Token | on |
Token ID | 319 |
Feature activation | -9.35e-02 |
Pos logit contributions | |
joy | +0.031 |
ez | +0.030 |
hered | +0.027 |
nightmares | +0.026 |
moment | +0.025 |
Neg logit contributions | |
ires | -0.030 |
ate | -0.028 |
iring | -0.028 |
um | -0.028 |
ruce | -0.026 |
Token | her |
Token ID | 607 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
joy | +0.202 |
ez | +0.195 |
hered | +0.173 |
nightmares | +0.166 |
moment | +0.160 |
Neg logit contributions | |
ires | -0.141 |
um | -0.138 |
ate | -0.136 |
iring | -0.130 |
ruce | -0.115 |
Token | finger |
Token ID | 7660 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
joy | +0.081 |
ez | +0.078 |
hered | +0.070 |
nightmares | +0.066 |
moment | +0.064 |
Neg logit contributions | |
ires | -0.058 |
um | -0.055 |
ate | -0.055 |
iring | -0.053 |
ruce | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
joy | +0.121 |
ez | +0.115 |
hered | +0.104 |
nightmares | +0.098 |
moment | +0.096 |
Neg logit contributions | |
ires | -0.108 |
ate | -0.102 |
iring | -0.102 |
um | -0.101 |
ruce | -0.091 |
Token | adm |
Token ID | 6178 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
joy | +0.077 |
ez | +0.073 |
hered | +0.067 |
nightmares | +0.062 |
moment | +0.062 |
Neg logit contributions | |
ires | -0.067 |
ate | -0.064 |
um | -0.063 |
iring | -0.062 |
ruce | -0.056 |
Token | ires |
Token ID | 2387 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
joy | +0.131 |
ez | +0.123 |
nightmares | +0.116 |
ymes | +0.113 |
ching | +0.111 |
Neg logit contributions | |
First | -0.051 |
stack | -0.043 |
bury | -0.041 |
Inside | -0.038 |
lly | -0.036 |
Token | it |
Token ID | 340 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
ires | +0.645 |
ate | +0.577 |
iring | +0.557 |
um | +0.553 |
Inside | +0.486 |
Neg logit contributions | |
joy | -1.138 |
ez | -1.106 |
hered | -1.084 |
oing | -0.955 |
moment | -0.940 |
Token | . |
Token ID | 13 |
Feature activation | -7.76e-03 |
Pos logit contributions | |
joy | +0.035 |
ez | +0.034 |
hered | +0.031 |
nightmares | +0.029 |
moment | +0.028 |
Neg logit contributions | |
ires | -0.028 |
ate | -0.026 |
iring | -0.026 |
um | -0.026 |
ruce | -0.023 |
Token | She |
Token ID | 1375 |
Feature activation | +5.36e-02 |
Pos logit contributions | |
joy | +0.015 |
ez | +0.014 |
hered | +0.013 |
nightmares | +0.012 |
moment | +0.012 |
Neg logit contributions | |
ires | -0.013 |
iring | -0.013 |
ate | -0.012 |
um | -0.012 |
ruce | -0.011 |
Token | feels |
Token ID | 5300 |
Feature activation | -8.79e-02 |
Pos logit contributions | |
iring | +0.110 |
ires | +0.107 |
um | +0.105 |
ate | +0.099 |
ruce | +0.097 |
Neg logit contributions | |
ez | -0.082 |
joy | -0.081 |
nightmares | -0.070 |
hered | -0.065 |
heaven | -0.065 |
Token | like |
Token ID | 588 |
Feature activation | +6.83e-02 |
Pos logit contributions | |
joy | +0.178 |
ez | +0.170 |
hered | +0.159 |
moment | +0.145 |
nightmares | +0.145 |
Neg logit contributions | |
ires | -0.144 |
ate | -0.136 |
iring | -0.134 |
um | -0.132 |
ruce | -0.118 |
Token | Tom |
Token ID | 4186 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
iring | +0.180 |
ate | +0.176 |
ires | +0.173 |
um | +0.161 |
ruce | +0.161 |
Neg logit contributions | |
joy | -0.121 |
ez | -0.110 |
nightmares | -0.103 |
moment | -0.098 |
pain | -0.098 |
Token | says |
Token ID | 1139 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
ires | +0.087 |
iring | +0.082 |
um | +0.080 |
ate | +0.080 |
ruce | +0.072 |
Neg logit contributions | |
joy | -0.116 |
ez | -0.113 |
hered | -0.102 |
nightmares | -0.098 |
moment | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | -6.87e-03 |
Pos logit contributions | |
ires | +0.212 |
um | +0.204 |
ate | +0.196 |
iring | +0.195 |
ruce | +0.166 |
Neg logit contributions | |
joy | -0.333 |
ez | -0.329 |
hered | -0.293 |
Charl | -0.285 |
nightmares | -0.280 |
Token | He |
Token ID | 679 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
joy | +0.014 |
ez | +0.013 |
hered | +0.012 |
nightmares | +0.011 |
moment | +0.011 |
Neg logit contributions | |
ires | -0.011 |
ate | -0.011 |
iring | -0.011 |
um | -0.011 |
ruce | -0.010 |
Token | adm |
Token ID | 6178 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
ires | +0.121 |
iring | +0.118 |
um | +0.115 |
ate | +0.113 |
ruce | +0.105 |
Neg logit contributions | |
joy | -0.111 |
ez | -0.105 |
hered | -0.094 |
nightmares | -0.088 |
moment | -0.085 |
Token | ires |
Token ID | 2387 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
joy | +0.152 |
nightmares | +0.135 |
ymes | +0.134 |
ez | +0.133 |
asleep | +0.130 |
Neg logit contributions | |
First | -0.057 |
lly | -0.043 |
replied | -0.040 |
lined | -0.039 |
Inside | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
ires | +0.710 |
ate | +0.647 |
iring | +0.630 |
um | +0.624 |
ruce | +0.538 |
Neg logit contributions | |
ez | -1.074 |
joy | -1.071 |
hered | -1.014 |
moment | -0.899 |
Charl | -0.889 |
Token | puzzle |
Token ID | 15027 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
ez | +0.071 |
joy | +0.069 |
hered | +0.061 |
Charl | +0.055 |
smart | +0.054 |
Neg logit contributions | |
ires | -0.068 |
ate | -0.064 |
um | -0.063 |
iring | -0.062 |
ruce | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -7.04e-03 |
Pos logit contributions | |
joy | +0.362 |
ez | +0.347 |
hered | +0.315 |
nightmares | +0.294 |
Charl | +0.284 |
Neg logit contributions | |
ires | -0.270 |
ate | -0.256 |
um | -0.250 |
iring | -0.249 |
ruce | -0.222 |
Token | He |
Token ID | 679 |
Feature activation | +6.31e-02 |
Pos logit contributions | |
joy | +0.013 |
ez | +0.012 |
hered | +0.011 |
moment | +0.010 |
nightmares | +0.010 |
Neg logit contributions | |
ires | -0.013 |
iring | -0.012 |
um | -0.012 |
ate | -0.012 |
ruce | -0.011 |
Token | thinks |
Token ID | 6834 |
Feature activation | -8.07e-03 |
Pos logit contributions | |
ires | +0.122 |
iring | +0.121 |
um | +0.118 |
ate | +0.114 |
ruce | +0.108 |
Neg logit contributions | |
joy | -0.106 |
ez | -0.099 |
hered | -0.087 |
nightmares | -0.084 |
moment | -0.080 |
Token | Mia |
Token ID | 32189 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
ires | +0.163 |
iring | +0.161 |
ate | +0.155 |
um | +0.150 |
ruce | +0.143 |
Neg logit contributions | |
joy | -0.150 |
ez | -0.135 |
moment | -0.125 |
nightmares | -0.120 |
hered | -0.119 |
Token | says |
Token ID | 1139 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
ires | +0.050 |
iring | +0.047 |
ate | +0.046 |
um | +0.045 |
ruce | +0.040 |
Neg logit contributions | |
joy | -0.090 |
ez | -0.086 |
hered | -0.080 |
nightmares | -0.076 |
moment | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | -6.97e-03 |
Pos logit contributions | |
ires | +0.185 |
um | +0.179 |
iring | +0.166 |
ate | +0.162 |
outed | +0.139 |
Neg logit contributions | |
ez | -0.356 |
joy | -0.348 |
Charl | -0.316 |
nightmares | -0.308 |
hered | -0.308 |
Token | She |
Token ID | 1375 |
Feature activation | +5.14e-02 |
Pos logit contributions | |
joy | +0.015 |
ez | +0.015 |
hered | +0.013 |
nightmares | +0.013 |
moment | +0.012 |
Neg logit contributions | |
ires | -0.011 |
um | -0.010 |
iring | -0.010 |
ate | -0.010 |
ruce | -0.009 |
Token | adm |
Token ID | 6178 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
ires | +0.101 |
iring | +0.098 |
um | +0.096 |
ate | +0.094 |
ruce | +0.088 |
Neg logit contributions | |
joy | -0.087 |
ez | -0.081 |
hered | -0.072 |
nightmares | -0.068 |
moment | -0.066 |
Token | ires |
Token ID | 2387 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
joy | +0.154 |
ymes | +0.136 |
nightmares | +0.136 |
ez | +0.135 |
asleep | +0.130 |
Neg logit contributions | |
First | -0.058 |
lly | -0.046 |
bury | -0.041 |
lined | -0.041 |
replied | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
ires | +0.727 |
ate | +0.662 |
iring | +0.650 |
um | +0.648 |
ruce | +0.565 |
Neg logit contributions | |
joy | -1.053 |
ez | -1.046 |
hered | -0.982 |
moment | -0.876 |
nightmares | -0.861 |
Token | puzzle |
Token ID | 15027 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
ez | +0.072 |
joy | +0.070 |
hered | +0.061 |
Charl | +0.056 |
nightmares | +0.056 |
Neg logit contributions | |
ires | -0.069 |
ate | -0.065 |
um | -0.064 |
iring | -0.063 |
ruce | -0.058 |
Token | too |
Token ID | 1165 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
joy | +0.366 |
ez | +0.351 |
hered | +0.319 |
nightmares | +0.298 |
Charl | +0.288 |
Neg logit contributions | |
ires | -0.269 |
ate | -0.256 |
um | -0.250 |
iring | -0.248 |
ruce | -0.221 |
Token | . |
Token ID | 13 |
Feature activation | -6.91e-03 |
Pos logit contributions | |
joy | +0.103 |
ez | +0.099 |
hered | +0.089 |
nightmares | +0.084 |
moment | +0.082 |
Neg logit contributions | |
ires | -0.089 |
ate | -0.084 |
iring | -0.084 |
um | -0.083 |
ruce | -0.074 |
Token | She |
Token ID | 1375 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
joy | +0.013 |
ez | +0.012 |
hered | +0.011 |
nightmares | +0.010 |
moment | +0.010 |
Neg logit contributions | |
ires | -0.012 |
iring | -0.011 |
ate | -0.011 |
um | -0.011 |
ruce | -0.010 |
Token | careful |
Token ID | 8161 |
Feature activation | -6.09e-03 |
Pos logit contributions | |
joy | +0.004 |
ez | +0.004 |
hered | +0.004 |
nightmares | +0.003 |
moment | +0.003 |
Neg logit contributions | |
ires | -0.003 |
ate | -0.003 |
iring | -0.003 |
um | -0.003 |
ruce | -0.002 |
Token | next |
Token ID | 1306 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
joy | +0.013 |
ez | +0.012 |
hered | +0.011 |
nightmares | +0.010 |
moment | +0.010 |
Neg logit contributions | |
ires | -0.010 |
iring | -0.009 |
ate | -0.009 |
um | -0.009 |
ruce | -0.008 |
Token | time |
Token ID | 640 |
Feature activation | -7.40e-02 |
Pos logit contributions | |
um | +0.128 |
ires | +0.127 |
ate | +0.124 |
iring | +0.117 |
ruce | +0.106 |
Neg logit contributions | |
joy | -0.187 |
ez | -0.180 |
hered | -0.169 |
Charl | -0.153 |
heaven | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
joy | +0.158 |
ez | +0.153 |
hered | +0.136 |
nightmares | +0.130 |
moment | +0.127 |
Neg logit contributions | |
ires | -0.112 |
ate | -0.107 |
um | -0.106 |
iring | -0.104 |
ruce | -0.090 |
Token | W |
Token ID | 370 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
joy | +0.039 |
ez | +0.037 |
hered | +0.035 |
nightmares | +0.033 |
moment | +0.032 |
Neg logit contributions | |
ires | -0.028 |
ate | -0.027 |
um | -0.027 |
iring | -0.026 |
Inside | -0.024 |
Token | ires |
Token ID | 2387 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
ez | +0.195 |
joy | +0.193 |
hered | +0.165 |
nightmares | +0.164 |
failure | +0.160 |
Neg logit contributions | |
iring | -0.117 |
ires | -0.115 |
ate | -0.112 |
um | -0.109 |
ruce | -0.100 |
Token | can |
Token ID | 460 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
ires | +0.742 |
ate | +0.716 |
um | +0.691 |
iring | +0.672 |
ruce | +0.618 |
Neg logit contributions | |
joy | -1.035 |
ez | -0.896 |
hered | -0.856 |
moment | -0.792 |
knowing | -0.778 |
Token | be |
Token ID | 307 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
ez | +0.036 |
joy | +0.034 |
hered | +0.033 |
nightmares | +0.031 |
moment | +0.030 |
Neg logit contributions | |
ires | -0.035 |
iring | -0.032 |
um | -0.031 |
ate | -0.031 |
outed | -0.028 |
Token | dangerous |
Token ID | 4923 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
ate | +0.069 |
ires | +0.068 |
iring | +0.064 |
um | +0.064 |
her | +0.058 |
Neg logit contributions | |
joy | -0.107 |
ez | -0.098 |
hered | -0.097 |
moment | -0.088 |
nightmares | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
ires | +0.046 |
iring | +0.044 |
ate | +0.043 |
um | +0.042 |
ruce | +0.036 |
Neg logit contributions | |
joy | -0.069 |
ez | -0.067 |
hered | -0.062 |
moment | -0.056 |
Charl | -0.056 |
Token | I |
Token ID | 314 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
joy | +0.035 |
ez | +0.033 |
hered | +0.030 |
moment | +0.028 |
nightmares | +0.028 |
Neg logit contributions | |
ires | -0.029 |
ate | -0.027 |
um | -0.027 |
iring | -0.026 |
Inside | -0.024 |
Token | told |
Token ID | 1297 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
ires | +0.002 |
ate | +0.002 |
um | +0.002 |
iring | +0.002 |
ruce | +0.002 |
Neg logit contributions | |
joy | -0.004 |
ez | -0.004 |
hered | -0.003 |
nightmares | -0.003 |
heaven | -0.003 |
Token | Fin |
Token ID | 4463 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
ires | +0.089 |
ate | +0.085 |
iring | +0.085 |
um | +0.084 |
ruce | +0.076 |
Neg logit contributions | |
joy | -0.094 |
ez | -0.090 |
hered | -0.082 |
nightmares | -0.076 |
moment | -0.075 |
Token | that |
Token ID | 326 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
ires | +0.036 |
ate | +0.036 |
um | +0.033 |
iring | +0.033 |
ered | +0.028 |
Neg logit contributions | |
joy | -0.077 |
ez | -0.073 |
hered | -0.068 |
Charl | -0.064 |
nightmares | -0.062 |
Token | he |
Token ID | 339 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
ires | +0.180 |
iring | +0.172 |
ate | +0.172 |
ruce | +0.144 |
Inside | +0.139 |
Neg logit contributions | |
joy | -0.384 |
ez | -0.378 |
hered | -0.366 |
Charl | -0.327 |
tale | -0.321 |
Token | adm |
Token ID | 6178 |
Feature activation | -6.57e-02 |
Pos logit contributions | |
joy | +0.023 |
ez | +0.021 |
hered | +0.020 |
nightmares | +0.018 |
moment | +0.018 |
Neg logit contributions | |
ires | -0.016 |
ate | -0.015 |
iring | -0.015 |
um | -0.015 |
ruce | -0.013 |
Token | ires |
Token ID | 2387 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
joy | +0.162 |
ez | +0.139 |
nightmares | +0.133 |
smart | +0.131 |
ching | +0.127 |
Neg logit contributions | |
First | -0.067 |
stack | -0.054 |
Later | -0.052 |
brushes | -0.047 |
elevator | -0.044 |
Token | how |
Token ID | 703 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
ires | +0.647 |
iring | +0.581 |
ate | +0.564 |
um | +0.552 |
Inside | +0.494 |
Neg logit contributions | |
joy | -1.102 |
ez | -1.063 |
hered | -0.992 |
moment | -0.926 |
ñ | -0.918 |
Token | brave |
Token ID | 14802 |
Feature activation | -7.64e-02 |
Pos logit contributions | |
ires | +0.153 |
iring | +0.144 |
ate | +0.144 |
um | +0.142 |
ruce | +0.127 |
Neg logit contributions | |
joy | -0.190 |
ez | -0.183 |
hered | -0.167 |
nightmares | -0.156 |
moment | -0.154 |
Token | he |
Token ID | 339 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
joy | +0.154 |
ez | +0.148 |
hered | +0.134 |
nightmares | +0.127 |
moment | +0.125 |
Neg logit contributions | |
ires | -0.125 |
iring | -0.118 |
ate | -0.117 |
um | -0.116 |
ruce | -0.104 |
Token | is |
Token ID | 318 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
joy | +0.026 |
ez | +0.025 |
hered | +0.023 |
nightmares | +0.021 |
moment | +0.021 |
Neg logit contributions | |
ires | -0.018 |
ate | -0.017 |
iring | -0.017 |
um | -0.016 |
ruce | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
joy | +0.048 |
ez | +0.046 |
hered | +0.043 |
Charl | +0.041 |
ñ | +0.039 |
Neg logit contributions | |
ires | -0.024 |
ate | -0.023 |
iring | -0.021 |
um | -0.020 |
Inside | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
joy | +0.009 |
ez | +0.009 |
hered | +0.008 |
moment | +0.007 |
nightmares | +0.007 |
Neg logit contributions | |
ires | -0.014 |
iring | -0.013 |
ate | -0.013 |
um | -0.013 |
ruce | -0.012 |
Token | green |
Token ID | 4077 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
ires | +0.167 |
um | +0.158 |
iring | +0.157 |
ate | +0.156 |
ruce | +0.137 |
Neg logit contributions | |
joy | -0.232 |
ez | -0.231 |
hered | -0.210 |
nightmares | -0.193 |
Charl | -0.184 |
Token | apple |
Token ID | 17180 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
ires | +0.290 |
ate | +0.276 |
um | +0.275 |
iring | +0.270 |
ruce | +0.232 |
Neg logit contributions | |
joy | -0.410 |
ez | -0.394 |
hered | -0.361 |
nightmares | -0.332 |
Charl | -0.326 |
Token | . |
Token ID | 13 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
ires | +0.106 |
um | +0.102 |
ate | +0.101 |
iring | +0.098 |
ruce | +0.086 |
Neg logit contributions | |
joy | -0.166 |
ez | -0.155 |
hered | -0.140 |
nightmares | -0.134 |
moment | -0.130 |
Token | Y |
Token ID | 575 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
ires | +0.019 |
ate | +0.018 |
um | +0.018 |
iring | +0.018 |
ruce | +0.016 |
Neg logit contributions | |
joy | -0.026 |
ez | -0.025 |
hered | -0.023 |
nightmares | -0.022 |
moment | -0.021 |
Token | um |
Token ID | 388 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
joy | +0.102 |
ez | +0.100 |
hered | +0.099 |
ñ | +0.089 |
tale | +0.089 |
Neg logit contributions | |
ires | -0.026 |
iring | -0.021 |
ate | -0.021 |
ruce | -0.019 |
outed | -0.016 |
Token | ! |
Token ID | 0 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
ires | +0.636 |
um | +0.635 |
ate | +0.621 |
iring | +0.600 |
lly | +0.532 |
Neg logit contributions | |
joy | -1.051 |
ez | -1.003 |
hered | -0.948 |
nightmares | -0.893 |
moment | -0.861 |
Token | |
Token ID | 220 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
joy | +0.047 |
ez | +0.045 |
hered | +0.041 |
nightmares | +0.039 |
moment | +0.038 |
Neg logit contributions | |
ires | -0.041 |
ate | -0.039 |
um | -0.039 |
iring | -0.039 |
ruce | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +2.45e-03 |
Pos logit contributions | |
joy | +0.117 |
ez | +0.111 |
hered | +0.099 |
moment | +0.092 |
nightmares | +0.091 |
Neg logit contributions | |
ires | -0.120 |
iring | -0.115 |
ate | -0.113 |
um | -0.112 |
ruce | -0.105 |
Token |
|
Token ID | 198 |
Feature activation | +2.96e-04 |
Pos logit contributions | |
ires | +0.005 |
iring | +0.005 |
ate | +0.004 |
um | +0.004 |
ruce | +0.004 |
Neg logit contributions | |
joy | -0.004 |
ez | -0.004 |
hered | -0.003 |
moment | -0.003 |
heaven | -0.003 |
Token | He |
Token ID | 1544 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
ires | +0.001 |
iring | +0.000 |
ate | +0.000 |
um | +0.000 |
ruce | +0.000 |
Neg logit contributions | |
joy | -0.001 |
ez | -0.001 |
hered | -0.001 |
moment | -0.000 |
nightmares | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
ires | +0.239 |
ate | +0.227 |
iring | +0.223 |
um | +0.223 |
ruce | +0.194 |
Neg logit contributions | |
joy | -0.346 |
ez | -0.332 |
hered | -0.299 |
nightmares | -0.286 |
moment | -0.273 |
Token |
|
Token ID | 198 |
Feature activation | +4.87e-02 |
Pos logit contributions | |
ires | +0.073 |
ate | +0.069 |
um | +0.068 |
iring | +0.068 |
ruce | +0.060 |
Neg logit contributions | |
joy | -0.099 |
ez | -0.095 |
hered | -0.087 |
nightmares | -0.082 |
moment | -0.080 |
Token |
|
Token ID | 198 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
ires | +0.091 |
iring | +0.087 |
ate | +0.085 |
um | +0.084 |
ruce | +0.078 |
Neg logit contributions | |
joy | -0.086 |
ez | -0.084 |
hered | -0.070 |
moment | -0.069 |
nightmares | -0.068 |
Token | " |
Token ID | 1 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
ires | +0.057 |
iring | +0.054 |
ate | +0.053 |
um | +0.053 |
ruce | +0.048 |
Neg logit contributions | |
joy | -0.072 |
ez | -0.070 |
hered | -0.063 |
moment | -0.059 |
nightmares | -0.059 |
Token | M |
Token ID | 44 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
ires | +0.075 |
ate | +0.073 |
um | +0.071 |
iring | +0.069 |
ruce | +0.063 |
Neg logit contributions | |
joy | -0.099 |
ez | -0.093 |
nightmares | -0.083 |
hered | -0.082 |
breathing | -0.079 |
Token | um |
Token ID | 388 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
joy | +0.047 |
ez | +0.044 |
hered | +0.041 |
tale | +0.040 |
nightmares | +0.040 |
Neg logit contributions | |
ate | -0.027 |
ires | -0.027 |
ruce | -0.025 |
iring | -0.025 |
her | -0.021 |
Token | ! |
Token ID | 0 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
um | +0.605 |
ires | +0.599 |
ate | +0.590 |
iring | +0.564 |
ruce | +0.481 |
Neg logit contributions | |
joy | -1.052 |
ez | -1.017 |
hered | -0.936 |
nightmares | -0.903 |
heaven | -0.864 |
Token | Look |
Token ID | 6803 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
joy | +0.032 |
ez | +0.031 |
hered | +0.028 |
nightmares | +0.027 |
moment | +0.026 |
Neg logit contributions | |
ires | -0.030 |
um | -0.029 |
ate | -0.029 |
iring | -0.028 |
Inside | -0.026 |
Token | at |
Token ID | 379 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
joy | +0.028 |
ez | +0.025 |
hered | +0.022 |
moment | +0.021 |
nightmares | +0.021 |
Neg logit contributions | |
ires | -0.029 |
iring | -0.028 |
ate | -0.027 |
ruce | -0.025 |
um | -0.024 |
Token | this |
Token ID | 428 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
ires | +0.310 |
iring | +0.292 |
ate | +0.290 |
um | +0.287 |
ruce | +0.260 |
Neg logit contributions | |
joy | -0.391 |
ez | -0.377 |
hered | -0.343 |
nightmares | -0.328 |
moment | -0.321 |
Token | tree |
Token ID | 5509 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
ires | +0.026 |
ate | +0.024 |
iring | +0.024 |
um | +0.024 |
ruce | +0.021 |
Neg logit contributions | |
joy | -0.036 |
ez | -0.035 |
hered | -0.031 |
nightmares | -0.030 |
moment | -0.029 |
Token | hat |
Token ID | 6877 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
ires | +0.147 |
ate | +0.145 |
um | +0.143 |
iring | +0.135 |
ruce | +0.118 |
Neg logit contributions | |
ez | -0.224 |
joy | -0.217 |
hered | -0.202 |
nightmares | -0.182 |
smart | -0.180 |
Token | . |
Token ID | 13 |
Feature activation | +7.36e-03 |
Pos logit contributions | |
ires | +0.086 |
ate | +0.082 |
um | +0.082 |
iring | +0.081 |
ruce | +0.071 |
Neg logit contributions | |
joy | -0.125 |
ez | -0.118 |
hered | -0.106 |
nightmares | -0.103 |
moment | -0.100 |
Token |
|
Token ID | 198 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
ires | +0.012 |
ate | +0.011 |
iring | +0.011 |
um | +0.011 |
ruce | +0.010 |
Neg logit contributions | |
joy | -0.015 |
ez | -0.015 |
hered | -0.013 |
nightmares | -0.013 |
moment | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
ires | +0.051 |
iring | +0.049 |
um | +0.047 |
ate | +0.047 |
ruce | +0.044 |
Neg logit contributions | |
joy | -0.044 |
ez | -0.042 |
moment | -0.035 |
hered | -0.035 |
heaven | -0.034 |
Token | M |
Token ID | 44 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
ires | +0.034 |
iring | +0.032 |
ate | +0.032 |
um | +0.031 |
ruce | +0.028 |
Neg logit contributions | |
joy | -0.041 |
ez | -0.040 |
hered | -0.036 |
moment | -0.033 |
nightmares | -0.033 |
Token | um |
Token ID | 388 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
joy | +0.056 |
ez | +0.052 |
nightmares | +0.047 |
hered | +0.047 |
tale | +0.046 |
Neg logit contributions | |
ate | -0.030 |
ires | -0.028 |
ruce | -0.028 |
iring | -0.026 |
Inside | -0.024 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
ires | +0.722 |
iring | +0.676 |
ate | +0.675 |
um | +0.663 |
ruce | +0.595 |
Neg logit contributions | |
joy | -0.980 |
ez | -0.945 |
hered | -0.860 |
nightmares | -0.810 |
moment | -0.800 |
Token | the |
Token ID | 262 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
um | +0.304 |
ires | +0.286 |
ate | +0.266 |
iring | +0.262 |
Inside | +0.236 |
Neg logit contributions | |
ez | -0.546 |
joy | -0.527 |
hered | -0.494 |
nightmares | -0.489 |
tale | -0.462 |
Token | priest |
Token ID | 11503 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
joy | +0.049 |
ez | +0.048 |
hered | +0.044 |
nightmares | +0.041 |
moment | +0.039 |
Neg logit contributions | |
ires | -0.037 |
ate | -0.034 |
um | -0.034 |
iring | -0.034 |
ruce | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
joy | +0.025 |
ez | +0.024 |
hered | +0.023 |
nightmares | +0.021 |
Charl | +0.021 |
Neg logit contributions | |
ires | -0.016 |
um | -0.015 |
ate | -0.015 |
iring | -0.014 |
ruce | -0.012 |
Token | " |
Token ID | 366 |
Feature activation | -6.24e-02 |
Pos logit contributions | |
ires | +0.036 |
um | +0.034 |
iring | +0.034 |
ate | +0.033 |
Inside | +0.031 |
Neg logit contributions | |
ez | -0.049 |
joy | -0.047 |
nightmares | -0.044 |
hered | -0.044 |
heaven | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +6.46e-02 |
Pos logit contributions | |
joy | +0.729 |
ez | +0.684 |
hered | +0.621 |
nightmares | +0.597 |
moment | +0.574 |
Neg logit contributions | |
ires | -0.533 |
iring | -0.510 |
um | -0.507 |
ate | -0.505 |
ruce | -0.447 |
Token |
|
Token ID | 198 |
Feature activation | +6.02e-02 |
Pos logit contributions | |
ires | +0.102 |
ate | +0.096 |
um | +0.095 |
iring | +0.095 |
ruce | +0.084 |
Neg logit contributions | |
joy | -0.133 |
ez | -0.128 |
hered | -0.117 |
nightmares | -0.110 |
moment | -0.108 |
Token |
|
Token ID | 198 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
ires | +0.107 |
iring | +0.102 |
ate | +0.101 |
um | +0.099 |
ruce | +0.091 |
Neg logit contributions | |
joy | -0.113 |
ez | -0.109 |
hered | -0.095 |
moment | -0.090 |
nightmares | -0.090 |
Token | " |
Token ID | 1 |
Feature activation | +5.24e-02 |
Pos logit contributions | |
ires | +0.065 |
iring | +0.061 |
ate | +0.061 |
um | +0.060 |
ruce | +0.054 |
Neg logit contributions | |
joy | -0.084 |
ez | -0.081 |
hered | -0.073 |
nightmares | -0.069 |
moment | -0.068 |
Token | Y |
Token ID | 56 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
ires | +0.087 |
ate | +0.084 |
um | +0.082 |
iring | +0.081 |
ruce | +0.073 |
Neg logit contributions | |
joy | -0.105 |
ez | -0.099 |
hered | -0.087 |
nightmares | -0.087 |
breathing | -0.083 |
Token | um |
Token ID | 388 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
ires | +0.023 |
ate | +0.021 |
iring | +0.021 |
ruce | +0.016 |
ised | +0.016 |
Neg logit contributions | |
hered | -0.047 |
joy | -0.047 |
ez | -0.046 |
ñ | -0.043 |
rav | -0.041 |
Token | ! |
Token ID | 0 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
um | +0.577 |
ires | +0.548 |
ate | +0.520 |
iring | +0.520 |
ruce | +0.423 |
Neg logit contributions | |
joy | -1.094 |
ez | -1.064 |
hered | -0.988 |
nightmares | -0.950 |
Charl | -0.928 |
Token | This |
Token ID | 770 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
joy | +0.042 |
ez | +0.040 |
hered | +0.036 |
nightmares | +0.034 |
moment | +0.033 |
Neg logit contributions | |
ires | -0.043 |
ate | -0.041 |
um | -0.041 |
iring | -0.041 |
ruce | -0.037 |
Token | pizza |
Token ID | 14256 |
Feature activation | -9.01e-02 |
Pos logit contributions | |
joy | +0.041 |
ez | +0.039 |
hered | +0.035 |
nightmares | +0.033 |
knowing | +0.032 |
Neg logit contributions | |
ires | -0.029 |
ate | -0.029 |
um | -0.029 |
iring | -0.027 |
ruce | -0.025 |
Token | is |
Token ID | 318 |
Feature activation | -7.86e-03 |
Pos logit contributions | |
joy | +0.162 |
ez | +0.159 |
hered | +0.143 |
moment | +0.134 |
nightmares | +0.132 |
Neg logit contributions | |
ires | -0.160 |
iring | -0.153 |
ate | -0.150 |
um | -0.149 |
ruce | -0.136 |
Token | delicious |
Token ID | 12625 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
joy | +0.015 |
ez | +0.015 |
hered | +0.013 |
nightmares | +0.013 |
moment | +0.013 |
Neg logit contributions | |
ires | -0.013 |
iring | -0.012 |
ate | -0.012 |
um | -0.012 |
ruce | -0.011 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
ires | +0.012 |
iring | +0.011 |
um | +0.011 |
ate | +0.011 |
ruce | +0.010 |
Neg logit contributions | |
joy | -0.018 |
ez | -0.018 |
hered | -0.016 |
nightmares | -0.015 |
Charl | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
ires | +0.435 |
um | +0.419 |
ate | +0.410 |
iring | +0.403 |
ruce | +0.346 |
Neg logit contributions | |
joy | -0.637 |
ez | -0.620 |
hered | -0.564 |
nightmares | -0.534 |
moment | -0.525 |
Token |
|
Token ID | 198 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
ires | +0.043 |
um | +0.042 |
ate | +0.040 |
iring | +0.040 |
Inside | +0.036 |
Neg logit contributions | |
joy | -0.056 |
ez | -0.056 |
hered | -0.050 |
nightmares | -0.049 |
moment | -0.046 |
Token |
|
Token ID | 198 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
ires | +0.054 |
ate | +0.052 |
um | +0.052 |
iring | +0.051 |
ruce | +0.046 |
Neg logit contributions | |
joy | -0.066 |
ez | -0.063 |
hered | -0.057 |
nightmares | -0.054 |
moment | -0.053 |
Token | M |
Token ID | 44 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
ires | +0.037 |
um | +0.036 |
ate | +0.036 |
iring | +0.035 |
ruce | +0.031 |
Neg logit contributions | |
joy | -0.049 |
ez | -0.047 |
hered | -0.042 |
nightmares | -0.041 |
moment | -0.039 |
Token | um |
Token ID | 388 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
joy | +0.063 |
ez | +0.060 |
hered | +0.054 |
nightmares | +0.053 |
tale | +0.052 |
Neg logit contributions | |
ate | -0.027 |
ires | -0.026 |
ruce | -0.024 |
iring | -0.023 |
Inside | -0.019 |
Token | thought |
Token ID | 1807 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
ires | +0.604 |
um | +0.569 |
ate | +0.564 |
iring | +0.562 |
ruce | +0.485 |
Neg logit contributions | |
joy | -1.068 |
ez | -1.025 |
hered | -0.946 |
nightmares | -0.907 |
moment | -0.888 |
Token | for |
Token ID | 329 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
ires | +0.079 |
um | +0.075 |
iring | +0.073 |
ate | +0.071 |
ruce | +0.065 |
Neg logit contributions | |
joy | -0.128 |
ez | -0.123 |
hered | -0.113 |
nightmares | -0.105 |
Charl | -0.103 |
Token | a |
Token ID | 257 |
Feature activation | -9.59e-03 |
Pos logit contributions | |
ires | +0.049 |
ate | +0.047 |
um | +0.045 |
iring | +0.045 |
Inside | +0.040 |
Neg logit contributions | |
ez | -0.090 |
joy | -0.090 |
hered | -0.083 |
oing | -0.074 |
nightmares | -0.074 |
Token | minute |
Token ID | 5664 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
ez | +0.016 |
joy | +0.013 |
es | +0.012 |
hered | +0.012 |
tern | +0.011 |
Neg logit contributions | |
cool | -0.017 |
um | -0.017 |
Inside | -0.017 |
open | -0.017 |
interesting | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
joy | +0.608 |
ez | +0.599 |
hered | +0.542 |
Charl | +0.512 |
nightmares | +0.509 |
Neg logit contributions | |
um | -0.321 |
ires | -0.315 |
iring | -0.296 |
ate | -0.294 |
Inside | -0.256 |
Token | dog |
Token ID | 3290 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
um | +0.097 |
ires | +0.090 |
ate | +0.085 |
iring | +0.083 |
ruce | +0.073 |
Neg logit contributions | |
joy | -0.141 |
ez | -0.134 |
hered | -0.123 |
Charl | -0.120 |
nightmares | -0.117 |
Token | food |
Token ID | 2057 |
Feature activation | -8.89e-02 |
Pos logit contributions | |
joy | +0.475 |
ez | +0.431 |
hered | +0.400 |
moment | +0.388 |
nightmares | +0.387 |
Neg logit contributions | |
ires | -0.244 |
um | -0.242 |
ate | -0.229 |
iring | -0.222 |
lly | -0.195 |
Token | . |
Token ID | 13 |
Feature activation | +2.25e-03 |
Pos logit contributions | |
joy | +0.195 |
ez | +0.186 |
hered | +0.167 |
nightmares | +0.160 |
moment | +0.154 |
Neg logit contributions | |
ires | -0.130 |
um | -0.124 |
ate | -0.122 |
iring | -0.122 |
ruce | -0.106 |
Token | " |
Token ID | 366 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
ires | +0.003 |
um | +0.003 |
ate | +0.003 |
iring | +0.003 |
Inside | +0.003 |
Neg logit contributions | |
joy | -0.005 |
ez | -0.005 |
hered | -0.004 |
nightmares | -0.004 |
moment | -0.004 |
Token | Y |
Token ID | 56 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
joy | +0.099 |
nightmares | +0.089 |
ez | +0.085 |
breathing | +0.083 |
hered | +0.081 |
Neg logit contributions | |
um | -0.060 |
ate | -0.059 |
ires | -0.059 |
iring | -0.052 |
her | -0.051 |
Token | um |
Token ID | 388 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
ires | +0.016 |
ate | +0.015 |
iring | +0.013 |
ruce | +0.011 |
outed | +0.010 |
Neg logit contributions | |
ez | -0.040 |
joy | -0.040 |
hered | -0.040 |
ñ | -0.035 |
moment | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
um | +0.571 |
ires | +0.512 |
iring | +0.506 |
ate | +0.487 |
lly | +0.413 |
Neg logit contributions | |
joy | -1.101 |
ez | -1.070 |
hered | -0.982 |
nightmares | -0.971 |
Charl | -0.934 |
Token | this |
Token ID | 428 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
ires | +0.030 |
um | +0.029 |
iring | +0.028 |
ate | +0.027 |
Inside | +0.025 |
Neg logit contributions | |
joy | -0.054 |
ez | -0.052 |
hered | -0.047 |
nightmares | -0.044 |
breathing | -0.043 |
Token | is |
Token ID | 318 |
Feature activation | -9.12e-03 |
Pos logit contributions | |
um | +0.019 |
ires | +0.019 |
ate | +0.019 |
iring | +0.017 |
lly | +0.017 |
Neg logit contributions | |
joy | -0.029 |
ez | -0.027 |
hered | -0.025 |
nightmares | -0.023 |
smart | -0.023 |
Token | delicious |
Token ID | 12625 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
joy | +0.020 |
ez | +0.019 |
hered | +0.017 |
nightmares | +0.016 |
moment | +0.016 |
Neg logit contributions | |
ires | -0.014 |
ate | -0.013 |
iring | -0.013 |
um | -0.013 |
ruce | -0.011 |
Token | !" |
Token ID | 2474 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
ires | +0.053 |
um | +0.051 |
ate | +0.050 |
iring | +0.049 |
ruce | +0.043 |
Neg logit contributions | |
joy | -0.083 |
ez | -0.080 |
hered | -0.073 |
nightmares | -0.070 |
moment | -0.067 |
Token | him |
Token ID | 683 |
Feature activation | +9.45e-02 |
Pos logit contributions | |
joy | +0.146 |
ez | +0.135 |
hered | +0.125 |
nightmares | +0.115 |
moment | +0.115 |
Neg logit contributions | |
ires | -0.093 |
um | -0.089 |
iring | -0.087 |
ate | -0.087 |
ruce | -0.076 |
Token | fly |
Token ID | 6129 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
ires | +0.137 |
iring | +0.127 |
um | +0.126 |
ate | +0.125 |
ruce | +0.111 |
Neg logit contributions | |
joy | -0.205 |
ez | -0.200 |
hered | -0.184 |
nightmares | -0.174 |
moment | -0.173 |
Token | . |
Token ID | 13 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
ires | +0.020 |
ate | +0.019 |
iring | +0.018 |
um | +0.018 |
ruce | +0.016 |
Neg logit contributions | |
joy | -0.030 |
ez | -0.029 |
hered | -0.026 |
nightmares | -0.025 |
heaven | -0.024 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
ires | +0.020 |
ate | +0.019 |
um | +0.019 |
iring | +0.019 |
ruce | +0.016 |
Neg logit contributions | |
joy | -0.027 |
ez | -0.026 |
hered | -0.024 |
nightmares | -0.023 |
moment | -0.022 |
Token | M |
Token ID | 44 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
ate | +0.072 |
ires | +0.071 |
um | +0.068 |
iring | +0.066 |
Inside | +0.060 |
Neg logit contributions | |
joy | -0.122 |
ez | -0.120 |
hered | -0.109 |
nightmares | -0.109 |
knowing | -0.105 |
Token | um |
Token ID | 388 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
joy | +0.067 |
ez | +0.063 |
moment | +0.057 |
nightmares | +0.055 |
tale | +0.055 |
Neg logit contributions | |
ate | -0.043 |
ires | -0.043 |
iring | -0.041 |
ruce | -0.036 |
ised | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
ires | +0.780 |
ate | +0.735 |
iring | +0.735 |
um | +0.727 |
ruce | +0.657 |
Neg logit contributions | |
joy | -0.894 |
ez | -0.859 |
hered | -0.776 |
nightmares | -0.728 |
moment | -0.718 |
Token | Dad |
Token ID | 17415 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
joy | +0.091 |
ez | +0.088 |
hered | +0.079 |
nightmares | +0.077 |
moment | +0.074 |
Neg logit contributions | |
ires | -0.062 |
um | -0.060 |
ate | -0.059 |
iring | -0.059 |
ruce | -0.051 |
Token | had |
Token ID | 550 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
ires | +0.355 |
um | +0.341 |
ate | +0.341 |
iring | +0.331 |
ruce | +0.297 |
Neg logit contributions | |
joy | -0.469 |
ez | -0.447 |
hered | -0.405 |
nightmares | -0.396 |
moment | -0.381 |
Token | a |
Token ID | 257 |
Feature activation | -7.49e-03 |
Pos logit contributions | |
joy | +0.287 |
ez | +0.278 |
hered | +0.255 |
nightmares | +0.239 |
moment | +0.237 |
Neg logit contributions | |
ires | -0.212 |
ate | -0.202 |
um | -0.200 |
iring | -0.198 |
ruce | -0.177 |
Token | new |
Token ID | 649 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
ez | +0.019 |
joy | +0.018 |
hered | +0.017 |
nightmares | +0.016 |
smart | +0.016 |
Neg logit contributions | |
ires | -0.008 |
um | -0.008 |
ate | -0.007 |
iring | -0.007 |
ruce | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
joy | +0.269 |
ez | +0.259 |
hered | +0.236 |
nightmares | +0.223 |
heaven | +0.214 |
Neg logit contributions | |
ires | -0.201 |
ate | -0.191 |
um | -0.188 |
iring | -0.186 |
ruce | -0.163 |
Token |
|
Token ID | 198 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
ires | +0.028 |
ate | +0.026 |
um | +0.026 |
iring | +0.026 |
ruce | +0.023 |
Neg logit contributions | |
joy | -0.033 |
ez | -0.032 |
hered | -0.029 |
nightmares | -0.027 |
moment | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
ires | +0.057 |
iring | +0.054 |
ate | +0.053 |
um | +0.053 |
ruce | +0.048 |
Neg logit contributions | |
joy | -0.066 |
ez | -0.063 |
hered | -0.056 |
nightmares | -0.053 |
moment | -0.053 |
Token | " |
Token ID | 1 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
ires | +0.039 |
ate | +0.037 |
iring | +0.037 |
um | +0.037 |
ruce | +0.033 |
Neg logit contributions | |
joy | -0.052 |
ez | -0.050 |
hered | -0.046 |
nightmares | -0.043 |
moment | -0.042 |
Token | M |
Token ID | 44 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
ires | +0.068 |
ate | +0.066 |
um | +0.064 |
iring | +0.064 |
ruce | +0.056 |
Neg logit contributions | |
joy | -0.080 |
ez | -0.075 |
hered | -0.068 |
nightmares | -0.066 |
breathing | -0.063 |
Token | um |
Token ID | 388 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
joy | +0.071 |
ez | +0.066 |
hered | +0.060 |
nightmares | +0.059 |
moment | +0.059 |
Neg logit contributions | |
ires | -0.037 |
ate | -0.036 |
ruce | -0.034 |
iring | -0.033 |
outed | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
um | +0.590 |
ires | +0.579 |
ate | +0.571 |
iring | +0.545 |
lly | +0.467 |
Neg logit contributions | |
joy | -1.027 |
ez | -0.997 |
hered | -0.929 |
nightmares | -0.891 |
heaven | -0.853 |
Token | what |
Token ID | 644 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
ires | +0.047 |
um | +0.044 |
ate | +0.044 |
iring | +0.044 |
ruce | +0.039 |
Neg logit contributions | |
joy | -0.069 |
ez | -0.067 |
hered | -0.061 |
nightmares | -0.057 |
moment | -0.055 |
Token | are |
Token ID | 389 |
Feature activation | -8.93e-02 |
Pos logit contributions | |
ires | +0.028 |
ate | +0.028 |
um | +0.026 |
iring | +0.025 |
ruce | +0.024 |
Neg logit contributions | |
joy | -0.040 |
ez | -0.038 |
hered | -0.036 |
knowing | -0.032 |
moment | -0.031 |
Token | these |
Token ID | 777 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
joy | +0.101 |
ez | +0.090 |
nightmares | +0.087 |
risking | +0.080 |
knowing | +0.079 |
Neg logit contributions | |
ate | -0.185 |
outed | -0.179 |
ires | -0.178 |
um | -0.178 |
iring | -0.177 |
Token | ?" |
Token ID | 1701 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
ires | +0.075 |
ate | +0.069 |
iring | +0.068 |
um | +0.066 |
ruce | +0.061 |
Neg logit contributions | |
joy | -0.134 |
ez | -0.128 |
hered | -0.119 |
nightmares | -0.117 |
moment | -0.115 |
Token | time |
Token ID | 640 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
ires | +0.125 |
iring | +0.118 |
ate | +0.117 |
um | +0.117 |
ruce | +0.105 |
Neg logit contributions | |
joy | -0.162 |
ez | -0.157 |
hered | -0.139 |
nightmares | -0.137 |
moment | -0.134 |
Token | to |
Token ID | 284 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
joy | +0.106 |
ez | +0.103 |
hered | +0.093 |
nightmares | +0.088 |
moment | +0.085 |
Neg logit contributions | |
ires | -0.083 |
ate | -0.077 |
um | -0.077 |
iring | -0.077 |
ruce | -0.069 |
Token | eat |
Token ID | 4483 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
ires | +0.019 |
iring | +0.018 |
ate | +0.017 |
um | +0.017 |
ruce | +0.015 |
Neg logit contributions | |
joy | -0.026 |
ez | -0.025 |
hered | -0.023 |
nightmares | -0.022 |
moment | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -5.25e-03 |
Pos logit contributions | |
ires | +0.073 |
ate | +0.069 |
um | +0.069 |
iring | +0.068 |
ruce | +0.061 |
Neg logit contributions | |
joy | -0.095 |
ez | -0.092 |
hered | -0.084 |
nightmares | -0.078 |
moment | -0.077 |
Token | Y |
Token ID | 575 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
joy | +0.011 |
ez | +0.010 |
hered | +0.009 |
nightmares | +0.009 |
moment | +0.009 |
Neg logit contributions | |
ires | -0.008 |
iring | -0.008 |
ate | -0.008 |
um | -0.008 |
ruce | -0.007 |
Token | um |
Token ID | 388 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
joy | +0.154 |
ez | +0.150 |
hered | +0.147 |
ñ | +0.134 |
moment | +0.133 |
Neg logit contributions | |
ires | -0.045 |
iring | -0.038 |
ate | -0.035 |
ruce | -0.034 |
outed | -0.026 |
Token | ! |
Token ID | 0 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
um | +0.605 |
ires | +0.603 |
ate | +0.588 |
iring | +0.564 |
lly | +0.504 |
Neg logit contributions | |
joy | -1.043 |
ez | -0.978 |
hered | -0.928 |
nightmares | -0.870 |
moment | -0.845 |
Token | It |
Token ID | 632 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
joy | +0.062 |
ez | +0.059 |
hered | +0.054 |
nightmares | +0.051 |
moment | +0.050 |
Neg logit contributions | |
ires | -0.051 |
ate | -0.048 |
um | -0.048 |
iring | -0.048 |
ruce | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
ires | +0.046 |
ate | +0.044 |
um | +0.044 |
iring | +0.044 |
ruce | +0.039 |
Neg logit contributions | |
joy | -0.071 |
ez | -0.065 |
hered | -0.060 |
nightmares | -0.056 |
moment | -0.055 |
Token | the |
Token ID | 262 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
ires | +0.036 |
ate | +0.034 |
iring | +0.034 |
um | +0.034 |
ruce | +0.030 |
Neg logit contributions | |
joy | -0.046 |
ez | -0.044 |
hered | -0.041 |
nightmares | -0.038 |
moment | -0.038 |
Token | best |
Token ID | 1266 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
joy | +0.066 |
ez | +0.065 |
hered | +0.059 |
nightmares | +0.055 |
heaven | +0.053 |
Neg logit contributions | |
ires | -0.042 |
um | -0.039 |
iring | -0.039 |
ate | -0.039 |
ruce | -0.033 |
Token | look |
Token ID | 804 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
ires | +0.152 |
um | +0.144 |
ate | +0.144 |
iring | +0.142 |
ruce | +0.123 |
Neg logit contributions | |
joy | -0.208 |
ez | -0.203 |
hered | -0.186 |
nightmares | -0.173 |
Charl | -0.169 |
Token | ." |
Token ID | 526 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
ires | +0.161 |
ate | +0.152 |
iring | +0.151 |
um | +0.150 |
ruce | +0.133 |
Neg logit contributions | |
joy | -0.239 |
ez | -0.230 |
hered | -0.209 |
nightmares | -0.200 |
moment | -0.197 |
Token |
|
Token ID | 198 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
ires | +0.018 |
um | +0.018 |
ate | +0.017 |
iring | +0.017 |
Inside | +0.015 |
Neg logit contributions | |
joy | -0.026 |
ez | -0.026 |
hered | -0.024 |
nightmares | -0.023 |
moment | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
ires | +0.035 |
iring | +0.033 |
ate | +0.033 |
um | +0.032 |
ruce | +0.029 |
Neg logit contributions | |
joy | -0.038 |
ez | -0.036 |
hered | -0.032 |
nightmares | -0.030 |
moment | -0.030 |
Token | M |
Token ID | 44 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
ires | +0.023 |
ate | +0.022 |
um | +0.022 |
iring | +0.022 |
ruce | +0.019 |
Neg logit contributions | |
joy | -0.029 |
ez | -0.028 |
hered | -0.025 |
nightmares | -0.024 |
moment | -0.023 |
Token | um |
Token ID | 388 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
joy | +0.073 |
ez | +0.069 |
hered | +0.063 |
nightmares | +0.062 |
moment | +0.060 |
Neg logit contributions | |
ate | -0.036 |
ires | -0.034 |
iring | -0.032 |
ruce | -0.031 |
um | -0.026 |
Token | went |
Token ID | 1816 |
Feature activation | +8.83e-02 |
Pos logit contributions | |
ires | +0.650 |
um | +0.617 |
ate | +0.612 |
iring | +0.608 |
ruce | +0.531 |
Neg logit contributions | |
joy | -1.005 |
ez | -0.960 |
hered | -0.884 |
nightmares | -0.847 |
moment | -0.828 |
Token | outside |
Token ID | 2354 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
ires | +0.133 |
ate | +0.125 |
iring | +0.125 |
um | +0.123 |
ruce | +0.109 |
Neg logit contributions | |
joy | -0.190 |
ez | -0.182 |
hered | -0.167 |
nightmares | -0.158 |
moment | -0.156 |
Token | , |
Token ID | 11 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
um | +0.165 |
ires | +0.164 |
iring | +0.159 |
ate | +0.159 |
Inside | +0.154 |
Neg logit contributions | |
ez | -0.393 |
joy | -0.375 |
hered | -0.352 |
nightmares | -0.350 |
liar | -0.349 |
Token | and |
Token ID | 290 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
ires | +0.043 |
iring | +0.041 |
um | +0.041 |
ate | +0.041 |
outed | +0.036 |
Neg logit contributions | |
ez | -0.051 |
joy | -0.051 |
hered | -0.046 |
nightmares | -0.044 |
tale | -0.041 |
Token | found |
Token ID | 1043 |
Feature activation | +9.77e-02 |
Pos logit contributions | |
joy | +0.097 |
ez | +0.094 |
hered | +0.086 |
nightmares | +0.081 |
moment | +0.080 |
Neg logit contributions | |
ires | -0.069 |
ate | -0.064 |
iring | -0.064 |
um | -0.064 |
ruce | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +6.80e-03 |
Pos logit contributions | |
ires | +0.426 |
ate | +0.401 |
iring | +0.399 |
um | +0.392 |
ruce | +0.361 |
Neg logit contributions | |
joy | -0.448 |
ez | -0.426 |
hered | -0.390 |
moment | -0.364 |
nightmares | -0.359 |
Token |
|
Token ID | 198 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
ires | +0.011 |
iring | +0.011 |
ate | +0.011 |
um | +0.011 |
ruce | +0.010 |
Neg logit contributions | |
joy | -0.013 |
ez | -0.013 |
hered | -0.011 |
nightmares | -0.011 |
moment | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
ires | +0.041 |
iring | +0.040 |
um | +0.039 |
ate | +0.039 |
ruce | +0.036 |
Neg logit contributions | |
joy | -0.034 |
ez | -0.033 |
moment | -0.027 |
hered | -0.027 |
heaven | -0.026 |
Token | " |
Token ID | 1 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
ires | +0.025 |
iring | +0.024 |
ate | +0.023 |
um | +0.023 |
ruce | +0.021 |
Neg logit contributions | |
joy | -0.029 |
ez | -0.028 |
hered | -0.025 |
moment | -0.023 |
heaven | -0.023 |
Token | Y |
Token ID | 56 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
ires | +0.055 |
ate | +0.053 |
um | +0.051 |
iring | +0.051 |
ruce | +0.046 |
Neg logit contributions | |
joy | -0.062 |
ez | -0.059 |
hered | -0.053 |
nightmares | -0.051 |
breathing | -0.048 |
Token | um |
Token ID | 388 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
ires | +0.019 |
iring | +0.017 |
ate | +0.016 |
outed | +0.014 |
ised | +0.014 |
Neg logit contributions | |
hered | -0.032 |
joy | -0.030 |
ez | -0.030 |
ñ | -0.028 |
rav | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
ires | +0.626 |
um | +0.622 |
ate | +0.595 |
iring | +0.590 |
ruce | +0.504 |
Neg logit contributions | |
joy | -1.009 |
ez | -0.983 |
hered | -0.900 |
nightmares | -0.866 |
moment | -0.838 |
Token | y |
Token ID | 331 |
Feature activation | +7.72e-02 |
Pos logit contributions | |
ires | +0.038 |
ate | +0.036 |
iring | +0.036 |
um | +0.036 |
ruce | +0.032 |
Neg logit contributions | |
joy | -0.046 |
ez | -0.044 |
hered | -0.040 |
nightmares | -0.038 |
moment | -0.037 |
Token | um |
Token ID | 388 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
ires | +0.066 |
iring | +0.057 |
ate | +0.055 |
outed | +0.048 |
her | +0.046 |
Neg logit contributions | |
ez | -0.199 |
joy | -0.198 |
hered | -0.196 |
moment | -0.171 |
ñ | -0.170 |
Token | , |
Token ID | 11 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
ires | +0.612 |
um | +0.591 |
ate | +0.579 |
iring | +0.572 |
ruce | +0.489 |
Neg logit contributions | |
joy | -1.022 |
ez | -0.986 |
hered | -0.910 |
nightmares | -0.867 |
moment | -0.842 |
Token | these |
Token ID | 777 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
ires | +0.028 |
ate | +0.027 |
iring | +0.027 |
um | +0.026 |
ruce | +0.024 |
Neg logit contributions | |
joy | -0.035 |
ez | -0.033 |
hered | -0.030 |
moment | -0.028 |
nightmares | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +2.88e-03 |
Pos logit contributions | |
joy | +0.207 |
ez | +0.204 |
hered | +0.181 |
nightmares | +0.172 |
heaven | +0.164 |
Neg logit contributions | |
ires | -0.150 |
ate | -0.145 |
iring | -0.141 |
um | -0.138 |
ruce | -0.125 |
Token |
|
Token ID | 198 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
ires | +0.004 |
ate | +0.004 |
iring | +0.004 |
um | +0.004 |
ruce | +0.004 |
Neg logit contributions | |
joy | -0.006 |
ez | -0.006 |
hered | -0.005 |
nightmares | -0.005 |
moment | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
ires | +0.040 |
iring | +0.038 |
ate | +0.037 |
um | +0.037 |
ruce | +0.034 |
Neg logit contributions | |
joy | -0.038 |
ez | -0.036 |
hered | -0.031 |
moment | -0.031 |
nightmares | -0.029 |
Token | " |
Token ID | 1 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
ires | +0.026 |
iring | +0.025 |
ate | +0.024 |
um | +0.024 |
ruce | +0.022 |
Neg logit contributions | |
joy | -0.034 |
ez | -0.033 |
hered | -0.029 |
moment | -0.028 |
nightmares | -0.027 |
Token | Y |
Token ID | 56 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ires | +0.059 |
ate | +0.057 |
um | +0.056 |
iring | +0.056 |
ruce | +0.050 |
Neg logit contributions | |
joy | -0.071 |
ez | -0.068 |
hered | -0.060 |
nightmares | -0.059 |
breathing | -0.056 |
Token | um |
Token ID | 388 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
ires | +0.017 |
ate | +0.015 |
iring | +0.015 |
outed | +0.012 |
ruce | +0.012 |
Neg logit contributions | |
hered | -0.024 |
joy | -0.023 |
ez | -0.022 |
ñ | -0.022 |
rav | -0.021 |
Token | !" |
Token ID | 2474 |
Feature activation | +7.93e-02 |
Pos logit contributions | |
ires | +0.651 |
um | +0.628 |
iring | +0.617 |
ate | +0.616 |
ruce | +0.531 |
Neg logit contributions | |
joy | -0.994 |
ez | -0.965 |
hered | -0.884 |
nightmares | -0.840 |
moment | -0.818 |
Token | she |
Token ID | 673 |
Feature activation | -7.35e-02 |
Pos logit contributions | |
ires | +0.129 |
iring | +0.122 |
ate | +0.121 |
um | +0.116 |
ruce | +0.108 |
Neg logit contributions | |
joy | -0.162 |
ez | -0.155 |
hered | -0.141 |
nightmares | -0.132 |
moment | -0.131 |
Token | exclaimed |
Token ID | 33552 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
joy | +0.164 |
ez | +0.162 |
hered | +0.148 |
nightmares | +0.143 |
moment | +0.138 |
Neg logit contributions | |
ires | -0.105 |
iring | -0.099 |
ate | -0.093 |
um | -0.091 |
ruce | -0.082 |
Token | with |
Token ID | 351 |
Feature activation | +8.68e-02 |
Pos logit contributions | |
um | +0.267 |
ires | +0.259 |
iring | +0.246 |
ate | +0.229 |
outed | +0.200 |
Neg logit contributions | |
ez | -0.532 |
joy | -0.515 |
hered | -0.470 |
nightmares | -0.468 |
Charl | -0.464 |
Token | delight |
Token ID | 10974 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
ires | +0.155 |
ate | +0.147 |
iring | +0.147 |
um | +0.145 |
ruce | +0.132 |
Neg logit contributions | |
joy | -0.163 |
ez | -0.156 |
hered | -0.140 |
nightmares | -0.132 |
moment | -0.129 |
Token | her |
Token ID | 607 |
Feature activation | -5.94e-03 |
Pos logit contributions | |
ires | +0.380 |
um | +0.366 |
ate | +0.350 |
iring | +0.346 |
ruce | +0.295 |
Neg logit contributions | |
joy | -0.588 |
ez | -0.578 |
hered | -0.534 |
moment | -0.491 |
nightmares | -0.488 |
Token | mum |
Token ID | 25682 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
joy | +0.014 |
ez | +0.013 |
hered | +0.012 |
nightmares | +0.012 |
moment | +0.011 |
Neg logit contributions | |
ires | -0.008 |
ate | -0.007 |
iring | -0.007 |
um | -0.007 |
ruce | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
ires | +0.321 |
um | +0.306 |
ate | +0.302 |
iring | +0.296 |
ruce | +0.253 |
Neg logit contributions | |
joy | -0.500 |
ez | -0.496 |
hered | -0.455 |
nightmares | -0.427 |
Charl | -0.407 |
Token | " |
Token ID | 366 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
ires | +0.061 |
um | +0.060 |
iring | +0.059 |
ate | +0.057 |
Inside | +0.053 |
Neg logit contributions | |
ez | -0.089 |
joy | -0.084 |
nightmares | -0.079 |
hered | -0.078 |
heaven | -0.072 |
Token | M |
Token ID | 44 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
joy | +0.085 |
ez | +0.083 |
hered | +0.074 |
nightmares | +0.073 |
heaven | +0.068 |
Neg logit contributions | |
ires | -0.073 |
ate | -0.071 |
um | -0.070 |
iring | -0.070 |
ruce | -0.062 |
Token | um |
Token ID | 388 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
joy | +0.081 |
ez | +0.076 |
tale | +0.069 |
hered | +0.068 |
nightmares | +0.067 |
Neg logit contributions | |
ate | -0.047 |
ires | -0.046 |
ruce | -0.044 |
iring | -0.042 |
prett | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
ires | +0.587 |
um | +0.585 |
ate | +0.569 |
iring | +0.543 |
ruce | +0.456 |
Neg logit contributions | |
joy | -1.039 |
ez | -1.011 |
hered | -0.932 |
nightmares | -0.891 |
heaven | -0.852 |
Token | why |
Token ID | 1521 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
ires | +0.039 |
um | +0.037 |
ate | +0.036 |
iring | +0.036 |
ruce | +0.031 |
Neg logit contributions | |
joy | -0.060 |
ez | -0.059 |
hered | -0.054 |
nightmares | -0.051 |
moment | -0.049 |
Token | do |
Token ID | 466 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
ires | +0.138 |
ate | +0.133 |
iring | +0.127 |
um | +0.125 |
ruce | +0.115 |
Neg logit contributions | |
joy | -0.199 |
ez | -0.188 |
hered | -0.179 |
nightmares | -0.153 |
Charl | -0.152 |
Token | I |
Token ID | 314 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
ires | +0.223 |
ate | +0.211 |
um | +0.208 |
iring | +0.204 |
lly | +0.186 |
Neg logit contributions | |
joy | -0.403 |
ez | -0.378 |
hered | -0.347 |
knowing | -0.326 |
moment | -0.321 |
Token | feel |
Token ID | 1254 |
Feature activation | -8.32e-02 |
Pos logit contributions | |
ires | +0.046 |
ate | +0.044 |
iring | +0.044 |
um | +0.044 |
ruce | +0.039 |
Neg logit contributions | |
joy | -0.053 |
ez | -0.051 |
hered | -0.046 |
nightmares | -0.043 |
moment | -0.043 |
Token | that |
Token ID | 326 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
joy | +0.011 |
ez | +0.011 |
hered | +0.010 |
nightmares | +0.009 |
moment | +0.009 |
Neg logit contributions | |
ires | -0.009 |
ate | -0.008 |
iring | -0.008 |
um | -0.008 |
ruce | -0.007 |
Token | day |
Token ID | 1110 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
ires | +0.307 |
ate | +0.292 |
um | +0.285 |
iring | +0.277 |
ruce | +0.248 |
Neg logit contributions | |
joy | -0.493 |
ez | -0.458 |
hered | -0.445 |
Charl | -0.400 |
oing | -0.384 |
Token | . |
Token ID | 13 |
Feature activation | +3.15e-03 |
Pos logit contributions | |
joy | +0.108 |
ez | +0.103 |
hered | +0.092 |
oing | +0.091 |
nightmares | +0.090 |
Neg logit contributions | |
um | -0.054 |
ate | -0.047 |
ires | -0.047 |
iring | -0.044 |
Inside | -0.041 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
ires | +0.004 |
um | +0.004 |
ate | +0.004 |
iring | +0.004 |
Inside | +0.004 |
Neg logit contributions | |
joy | -0.007 |
ez | -0.007 |
hered | -0.006 |
nightmares | -0.006 |
moment | -0.006 |
Token | M |
Token ID | 44 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
ate | +0.067 |
ires | +0.066 |
um | +0.064 |
iring | +0.062 |
Inside | +0.056 |
Neg logit contributions | |
joy | -0.109 |
ez | -0.105 |
hered | -0.096 |
nightmares | -0.095 |
knowing | -0.093 |
Token | um |
Token ID | 388 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
joy | +0.082 |
ez | +0.077 |
moment | +0.069 |
nightmares | +0.069 |
hered | +0.068 |
Neg logit contributions | |
ires | -0.053 |
ate | -0.052 |
iring | -0.050 |
ruce | -0.044 |
ised | -0.039 |
Token | woke |
Token ID | 19092 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
ires | +0.764 |
ate | +0.720 |
iring | +0.719 |
um | +0.712 |
ruce | +0.642 |
Neg logit contributions | |
joy | -0.894 |
ez | -0.858 |
hered | -0.776 |
nightmares | -0.730 |
moment | -0.719 |
Token | up |
Token ID | 510 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
joy | +0.537 |
ez | +0.493 |
heaven | +0.464 |
knowing | +0.456 |
asleep | +0.454 |
Neg logit contributions | |
ires | -0.282 |
ate | -0.271 |
iring | -0.254 |
um | -0.229 |
ruce | -0.227 |
Token | early |
Token ID | 1903 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
ires | +0.351 |
um | +0.334 |
iring | +0.334 |
ate | +0.327 |
ruce | +0.274 |
Neg logit contributions | |
ez | -0.584 |
joy | -0.568 |
hered | -0.524 |
nightmares | -0.490 |
smart | -0.487 |
Token | . |
Token ID | 13 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
ires | +0.025 |
um | +0.023 |
ate | +0.023 |
iring | +0.023 |
ruce | +0.021 |
Neg logit contributions | |
joy | -0.032 |
ez | -0.031 |
hered | -0.028 |
nightmares | -0.027 |
moment | -0.026 |
Token | She |
Token ID | 1375 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
joy | +0.068 |
ez | +0.067 |
hered | +0.061 |
nightmares | +0.058 |
moment | +0.056 |
Neg logit contributions | |
ires | -0.043 |
um | -0.040 |
ate | -0.040 |
iring | -0.040 |
ruce | -0.035 |
Token | down |
Token ID | 866 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
ez | +0.055 |
joy | +0.054 |
hered | +0.047 |
nightmares | +0.046 |
heaven | +0.044 |
Neg logit contributions | |
ires | -0.032 |
um | -0.032 |
iring | -0.031 |
ate | -0.030 |
lly | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +3.65e-02 |
Pos logit contributions | |
ires | +0.178 |
ate | +0.171 |
um | +0.168 |
iring | +0.166 |
ruce | +0.143 |
Neg logit contributions | |
ez | -0.260 |
joy | -0.257 |
hered | -0.232 |
nightmares | -0.220 |
ñ | -0.213 |
Token | street |
Token ID | 4675 |
Feature activation | +8.68e-02 |
Pos logit contributions | |
ires | +0.047 |
um | +0.044 |
ate | +0.044 |
iring | +0.044 |
ruce | +0.037 |
Neg logit contributions | |
joy | -0.085 |
ez | -0.083 |
hered | -0.076 |
nightmares | -0.072 |
moment | -0.069 |
Token | , |
Token ID | 11 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
ires | +0.130 |
um | +0.125 |
ate | +0.125 |
iring | +0.124 |
ruce | +0.107 |
Neg logit contributions | |
joy | -0.178 |
ez | -0.178 |
hered | -0.158 |
nightmares | -0.153 |
Charl | -0.145 |
Token | adm |
Token ID | 6178 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
ires | +0.086 |
iring | +0.081 |
ate | +0.080 |
um | +0.079 |
ruce | +0.070 |
Neg logit contributions | |
ez | -0.112 |
joy | -0.112 |
hered | -0.102 |
nightmares | -0.095 |
tale | -0.092 |
Token | iring |
Token ID | 3428 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
joy | +0.080 |
ez | +0.070 |
knowing | +0.067 |
nightmares | +0.067 |
ching | +0.066 |
Neg logit contributions | |
First | -0.022 |
Inside | -0.018 |
ruce | -0.016 |
Later | -0.015 |
bury | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
ires | +0.654 |
iring | +0.608 |
ate | +0.605 |
um | +0.593 |
ruce | +0.513 |
Neg logit contributions | |
joy | -0.996 |
ez | -0.974 |
hered | -0.910 |
nightmares | -0.823 |
moment | -0.808 |
Token | big |
Token ID | 1263 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
joy | +0.027 |
ez | +0.026 |
hered | +0.024 |
nightmares | +0.022 |
heaven | +0.022 |
Neg logit contributions | |
ires | -0.016 |
iring | -0.015 |
um | -0.015 |
ate | -0.015 |
ruce | -0.013 |
Token | billboard |
Token ID | 41448 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
um | +0.238 |
ires | +0.238 |
iring | +0.218 |
ate | +0.218 |
windows | +0.191 |
Neg logit contributions | |
ez | -0.462 |
joy | -0.449 |
hered | -0.417 |
nightmares | -0.391 |
Charl | -0.391 |
Token | ahead |
Token ID | 4058 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
joy | +0.225 |
ez | +0.217 |
hered | +0.194 |
nightmares | +0.185 |
moment | +0.179 |
Neg logit contributions | |
ires | -0.159 |
iring | -0.151 |
um | -0.150 |
ate | -0.148 |
ruce | -0.131 |
Token | . |
Token ID | 13 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
ires | +0.126 |
um | +0.124 |
iring | +0.119 |
ate | +0.118 |
ruce | +0.103 |
Neg logit contributions | |
joy | -0.196 |
ez | -0.189 |
hered | -0.172 |
nightmares | -0.166 |
Charl | -0.159 |
Token | , |
Token ID | 11 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
joy | +0.236 |
ez | +0.209 |
hered | +0.199 |
moment | +0.187 |
nightmares | +0.185 |
Neg logit contributions | |
ires | -0.148 |
ate | -0.139 |
iring | -0.135 |
um | -0.130 |
ruce | -0.127 |
Token | the |
Token ID | 262 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
ires | +0.017 |
iring | +0.017 |
um | +0.016 |
ate | +0.016 |
Inside | +0.015 |
Neg logit contributions | |
ez | -0.028 |
joy | -0.025 |
hered | -0.024 |
oing | -0.022 |
nightmares | -0.021 |
Token | King |
Token ID | 2677 |
Feature activation | -9.34e-02 |
Pos logit contributions | |
joy | +0.057 |
ez | +0.056 |
hered | +0.048 |
nightmares | +0.043 |
moment | +0.041 |
Neg logit contributions | |
ires | -0.085 |
ate | -0.082 |
iring | -0.081 |
um | -0.081 |
ruce | -0.074 |
Token | was |
Token ID | 373 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
joy | +0.195 |
ez | +0.188 |
hered | +0.168 |
nightmares | +0.165 |
heaven | +0.158 |
Neg logit contributions | |
ires | -0.139 |
ate | -0.137 |
um | -0.137 |
iring | -0.136 |
ruce | -0.117 |
Token | over |
Token ID | 625 |
Feature activation | +7.55e-02 |
Pos logit contributions | |
ires | +0.026 |
iring | +0.024 |
ate | +0.024 |
um | +0.024 |
ruce | +0.021 |
Neg logit contributions | |
joy | -0.032 |
ez | -0.030 |
hered | -0.028 |
nightmares | -0.026 |
moment | -0.025 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
um | +0.211 |
ires | +0.211 |
iring | +0.208 |
ate | +0.206 |
Inside | +0.198 |
Neg logit contributions | |
ez | -0.055 |
ñ | -0.032 |
hered | -0.030 |
nightmares | -0.029 |
Charl | -0.029 |
Token | ed |
Token ID | 276 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
joy | +1.223 |
nightmares | +1.221 |
failure | +1.196 |
attacking | +1.195 |
foe | +1.184 |
Neg logit contributions | |
um | -0.717 |
iring | -0.641 |
ires | -0.637 |
ate | -0.611 |
am | -0.516 |
Token | . |
Token ID | 13 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
ez | +0.186 |
joy | +0.173 |
hered | +0.163 |
nightmares | +0.162 |
Charl | +0.156 |
Neg logit contributions | |
ires | -0.104 |
iring | -0.104 |
um | -0.104 |
ate | -0.097 |
Inside | -0.079 |
Token | He |
Token ID | 679 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
joy | +0.079 |
ez | +0.077 |
hered | +0.071 |
nightmares | +0.067 |
moment | +0.066 |
Neg logit contributions | |
ires | -0.050 |
ate | -0.047 |
um | -0.046 |
iring | -0.046 |
ruce | -0.039 |
Token | put |
Token ID | 1234 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
ires | +0.100 |
iring | +0.093 |
ate | +0.089 |
um | +0.088 |
ruce | +0.082 |
Neg logit contributions | |
joy | -0.097 |
ez | -0.095 |
hered | -0.086 |
moment | -0.079 |
knowing | -0.079 |
Token | on |
Token ID | 319 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
ires | +0.366 |
ate | +0.349 |
um | +0.344 |
iring | +0.343 |
ruce | +0.305 |
Neg logit contributions | |
joy | -0.455 |
ez | -0.439 |
hered | -0.399 |
nightmares | -0.380 |
moment | -0.366 |
Token | him |
Token ID | 683 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
ires | +0.049 |
um | +0.045 |
ate | +0.045 |
iring | +0.043 |
ruce | +0.038 |
Neg logit contributions | |
joy | -0.096 |
ez | -0.095 |
hered | -0.088 |
nightmares | -0.084 |
Charl | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
ires | +0.029 |
iring | +0.028 |
ate | +0.027 |
um | +0.026 |
Inside | +0.023 |
Neg logit contributions | |
joy | -0.060 |
ez | -0.058 |
hered | -0.053 |
Charl | -0.051 |
nightmares | -0.050 |
Token | Moe |
Token ID | 46488 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
joy | +0.052 |
ez | +0.044 |
hered | +0.040 |
moment | +0.037 |
heaven | +0.036 |
Neg logit contributions | |
ate | -0.054 |
ires | -0.054 |
iring | -0.053 |
um | -0.050 |
ruce | -0.049 |
Token | was |
Token ID | 373 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
ires | +0.245 |
ate | +0.243 |
um | +0.241 |
iring | +0.235 |
ruce | +0.210 |
Neg logit contributions | |
joy | -0.359 |
ez | -0.339 |
hered | -0.306 |
nightmares | -0.299 |
moment | -0.290 |
Token | over |
Token ID | 625 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
ires | +0.023 |
iring | +0.023 |
ate | +0.023 |
um | +0.022 |
ruce | +0.019 |
Neg logit contributions | |
joy | -0.043 |
ez | -0.042 |
hered | -0.038 |
nightmares | -0.037 |
moment | -0.036 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
ires | +0.204 |
iring | +0.202 |
um | +0.200 |
ate | +0.196 |
Inside | +0.190 |
Neg logit contributions | |
ez | -0.065 |
lull | -0.039 |
ñ | -0.038 |
Charl | -0.037 |
nightmares | -0.036 |
Token | ed |
Token ID | 276 |
Feature activation | -7.90e-02 |
Pos logit contributions | |
joy | +1.298 |
nightmares | +1.243 |
failure | +1.237 |
heaven | +1.213 |
Charl | +1.198 |
Neg logit contributions | |
um | -0.646 |
iring | -0.605 |
ires | -0.593 |
ate | -0.583 |
am | -0.532 |
Token | to |
Token ID | 284 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
ez | +0.196 |
joy | +0.180 |
hered | +0.173 |
nightmares | +0.169 |
Charl | +0.167 |
Neg logit contributions | |
iring | -0.093 |
ires | -0.089 |
um | -0.087 |
ate | -0.082 |
up | -0.067 |
Token | see |
Token ID | 766 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
joy | +0.056 |
ez | +0.054 |
hered | +0.048 |
nightmares | +0.047 |
heaven | +0.045 |
Neg logit contributions | |
ires | -0.040 |
ate | -0.039 |
um | -0.039 |
iring | -0.037 |
ruce | -0.033 |
Token | them |
Token ID | 606 |
Feature activation | +4.65e-02 |
Pos logit contributions | |
ires | +0.172 |
iring | +0.160 |
ate | +0.160 |
um | +0.155 |
ruce | +0.135 |
Neg logit contributions | |
joy | -0.250 |
ez | -0.245 |
hered | -0.227 |
nightmares | -0.211 |
moment | -0.205 |
Token | again |
Token ID | 757 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
iring | +0.065 |
ires | +0.063 |
ate | +0.060 |
um | +0.059 |
ruce | +0.054 |
Neg logit contributions | |
ez | -0.101 |
joy | -0.100 |
Charl | -0.088 |
hered | -0.085 |
nightmares | -0.083 |
Token | noises |
Token ID | 26782 |
Feature activation | -7.13e-02 |
Pos logit contributions | |
ires | +0.170 |
ate | +0.167 |
iring | +0.157 |
um | +0.154 |
windows | +0.136 |
Neg logit contributions | |
joy | -0.256 |
ez | -0.244 |
hered | -0.231 |
Charl | -0.217 |
Patch | -0.210 |
Token | . |
Token ID | 13 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
joy | +0.180 |
ez | +0.165 |
hered | +0.147 |
moment | +0.143 |
nightmares | +0.142 |
Neg logit contributions | |
ires | -0.081 |
iring | -0.078 |
um | -0.075 |
ate | -0.075 |
her | -0.068 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -7.37e-02 |
Pos logit contributions | |
joy | +0.058 |
ez | +0.056 |
hered | +0.050 |
nightmares | +0.047 |
moment | +0.047 |
Neg logit contributions | |
ires | -0.051 |
ate | -0.049 |
iring | -0.048 |
um | -0.048 |
ruce | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
joy | +0.161 |
ez | +0.149 |
hered | +0.139 |
nightmares | +0.134 |
moment | +0.130 |
Neg logit contributions | |
ate | -0.103 |
ires | -0.103 |
um | -0.102 |
iring | -0.098 |
ruce | -0.087 |
Token | over |
Token ID | 625 |
Feature activation | +7.80e-02 |
Pos logit contributions | |
ires | +0.023 |
ate | +0.023 |
iring | +0.022 |
um | +0.022 |
ruce | +0.019 |
Neg logit contributions | |
joy | -0.041 |
ez | -0.040 |
hered | -0.037 |
nightmares | -0.035 |
moment | -0.034 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
ires | +0.213 |
um | +0.210 |
iring | +0.209 |
ate | +0.206 |
Inside | +0.192 |
Neg logit contributions | |
ez | -0.064 |
ñ | -0.038 |
hered | -0.038 |
ching | -0.034 |
liar | -0.033 |
Token | ed |
Token ID | 276 |
Feature activation | -8.00e-02 |
Pos logit contributions | |
joy | +1.191 |
ez | +1.102 |
failure | +1.096 |
nightmares | +1.090 |
attacking | +1.079 |
Neg logit contributions | |
um | -0.798 |
ires | -0.749 |
ate | -0.726 |
iring | -0.724 |
outed | -0.623 |
Token | and |
Token ID | 290 |
Feature activation | -5.78e-02 |
Pos logit contributions | |
ez | +0.181 |
joy | +0.174 |
hered | +0.160 |
nightmares | +0.154 |
ñ | +0.150 |
Neg logit contributions | |
ires | -0.108 |
um | -0.106 |
iring | -0.106 |
ate | -0.102 |
ruce | -0.081 |
Token | promised |
Token ID | 8072 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
joy | +0.122 |
ez | +0.117 |
hered | +0.107 |
nightmares | +0.101 |
moment | +0.100 |
Neg logit contributions | |
ires | -0.090 |
iring | -0.084 |
ate | -0.083 |
um | -0.083 |
ruce | -0.074 |
Token | to |
Token ID | 284 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
ez | +0.306 |
joy | +0.305 |
hered | +0.270 |
nightmares | +0.254 |
heaven | +0.245 |
Neg logit contributions | |
ires | -0.194 |
um | -0.191 |
ate | -0.190 |
iring | -0.187 |
ruce | -0.155 |
Token | keep |
Token ID | 1394 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
joy | +0.046 |
ez | +0.045 |
hered | +0.041 |
moment | +0.039 |
nightmares | +0.039 |
Neg logit contributions | |
ires | -0.043 |
iring | -0.041 |
um | -0.038 |
ate | -0.038 |
outed | -0.036 |
Token | player |
Token ID | 2137 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
joy | +0.102 |
ez | +0.094 |
hered | +0.081 |
Charl | +0.081 |
Sleep | +0.079 |
Neg logit contributions | |
um | -0.066 |
ires | -0.066 |
ate | -0.063 |
iring | -0.061 |
ruce | -0.056 |
Token | ! |
Token ID | 0 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
joy | +0.502 |
ez | +0.471 |
Charl | +0.424 |
hered | +0.424 |
Sleep | +0.418 |
Neg logit contributions | |
ires | -0.246 |
ate | -0.231 |
um | -0.230 |
iring | -0.219 |
ruce | -0.196 |
Token | Emily |
Token ID | 17608 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
joy | +0.078 |
ez | +0.074 |
hered | +0.067 |
nightmares | +0.063 |
moment | +0.061 |
Neg logit contributions | |
ires | -0.075 |
ate | -0.071 |
iring | -0.071 |
um | -0.071 |
ruce | -0.064 |
Token | was |
Token ID | 373 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
joy | +0.422 |
ez | +0.395 |
hered | +0.360 |
nightmares | +0.347 |
moment | +0.339 |
Neg logit contributions | |
ires | -0.281 |
ate | -0.279 |
um | -0.274 |
iring | -0.267 |
ruce | -0.244 |
Token | over |
Token ID | 625 |
Feature activation | +7.55e-02 |
Pos logit contributions | |
ires | +0.025 |
iring | +0.024 |
ate | +0.024 |
um | +0.023 |
ruce | +0.020 |
Neg logit contributions | |
joy | -0.049 |
ez | -0.047 |
hered | -0.044 |
nightmares | -0.041 |
moment | -0.041 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
ires | +0.213 |
iring | +0.211 |
um | +0.210 |
ate | +0.206 |
Inside | +0.193 |
Neg logit contributions | |
ez | -0.056 |
hered | -0.028 |
ñ | -0.028 |
liar | -0.027 |
oscope | -0.026 |
Token | ed |
Token ID | 276 |
Feature activation | -7.79e-02 |
Pos logit contributions | |
joy | +1.238 |
nightmares | +1.170 |
failure | +1.160 |
foe | +1.142 |
attacking | +1.141 |
Neg logit contributions | |
um | -0.732 |
iring | -0.691 |
ires | -0.677 |
ate | -0.666 |
outed | -0.555 |
Token | as |
Token ID | 355 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
ez | +0.189 |
joy | +0.177 |
hered | +0.165 |
nightmares | +0.163 |
liar | +0.159 |
Neg logit contributions | |
iring | -0.096 |
um | -0.094 |
ires | -0.093 |
ate | -0.088 |
up | -0.068 |
Token | she |
Token ID | 673 |
Feature activation | -8.30e-02 |
Pos logit contributions | |
ires | +0.040 |
ate | +0.037 |
iring | +0.036 |
um | +0.032 |
Inside | +0.029 |
Neg logit contributions | |
ez | -0.068 |
joy | -0.065 |
hered | -0.064 |
ñ | -0.056 |
oscope | -0.056 |
Token | recorded |
Token ID | 6264 |
Feature activation | -9.20e-02 |
Pos logit contributions | |
joy | +0.158 |
ez | +0.153 |
hered | +0.138 |
nightmares | +0.129 |
moment | +0.126 |
Neg logit contributions | |
ires | -0.147 |
iring | -0.138 |
ate | -0.136 |
um | -0.136 |
ruce | -0.122 |
Token | the |
Token ID | 262 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
joy | +0.220 |
ez | +0.220 |
hered | +0.206 |
Charl | +0.193 |
Sleep | +0.192 |
Neg logit contributions | |
ires | -0.102 |
ate | -0.097 |
um | -0.093 |
iring | -0.092 |
all | -0.078 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
joy | +0.048 |
ez | +0.042 |
hered | +0.037 |
knowing | +0.036 |
moment | +0.036 |
Neg logit contributions | |
ires | -0.061 |
ate | -0.059 |
um | -0.058 |
iring | -0.057 |
ruce | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
ez | +0.314 |
joy | +0.291 |
hered | +0.291 |
Charl | +0.275 |
ching | +0.259 |
Neg logit contributions | |
ires | -0.159 |
iring | -0.155 |
ate | -0.145 |
um | -0.143 |
Inside | -0.111 |
Token | She |
Token ID | 1375 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
ez | +0.077 |
joy | +0.076 |
hered | +0.072 |
nightmares | +0.068 |
tale | +0.066 |
Neg logit contributions | |
ires | -0.044 |
um | -0.040 |
ate | -0.039 |
iring | -0.038 |
Inside | -0.036 |
Token | was |
Token ID | 373 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
ires | +0.069 |
ate | +0.066 |
iring | +0.065 |
um | +0.065 |
ruce | +0.059 |
Neg logit contributions | |
joy | -0.073 |
ez | -0.069 |
hered | -0.063 |
nightmares | -0.059 |
moment | -0.058 |
Token | over |
Token ID | 625 |
Feature activation | +7.32e-02 |
Pos logit contributions | |
ate | +0.021 |
ires | +0.021 |
iring | +0.021 |
um | +0.021 |
ruce | +0.018 |
Neg logit contributions | |
joy | -0.029 |
ez | -0.029 |
hered | -0.027 |
nightmares | -0.025 |
ñ | -0.025 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
ires | +0.211 |
iring | +0.206 |
um | +0.206 |
ate | +0.202 |
Inside | +0.188 |
Neg logit contributions | |
ez | -0.051 |
ñ | -0.035 |
hered | -0.032 |
lull | -0.028 |
Charl | -0.027 |
Token | ed |
Token ID | 276 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
joy | +1.209 |
nightmares | +1.183 |
heaven | +1.157 |
ez | +1.145 |
failure | +1.139 |
Neg logit contributions | |
um | -0.749 |
ires | -0.704 |
ate | -0.697 |
iring | -0.692 |
am | -0.583 |
Token | that |
Token ID | 326 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
ez | +0.183 |
joy | +0.176 |
hered | +0.165 |
nightmares | +0.158 |
ñ | +0.156 |
Neg logit contributions | |
ires | -0.104 |
iring | -0.103 |
um | -0.101 |
ate | -0.096 |
Inside | -0.074 |
Token | her |
Token ID | 607 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
ires | +0.153 |
iring | +0.143 |
um | +0.143 |
all | +0.138 |
ate | +0.136 |
Neg logit contributions | |
ez | -0.386 |
joy | -0.366 |
hered | -0.356 |
ñ | -0.351 |
oing | -0.339 |
Token | pet |
Token ID | 4273 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
ez | +0.102 |
joy | +0.102 |
hered | +0.091 |
nightmares | +0.087 |
moment | +0.084 |
Neg logit contributions | |
ires | -0.061 |
um | -0.057 |
iring | -0.057 |
ate | -0.057 |
ruce | -0.047 |
Token | was |
Token ID | 373 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
ez | +0.042 |
joy | +0.042 |
nightmares | +0.036 |
hered | +0.035 |
breathing | +0.034 |
Neg logit contributions | |
um | -0.034 |
ires | -0.034 |
iring | -0.034 |
ate | -0.033 |
lly | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
joy | +0.108 |
ez | +0.104 |
hered | +0.093 |
nightmares | +0.092 |
Charl | +0.090 |
Neg logit contributions | |
ate | -0.067 |
ires | -0.067 |
um | -0.064 |
iring | -0.063 |
ered | -0.054 |
Token | O |
Token ID | 440 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
joy | +0.076 |
ez | +0.073 |
hered | +0.066 |
nightmares | +0.061 |
moment | +0.061 |
Neg logit contributions | |
ires | -0.060 |
ate | -0.057 |
iring | -0.056 |
um | -0.056 |
ruce | -0.051 |
Token | lly |
Token ID | 12810 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
ires | +0.124 |
ate | +0.106 |
iring | +0.105 |
um | +0.101 |
ruce | +0.079 |
Neg logit contributions | |
joy | -0.455 |
ez | -0.442 |
hered | -0.413 |
moment | -0.395 |
nightmares | -0.389 |
Token | was |
Token ID | 373 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
ires | +0.488 |
ate | +0.485 |
um | +0.478 |
iring | +0.463 |
ruce | +0.404 |
Neg logit contributions | |
joy | -0.732 |
ez | -0.693 |
hered | -0.636 |
nightmares | -0.618 |
heaven | -0.595 |
Token | over |
Token ID | 625 |
Feature activation | +7.48e-02 |
Pos logit contributions | |
ires | +0.015 |
ate | +0.015 |
um | +0.014 |
iring | +0.014 |
ruce | +0.012 |
Neg logit contributions | |
joy | -0.028 |
ez | -0.028 |
hered | -0.026 |
nightmares | -0.025 |
moment | -0.024 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ires | +0.221 |
iring | +0.219 |
um | +0.216 |
ate | +0.215 |
Inside | +0.198 |
Neg logit contributions | |
ez | -0.046 |
hered | -0.023 |
ñ | -0.022 |
nightmares | -0.019 |
Charl | -0.018 |
Token | ed |
Token ID | 276 |
Feature activation | -7.81e-02 |
Pos logit contributions | |
joy | +1.202 |
nightmares | +1.171 |
heaven | +1.145 |
failure | +1.123 |
ez | +1.116 |
Neg logit contributions | |
um | -0.773 |
ires | -0.714 |
iring | -0.711 |
ate | -0.698 |
am | -0.587 |
Token | with |
Token ID | 351 |
Feature activation | +7.55e-02 |
Pos logit contributions | |
ez | +0.169 |
joy | +0.166 |
hered | +0.152 |
nightmares | +0.147 |
heaven | +0.141 |
Neg logit contributions | |
ires | -0.112 |
iring | -0.109 |
um | -0.109 |
ate | -0.106 |
ruce | -0.086 |
Token | his |
Token ID | 465 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
ires | +0.113 |
ate | +0.107 |
iring | +0.106 |
um | +0.105 |
ruce | +0.093 |
Neg logit contributions | |
joy | -0.162 |
ez | -0.156 |
hered | -0.144 |
nightmares | -0.135 |
moment | -0.134 |
Token | prize |
Token ID | 11596 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
ez | +0.050 |
joy | +0.047 |
hered | +0.044 |
heaven | +0.042 |
nightmares | +0.041 |
Neg logit contributions | |
ires | -0.031 |
um | -0.030 |
ate | -0.029 |
iring | -0.028 |
ruce | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | +8.90e-03 |
Pos logit contributions | |
ez | +0.081 |
joy | +0.080 |
hered | +0.070 |
nightmares | +0.068 |
Charl | +0.067 |
Neg logit contributions | |
ires | -0.044 |
um | -0.043 |
iring | -0.043 |
ate | -0.040 |
lly | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
joy | +0.161 |
ez | +0.153 |
hered | +0.138 |
nightmares | +0.128 |
moment | +0.127 |
Neg logit contributions | |
ires | -0.154 |
iring | -0.145 |
ate | -0.144 |
um | -0.143 |
ruce | -0.132 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
joy | +0.020 |
ez | +0.019 |
hered | +0.017 |
nightmares | +0.016 |
moment | +0.016 |
Neg logit contributions | |
ires | -0.018 |
ate | -0.017 |
iring | -0.017 |
um | -0.017 |
ruce | -0.015 |
Token | Jack |
Token ID | 14295 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
joy | +0.021 |
ez | +0.020 |
hered | +0.018 |
nightmares | +0.018 |
moment | +0.017 |
Neg logit contributions | |
ires | -0.016 |
ate | -0.015 |
iring | -0.015 |
um | -0.015 |
ruce | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
ires | +0.025 |
um | +0.025 |
ate | +0.025 |
iring | +0.024 |
ruce | +0.021 |
Neg logit contributions | |
joy | -0.041 |
ez | -0.038 |
hered | -0.036 |
nightmares | -0.034 |
heaven | -0.033 |
Token | over |
Token ID | 625 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
ires | +0.017 |
ate | +0.017 |
um | +0.017 |
iring | +0.016 |
ruce | +0.014 |
Neg logit contributions | |
joy | -0.027 |
ez | -0.026 |
hered | -0.024 |
nightmares | -0.023 |
moment | -0.022 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ires | +0.118 |
um | +0.115 |
iring | +0.114 |
ate | +0.109 |
Inside | +0.095 |
Neg logit contributions | |
ez | -0.146 |
hered | -0.130 |
joy | -0.128 |
nightmares | -0.123 |
ñ | -0.122 |
Token | ed |
Token ID | 276 |
Feature activation | -7.99e-02 |
Pos logit contributions | |
joy | +1.114 |
ez | +1.057 |
nightmares | +1.048 |
heaven | +1.024 |
failure | +0.991 |
Neg logit contributions | |
um | -0.858 |
ires | -0.815 |
ate | -0.797 |
iring | -0.778 |
am | -0.665 |
Token | . |
Token ID | 13 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
joy | +0.170 |
ez | +0.168 |
hered | +0.152 |
nightmares | +0.146 |
heaven | +0.140 |
Neg logit contributions | |
ires | -0.118 |
um | -0.115 |
iring | -0.115 |
ate | -0.112 |
ruce | -0.093 |
Token | He |
Token ID | 679 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
joy | +0.067 |
ez | +0.064 |
hered | +0.058 |
nightmares | +0.055 |
moment | +0.055 |
Neg logit contributions | |
ires | -0.048 |
ate | -0.045 |
iring | -0.045 |
um | -0.044 |
ruce | -0.040 |
Token | had |
Token ID | 550 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
ires | +0.090 |
iring | +0.084 |
ate | +0.082 |
um | +0.080 |
ruce | +0.073 |
Neg logit contributions | |
joy | -0.112 |
ez | -0.110 |
hered | -0.099 |
nightmares | -0.093 |
moment | -0.093 |
Token | conquered |
Token ID | 29346 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
joy | +0.296 |
ez | +0.289 |
hered | +0.269 |
nightmares | +0.247 |
moment | +0.244 |
Neg logit contributions | |
ate | -0.226 |
ires | -0.223 |
um | -0.217 |
iring | -0.210 |
ruce | -0.196 |
Token | happen |
Token ID | 1645 |
Feature activation | -7.80e-02 |
Pos logit contributions | |
joy | +0.065 |
ez | +0.062 |
hered | +0.056 |
nightmares | +0.052 |
Charl | +0.050 |
Neg logit contributions | |
ires | -0.043 |
ate | -0.041 |
iring | -0.040 |
um | -0.039 |
ruce | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
joy | +0.169 |
ez | +0.169 |
hered | +0.153 |
nightmares | +0.141 |
Charl | +0.134 |
Neg logit contributions | |
ires | -0.118 |
ate | -0.113 |
iring | -0.111 |
um | -0.110 |
ruce | -0.092 |
Token | Jane |
Token ID | 12091 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
joy | +0.069 |
ez | +0.067 |
hered | +0.061 |
nightmares | +0.058 |
moment | +0.057 |
Neg logit contributions | |
ires | -0.051 |
ate | -0.048 |
um | -0.048 |
iring | -0.048 |
ruce | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
ate | +0.090 |
ires | +0.087 |
um | +0.087 |
iring | +0.083 |
lly | +0.073 |
Neg logit contributions | |
joy | -0.145 |
ez | -0.128 |
hered | -0.123 |
nightmares | -0.117 |
heaven | -0.114 |
Token | over |
Token ID | 625 |
Feature activation | +7.61e-02 |
Pos logit contributions | |
ate | +0.015 |
ires | +0.015 |
iring | +0.015 |
um | +0.015 |
ruce | +0.012 |
Neg logit contributions | |
joy | -0.034 |
ez | -0.033 |
hered | -0.032 |
nightmares | -0.029 |
moment | -0.029 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ires | +0.204 |
um | +0.200 |
iring | +0.199 |
ate | +0.196 |
Inside | +0.181 |
Neg logit contributions | |
ez | -0.070 |
hered | -0.047 |
ñ | -0.047 |
oing | -0.041 |
ching | -0.037 |
Token | ed |
Token ID | 276 |
Feature activation | -7.78e-02 |
Pos logit contributions | |
joy | +1.230 |
attacking | +1.187 |
nightmares | +1.180 |
failure | +1.173 |
ez | +1.172 |
Neg logit contributions | |
um | -0.733 |
ires | -0.692 |
ate | -0.679 |
iring | -0.654 |
Inside | -0.521 |
Token | , |
Token ID | 11 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
ez | +0.180 |
joy | +0.172 |
hered | +0.159 |
nightmares | +0.152 |
ñ | +0.150 |
Neg logit contributions | |
ires | -0.104 |
iring | -0.103 |
um | -0.102 |
ate | -0.098 |
ised | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
ires | +0.015 |
iring | +0.014 |
um | +0.014 |
ate | +0.014 |
Inside | +0.012 |
Neg logit contributions | |
ez | -0.028 |
joy | -0.027 |
hered | -0.026 |
ñ | -0.024 |
liar | -0.023 |
Token | thanked |
Token ID | 26280 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
joy | +0.120 |
ez | +0.115 |
hered | +0.104 |
nightmares | +0.099 |
moment | +0.095 |
Neg logit contributions | |
ires | -0.093 |
ate | -0.089 |
iring | -0.089 |
um | -0.089 |
ruce | -0.077 |
Token | Pixie |
Token ID | 46687 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
ez | +0.040 |
joy | +0.040 |
hered | +0.036 |
nightmares | +0.033 |
moment | +0.033 |
Neg logit contributions | |
ires | -0.027 |
um | -0.025 |
ate | -0.025 |
iring | -0.024 |
ruce | -0.020 |
Token | The |
Token ID | 464 |
Feature activation | -5.26e-04 |
Pos logit contributions | |
joy | +0.022 |
ez | +0.021 |
hered | +0.019 |
nightmares | +0.018 |
moment | +0.018 |
Neg logit contributions | |
ires | -0.015 |
ate | -0.014 |
iring | -0.014 |
um | -0.014 |
ruce | -0.012 |
Token | oct |
Token ID | 19318 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
joy | +0.001 |
ez | +0.001 |
hered | +0.001 |
nightmares | +0.001 |
moment | +0.001 |
Neg logit contributions | |
ires | -0.001 |
ate | -0.001 |
iring | -0.001 |
um | -0.001 |
ruce | -0.001 |
Token | opus |
Token ID | 25790 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
ires | +0.058 |
ate | +0.055 |
um | +0.055 |
iring | +0.055 |
ruce | +0.050 |
Neg logit contributions | |
joy | -0.060 |
ez | -0.057 |
hered | -0.051 |
nightmares | -0.046 |
moment | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
joy | +0.027 |
ez | +0.024 |
hered | +0.023 |
nightmares | +0.022 |
heaven | +0.021 |
Neg logit contributions | |
um | -0.017 |
ate | -0.017 |
ires | -0.017 |
iring | -0.016 |
lly | -0.015 |
Token | over |
Token ID | 625 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
ires | +0.018 |
iring | +0.017 |
ate | +0.017 |
um | +0.017 |
ruce | +0.015 |
Neg logit contributions | |
joy | -0.031 |
ez | -0.030 |
hered | -0.028 |
nightmares | -0.026 |
moment | -0.026 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
um | +0.197 |
ires | +0.196 |
iring | +0.196 |
ate | +0.192 |
Inside | +0.185 |
Neg logit contributions | |
ez | -0.057 |
ñ | -0.047 |
Charl | -0.041 |
hered | -0.037 |
nightmares | -0.034 |
Token | ed |
Token ID | 276 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
joy | +1.173 |
nightmares | +1.146 |
heaven | +1.119 |
ez | +1.091 |
forgetting | +1.068 |
Neg logit contributions | |
um | -0.818 |
ires | -0.763 |
ate | -0.752 |
iring | -0.751 |
am | -0.641 |
Token | , |
Token ID | 11 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
ez | +0.184 |
joy | +0.176 |
hered | +0.165 |
nightmares | +0.164 |
Charl | +0.162 |
Neg logit contributions | |
iring | -0.099 |
um | -0.099 |
ires | -0.097 |
ate | -0.091 |
ised | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
iring | +0.012 |
ires | +0.011 |
um | +0.011 |
ate | +0.011 |
Inside | +0.010 |
Neg logit contributions | |
ez | -0.024 |
joy | -0.022 |
hered | -0.022 |
nightmares | -0.021 |
ñ | -0.021 |
Token | he |
Token ID | 339 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
joy | +0.134 |
ez | +0.130 |
hered | +0.119 |
nightmares | +0.115 |
Charl | +0.111 |
Neg logit contributions | |
ires | -0.084 |
iring | -0.083 |
ate | -0.082 |
um | -0.081 |
lly | -0.068 |
Token | thanked |
Token ID | 26280 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
joy | +0.040 |
ez | +0.038 |
hered | +0.035 |
nightmares | +0.034 |
heaven | +0.033 |
Neg logit contributions | |
ires | -0.028 |
ate | -0.028 |
um | -0.027 |
iring | -0.027 |
ruce | -0.024 |
Token | rocket |
Token ID | 10701 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
joy | +0.084 |
ez | +0.082 |
hered | +0.077 |
smart | +0.069 |
liar | +0.068 |
Neg logit contributions | |
ires | -0.055 |
iring | -0.053 |
um | -0.051 |
ate | -0.051 |
ruce | -0.046 |
Token | 's |
Token ID | 338 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
joy | +0.133 |
ez | +0.126 |
hered | +0.113 |
Charl | +0.109 |
nightmares | +0.107 |
Neg logit contributions | |
ires | -0.082 |
ate | -0.082 |
iring | -0.081 |
um | -0.079 |
lly | -0.070 |
Token | friends |
Token ID | 2460 |
Feature activation | +3.85e-02 |
Pos logit contributions | |
joy | +0.345 |
ez | +0.336 |
hered | +0.312 |
Charl | +0.275 |
ñ | +0.272 |
Neg logit contributions | |
ires | -0.310 |
iring | -0.299 |
um | -0.294 |
ate | -0.290 |
ruce | -0.259 |
Token | were |
Token ID | 547 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
ate | +0.041 |
um | +0.040 |
lly | +0.040 |
iring | +0.039 |
ires | +0.038 |
Neg logit contributions | |
joy | -0.087 |
ez | -0.084 |
hered | -0.076 |
nightmares | -0.076 |
ñ | -0.075 |
Token | over |
Token ID | 625 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
joy | +0.071 |
ez | +0.069 |
hered | +0.063 |
nightmares | +0.059 |
moment | +0.058 |
Neg logit contributions | |
ires | -0.054 |
iring | -0.053 |
um | -0.053 |
ate | -0.053 |
ruce | -0.046 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ires | +0.149 |
iring | +0.147 |
um | +0.145 |
ate | +0.143 |
Inside | +0.131 |
Neg logit contributions | |
ez | -0.121 |
ñ | -0.105 |
hered | -0.102 |
Charl | -0.098 |
liar | -0.098 |
Token | ed |
Token ID | 276 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
joy | +1.265 |
failure | +1.234 |
nightmares | +1.223 |
liar | +1.211 |
heaven | +1.204 |
Neg logit contributions | |
um | -0.649 |
iring | -0.601 |
ires | -0.597 |
ate | -0.586 |
am | -0.534 |
Token | to |
Token ID | 284 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
ez | +0.188 |
joy | +0.177 |
hered | +0.166 |
Charl | +0.165 |
ñ | +0.164 |
Neg logit contributions | |
iring | -0.088 |
ires | -0.086 |
um | -0.083 |
ate | -0.080 |
up | -0.064 |
Token | have |
Token ID | 423 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
joy | +0.053 |
ez | +0.051 |
hered | +0.046 |
nightmares | +0.044 |
Charl | +0.043 |
Neg logit contributions | |
ires | -0.037 |
ate | -0.036 |
um | -0.035 |
iring | -0.033 |
ered | -0.031 |
Token | him |
Token ID | 683 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
joy | +0.303 |
ez | +0.291 |
hered | +0.275 |
oing | +0.269 |
moment | +0.266 |
Neg logit contributions | |
ate | -0.191 |
um | -0.172 |
ires | -0.161 |
lly | -0.158 |
iring | -0.156 |
Token | back |
Token ID | 736 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
ate | +0.023 |
um | +0.022 |
iring | +0.022 |
ires | +0.021 |
Inside | +0.020 |
Neg logit contributions | |
ez | -0.055 |
joy | -0.055 |
Charl | -0.051 |
hered | -0.051 |
liar | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
joy | +0.020 |
ez | +0.019 |
hered | +0.017 |
nightmares | +0.016 |
moment | +0.016 |
Neg logit contributions | |
ires | -0.018 |
iring | -0.017 |
ate | -0.017 |
um | -0.017 |
ruce | -0.015 |
Token | Cl |
Token ID | 2601 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
joy | +0.020 |
ez | +0.019 |
hered | +0.017 |
nightmares | +0.017 |
moment | +0.016 |
Neg logit contributions | |
ires | -0.018 |
ate | -0.017 |
um | -0.017 |
iring | -0.017 |
ruce | -0.015 |
Token | aire |
Token ID | 7626 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
ez | +0.256 |
joy | +0.243 |
nightmares | +0.224 |
breathing | +0.217 |
smart | +0.214 |
Neg logit contributions | |
um | -0.234 |
ate | -0.229 |
ires | -0.228 |
iring | -0.209 |
outed | -0.182 |
Token | was |
Token ID | 373 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
joy | +0.305 |
ez | +0.282 |
hered | +0.261 |
nightmares | +0.255 |
heaven | +0.247 |
Neg logit contributions | |
um | -0.207 |
ate | -0.206 |
ires | -0.204 |
iring | -0.195 |
lly | -0.175 |
Token | over |
Token ID | 625 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
um | +0.018 |
ate | +0.017 |
iring | +0.017 |
ires | +0.017 |
ruce | +0.014 |
Neg logit contributions | |
joy | -0.032 |
ez | -0.032 |
hered | -0.029 |
nightmares | -0.028 |
ñ | -0.027 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ires | +0.215 |
um | +0.215 |
iring | +0.214 |
ate | +0.208 |
Inside | +0.196 |
Neg logit contributions | |
ez | -0.055 |
ñ | -0.035 |
hered | -0.029 |
oscope | -0.026 |
liar | -0.026 |
Token | ed |
Token ID | 276 |
Feature activation | -7.71e-02 |
Pos logit contributions | |
nightmares | +1.218 |
joy | +1.208 |
attacking | +1.194 |
ez | +1.187 |
breathing | +1.162 |
Neg logit contributions | |
um | -0.756 |
ires | -0.667 |
ate | -0.664 |
iring | -0.652 |
ised | -0.521 |
Token | - |
Token ID | 532 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
ez | +0.182 |
joy | +0.172 |
hered | +0.160 |
nightmares | +0.155 |
ñ | +0.152 |
Neg logit contributions | |
um | -0.101 |
iring | -0.100 |
ires | -0.100 |
ate | -0.093 |
ised | -0.072 |
Token | her |
Token ID | 607 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
um | +0.052 |
ires | +0.052 |
ate | +0.049 |
iring | +0.049 |
Inside | +0.044 |
Neg logit contributions | |
ez | -0.084 |
joy | -0.077 |
hered | -0.073 |
nightmares | -0.070 |
ñ | -0.068 |
Token | mug |
Token ID | 25152 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
ez | +0.103 |
joy | +0.095 |
hered | +0.089 |
smart | +0.086 |
ñ | +0.083 |
Neg logit contributions | |
ires | -0.059 |
um | -0.058 |
iring | -0.054 |
ate | -0.052 |
ruce | -0.048 |
Token | had |
Token ID | 550 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
joy | +0.402 |
ez | +0.384 |
hered | +0.344 |
nightmares | +0.338 |
heaven | +0.328 |
Neg logit contributions | |
um | -0.211 |
ate | -0.201 |
ires | -0.199 |
iring | -0.196 |
lly | -0.172 |
Token | . |
Token ID | 13 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
joy | +0.436 |
ez | +0.405 |
hered | +0.365 |
nightmares | +0.350 |
Sleep | +0.342 |
Neg logit contributions | |
ate | -0.227 |
um | -0.225 |
ires | -0.223 |
iring | -0.216 |
lly | -0.202 |
Token | The |
Token ID | 383 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
joy | +0.074 |
ez | +0.072 |
hered | +0.066 |
nightmares | +0.063 |
moment | +0.061 |
Neg logit contributions | |
ires | -0.055 |
ate | -0.052 |
um | -0.052 |
iring | -0.051 |
ruce | -0.045 |
Token | princess |
Token ID | 21752 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
ez | +0.066 |
joy | +0.065 |
hered | +0.059 |
nightmares | +0.054 |
Charl | +0.052 |
Neg logit contributions | |
ires | -0.051 |
um | -0.048 |
ate | -0.048 |
iring | -0.047 |
ruce | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
joy | +0.411 |
ez | +0.386 |
hered | +0.356 |
nightmares | +0.342 |
heaven | +0.328 |
Neg logit contributions | |
ires | -0.271 |
ate | -0.268 |
um | -0.261 |
iring | -0.259 |
ruce | -0.223 |
Token | over |
Token ID | 625 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
ires | +0.017 |
ate | +0.016 |
iring | +0.016 |
um | +0.016 |
ruce | +0.014 |
Neg logit contributions | |
joy | -0.029 |
ez | -0.028 |
hered | -0.026 |
nightmares | -0.025 |
moment | -0.024 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ires | +0.188 |
iring | +0.186 |
um | +0.186 |
ate | +0.180 |
Inside | +0.167 |
Neg logit contributions | |
ez | -0.076 |
ñ | -0.053 |
hered | -0.051 |
nightmares | -0.048 |
Charl | -0.046 |
Token | ed |
Token ID | 276 |
Feature activation | -7.90e-02 |
Pos logit contributions | |
joy | +1.189 |
nightmares | +1.186 |
heaven | +1.131 |
failure | +1.131 |
ez | +1.127 |
Neg logit contributions | |
um | -0.780 |
ires | -0.722 |
ate | -0.712 |
iring | -0.695 |
am | -0.559 |
Token | ! |
Token ID | 0 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
ez | +0.180 |
joy | +0.172 |
hered | +0.158 |
nightmares | +0.157 |
ñ | +0.149 |
Neg logit contributions | |
ires | -0.107 |
um | -0.106 |
iring | -0.105 |
ate | -0.101 |
Inside | -0.080 |
Token | She |
Token ID | 1375 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
joy | +0.096 |
ez | +0.093 |
hered | +0.084 |
nightmares | +0.081 |
moment | +0.078 |
Neg logit contributions | |
ires | -0.073 |
ate | -0.069 |
um | -0.069 |
iring | -0.068 |
ruce | -0.059 |
Token | was |
Token ID | 373 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
ires | +0.061 |
iring | +0.057 |
ate | +0.056 |
um | +0.055 |
ruce | +0.051 |
Neg logit contributions | |
joy | -0.064 |
ez | -0.063 |
hered | -0.056 |
moment | -0.052 |
nightmares | -0.052 |
Token | so |
Token ID | 523 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
ires | +0.018 |
ate | +0.017 |
iring | +0.017 |
um | +0.017 |
ruce | +0.014 |
Neg logit contributions | |
joy | -0.030 |
ez | -0.029 |
hered | -0.027 |
nightmares | -0.026 |
moment | -0.025 |
Token | rock |
Token ID | 3881 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
ires | +0.223 |
ate | +0.221 |
um | +0.219 |
iring | +0.205 |
ruce | +0.182 |
Neg logit contributions | |
joy | -0.298 |
ez | -0.296 |
hered | -0.264 |
nightmares | -0.251 |
Charl | -0.245 |
Token | . |
Token ID | 13 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
ires | +0.128 |
ate | +0.123 |
iring | +0.121 |
um | +0.120 |
ruce | +0.106 |
Neg logit contributions | |
joy | -0.186 |
ez | -0.176 |
hered | -0.159 |
nightmares | -0.152 |
Charl | -0.148 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.35e-03 |
Pos logit contributions | |
joy | +0.053 |
ez | +0.048 |
hered | +0.043 |
moment | +0.041 |
heaven | +0.040 |
Neg logit contributions | |
ires | -0.059 |
iring | -0.057 |
ate | -0.057 |
um | -0.054 |
ruce | -0.051 |
Token | was |
Token ID | 373 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
ires | +0.004 |
iring | +0.004 |
ate | +0.004 |
um | +0.004 |
ruce | +0.004 |
Neg logit contributions | |
joy | -0.004 |
ez | -0.004 |
hered | -0.004 |
moment | -0.003 |
knowing | -0.003 |
Token | over |
Token ID | 625 |
Feature activation | +7.54e-02 |
Pos logit contributions | |
ires | +0.025 |
ate | +0.023 |
iring | +0.023 |
um | +0.023 |
ruce | +0.021 |
Neg logit contributions | |
joy | -0.036 |
ez | -0.034 |
hered | -0.031 |
nightmares | -0.029 |
moment | -0.029 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ires | +0.225 |
iring | +0.222 |
um | +0.222 |
ate | +0.219 |
Inside | +0.208 |
Neg logit contributions | |
ez | -0.050 |
hered | -0.022 |
ching | -0.018 |
ñ | -0.017 |
nightmares | -0.017 |
Token | ed |
Token ID | 276 |
Feature activation | -7.63e-02 |
Pos logit contributions | |
joy | +1.198 |
failure | +1.155 |
nightmares | +1.154 |
ez | +1.140 |
attacking | +1.135 |
Neg logit contributions | |
um | -0.754 |
ires | -0.709 |
ate | -0.704 |
iring | -0.684 |
Inside | -0.555 |
Token | and |
Token ID | 290 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
ez | +0.170 |
joy | +0.162 |
hered | +0.150 |
nightmares | +0.145 |
liar | +0.138 |
Neg logit contributions | |
ires | -0.108 |
iring | -0.106 |
um | -0.105 |
ate | -0.103 |
ruce | -0.083 |
Token | knew |
Token ID | 2993 |
Feature activation | -9.81e-02 |
Pos logit contributions | |
joy | +0.112 |
ez | +0.105 |
hered | +0.097 |
knowing | +0.093 |
moment | +0.090 |
Neg logit contributions | |
ires | -0.105 |
iring | -0.097 |
ate | -0.097 |
um | -0.095 |
ruce | -0.088 |
Token | it |
Token ID | 340 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
joy | +0.213 |
ez | +0.205 |
hered | +0.187 |
nightmares | +0.178 |
moment | +0.175 |
Neg logit contributions | |
ires | -0.145 |
ate | -0.136 |
iring | -0.136 |
um | -0.135 |
ruce | -0.119 |
Token | was |
Token ID | 373 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
joy | +0.057 |
ez | +0.055 |
hered | +0.050 |
nightmares | +0.047 |
heaven | +0.047 |
Neg logit contributions | |
ate | -0.027 |
ires | -0.026 |
iring | -0.026 |
um | -0.025 |
ruce | -0.022 |
Token | 't |
Token ID | 470 |
Feature activation | +8.69e-03 |
Pos logit contributions | |
ires | +0.135 |
ate | +0.128 |
um | +0.128 |
iring | +0.127 |
ruce | +0.113 |
Neg logit contributions | |
joy | -0.182 |
ez | -0.173 |
hered | -0.156 |
nightmares | -0.149 |
moment | -0.146 |
Token | help |
Token ID | 1037 |
Feature activation | -8.46e-02 |
Pos logit contributions | |
ires | +0.013 |
ate | +0.013 |
um | +0.013 |
iring | +0.013 |
ruce | +0.011 |
Neg logit contributions | |
joy | -0.018 |
ez | -0.017 |
hered | -0.016 |
nightmares | -0.015 |
moment | -0.015 |
Token | but |
Token ID | 475 |
Feature activation | -5.75e-02 |
Pos logit contributions | |
joy | +0.166 |
ez | +0.160 |
hered | +0.143 |
nightmares | +0.133 |
heaven | +0.130 |
Neg logit contributions | |
ires | -0.145 |
iring | -0.140 |
ate | -0.137 |
um | -0.136 |
ruce | -0.122 |
Token | feel |
Token ID | 1254 |
Feature activation | -8.82e-02 |
Pos logit contributions | |
joy | +0.109 |
ez | +0.105 |
hered | +0.092 |
nightmares | +0.087 |
heaven | +0.082 |
Neg logit contributions | |
ires | -0.101 |
iring | -0.097 |
ate | -0.096 |
um | -0.096 |
ruce | -0.086 |
Token | over |
Token ID | 625 |
Feature activation | +7.75e-02 |
Pos logit contributions | |
joy | +0.187 |
ez | +0.183 |
hered | +0.174 |
ñ | +0.172 |
rav | +0.160 |
Neg logit contributions | |
ires | -0.122 |
iring | -0.120 |
ate | -0.116 |
um | -0.109 |
ised | -0.101 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ires | +0.150 |
iring | +0.146 |
ate | +0.141 |
um | +0.140 |
ised | +0.123 |
Neg logit contributions | |
ez | -0.124 |
ñ | -0.111 |
hered | -0.111 |
Charl | -0.104 |
tale | -0.104 |
Token | ed |
Token ID | 276 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
joy | +1.231 |
nightmares | +1.182 |
heaven | +1.163 |
failure | +1.149 |
foe | +1.140 |
Neg logit contributions | |
um | -0.748 |
ires | -0.699 |
iring | -0.685 |
ate | -0.681 |
am | -0.586 |
Token | . |
Token ID | 13 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
ez | +0.175 |
joy | +0.170 |
hered | +0.159 |
nightmares | +0.150 |
Charl | +0.148 |
Neg logit contributions | |
ires | -0.108 |
iring | -0.107 |
um | -0.104 |
ate | -0.102 |
ruce | -0.079 |
Token | He |
Token ID | 679 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
joy | +0.071 |
ez | +0.068 |
hered | +0.061 |
nightmares | +0.058 |
moment | +0.058 |
Neg logit contributions | |
ires | -0.052 |
ate | -0.049 |
iring | -0.048 |
um | -0.048 |
ruce | -0.043 |
Token | felt |
Token ID | 2936 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
ires | +0.091 |
ate | +0.087 |
um | +0.086 |
iring | +0.086 |
ruce | +0.077 |
Neg logit contributions | |
joy | -0.111 |
ez | -0.106 |
hered | -0.096 |
nightmares | -0.091 |
moment | -0.089 |
Token | so |
Token ID | 523 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
joy | +0.140 |
ez | +0.136 |
hered | +0.124 |
nightmares | +0.117 |
ñ | +0.115 |
Neg logit contributions | |
ires | -0.100 |
ate | -0.096 |
iring | -0.095 |
um | -0.094 |
ruce | -0.083 |
Token | win |
Token ID | 1592 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
ez | +0.090 |
joy | +0.083 |
hered | +0.076 |
nightmares | +0.072 |
Charl | +0.072 |
Neg logit contributions | |
ires | -0.045 |
ate | -0.043 |
um | -0.043 |
iring | -0.038 |
her | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
joy | +0.112 |
ez | +0.105 |
nightmares | +0.095 |
hered | +0.093 |
Charl | +0.091 |
Neg logit contributions | |
um | -0.062 |
ate | -0.062 |
ires | -0.060 |
iring | -0.056 |
ered | -0.053 |
Token | Jane |
Token ID | 12091 |
Feature activation | +6.61e-02 |
Pos logit contributions | |
joy | +0.069 |
ez | +0.068 |
hered | +0.062 |
nightmares | +0.060 |
moment | +0.056 |
Neg logit contributions | |
ires | -0.054 |
um | -0.051 |
ate | -0.050 |
iring | -0.049 |
Inside | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
um | +0.091 |
ate | +0.090 |
ires | +0.086 |
iring | +0.083 |
lly | +0.077 |
Neg logit contributions | |
joy | -0.149 |
ez | -0.134 |
hered | -0.127 |
nightmares | -0.124 |
heaven | -0.122 |
Token | over |
Token ID | 625 |
Feature activation | +7.75e-02 |
Pos logit contributions | |
um | +0.018 |
ate | +0.018 |
iring | +0.018 |
ires | +0.017 |
her | +0.014 |
Neg logit contributions | |
joy | -0.037 |
ez | -0.037 |
hered | -0.034 |
nightmares | -0.033 |
disturbances | -0.032 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
um | +0.214 |
ires | +0.214 |
iring | +0.213 |
ate | +0.208 |
Inside | +0.192 |
Neg logit contributions | |
ez | -0.062 |
ñ | -0.040 |
hered | -0.035 |
oscope | -0.032 |
villains | -0.032 |
Token | ed |
Token ID | 276 |
Feature activation | -7.72e-02 |
Pos logit contributions | |
joy | +1.229 |
nightmares | +1.223 |
attacking | +1.191 |
ez | +1.171 |
failure | +1.171 |
Neg logit contributions | |
um | -0.739 |
ate | -0.651 |
ires | -0.650 |
iring | -0.639 |
outed | -0.503 |
Token | ! |
Token ID | 0 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
ez | +0.182 |
joy | +0.172 |
hered | +0.159 |
nightmares | +0.157 |
ñ | +0.151 |
Neg logit contributions | |
um | -0.102 |
ires | -0.101 |
iring | -0.101 |
ate | -0.095 |
ised | -0.074 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
ez | +0.096 |
joy | +0.095 |
hered | +0.087 |
nightmares | +0.085 |
heaven | +0.079 |
Neg logit contributions | |
ires | -0.066 |
um | -0.063 |
ate | -0.062 |
iring | -0.060 |
Inside | -0.054 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
ate | +0.035 |
um | +0.033 |
ires | +0.032 |
iring | +0.031 |
Inside | +0.030 |
Neg logit contributions | |
ez | -0.065 |
joy | -0.064 |
nightmares | -0.060 |
failure | -0.058 |
liar | -0.058 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
joy | +0.224 |
ez | +0.197 |
moment | +0.191 |
hered | +0.181 |
nightmares | +0.177 |
Neg logit contributions | |
ires | -0.167 |
ate | -0.166 |
iring | -0.157 |
um | -0.153 |
ruce | -0.142 |
Token | S |
Token ID | 50 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
joy | +0.022 |
ez | +0.021 |
hered | +0.019 |
moment | +0.017 |
nightmares | +0.017 |
Neg logit contributions | |
ires | -0.021 |
iring | -0.020 |
ate | -0.020 |
um | -0.019 |
ruce | -0.018 |
Token | ally |
Token ID | 453 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
joy | +0.137 |
ez | +0.132 |
hered | +0.124 |
moment | +0.117 |
nightmares | +0.116 |
Neg logit contributions | |
ires | -0.054 |
ate | -0.049 |
iring | -0.049 |
um | -0.046 |
ruce | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
um | +0.295 |
ate | +0.292 |
ires | +0.286 |
iring | +0.272 |
lly | +0.243 |
Neg logit contributions | |
joy | -0.436 |
ez | -0.403 |
hered | -0.373 |
nightmares | -0.366 |
heaven | -0.353 |
Token | so |
Token ID | 523 |
Feature activation | +5.75e-02 |
Pos logit contributions | |
um | +0.016 |
ate | +0.016 |
iring | +0.015 |
ires | +0.015 |
her | +0.013 |
Neg logit contributions | |
joy | -0.036 |
ez | -0.035 |
hered | -0.033 |
nightmares | -0.032 |
ñ | -0.031 |
Token | over |
Token ID | 625 |
Feature activation | +7.84e-02 |
Pos logit contributions | |
ires | +0.101 |
um | +0.097 |
iring | +0.097 |
ate | +0.097 |
ruce | +0.085 |
Neg logit contributions | |
joy | -0.109 |
ez | -0.104 |
hered | -0.094 |
nightmares | -0.090 |
moment | -0.085 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
um | +0.112 |
ires | +0.112 |
iring | +0.109 |
ate | +0.105 |
Inside | +0.092 |
Neg logit contributions | |
ez | -0.167 |
hered | -0.144 |
ñ | -0.143 |
liar | -0.136 |
nightmares | -0.135 |
Token | ed |
Token ID | 276 |
Feature activation | -7.62e-02 |
Pos logit contributions | |
joy | +1.165 |
nightmares | +1.120 |
ez | +1.096 |
heaven | +1.067 |
failure | +1.059 |
Neg logit contributions | |
um | -0.855 |
ires | -0.801 |
ate | -0.770 |
iring | -0.769 |
ised | -0.642 |
Token | with |
Token ID | 351 |
Feature activation | +7.66e-02 |
Pos logit contributions | |
ez | +0.172 |
joy | +0.171 |
hered | +0.154 |
nightmares | +0.149 |
ñ | +0.142 |
Neg logit contributions | |
ires | -0.104 |
um | -0.103 |
iring | -0.101 |
ate | -0.097 |
ruce | -0.079 |
Token | her |
Token ID | 607 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
ires | +0.120 |
iring | +0.112 |
ate | +0.112 |
um | +0.111 |
ruce | +0.099 |
Neg logit contributions | |
joy | -0.160 |
ez | -0.155 |
hered | -0.142 |
nightmares | -0.133 |
moment | -0.131 |
Token | present |
Token ID | 1944 |
Feature activation | +5.20e-03 |
Pos logit contributions | |
ez | +0.105 |
joy | +0.099 |
hered | +0.094 |
nightmares | +0.084 |
Charl | +0.084 |
Neg logit contributions | |
ires | -0.056 |
um | -0.055 |
iring | -0.051 |
ate | -0.049 |
ruce | -0.042 |
Token | that |
Token ID | 326 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
um | +0.007 |
iring | +0.006 |
ires | +0.006 |
ate | +0.006 |
Inside | +0.005 |
Neg logit contributions | |
ez | -0.013 |
joy | -0.013 |
nightmares | -0.011 |
hered | -0.010 |
heaven | -0.010 |
Token | park |
Token ID | 3952 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
joy | +0.083 |
ez | +0.079 |
hered | +0.073 |
nightmares | +0.070 |
moment | +0.069 |
Neg logit contributions | |
ires | -0.050 |
iring | -0.047 |
ate | -0.046 |
um | -0.046 |
ruce | -0.040 |
Token | ! |
Token ID | 0 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
ires | +0.145 |
ate | +0.144 |
um | +0.143 |
iring | +0.142 |
ruce | +0.119 |
Neg logit contributions | |
ez | -0.240 |
joy | -0.239 |
hered | -0.213 |
nightmares | -0.208 |
Charl | -0.202 |
Token | He |
Token ID | 679 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
joy | +0.085 |
ez | +0.082 |
hered | +0.074 |
nightmares | +0.070 |
moment | +0.069 |
Neg logit contributions | |
ires | -0.065 |
ate | -0.061 |
iring | -0.061 |
um | -0.060 |
ruce | -0.054 |
Token | was |
Token ID | 373 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
ires | +0.103 |
iring | +0.095 |
ate | +0.093 |
um | +0.092 |
ruce | +0.085 |
Neg logit contributions | |
joy | -0.103 |
ez | -0.100 |
hered | -0.089 |
moment | -0.083 |
nightmares | -0.082 |
Token | over |
Token ID | 625 |
Feature activation | +7.75e-02 |
Pos logit contributions | |
ires | +0.021 |
iring | +0.020 |
ate | +0.020 |
um | +0.019 |
ruce | +0.017 |
Neg logit contributions | |
joy | -0.036 |
ez | -0.035 |
hered | -0.032 |
nightmares | -0.031 |
moment | -0.030 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
ires | +0.229 |
iring | +0.227 |
um | +0.226 |
ate | +0.224 |
Inside | +0.213 |
Neg logit contributions | |
ez | -0.046 |
hered | -0.023 |
nightmares | -0.017 |
ñ | -0.017 |
Charl | -0.017 |
Token | ed |
Token ID | 276 |
Feature activation | -7.72e-02 |
Pos logit contributions | |
joy | +1.164 |
nightmares | +1.137 |
ez | +1.108 |
heaven | +1.103 |
failure | +1.101 |
Neg logit contributions | |
um | -0.778 |
ires | -0.734 |
iring | -0.709 |
ate | -0.708 |
am | -0.622 |
Token | . |
Token ID | 13 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
ez | +0.168 |
joy | +0.161 |
hered | +0.148 |
nightmares | +0.147 |
heaven | +0.140 |
Neg logit contributions | |
ires | -0.109 |
iring | -0.108 |
um | -0.107 |
ate | -0.104 |
ruce | -0.084 |
Token | He |
Token ID | 679 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
joy | +0.059 |
ez | +0.055 |
hered | +0.050 |
moment | +0.047 |
nightmares | +0.047 |
Neg logit contributions | |
ires | -0.047 |
ate | -0.045 |
iring | -0.044 |
um | -0.044 |
ruce | -0.040 |
Token | thanked |
Token ID | 26280 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
ires | +0.106 |
iring | +0.098 |
ate | +0.096 |
um | +0.093 |
ruce | +0.087 |
Neg logit contributions | |
joy | -0.100 |
ez | -0.097 |
hered | -0.087 |
moment | -0.079 |
knowing | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
ez | +0.039 |
joy | +0.038 |
hered | +0.035 |
nightmares | +0.033 |
ching | +0.032 |
Neg logit contributions | |
ires | -0.022 |
um | -0.022 |
ate | -0.021 |
iring | -0.020 |
Inside | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
joy | +0.156 |
ez | +0.148 |
hered | +0.135 |
nightmares | +0.123 |
moment | +0.122 |
Neg logit contributions | |
ires | -0.151 |
iring | -0.143 |
ate | -0.142 |
um | -0.141 |
ruce | -0.130 |
Token | L |
Token ID | 43 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
joy | +0.024 |
ez | +0.023 |
hered | +0.021 |
nightmares | +0.021 |
breathing | +0.020 |
Neg logit contributions | |
ires | -0.015 |
um | -0.015 |
ate | -0.015 |
iring | -0.014 |
ruce | -0.013 |
Token | illy |
Token ID | 6548 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
ires | +0.015 |
ate | +0.014 |
um | +0.013 |
iring | +0.013 |
ruce | +0.011 |
Neg logit contributions | |
joy | -0.030 |
ez | -0.028 |
hered | -0.027 |
nightmares | -0.026 |
moment | -0.025 |
Token | was |
Token ID | 373 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
um | +0.165 |
ate | +0.165 |
ires | +0.165 |
iring | +0.157 |
ruce | +0.140 |
Neg logit contributions | |
joy | -0.224 |
ez | -0.213 |
hered | -0.193 |
nightmares | -0.185 |
moment | -0.179 |
Token | over |
Token ID | 625 |
Feature activation | +7.91e-02 |
Pos logit contributions | |
ate | +0.023 |
ires | +0.023 |
um | +0.023 |
iring | +0.022 |
her | +0.019 |
Neg logit contributions | |
joy | -0.039 |
ez | -0.039 |
hered | -0.036 |
nightmares | -0.034 |
moment | -0.033 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
ires | +0.207 |
um | +0.204 |
iring | +0.202 |
ate | +0.199 |
Inside | +0.190 |
Neg logit contributions | |
ez | -0.072 |
ñ | -0.047 |
hered | -0.047 |
ching | -0.041 |
oing | -0.041 |
Token | ed |
Token ID | 276 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
joy | +1.196 |
nightmares | +1.144 |
ez | +1.123 |
heaven | +1.118 |
failure | +1.118 |
Neg logit contributions | |
um | -0.770 |
ires | -0.723 |
ate | -0.710 |
iring | -0.706 |
outed | -0.582 |
Token | and |
Token ID | 290 |
Feature activation | -6.06e-02 |
Pos logit contributions | |
ez | +0.176 |
joy | +0.170 |
hered | +0.156 |
nightmares | +0.151 |
ñ | +0.146 |
Neg logit contributions | |
ires | -0.105 |
iring | -0.104 |
um | -0.103 |
ate | -0.098 |
ised | -0.078 |
Token | felt |
Token ID | 2936 |
Feature activation | -6.55e-02 |
Pos logit contributions | |
joy | +0.129 |
ez | +0.124 |
hered | +0.113 |
nightmares | +0.107 |
moment | +0.105 |
Neg logit contributions | |
ires | -0.092 |
ate | -0.087 |
iring | -0.087 |
um | -0.086 |
ruce | -0.076 |
Token | so |
Token ID | 523 |
Feature activation | +6.11e-02 |
Pos logit contributions | |
joy | +0.141 |
ez | +0.139 |
hered | +0.127 |
nightmares | +0.120 |
moment | +0.117 |
Neg logit contributions | |
ires | -0.091 |
ate | -0.090 |
iring | -0.088 |
um | -0.087 |
ruce | -0.076 |
Token | lucky |
Token ID | 9670 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
ires | +0.116 |
ate | +0.110 |
iring | +0.110 |
um | +0.109 |
ruce | +0.101 |
Neg logit contributions | |
joy | -0.107 |
ez | -0.101 |
hered | -0.090 |
moment | -0.084 |
nightmares | -0.084 |
Token | beach |
Token ID | 10481 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
joy | +0.077 |
moment | +0.069 |
ez | +0.062 |
payoff | +0.062 |
hered | +0.061 |
Neg logit contributions | |
ires | -0.050 |
iring | -0.045 |
um | -0.042 |
ate | -0.041 |
outed | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
ires | +0.331 |
ate | +0.317 |
iring | +0.313 |
um | +0.312 |
ruce | +0.271 |
Neg logit contributions | |
joy | -0.476 |
ez | -0.473 |
hered | -0.426 |
nightmares | -0.405 |
moment | -0.387 |
Token | Maggie |
Token ID | 33014 |
Feature activation | +4.95e-03 |
Pos logit contributions | |
ires | +0.025 |
ate | +0.023 |
iring | +0.023 |
um | +0.023 |
ruce | +0.021 |
Neg logit contributions | |
joy | -0.029 |
ez | -0.027 |
hered | -0.025 |
moment | -0.024 |
nightmares | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
ires | +0.008 |
iring | +0.007 |
ate | +0.007 |
um | +0.007 |
ruce | +0.006 |
Neg logit contributions | |
joy | -0.010 |
ez | -0.010 |
hered | -0.009 |
nightmares | -0.009 |
moment | -0.008 |
Token | over |
Token ID | 625 |
Feature activation | +7.98e-02 |
Pos logit contributions | |
ires | +0.029 |
ate | +0.027 |
iring | +0.027 |
um | +0.026 |
ruce | +0.024 |
Neg logit contributions | |
joy | -0.035 |
ez | -0.033 |
hered | -0.030 |
nightmares | -0.028 |
moment | -0.028 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
ires | +0.243 |
um | +0.239 |
iring | +0.238 |
ate | +0.238 |
Inside | +0.224 |
Neg logit contributions | |
ez | -0.047 |
hered | -0.024 |
ñ | -0.018 |
nightmares | -0.016 |
ching | -0.013 |
Token | ed |
Token ID | 276 |
Feature activation | -7.77e-02 |
Pos logit contributions | |
joy | +1.202 |
ez | +1.145 |
nightmares | +1.141 |
failure | +1.138 |
attacking | +1.114 |
Neg logit contributions | |
um | -0.790 |
ires | -0.727 |
iring | -0.716 |
ate | -0.692 |
am | -0.597 |
Token | . |
Token ID | 13 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
ez | +0.183 |
joy | +0.170 |
hered | +0.162 |
nightmares | +0.154 |
ñ | +0.152 |
Neg logit contributions | |
ires | -0.101 |
iring | -0.100 |
um | -0.099 |
ate | -0.095 |
ruce | -0.075 |
Token | She |
Token ID | 1375 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
joy | +0.062 |
ez | +0.055 |
hered | +0.049 |
moment | +0.049 |
nightmares | +0.047 |
Neg logit contributions | |
ires | -0.054 |
ate | -0.052 |
iring | -0.052 |
um | -0.050 |
ruce | -0.048 |
Token | jumped |
Token ID | 11687 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
ires | +0.072 |
iring | +0.067 |
ate | +0.062 |
um | +0.060 |
ruce | +0.056 |
Neg logit contributions | |
joy | -0.076 |
ez | -0.075 |
hered | -0.067 |
knowing | -0.063 |
moment | -0.062 |
Token | in |
Token ID | 287 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
joy | +0.024 |
ez | +0.023 |
hered | +0.022 |
moment | +0.020 |
knowing | +0.020 |
Neg logit contributions | |
ires | -0.015 |
iring | -0.014 |
ate | -0.013 |
um | -0.013 |
ruce | -0.012 |
Token | mountain |
Token ID | 8598 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
joy | +0.090 |
ez | +0.090 |
hered | +0.083 |
nightmares | +0.077 |
heaven | +0.075 |
Neg logit contributions | |
ires | -0.046 |
um | -0.044 |
iring | -0.043 |
ate | -0.042 |
ruce | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
ires | +0.021 |
ate | +0.020 |
iring | +0.020 |
um | +0.020 |
ruce | +0.017 |
Neg logit contributions | |
joy | -0.031 |
ez | -0.031 |
hered | -0.028 |
nightmares | -0.027 |
Charl | -0.026 |
Token | He |
Token ID | 679 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
joy | +0.071 |
ez | +0.068 |
hered | +0.062 |
nightmares | +0.059 |
moment | +0.058 |
Neg logit contributions | |
ires | -0.051 |
iring | -0.048 |
ate | -0.047 |
um | -0.047 |
ruce | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
ires | +0.095 |
iring | +0.089 |
um | +0.087 |
ate | +0.087 |
ruce | +0.079 |
Neg logit contributions | |
joy | -0.106 |
ez | -0.104 |
hered | -0.092 |
nightmares | -0.087 |
moment | -0.087 |
Token | over |
Token ID | 625 |
Feature activation | +7.84e-02 |
Pos logit contributions | |
ires | +0.023 |
ate | +0.023 |
iring | +0.023 |
um | +0.022 |
ruce | +0.019 |
Neg logit contributions | |
joy | -0.039 |
ez | -0.038 |
hered | -0.036 |
nightmares | -0.034 |
moment | -0.033 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
ires | +0.188 |
iring | +0.186 |
um | +0.185 |
ate | +0.183 |
ised | +0.169 |
Neg logit contributions | |
ez | -0.089 |
hered | -0.072 |
ñ | -0.066 |
Charl | -0.065 |
tale | -0.063 |
Token | ed |
Token ID | 276 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
joy | +1.182 |
nightmares | +1.149 |
heaven | +1.148 |
ez | +1.122 |
failure | +1.101 |
Neg logit contributions | |
um | -0.753 |
ires | -0.722 |
iring | -0.709 |
ate | -0.691 |
am | -0.609 |
Token | . |
Token ID | 13 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
ez | +0.176 |
joy | +0.169 |
hered | +0.158 |
nightmares | +0.153 |
heaven | +0.149 |
Neg logit contributions | |
iring | -0.106 |
ires | -0.106 |
um | -0.102 |
ate | -0.101 |
ised | -0.081 |
Token | He |
Token ID | 679 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
joy | +0.068 |
ez | +0.065 |
hered | +0.059 |
nightmares | +0.056 |
moment | +0.055 |
Neg logit contributions | |
ires | -0.048 |
iring | -0.045 |
ate | -0.045 |
um | -0.045 |
ruce | -0.040 |
Token | looked |
Token ID | 3114 |
Feature activation | +6.33e-02 |
Pos logit contributions | |
ires | +0.100 |
iring | +0.094 |
ate | +0.091 |
um | +0.090 |
ruce | +0.083 |
Neg logit contributions | |
joy | -0.102 |
ez | -0.101 |
hered | -0.088 |
moment | -0.083 |
nightmares | -0.082 |
Token | back |
Token ID | 736 |
Feature activation | -9.89e-02 |
Pos logit contributions | |
ires | +0.091 |
ate | +0.087 |
um | +0.084 |
iring | +0.083 |
ruce | +0.075 |
Neg logit contributions | |
joy | -0.146 |
ez | -0.130 |
hered | -0.119 |
moment | -0.118 |
knowing | -0.117 |
Token | , |
Token ID | 11 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
ez | +0.192 |
joy | +0.192 |
hered | +0.170 |
tale | +0.162 |
oing | +0.161 |
Neg logit contributions | |
um | -0.082 |
ires | -0.074 |
ate | -0.073 |
iring | -0.072 |
Inside | -0.068 |
Token | she |
Token ID | 673 |
Feature activation | -7.80e-02 |
Pos logit contributions | |
ires | +0.027 |
ate | +0.026 |
iring | +0.026 |
um | +0.026 |
ruce | +0.023 |
Neg logit contributions | |
joy | -0.038 |
ez | -0.038 |
hered | -0.034 |
nightmares | -0.031 |
tale | -0.030 |
Token | decided |
Token ID | 3066 |
Feature activation | +9.25e-02 |
Pos logit contributions | |
ez | +0.134 |
joy | +0.132 |
hered | +0.115 |
nightmares | +0.113 |
moment | +0.104 |
Neg logit contributions | |
ires | -0.157 |
iring | -0.146 |
ate | -0.141 |
um | -0.139 |
ruce | -0.129 |
Token | to |
Token ID | 284 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
ires | +0.123 |
ate | +0.118 |
um | +0.117 |
iring | +0.117 |
ruce | +0.104 |
Neg logit contributions | |
ez | -0.208 |
joy | -0.203 |
hered | -0.187 |
nightmares | -0.169 |
moment | -0.167 |
Token | tidy |
Token ID | 43044 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
ez | +0.039 |
heaven | +0.037 |
hered | +0.037 |
joy | +0.036 |
nightmares | +0.036 |
Neg logit contributions | |
ires | -0.042 |
iring | -0.040 |
um | -0.036 |
outed | -0.035 |
ate | -0.034 |
Token | up |
Token ID | 510 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
ires | +0.304 |
iring | +0.288 |
um | +0.286 |
ate | +0.269 |
ruce | +0.247 |
Neg logit contributions | |
joy | -0.372 |
ez | -0.354 |
nightmares | -0.314 |
hered | -0.311 |
moment | -0.308 |
Token | her |
Token ID | 607 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
ires | +0.386 |
ate | +0.371 |
iring | +0.358 |
um | +0.354 |
ruce | +0.316 |
Neg logit contributions | |
joy | -0.553 |
ez | -0.538 |
hered | -0.497 |
nightmares | -0.457 |
moment | -0.447 |
Token | room |
Token ID | 2119 |
Feature activation | +9.03e-02 |
Pos logit contributions | |
joy | +0.092 |
ez | +0.087 |
nightmares | +0.080 |
hered | +0.079 |
moment | +0.078 |
Neg logit contributions | |
ires | -0.061 |
um | -0.057 |
ate | -0.057 |
iring | -0.057 |
ruce | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
ires | +0.148 |
ate | +0.141 |
iring | +0.140 |
um | +0.139 |
ruce | +0.124 |
Neg logit contributions | |
joy | -0.180 |
ez | -0.174 |
hered | -0.157 |
nightmares | -0.147 |
moment | -0.144 |
Token | She |
Token ID | 1375 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
joy | +0.043 |
ez | +0.039 |
hered | +0.036 |
nightmares | +0.034 |
moment | +0.034 |
Neg logit contributions | |
ires | -0.039 |
iring | -0.037 |
ate | -0.037 |
um | -0.036 |
ruce | -0.033 |
Token | put |
Token ID | 1234 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
ires | +0.093 |
iring | +0.088 |
um | +0.078 |
ate | +0.073 |
ruce | +0.066 |
Neg logit contributions | |
ez | -0.079 |
joy | -0.072 |
moment | -0.066 |
nightmares | -0.065 |
hered | -0.064 |
Token | know |
Token ID | 760 |
Feature activation | -6.97e-03 |
Pos logit contributions | |
ires | +0.019 |
iring | +0.018 |
ate | +0.018 |
um | +0.018 |
ruce | +0.016 |
Neg logit contributions | |
joy | -0.019 |
ez | -0.019 |
hered | -0.017 |
nightmares | -0.016 |
moment | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
joy | +0.015 |
ez | +0.015 |
hered | +0.014 |
moment | +0.012 |
nightmares | +0.012 |
Neg logit contributions | |
ires | -0.011 |
iring | -0.010 |
ate | -0.010 |
um | -0.009 |
ruce | -0.008 |
Token | It |
Token ID | 632 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
joy | +0.046 |
ez | +0.044 |
hered | +0.038 |
nightmares | +0.036 |
moment | +0.035 |
Neg logit contributions | |
ires | -0.058 |
um | -0.056 |
iring | -0.056 |
ate | -0.056 |
ruce | -0.053 |
Token | looks |
Token ID | 3073 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
ires | +0.043 |
ate | +0.041 |
iring | +0.040 |
um | +0.040 |
ruce | +0.037 |
Neg logit contributions | |
joy | -0.038 |
ez | -0.035 |
hered | -0.031 |
nightmares | -0.028 |
moment | -0.028 |
Token | really |
Token ID | 1107 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ires | +0.050 |
um | +0.048 |
ate | +0.048 |
iring | +0.047 |
ruce | +0.042 |
Neg logit contributions | |
joy | -0.069 |
ez | -0.066 |
hered | -0.059 |
nightmares | -0.058 |
moment | -0.056 |
Token | unusual |
Token ID | 8468 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
um | +0.235 |
iring | +0.233 |
ires | +0.227 |
ate | +0.218 |
ruce | +0.207 |
Neg logit contributions | |
joy | -0.270 |
ez | -0.245 |
nightmares | -0.233 |
breathing | -0.224 |
moment | -0.223 |
Token | !" |
Token ID | 2474 |
Feature activation | +7.12e-02 |
Pos logit contributions | |
ires | +0.394 |
iring | +0.370 |
ate | +0.369 |
um | +0.362 |
ruce | +0.318 |
Neg logit contributions | |
joy | -0.513 |
ez | -0.495 |
hered | -0.458 |
nightmares | -0.415 |
moment | -0.411 |
Token |
|
Token ID | 198 |
Feature activation | -7.58e-03 |
Pos logit contributions | |
ires | +0.102 |
um | +0.096 |
ate | +0.095 |
iring | +0.095 |
Inside | +0.084 |
Neg logit contributions | |
joy | -0.157 |
ez | -0.153 |
hered | -0.141 |
nightmares | -0.135 |
tale | -0.127 |
Token |
|
Token ID | 198 |
Feature activation | -7.92e-03 |
Pos logit contributions | |
joy | +0.012 |
ez | +0.012 |
hered | +0.010 |
moment | +0.010 |
Charl | +0.009 |
Neg logit contributions | |
ires | -0.015 |
iring | -0.014 |
um | -0.014 |
ate | -0.014 |
outed | -0.013 |
Token | John |
Token ID | 7554 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
joy | +0.012 |
ez | +0.012 |
hered | +0.010 |
moment | +0.010 |
Charl | +0.010 |
Neg logit contributions | |
ires | -0.016 |
iring | -0.015 |
ate | -0.015 |
um | -0.015 |
ruce | -0.014 |
Token | decided |
Token ID | 3066 |
Feature activation | +9.02e-02 |
Pos logit contributions | |
ez | +0.110 |
joy | +0.108 |
hered | +0.096 |
moment | +0.090 |
nightmares | +0.087 |
Neg logit contributions | |
ires | -0.115 |
iring | -0.107 |
ate | -0.101 |
um | -0.098 |
ruce | -0.095 |
Token | pour |
Token ID | 12797 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
joy | +0.054 |
ez | +0.054 |
hered | +0.048 |
moment | +0.047 |
nightmares | +0.047 |
Neg logit contributions | |
ires | -0.042 |
iring | -0.040 |
um | -0.039 |
ate | -0.037 |
ruce | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
ires | +0.182 |
iring | +0.171 |
ate | +0.171 |
um | +0.167 |
ruce | +0.151 |
Neg logit contributions | |
joy | -0.215 |
ez | -0.207 |
hered | -0.187 |
nightmares | -0.177 |
moment | -0.176 |
Token | The |
Token ID | 383 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
joy | +0.052 |
ez | +0.048 |
hered | +0.044 |
moment | +0.042 |
nightmares | +0.040 |
Neg logit contributions | |
ires | -0.056 |
iring | -0.054 |
ate | -0.052 |
um | -0.051 |
ruce | -0.048 |
Token | milk |
Token ID | 7545 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
joy | +0.057 |
ez | +0.054 |
hered | +0.049 |
moment | +0.048 |
nightmares | +0.048 |
Neg logit contributions | |
ires | -0.040 |
ate | -0.038 |
iring | -0.037 |
um | -0.036 |
ruce | -0.033 |
Token | bub |
Token ID | 10015 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
joy | +0.126 |
ez | +0.123 |
hered | +0.112 |
moment | +0.105 |
nightmares | +0.105 |
Neg logit contributions | |
ires | -0.101 |
ate | -0.094 |
iring | -0.094 |
um | -0.092 |
ruce | -0.083 |
Token | bled |
Token ID | 9342 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
ires | +0.106 |
iring | +0.094 |
ate | +0.094 |
um | +0.092 |
ruce | +0.084 |
Neg logit contributions | |
ez | -0.178 |
joy | -0.174 |
hered | -0.165 |
moment | -0.152 |
breathing | -0.149 |
Token | and |
Token ID | 290 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
ires | +0.337 |
um | +0.329 |
iring | +0.325 |
ate | +0.317 |
ruce | +0.276 |
Neg logit contributions | |
joy | -0.554 |
ez | -0.524 |
hered | -0.484 |
nightmares | -0.454 |
heaven | -0.449 |
Token | ste |
Token ID | 2876 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
joy | +0.117 |
ez | +0.110 |
hered | +0.102 |
nightmares | +0.095 |
heaven | +0.092 |
Neg logit contributions | |
ires | -0.079 |
ate | -0.075 |
um | -0.075 |
iring | -0.074 |
ruce | -0.066 |
Token | amed |
Token ID | 2434 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
ez | +0.102 |
joy | +0.094 |
hered | +0.089 |
moment | +0.081 |
nightmares | +0.078 |
Neg logit contributions | |
ires | -0.089 |
iring | -0.081 |
ate | -0.076 |
um | -0.076 |
ruce | -0.074 |
Token | as |
Token ID | 355 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
ires | +0.152 |
iring | +0.145 |
um | +0.145 |
ate | +0.144 |
ruce | +0.122 |
Neg logit contributions | |
joy | -0.242 |
ez | -0.232 |
hered | -0.217 |
nightmares | -0.205 |
moment | -0.200 |
Token | it |
Token ID | 340 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
ires | +0.042 |
iring | +0.040 |
ate | +0.040 |
um | +0.039 |
ruce | +0.032 |
Neg logit contributions | |
ez | -0.067 |
joy | -0.066 |
hered | -0.061 |
tale | -0.055 |
nightmares | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
joy | +0.022 |
ez | +0.022 |
hered | +0.019 |
moment | +0.018 |
nightmares | +0.017 |
Neg logit contributions | |
ires | -0.024 |
iring | -0.023 |
ate | -0.022 |
um | -0.022 |
ruce | -0.020 |
Token | The |
Token ID | 464 |
Feature activation | +1.57e-03 |
Pos logit contributions | |
joy | +0.026 |
ez | +0.024 |
hered | +0.022 |
nightmares | +0.021 |
moment | +0.021 |
Neg logit contributions | |
ires | -0.020 |
ate | -0.018 |
iring | -0.018 |
um | -0.018 |
ruce | -0.016 |
Token | old |
Token ID | 1468 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
um | +0.002 |
ires | +0.002 |
ate | +0.002 |
iring | +0.002 |
old | +0.002 |
Neg logit contributions | |
ez | -0.004 |
joy | -0.004 |
hered | -0.003 |
nightmares | -0.003 |
Charl | -0.003 |
Token | man |
Token ID | 582 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
um | +0.282 |
ate | +0.275 |
ires | +0.260 |
iring | +0.227 |
elevator | +0.211 |
Neg logit contributions | |
ez | -0.606 |
hered | -0.556 |
joy | -0.544 |
nightmares | -0.540 |
unnecessary | -0.517 |
Token | walked |
Token ID | 6807 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
joy | +0.071 |
ez | +0.067 |
hered | +0.061 |
nightmares | +0.060 |
moment | +0.057 |
Neg logit contributions | |
ires | -0.050 |
ate | -0.049 |
um | -0.048 |
iring | -0.047 |
ruce | -0.041 |
Token | up |
Token ID | 510 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
joy | +0.031 |
ez | +0.029 |
hered | +0.027 |
moment | +0.025 |
nightmares | +0.025 |
Neg logit contributions | |
ires | -0.022 |
ate | -0.020 |
iring | -0.020 |
um | -0.020 |
ruce | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
ires | +0.323 |
ate | +0.319 |
um | +0.315 |
iring | +0.304 |
Inside | +0.289 |
Neg logit contributions | |
ez | -0.593 |
joy | -0.553 |
nightmares | -0.518 |
hered | -0.509 |
tale | -0.493 |
Token | the |
Token ID | 262 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
ez | +0.054 |
joy | +0.053 |
hered | +0.048 |
nightmares | +0.048 |
tale | +0.044 |
Neg logit contributions | |
ires | -0.033 |
ate | -0.032 |
um | -0.030 |
iring | -0.030 |
Inside | -0.026 |
Token | child |
Token ID | 1200 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
ez | +0.085 |
joy | +0.080 |
hered | +0.074 |
nightmares | +0.071 |
smart | +0.067 |
Neg logit contributions | |
ires | -0.060 |
iring | -0.059 |
um | -0.058 |
ate | -0.056 |
windows | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
joy | +0.431 |
ez | +0.425 |
hered | +0.377 |
nightmares | +0.377 |
liar | +0.361 |
Neg logit contributions | |
ires | -0.293 |
um | -0.289 |
ate | -0.282 |
iring | -0.276 |
Inside | -0.247 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
joy | +0.151 |
ez | +0.146 |
hered | +0.135 |
nightmares | +0.130 |
moment | +0.127 |
Neg logit contributions | |
ires | -0.069 |
ate | -0.065 |
iring | -0.064 |
um | -0.063 |
ruce | -0.053 |
Token | leash |
Token ID | 40529 |
Feature activation | -9.08e-02 |
Pos logit contributions | |
joy | +0.044 |
ez | +0.041 |
hered | +0.037 |
Charl | +0.034 |
knowing | +0.034 |
Neg logit contributions | |
ires | -0.033 |
um | -0.031 |
iring | -0.031 |
ate | -0.031 |
outed | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
joy | +0.158 |
ez | +0.156 |
hered | +0.143 |
nightmares | +0.134 |
moment | +0.131 |
Neg logit contributions | |
ires | -0.164 |
ate | -0.155 |
iring | -0.154 |
um | -0.152 |
ruce | -0.141 |
Token | Do |
Token ID | 2141 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
joy | +0.038 |
ez | +0.036 |
hered | +0.033 |
nightmares | +0.030 |
moment | +0.030 |
Neg logit contributions | |
ires | -0.042 |
ate | -0.040 |
iring | -0.040 |
um | -0.040 |
ruce | -0.036 |
Token | you |
Token ID | 345 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
joy | +0.411 |
ez | +0.404 |
hered | +0.365 |
nightmares | +0.353 |
moment | +0.337 |
Neg logit contributions | |
ires | -0.386 |
iring | -0.371 |
ate | -0.370 |
um | -0.362 |
ruce | -0.330 |
Token | understand |
Token ID | 1833 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
ires | +0.032 |
iring | +0.030 |
ate | +0.030 |
um | +0.030 |
ruce | +0.027 |
Neg logit contributions | |
joy | -0.028 |
ez | -0.028 |
hered | -0.024 |
nightmares | -0.023 |
moment | -0.022 |
Token | ?" |
Token ID | 1701 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
ez | +0.364 |
joy | +0.346 |
nightmares | +0.331 |
hered | +0.327 |
heaven | +0.325 |
Neg logit contributions | |
ires | -0.154 |
iring | -0.152 |
um | -0.135 |
ate | -0.128 |
outed | -0.118 |
Token | uncle |
Token ID | 7711 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
ires | +0.512 |
ate | +0.482 |
iring | +0.480 |
um | +0.476 |
ruce | +0.425 |
Neg logit contributions | |
joy | -0.666 |
ez | -0.640 |
hered | -0.582 |
nightmares | -0.550 |
moment | -0.541 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
ires | +0.025 |
ate | +0.024 |
iring | +0.023 |
um | +0.023 |
ruce | +0.021 |
Neg logit contributions | |
joy | -0.033 |
ez | -0.032 |
hered | -0.029 |
nightmares | -0.027 |
moment | -0.027 |
Token | said |
Token ID | 531 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
joy | +0.095 |
ez | +0.092 |
hered | +0.085 |
nightmares | +0.082 |
moment | +0.081 |
Neg logit contributions | |
ires | -0.038 |
ate | -0.035 |
um | -0.035 |
iring | -0.034 |
ruce | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
um | +0.202 |
ires | +0.197 |
iring | +0.189 |
ate | +0.169 |
outed | +0.150 |
Neg logit contributions | |
ez | -0.451 |
joy | -0.443 |
nightmares | -0.404 |
Charl | -0.402 |
hered | -0.398 |
Token |
|
Token ID | 198 |
Feature activation | -4.29e-04 |
Pos logit contributions | |
joy | +0.047 |
ez | +0.045 |
hered | +0.041 |
nightmares | +0.039 |
moment | +0.038 |
Neg logit contributions | |
ires | -0.038 |
ate | -0.036 |
iring | -0.036 |
um | -0.036 |
ruce | -0.032 |
Token | help |
Token ID | 1037 |
Feature activation | -8.48e-02 |
Pos logit contributions | |
joy | +0.350 |
ez | +0.348 |
hered | +0.313 |
moment | +0.286 |
ñ | +0.282 |
Neg logit contributions | |
ires | -0.215 |
ate | -0.207 |
iring | -0.197 |
um | -0.192 |
ruce | -0.171 |
Token | from |
Token ID | 422 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
joy | +0.197 |
ez | +0.190 |
hered | +0.171 |
moment | +0.156 |
nightmares | +0.156 |
Neg logit contributions | |
ires | -0.124 |
ate | -0.118 |
iring | -0.115 |
um | -0.111 |
ruce | -0.095 |
Token | others |
Token ID | 1854 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
ires | +0.027 |
ate | +0.026 |
iring | +0.025 |
um | +0.024 |
ruce | +0.021 |
Neg logit contributions | |
joy | -0.034 |
ez | -0.033 |
hered | -0.030 |
moment | -0.027 |
ñ | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
joy | +0.067 |
ez | +0.065 |
hered | +0.058 |
ñ | +0.055 |
liar | +0.052 |
Neg logit contributions | |
ires | -0.033 |
ate | -0.030 |
iring | -0.029 |
um | -0.026 |
her | -0.023 |
Token | learn |
Token ID | 2193 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
joy | +0.053 |
ez | +0.051 |
hered | +0.047 |
nightmares | +0.044 |
heaven | +0.044 |
Neg logit contributions | |
ires | -0.050 |
iring | -0.047 |
um | -0.047 |
ate | -0.046 |
ruce | -0.042 |
Token | new |
Token ID | 649 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
joy | +0.139 |
ez | +0.136 |
hered | +0.124 |
moment | +0.110 |
nightmares | +0.109 |
Neg logit contributions | |
ires | -0.092 |
ate | -0.088 |
iring | -0.085 |
um | -0.081 |
ruce | -0.073 |
Token | things |
Token ID | 1243 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
ate | +0.039 |
ires | +0.038 |
iring | +0.032 |
um | +0.031 |
ised | +0.026 |
Neg logit contributions | |
joy | -0.087 |
ez | -0.087 |
hered | -0.084 |
oled | -0.074 |
Charl | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
joy | +0.068 |
ez | +0.066 |
hered | +0.058 |
nightmares | +0.054 |
heaven | +0.052 |
Neg logit contributions | |
ires | -0.044 |
ate | -0.042 |
iring | -0.041 |
um | -0.038 |
ruce | -0.035 |
Token | And |
Token ID | 843 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
joy | +0.066 |
ez | +0.062 |
hered | +0.057 |
nightmares | +0.054 |
moment | +0.053 |
Neg logit contributions | |
ires | -0.057 |
ate | -0.054 |
um | -0.052 |
iring | -0.052 |
Inside | -0.046 |
Token | if |
Token ID | 611 |
Feature activation | -7.86e-03 |
Pos logit contributions | |
joy | +0.387 |
ez | +0.373 |
hered | +0.336 |
nightmares | +0.315 |
moment | +0.308 |
Neg logit contributions | |
ires | -0.332 |
ate | -0.313 |
iring | -0.313 |
um | -0.308 |
ruce | -0.277 |
Token | we |
Token ID | 356 |
Feature activation | -7.65e-02 |
Pos logit contributions | |
ez | +0.016 |
joy | +0.016 |
hered | +0.014 |
nightmares | +0.013 |
Charl | +0.013 |
Neg logit contributions | |
ires | -0.013 |
ate | -0.012 |
iring | -0.012 |
um | -0.012 |
ruce | -0.010 |
Token | her |
Token ID | 607 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
ires | +0.035 |
um | +0.034 |
iring | +0.033 |
ate | +0.033 |
ruce | +0.032 |
Neg logit contributions | |
joy | -0.034 |
ez | -0.031 |
nightmares | -0.027 |
heaven | -0.027 |
hered | -0.027 |
Token | special |
Token ID | 2041 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ez | +0.091 |
joy | +0.091 |
hered | +0.083 |
nightmares | +0.076 |
moment | +0.075 |
Neg logit contributions | |
ires | -0.070 |
iring | -0.065 |
ate | -0.065 |
um | -0.064 |
ruce | -0.056 |
Token | object |
Token ID | 2134 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
ires | +0.241 |
ate | +0.230 |
um | +0.228 |
iring | +0.225 |
ruce | +0.200 |
Neg logit contributions | |
joy | -0.285 |
ez | -0.285 |
hered | -0.255 |
nightmares | -0.243 |
heaven | -0.230 |
Token | . |
Token ID | 13 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
ires | +0.199 |
iring | +0.190 |
um | +0.190 |
ate | +0.180 |
ruce | +0.149 |
Neg logit contributions | |
joy | -0.327 |
ez | -0.324 |
hered | -0.292 |
nightmares | -0.273 |
moment | -0.268 |
Token | She |
Token ID | 1375 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
ez | +0.075 |
joy | +0.072 |
hered | +0.069 |
attacking | +0.067 |
roaring | +0.066 |
Neg logit contributions | |
ires | -0.041 |
um | -0.039 |
ate | -0.039 |
Inside | -0.036 |
iring | -0.036 |
Token | was |
Token ID | 373 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
ires | +0.068 |
ate | +0.065 |
um | +0.064 |
iring | +0.064 |
ruce | +0.057 |
Neg logit contributions | |
joy | -0.079 |
ez | -0.075 |
hered | -0.069 |
nightmares | -0.064 |
moment | -0.063 |
Token | very |
Token ID | 845 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
ires | +0.022 |
ate | +0.022 |
iring | +0.021 |
um | +0.021 |
ruce | +0.018 |
Neg logit contributions | |
joy | -0.036 |
ez | -0.036 |
hered | -0.033 |
nightmares | -0.031 |
moment | -0.030 |
Token | proud |
Token ID | 6613 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
joy | +0.044 |
ez | +0.038 |
knowing | +0.034 |
moment | +0.034 |
hered | +0.033 |
Neg logit contributions | |
ires | -0.056 |
ate | -0.053 |
um | -0.053 |
iring | -0.052 |
ruce | -0.050 |
Token | of |
Token ID | 286 |
Feature activation | +9.56e-02 |
Pos logit contributions | |
ires | +0.025 |
iring | +0.024 |
um | +0.023 |
ate | +0.023 |
ruce | +0.019 |
Neg logit contributions | |
joy | -0.050 |
ez | -0.050 |
hered | -0.046 |
nightmares | -0.043 |
moment | -0.043 |
Token | herself |
Token ID | 5223 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
ires | +0.130 |
iring | +0.118 |
ate | +0.118 |
um | +0.106 |
ered | +0.090 |
Neg logit contributions | |
ez | -0.228 |
joy | -0.225 |
hered | -0.220 |
oing | -0.198 |
tern | -0.191 |
Token | . |
Token ID | 13 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
um | +0.046 |
iring | +0.041 |
ires | +0.039 |
all | +0.034 |
ate | +0.033 |
Neg logit contributions | |
ez | -0.146 |
joy | -0.140 |
hered | -0.128 |
Bad | -0.125 |
nightmares | -0.124 |
Token | , |
Token ID | 11 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
ez | +0.148 |
joy | +0.147 |
hered | +0.128 |
nightmares | +0.127 |
breathing | +0.124 |
Neg logit contributions | |
um | -0.076 |
ires | -0.071 |
ate | -0.070 |
iring | -0.068 |
Inside | -0.065 |
Token | Jack |
Token ID | 3619 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
ires | +0.045 |
iring | +0.044 |
um | +0.043 |
ate | +0.043 |
Inside | +0.038 |
Neg logit contributions | |
ez | -0.069 |
joy | -0.067 |
hered | -0.061 |
nightmares | -0.057 |
breathing | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
ires | +0.057 |
ate | +0.054 |
um | +0.054 |
iring | +0.053 |
ruce | +0.047 |
Neg logit contributions | |
joy | -0.077 |
ez | -0.074 |
hered | -0.068 |
nightmares | -0.064 |
moment | -0.063 |
Token | Jane |
Token ID | 12091 |
Feature activation | +7.70e-02 |
Pos logit contributions | |
joy | +0.089 |
ez | +0.088 |
hered | +0.082 |
nightmares | +0.074 |
oing | +0.073 |
Neg logit contributions | |
ate | -0.058 |
ires | -0.058 |
um | -0.057 |
iring | -0.054 |
Inside | -0.049 |
Token | decided |
Token ID | 3066 |
Feature activation | +9.98e-02 |
Pos logit contributions | |
ires | +0.130 |
iring | +0.122 |
ate | +0.120 |
um | +0.117 |
ruce | +0.106 |
Neg logit contributions | |
joy | -0.153 |
ez | -0.150 |
hered | -0.137 |
nightmares | -0.126 |
moment | -0.125 |
Token | to |
Token ID | 284 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
ires | +0.140 |
ate | +0.135 |
iring | +0.133 |
um | +0.133 |
ruce | +0.117 |
Neg logit contributions | |
ez | -0.221 |
joy | -0.215 |
hered | -0.195 |
nightmares | -0.183 |
heaven | -0.178 |
Token | play |
Token ID | 711 |
Feature activation | +7.39e-02 |
Pos logit contributions | |
ez | +0.029 |
joy | +0.028 |
hered | +0.027 |
nightmares | +0.025 |
heaven | +0.025 |
Neg logit contributions | |
ires | -0.026 |
iring | -0.025 |
um | -0.022 |
outed | -0.022 |
ate | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
ires | +0.104 |
ate | +0.102 |
iring | +0.096 |
um | +0.095 |
ruce | +0.087 |
Neg logit contributions | |
joy | -0.164 |
ez | -0.158 |
hered | -0.145 |
nightmares | -0.136 |
moment | -0.134 |
Token | game |
Token ID | 983 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
joy | +0.026 |
ez | +0.025 |
hered | +0.023 |
nightmares | +0.022 |
moment | +0.021 |
Neg logit contributions | |
ires | -0.021 |
ate | -0.020 |
iring | -0.020 |
um | -0.020 |
ruce | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
joy | +0.021 |
ez | +0.020 |
hered | +0.019 |
nightmares | +0.018 |
moment | +0.017 |
Neg logit contributions | |
ires | -0.016 |
iring | -0.015 |
ate | -0.015 |
um | -0.015 |
ruce | -0.013 |
Token | Jane |
Token ID | 12091 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
joy | +0.040 |
ez | +0.038 |
hered | +0.035 |
nightmares | +0.033 |
moment | +0.033 |
Neg logit contributions | |
ires | -0.031 |
iring | -0.030 |
ate | -0.030 |
um | -0.029 |
ruce | -0.026 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
joy | +0.029 |
moment | +0.023 |
hered | +0.022 |
ez | +0.022 |
heaven | +0.020 |
Neg logit contributions | |
ires | -0.044 |
iring | -0.043 |
ate | -0.042 |
um | -0.040 |
ered | -0.038 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
ires | +0.035 |
iring | +0.033 |
ate | +0.031 |
um | +0.031 |
ruce | +0.028 |
Neg logit contributions | |
joy | -0.026 |
ez | -0.026 |
hered | -0.022 |
moment | -0.021 |
knowing | -0.021 |
Token | her |
Token ID | 607 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
ires | +0.321 |
um | +0.317 |
ate | +0.313 |
iring | +0.295 |
ruce | +0.264 |
Neg logit contributions | |
joy | -0.497 |
ez | -0.489 |
hered | -0.437 |
nightmares | -0.419 |
tale | -0.401 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
joy | +0.085 |
ez | +0.076 |
hered | +0.072 |
moment | +0.071 |
nightmares | +0.071 |
Neg logit contributions | |
ires | -0.060 |
ate | -0.055 |
iring | -0.055 |
um | -0.052 |
ruce | -0.048 |
Token | what |
Token ID | 644 |
Feature activation | +5.88e-03 |
Pos logit contributions | |
ires | +0.255 |
ate | +0.243 |
iring | +0.241 |
um | +0.233 |
ruce | +0.220 |
Neg logit contributions | |
joy | -0.245 |
ez | -0.220 |
hered | -0.202 |
moment | -0.200 |
knowing | -0.192 |
Token | it |
Token ID | 340 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ate | +0.008 |
ires | +0.007 |
um | +0.007 |
iring | +0.007 |
lly | +0.007 |
Neg logit contributions | |
joy | -0.014 |
ez | -0.013 |
hered | -0.013 |
nightmares | -0.011 |
worrying | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
ez | +0.029 |
joy | +0.029 |
moment | +0.027 |
hered | +0.026 |
nightmares | +0.025 |
Neg logit contributions | |
ires | -0.040 |
um | -0.038 |
iring | -0.037 |
ate | -0.036 |
ruce | -0.032 |
Token | for |
Token ID | 329 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
ires | +0.046 |
um | +0.044 |
ate | +0.043 |
iring | +0.042 |
ruce | +0.040 |
Neg logit contributions | |
joy | -0.033 |
ez | -0.030 |
knowing | -0.030 |
breathing | -0.029 |
moment | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
ires | +0.042 |
ate | +0.039 |
um | +0.039 |
iring | +0.039 |
ruce | +0.034 |
Neg logit contributions | |
joy | -0.054 |
ez | -0.051 |
hered | -0.047 |
nightmares | -0.044 |
moment | -0.044 |
Token | " |
Token ID | 366 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
joy | +0.032 |
ez | +0.027 |
hered | +0.026 |
moment | +0.026 |
heaven | +0.023 |
Neg logit contributions | |
ires | -0.042 |
iring | -0.041 |
ate | -0.039 |
um | -0.037 |
ruce | -0.036 |
Token | It |
Token ID | 1026 |
Feature activation | -4.96e-03 |
Pos logit contributions | |
joy | +0.127 |
ez | +0.124 |
hered | +0.112 |
moment | +0.107 |
Charl | +0.103 |
Neg logit contributions | |
ires | -0.115 |
iring | -0.108 |
ate | -0.105 |
um | -0.103 |
ruce | -0.095 |
Token | your |
Token ID | 534 |
Feature activation | +9.75e-03 |
Pos logit contributions | |
joy | +0.285 |
ez | +0.274 |
hered | +0.253 |
moment | +0.234 |
nightmares | +0.232 |
Neg logit contributions | |
ires | -0.206 |
ate | -0.197 |
um | -0.191 |
iring | -0.190 |
ruce | -0.170 |
Token | doll |
Token ID | 3654 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
ires | +0.015 |
iring | +0.015 |
um | +0.015 |
ate | +0.015 |
ruce | +0.013 |
Neg logit contributions | |
joy | -0.020 |
ez | -0.019 |
hered | -0.017 |
nightmares | -0.017 |
moment | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
joy | +0.217 |
ez | +0.213 |
hered | +0.194 |
nightmares | +0.184 |
moment | +0.180 |
Neg logit contributions | |
ires | -0.184 |
iring | -0.172 |
ate | -0.171 |
um | -0.170 |
ruce | -0.154 |
Token | Go |
Token ID | 1514 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
ez | +0.035 |
joy | +0.035 |
hered | +0.030 |
heaven | +0.029 |
nightmares | +0.028 |
Neg logit contributions | |
ires | -0.047 |
iring | -0.046 |
ate | -0.045 |
um | -0.045 |
ruce | -0.043 |
Token | away |
Token ID | 1497 |
Feature activation | -8.04e-02 |
Pos logit contributions | |
joy | +0.101 |
ez | +0.098 |
hered | +0.093 |
moment | +0.083 |
heaven | +0.080 |
Neg logit contributions | |
ires | -0.134 |
iring | -0.129 |
um | -0.119 |
ate | -0.117 |
ruce | -0.111 |
Token | !" |
Token ID | 2474 |
Feature activation | +7.51e-02 |
Pos logit contributions | |
joy | +0.168 |
ez | +0.158 |
hered | +0.147 |
moment | +0.142 |
nightmares | +0.141 |
Neg logit contributions | |
ires | -0.125 |
iring | -0.115 |
um | -0.115 |
ate | -0.114 |
ruce | -0.106 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
ires | +0.145 |
iring | +0.140 |
ate | +0.138 |
um | +0.133 |
ruce | +0.125 |
Neg logit contributions | |
joy | -0.130 |
ez | -0.125 |
hered | -0.112 |
nightmares | -0.103 |
heaven | -0.103 |
Token | said |
Token ID | 531 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
joy | +0.085 |
ez | +0.083 |
hered | +0.077 |
nightmares | +0.073 |
moment | +0.072 |
Neg logit contributions | |
ires | -0.043 |
iring | -0.039 |
ate | -0.039 |
um | -0.038 |
ruce | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
um | +0.192 |
ires | +0.173 |
iring | +0.167 |
ate | +0.157 |
ered | +0.146 |
Neg logit contributions | |
ez | -0.453 |
joy | -0.453 |
Charl | -0.406 |
nightmares | -0.403 |
tale | -0.400 |
Token | He |
Token ID | 679 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
joy | +0.045 |
ez | +0.043 |
hered | +0.039 |
nightmares | +0.036 |
moment | +0.036 |
Neg logit contributions | |
ires | -0.041 |
ate | -0.038 |
iring | -0.038 |
um | -0.038 |
ruce | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
ires | +0.105 |
iring | +0.098 |
ate | +0.096 |
um | +0.096 |
ruce | +0.087 |
Neg logit contributions | |
joy | -0.116 |
ez | -0.113 |
hered | -0.103 |
nightmares | -0.094 |
moment | -0.093 |
Token | of |
Token ID | 286 |
Feature activation | +9.84e-02 |
Pos logit contributions | |
joy | +0.602 |
ez | +0.590 |
hered | +0.545 |
nightmares | +0.509 |
moment | +0.505 |
Neg logit contributions | |
ires | -0.274 |
iring | -0.255 |
um | -0.252 |
ate | -0.242 |
ruce | -0.209 |
Token | singing |
Token ID | 13777 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
ires | +0.150 |
ate | +0.142 |
iring | +0.136 |
um | +0.122 |
ruce | +0.121 |
Neg logit contributions | |
ez | -0.208 |
hered | -0.196 |
joy | -0.191 |
oing | -0.172 |
ching | -0.167 |
Token | and |
Token ID | 290 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
joy | +0.578 |
ez | +0.575 |
hered | +0.520 |
liar | +0.475 |
Charl | +0.469 |
Neg logit contributions | |
ires | -0.303 |
ate | -0.301 |
um | -0.289 |
iring | -0.285 |
ruce | -0.240 |
Token | dancing |
Token ID | 15360 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
joy | +0.096 |
ez | +0.095 |
hered | +0.089 |
liar | +0.074 |
Charl | +0.072 |
Neg logit contributions | |
ires | -0.112 |
ate | -0.110 |
iring | -0.107 |
um | -0.099 |
ruce | -0.096 |
Token | on |
Token ID | 319 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
ez | +0.488 |
joy | +0.488 |
hered | +0.439 |
nightmares | +0.409 |
heaven | +0.402 |
Neg logit contributions | |
ires | -0.256 |
ate | -0.253 |
iring | -0.239 |
um | -0.234 |
ruce | -0.202 |
Token | the |
Token ID | 262 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
ez | +0.225 |
joy | +0.225 |
hered | +0.217 |
liar | +0.190 |
oing | +0.187 |
Neg logit contributions | |
ires | -0.136 |
ate | -0.131 |
iring | -0.121 |
um | -0.119 |
ruce | -0.108 |
Token | stage |
Token ID | 3800 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
joy | +0.079 |
ez | +0.077 |
hered | +0.071 |
nightmares | +0.065 |
smart | +0.063 |
Neg logit contributions | |
ires | -0.054 |
ate | -0.051 |
iring | -0.050 |
um | -0.049 |
ruce | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
joy | +0.116 |
ez | +0.114 |
hered | +0.104 |
nightmares | +0.094 |
Charl | +0.094 |
Neg logit contributions | |
ires | -0.074 |
ate | -0.071 |
iring | -0.068 |
um | -0.068 |
ruce | -0.059 |
Token | The |
Token ID | 383 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
joy | +0.069 |
ez | +0.067 |
hered | +0.063 |
nightmares | +0.057 |
moment | +0.056 |
Neg logit contributions | |
ires | -0.046 |
um | -0.043 |
ate | -0.042 |
iring | -0.042 |
Inside | -0.037 |
Token | theater |
Token ID | 13766 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
joy | +0.065 |
ez | +0.064 |
hered | +0.058 |
nightmares | +0.053 |
heaven | +0.052 |
Neg logit contributions | |
ires | -0.047 |
ate | -0.044 |
um | -0.043 |
iring | -0.043 |
ruce | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
ires | +0.051 |
ate | +0.049 |
um | +0.049 |
iring | +0.048 |
ruce | +0.042 |
Neg logit contributions | |
joy | -0.082 |
ez | -0.079 |
hered | -0.072 |
nightmares | -0.067 |
heaven | -0.066 |
Token | milk |
Token ID | 7545 |
Feature activation | -6.25e-02 |
Pos logit contributions | |
ires | +0.023 |
ate | +0.022 |
iring | +0.022 |
um | +0.021 |
ruce | +0.019 |
Neg logit contributions | |
joy | -0.025 |
ez | -0.024 |
hered | -0.022 |
nightmares | -0.021 |
moment | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
joy | +0.125 |
ez | +0.120 |
hered | +0.109 |
nightmares | +0.103 |
moment | +0.101 |
Neg logit contributions | |
ires | -0.103 |
ate | -0.097 |
iring | -0.097 |
um | -0.096 |
ruce | -0.086 |
Token | poured |
Token ID | 19036 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
joy | +0.110 |
ez | +0.107 |
hered | +0.096 |
nightmares | +0.092 |
moment | +0.092 |
Neg logit contributions | |
ires | -0.088 |
iring | -0.083 |
ate | -0.081 |
um | -0.080 |
ruce | -0.073 |
Token | it |
Token ID | 340 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
ires | +0.211 |
ate | +0.199 |
um | +0.199 |
iring | +0.198 |
ruce | +0.176 |
Neg logit contributions | |
joy | -0.272 |
ez | -0.261 |
hered | -0.237 |
nightmares | -0.225 |
moment | -0.218 |
Token | over |
Token ID | 625 |
Feature activation | +7.82e-02 |
Pos logit contributions | |
ez | +0.037 |
joy | +0.037 |
hered | +0.035 |
moment | +0.033 |
nightmares | +0.031 |
Neg logit contributions | |
ires | -0.034 |
iring | -0.032 |
ate | -0.031 |
um | -0.030 |
ruce | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
ires | +0.114 |
ate | +0.110 |
um | +0.109 |
iring | +0.106 |
ruce | +0.093 |
Neg logit contributions | |
ez | -0.167 |
joy | -0.166 |
hered | -0.153 |
nightmares | -0.145 |
Charl | -0.138 |
Token | cereal |
Token ID | 33158 |
Feature activation | +6.88e-03 |
Pos logit contributions | |
joy | +0.080 |
ez | +0.077 |
hered | +0.071 |
nightmares | +0.068 |
moment | +0.067 |
Neg logit contributions | |
ires | -0.050 |
ate | -0.046 |
iring | -0.046 |
um | -0.045 |
ruce | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
ires | +0.012 |
iring | +0.011 |
ate | +0.011 |
um | +0.011 |
ruce | +0.010 |
Neg logit contributions | |
joy | -0.013 |
ez | -0.012 |
hered | -0.011 |
moment | -0.011 |
nightmares | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -5.04e-03 |
Pos logit contributions | |
joy | +0.043 |
ez | +0.040 |
hered | +0.036 |
moment | +0.034 |
nightmares | +0.031 |
Neg logit contributions | |
ires | -0.049 |
iring | -0.048 |
ate | -0.046 |
um | -0.046 |
ruce | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -5.43e-03 |
Pos logit contributions | |
joy | +0.007 |
ez | +0.007 |
hered | +0.006 |
moment | +0.006 |
Charl | +0.005 |
Neg logit contributions | |
ires | -0.011 |
iring | -0.010 |
ate | -0.010 |
um | -0.010 |
ruce | -0.009 |
Token | They |
Token ID | 2990 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
joy | +0.011 |
ez | +0.010 |
hered | +0.010 |
moment | +0.009 |
Charl | +0.009 |
Neg logit contributions | |
ires | -0.009 |
iring | -0.008 |
ate | -0.008 |
um | -0.008 |
ruce | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
joy | +0.018 |
ez | +0.017 |
hered | +0.016 |
Charl | +0.016 |
Sleep | +0.015 |
Neg logit contributions | |
ires | -0.011 |
ate | -0.011 |
um | -0.010 |
iring | -0.010 |
ered | -0.009 |
Token | They |
Token ID | 1119 |
Feature activation | -6.27e-03 |
Pos logit contributions | |
joy | +0.036 |
ez | +0.032 |
moment | +0.029 |
nightmares | +0.028 |
hered | +0.028 |
Neg logit contributions | |
ires | -0.042 |
ate | -0.042 |
iring | -0.042 |
um | -0.040 |
ruce | -0.038 |
Token | had |
Token ID | 550 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
joy | +0.011 |
ez | +0.011 |
hered | +0.010 |
nightmares | +0.009 |
moment | +0.009 |
Neg logit contributions | |
ires | -0.012 |
iring | -0.011 |
um | -0.011 |
ate | -0.011 |
ruce | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
joy | +0.358 |
ez | +0.355 |
hered | +0.332 |
nightmares | +0.302 |
moment | +0.301 |
Neg logit contributions | |
ate | -0.153 |
ires | -0.151 |
iring | -0.138 |
um | -0.138 |
ruce | -0.129 |
Token | fun |
Token ID | 1257 |
Feature activation | +7.69e-02 |
Pos logit contributions | |
ez | +0.053 |
joy | +0.047 |
hered | +0.045 |
smart | +0.045 |
keeping | +0.043 |
Neg logit contributions | |
um | -0.018 |
cool | -0.017 |
lly | -0.017 |
ate | -0.017 |
ires | -0.016 |
Token | day |
Token ID | 1110 |
Feature activation | -7.81e-02 |
Pos logit contributions | |
ires | +0.124 |
ate | +0.121 |
um | +0.120 |
iring | +0.113 |
ruce | +0.108 |
Neg logit contributions | |
joy | -0.154 |
ez | -0.147 |
hered | -0.132 |
nightmares | -0.126 |
heaven | -0.123 |
Token | at |
Token ID | 379 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
joy | +0.184 |
ez | +0.177 |
hered | +0.161 |
moment | +0.158 |
nightmares | +0.157 |
Neg logit contributions | |
ires | -0.097 |
um | -0.088 |
iring | -0.088 |
ate | -0.088 |
ruce | -0.075 |
Token | the |
Token ID | 262 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
ires | +0.274 |
ate | +0.270 |
um | +0.253 |
iring | +0.253 |
ruce | +0.230 |
Neg logit contributions | |
joy | -0.377 |
ez | -0.367 |
hered | -0.331 |
nightmares | -0.315 |
heaven | -0.300 |
Token | park |
Token ID | 3952 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
joy | +0.090 |
ez | +0.085 |
hered | +0.080 |
moment | +0.076 |
nightmares | +0.076 |
Neg logit contributions | |
ires | -0.051 |
iring | -0.048 |
ate | -0.046 |
um | -0.046 |
ruce | -0.040 |
Token | with |
Token ID | 351 |
Feature activation | +8.00e-02 |
Pos logit contributions | |
ires | +0.127 |
ate | +0.125 |
um | +0.119 |
iring | +0.118 |
her | +0.100 |
Neg logit contributions | |
ez | -0.277 |
joy | -0.263 |
Charl | -0.240 |
hered | -0.239 |
nightmares | -0.233 |
Token | the |
Token ID | 262 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
ires | +0.126 |
ate | +0.119 |
iring | +0.119 |
um | +0.118 |
ruce | +0.105 |
Neg logit contributions | |
joy | -0.167 |
ez | -0.161 |
hered | -0.147 |
nightmares | -0.138 |
moment | -0.136 |
Token | They |
Token ID | 1119 |
Feature activation | -1.83e-03 |
Pos logit contributions | |
um | +0.099 |
ires | +0.091 |
iring | +0.088 |
Inside | +0.085 |
ate | +0.085 |
Neg logit contributions | |
joy | -0.196 |
ez | -0.187 |
nightmares | -0.171 |
worrying | -0.167 |
hered | -0.166 |
Token | both |
Token ID | 1111 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
joy | +0.004 |
ez | +0.004 |
hered | +0.003 |
moment | +0.003 |
nightmares | +0.003 |
Neg logit contributions | |
ires | -0.003 |
iring | -0.003 |
ate | -0.003 |
um | -0.003 |
ruce | -0.003 |
Token | went |
Token ID | 1816 |
Feature activation | +7.45e-02 |
Pos logit contributions | |
ires | +0.281 |
ate | +0.270 |
um | +0.269 |
iring | +0.263 |
ruce | +0.236 |
Neg logit contributions | |
joy | -0.402 |
ez | -0.380 |
hered | -0.349 |
nightmares | -0.335 |
moment | -0.325 |
Token | back |
Token ID | 736 |
Feature activation | -9.67e-02 |
Pos logit contributions | |
ires | +0.136 |
ate | +0.127 |
iring | +0.127 |
um | +0.127 |
ruce | +0.115 |
Neg logit contributions | |
joy | -0.139 |
ez | -0.125 |
hered | -0.118 |
moment | -0.110 |
nightmares | -0.108 |
Token | down |
Token ID | 866 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
joy | +0.191 |
ez | +0.177 |
hered | +0.164 |
moment | +0.154 |
nightmares | +0.153 |
Neg logit contributions | |
ires | -0.166 |
um | -0.154 |
iring | -0.154 |
ate | -0.154 |
ruce | -0.139 |
Token | the |
Token ID | 262 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
ires | +0.051 |
ate | +0.050 |
iring | +0.049 |
um | +0.048 |
Inside | +0.042 |
Neg logit contributions | |
ez | -0.086 |
joy | -0.080 |
nightmares | -0.074 |
hered | -0.074 |
ñ | -0.071 |
Token | hill |
Token ID | 12788 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
joy | +0.070 |
ez | +0.063 |
moment | +0.062 |
hered | +0.060 |
nightmares | +0.056 |
Neg logit contributions | |
ires | -0.067 |
ate | -0.062 |
iring | -0.062 |
um | -0.061 |
outed | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
ires | +0.038 |
um | +0.038 |
ate | +0.038 |
iring | +0.037 |
her | +0.031 |
Neg logit contributions | |
ez | -0.062 |
joy | -0.059 |
nightmares | -0.054 |
hered | -0.053 |
Charl | -0.051 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
ires | +0.027 |
iring | +0.026 |
ate | +0.026 |
um | +0.024 |
Inside | +0.023 |
Neg logit contributions | |
ez | -0.037 |
joy | -0.033 |
nightmares | -0.032 |
hered | -0.032 |
ñ | -0.031 |
Token | with |
Token ID | 351 |
Feature activation | +8.30e-02 |
Pos logit contributions | |
ez | +0.316 |
joy | +0.300 |
hered | +0.284 |
Charl | +0.281 |
liar | +0.266 |
Neg logit contributions | |
iring | -0.135 |
ires | -0.134 |
ate | -0.119 |
um | -0.115 |
Inside | -0.091 |
Token | their |
Token ID | 511 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
ires | +0.145 |
iring | +0.136 |
ate | +0.135 |
um | +0.132 |
ruce | +0.119 |
Neg logit contributions | |
joy | -0.162 |
ez | -0.156 |
hered | -0.143 |
moment | -0.132 |
nightmares | -0.129 |
Token | friends |
Token ID | 2460 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
ires | +0.162 |
ate | +0.158 |
iring | +0.150 |
um | +0.146 |
ruce | +0.144 |
Neg logit contributions | |
joy | -0.184 |
ez | -0.172 |
hered | -0.164 |
moment | -0.143 |
oing | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
ires | +0.051 |
um | +0.048 |
iring | +0.048 |
ate | +0.048 |
ruce | +0.041 |
Neg logit contributions | |
joy | -0.092 |
ez | -0.089 |
hered | -0.081 |
nightmares | -0.078 |
moment | -0.076 |
Token | They |
Token ID | 1119 |
Feature activation | -6.66e-03 |
Pos logit contributions | |
joy | +0.052 |
ez | +0.046 |
hered | +0.043 |
moment | +0.042 |
nightmares | +0.040 |
Neg logit contributions | |
ires | -0.041 |
iring | -0.040 |
ate | -0.038 |
um | -0.037 |
ruce | -0.035 |
Token | had |
Token ID | 550 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
joy | +0.013 |
ez | +0.013 |
hered | +0.011 |
nightmares | +0.011 |
moment | +0.010 |
Neg logit contributions | |
ires | -0.011 |
iring | -0.011 |
ate | -0.011 |
um | -0.011 |
ruce | -0.009 |
Token | been |
Token ID | 587 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
joy | +0.334 |
ez | +0.332 |
hered | +0.309 |
moment | +0.282 |
nightmares | +0.281 |
Neg logit contributions | |
ate | -0.164 |
ires | -0.159 |
um | -0.154 |
iring | -0.153 |
her | -0.139 |
Token | having |
Token ID | 1719 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
joy | +0.351 |
ez | +0.350 |
hered | +0.312 |
moment | +0.295 |
nightmares | +0.291 |
Neg logit contributions | |
ate | -0.297 |
ires | -0.295 |
iring | -0.291 |
um | -0.281 |
her | -0.257 |
Token | so |
Token ID | 523 |
Feature activation | +6.10e-02 |
Pos logit contributions | |
ez | +0.478 |
joy | +0.466 |
hered | +0.446 |
oing | +0.419 |
moment | +0.406 |
Neg logit contributions | |
ate | -0.175 |
ires | -0.173 |
um | -0.169 |
iring | -0.161 |
ruce | -0.152 |
Token | much |
Token ID | 881 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
ires | +0.118 |
iring | +0.112 |
um | +0.112 |
ate | +0.112 |
ruce | +0.102 |
Neg logit contributions | |
joy | -0.105 |
ez | -0.100 |
hered | -0.090 |
nightmares | -0.082 |
moment | -0.082 |
Token | fun |
Token ID | 1257 |
Feature activation | +7.77e-02 |
Pos logit contributions | |
joy | +0.300 |
ez | +0.293 |
hered | +0.270 |
nightmares | +0.253 |
moment | +0.249 |
Neg logit contributions | |
ires | -0.137 |
ate | -0.126 |
um | -0.125 |
iring | -0.123 |
ruce | -0.108 |
Token | . |
Token ID | 13 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
ires | +0.096 |
iring | +0.088 |
ate | +0.085 |
um | +0.079 |
ruce | +0.071 |
Neg logit contributions | |
joy | -0.183 |
ez | -0.180 |
hered | -0.170 |
severe | -0.159 |
Charl | -0.158 |
Token |
|
Token ID | 198 |
Feature activation | -2.90e-03 |
Pos logit contributions | |
joy | +0.049 |
ez | +0.043 |
hered | +0.040 |
moment | +0.039 |
nightmares | +0.037 |
Neg logit contributions | |
ires | -0.043 |
iring | -0.042 |
ate | -0.041 |
um | -0.039 |
ruce | -0.038 |
Token | realized |
Token ID | 6939 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
ires | +0.083 |
iring | +0.079 |
ate | +0.077 |
um | +0.077 |
ruce | +0.071 |
Neg logit contributions | |
joy | -0.074 |
ez | -0.073 |
hered | -0.064 |
nightmares | -0.060 |
moment | -0.058 |
Token | that |
Token ID | 326 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
joy | +0.116 |
ez | +0.112 |
hered | +0.104 |
nightmares | +0.099 |
moment | +0.098 |
Neg logit contributions | |
ires | -0.055 |
ate | -0.051 |
iring | -0.050 |
um | -0.050 |
ruce | -0.042 |
Token | being |
Token ID | 852 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
ires | +0.191 |
ate | +0.182 |
iring | +0.182 |
um | +0.159 |
ruce | +0.151 |
Neg logit contributions | |
joy | -0.325 |
ez | -0.324 |
hered | -0.322 |
Charl | -0.286 |
oing | -0.277 |
Token | fair |
Token ID | 3148 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
joy | +0.623 |
ez | +0.594 |
hered | +0.550 |
moment | +0.513 |
nightmares | +0.504 |
Neg logit contributions | |
ires | -0.480 |
ate | -0.456 |
iring | -0.455 |
um | -0.438 |
ruce | -0.391 |
Token | meant |
Token ID | 4001 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
ires | +0.181 |
iring | +0.173 |
ate | +0.172 |
um | +0.166 |
ruce | +0.143 |
Neg logit contributions | |
joy | -0.231 |
ez | -0.227 |
hered | -0.201 |
nightmares | -0.189 |
heaven | -0.188 |
Token | being |
Token ID | 852 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
ez | +0.056 |
joy | +0.056 |
hered | +0.051 |
Charl | +0.047 |
oing | +0.047 |
Neg logit contributions | |
ires | -0.036 |
iring | -0.034 |
ate | -0.033 |
um | -0.032 |
Inside | -0.028 |
Token | balanced |
Token ID | 12974 |
Feature activation | +5.26e-03 |
Pos logit contributions | |
joy | +0.662 |
ez | +0.637 |
hered | +0.588 |
moment | +0.550 |
nightmares | +0.545 |
Neg logit contributions | |
ires | -0.443 |
ate | -0.428 |
iring | -0.419 |
um | -0.394 |
ruce | -0.346 |
Token | and |
Token ID | 290 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
ires | +0.008 |
ate | +0.007 |
iring | +0.007 |
um | +0.007 |
ruce | +0.006 |
Neg logit contributions | |
joy | -0.012 |
ez | -0.011 |
hered | -0.010 |
nightmares | -0.010 |
heaven | -0.009 |
Token | that |
Token ID | 326 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
joy | +0.115 |
ez | +0.111 |
hered | +0.100 |
nightmares | +0.092 |
moment | +0.089 |
Neg logit contributions | |
ires | -0.114 |
ate | -0.109 |
iring | -0.107 |
um | -0.103 |
ruce | -0.095 |
Token | it |
Token ID | 340 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
ires | +0.215 |
ate | +0.211 |
iring | +0.202 |
um | +0.187 |
ruce | +0.177 |
Neg logit contributions | |
joy | -0.314 |
ez | -0.306 |
hered | -0.289 |
Charl | -0.262 |
oing | -0.252 |
Token | was |
Token ID | 373 |
Feature activation | +5.37e-03 |
Pos logit contributions | |
joy | +0.057 |
ez | +0.053 |
hered | +0.048 |
nightmares | +0.044 |
moment | +0.044 |
Neg logit contributions | |
ires | -0.048 |
ate | -0.046 |
iring | -0.045 |
um | -0.044 |
ruce | -0.041 |
Token | distance |
Token ID | 5253 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
ires | +0.078 |
um | +0.075 |
ate | +0.073 |
iring | +0.073 |
ruce | +0.062 |
Neg logit contributions | |
joy | -0.110 |
ez | -0.109 |
hered | -0.099 |
nightmares | -0.091 |
heaven | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | +3.61e-03 |
Pos logit contributions | |
joy | +0.464 |
ez | +0.455 |
hered | +0.410 |
nightmares | +0.386 |
Charl | +0.375 |
Neg logit contributions | |
ires | -0.339 |
iring | -0.322 |
um | -0.321 |
ate | -0.317 |
ruce | -0.278 |
Token | Everyone |
Token ID | 11075 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
ires | +0.006 |
ate | +0.005 |
um | +0.005 |
iring | +0.005 |
ruce | +0.005 |
Neg logit contributions | |
joy | -0.008 |
ez | -0.007 |
hered | -0.007 |
nightmares | -0.006 |
moment | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
ires | +0.248 |
ate | +0.235 |
iring | +0.233 |
um | +0.229 |
ruce | +0.204 |
Neg logit contributions | |
joy | -0.363 |
ez | -0.343 |
hered | -0.317 |
nightmares | -0.301 |
heaven | -0.291 |
Token | over |
Token ID | 625 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
ires | +0.081 |
iring | +0.079 |
ate | +0.078 |
um | +0.077 |
ruce | +0.068 |
Neg logit contributions | |
joy | -0.104 |
ez | -0.101 |
hered | -0.091 |
nightmares | -0.087 |
moment | -0.084 |
Token | joy |
Token ID | 2633 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
ires | +0.164 |
iring | +0.159 |
um | +0.158 |
ate | +0.151 |
ised | +0.132 |
Neg logit contributions | |
ez | -0.211 |
hered | -0.178 |
ñ | -0.174 |
Charl | -0.169 |
nightmares | -0.168 |
Token | ed |
Token ID | 276 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
joy | +1.096 |
ez | +1.056 |
nightmares | +1.017 |
heaven | +1.003 |
Charl | +0.975 |
Neg logit contributions | |
um | -0.815 |
ires | -0.799 |
iring | -0.777 |
ate | -0.753 |
outed | -0.656 |
Token | that |
Token ID | 326 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
ez | +0.151 |
joy | +0.142 |
hered | +0.131 |
nightmares | +0.129 |
Charl | +0.124 |
Neg logit contributions | |
ires | -0.087 |
iring | -0.086 |
um | -0.085 |
ate | -0.080 |
ruce | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
ires | +0.205 |
um | +0.196 |
iring | +0.192 |
ate | +0.189 |
ered | +0.168 |
Neg logit contributions | |
ez | -0.383 |
joy | -0.367 |
hered | -0.339 |
Charl | -0.324 |
oing | -0.315 |
Token | cow |
Token ID | 9875 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
ez | +0.039 |
joy | +0.038 |
hered | +0.035 |
nightmares | +0.031 |
heaven | +0.031 |
Neg logit contributions | |
ires | -0.028 |
um | -0.027 |
iring | -0.027 |
ate | -0.027 |
ruce | -0.023 |
Token | had |
Token ID | 550 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
joy | +0.317 |
ez | +0.298 |
hered | +0.270 |
nightmares | +0.264 |
Charl | +0.260 |
Neg logit contributions | |
ires | -0.178 |
um | -0.176 |
iring | -0.175 |
ate | -0.174 |
lly | -0.157 |
Token | wind |
Token ID | 2344 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
joy | +0.055 |
ez | +0.053 |
hered | +0.048 |
nightmares | +0.046 |
moment | +0.045 |
Neg logit contributions | |
ires | -0.044 |
ate | -0.042 |
iring | -0.041 |
um | -0.041 |
ruce | -0.037 |
Token | is |
Token ID | 318 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
ires | +0.053 |
ate | +0.052 |
iring | +0.050 |
um | +0.048 |
ruce | +0.043 |
Neg logit contributions | |
joy | -0.084 |
ez | -0.074 |
hered | -0.067 |
nightmares | -0.066 |
Charl | -0.065 |
Token | just |
Token ID | 655 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
joy | +0.045 |
ez | +0.042 |
hered | +0.038 |
nightmares | +0.035 |
moment | +0.034 |
Neg logit contributions | |
ires | -0.034 |
iring | -0.033 |
ate | -0.032 |
um | -0.032 |
ruce | -0.029 |
Token | singing |
Token ID | 13777 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
ires | +0.148 |
iring | +0.142 |
ate | +0.138 |
um | +0.130 |
ruce | +0.117 |
Neg logit contributions | |
joy | -0.223 |
ez | -0.206 |
hered | -0.193 |
Charl | -0.176 |
ñ | -0.175 |
Token | a |
Token ID | 257 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
joy | +0.623 |
ez | +0.552 |
hered | +0.521 |
forgetting | +0.478 |
Charl | +0.475 |
Neg logit contributions | |
ires | -0.313 |
ate | -0.300 |
iring | -0.296 |
um | -0.295 |
her | -0.250 |
Token | lull |
Token ID | 42858 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
joy | +0.054 |
ez | +0.052 |
hered | +0.046 |
nightmares | +0.042 |
smart | +0.041 |
Neg logit contributions | |
um | -0.028 |
iring | -0.028 |
ires | -0.027 |
ate | -0.026 |
ruce | -0.024 |
Token | aby |
Token ID | 3930 |
Feature activation | -8.17e-02 |
Pos logit contributions | |
joy | +0.683 |
ez | +0.650 |
hered | +0.580 |
nightmares | +0.540 |
moment | +0.530 |
Neg logit contributions | |
ires | -0.776 |
ate | -0.739 |
iring | -0.736 |
um | -0.731 |
ruce | -0.670 |
Token | to |
Token ID | 284 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
joy | +0.211 |
ez | +0.185 |
hered | +0.168 |
Charl | +0.163 |
nightmares | +0.160 |
Neg logit contributions | |
ires | -0.099 |
um | -0.095 |
iring | -0.095 |
ate | -0.089 |
?" | -0.076 |
Token | help |
Token ID | 1037 |
Feature activation | -8.56e-02 |
Pos logit contributions | |
joy | +0.061 |
ez | +0.055 |
hered | +0.052 |
nightmares | +0.047 |
moment | +0.047 |
Neg logit contributions | |
ires | -0.042 |
ate | -0.040 |
iring | -0.039 |
um | -0.038 |
ruce | -0.035 |
Token | you |
Token ID | 345 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
joy | +0.206 |
ez | +0.197 |
hered | +0.182 |
nightmares | +0.173 |
moment | +0.171 |
Neg logit contributions | |
ires | -0.109 |
ate | -0.102 |
iring | -0.101 |
um | -0.100 |
ruce | -0.086 |
Token | sleep |
Token ID | 3993 |
Feature activation | -9.52e-02 |
Pos logit contributions | |
ate | +0.017 |
ires | +0.017 |
iring | +0.017 |
um | +0.016 |
ruce | +0.014 |
Neg logit contributions | |
joy | -0.027 |
ez | -0.025 |
hered | -0.022 |
Charl | -0.020 |
nightmares | -0.020 |
Token | Mom |
Token ID | 11254 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
joy | +0.068 |
ez | +0.064 |
hered | +0.058 |
nightmares | +0.055 |
moment | +0.054 |
Neg logit contributions | |
ires | -0.057 |
ate | -0.054 |
iring | -0.054 |
um | -0.053 |
ruce | -0.049 |
Token | hugged |
Token ID | 41130 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
ires | +0.371 |
um | +0.367 |
ate | +0.366 |
iring | +0.359 |
ruce | +0.327 |
Neg logit contributions | |
joy | -0.431 |
ez | -0.404 |
hered | -0.365 |
nightmares | -0.355 |
heaven | -0.340 |
Token | her |
Token ID | 607 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
joy | +0.189 |
ez | +0.182 |
hered | +0.164 |
nightmares | +0.154 |
moment | +0.151 |
Neg logit contributions | |
ires | -0.141 |
um | -0.133 |
ate | -0.131 |
iring | -0.130 |
ruce | -0.115 |
Token | back |
Token ID | 736 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
ez | +0.123 |
joy | +0.119 |
hered | +0.109 |
Charl | +0.107 |
nightmares | +0.105 |
Neg logit contributions | |
um | -0.060 |
up | -0.055 |
ires | -0.055 |
ate | -0.053 |
iring | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
ez | +0.241 |
joy | +0.240 |
hered | +0.228 |
Charl | +0.209 |
nightmares | +0.205 |
Neg logit contributions | |
ires | -0.132 |
um | -0.125 |
iring | -0.119 |
ate | -0.117 |
outed | -0.097 |
Token | forg |
Token ID | 5784 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
joy | +0.149 |
ez | +0.144 |
hered | +0.132 |
nightmares | +0.126 |
moment | +0.124 |
Neg logit contributions | |
ires | -0.080 |
ate | -0.075 |
iring | -0.075 |
um | -0.074 |
ruce | -0.064 |
Token | ave |
Token ID | 1015 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
joy | +0.720 |
ez | +0.648 |
hered | +0.604 |
nightmares | +0.596 |
moment | +0.591 |
Neg logit contributions | |
ires | -0.469 |
ate | -0.432 |
iring | -0.431 |
um | -0.429 |
outed | -0.405 |
Token | her |
Token ID | 607 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
joy | +0.044 |
ez | +0.042 |
hered | +0.038 |
nightmares | +0.036 |
moment | +0.035 |
Neg logit contributions | |
ires | -0.040 |
ate | -0.037 |
iring | -0.037 |
um | -0.037 |
ruce | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
ez | +0.132 |
joy | +0.129 |
Charl | +0.124 |
hered | +0.122 |
Patch | +0.114 |
Neg logit contributions | |
ires | -0.051 |
um | -0.047 |
iring | -0.045 |
ate | -0.043 |
up | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
joy | +0.066 |
ez | +0.061 |
hered | +0.055 |
nightmares | +0.053 |
moment | +0.052 |
Neg logit contributions | |
ires | -0.056 |
ate | -0.054 |
iring | -0.054 |
um | -0.053 |
ruce | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
joy | +0.022 |
ez | +0.021 |
hered | +0.018 |
moment | +0.018 |
breathing | +0.017 |
Neg logit contributions | |
ires | -0.030 |
iring | -0.029 |
ate | -0.028 |
um | -0.028 |
ruce | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +5.99e-04 |
Pos logit contributions | |
ires | +0.003 |
iring | +0.002 |
um | +0.002 |
outed | +0.002 |
ate | +0.002 |
Neg logit contributions | |
joy | -0.002 |
ez | -0.002 |
hered | -0.001 |
moment | -0.001 |
smart | -0.001 |
Token | " |
Token ID | 1 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
ires | +0.001 |
iring | +0.001 |
um | +0.001 |
ate | +0.001 |
outed | +0.001 |
Neg logit contributions | |
joy | -0.001 |
ez | -0.001 |
hered | -0.001 |
smart | -0.001 |
moment | -0.001 |
Token | Why |
Token ID | 5195 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
ires | +0.034 |
ate | +0.034 |
iring | +0.033 |
um | +0.032 |
ruce | +0.029 |
Neg logit contributions | |
joy | -0.043 |
ez | -0.042 |
nightmares | -0.036 |
hered | -0.036 |
knowing | -0.035 |
Token | are |
Token ID | 389 |
Feature activation | -8.97e-02 |
Pos logit contributions | |
joy | +0.154 |
ez | +0.147 |
hered | +0.132 |
nightmares | +0.121 |
heaven | +0.121 |
Neg logit contributions | |
ires | -0.117 |
ate | -0.114 |
um | -0.109 |
iring | -0.108 |
ruce | -0.097 |
Token | you |
Token ID | 345 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
joy | +0.129 |
ez | +0.112 |
nightmares | +0.110 |
hered | +0.106 |
forgetting | +0.102 |
Neg logit contributions | |
ate | -0.184 |
ires | -0.178 |
um | -0.177 |
iring | -0.172 |
ruce | -0.159 |
Token | crying |
Token ID | 13774 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
ires | +0.041 |
ate | +0.039 |
um | +0.039 |
iring | +0.038 |
ruce | +0.036 |
Neg logit contributions | |
joy | -0.026 |
ez | -0.024 |
hered | -0.021 |
nightmares | -0.019 |
moment | -0.019 |
Token | ?" |
Token ID | 1701 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
joy | +0.732 |
ez | +0.698 |
hered | +0.669 |
moment | +0.631 |
nightmares | +0.624 |
Neg logit contributions | |
ires | -0.385 |
ate | -0.343 |
iring | -0.321 |
ruce | -0.315 |
um | -0.310 |
Token | she |
Token ID | 673 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
ires | +0.679 |
ate | +0.669 |
iring | +0.661 |
um | +0.654 |
ruce | +0.614 |
Neg logit contributions | |
joy | -0.500 |
ez | -0.436 |
nightmares | -0.390 |
hered | -0.389 |
moment | -0.381 |
Token | asks |
Token ID | 7893 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
joy | +0.169 |
ez | +0.166 |
hered | +0.153 |
nightmares | +0.151 |
moment | +0.143 |
Neg logit contributions | |
ires | -0.112 |
iring | -0.106 |
um | -0.106 |
ate | -0.103 |
ruce | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | -8.67e-03 |
Pos logit contributions | |
ires | +0.249 |
um | +0.231 |
iring | +0.229 |
ate | +0.218 |
outed | +0.180 |
Neg logit contributions | |
ez | -0.451 |
joy | -0.438 |
hered | -0.393 |
Charl | -0.388 |
nightmares | -0.387 |
Token | " |
Token ID | 366 |
Feature activation | -6.97e-02 |
Pos logit contributions | |
joy | +0.020 |
ez | +0.020 |
hered | +0.018 |
nightmares | +0.018 |
ñ | +0.017 |
Neg logit contributions | |
ires | -0.011 |
ate | -0.010 |
iring | -0.010 |
um | -0.010 |
Inside | -0.010 |