Token | the |
Token ID | 262 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
day | +0.061 |
night | +0.060 |
time | +0.058 |
nights | +0.058 |
morning | +0.057 |
Neg logit contributions | |
boys | -0.046 |
paramed | -0.045 |
chefs | -0.043 |
Lee | -0.040 |
groom | -0.039 |
Token | ambulance |
Token ID | 22536 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
paramed | +0.063 |
boys | +0.061 |
chefs | +0.059 |
Lee | +0.058 |
printer | +0.056 |
Neg logit contributions | |
day | -0.091 |
nights | -0.091 |
night | -0.089 |
fy | -0.086 |
time | -0.085 |
Token | he |
Token ID | 339 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
boys | +0.214 |
paramed | +0.214 |
chefs | +0.210 |
Lee | +0.200 |
groom | +0.195 |
Neg logit contributions | |
day | -0.194 |
night | -0.190 |
nights | -0.186 |
morning | -0.180 |
afternoon | -0.176 |
Token | saw |
Token ID | 2497 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
day | +0.201 |
night | +0.199 |
nights | +0.196 |
morning | +0.190 |
time | +0.190 |
Neg logit contributions | |
paramed | -0.114 |
boys | -0.112 |
chefs | -0.108 |
Lee | -0.105 |
reporter | -0.096 |
Token | a |
Token ID | 257 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
boys | +0.102 |
paramed | +0.101 |
chefs | +0.095 |
Lee | +0.093 |
groom | +0.089 |
Neg logit contributions | |
day | -0.123 |
night | -0.121 |
nights | -0.119 |
time | -0.115 |
morning | -0.114 |
Token | paramed |
Token ID | 34581 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
boys | +0.072 |
chefs | +0.067 |
Lee | +0.065 |
wrists | +0.065 |
oa | +0.062 |
Neg logit contributions | |
day | -0.061 |
time | -0.057 |
night | -0.057 |
morning | -0.053 |
evening | -0.051 |
Token | ic |
Token ID | 291 |
Feature activation | +5.17e-02 |
Pos logit contributions | |
boys | +0.591 |
paramed | +0.582 |
chefs | +0.574 |
Lee | +0.566 |
printer | +0.539 |
Neg logit contributions | |
day | -0.361 |
night | -0.353 |
nights | -0.352 |
morning | -0.327 |
afternoon | -0.326 |
Token | wearing |
Token ID | 5762 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
paramed | +0.107 |
boys | +0.105 |
chefs | +0.103 |
Lee | +0.102 |
printer | +0.099 |
Neg logit contributions | |
nights | -0.055 |
day | -0.055 |
night | -0.054 |
fy | -0.051 |
Along | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
day | +0.175 |
night | +0.171 |
time | +0.169 |
nights | +0.164 |
morning | +0.164 |
Neg logit contributions | |
boys | -0.151 |
paramed | -0.144 |
chefs | -0.143 |
Lee | -0.138 |
principal | -0.132 |
Token | big |
Token ID | 1263 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
boys | +0.072 |
chefs | +0.067 |
Lee | +0.067 |
paramed | +0.063 |
principal | +0.062 |
Neg logit contributions | |
day | -0.055 |
time | -0.054 |
night | -0.052 |
morning | -0.050 |
fy | -0.049 |
Token | coat |
Token ID | 13209 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
day | +0.026 |
time | +0.025 |
night | +0.025 |
morning | +0.024 |
evening | +0.023 |
Neg logit contributions | |
boys | -0.028 |
chefs | -0.027 |
principal | -0.024 |
Lee | -0.023 |
oa | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
boys | +0.044 |
Lee | +0.039 |
paramed | +0.039 |
chefs | +0.037 |
groom | +0.037 |
Neg logit contributions | |
nights | -0.054 |
fy | -0.052 |
day | -0.052 |
night | -0.051 |
afternoon | -0.051 |
Token | she |
Token ID | 673 |
Feature activation | -9.79e-02 |
Pos logit contributions | |
boys | +0.188 |
paramed | +0.184 |
chefs | +0.180 |
Lee | +0.172 |
principal | +0.170 |
Neg logit contributions | |
day | -0.180 |
night | -0.179 |
nights | -0.172 |
morning | -0.169 |
time | -0.163 |
Token | noticed |
Token ID | 6810 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
night | +0.173 |
day | +0.172 |
nights | +0.170 |
time | +0.160 |
morning | +0.160 |
Neg logit contributions | |
paramed | -0.144 |
chefs | -0.139 |
boys | -0.135 |
Lee | -0.134 |
principal | -0.123 |
Token | a |
Token ID | 257 |
Feature activation | +5.24e-02 |
Pos logit contributions | |
paramed | +0.137 |
boys | +0.136 |
chefs | +0.133 |
Lee | +0.130 |
principal | +0.123 |
Neg logit contributions | |
day | -0.186 |
night | -0.184 |
nights | -0.180 |
morning | -0.174 |
time | -0.173 |
Token | little |
Token ID | 1310 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
chefs | +0.102 |
boys | +0.101 |
Lee | +0.100 |
paramed | +0.096 |
principal | +0.093 |
Neg logit contributions | |
day | -0.072 |
night | -0.067 |
time | -0.066 |
morning | -0.061 |
evening | -0.060 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
day | +0.077 |
night | +0.076 |
morning | +0.071 |
nights | +0.069 |
afternoon | +0.068 |
Neg logit contributions | |
paramed | -0.069 |
boys | -0.068 |
chefs | -0.068 |
principal | -0.066 |
Lee | -0.066 |
Token | m |
Token ID | 76 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
paramed | +0.526 |
boys | +0.520 |
chefs | +0.508 |
Lee | +0.502 |
principal | +0.491 |
Neg logit contributions | |
day | -0.441 |
night | -0.420 |
nights | -0.412 |
afternoon | -0.395 |
morning | -0.393 |
Token | unk |
Token ID | 2954 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
day | +0.019 |
night | +0.019 |
nights | +0.019 |
morning | +0.018 |
afternoon | +0.018 |
Neg logit contributions | |
boys | -0.018 |
paramed | -0.018 |
chefs | -0.017 |
Lee | -0.017 |
printer | -0.016 |
Token | sc |
Token ID | 629 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
boys | +0.176 |
paramed | +0.175 |
chefs | +0.171 |
Lee | +0.167 |
principal | +0.159 |
Neg logit contributions | |
day | -0.170 |
night | -0.166 |
nights | -0.164 |
morning | -0.157 |
time | -0.156 |
Token | ur |
Token ID | 333 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
day | +0.012 |
night | +0.012 |
nights | +0.012 |
morning | +0.011 |
time | +0.011 |
Neg logit contributions | |
boys | -0.022 |
paramed | -0.022 |
chefs | -0.022 |
Lee | -0.021 |
principal | -0.020 |
Token | rying |
Token ID | 14992 |
Feature activation | -5.25e-02 |
Pos logit contributions | |
nights | +0.102 |
day | +0.102 |
night | +0.099 |
afternoon | +0.090 |
ie | +0.090 |
Neg logit contributions | |
paramed | -0.217 |
Lee | -0.209 |
boys | -0.207 |
chefs | -0.206 |
printer | -0.203 |
Token | and |
Token ID | 290 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
boys | +0.205 |
paramed | +0.203 |
Lee | +0.196 |
chefs | +0.195 |
printer | +0.191 |
Neg logit contributions | |
nights | -0.101 |
night | -0.097 |
day | -0.097 |
afternoon | -0.090 |
morning | -0.090 |
Token | wearing |
Token ID | 5762 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
day | +0.019 |
night | +0.019 |
nights | +0.018 |
morning | +0.018 |
time | +0.017 |
Neg logit contributions | |
boys | -0.015 |
paramed | -0.015 |
chefs | -0.014 |
Lee | -0.014 |
reporter | -0.013 |
Token | gloves |
Token ID | 18051 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
day | +0.171 |
night | +0.169 |
nights | +0.165 |
time | +0.165 |
morning | +0.161 |
Neg logit contributions | |
boys | -0.152 |
paramed | -0.149 |
chefs | -0.144 |
Lee | -0.143 |
principal | -0.138 |
Token | . |
Token ID | 13 |
Feature activation | -2.14e-03 |
Pos logit contributions | |
boys | +0.034 |
paramed | +0.034 |
chefs | +0.033 |
Lee | +0.033 |
principal | +0.031 |
Neg logit contributions | |
day | -0.019 |
night | -0.019 |
nights | -0.018 |
morning | -0.017 |
time | -0.017 |
Token | The |
Token ID | 383 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
day | +0.003 |
nights | +0.003 |
night | +0.003 |
afternoon | +0.003 |
morning | +0.003 |
Neg logit contributions | |
boys | -0.004 |
paramed | -0.004 |
chefs | -0.004 |
Lee | -0.004 |
principal | -0.003 |
Token | paramed |
Token ID | 34581 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
boys | +0.047 |
chefs | +0.040 |
wrists | +0.038 |
Lee | +0.038 |
reporter | +0.035 |
Neg logit contributions | |
day | -0.104 |
night | -0.100 |
morning | -0.098 |
nights | -0.097 |
time | -0.097 |
Token | ic |
Token ID | 291 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
boys | +0.614 |
chefs | +0.598 |
paramed | +0.593 |
Lee | +0.587 |
groom | +0.564 |
Neg logit contributions | |
day | -0.334 |
nights | -0.328 |
night | -0.323 |
afternoon | -0.301 |
morning | -0.297 |
Token | was |
Token ID | 373 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
boys | +0.085 |
chefs | +0.081 |
paramed | +0.079 |
Lee | +0.079 |
principal | +0.075 |
Neg logit contributions | |
day | -0.074 |
night | -0.071 |
nights | -0.071 |
time | -0.067 |
morning | -0.067 |
Token | very |
Token ID | 845 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
boys | +0.058 |
Lee | +0.055 |
paramed | +0.055 |
chefs | +0.055 |
principal | +0.052 |
Neg logit contributions | |
day | -0.058 |
night | -0.057 |
nights | -0.056 |
time | -0.055 |
morning | -0.054 |
Token | polite |
Token ID | 23507 |
Feature activation | +6.41e-02 |
Pos logit contributions | |
day | +0.030 |
night | +0.030 |
nights | +0.030 |
morning | +0.028 |
afternoon | +0.028 |
Neg logit contributions | |
boys | -0.031 |
paramed | -0.031 |
chefs | -0.030 |
Lee | -0.029 |
reporter | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -9.60e-03 |
Pos logit contributions | |
boys | +0.103 |
paramed | +0.101 |
chefs | +0.100 |
Lee | +0.100 |
reporter | +0.093 |
Neg logit contributions | |
day | -0.097 |
nights | -0.096 |
night | -0.096 |
morning | -0.091 |
afternoon | -0.091 |
Token | around |
Token ID | 1088 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
day | +0.062 |
nights | +0.062 |
night | +0.061 |
morning | +0.057 |
afternoon | +0.056 |
Neg logit contributions | |
boys | -0.103 |
paramed | -0.101 |
chefs | -0.098 |
Lee | -0.098 |
printer | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | +7.35e-03 |
Pos logit contributions | |
day | +0.055 |
night | +0.054 |
nights | +0.053 |
morning | +0.051 |
time | +0.051 |
Neg logit contributions | |
boys | -0.062 |
paramed | -0.062 |
chefs | -0.060 |
Lee | -0.059 |
principal | -0.056 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
boys | +0.014 |
paramed | +0.013 |
chefs | +0.013 |
groom | +0.012 |
reporter | +0.012 |
Neg logit contributions | |
day | -0.010 |
night | -0.010 |
nights | -0.009 |
Along | -0.009 |
morning | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
night | +0.063 |
day | +0.062 |
nights | +0.061 |
time | +0.058 |
morning | +0.058 |
Neg logit contributions | |
boys | -0.087 |
paramed | -0.086 |
chefs | -0.085 |
principal | -0.083 |
reporter | -0.081 |
Token | the |
Token ID | 262 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
boys | +0.203 |
paramed | +0.194 |
chefs | +0.192 |
Lee | +0.185 |
principal | +0.184 |
Neg logit contributions | |
day | -0.149 |
night | -0.149 |
nights | -0.142 |
morning | -0.139 |
time | -0.135 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
boys | +0.079 |
paramed | +0.075 |
chefs | +0.075 |
Lee | +0.073 |
reporter | +0.071 |
Neg logit contributions | |
day | -0.094 |
night | -0.091 |
nights | -0.089 |
morning | -0.087 |
time | -0.086 |
Token | m |
Token ID | 76 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
boys | +0.558 |
paramed | +0.550 |
chefs | +0.545 |
Lee | +0.540 |
principal | +0.534 |
Neg logit contributions | |
day | -0.410 |
night | -0.371 |
nights | -0.366 |
afternoon | -0.365 |
morning | -0.356 |
Token | unk |
Token ID | 2954 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
day | +0.028 |
night | +0.027 |
nights | +0.027 |
morning | +0.026 |
afternoon | +0.026 |
Neg logit contributions | |
paramed | -0.019 |
boys | -0.019 |
chefs | -0.018 |
Lee | -0.018 |
printer | -0.016 |
Token | stopped |
Token ID | 5025 |
Feature activation | +6.29e-03 |
Pos logit contributions | |
boys | +0.148 |
paramed | +0.146 |
chefs | +0.141 |
Lee | +0.138 |
printer | +0.129 |
Neg logit contributions | |
day | -0.194 |
night | -0.191 |
nights | -0.190 |
morning | -0.182 |
time | -0.181 |
Token | and |
Token ID | 290 |
Feature activation | +2.18e-03 |
Pos logit contributions | |
boys | +0.010 |
paramed | +0.010 |
chefs | +0.010 |
Lee | +0.010 |
principal | +0.009 |
Neg logit contributions | |
day | -0.009 |
night | -0.009 |
nights | -0.009 |
morning | -0.009 |
time | -0.009 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
chefs | +0.004 |
boys | +0.004 |
paramed | +0.003 |
principal | +0.003 |
Lee | +0.003 |
Neg logit contributions | |
night | -0.003 |
day | -0.003 |
morning | -0.003 |
nights | -0.003 |
evening | -0.003 |
Token | after |
Token ID | 706 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
boys | +0.054 |
paramed | +0.053 |
chefs | +0.052 |
Lee | +0.052 |
reporter | +0.049 |
Neg logit contributions | |
day | -0.039 |
night | -0.038 |
nights | -0.038 |
morning | -0.036 |
afternoon | -0.035 |
Token | that |
Token ID | 326 |
Feature activation | +6.83e-02 |
Pos logit contributions | |
day | +0.068 |
nights | +0.067 |
night | +0.067 |
fy | +0.066 |
morning | +0.064 |
Neg logit contributions | |
boys | -0.055 |
paramed | -0.053 |
chefs | -0.052 |
Lee | -0.051 |
printer | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-03 |
Pos logit contributions | |
boys | +0.143 |
paramed | +0.136 |
Lee | +0.136 |
chefs | +0.134 |
printer | +0.132 |
Neg logit contributions | |
nights | -0.068 |
fy | -0.066 |
night | -0.064 |
Along | -0.064 |
day | -0.063 |
Token | Whenever |
Token ID | 21326 |
Feature activation | -9.91e-02 |
Pos logit contributions | |
nights | +0.003 |
day | +0.003 |
night | +0.003 |
fy | +0.003 |
afternoon | +0.003 |
Neg logit contributions | |
boys | -0.003 |
paramed | -0.003 |
Lee | -0.003 |
chefs | -0.002 |
principal | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
day | +0.197 |
night | +0.192 |
time | +0.185 |
nights | +0.185 |
morning | +0.183 |
Neg logit contributions | |
paramed | -0.140 |
boys | -0.128 |
chefs | -0.121 |
groom | -0.118 |
wrists | -0.118 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
boys | +0.072 |
Lee | +0.065 |
reporter | +0.065 |
paramed | +0.064 |
chefs | +0.062 |
Neg logit contributions | |
day | -0.126 |
night | -0.119 |
morning | -0.117 |
time | -0.116 |
afternoon | -0.115 |
Token | m |
Token ID | 76 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
boys | +0.560 |
paramed | +0.542 |
wrists | +0.533 |
reporter | +0.531 |
groom | +0.530 |
Neg logit contributions | |
day | -0.427 |
night | -0.377 |
afternoon | -0.370 |
morning | -0.370 |
nights | -0.353 |
Token | unk |
Token ID | 2954 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
day | +0.026 |
night | +0.025 |
nights | +0.025 |
time | +0.024 |
morning | +0.024 |
Neg logit contributions | |
boys | -0.020 |
paramed | -0.019 |
chefs | -0.019 |
Lee | -0.018 |
reporter | -0.018 |
Token | needed |
Token ID | 2622 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
boys | +0.152 |
paramed | +0.150 |
chefs | +0.145 |
Lee | +0.143 |
reporter | +0.133 |
Neg logit contributions | |
day | -0.195 |
night | -0.192 |
nights | -0.190 |
morning | -0.182 |
time | -0.181 |
Token | help |
Token ID | 1037 |
Feature activation | +6.93e-02 |
Pos logit contributions | |
day | +0.033 |
night | +0.032 |
nights | +0.032 |
time | +0.032 |
morning | +0.031 |
Neg logit contributions | |
boys | -0.029 |
paramed | -0.028 |
Lee | -0.027 |
chefs | -0.027 |
principal | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
paramed | +0.117 |
boys | +0.116 |
chefs | +0.113 |
Lee | +0.111 |
reporter | +0.106 |
Neg logit contributions | |
day | -0.104 |
night | -0.101 |
nights | -0.100 |
time | -0.096 |
morning | -0.095 |
Token | you |
Token ID | 345 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
paramed | +0.119 |
boys | +0.107 |
chefs | +0.107 |
reporter | +0.103 |
Lee | +0.103 |
Neg logit contributions | |
day | -0.074 |
night | -0.071 |
time | -0.069 |
nights | -0.069 |
morning | -0.066 |
Token | ". |
Token ID | 1911 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
day | +0.089 |
night | +0.087 |
nights | +0.087 |
morning | +0.082 |
afternoon | +0.082 |
Neg logit contributions | |
boys | -0.097 |
paramed | -0.095 |
chefs | -0.093 |
Lee | -0.091 |
groom | -0.086 |
Token |
|
Token ID | 198 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
boys | +0.298 |
paramed | +0.294 |
chefs | +0.285 |
Lee | +0.284 |
principal | +0.267 |
Neg logit contributions | |
day | -0.275 |
night | -0.270 |
nights | -0.270 |
morning | -0.255 |
time | -0.254 |
Token |
|
Token ID | 198 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
boys | +0.041 |
paramed | +0.041 |
chefs | +0.039 |
Lee | +0.039 |
principal | +0.037 |
Neg logit contributions | |
day | -0.034 |
nights | -0.034 |
night | -0.033 |
fy | -0.031 |
morning | -0.031 |
Token | The |
Token ID | 464 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
boys | +0.033 |
paramed | +0.033 |
chefs | +0.033 |
Lee | +0.032 |
groom | +0.030 |
Neg logit contributions | |
day | -0.042 |
nights | -0.041 |
night | -0.041 |
afternoon | -0.040 |
time | -0.039 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
boys | +0.024 |
Lee | +0.022 |
chefs | +0.022 |
reporter | +0.022 |
paramed | +0.021 |
Neg logit contributions | |
day | -0.030 |
night | -0.028 |
time | -0.027 |
morning | -0.027 |
afternoon | -0.027 |
Token | m |
Token ID | 76 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
boys | +0.591 |
paramed | +0.568 |
chefs | +0.559 |
groom | +0.559 |
Lee | +0.557 |
Neg logit contributions | |
day | -0.397 |
night | -0.345 |
afternoon | -0.345 |
morning | -0.337 |
nights | -0.325 |
Token | unk |
Token ID | 2954 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
day | +0.027 |
night | +0.026 |
nights | +0.026 |
time | +0.026 |
morning | +0.025 |
Neg logit contributions | |
boys | -0.022 |
paramed | -0.021 |
chefs | -0.020 |
Lee | -0.020 |
reporter | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -6.85e-03 |
Pos logit contributions | |
paramed | +0.147 |
boys | +0.147 |
chefs | +0.146 |
Lee | +0.140 |
reporter | +0.131 |
Neg logit contributions | |
day | -0.192 |
night | -0.188 |
nights | -0.186 |
morning | -0.180 |
time | -0.178 |
Token | Sam |
Token ID | 3409 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
day | +0.014 |
night | +0.013 |
afternoon | +0.013 |
nights | +0.012 |
morning | +0.012 |
Neg logit contributions | |
paramed | -0.009 |
chefs | -0.009 |
reporter | -0.009 |
boys | -0.009 |
principal | -0.008 |
Token | were |
Token ID | 547 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
boys | +0.082 |
paramed | +0.081 |
chefs | +0.078 |
Lee | +0.076 |
printer | +0.071 |
Neg logit contributions | |
day | -0.106 |
night | -0.104 |
nights | -0.104 |
morning | -0.099 |
afternoon | -0.099 |
Token |
|
Token ID | 198 |
Feature activation | -4.52e-03 |
Pos logit contributions | |
day | +0.004 |
night | +0.004 |
nights | +0.004 |
morning | +0.003 |
time | +0.003 |
Neg logit contributions | |
boys | -0.005 |
paramed | -0.005 |
chefs | -0.005 |
Lee | -0.004 |
groom | -0.004 |
Token | L |
Token ID | 43 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
day | +0.008 |
nights | +0.008 |
night | +0.008 |
afternoon | +0.007 |
time | +0.007 |
Neg logit contributions | |
boys | -0.007 |
paramed | -0.006 |
Lee | -0.006 |
chefs | -0.006 |
groom | -0.006 |
Token | ily |
Token ID | 813 |
Feature activation | -2.62e-04 |
Pos logit contributions | |
night | +0.077 |
nights | +0.076 |
day | +0.075 |
fy | +0.074 |
morning | +0.073 |
Neg logit contributions | |
chefs | -0.055 |
paramed | -0.050 |
boys | -0.049 |
principal | -0.047 |
Lee | -0.047 |
Token | asked |
Token ID | 1965 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
day | +0.000 |
night | +0.000 |
nights | +0.000 |
ie | +0.000 |
time | +0.000 |
Neg logit contributions | |
chefs | -0.001 |
bluff | -0.001 |
paramed | -0.001 |
liar | -0.001 |
Lee | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +3.55e-02 |
Pos logit contributions | |
wrists | +0.080 |
groom | +0.078 |
reporter | +0.076 |
paramed | +0.075 |
boys | +0.074 |
Neg logit contributions | |
day | -0.060 |
night | -0.059 |
ie | -0.056 |
morning | -0.054 |
time | -0.054 |
Token | paramed |
Token ID | 34581 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
boys | +0.045 |
wrists | +0.045 |
chefs | +0.041 |
ah | +0.038 |
Lee | +0.038 |
Neg logit contributions | |
day | -0.074 |
time | -0.071 |
night | -0.069 |
morning | -0.069 |
nights | -0.066 |
Token | ic |
Token ID | 291 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
boys | +0.583 |
chefs | +0.569 |
Lee | +0.560 |
paramed | +0.557 |
groom | +0.543 |
Neg logit contributions | |
day | -0.337 |
nights | -0.329 |
night | -0.322 |
afternoon | -0.303 |
fy | -0.295 |
Token | , |
Token ID | 11 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
boys | +0.082 |
chefs | +0.078 |
paramed | +0.077 |
wrists | +0.076 |
Lee | +0.076 |
Neg logit contributions | |
day | -0.051 |
night | -0.048 |
nights | -0.046 |
morning | -0.045 |
time | -0.045 |
Token | " |
Token ID | 366 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
boys | +0.122 |
wrists | +0.105 |
chefs | +0.101 |
principal | +0.098 |
reporter | +0.095 |
Neg logit contributions | |
day | -0.153 |
Along | -0.144 |
morning | -0.142 |
night | -0.141 |
time | -0.140 |
Token | Is |
Token ID | 3792 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
boys | +0.289 |
Lee | +0.274 |
wrists | +0.272 |
chefs | +0.256 |
liar | +0.255 |
Neg logit contributions | |
day | -0.386 |
nights | -0.378 |
night | -0.372 |
morning | -0.360 |
afternoon | -0.357 |
Token | the |
Token ID | 262 |
Feature activation | +3.55e-02 |
Pos logit contributions | |
paramed | +0.071 |
boys | +0.068 |
chefs | +0.067 |
Lee | +0.067 |
reporter | +0.065 |
Neg logit contributions | |
day | -0.082 |
night | -0.081 |
nights | -0.078 |
time | -0.077 |
morning | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
day | +0.015 |
night | +0.015 |
nights | +0.014 |
time | +0.014 |
morning | +0.014 |
Neg logit contributions | |
boys | -0.027 |
paramed | -0.027 |
chefs | -0.026 |
Lee | -0.026 |
principal | -0.025 |
Token | The |
Token ID | 383 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
night | +0.069 |
day | +0.069 |
time | +0.068 |
morning | +0.067 |
nights | +0.066 |
Neg logit contributions | |
boys | -0.064 |
paramed | -0.059 |
chefs | -0.056 |
groom | -0.054 |
dads | -0.053 |
Token | driver |
Token ID | 4639 |
Feature activation | +9.14e-02 |
Pos logit contributions | |
boys | +0.048 |
chefs | +0.045 |
Lee | +0.044 |
paramed | +0.044 |
wrists | +0.041 |
Neg logit contributions | |
day | -0.067 |
night | -0.066 |
nights | -0.064 |
time | -0.063 |
morning | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
boys | +0.128 |
paramed | +0.127 |
chefs | +0.126 |
Lee | +0.121 |
groom | +0.115 |
Neg logit contributions | |
day | -0.159 |
night | -0.157 |
nights | -0.155 |
time | -0.151 |
morning | -0.147 |
Token | the |
Token ID | 262 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
night | +0.075 |
day | +0.072 |
morning | +0.071 |
time | +0.069 |
nights | +0.069 |
Neg logit contributions | |
boys | -0.061 |
paramed | -0.060 |
chefs | -0.053 |
Fighting | -0.052 |
Lincoln | -0.052 |
Token | paramed |
Token ID | 34581 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
boys | +0.046 |
chefs | +0.044 |
Lee | +0.043 |
paramed | +0.042 |
reporter | +0.040 |
Neg logit contributions | |
day | -0.060 |
night | -0.058 |
nights | -0.057 |
time | -0.056 |
morning | -0.055 |
Token | ic |
Token ID | 291 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
boys | +0.542 |
paramed | +0.532 |
chefs | +0.525 |
Lee | +0.516 |
groom | +0.494 |
Neg logit contributions | |
day | -0.380 |
night | -0.372 |
nights | -0.371 |
morning | -0.346 |
afternoon | -0.346 |
Token | jumped |
Token ID | 11687 |
Feature activation | -9.98e-02 |
Pos logit contributions | |
paramed | +0.060 |
boys | +0.060 |
chefs | +0.058 |
Lee | +0.057 |
reporter | +0.054 |
Neg logit contributions | |
day | -0.059 |
night | -0.058 |
nights | -0.058 |
morning | -0.055 |
afternoon | -0.055 |
Token | out |
Token ID | 503 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
nights | +0.142 |
day | +0.142 |
night | +0.137 |
afternoon | +0.132 |
morning | +0.130 |
Neg logit contributions | |
boys | -0.174 |
paramed | -0.170 |
chefs | -0.166 |
Lee | -0.165 |
printer | -0.154 |
Token | . |
Token ID | 13 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
fy | +0.034 |
nights | +0.033 |
day | +0.033 |
found | +0.032 |
afternoon | +0.032 |
Neg logit contributions | |
paramed | -0.076 |
chefs | -0.076 |
Lee | -0.076 |
boys | -0.075 |
reporter | -0.071 |
Token | They |
Token ID | 1119 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
night | +0.052 |
day | +0.051 |
nights | +0.050 |
time | +0.049 |
morning | +0.049 |
Neg logit contributions | |
boys | -0.060 |
paramed | -0.058 |
chefs | -0.055 |
groom | -0.054 |
Lee | -0.053 |
Token | to |
Token ID | 284 |
Feature activation | +6.41e-02 |
Pos logit contributions | |
day | +0.170 |
night | +0.167 |
nights | +0.163 |
morning | +0.158 |
time | +0.157 |
Neg logit contributions | |
boys | -0.190 |
paramed | -0.189 |
chefs | -0.183 |
Lee | -0.181 |
reporter | -0.174 |
Token | be |
Token ID | 307 |
Feature activation | +7.65e-02 |
Pos logit contributions | |
boys | +0.091 |
Lee | +0.089 |
chefs | +0.088 |
paramed | +0.087 |
reporter | +0.085 |
Neg logit contributions | |
day | -0.115 |
night | -0.113 |
time | -0.108 |
nights | -0.107 |
morning | -0.107 |
Token | okay |
Token ID | 8788 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
boys | +0.118 |
Lee | +0.117 |
paramed | +0.115 |
reporter | +0.111 |
principal | +0.110 |
Neg logit contributions | |
day | -0.126 |
night | -0.124 |
nights | -0.123 |
time | -0.122 |
morning | -0.118 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
wrists | +0.070 |
boys | +0.066 |
paramed | +0.066 |
reporter | +0.063 |
tripod | +0.062 |
Neg logit contributions | |
day | -0.073 |
time | -0.069 |
night | -0.068 |
ie | -0.066 |
of | -0.066 |
Token | The |
Token ID | 383 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
boys | +0.310 |
chefs | +0.308 |
groom | +0.295 |
liar | +0.288 |
paramed | +0.286 |
Neg logit contributions | |
day | -0.210 |
morning | -0.198 |
night | -0.197 |
Along | -0.191 |
evening | -0.178 |
Token | paramed |
Token ID | 34581 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
boys | +0.040 |
wrists | +0.038 |
chefs | +0.036 |
Lee | +0.033 |
ah | +0.031 |
Neg logit contributions | |
day | -0.068 |
night | -0.063 |
morning | -0.063 |
time | -0.063 |
nights | -0.060 |
Token | ic |
Token ID | 291 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
boys | +0.614 |
chefs | +0.595 |
Lee | +0.586 |
paramed | +0.577 |
groom | +0.572 |
Neg logit contributions | |
day | -0.306 |
nights | -0.294 |
night | -0.290 |
afternoon | -0.274 |
morning | -0.264 |
Token | smiled |
Token ID | 13541 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
boys | +0.064 |
chefs | +0.061 |
Lee | +0.059 |
wrists | +0.056 |
paramed | +0.056 |
Neg logit contributions | |
day | -0.061 |
night | -0.057 |
nights | -0.056 |
ie | -0.055 |
time | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
boys | +0.037 |
paramed | +0.037 |
chefs | +0.036 |
Lee | +0.035 |
printer | +0.033 |
Neg logit contributions | |
day | -0.033 |
night | -0.032 |
nights | -0.032 |
morning | -0.030 |
time | -0.030 |
Token | said |
Token ID | 531 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
day | +0.050 |
night | +0.049 |
morning | +0.047 |
nights | +0.045 |
evening | +0.045 |
Neg logit contributions | |
principal | -0.058 |
boys | -0.057 |
jury | -0.055 |
chefs | -0.054 |
liar | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | +7.72e-02 |
Pos logit contributions | |
boys | +0.025 |
principal | +0.024 |
chefs | +0.024 |
wrists | +0.023 |
reporter | +0.023 |
Neg logit contributions | |
day | -0.010 |
morning | -0.009 |
night | -0.009 |
ly | -0.009 |
time | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
paramed | +0.105 |
Lee | +0.100 |
chefs | +0.099 |
boys | +0.099 |
principal | +0.097 |
Neg logit contributions | |
day | -0.097 |
night | -0.096 |
nights | -0.092 |
time | -0.091 |
morning | -0.090 |
Token | so |
Token ID | 523 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
paramed | +0.060 |
boys | +0.059 |
Lee | +0.057 |
chefs | +0.056 |
principal | +0.055 |
Neg logit contributions | |
day | -0.063 |
time | -0.062 |
night | -0.062 |
nights | -0.059 |
morning | -0.058 |
Token | clever |
Token ID | 14169 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
paramed | +0.027 |
boys | +0.027 |
chefs | +0.026 |
Lee | +0.025 |
printer | +0.024 |
Neg logit contributions | |
nights | -0.027 |
day | -0.026 |
night | -0.026 |
morning | -0.025 |
afternoon | -0.025 |
Token | of |
Token ID | 286 |
Feature activation | -9.28e-02 |
Pos logit contributions | |
boys | +0.067 |
chefs | +0.064 |
paramed | +0.064 |
Lee | +0.062 |
printer | +0.058 |
Neg logit contributions | |
day | -0.077 |
nights | -0.077 |
night | -0.076 |
afternoon | -0.073 |
morning | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
day | +0.138 |
night | +0.134 |
nights | +0.132 |
time | +0.128 |
morning | +0.126 |
Neg logit contributions | |
paramed | -0.165 |
boys | -0.161 |
Lee | -0.154 |
chefs | -0.153 |
principal | -0.148 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
boys | +0.058 |
paramed | +0.053 |
Lee | +0.053 |
reporter | +0.051 |
chefs | +0.051 |
Neg logit contributions | |
day | -0.107 |
night | -0.101 |
time | -0.099 |
morning | -0.099 |
afternoon | -0.097 |
Token | m |
Token ID | 76 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
paramed | +0.495 |
boys | +0.491 |
Lee | +0.467 |
principal | +0.462 |
wrists | +0.462 |
Neg logit contributions | |
day | -0.452 |
night | -0.415 |
nights | -0.400 |
morning | -0.399 |
afternoon | -0.398 |
Token | unk |
Token ID | 2954 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
day | +0.036 |
night | +0.035 |
time | +0.034 |
nights | +0.034 |
morning | +0.033 |
Neg logit contributions | |
boys | -0.035 |
paramed | -0.033 |
chefs | -0.032 |
Lee | -0.032 |
reporter | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | +7.73e-02 |
Pos logit contributions | |
boys | +0.164 |
paramed | +0.162 |
chefs | +0.157 |
Lee | +0.155 |
printer | +0.146 |
Neg logit contributions | |
day | -0.159 |
night | -0.156 |
nights | -0.156 |
morning | -0.148 |
afternoon | -0.147 |
Token | hide |
Token ID | 7808 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
paramed | +0.119 |
boys | +0.119 |
chefs | +0.118 |
Lee | +0.114 |
printer | +0.108 |
Neg logit contributions | |
day | -0.129 |
night | -0.127 |
time | -0.122 |
morning | -0.120 |
nights | -0.118 |
Token | from |
Token ID | 422 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
paramed | +0.073 |
boys | +0.070 |
Lee | +0.067 |
principal | +0.067 |
reporter | +0.067 |
Neg logit contributions | |
day | -0.056 |
night | -0.055 |
nights | -0.055 |
time | -0.052 |
fy | -0.051 |
Token | a |
Token ID | 257 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
paramed | +0.062 |
boys | +0.061 |
Lee | +0.059 |
chefs | +0.057 |
reporter | +0.057 |
Neg logit contributions | |
day | -0.078 |
night | -0.078 |
nights | -0.076 |
time | -0.073 |
morning | -0.073 |
Token | big |
Token ID | 1263 |
Feature activation | -4.67e-03 |
Pos logit contributions | |
boys | +0.092 |
Lee | +0.087 |
paramed | +0.086 |
chefs | +0.084 |
wrists | +0.083 |
Neg logit contributions | |
day | -0.065 |
night | -0.061 |
time | -0.060 |
morning | -0.056 |
afternoon | -0.056 |
Token | jar |
Token ID | 17379 |
Feature activation | +9.86e-02 |
Pos logit contributions | |
day | +0.008 |
night | +0.008 |
time | +0.007 |
morning | +0.007 |
afternoon | +0.007 |
Neg logit contributions | |
boys | -0.008 |
paramed | -0.007 |
chefs | -0.007 |
groom | -0.007 |
Fighting | -0.007 |
Token | of |
Token ID | 286 |
Feature activation | -8.94e-02 |
Pos logit contributions | |
boys | +0.231 |
paramed | +0.225 |
chefs | +0.223 |
Lee | +0.219 |
printer | +0.217 |
Neg logit contributions | |
nights | -0.076 |
night | -0.075 |
day | -0.072 |
morning | -0.067 |
evening | -0.065 |
Token | chocolate |
Token ID | 11311 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
day | +0.154 |
night | +0.152 |
nights | +0.147 |
time | +0.143 |
morning | +0.143 |
Neg logit contributions | |
boys | -0.138 |
paramed | -0.133 |
Lee | -0.129 |
chefs | -0.124 |
reporter | -0.122 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
boys | +0.090 |
paramed | +0.089 |
chefs | +0.085 |
Lee | +0.084 |
principal | +0.080 |
Neg logit contributions | |
day | -0.091 |
night | -0.090 |
nights | -0.088 |
morning | -0.085 |
time | -0.084 |
Token | cookies |
Token ID | 14746 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
paramed | +0.414 |
boys | +0.411 |
Lee | +0.383 |
chefs | +0.375 |
reporter | +0.371 |
Neg logit contributions | |
day | -0.517 |
night | -0.503 |
nights | -0.498 |
time | -0.481 |
afternoon | -0.479 |
Token | hidden |
Token ID | 7104 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
paramed | +0.252 |
boys | +0.251 |
chefs | +0.242 |
Lee | +0.239 |
principal | +0.230 |
Neg logit contributions | |
day | -0.186 |
night | -0.181 |
nights | -0.177 |
time | -0.169 |
morning | -0.168 |
Token | inside |
Token ID | 2641 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
day | +0.072 |
night | +0.071 |
nights | +0.070 |
morning | +0.068 |
evening | +0.067 |
Neg logit contributions | |
boys | -0.092 |
chefs | -0.089 |
Lee | -0.085 |
paramed | -0.083 |
printer | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | +6.58e-04 |
Pos logit contributions | |
boys | +0.071 |
paramed | +0.070 |
chefs | +0.070 |
Lee | +0.068 |
printer | +0.065 |
Neg logit contributions | |
day | -0.052 |
nights | -0.051 |
night | -0.051 |
fy | -0.048 |
morning | -0.048 |
Token | Both |
Token ID | 5747 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
boys | +0.001 |
paramed | +0.001 |
chefs | +0.001 |
Lee | +0.001 |
groom | +0.001 |
Neg logit contributions | |
day | -0.001 |
night | -0.001 |
nights | -0.001 |
morning | -0.001 |
time | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
day | +0.038 |
nights | +0.037 |
night | +0.037 |
afternoon | +0.036 |
morning | +0.036 |
Neg logit contributions | |
boys | -0.025 |
paramed | -0.024 |
chefs | -0.023 |
Lee | -0.023 |
groom | -0.021 |
Token | special |
Token ID | 2041 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
boys | +0.077 |
Lee | +0.072 |
chefs | +0.072 |
paramed | +0.071 |
principal | +0.070 |
Neg logit contributions | |
day | -0.063 |
night | -0.060 |
time | -0.059 |
fy | -0.057 |
morning | -0.056 |
Token | treat |
Token ID | 2190 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
day | +0.117 |
night | +0.112 |
time | +0.109 |
morning | +0.108 |
nights | +0.104 |
Neg logit contributions | |
boys | -0.107 |
chefs | -0.100 |
paramed | -0.099 |
Lee | -0.098 |
principal | -0.095 |
Token | for |
Token ID | 329 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
boys | +0.279 |
paramed | +0.277 |
chefs | +0.270 |
Lee | +0.265 |
principal | +0.256 |
Neg logit contributions | |
day | -0.158 |
night | -0.154 |
nights | -0.151 |
morning | -0.143 |
time | -0.141 |
Token | the |
Token ID | 262 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
boys | +0.054 |
paramed | +0.053 |
chefs | +0.051 |
Lee | +0.050 |
reporter | +0.047 |
Neg logit contributions | |
day | -0.065 |
night | -0.064 |
nights | -0.064 |
morning | -0.061 |
time | -0.061 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
boys | +0.060 |
paramed | +0.059 |
chefs | +0.057 |
Lee | +0.056 |
reporter | +0.052 |
Neg logit contributions | |
day | -0.095 |
night | -0.093 |
nights | -0.092 |
morning | -0.089 |
time | -0.089 |
Token | m |
Token ID | 76 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
boys | +0.470 |
paramed | +0.467 |
chefs | +0.444 |
Lee | +0.443 |
principal | +0.432 |
Neg logit contributions | |
day | -0.460 |
night | -0.438 |
nights | -0.428 |
morning | -0.417 |
afternoon | -0.415 |
Token | unk |
Token ID | 2954 |
Feature activation | +9.94e-02 |
Pos logit contributions | |
nights | +0.035 |
night | +0.035 |
day | +0.035 |
fy | +0.033 |
afternoon | +0.033 |
Neg logit contributions | |
paramed | -0.029 |
boys | -0.028 |
chefs | -0.028 |
Lee | -0.027 |
printer | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -5.96e-04 |
Pos logit contributions | |
boys | +0.177 |
paramed | +0.171 |
Lee | +0.167 |
chefs | +0.167 |
printer | +0.159 |
Neg logit contributions | |
nights | -0.136 |
day | -0.133 |
night | -0.131 |
afternoon | -0.125 |
Along | -0.124 |
Token | She |
Token ID | 1375 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
day | +0.001 |
nights | +0.001 |
night | +0.001 |
afternoon | +0.001 |
time | +0.001 |
Neg logit contributions | |
paramed | -0.001 |
boys | -0.001 |
Lee | -0.001 |
chefs | -0.001 |
printer | -0.001 |
Token | said |
Token ID | 531 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
day | +0.020 |
night | +0.019 |
nights | +0.019 |
time | +0.018 |
fy | +0.018 |
Neg logit contributions | |
paramed | -0.020 |
Lee | -0.019 |
boys | -0.019 |
chefs | -0.019 |
principal | -0.018 |
Token | place |
Token ID | 1295 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
day | +0.058 |
night | +0.058 |
time | +0.058 |
nights | +0.054 |
afternoon | +0.053 |
Neg logit contributions | |
paramed | -0.048 |
boys | -0.047 |
principal | -0.043 |
Lee | -0.043 |
chefs | -0.042 |
Token | ! |
Token ID | 0 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
day | +0.016 |
night | +0.015 |
nights | +0.015 |
time | +0.015 |
morning | +0.014 |
Neg logit contributions | |
paramed | -0.020 |
boys | -0.020 |
chefs | -0.019 |
Lee | -0.019 |
groom | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
boys | +0.049 |
paramed | +0.049 |
chefs | +0.047 |
Lee | +0.046 |
groom | +0.045 |
Neg logit contributions | |
day | -0.040 |
night | -0.039 |
nights | -0.039 |
morning | -0.037 |
time | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
boys | +0.032 |
paramed | +0.031 |
chefs | +0.031 |
Lee | +0.030 |
groom | +0.029 |
Neg logit contributions | |
day | -0.022 |
night | -0.021 |
nights | -0.021 |
time | -0.020 |
morning | -0.020 |
Token | The |
Token ID | 464 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
boys | +0.030 |
Lee | +0.029 |
chefs | +0.028 |
paramed | +0.028 |
wrists | +0.027 |
Neg logit contributions | |
day | -0.022 |
nights | -0.022 |
time | -0.021 |
Along | -0.021 |
afternoon | -0.021 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
boys | +0.020 |
paramed | +0.019 |
Lee | +0.019 |
chefs | +0.019 |
reporter | +0.018 |
Neg logit contributions | |
day | -0.025 |
night | -0.024 |
nights | -0.024 |
morning | -0.023 |
time | -0.023 |
Token | mun |
Token ID | 6199 |
Feature activation | +7.05e-02 |
Pos logit contributions | |
boys | +0.537 |
paramed | +0.532 |
chefs | +0.515 |
Lee | +0.511 |
groom | +0.500 |
Neg logit contributions | |
day | -0.405 |
night | -0.371 |
nights | -0.365 |
afternoon | -0.361 |
morning | -0.353 |
Token | ks |
Token ID | 591 |
Feature activation | +4.33e-02 |
Pos logit contributions | |
boys | +0.091 |
paramed | +0.089 |
chefs | +0.085 |
Lee | +0.084 |
principal | +0.079 |
Neg logit contributions | |
day | -0.133 |
night | -0.130 |
nights | -0.130 |
time | -0.126 |
morning | -0.124 |
Token | didn |
Token ID | 1422 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
boys | +0.073 |
paramed | +0.072 |
chefs | +0.071 |
Lee | +0.070 |
principal | +0.066 |
Neg logit contributions | |
day | -0.064 |
night | -0.063 |
nights | -0.062 |
time | -0.060 |
morning | -0.059 |
Token | 't |
Token ID | 470 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
nights | +0.042 |
day | +0.041 |
night | +0.040 |
afternoon | +0.039 |
fy | +0.039 |
Neg logit contributions | |
paramed | -0.037 |
boys | -0.037 |
Lee | -0.035 |
chefs | -0.034 |
printer | -0.033 |
Token | know |
Token ID | 760 |
Feature activation | +5.98e-02 |
Pos logit contributions | |
boys | +0.018 |
paramed | +0.018 |
chefs | +0.017 |
Lee | +0.017 |
reporter | +0.016 |
Neg logit contributions | |
day | -0.021 |
night | -0.021 |
nights | -0.021 |
morning | -0.020 |
afternoon | -0.020 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
nights | +0.102 |
day | +0.102 |
night | +0.100 |
afternoon | +0.096 |
morning | +0.096 |
Neg logit contributions | |
boys | -0.095 |
paramed | -0.091 |
chefs | -0.090 |
Lee | -0.089 |
groom | -0.083 |
Token | |
Token ID | 220 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
paramed | +0.283 |
boys | +0.283 |
Lee | +0.280 |
chefs | +0.267 |
printer | +0.254 |
Neg logit contributions | |
nights | -0.316 |
day | -0.315 |
night | -0.309 |
fy | -0.300 |
time | -0.298 |
Token |
|
Token ID | 198 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
day | +0.017 |
night | +0.017 |
nights | +0.017 |
morning | +0.016 |
afternoon | +0.016 |
Neg logit contributions | |
boys | -0.020 |
paramed | -0.020 |
chefs | -0.019 |
Lee | -0.019 |
principal | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
boys | +0.034 |
paramed | +0.033 |
chefs | +0.032 |
Lee | +0.032 |
printer | +0.030 |
Neg logit contributions | |
day | -0.028 |
night | -0.027 |
nights | -0.027 |
morning | -0.025 |
afternoon | -0.025 |
Token | The |
Token ID | 464 |
Feature activation | +5.85e-03 |
Pos logit contributions | |
boys | +0.032 |
chefs | +0.032 |
Lee | +0.032 |
paramed | +0.031 |
printer | +0.030 |
Neg logit contributions | |
day | -0.026 |
nights | -0.025 |
night | -0.024 |
afternoon | -0.024 |
Along | -0.024 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
boys | +0.009 |
Lee | +0.008 |
chefs | +0.008 |
reporter | +0.008 |
paramed | +0.008 |
Neg logit contributions | |
day | -0.011 |
night | -0.010 |
time | -0.010 |
morning | -0.010 |
afternoon | -0.009 |
Token | m |
Token ID | 76 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
boys | +0.535 |
paramed | +0.526 |
Lee | +0.510 |
chefs | +0.509 |
principal | +0.498 |
Neg logit contributions | |
day | -0.409 |
night | -0.370 |
afternoon | -0.365 |
nights | -0.358 |
morning | -0.354 |
Token | unk |
Token ID | 2954 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
day | +0.041 |
night | +0.040 |
nights | +0.040 |
time | +0.039 |
morning | +0.039 |
Neg logit contributions | |
boys | -0.032 |
paramed | -0.030 |
chefs | -0.030 |
Lee | -0.029 |
reporter | -0.028 |
Token | finished |
Token ID | 5201 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
boys | +0.130 |
paramed | +0.129 |
chefs | +0.127 |
Lee | +0.122 |
reporter | +0.114 |
Neg logit contributions | |
day | -0.181 |
night | -0.178 |
nights | -0.176 |
morning | -0.170 |
afternoon | -0.169 |
Token | the |
Token ID | 262 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
boys | +0.144 |
paramed | +0.142 |
chefs | +0.137 |
reporter | +0.132 |
Lee | +0.131 |
Neg logit contributions | |
day | -0.145 |
night | -0.141 |
nights | -0.135 |
time | -0.134 |
morning | -0.133 |
Token | treat |
Token ID | 2190 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
paramed | +0.053 |
boys | +0.053 |
chefs | +0.050 |
Lee | +0.049 |
printer | +0.046 |
Neg logit contributions | |
nights | -0.089 |
day | -0.088 |
night | -0.088 |
fy | -0.084 |
morning | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
boys | +0.039 |
paramed | +0.038 |
Lee | +0.036 |
chefs | +0.036 |
printer | +0.035 |
Neg logit contributions | |
nights | -0.031 |
day | -0.029 |
night | -0.029 |
fy | -0.028 |
afternoon | -0.028 |
Token | It |
Token ID | 632 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
day | +0.017 |
night | +0.017 |
nights | +0.017 |
morning | +0.016 |
Along | +0.016 |
Neg logit contributions | |
boys | -0.022 |
paramed | -0.021 |
chefs | -0.021 |
Lee | -0.020 |
groom | -0.020 |
Token | was |
Token ID | 373 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
boys | +0.143 |
paramed | +0.142 |
chefs | +0.139 |
Lee | +0.137 |
principal | +0.130 |
Neg logit contributions | |
day | -0.146 |
night | -0.143 |
nights | -0.141 |
time | -0.135 |
morning | -0.135 |
Token | a |
Token ID | 257 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
boys | +0.055 |
paramed | +0.054 |
chefs | +0.052 |
Lee | +0.051 |
principal | +0.049 |
Neg logit contributions | |
day | -0.070 |
night | -0.069 |
time | -0.067 |
nights | -0.067 |
morning | -0.065 |
Token | small |
Token ID | 1402 |
Feature activation | -8.47e-03 |
Pos logit contributions | |
boys | +0.070 |
chefs | +0.067 |
Lee | +0.066 |
paramed | +0.065 |
wrists | +0.063 |
Neg logit contributions | |
day | -0.059 |
night | -0.056 |
time | -0.054 |
morning | -0.052 |
evening | -0.051 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
day | +0.015 |
night | +0.015 |
morning | +0.014 |
time | +0.013 |
nights | +0.013 |
Neg logit contributions | |
boys | -0.013 |
paramed | -0.013 |
chefs | -0.013 |
principal | -0.012 |
Lee | -0.012 |
Token | m |
Token ID | 76 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
paramed | +0.503 |
boys | +0.502 |
chefs | +0.481 |
Lee | +0.478 |
principal | +0.465 |
Neg logit contributions | |
day | -0.418 |
night | -0.401 |
nights | -0.392 |
morning | -0.376 |
afternoon | -0.375 |
Token | unk |
Token ID | 2954 |
Feature activation | +9.41e-02 |
Pos logit contributions | |
day | +0.046 |
nights | +0.046 |
night | +0.046 |
morning | +0.044 |
fy | +0.044 |
Neg logit contributions | |
paramed | -0.033 |
boys | -0.033 |
chefs | -0.032 |
Lee | -0.031 |
printer | -0.029 |
Token | ! |
Token ID | 0 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
boys | +0.173 |
paramed | +0.172 |
chefs | +0.168 |
Lee | +0.165 |
principal | +0.158 |
Neg logit contributions | |
day | -0.124 |
night | -0.121 |
nights | -0.120 |
morning | -0.113 |
time | -0.112 |
Token | The |
Token ID | 383 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
boys | +0.041 |
paramed | +0.040 |
chefs | +0.039 |
Lee | +0.038 |
reporter | +0.037 |
Neg logit contributions | |
day | -0.033 |
night | -0.033 |
nights | -0.032 |
morning | -0.031 |
time | -0.031 |
Token | princess |
Token ID | 21752 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
boys | +0.083 |
Lee | +0.077 |
paramed | +0.076 |
reporter | +0.075 |
chefs | +0.075 |
Neg logit contributions | |
day | -0.081 |
night | -0.077 |
time | -0.074 |
morning | -0.074 |
nights | -0.072 |
Token | boy |
Token ID | 2933 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
boys | +0.060 |
reporter | +0.056 |
Lee | +0.053 |
groom | +0.052 |
paramed | +0.050 |
Neg logit contributions | |
day | -0.113 |
night | -0.105 |
time | -0.102 |
morning | -0.102 |
afternoon | -0.102 |
Token | finally |
Token ID | 3443 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
chefs | +0.214 |
paramed | +0.214 |
boys | +0.209 |
Lee | +0.202 |
reporter | +0.192 |
Neg logit contributions | |
day | -0.225 |
night | -0.222 |
nights | -0.218 |
time | -0.208 |
morning | -0.208 |
Token | understood |
Token ID | 7247 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
day | +0.048 |
night | +0.048 |
time | +0.046 |
nights | +0.045 |
morning | +0.044 |
Neg logit contributions | |
paramed | -0.045 |
chefs | -0.045 |
boys | -0.045 |
Lee | -0.043 |
printer | -0.040 |
Token | why |
Token ID | 1521 |
Feature activation | +9.76e-02 |
Pos logit contributions | |
boys | +0.212 |
paramed | +0.208 |
chefs | +0.198 |
Lee | +0.190 |
reporter | +0.188 |
Neg logit contributions | |
day | -0.209 |
night | -0.202 |
time | -0.199 |
nights | -0.193 |
morning | -0.188 |
Token | the |
Token ID | 262 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
boys | +0.149 |
paramed | +0.146 |
chefs | +0.141 |
Lee | +0.134 |
reporter | +0.133 |
Neg logit contributions | |
day | -0.167 |
night | -0.162 |
nights | -0.158 |
time | -0.156 |
morning | -0.153 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
boys | +0.055 |
reporter | +0.049 |
Lee | +0.046 |
chefs | +0.045 |
groom | +0.044 |
Neg logit contributions | |
day | -0.117 |
night | -0.109 |
morning | -0.107 |
time | -0.106 |
afternoon | -0.105 |
Token | m |
Token ID | 76 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
boys | +0.624 |
wrists | +0.597 |
groom | +0.596 |
paramed | +0.596 |
reporter | +0.588 |
Neg logit contributions | |
day | -0.329 |
afternoon | -0.271 |
night | -0.271 |
morning | -0.264 |
time | -0.252 |
Token | unk |
Token ID | 2954 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
day | +0.040 |
night | +0.038 |
time | +0.037 |
nights | +0.037 |
morning | +0.037 |
Neg logit contributions | |
boys | -0.035 |
paramed | -0.033 |
chefs | -0.032 |
reporter | -0.032 |
Lee | -0.031 |
Token | had |
Token ID | 550 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
boys | +0.155 |
paramed | +0.155 |
chefs | +0.151 |
Lee | +0.147 |
principal | +0.139 |
Neg logit contributions | |
day | -0.160 |
night | -0.156 |
nights | -0.152 |
time | -0.147 |
morning | -0.147 |
Token | been |
Token ID | 587 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
day | +0.023 |
night | +0.023 |
time | +0.023 |
nights | +0.022 |
morning | +0.021 |
Neg logit contributions | |
paramed | -0.024 |
boys | -0.023 |
Lee | -0.023 |
chefs | -0.022 |
principal | -0.021 |
Token | so |
Token ID | 523 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
time | +0.078 |
day | +0.076 |
night | +0.071 |
nights | +0.070 |
afternoon | +0.068 |
Neg logit contributions | |
paramed | -0.085 |
boys | -0.080 |
Lee | -0.079 |
chefs | -0.074 |
wrists | -0.073 |
Token | bird |
Token ID | 6512 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
paramed | +0.082 |
boys | +0.082 |
Lee | +0.079 |
chefs | +0.079 |
principal | +0.075 |
Neg logit contributions | |
day | -0.083 |
night | -0.081 |
nights | -0.079 |
time | -0.077 |
morning | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | -6.39e-03 |
Pos logit contributions | |
boys | +0.160 |
paramed | +0.148 |
Lee | +0.147 |
chefs | +0.145 |
printer | +0.139 |
Neg logit contributions | |
nights | -0.124 |
night | -0.118 |
day | -0.118 |
fy | -0.114 |
afternoon | -0.113 |
Token | Luckily |
Token ID | 24262 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
day | +0.008 |
Along | +0.008 |
night | +0.008 |
morning | +0.008 |
nights | +0.008 |
Neg logit contributions | |
boys | -0.012 |
paramed | -0.011 |
chefs | -0.011 |
groom | -0.011 |
reporter | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +9.57e-02 |
Pos logit contributions | |
day | +0.038 |
night | +0.037 |
nights | +0.037 |
morning | +0.035 |
time | +0.035 |
Neg logit contributions | |
paramed | -0.045 |
boys | -0.045 |
chefs | -0.044 |
Lee | -0.042 |
reporter | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
boys | +0.157 |
paramed | +0.156 |
chefs | +0.148 |
wrists | +0.143 |
Lee | +0.142 |
Neg logit contributions | |
day | -0.154 |
night | -0.150 |
nights | -0.146 |
morning | -0.142 |
time | -0.138 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
paramed | +0.080 |
boys | +0.080 |
Lee | +0.078 |
chefs | +0.077 |
principal | +0.074 |
Neg logit contributions | |
day | -0.083 |
night | -0.080 |
nights | -0.077 |
time | -0.076 |
morning | -0.075 |
Token | m |
Token ID | 76 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
paramed | +0.508 |
Lee | +0.495 |
boys | +0.493 |
principal | +0.491 |
chefs | +0.483 |
Neg logit contributions | |
day | -0.430 |
night | -0.381 |
ie | -0.373 |
afternoon | -0.373 |
nights | -0.371 |
Token | unk |
Token ID | 2954 |
Feature activation | +9.37e-02 |
Pos logit contributions | |
day | +0.042 |
nights | +0.042 |
night | +0.042 |
morning | +0.040 |
afternoon | +0.040 |
Neg logit contributions | |
boys | -0.036 |
paramed | -0.036 |
chefs | -0.034 |
Lee | -0.034 |
printer | -0.032 |
Token | said |
Token ID | 531 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
paramed | +0.136 |
chefs | +0.134 |
boys | +0.133 |
Lee | +0.128 |
principal | +0.121 |
Neg logit contributions | |
day | -0.163 |
night | -0.159 |
nights | -0.155 |
time | -0.153 |
morning | -0.151 |
Token | they |
Token ID | 484 |
Feature activation | +4.24e-03 |
Pos logit contributions | |
boys | +0.045 |
chefs | +0.043 |
paramed | +0.043 |
principal | +0.042 |
Lee | +0.041 |
Neg logit contributions | |
day | -0.037 |
night | -0.035 |
time | -0.035 |
morning | -0.035 |
nights | -0.034 |
Token | should |
Token ID | 815 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
paramed | +0.008 |
boys | +0.007 |
chefs | +0.007 |
Lee | +0.007 |
principal | +0.007 |
Neg logit contributions | |
day | -0.006 |
night | -0.006 |
nights | -0.006 |
time | -0.006 |
morning | -0.005 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
boys | +0.076 |
chefs | +0.072 |
paramed | +0.072 |
Lee | +0.071 |
principal | +0.068 |
Neg logit contributions | |
day | -0.067 |
night | -0.064 |
time | -0.061 |
morning | -0.059 |
afternoon | -0.058 |
Token | m |
Token ID | 76 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
paramed | +0.515 |
boys | +0.513 |
chefs | +0.491 |
Lee | +0.487 |
groom | +0.480 |
Neg logit contributions | |
day | -0.398 |
night | -0.379 |
nights | -0.370 |
time | -0.356 |
afternoon | -0.355 |
Token | unk |
Token ID | 2954 |
Feature activation | +9.49e-02 |
Pos logit contributions | |
day | +0.051 |
nights | +0.050 |
night | +0.050 |
morning | +0.048 |
afternoon | +0.048 |
Neg logit contributions | |
boys | -0.033 |
paramed | -0.033 |
chefs | -0.032 |
Lee | -0.031 |
printer | -0.029 |
Token | ! |
Token ID | 0 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
boys | +0.174 |
paramed | +0.172 |
chefs | +0.168 |
Lee | +0.166 |
reporter | +0.158 |
Neg logit contributions | |
day | -0.126 |
night | -0.123 |
nights | -0.122 |
morning | -0.115 |
time | -0.114 |
Token | The |
Token ID | 383 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
boys | +0.047 |
chefs | +0.044 |
paramed | +0.043 |
groom | +0.041 |
Lee | +0.041 |
Neg logit contributions | |
day | -0.043 |
night | -0.042 |
morning | -0.041 |
Along | -0.040 |
time | -0.040 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
boys | +0.061 |
chefs | +0.055 |
reporter | +0.055 |
wrists | +0.052 |
principal | +0.052 |
Neg logit contributions | |
day | -0.106 |
night | -0.100 |
time | -0.097 |
morning | -0.097 |
afternoon | -0.093 |
Token | m |
Token ID | 76 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
boys | +0.536 |
paramed | +0.533 |
groom | +0.525 |
wrists | +0.518 |
principal | +0.517 |
Neg logit contributions | |
day | -0.390 |
night | -0.344 |
nights | -0.338 |
afternoon | -0.337 |
time | -0.329 |
Token | unk |
Token ID | 2954 |
Feature activation | +9.59e-02 |
Pos logit contributions | |
day | +0.050 |
night | +0.049 |
nights | +0.049 |
time | +0.047 |
morning | +0.047 |
Neg logit contributions | |
boys | -0.036 |
paramed | -0.035 |
chefs | -0.034 |
Lee | -0.033 |
principal | -0.031 |
Token | said |
Token ID | 531 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
paramed | +0.147 |
boys | +0.146 |
chefs | +0.146 |
Lee | +0.140 |
principal | +0.134 |
Neg logit contributions | |
day | -0.157 |
night | -0.152 |
nights | -0.150 |
time | -0.145 |
morning | -0.144 |
Token | to |
Token ID | 284 |
Feature activation | +7.59e-02 |
Pos logit contributions | |
boys | +0.054 |
chefs | +0.053 |
principal | +0.052 |
wrists | +0.050 |
paramed | +0.049 |
Neg logit contributions | |
of | -0.028 |
day | -0.028 |
morning | -0.026 |
night | -0.025 |
time | -0.025 |
Token | Sally |
Token ID | 25737 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
boys | +0.118 |
paramed | +0.114 |
chefs | +0.114 |
Lee | +0.107 |
reporter | +0.105 |
Neg logit contributions | |
day | -0.131 |
night | -0.126 |
time | -0.123 |
morning | -0.121 |
nights | -0.119 |
Token | frog |
Token ID | 21264 |
Feature activation | +8.32e-02 |
Pos logit contributions | |
boys | +0.068 |
Lee | +0.066 |
reporter | +0.066 |
paramed | +0.064 |
chefs | +0.063 |
Neg logit contributions | |
day | -0.101 |
night | -0.092 |
time | -0.090 |
morning | -0.089 |
afternoon | -0.089 |
Token | agreed |
Token ID | 4987 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
paramed | +0.117 |
chefs | +0.116 |
boys | +0.116 |
Lee | +0.114 |
principal | +0.107 |
Neg logit contributions | |
day | -0.148 |
night | -0.144 |
nights | -0.141 |
time | -0.138 |
morning | -0.137 |
Token | , |
Token ID | 11 |
Feature activation | +9.98e-02 |
Pos logit contributions | |
boys | +0.192 |
paramed | +0.190 |
chefs | +0.185 |
Lee | +0.181 |
principal | +0.173 |
Neg logit contributions | |
day | -0.165 |
night | -0.161 |
nights | -0.159 |
time | -0.152 |
morning | -0.152 |
Token | and |
Token ID | 290 |
Feature activation | -9.75e-03 |
Pos logit contributions | |
boys | +0.176 |
chefs | +0.172 |
paramed | +0.172 |
Lee | +0.164 |
principal | +0.160 |
Neg logit contributions | |
day | -0.143 |
night | -0.138 |
nights | -0.137 |
morning | -0.131 |
time | -0.130 |
Token | the |
Token ID | 262 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
day | +0.017 |
night | +0.016 |
nights | +0.016 |
morning | +0.015 |
afternoon | +0.015 |
Neg logit contributions | |
paramed | -0.015 |
boys | -0.014 |
chefs | -0.014 |
principal | -0.014 |
Lee | -0.014 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
boys | +0.084 |
Lee | +0.083 |
reporter | +0.082 |
paramed | +0.080 |
chefs | +0.080 |
Neg logit contributions | |
day | -0.080 |
night | -0.073 |
time | -0.070 |
morning | -0.070 |
afternoon | -0.068 |
Token | m |
Token ID | 76 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
boys | +0.551 |
paramed | +0.541 |
Lee | +0.537 |
principal | +0.536 |
chefs | +0.533 |
Neg logit contributions | |
day | -0.385 |
night | -0.343 |
afternoon | -0.328 |
morning | -0.327 |
nights | -0.320 |
Token | unk |
Token ID | 2954 |
Feature activation | +9.49e-02 |
Pos logit contributions | |
day | +0.044 |
night | +0.044 |
nights | +0.043 |
time | +0.042 |
morning | +0.042 |
Neg logit contributions | |
boys | -0.034 |
paramed | -0.033 |
chefs | -0.032 |
Lee | -0.032 |
principal | -0.030 |
Token | agreed |
Token ID | 4987 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
boys | +0.131 |
paramed | +0.131 |
chefs | +0.128 |
Lee | +0.125 |
principal | +0.117 |
Neg logit contributions | |
day | -0.169 |
night | -0.166 |
nights | -0.164 |
morning | -0.158 |
time | -0.157 |
Token | too |
Token ID | 1165 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
boys | +0.211 |
paramed | +0.208 |
chefs | +0.203 |
Lee | +0.201 |
reporter | +0.191 |
Neg logit contributions | |
day | -0.147 |
night | -0.145 |
nights | -0.145 |
morning | -0.135 |
afternoon | -0.135 |
Token | . |
Token ID | 13 |
Feature activation | +2.03e-03 |
Pos logit contributions | |
boys | +0.067 |
Lee | +0.063 |
paramed | +0.063 |
chefs | +0.063 |
reporter | +0.059 |
Neg logit contributions | |
nights | -0.047 |
day | -0.047 |
night | -0.046 |
afternoon | -0.044 |
morning | -0.044 |
Token | animal |
Token ID | 5044 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
day | +0.003 |
night | +0.003 |
time | +0.003 |
morning | +0.003 |
nights | +0.003 |
Neg logit contributions | |
boys | -0.004 |
paramed | -0.003 |
chefs | -0.003 |
principal | -0.003 |
Fighting | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -7.50e-03 |
Pos logit contributions | |
boys | +0.025 |
paramed | +0.024 |
chefs | +0.024 |
Lee | +0.023 |
printer | +0.022 |
Neg logit contributions | |
day | -0.022 |
nights | -0.022 |
night | -0.022 |
morning | -0.021 |
afternoon | -0.021 |
Token | It |
Token ID | 632 |
Feature activation | +9.34e-02 |
Pos logit contributions | |
day | +0.011 |
night | +0.011 |
Along | +0.011 |
morning | +0.011 |
time | +0.010 |
Neg logit contributions | |
boys | -0.013 |
chefs | -0.012 |
paramed | -0.012 |
groom | -0.012 |
Lee | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
paramed | +0.157 |
chefs | +0.155 |
boys | +0.154 |
Lee | +0.149 |
principal | +0.146 |
Neg logit contributions | |
day | -0.143 |
night | -0.138 |
nights | -0.134 |
time | -0.132 |
morning | -0.129 |
Token | a |
Token ID | 257 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
boys | +0.062 |
paramed | +0.062 |
chefs | +0.060 |
Lee | +0.059 |
principal | +0.057 |
Neg logit contributions | |
day | -0.069 |
night | -0.068 |
time | -0.066 |
nights | -0.066 |
morning | -0.064 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
boys | +0.076 |
chefs | +0.072 |
paramed | +0.072 |
Lee | +0.071 |
principal | +0.068 |
Neg logit contributions | |
day | -0.067 |
night | -0.064 |
time | -0.061 |
morning | -0.059 |
afternoon | -0.058 |
Token | m |
Token ID | 76 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
paramed | +0.515 |
boys | +0.513 |
chefs | +0.491 |
Lee | +0.487 |
groom | +0.480 |
Neg logit contributions | |
day | -0.398 |
night | -0.379 |
nights | -0.370 |
time | -0.356 |
afternoon | -0.355 |
Token | unk |
Token ID | 2954 |
Feature activation | +9.49e-02 |
Pos logit contributions | |
day | +0.051 |
nights | +0.050 |
night | +0.050 |
morning | +0.048 |
afternoon | +0.048 |
Neg logit contributions | |
boys | -0.033 |
paramed | -0.033 |
chefs | -0.032 |
Lee | -0.031 |
printer | -0.029 |
Token | ! |
Token ID | 0 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
boys | +0.174 |
paramed | +0.172 |
chefs | +0.168 |
Lee | +0.166 |
reporter | +0.158 |
Neg logit contributions | |
day | -0.126 |
night | -0.123 |
nights | -0.122 |
morning | -0.115 |
time | -0.114 |
Token | The |
Token ID | 383 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
boys | +0.047 |
chefs | +0.044 |
paramed | +0.043 |
groom | +0.041 |
Lee | +0.041 |
Neg logit contributions | |
day | -0.043 |
night | -0.042 |
morning | -0.041 |
Along | -0.040 |
time | -0.040 |
Token | chip |
Token ID | 11594 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
boys | +0.061 |
chefs | +0.055 |
reporter | +0.055 |
wrists | +0.052 |
principal | +0.052 |
Neg logit contributions | |
day | -0.106 |
night | -0.100 |
time | -0.097 |
morning | -0.097 |
afternoon | -0.093 |
Token | is |
Token ID | 318 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
paramed | +0.023 |
boys | +0.023 |
Lee | +0.022 |
chefs | +0.022 |
principal | +0.021 |
Neg logit contributions | |
day | -0.022 |
night | -0.021 |
nights | -0.021 |
morning | -0.020 |
time | -0.020 |
Token | inside |
Token ID | 2641 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
boys | +0.043 |
paramed | +0.042 |
Lee | +0.041 |
chefs | +0.039 |
principal | +0.039 |
Neg logit contributions | |
day | -0.040 |
night | -0.039 |
time | -0.038 |
nights | -0.037 |
morning | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -6.36e-02 |
Pos logit contributions | |
day | +0.026 |
night | +0.025 |
nights | +0.025 |
morning | +0.024 |
time | +0.023 |
Neg logit contributions | |
boys | -0.034 |
paramed | -0.033 |
chefs | -0.032 |
Lee | -0.032 |
reporter | -0.030 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
day | +0.102 |
night | +0.101 |
nights | +0.101 |
morning | +0.095 |
afternoon | +0.095 |
Neg logit contributions | |
boys | -0.098 |
paramed | -0.097 |
chefs | -0.094 |
Lee | -0.093 |
printer | -0.087 |
Token | One |
Token ID | 3198 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
nights | +0.294 |
fy | +0.286 |
day | +0.277 |
night | +0.271 |
afternoon | +0.265 |
Neg logit contributions | |
Lee | -0.275 |
boys | -0.275 |
chefs | -0.273 |
paramed | -0.266 |
principal | -0.250 |
Token | day |
Token ID | 1110 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
fy | +0.214 |
Along | +0.199 |
lessly | +0.198 |
ight | +0.197 |
ly | +0.195 |
Neg logit contributions | |
dads | -0.409 |
mixer | -0.409 |
dad | -0.408 |
band | -0.398 |
paramed | -0.398 |
Token | , |
Token ID | 11 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
nights | +0.616 |
fy | +0.597 |
found | +0.554 |
night | +0.548 |
ingle | +0.547 |
Neg logit contributions | |
Lee | -0.327 |
chefs | -0.307 |
paramed | -0.290 |
boys | -0.286 |
printer | -0.271 |
Token | a |
Token ID | 257 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
boys | +0.041 |
paramed | +0.040 |
chefs | +0.040 |
Lee | +0.039 |
printer | +0.035 |
Neg logit contributions | |
nights | -0.068 |
day | -0.067 |
night | -0.066 |
fy | -0.065 |
morning | -0.063 |
Token | boy |
Token ID | 2933 |
Feature activation | +8.34e-02 |
Pos logit contributions | |
day | +0.031 |
night | +0.029 |
time | +0.028 |
morning | +0.028 |
nights | +0.027 |
Neg logit contributions | |
boys | -0.048 |
paramed | -0.047 |
Lee | -0.046 |
chefs | -0.045 |
wrists | -0.044 |
Token | named |
Token ID | 3706 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
boys | +0.081 |
paramed | +0.080 |
chefs | +0.077 |
Lee | +0.074 |
printer | +0.069 |
Neg logit contributions | |
nights | -0.179 |
day | -0.177 |
night | -0.176 |
fy | -0.173 |
afternoon | -0.170 |
Token | Tim |
Token ID | 5045 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
nights | +0.175 |
day | +0.165 |
fy | +0.163 |
night | +0.163 |
afternoon | +0.163 |
Neg logit contributions | |
boys | -0.061 |
chefs | -0.061 |
Lee | -0.057 |
paramed | -0.056 |
printer | -0.050 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
fy | +0.025 |
Along | +0.024 |
ight | +0.023 |
ingle | +0.023 |
dark | +0.023 |
Neg logit contributions | |
chefs | -0.026 |
boys | -0.023 |
spies | -0.022 |
reporter | -0.022 |
printer | -0.022 |
Token | ane |
Token ID | 1531 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
nights | +0.304 |
fy | +0.295 |
night | +0.292 |
day | +0.284 |
afternoon | +0.275 |
Neg logit contributions | |
boys | -0.379 |
paramed | -0.361 |
Lee | -0.361 |
chefs | -0.351 |
principal | -0.347 |
Token | . |
Token ID | 13 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
nights | +0.082 |
day | +0.081 |
night | +0.081 |
fy | +0.078 |
Along | +0.078 |
Neg logit contributions | |
boys | -0.090 |
paramed | -0.087 |
chefs | -0.083 |
Lee | -0.083 |
printer | -0.081 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
day | +0.098 |
night | +0.097 |
nights | +0.097 |
morning | +0.091 |
afternoon | +0.091 |
Neg logit contributions | |
boys | -0.099 |
paramed | -0.098 |
chefs | -0.095 |
Lee | -0.095 |
printer | -0.089 |
Token | One |
Token ID | 3198 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
fy | +0.354 |
nights | +0.350 |
spine | +0.318 |
ly | +0.317 |
rated | +0.315 |
Neg logit contributions | |
Lee | -0.247 |
chefs | -0.238 |
boys | -0.231 |
paramed | -0.222 |
principal | -0.212 |
Token | day |
Token ID | 1110 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
fy | +0.237 |
Along | +0.222 |
ly | +0.215 |
lessly | +0.215 |
ets | +0.211 |
Neg logit contributions | |
dads | -0.385 |
qu | -0.383 |
dad | -0.383 |
band | -0.379 |
paramed | -0.375 |
Token | , |
Token ID | 11 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
nights | +0.625 |
fy | +0.612 |
found | +0.565 |
ingle | +0.557 |
rated | +0.555 |
Neg logit contributions | |
Lee | -0.306 |
chefs | -0.284 |
paramed | -0.281 |
boys | -0.263 |
printer | -0.252 |
Token | a |
Token ID | 257 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
boys | +0.038 |
paramed | +0.037 |
chefs | +0.037 |
Lee | +0.036 |
printer | +0.032 |
Neg logit contributions | |
nights | -0.068 |
day | -0.066 |
night | -0.065 |
fy | -0.065 |
afternoon | -0.063 |
Token | turkey |
Token ID | 26876 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
day | +0.035 |
night | +0.034 |
nights | +0.032 |
time | +0.032 |
morning | +0.032 |
Neg logit contributions | |
boys | -0.053 |
paramed | -0.052 |
chefs | -0.051 |
Lee | -0.051 |
principal | -0.048 |
Token | had |
Token ID | 550 |
Feature activation | -7.41e-02 |
Pos logit contributions | |
boys | +0.069 |
paramed | +0.067 |
chefs | +0.066 |
Lee | +0.063 |
printer | +0.062 |
Neg logit contributions | |
nights | -0.086 |
day | -0.083 |
fy | -0.083 |
night | -0.082 |
afternoon | -0.080 |
Token | an |
Token ID | 281 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
nights | +0.134 |
day | +0.132 |
fy | +0.131 |
night | +0.130 |
afternoon | +0.126 |
Neg logit contributions | |
boys | -0.097 |
paramed | -0.096 |
chefs | -0.093 |
printer | -0.088 |
Lee | -0.087 |
Token | did |
Token ID | 750 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
nights | +0.144 |
fy | +0.141 |
day | +0.139 |
night | +0.138 |
Along | +0.134 |
Neg logit contributions | |
boys | -0.104 |
paramed | -0.100 |
Lee | -0.100 |
chefs | -0.099 |
printer | -0.090 |
Token | it |
Token ID | 340 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
day | +0.115 |
nights | +0.114 |
night | +0.113 |
morning | +0.109 |
afternoon | +0.109 |
Neg logit contributions | |
boys | -0.058 |
paramed | -0.056 |
chefs | -0.054 |
Lee | -0.053 |
principal | -0.049 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
nights | +0.045 |
day | +0.043 |
night | +0.043 |
fy | +0.041 |
Along | +0.041 |
Neg logit contributions | |
boys | -0.052 |
chefs | -0.049 |
Lee | -0.049 |
paramed | -0.049 |
principal | -0.046 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
boys | +0.197 |
paramed | +0.194 |
chefs | +0.187 |
Lee | +0.183 |
printer | +0.173 |
Neg logit contributions | |
day | -0.220 |
night | -0.216 |
nights | -0.214 |
morning | -0.206 |
time | -0.205 |
Token | One |
Token ID | 3198 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
nights | +0.325 |
fy | +0.319 |
spine | +0.296 |
day | +0.293 |
night | +0.290 |
Neg logit contributions | |
Lee | -0.246 |
chefs | -0.237 |
boys | -0.237 |
paramed | -0.225 |
principal | -0.215 |
Token | day |
Token ID | 1110 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
fy | +0.137 |
Along | +0.131 |
nights | +0.130 |
spine | +0.120 |
lessly | +0.108 |
Neg logit contributions | |
dads | -0.462 |
qu | -0.461 |
paramed | -0.460 |
mixer | -0.457 |
printer | -0.454 |
Token | , |
Token ID | 11 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
nights | +0.553 |
fy | +0.533 |
night | +0.505 |
day | +0.501 |
Along | +0.492 |
Neg logit contributions | |
Lee | -0.362 |
chefs | -0.349 |
boys | -0.346 |
paramed | -0.345 |
printer | -0.324 |
Token | Tim |
Token ID | 5045 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
boys | +0.049 |
paramed | +0.049 |
chefs | +0.046 |
Lee | +0.045 |
principal | +0.043 |
Neg logit contributions | |
day | -0.071 |
night | -0.070 |
nights | -0.068 |
time | -0.067 |
morning | -0.067 |
Token | my |
Token ID | 1820 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
fy | +0.050 |
nights | +0.050 |
Along | +0.047 |
lessly | +0.046 |
way | +0.046 |
Neg logit contributions | |
boys | -0.040 |
chefs | -0.040 |
Lee | -0.038 |
paramed | -0.036 |
printer | -0.035 |
Token | went |
Token ID | 1816 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
nights | +0.242 |
fy | +0.238 |
day | +0.235 |
Along | +0.230 |
night | +0.229 |
Neg logit contributions | |
boys | -0.186 |
chefs | -0.177 |
paramed | -0.176 |
Lee | -0.172 |
printer | -0.165 |
Token | to |
Token ID | 284 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
nights | +0.225 |
ight | +0.220 |
fy | +0.217 |
rated | +0.216 |
ingle | +0.207 |
Neg logit contributions | |
Lee | -0.107 |
boys | -0.101 |
chefs | -0.096 |
paramed | -0.096 |
printer | -0.091 |
Token | their |
Token ID | 511 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
day | +0.257 |
night | +0.253 |
nights | +0.252 |
morning | +0.241 |
afternoon | +0.240 |
Neg logit contributions | |
boys | -0.188 |
paramed | -0.184 |
chefs | -0.178 |
Lee | -0.176 |
printer | -0.164 |
Token | faces |
Token ID | 6698 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
nights | +0.083 |
Along | +0.081 |
night | +0.081 |
day | +0.081 |
fy | +0.080 |
Neg logit contributions | |
boys | -0.072 |
Lee | -0.065 |
paramed | -0.065 |
chefs | -0.065 |
printer | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -6.73e-02 |
Pos logit contributions | |
boys | +0.037 |
chefs | +0.037 |
Lee | +0.036 |
paramed | +0.036 |
reporter | +0.034 |
Neg logit contributions | |
nights | -0.026 |
day | -0.026 |
night | -0.026 |
fy | -0.025 |
Along | -0.024 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
day | +0.107 |
nights | +0.106 |
night | +0.105 |
afternoon | +0.099 |
morning | +0.099 |
Neg logit contributions | |
boys | -0.105 |
paramed | -0.104 |
Lee | -0.101 |
chefs | -0.101 |
principal | -0.094 |
Token | One |
Token ID | 3198 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
nights | +0.343 |
fy | +0.343 |
spine | +0.312 |
day | +0.310 |
ly | +0.309 |
Neg logit contributions | |
Lee | -0.256 |
chefs | -0.247 |
boys | -0.244 |
paramed | -0.235 |
principal | -0.223 |
Token | day |
Token ID | 1110 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
fy | +0.214 |
Along | +0.209 |
nights | +0.203 |
ly | +0.199 |
ight | +0.198 |
Neg logit contributions | |
dads | -0.408 |
qu | -0.397 |
dad | -0.395 |
band | -0.395 |
fountain | -0.394 |
Token | , |
Token ID | 11 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
nights | +0.609 |
fy | +0.583 |
found | +0.543 |
night | +0.542 |
day | +0.537 |
Neg logit contributions | |
Lee | -0.323 |
chefs | -0.303 |
paramed | -0.295 |
boys | -0.281 |
printer | -0.263 |
Token | Mum |
Token ID | 38367 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
boys | +0.032 |
chefs | +0.032 |
paramed | +0.031 |
Lee | +0.031 |
printer | +0.028 |
Neg logit contributions | |
nights | -0.067 |
fy | -0.064 |
day | -0.064 |
night | -0.063 |
afternoon | -0.061 |
Token | was |
Token ID | 373 |
Feature activation | -9.50e-03 |
Pos logit contributions | |
boys | +0.054 |
paramed | +0.053 |
wrists | +0.050 |
principal | +0.050 |
chefs | +0.050 |
Neg logit contributions | |
day | -0.050 |
night | -0.049 |
time | -0.046 |
morning | -0.046 |
nights | -0.045 |
Token | looking |
Token ID | 2045 |
Feature activation | -8.72e-02 |
Pos logit contributions | |
day | +0.016 |
nights | +0.016 |
night | +0.016 |
fy | +0.015 |
afternoon | +0.015 |
Neg logit contributions | |
paramed | -0.014 |
boys | -0.014 |
chefs | -0.014 |
Lee | -0.013 |
printer | -0.012 |
Token | at |
Token ID | 379 |
Feature activation | -6.50e-02 |
Pos logit contributions | |
nights | +0.137 |
fy | +0.133 |
day | +0.132 |
night | +0.129 |
ight | +0.127 |
Neg logit contributions | |
chefs | -0.137 |
paramed | -0.136 |
Lee | -0.128 |
boys | -0.128 |
printer | -0.120 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
day | +0.244 |
night | +0.240 |
nights | +0.239 |
morning | +0.229 |
afternoon | +0.228 |
Neg logit contributions | |
boys | -0.170 |
paramed | -0.168 |
chefs | -0.162 |
Lee | -0.159 |
printer | -0.148 |
Token | named |
Token ID | 3706 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
paramed | +0.027 |
chefs | +0.026 |
boys | +0.026 |
Lee | +0.025 |
principal | +0.024 |
Neg logit contributions | |
day | -0.031 |
night | -0.030 |
nights | -0.029 |
time | -0.029 |
morning | -0.029 |
Token | Lily |
Token ID | 20037 |
Feature activation | -8.76e-03 |
Pos logit contributions | |
day | +0.156 |
night | +0.154 |
nights | +0.154 |
morning | +0.148 |
afternoon | +0.147 |
Neg logit contributions | |
boys | -0.092 |
paramed | -0.090 |
chefs | -0.086 |
Lee | -0.086 |
principal | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -7.02e-02 |
Pos logit contributions | |
nights | +0.012 |
day | +0.012 |
night | +0.012 |
fy | +0.011 |
morning | +0.011 |
Neg logit contributions | |
boys | -0.016 |
paramed | -0.015 |
Lee | -0.015 |
chefs | -0.015 |
groom | -0.014 |
Token | One |
Token ID | 1881 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
nights | +0.112 |
day | +0.109 |
night | +0.108 |
fy | +0.105 |
afternoon | +0.102 |
Neg logit contributions | |
boys | -0.110 |
Lee | -0.110 |
paramed | -0.109 |
chefs | -0.106 |
principal | -0.099 |
Token | day |
Token ID | 1110 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
ides | +0.167 |
ently | +0.165 |
ight | +0.157 |
ets | +0.154 |
spine | +0.152 |
Neg logit contributions | |
paramed | -0.267 |
sprayed | -0.259 |
duck | -0.254 |
qu | -0.251 |
printer | -0.247 |
Token | , |
Token ID | 11 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
nights | +0.553 |
fy | +0.526 |
day | +0.517 |
night | +0.515 |
ight | +0.509 |
Neg logit contributions | |
Lee | -0.358 |
paramed | -0.345 |
chefs | -0.338 |
boys | -0.325 |
printer | -0.300 |
Token | she |
Token ID | 673 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
boys | +0.075 |
Lee | +0.073 |
chefs | +0.073 |
paramed | +0.073 |
groom | +0.068 |
Neg logit contributions | |
nights | -0.073 |
night | -0.071 |
day | -0.071 |
fy | -0.070 |
afternoon | -0.068 |
Token | went |
Token ID | 1816 |
Feature activation | -9.67e-02 |
Pos logit contributions | |
day | +0.310 |
night | +0.305 |
nights | +0.304 |
morning | +0.292 |
afternoon | +0.290 |
Neg logit contributions | |
boys | -0.201 |
paramed | -0.197 |
chefs | -0.190 |
Lee | -0.186 |
printer | -0.173 |
Token | to |
Token ID | 284 |
Feature activation | +4.00e-02 |
Pos logit contributions | |
nights | +0.181 |
fy | +0.173 |
ight | +0.172 |
day | +0.170 |
night | +0.169 |
Neg logit contributions | |
boys | -0.125 |
Lee | -0.121 |
chefs | -0.119 |
paramed | -0.117 |
printer | -0.111 |
Token | the |
Token ID | 262 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
boys | +0.050 |
paramed | +0.050 |
chefs | +0.047 |
Lee | +0.046 |
reporter | +0.043 |
Neg logit contributions | |
day | -0.078 |
night | -0.077 |
nights | -0.075 |
morning | -0.073 |
time | -0.073 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
day | +0.244 |
night | +0.240 |
nights | +0.239 |
morning | +0.229 |
afternoon | +0.228 |
Neg logit contributions | |
boys | -0.170 |
paramed | -0.168 |
chefs | -0.162 |
Lee | -0.159 |
printer | -0.148 |
Token | named |
Token ID | 3706 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
paramed | +0.027 |
chefs | +0.026 |
boys | +0.026 |
Lee | +0.025 |
principal | +0.024 |
Neg logit contributions | |
day | -0.031 |
night | -0.030 |
nights | -0.029 |
time | -0.029 |
morning | -0.029 |
Token | Lily |
Token ID | 20037 |
Feature activation | -8.76e-03 |
Pos logit contributions | |
day | +0.156 |
night | +0.154 |
nights | +0.154 |
morning | +0.148 |
afternoon | +0.147 |
Neg logit contributions | |
boys | -0.092 |
paramed | -0.090 |
chefs | -0.086 |
Lee | -0.086 |
principal | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -7.02e-02 |
Pos logit contributions | |
nights | +0.012 |
day | +0.012 |
night | +0.012 |
fy | +0.011 |
morning | +0.011 |
Neg logit contributions | |
boys | -0.016 |
paramed | -0.015 |
Lee | -0.015 |
chefs | -0.015 |
groom | -0.014 |
Token | One |
Token ID | 1881 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
nights | +0.112 |
day | +0.109 |
night | +0.108 |
fy | +0.105 |
afternoon | +0.102 |
Neg logit contributions | |
boys | -0.110 |
Lee | -0.110 |
paramed | -0.109 |
chefs | -0.106 |
principal | -0.099 |
Token | day |
Token ID | 1110 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
ides | +0.167 |
ently | +0.165 |
ight | +0.157 |
ets | +0.154 |
spine | +0.152 |
Neg logit contributions | |
paramed | -0.267 |
sprayed | -0.259 |
duck | -0.254 |
qu | -0.251 |
printer | -0.247 |
Token | , |
Token ID | 11 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
nights | +0.553 |
fy | +0.526 |
day | +0.517 |
night | +0.515 |
ight | +0.509 |
Neg logit contributions | |
Lee | -0.358 |
paramed | -0.345 |
chefs | -0.338 |
boys | -0.325 |
printer | -0.300 |
Token | Lily |
Token ID | 20037 |
Feature activation | +9.22e-03 |
Pos logit contributions | |
boys | +0.075 |
Lee | +0.073 |
chefs | +0.073 |
paramed | +0.073 |
groom | +0.068 |
Neg logit contributions | |
nights | -0.073 |
night | -0.071 |
day | -0.071 |
fy | -0.070 |
afternoon | -0.068 |
Token | went |
Token ID | 1816 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
boys | +0.013 |
paramed | +0.012 |
chefs | +0.012 |
Lee | +0.011 |
printer | +0.011 |
Neg logit contributions | |
nights | -0.016 |
day | -0.016 |
night | -0.016 |
fy | -0.016 |
afternoon | -0.015 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
nights | +0.196 |
ight | +0.192 |
fy | +0.189 |
rated | +0.184 |
time | +0.182 |
Neg logit contributions | |
boys | -0.123 |
Lee | -0.120 |
chefs | -0.117 |
paramed | -0.112 |
printer | -0.108 |
Token | to |
Token ID | 284 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
fy | +0.317 |
nights | +0.307 |
ight | +0.303 |
rated | +0.285 |
day | +0.282 |
Neg logit contributions | |
Lee | -0.170 |
chefs | -0.166 |
boys | -0.165 |
paramed | -0.160 |
printer | -0.144 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
day | +0.244 |
night | +0.240 |
nights | +0.239 |
morning | +0.229 |
afternoon | +0.228 |
Neg logit contributions | |
boys | -0.170 |
paramed | -0.168 |
chefs | -0.162 |
Lee | -0.159 |
printer | -0.148 |
Token | named |
Token ID | 3706 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
paramed | +0.027 |
chefs | +0.026 |
boys | +0.026 |
Lee | +0.025 |
principal | +0.024 |
Neg logit contributions | |
day | -0.031 |
night | -0.030 |
nights | -0.029 |
time | -0.029 |
morning | -0.029 |
Token | Lily |
Token ID | 20037 |
Feature activation | -8.76e-03 |
Pos logit contributions | |
day | +0.156 |
night | +0.154 |
nights | +0.154 |
morning | +0.148 |
afternoon | +0.147 |
Neg logit contributions | |
boys | -0.092 |
paramed | -0.090 |
chefs | -0.086 |
Lee | -0.086 |
principal | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -7.02e-02 |
Pos logit contributions | |
nights | +0.012 |
day | +0.012 |
night | +0.012 |
fy | +0.011 |
morning | +0.011 |
Neg logit contributions | |
boys | -0.016 |
paramed | -0.015 |
Lee | -0.015 |
chefs | -0.015 |
groom | -0.014 |
Token | One |
Token ID | 1881 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
nights | +0.112 |
day | +0.109 |
night | +0.108 |
fy | +0.105 |
afternoon | +0.102 |
Neg logit contributions | |
boys | -0.110 |
Lee | -0.110 |
paramed | -0.109 |
chefs | -0.106 |
principal | -0.099 |
Token | day |
Token ID | 1110 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
ides | +0.167 |
ently | +0.165 |
ight | +0.157 |
ets | +0.154 |
spine | +0.152 |
Neg logit contributions | |
paramed | -0.267 |
sprayed | -0.259 |
duck | -0.254 |
qu | -0.251 |
printer | -0.247 |
Token | , |
Token ID | 11 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
nights | +0.553 |
fy | +0.526 |
day | +0.517 |
night | +0.515 |
ight | +0.509 |
Neg logit contributions | |
Lee | -0.358 |
paramed | -0.345 |
chefs | -0.338 |
boys | -0.325 |
printer | -0.300 |
Token | Lily |
Token ID | 20037 |
Feature activation | +9.22e-03 |
Pos logit contributions | |
boys | +0.075 |
Lee | +0.073 |
chefs | +0.073 |
paramed | +0.073 |
groom | +0.068 |
Neg logit contributions | |
nights | -0.073 |
night | -0.071 |
day | -0.071 |
fy | -0.070 |
afternoon | -0.068 |
Token | 's |
Token ID | 338 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
boys | +0.013 |
paramed | +0.012 |
chefs | +0.012 |
Lee | +0.011 |
printer | +0.011 |
Neg logit contributions | |
nights | -0.016 |
day | -0.016 |
night | -0.016 |
fy | -0.016 |
afternoon | -0.015 |
Token | mom |
Token ID | 1995 |
Feature activation | +8.75e-02 |
Pos logit contributions | |
paramed | +0.032 |
boys | +0.032 |
chefs | +0.030 |
Lee | +0.029 |
printer | +0.028 |
Neg logit contributions | |
nights | -0.102 |
fy | -0.099 |
night | -0.096 |
day | -0.095 |
afternoon | -0.094 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
boys | +0.135 |
paramed | +0.134 |
chefs | +0.131 |
Lee | +0.130 |
principal | +0.124 |
Neg logit contributions | |
day | -0.145 |
night | -0.142 |
nights | -0.136 |
time | -0.134 |
morning | -0.134 |
Token | good |
Token ID | 922 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
day | +0.058 |
night | +0.057 |
nights | +0.056 |
time | +0.054 |
morning | +0.054 |
Neg logit contributions | |
boys | -0.055 |
paramed | -0.054 |
Lee | -0.053 |
chefs | -0.052 |
reporter | -0.049 |
Token | friends |
Token ID | 2460 |
Feature activation | -4.68e-03 |
Pos logit contributions | |
boys | +0.052 |
paramed | +0.051 |
chefs | +0.049 |
Lee | +0.049 |
reporter | +0.047 |
Neg logit contributions | |
day | -0.043 |
night | -0.042 |
nights | -0.042 |
morning | -0.040 |
time | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
day | +0.006 |
night | +0.005 |
nights | +0.005 |
time | +0.005 |
morning | +0.005 |
Neg logit contributions | |
paramed | -0.010 |
boys | -0.009 |
chefs | -0.009 |
Lee | -0.009 |
printer | -0.009 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
nights | +0.070 |
night | +0.068 |
day | +0.068 |
fy | +0.065 |
afternoon | +0.064 |
Neg logit contributions | |
boys | -0.068 |
Lee | -0.067 |
paramed | -0.067 |
chefs | -0.065 |
principal | -0.061 |
Token | One |
Token ID | 3198 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
fy | +0.337 |
nights | +0.333 |
spine | +0.306 |
ly | +0.305 |
day | +0.303 |
Neg logit contributions | |
Lee | -0.235 |
boys | -0.230 |
chefs | -0.226 |
paramed | -0.212 |
principal | -0.205 |
Token | day |
Token ID | 1110 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
fy | +0.214 |
lessly | +0.206 |
spine | +0.205 |
Along | +0.205 |
ly | +0.199 |
Neg logit contributions | |
dads | -0.398 |
qu | -0.385 |
dad | -0.383 |
band | -0.379 |
mean | -0.378 |
Token | , |
Token ID | 11 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
nights | +0.633 |
fy | +0.620 |
found | +0.579 |
des | +0.575 |
ingle | +0.573 |
Neg logit contributions | |
Lee | -0.280 |
chefs | -0.254 |
paramed | -0.244 |
boys | -0.236 |
ha | -0.231 |
Token | a |
Token ID | 257 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
chefs | +0.037 |
boys | +0.037 |
Lee | +0.036 |
paramed | +0.035 |
printer | +0.032 |
Neg logit contributions | |
nights | -0.077 |
fy | -0.073 |
night | -0.071 |
day | -0.071 |
afternoon | -0.069 |
Token | little |
Token ID | 1310 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
day | +0.027 |
night | +0.026 |
nights | +0.025 |
time | +0.025 |
morning | +0.025 |
Neg logit contributions | |
boys | -0.040 |
paramed | -0.039 |
chefs | -0.038 |
Lee | -0.038 |
principal | -0.036 |
Token | girl |
Token ID | 2576 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
day | +0.214 |
night | +0.208 |
morning | +0.199 |
nights | +0.197 |
afternoon | +0.195 |
Neg logit contributions | |
boys | -0.141 |
paramed | -0.136 |
chefs | -0.131 |
Lee | -0.128 |
principal | -0.126 |
Token | named |
Token ID | 3706 |
Feature activation | -7.11e-02 |
Pos logit contributions | |
boys | +0.020 |
paramed | +0.019 |
chefs | +0.019 |
Lee | +0.018 |
printer | +0.017 |
Neg logit contributions | |
day | -0.044 |
nights | -0.044 |
night | -0.044 |
fy | -0.043 |
morning | -0.042 |
Token | big |
Token ID | 1263 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
day | +0.114 |
night | +0.109 |
time | +0.106 |
morning | +0.103 |
nights | +0.102 |
Neg logit contributions | |
boys | -0.142 |
paramed | -0.138 |
Lee | -0.135 |
chefs | -0.134 |
principal | -0.126 |
Token | hug |
Token ID | 16225 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
day | +0.186 |
night | +0.180 |
time | +0.174 |
morning | +0.171 |
afternoon | +0.164 |
Neg logit contributions | |
boys | -0.174 |
paramed | -0.166 |
chefs | -0.160 |
Lee | -0.154 |
principal | -0.149 |
Token | . |
Token ID | 13 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
boys | +0.045 |
paramed | +0.044 |
chefs | +0.043 |
Lee | +0.042 |
principal | +0.041 |
Neg logit contributions | |
day | -0.027 |
night | -0.026 |
nights | -0.025 |
morning | -0.025 |
time | -0.024 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
day | +0.091 |
nights | +0.090 |
night | +0.090 |
afternoon | +0.084 |
time | +0.084 |
Neg logit contributions | |
boys | -0.088 |
paramed | -0.088 |
Lee | -0.086 |
chefs | -0.085 |
principal | -0.079 |
Token | One |
Token ID | 3198 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
nights | +0.329 |
fy | +0.326 |
day | +0.300 |
spine | +0.299 |
night | +0.295 |
Neg logit contributions | |
Lee | -0.254 |
boys | -0.245 |
chefs | -0.244 |
paramed | -0.229 |
reporter | -0.218 |
Token | day |
Token ID | 1110 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
fy | +0.203 |
lessly | +0.200 |
Along | +0.198 |
spine | +0.194 |
nights | +0.190 |
Neg logit contributions | |
dads | -0.404 |
qu | -0.403 |
dad | -0.397 |
band | -0.394 |
mixer | -0.393 |
Token | , |
Token ID | 11 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
nights | +0.609 |
fy | +0.587 |
found | +0.545 |
ingle | +0.542 |
night | +0.542 |
Neg logit contributions | |
Lee | -0.309 |
chefs | -0.284 |
paramed | -0.276 |
boys | -0.273 |
printer | -0.252 |
Token | there |
Token ID | 612 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
boys | +0.040 |
chefs | +0.039 |
Lee | +0.039 |
paramed | +0.038 |
printer | +0.034 |
Neg logit contributions | |
nights | -0.078 |
fy | -0.075 |
day | -0.074 |
night | -0.074 |
afternoon | -0.071 |
Token | was |
Token ID | 373 |
Feature activation | -7.65e-03 |
Pos logit contributions | |
fy | +0.189 |
nights | +0.188 |
night | +0.176 |
day | +0.175 |
Along | +0.174 |
Neg logit contributions | |
boys | -0.111 |
paramed | -0.109 |
Lee | -0.109 |
chefs | -0.108 |
printer | -0.095 |
Token | a |
Token ID | 257 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
fy | +0.019 |
rated | +0.019 |
nights | +0.019 |
spine | +0.019 |
ingle | +0.018 |
Neg logit contributions | |
chefs | -0.005 |
paramed | -0.004 |
boys | -0.004 |
printer | -0.004 |
Lee | -0.003 |
Token | slow |
Token ID | 3105 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
day | +0.017 |
night | +0.015 |
time | +0.014 |
morning | +0.014 |
afternoon | +0.014 |
Neg logit contributions | |
boys | -0.030 |
Lee | -0.028 |
paramed | -0.028 |
wrists | -0.028 |
chefs | -0.027 |
Token | are |
Token ID | 389 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
Lee | +0.090 |
chefs | +0.082 |
paramed | +0.082 |
boys | +0.081 |
principal | +0.071 |
Neg logit contributions | |
nights | -0.341 |
fy | -0.329 |
ight | -0.324 |
day | -0.324 |
way | -0.319 |
Token | happy |
Token ID | 3772 |
Feature activation | -6.70e-02 |
Pos logit contributions | |
paramed | +0.051 |
boys | +0.050 |
Lee | +0.050 |
chefs | +0.046 |
reporter | +0.045 |
Neg logit contributions | |
day | -0.056 |
night | -0.054 |
time | -0.052 |
morning | -0.051 |
nights | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
day | +0.087 |
night | +0.085 |
nights | +0.083 |
morning | +0.079 |
time | +0.078 |
Neg logit contributions | |
boys | -0.126 |
paramed | -0.126 |
chefs | -0.122 |
Lee | -0.120 |
principal | -0.115 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
nights | +0.106 |
day | +0.106 |
night | +0.105 |
afternoon | +0.099 |
time | +0.099 |
Neg logit contributions | |
boys | -0.094 |
paramed | -0.094 |
Lee | -0.092 |
chefs | -0.090 |
reporter | -0.084 |
Token | One |
Token ID | 3198 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
nights | +0.324 |
fy | +0.324 |
spine | +0.294 |
day | +0.294 |
ly | +0.291 |
Neg logit contributions | |
Lee | -0.252 |
chefs | -0.248 |
boys | -0.246 |
paramed | -0.234 |
principal | -0.223 |
Token | day |
Token ID | 1110 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
fy | +0.213 |
Along | +0.204 |
ight | +0.193 |
lessly | +0.192 |
spine | +0.189 |
Neg logit contributions | |
dads | -0.418 |
qu | -0.403 |
dad | -0.400 |
band | -0.400 |
fountain | -0.394 |
Token | , |
Token ID | 11 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
nights | +0.627 |
fy | +0.608 |
found | +0.567 |
ight | +0.560 |
rated | +0.559 |
Neg logit contributions | |
Lee | -0.290 |
chefs | -0.270 |
paramed | -0.255 |
boys | -0.246 |
printer | -0.232 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
boys | +0.041 |
chefs | +0.040 |
paramed | +0.039 |
Lee | +0.039 |
printer | +0.035 |
Neg logit contributions | |
nights | -0.069 |
day | -0.066 |
night | -0.065 |
fy | -0.065 |
afternoon | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | -6.89e-02 |
Pos logit contributions | |
boys | +0.040 |
chefs | +0.038 |
paramed | +0.038 |
Lee | +0.037 |
printer | +0.035 |
Neg logit contributions | |
nights | -0.047 |
day | -0.046 |
fy | -0.046 |
night | -0.046 |
Along | -0.044 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.48e-03 |
Pos logit contributions | |
day | +0.102 |
night | +0.100 |
nights | +0.094 |
morning | +0.094 |
time | +0.092 |
Neg logit contributions | |
boys | -0.123 |
paramed | -0.121 |
chefs | -0.114 |
Lee | -0.114 |
principal | -0.113 |
Token | went |
Token ID | 1816 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
boys | +0.002 |
paramed | +0.002 |
chefs | +0.002 |
Lee | +0.002 |
principal | +0.002 |
Neg logit contributions | |
day | -0.003 |
night | -0.002 |
nights | -0.002 |
morning | -0.002 |
time | -0.002 |
Token | cold |
Token ID | 4692 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
day | +0.224 |
night | +0.218 |
nights | +0.212 |
time | +0.205 |
morning | +0.204 |
Neg logit contributions | |
paramed | -0.171 |
boys | -0.168 |
chefs | -0.160 |
Lee | -0.160 |
reporter | -0.151 |
Token | winter |
Token ID | 7374 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
Along | +0.328 |
fy | +0.325 |
nights | +0.315 |
found | +0.314 |
day | +0.304 |
Neg logit contributions | |
chefs | -0.367 |
boys | -0.364 |
paramed | -0.360 |
Lee | -0.352 |
printer | -0.324 |
Token | snow |
Token ID | 6729 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
Along | +0.351 |
fy | +0.350 |
found | +0.349 |
nights | +0.338 |
ight | +0.331 |
Neg logit contributions | |
boys | -0.462 |
paramed | -0.454 |
chefs | -0.440 |
Lee | -0.439 |
printer | -0.412 |
Token | . |
Token ID | 13 |
Feature activation | -6.48e-02 |
Pos logit contributions | |
fy | +0.233 |
found | +0.230 |
nights | +0.228 |
Along | +0.228 |
day | +0.220 |
Neg logit contributions | |
boys | -0.239 |
paramed | -0.229 |
Lee | -0.221 |
chefs | -0.220 |
printer | -0.205 |
Token | One |
Token ID | 1881 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
nights | +0.113 |
night | +0.109 |
day | +0.108 |
fy | +0.107 |
afternoon | +0.104 |
Neg logit contributions | |
Lee | -0.091 |
paramed | -0.089 |
boys | -0.086 |
chefs | -0.083 |
printer | -0.076 |
Token | day |
Token ID | 1110 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
Along | +0.144 |
ets | +0.136 |
ides | +0.135 |
spine | +0.135 |
ight | +0.130 |
Neg logit contributions | |
paramed | -0.294 |
duck | -0.289 |
mixer | -0.280 |
qu | -0.269 |
sprayed | -0.269 |
Token | , |
Token ID | 11 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
day | +0.465 |
nights | +0.464 |
night | +0.457 |
fy | +0.440 |
afternoon | +0.437 |
Neg logit contributions | |
paramed | -0.430 |
boys | -0.426 |
Lee | -0.420 |
chefs | -0.417 |
reporter | -0.381 |
Token | she |
Token ID | 673 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
boys | +0.079 |
paramed | +0.078 |
chefs | +0.076 |
Lee | +0.075 |
printer | +0.070 |
Neg logit contributions | |
day | -0.087 |
night | -0.086 |
nights | -0.085 |
morning | -0.081 |
afternoon | -0.081 |
Token | went |
Token ID | 1816 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
day | +0.310 |
nights | +0.307 |
night | +0.305 |
afternoon | +0.293 |
fy | +0.293 |
Neg logit contributions | |
boys | -0.191 |
paramed | -0.186 |
chefs | -0.180 |
Lee | -0.173 |
groom | -0.164 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
nights | +0.183 |
ight | +0.177 |
fy | +0.176 |
day | +0.175 |
afternoon | +0.172 |
Neg logit contributions | |
boys | -0.119 |
paramed | -0.117 |
Lee | -0.117 |
chefs | -0.116 |
principal | -0.105 |
Token | to |
Token ID | 284 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
fy | +0.294 |
nights | +0.283 |
ight | +0.279 |
rated | +0.265 |
day | +0.265 |
Neg logit contributions | |
Lee | -0.161 |
chefs | -0.157 |
boys | -0.157 |
paramed | -0.155 |
printer | -0.139 |
Token | her |
Token ID | 607 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
boys | +0.075 |
paramed | +0.071 |
chefs | +0.070 |
printer | +0.069 |
Lee | +0.067 |
Neg logit contributions | |
nights | -0.076 |
fy | -0.075 |
night | -0.072 |
rated | -0.070 |
morning | -0.069 |
Token | heart |
Token ID | 2612 |
Feature activation | +2.63e-03 |
Pos logit contributions | |
nights | +0.294 |
Along | +0.282 |
fy | +0.282 |
night | +0.280 |
day | +0.279 |
Neg logit contributions | |
boys | -0.217 |
Lee | -0.206 |
chefs | -0.204 |
paramed | -0.204 |
groom | -0.199 |
Token | . |
Token ID | 13 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
boys | +0.004 |
paramed | +0.004 |
chefs | +0.004 |
Lee | +0.004 |
groom | +0.004 |
Neg logit contributions | |
day | -0.004 |
nights | -0.004 |
night | -0.004 |
morning | -0.003 |
afternoon | -0.003 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
nights | +0.082 |
night | +0.080 |
day | +0.080 |
fy | +0.077 |
afternoon | +0.075 |
Neg logit contributions | |
Lee | -0.076 |
paramed | -0.075 |
boys | -0.073 |
chefs | -0.070 |
principal | -0.068 |
Token | One |
Token ID | 3198 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
fy | +0.333 |
nights | +0.331 |
spine | +0.302 |
day | +0.300 |
ly | +0.299 |
Neg logit contributions | |
Lee | -0.248 |
boys | -0.237 |
chefs | -0.234 |
paramed | -0.225 |
principal | -0.212 |
Token | day |
Token ID | 1110 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
fy | +0.190 |
Along | +0.183 |
spine | +0.180 |
nights | +0.178 |
lessly | +0.176 |
Neg logit contributions | |
dads | -0.406 |
qu | -0.402 |
mixer | -0.394 |
paramed | -0.393 |
band | -0.391 |
Token | , |
Token ID | 11 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
nights | +0.588 |
fy | +0.568 |
found | +0.531 |
night | +0.530 |
ingle | +0.526 |
Neg logit contributions | |
Lee | -0.322 |
paramed | -0.302 |
chefs | -0.298 |
boys | -0.286 |
printer | -0.267 |
Token | a |
Token ID | 257 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
boys | +0.045 |
paramed | +0.044 |
chefs | +0.044 |
Lee | +0.043 |
printer | +0.039 |
Neg logit contributions | |
nights | -0.081 |
day | -0.079 |
night | -0.079 |
fy | -0.077 |
morning | -0.075 |
Token | huge |
Token ID | 3236 |
Feature activation | -7.49e-02 |
Pos logit contributions | |
day | +0.027 |
night | +0.026 |
nights | +0.025 |
time | +0.024 |
morning | +0.024 |
Neg logit contributions | |
boys | -0.044 |
paramed | -0.043 |
chefs | -0.042 |
Lee | -0.042 |
principal | -0.040 |
Token | human |
Token ID | 1692 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
day | +0.149 |
night | +0.142 |
morning | +0.137 |
time | +0.135 |
afternoon | +0.132 |
Neg logit contributions | |
boys | -0.105 |
paramed | -0.098 |
chefs | -0.097 |
Lee | -0.093 |
wrists | -0.092 |
Token | was |
Token ID | 373 |
Feature activation | -3.92e-03 |
Pos logit contributions | |
paramed | +0.014 |
boys | +0.014 |
chefs | +0.013 |
Lee | +0.013 |
wrists | +0.013 |
Neg logit contributions | |
day | -0.020 |
night | -0.020 |
morning | -0.019 |
nights | -0.019 |
time | -0.019 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
day | +0.246 |
night | +0.240 |
nights | +0.232 |
morning | +0.230 |
time | +0.226 |
Neg logit contributions | |
boys | -0.177 |
paramed | -0.175 |
chefs | -0.166 |
Lee | -0.165 |
principal | -0.158 |
Token | named |
Token ID | 3706 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
paramed | +0.025 |
boys | +0.025 |
chefs | +0.024 |
Lee | +0.024 |
principal | +0.023 |
Neg logit contributions | |
day | -0.037 |
night | -0.036 |
nights | -0.035 |
morning | -0.034 |
time | -0.034 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.91e-03 |
Pos logit contributions | |
day | +0.144 |
night | +0.142 |
nights | +0.140 |
morning | +0.136 |
time | +0.135 |
Neg logit contributions | |
boys | -0.090 |
paramed | -0.089 |
chefs | -0.086 |
Lee | -0.082 |
reporter | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
nights | +0.003 |
day | +0.003 |
night | +0.003 |
afternoon | +0.003 |
fy | +0.003 |
Neg logit contributions | |
boys | -0.003 |
Lee | -0.003 |
paramed | -0.003 |
chefs | -0.003 |
groom | -0.003 |
Token | One |
Token ID | 1881 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
nights | +0.104 |
day | +0.103 |
night | +0.102 |
fy | +0.096 |
afternoon | +0.096 |
Neg logit contributions | |
boys | -0.107 |
paramed | -0.106 |
Lee | -0.104 |
chefs | -0.102 |
principal | -0.096 |
Token | day |
Token ID | 1110 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
ides | +0.183 |
ently | +0.179 |
ets | +0.167 |
ight | +0.163 |
spine | +0.163 |
Neg logit contributions | |
paramed | -0.261 |
sprayed | -0.258 |
duck | -0.256 |
qu | -0.248 |
printer | -0.240 |
Token | , |
Token ID | 11 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
nights | +0.551 |
fy | +0.521 |
night | +0.516 |
day | +0.515 |
ight | +0.510 |
Neg logit contributions | |
Lee | -0.348 |
paramed | -0.338 |
chefs | -0.334 |
boys | -0.322 |
printer | -0.294 |
Token | she |
Token ID | 673 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
boys | +0.064 |
chefs | +0.064 |
Lee | +0.063 |
paramed | +0.063 |
groom | +0.059 |
Neg logit contributions | |
nights | -0.076 |
night | -0.074 |
day | -0.073 |
fy | -0.073 |
afternoon | -0.070 |
Token | was |
Token ID | 373 |
Feature activation | -2.44e-03 |
Pos logit contributions | |
day | +0.303 |
nights | +0.300 |
night | +0.298 |
morning | +0.286 |
afternoon | +0.285 |
Neg logit contributions | |
boys | -0.203 |
paramed | -0.198 |
chefs | -0.192 |
Lee | -0.187 |
printer | -0.176 |
Token | playing |
Token ID | 2712 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
day | +0.004 |
night | +0.004 |
nights | +0.004 |
morning | +0.004 |
time | +0.004 |
Neg logit contributions | |
boys | -0.004 |
paramed | -0.004 |
chefs | -0.004 |
Lee | -0.004 |
reporter | -0.003 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
day | +0.150 |
night | +0.139 |
time | +0.134 |
morning | +0.133 |
nights | +0.132 |
Neg logit contributions | |
paramed | -0.256 |
boys | -0.250 |
Lee | -0.247 |
reporter | -0.237 |
wrists | -0.236 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
day | +0.246 |
night | +0.240 |
nights | +0.232 |
morning | +0.230 |
time | +0.226 |
Neg logit contributions | |
boys | -0.177 |
paramed | -0.175 |
chefs | -0.166 |
Lee | -0.165 |
principal | -0.158 |
Token | named |
Token ID | 3706 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
paramed | +0.025 |
boys | +0.025 |
chefs | +0.024 |
Lee | +0.024 |
principal | +0.023 |
Neg logit contributions | |
day | -0.037 |
night | -0.036 |
nights | -0.035 |
morning | -0.034 |
time | -0.034 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.91e-03 |
Pos logit contributions | |
day | +0.144 |
night | +0.142 |
nights | +0.140 |
morning | +0.136 |
time | +0.135 |
Neg logit contributions | |
boys | -0.090 |
paramed | -0.089 |
chefs | -0.086 |
Lee | -0.082 |
reporter | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
nights | +0.003 |
day | +0.003 |
night | +0.003 |
afternoon | +0.003 |
fy | +0.003 |
Neg logit contributions | |
boys | -0.003 |
Lee | -0.003 |
paramed | -0.003 |
chefs | -0.003 |
groom | -0.003 |
Token | One |
Token ID | 1881 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
nights | +0.104 |
day | +0.103 |
night | +0.102 |
fy | +0.096 |
afternoon | +0.096 |
Neg logit contributions | |
boys | -0.107 |
paramed | -0.106 |
Lee | -0.104 |
chefs | -0.102 |
principal | -0.096 |
Token | day |
Token ID | 1110 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
ides | +0.183 |
ently | +0.179 |
ets | +0.167 |
ight | +0.163 |
spine | +0.163 |
Neg logit contributions | |
paramed | -0.261 |
sprayed | -0.258 |
duck | -0.256 |
qu | -0.248 |
printer | -0.240 |
Token | , |
Token ID | 11 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
nights | +0.551 |
fy | +0.521 |
night | +0.516 |
day | +0.515 |
ight | +0.510 |
Neg logit contributions | |
Lee | -0.348 |
paramed | -0.338 |
chefs | -0.334 |
boys | -0.322 |
printer | -0.294 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
boys | +0.064 |
chefs | +0.064 |
Lee | +0.063 |
paramed | +0.063 |
groom | +0.059 |
Neg logit contributions | |
nights | -0.076 |
night | -0.074 |
day | -0.073 |
fy | -0.073 |
afternoon | -0.070 |
Token | went |
Token ID | 1816 |
Feature activation | -9.93e-02 |
Pos logit contributions | |
boys | +0.014 |
chefs | +0.013 |
paramed | +0.013 |
Lee | +0.013 |
printer | +0.012 |
Neg logit contributions | |
nights | -0.019 |
day | -0.019 |
fy | -0.019 |
night | -0.018 |
afternoon | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
nights | +0.199 |
ight | +0.196 |
fy | +0.192 |
rated | +0.188 |
time | +0.184 |
Neg logit contributions | |
boys | -0.114 |
Lee | -0.113 |
chefs | -0.108 |
paramed | -0.104 |
principal | -0.100 |
Token | the |
Token ID | 262 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
boys | +0.043 |
paramed | +0.043 |
chefs | +0.041 |
Lee | +0.040 |
reporter | +0.037 |
Neg logit contributions | |
day | -0.074 |
night | -0.073 |
nights | -0.073 |
morning | -0.070 |
time | -0.070 |
Token | her |
Token ID | 607 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
day | +0.257 |
night | +0.253 |
nights | +0.252 |
morning | +0.241 |
afternoon | +0.240 |
Neg logit contributions | |
boys | -0.188 |
paramed | -0.184 |
chefs | -0.178 |
Lee | -0.176 |
printer | -0.164 |
Token | face |
Token ID | 1986 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
nights | +0.312 |
Along | +0.308 |
fy | +0.303 |
ingle | +0.303 |
way | +0.299 |
Neg logit contributions | |
boys | -0.229 |
wrists | -0.222 |
Lee | -0.212 |
printer | -0.208 |
principal | -0.205 |
Token | . |
Token ID | 13 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
boys | +0.068 |
paramed | +0.067 |
chefs | +0.066 |
Lee | +0.065 |
groom | +0.062 |
Neg logit contributions | |
day | -0.043 |
night | -0.042 |
nights | -0.042 |
morning | -0.039 |
afternoon | -0.038 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
nights | +0.081 |
day | +0.080 |
night | +0.080 |
fy | +0.076 |
afternoon | +0.075 |
Neg logit contributions | |
Lee | -0.079 |
paramed | -0.079 |
boys | -0.078 |
chefs | -0.075 |
principal | -0.071 |
Token | One |
Token ID | 3198 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
nights | +0.348 |
fy | +0.348 |
spine | +0.317 |
ly | +0.315 |
day | +0.312 |
Neg logit contributions | |
Lee | -0.243 |
boys | -0.227 |
chefs | -0.226 |
paramed | -0.217 |
principal | -0.205 |
Token | day |
Token ID | 1110 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
fy | +0.194 |
Along | +0.194 |
nights | +0.193 |
spine | +0.187 |
lessly | +0.187 |
Neg logit contributions | |
dads | -0.400 |
mixer | -0.395 |
qu | -0.392 |
dad | -0.388 |
band | -0.386 |
Token | , |
Token ID | 11 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
nights | +0.609 |
fy | +0.584 |
found | +0.546 |
ingle | +0.540 |
night | +0.540 |
Neg logit contributions | |
Lee | -0.308 |
paramed | -0.283 |
chefs | -0.282 |
boys | -0.260 |
ha | -0.250 |
Token | a |
Token ID | 257 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
boys | +0.033 |
chefs | +0.033 |
Lee | +0.033 |
paramed | +0.032 |
printer | +0.028 |
Neg logit contributions | |
nights | -0.074 |
fy | -0.071 |
day | -0.070 |
night | -0.070 |
afternoon | -0.067 |
Token | little |
Token ID | 1310 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
day | +0.028 |
night | +0.027 |
nights | +0.025 |
time | +0.025 |
morning | +0.025 |
Neg logit contributions | |
boys | -0.045 |
paramed | -0.044 |
chefs | -0.042 |
Lee | -0.042 |
principal | -0.040 |
Token | boy |
Token ID | 2933 |
Feature activation | +9.11e-02 |
Pos logit contributions | |
day | +0.208 |
night | +0.202 |
morning | +0.193 |
afternoon | +0.188 |
nights | +0.188 |
Neg logit contributions | |
boys | -0.150 |
paramed | -0.144 |
chefs | -0.139 |
Lee | -0.136 |
principal | -0.136 |
Token | 's |
Token ID | 338 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
boys | +0.086 |
paramed | +0.084 |
chefs | +0.081 |
Lee | +0.077 |
printer | +0.072 |
Neg logit contributions | |
nights | -0.199 |
day | -0.198 |
night | -0.196 |
fy | -0.192 |
afternoon | -0.189 |
Token | cars |
Token ID | 5006 |
Feature activation | -8.67e-02 |
Pos logit contributions | |
day | +0.097 |
night | +0.095 |
nights | +0.093 |
morning | +0.090 |
time | +0.089 |
Neg logit contributions | |
paramed | -0.093 |
boys | -0.093 |
chefs | -0.090 |
Lee | -0.088 |
reporter | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
day | +0.119 |
night | +0.117 |
nights | +0.117 |
morning | +0.110 |
afternoon | +0.109 |
Neg logit contributions | |
boys | -0.153 |
paramed | -0.151 |
chefs | -0.147 |
Lee | -0.145 |
printer | -0.138 |
Token | trucks |
Token ID | 13960 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
day | +0.136 |
night | +0.132 |
nights | +0.130 |
morning | +0.124 |
time | +0.124 |
Neg logit contributions | |
paramed | -0.116 |
boys | -0.106 |
chefs | -0.105 |
Lee | -0.101 |
reporter | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
day | +0.047 |
night | +0.046 |
nights | +0.045 |
morning | +0.043 |
time | +0.042 |
Neg logit contributions | |
paramed | -0.070 |
boys | -0.070 |
chefs | -0.068 |
Lee | -0.067 |
groom | -0.064 |
Token | One |
Token ID | 1881 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
day | +0.099 |
nights | +0.097 |
night | +0.097 |
morning | +0.091 |
afternoon | +0.091 |
Neg logit contributions | |
boys | -0.099 |
paramed | -0.099 |
chefs | -0.095 |
Lee | -0.095 |
principal | -0.089 |
Token | day |
Token ID | 1110 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
ight | +0.097 |
fy | +0.094 |
nights | +0.087 |
Along | +0.085 |
ingle | +0.079 |
Neg logit contributions | |
paramed | -0.350 |
printer | -0.349 |
Lee | -0.340 |
mixer | -0.337 |
chefs | -0.332 |
Token | , |
Token ID | 11 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
day | +0.459 |
nights | +0.459 |
night | +0.451 |
fy | +0.431 |
afternoon | +0.430 |
Neg logit contributions | |
boys | -0.434 |
paramed | -0.432 |
chefs | -0.424 |
Lee | -0.421 |
printer | -0.389 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
boys | +0.076 |
paramed | +0.075 |
chefs | +0.070 |
Lee | +0.069 |
principal | +0.067 |
Neg logit contributions | |
day | -0.091 |
night | -0.089 |
nights | -0.087 |
morning | -0.085 |
time | -0.084 |
Token | my |
Token ID | 1820 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
nights | +0.030 |
fy | +0.029 |
day | +0.029 |
Along | +0.028 |
night | +0.028 |
Neg logit contributions | |
boys | -0.030 |
chefs | -0.029 |
Lee | -0.028 |
paramed | -0.028 |
principal | -0.027 |
Token | 's |
Token ID | 338 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
day | +0.243 |
night | +0.240 |
nights | +0.234 |
morning | +0.227 |
time | +0.226 |
Neg logit contributions | |
paramed | -0.200 |
boys | -0.198 |
chefs | -0.189 |
Lee | -0.188 |
groom | -0.176 |
Token | mom |
Token ID | 1995 |
Feature activation | +9.71e-02 |
Pos logit contributions | |
boys | +0.041 |
paramed | +0.040 |
chefs | +0.039 |
Lee | +0.038 |
printer | +0.035 |
Neg logit contributions | |
nights | -0.091 |
night | -0.089 |
day | -0.088 |
fy | -0.087 |
afternoon | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
day | +0.247 |
nights | +0.245 |
night | +0.244 |
fy | +0.236 |
morning | +0.234 |
Neg logit contributions | |
boys | -0.127 |
paramed | -0.120 |
chefs | -0.117 |
Lee | -0.115 |
printer | -0.106 |
Token | swings |
Token ID | 26728 |
Feature activation | -5.75e-03 |
Pos logit contributions | |
day | +0.021 |
nights | +0.021 |
night | +0.021 |
fy | +0.020 |
morning | +0.020 |
Neg logit contributions | |
boys | -0.017 |
paramed | -0.016 |
chefs | -0.016 |
Lee | -0.015 |
printer | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
nights | +0.008 |
day | +0.008 |
night | +0.008 |
fy | +0.007 |
morning | +0.007 |
Neg logit contributions | |
boys | -0.010 |
Lee | -0.010 |
paramed | -0.010 |
chefs | -0.009 |
printer | -0.009 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
nights | +0.081 |
night | +0.080 |
day | +0.079 |
fy | +0.076 |
afternoon | +0.075 |
Neg logit contributions | |
boys | -0.073 |
Lee | -0.072 |
paramed | -0.072 |
chefs | -0.069 |
principal | -0.065 |
Token | One |
Token ID | 3198 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
nights | +0.326 |
fy | +0.321 |
day | +0.297 |
night | +0.296 |
ly | +0.294 |
Neg logit contributions | |
Lee | -0.241 |
boys | -0.239 |
chefs | -0.237 |
paramed | -0.226 |
principal | -0.213 |
Token | day |
Token ID | 1110 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
fy | +0.213 |
nights | +0.211 |
lessly | +0.202 |
ly | +0.201 |
Along | +0.200 |
Neg logit contributions | |
dads | -0.422 |
dad | -0.398 |
qu | -0.389 |
band | -0.388 |
mean | -0.385 |
Token | , |
Token ID | 11 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
nights | +0.647 |
fy | +0.618 |
found | +0.577 |
adows | +0.571 |
rated | +0.571 |
Neg logit contributions | |
Lee | -0.278 |
chefs | -0.251 |
paramed | -0.238 |
boys | -0.222 |
ha | -0.220 |
Token | Little |
Token ID | 7703 |
Feature activation | -9.09e-02 |
Pos logit contributions | |
chefs | +0.036 |
boys | +0.035 |
paramed | +0.034 |
Lee | +0.034 |
printer | +0.031 |
Neg logit contributions | |
nights | -0.074 |
fy | -0.071 |
day | -0.070 |
night | -0.070 |
afternoon | -0.067 |
Token | Billy |
Token ID | 15890 |
Feature activation | +8.76e-03 |
Pos logit contributions | |
day | +0.170 |
night | +0.167 |
nights | +0.163 |
morning | +0.160 |
time | +0.158 |
Neg logit contributions | |
boys | -0.122 |
paramed | -0.121 |
chefs | -0.116 |
Lee | -0.111 |
principal | -0.108 |
Token | was |
Token ID | 373 |
Feature activation | -4.60e-03 |
Pos logit contributions | |
boys | +0.012 |
chefs | +0.011 |
paramed | +0.011 |
Lee | +0.011 |
printer | +0.010 |
Neg logit contributions | |
nights | -0.015 |
fy | -0.015 |
day | -0.015 |
night | -0.015 |
Along | -0.014 |
Token | feeling |
Token ID | 4203 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
fy | +0.009 |
nights | +0.009 |
night | +0.008 |
day | +0.008 |
ight | +0.008 |
Neg logit contributions | |
paramed | -0.006 |
chefs | -0.006 |
boys | -0.006 |
Lee | -0.005 |
printer | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
day | +0.097 |
night | +0.096 |
nights | +0.095 |
morning | +0.091 |
time | +0.091 |
Neg logit contributions | |
boys | -0.072 |
paramed | -0.071 |
chefs | -0.069 |
Lee | -0.067 |
reporter | -0.063 |
Token | smile |
Token ID | 8212 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
day | +0.070 |
night | +0.067 |
time | +0.065 |
morning | +0.064 |
afternoon | +0.062 |
Neg logit contributions | |
boys | -0.079 |
paramed | -0.078 |
chefs | -0.075 |
Lee | -0.074 |
principal | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
boys | +0.131 |
paramed | +0.129 |
Lee | +0.128 |
chefs | +0.126 |
groom | +0.123 |
Neg logit contributions | |
nights | -0.100 |
day | -0.097 |
night | -0.095 |
fy | -0.093 |
Along | -0.092 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
nights | +0.081 |
day | +0.079 |
night | +0.079 |
fy | +0.075 |
afternoon | +0.074 |
Neg logit contributions | |
Lee | -0.080 |
boys | -0.079 |
paramed | -0.079 |
chefs | -0.077 |
principal | -0.072 |
Token | One |
Token ID | 3198 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
nights | +0.323 |
fy | +0.319 |
day | +0.295 |
spine | +0.293 |
night | +0.291 |
Neg logit contributions | |
Lee | -0.253 |
boys | -0.245 |
chefs | -0.244 |
paramed | -0.231 |
reporter | -0.219 |
Token | day |
Token ID | 1110 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
fy | +0.192 |
Along | +0.187 |
lessly | +0.185 |
spine | +0.185 |
nights | +0.182 |
Neg logit contributions | |
qu | -0.405 |
dads | -0.404 |
band | -0.393 |
mixer | -0.392 |
dad | -0.391 |
Token | , |
Token ID | 11 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
nights | +0.610 |
fy | +0.583 |
night | +0.543 |
ingle | +0.543 |
found | +0.541 |
Neg logit contributions | |
Lee | -0.305 |
chefs | -0.279 |
paramed | -0.273 |
boys | -0.266 |
printer | -0.245 |
Token | m |
Token ID | 285 |
Feature activation | +4.99e-02 |
Pos logit contributions | |
boys | +0.038 |
Lee | +0.038 |
chefs | +0.038 |
paramed | +0.037 |
printer | +0.032 |
Neg logit contributions | |
nights | -0.083 |
fy | -0.079 |
night | -0.078 |
day | -0.077 |
afternoon | -0.075 |
Token | ama |
Token ID | 1689 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
paramed | +0.074 |
Lee | +0.069 |
boys | +0.064 |
ha | +0.059 |
principal | +0.059 |
Neg logit contributions | |
nights | -0.091 |
night | -0.089 |
day | -0.087 |
time | -0.087 |
afternoon | -0.086 |
Token | and |
Token ID | 290 |
Feature activation | -6.21e-02 |
Pos logit contributions | |
day | +0.041 |
night | +0.040 |
nights | +0.039 |
morning | +0.038 |
time | +0.037 |
Neg logit contributions | |
boys | -0.042 |
paramed | -0.042 |
chefs | -0.040 |
Lee | -0.040 |
principal | -0.038 |
Token | her |
Token ID | 607 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
day | +0.100 |
night | +0.100 |
morning | +0.094 |
time | +0.090 |
evening | +0.090 |
Neg logit contributions | |
boys | -0.104 |
principal | -0.100 |
paramed | -0.097 |
wrists | -0.094 |
printer | -0.094 |
Token | much |
Token ID | 881 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
nights | +0.057 |
Along | +0.057 |
spine | +0.055 |
fy | +0.054 |
ingle | +0.052 |
Neg logit contributions | |
groom | -0.035 |
paramed | -0.035 |
dads | -0.034 |
chefs | -0.033 |
qu | -0.033 |
Token | fun |
Token ID | 1257 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
Along | +0.220 |
nights | +0.220 |
night | +0.211 |
day | +0.210 |
fy | +0.209 |
Neg logit contributions | |
chefs | -0.219 |
paramed | -0.217 |
boys | -0.215 |
Lee | -0.199 |
reporter | -0.197 |
Token | ." |
Token ID | 526 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
nights | +0.125 |
day | +0.124 |
night | +0.124 |
fy | +0.121 |
Along | +0.118 |
Neg logit contributions | |
boys | -0.163 |
chefs | -0.158 |
paramed | -0.157 |
Lee | -0.154 |
reporter | -0.147 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
boys | +0.187 |
paramed | +0.185 |
chefs | +0.179 |
Lee | +0.178 |
principal | +0.168 |
Neg logit contributions | |
day | -0.173 |
night | -0.171 |
nights | -0.170 |
morning | -0.161 |
afternoon | -0.160 |
Token | One |
Token ID | 3198 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
nights | +0.322 |
fy | +0.318 |
day | +0.295 |
night | +0.290 |
spine | +0.288 |
Neg logit contributions | |
Lee | -0.245 |
boys | -0.235 |
chefs | -0.231 |
paramed | -0.226 |
principal | -0.213 |
Token | day |
Token ID | 1110 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
Along | +0.143 |
fy | +0.137 |
nights | +0.130 |
spine | +0.127 |
lessly | +0.122 |
Neg logit contributions | |
paramed | -0.444 |
qu | -0.439 |
dads | -0.438 |
mixer | -0.432 |
bully | -0.431 |
Token | , |
Token ID | 11 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
nights | +0.561 |
fy | +0.540 |
night | +0.510 |
day | +0.505 |
found | +0.504 |
Neg logit contributions | |
Lee | -0.338 |
chefs | -0.321 |
paramed | -0.318 |
boys | -0.314 |
printer | -0.289 |
Token | Ken |
Token ID | 7148 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
boys | +0.045 |
paramed | +0.044 |
chefs | +0.043 |
Lee | +0.043 |
reporter | +0.038 |
Neg logit contributions | |
nights | -0.082 |
day | -0.081 |
night | -0.081 |
fy | -0.079 |
morning | -0.077 |
Token | and |
Token ID | 290 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
nights | +0.102 |
fy | +0.098 |
day | +0.097 |
night | +0.095 |
afternoon | +0.093 |
Neg logit contributions | |
boys | -0.081 |
Lee | -0.078 |
paramed | -0.077 |
chefs | -0.074 |
reporter | -0.071 |
Token | his |
Token ID | 465 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
day | +0.073 |
night | +0.069 |
morning | +0.064 |
nights | +0.064 |
time | +0.064 |
Neg logit contributions | |
boys | -0.119 |
paramed | -0.117 |
chefs | -0.112 |
printer | -0.110 |
wrists | -0.110 |
Token | Mom |
Token ID | 11254 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
day | +0.233 |
nights | +0.232 |
night | +0.231 |
morning | +0.222 |
afternoon | +0.221 |
Neg logit contributions | |
boys | -0.122 |
paramed | -0.118 |
chefs | -0.114 |
Lee | -0.111 |
printer | -0.102 |
Token | toys |
Token ID | 14958 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
boys | +0.025 |
paramed | +0.025 |
Lee | +0.024 |
chefs | +0.024 |
principal | +0.022 |
Neg logit contributions | |
day | -0.042 |
night | -0.042 |
nights | -0.041 |
morning | -0.040 |
time | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | -7.35e-02 |
Pos logit contributions | |
day | +0.135 |
night | +0.132 |
nights | +0.131 |
morning | +0.124 |
time | +0.123 |
Neg logit contributions | |
boys | -0.182 |
paramed | -0.180 |
chefs | -0.175 |
Lee | -0.173 |
reporter | -0.165 |
Token | books |
Token ID | 3835 |
Feature activation | -6.05e-02 |
Pos logit contributions | |
day | +0.128 |
night | +0.125 |
nights | +0.122 |
morning | +0.118 |
afternoon | +0.117 |
Neg logit contributions | |
paramed | -0.111 |
boys | -0.107 |
chefs | -0.104 |
Lee | -0.103 |
reporter | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
day | +0.076 |
nights | +0.074 |
night | +0.074 |
morning | +0.069 |
afternoon | +0.069 |
Neg logit contributions | |
boys | -0.115 |
paramed | -0.113 |
chefs | -0.110 |
Lee | -0.109 |
reporter | -0.104 |
Token | One |
Token ID | 1881 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
day | +0.097 |
nights | +0.097 |
night | +0.096 |
fy | +0.091 |
afternoon | +0.091 |
Neg logit contributions | |
boys | -0.100 |
paramed | -0.099 |
Lee | -0.097 |
chefs | -0.096 |
principal | -0.089 |
Token | day |
Token ID | 1110 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
ight | +0.086 |
fy | +0.083 |
Along | +0.083 |
nights | +0.079 |
ingle | +0.071 |
Neg logit contributions | |
paramed | -0.365 |
printer | -0.360 |
mixer | -0.355 |
Lee | -0.354 |
chefs | -0.351 |
Token | , |
Token ID | 11 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
nights | +0.475 |
day | +0.470 |
night | +0.463 |
fy | +0.451 |
afternoon | +0.445 |
Neg logit contributions | |
boys | -0.414 |
paramed | -0.413 |
Lee | -0.410 |
chefs | -0.408 |
printer | -0.370 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.92e-03 |
Pos logit contributions | |
boys | +0.073 |
paramed | +0.072 |
chefs | +0.068 |
Lee | +0.067 |
reporter | +0.064 |
Neg logit contributions | |
day | -0.087 |
night | -0.085 |
nights | -0.084 |
morning | -0.081 |
time | -0.081 |
Token | has |
Token ID | 468 |
Feature activation | +1.69e-03 |
Pos logit contributions | |
boys | +0.010 |
paramed | +0.010 |
chefs | +0.010 |
Lee | +0.009 |
reporter | +0.009 |
Neg logit contributions | |
day | -0.012 |
night | -0.012 |
nights | -0.012 |
morning | -0.011 |
time | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
boys | +0.002 |
paramed | +0.002 |
chefs | +0.002 |
Lee | +0.002 |
reporter | +0.002 |
Neg logit contributions | |
day | -0.003 |
night | -0.003 |
nights | -0.003 |
morning | -0.003 |
time | -0.003 |
Token | new |
Token ID | 649 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
day | +0.020 |
night | +0.019 |
time | +0.019 |
morning | +0.018 |
evening | +0.017 |
Neg logit contributions | |
boys | -0.025 |
chefs | -0.024 |
paramed | -0.023 |
Lee | -0.023 |
principal | -0.022 |
Token | man |
Token ID | 582 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
boys | +0.055 |
wrists | +0.048 |
chefs | +0.048 |
paramed | +0.045 |
groom | +0.044 |
Neg logit contributions | |
day | -0.075 |
night | -0.069 |
time | -0.069 |
morning | -0.067 |
ly | -0.064 |
Token | didn |
Token ID | 1422 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
chefs | +0.240 |
paramed | +0.239 |
Lee | +0.236 |
groom | +0.230 |
boys | +0.225 |
Neg logit contributions | |
day | -0.235 |
night | -0.219 |
time | -0.218 |
nights | -0.213 |
way | -0.205 |
Token | 't |
Token ID | 470 |
Feature activation | -4.84e-03 |
Pos logit contributions | |
nights | +0.065 |
day | +0.065 |
night | +0.064 |
afternoon | +0.061 |
morning | +0.061 |
Neg logit contributions | |
boys | -0.063 |
paramed | -0.062 |
Lee | -0.059 |
chefs | -0.059 |
principal | -0.056 |
Token | like |
Token ID | 588 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
day | +0.008 |
night | +0.008 |
nights | +0.008 |
time | +0.008 |
morning | +0.008 |
Neg logit contributions | |
boys | -0.007 |
paramed | -0.007 |
chefs | -0.007 |
Lee | -0.007 |
printer | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
day | +0.020 |
night | +0.019 |
time | +0.019 |
nights | +0.018 |
morning | +0.018 |
Neg logit contributions | |
paramed | -0.016 |
boys | -0.016 |
chefs | -0.014 |
Lee | -0.014 |
groom | -0.014 |
Token | fight |
Token ID | 1907 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
chefs | +0.101 |
paramed | +0.101 |
printer | +0.101 |
wrists | +0.096 |
Lee | +0.096 |
Neg logit contributions | |
day | -0.113 |
time | -0.110 |
night | -0.102 |
ie | -0.100 |
morning | -0.094 |
Token | and |
Token ID | 290 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
paramed | +0.325 |
boys | +0.321 |
chefs | +0.312 |
Lee | +0.303 |
groom | +0.300 |
Neg logit contributions | |
day | -0.254 |
night | -0.247 |
nights | -0.239 |
time | -0.237 |
morning | -0.230 |
Token | he |
Token ID | 339 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
day | +0.048 |
night | +0.047 |
morning | +0.044 |
nights | +0.044 |
time | +0.043 |
Neg logit contributions | |
paramed | -0.046 |
chefs | -0.045 |
boys | -0.044 |
Lee | -0.042 |
printer | -0.041 |
Token | tried |
Token ID | 3088 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
day | +0.175 |
time | +0.165 |
night | +0.164 |
ie | +0.153 |
morning | +0.152 |
Neg logit contributions | |
chefs | -0.215 |
paramed | -0.206 |
Lee | -0.199 |
principal | -0.194 |
printer | -0.193 |
Token | to |
Token ID | 284 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
paramed | +0.051 |
groom | +0.048 |
boys | +0.047 |
chefs | +0.046 |
printer | +0.046 |
Neg logit contributions | |
day | -0.028 |
time | -0.027 |
night | -0.024 |
way | -0.022 |
morning | -0.022 |
Token | get |
Token ID | 651 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
Lee | +0.101 |
chefs | +0.101 |
paramed | +0.100 |
printer | +0.095 |
reporter | +0.092 |
Neg logit contributions | |
day | -0.110 |
time | -0.110 |
night | -0.101 |
morning | -0.095 |
afternoon | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +8.82e-02 |
Pos logit contributions | |
day | +0.101 |
night | +0.099 |
morning | +0.090 |
time | +0.089 |
afternoon | +0.085 |
Neg logit contributions | |
paramed | -0.192 |
boys | -0.182 |
chefs | -0.180 |
principal | -0.178 |
groom | -0.172 |
Token | I |
Token ID | 314 |
Feature activation | +7.84e-02 |
Pos logit contributions | |
boys | +0.139 |
paramed | +0.138 |
Lee | +0.130 |
principal | +0.130 |
chefs | +0.127 |
Neg logit contributions | |
day | -0.150 |
night | -0.147 |
morning | -0.138 |
time | -0.138 |
evening | -0.131 |
Token | want |
Token ID | 765 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
boys | +0.130 |
paramed | +0.125 |
Lee | +0.125 |
chefs | +0.118 |
wrists | +0.117 |
Neg logit contributions | |
day | -0.128 |
night | -0.126 |
time | -0.123 |
morning | -0.116 |
fy | -0.116 |
Token | that |
Token ID | 326 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
paramed | +0.019 |
boys | +0.018 |
Lee | +0.018 |
chefs | +0.017 |
reporter | +0.017 |
Neg logit contributions | |
day | -0.019 |
night | -0.018 |
nights | -0.017 |
morning | -0.017 |
time | -0.017 |
Token | glass |
Token ID | 5405 |
Feature activation | +7.08e-02 |
Pos logit contributions | |
paramed | +0.082 |
boys | +0.080 |
chefs | +0.078 |
reporter | +0.074 |
principal | +0.073 |
Neg logit contributions | |
day | -0.078 |
night | -0.076 |
morning | -0.072 |
time | -0.069 |
afternoon | -0.068 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
boys | +0.115 |
paramed | +0.115 |
chefs | +0.110 |
Lee | +0.109 |
principal | +0.103 |
Neg logit contributions | |
day | -0.109 |
night | -0.107 |
nights | -0.106 |
morning | -0.101 |
time | -0.101 |
Token |
|
Token ID | 198 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
paramed | +0.294 |
boys | +0.287 |
Lee | +0.285 |
chefs | +0.275 |
printer | +0.264 |
Neg logit contributions | |
nights | -0.288 |
day | -0.285 |
night | -0.281 |
fy | -0.274 |
afternoon | -0.268 |
Token |
|
Token ID | 198 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
paramed | +0.018 |
boys | +0.018 |
Lee | +0.017 |
chefs | +0.017 |
printer | +0.016 |
Neg logit contributions | |
nights | -0.016 |
day | -0.016 |
night | -0.016 |
fy | -0.015 |
afternoon | -0.015 |
Token | Mom |
Token ID | 29252 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
boys | +0.017 |
chefs | +0.016 |
paramed | +0.016 |
Lee | +0.016 |
principal | +0.015 |
Neg logit contributions | |
day | -0.018 |
night | -0.017 |
nights | -0.017 |
morning | -0.016 |
afternoon | -0.016 |
Token | my |
Token ID | 1820 |
Feature activation | -9.06e-02 |
Pos logit contributions | |
boys | +0.072 |
paramed | +0.071 |
chefs | +0.071 |
Lee | +0.068 |
principal | +0.067 |
Neg logit contributions | |
day | -0.038 |
night | -0.037 |
nights | -0.035 |
morning | -0.034 |
time | -0.033 |
Token | said |
Token ID | 531 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
ie | +0.124 |
day | +0.120 |
ly | +0.119 |
of | +0.115 |
night | +0.115 |
Neg logit contributions | |
chefs | -0.168 |
liar | -0.164 |
wasteful | -0.157 |
jury | -0.154 |
principal | -0.151 |
Token | want |
Token ID | 765 |
Feature activation | +6.26e-03 |
Pos logit contributions | |
nights | +0.136 |
day | +0.128 |
night | +0.127 |
Along | +0.124 |
afternoon | +0.124 |
Neg logit contributions | |
boys | -0.100 |
chefs | -0.097 |
paramed | -0.097 |
groom | -0.090 |
Lee | -0.089 |
Token | to |
Token ID | 284 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
paramed | +0.011 |
boys | +0.011 |
chefs | +0.011 |
Lee | +0.011 |
reporter | +0.010 |
Neg logit contributions | |
day | -0.010 |
night | -0.009 |
morning | -0.009 |
nights | -0.009 |
time | -0.009 |
Token | play |
Token ID | 711 |
Feature activation | -8.63e-02 |
Pos logit contributions | |
paramed | +0.109 |
chefs | +0.107 |
boys | +0.106 |
Lee | +0.105 |
reporter | +0.100 |
Neg logit contributions | |
day | -0.097 |
night | -0.095 |
time | -0.090 |
morning | -0.089 |
nights | -0.088 |
Token | with |
Token ID | 351 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
day | +0.112 |
night | +0.104 |
time | +0.103 |
morning | +0.098 |
afternoon | +0.092 |
Neg logit contributions | |
paramed | -0.184 |
Lee | -0.169 |
dads | -0.167 |
chefs | -0.164 |
boys | -0.162 |
Token | me |
Token ID | 502 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
paramed | +0.007 |
boys | +0.006 |
chefs | +0.006 |
Lee | +0.006 |
reporter | +0.006 |
Neg logit contributions | |
day | -0.005 |
night | -0.005 |
nights | -0.005 |
time | -0.005 |
morning | -0.005 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
day | +0.111 |
night | +0.109 |
nights | +0.105 |
morning | +0.104 |
time | +0.103 |
Neg logit contributions | |
paramed | -0.072 |
boys | -0.068 |
chefs | -0.066 |
Lee | -0.064 |
reporter | -0.062 |
Token | asked |
Token ID | 1965 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
boys | +0.238 |
chefs | +0.235 |
paramed | +0.226 |
Lee | +0.205 |
reporter | +0.205 |
Neg logit contributions | |
day | -0.273 |
night | -0.265 |
morning | -0.257 |
time | -0.249 |
evening | -0.243 |
Token | Lily |
Token ID | 20037 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
boys | +0.069 |
chefs | +0.066 |
paramed | +0.065 |
reporter | +0.061 |
principal | +0.060 |
Neg logit contributions | |
day | -0.084 |
night | -0.082 |
time | -0.079 |
morning | -0.078 |
nights | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
boys | +0.102 |
paramed | +0.100 |
chefs | +0.098 |
Lee | +0.097 |
principal | +0.093 |
Neg logit contributions | |
day | -0.059 |
night | -0.058 |
nights | -0.058 |
morning | -0.054 |
afternoon | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | -1.03e-03 |
Pos logit contributions | |
day | +0.014 |
Along | +0.013 |
night | +0.013 |
morning | +0.013 |
time | +0.013 |
Neg logit contributions | |
boys | -0.032 |
chefs | -0.031 |
paramed | -0.030 |
groom | -0.029 |
wasteful | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -7.58e-04 |
Pos logit contributions | |
day | +0.001 |
night | +0.001 |
time | +0.001 |
nights | +0.001 |
morning | +0.001 |
Neg logit contributions | |
boys | -0.002 |
chefs | -0.002 |
paramed | -0.002 |
Lee | -0.002 |
groom | -0.002 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
boys | +0.045 |
Lee | +0.043 |
chefs | +0.042 |
paramed | +0.042 |
principal | +0.040 |
Neg logit contributions | |
day | -0.038 |
night | -0.037 |
nights | -0.035 |
morning | -0.035 |
time | -0.034 |
Token | that |
Token ID | 326 |
Feature activation | +6.17e-02 |
Pos logit contributions | |
day | +0.020 |
night | +0.020 |
nights | +0.020 |
morning | +0.019 |
time | +0.019 |
Neg logit contributions | |
boys | -0.023 |
paramed | -0.023 |
chefs | -0.022 |
Lee | -0.022 |
groom | -0.021 |
Token | she |
Token ID | 673 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
boys | +0.118 |
paramed | +0.117 |
chefs | +0.114 |
Lee | +0.112 |
reporter | +0.107 |
Neg logit contributions | |
day | -0.078 |
night | -0.077 |
nights | -0.076 |
morning | -0.072 |
afternoon | -0.071 |
Token | said |
Token ID | 531 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
day | +0.151 |
night | +0.151 |
fy | +0.147 |
nights | +0.145 |
time | +0.134 |
Neg logit contributions | |
Lee | -0.187 |
paramed | -0.184 |
chefs | -0.175 |
boys | -0.173 |
wrists | -0.169 |
Token | , |
Token ID | 11 |
Feature activation | +9.84e-02 |
Pos logit contributions | |
boys | +0.062 |
chefs | +0.059 |
paramed | +0.057 |
principal | +0.056 |
Lee | +0.056 |
Neg logit contributions | |
day | -0.045 |
night | -0.043 |
morning | -0.042 |
of | -0.041 |
time | -0.040 |
Token | " |
Token ID | 366 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
boys | +0.124 |
chefs | +0.118 |
paramed | +0.101 |
principal | +0.096 |
Lee | +0.095 |
Neg logit contributions | |
day | -0.201 |
night | -0.192 |
Along | -0.190 |
morning | -0.189 |
time | -0.183 |
Token | Thank |
Token ID | 10449 |
Feature activation | +6.11e-02 |
Pos logit contributions | |
paramed | +0.228 |
boys | +0.227 |
chefs | +0.212 |
Lee | +0.206 |
groom | +0.197 |
Neg logit contributions | |
day | -0.520 |
night | -0.514 |
nights | -0.514 |
time | -0.496 |
fy | -0.493 |
Token | you |
Token ID | 345 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
boys | +0.115 |
paramed | +0.114 |
chefs | +0.111 |
Lee | +0.109 |
groom | +0.105 |
Neg logit contributions | |
day | -0.079 |
night | -0.077 |
nights | -0.076 |
morning | -0.073 |
time | -0.072 |
Token | for |
Token ID | 329 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
day | +0.084 |
night | +0.082 |
morning | +0.077 |
nights | +0.076 |
time | +0.076 |
Neg logit contributions | |
boys | -0.128 |
chefs | -0.126 |
paramed | -0.124 |
wrists | -0.123 |
groom | -0.122 |
Token | answering |
Token ID | 18877 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
wrists | +0.055 |
Lee | +0.055 |
boys | +0.054 |
paramed | +0.051 |
groom | +0.051 |
Neg logit contributions | |
day | -0.052 |
night | -0.048 |
nights | -0.047 |
time | -0.046 |
days | -0.045 |
Token | our |
Token ID | 674 |
Feature activation | +8.51e-02 |
Pos logit contributions | |
time | +0.080 |
day | +0.080 |
night | +0.078 |
morning | +0.074 |
nights | +0.072 |
Neg logit contributions | |
boys | -0.105 |
paramed | -0.104 |
chefs | -0.101 |
reporter | -0.099 |
liar | -0.094 |
Token | Jimmy |
Token ID | 12963 |
Feature activation | +9.14e-02 |
Pos logit contributions | |
boys | +0.141 |
paramed | +0.137 |
chefs | +0.131 |
Lee | +0.127 |
reporter | +0.123 |
Neg logit contributions | |
day | -0.166 |
night | -0.165 |
nights | -0.161 |
morning | -0.157 |
time | -0.155 |
Token | had |
Token ID | 550 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
paramed | +0.151 |
chefs | +0.145 |
boys | +0.141 |
Lee | +0.140 |
principal | +0.134 |
Neg logit contributions | |
night | -0.139 |
day | -0.138 |
time | -0.135 |
nights | -0.135 |
morning | -0.128 |
Token | an |
Token ID | 281 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
time | +0.038 |
night | +0.035 |
day | +0.033 |
morning | +0.032 |
nights | +0.031 |
Neg logit contributions | |
boys | -0.039 |
Lee | -0.037 |
reporter | -0.035 |
paramed | -0.034 |
itor | -0.033 |
Token | idea |
Token ID | 2126 |
Feature activation | +5.64e-02 |
Pos logit contributions | |
boys | +0.063 |
Lee | +0.060 |
paramed | +0.060 |
chefs | +0.059 |
groom | +0.058 |
Neg logit contributions | |
day | -0.031 |
night | -0.030 |
time | -0.029 |
morning | -0.028 |
nights | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
boys | +0.100 |
paramed | +0.098 |
Lee | +0.097 |
chefs | +0.096 |
reporter | +0.091 |
Neg logit contributions | |
nights | -0.077 |
night | -0.075 |
day | -0.074 |
afternoon | -0.071 |
fy | -0.071 |
Token | " |
Token ID | 366 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
day | +0.013 |
night | +0.012 |
nights | +0.012 |
morning | +0.012 |
time | +0.011 |
Neg logit contributions | |
boys | -0.021 |
paramed | -0.020 |
chefs | -0.020 |
Lee | -0.019 |
groom | -0.019 |
Token | Mom |
Token ID | 29252 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
boys | +0.253 |
paramed | +0.247 |
chefs | +0.236 |
Lee | +0.232 |
printer | +0.212 |
Neg logit contributions | |
day | -0.477 |
night | -0.470 |
nights | -0.468 |
morning | -0.451 |
afternoon | -0.450 |
Token | my |
Token ID | 1820 |
Feature activation | -8.29e-02 |
Pos logit contributions | |
paramed | +0.093 |
boys | +0.092 |
chefs | +0.089 |
reporter | +0.088 |
groom | +0.087 |
Neg logit contributions | |
day | -0.045 |
night | -0.043 |
nights | -0.039 |
morning | -0.039 |
time | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | +9.21e-02 |
Pos logit contributions | |
day | +0.113 |
night | +0.112 |
time | +0.104 |
morning | +0.103 |
fy | +0.096 |
Neg logit contributions | |
paramed | -0.155 |
principal | -0.146 |
boys | -0.145 |
wrists | -0.144 |
chefs | -0.143 |
Token | I |
Token ID | 314 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
boys | +0.136 |
paramed | +0.132 |
wrists | +0.123 |
Lee | +0.121 |
reporter | +0.118 |
Neg logit contributions | |
day | -0.165 |
night | -0.162 |
time | -0.155 |
morning | -0.153 |
nights | -0.150 |
Token | know |
Token ID | 760 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
boys | +0.129 |
paramed | +0.124 |
Lee | +0.122 |
wrists | +0.120 |
chefs | +0.119 |
Neg logit contributions | |
day | -0.137 |
night | -0.134 |
time | -0.134 |
nights | -0.125 |
morning | -0.125 |
Token | was |
Token ID | 373 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
nights | +0.111 |
day | +0.110 |
night | +0.110 |
Along | +0.106 |
fy | +0.106 |
Neg logit contributions | |
boys | -0.126 |
chefs | -0.123 |
paramed | -0.122 |
Lee | -0.120 |
printer | -0.117 |
Token | about |
Token ID | 546 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
boys | +0.023 |
paramed | +0.023 |
chefs | +0.022 |
Lee | +0.021 |
reporter | +0.020 |
Neg logit contributions | |
day | -0.027 |
night | -0.026 |
nights | -0.026 |
morning | -0.025 |
time | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | +5.36e-02 |
Pos logit contributions | |
boys | +0.022 |
paramed | +0.022 |
chefs | +0.021 |
Lee | +0.021 |
reporter | +0.019 |
Neg logit contributions | |
day | -0.030 |
night | -0.030 |
nights | -0.030 |
morning | -0.028 |
time | -0.028 |
Token | happen |
Token ID | 1645 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
boys | +0.074 |
paramed | +0.074 |
chefs | +0.071 |
Lee | +0.070 |
reporter | +0.065 |
Neg logit contributions | |
day | -0.096 |
night | -0.094 |
nights | -0.093 |
morning | -0.090 |
time | -0.090 |
Token | ! |
Token ID | 0 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
boys | +0.031 |
paramed | +0.031 |
chefs | +0.030 |
Lee | +0.030 |
printer | +0.028 |
Neg logit contributions | |
day | -0.024 |
night | -0.024 |
nights | -0.024 |
morning | -0.022 |
afternoon | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +9.76e-03 |
Pos logit contributions | |
Lee | +0.039 |
paramed | +0.039 |
boys | +0.039 |
chefs | +0.037 |
principal | +0.036 |
Neg logit contributions | |
nights | -0.038 |
day | -0.038 |
night | -0.038 |
afternoon | -0.036 |
fy | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
paramed | +0.015 |
boys | +0.014 |
Lee | +0.014 |
chefs | +0.013 |
principal | +0.013 |
Neg logit contributions | |
nights | -0.016 |
night | -0.015 |
day | -0.015 |
fy | -0.015 |
morning | -0.014 |
Token | When |
Token ID | 2215 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
paramed | +0.020 |
boys | +0.020 |
Lee | +0.019 |
chefs | +0.019 |
principal | +0.018 |
Neg logit contributions | |
night | -0.025 |
day | -0.025 |
nights | -0.024 |
morning | -0.023 |
fy | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
day | +0.285 |
nights | +0.281 |
night | +0.280 |
afternoon | +0.267 |
morning | +0.267 |
Neg logit contributions | |
boys | -0.215 |
paramed | -0.211 |
chefs | -0.204 |
Lee | -0.202 |
printer | -0.188 |
Token | painting |
Token ID | 12036 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
paramed | +0.084 |
boys | +0.083 |
printer | +0.083 |
chefs | +0.080 |
reporter | +0.079 |
Neg logit contributions | |
nights | -0.077 |
fy | -0.077 |
Along | -0.076 |
ingle | -0.076 |
adows | -0.075 |
Token | was |
Token ID | 373 |
Feature activation | +5.17e-02 |
Pos logit contributions | |
boys | +0.034 |
paramed | +0.034 |
chefs | +0.033 |
Lee | +0.032 |
reporter | +0.031 |
Neg logit contributions | |
day | -0.036 |
night | -0.036 |
nights | -0.035 |
morning | -0.034 |
time | -0.034 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
nights | +0.251 |
day | +0.249 |
night | +0.246 |
afternoon | +0.235 |
morning | +0.234 |
Neg logit contributions | |
boys | -0.249 |
paramed | -0.234 |
chefs | -0.233 |
Lee | -0.233 |
printer | -0.215 |
Token |
|
Token ID | 198 |
Feature activation | -9.05e-04 |
Pos logit contributions | |
boys | +0.260 |
paramed | +0.259 |
Lee | +0.250 |
chefs | +0.245 |
printer | +0.228 |
Neg logit contributions | |
day | -0.296 |
night | -0.293 |
nights | -0.292 |
fy | -0.277 |
afternoon | -0.277 |
Token |
|
Token ID | 198 |
Feature activation | -1.86e-04 |
Pos logit contributions | |
day | +0.001 |
night | +0.001 |
nights | +0.001 |
morning | +0.001 |
afternoon | +0.001 |
Neg logit contributions | |
boys | -0.002 |
paramed | -0.002 |
chefs | -0.001 |
Lee | -0.001 |
principal | -0.001 |
Token | Dog |
Token ID | 32942 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
day | +0.000 |
nights | +0.000 |
night | +0.000 |
time | +0.000 |
afternoon | +0.000 |
Neg logit contributions | |
boys | -0.000 |
chefs | -0.000 |
paramed | -0.000 |
Lee | -0.000 |
printer | -0.000 |
Token | bark |
Token ID | 21405 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
boys | +0.028 |
paramed | +0.027 |
Lee | +0.026 |
chefs | +0.026 |
printer | +0.024 |
Neg logit contributions | |
day | -0.034 |
night | -0.033 |
nights | -0.033 |
morning | -0.031 |
afternoon | -0.031 |
Token | ed |
Token ID | 276 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
boys | +0.103 |
paramed | +0.102 |
chefs | +0.100 |
Lee | +0.099 |
principal | +0.095 |
Neg logit contributions | |
day | -0.063 |
night | -0.062 |
nights | -0.062 |
morning | -0.058 |
time | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
boys | +0.091 |
paramed | +0.090 |
chefs | +0.088 |
Lee | +0.086 |
principal | +0.082 |
Neg logit contributions | |
day | -0.073 |
night | -0.071 |
nights | -0.071 |
morning | -0.067 |
time | -0.067 |
Token | as |
Token ID | 355 |
Feature activation | -9.18e-02 |
Pos logit contributions | |
boys | +0.116 |
chefs | +0.113 |
paramed | +0.110 |
principal | +0.107 |
liar | +0.103 |
Neg logit contributions | |
day | -0.154 |
night | -0.147 |
Along | -0.146 |
morning | -0.145 |
nights | -0.145 |
Token | if |
Token ID | 611 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
nights | +0.154 |
day | +0.153 |
night | +0.152 |
fy | +0.148 |
afternoon | +0.147 |
Neg logit contributions | |
boys | -0.131 |
paramed | -0.128 |
Lee | -0.125 |
chefs | -0.121 |
groom | -0.115 |
Token | to |
Token ID | 284 |
Feature activation | +6.80e-02 |
Pos logit contributions | |
boys | +0.055 |
paramed | +0.055 |
chefs | +0.054 |
Lee | +0.050 |
principal | +0.050 |
Neg logit contributions | |
day | -0.078 |
night | -0.076 |
nights | -0.075 |
morning | -0.073 |
time | -0.072 |
Token | say |
Token ID | 910 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
boys | +0.095 |
paramed | +0.092 |
Lee | +0.091 |
chefs | +0.089 |
groom | +0.085 |
Neg logit contributions | |
nights | -0.120 |
day | -0.119 |
night | -0.118 |
fy | -0.115 |
afternoon | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
day | +0.127 |
nights | +0.127 |
night | +0.124 |
morning | +0.118 |
afternoon | +0.118 |
Neg logit contributions | |
boys | -0.146 |
paramed | -0.142 |
chefs | -0.138 |
Lee | -0.138 |
groom | -0.130 |
Token | She |
Token ID | 1375 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
nights | +0.041 |
day | +0.041 |
night | +0.040 |
fy | +0.038 |
afternoon | +0.038 |
Neg logit contributions | |
paramed | -0.051 |
Lee | -0.050 |
chefs | -0.050 |
boys | -0.050 |
principal | -0.047 |
Token | learned |
Token ID | 4499 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
day | +0.049 |
night | +0.048 |
time | +0.046 |
nights | +0.046 |
morning | +0.044 |
Neg logit contributions | |
paramed | -0.044 |
chefs | -0.041 |
Lee | -0.041 |
boys | -0.040 |
printer | -0.038 |
Token | that |
Token ID | 326 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
boys | +0.088 |
paramed | +0.086 |
chefs | +0.084 |
Lee | +0.082 |
printer | +0.078 |
Neg logit contributions | |
day | -0.104 |
night | -0.103 |
nights | -0.099 |
morning | -0.097 |
time | -0.097 |
Token | the |
Token ID | 262 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
boys | +0.080 |
paramed | +0.079 |
chefs | +0.077 |
Lee | +0.075 |
printer | +0.071 |
Neg logit contributions | |
day | -0.084 |
night | -0.083 |
nights | -0.082 |
morning | -0.078 |
time | -0.078 |
Token | clown |
Token ID | 25573 |
Feature activation | +6.83e-02 |
Pos logit contributions | |
boys | +0.052 |
paramed | +0.051 |
chefs | +0.049 |
Lee | +0.049 |
groom | +0.045 |
Neg logit contributions | |
day | -0.074 |
night | -0.071 |
time | -0.069 |
nights | -0.069 |
morning | -0.068 |
Token | was |
Token ID | 373 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
boys | +0.099 |
paramed | +0.095 |
chefs | +0.092 |
Lee | +0.091 |
reporter | +0.086 |
Neg logit contributions | |
day | -0.116 |
nights | -0.116 |
night | -0.115 |
morning | -0.110 |
afternoon | -0.110 |
Token | not |
Token ID | 407 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
paramed | +0.041 |
boys | +0.040 |
Lee | +0.039 |
chefs | +0.038 |
printer | +0.036 |
Neg logit contributions | |
day | -0.054 |
night | -0.053 |
nights | -0.051 |
time | -0.051 |
morning | -0.050 |
Token | scary |
Token ID | 14343 |
Feature activation | +8.14e-03 |
Pos logit contributions | |
boys | +0.085 |
paramed | +0.085 |
chefs | +0.081 |
Lee | +0.081 |
principal | +0.077 |
Neg logit contributions | |
day | -0.091 |
night | -0.089 |
nights | -0.087 |
time | -0.085 |
morning | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
boys | +0.013 |
chefs | +0.013 |
Lee | +0.013 |
paramed | +0.012 |
reporter | +0.012 |
Neg logit contributions | |
nights | -0.012 |
day | -0.012 |
night | -0.012 |
Along | -0.012 |
fy | -0.012 |
Token | He |
Token ID | 679 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
nights | +0.040 |
day | +0.040 |
night | +0.039 |
fy | +0.037 |
afternoon | +0.037 |
Neg logit contributions | |
paramed | -0.041 |
Lee | -0.041 |
chefs | -0.041 |
boys | -0.040 |
principal | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
paramed | +0.034 |
Lee | +0.034 |
chefs | +0.033 |
boys | +0.033 |
principal | +0.031 |
Neg logit contributions | |
nights | -0.040 |
day | -0.039 |
fy | -0.039 |
night | -0.038 |
afternoon | -0.037 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
day | +0.068 |
night | +0.067 |
nights | +0.065 |
morning | +0.063 |
afternoon | +0.062 |
Neg logit contributions | |
paramed | -0.053 |
boys | -0.052 |
chefs | -0.050 |
Lee | -0.048 |
reporter | -0.046 |
Token | lots |
Token ID | 6041 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
boys | +0.089 |
wrists | +0.081 |
paramed | +0.081 |
Lee | +0.081 |
argue | +0.079 |
Neg logit contributions | |
day | -0.064 |
night | -0.063 |
of | -0.058 |
morning | -0.058 |
nights | -0.057 |
Token | of |
Token ID | 286 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
Along | +0.036 |
spine | +0.035 |
night | +0.035 |
nights | +0.034 |
rated | +0.033 |
Neg logit contributions | |
paramed | -0.107 |
chefs | -0.103 |
boys | -0.100 |
Lee | -0.099 |
principal | -0.097 |
Token | tasty |
Token ID | 25103 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
day | +0.167 |
night | +0.158 |
nights | +0.158 |
time | +0.156 |
morning | +0.150 |
Neg logit contributions | |
boys | -0.187 |
Lee | -0.176 |
paramed | -0.169 |
chefs | -0.168 |
reporter | -0.166 |
Token | cakes |
Token ID | 30849 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
day | +0.043 |
night | +0.043 |
nights | +0.042 |
morning | +0.041 |
time | +0.040 |
Neg logit contributions | |
boys | -0.033 |
paramed | -0.032 |
chefs | -0.031 |
Lee | -0.031 |
reporter | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
boys | +0.258 |
paramed | +0.250 |
chefs | +0.249 |
Lee | +0.247 |
reporter | +0.231 |
Neg logit contributions | |
nights | -0.189 |
day | -0.188 |
night | -0.187 |
morning | -0.176 |
afternoon | -0.174 |
Token | He |
Token ID | 679 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
nights | +0.063 |
day | +0.062 |
night | +0.061 |
fy | +0.059 |
afternoon | +0.058 |
Neg logit contributions | |
Lee | -0.067 |
boys | -0.065 |
paramed | -0.064 |
chefs | -0.064 |
principal | -0.061 |
Token | wanted |
Token ID | 2227 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
day | +0.058 |
night | +0.055 |
nights | +0.055 |
time | +0.053 |
afternoon | +0.051 |
Neg logit contributions | |
Lee | -0.061 |
paramed | -0.061 |
chefs | -0.059 |
boys | -0.058 |
principal | -0.056 |
Token | one |
Token ID | 530 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
day | +0.063 |
time | +0.061 |
night | +0.060 |
nights | +0.056 |
morning | +0.055 |
Neg logit contributions | |
paramed | -0.097 |
boys | -0.096 |
reporter | -0.092 |
Lee | -0.089 |
wrists | -0.089 |
Token | of |
Token ID | 286 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
nights | +0.073 |
night | +0.069 |
Along | +0.067 |
fy | +0.067 |
day | +0.067 |
Neg logit contributions | |
boys | -0.060 |
chefs | -0.059 |
Lee | -0.059 |
paramed | -0.058 |
printer | -0.053 |
Token | was |
Token ID | 373 |
Feature activation | +4.03e-03 |
Pos logit contributions | |
nights | +0.116 |
day | +0.113 |
fy | +0.112 |
night | +0.112 |
Along | +0.108 |
Neg logit contributions | |
boys | -0.088 |
paramed | -0.086 |
Lee | -0.084 |
chefs | -0.084 |
printer | -0.077 |
Token | a |
Token ID | 257 |
Feature activation | +8.90e-04 |
Pos logit contributions | |
chefs | +0.003 |
boys | +0.003 |
paramed | +0.002 |
printer | +0.002 |
Lee | +0.002 |
Neg logit contributions | |
fy | -0.010 |
nights | -0.010 |
spine | -0.010 |
rated | -0.009 |
way | -0.009 |
Token | little |
Token ID | 1310 |
Feature activation | -9.05e-02 |
Pos logit contributions | |
boys | +0.002 |
Lee | +0.002 |
wrists | +0.002 |
paramed | +0.002 |
chefs | +0.002 |
Neg logit contributions | |
day | -0.001 |
night | -0.001 |
time | -0.001 |
morning | -0.001 |
afternoon | -0.001 |
Token | boy |
Token ID | 2933 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
day | +0.164 |
night | +0.160 |
morning | +0.150 |
afternoon | +0.144 |
evening | +0.142 |
Neg logit contributions | |
boys | -0.137 |
paramed | -0.130 |
principal | -0.128 |
wrists | -0.127 |
Lee | -0.124 |
Token | named |
Token ID | 3706 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
boys | +0.098 |
paramed | +0.094 |
chefs | +0.089 |
Lee | +0.086 |
printer | +0.079 |
Neg logit contributions | |
nights | -0.233 |
day | -0.232 |
night | -0.231 |
fy | -0.224 |
morning | -0.223 |
Token | Tim |
Token ID | 5045 |
Feature activation | +7.18e-03 |
Pos logit contributions | |
nights | +0.132 |
day | +0.125 |
afternoon | +0.124 |
night | +0.124 |
fy | +0.123 |
Neg logit contributions | |
chefs | -0.043 |
boys | -0.043 |
Lee | -0.040 |
paramed | -0.036 |
printer | -0.032 |
Token | my |
Token ID | 1820 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
boys | +0.010 |
chefs | +0.010 |
Lee | +0.009 |
paramed | +0.009 |
reporter | +0.008 |
Neg logit contributions | |
nights | -0.013 |
fy | -0.013 |
Along | -0.012 |
olly | -0.011 |
andel | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
nights | +0.154 |
fy | +0.146 |
day | +0.144 |
night | +0.143 |
morning | +0.139 |
Neg logit contributions | |
boys | -0.204 |
paramed | -0.199 |
chefs | -0.194 |
Lee | -0.190 |
printer | -0.184 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
nights | +0.115 |
night | +0.110 |
day | +0.110 |
fy | +0.109 |
afternoon | +0.106 |
Neg logit contributions | |
boys | -0.053 |
paramed | -0.052 |
chefs | -0.052 |
Lee | -0.052 |
printer | -0.043 |
Token | my |
Token ID | 1820 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
boys | +0.025 |
chefs | +0.024 |
paramed | +0.024 |
Lee | +0.023 |
printer | +0.022 |
Neg logit contributions | |
nights | -0.024 |
fy | -0.023 |
Along | -0.023 |
day | -0.022 |
night | -0.022 |
Token | loved |
Token ID | 6151 |
Feature activation | -6.40e-02 |
Pos logit contributions | |
nights | +0.194 |
day | +0.189 |
night | +0.188 |
fy | +0.184 |
morning | +0.181 |
Neg logit contributions | |
boys | -0.156 |
paramed | -0.151 |
chefs | -0.147 |
Lee | -0.143 |
printer | -0.137 |
Token | said |
Token ID | 531 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
day | +0.001 |
night | +0.001 |
time | +0.001 |
nights | +0.001 |
morning | +0.001 |
Neg logit contributions | |
paramed | -0.001 |
chefs | -0.001 |
Lee | -0.001 |
boys | -0.001 |
principal | -0.001 |
Token | that |
Token ID | 326 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
boys | +0.042 |
chefs | +0.041 |
paramed | +0.040 |
Lee | +0.039 |
principal | +0.039 |
Neg logit contributions | |
day | -0.035 |
time | -0.034 |
of | -0.033 |
morning | -0.032 |
night | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
paramed | +0.098 |
boys | +0.088 |
chefs | +0.085 |
Lee | +0.084 |
wrists | +0.084 |
Neg logit contributions | |
day | -0.102 |
night | -0.093 |
time | -0.089 |
nights | -0.089 |
morning | -0.086 |
Token | other |
Token ID | 584 |
Feature activation | -7.03e-02 |
Pos logit contributions | |
paramed | +0.074 |
boys | +0.074 |
Lee | +0.070 |
chefs | +0.069 |
reporter | +0.065 |
Neg logit contributions | |
day | -0.076 |
night | -0.072 |
time | -0.069 |
nights | -0.069 |
morning | -0.069 |
Token | team |
Token ID | 1074 |
Feature activation | +9.21e-02 |
Pos logit contributions | |
day | +0.128 |
night | +0.122 |
nights | +0.121 |
time | +0.119 |
morning | +0.116 |
Neg logit contributions | |
paramed | -0.102 |
boys | -0.099 |
chefs | -0.095 |
Lee | -0.093 |
reporter | -0.090 |
Token | couldn |
Token ID | 3521 |
Feature activation | -9.00e-02 |
Pos logit contributions | |
paramed | +0.151 |
boys | +0.149 |
chefs | +0.146 |
Lee | +0.143 |
groom | +0.137 |
Neg logit contributions | |
day | -0.146 |
night | -0.140 |
nights | -0.137 |
time | -0.134 |
morning | -0.132 |
Token | 't |
Token ID | 470 |
Feature activation | +4.84e-03 |
Pos logit contributions | |
day | +0.137 |
night | +0.135 |
nights | +0.135 |
morning | +0.127 |
afternoon | +0.127 |
Neg logit contributions | |
boys | -0.147 |
paramed | -0.145 |
chefs | -0.140 |
Lee | -0.139 |
printer | -0.131 |
Token | join |
Token ID | 4654 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
boys | +0.007 |
paramed | +0.007 |
chefs | +0.007 |
Lee | +0.007 |
printer | +0.007 |
Neg logit contributions | |
day | -0.008 |
night | -0.008 |
nights | -0.008 |
time | -0.007 |
morning | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -4.54e-03 |
Pos logit contributions | |
paramed | +0.031 |
boys | +0.030 |
chefs | +0.029 |
Lee | +0.028 |
reporter | +0.027 |
Neg logit contributions | |
day | -0.038 |
night | -0.037 |
nights | -0.037 |
time | -0.036 |
morning | -0.036 |
Token | They |
Token ID | 1119 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
day | +0.006 |
night | +0.006 |
Along | +0.006 |
time | +0.006 |
morning | +0.006 |
Neg logit contributions | |
boys | -0.008 |
paramed | -0.008 |
chefs | -0.008 |
groom | -0.008 |
reporter | -0.008 |
Token | said |
Token ID | 531 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
chefs | +0.096 |
paramed | +0.094 |
boys | +0.091 |
Lee | +0.091 |
principal | +0.087 |
Neg logit contributions | |
day | -0.079 |
night | -0.077 |
time | -0.073 |
nights | -0.070 |
morning | -0.068 |
Token | she |
Token ID | 673 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
day | +0.066 |
night | +0.065 |
nights | +0.065 |
morning | +0.061 |
time | +0.061 |
Neg logit contributions | |
boys | -0.062 |
paramed | -0.061 |
chefs | -0.059 |
Lee | -0.059 |
reporter | -0.055 |
Token | loved |
Token ID | 6151 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
day | +0.226 |
night | +0.221 |
nights | +0.217 |
morning | +0.208 |
time | +0.207 |
Neg logit contributions | |
boys | -0.221 |
paramed | -0.219 |
chefs | -0.212 |
Lee | -0.211 |
principal | -0.197 |
Token | to |
Token ID | 284 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
day | +0.098 |
night | +0.096 |
nights | +0.093 |
morning | +0.090 |
time | +0.089 |
Neg logit contributions | |
boys | -0.096 |
paramed | -0.095 |
Lee | -0.090 |
chefs | -0.089 |
reporter | -0.086 |
Token | peek |
Token ID | 27185 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
boys | +0.094 |
paramed | +0.093 |
chefs | +0.090 |
Lee | +0.089 |
reporter | +0.084 |
Neg logit contributions | |
day | -0.092 |
night | -0.090 |
nights | -0.089 |
morning | -0.085 |
time | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
day | +0.075 |
nights | +0.073 |
night | +0.071 |
afternoon | +0.071 |
evening | +0.069 |
Neg logit contributions | |
boys | -0.085 |
chefs | -0.083 |
paramed | -0.081 |
Lee | -0.078 |
groom | -0.078 |
Token | She |
Token ID | 1375 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
nights | +0.060 |
fy | +0.057 |
ight | +0.054 |
day | +0.054 |
night | +0.053 |
Neg logit contributions | |
Lee | -0.082 |
paramed | -0.075 |
boys | -0.073 |
wrists | -0.072 |
principal | -0.071 |
Token | would |
Token ID | 561 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
nights | +0.092 |
day | +0.092 |
night | +0.091 |
fy | +0.089 |
morning | +0.087 |
Neg logit contributions | |
boys | -0.063 |
paramed | -0.060 |
chefs | -0.057 |
Lee | -0.057 |
groom | -0.056 |
Token | peek |
Token ID | 27185 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
day | +0.037 |
nights | +0.037 |
night | +0.037 |
afternoon | +0.035 |
morning | +0.035 |
Neg logit contributions | |
boys | -0.033 |
paramed | -0.032 |
chefs | -0.031 |
Lee | -0.030 |
groom | -0.029 |
Token | in |
Token ID | 287 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
day | +0.072 |
nights | +0.068 |
night | +0.068 |
afternoon | +0.067 |
time | +0.066 |
Neg logit contributions | |
boys | -0.090 |
chefs | -0.089 |
paramed | -0.087 |
Lee | -0.086 |
groom | -0.083 |
Token | the |
Token ID | 262 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
day | +0.095 |
nights | +0.094 |
night | +0.094 |
fy | +0.091 |
morning | +0.090 |
Neg logit contributions | |
boys | -0.058 |
paramed | -0.057 |
chefs | -0.056 |
Lee | -0.055 |
printer | -0.051 |
Token | trees |
Token ID | 7150 |
Feature activation | -7.80e-02 |
Pos logit contributions | |
paramed | +0.019 |
printer | +0.018 |
principal | +0.017 |
Lee | +0.017 |
chefs | +0.017 |
Neg logit contributions | |
nights | -0.038 |
fy | -0.037 |
t | -0.036 |
day | -0.036 |
ight | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | +7.32e-02 |
Pos logit contributions | |
ight | +0.140 |
fy | +0.140 |
rated | +0.138 |
found | +0.138 |
adow | +0.137 |
Neg logit contributions | |
paramed | -0.086 |
Lee | -0.083 |
boys | -0.079 |
chefs | -0.077 |
printer | -0.072 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
boys | +0.097 |
paramed | +0.094 |
chefs | +0.090 |
Lee | +0.087 |
principal | +0.085 |
Neg logit contributions | |
day | -0.142 |
night | -0.140 |
nights | -0.136 |
morning | -0.134 |
time | -0.133 |
Token | started |
Token ID | 2067 |
Feature activation | -5.74e-03 |
Pos logit contributions | |
paramed | +0.060 |
chefs | +0.058 |
Lee | +0.057 |
boys | +0.057 |
reporter | +0.054 |
Neg logit contributions | |
night | -0.073 |
day | -0.073 |
nights | -0.071 |
time | -0.071 |
morning | -0.068 |
Token | to |
Token ID | 284 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
day | +0.008 |
night | +0.008 |
time | +0.007 |
nights | +0.007 |
morning | +0.007 |
Neg logit contributions | |
boys | -0.010 |
chefs | -0.010 |
reporter | -0.010 |
paramed | -0.010 |
Lee | -0.010 |
Token | feel |
Token ID | 1254 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
chefs | +0.100 |
paramed | +0.097 |
reporter | +0.096 |
Lee | +0.095 |
principal | +0.094 |
Neg logit contributions | |
night | -0.102 |
day | -0.101 |
time | -0.096 |
morning | -0.093 |
evening | -0.091 |
Token | dizz |
Token ID | 29527 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
boys | +0.218 |
paramed | +0.216 |
chefs | +0.209 |
Lee | +0.206 |
printer | +0.194 |
Neg logit contributions | |
day | -0.220 |
night | -0.217 |
nights | -0.215 |
morning | -0.205 |
afternoon | -0.204 |
Token | y |
Token ID | 88 |
Feature activation | -5.74e-02 |
Pos logit contributions | |
Along | +0.058 |
nights | +0.057 |
trek | +0.056 |
journey | +0.054 |
day | +0.052 |
Neg logit contributions | |
chefs | -0.040 |
paramed | -0.040 |
Lee | -0.039 |
boys | -0.038 |
wasteful | -0.034 |
Token | from |
Token ID | 422 |
Feature activation | -3.06e-04 |
Pos logit contributions | |
nights | +0.091 |
Along | +0.090 |
found | +0.088 |
day | +0.085 |
fy | +0.085 |
Neg logit contributions | |
boys | -0.090 |
chefs | -0.083 |
Lee | -0.081 |
paramed | -0.081 |
fighting | -0.075 |
Token | lifting |
Token ID | 16842 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
nights | +0.001 |
day | +0.001 |
fy | +0.001 |
night | +0.001 |
Along | +0.001 |
Neg logit contributions | |
boys | -0.000 |
Lee | -0.000 |
paramed | -0.000 |
chefs | -0.000 |
groom | -0.000 |
Token | Mitt |
Token ID | 16627 |
Feature activation | +8.77e-02 |
Pos logit contributions | |
day | +0.041 |
night | +0.040 |
nights | +0.039 |
time | +0.038 |
morning | +0.038 |
Neg logit contributions | |
boys | -0.040 |
paramed | -0.040 |
chefs | -0.038 |
Lee | -0.038 |
reporter | -0.036 |
Token | ens |
Token ID | 641 |
Feature activation | -8.02e-03 |
Pos logit contributions | |
boys | +0.153 |
paramed | +0.150 |
Lee | +0.146 |
chefs | +0.146 |
reporter | +0.136 |
Neg logit contributions | |
day | -0.125 |
nights | -0.123 |
night | -0.122 |
afternoon | -0.115 |
morning | -0.114 |
Token | and |
Token ID | 290 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
nights | +0.189 |
day | +0.183 |
fy | +0.181 |
night | +0.180 |
time | +0.176 |
Neg logit contributions | |
boys | -0.174 |
Lee | -0.172 |
chefs | -0.164 |
paramed | -0.159 |
reporter | -0.155 |
Token | walked |
Token ID | 6807 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
day | +0.051 |
nights | +0.050 |
time | +0.049 |
night | +0.049 |
fy | +0.049 |
Neg logit contributions | |
boys | -0.036 |
Lee | -0.035 |
paramed | -0.034 |
chefs | -0.034 |
groom | -0.032 |
Token | until |
Token ID | 1566 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
nights | +0.163 |
fy | +0.159 |
found | +0.158 |
day | +0.158 |
ight | +0.157 |
Neg logit contributions | |
boys | -0.191 |
Lee | -0.190 |
chefs | -0.181 |
reporter | -0.170 |
paramed | -0.169 |
Token | she |
Token ID | 673 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
fy | +0.208 |
ight | +0.203 |
found | +0.202 |
adow | +0.195 |
nights | +0.193 |
Neg logit contributions | |
Lee | -0.157 |
paramed | -0.148 |
boys | -0.142 |
principal | -0.132 |
chefs | -0.131 |
Token | was |
Token ID | 373 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
day | +0.244 |
nights | +0.242 |
night | +0.238 |
fy | +0.234 |
afternoon | +0.234 |
Neg logit contributions | |
boys | -0.130 |
paramed | -0.126 |
chefs | -0.121 |
Lee | -0.120 |
groom | -0.112 |
Token | very |
Token ID | 845 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
boys | +0.039 |
chefs | +0.038 |
paramed | +0.038 |
Lee | +0.038 |
groom | +0.036 |
Neg logit contributions | |
nights | -0.042 |
day | -0.041 |
fy | -0.040 |
night | -0.040 |
afternoon | -0.039 |
Token | tired |
Token ID | 10032 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
nights | +0.037 |
day | +0.036 |
night | +0.034 |
Along | +0.034 |
afternoon | +0.034 |
Neg logit contributions | |
chefs | -0.032 |
paramed | -0.032 |
boys | -0.031 |
Lee | -0.030 |
groom | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
found | +0.061 |
Along | +0.060 |
nights | +0.060 |
fy | +0.059 |
afternoon | +0.057 |
Neg logit contributions | |
Lee | -0.053 |
boys | -0.052 |
chefs | -0.051 |
paramed | -0.050 |
reporter | -0.045 |
Token | She |
Token ID | 1375 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
nights | +0.054 |
iest | +0.052 |
night | +0.052 |
fy | +0.052 |
day | +0.052 |
Neg logit contributions | |
Lee | -0.040 |
paramed | -0.037 |
principal | -0.033 |
chefs | -0.033 |
boys | -0.033 |
Token | was |
Token ID | 373 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
day | +0.060 |
nights | +0.059 |
night | +0.059 |
morning | +0.057 |
afternoon | +0.057 |
Neg logit contributions | |
boys | -0.043 |
paramed | -0.042 |
chefs | -0.040 |
Lee | -0.039 |
groom | -0.038 |
Token | strong |
Token ID | 1913 |
Feature activation | +4.35e-02 |
Pos logit contributions | |
boys | +0.038 |
paramed | +0.038 |
chefs | +0.037 |
Lee | +0.036 |
groom | +0.034 |
Neg logit contributions | |
day | -0.044 |
nights | -0.044 |
night | -0.043 |
fy | -0.041 |
morning | -0.041 |
Token | at |
Token ID | 379 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
paramed | +0.007 |
boys | +0.007 |
chefs | +0.007 |
Lee | +0.007 |
principal | +0.006 |
Neg logit contributions | |
day | -0.006 |
night | -0.005 |
nights | -0.005 |
time | -0.005 |
morning | -0.005 |
Token | all |
Token ID | 477 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
day | +0.063 |
night | +0.062 |
nights | +0.060 |
time | +0.059 |
morning | +0.058 |
Neg logit contributions | |
paramed | -0.052 |
boys | -0.050 |
chefs | -0.048 |
Lee | -0.047 |
reporter | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
Along | +0.045 |
fy | +0.044 |
nights | +0.044 |
night | +0.043 |
day | +0.043 |
Neg logit contributions | |
boys | -0.023 |
chefs | -0.021 |
Lee | -0.021 |
paramed | -0.021 |
printer | -0.020 |
Token | buttons |
Token ID | 12163 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
boys | +0.069 |
chefs | +0.068 |
paramed | +0.065 |
Lee | +0.063 |
printer | +0.063 |
Neg logit contributions | |
nights | -0.070 |
fy | -0.069 |
Along | -0.068 |
day | -0.068 |
night | -0.067 |
Token | and |
Token ID | 290 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
day | +0.018 |
nights | +0.018 |
night | +0.018 |
afternoon | +0.017 |
morning | +0.017 |
Neg logit contributions | |
boys | -0.019 |
paramed | -0.018 |
chefs | -0.018 |
Lee | -0.018 |
printer | -0.017 |
Token | lights |
Token ID | 7588 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
night | +0.047 |
day | +0.047 |
nights | +0.045 |
morning | +0.044 |
time | +0.044 |
Neg logit contributions | |
paramed | -0.040 |
boys | -0.036 |
chefs | -0.034 |
reporter | -0.034 |
Lee | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
nights | +0.081 |
day | +0.080 |
night | +0.078 |
afternoon | +0.075 |
Along | +0.075 |
Neg logit contributions | |
boys | -0.083 |
paramed | -0.079 |
chefs | -0.078 |
Lee | -0.078 |
groom | -0.073 |
Token | They |
Token ID | 1119 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
day | +0.046 |
night | +0.045 |
nights | +0.044 |
morning | +0.043 |
Along | +0.043 |
Neg logit contributions | |
boys | -0.044 |
paramed | -0.044 |
chefs | -0.042 |
groom | -0.040 |
Lee | -0.040 |
Token | are |
Token ID | 389 |
Feature activation | +9.21e-02 |
Pos logit contributions | |
paramed | +0.055 |
boys | +0.055 |
chefs | +0.052 |
Lee | +0.052 |
principal | +0.048 |
Neg logit contributions | |
day | -0.076 |
night | -0.075 |
nights | -0.074 |
morning | -0.071 |
time | -0.071 |
Token | incredible |
Token ID | 8082 |
Feature activation | -5.27e-02 |
Pos logit contributions | |
paramed | +0.153 |
boys | +0.148 |
Lee | +0.142 |
chefs | +0.139 |
principal | +0.136 |
Neg logit contributions | |
day | -0.146 |
night | -0.143 |
nights | -0.139 |
time | -0.137 |
morning | -0.133 |
Token | ," |
Token ID | 553 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
day | +0.079 |
night | +0.078 |
nights | +0.077 |
morning | +0.073 |
time | +0.073 |
Neg logit contributions | |
paramed | -0.089 |
boys | -0.088 |
chefs | -0.084 |
Lee | -0.083 |
principal | -0.080 |
Token | in |
Token ID | 287 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
day | +0.161 |
nights | +0.158 |
night | +0.158 |
morning | +0.152 |
afternoon | +0.150 |
Neg logit contributions | |
boys | -0.138 |
chefs | -0.134 |
paramed | -0.134 |
Lee | -0.128 |
groom | -0.122 |
Token | trouble |
Token ID | 5876 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
day | +0.050 |
nights | +0.050 |
night | +0.049 |
fy | +0.048 |
afternoon | +0.048 |
Neg logit contributions | |
boys | -0.036 |
paramed | -0.035 |
chefs | -0.034 |
Lee | -0.034 |
printer | -0.031 |
Token | again |
Token ID | 757 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
boys | +0.237 |
chefs | +0.226 |
paramed | +0.226 |
Lee | +0.224 |
principal | +0.210 |
Neg logit contributions | |
day | -0.177 |
nights | -0.174 |
night | -0.174 |
Along | -0.167 |
fy | -0.166 |
Token | . |
Token ID | 13 |
Feature activation | +6.64e-03 |
Pos logit contributions | |
boys | +0.188 |
chefs | +0.181 |
Lee | +0.180 |
paramed | +0.176 |
groom | +0.167 |
Neg logit contributions | |
day | -0.140 |
nights | -0.140 |
night | -0.137 |
fy | -0.136 |
Along | -0.132 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
paramed | +0.010 |
Lee | +0.010 |
boys | +0.010 |
chefs | +0.009 |
principal | +0.009 |
Neg logit contributions | |
nights | -0.011 |
day | -0.011 |
night | -0.010 |
fy | -0.010 |
afternoon | -0.010 |
Token | Once |
Token ID | 7454 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
nights | +0.237 |
fy | +0.234 |
day | +0.222 |
night | +0.218 |
afternoon | +0.214 |
Neg logit contributions | |
Lee | -0.206 |
boys | -0.203 |
chefs | -0.198 |
paramed | -0.194 |
reporter | -0.181 |
Token | upon |
Token ID | 2402 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
nights | +0.230 |
fy | +0.222 |
Along | +0.187 |
des | +0.183 |
abbling | +0.179 |
Neg logit contributions | |
Lee | -0.432 |
dad | -0.406 |
dads | -0.398 |
paramed | -0.394 |
mixer | -0.389 |
Token | a |
Token ID | 257 |
Feature activation | -6.00e-03 |
Pos logit contributions | |
nights | +0.625 |
fy | +0.617 |
spine | +0.592 |
day | +0.583 |
way | +0.577 |
Neg logit contributions | |
chefs | -0.158 |
boys | -0.145 |
paramed | -0.122 |
Lee | -0.119 |
groom | -0.102 |
Token | time |
Token ID | 640 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
nights | +0.005 |
lessly | +0.004 |
iest | +0.004 |
ingle | +0.004 |
dark | +0.004 |
Neg logit contributions | |
boys | -0.013 |
paramed | -0.012 |
chefs | -0.012 |
qu | -0.012 |
Lee | -0.012 |
Token | there |
Token ID | 612 |
Feature activation | -6.85e-02 |
Pos logit contributions | |
nights | +0.437 |
fy | +0.423 |
night | +0.417 |
day | +0.408 |
afternoon | +0.402 |
Neg logit contributions | |
Lee | -0.302 |
boys | -0.295 |
paramed | -0.282 |
chefs | -0.280 |
printer | -0.249 |
Token | were |
Token ID | 547 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
nights | +0.116 |
day | +0.115 |
night | +0.114 |
fy | +0.111 |
morning | +0.109 |
Neg logit contributions | |
boys | -0.097 |
paramed | -0.095 |
Lee | -0.092 |
chefs | -0.092 |
printer | -0.085 |
Token | together |
Token ID | 1978 |
Feature activation | +2.84e-03 |
Pos logit contributions | |
day | +0.120 |
time | +0.111 |
night | +0.107 |
afternoon | +0.105 |
morning | +0.104 |
Neg logit contributions | |
paramed | -0.186 |
Lee | -0.169 |
principal | -0.167 |
wasteful | -0.163 |
groom | -0.162 |
Token | . |
Token ID | 13 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
boys | +0.005 |
chefs | +0.005 |
paramed | +0.005 |
Lee | +0.005 |
printer | +0.005 |
Neg logit contributions | |
nights | -0.003 |
day | -0.003 |
night | -0.003 |
fy | -0.003 |
morning | -0.003 |
Token | |
Token ID | 220 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
nights | +0.058 |
fy | +0.057 |
night | +0.054 |
day | +0.054 |
ight | +0.052 |
Neg logit contributions | |
Lee | -0.049 |
boys | -0.047 |
paramed | -0.046 |
chefs | -0.044 |
printer | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
nights | +0.064 |
night | +0.062 |
day | +0.061 |
fy | +0.061 |
afternoon | +0.059 |
Neg logit contributions | |
boys | -0.054 |
Lee | -0.053 |
paramed | -0.053 |
chefs | -0.052 |
principal | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
nights | +0.027 |
fy | +0.027 |
night | +0.026 |
day | +0.025 |
spine | +0.025 |
Neg logit contributions | |
boys | -0.020 |
Lee | -0.019 |
paramed | -0.019 |
chefs | -0.018 |
principal | -0.018 |
Token | One |
Token ID | 3198 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
nights | +0.017 |
night | +0.017 |
day | +0.016 |
fy | +0.016 |
morning | +0.015 |
Neg logit contributions | |
boys | -0.026 |
paramed | -0.026 |
chefs | -0.024 |
Lee | -0.024 |
groom | -0.024 |
Token | day |
Token ID | 1110 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
ently | +0.160 |
ight | +0.148 |
spine | +0.145 |
ides | +0.144 |
lessly | +0.141 |
Neg logit contributions | |
paramed | -0.323 |
qu | -0.319 |
dads | -0.317 |
printer | -0.316 |
duck | -0.314 |
Token | , |
Token ID | 11 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
day | +0.345 |
night | +0.338 |
time | +0.322 |
morning | +0.320 |
nights | +0.320 |
Neg logit contributions | |
boys | -0.451 |
paramed | -0.443 |
chefs | -0.423 |
principal | -0.419 |
Lee | -0.410 |
Token | she |
Token ID | 673 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
boys | +0.121 |
paramed | +0.119 |
principal | +0.112 |
chefs | +0.109 |
reporter | +0.109 |
Neg logit contributions | |
day | -0.126 |
night | -0.121 |
morning | -0.117 |
time | -0.116 |
nights | -0.112 |
Token | and |
Token ID | 290 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
day | +0.241 |
night | +0.237 |
nights | +0.228 |
time | +0.227 |
morning | +0.222 |
Neg logit contributions | |
paramed | -0.190 |
boys | -0.183 |
Lee | -0.178 |
chefs | -0.176 |
reporter | -0.163 |
Token | her |
Token ID | 607 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
day | +0.044 |
night | +0.041 |
morning | +0.038 |
time | +0.037 |
afternoon | +0.036 |
Neg logit contributions | |
chefs | -0.074 |
paramed | -0.073 |
boys | -0.072 |
reporter | -0.068 |
liar | -0.068 |
Token | strongest |
Token ID | 12841 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
paramed | +0.038 |
boys | +0.036 |
Lee | +0.035 |
chefs | +0.033 |
wrists | +0.032 |
Neg logit contributions | |
day | -0.036 |
night | -0.034 |
time | -0.032 |
morning | -0.032 |
afternoon | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
day | +0.017 |
night | +0.016 |
time | +0.016 |
nights | +0.016 |
morning | +0.015 |
Neg logit contributions | |
paramed | -0.018 |
boys | -0.018 |
Lee | -0.017 |
chefs | -0.016 |
wrists | -0.016 |
Token | So |
Token ID | 1406 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
nights | +0.049 |
night | +0.048 |
day | +0.047 |
fy | +0.046 |
afternoon | +0.043 |
Neg logit contributions | |
Lee | -0.072 |
boys | -0.071 |
chefs | -0.071 |
paramed | -0.070 |
principal | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
fy | +0.066 |
nights | +0.063 |
dark | +0.060 |
ingle | +0.060 |
abbling | +0.060 |
Neg logit contributions | |
Lee | -0.056 |
chefs | -0.054 |
dads | -0.052 |
boys | -0.051 |
reporter | -0.048 |
Token | every |
Token ID | 790 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
chefs | +0.112 |
Lee | +0.110 |
dads | +0.105 |
boys | +0.103 |
reporter | +0.100 |
Neg logit contributions | |
fy | -0.111 |
nights | -0.108 |
night | -0.102 |
abbling | -0.099 |
day | -0.099 |
Token | day |
Token ID | 1110 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
nights | +0.057 |
fy | +0.053 |
adows | +0.044 |
ight | +0.040 |
spine | +0.040 |
Neg logit contributions | |
printer | -0.351 |
Lee | -0.349 |
paramed | -0.345 |
groom | -0.344 |
reporter | -0.340 |
Token | he |
Token ID | 339 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
fy | +0.422 |
nights | +0.420 |
night | +0.410 |
day | +0.407 |
Along | +0.397 |
Neg logit contributions | |
chefs | -0.314 |
Lee | -0.307 |
boys | -0.294 |
paramed | -0.281 |
dads | -0.276 |
Token | would |
Token ID | 561 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
day | +0.187 |
night | +0.183 |
nights | +0.177 |
time | +0.173 |
morning | +0.169 |
Neg logit contributions | |
paramed | -0.214 |
Lee | -0.203 |
boys | -0.201 |
chefs | -0.197 |
principal | -0.190 |
Token | wake |
Token ID | 7765 |
Feature activation | -2.73e-03 |
Pos logit contributions | |
day | +0.033 |
night | +0.032 |
nights | +0.031 |
time | +0.031 |
morning | +0.030 |
Neg logit contributions | |
paramed | -0.031 |
boys | -0.031 |
chefs | -0.030 |
Lee | -0.029 |
principal | -0.028 |
Token | up |
Token ID | 510 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
day | +0.003 |
night | +0.003 |
time | +0.003 |
morning | +0.002 |
nights | +0.002 |
Neg logit contributions | |
paramed | -0.006 |
boys | -0.006 |
chefs | -0.006 |
Lee | -0.006 |
liar | -0.006 |
Token | early |
Token ID | 1903 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
day | +0.051 |
night | +0.050 |
nights | +0.050 |
morning | +0.047 |
time | +0.047 |
Neg logit contributions | |
boys | -0.060 |
paramed | -0.059 |
chefs | -0.058 |
Lee | -0.057 |
reporter | -0.054 |
Token | her |
Token ID | 607 |
Feature activation | -9.54e-02 |
Pos logit contributions | |
boys | +0.103 |
paramed | +0.099 |
chefs | +0.095 |
Lee | +0.094 |
groom | +0.089 |
Neg logit contributions | |
day | -0.106 |
nights | -0.105 |
night | -0.104 |
fy | -0.100 |
morning | -0.099 |
Token | way |
Token ID | 835 |
Feature activation | -6.71e-02 |
Pos logit contributions | |
nights | +0.140 |
fy | +0.135 |
day | +0.134 |
night | +0.133 |
Along | +0.131 |
Neg logit contributions | |
boys | -0.161 |
Lee | -0.150 |
chefs | -0.150 |
paramed | -0.146 |
principal | -0.140 |
Token | . |
Token ID | 13 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
nights | +0.110 |
fy | +0.102 |
Along | +0.101 |
afternoon | +0.100 |
day | +0.100 |
Neg logit contributions | |
boys | -0.099 |
Lee | -0.087 |
paramed | -0.087 |
printer | -0.087 |
chefs | -0.084 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
boys | +0.019 |
paramed | +0.019 |
chefs | +0.018 |
Lee | +0.018 |
printer | +0.017 |
Neg logit contributions | |
day | -0.015 |
night | -0.015 |
nights | -0.015 |
morning | -0.014 |
afternoon | -0.014 |
Token | Once |
Token ID | 7454 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
nights | +0.219 |
day | +0.212 |
fy | +0.211 |
night | +0.209 |
afternoon | +0.201 |
Neg logit contributions | |
boys | -0.214 |
Lee | -0.208 |
paramed | -0.205 |
chefs | -0.205 |
reporter | -0.191 |
Token | upon |
Token ID | 2402 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
nights | +0.210 |
fy | +0.205 |
des | +0.170 |
Along | +0.165 |
afternoon | +0.162 |
Neg logit contributions | |
Lee | -0.427 |
dad | -0.399 |
dads | -0.396 |
boys | -0.395 |
paramed | -0.392 |
Token | a |
Token ID | 257 |
Feature activation | +4.69e-03 |
Pos logit contributions | |
nights | +0.581 |
fy | +0.572 |
day | +0.548 |
night | +0.542 |
spine | +0.542 |
Neg logit contributions | |
chefs | -0.172 |
boys | -0.169 |
paramed | -0.142 |
Lee | -0.139 |
groom | -0.120 |
Token | time |
Token ID | 640 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
boys | +0.010 |
paramed | +0.010 |
chefs | +0.010 |
qu | +0.010 |
printer | +0.010 |
Neg logit contributions | |
nights | -0.004 |
lessly | -0.003 |
iest | -0.003 |
night | -0.003 |
ingle | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
nights | +0.437 |
fy | +0.422 |
night | +0.413 |
day | +0.401 |
afternoon | +0.400 |
Neg logit contributions | |
Lee | -0.282 |
boys | -0.275 |
paramed | -0.259 |
chefs | -0.257 |
printer | -0.229 |
Token | there |
Token ID | 612 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
chefs | +0.053 |
boys | +0.051 |
Lee | +0.049 |
paramed | +0.047 |
fighting | +0.045 |
Neg logit contributions | |
nights | -0.128 |
fy | -0.125 |
ingle | -0.116 |
found | -0.114 |
night | -0.114 |
Token | was |
Token ID | 373 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
day | +0.101 |
nights | +0.100 |
night | +0.099 |
morning | +0.094 |
fy | +0.094 |
Neg logit contributions | |
boys | -0.091 |
paramed | -0.089 |
chefs | -0.087 |
Lee | -0.086 |
printer | -0.080 |
Token | played |
Token ID | 2826 |
Feature activation | -9.08e-02 |
Pos logit contributions | |
paramed | +0.054 |
chefs | +0.052 |
Lee | +0.051 |
boys | +0.050 |
principal | +0.050 |
Neg logit contributions | |
day | -0.052 |
night | -0.051 |
nights | -0.048 |
time | -0.047 |
morning | -0.047 |
Token | together |
Token ID | 1978 |
Feature activation | -4.46e-03 |
Pos logit contributions | |
day | +0.128 |
time | +0.117 |
afternoon | +0.114 |
night | +0.113 |
morning | +0.110 |
Neg logit contributions | |
paramed | -0.188 |
groom | -0.173 |
Lee | -0.169 |
principal | -0.166 |
reporter | -0.164 |
Token | until |
Token ID | 1566 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
nights | +0.006 |
day | +0.006 |
night | +0.006 |
fy | +0.006 |
morning | +0.006 |
Neg logit contributions | |
boys | -0.007 |
chefs | -0.007 |
paramed | -0.007 |
Lee | -0.007 |
printer | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
day | +0.268 |
night | +0.267 |
nights | +0.256 |
morning | +0.254 |
time | +0.251 |
Neg logit contributions | |
boys | -0.166 |
paramed | -0.164 |
chefs | -0.156 |
Lee | -0.150 |
principal | -0.143 |
Token | was |
Token ID | 373 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
boys | +0.027 |
paramed | +0.026 |
chefs | +0.025 |
Lee | +0.024 |
groom | +0.023 |
Neg logit contributions | |
nights | -0.036 |
day | -0.036 |
night | -0.035 |
fy | -0.035 |
Along | -0.034 |
Token | time |
Token ID | 640 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
chefs | +0.051 |
boys | +0.051 |
paramed | +0.048 |
Lee | +0.048 |
groom | +0.047 |
Neg logit contributions | |
found | -0.012 |
nights | -0.012 |
fy | -0.011 |
ingle | -0.010 |
ly | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +6.13e-02 |
Pos logit contributions | |
nights | +0.407 |
fy | +0.378 |
day | +0.377 |
night | +0.374 |
afternoon | +0.371 |
Neg logit contributions | |
Lee | -0.254 |
boys | -0.252 |
chefs | -0.247 |
paramed | -0.237 |
printer | -0.235 |
Token | go |
Token ID | 467 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
boys | +0.103 |
paramed | +0.101 |
chefs | +0.098 |
Lee | +0.096 |
groom | +0.093 |
Neg logit contributions | |
nights | -0.091 |
day | -0.090 |
night | -0.088 |
fy | -0.084 |
afternoon | -0.084 |
Token | home |
Token ID | 1363 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
nights | +0.066 |
ight | +0.062 |
fy | +0.061 |
day | +0.059 |
found | +0.059 |
Neg logit contributions | |
boys | -0.071 |
chefs | -0.067 |
Lee | -0.067 |
paramed | -0.066 |
dads | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
day | +0.012 |
night | +0.012 |
nights | +0.012 |
morning | +0.011 |
time | +0.011 |
Neg logit contributions | |
boys | -0.023 |
paramed | -0.023 |
chefs | -0.022 |
Lee | -0.022 |
printer | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
nights | +0.053 |
day | +0.050 |
adows | +0.050 |
night | +0.050 |
iest | +0.050 |
Neg logit contributions | |
paramed | -0.022 |
Lee | -0.020 |
boys | -0.020 |
chefs | -0.019 |
printer | -0.018 |