ID | 1119 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
eps | +0.003 |
bo | +0.003 |
o | +0.003 |
bie | +0.003 |
favorite | +0.003 |
Neg logit contributions | |
Ivy | -0.002 |
Holly | -0.002 |
Iris | -0.002 |
Beat | -0.002 |
Wendy | -0.002 |
Token | found |
Token ID | 1043 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
Ivy | +0.062 |
Holly | +0.058 |
Iris | +0.058 |
Beat | +0.056 |
Fiona | +0.056 |
Neg logit contributions | |
favorite | -0.033 |
o | -0.032 |
bie | -0.031 |
bo | -0.029 |
ams | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
Ivy | +0.059 |
Iris | +0.057 |
Holly | +0.055 |
Beat | +0.054 |
Fiona | +0.053 |
Neg logit contributions | |
o | -0.026 |
bie | -0.025 |
favorite | -0.025 |
eps | -0.021 |
bo | -0.021 |
Token | big |
Token ID | 1263 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
Ivy | +0.081 |
Iris | +0.077 |
Holly | +0.075 |
ner | +0.075 |
Beat | +0.074 |
Neg logit contributions | |
favorite | -0.019 |
o | -0.017 |
bie | -0.016 |
bright | -0.012 |
eps | -0.011 |
Token | shelter |
Token ID | 11772 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
favorite | +0.023 |
bright | +0.021 |
black | +0.020 |
big | +0.020 |
o | +0.020 |
Neg logit contributions | |
Ivy | -0.043 |
Iris | -0.040 |
ner | -0.040 |
Beat | -0.039 |
Holly | -0.039 |
Token | made |
Token ID | 925 |
Feature activation | -6.93e-03 |
Pos logit contributions | |
Ivy | +0.326 |
Iris | +0.309 |
Holly | +0.299 |
Beat | +0.294 |
ner | +0.292 |
Neg logit contributions | |
favorite | -0.163 |
o | -0.163 |
bie | -0.154 |
eps | -0.132 |
bo | -0.127 |
Token | of |
Token ID | 286 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
Ivy | +0.015 |
Iris | +0.014 |
Holly | +0.013 |
Beat | +0.013 |
ner | +0.013 |
Neg logit contributions | |
o | -0.010 |
favorite | -0.010 |
bie | -0.009 |
eps | -0.008 |
bright | -0.008 |
Token | sticks |
Token ID | 16461 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
Ivy | +0.099 |
ner | +0.099 |
Iris | +0.094 |
Holly | +0.089 |
Beat | +0.087 |
Neg logit contributions | |
favorite | -0.052 |
bright | -0.051 |
big | -0.050 |
hot | -0.048 |
black | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
Ivy | +0.048 |
Iris | +0.047 |
Beat | +0.044 |
Holly | +0.044 |
ner | +0.043 |
Neg logit contributions | |
favorite | -0.018 |
o | -0.018 |
bie | -0.015 |
bright | -0.014 |
big | -0.013 |
Token | leaves |
Token ID | 5667 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
Ivy | +0.054 |
Iris | +0.051 |
Holly | +0.049 |
ner | +0.048 |
Beat | +0.048 |
Neg logit contributions | |
favorite | -0.039 |
o | -0.037 |
bie | -0.035 |
bright | -0.034 |
eps | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -5.75e-03 |
Pos logit contributions | |
favorite | +0.003 |
o | +0.002 |
bie | +0.002 |
eps | +0.002 |
bright | +0.002 |
Neg logit contributions | |
Ivy | -0.006 |
Iris | -0.005 |
Holly | -0.005 |
Beat | -0.005 |
ner | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -9.50e-04 |
Pos logit contributions | |
Ivy | +0.014 |
Iris | +0.014 |
ner | +0.013 |
Holly | +0.013 |
Beat | +0.013 |
Neg logit contributions | |
o | -0.006 |
favorite | -0.006 |
bie | -0.005 |
bo | -0.004 |
bright | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +4.30e-03 |
Pos logit contributions | |
Ivy | +0.002 |
Iris | +0.002 |
Holly | +0.002 |
Beat | +0.002 |
ner | +0.002 |
Neg logit contributions | |
o | -0.001 |
favorite | -0.001 |
bie | -0.001 |
eps | -0.001 |
bo | -0.001 |
Token | Then |
Token ID | 6423 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
o | +0.006 |
favorite | +0.006 |
bie | +0.005 |
eps | +0.005 |
bo | +0.005 |
Neg logit contributions | |
Ivy | -0.009 |
Holly | -0.008 |
Iris | -0.008 |
ner | -0.008 |
Beat | -0.008 |
Token | something |
Token ID | 1223 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
favorite | +0.044 |
o | +0.042 |
bie | +0.040 |
bright | +0.036 |
favourite | +0.033 |
Neg logit contributions | |
Ivy | -0.074 |
Iris | -0.070 |
ner | -0.069 |
Holly | -0.068 |
Beat | -0.067 |
Token | worse |
Token ID | 4785 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
Ivy | +0.056 |
Iris | +0.053 |
ner | +0.053 |
Holly | +0.052 |
Beat | +0.052 |
Neg logit contributions | |
favorite | -0.028 |
o | -0.025 |
bright | -0.022 |
bie | -0.022 |
eps | -0.022 |
Token | happened |
Token ID | 3022 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
Ivy | +0.327 |
Iris | +0.309 |
ner | +0.306 |
Holly | +0.298 |
Fiona | +0.281 |
Neg logit contributions | |
o | -0.156 |
favorite | -0.156 |
saw | -0.136 |
at | -0.134 |
bie | -0.129 |
Token | . |
Token ID | 13 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
Ivy | +0.047 |
Iris | +0.045 |
Holly | +0.044 |
Beat | +0.043 |
ner | +0.043 |
Neg logit contributions | |
favorite | -0.017 |
o | -0.017 |
bie | -0.016 |
eps | -0.013 |
bo | -0.013 |
Token | Daddy |
Token ID | 26926 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
favorite | +0.008 |
o | +0.008 |
bie | +0.007 |
bright | +0.007 |
hot | +0.007 |
Neg logit contributions | |
Ivy | -0.010 |
Iris | -0.010 |
Holly | -0.010 |
Beat | -0.010 |
ner | -0.010 |
Token | left |
Token ID | 1364 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
o | +0.032 |
favorite | +0.030 |
bie | +0.030 |
eps | +0.026 |
bo | +0.025 |
Neg logit contributions | |
Ivy | -0.068 |
Iris | -0.064 |
Holly | -0.062 |
Beat | -0.061 |
ner | -0.060 |
Token | the |
Token ID | 262 |
Feature activation | -7.74e-03 |
Pos logit contributions | |
Ivy | +0.077 |
Iris | +0.074 |
Beat | +0.072 |
ner | +0.071 |
Holly | +0.071 |
Neg logit contributions | |
favorite | -0.058 |
o | -0.056 |
bie | -0.054 |
bright | -0.050 |
hot | -0.048 |
Token | believe |
Token ID | 1975 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
Ivy | +0.059 |
Iris | +0.056 |
Holly | +0.054 |
ner | +0.053 |
Beat | +0.053 |
Neg logit contributions | |
o | -0.036 |
favorite | -0.036 |
bie | -0.034 |
eps | -0.030 |
bo | -0.030 |
Token | him |
Token ID | 683 |
Feature activation | -1.79e-03 |
Pos logit contributions | |
favorite | +0.017 |
o | +0.017 |
bie | +0.016 |
eps | +0.015 |
bo | +0.014 |
Neg logit contributions | |
Ivy | -0.020 |
Iris | -0.018 |
Holly | -0.018 |
Beat | -0.017 |
ner | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -3.42e-03 |
Pos logit contributions | |
Ivy | +0.004 |
Iris | +0.004 |
Holly | +0.004 |
Beat | +0.004 |
ner | +0.004 |
Neg logit contributions | |
o | -0.002 |
favorite | -0.002 |
bie | -0.002 |
eps | -0.001 |
v | -0.001 |
Token | So |
Token ID | 1406 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
Ivy | +0.008 |
Iris | +0.007 |
Holly | +0.007 |
ner | +0.007 |
Beat | +0.007 |
Neg logit contributions | |
o | -0.004 |
favorite | -0.004 |
bie | -0.004 |
bo | -0.004 |
eps | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
Ivy | +0.074 |
Iris | +0.069 |
Holly | +0.067 |
Beat | +0.066 |
ner | +0.064 |
Neg logit contributions | |
favorite | -0.055 |
o | -0.055 |
bie | -0.053 |
eps | -0.049 |
v | -0.047 |
Token | Woo |
Token ID | 39832 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
Ivy | +0.080 |
Iris | +0.075 |
Holly | +0.072 |
Beat | +0.071 |
ner | +0.070 |
Neg logit contributions | |
favorite | -0.059 |
o | -0.059 |
bie | -0.057 |
eps | -0.052 |
bo | -0.050 |
Token | ly |
Token ID | 306 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
Ivy | +0.276 |
Iris | +0.261 |
ner | +0.249 |
Holly | +0.240 |
Fiona | +0.239 |
Neg logit contributions | |
favorite | -0.219 |
o | -0.209 |
bie | -0.198 |
bo | -0.189 |
ams | -0.170 |
Token | decided |
Token ID | 3066 |
Feature activation | -2.41e-03 |
Pos logit contributions | |
Ivy | +0.044 |
Iris | +0.042 |
ner | +0.041 |
Holly | +0.040 |
Beat | +0.040 |
Neg logit contributions | |
o | -0.021 |
favorite | -0.020 |
bie | -0.020 |
bo | -0.016 |
eps | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
Ivy | +0.007 |
Iris | +0.006 |
Holly | +0.006 |
ner | +0.006 |
Beat | +0.006 |
Neg logit contributions | |
favorite | -0.002 |
o | -0.002 |
bie | -0.001 |
bright | -0.001 |
drawn | -0.001 |
Token | be |
Token ID | 307 |
Feature activation | -5.54e-03 |
Pos logit contributions | |
Ivy | +0.046 |
Iris | +0.044 |
ner | +0.044 |
Holly | +0.042 |
Beat | +0.042 |
Neg logit contributions | |
favorite | -0.028 |
o | -0.028 |
bie | -0.026 |
bright | -0.025 |
bo | -0.023 |
Token | brave |
Token ID | 14802 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
Ivy | +0.013 |
Iris | +0.013 |
Holly | +0.012 |
Beat | +0.012 |
Fiona | +0.012 |
Neg logit contributions | |
bie | -0.006 |
o | -0.006 |
favorite | -0.006 |
eps | -0.005 |
bo | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | -8.98e-03 |
Pos logit contributions | |
o | +0.009 |
favorite | +0.009 |
bie | +0.009 |
eps | +0.007 |
bo | +0.007 |
Neg logit contributions | |
Ivy | -0.015 |
Iris | -0.014 |
Holly | -0.013 |
Beat | -0.013 |
ner | -0.013 |
Token | gone |
Token ID | 3750 |
Feature activation | -6.55e-02 |
Pos logit contributions | |
Ivy | +0.020 |
Iris | +0.019 |
Beat | +0.018 |
Holly | +0.018 |
Fiona | +0.018 |
Neg logit contributions | |
o | -0.011 |
bie | -0.011 |
favorite | -0.011 |
v | -0.009 |
bo | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
Ivy | +0.149 |
Iris | +0.141 |
Holly | +0.136 |
ner | +0.134 |
Beat | +0.132 |
Neg logit contributions | |
favorite | -0.083 |
o | -0.080 |
bie | -0.077 |
bright | -0.069 |
eps | -0.067 |
Token | everything |
Token ID | 2279 |
Feature activation | -8.80e-03 |
Pos logit contributions | |
favorite | +0.027 |
o | +0.027 |
bie | +0.026 |
eps | +0.023 |
bo | +0.022 |
Neg logit contributions | |
Ivy | -0.040 |
Iris | -0.038 |
Holly | -0.036 |
Beat | -0.036 |
ner | -0.035 |
Token | felt |
Token ID | 2936 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
Ivy | +0.020 |
Iris | +0.019 |
Holly | +0.018 |
Beat | +0.018 |
ner | +0.017 |
Neg logit contributions | |
favorite | -0.011 |
o | -0.011 |
bie | -0.011 |
eps | -0.009 |
bo | -0.009 |
Token | worse |
Token ID | 4785 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
Ivy | +0.049 |
Iris | +0.047 |
Holly | +0.045 |
Beat | +0.045 |
ner | +0.044 |
Neg logit contributions | |
o | -0.021 |
favorite | -0.021 |
bie | -0.020 |
eps | -0.017 |
bo | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
Ivy | +0.346 |
Iris | +0.330 |
Holly | +0.320 |
Beat | +0.316 |
ner | +0.313 |
Neg logit contributions | |
o | -0.138 |
favorite | -0.137 |
bie | -0.130 |
eps | -0.109 |
bo | -0.104 |
Token | The |
Token ID | 383 |
Feature activation | -8.60e-03 |
Pos logit contributions | |
favorite | +0.012 |
o | +0.012 |
bie | +0.012 |
eps | +0.010 |
bo | +0.010 |
Neg logit contributions | |
Ivy | -0.023 |
Iris | -0.022 |
Holly | -0.021 |
Beat | -0.021 |
ner | -0.021 |
Token | bad |
Token ID | 2089 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
Ivy | +0.016 |
Iris | +0.015 |
Holly | +0.014 |
Beat | +0.014 |
ner | +0.014 |
Neg logit contributions | |
o | -0.015 |
bie | -0.014 |
favorite | -0.014 |
eps | -0.012 |
bo | -0.012 |
Token | ending |
Token ID | 7464 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
Ivy | +0.099 |
Iris | +0.091 |
Holly | +0.090 |
ner | +0.086 |
Beat | +0.084 |
Neg logit contributions | |
favorite | -0.087 |
o | -0.082 |
bie | -0.081 |
bright | -0.075 |
eps | -0.073 |
Token | was |
Token ID | 373 |
Feature activation | -6.66e-04 |
Pos logit contributions | |
Ivy | +0.140 |
Iris | +0.134 |
Holly | +0.127 |
ner | +0.126 |
Beat | +0.124 |
Neg logit contributions | |
o | -0.057 |
favorite | -0.055 |
bie | -0.049 |
bright | -0.043 |
saw | -0.042 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
bie | +0.018 |
o | +0.017 |
favorite | +0.017 |
v | +0.014 |
ams | +0.014 |
Neg logit contributions | |
Ivy | -0.028 |
Iris | -0.027 |
Holly | -0.025 |
Beat | -0.025 |
ner | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
Ivy | +0.045 |
Iris | +0.043 |
Holly | +0.041 |
Beat | +0.041 |
ner | +0.041 |
Neg logit contributions | |
favorite | -0.013 |
o | -0.013 |
bie | -0.012 |
eps | -0.009 |
bo | -0.009 |
Token | They |
Token ID | 1119 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
favorite | +0.027 |
o | +0.024 |
bie | +0.023 |
bright | +0.021 |
big | +0.020 |
Neg logit contributions | |
Ivy | -0.034 |
ner | -0.033 |
Iris | -0.032 |
Holly | -0.030 |
Beat | -0.029 |
Token | did |
Token ID | 750 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
Ivy | +0.031 |
Iris | +0.029 |
Beat | +0.029 |
Holly | +0.028 |
Fiona | +0.027 |
Neg logit contributions | |
o | -0.019 |
bie | -0.019 |
favorite | -0.019 |
ams | -0.016 |
tallest | -0.016 |
Token | not |
Token ID | 407 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
favorite | +0.026 |
o | +0.026 |
bie | +0.025 |
eps | +0.022 |
bo | +0.021 |
Neg logit contributions | |
Ivy | -0.045 |
Iris | -0.042 |
Holly | -0.041 |
Beat | -0.040 |
ner | -0.040 |
Token | like |
Token ID | 588 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
favorite | +0.072 |
bie | +0.069 |
o | +0.068 |
tallest | +0.061 |
favourite | +0.060 |
Neg logit contributions | |
Ivy | -0.099 |
Iris | -0.093 |
Beat | -0.089 |
Holly | -0.088 |
Fiona | -0.084 |
Token | camp |
Token ID | 1413 |
Feature activation | +3.56e-04 |
Pos logit contributions | |
favorite | +0.163 |
o | +0.162 |
bie | +0.155 |
eps | +0.136 |
bo | +0.131 |
Neg logit contributions | |
Ivy | -0.261 |
Iris | -0.247 |
Holly | -0.237 |
Beat | -0.234 |
ner | -0.233 |
Token | . |
Token ID | 13 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
o | +0.000 |
favorite | +0.000 |
bie | +0.000 |
eps | +0.000 |
bo | +0.000 |
Neg logit contributions | |
Ivy | -0.001 |
Iris | -0.001 |
Holly | -0.001 |
Beat | -0.001 |
ner | -0.001 |
Token | They |
Token ID | 1119 |
Feature activation | -9.29e-03 |
Pos logit contributions | |
favorite | +0.039 |
o | +0.038 |
bie | +0.036 |
eps | +0.033 |
bo | +0.032 |
Neg logit contributions | |
Ivy | -0.039 |
Iris | -0.036 |
ner | -0.035 |
Holly | -0.035 |
Beat | -0.034 |
Token | did |
Token ID | 750 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
Ivy | +0.020 |
Iris | +0.019 |
Beat | +0.018 |
Holly | +0.018 |
Fiona | +0.017 |
Neg logit contributions | |
o | -0.012 |
bie | -0.012 |
favorite | -0.012 |
ams | -0.010 |
tallest | -0.010 |
Token | not |
Token ID | 407 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
o | +0.031 |
favorite | +0.030 |
bie | +0.030 |
eps | +0.026 |
v | +0.025 |
Neg logit contributions | |
Ivy | -0.046 |
Iris | -0.044 |
Holly | -0.042 |
Beat | -0.041 |
ner | -0.040 |
Token | thankful |
Token ID | 25535 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
Ivy | +0.006 |
Iris | +0.006 |
Holly | +0.005 |
ner | +0.005 |
Beat | +0.005 |
Neg logit contributions | |
favorite | -0.003 |
o | -0.003 |
bie | -0.003 |
pter | -0.002 |
bo | -0.002 |
Token | for |
Token ID | 329 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
favorite | +0.014 |
o | +0.014 |
bie | +0.014 |
eps | +0.012 |
bo | +0.012 |
Neg logit contributions | |
Ivy | -0.024 |
Iris | -0.023 |
Holly | -0.022 |
Beat | -0.022 |
ner | -0.021 |
Token | Sally |
Token ID | 25737 |
Feature activation | +6.68e-03 |
Pos logit contributions | |
Ivy | +0.025 |
Iris | +0.024 |
ner | +0.023 |
Holly | +0.023 |
Beat | +0.022 |
Neg logit contributions | |
favorite | -0.016 |
o | -0.015 |
bie | -0.014 |
bright | -0.013 |
eps | -0.013 |
Token | â |
Token ID | 22940 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
favorite | +0.007 |
o | +0.007 |
bie | +0.006 |
eps | +0.006 |
bo | +0.005 |
Neg logit contributions | |
Ivy | -0.017 |
Iris | -0.016 |
Holly | -0.015 |
Beat | -0.015 |
ner | -0.015 |
Token | € |
Token ID | 26391 |
Feature activation | -1.54e-03 |
Pos logit contributions | |
Ivy | +0.074 |
Iris | +0.069 |
Holly | +0.067 |
Beat | +0.065 |
ner | +0.065 |
Neg logit contributions | |
o | -0.060 |
favorite | -0.060 |
bie | -0.057 |
bo | -0.052 |
eps | -0.051 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
Ivy | +0.003 |
Beat | +0.003 |
Iris | +0.003 |
ner | +0.002 |
Holly | +0.002 |
Neg logit contributions | |
favorite | -0.002 |
bie | -0.002 |
o | -0.002 |
bo | -0.002 |
vice | -0.002 |
Token | s |
Token ID | 82 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
favorite | +0.138 |
o | +0.127 |
bie | +0.127 |
eps | +0.119 |
vice | +0.110 |
Neg logit contributions | |
Ivy | -0.216 |
Iris | -0.201 |
Holly | -0.198 |
ner | -0.193 |
Beat | -0.191 |
Token | enormous |
Token ID | 9812 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
o | +0.030 |
favorite | +0.030 |
bie | +0.029 |
eps | +0.027 |
bo | +0.026 |
Neg logit contributions | |
Ivy | -0.035 |
Iris | -0.033 |
Holly | -0.031 |
Beat | -0.031 |
ner | -0.030 |
Token | gift |
Token ID | 6979 |
Feature activation | -5.55e-02 |
Pos logit contributions | |
favorite | +0.051 |
o | +0.047 |
bie | +0.045 |
bright | +0.043 |
tallest | +0.040 |
Neg logit contributions | |
Ivy | -0.070 |
Iris | -0.065 |
Holly | -0.063 |
ner | -0.061 |
Beat | -0.060 |
Token | of |
Token ID | 286 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
Ivy | +0.148 |
Iris | +0.143 |
Holly | +0.136 |
ner | +0.136 |
Beat | +0.136 |
Neg logit contributions | |
favorite | -0.047 |
o | -0.046 |
bie | -0.040 |
eps | -0.031 |
bright | -0.031 |
Token | lasers |
Token ID | 34628 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
Ivy | +0.038 |
Iris | +0.036 |
ner | +0.035 |
Holly | +0.034 |
Beat | +0.034 |
Neg logit contributions | |
favorite | -0.022 |
o | -0.021 |
bie | -0.020 |
bright | -0.018 |
eps | -0.018 |
Token | He |
Token ID | 679 |
Feature activation | +2.92e-03 |
Pos logit contributions | |
Ivy | +0.004 |
Iris | +0.004 |
Holly | +0.003 |
Beat | +0.003 |
ner | +0.003 |
Neg logit contributions | |
favorite | -0.002 |
o | -0.002 |
bie | -0.002 |
eps | -0.002 |
bo | -0.001 |
Token | wished |
Token ID | 16555 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
o | +0.004 |
favorite | +0.004 |
bie | +0.004 |
tallest | +0.003 |
eps | +0.003 |
Neg logit contributions | |
Ivy | -0.006 |
Iris | -0.006 |
Holly | -0.006 |
Beat | -0.006 |
ner | -0.005 |
Token | he |
Token ID | 339 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
Ivy | +0.100 |
Iris | +0.094 |
Holly | +0.090 |
Beat | +0.089 |
ner | +0.088 |
Neg logit contributions | |
favorite | -0.071 |
o | -0.071 |
bie | -0.068 |
eps | -0.061 |
bo | -0.059 |
Token | had |
Token ID | 550 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
favorite | +0.026 |
o | +0.026 |
bie | +0.024 |
eps | +0.021 |
bo | +0.020 |
Neg logit contributions | |
Ivy | -0.056 |
Iris | -0.054 |
Holly | -0.052 |
Beat | -0.051 |
ner | -0.051 |
Token | not |
Token ID | 407 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
bie | +0.018 |
o | +0.018 |
favorite | +0.017 |
tallest | +0.016 |
int | +0.016 |
Neg logit contributions | |
Ivy | -0.025 |
Iris | -0.023 |
Holly | -0.023 |
Beat | -0.021 |
Fiona | -0.021 |
Token | drawn |
Token ID | 7428 |
Feature activation | +9.58e-02 |
Pos logit contributions | |
favorite | +0.041 |
bie | +0.040 |
o | +0.039 |
eps | +0.034 |
bo | +0.034 |
Neg logit contributions | |
Ivy | -0.067 |
Iris | -0.063 |
Holly | -0.061 |
ner | -0.060 |
Beat | -0.059 |
Token | on |
Token ID | 319 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
bie | +0.150 |
o | +0.148 |
favorite | +0.141 |
v | +0.129 |
eps | +0.128 |
Neg logit contributions | |
Ivy | -0.177 |
Iris | -0.167 |
Holly | -0.162 |
ner | -0.159 |
ion | -0.152 |
Token | the |
Token ID | 262 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
o | +0.027 |
bie | +0.026 |
favorite | +0.025 |
eps | +0.023 |
int | +0.023 |
Neg logit contributions | |
Ivy | -0.030 |
Iris | -0.028 |
Holly | -0.027 |
Beat | -0.026 |
ner | -0.026 |
Token | wall |
Token ID | 3355 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
o | +0.077 |
bie | +0.074 |
favorite | +0.069 |
ams | +0.065 |
bo | +0.064 |
Neg logit contributions | |
Ivy | -0.066 |
Iris | -0.062 |
ner | -0.060 |
Holly | -0.059 |
Beat | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +3.91e-03 |
Pos logit contributions | |
bie | +0.061 |
o | +0.059 |
favorite | +0.059 |
eps | +0.052 |
favorites | +0.048 |
Neg logit contributions | |
Ivy | -0.094 |
ner | -0.088 |
Iris | -0.087 |
Holly | -0.086 |
Beat | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | +8.48e-03 |
Pos logit contributions | |
bie | +0.005 |
favorite | +0.005 |
o | +0.005 |
eps | +0.005 |
ams | +0.005 |
Neg logit contributions | |
Ivy | -0.008 |
Iris | -0.008 |
Holly | -0.008 |
Beat | -0.008 |
Fiona | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +3.96e-04 |
Pos logit contributions | |
Ivy | +0.005 |
Iris | +0.004 |
Holly | +0.004 |
Beat | +0.004 |
Fiona | +0.004 |
Neg logit contributions | |
o | -0.002 |
favorite | -0.002 |
bie | -0.002 |
eps | -0.002 |
bo | -0.002 |
Token | but |
Token ID | 475 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
favorite | +0.001 |
o | +0.001 |
bie | +0.001 |
eps | +0.000 |
bo | +0.000 |
Neg logit contributions | |
Ivy | -0.001 |
Iris | -0.001 |
Holly | -0.001 |
Beat | -0.001 |
ner | -0.001 |
Token | it |
Token ID | 340 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
favorite | +0.055 |
o | +0.055 |
bie | +0.053 |
eps | +0.047 |
bo | +0.045 |
Neg logit contributions | |
Ivy | -0.083 |
Iris | -0.078 |
Holly | -0.075 |
Beat | -0.074 |
ner | -0.073 |
Token | was |
Token ID | 373 |
Feature activation | +9.86e-03 |
Pos logit contributions | |
Ivy | +0.035 |
Iris | +0.032 |
Holly | +0.032 |
Beat | +0.031 |
Fiona | +0.030 |
Neg logit contributions | |
favorite | -0.024 |
o | -0.024 |
bie | -0.024 |
eps | -0.021 |
ams | -0.021 |
Token | deeper |
Token ID | 9211 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
favorite | +0.011 |
o | +0.011 |
bie | +0.010 |
eps | +0.009 |
bo | +0.008 |
Neg logit contributions | |
Ivy | -0.024 |
Iris | -0.023 |
Holly | -0.022 |
Beat | -0.022 |
ner | -0.022 |
Token | than |
Token ID | 621 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
Ivy | +0.163 |
Iris | +0.153 |
Holly | +0.149 |
Beat | +0.148 |
Fiona | +0.145 |
Neg logit contributions | |
o | -0.110 |
favorite | -0.106 |
bie | -0.105 |
eps | -0.093 |
bo | -0.092 |
Token | he |
Token ID | 339 |
Feature activation | +7.49e-02 |
Pos logit contributions | |
o | +0.175 |
favorite | +0.175 |
bie | +0.170 |
eps | +0.152 |
v | +0.148 |
Neg logit contributions | |
Ivy | -0.254 |
Iris | -0.237 |
Holly | -0.232 |
Beat | -0.226 |
ner | -0.226 |
Token | thought |
Token ID | 1807 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
favorite | +0.104 |
bie | +0.100 |
o | +0.097 |
ams | +0.087 |
eps | +0.086 |
Neg logit contributions | |
Ivy | -0.154 |
Holly | -0.143 |
Iris | -0.140 |
Beat | -0.137 |
ner | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | +6.86e-03 |
Pos logit contributions | |
favorite | +0.044 |
o | +0.044 |
bie | +0.042 |
eps | +0.036 |
bo | +0.035 |
Neg logit contributions | |
Ivy | -0.086 |
Iris | -0.082 |
Holly | -0.079 |
Beat | -0.078 |
ner | -0.077 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
favorite | +0.009 |
o | +0.009 |
bie | +0.009 |
eps | +0.008 |
v | +0.007 |
Neg logit contributions | |
Ivy | -0.015 |
Iris | -0.014 |
Holly | -0.014 |
Beat | -0.014 |
ner | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +7.12e-03 |
Pos logit contributions | |
favorite | +0.022 |
o | +0.022 |
bie | +0.020 |
bright | +0.018 |
eps | +0.017 |
Neg logit contributions | |
Ivy | -0.039 |
Iris | -0.037 |
Holly | -0.035 |
Beat | -0.035 |
ner | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +5.31e-04 |
Pos logit contributions | |
favorite | +0.032 |
bie | +0.031 |
o | +0.029 |
tallest | +0.027 |
ams | +0.027 |
Neg logit contributions | |
Ivy | -0.049 |
Iris | -0.046 |
Holly | -0.045 |
Beat | -0.044 |
ner | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | +6.60e-03 |
Pos logit contributions | |
favorite | +0.001 |
o | +0.001 |
bie | +0.001 |
eps | +0.000 |
bo | +0.000 |
Neg logit contributions | |
Ivy | -0.001 |
Iris | -0.001 |
Holly | -0.001 |
Beat | -0.001 |
ner | -0.001 |
Token | Jack |
Token ID | 14295 |
Feature activation | -2.08e-03 |
Pos logit contributions | |
favorite | +0.010 |
o | +0.010 |
bie | +0.010 |
eps | +0.009 |
bo | +0.009 |
Neg logit contributions | |
Ivy | -0.013 |
Iris | -0.012 |
Holly | -0.012 |
Beat | -0.011 |
ner | -0.011 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
Ivy | +0.005 |
Holly | +0.004 |
Iris | +0.004 |
Beat | +0.004 |
Fiona | +0.004 |
Neg logit contributions | |
favorite | -0.002 |
o | -0.002 |
bie | -0.002 |
tallest | -0.002 |
eps | -0.002 |
Token | 't |
Token ID | 470 |
Feature activation | -6.21e-03 |
Pos logit contributions | |
Ivy | +0.033 |
Iris | +0.032 |
Beat | +0.030 |
Holly | +0.030 |
Fiona | +0.029 |
Neg logit contributions | |
o | -0.028 |
favorite | -0.027 |
big | -0.026 |
very | -0.026 |
huge | -0.025 |
Token | like |
Token ID | 588 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Ivy | +0.016 |
Iris | +0.015 |
Holly | +0.014 |
Beat | +0.014 |
ner | +0.014 |
Neg logit contributions | |
o | -0.006 |
favorite | -0.006 |
bie | -0.006 |
eps | -0.005 |
bo | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
o | +0.190 |
bie | +0.186 |
favorite | +0.183 |
eps | +0.173 |
bo | +0.168 |
Neg logit contributions | |
Ivy | -0.210 |
Holly | -0.191 |
Iris | -0.190 |
ner | -0.183 |
Fiona | -0.178 |
Token | sauce |
Token ID | 10746 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
o | +0.072 |
eps | +0.067 |
bie | +0.065 |
bo | +0.063 |
pter | +0.063 |
Neg logit contributions | |
Ivy | -0.050 |
happen | -0.048 |
Holly | -0.047 |
ner | -0.042 |
policy | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +6.29e-03 |
Pos logit contributions | |
o | +0.068 |
favorite | +0.068 |
bie | +0.066 |
eps | +0.057 |
bo | +0.054 |
Neg logit contributions | |
Ivy | -0.125 |
Iris | -0.118 |
Holly | -0.115 |
Beat | -0.112 |
ner | -0.112 |
Token | He |
Token ID | 679 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
o | +0.009 |
bie | +0.009 |
favorite | +0.009 |
eps | +0.008 |
bo | +0.008 |
Neg logit contributions | |
Ivy | -0.013 |
Iris | -0.012 |
Holly | -0.012 |
Beat | -0.011 |
ner | -0.011 |
Token | didn |
Token ID | 1422 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
o | +0.014 |
bie | +0.014 |
favorite | +0.014 |
tallest | +0.012 |
ams | +0.012 |
Neg logit contributions | |
Ivy | -0.023 |
Holly | -0.021 |
Iris | -0.021 |
Fiona | -0.020 |
Beat | -0.020 |
Token | L |
Token ID | 43 |
Feature activation | +5.43e-03 |
Pos logit contributions | |
Ivy | +0.183 |
Iris | +0.174 |
Holly | +0.167 |
Beat | +0.165 |
ner | +0.164 |
Neg logit contributions | |
o | -0.095 |
favorite | -0.095 |
bie | -0.091 |
eps | -0.079 |
bo | -0.075 |
Token | ily |
Token ID | 813 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
o | +0.009 |
favorite | +0.009 |
bie | +0.009 |
eps | +0.008 |
v | +0.008 |
Neg logit contributions | |
Ivy | -0.010 |
Iris | -0.009 |
Beat | -0.009 |
ner | -0.009 |
Holly | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
Ivy | +0.073 |
Iris | +0.070 |
Holly | +0.067 |
ner | +0.066 |
Beat | +0.066 |
Neg logit contributions | |
o | -0.032 |
favorite | -0.032 |
bie | -0.030 |
eps | -0.025 |
bo | -0.025 |
Token | Ben |
Token ID | 3932 |
Feature activation | +5.25e-03 |
Pos logit contributions | |
favorite | +0.025 |
o | +0.025 |
bie | +0.024 |
eps | +0.022 |
bo | +0.021 |
Neg logit contributions | |
Ivy | -0.034 |
Iris | -0.032 |
Holly | -0.031 |
Beat | -0.031 |
ner | -0.030 |
Token | were |
Token ID | 547 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
o | +0.006 |
favorite | +0.006 |
bie | +0.005 |
eps | +0.004 |
bo | +0.004 |
Neg logit contributions | |
Ivy | -0.013 |
Iris | -0.012 |
Holly | -0.012 |
ner | -0.012 |
Beat | -0.012 |
Token | going |
Token ID | 1016 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
v | +0.027 |
bo | +0.027 |
vice | +0.025 |
o | +0.025 |
pter | +0.025 |
Neg logit contributions | |
Ivy | -0.034 |
Holly | -0.033 |
Iris | -0.032 |
Beat | -0.030 |
Fiona | -0.030 |
Token | on |
Token ID | 319 |
Feature activation | +5.47e-03 |
Pos logit contributions | |
bie | +0.033 |
o | +0.032 |
bo | +0.032 |
favorite | +0.031 |
eps | +0.030 |
Neg logit contributions | |
Ivy | -0.040 |
Iris | -0.040 |
Holly | -0.039 |
ner | -0.039 |
Beat | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
o | +0.006 |
bie | +0.006 |
favorite | +0.006 |
eps | +0.005 |
bo | +0.005 |
Neg logit contributions | |
Ivy | -0.013 |
Iris | -0.013 |
Holly | -0.012 |
Beat | -0.012 |
Fiona | -0.012 |
Token | trip |
Token ID | 5296 |
Feature activation | -4.21e-03 |
Pos logit contributions | |
o | +0.038 |
ams | +0.037 |
bie | +0.036 |
int | +0.035 |
aw | +0.035 |
Neg logit contributions | |
Ivy | -0.045 |
Iris | -0.043 |
Holly | -0.041 |
Beat | -0.041 |
Fiona | -0.040 |
Token | with |
Token ID | 351 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
Ivy | +0.009 |
Iris | +0.008 |
Holly | +0.008 |
ner | +0.008 |
Beat | +0.008 |
Neg logit contributions | |
bie | -0.006 |
favorite | -0.006 |
o | -0.006 |
eps | -0.005 |
bo | -0.005 |
Token | Mom |
Token ID | 11254 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
o | +0.019 |
bie | +0.018 |
favorite | +0.018 |
bo | +0.017 |
eps | +0.017 |
Neg logit contributions | |
Ivy | -0.020 |
Holly | -0.019 |
Iris | -0.019 |
Beat | -0.018 |
Fiona | -0.018 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
Ivy | +0.059 |
Iris | +0.056 |
ner | +0.055 |
Beat | +0.054 |
Holly | +0.053 |
Neg logit contributions | |
favorite | -0.042 |
o | -0.040 |
bie | -0.038 |
bright | -0.035 |
black | -0.033 |
Token | found |
Token ID | 1043 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
Ivy | +0.061 |
Holly | +0.055 |
Iris | +0.055 |
Fiona | +0.053 |
ner | +0.052 |
Neg logit contributions | |
favorite | -0.032 |
o | -0.031 |
bie | -0.030 |
eps | -0.029 |
tallest | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
Ivy | +0.033 |
Iris | +0.031 |
Beat | +0.029 |
ner | +0.029 |
Holly | +0.029 |
Neg logit contributions | |
favorite | -0.036 |
o | -0.036 |
bie | -0.034 |
eps | -0.031 |
bo | -0.030 |
Token | can |
Token ID | 460 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
Ivy | +0.065 |
Iris | +0.062 |
Holly | +0.060 |
Beat | +0.059 |
ner | +0.059 |
Neg logit contributions | |
favorite | -0.020 |
o | -0.020 |
bie | -0.018 |
eps | -0.015 |
bo | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
Ivy | +0.046 |
Iris | +0.042 |
Holly | +0.041 |
ner | +0.040 |
Beat | +0.039 |
Neg logit contributions | |
favorite | -0.038 |
o | -0.037 |
bie | -0.037 |
eps | -0.034 |
bo | -0.033 |
Token | y |
Token ID | 331 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
Ivy | +0.055 |
Iris | +0.053 |
Beat | +0.051 |
ner | +0.050 |
Holly | +0.050 |
Neg logit contributions | |
favorite | -0.033 |
o | -0.032 |
bie | -0.031 |
bright | -0.027 |
eps | -0.027 |
Token | ummy |
Token ID | 13513 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
o | +0.077 |
favorite | +0.076 |
bie | +0.074 |
eps | +0.066 |
bo | +0.066 |
Neg logit contributions | |
Ivy | -0.077 |
Iris | -0.072 |
Holly | -0.068 |
Beat | -0.068 |
ner | -0.067 |
Token | cat |
Token ID | 3797 |
Feature activation | -9.02e-03 |
Pos logit contributions | |
favorite | +0.070 |
o | +0.069 |
bie | +0.066 |
eps | +0.058 |
bo | +0.056 |
Neg logit contributions | |
Ivy | -0.104 |
Iris | -0.099 |
Holly | -0.095 |
Beat | -0.095 |
ner | -0.093 |
Token | food |
Token ID | 2057 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
Ivy | +0.022 |
Iris | +0.021 |
Holly | +0.020 |
Beat | +0.020 |
ner | +0.020 |
Neg logit contributions | |
o | -0.010 |
favorite | -0.010 |
bie | -0.009 |
eps | -0.008 |
bo | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
Ivy | +0.074 |
Iris | +0.072 |
Beat | +0.070 |
ner | +0.069 |
Holly | +0.068 |
Neg logit contributions | |
o | -0.028 |
favorite | -0.027 |
bie | -0.024 |
eps | -0.019 |
bright | -0.019 |
Token | opened |
Token ID | 4721 |
Feature activation | -8.14e-03 |
Pos logit contributions | |
Ivy | +0.027 |
Iris | +0.026 |
Beat | +0.026 |
ner | +0.025 |
Holly | +0.025 |
Neg logit contributions | |
favorite | -0.018 |
o | -0.017 |
bie | -0.017 |
bright | -0.014 |
eps | -0.014 |
Token | can |
Token ID | 460 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
Ivy | +0.014 |
ner | +0.014 |
Iris | +0.013 |
Holly | +0.013 |
arises | +0.013 |
Neg logit contributions | |
bie | -0.017 |
o | -0.016 |
favorite | -0.016 |
ams | -0.015 |
tallest | -0.015 |
Token | write |
Token ID | 3551 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
Ivy | +0.030 |
Iris | +0.028 |
ner | +0.028 |
Holly | +0.027 |
Beat | +0.026 |
Neg logit contributions | |
o | -0.026 |
bie | -0.026 |
favorite | -0.026 |
ams | -0.022 |
eps | -0.022 |
Token | my |
Token ID | 616 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
o | +0.033 |
bie | +0.033 |
favorite | +0.032 |
vice | +0.028 |
bo | +0.027 |
Neg logit contributions | |
Ivy | -0.045 |
Iris | -0.043 |
Holly | -0.041 |
ner | -0.040 |
Beat | -0.040 |
Token | name |
Token ID | 1438 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
happen | +0.051 |
Ivy | +0.051 |
Iris | +0.049 |
Holly | +0.048 |
apology | +0.048 |
Neg logit contributions | |
o | -0.067 |
bie | -0.063 |
vice | -0.060 |
v | -0.056 |
ams | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +5.41e-03 |
Pos logit contributions | |
favorite | +0.045 |
o | +0.045 |
bie | +0.045 |
eps | +0.038 |
vice | +0.036 |
Neg logit contributions | |
Ivy | -0.071 |
Iris | -0.067 |
Holly | -0.064 |
ner | -0.063 |
Beat | -0.062 |
Token | T |
Token ID | 309 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
favorite | +0.010 |
o | +0.010 |
bie | +0.010 |
eps | +0.009 |
bo | +0.009 |
Neg logit contributions | |
Ivy | -0.009 |
Iris | -0.008 |
Holly | -0.008 |
Beat | -0.008 |
ner | -0.008 |
Token | - |
Token ID | 12 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
favorite | +0.086 |
o | +0.085 |
bie | +0.083 |
eps | +0.074 |
bo | +0.071 |
Neg logit contributions | |
Ivy | -0.114 |
Iris | -0.107 |
Holly | -0.103 |
Beat | -0.102 |
ner | -0.102 |
Token | O |
Token ID | 46 |
Feature activation | -8.29e-02 |
Pos logit contributions | |
bie | +0.055 |
eps | +0.051 |
tallest | +0.051 |
sweaty | +0.051 |
favorite | +0.051 |
Neg logit contributions | |
Ivy | -0.068 |
ner | -0.066 |
ion | -0.065 |
Iris | -0.065 |
Holly | -0.063 |
Token | - |
Token ID | 12 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
ner | +0.134 |
Ivy | +0.132 |
ion | +0.123 |
Holly | +0.120 |
Iris | +0.120 |
Neg logit contributions | |
favorite | -0.131 |
bright | -0.117 |
tallest | -0.115 |
huge | -0.113 |
favourite | -0.112 |
Token | M |
Token ID | 44 |
Feature activation | +6.26e-02 |
Pos logit contributions | |
favorite | +0.045 |
bie | +0.045 |
tallest | +0.043 |
eps | +0.043 |
sweaty | +0.041 |
Neg logit contributions | |
Ivy | -0.062 |
ner | -0.060 |
ion | -0.059 |
Iris | -0.059 |
Holly | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | +5.20e-03 |
Pos logit contributions | |
favorite | +0.097 |
o | +0.095 |
bie | +0.093 |
eps | +0.087 |
vice | +0.083 |
Neg logit contributions | |
Ivy | -0.117 |
Iris | -0.109 |
Holly | -0.107 |
ner | -0.107 |
Beat | -0.103 |
Token | named |
Token ID | 3706 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
Ivy | +0.052 |
Iris | +0.049 |
Holly | +0.047 |
Beat | +0.047 |
ner | +0.045 |
Neg logit contributions | |
favorite | -0.035 |
bie | -0.032 |
o | -0.031 |
eps | -0.029 |
ams | -0.028 |
Token | Lily |
Token ID | 20037 |
Feature activation | -7.47e-02 |
Pos logit contributions | |
Ivy | +0.045 |
Iris | +0.041 |
ner | +0.040 |
Holly | +0.039 |
Beat | +0.038 |
Neg logit contributions | |
o | -0.060 |
favorite | -0.057 |
bie | -0.057 |
bo | -0.051 |
bright | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
Ivy | +0.173 |
Iris | +0.165 |
Holly | +0.159 |
Beat | +0.158 |
ner | +0.157 |
Neg logit contributions | |
favorite | -0.088 |
bie | -0.080 |
o | -0.076 |
eps | -0.070 |
ams | -0.069 |
Token | Lily |
Token ID | 20037 |
Feature activation | -7.42e-02 |
Pos logit contributions | |
Ivy | +0.036 |
Iris | +0.034 |
Holly | +0.032 |
Beat | +0.032 |
ner | +0.032 |
Neg logit contributions | |
o | -0.021 |
favorite | -0.021 |
bie | -0.020 |
eps | -0.017 |
bo | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
Ivy | +0.173 |
Iris | +0.164 |
Holly | +0.158 |
Beat | +0.156 |
ner | +0.155 |
Neg logit contributions | |
favorite | -0.087 |
o | -0.086 |
bie | -0.083 |
eps | -0.072 |
bo | -0.068 |
Token | pretending |
Token ID | 23662 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
Ivy | +0.046 |
Iris | +0.044 |
Holly | +0.043 |
Beat | +0.042 |
Fiona | +0.041 |
Neg logit contributions | |
bie | -0.024 |
favorite | -0.024 |
o | -0.023 |
ams | -0.021 |
eps | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -2.88e-03 |
Pos logit contributions | |
favorite | +0.053 |
o | +0.053 |
bie | +0.051 |
eps | +0.044 |
bo | +0.042 |
Neg logit contributions | |
Ivy | -0.110 |
Iris | -0.105 |
Holly | -0.101 |
Beat | -0.100 |
ner | -0.099 |
Token | be |
Token ID | 307 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
Ivy | +0.006 |
Iris | +0.006 |
Holly | +0.006 |
Beat | +0.005 |
ner | +0.005 |
Neg logit contributions | |
favorite | -0.004 |
o | -0.004 |
bie | -0.004 |
bo | -0.003 |
eps | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
bie | +0.019 |
favorite | +0.019 |
o | +0.019 |
eps | +0.016 |
bo | +0.015 |
Neg logit contributions | |
Ivy | -0.039 |
Iris | -0.037 |
Beat | -0.036 |
Holly | -0.036 |
Fiona | -0.035 |
Token | bird |
Token ID | 6512 |
Feature activation | -3.97e-03 |
Pos logit contributions | |
Ivy | +0.029 |
Iris | +0.027 |
Holly | +0.026 |
Beat | +0.026 |
ner | +0.025 |
Neg logit contributions | |
o | -0.023 |
favorite | -0.023 |
bie | -0.022 |
eps | -0.020 |
bo | -0.019 |
Token | too |
Token ID | 1165 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
Ivy | +0.008 |
Iris | +0.008 |
Holly | +0.007 |
Beat | +0.007 |
ner | +0.007 |
Neg logit contributions | |
bie | -0.006 |
favorite | -0.005 |
eps | -0.005 |
o | -0.005 |
vice | -0.005 |
Token | it |
Token ID | 340 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
Ivy | +0.052 |
Iris | +0.049 |
Holly | +0.046 |
ner | +0.046 |
Beat | +0.045 |
Neg logit contributions | |
favorite | -0.047 |
o | -0.045 |
bie | -0.044 |
eps | -0.040 |
bright | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +3.84e-03 |
Pos logit contributions | |
Ivy | +0.073 |
Iris | +0.069 |
Holly | +0.067 |
Beat | +0.066 |
ner | +0.066 |
Neg logit contributions | |
favorite | -0.032 |
o | -0.031 |
bie | -0.029 |
eps | -0.025 |
bo | -0.024 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -7.72e-02 |
Pos logit contributions | |
favorite | +0.004 |
o | +0.004 |
bie | +0.004 |
eps | +0.004 |
v | +0.003 |
Neg logit contributions | |
Ivy | -0.009 |
Iris | -0.009 |
Holly | -0.009 |
Beat | -0.008 |
ner | -0.008 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
Ivy | +0.182 |
Iris | +0.172 |
Holly | +0.169 |
Beat | +0.166 |
ner | +0.164 |
Neg logit contributions | |
favorite | -0.090 |
bie | -0.084 |
o | -0.084 |
eps | -0.073 |
ams | -0.069 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.40e-03 |
Pos logit contributions | |
o | +0.050 |
favorite | +0.050 |
eps | +0.048 |
v | +0.047 |
bo | +0.047 |
Neg logit contributions | |
Ivy | -0.065 |
Iris | -0.061 |
Holly | -0.061 |
Beat | -0.059 |
Fiona | -0.058 |
Token | a |
Token ID | 257 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
bie | +0.002 |
o | +0.002 |
favorite | +0.002 |
ams | +0.002 |
tallest | +0.002 |
Neg logit contributions | |
Ivy | -0.003 |
Iris | -0.003 |
Holly | -0.003 |
Beat | -0.003 |
ner | -0.003 |
Token | time |
Token ID | 640 |
Feature activation | +5.36e-02 |
Pos logit contributions | |
tallest | +0.087 |
v | +0.086 |
ams | +0.084 |
o | +0.082 |
eps | +0.080 |
Neg logit contributions | |
Ivy | -0.078 |
Iris | -0.067 |
Holly | -0.067 |
iona | -0.063 |
Beat | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
favorite | +0.080 |
o | +0.080 |
bie | +0.077 |
eps | +0.070 |
bo | +0.068 |
Neg logit contributions | |
Ivy | -0.107 |
Iris | -0.100 |
Holly | -0.097 |
Beat | -0.095 |
ner | -0.094 |
Token | there |
Token ID | 612 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
favorite | +0.018 |
o | +0.018 |
bie | +0.018 |
ams | +0.016 |
eps | +0.016 |
Neg logit contributions | |
Ivy | -0.025 |
Iris | -0.024 |
Holly | -0.023 |
Beat | -0.023 |
ner | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
Ivy | +0.033 |
Iris | +0.031 |
Holly | +0.030 |
Beat | +0.030 |
ner | +0.029 |
Neg logit contributions | |
favorite | -0.020 |
o | -0.020 |
bie | -0.019 |
eps | -0.017 |
bo | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | -9.39e-03 |
Pos logit contributions | |
o | +0.026 |
bie | +0.026 |
favorite | +0.024 |
eps | +0.021 |
bo | +0.021 |
Neg logit contributions | |
Ivy | -0.065 |
Iris | -0.061 |
Holly | -0.060 |
Beat | -0.059 |
Fiona | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -6.83e-03 |
Pos logit contributions | |
Ivy | +0.017 |
Iris | +0.016 |
Holly | +0.016 |
ner | +0.016 |
Beat | +0.016 |
Neg logit contributions | |
bie | -0.008 |
o | -0.007 |
favorite | -0.007 |
eps | -0.006 |
bo | -0.006 |
Token | It |
Token ID | 632 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
Ivy | +0.014 |
Beat | +0.014 |
Iris | +0.013 |
ner | +0.013 |
Holly | +0.013 |
Neg logit contributions | |
favorite | -0.010 |
o | -0.008 |
bie | -0.008 |
bright | -0.007 |
black | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
Ivy | +0.101 |
Iris | +0.096 |
Holly | +0.092 |
ner | +0.092 |
Beat | +0.092 |
Neg logit contributions | |
o | -0.048 |
favorite | -0.048 |
bie | -0.043 |
eps | -0.039 |
bo | -0.038 |
Token | a |
Token ID | 257 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
Ivy | +0.055 |
Iris | +0.052 |
Holly | +0.051 |
ner | +0.050 |
Beat | +0.050 |
Neg logit contributions | |
favorite | -0.016 |
o | -0.015 |
bie | -0.014 |
eps | -0.011 |
bright | -0.011 |
Token | knob |
Token ID | 36842 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
Ivy | +0.101 |
Iris | +0.096 |
Holly | +0.093 |
Beat | +0.092 |
ner | +0.092 |
Neg logit contributions | |
favorite | -0.044 |
o | -0.044 |
bie | -0.041 |
eps | -0.035 |
bo | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
favorite | +0.018 |
o | +0.018 |
bie | +0.018 |
eps | +0.015 |
bo | +0.014 |
Neg logit contributions | |
Ivy | -0.037 |
Iris | -0.035 |
Holly | -0.034 |
ner | -0.034 |
Beat | -0.034 |
Token | It |
Token ID | 632 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
Ivy | +0.022 |
Beat | +0.022 |
Iris | +0.022 |
ner | +0.021 |
Holly | +0.020 |
Neg logit contributions | |
favorite | -0.018 |
o | -0.016 |
bie | -0.016 |
aw | -0.015 |
bright | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
Ivy | +0.099 |
Iris | +0.094 |
Holly | +0.090 |
Beat | +0.089 |
ner | +0.088 |
Neg logit contributions | |
favorite | -0.057 |
o | -0.056 |
bie | -0.054 |
eps | -0.047 |
bo | -0.045 |
Token | round |
Token ID | 2835 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
Ivy | +0.055 |
Iris | +0.053 |
Holly | +0.051 |
Beat | +0.051 |
ner | +0.050 |
Neg logit contributions | |
o | -0.018 |
favorite | -0.018 |
bie | -0.017 |
eps | -0.014 |
bo | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
favorite | +0.021 |
o | +0.021 |
bie | +0.020 |
eps | +0.017 |
bo | +0.016 |
Neg logit contributions | |
Ivy | -0.038 |
Iris | -0.036 |
Holly | -0.034 |
Beat | -0.034 |
ner | -0.034 |
Token | shiny |
Token ID | 22441 |
Feature activation | -9.08e-03 |
Pos logit contributions | |
Ivy | +0.042 |
Iris | +0.040 |
Holly | +0.038 |
Beat | +0.038 |
ner | +0.038 |
Neg logit contributions | |
o | -0.017 |
favorite | -0.016 |
bie | -0.016 |
eps | -0.013 |
bo | -0.012 |
Token | different |
Token ID | 1180 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
Ivy | +0.052 |
Iris | +0.048 |
Holly | +0.047 |
ner | +0.045 |
Fiona | +0.045 |
Neg logit contributions | |
o | -0.029 |
bie | -0.028 |
favorite | -0.028 |
eps | -0.023 |
bo | -0.023 |
Token | colors |
Token ID | 7577 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
o | +0.005 |
favorite | +0.005 |
bie | +0.005 |
eps | +0.004 |
bo | +0.004 |
Neg logit contributions | |
Ivy | -0.008 |
Iris | -0.007 |
Holly | -0.007 |
ner | -0.007 |
Fiona | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
favorite | +0.040 |
o | +0.040 |
bie | +0.036 |
bright | +0.031 |
eps | +0.031 |
Neg logit contributions | |
Ivy | -0.089 |
Iris | -0.086 |
Beat | -0.083 |
Holly | -0.082 |
ner | -0.080 |
Token | patterns |
Token ID | 7572 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
Ivy | +0.043 |
Iris | +0.040 |
Holly | +0.039 |
Beat | +0.038 |
ner | +0.038 |
Neg logit contributions | |
o | -0.025 |
favorite | -0.024 |
bie | -0.024 |
eps | -0.021 |
bo | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Ivy | +0.028 |
Iris | +0.027 |
Beat | +0.026 |
Holly | +0.026 |
ner | +0.026 |
Neg logit contributions | |
favorite | -0.009 |
o | -0.009 |
bie | -0.008 |
eps | -0.007 |
v | -0.006 |
Token | They |
Token ID | 1119 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
Ivy | +0.028 |
Iris | +0.026 |
Beat | +0.025 |
ner | +0.025 |
Holly | +0.025 |
Neg logit contributions | |
favorite | -0.019 |
o | -0.018 |
bie | -0.018 |
eps | -0.015 |
bo | -0.015 |
Token | are |
Token ID | 389 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
Ivy | +0.074 |
Iris | +0.071 |
Holly | +0.068 |
Beat | +0.068 |
ner | +0.067 |
Neg logit contributions | |
favorite | -0.037 |
o | -0.036 |
bie | -0.036 |
eps | -0.031 |
bright | -0.029 |
Token | attractive |
Token ID | 10966 |
Feature activation | +1.40e-03 |
Pos logit contributions | |
Ivy | +0.050 |
Iris | +0.048 |
Holly | +0.047 |
Fiona | +0.045 |
Beat | +0.045 |
Neg logit contributions | |
o | -0.020 |
bie | -0.019 |
favorite | -0.018 |
bo | -0.017 |
v | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
o | +0.002 |
favorite | +0.002 |
bie | +0.002 |
eps | +0.001 |
bo | +0.001 |
Neg logit contributions | |
Ivy | -0.003 |
Iris | -0.003 |
Holly | -0.003 |
Beat | -0.003 |
ner | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -6.85e-03 |
Pos logit contributions | |
Ivy | +0.055 |
Iris | +0.052 |
Holly | +0.050 |
Beat | +0.050 |
ner | +0.049 |
Neg logit contributions | |
favorite | -0.023 |
o | -0.023 |
bie | -0.022 |
eps | -0.018 |
bo | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +1.14e-03 |
Pos logit contributions | |
Ivy | +0.017 |
Iris | +0.017 |
Holly | +0.016 |
Beat | +0.016 |
ner | +0.016 |
Neg logit contributions | |
o | -0.007 |
favorite | -0.007 |
bie | -0.006 |
eps | -0.005 |
bo | -0.005 |
Token | " |
Token ID | 1 |
Feature activation | -2.18e-03 |
Pos logit contributions | |
o | +0.001 |
favorite | +0.001 |
bo | +0.001 |
bie | +0.001 |
eps | +0.001 |
Neg logit contributions | |
Ivy | -0.001 |
Iris | -0.001 |
Holly | -0.001 |
ner | -0.001 |
Beat | -0.001 |
Token | I |
Token ID | 40 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
Ivy | +0.005 |
Iris | +0.004 |
Holly | +0.004 |
Beat | +0.004 |
ner | +0.004 |
Neg logit contributions | |
favorite | -0.003 |
bie | -0.003 |
o | -0.003 |
eps | -0.003 |
ams | -0.003 |
Token | 'm |
Token ID | 1101 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
Ivy | +0.040 |
Iris | +0.038 |
Holly | +0.037 |
Beat | +0.036 |
ner | +0.035 |
Neg logit contributions | |
favorite | -0.030 |
o | -0.029 |
bie | -0.029 |
eps | -0.026 |
ams | -0.026 |
Token | Tim |
Token ID | 5045 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
Ivy | +0.077 |
Iris | +0.072 |
Holly | +0.069 |
Beat | +0.068 |
ner | +0.068 |
Neg logit contributions | |
favorite | -0.062 |
o | -0.062 |
bie | -0.059 |
eps | -0.053 |
bo | -0.052 |
Token | my |
Token ID | 1820 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
Ivy | +0.089 |
Iris | +0.084 |
Holly | +0.082 |
Beat | +0.081 |
ner | +0.081 |
Neg logit contributions | |
favorite | -0.033 |
bie | -0.029 |
o | -0.028 |
eps | -0.026 |
vice | -0.024 |
Token | ," |
Token ID | 553 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
favorite | +0.048 |
bie | +0.046 |
o | +0.045 |
eps | +0.044 |
vice | +0.043 |
Neg logit contributions | |
Ivy | -0.049 |
Iris | -0.045 |
Holly | -0.043 |
ner | -0.043 |
Beat | -0.042 |
Token | said |
Token ID | 531 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
Ivy | +0.035 |
Iris | +0.033 |
Holly | +0.032 |
ner | +0.032 |
Beat | +0.031 |
Neg logit contributions | |
o | -0.019 |
favorite | -0.019 |
bie | -0.018 |
eps | -0.016 |
bo | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
o | +0.023 |
favorite | +0.023 |
bie | +0.022 |
bo | +0.020 |
eps | +0.019 |
Neg logit contributions | |
Ivy | -0.020 |
Iris | -0.019 |
ner | -0.018 |
Holly | -0.018 |
Beat | -0.018 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
Ivy | +0.050 |
Iris | +0.048 |
Holly | +0.046 |
ner | +0.045 |
Beat | +0.045 |
Neg logit contributions | |
favorite | -0.029 |
o | -0.028 |
bie | -0.026 |
eps | -0.023 |
bo | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -7.81e-03 |
Pos logit contributions | |
Ivy | +0.029 |
Iris | +0.028 |
Holly | +0.027 |
Beat | +0.026 |
Fiona | +0.026 |
Neg logit contributions | |
favorite | -0.017 |
eps | -0.017 |
bie | -0.017 |
tallest | -0.017 |
ams | -0.015 |
Token | " |
Token ID | 366 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
Ivy | +0.019 |
Iris | +0.018 |
Holly | +0.017 |
Beat | +0.017 |
ner | +0.017 |
Neg logit contributions | |
o | -0.009 |
favorite | -0.008 |
bie | -0.008 |
eps | -0.006 |
bo | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
Ivy | +0.012 |
Iris | +0.011 |
ner | +0.011 |
Holly | +0.011 |
Beat | +0.011 |
Neg logit contributions | |
favorite | -0.007 |
o | -0.007 |
bie | -0.007 |
bright | -0.006 |
eps | -0.006 |
Token | bright |
Token ID | 6016 |
Feature activation | +9.22e-02 |
Pos logit contributions | |
o | +0.053 |
favorite | +0.051 |
bie | +0.050 |
eps | +0.043 |
v | +0.042 |
Neg logit contributions | |
Ivy | -0.094 |
Iris | -0.089 |
Holly | -0.086 |
Beat | -0.085 |
ner | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
favorite | +0.103 |
o | +0.101 |
bie | +0.097 |
eps | +0.081 |
bright | +0.081 |
Neg logit contributions | |
Ivy | -0.222 |
Iris | -0.210 |
Holly | -0.203 |
ner | -0.200 |
Beat | -0.199 |
Token | icy |
Token ID | 30284 |
Feature activation | +3.47e-03 |
Pos logit contributions | |
o | +0.010 |
bie | +0.010 |
favorite | +0.009 |
eps | +0.007 |
bo | +0.007 |
Neg logit contributions | |
Ivy | -0.028 |
Iris | -0.027 |
Holly | -0.026 |
Beat | -0.025 |
ner | -0.025 |
Token | day |
Token ID | 1110 |
Feature activation | +9.25e-03 |
Pos logit contributions | |
favorite | +0.005 |
o | +0.005 |
bie | +0.004 |
eps | +0.004 |
bo | +0.004 |
Neg logit contributions | |
Ivy | -0.008 |
Iris | -0.007 |
Holly | -0.007 |
Beat | -0.007 |
ner | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
favorite | +0.010 |
o | +0.010 |
bright | +0.009 |
bie | +0.009 |
at | +0.008 |
Neg logit contributions | |
Ivy | -0.022 |
Iris | -0.021 |
Holly | -0.020 |
ner | -0.019 |
Beat | -0.019 |
Token | Sarah |
Token ID | 10490 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
Ivy | +0.032 |
Iris | +0.029 |
Holly | +0.029 |
ner | +0.028 |
Beat | +0.027 |
Neg logit contributions | |
favorite | -0.032 |
o | -0.030 |
bie | -0.029 |
bright | -0.027 |
eps | -0.027 |
Token | went |
Token ID | 1816 |
Feature activation | +5.37e-03 |
Pos logit contributions | |
Ivy | +0.026 |
Iris | +0.024 |
Holly | +0.024 |
Beat | +0.024 |
ner | +0.023 |
Neg logit contributions | |
favorite | -0.012 |
o | -0.011 |
bie | -0.011 |
eps | -0.009 |
v | -0.009 |
Token | outside |
Token ID | 2354 |
Feature activation | -6.27e-04 |
Pos logit contributions | |
favorite | +0.004 |
o | +0.004 |
bie | +0.003 |
big | +0.003 |
very | +0.003 |
Neg logit contributions | |
Ivy | -0.015 |
Iris | -0.014 |
ner | -0.014 |
Beat | -0.014 |
arises | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | -2.18e-03 |
Pos logit contributions | |
Ivy | +0.001 |
Iris | +0.001 |
Holly | +0.001 |
Beat | +0.001 |
ner | +0.001 |
Neg logit contributions | |
favorite | -0.001 |
o | -0.001 |
bie | -0.001 |
eps | -0.001 |
bo | -0.001 |
Token | dance |
Token ID | 9280 |
Feature activation | +1.82e-03 |
Pos logit contributions | |
Ivy | +0.005 |
Iris | +0.005 |
ner | +0.005 |
Holly | +0.005 |
Beat | +0.005 |
Neg logit contributions | |
favorite | -0.003 |
o | -0.002 |
bie | -0.002 |
bright | -0.002 |
bo | -0.002 |
Token | water |
Token ID | 1660 |
Feature activation | +8.28e-03 |
Pos logit contributions | |
o | +0.022 |
favorite | +0.022 |
bie | +0.021 |
eps | +0.019 |
v | +0.018 |
Neg logit contributions | |
Ivy | -0.022 |
Iris | -0.020 |
Holly | -0.020 |
Beat | -0.019 |
Fiona | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -3.62e-03 |
Pos logit contributions | |
favorite | +0.009 |
bie | +0.009 |
o | +0.008 |
eps | +0.007 |
bo | +0.007 |
Neg logit contributions | |
Ivy | -0.020 |
Holly | -0.018 |
Iris | -0.018 |
ner | -0.018 |
Fiona | -0.018 |
Token | It |
Token ID | 632 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
Ivy | +0.007 |
Iris | +0.006 |
Holly | +0.006 |
Beat | +0.006 |
ner | +0.006 |
Neg logit contributions | |
favorite | -0.006 |
o | -0.006 |
bie | -0.006 |
eps | -0.005 |
bo | -0.005 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
Ivy | +0.086 |
Holly | +0.080 |
Iris | +0.080 |
Fiona | +0.076 |
Beat | +0.075 |
Neg logit contributions | |
bie | -0.057 |
favorite | -0.056 |
o | -0.055 |
eps | -0.050 |
ams | -0.050 |
Token | happy |
Token ID | 3772 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
o | +0.020 |
bie | +0.019 |
bo | +0.017 |
v | +0.016 |
favorite | +0.016 |
Neg logit contributions | |
Ivy | -0.040 |
Iris | -0.038 |
Holly | -0.038 |
Beat | -0.037 |
Fiona | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
o | +0.026 |
favorite | +0.026 |
bie | +0.025 |
eps | +0.023 |
bo | +0.022 |
Neg logit contributions | |
Ivy | -0.050 |
Iris | -0.047 |
Holly | -0.046 |
Beat | -0.045 |
ner | -0.044 |
Token | fun |
Token ID | 1257 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
Ivy | +0.041 |
Iris | +0.039 |
Holly | +0.038 |
Beat | +0.037 |
Fiona | +0.037 |
Neg logit contributions | |
bie | -0.017 |
o | -0.017 |
favorite | -0.016 |
eps | -0.015 |
bo | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -7.29e-03 |
Pos logit contributions | |
Ivy | +0.047 |
Iris | +0.044 |
Holly | +0.043 |
Beat | +0.042 |
ner | +0.042 |
Neg logit contributions | |
o | -0.018 |
favorite | -0.018 |
bie | -0.017 |
eps | -0.015 |
bo | -0.014 |
Token | L |
Token ID | 406 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
Ivy | +0.014 |
Iris | +0.013 |
Holly | +0.012 |
Beat | +0.012 |
ner | +0.012 |
Neg logit contributions | |
o | -0.012 |
favorite | -0.012 |
bie | -0.012 |
eps | -0.011 |
bo | -0.010 |
Token | ila |
Token ID | 10102 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
o | +0.017 |
bie | +0.016 |
favorite | +0.016 |
eps | +0.014 |
v | +0.014 |
Neg logit contributions | |
Ivy | -0.021 |
Iris | -0.020 |
Beat | -0.019 |
Holly | -0.019 |
ner | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -6.21e-04 |
Pos logit contributions | |
o | +0.023 |
favorite | +0.023 |
bie | +0.022 |
eps | +0.019 |
bo | +0.018 |
Neg logit contributions | |
Ivy | -0.044 |
Iris | -0.042 |
Holly | -0.040 |
Beat | -0.040 |
ner | -0.039 |
Token | sweater |
Token ID | 35685 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
favorite | +0.170 |
o | +0.153 |
bie | +0.148 |
bright | +0.147 |
favourite | +0.143 |
Neg logit contributions | |
Ivy | -0.142 |
Iris | -0.133 |
ner | -0.125 |
Beat | -0.123 |
Holly | -0.118 |
Token | , |
Token ID | 11 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
Ivy | +0.058 |
Iris | +0.055 |
Holly | +0.054 |
Beat | +0.053 |
ner | +0.052 |
Neg logit contributions | |
o | -0.024 |
favorite | -0.024 |
bie | -0.023 |
eps | -0.019 |
bo | -0.018 |
Token | but |
Token ID | 475 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
Ivy | +0.043 |
Iris | +0.040 |
Holly | +0.039 |
Beat | +0.038 |
Fiona | +0.037 |
Neg logit contributions | |
o | -0.034 |
favorite | -0.033 |
bie | -0.033 |
eps | -0.030 |
bo | -0.029 |
Token | it |
Token ID | 340 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
favorite | +0.042 |
o | +0.041 |
bie | +0.041 |
bright | +0.036 |
hot | +0.035 |
Neg logit contributions | |
Ivy | -0.046 |
Iris | -0.044 |
ner | -0.042 |
Beat | -0.042 |
Holly | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | -1.59e-03 |
Pos logit contributions | |
Ivy | +0.071 |
Iris | +0.066 |
Holly | +0.064 |
Beat | +0.063 |
Fiona | +0.062 |
Neg logit contributions | |
o | -0.046 |
favorite | -0.046 |
bie | -0.045 |
eps | -0.040 |
v | -0.039 |
Token | better |
Token ID | 1365 |
Feature activation | -8.95e-02 |
Pos logit contributions | |
Ivy | +0.004 |
Iris | +0.004 |
Holly | +0.003 |
Beat | +0.003 |
Fiona | +0.003 |
Neg logit contributions | |
o | -0.002 |
favorite | -0.002 |
bie | -0.002 |
eps | -0.001 |
v | -0.001 |
Token | than |
Token ID | 621 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
Ivy | +0.181 |
Iris | +0.171 |
Holly | +0.168 |
Fiona | +0.163 |
Beat | +0.163 |
Neg logit contributions | |
o | -0.119 |
bie | -0.113 |
favorite | -0.112 |
eps | -0.110 |
bo | -0.106 |
Token | being |
Token ID | 852 |
Feature activation | -9.06e-03 |
Pos logit contributions | |
favorite | +0.168 |
o | +0.166 |
bie | +0.165 |
bright | +0.142 |
hot | +0.137 |
Neg logit contributions | |
Ivy | -0.216 |
Iris | -0.209 |
Beat | -0.197 |
Holly | -0.193 |
ner | -0.191 |
Token | embarrassed |
Token ID | 21100 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
Ivy | +0.021 |
Iris | +0.020 |
Holly | +0.019 |
Beat | +0.019 |
ner | +0.019 |
Neg logit contributions | |
favorite | -0.011 |
o | -0.011 |
bie | -0.011 |
eps | -0.009 |
tallest | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +7.43e-03 |
Pos logit contributions | |
Ivy | +0.048 |
Iris | +0.046 |
Holly | +0.044 |
Beat | +0.044 |
ner | +0.044 |
Neg logit contributions | |
favorite | -0.015 |
o | -0.015 |
bie | -0.014 |
eps | -0.011 |
bright | -0.010 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.59e-02 |
Pos logit contributions | |
favorite | +0.012 |
o | +0.012 |
bie | +0.011 |
eps | +0.010 |
bo | +0.010 |
Neg logit contributions | |
Ivy | -0.014 |
Iris | -0.013 |
Holly | -0.013 |
Beat | -0.013 |
ner | -0.013 |
Token | harder |
Token ID | 7069 |
Feature activation | -6.53e-02 |
Pos logit contributions | |
Ivy | +0.020 |
Iris | +0.019 |
Holly | +0.018 |
Beat | +0.018 |
ner | +0.017 |
Neg logit contributions | |
o | -0.012 |
favorite | -0.012 |
bie | -0.012 |
eps | -0.010 |
bo | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-03 |
Pos logit contributions | |
Ivy | +0.153 |
Iris | +0.145 |
Holly | +0.140 |
Beat | +0.138 |
ner | +0.137 |
Neg logit contributions | |
favorite | -0.076 |
o | -0.075 |
bie | -0.072 |
eps | -0.062 |
bo | -0.059 |
Token |
|
Token ID | 198 |
Feature activation | +9.13e-03 |
Pos logit contributions | |
Ivy | +0.002 |
Iris | +0.002 |
Holly | +0.002 |
Beat | +0.002 |
ner | +0.002 |
Neg logit contributions | |
o | -0.001 |
favorite | -0.001 |
bie | -0.001 |
eps | -0.001 |
bo | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
favorite | +0.011 |
o | +0.011 |
bie | +0.010 |
eps | +0.009 |
tallest | +0.009 |
Neg logit contributions | |
Ivy | -0.021 |
Iris | -0.020 |
Holly | -0.019 |
Beat | -0.019 |
Fiona | -0.019 |
Token | The |
Token ID | 464 |
Feature activation | +3.11e-03 |
Pos logit contributions | |
o | +0.020 |
favorite | +0.020 |
bie | +0.019 |
eps | +0.017 |
bo | +0.017 |
Neg logit contributions | |
Ivy | -0.032 |
Iris | -0.030 |
Holly | -0.029 |
Beat | -0.028 |
ner | -0.028 |
Token | lesson |
Token ID | 11483 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
o | +0.005 |
favorite | +0.004 |
bie | +0.004 |
v | +0.004 |
eps | +0.004 |
Neg logit contributions | |
Ivy | -0.006 |
Iris | -0.006 |
Holly | -0.006 |
Beat | -0.006 |
ner | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
Ivy | +0.264 |
Iris | +0.243 |
Holly | +0.241 |
pan | +0.231 |
Buffy | +0.230 |
Neg logit contributions | |
o | -0.117 |
at | -0.104 |
aw | -0.102 |
favorite | -0.101 |
saw | -0.089 |
Token | little |
Token ID | 1310 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
Ivy | +0.024 |
Iris | +0.023 |
Holly | +0.022 |
Beat | +0.022 |
ner | +0.022 |
Neg logit contributions | |
o | -0.016 |
favorite | -0.015 |
bie | -0.015 |
eps | -0.013 |
bo | -0.013 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
Ivy | +0.043 |
Iris | +0.040 |
Holly | +0.039 |
Beat | +0.038 |
ner | +0.038 |
Neg logit contributions | |
favorite | -0.029 |
o | -0.029 |
bie | -0.028 |
eps | -0.025 |
bo | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
Ivy | +0.036 |
Iris | +0.034 |
Holly | +0.033 |
Beat | +0.032 |
ner | +0.032 |
Neg logit contributions | |
o | -0.020 |
favorite | -0.020 |
bie | -0.019 |
eps | -0.017 |
bo | -0.016 |
Token | trying |
Token ID | 2111 |
Feature activation | -5.59e-03 |
Pos logit contributions | |
Ivy | +0.036 |
Iris | +0.035 |
Holly | +0.033 |
Beat | +0.033 |
ner | +0.033 |
Neg logit contributions | |
o | -0.015 |
favorite | -0.015 |
bie | -0.014 |
eps | -0.012 |
bo | -0.012 |
Token | song |
Token ID | 3496 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
Ivy | +0.022 |
Iris | +0.021 |
Holly | +0.020 |
Beat | +0.019 |
ner | +0.019 |
Neg logit contributions | |
o | -0.022 |
favorite | -0.021 |
bie | -0.021 |
eps | -0.019 |
bo | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
Ivy | +0.102 |
Iris | +0.097 |
Holly | +0.094 |
Beat | +0.092 |
ner | +0.091 |
Neg logit contributions | |
o | -0.062 |
favorite | -0.062 |
bie | -0.061 |
eps | -0.054 |
bo | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | -1.30e-03 |
Pos logit contributions | |
Ivy | +0.025 |
Iris | +0.023 |
Holly | +0.023 |
Beat | +0.022 |
ner | +0.022 |
Neg logit contributions | |
o | -0.011 |
favorite | -0.011 |
bie | -0.010 |
eps | -0.008 |
bo | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +4.08e-03 |
Pos logit contributions | |
Ivy | +0.003 |
Iris | +0.003 |
Holly | +0.003 |
Beat | +0.003 |
ner | +0.003 |
Neg logit contributions | |
o | -0.001 |
favorite | -0.001 |
bie | -0.001 |
eps | -0.001 |
bo | -0.001 |
Token | The |
Token ID | 464 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
o | +0.005 |
favorite | +0.004 |
eps | +0.004 |
bie | +0.004 |
bright | +0.004 |
Neg logit contributions | |
Ivy | -0.010 |
Iris | -0.009 |
Holly | -0.009 |
Beat | -0.009 |
ner | -0.009 |
Token | kitten |
Token ID | 42143 |
Feature activation | -9.01e-02 |
Pos logit contributions | |
Ivy | +0.021 |
Iris | +0.020 |
Holly | +0.019 |
Beat | +0.019 |
ner | +0.019 |
Neg logit contributions | |
favorite | -0.016 |
o | -0.015 |
bie | -0.014 |
bright | -0.014 |
eps | -0.013 |
Token | felt |
Token ID | 2936 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
Ivy | +0.207 |
Iris | +0.196 |
Holly | +0.189 |
Beat | +0.187 |
ner | +0.185 |
Neg logit contributions | |
favorite | -0.108 |
o | -0.108 |
bie | -0.103 |
eps | -0.089 |
bo | -0.086 |
Token | much |
Token ID | 881 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
Ivy | +0.065 |
Iris | +0.062 |
Holly | +0.060 |
Beat | +0.059 |
ner | +0.058 |
Neg logit contributions | |
o | -0.023 |
favorite | -0.022 |
bie | -0.022 |
eps | -0.018 |
v | -0.016 |
Token | better |
Token ID | 1365 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Ivy | +0.027 |
Iris | +0.025 |
Holly | +0.024 |
Beat | +0.023 |
ner | +0.023 |
Neg logit contributions | |
favorite | -0.026 |
o | -0.025 |
bie | -0.024 |
eps | -0.022 |
bright | -0.022 |
Token | as |
Token ID | 355 |
Feature activation | +4.10e-02 |
Pos logit contributions | |
Ivy | +0.221 |
Iris | +0.207 |
Holly | +0.204 |
ner | +0.203 |
Beat | +0.199 |
Neg logit contributions | |
bie | -0.162 |
o | -0.161 |
favorite | -0.160 |
eps | -0.150 |
bo | -0.142 |
Token | he |
Token ID | 339 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
favorite | +0.057 |
o | +0.055 |
bie | +0.052 |
bright | +0.047 |
eps | +0.045 |
Neg logit contributions | |
Ivy | -0.088 |
Iris | -0.083 |
Holly | -0.080 |
ner | -0.078 |
Beat | -0.078 |
Token | spraying |
Token ID | 40170 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
Ivy | +0.053 |
Iris | +0.050 |
Holly | +0.048 |
Beat | +0.047 |
ner | +0.047 |
Neg logit contributions | |
o | -0.032 |
favorite | -0.032 |
bie | -0.030 |
eps | -0.026 |
bo | -0.026 |
Token | up |
Token ID | 510 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
Ivy | +0.069 |
Iris | +0.066 |
ner | +0.064 |
Beat | +0.063 |
Holly | +0.062 |
Neg logit contributions | |
favorite | -0.037 |
o | -0.037 |
bie | -0.034 |
bright | -0.031 |
hot | -0.030 |
Token | so |
Token ID | 523 |
Feature activation | +6.35e-03 |
Pos logit contributions | |
Ivy | +0.064 |
Iris | +0.060 |
Holly | +0.058 |
Beat | +0.058 |
ner | +0.057 |
Neg logit contributions | |
favorite | -0.033 |
o | -0.033 |
bie | -0.031 |
eps | -0.027 |
bo | -0.026 |
Token | high |
Token ID | 1029 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
favorite | +0.010 |
o | +0.010 |
bie | +0.010 |
bright | +0.009 |
eps | +0.009 |
Neg logit contributions | |
Ivy | -0.012 |
Iris | -0.011 |
Holly | -0.011 |
Beat | -0.010 |
ner | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -3.31e-04 |
Pos logit contributions | |
favorite | +0.026 |
o | +0.024 |
bie | +0.022 |
bright | +0.020 |
bo | +0.020 |
Neg logit contributions | |
Ivy | -0.060 |
Iris | -0.058 |
Beat | -0.056 |
Holly | -0.056 |
ner | -0.056 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.92e-02 |
Pos logit contributions | |
Ivy | +0.001 |
Iris | +0.001 |
Holly | +0.001 |
Beat | +0.001 |
ner | +0.001 |
Neg logit contributions | |
favorite | -0.000 |
o | -0.000 |
bie | -0.000 |
eps | -0.000 |
bo | -0.000 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
Ivy | +0.216 |
Iris | +0.206 |
Holly | +0.197 |
Beat | +0.196 |
ner | +0.194 |
Neg logit contributions | |
o | -0.099 |
favorite | -0.092 |
bie | -0.090 |
bo | -0.074 |
eps | -0.074 |
Token | upon |
Token ID | 2402 |
Feature activation | -3.25e-03 |
Pos logit contributions | |
o | +0.042 |
favorite | +0.042 |
eps | +0.039 |
bie | +0.038 |
v | +0.038 |
Neg logit contributions | |
Ivy | -0.059 |
Iris | -0.055 |
Holly | -0.055 |
Beat | -0.053 |
Fiona | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | +4.65e-02 |
Pos logit contributions | |
Ivy | +0.008 |
Iris | +0.007 |
Holly | +0.007 |
Beat | +0.007 |
ner | +0.007 |
Neg logit contributions | |
o | -0.004 |
bie | -0.004 |
favorite | -0.004 |
ams | -0.003 |
eps | -0.003 |
Token | time |
Token ID | 640 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
v | +0.073 |
ams | +0.072 |
tallest | +0.072 |
o | +0.072 |
eps | +0.068 |
Neg logit contributions | |
Ivy | -0.085 |
Iris | -0.076 |
Holly | -0.074 |
Beat | -0.072 |
Fiona | -0.071 |
Token | there |
Token ID | 612 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
o | +0.067 |
favorite | +0.066 |
bie | +0.063 |
eps | +0.053 |
bright | +0.052 |
Neg logit contributions | |
Ivy | -0.104 |
Iris | -0.098 |
Holly | -0.094 |
ner | -0.093 |
Beat | -0.092 |
Token |
|
Token ID | 198 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
favorite | +0.010 |
o | +0.010 |
bie | +0.009 |
eps | +0.008 |
bo | +0.008 |
Neg logit contributions | |
Ivy | -0.017 |
Iris | -0.016 |
Holly | -0.015 |
Beat | -0.015 |
ner | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
favorite | +0.014 |
o | +0.014 |
bie | +0.013 |
eps | +0.011 |
bo | +0.011 |
Neg logit contributions | |
Ivy | -0.024 |
Iris | -0.023 |
Holly | -0.022 |
Beat | -0.022 |
ner | -0.021 |
Token | The |
Token ID | 464 |
Feature activation | -1.01e-03 |
Pos logit contributions | |
o | +0.021 |
favorite | +0.020 |
bie | +0.020 |
eps | +0.018 |
bo | +0.017 |
Neg logit contributions | |
Ivy | -0.029 |
Iris | -0.027 |
Holly | -0.026 |
Beat | -0.026 |
ner | -0.025 |
Token | end |
Token ID | 886 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
Ivy | +0.002 |
Iris | +0.002 |
Holly | +0.002 |
Beat | +0.002 |
ner | +0.002 |
Neg logit contributions | |
o | -0.002 |
bie | -0.002 |
favorite | -0.001 |
bo | -0.001 |
eps | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +4.75e-03 |
Pos logit contributions | |
Ivy | +0.075 |
Iris | +0.072 |
Fiona | +0.070 |
ion | +0.069 |
Beat | +0.069 |
Neg logit contributions | |
favorite | -0.049 |
bie | -0.047 |
eps | -0.046 |
tallest | -0.045 |
o | -0.043 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.01e-02 |
Pos logit contributions | |
favorite | +0.006 |
o | +0.005 |
bie | +0.005 |
eps | +0.005 |
v | +0.005 |
Neg logit contributions | |
Ivy | -0.011 |
Iris | -0.010 |
Holly | -0.010 |
Beat | -0.010 |
ner | -0.010 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
Ivy | +0.218 |
Iris | +0.207 |
Holly | +0.200 |
Beat | +0.198 |
ner | +0.196 |
Neg logit contributions | |
o | -0.097 |
favorite | -0.097 |
bie | -0.092 |
eps | -0.078 |
bo | -0.075 |
Token | there |
Token ID | 612 |
Feature activation | -3.79e-03 |
Pos logit contributions | |
favorite | +0.034 |
o | +0.034 |
bie | +0.032 |
eps | +0.031 |
bo | +0.030 |
Neg logit contributions | |
Ivy | -0.054 |
Iris | -0.051 |
Holly | -0.050 |
Beat | -0.049 |
ner | -0.048 |
Token | was |
Token ID | 373 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
Ivy | +0.008 |
Iris | +0.007 |
Holly | +0.007 |
Beat | +0.007 |
Fiona | +0.007 |
Neg logit contributions | |
favorite | -0.006 |
o | -0.005 |
bie | -0.005 |
eps | -0.005 |
v | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -2.81e-03 |
Pos logit contributions | |
o | +0.033 |
bie | +0.032 |
favorite | +0.031 |
eps | +0.027 |
bo | +0.026 |
Neg logit contributions | |
Ivy | -0.087 |
Iris | -0.083 |
Holly | -0.081 |
Beat | -0.080 |
Fiona | -0.078 |
Token | modern |
Token ID | 3660 |
Feature activation | +5.36e-03 |
Pos logit contributions | |
Ivy | +0.007 |
Iris | +0.006 |
Holly | +0.006 |
Beat | +0.006 |
ner | +0.006 |
Neg logit contributions | |
o | -0.003 |
favorite | -0.003 |
bie | -0.003 |
eps | -0.003 |
ams | -0.003 |