Token | he |
Token ID | 339 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
Proced | +0.035 |
yip | +0.033 |
Fn | +0.031 |
Address | +0.031 |
aning | +0.031 |
Neg logit contributions | |
eyed | -0.027 |
didn | -0.027 |
CBS | -0.027 |
atra | -0.026 |
photos | -0.026 |
Token | think |
Token ID | 892 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
Proced | +0.166 |
yip | +0.157 |
Address | +0.146 |
aning | +0.146 |
Fn | +0.145 |
Neg logit contributions | |
eyed | -0.139 |
didn | -0.137 |
CBS | -0.136 |
atra | -0.133 |
photos | -0.132 |
Token | he |
Token ID | 339 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
Proced | +0.030 |
yip | +0.029 |
Fn | +0.027 |
Address | +0.027 |
aning | +0.026 |
Neg logit contributions | |
eyed | -0.026 |
CBS | -0.025 |
didn | -0.025 |
atra | -0.025 |
photos | -0.024 |
Token | is |
Token ID | 318 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
Proced | +0.186 |
yip | +0.179 |
aning | +0.165 |
Address | +0.165 |
Fn | +0.164 |
Neg logit contributions | |
eyed | -0.156 |
didn | -0.154 |
CBS | -0.153 |
atra | -0.149 |
photos | -0.148 |
Token | fool |
Token ID | 9192 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
Proced | +0.106 |
yip | +0.101 |
Address | +0.093 |
aning | +0.093 |
Fn | +0.093 |
Neg logit contributions | |
eyed | -0.094 |
CBS | -0.093 |
didn | -0.093 |
atra | -0.091 |
wasn | -0.090 |
Token | ing |
Token ID | 278 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
Proced | +0.044 |
yip | +0.040 |
Address | +0.030 |
aning | +0.030 |
Fn | +0.030 |
Neg logit contributions | |
eyed | -0.171 |
didn | -0.170 |
CBS | -0.170 |
atra | -0.167 |
wasn | -0.167 |
Token | ? |
Token ID | 30 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
Proced | +0.485 |
yip | +0.467 |
Fn | +0.426 |
Address | +0.425 |
aning | +0.424 |
Neg logit contributions | |
eyed | -0.441 |
didn | -0.438 |
CBS | -0.437 |
atra | -0.428 |
wasn | -0.424 |
Token |
|
Token ID | 198 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Proced | +0.023 |
yip | +0.021 |
Fn | +0.020 |
Address | +0.019 |
Agg | +0.019 |
Neg logit contributions | |
eyed | -0.022 |
didn | -0.022 |
CBS | -0.022 |
atra | -0.022 |
hadn | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Proced | +0.065 |
yip | +0.060 |
Address | +0.056 |
Agg | +0.055 |
aning | +0.055 |
Neg logit contributions | |
eyed | -0.086 |
didn | -0.085 |
CBS | -0.085 |
atra | -0.084 |
wasn | -0.083 |
Token | David |
Token ID | 11006 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
Proced | +0.067 |
yip | +0.063 |
Address | +0.061 |
Agg | +0.061 |
AMS | +0.059 |
Neg logit contributions | |
eyed | -0.074 |
didn | -0.072 |
atra | -0.072 |
hadn | -0.072 |
wasn | -0.072 |
Token | Cron |
Token ID | 31683 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
eyed | +0.023 |
CBS | +0.023 |
didn | +0.023 |
atra | +0.022 |
wasn | +0.022 |
Neg logit contributions | |
Proced | -0.026 |
yip | -0.025 |
Fn | -0.023 |
Address | -0.023 |
aning | -0.023 |
Token | Sunderland |
Token ID | 35706 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
Proced | +0.079 |
yip | +0.075 |
Fn | +0.070 |
aning | +0.069 |
Address | +0.069 |
Neg logit contributions | |
CBS | -0.063 |
eyed | -0.063 |
didn | -0.063 |
atra | -0.062 |
photos | -0.060 |
Token | 's |
Token ID | 338 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
Proced | +0.044 |
yip | +0.042 |
Fn | +0.039 |
Address | +0.039 |
aning | +0.039 |
Neg logit contributions | |
eyed | -0.023 |
CBS | -0.023 |
didn | -0.023 |
atra | -0.022 |
photos | -0.022 |
Token | owner |
Token ID | 4870 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
eyed | +0.014 |
CBS | +0.014 |
didn | +0.014 |
atra | +0.014 |
photos | +0.013 |
Neg logit contributions | |
Proced | -0.017 |
yip | -0.016 |
Fn | -0.016 |
Address | -0.015 |
aning | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
Proced | +0.011 |
yip | +0.010 |
Fn | +0.010 |
Address | +0.009 |
aning | +0.009 |
Neg logit contributions | |
eyed | -0.011 |
didn | -0.011 |
CBS | -0.011 |
atra | -0.011 |
wasn | -0.011 |
Token | Given |
Token ID | 11259 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
Proced | +0.038 |
yip | +0.034 |
Fn | +0.033 |
Address | +0.033 |
Finding | +0.032 |
Neg logit contributions | |
eyed | -0.035 |
CBS | -0.034 |
didn | -0.034 |
atra | -0.033 |
wasn | -0.033 |
Token | Fab |
Token ID | 14236 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
Proced | +0.163 |
yip | +0.157 |
Fn | +0.146 |
Address | +0.145 |
aning | +0.143 |
Neg logit contributions | |
eyed | -0.114 |
CBS | -0.113 |
didn | -0.111 |
atra | -0.110 |
photos | -0.106 |
Token | io |
Token ID | 952 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Proced | +0.442 |
yip | +0.424 |
Address | +0.391 |
Agg | +0.389 |
aning | +0.386 |
Neg logit contributions | |
eyed | -0.475 |
CBS | -0.471 |
didn | -0.464 |
atra | -0.459 |
wasn | -0.453 |
Token | Cape |
Token ID | 15725 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
Proced | +0.075 |
yip | +0.071 |
Agg | +0.066 |
Fn | +0.066 |
Address | +0.066 |
Neg logit contributions | |
CBS | -0.065 |
eyed | -0.065 |
didn | -0.064 |
atra | -0.063 |
photos | -0.062 |
Token | llo |
Token ID | 18798 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
Proced | +0.098 |
yip | +0.093 |
Fn | +0.085 |
aning | +0.085 |
Address | +0.085 |
Neg logit contributions | |
eyed | -0.090 |
CBS | -0.089 |
didn | -0.088 |
photos | -0.086 |
atra | -0.086 |
Token | 's |
Token ID | 338 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
Proced | +0.126 |
yip | +0.123 |
Fn | +0.115 |
Address | +0.115 |
aning | +0.114 |
Neg logit contributions | |
eyed | -0.070 |
CBS | -0.069 |
didn | -0.068 |
atra | -0.066 |
photos | -0.066 |
Token | recent |
Token ID | 2274 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
Proced | +0.030 |
yip | +0.029 |
Address | +0.027 |
Fn | +0.027 |
aning | +0.027 |
Neg logit contributions | |
eyed | -0.029 |
CBS | -0.028 |
didn | -0.028 |
atra | -0.028 |
photos | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
eyed | +0.115 |
didn | +0.114 |
CBS | +0.113 |
atra | +0.112 |
wasn | +0.111 |
Neg logit contributions | |
Proced | -0.095 |
yip | -0.091 |
Fn | -0.082 |
Address | -0.082 |
Agg | -0.081 |
Token | reconnect |
Token ID | 37671 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
Proced | +0.076 |
yip | +0.072 |
Fn | +0.068 |
Address | +0.067 |
aning | +0.066 |
Neg logit contributions | |
eyed | -0.063 |
didn | -0.063 |
CBS | -0.062 |
atra | -0.060 |
politics | -0.059 |
Token | – |
Token ID | 784 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
eyed | +0.152 |
didn | +0.150 |
CBS | +0.150 |
atra | +0.147 |
wasn | +0.147 |
Neg logit contributions | |
Proced | -0.151 |
yip | -0.145 |
Fn | -0.132 |
Address | -0.131 |
aning | -0.130 |
Token | no |
Token ID | 645 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
eyed | +0.086 |
CBS | +0.085 |
atra | +0.084 |
didn | +0.084 |
photos | +0.081 |
Neg logit contributions | |
Proced | -0.106 |
yip | -0.102 |
Fn | -0.096 |
Address | -0.093 |
aning | -0.093 |
Token | big |
Token ID | 1263 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
eyed | +0.066 |
CBS | +0.065 |
didn | +0.065 |
atra | +0.064 |
hadn | +0.063 |
Neg logit contributions | |
Proced | -0.066 |
yip | -0.064 |
Fn | -0.059 |
Address | -0.058 |
aning | -0.058 |
Token | gie |
Token ID | 22699 |
Feature activation | +6.78e-01 |
Pos logit contributions | |
CBS | +0.114 |
eyed | +0.114 |
didn | +0.113 |
atra | +0.112 |
wasn | +0.112 |
Neg logit contributions | |
Proced | -0.022 |
yip | -0.019 |
Fn | -0.014 |
aning | -0.014 |
Address | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -6.77e-03 |
Pos logit contributions | |
Proced | +0.417 |
yip | +0.399 |
Fn | +0.361 |
Address | +0.357 |
aning | +0.356 |
Neg logit contributions | |
eyed | -0.497 |
didn | -0.490 |
CBS | -0.489 |
atra | -0.482 |
wasn | -0.476 |
Token | At |
Token ID | 1629 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
eyed | +0.005 |
CBS | +0.005 |
didn | +0.004 |
atra | +0.004 |
hadn | +0.004 |
Neg logit contributions | |
Proced | -0.005 |
yip | -0.004 |
Fn | -0.004 |
Address | -0.004 |
Finding | -0.004 |
Token | this |
Token ID | 428 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
eyed | +0.107 |
CBS | +0.106 |
didn | +0.105 |
atra | +0.103 |
photos | +0.102 |
Neg logit contributions | |
Proced | -0.132 |
yip | -0.127 |
Fn | -0.119 |
Address | -0.117 |
aning | -0.116 |
Token | moment |
Token ID | 2589 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
eyed | +0.014 |
didn | +0.013 |
CBS | +0.013 |
atra | +0.013 |
wasn | +0.013 |
Neg logit contributions | |
Proced | -0.015 |
yip | -0.015 |
Address | -0.014 |
Fn | -0.014 |
aning | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
Proced | +0.087 |
yip | +0.085 |
Fn | +0.078 |
Address | +0.077 |
aning | +0.076 |
Neg logit contributions | |
eyed | -0.082 |
CBS | -0.081 |
didn | -0.081 |
atra | -0.080 |
wasn | -0.078 |
Token | as |
Token ID | 355 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
Proced | +0.082 |
yip | +0.079 |
Address | +0.072 |
aning | +0.072 |
Fn | +0.071 |
Neg logit contributions | |
eyed | -0.081 |
didn | -0.080 |
CBS | -0.080 |
atra | -0.079 |
wasn | -0.078 |
Token | it |
Token ID | 340 |
Feature activation | -6.67e-03 |
Pos logit contributions | |
eyed | +0.020 |
didn | +0.019 |
CBS | +0.019 |
atra | +0.019 |
photos | +0.018 |
Neg logit contributions | |
Proced | -0.022 |
yip | -0.021 |
aning | -0.019 |
Fn | -0.019 |
Address | -0.019 |
Token | is |
Token ID | 318 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
eyed | +0.004 |
CBS | +0.004 |
didn | +0.004 |
atra | +0.004 |
photos | +0.004 |
Neg logit contributions | |
Proced | -0.005 |
yip | -0.005 |
aning | -0.004 |
Address | -0.004 |
Fn | -0.004 |
Token | dem |
Token ID | 1357 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
Proced | +0.013 |
yip | +0.013 |
aning | +0.012 |
Fn | +0.012 |
Address | +0.012 |
Neg logit contributions | |
eyed | -0.012 |
didn | -0.012 |
CBS | -0.011 |
atra | -0.011 |
photos | -0.011 |
Token | e |
Token ID | 68 |
Feature activation | +5.45e-02 |
Pos logit contributions | |
CBS | +0.084 |
didn | +0.084 |
eyed | +0.083 |
hadn | +0.082 |
wasn | +0.082 |
Neg logit contributions | |
Proced | -0.087 |
yip | -0.083 |
aning | -0.078 |
Address | -0.076 |
Fn | -0.076 |
Token | aning |
Token ID | 7574 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
Proced | +0.003 |
yip | +0.001 |
aning | +0.000 |
Address | -0.002 |
Agg | -0.002 |
Neg logit contributions | |
eyed | -0.071 |
CBS | -0.070 |
didn | -0.070 |
atra | -0.070 |
wasn | -0.069 |
Token | for |
Token ID | 329 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
Proced | +0.405 |
yip | +0.386 |
aning | +0.348 |
Fn | +0.346 |
Address | +0.346 |
Neg logit contributions | |
eyed | -0.501 |
didn | -0.496 |
CBS | -0.495 |
atra | -0.486 |
wasn | -0.483 |
Token | women |
Token ID | 1466 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
Proced | +0.108 |
yip | +0.104 |
Fn | +0.096 |
aning | +0.096 |
Address | +0.095 |
Neg logit contributions | |
eyed | -0.084 |
didn | -0.083 |
CBS | -0.083 |
atra | -0.082 |
photos | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | -4.80e-04 |
Pos logit contributions | |
eyed | +0.043 |
didn | +0.043 |
CBS | +0.043 |
atra | +0.042 |
wasn | +0.041 |
Neg logit contributions | |
Proced | -0.037 |
yip | -0.036 |
aning | -0.032 |
Fn | -0.032 |
Address | -0.032 |
Token | but |
Token ID | 475 |
Feature activation | -9.79e-02 |
Pos logit contributions | |
eyed | +0.000 |
CBS | +0.000 |
didn | +0.000 |
atra | +0.000 |
hadn | +0.000 |
Neg logit contributions | |
Proced | -0.000 |
yip | -0.000 |
aning | -0.000 |
Fn | -0.000 |
Address | -0.000 |
Token | according |
Token ID | 1864 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
eyed | +0.057 |
CBS | +0.056 |
didn | +0.056 |
atra | +0.056 |
photos | +0.054 |
Neg logit contributions | |
Proced | -0.075 |
yip | -0.072 |
Fn | -0.067 |
Agg | -0.066 |
aning | -0.066 |
Token | survivors |
Token ID | 13644 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
Proced | +0.045 |
yip | +0.042 |
Fn | +0.039 |
Agg | +0.039 |
Address | +0.039 |
Neg logit contributions | |
eyed | -0.040 |
didn | -0.040 |
CBS | -0.040 |
atra | -0.038 |
wasn | -0.038 |
Token | ' |
Token ID | 6 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
Proced | +0.084 |
yip | +0.079 |
Fn | +0.072 |
aning | +0.072 |
Address | +0.072 |
Neg logit contributions | |
eyed | -0.085 |
didn | -0.084 |
CBS | -0.084 |
atra | -0.082 |
wasn | -0.081 |
Token | skills |
Token ID | 4678 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
eyed | +0.086 |
didn | +0.085 |
CBS | +0.085 |
atra | +0.083 |
wasn | +0.081 |
Neg logit contributions | |
Proced | -0.098 |
yip | -0.090 |
Fn | -0.086 |
Address | -0.084 |
Agg | -0.084 |
Token | ( |
Token ID | 357 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
eyed | +0.163 |
didn | +0.162 |
CBS | +0.161 |
wasn | +0.157 |
atra | +0.157 |
Neg logit contributions | |
Proced | -0.162 |
yip | -0.155 |
Fn | -0.140 |
Address | -0.140 |
aning | -0.140 |
Token | f |
Token ID | 69 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
eyed | +0.007 |
didn | +0.007 |
CBS | +0.007 |
wasn | +0.007 |
atra | +0.007 |
Neg logit contributions | |
Proced | -0.009 |
yip | -0.009 |
Agg | -0.008 |
Address | -0.008 |
aning | -0.008 |
Token | ishing |
Token ID | 3929 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
eyed | +0.114 |
didn | +0.114 |
CBS | +0.114 |
wasn | +0.112 |
atra | +0.111 |
Neg logit contributions | |
Proced | -0.105 |
yip | -0.101 |
aning | -0.094 |
Agg | -0.092 |
Address | -0.092 |
Token | , |
Token ID | 11 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
Proced | +0.437 |
yip | +0.417 |
Fn | +0.381 |
Address | +0.377 |
aning | +0.377 |
Neg logit contributions | |
eyed | -0.440 |
didn | -0.436 |
CBS | -0.435 |
atra | -0.424 |
wasn | -0.418 |
Token | cooking |
Token ID | 10801 |
Feature activation | -3.75e-03 |
Pos logit contributions | |
Proced | +0.038 |
yip | +0.036 |
aning | +0.033 |
Fn | +0.033 |
Agg | +0.033 |
Neg logit contributions | |
eyed | -0.039 |
didn | -0.039 |
CBS | -0.039 |
atra | -0.038 |
wasn | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | +5.92e-02 |
Pos logit contributions | |
eyed | +0.002 |
didn | +0.002 |
CBS | +0.002 |
atra | +0.002 |
wasn | +0.002 |
Neg logit contributions | |
Proced | -0.003 |
yip | -0.002 |
Fn | -0.002 |
Address | -0.002 |
aning | -0.002 |
Token | etc |
Token ID | 3503 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
Proced | +0.040 |
yip | +0.038 |
aning | +0.035 |
Fn | +0.035 |
Agg | +0.034 |
Neg logit contributions | |
eyed | -0.040 |
didn | -0.040 |
CBS | -0.040 |
atra | -0.039 |
wasn | -0.038 |
Token | .), |
Token ID | 12179 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
eyed | +0.006 |
didn | +0.006 |
CBS | +0.006 |
wasn | +0.005 |
atra | +0.005 |
Neg logit contributions | |
Proced | -0.016 |
yip | -0.016 |
Address | -0.015 |
Fn | -0.015 |
Agg | -0.015 |
Token | - |
Token ID | 12 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
Proced | +0.077 |
yip | +0.073 |
Fn | +0.068 |
Address | +0.068 |
Agg | +0.067 |
Neg logit contributions | |
eyed | -0.065 |
didn | -0.064 |
CBS | -0.063 |
wasn | -0.062 |
atra | -0.062 |
Token | Super |
Token ID | 12442 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Proced | +0.009 |
yip | +0.009 |
Address | +0.008 |
Agg | +0.008 |
AMS | +0.008 |
Neg logit contributions | |
eyed | -0.010 |
didn | -0.010 |
wasn | -0.010 |
CBS | -0.010 |
hadn | -0.010 |
Token | bowl |
Token ID | 36859 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
Proced | +0.093 |
yip | +0.089 |
Fn | +0.084 |
Agg | +0.083 |
Address | +0.083 |
Neg logit contributions | |
eyed | -0.066 |
didn | -0.065 |
CBS | -0.064 |
hadn | -0.064 |
wasn | -0.063 |
Token | Advertising |
Token ID | 40113 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
Proced | +0.026 |
yip | +0.025 |
Fn | +0.023 |
Address | +0.023 |
Agg | +0.022 |
Neg logit contributions | |
eyed | -0.025 |
didn | -0.025 |
CBS | -0.025 |
atra | -0.024 |
wasn | -0.024 |
Token | - |
Token ID | 12 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Proced | +0.044 |
yip | +0.043 |
Address | +0.039 |
Fn | +0.039 |
Agg | +0.039 |
Neg logit contributions | |
eyed | -0.031 |
didn | -0.030 |
CBS | -0.030 |
wasn | -0.030 |
atra | -0.029 |
Token | Aust |
Token ID | 15160 |
Feature activation | +6.50e-01 |
Pos logit contributions | |
Proced | +0.074 |
yip | +0.070 |
Address | +0.065 |
Agg | +0.065 |
AMS | +0.063 |
Neg logit contributions | |
eyed | -0.083 |
didn | -0.081 |
wasn | -0.080 |
CBS | -0.080 |
hadn | -0.080 |
Token | rail |
Token ID | 30224 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
Proced | +0.498 |
yip | +0.482 |
aning | +0.443 |
Address | +0.443 |
Fn | +0.442 |
Neg logit contributions | |
eyed | -0.373 |
didn | -0.370 |
CBS | -0.366 |
wasn | -0.359 |
hadn | -0.357 |
Token | ian |
Token ID | 666 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
Proced | +0.015 |
yip | +0.014 |
Address | +0.013 |
Fn | +0.013 |
aning | +0.013 |
Neg logit contributions | |
eyed | -0.015 |
didn | -0.014 |
CBS | -0.014 |
wasn | -0.014 |
hadn | -0.014 |
Token | Teen |
Token ID | 21432 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
Proced | +0.204 |
yip | +0.193 |
Fn | +0.178 |
Address | +0.177 |
Agg | +0.176 |
Neg logit contributions | |
eyed | -0.195 |
didn | -0.193 |
CBS | -0.192 |
wasn | -0.188 |
hadn | -0.187 |
Token | Party |
Token ID | 3615 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
Proced | +0.253 |
yip | +0.241 |
Fn | +0.222 |
Agg | +0.220 |
Address | +0.220 |
Neg logit contributions | |
eyed | -0.234 |
didn | -0.232 |
CBS | -0.231 |
wasn | -0.226 |
hadn | -0.225 |
Token | - |
Token ID | 12 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
Proced | +0.143 |
yip | +0.139 |
Address | +0.128 |
Fn | +0.128 |
Agg | +0.128 |
Neg logit contributions | |
eyed | -0.089 |
didn | -0.088 |
CBS | -0.088 |
wasn | -0.086 |
atra | -0.085 |
Token | lett |
Token ID | 15503 |
Feature activation | -8.66e-05 |
Pos logit contributions | |
Proced | +0.182 |
yip | +0.174 |
aning | +0.165 |
Fn | +0.161 |
Address | +0.160 |
Neg logit contributions | |
CBS | -0.141 |
eyed | -0.141 |
didn | -0.139 |
wasn | -0.136 |
hadn | -0.135 |
Token | Foundation |
Token ID | 5693 |
Feature activation | +1.09e-03 |
Pos logit contributions | |
eyed | +0.000 |
CBS | +0.000 |
didn | +0.000 |
atra | +0.000 |
wasn | +0.000 |
Neg logit contributions | |
Proced | -0.000 |
yip | -0.000 |
Address | -0.000 |
Fn | -0.000 |
aning | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
Proced | +0.001 |
yip | +0.001 |
Address | +0.001 |
Fn | +0.001 |
Agg | +0.001 |
Neg logit contributions | |
eyed | -0.001 |
didn | -0.001 |
CBS | -0.001 |
atra | -0.001 |
wasn | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
Proced | +0.021 |
yip | +0.019 |
Address | +0.018 |
Agg | +0.018 |
aning | +0.017 |
Neg logit contributions | |
eyed | -0.027 |
didn | -0.027 |
CBS | -0.027 |
atra | -0.027 |
wasn | -0.026 |
Token | F |
Token ID | 37 |
Feature activation | -8.50e-02 |
Pos logit contributions | |
Proced | +0.016 |
yip | +0.015 |
Address | +0.014 |
Agg | +0.014 |
AMS | +0.014 |
Neg logit contributions | |
eyed | -0.017 |
didn | -0.016 |
wasn | -0.016 |
hadn | -0.016 |
atra | -0.016 |
Token | ounded |
Token ID | 6302 |
Feature activation | +6.48e-01 |
Pos logit contributions | |
eyed | +0.066 |
didn | +0.065 |
CBS | +0.065 |
wasn | +0.064 |
atra | +0.063 |
Neg logit contributions | |
Proced | -0.048 |
yip | -0.046 |
Address | -0.041 |
aning | -0.041 |
Agg | -0.041 |
Token | in |
Token ID | 287 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
Proced | +0.423 |
yip | +0.401 |
Address | +0.368 |
Fn | +0.365 |
Agg | +0.363 |
Neg logit contributions | |
eyed | -0.451 |
didn | -0.446 |
CBS | -0.442 |
atra | -0.435 |
wasn | -0.432 |
Token | 1966 |
Token ID | 19322 |
Feature activation | -6.50e-02 |
Pos logit contributions | |
eyed | +0.012 |
didn | +0.012 |
CBS | +0.012 |
atra | +0.011 |
hadn | +0.011 |
Neg logit contributions | |
Proced | -0.014 |
yip | -0.013 |
Fn | -0.013 |
Address | -0.012 |
aning | -0.012 |
Token | by |
Token ID | 416 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
eyed | +0.044 |
didn | +0.044 |
CBS | +0.044 |
wasn | +0.043 |
atra | +0.043 |
Neg logit contributions | |
Proced | -0.043 |
yip | -0.041 |
Address | -0.038 |
Fn | -0.037 |
aning | -0.037 |
Token | William |
Token ID | 3977 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
Proced | +0.080 |
yip | +0.076 |
Fn | +0.071 |
Address | +0.070 |
aning | +0.070 |
Neg logit contributions | |
eyed | -0.066 |
didn | -0.065 |
CBS | -0.064 |
atra | -0.063 |
wasn | -0.062 |
Token | Hew |
Token ID | 30446 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
Proced | +0.059 |
yip | +0.056 |
Fn | +0.052 |
Address | +0.051 |
aning | +0.051 |
Neg logit contributions | |
eyed | -0.062 |
didn | -0.061 |
CBS | -0.060 |
atra | -0.059 |
wasn | -0.059 |
Token | to |
Token ID | 284 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
eyed | +0.025 |
didn | +0.025 |
CBS | +0.025 |
atra | +0.025 |
wasn | +0.025 |
Neg logit contributions | |
Proced | -0.017 |
yip | -0.016 |
Address | -0.014 |
Fn | -0.014 |
aning | -0.014 |
Token | be |
Token ID | 307 |
Feature activation | +3.11e-03 |
Pos logit contributions | |
Proced | +0.015 |
yip | +0.014 |
Address | +0.013 |
aning | +0.013 |
Fn | +0.013 |
Neg logit contributions | |
eyed | -0.011 |
didn | -0.011 |
CBS | -0.011 |
photos | -0.011 |
wasn | -0.011 |
Token | part |
Token ID | 636 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
Proced | +0.002 |
yip | +0.002 |
Address | +0.002 |
aning | +0.002 |
Fn | +0.002 |
Neg logit contributions | |
eyed | -0.002 |
didn | -0.002 |
CBS | -0.002 |
wasn | -0.002 |
photos | -0.002 |
Token | of |
Token ID | 286 |
Feature activation | -8.57e-02 |
Pos logit contributions | |
didn | +0.080 |
eyed | +0.080 |
CBS | +0.079 |
atra | +0.078 |
wasn | +0.077 |
Neg logit contributions | |
Proced | -0.079 |
yip | -0.074 |
aning | -0.069 |
Address | -0.068 |
Fn | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
eyed | +0.046 |
didn | +0.046 |
CBS | +0.045 |
wasn | +0.044 |
atra | +0.044 |
Neg logit contributions | |
Proced | -0.070 |
yip | -0.067 |
Fn | -0.062 |
Address | -0.062 |
aning | -0.062 |
Token | Fab |
Token ID | 14236 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
eyed | +0.086 |
didn | +0.085 |
CBS | +0.085 |
wasn | +0.083 |
atra | +0.082 |
Neg logit contributions | |
Proced | -0.091 |
yip | -0.085 |
Fn | -0.079 |
Address | -0.079 |
aning | -0.078 |
Token | Cities |
Token ID | 20830 |
Feature activation | -5.86e-03 |
Pos logit contributions | |
Proced | +0.436 |
yip | +0.419 |
Address | +0.383 |
Agg | +0.381 |
Fn | +0.380 |
Neg logit contributions | |
eyed | -0.429 |
didn | -0.421 |
CBS | -0.420 |
atra | -0.411 |
wasn | -0.410 |
Token | movement |
Token ID | 3356 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
eyed | +0.004 |
didn | +0.004 |
CBS | +0.004 |
atra | +0.004 |
photos | +0.004 |
Neg logit contributions | |
Proced | -0.004 |
yip | -0.004 |
Address | -0.003 |
Fn | -0.003 |
Agg | -0.003 |
Token | is |
Token ID | 318 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
eyed | +0.062 |
CBS | +0.061 |
didn | +0.060 |
atra | +0.060 |
photos | +0.058 |
Neg logit contributions | |
Proced | -0.074 |
yip | -0.071 |
Address | -0.066 |
aning | -0.065 |
Fn | -0.065 |
Token | to |
Token ID | 284 |
Feature activation | +6.16e-02 |
Pos logit contributions | |
Proced | +0.027 |
yip | +0.026 |
Address | +0.023 |
Fn | +0.023 |
aning | +0.023 |
Neg logit contributions | |
eyed | -0.033 |
didn | -0.033 |
CBS | -0.033 |
atra | -0.032 |
wasn | -0.032 |
Token | produce |
Token ID | 4439 |
Feature activation | +8.34e-02 |
Pos logit contributions | |
Proced | +0.041 |
yip | +0.039 |
aning | +0.036 |
Address | +0.035 |
Fn | +0.035 |
Neg logit contributions | |
eyed | -0.043 |
didn | -0.043 |
CBS | -0.042 |
photos | -0.041 |
atra | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
Proced | +0.066 |
yip | +0.063 |
Fn | +0.058 |
Address | +0.058 |
Agg | +0.057 |
Neg logit contributions | |
eyed | -0.058 |
CBS | -0.057 |
didn | -0.057 |
atra | -0.056 |
wasn | -0.055 |
Token | cant |
Token ID | 18548 |
Feature activation | -5.48e-03 |
Pos logit contributions | |
eyed | +0.022 |
didn | +0.021 |
CBS | +0.021 |
atra | +0.021 |
wasn | +0.021 |
Neg logit contributions | |
Proced | -0.016 |
yip | -0.015 |
Address | -0.014 |
Fn | -0.014 |
aning | -0.014 |
Token | een |
Token ID | 6429 |
Feature activation | +8.53e-02 |
Pos logit contributions | |
CBS | +0.005 |
eyed | +0.005 |
didn | +0.005 |
atra | +0.004 |
wasn | +0.004 |
Neg logit contributions | |
Proced | -0.003 |
yip | -0.003 |
aning | -0.002 |
Address | -0.002 |
Fn | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
Proced | +0.069 |
yip | +0.067 |
Address | +0.063 |
Fn | +0.062 |
aning | +0.062 |
Neg logit contributions | |
eyed | -0.045 |
didn | -0.044 |
CBS | -0.044 |
atra | -0.043 |
photos | -0.042 |
Token | mill |
Token ID | 3939 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
eyed | +0.071 |
didn | +0.071 |
CBS | +0.070 |
atra | +0.069 |
photos | +0.068 |
Neg logit contributions | |
Proced | -0.076 |
yip | -0.072 |
Address | -0.068 |
aning | -0.067 |
Fn | -0.067 |
Token | ing |
Token ID | 278 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
Proced | +0.056 |
yip | +0.053 |
Address | +0.045 |
aning | +0.045 |
Fn | +0.044 |
Neg logit contributions | |
eyed | -0.123 |
didn | -0.122 |
CBS | -0.122 |
atra | -0.120 |
wasn | -0.120 |
Token | with |
Token ID | 351 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
Proced | +0.445 |
yip | +0.423 |
Address | +0.391 |
aning | +0.387 |
Agg | +0.384 |
Neg logit contributions | |
didn | -0.411 |
eyed | -0.411 |
CBS | -0.410 |
atra | -0.405 |
wasn | -0.396 |
Token | activity |
Token ID | 3842 |
Feature activation | -3.77e-03 |
Pos logit contributions | |
Proced | +0.070 |
yip | +0.067 |
Address | +0.062 |
Fn | +0.062 |
aning | +0.062 |
Neg logit contributions | |
eyed | -0.060 |
CBS | -0.060 |
didn | -0.060 |
atra | -0.059 |
wasn | -0.057 |
Token | when |
Token ID | 618 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
eyed | +0.003 |
didn | +0.003 |
CBS | +0.003 |
atra | +0.003 |
wasn | +0.002 |
Neg logit contributions | |
Proced | -0.002 |
yip | -0.002 |
Address | -0.002 |
aning | -0.002 |
Fn | -0.002 |
Token | PM |
Token ID | 3122 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
eyed | +0.015 |
CBS | +0.015 |
didn | +0.015 |
atra | +0.015 |
wasn | +0.014 |
Neg logit contributions | |
Proced | -0.015 |
yip | -0.014 |
Fn | -0.013 |
Address | -0.013 |
Agg | -0.013 |
Token | Modi |
Token ID | 14637 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
Proced | +0.169 |
yip | +0.159 |
Address | +0.146 |
Fn | +0.145 |
Agg | +0.143 |
Neg logit contributions | |
eyed | -0.220 |
CBS | -0.216 |
didn | -0.216 |
atra | -0.211 |
photos | -0.210 |
Token | Dar |
Token ID | 32708 |
Feature activation | +4.65e-02 |
Pos logit contributions | |
eyed | +0.100 |
atra | +0.097 |
didn | +0.097 |
CBS | +0.097 |
wasn | +0.096 |
Neg logit contributions | |
Proced | -0.081 |
yip | -0.076 |
Address | -0.073 |
Agg | -0.072 |
Finding | -0.070 |
Token | ren |
Token ID | 918 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
Proced | +0.025 |
yip | +0.024 |
aning | +0.022 |
Address | +0.021 |
Agg | +0.021 |
Neg logit contributions | |
eyed | -0.037 |
CBS | -0.036 |
didn | -0.036 |
atra | -0.036 |
wasn | -0.035 |
Token | must |
Token ID | 1276 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
eyed | +0.013 |
CBS | +0.012 |
didn | +0.012 |
atra | +0.012 |
photos | +0.012 |
Neg logit contributions | |
Proced | -0.018 |
yip | -0.017 |
Fn | -0.016 |
aning | -0.016 |
Address | -0.016 |
Token | be |
Token ID | 307 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
Proced | +0.022 |
yip | +0.021 |
aning | +0.020 |
Address | +0.019 |
Fn | +0.019 |
Neg logit contributions | |
eyed | -0.024 |
CBS | -0.024 |
didn | -0.024 |
photos | -0.023 |
atra | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | +4.46e-03 |
Pos logit contributions | |
Proced | +0.019 |
yip | +0.018 |
aning | +0.016 |
Fn | +0.016 |
Address | +0.016 |
Neg logit contributions | |
eyed | -0.020 |
CBS | -0.019 |
didn | -0.019 |
atra | -0.019 |
photos | -0.019 |
Token | Fab |
Token ID | 14236 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
Proced | +0.003 |
yip | +0.003 |
Fn | +0.003 |
aning | +0.003 |
Address | +0.003 |
Neg logit contributions | |
eyed | -0.003 |
CBS | -0.003 |
didn | -0.003 |
atra | -0.003 |
wasn | -0.003 |
Token | io |
Token ID | 952 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
Proced | +0.434 |
yip | +0.417 |
Address | +0.380 |
Agg | +0.378 |
Fn | +0.378 |
Neg logit contributions | |
eyed | -0.428 |
CBS | -0.421 |
didn | -0.421 |
atra | -0.412 |
wasn | -0.411 |
Token | Cape |
Token ID | 15725 |
Feature activation | +9.56e-02 |
Pos logit contributions | |
Proced | +0.035 |
yip | +0.034 |
Address | +0.031 |
Fn | +0.031 |
Agg | +0.031 |
Neg logit contributions | |
eyed | -0.020 |
didn | -0.020 |
CBS | -0.020 |
atra | -0.019 |
wasn | -0.019 |
Token | llo |
Token ID | 18798 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
Proced | +0.067 |
yip | +0.064 |
aning | +0.059 |
Address | +0.059 |
Fn | +0.058 |
Neg logit contributions | |
eyed | -0.062 |
CBS | -0.061 |
didn | -0.061 |
atra | -0.059 |
wasn | -0.059 |
Token | 's |
Token ID | 338 |
Feature activation | +7.75e-02 |
Pos logit contributions | |
Proced | +0.103 |
yip | +0.100 |
Address | +0.093 |
Fn | +0.093 |
aning | +0.093 |
Neg logit contributions | |
eyed | -0.058 |
CBS | -0.057 |
didn | -0.057 |
atra | -0.056 |
wasn | -0.055 |
Token | thoughts |
Token ID | 6066 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
Proced | +0.059 |
yip | +0.055 |
Fn | +0.052 |
Address | +0.052 |
aning | +0.052 |
Neg logit contributions | |
CBS | -0.046 |
eyed | -0.046 |
didn | -0.045 |
atra | -0.045 |
photos | -0.043 |
Token | ,[ |
Token ID | 17414 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
Proced | +0.004 |
yip | +0.004 |
Fn | +0.003 |
Address | +0.003 |
aning | +0.003 |
Neg logit contributions | |
eyed | -0.004 |
didn | -0.004 |
CBS | -0.004 |
atra | -0.004 |
wasn | -0.004 |
Token | 9 |
Token ID | 24 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
Proced | +0.189 |
yip | +0.182 |
Address | +0.167 |
Fn | +0.167 |
aning | +0.166 |
Neg logit contributions | |
eyed | -0.146 |
didn | -0.144 |
CBS | -0.144 |
wasn | -0.141 |
atra | -0.140 |
Token | ] |
Token ID | 60 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
didn | +0.043 |
eyed | +0.043 |
wasn | +0.042 |
CBS | +0.042 |
hadn | +0.041 |
Neg logit contributions | |
Proced | -0.114 |
yip | -0.112 |
Address | -0.104 |
Agg | -0.104 |
aning | -0.104 |
Token | and |
Token ID | 290 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
eyed | +0.067 |
didn | +0.066 |
CBS | +0.066 |
atra | +0.064 |
wasn | +0.064 |
Neg logit contributions | |
Proced | -0.071 |
yip | -0.068 |
Fn | -0.062 |
Address | -0.062 |
aning | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | -2.46e-03 |
Pos logit contributions | |
Proced | +0.027 |
yip | +0.025 |
Fn | +0.023 |
Address | +0.023 |
Agg | +0.023 |
Neg logit contributions | |
eyed | -0.023 |
CBS | -0.022 |
didn | -0.022 |
atra | -0.022 |
photos | -0.021 |
Token | CRC |
Token ID | 45623 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
eyed | +0.002 |
didn | +0.002 |
CBS | +0.002 |
atra | +0.002 |
wasn | +0.002 |
Neg logit contributions | |
Proced | -0.002 |
yip | -0.002 |
Fn | -0.001 |
Address | -0.001 |
aning | -0.001 |
Token | - |
Token ID | 12 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
Proced | +0.407 |
yip | +0.389 |
Address | +0.354 |
Fn | +0.352 |
Agg | +0.348 |
Neg logit contributions | |
eyed | -0.453 |
didn | -0.447 |
CBS | -0.444 |
atra | -0.436 |
wasn | -0.433 |
Token | 32 |
Token ID | 2624 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
eyed | +0.030 |
didn | +0.030 |
CBS | +0.030 |
atra | +0.029 |
wasn | +0.029 |
Neg logit contributions | |
Proced | -0.024 |
yip | -0.023 |
Address | -0.021 |
Agg | -0.021 |
Fn | -0.021 |
Token | checks |
Token ID | 8794 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
Proced | +0.102 |
yip | +0.098 |
Address | +0.087 |
Fn | +0.086 |
Agg | +0.086 |
Neg logit contributions | |
eyed | -0.138 |
didn | -0.136 |
CBS | -0.136 |
wasn | -0.134 |
atra | -0.133 |
Token | um |
Token ID | 388 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
Proced | +0.105 |
yip | +0.100 |
Address | +0.089 |
Fn | +0.089 |
Agg | +0.088 |
Neg logit contributions | |
eyed | -0.139 |
didn | -0.137 |
CBS | -0.137 |
wasn | -0.135 |
atra | -0.134 |
Token | for |
Token ID | 329 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
eyed | +0.073 |
didn | +0.072 |
CBS | +0.072 |
wasn | +0.071 |
atra | +0.071 |
Neg logit contributions | |
Proced | -0.062 |
yip | -0.059 |
Address | -0.053 |
Fn | -0.053 |
Agg | -0.053 |
Token | Old |
Token ID | 5706 |
Feature activation | +8.05e-02 |
Pos logit contributions | |
Proced | +0.118 |
yip | +0.111 |
Fn | +0.103 |
Address | +0.102 |
Agg | +0.102 |
Neg logit contributions | |
eyed | -0.109 |
didn | -0.108 |
CBS | -0.108 |
wasn | -0.105 |
hadn | -0.104 |
Token | - |
Token ID | 12 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
Proced | +0.057 |
yip | +0.055 |
Fn | +0.051 |
Address | +0.050 |
aning | +0.050 |
Neg logit contributions | |
eyed | -0.051 |
didn | -0.050 |
CBS | -0.050 |
wasn | -0.049 |
hadn | -0.049 |
Token | E |
Token ID | 36 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
Proced | +0.015 |
yip | +0.014 |
Address | +0.013 |
Agg | +0.013 |
Fn | +0.013 |
Neg logit contributions | |
eyed | -0.017 |
didn | -0.017 |
CBS | -0.017 |
wasn | -0.017 |
hadn | -0.017 |
Token | pt |
Token ID | 457 |
Feature activation | -7.84e-02 |
Pos logit contributions | |
Proced | +0.090 |
yip | +0.086 |
aning | +0.079 |
Address | +0.078 |
Fn | +0.078 |
Neg logit contributions | |
eyed | -0.087 |
didn | -0.087 |
CBS | -0.087 |
wasn | -0.085 |
atra | -0.085 |
Token | sch |
Token ID | 20601 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
eyed | +0.044 |
didn | +0.043 |
CBS | +0.043 |
wasn | +0.042 |
atra | +0.042 |
Neg logit contributions | |
Proced | -0.062 |
yip | -0.060 |
Address | -0.055 |
Fn | -0.055 |
Agg | -0.055 |
Token | tein |
Token ID | 22006 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
Proced | +0.054 |
yip | +0.050 |
Address | +0.041 |
Fn | +0.041 |
aning | +0.040 |
Neg logit contributions | |
eyed | -0.147 |
didn | -0.146 |
CBS | -0.146 |
wasn | -0.144 |
atra | -0.144 |
Token | Pt |
Token ID | 34352 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
Proced | +0.493 |
yip | +0.474 |
Fn | +0.439 |
Address | +0.438 |
aning | +0.435 |
Neg logit contributions | |
eyed | -0.354 |
didn | -0.348 |
CBS | -0.347 |
wasn | -0.338 |
atra | -0.336 |
Token | 3 |
Token ID | 18 |
Feature activation | -3.26e-03 |
Pos logit contributions | |
eyed | +0.124 |
didn | +0.123 |
CBS | +0.122 |
wasn | +0.120 |
atra | +0.120 |
Neg logit contributions | |
Proced | -0.100 |
yip | -0.095 |
Fn | -0.085 |
Address | -0.085 |
aning | -0.085 |
Token | - |
Token ID | 12 |
Feature activation | +6.34e-02 |
Pos logit contributions | |
eyed | +0.002 |
CBS | +0.002 |
didn | +0.002 |
wasn | +0.002 |
atra | +0.002 |
Neg logit contributions | |
Proced | -0.003 |
yip | -0.002 |
Address | -0.002 |
Fn | -0.002 |
Agg | -0.002 |
Token | Ant |
Token ID | 13217 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
Proced | +0.039 |
yip | +0.037 |
Address | +0.035 |
Agg | +0.034 |
aning | +0.034 |
Neg logit contributions | |
eyed | -0.046 |
didn | -0.045 |
CBS | -0.044 |
wasn | -0.044 |
hadn | -0.044 |
Token | s |
Token ID | 82 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
Proced | +0.286 |
yip | +0.274 |
Fn | +0.254 |
Address | +0.253 |
Agg | +0.252 |
Neg logit contributions | |
eyed | -0.215 |
didn | -0.213 |
CBS | -0.212 |
wasn | -0.206 |
hadn | -0.206 |
Token | at |
Token ID | 379 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
Proced | +0.095 |
yip | +0.092 |
Address | +0.084 |
Fn | +0.083 |
aning | +0.083 |
Neg logit contributions | |
CBS | -0.085 |
eyed | -0.085 |
didn | -0.084 |
atra | -0.082 |
photos | -0.082 |
Token | work |
Token ID | 670 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
eyed | +0.107 |
didn | +0.106 |
CBS | +0.106 |
atra | +0.104 |
wasn | +0.103 |
Neg logit contributions | |
Proced | -0.130 |
yip | -0.125 |
Fn | -0.117 |
Address | -0.115 |
Agg | -0.114 |
Token | . |
Token ID | 13 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
eyed | +0.092 |
CBS | +0.091 |
didn | +0.091 |
atra | +0.089 |
wasn | +0.088 |
Neg logit contributions | |
Proced | -0.096 |
yip | -0.092 |
Fn | -0.084 |
Address | -0.083 |
aning | -0.083 |
Token | No |
Token ID | 1400 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
eyed | +0.040 |
CBS | +0.039 |
didn | +0.039 |
atra | +0.038 |
wasn | +0.037 |
Neg logit contributions | |
Proced | -0.040 |
yip | -0.037 |
Fn | -0.035 |
Address | -0.035 |
Finding | -0.034 |
Token | big |
Token ID | 1263 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
eyed | +0.237 |
didn | +0.234 |
CBS | +0.234 |
atra | +0.231 |
wasn | +0.227 |
Neg logit contributions | |
Proced | -0.234 |
yip | -0.223 |
Fn | -0.206 |
aning | -0.203 |
Address | -0.203 |
Token | gie |
Token ID | 22699 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
eyed | +0.141 |
CBS | +0.140 |
didn | +0.140 |
atra | +0.138 |
wasn | +0.137 |
Neg logit contributions | |
Proced | -0.088 |
yip | -0.083 |
Fn | -0.075 |
aning | -0.074 |
Address | -0.074 |
Token | , |
Token ID | 11 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
Proced | +0.393 |
yip | +0.375 |
Fn | +0.341 |
Address | +0.339 |
aning | +0.338 |
Neg logit contributions | |
eyed | -0.449 |
didn | -0.445 |
CBS | -0.442 |
atra | -0.436 |
wasn | -0.431 |
Token | because |
Token ID | 780 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
eyed | +0.010 |
CBS | +0.010 |
didn | +0.010 |
atra | +0.010 |
photos | +0.010 |
Neg logit contributions | |
Proced | -0.014 |
yip | -0.014 |
Fn | -0.013 |
aning | -0.013 |
Address | -0.013 |
Token | I |
Token ID | 314 |
Feature activation | -7.13e-02 |
Pos logit contributions | |
eyed | +0.120 |
CBS | +0.118 |
didn | +0.117 |
atra | +0.116 |
photos | +0.113 |
Neg logit contributions | |
Proced | -0.148 |
yip | -0.142 |
Fn | -0.133 |
Address | -0.130 |
aning | -0.130 |
Token | got |
Token ID | 1392 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
eyed | +0.044 |
CBS | +0.043 |
didn | +0.042 |
atra | +0.042 |
photos | +0.041 |
Neg logit contributions | |
Proced | -0.052 |
yip | -0.049 |
aning | -0.046 |
Fn | -0.046 |
Address | -0.046 |
Token | your |
Token ID | 534 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
eyed | +0.062 |
didn | +0.061 |
CBS | +0.061 |
atra | +0.059 |
wasn | +0.058 |
Neg logit contributions | |
Proced | -0.074 |
yip | -0.071 |
Fn | -0.066 |
Address | -0.065 |
Agg | -0.065 |
Token | Recap |
Token ID | 46585 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
Proced | +0.039 |
yip | +0.037 |
Fn | +0.034 |
Address | +0.034 |
aning | +0.034 |
Neg logit contributions | |
eyed | -0.033 |
didn | -0.033 |
CBS | -0.032 |
wasn | -0.032 |
atra | -0.032 |
Token | - |
Token ID | 12 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
Proced | +0.094 |
yip | +0.090 |
Fn | +0.083 |
Address | +0.083 |
aning | +0.082 |
Neg logit contributions | |
eyed | -0.073 |
didn | -0.072 |
CBS | -0.071 |
wasn | -0.070 |
atra | -0.069 |
Token | E |
Token ID | 36 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Proced | +0.058 |
yip | +0.055 |
Address | +0.051 |
Agg | +0.051 |
Fn | +0.050 |
Neg logit contributions | |
eyed | -0.062 |
didn | -0.061 |
CBS | -0.060 |
wasn | -0.060 |
atra | -0.060 |
Token | ps |
Token ID | 862 |
Feature activation | +6.53e-02 |
Pos logit contributions | |
Proced | +0.090 |
yip | +0.086 |
aning | +0.079 |
Address | +0.079 |
Fn | +0.078 |
Neg logit contributions | |
eyed | -0.092 |
didn | -0.091 |
CBS | -0.091 |
wasn | -0.089 |
atra | -0.089 |
Token | ch |
Token ID | 354 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
Proced | +0.040 |
yip | +0.038 |
Fn | +0.035 |
Address | +0.035 |
Agg | +0.035 |
Neg logit contributions | |
eyed | -0.048 |
didn | -0.047 |
CBS | -0.047 |
wasn | -0.046 |
hadn | -0.046 |
Token | tein |
Token ID | 22006 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
eyed | +0.007 |
didn | +0.006 |
CBS | +0.006 |
wasn | +0.006 |
hadn | +0.006 |
Neg logit contributions | |
Proced | -0.008 |
yip | -0.007 |
aning | -0.007 |
Fn | -0.007 |
Address | -0.007 |
Token | Pt |
Token ID | 34352 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
Proced | +0.424 |
yip | +0.404 |
Fn | +0.373 |
Address | +0.371 |
aning | +0.369 |
Neg logit contributions | |
eyed | -0.407 |
didn | -0.403 |
CBS | -0.400 |
wasn | -0.392 |
atra | -0.390 |
Token | 1 |
Token ID | 352 |
Feature activation | +6.91e-02 |
Pos logit contributions | |
eyed | +0.084 |
didn | +0.083 |
CBS | +0.083 |
wasn | +0.081 |
atra | +0.081 |
Neg logit contributions | |
Proced | -0.071 |
yip | -0.067 |
Address | -0.061 |
Fn | -0.061 |
Agg | -0.060 |
Token | - |
Token ID | 12 |
Feature activation | +9.52e-02 |
Pos logit contributions | |
Proced | +0.054 |
yip | +0.052 |
Address | +0.048 |
Fn | +0.048 |
Agg | +0.048 |
Neg logit contributions | |
eyed | -0.039 |
didn | -0.038 |
CBS | -0.038 |
atra | -0.037 |
wasn | -0.037 |
Token | Greg |
Token ID | 25025 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
Proced | +0.061 |
yip | +0.058 |
Address | +0.054 |
Agg | +0.053 |
Fn | +0.053 |
Neg logit contributions | |
eyed | -0.067 |
didn | -0.066 |
CBS | -0.065 |
wasn | -0.065 |
atra | -0.065 |
Token | orio |
Token ID | 40974 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
Proced | +0.128 |
yip | +0.123 |
Fn | +0.114 |
Address | +0.113 |
aning | +0.113 |
Neg logit contributions | |
eyed | -0.114 |
didn | -0.112 |
CBS | -0.112 |
wasn | -0.109 |
atra | -0.109 |
Token | Guy |
Token ID | 13145 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
Proced | +0.352 |
yip | +0.337 |
Address | +0.312 |
Fn | +0.311 |
aning | +0.310 |
Neg logit contributions | |
eyed | -0.265 |
didn | -0.263 |
CBS | -0.261 |
wasn | -0.256 |
hadn | -0.254 |
Token | - |
Token ID | 12 |
Feature activation | +5.49e-02 |
Pos logit contributions | |
Proced | +0.154 |
yip | +0.149 |
Address | +0.138 |
Fn | +0.137 |
Agg | +0.137 |
Neg logit contributions | |
eyed | -0.105 |
didn | -0.104 |
CBS | -0.103 |
wasn | -0.101 |
atra | -0.101 |
Token | E |
Token ID | 36 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
Proced | +0.035 |
yip | +0.033 |
Address | +0.031 |
Agg | +0.031 |
Fn | +0.030 |
Neg logit contributions | |
eyed | -0.039 |
didn | -0.038 |
wasn | -0.038 |
CBS | -0.038 |
atra | -0.038 |
Token | ps |
Token ID | 862 |
Feature activation | +6.72e-03 |
Pos logit contributions | |
Proced | +0.089 |
yip | +0.085 |
aning | +0.076 |
Address | +0.076 |
Fn | +0.075 |
Neg logit contributions | |
eyed | -0.119 |
didn | -0.118 |
CBS | -0.118 |
wasn | -0.116 |
atra | -0.116 |
Token | ch |
Token ID | 354 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
Proced | +0.005 |
yip | +0.005 |
Fn | +0.005 |
Address | +0.005 |
aning | +0.004 |
Neg logit contributions | |
eyed | -0.004 |
didn | -0.004 |
CBS | -0.004 |
wasn | -0.004 |
hadn | -0.004 |
Token | tein |
Token ID | 22006 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
eyed | +0.047 |
didn | +0.047 |
CBS | +0.047 |
wasn | +0.047 |
hadn | +0.047 |
Neg logit contributions | |
Proced | -0.007 |
yip | -0.006 |
aning | -0.004 |
Address | -0.004 |
Agg | -0.004 |
Token | Pt |
Token ID | 34352 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
Proced | +0.495 |
yip | +0.475 |
Fn | +0.442 |
Address | +0.441 |
aning | +0.439 |
Neg logit contributions | |
eyed | -0.328 |
didn | -0.323 |
CBS | -0.321 |
wasn | -0.312 |
atra | -0.311 |
Token | 2 |
Token ID | 362 |
Feature activation | +7.74e-02 |
Pos logit contributions | |
eyed | +0.122 |
didn | +0.121 |
CBS | +0.120 |
wasn | +0.118 |
atra | +0.118 |
Neg logit contributions | |
Proced | -0.104 |
yip | -0.099 |
Address | -0.089 |
Fn | -0.089 |
Agg | -0.089 |
Token | 2008 |
Token ID | 3648 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
Proced | +0.065 |
yip | +0.063 |
Address | +0.058 |
Fn | +0.058 |
Agg | +0.058 |
Neg logit contributions | |
eyed | -0.040 |
CBS | -0.039 |
didn | -0.039 |
wasn | -0.038 |
atra | -0.038 |
Token | 01 |
Token ID | 486 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
Proced | +0.032 |
yip | +0.030 |
Address | +0.028 |
Agg | +0.028 |
Fn | +0.028 |
Neg logit contributions | |
eyed | -0.029 |
didn | -0.029 |
CBS | -0.029 |
atra | -0.028 |
wasn | -0.028 |
Token | 10 |
Token ID | 940 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
Proced | +0.020 |
yip | +0.020 |
Fn | +0.018 |
aning | +0.018 |
Address | +0.018 |
Neg logit contributions | |
eyed | -0.021 |
CBS | -0.021 |
didn | -0.021 |
wasn | -0.021 |
hadn | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | +7.91e-02 |
Pos logit contributions | |
Proced | +0.140 |
yip | +0.132 |
Fn | +0.123 |
aning | +0.122 |
Address | +0.122 |
Neg logit contributions | |
eyed | -0.138 |
didn | -0.137 |
CBS | -0.136 |
atra | -0.133 |
wasn | -0.131 |
Token | three |
Token ID | 1115 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
Proced | +0.054 |
yip | +0.051 |
Fn | +0.048 |
aning | +0.047 |
Address | +0.047 |
Neg logit contributions | |
eyed | -0.053 |
didn | -0.052 |
CBS | -0.052 |
atra | -0.050 |
wasn | -0.050 |
Token | - |
Token ID | 12 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
Proced | +0.035 |
yip | +0.033 |
Fn | +0.031 |
Address | +0.031 |
aning | +0.030 |
Neg logit contributions | |
eyed | -0.032 |
didn | -0.032 |
CBS | -0.032 |
atra | -0.031 |
wasn | -0.031 |
Token | quarter |
Token ID | 24385 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
Proced | +0.026 |
yip | +0.025 |
Address | +0.023 |
aning | +0.023 |
Fn | +0.022 |
Neg logit contributions | |
CBS | -0.024 |
didn | -0.024 |
eyed | -0.024 |
atra | -0.023 |
wasn | -0.023 |
Token | - |
Token ID | 12 |
Feature activation | +6.14e-02 |
Pos logit contributions | |
Proced | +0.021 |
yip | +0.021 |
Fn | +0.019 |
Address | +0.019 |
aning | +0.019 |
Neg logit contributions | |
eyed | -0.022 |
didn | -0.022 |
CBS | -0.022 |
atra | -0.022 |
wasn | -0.022 |
Token | length |
Token ID | 13664 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
Proced | +0.036 |
yip | +0.035 |
Address | +0.031 |
aning | +0.031 |
Fn | +0.031 |
Neg logit contributions | |
eyed | -0.046 |
didn | -0.046 |
CBS | -0.046 |
wasn | -0.045 |
atra | -0.045 |
Token | sleeves |
Token ID | 27409 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
Proced | +0.423 |
yip | +0.402 |
Fn | +0.374 |
Address | +0.369 |
aning | +0.368 |
Neg logit contributions | |
eyed | -0.395 |
didn | -0.393 |
CBS | -0.392 |
atra | -0.381 |
wasn | -0.376 |
Token | , |
Token ID | 11 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
Proced | +0.087 |
yip | +0.084 |
Address | +0.076 |
Fn | +0.076 |
aning | +0.076 |
Neg logit contributions | |
eyed | -0.086 |
didn | -0.085 |
CBS | -0.085 |
wasn | -0.083 |
atra | -0.083 |
Token | a |
Token ID | 257 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
Proced | +0.076 |
yip | +0.072 |
Fn | +0.067 |
Address | +0.066 |
aning | +0.066 |
Neg logit contributions | |
eyed | -0.072 |
didn | -0.071 |
CBS | -0.071 |
atra | -0.069 |
wasn | -0.069 |
Token | high |
Token ID | 1029 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
Proced | +0.083 |
yip | +0.078 |
Fn | +0.073 |
Address | +0.071 |
aning | +0.071 |
Neg logit contributions | |
eyed | -0.080 |
didn | -0.079 |
CBS | -0.079 |
atra | -0.077 |
wasn | -0.077 |
Token | neck |
Token ID | 7393 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
Proced | +0.106 |
yip | +0.102 |
Fn | +0.094 |
aning | +0.094 |
Address | +0.094 |
Neg logit contributions | |
eyed | -0.082 |
CBS | -0.081 |
didn | -0.081 |
atra | -0.078 |
wasn | -0.078 |
Token | of |
Token ID | 286 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
Proced | +0.218 |
yip | +0.210 |
Address | +0.191 |
Fn | +0.190 |
Agg | +0.189 |
Neg logit contributions | |
eyed | -0.207 |
didn | -0.204 |
CBS | -0.203 |
wasn | -0.200 |
atra | -0.199 |
Token | basketball |
Token ID | 9669 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
Proced | +0.032 |
yip | +0.031 |
Address | +0.029 |
Fn | +0.029 |
Finding | +0.029 |
Neg logit contributions | |
eyed | -0.028 |
didn | -0.027 |
CBS | -0.027 |
atra | -0.027 |
wasn | -0.026 |
Token | on |
Token ID | 319 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
eyed | +0.118 |
didn | +0.117 |
CBS | +0.116 |
atra | +0.114 |
wasn | +0.111 |
Neg logit contributions | |
Proced | -0.108 |
yip | -0.103 |
Fn | -0.095 |
Address | -0.095 |
aning | -0.094 |
Token | Al |
Token ID | 978 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
Proced | +0.043 |
yip | +0.040 |
Address | +0.038 |
Fn | +0.038 |
aning | +0.037 |
Neg logit contributions | |
eyed | -0.037 |
didn | -0.036 |
CBS | -0.036 |
atra | -0.035 |
wasn | -0.035 |
Token | cat |
Token ID | 9246 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
Proced | +0.082 |
yip | +0.080 |
Address | +0.072 |
aning | +0.071 |
Agg | +0.071 |
Neg logit contributions | |
eyed | -0.123 |
didn | -0.122 |
CBS | -0.122 |
wasn | -0.120 |
hadn | -0.120 |
Token | raz |
Token ID | 3247 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
Proced | +0.066 |
yip | +0.061 |
aning | +0.053 |
Agg | +0.052 |
Address | +0.052 |
Neg logit contributions | |
eyed | -0.170 |
didn | -0.169 |
CBS | -0.169 |
wasn | -0.167 |
hadn | -0.166 |
Token | . |
Token ID | 13 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
Proced | +0.381 |
yip | +0.361 |
Address | +0.331 |
Fn | +0.327 |
aning | +0.325 |
Neg logit contributions | |
eyed | -0.435 |
didn | -0.431 |
CBS | -0.428 |
atra | -0.421 |
wasn | -0.416 |
Token | Only |
Token ID | 5514 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
Proced | +0.084 |
yip | +0.076 |
Address | +0.073 |
Fn | +0.073 |
Agg | +0.072 |
Neg logit contributions | |
eyed | -0.095 |
didn | -0.093 |
CBS | -0.092 |
atra | -0.090 |
wasn | -0.090 |
Token | the |
Token ID | 262 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
eyed | +0.076 |
CBS | +0.074 |
didn | +0.074 |
atra | +0.073 |
turn | +0.071 |
Neg logit contributions | |
Proced | -0.083 |
yip | -0.079 |
Address | -0.074 |
Fn | -0.073 |
aning | -0.073 |
Token | beginning |
Token ID | 3726 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
Proced | +0.025 |
yip | +0.023 |
Address | +0.022 |
Fn | +0.021 |
aning | +0.021 |
Neg logit contributions | |
eyed | -0.035 |
didn | -0.035 |
CBS | -0.035 |
atra | -0.034 |
wasn | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +8.66e-02 |
Pos logit contributions | |
Proced | +0.214 |
yip | +0.205 |
Address | +0.186 |
Fn | +0.185 |
aning | +0.185 |
Neg logit contributions | |
eyed | -0.224 |
didn | -0.221 |
CBS | -0.221 |
atra | -0.216 |
wasn | -0.216 |
Token | , |
Token ID | 11 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
eyed | +0.067 |
didn | +0.067 |
CBS | +0.066 |
atra | +0.066 |
wasn | +0.064 |
Neg logit contributions | |
Proced | -0.063 |
yip | -0.060 |
aning | -0.055 |
Address | -0.054 |
Fn | -0.054 |
Token | or |
Token ID | 393 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
Proced | +0.038 |
yip | +0.036 |
aning | +0.034 |
Fn | +0.033 |
Address | +0.033 |
Neg logit contributions | |
eyed | -0.030 |
CBS | -0.029 |
didn | -0.029 |
atra | -0.028 |
photos | -0.028 |
Token | am |
Token ID | 716 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
eyed | +0.026 |
CBS | +0.025 |
didn | +0.025 |
atra | +0.025 |
photos | +0.024 |
Neg logit contributions | |
Proced | -0.037 |
yip | -0.034 |
aning | -0.033 |
Fn | -0.033 |
Address | -0.032 |
Token | I |
Token ID | 314 |
Feature activation | +8.60e-03 |
Pos logit contributions | |
Proced | +0.108 |
yip | +0.102 |
aning | +0.096 |
Fn | +0.095 |
Address | +0.095 |
Neg logit contributions | |
eyed | -0.081 |
didn | -0.080 |
CBS | -0.080 |
hadn | -0.077 |
atra | -0.077 |
Token | fool |
Token ID | 9192 |
Feature activation | +9.56e-02 |
Pos logit contributions | |
Proced | +0.006 |
yip | +0.006 |
aning | +0.005 |
Fn | +0.005 |
Address | +0.005 |
Neg logit contributions | |
eyed | -0.005 |
didn | -0.005 |
CBS | -0.005 |
atra | -0.005 |
photos | -0.005 |
Token | ing |
Token ID | 278 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
Proced | +0.024 |
yip | +0.021 |
aning | +0.017 |
Address | +0.016 |
Agg | +0.016 |
Neg logit contributions | |
eyed | -0.104 |
CBS | -0.103 |
didn | -0.103 |
atra | -0.102 |
wasn | -0.101 |
Token | myself |
Token ID | 3589 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
Proced | +0.410 |
yip | +0.392 |
Fn | +0.363 |
Agg | +0.360 |
aning | +0.358 |
Neg logit contributions | |
atra | -0.394 |
eyed | -0.393 |
didn | -0.393 |
CBS | -0.392 |
photos | -0.377 |
Token | ? |
Token ID | 30 |
Feature activation | -8.41e-02 |
Pos logit contributions | |
Proced | +0.063 |
yip | +0.061 |
aning | +0.056 |
Fn | +0.056 |
Address | +0.055 |
Neg logit contributions | |
eyed | -0.061 |
CBS | -0.061 |
didn | -0.060 |
atra | -0.060 |
photos | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | +9.17e-02 |
Pos logit contributions | |
eyed | +0.062 |
didn | +0.061 |
CBS | +0.061 |
atra | +0.060 |
wasn | +0.059 |
Neg logit contributions | |
Proced | -0.052 |
yip | -0.049 |
Fn | -0.045 |
Address | -0.044 |
aning | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
Proced | +0.053 |
yip | +0.049 |
Address | +0.045 |
Agg | +0.045 |
aning | +0.045 |
Neg logit contributions | |
eyed | -0.071 |
didn | -0.069 |
CBS | -0.069 |
atra | -0.069 |
wasn | -0.068 |
Token | All |
Token ID | 3237 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
Proced | +0.053 |
yip | +0.049 |
Agg | +0.047 |
Address | +0.047 |
AMS | +0.046 |
Neg logit contributions | |
eyed | -0.056 |
atra | -0.055 |
didn | -0.055 |
hadn | -0.054 |
wasn | -0.054 |
Token | : |
Token ID | 25 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
Proced | +0.029 |
yip | +0.026 |
Address | +0.023 |
Fn | +0.023 |
aning | +0.023 |
Neg logit contributions | |
eyed | -0.061 |
didn | -0.060 |
CBS | -0.060 |
atra | -0.059 |
wasn | -0.059 |
Token | Great |
Token ID | 3878 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
Proced | +0.159 |
yip | +0.149 |
Fn | +0.140 |
Address | +0.138 |
Agg | +0.137 |
Neg logit contributions | |
eyed | -0.147 |
didn | -0.144 |
CBS | -0.143 |
atra | -0.140 |
wasn | -0.140 |
Token | full |
Token ID | 12853 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
Proced | +0.023 |
yip | +0.022 |
Address | +0.021 |
Fn | +0.021 |
Agg | +0.021 |
Neg logit contributions | |
eyed | -0.020 |
CBS | -0.020 |
didn | -0.020 |
atra | -0.020 |
wasn | -0.020 |
Token | Taco |
Token ID | 40129 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Proced | +0.027 |
yip | +0.026 |
Address | +0.024 |
Fn | +0.024 |
Agg | +0.023 |
Neg logit contributions | |
eyed | -0.025 |
didn | -0.025 |
CBS | -0.025 |
wasn | -0.024 |
atra | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | +7.94e-02 |
Pos logit contributions | |
Proced | +0.077 |
yip | +0.073 |
Address | +0.068 |
Fn | +0.068 |
aning | +0.067 |
Neg logit contributions | |
eyed | -0.075 |
didn | -0.075 |
CBS | -0.074 |
wasn | -0.072 |
atra | -0.072 |
Token | Sab |
Token ID | 9910 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
Proced | +0.056 |
yip | +0.053 |
Fn | +0.049 |
Address | +0.049 |
Agg | +0.049 |
Neg logit contributions | |
eyed | -0.052 |
didn | -0.051 |
CBS | -0.050 |
atra | -0.049 |
wasn | -0.049 |
Token | etta |
Token ID | 15253 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
Proced | +0.425 |
yip | +0.409 |
Address | +0.374 |
Fn | +0.372 |
aning | +0.372 |
Neg logit contributions | |
eyed | -0.377 |
didn | -0.374 |
CBS | -0.372 |
wasn | -0.365 |
hadn | -0.362 |
Token | , |
Token ID | 11 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
eyed | +0.213 |
didn | +0.211 |
CBS | +0.209 |
wasn | +0.204 |
atra | +0.204 |
Neg logit contributions | |
Proced | -0.254 |
yip | -0.243 |
Address | -0.225 |
Fn | -0.224 |
aning | -0.222 |
Token | Café |
Token ID | 42151 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
Proced | +0.046 |
yip | +0.043 |
Fn | +0.040 |
Address | +0.040 |
Agg | +0.039 |
Neg logit contributions | |
eyed | -0.043 |
didn | -0.042 |
CBS | -0.042 |
atra | -0.041 |
wasn | -0.041 |
Token | Z |
Token ID | 1168 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
eyed | +0.205 |
didn | +0.202 |
CBS | +0.202 |
wasn | +0.197 |
atra | +0.197 |
Neg logit contributions | |
Proced | -0.209 |
yip | -0.199 |
Fn | -0.183 |
Address | -0.182 |
Agg | -0.181 |
Token | ol |
Token ID | 349 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
Proced | +0.129 |
yip | +0.125 |
Address | +0.113 |
Fn | +0.113 |
aning | +0.113 |
Neg logit contributions | |
eyed | -0.125 |
didn | -0.124 |
CBS | -0.124 |
wasn | -0.121 |
hadn | -0.120 |
Token | Marshal |
Token ID | 37899 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
eyed | +0.168 |
CBS | +0.168 |
didn | +0.167 |
atra | +0.163 |
wasn | +0.162 |
Neg logit contributions | |
Proced | -0.134 |
yip | -0.125 |
Fn | -0.114 |
Address | -0.114 |
aning | -0.113 |
Token | Rod |
Token ID | 6882 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Proced | +0.086 |
yip | +0.082 |
Fn | +0.076 |
Address | +0.075 |
aning | +0.074 |
Neg logit contributions | |
eyed | -0.086 |
CBS | -0.085 |
didn | -0.085 |
atra | -0.083 |
wasn | -0.083 |
Token | olf |
Token ID | 4024 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
Proced | +0.121 |
yip | +0.116 |
aning | +0.107 |
Fn | +0.107 |
Address | +0.106 |
Neg logit contributions | |
CBS | -0.101 |
eyed | -0.100 |
didn | -0.100 |
wasn | -0.097 |
hadn | -0.097 |
Token | o |
Token ID | 78 |
Feature activation | +7.16e-02 |
Pos logit contributions | |
eyed | +0.022 |
CBS | +0.022 |
didn | +0.022 |
wasn | +0.022 |
atra | +0.022 |
Neg logit contributions | |
Proced | -0.022 |
yip | -0.021 |
Fn | -0.020 |
Address | -0.020 |
aning | -0.019 |
Token | G |
Token ID | 402 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
Proced | +0.047 |
yip | +0.045 |
Fn | +0.041 |
Address | +0.041 |
Agg | +0.040 |
Neg logit contributions | |
eyed | -0.050 |
didn | -0.049 |
CBS | -0.049 |
atra | -0.048 |
wasn | -0.048 |
Token | raz |
Token ID | 3247 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
Proced | +0.055 |
yip | +0.052 |
Address | +0.044 |
aning | +0.044 |
Agg | +0.044 |
Neg logit contributions | |
eyed | -0.131 |
didn | -0.130 |
CBS | -0.130 |
wasn | -0.128 |
atra | -0.128 |
Token | iani |
Token ID | 25111 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
Proced | +0.389 |
yip | +0.377 |
aning | +0.342 |
Address | +0.341 |
Agg | +0.339 |
Neg logit contributions | |
eyed | -0.402 |
CBS | -0.398 |
didn | -0.397 |
wasn | -0.388 |
atra | -0.387 |
Token | , |
Token ID | 11 |
Feature activation | +7.45e-02 |
Pos logit contributions | |
Proced | +0.129 |
yip | +0.123 |
Fn | +0.118 |
Address | +0.116 |
aning | +0.116 |
Neg logit contributions | |
CBS | -0.072 |
eyed | -0.072 |
didn | -0.071 |
photos | -0.069 |
atra | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
Proced | +0.054 |
yip | +0.052 |
Fn | +0.049 |
Address | +0.048 |
Agg | +0.047 |
Neg logit contributions | |
eyed | -0.046 |
CBS | -0.046 |
didn | -0.045 |
atra | -0.044 |
photos | -0.044 |
Token | former |
Token ID | 1966 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
eyed | +0.036 |
didn | +0.036 |
CBS | +0.036 |
atra | +0.035 |
photos | +0.035 |
Neg logit contributions | |
Proced | -0.042 |
yip | -0.040 |
Fn | -0.038 |
Address | -0.037 |
aning | -0.037 |
Token | governor |
Token ID | 8153 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
Proced | +0.012 |
yip | +0.012 |
Fn | +0.011 |
Address | +0.011 |
aning | +0.011 |
Neg logit contributions | |
eyed | -0.010 |
didn | -0.010 |
CBS | -0.010 |
atra | -0.010 |
wasn | -0.010 |
Token | as |
Token ID | 355 |
Feature activation | -8.72e-03 |
Pos logit contributions | |
eyed | +0.021 |
didn | +0.021 |
CBS | +0.021 |
atra | +0.021 |
wasn | +0.021 |
Neg logit contributions | |
Proced | -0.020 |
yip | -0.019 |
Address | -0.017 |
aning | -0.017 |
Fn | -0.017 |
Token | Bac |
Token ID | 35583 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
eyed | +0.005 |
didn | +0.005 |
CBS | +0.005 |
atra | +0.005 |
photos | +0.005 |
Neg logit contributions | |
Proced | -0.007 |
yip | -0.006 |
Finding | -0.006 |
Fn | -0.006 |
Address | -0.006 |
Token | ile |
Token ID | 576 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
eyed | +0.048 |
didn | +0.048 |
CBS | +0.048 |
wasn | +0.047 |
atra | +0.047 |
Neg logit contributions | |
Proced | -0.030 |
yip | -0.029 |
Address | -0.025 |
aning | -0.025 |
Agg | -0.025 |
Token | 's |
Token ID | 338 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
Proced | +0.030 |
yip | +0.028 |
Address | +0.026 |
aning | +0.026 |
Fn | +0.025 |
Neg logit contributions | |
eyed | -0.034 |
CBS | -0.034 |
didn | -0.034 |
atra | -0.033 |
photos | -0.033 |
Token | Inn |
Token ID | 10849 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
eyed | +0.021 |
CBS | +0.020 |
didn | +0.020 |
atra | +0.020 |
photos | +0.020 |
Neg logit contributions | |
Proced | -0.019 |
yip | -0.018 |
Address | -0.016 |
aning | -0.016 |
Fn | -0.016 |
Token | oc |
Token ID | 420 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
Proced | +0.209 |
yip | +0.197 |
Address | +0.192 |
aning | +0.191 |
Finding | +0.189 |
Neg logit contributions | |
eyed | -0.056 |
CBS | -0.052 |
didn | -0.049 |
photos | -0.048 |
atra | -0.047 |
Token | ence |
Token ID | 594 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
eyed | +0.538 |
didn | +0.531 |
CBS | +0.531 |
wasn | +0.520 |
atra | +0.519 |
Neg logit contributions | |
Proced | -0.387 |
yip | -0.371 |
Address | -0.329 |
Fn | -0.325 |
aning | -0.325 |
Token | of |
Token ID | 286 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
Proced | +0.044 |
yip | +0.041 |
Address | +0.038 |
aning | +0.037 |
Fn | +0.037 |
Neg logit contributions | |
eyed | -0.043 |
CBS | -0.042 |
didn | -0.042 |
atra | -0.041 |
photos | -0.041 |
Token | Muslims |
Token ID | 7045 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
eyed | +0.016 |
didn | +0.015 |
CBS | +0.015 |
wasn | +0.015 |
atra | +0.015 |
Neg logit contributions | |
Proced | -0.027 |
yip | -0.026 |
Fn | -0.024 |
aning | -0.024 |
Finding | -0.024 |
Token | are |
Token ID | 389 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
eyed | +0.111 |
CBS | +0.109 |
didn | +0.109 |
atra | +0.107 |
photos | +0.106 |
Neg logit contributions | |
Proced | -0.104 |
yip | -0.097 |
Address | -0.090 |
aning | -0.089 |
Fn | -0.089 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
Proced | +0.132 |
yip | +0.123 |
aning | +0.116 |
Address | +0.114 |
Fn | +0.114 |
Neg logit contributions | |
eyed | -0.125 |
didn | -0.124 |
CBS | -0.121 |
atra | -0.121 |
photos | -0.119 |
Token | , |
Token ID | 11 |
Feature activation | -5.85e-02 |
Pos logit contributions | |
eyed | +0.190 |
didn | +0.188 |
CBS | +0.187 |
wasn | +0.185 |
atra | +0.183 |
Neg logit contributions | |
Proced | -0.203 |
yip | -0.196 |
Address | -0.178 |
Fn | -0.177 |
Agg | -0.177 |
Token | Cit |
Token ID | 15792 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
eyed | +0.038 |
didn | +0.037 |
CBS | +0.037 |
atra | +0.037 |
photos | +0.036 |
Neg logit contributions | |
Proced | -0.041 |
yip | -0.039 |
Fn | -0.036 |
Address | -0.036 |
aning | -0.036 |
Token | igroup |
Token ID | 47875 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
eyed | +0.396 |
didn | +0.393 |
CBS | +0.391 |
wasn | +0.388 |
hadn | +0.386 |
Neg logit contributions | |
Proced | -0.081 |
yip | -0.072 |
Address | -0.053 |
aning | -0.053 |
Agg | -0.053 |
Token | , |
Token ID | 11 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
eyed | +0.037 |
didn | +0.036 |
CBS | +0.036 |
wasn | +0.035 |
atra | +0.035 |
Neg logit contributions | |
Proced | -0.038 |
yip | -0.036 |
Address | -0.033 |
Fn | -0.033 |
Agg | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
eyed | +0.015 |
didn | +0.014 |
CBS | +0.014 |
atra | +0.014 |
photos | +0.014 |
Neg logit contributions | |
Proced | -0.018 |
yip | -0.017 |
Fn | -0.016 |
Address | -0.015 |
aning | -0.015 |
Token | Wells |
Token ID | 18292 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
eyed | +0.021 |
didn | +0.021 |
CBS | +0.020 |
atra | +0.020 |
photos | +0.020 |
Neg logit contributions | |
Proced | -0.036 |
yip | -0.034 |
Fn | -0.032 |
Address | -0.032 |
aning | -0.032 |
Token | Fargo |
Token ID | 28125 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
eyed | +0.353 |
didn | +0.344 |
CBS | +0.343 |
atra | +0.338 |
photos | +0.331 |
Neg logit contributions | |
Proced | -0.562 |
yip | -0.537 |
Address | -0.506 |
aning | -0.504 |
Fn | -0.502 |
Token | are |
Token ID | 389 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
eyed | +0.170 |
CBS | +0.167 |
didn | +0.165 |
atra | +0.165 |
photos | +0.161 |
Neg logit contributions | |
Proced | -0.170 |
yip | -0.159 |
Address | -0.148 |
Fn | -0.147 |
Agg | -0.147 |
Token | the |
Token ID | 262 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
Proced | +0.103 |
yip | +0.097 |
Fn | +0.091 |
Address | +0.091 |
aning | +0.091 |
Neg logit contributions | |
eyed | -0.091 |
didn | -0.089 |
CBS | -0.088 |
atra | -0.088 |
photos | -0.086 |
Token | largest |
Token ID | 4387 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
eyed | +0.094 |
CBS | +0.093 |
didn | +0.093 |
atra | +0.091 |
photos | +0.089 |
Neg logit contributions | |
Proced | -0.104 |
yip | -0.098 |
Address | -0.092 |
Fn | -0.091 |
aning | -0.091 |
Token | four |
Token ID | 1440 |
Feature activation | -5.84e-02 |
Pos logit contributions | |
Proced | +0.072 |
yip | +0.067 |
Address | +0.062 |
Agg | +0.062 |
Fn | +0.062 |
Neg logit contributions | |
eyed | -0.073 |
didn | -0.072 |
CBS | -0.072 |
atra | -0.071 |
photos | -0.070 |
Token | is |
Token ID | 318 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
Proced | +0.011 |
yip | +0.010 |
aning | +0.010 |
Address | +0.010 |
Fn | +0.009 |
Neg logit contributions | |
eyed | -0.009 |
CBS | -0.009 |
didn | -0.009 |
atra | -0.009 |
photos | -0.008 |
Token | Bank |
Token ID | 5018 |
Feature activation | -7.44e-02 |
Pos logit contributions | |
Proced | +0.044 |
yip | +0.041 |
Fn | +0.039 |
Address | +0.039 |
aning | +0.039 |
Neg logit contributions | |
eyed | -0.031 |
didn | -0.030 |
atra | -0.029 |
CBS | -0.029 |
photos | -0.029 |
Token | of |
Token ID | 286 |
Feature activation | -4.80e-04 |
Pos logit contributions | |
eyed | +0.051 |
didn | +0.050 |
CBS | +0.049 |
atra | +0.049 |
photos | +0.048 |
Neg logit contributions | |
Proced | -0.050 |
yip | -0.046 |
Address | -0.043 |
Fn | -0.042 |
aning | -0.042 |
Token | America |
Token ID | 2253 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
eyed | +0.000 |
CBS | +0.000 |
didn | +0.000 |
photos | +0.000 |
atra | +0.000 |
Neg logit contributions | |
Proced | -0.000 |
yip | -0.000 |
Fn | -0.000 |
Address | -0.000 |
Agg | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -3.67e-03 |
Pos logit contributions | |
eyed | +0.161 |
didn | +0.158 |
CBS | +0.158 |
atra | +0.155 |
wasn | +0.154 |
Neg logit contributions | |
Proced | -0.152 |
yip | -0.143 |
Address | -0.131 |
Fn | -0.131 |
Agg | -0.130 |
Token | Wells |
Token ID | 18292 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
eyed | +0.003 |
didn | +0.003 |
atra | +0.003 |
CBS | +0.003 |
hadn | +0.003 |
Neg logit contributions | |
Proced | -0.002 |
yip | -0.002 |
Address | -0.002 |
Fn | -0.002 |
Agg | -0.002 |
Token | Fargo |
Token ID | 28125 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
eyed | +0.370 |
CBS | +0.358 |
didn | +0.357 |
atra | +0.353 |
photos | +0.347 |
Neg logit contributions | |
Proced | -0.545 |
yip | -0.514 |
Address | -0.489 |
aning | -0.488 |
Fn | -0.482 |
Token | , |
Token ID | 11 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
eyed | +0.124 |
CBS | +0.120 |
didn | +0.120 |
atra | +0.120 |
photos | +0.117 |
Neg logit contributions | |
Proced | -0.149 |
yip | -0.138 |
Address | -0.132 |
Fn | -0.131 |
Agg | -0.129 |
Token | JPMorgan |
Token ID | 43318 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
eyed | +0.007 |
atra | +0.007 |
didn | +0.007 |
CBS | +0.007 |
photos | +0.007 |
Neg logit contributions | |
Proced | -0.009 |
yip | -0.008 |
Fn | -0.008 |
Address | -0.008 |
AMS | -0.008 |
Token | Chase |
Token ID | 16346 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
eyed | +0.197 |
didn | +0.192 |
CBS | +0.191 |
atra | +0.187 |
photos | +0.184 |
Neg logit contributions | |
Proced | -0.359 |
yip | -0.339 |
Address | -0.322 |
Fn | -0.322 |
aning | -0.320 |
Token | , |
Token ID | 11 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
eyed | +0.339 |
didn | +0.335 |
CBS | +0.334 |
atra | +0.328 |
wasn | +0.326 |
Neg logit contributions | |
Proced | -0.347 |
yip | -0.331 |
Address | -0.303 |
Fn | -0.302 |
aning | -0.300 |
Token | of |
Token ID | 286 |
Feature activation | -6.87e-03 |
Pos logit contributions | |
Proced | +0.104 |
yip | +0.095 |
aning | +0.092 |
Fn | +0.091 |
Address | +0.090 |
Neg logit contributions | |
eyed | -0.090 |
CBS | -0.089 |
didn | -0.088 |
atra | -0.087 |
hadn | -0.085 |
Token | these |
Token ID | 777 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
eyed | +0.004 |
CBS | +0.004 |
didn | +0.004 |
atra | +0.004 |
hadn | +0.004 |
Neg logit contributions | |
Proced | -0.006 |
yip | -0.005 |
Fn | -0.005 |
Address | -0.005 |
aning | -0.005 |
Token | attacks |
Token ID | 3434 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
Proced | +0.083 |
yip | +0.078 |
Address | +0.073 |
aning | +0.072 |
Fn | +0.072 |
Neg logit contributions | |
eyed | -0.070 |
CBS | -0.069 |
didn | -0.069 |
atra | -0.068 |
wasn | -0.066 |
Token | on |
Token ID | 319 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
eyed | +0.026 |
CBS | +0.026 |
didn | +0.026 |
atra | +0.026 |
photos | +0.025 |
Neg logit contributions | |
Proced | -0.026 |
yip | -0.025 |
aning | -0.023 |
Address | -0.023 |
Agg | -0.023 |
Token | Nina |
Token ID | 36300 |
Feature activation | -9.59e-02 |
Pos logit contributions | |
eyed | +0.038 |
didn | +0.037 |
CBS | +0.037 |
atra | +0.037 |
wasn | +0.036 |
Neg logit contributions | |
Proced | -0.058 |
yip | -0.055 |
aning | -0.052 |
Fn | -0.052 |
Address | -0.052 |
Token | Turner |
Token ID | 15406 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
eyed | +0.014 |
didn | +0.013 |
CBS | +0.013 |
atra | +0.011 |
wasn | +0.011 |
Neg logit contributions | |
Proced | -0.115 |
yip | -0.111 |
aning | -0.107 |
Address | -0.107 |
Fn | -0.107 |
Token | , |
Token ID | 11 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
eyed | +0.464 |
didn | +0.460 |
CBS | +0.459 |
atra | +0.452 |
wasn | +0.444 |
Neg logit contributions | |
Proced | -0.446 |
yip | -0.422 |
Address | -0.388 |
aning | -0.385 |
Fn | -0.384 |
Token | but |
Token ID | 475 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
eyed | +0.008 |
CBS | +0.007 |
didn | +0.007 |
atra | +0.007 |
photos | +0.007 |
Neg logit contributions | |
Proced | -0.008 |
yip | -0.007 |
aning | -0.007 |
Fn | -0.007 |
Address | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | -1.71e-03 |
Pos logit contributions | |
eyed | +0.066 |
CBS | +0.066 |
atra | +0.065 |
didn | +0.063 |
photos | +0.062 |
Neg logit contributions | |
Proced | -0.092 |
yip | -0.086 |
Fn | -0.082 |
aning | -0.081 |
Address | -0.081 |
Token | has |
Token ID | 468 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
eyed | +0.001 |
CBS | +0.001 |
atra | +0.001 |
photos | +0.001 |
didn | +0.001 |
Neg logit contributions | |
Proced | -0.001 |
yip | -0.001 |
aning | -0.001 |
Address | -0.001 |
Fn | -0.001 |
Token | inexpl |
Token ID | 33199 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
Proced | +0.024 |
yip | +0.023 |
Fn | +0.021 |
Address | +0.021 |
aning | +0.021 |
Neg logit contributions | |
eyed | -0.021 |
CBS | -0.021 |
atra | -0.021 |
didn | -0.021 |
photos | -0.020 |
Token | are |
Token ID | 389 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
eyed | +0.098 |
CBS | +0.096 |
didn | +0.096 |
atra | +0.095 |
photos | +0.094 |
Neg logit contributions | |
Proced | -0.083 |
yip | -0.078 |
Address | -0.071 |
Fn | -0.071 |
aning | -0.070 |
Token | Bank |
Token ID | 5018 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Proced | +0.081 |
yip | +0.075 |
Address | +0.071 |
Fn | +0.071 |
aning | +0.070 |
Neg logit contributions | |
eyed | -0.079 |
didn | -0.077 |
atra | -0.076 |
CBS | -0.075 |
photos | -0.074 |
Token | of |
Token ID | 286 |
Feature activation | -7.03e-02 |
Pos logit contributions | |
eyed | +0.076 |
didn | +0.075 |
CBS | +0.074 |
atra | +0.073 |
wasn | +0.072 |
Neg logit contributions | |
Proced | -0.075 |
yip | -0.069 |
Address | -0.065 |
Fn | -0.064 |
Agg | -0.064 |
Token | America |
Token ID | 2253 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
eyed | +0.025 |
didn | +0.024 |
CBS | +0.024 |
photos | +0.023 |
atra | +0.023 |
Neg logit contributions | |
Proced | -0.069 |
yip | -0.067 |
Address | -0.064 |
Fn | -0.064 |
Agg | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
eyed | +0.180 |
didn | +0.178 |
CBS | +0.177 |
atra | +0.174 |
wasn | +0.173 |
Neg logit contributions | |
Proced | -0.183 |
yip | -0.174 |
Address | -0.159 |
Fn | -0.159 |
Agg | -0.158 |
Token | Wells |
Token ID | 18292 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
eyed | +0.028 |
didn | +0.026 |
atra | +0.026 |
CBS | +0.026 |
photos | +0.025 |
Neg logit contributions | |
Proced | -0.046 |
yip | -0.042 |
Fn | -0.041 |
Address | -0.041 |
Agg | -0.041 |
Token | Fargo |
Token ID | 28125 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
eyed | +0.360 |
didn | +0.353 |
CBS | +0.350 |
atra | +0.345 |
wasn | +0.339 |
Neg logit contributions | |
Proced | -0.550 |
yip | -0.523 |
Address | -0.492 |
Fn | -0.491 |
aning | -0.489 |
Token | , |
Token ID | 11 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
eyed | +0.154 |
didn | +0.152 |
CBS | +0.152 |
wasn | +0.150 |
atra | +0.148 |
Neg logit contributions | |
Proced | -0.167 |
yip | -0.162 |
Address | -0.146 |
Fn | -0.146 |
Agg | -0.146 |
Token | Equ |
Token ID | 7889 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
eyed | +0.032 |
didn | +0.031 |
atra | +0.030 |
CBS | +0.030 |
photos | +0.030 |
Neg logit contributions | |
Proced | -0.044 |
yip | -0.041 |
Fn | -0.040 |
aning | -0.039 |
Address | -0.039 |
Token | ifax |
Token ID | 26590 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
Proced | +0.014 |
yip | +0.013 |
aning | +0.011 |
Address | +0.011 |
Agg | +0.011 |
Neg logit contributions | |
eyed | -0.023 |
didn | -0.023 |
CBS | -0.023 |
wasn | -0.022 |
hadn | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
Proced | +0.125 |
yip | +0.121 |
Address | +0.110 |
Fn | +0.110 |
Agg | +0.109 |
Neg logit contributions | |
eyed | -0.117 |
didn | -0.116 |
CBS | -0.116 |
wasn | -0.114 |
atra | -0.113 |
Token | C |
Token ID | 327 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Proced | +0.009 |
yip | +0.008 |
Fn | +0.008 |
Address | +0.008 |
Agg | +0.008 |
Neg logit contributions | |
eyed | -0.005 |
didn | -0.005 |
atra | -0.005 |
CBS | -0.005 |
photos | -0.005 |
Token | iti |
Token ID | 8846 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
eyed | +0.110 |
didn | +0.110 |
CBS | +0.107 |
atra | +0.106 |
wasn | +0.104 |
Neg logit contributions | |
Proced | -0.223 |
yip | -0.214 |
Address | -0.203 |
aning | -0.203 |
Agg | -0.201 |
Token | , |
Token ID | 11 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
eyed | +0.206 |
didn | +0.204 |
CBS | +0.203 |
wasn | +0.199 |
atra | +0.198 |
Neg logit contributions | |
Proced | -0.214 |
yip | -0.206 |
Address | -0.187 |
Fn | -0.187 |
Agg | -0.186 |
Token | Chase |
Token ID | 16346 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
Proced | +0.012 |
yip | +0.011 |
Fn | +0.010 |
Address | +0.010 |
aning | +0.010 |
Neg logit contributions | |
eyed | -0.010 |
didn | -0.010 |
CBS | -0.010 |
atra | -0.010 |
wasn | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +9.00e-03 |
Pos logit contributions | |
eyed | +0.308 |
didn | +0.305 |
CBS | +0.303 |
wasn | +0.297 |
atra | +0.297 |
Neg logit contributions | |
Proced | -0.315 |
yip | -0.302 |
Address | -0.275 |
Fn | -0.274 |
aning | -0.273 |
Token | Wells |
Token ID | 18292 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
Proced | +0.006 |
yip | +0.006 |
Fn | +0.006 |
Address | +0.006 |
aning | +0.006 |
Neg logit contributions | |
eyed | -0.006 |
didn | -0.006 |
CBS | -0.006 |
atra | -0.005 |
wasn | -0.005 |
Token | Fargo |
Token ID | 28125 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
eyed | +0.352 |
didn | +0.346 |
CBS | +0.343 |
atra | +0.336 |
wasn | +0.333 |
Neg logit contributions | |
Proced | -0.557 |
yip | -0.534 |
Address | -0.500 |
Fn | -0.497 |
aning | -0.497 |
Token | , |
Token ID | 11 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
eyed | +0.161 |
didn | +0.159 |
CBS | +0.159 |
wasn | +0.157 |
hadn | +0.155 |
Neg logit contributions | |
Proced | -0.180 |
yip | -0.176 |
Fn | -0.159 |
Address | -0.158 |
Agg | -0.158 |
Token | and |
Token ID | 290 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
eyed | +0.013 |
didn | +0.013 |
CBS | +0.012 |
atra | +0.012 |
wasn | +0.012 |
Neg logit contributions | |
Proced | -0.014 |
yip | -0.013 |
Fn | -0.012 |
Address | -0.012 |
aning | -0.012 |
Token | Ally |
Token ID | 43039 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
eyed | +0.035 |
didn | +0.034 |
CBS | +0.033 |
atra | +0.033 |
photos | +0.032 |
Neg logit contributions | |
Proced | -0.043 |
yip | -0.040 |
Fn | -0.038 |
Address | -0.038 |
aning | -0.037 |
Token | Bank |
Token ID | 5018 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
eyed | +0.159 |
didn | +0.156 |
CBS | +0.156 |
atra | +0.156 |
turn | +0.153 |
Neg logit contributions | |
Proced | -0.122 |
yip | -0.111 |
Address | -0.106 |
Agg | -0.104 |
aning | -0.102 |
Token | sexism |
Token ID | 24953 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
eyed | +0.106 |
CBS | +0.105 |
didn | +0.105 |
atra | +0.104 |
photos | +0.101 |
Neg logit contributions | |
Proced | -0.142 |
yip | -0.134 |
aning | -0.126 |
Address | -0.125 |
Fn | -0.125 |
Token | and |
Token ID | 290 |
Feature activation | -8.53e-02 |
Pos logit contributions | |
eyed | +0.196 |
CBS | +0.194 |
didn | +0.192 |
atra | +0.190 |
photos | +0.187 |
Neg logit contributions | |
Proced | -0.214 |
yip | -0.208 |
Address | -0.187 |
aning | -0.187 |
Fn | -0.187 |
Token | racism |
Token ID | 10713 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
eyed | +0.052 |
CBS | +0.051 |
didn | +0.051 |
atra | +0.050 |
photos | +0.049 |
Neg logit contributions | |
Proced | -0.063 |
yip | -0.061 |
aning | -0.056 |
Address | -0.056 |
Fn | -0.056 |
Token | against |
Token ID | 1028 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
eyed | +0.108 |
CBS | +0.107 |
atra | +0.105 |
didn | +0.105 |
photos | +0.102 |
Neg logit contributions | |
Proced | -0.132 |
yip | -0.127 |
aning | -0.116 |
Address | -0.116 |
Fn | -0.115 |
Token | Nina |
Token ID | 36300 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
eyed | +0.156 |
CBS | +0.155 |
didn | +0.154 |
atra | +0.153 |
photos | +0.148 |
Neg logit contributions | |
Proced | -0.233 |
yip | -0.220 |
aning | -0.209 |
Fn | -0.208 |
Address | -0.206 |
Token | Turner |
Token ID | 15406 |
Feature activation | -6.68e-01 |
Pos logit contributions | |
eyed | +0.017 |
didn | +0.016 |
CBS | +0.016 |
atra | +0.015 |
photos | +0.014 |
Neg logit contributions | |
Proced | -0.096 |
yip | -0.091 |
Fn | -0.089 |
Agg | -0.089 |
aning | -0.089 |
Token | is |
Token ID | 318 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
eyed | +0.418 |
CBS | +0.414 |
didn | +0.411 |
atra | +0.407 |
photos | +0.396 |
Neg logit contributions | |
Proced | -0.480 |
yip | -0.454 |
Address | -0.421 |
Fn | -0.421 |
aning | -0.421 |
Token | especially |
Token ID | 2592 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
eyed | +0.044 |
didn | +0.044 |
CBS | +0.043 |
atra | +0.043 |
photos | +0.042 |
Neg logit contributions | |
Proced | -0.055 |
yip | -0.052 |
aning | -0.049 |
Fn | -0.049 |
Address | -0.049 |
Token | noteworthy |
Token ID | 30902 |
Feature activation | -8.36e-02 |
Pos logit contributions | |
Proced | +0.035 |
yip | +0.033 |
aning | +0.031 |
Address | +0.031 |
Fn | +0.031 |
Neg logit contributions | |
eyed | -0.036 |
didn | -0.036 |
CBS | -0.036 |
atra | -0.035 |
photos | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -8.77e-02 |
Pos logit contributions | |
eyed | +0.055 |
CBS | +0.055 |
didn | +0.054 |
atra | +0.054 |
wasn | +0.053 |
Neg logit contributions | |
Proced | -0.057 |
yip | -0.055 |
aning | -0.050 |
Address | -0.050 |
Fn | -0.050 |
Token | considering |
Token ID | 6402 |
Feature activation | +9.29e-02 |
Pos logit contributions | |
eyed | +0.054 |
CBS | +0.053 |
didn | +0.052 |
atra | +0.052 |
photos | +0.051 |
Neg logit contributions | |
Proced | -0.064 |
yip | -0.061 |
aning | -0.057 |
Fn | -0.056 |
Address | -0.056 |
Token | devices |
Token ID | 4410 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
eyed | +0.259 |
didn | +0.256 |
CBS | +0.255 |
atra | +0.250 |
photos | +0.243 |
Neg logit contributions | |
Proced | -0.313 |
yip | -0.301 |
Fn | -0.281 |
Address | -0.279 |
aning | -0.276 |
Token | . |
Token ID | 13 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
eyed | +0.109 |
didn | +0.108 |
CBS | +0.107 |
atra | +0.106 |
wasn | +0.104 |
Neg logit contributions | |
Proced | -0.108 |
yip | -0.104 |
Fn | -0.095 |
Address | -0.095 |
aning | -0.094 |
Token | S |
Token ID | 311 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
eyed | +0.085 |
didn | +0.083 |
CBS | +0.082 |
atra | +0.082 |
wasn | +0.081 |
Neg logit contributions | |
Proced | -0.084 |
yip | -0.077 |
Fn | -0.073 |
Address | -0.073 |
Agg | -0.072 |
Token | ada |
Token ID | 4763 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
didn | +0.090 |
eyed | +0.090 |
CBS | +0.089 |
wasn | +0.088 |
atra | +0.087 |
Neg logit contributions | |
Proced | -0.082 |
yip | -0.080 |
Address | -0.072 |
aning | -0.072 |
Agg | -0.071 |
Token | o |
Token ID | 78 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
eyed | +0.215 |
didn | +0.212 |
CBS | +0.211 |
wasn | +0.206 |
atra | +0.205 |
Neg logit contributions | |
Proced | -0.229 |
yip | -0.221 |
Address | -0.201 |
Fn | -0.200 |
aning | -0.200 |
Token | Turner |
Token ID | 15406 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
eyed | +0.015 |
didn | +0.015 |
CBS | +0.015 |
atra | +0.014 |
photos | +0.014 |
Neg logit contributions | |
Proced | -0.018 |
yip | -0.017 |
Address | -0.016 |
Fn | -0.016 |
aning | -0.015 |
Token | Es |
Token ID | 8678 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
eyed | +0.427 |
didn | +0.422 |
CBS | +0.419 |
atra | +0.411 |
wasn | +0.404 |
Neg logit contributions | |
Proced | -0.462 |
yip | -0.440 |
Address | -0.406 |
Fn | -0.405 |
aning | -0.401 |
Token | q |
Token ID | 80 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
eyed | +0.139 |
didn | +0.137 |
CBS | +0.136 |
wasn | +0.132 |
atra | +0.131 |
Neg logit contributions | |
Proced | -0.165 |
yip | -0.159 |
Address | -0.147 |
Fn | -0.146 |
Agg | -0.144 |
Token | , |
Token ID | 11 |
Feature activation | -8.79e-02 |
Pos logit contributions | |
eyed | +0.204 |
didn | +0.201 |
CBS | +0.201 |
atra | +0.196 |
wasn | +0.195 |
Neg logit contributions | |
Proced | -0.194 |
yip | -0.186 |
Address | -0.169 |
Fn | -0.169 |
aning | -0.167 |
Token | director |
Token ID | 3437 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
eyed | +0.051 |
didn | +0.050 |
CBS | +0.050 |
atra | +0.049 |
photos | +0.048 |
Neg logit contributions | |
Proced | -0.068 |
yip | -0.065 |
Fn | -0.061 |
Address | -0.060 |
aning | -0.059 |
Token | of |
Token ID | 286 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
eyed | +0.166 |
didn | +0.164 |
CBS | +0.163 |
wasn | +0.160 |
atra | +0.159 |
Neg logit contributions | |
Proced | -0.196 |
yip | -0.189 |
Address | -0.173 |
Fn | -0.173 |
aning | -0.172 |
Token | 38 |
Token ID | 4353 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
eyed | +0.226 |
didn | +0.221 |
CBS | +0.219 |
atra | +0.217 |
turn | +0.213 |
Neg logit contributions | |
Proced | -0.247 |
yip | -0.230 |
Fn | -0.216 |
Address | -0.214 |
Agg | -0.214 |
Token | , |
Token ID | 11 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
Proced | +0.057 |
yip | +0.054 |
Address | +0.050 |
Fn | +0.050 |
Agg | +0.050 |
Neg logit contributions | |
eyed | -0.051 |
didn | -0.050 |
CBS | -0.050 |
wasn | -0.049 |
atra | -0.049 |
Token | 833 |
Token ID | 48634 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
Proced | +0.020 |
yip | +0.019 |
Address | +0.018 |
Fn | +0.018 |
Agg | +0.018 |
Neg logit contributions | |
eyed | -0.019 |
didn | -0.019 |
CBS | -0.019 |
wasn | -0.018 |
atra | -0.018 |
Token | complaints |
Token ID | 9687 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
Proced | +0.008 |
yip | +0.007 |
Fn | +0.007 |
Address | +0.007 |
Agg | +0.007 |
Neg logit contributions | |
eyed | -0.008 |
didn | -0.008 |
CBS | -0.008 |
atra | -0.008 |
photos | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +4.19e-04 |
Pos logit contributions | |
eyed | +0.030 |
CBS | +0.029 |
didn | +0.029 |
atra | +0.029 |
photos | +0.029 |
Neg logit contributions | |
Proced | -0.028 |
yip | -0.027 |
Address | -0.024 |
Fn | -0.024 |
Agg | -0.024 |
Token | Wells |
Token ID | 18292 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
Proced | +0.000 |
yip | +0.000 |
Fn | +0.000 |
Address | +0.000 |
aning | +0.000 |
Neg logit contributions | |
eyed | -0.000 |
didn | -0.000 |
CBS | -0.000 |
atra | -0.000 |
photos | -0.000 |
Token | Fargo |
Token ID | 28125 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
eyed | +0.339 |
didn | +0.331 |
CBS | +0.328 |
atra | +0.324 |
photos | +0.319 |
Neg logit contributions | |
Proced | -0.535 |
yip | -0.510 |
Address | -0.481 |
Fn | -0.480 |
Agg | -0.478 |
Token | drew |
Token ID | 9859 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
eyed | +0.169 |
CBS | +0.164 |
didn | +0.163 |
atra | +0.163 |
turn | +0.160 |
Neg logit contributions | |
Proced | -0.145 |
yip | -0.133 |
Fn | -0.125 |
Agg | -0.124 |
Address | -0.124 |
Token | 26 |
Token ID | 2608 |
Feature activation | +9.63e-02 |
Pos logit contributions | |
eyed | +0.197 |
didn | +0.192 |
CBS | +0.192 |
atra | +0.192 |
photos | +0.187 |
Neg logit contributions | |
Proced | -0.150 |
yip | -0.140 |
Fn | -0.130 |
Address | -0.128 |
aning | -0.128 |
Token | , |
Token ID | 11 |
Feature activation | -7.55e-03 |
Pos logit contributions | |
Proced | +0.063 |
yip | +0.060 |
Address | +0.055 |
Fn | +0.055 |
Agg | +0.055 |
Neg logit contributions | |
eyed | -0.066 |
didn | -0.066 |
CBS | -0.065 |
atra | -0.064 |
wasn | -0.064 |
Token | 055 |
Token ID | 47838 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
eyed | +0.005 |
didn | +0.005 |
CBS | +0.005 |
wasn | +0.005 |
atra | +0.005 |
Neg logit contributions | |
Proced | -0.005 |
yip | -0.005 |
Address | -0.005 |
Fn | -0.005 |
Agg | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
eyed | +0.190 |
CBS | +0.184 |
didn | +0.183 |
atra | +0.181 |
photos | +0.178 |
Neg logit contributions | |
Proced | -0.226 |
yip | -0.216 |
Fn | -0.202 |
aning | -0.199 |
Agg | -0.199 |
Token | mortgage |
Token ID | 13682 |
Feature activation | +5.76e-03 |
Pos logit contributions | |
eyed | +0.094 |
CBS | +0.092 |
didn | +0.092 |
photos | +0.090 |
atra | +0.090 |
Neg logit contributions | |
Proced | -0.076 |
yip | -0.071 |
Fn | -0.066 |
aning | -0.066 |
Finding | -0.065 |
Token | meltdown |
Token ID | 37824 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
Proced | +0.003 |
yip | +0.003 |
Fn | +0.003 |
aning | +0.003 |
Address | +0.003 |
Neg logit contributions | |
eyed | -0.005 |
CBS | -0.005 |
didn | -0.004 |
atra | -0.004 |
photos | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
eyed | +0.106 |
CBS | +0.105 |
didn | +0.105 |
atra | +0.103 |
photos | +0.102 |
Neg logit contributions | |
Proced | -0.107 |
yip | -0.103 |
Fn | -0.094 |
aning | -0.094 |
Address | -0.094 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
eyed | +0.017 |
CBS | +0.017 |
photos | +0.016 |
didn | +0.016 |
atra | +0.016 |
Neg logit contributions | |
Proced | -0.028 |
yip | -0.026 |
Fn | -0.025 |
Address | -0.025 |
aning | -0.025 |
Token | Cohn |
Token ID | 45005 |
Feature activation | -6.49e-01 |
Pos logit contributions | |
eyed | +0.074 |
didn | +0.072 |
CBS | +0.072 |
atra | +0.070 |
photos | +0.069 |
Neg logit contributions | |
Proced | -0.107 |
yip | -0.103 |
Fn | -0.097 |
Agg | -0.096 |
Address | -0.096 |
Token | as |
Token ID | 355 |
Feature activation | -8.11e-02 |
Pos logit contributions | |
eyed | +0.407 |
didn | +0.401 |
CBS | +0.400 |
atra | +0.388 |
photos | +0.387 |
Neg logit contributions | |
Proced | -0.470 |
yip | -0.448 |
Address | -0.415 |
aning | -0.413 |
Fn | -0.413 |
Token | � |
Token ID | 564 |
Feature activation | +6.70e-03 |
Pos logit contributions | |
eyed | +0.053 |
didn | +0.053 |
CBS | +0.052 |
photos | +0.051 |
atra | +0.051 |
Neg logit contributions | |
Proced | -0.056 |
yip | -0.054 |
Fn | -0.049 |
aning | -0.049 |
Address | -0.049 |
Token | � |
Token ID | 250 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
Proced | +0.005 |
yip | +0.005 |
Fn | +0.005 |
Address | +0.005 |
Agg | +0.005 |
Neg logit contributions | |
eyed | -0.004 |
CBS | -0.004 |
atra | -0.004 |
didn | -0.004 |
photos | -0.004 |
Token | the |
Token ID | 1169 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
eyed | +0.039 |
CBS | +0.039 |
didn | +0.038 |
wasn | +0.038 |
hadn | +0.037 |
Neg logit contributions | |
Proced | -0.045 |
yip | -0.044 |
Address | -0.040 |
aning | -0.040 |
Agg | -0.040 |
Token | root |
Token ID | 6808 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
CBS | +0.148 |
eyed | +0.147 |
didn | +0.147 |
photos | +0.143 |
atra | +0.141 |
Neg logit contributions | |
Proced | -0.172 |
yip | -0.165 |
Fn | -0.153 |
aning | -0.152 |
Address | -0.152 |
Token | part |
Token ID | 3911 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
Proced | +0.046 |
yip | +0.044 |
Address | +0.040 |
Fn | +0.040 |
aning | +0.040 |
Neg logit contributions | |
eyed | -0.046 |
didn | -0.045 |
CBS | -0.045 |
wasn | -0.044 |
atra | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
eyed | +0.032 |
didn | +0.032 |
CBS | +0.032 |
wasn | +0.031 |
atra | +0.031 |
Neg logit contributions | |
Proced | -0.032 |
yip | -0.030 |
Address | -0.028 |
aning | -0.028 |
Fn | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
eyed | +0.048 |
didn | +0.047 |
CBS | +0.047 |
wasn | +0.046 |
hadn | +0.046 |
Neg logit contributions | |
Proced | -0.057 |
yip | -0.054 |
Address | -0.050 |
Fn | -0.050 |
aning | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
eyed | +0.078 |
didn | +0.077 |
CBS | +0.077 |
atra | +0.076 |
wasn | +0.076 |
Neg logit contributions | |
Proced | -0.059 |
yip | -0.055 |
Address | -0.051 |
Agg | -0.051 |
aning | -0.050 |
Token | En |
Token ID | 4834 |
Feature activation | +8.10e-03 |
Pos logit contributions | |
eyed | +0.070 |
didn | +0.069 |
wasn | +0.068 |
hadn | +0.068 |
atra | +0.068 |
Neg logit contributions | |
Proced | -0.064 |
yip | -0.060 |
Address | -0.059 |
Agg | -0.058 |
Finding | -0.056 |
Token | vision |
Token ID | 10178 |
Feature activation | -6.45e-01 |
Pos logit contributions | |
Proced | +0.006 |
yip | +0.006 |
Address | +0.006 |
aning | +0.006 |
Fn | +0.006 |
Neg logit contributions | |
eyed | -0.005 |
didn | -0.005 |
CBS | -0.005 |
atra | -0.004 |
wasn | -0.004 |
Token | ed |
Token ID | 276 |
Feature activation | +4.65e-03 |
Pos logit contributions | |
eyed | +0.566 |
didn | +0.561 |
CBS | +0.559 |
atra | +0.550 |
wasn | +0.549 |
Neg logit contributions | |
Proced | -0.306 |
yip | -0.283 |
Address | -0.249 |
Fn | -0.248 |
aning | -0.246 |
Token | as |
Token ID | 355 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
Proced | +0.003 |
yip | +0.003 |
Address | +0.003 |
Fn | +0.003 |
aning | +0.003 |
Neg logit contributions | |
eyed | -0.003 |
didn | -0.003 |
CBS | -0.003 |
atra | -0.003 |
wasn | -0.003 |
Token | one |
Token ID | 530 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
eyed | +0.031 |
didn | +0.030 |
CBS | +0.030 |
atra | +0.029 |
wasn | +0.029 |
Neg logit contributions | |
Proced | -0.031 |
yip | -0.030 |
Address | -0.027 |
Fn | -0.027 |
aning | -0.027 |
Token | cohesive |
Token ID | 42403 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
eyed | +0.041 |
didn | +0.040 |
CBS | +0.040 |
wasn | +0.039 |
atra | +0.039 |
Neg logit contributions | |
Proced | -0.046 |
yip | -0.044 |
Address | -0.040 |
Fn | -0.040 |
Agg | -0.040 |
Token | public |
Token ID | 1171 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
Proced | +0.081 |
yip | +0.075 |
Fn | +0.070 |
Address | +0.070 |
aning | +0.069 |
Neg logit contributions | |
eyed | -0.082 |
didn | -0.082 |
CBS | -0.081 |
atra | -0.079 |
wasn | -0.078 |
Token | � |
Token ID | 447 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
Proced | +0.032 |
yip | +0.031 |
aning | +0.029 |
Fn | +0.029 |
Address | +0.028 |
Neg logit contributions | |
eyed | -0.029 |
CBS | -0.029 |
atra | -0.028 |
didn | -0.027 |
photos | -0.027 |
Token | � |
Token ID | 247 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
Proced | +0.041 |
yip | +0.040 |
Address | +0.037 |
Fn | +0.037 |
Agg | +0.037 |
Neg logit contributions | |
eyed | -0.021 |
CBS | -0.021 |
didn | -0.021 |
atra | -0.020 |
wasn | -0.020 |
Token | m |
Token ID | 76 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
CBS | +0.010 |
eyed | +0.010 |
atra | +0.010 |
didn | +0.010 |
hadn | +0.010 |
Neg logit contributions | |
Proced | -0.018 |
yip | -0.018 |
aning | -0.017 |
Fn | -0.017 |
Address | -0.017 |
Token | sorry |
Token ID | 7926 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
eyed | +0.088 |
CBS | +0.087 |
didn | +0.086 |
atra | +0.085 |
photos | +0.083 |
Neg logit contributions | |
Proced | -0.114 |
yip | -0.109 |
aning | -0.102 |
Address | -0.102 |
Fn | -0.101 |
Token | , |
Token ID | 11 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
eyed | +0.074 |
CBS | +0.073 |
didn | +0.072 |
atra | +0.072 |
wasn | +0.070 |
Neg logit contributions | |
Proced | -0.070 |
yip | -0.067 |
Address | -0.061 |
aning | -0.061 |
Fn | -0.061 |
Token | sir |
Token ID | 15967 |
Feature activation | -6.39e-01 |
Pos logit contributions | |
Proced | +0.039 |
yip | +0.038 |
Address | +0.035 |
aning | +0.035 |
Fn | +0.035 |
Neg logit contributions | |
eyed | -0.030 |
CBS | -0.030 |
atra | -0.030 |
didn | -0.029 |
photos | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | -6.91e-03 |
Pos logit contributions | |
eyed | +0.452 |
CBS | +0.446 |
didn | +0.445 |
atra | +0.438 |
wasn | +0.436 |
Neg logit contributions | |
Proced | -0.408 |
yip | -0.391 |
Address | -0.353 |
Fn | -0.352 |
aning | -0.352 |
Token | but |
Token ID | 475 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
eyed | +0.004 |
CBS | +0.004 |
atra | +0.004 |
didn | +0.004 |
photos | +0.004 |
Neg logit contributions | |
Proced | -0.005 |
yip | -0.005 |
Address | -0.005 |
aning | -0.005 |
Fn | -0.005 |
Token | you |
Token ID | 345 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
eyed | +0.090 |
CBS | +0.088 |
atra | +0.088 |
didn | +0.086 |
photos | +0.083 |
Neg logit contributions | |
Proced | -0.113 |
yip | -0.108 |
Fn | -0.101 |
aning | -0.100 |
Address | -0.100 |
Token | � |
Token ID | 447 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
Proced | +0.014 |
yip | +0.013 |
aning | +0.013 |
Fn | +0.012 |
Address | +0.012 |
Neg logit contributions | |
eyed | -0.015 |
CBS | -0.014 |
atra | -0.014 |
didn | -0.014 |
photos | -0.014 |
Token | � |
Token ID | 247 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
Proced | +0.014 |
yip | +0.014 |
Fn | +0.013 |
Address | +0.013 |
Agg | +0.013 |
Neg logit contributions | |
eyed | -0.009 |
CBS | -0.009 |
didn | -0.009 |
atra | -0.009 |
wasn | -0.008 |
Token | � |
Token ID | 247 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
eyed | +0.015 |
CBS | +0.015 |
didn | +0.015 |
atra | +0.014 |
wasn | +0.014 |
Neg logit contributions | |
Proced | -0.030 |
yip | -0.029 |
Fn | -0.027 |
Address | -0.027 |
Agg | -0.027 |
Token | ll |
Token ID | 297 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
eyed | +0.045 |
CBS | +0.045 |
atra | +0.044 |
didn | +0.043 |
wasn | +0.043 |
Neg logit contributions | |
Proced | -0.070 |
yip | -0.068 |
aning | -0.063 |
Address | -0.063 |
Fn | -0.063 |
Token | excuse |
Token ID | 12226 |
Feature activation | -3.91e-03 |
Pos logit contributions | |
Proced | +0.014 |
yip | +0.013 |
aning | +0.012 |
Address | +0.012 |
Fn | +0.012 |
Neg logit contributions | |
eyed | -0.014 |
CBS | -0.014 |
atra | -0.014 |
didn | -0.014 |
photos | -0.013 |
Token | me |
Token ID | 502 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
eyed | +0.003 |
CBS | +0.003 |
atra | +0.003 |
didn | +0.003 |
wasn | +0.003 |
Neg logit contributions | |
Proced | -0.002 |
yip | -0.002 |
Address | -0.002 |
Fn | -0.002 |
Agg | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
Proced | +0.012 |
yip | +0.012 |
Address | +0.011 |
Fn | +0.011 |
aning | +0.011 |
Neg logit contributions | |
eyed | -0.013 |
didn | -0.012 |
CBS | -0.012 |
atra | -0.012 |
wasn | -0.012 |
Token | sir |
Token ID | 15967 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
Proced | +0.028 |
yip | +0.027 |
aning | +0.025 |
Fn | +0.025 |
Address | +0.025 |
Neg logit contributions | |
eyed | -0.022 |
CBS | -0.022 |
atra | -0.022 |
didn | -0.022 |
photos | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | +3.10e-03 |
Pos logit contributions | |
eyed | +0.419 |
didn | +0.414 |
CBS | +0.413 |
wasn | +0.406 |
atra | +0.403 |
Neg logit contributions | |
Proced | -0.436 |
yip | -0.420 |
Address | -0.381 |
Fn | -0.380 |
Agg | -0.378 |
Token | I |
Token ID | 314 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
Proced | +0.002 |
yip | +0.002 |
aning | +0.002 |
Fn | +0.002 |
Address | +0.002 |
Neg logit contributions | |
eyed | -0.002 |
CBS | -0.002 |
atra | -0.002 |
didn | -0.002 |
photos | -0.002 |
Token | � |
Token ID | 447 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
eyed | +0.091 |
CBS | +0.090 |
atra | +0.089 |
didn | +0.089 |
photos | +0.086 |
Neg logit contributions | |
Proced | -0.090 |
yip | -0.086 |
aning | -0.079 |
Address | -0.079 |
Fn | -0.078 |
Token | � |
Token ID | 247 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
eyed | +0.020 |
CBS | +0.020 |
didn | +0.019 |
atra | +0.019 |
wasn | +0.019 |
Neg logit contributions | |
Proced | -0.054 |
yip | -0.052 |
Address | -0.049 |
Fn | -0.049 |
Agg | -0.049 |
Token | m |
Token ID | 76 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
CBS | +0.054 |
eyed | +0.053 |
atra | +0.052 |
didn | +0.051 |
wasn | +0.050 |
Neg logit contributions | |
Proced | -0.092 |
yip | -0.090 |
aning | -0.084 |
Address | -0.083 |
Fn | -0.083 |
Token | biggest |
Token ID | 4094 |
Feature activation | -8.00e-04 |
Pos logit contributions | |
eyed | +0.082 |
didn | +0.081 |
CBS | +0.081 |
atra | +0.079 |
wasn | +0.078 |
Neg logit contributions | |
Proced | -0.080 |
yip | -0.076 |
Address | -0.070 |
aning | -0.069 |
Fn | -0.069 |
Token | owner |
Token ID | 4870 |
Feature activation | -5.74e-02 |
Pos logit contributions | |
eyed | +0.000 |
didn | +0.000 |
CBS | +0.000 |
atra | +0.000 |
wasn | +0.000 |
Neg logit contributions | |
Proced | -0.001 |
yip | -0.001 |
Address | -0.001 |
Agg | -0.001 |
aning | -0.001 |
Token | of |
Token ID | 286 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
eyed | +0.038 |
didn | +0.037 |
CBS | +0.037 |
atra | +0.036 |
wasn | +0.036 |
Neg logit contributions | |
Proced | -0.039 |
yip | -0.038 |
Address | -0.034 |
Fn | -0.034 |
Agg | -0.034 |
Token | coal |
Token ID | 5655 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
eyed | +0.014 |
didn | +0.014 |
CBS | +0.014 |
atra | +0.013 |
wasn | +0.013 |
Neg logit contributions | |
Proced | -0.020 |
yip | -0.020 |
aning | -0.018 |
Fn | -0.018 |
Agg | -0.018 |
Token | - |
Token ID | 12 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
eyed | +0.281 |
didn | +0.279 |
CBS | +0.278 |
atra | +0.272 |
wasn | +0.267 |
Neg logit contributions | |
Proced | -0.302 |
yip | -0.288 |
Address | -0.267 |
Agg | -0.266 |
Fn | -0.264 |
Token | fired |
Token ID | 26803 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
CBS | +0.050 |
didn | +0.049 |
atra | +0.049 |
eyed | +0.049 |
wasn | +0.048 |
Neg logit contributions | |
Proced | -0.162 |
yip | -0.159 |
Address | -0.151 |
Agg | -0.151 |
aning | -0.149 |
Token | power |
Token ID | 1176 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
eyed | +0.424 |
didn | +0.422 |
CBS | +0.422 |
atra | +0.419 |
wasn | +0.403 |
Neg logit contributions | |
Proced | -0.414 |
yip | -0.391 |
Agg | -0.364 |
Address | -0.364 |
aning | -0.364 |
Token | station |
Token ID | 4429 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
eyed | +0.079 |
didn | +0.078 |
CBS | +0.078 |
atra | +0.076 |
wasn | +0.074 |
Neg logit contributions | |
Proced | -0.129 |
yip | -0.123 |
Agg | -0.117 |
Address | -0.116 |
aning | -0.115 |
Token | – |
Token ID | 784 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
eyed | +0.214 |
didn | +0.212 |
CBS | +0.211 |
atra | +0.207 |
wasn | +0.204 |
Neg logit contributions | |
Proced | -0.232 |
yip | -0.221 |
Fn | -0.203 |
Address | -0.203 |
Agg | -0.202 |
Token | said |
Token ID | 531 |
Feature activation | -9.36e-02 |
Pos logit contributions | |
eyed | +0.097 |
CBS | +0.097 |
didn | +0.094 |
atra | +0.094 |
photos | +0.094 |
Neg logit contributions | |
Proced | -0.117 |
yip | -0.113 |
Address | -0.105 |
aning | -0.104 |
Fn | -0.104 |
Token | that |
Token ID | 326 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
eyed | +0.057 |
CBS | +0.056 |
didn | +0.056 |
atra | +0.055 |
photos | +0.053 |
Neg logit contributions | |
Proced | -0.069 |
yip | -0.066 |
aning | -0.061 |
Address | -0.061 |
Fn | -0.061 |
Token | top |
Token ID | 1353 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
eyed | +0.033 |
didn | +0.033 |
CBS | +0.033 |
atra | +0.032 |
photos | +0.031 |
Neg logit contributions | |
Proced | -0.042 |
yip | -0.039 |
Fn | -0.037 |
aning | -0.037 |
Address | -0.037 |
Token | executives |
Token ID | 12353 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
Proced | +0.062 |
yip | +0.059 |
Fn | +0.056 |
aning | +0.056 |
Address | +0.055 |
Neg logit contributions | |
eyed | -0.037 |
CBS | -0.036 |
didn | -0.036 |
photos | -0.035 |
atra | -0.035 |
Token | : |
Token ID | 25 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
eyed | +0.220 |
didn | +0.218 |
CBS | +0.217 |
wasn | +0.214 |
atra | +0.213 |
Neg logit contributions | |
Proced | -0.194 |
yip | -0.186 |
Address | -0.168 |
Fn | -0.167 |
Agg | -0.166 |
Token | President |
Token ID | 1992 |
Feature activation | -9.86e-02 |
Pos logit contributions | |
Proced | +0.102 |
yip | +0.096 |
Fn | +0.090 |
aning | +0.090 |
Agg | +0.089 |
Neg logit contributions | |
eyed | -0.079 |
CBS | -0.077 |
didn | -0.077 |
atra | -0.075 |
photos | -0.074 |
Token | Gary |
Token ID | 10936 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
eyed | +0.057 |
CBS | +0.056 |
didn | +0.055 |
photos | +0.054 |
atra | +0.054 |
Neg logit contributions | |
Proced | -0.076 |
yip | -0.072 |
Fn | -0.068 |
aning | -0.067 |
Address | -0.067 |
Token | Cohn |
Token ID | 45005 |
Feature activation | -6.25e-01 |
Pos logit contributions | |
eyed | +0.069 |
CBS | +0.067 |
didn | +0.065 |
atra | +0.061 |
photos | +0.058 |
Neg logit contributions | |
Proced | -0.329 |
yip | -0.315 |
Fn | -0.303 |
aning | -0.302 |
Address | -0.301 |
Token | ; |
Token ID | 26 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
eyed | +0.399 |
CBS | +0.393 |
didn | +0.390 |
atra | +0.383 |
photos | +0.380 |
Neg logit contributions | |
Proced | -0.442 |
yip | -0.421 |
Address | -0.389 |
Fn | -0.389 |
aning | -0.388 |
Token | David |
Token ID | 3271 |
Feature activation | -6.59e-02 |
Pos logit contributions | |
Proced | +0.030 |
yip | +0.029 |
Fn | +0.027 |
Address | +0.027 |
aning | +0.026 |
Neg logit contributions | |
eyed | -0.023 |
didn | -0.022 |
CBS | -0.022 |
atra | -0.022 |
photos | -0.021 |
Token | V |
Token ID | 569 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
eyed | +0.043 |
CBS | +0.042 |
didn | +0.042 |
atra | +0.041 |
wasn | +0.041 |
Neg logit contributions | |
Proced | -0.047 |
yip | -0.044 |
Fn | -0.041 |
aning | -0.041 |
Address | -0.040 |
Token | ini |
Token ID | 5362 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
didn | +0.183 |
eyed | +0.183 |
CBS | +0.182 |
photos | +0.178 |
wasn | +0.177 |
Neg logit contributions | |
Proced | -0.199 |
yip | -0.193 |
aning | -0.180 |
Fn | -0.178 |
Address | -0.178 |
Token | ar |
Token ID | 283 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
eyed | +0.102 |
didn | +0.100 |
CBS | +0.100 |
wasn | +0.098 |
atra | +0.097 |
Neg logit contributions | |
Proced | -0.114 |
yip | -0.109 |
Address | -0.100 |
Fn | -0.100 |
Agg | -0.099 |
Token | т |
Token ID | 20375 |
Feature activation | -8.32e-02 |
Pos logit contributions | |
eyed | +0.078 |
CBS | +0.078 |
didn | +0.078 |
atra | +0.076 |
wasn | +0.076 |
Neg logit contributions | |
Proced | -0.040 |
yip | -0.038 |
aning | -0.033 |
Address | -0.033 |
Agg | -0.032 |
Token | ь |
Token ID | 45367 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
eyed | +0.060 |
CBS | +0.059 |
didn | +0.059 |
atra | +0.057 |
wasn | +0.057 |
Neg logit contributions | |
Proced | -0.053 |
yip | -0.050 |
Address | -0.046 |
aning | -0.045 |
Agg | -0.045 |
Token | � |
Token ID | 141 |
Feature activation | +6.20e-02 |
Pos logit contributions | |
eyed | +0.229 |
didn | +0.226 |
CBS | +0.226 |
atra | +0.221 |
wasn | +0.221 |
Neg logit contributions | |
Proced | -0.191 |
yip | -0.182 |
Address | -0.164 |
Fn | -0.164 |
aning | -0.163 |
Token | � |
Token ID | 236 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
Proced | +0.038 |
yip | +0.036 |
aning | +0.032 |
Fn | +0.032 |
Address | +0.032 |
Neg logit contributions | |
eyed | -0.046 |
didn | -0.045 |
CBS | -0.045 |
atra | -0.044 |
wasn | -0.044 |
Token | � |
Token ID | 12466 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
eyed | +0.258 |
didn | +0.256 |
CBS | +0.255 |
wasn | +0.250 |
atra | +0.249 |
Neg logit contributions | |
Proced | -0.198 |
yip | -0.188 |
Address | -0.169 |
Fn | -0.169 |
aning | -0.168 |
Token | � |
Token ID | 112 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
eyed | +0.181 |
didn | +0.178 |
CBS | +0.178 |
atra | +0.175 |
wasn | +0.174 |
Neg logit contributions | |
Proced | -0.174 |
yip | -0.167 |
Fn | -0.152 |
Address | -0.151 |
aning | -0.151 |
Token | а |
Token ID | 16142 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
eyed | +0.455 |
didn | +0.450 |
CBS | +0.449 |
wasn | +0.440 |
atra | +0.438 |
Neg logit contributions | |
Proced | -0.366 |
yip | -0.348 |
Address | -0.313 |
Fn | -0.312 |
aning | -0.311 |
Token | т |
Token ID | 20375 |
Feature activation | -9.95e-02 |
Pos logit contributions | |
eyed | +0.360 |
didn | +0.356 |
CBS | +0.355 |
atra | +0.349 |
wasn | +0.349 |
Neg logit contributions | |
Proced | -0.201 |
yip | -0.186 |
Address | -0.165 |
Fn | -0.164 |
aning | -0.162 |
Token | |
Token ID | 220 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
eyed | +0.073 |
CBS | +0.073 |
didn | +0.073 |
wasn | +0.071 |
atra | +0.071 |
Neg logit contributions | |
Proced | -0.061 |
yip | -0.058 |
Address | -0.052 |
aning | -0.052 |
Fn | -0.052 |
Token | с |
Token ID | 21727 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
eyed | +0.333 |
didn | +0.329 |
CBS | +0.328 |
wasn | +0.323 |
atra | +0.322 |
Neg logit contributions | |
Proced | -0.191 |
yip | -0.179 |
Address | -0.158 |
aning | -0.157 |
Agg | -0.156 |
Token | м |
Token ID | 43108 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
eyed | +0.104 |
didn | +0.102 |
CBS | +0.102 |
wasn | +0.100 |
atra | +0.100 |
Neg logit contributions | |
Proced | -0.071 |
yip | -0.067 |
Address | -0.059 |
Fn | -0.059 |
Agg | -0.059 |
Token | oliath |
Token ID | 43009 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Proced | +0.046 |
yip | +0.044 |
Address | +0.038 |
aning | +0.038 |
Agg | +0.038 |
Neg logit contributions | |
eyed | -0.075 |
didn | -0.075 |
CBS | -0.075 |
wasn | -0.073 |
atra | -0.073 |
Token | TBA |
Token ID | 48591 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
Proced | +0.121 |
yip | +0.114 |
Fn | +0.103 |
Address | +0.103 |
Agg | +0.102 |
Neg logit contributions | |
eyed | -0.144 |
didn | -0.142 |
CBS | -0.141 |
atra | -0.139 |
wasn | -0.138 |
Token | TBA |
Token ID | 48591 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
Proced | +0.144 |
yip | +0.135 |
Fn | +0.123 |
Address | +0.123 |
Agg | +0.122 |
Neg logit contributions | |
eyed | -0.160 |
didn | -0.158 |
CBS | -0.156 |
atra | -0.154 |
wasn | -0.154 |
Token |
|
Token ID | 198 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
Proced | +0.182 |
yip | +0.172 |
Fn | +0.161 |
Address | +0.160 |
Agg | +0.159 |
Neg logit contributions | |
eyed | -0.128 |
didn | -0.125 |
CBS | -0.124 |
atra | -0.121 |
wasn | -0.121 |
Token |
|
Token ID | 198 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
Proced | +0.024 |
yip | +0.023 |
Address | +0.021 |
Agg | +0.021 |
aning | +0.021 |
Neg logit contributions | |
eyed | -0.032 |
didn | -0.031 |
CBS | -0.031 |
atra | -0.031 |
wasn | -0.031 |
Token | CBS |
Token ID | 22923 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
Proced | +0.023 |
yip | +0.021 |
Address | +0.020 |
Agg | +0.020 |
AMS | +0.020 |
Neg logit contributions | |
eyed | -0.027 |
didn | -0.026 |
wasn | -0.026 |
hadn | -0.026 |
atra | -0.026 |
Token | All |
Token ID | 1439 |
Feature activation | +7.91e-02 |
Pos logit contributions | |
eyed | +0.391 |
didn | +0.383 |
atra | +0.379 |
CBS | +0.376 |
photos | +0.371 |
Neg logit contributions | |
Proced | -0.434 |
yip | -0.399 |
Fn | -0.377 |
Address | -0.375 |
Agg | -0.374 |
Token | Access |
Token ID | 8798 |
Feature activation | +8.97e-02 |
Pos logit contributions | |
Proced | +0.048 |
yip | +0.043 |
Address | +0.041 |
aning | +0.041 |
Fn | +0.040 |
Neg logit contributions | |
eyed | -0.059 |
didn | -0.059 |
atra | -0.058 |
CBS | -0.058 |
wasn | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
Proced | +0.062 |
yip | +0.058 |
Fn | +0.054 |
Address | +0.054 |
Agg | +0.053 |
Neg logit contributions | |
eyed | -0.060 |
didn | -0.059 |
atra | -0.058 |
CBS | -0.057 |
turn | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
Proced | +0.025 |
yip | +0.023 |
Address | +0.022 |
Agg | +0.021 |
aning | +0.021 |
Neg logit contributions | |
eyed | -0.033 |
didn | -0.033 |
CBS | -0.033 |
atra | -0.032 |
wasn | -0.032 |
Token | All |
Token ID | 3237 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
Proced | +0.031 |
yip | +0.030 |
Address | +0.029 |
Agg | +0.028 |
AMS | +0.028 |
Neg logit contributions | |
eyed | -0.032 |
didn | -0.032 |
wasn | -0.032 |
hadn | -0.031 |
atra | -0.031 |
Token | FR |
Token ID | 8782 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
Proced | +0.034 |
yip | +0.033 |
Fn | +0.030 |
Address | +0.030 |
Agg | +0.029 |
Neg logit contributions | |
eyed | -0.038 |
didn | -0.038 |
CBS | -0.037 |
wasn | -0.036 |
atra | -0.036 |
Token | ANC |
Token ID | 20940 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
Proced | +0.015 |
yip | +0.014 |
Address | +0.012 |
Agg | +0.011 |
Fn | +0.011 |
Neg logit contributions | |
eyed | -0.036 |
didn | -0.036 |
CBS | -0.035 |
wasn | -0.035 |
atra | -0.035 |
Token | IS |
Token ID | 1797 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
Proced | +0.169 |
Address | +0.149 |
yip | +0.148 |
AMS | +0.143 |
Agg | +0.142 |
Neg logit contributions | |
eyed | -0.300 |
didn | -0.295 |
atra | -0.289 |
wasn | -0.289 |
hadn | -0.288 |
Token | CO |
Token ID | 8220 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
Proced | +0.246 |
yip | +0.237 |
Address | +0.227 |
Agg | +0.224 |
Fn | +0.224 |
Neg logit contributions | |
eyed | -0.093 |
didn | -0.091 |
CBS | -0.087 |
wasn | -0.086 |
hadn | -0.085 |
Token | ( |
Token ID | 357 |
Feature activation | +5.08e-03 |
Pos logit contributions | |
eyed | +0.083 |
didn | +0.082 |
CBS | +0.082 |
wasn | +0.080 |
atra | +0.079 |
Neg logit contributions | |
Proced | -0.143 |
yip | -0.139 |
Address | -0.128 |
Fn | -0.128 |
Agg | -0.128 |
Token | CBS |
Token ID | 22923 |
Feature activation | -6.03e-01 |
Pos logit contributions | |
Proced | +0.004 |
yip | +0.004 |
Address | +0.004 |
Agg | +0.004 |
AMS | +0.004 |
Neg logit contributions | |
eyed | -0.002 |
didn | -0.002 |
wasn | -0.002 |
hadn | -0.002 |
atra | -0.002 |
Token | SF |
Token ID | 14362 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
didn | +0.408 |
eyed | +0.407 |
atra | +0.399 |
hadn | +0.394 |
wasn | +0.393 |
Neg logit contributions | |
Proced | -0.408 |
yip | -0.355 |
Agg | -0.352 |
Fn | -0.350 |
Address | -0.348 |
Token | ) |
Token ID | 8 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
eyed | +0.021 |
didn | +0.021 |
atra | +0.020 |
wasn | +0.020 |
CBS | +0.020 |
Neg logit contributions | |
Proced | -0.041 |
yip | -0.038 |
Agg | -0.037 |
Address | -0.037 |
AMS | -0.036 |
Token | — |
Token ID | 851 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
eyed | +0.070 |
didn | +0.068 |
CBS | +0.066 |
atra | +0.065 |
wasn | +0.065 |
Neg logit contributions | |
Proced | -0.172 |
yip | -0.164 |
Address | -0.156 |
Fn | -0.155 |
Agg | -0.155 |
Token | Both |
Token ID | 5747 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
eyed | +0.098 |
didn | +0.096 |
CBS | +0.094 |
atra | +0.093 |
wasn | +0.093 |
Neg logit contributions | |
Proced | -0.118 |
yip | -0.109 |
Fn | -0.104 |
Address | -0.103 |
Agg | -0.103 |
Token | Google |
Token ID | 3012 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
Proced | +0.169 |
yip | +0.161 |
Fn | +0.150 |
Address | +0.149 |
aning | +0.148 |
Neg logit contributions | |
eyed | -0.130 |
didn | -0.129 |
CBS | -0.126 |
atra | -0.124 |
wasn | -0.123 |
Token | , |
Token ID | 11 |
Feature activation | +7.70e-02 |
Pos logit contributions | |
Proced | +0.079 |
yip | +0.076 |
Address | +0.069 |
Agg | +0.068 |
Fn | +0.068 |
Neg logit contributions | |
eyed | -0.086 |
didn | -0.084 |
CBS | -0.084 |
atra | -0.083 |
wasn | -0.082 |
Token | 055 |
Token ID | 47838 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
Proced | +0.056 |
yip | +0.055 |
Address | +0.050 |
Fn | +0.050 |
Agg | +0.050 |
Neg logit contributions | |
eyed | -0.047 |
didn | -0.046 |
CBS | -0.046 |
atra | -0.045 |
wasn | -0.045 |
Token | complaints |
Token ID | 9687 |
Feature activation | +8.28e-03 |
Pos logit contributions | |
eyed | +0.117 |
didn | +0.114 |
CBS | +0.114 |
atra | +0.112 |
photos | +0.110 |
Neg logit contributions | |
Proced | -0.123 |
yip | -0.117 |
Fn | -0.108 |
Address | -0.108 |
Agg | -0.107 |
Token | filed |
Token ID | 5717 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
Proced | +0.006 |
yip | +0.006 |
Fn | +0.005 |
Agg | +0.005 |
aning | +0.005 |
Neg logit contributions | |
eyed | -0.005 |
didn | -0.005 |
CBS | -0.005 |
atra | -0.005 |
photos | -0.005 |
Token | against |
Token ID | 1028 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
eyed | +0.164 |
didn | +0.162 |
CBS | +0.161 |
atra | +0.158 |
wasn | +0.158 |
Neg logit contributions | |
Proced | -0.163 |
yip | -0.157 |
Address | -0.142 |
Agg | -0.142 |
Fn | -0.142 |
Token | Wells |
Token ID | 18292 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
eyed | +0.111 |
didn | +0.108 |
CBS | +0.107 |
atra | +0.106 |
wasn | +0.105 |
Neg logit contributions | |
Proced | -0.154 |
yip | -0.145 |
Fn | -0.136 |
Address | -0.136 |
aning | -0.135 |
Token | Fargo |
Token ID | 28125 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
eyed | +0.315 |
didn | +0.309 |
CBS | +0.305 |
atra | +0.299 |
wasn | +0.296 |
Neg logit contributions | |
Proced | -0.493 |
yip | -0.469 |
Address | -0.441 |
Fn | -0.440 |
aning | -0.440 |
Token | — |
Token ID | 960 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
eyed | +0.116 |
didn | +0.114 |
CBS | +0.114 |
atra | +0.112 |
wasn | +0.111 |
Neg logit contributions | |
Proced | -0.118 |
yip | -0.113 |
Address | -0.103 |
Agg | -0.103 |
Fn | -0.103 |
Token | which |
Token ID | 4758 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
eyed | +0.134 |
didn | +0.131 |
atra | +0.131 |
wasn | +0.131 |
hadn | +0.130 |
Neg logit contributions | |
Proced | -0.143 |
yip | -0.139 |
Agg | -0.129 |
Address | -0.128 |
aning | -0.127 |
Token | is |
Token ID | 318 |
Feature activation | +4.35e-02 |
Pos logit contributions | |
Proced | +0.093 |
yip | +0.088 |
Fn | +0.082 |
Address | +0.081 |
Agg | +0.081 |
Neg logit contributions | |
eyed | -0.091 |
CBS | -0.088 |
didn | -0.087 |
atra | -0.087 |
photos | -0.085 |
Token | accused |
Token ID | 5371 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
Proced | +0.031 |
yip | +0.029 |
Fn | +0.027 |
aning | +0.027 |
Address | +0.027 |
Neg logit contributions | |
eyed | -0.028 |
didn | -0.027 |
atra | -0.027 |
CBS | -0.027 |
photos | -0.026 |
Token | S |
Token ID | 50 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
eyed | +0.004 |
didn | +0.004 |
atra | +0.004 |
wasn | +0.004 |
hadn | +0.004 |
Neg logit contributions | |
Proced | -0.004 |
yip | -0.004 |
Agg | -0.004 |
Address | -0.004 |
Finding | -0.004 |
Token | par |
Token ID | 1845 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
didn | +0.016 |
eyed | +0.016 |
CBS | +0.016 |
wasn | +0.015 |
atra | +0.015 |
Neg logit contributions | |
Proced | -0.022 |
yip | -0.021 |
aning | -0.020 |
AMS | -0.020 |
Address | -0.019 |
Token | ks |
Token ID | 591 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
eyed | +0.128 |
didn | +0.126 |
CBS | +0.124 |
wasn | +0.122 |
atra | +0.121 |
Neg logit contributions | |
Proced | -0.145 |
yip | -0.139 |
aning | -0.131 |
Address | -0.130 |
Agg | -0.129 |
Token | , |
Token ID | 11 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
Proced | +0.094 |
yip | +0.089 |
Fn | +0.085 |
aning | +0.084 |
Address | +0.084 |
Neg logit contributions | |
eyed | -0.067 |
CBS | -0.067 |
didn | -0.065 |
photos | -0.065 |
atra | -0.065 |
Token | a |
Token ID | 257 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
Proced | +0.011 |
yip | +0.011 |
Fn | +0.010 |
Agg | +0.010 |
aning | +0.010 |
Neg logit contributions | |
eyed | -0.010 |
didn | -0.010 |
CBS | -0.010 |
atra | -0.010 |
photos | -0.010 |
Token | Cohn |
Token ID | 45005 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
Proced | +0.034 |
yip | +0.032 |
Fn | +0.030 |
Address | +0.029 |
Finding | +0.029 |
Neg logit contributions | |
eyed | -0.029 |
didn | -0.029 |
CBS | -0.029 |
atra | -0.028 |
photos | -0.028 |
Token | prot |
Token ID | 1237 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
CBS | +0.316 |
eyed | +0.316 |
didn | +0.315 |
atra | +0.301 |
photos | +0.301 |
Neg logit contributions | |
Proced | -0.480 |
yip | -0.461 |
Fn | -0.431 |
Address | -0.430 |
aning | -0.429 |
Token | é |
Token ID | 2634 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
CBS | +0.204 |
didn | +0.200 |
eyed | +0.199 |
wasn | +0.194 |
atra | +0.193 |
Neg logit contributions | |
Proced | -0.254 |
yip | -0.251 |
aning | -0.233 |
Address | -0.231 |
Agg | -0.229 |
Token | g |
Token ID | 70 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
CBS | +0.166 |
eyed | +0.163 |
didn | +0.162 |
atra | +0.161 |
hadn | +0.158 |
Neg logit contributions | |
Proced | -0.134 |
yip | -0.130 |
aning | -0.117 |
Address | -0.117 |
Fn | -0.116 |
Token | é |
Token ID | 2634 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
Proced | +0.032 |
yip | +0.031 |
Address | +0.029 |
aning | +0.029 |
Agg | +0.029 |
Neg logit contributions | |
CBS | -0.022 |
eyed | -0.022 |
didn | -0.022 |
atra | -0.021 |
wasn | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
eyed | +0.139 |
CBS | +0.138 |
didn | +0.138 |
atra | +0.135 |
wasn | +0.134 |
Neg logit contributions | |
Proced | -0.143 |
yip | -0.137 |
Fn | -0.125 |
Address | -0.125 |
aning | -0.125 |
Token | the |
Token ID | 262 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
Proced | +0.022 |
yip | +0.020 |
Fn | +0.019 |
aning | +0.019 |
Agg | +0.019 |
Neg logit contributions | |
eyed | -0.018 |
CBS | -0.017 |
didn | -0.017 |
atra | -0.017 |
photos | -0.017 |
Token | King |
Token ID | 2677 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
eyed | +0.012 |
CBS | +0.012 |
didn | +0.012 |
atra | +0.012 |
photos | +0.012 |
Neg logit contributions | |
Proced | -0.013 |
yip | -0.012 |
aning | -0.012 |
Fn | -0.012 |
Agg | -0.012 |
Token | 's |
Token ID | 338 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
eyed | +0.026 |
didn | +0.025 |
CBS | +0.025 |
photos | +0.025 |
atra | +0.024 |
Neg logit contributions | |
Proced | -0.042 |
yip | -0.040 |
aning | -0.037 |
Agg | -0.037 |
Fn | -0.037 |
Token | judgment |
Token ID | 8492 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
Proced | +0.034 |
yip | +0.032 |
aning | +0.030 |
Fn | +0.030 |
Address | +0.030 |
Neg logit contributions | |
eyed | -0.021 |
CBS | -0.021 |
didn | -0.021 |
atra | -0.020 |
photos | -0.020 |
Token | )." |
Token ID | 21387 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
eyed | +0.100 |
didn | +0.099 |
CBS | +0.098 |
atra | +0.096 |
photos | +0.095 |
Neg logit contributions | |
Proced | -0.184 |
yip | -0.179 |
aning | -0.166 |
Address | -0.166 |
Fn | -0.166 |
Token | Sah |
Token ID | 32194 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
eyed | +0.279 |
didn | +0.274 |
CBS | +0.273 |
atra | +0.268 |
wasn | +0.268 |
Neg logit contributions | |
Proced | -0.276 |
yip | -0.260 |
Address | -0.241 |
Agg | -0.241 |
Fn | -0.240 |
Token | ih |
Token ID | 4449 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Proced | +0.172 |
yip | +0.162 |
Agg | +0.142 |
Address | +0.142 |
aning | +0.141 |
Neg logit contributions | |
eyed | -0.293 |
CBS | -0.291 |
didn | -0.291 |
atra | -0.285 |
wasn | -0.285 |
Token | al |
Token ID | 435 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
Proced | +0.136 |
yip | +0.128 |
Address | +0.119 |
Fn | +0.119 |
aning | +0.119 |
Neg logit contributions | |
eyed | -0.129 |
didn | -0.126 |
CBS | -0.126 |
atra | -0.123 |
wasn | -0.123 |
Token | - |
Token ID | 12 |
Feature activation | -6.43e-03 |
Pos logit contributions | |
Proced | +0.171 |
yip | +0.166 |
Address | +0.154 |
Fn | +0.154 |
Agg | +0.153 |
Neg logit contributions | |
eyed | -0.101 |
didn | -0.099 |
CBS | -0.099 |
wasn | -0.096 |
atra | -0.096 |
Token | B |
Token ID | 33 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
eyed | +0.004 |
didn | +0.004 |
CBS | +0.003 |
wasn | +0.003 |
atra | +0.003 |
Neg logit contributions | |
Proced | -0.005 |
yip | -0.005 |
Address | -0.005 |
Agg | -0.005 |
Fn | -0.004 |
Token | uk |
Token ID | 2724 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
eyed | +0.055 |
CBS | +0.055 |
didn | +0.055 |
wasn | +0.053 |
hadn | +0.053 |
Neg logit contributions | |
Proced | -0.078 |
yip | -0.076 |
aning | -0.072 |
Agg | -0.071 |
Address | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Proced | +0.061 |
yip | +0.056 |
Address | +0.053 |
Agg | +0.052 |
aning | +0.052 |
Neg logit contributions | |
eyed | -0.081 |
didn | -0.080 |
CBS | -0.079 |
atra | -0.079 |
wasn | -0.078 |
Token | No |
Token ID | 2949 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
Proced | +0.073 |
yip | +0.068 |
Address | +0.066 |
Agg | +0.065 |
Finding | +0.064 |
Neg logit contributions | |
eyed | -0.083 |
didn | -0.080 |
wasn | -0.080 |
atra | -0.080 |
hadn | -0.080 |
Token | breakfast |
Token ID | 12607 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
eyed | +0.089 |
didn | +0.088 |
CBS | +0.088 |
atra | +0.086 |
wasn | +0.085 |
Neg logit contributions | |
Proced | -0.091 |
yip | -0.087 |
Fn | -0.080 |
aning | -0.080 |
Address | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Proced | +0.007 |
yip | +0.007 |
Fn | +0.006 |
Address | +0.006 |
aning | +0.006 |
Neg logit contributions | |
eyed | -0.007 |
CBS | -0.007 |
didn | -0.007 |
atra | -0.007 |
photos | -0.006 |
Token | lighter |
Token ID | 14871 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
Proced | +0.087 |
yip | +0.083 |
Fn | +0.079 |
aning | +0.078 |
Address | +0.078 |
Neg logit contributions | |
eyed | -0.051 |
CBS | -0.050 |
didn | -0.050 |
atra | -0.049 |
photos | -0.047 |
Token | lunch |
Token ID | 9965 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
Proced | +0.009 |
yip | +0.009 |
Fn | +0.008 |
aning | +0.008 |
Address | +0.008 |
Neg logit contributions | |
eyed | -0.009 |
didn | -0.009 |
CBS | -0.009 |
atra | -0.009 |
hadn | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Proced | +0.252 |
yip | +0.242 |
Fn | +0.222 |
Address | +0.222 |
aning | +0.221 |
Neg logit contributions | |
eyed | -0.247 |
didn | -0.243 |
CBS | -0.243 |
atra | -0.238 |
wasn | -0.235 |
Token | and |
Token ID | 290 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
Proced | +0.080 |
yip | +0.076 |
Fn | +0.072 |
aning | +0.072 |
Address | +0.071 |
Neg logit contributions | |
eyed | -0.062 |
CBS | -0.061 |
didn | -0.061 |
atra | -0.060 |
photos | -0.058 |
Token | larger |
Token ID | 4025 |
Feature activation | +8.82e-02 |
Pos logit contributions | |
Proced | +0.041 |
yip | +0.039 |
Fn | +0.037 |
aning | +0.036 |
Address | +0.036 |
Neg logit contributions | |
eyed | -0.033 |
CBS | -0.032 |
didn | -0.032 |
atra | -0.032 |
photos | -0.031 |
Token | dinner |
Token ID | 8073 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
Proced | +0.055 |
yip | +0.052 |
Fn | +0.049 |
aning | +0.048 |
Address | +0.048 |
Neg logit contributions | |
eyed | -0.063 |
didn | -0.063 |
CBS | -0.062 |
atra | -0.061 |
hadn | -0.060 |
Token | also |
Token ID | 635 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
Proced | +0.111 |
yip | +0.104 |
Fn | +0.098 |
Address | +0.097 |
aning | +0.096 |
Neg logit contributions | |
eyed | -0.123 |
CBS | -0.121 |
didn | -0.120 |
atra | -0.119 |
photos | -0.116 |
Token | rain |
Token ID | 6290 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
eyed | +0.009 |
CBS | +0.009 |
didn | +0.009 |
atra | +0.008 |
photos | +0.008 |
Neg logit contributions | |
Proced | -0.011 |
yip | -0.010 |
Fn | -0.009 |
aning | -0.009 |
Agg | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
eyed | +0.011 |
CBS | +0.011 |
didn | +0.011 |
atra | +0.010 |
wasn | +0.010 |
Neg logit contributions | |
Proced | -0.012 |
yip | -0.012 |
aning | -0.010 |
Address | -0.010 |
Fn | -0.010 |
Token | at |
Token ID | 379 |
Feature activation | -5.86e-02 |
Pos logit contributions | |
eyed | +0.013 |
CBS | +0.012 |
didn | +0.012 |
atra | +0.012 |
photos | +0.012 |
Neg logit contributions | |
Proced | -0.017 |
yip | -0.016 |
Fn | -0.015 |
aning | -0.015 |
Agg | -0.015 |
Token | night |
Token ID | 1755 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
eyed | +0.026 |
CBS | +0.026 |
didn | +0.025 |
hadn | +0.025 |
atra | +0.024 |
Neg logit contributions | |
Proced | -0.053 |
yip | -0.052 |
aning | -0.048 |
Fn | -0.048 |
Agg | -0.048 |
Token | in |
Token ID | 287 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
eyed | +0.255 |
didn | +0.253 |
CBS | +0.252 |
atra | +0.247 |
wasn | +0.246 |
Neg logit contributions | |
Proced | -0.226 |
yip | -0.215 |
Address | -0.194 |
Fn | -0.194 |
aning | -0.193 |
Token | Rad |
Token ID | 5325 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
eyed | +0.010 |
CBS | +0.010 |
didn | +0.010 |
atra | +0.010 |
wasn | +0.010 |
Neg logit contributions | |
Proced | -0.013 |
yip | -0.012 |
Fn | -0.011 |
Address | -0.011 |
aning | -0.011 |
Token | mobile |
Token ID | 24896 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
Proced | +0.267 |
yip | +0.254 |
Address | +0.233 |
aning | +0.233 |
Fn | +0.232 |
Neg logit contributions | |
eyed | -0.224 |
didn | -0.223 |
CBS | -0.223 |
wasn | -0.216 |
atra | -0.215 |
Token | inspired |
Token ID | 7867 |
Feature activation | +7.26e-02 |
Pos logit contributions | |
Proced | +0.083 |
yip | +0.080 |
Address | +0.072 |
Fn | +0.072 |
Agg | +0.072 |
Neg logit contributions | |
eyed | -0.083 |
didn | -0.082 |
CBS | -0.081 |
atra | -0.079 |
wasn | -0.079 |
Token | us |
Token ID | 514 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
Proced | +0.052 |
yip | +0.049 |
Fn | +0.045 |
Address | +0.045 |
aning | +0.045 |
Neg logit contributions | |
eyed | -0.047 |
didn | -0.046 |
CBS | -0.046 |
atra | -0.045 |
wasn | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
Proced | +0.065 |
yip | +0.062 |
Address | +0.054 |
Fn | +0.054 |
Agg | +0.054 |
Neg logit contributions | |
eyed | -0.097 |
didn | -0.096 |
CBS | -0.096 |
wasn | -0.094 |
atra | -0.094 |
Token | make |
Token ID | 787 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
Proced | +0.025 |
yip | +0.023 |
Address | +0.021 |
Fn | +0.021 |
aning | +0.021 |
Neg logit contributions | |
eyed | -0.028 |
didn | -0.028 |
CBS | -0.028 |
atra | -0.027 |
wasn | -0.027 |
Token | more |
Token ID | 517 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
Proced | +0.051 |
yip | +0.050 |
Fn | +0.046 |
aning | +0.046 |
Agg | +0.046 |
Neg logit contributions | |
eyed | -0.038 |
didn | -0.037 |
CBS | -0.037 |
atra | -0.036 |
photos | -0.036 |
Token | than |
Token ID | 621 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
eyed | +0.013 |
didn | +0.012 |
CBS | +0.012 |
atra | +0.012 |
photos | +0.011 |
Neg logit contributions | |
Proced | -0.071 |
yip | -0.069 |
Agg | -0.066 |
Address | -0.066 |
Fn | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | -8.68e-02 |
Pos logit contributions | |
eyed | +0.256 |
CBS | +0.247 |
didn | +0.247 |
atra | +0.243 |
photos | +0.239 |
Neg logit contributions | |
Proced | -0.347 |
yip | -0.334 |
Fn | -0.308 |
Agg | -0.307 |
Address | -0.306 |
Token | entire |
Token ID | 2104 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
eyed | +0.060 |
didn | +0.059 |
CBS | +0.059 |
atra | +0.058 |
wasn | +0.057 |
Neg logit contributions | |
Proced | -0.057 |
yip | -0.054 |
Agg | -0.049 |
Fn | -0.049 |
Address | -0.049 |
Token | EFF |
Token ID | 33659 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
Proced | +0.095 |
yip | +0.089 |
Fn | +0.082 |
Agg | +0.081 |
Address | +0.081 |
Neg logit contributions | |
eyed | -0.101 |
didn | -0.100 |
CBS | -0.099 |
photos | -0.097 |
atra | -0.097 |
Token | budget |
Token ID | 4466 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
Proced | +0.033 |
yip | +0.031 |
Agg | +0.028 |
Address | +0.028 |
Fn | +0.028 |
Neg logit contributions | |
eyed | -0.036 |
didn | -0.035 |
CBS | -0.035 |
photos | -0.034 |
atra | -0.034 |
Token | for |
Token ID | 329 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
Proced | +0.240 |
yip | +0.229 |
Agg | +0.206 |
Address | +0.206 |
Fn | +0.204 |
Neg logit contributions | |
eyed | -0.279 |
CBS | -0.272 |
didn | -0.272 |
atra | -0.268 |
photos | -0.265 |
Token | a |
Token ID | 257 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
Proced | +0.090 |
yip | +0.084 |
Fn | +0.079 |
Finding | +0.078 |
aning | +0.078 |
Neg logit contributions | |
eyed | -0.072 |
CBS | -0.070 |
didn | -0.069 |
atra | -0.068 |
photos | -0.068 |
Token | whole |
Token ID | 2187 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
Proced | +0.025 |
yip | +0.024 |
Fn | +0.021 |
aning | +0.021 |
Finding | +0.021 |
Neg logit contributions | |
eyed | -0.030 |
didn | -0.029 |
CBS | -0.029 |
atra | -0.028 |
photos | -0.028 |
Token | year |
Token ID | 614 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
Proced | +0.073 |
yip | +0.068 |
Fn | +0.061 |
Address | +0.061 |
aning | +0.061 |
Neg logit contributions | |
eyed | -0.108 |
didn | -0.106 |
CBS | -0.106 |
atra | -0.104 |
wasn | -0.104 |
Token | , |
Token ID | 11 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
Proced | +0.034 |
yip | +0.033 |
Address | +0.029 |
Agg | +0.029 |
aning | +0.029 |
Neg logit contributions | |
eyed | -0.039 |
CBS | -0.039 |
didn | -0.039 |
atra | -0.038 |
photos | -0.038 |
Token | or |
Token ID | 393 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
Proced | +0.145 |
yip | +0.137 |
aning | +0.129 |
Address | +0.128 |
Fn | +0.128 |
Neg logit contributions | |
eyed | -0.121 |
CBS | -0.121 |
didn | -0.120 |
atra | -0.118 |
hadn | -0.114 |
Token | overcome |
Token ID | 10980 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
eyed | +0.026 |
didn | +0.025 |
CBS | +0.025 |
atra | +0.024 |
photos | +0.024 |
Neg logit contributions | |
Proced | -0.028 |
yip | -0.026 |
Address | -0.025 |
aning | -0.025 |
Fn | -0.025 |
Token | nature |
Token ID | 3450 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
eyed | +0.069 |
didn | +0.069 |
CBS | +0.068 |
atra | +0.067 |
photos | +0.066 |
Neg logit contributions | |
Proced | -0.078 |
yip | -0.074 |
aning | -0.069 |
Fn | -0.069 |
Agg | -0.069 |
Token | by |
Token ID | 416 |
Feature activation | +6.18e-02 |
Pos logit contributions | |
Proced | +0.118 |
yip | +0.111 |
aning | +0.104 |
Address | +0.103 |
Fn | +0.103 |
Neg logit contributions | |
eyed | -0.111 |
CBS | -0.110 |
didn | -0.108 |
atra | -0.107 |
photos | -0.105 |
Token | training |
Token ID | 3047 |
Feature activation | +8.74e-02 |
Pos logit contributions | |
Proced | +0.043 |
yip | +0.039 |
aning | +0.037 |
Agg | +0.037 |
Address | +0.037 |
Neg logit contributions | |
eyed | -0.041 |
didn | -0.041 |
CBS | -0.040 |
atra | -0.040 |
photos | -0.039 |
Token | them |
Token ID | 606 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
Proced | +0.061 |
yip | +0.057 |
aning | +0.054 |
Fn | +0.054 |
Agg | +0.053 |
Neg logit contributions | |
eyed | -0.056 |
CBS | -0.056 |
didn | -0.056 |
atra | -0.056 |
turn | -0.053 |
Token | as |
Token ID | 355 |
Feature activation | +6.22e-02 |
Pos logit contributions | |
Proced | +0.210 |
yip | +0.194 |
aning | +0.181 |
Agg | +0.179 |
Address | +0.179 |
Neg logit contributions | |
eyed | -0.277 |
CBS | -0.276 |
didn | -0.273 |
atra | -0.271 |
photos | -0.265 |
Token | boys |
Token ID | 6510 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
Proced | +0.046 |
yip | +0.043 |
aning | +0.041 |
Agg | +0.041 |
Fn | +0.041 |
Neg logit contributions | |
eyed | -0.038 |
didn | -0.037 |
CBS | -0.037 |
atra | -0.037 |
hadn | -0.036 |
Token | has |
Token ID | 468 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
eyed | +0.102 |
CBS | +0.101 |
didn | +0.100 |
atra | +0.100 |
photos | +0.096 |
Neg logit contributions | |
Proced | -0.128 |
yip | -0.123 |
aning | -0.114 |
Address | -0.114 |
Fn | -0.114 |
Token | almost |
Token ID | 2048 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
Proced | +0.086 |
yip | +0.082 |
Address | +0.078 |
aning | +0.078 |
Fn | +0.077 |
Neg logit contributions | |
eyed | -0.066 |
didn | -0.065 |
CBS | -0.065 |
atra | -0.064 |
photos | -0.062 |
Token | extinguished |
Token ID | 42774 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
Proced | +0.145 |
yip | +0.139 |
aning | +0.130 |
Address | +0.130 |
Agg | +0.129 |
Neg logit contributions | |
eyed | -0.115 |
didn | -0.114 |
CBS | -0.113 |
atra | -0.111 |
photos | -0.110 |
Token | to |
Token ID | 284 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
Proced | +0.063 |
yip | +0.061 |
Address | +0.054 |
aning | +0.054 |
Fn | +0.054 |
Neg logit contributions | |
eyed | -0.076 |
didn | -0.074 |
CBS | -0.074 |
atra | -0.073 |
wasn | -0.072 |
Token | relatives |
Token ID | 12772 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
Proced | +0.080 |
yip | +0.076 |
Address | +0.071 |
Fn | +0.070 |
aning | +0.070 |
Neg logit contributions | |
eyed | -0.070 |
didn | -0.068 |
CBS | -0.068 |
atra | -0.067 |
wasn | -0.066 |
Token | overseas |
Token ID | 11292 |
Feature activation | -8.61e-04 |
Pos logit contributions | |
Proced | +0.088 |
yip | +0.083 |
Address | +0.077 |
Fn | +0.077 |
Agg | +0.076 |
Neg logit contributions | |
eyed | -0.086 |
didn | -0.085 |
CBS | -0.085 |
atra | -0.083 |
wasn | -0.082 |
Token | using |
Token ID | 1262 |
Feature activation | +8.53e-02 |
Pos logit contributions | |
eyed | +0.001 |
didn | +0.001 |
CBS | +0.001 |
atra | +0.001 |
wasn | +0.001 |
Neg logit contributions | |
Proced | -0.001 |
yip | -0.001 |
Address | -0.000 |
Fn | -0.000 |
aning | -0.000 |
Token | @ |
Token ID | 2488 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
Proced | +0.062 |
yip | +0.058 |
Fn | +0.054 |
Address | +0.054 |
aning | +0.053 |
Neg logit contributions | |
eyed | -0.054 |
didn | -0.053 |
CBS | -0.053 |
atra | -0.052 |
wasn | -0.051 |
Token | Pay |
Token ID | 19197 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
eyed | +0.088 |
didn | +0.086 |
wasn | +0.085 |
hadn | +0.085 |
CBS | +0.084 |
Neg logit contributions | |
Proced | -0.101 |
yip | -0.099 |
Address | -0.091 |
Agg | -0.091 |
AMS | -0.089 |
Token | pal |
Token ID | 18596 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
Proced | +0.036 |
yip | +0.036 |
Address | +0.033 |
Agg | +0.033 |
AMS | +0.032 |
Neg logit contributions | |
eyed | -0.030 |
didn | -0.030 |
hadn | -0.029 |
wasn | -0.029 |
CBS | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
eyed | +0.152 |
didn | +0.149 |
CBS | +0.149 |
atra | +0.146 |
wasn | +0.145 |
Neg logit contributions | |
Proced | -0.141 |
yip | -0.136 |
Address | -0.126 |
Agg | -0.124 |
Fn | -0.123 |
Token | Last |
Token ID | 4586 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
eyed | +0.038 |
CBS | +0.036 |
didn | +0.036 |
atra | +0.036 |
wasn | +0.035 |
Neg logit contributions | |
Proced | -0.042 |
yip | -0.039 |
Address | -0.037 |
Fn | -0.036 |
Agg | -0.036 |
Token | fall |
Token ID | 2121 |
Feature activation | +5.76e-02 |
Pos logit contributions | |
Proced | +0.069 |
yip | +0.065 |
Address | +0.059 |
Fn | +0.059 |
Agg | +0.059 |
Neg logit contributions | |
eyed | -0.077 |
CBS | -0.077 |
didn | -0.077 |
atra | -0.076 |
photos | -0.074 |
Token | , |
Token ID | 11 |
Feature activation | +5.96e-03 |
Pos logit contributions | |
Proced | +0.041 |
yip | +0.039 |
Address | +0.036 |
Fn | +0.036 |
aning | +0.036 |
Neg logit contributions | |
eyed | -0.037 |
CBS | -0.036 |
didn | -0.036 |
atra | -0.035 |
photos | -0.035 |
Token | Gaza |
Token ID | 48790 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
Proced | +0.012 |
yip | +0.012 |
Address | +0.011 |
Agg | +0.011 |
aning | +0.011 |
Neg logit contributions | |
eyed | -0.013 |
didn | -0.013 |
atra | -0.013 |
wasn | -0.012 |
CBS | -0.012 |
Token | Under |
Token ID | 4698 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
Proced | +0.155 |
yip | +0.148 |
Address | +0.137 |
Fn | +0.135 |
Agg | +0.135 |
Neg logit contributions | |
eyed | -0.151 |
didn | -0.150 |
CBS | -0.148 |
atra | -0.147 |
wasn | -0.145 |
Token | Attack |
Token ID | 8307 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
didn | +0.035 |
eyed | +0.035 |
CBS | +0.035 |
atra | +0.035 |
hadn | +0.034 |
Neg logit contributions | |
Proced | -0.025 |
yip | -0.024 |
Fn | -0.022 |
Address | -0.022 |
aning | -0.021 |
Token | Today |
Token ID | 6288 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
Proced | +0.186 |
yip | +0.176 |
Address | +0.163 |
Fn | +0.162 |
Agg | +0.160 |
Neg logit contributions | |
eyed | -0.201 |
didn | -0.200 |
CBS | -0.199 |
atra | -0.197 |
wasn | -0.194 |
Token | . |
Token ID | 13 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
eyed | +0.128 |
didn | +0.127 |
CBS | +0.126 |
atra | +0.125 |
wasn | +0.123 |
Neg logit contributions | |
Proced | -0.120 |
yip | -0.114 |
Address | -0.105 |
Fn | -0.104 |
aning | -0.103 |
Token | � |
Token ID | 447 |
Feature activation | +7.26e-02 |
Pos logit contributions | |
Proced | +0.028 |
yip | +0.025 |
Fn | +0.024 |
Address | +0.024 |
aning | +0.024 |
Neg logit contributions | |
eyed | -0.028 |
didn | -0.028 |
CBS | -0.028 |
atra | -0.028 |
hadn | -0.027 |
Token | � |
Token ID | 251 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
Proced | +0.055 |
yip | +0.053 |
Fn | +0.049 |
Address | +0.049 |
aning | +0.048 |
Neg logit contributions | |
eyed | -0.043 |
didn | -0.042 |
CBS | -0.042 |
wasn | -0.041 |
atra | -0.041 |
Token | In |
Token ID | 554 |
Feature activation | -8.43e-02 |
Pos logit contributions | |
Proced | +0.037 |
yip | +0.035 |
Fn | +0.033 |
Address | +0.032 |
aning | +0.032 |
Neg logit contributions | |
eyed | -0.038 |
didn | -0.038 |
CBS | -0.037 |
atra | -0.037 |
hadn | -0.037 |
Token | reality |
Token ID | 3950 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
eyed | +0.046 |
didn | +0.046 |
atra | +0.046 |
CBS | +0.046 |
wasn | +0.044 |
Neg logit contributions | |
Proced | -0.066 |
yip | -0.064 |
Fn | -0.059 |
aning | -0.059 |
Address | -0.058 |
Token | the |
Token ID | 262 |
Feature activation | -9.83e-02 |
Pos logit contributions | |
eyed | +0.189 |
didn | +0.187 |
CBS | +0.186 |
atra | +0.182 |
wasn | +0.181 |
Neg logit contributions | |
Proced | -0.203 |
yip | -0.195 |
Fn | -0.178 |
Address | -0.178 |
aning | -0.177 |
Token | photo |
Token ID | 4590 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
eyed | +0.044 |
atra | +0.044 |
didn | +0.044 |
CBS | +0.043 |
turn | +0.041 |
Neg logit contributions | |
Proced | -0.087 |
yip | -0.084 |
Fn | -0.080 |
aning | -0.079 |
Address | -0.079 |
Token | this |
Token ID | 428 |
Feature activation | +3.95e-03 |
Pos logit contributions | |
Proced | +0.059 |
yip | +0.055 |
Fn | +0.049 |
Address | +0.048 |
aning | +0.048 |
Neg logit contributions | |
eyed | -0.094 |
didn | -0.093 |
CBS | -0.093 |
atra | -0.091 |
wasn | -0.091 |
Token | year |
Token ID | 614 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
Proced | +0.002 |
yip | +0.002 |
Fn | +0.002 |
Address | +0.002 |
Agg | +0.002 |
Neg logit contributions | |
eyed | -0.003 |
didn | -0.003 |
CBS | -0.003 |
atra | -0.003 |
wasn | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
Proced | +0.038 |
yip | +0.036 |
Address | +0.033 |
Fn | +0.033 |
aning | +0.033 |
Neg logit contributions | |
eyed | -0.037 |
didn | -0.036 |
CBS | -0.036 |
atra | -0.035 |
wasn | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -6.57e-02 |
Pos logit contributions | |
Proced | +0.022 |
yip | +0.021 |
Fn | +0.020 |
aning | +0.020 |
Address | +0.020 |
Neg logit contributions | |
eyed | -0.013 |
didn | -0.013 |
CBS | -0.013 |
atra | -0.013 |
wasn | -0.013 |
Token | hugely |
Token ID | 22641 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
eyed | +0.041 |
didn | +0.041 |
CBS | +0.040 |
atra | +0.040 |
wasn | +0.040 |
Neg logit contributions | |
Proced | -0.047 |
yip | -0.044 |
Fn | -0.042 |
aning | -0.041 |
Address | -0.041 |
Token | popular |
Token ID | 2968 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
Proced | +0.158 |
yip | +0.150 |
Address | +0.135 |
Fn | +0.135 |
Agg | +0.135 |
Neg logit contributions | |
eyed | -0.212 |
didn | -0.212 |
CBS | -0.209 |
atra | -0.206 |
photos | -0.204 |
Token | Halloween |
Token ID | 16032 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
Proced | +0.125 |
yip | +0.115 |
Fn | +0.107 |
aning | +0.106 |
Finding | +0.106 |
Neg logit contributions | |
eyed | -0.125 |
didn | -0.125 |
CBS | -0.123 |
atra | -0.122 |
wasn | -0.121 |
Token | Horror |
Token ID | 22732 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
Proced | +0.084 |
yip | +0.078 |
Fn | +0.072 |
Address | +0.072 |
Agg | +0.072 |
Neg logit contributions | |
eyed | -0.089 |
didn | -0.088 |
CBS | -0.087 |
atra | -0.086 |
photos | -0.085 |
Token | Nights |
Token ID | 27760 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
Proced | +0.236 |
yip | +0.219 |
Fn | +0.204 |
Address | +0.203 |
Agg | +0.202 |
Neg logit contributions | |
eyed | -0.238 |
didn | -0.236 |
CBS | -0.233 |
atra | -0.228 |
wasn | -0.226 |
Token | , |
Token ID | 11 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
eyed | +0.032 |
didn | +0.031 |
CBS | +0.031 |
atra | +0.030 |
photos | +0.030 |
Neg logit contributions | |
Proced | -0.029 |
yip | -0.027 |
Fn | -0.025 |
Address | -0.025 |
aning | -0.025 |
Token | I |
Token ID | 314 |
Feature activation | -6.57e-02 |
Pos logit contributions | |
Proced | +0.010 |
yip | +0.010 |
Fn | +0.009 |
Address | +0.009 |
Agg | +0.009 |
Neg logit contributions | |
eyed | -0.008 |
didn | -0.008 |
CBS | -0.008 |
atra | -0.008 |
wasn | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +5.58e-02 |
Pos logit contributions | |
eyed | +0.086 |
didn | +0.085 |
CBS | +0.085 |
wasn | +0.083 |
atra | +0.083 |
Neg logit contributions | |
Proced | -0.089 |
yip | -0.086 |
Address | -0.078 |
Fn | -0.078 |
aning | -0.078 |
Token | neoliberal |
Token ID | 31972 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
Proced | +0.041 |
yip | +0.039 |
Fn | +0.036 |
Address | +0.036 |
Agg | +0.036 |
Neg logit contributions | |
CBS | -0.034 |
eyed | -0.034 |
didn | -0.034 |
atra | -0.033 |
photos | -0.033 |
Token | policies |
Token ID | 4788 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
Proced | +0.016 |
yip | +0.015 |
Address | +0.014 |
Fn | +0.014 |
aning | +0.014 |
Neg logit contributions | |
CBS | -0.014 |
eyed | -0.013 |
didn | -0.013 |
atra | -0.013 |
photos | -0.013 |
Token | had |
Token ID | 550 |
Feature activation | -9.67e-02 |
Pos logit contributions | |
eyed | +0.044 |
CBS | +0.044 |
atra | +0.043 |
didn | +0.043 |
photos | +0.043 |
Neg logit contributions | |
Proced | -0.060 |
yip | -0.057 |
aning | -0.054 |
Address | -0.053 |
Agg | -0.052 |
Token | dec |
Token ID | 875 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
eyed | +0.057 |
CBS | +0.057 |
didn | +0.056 |
atra | +0.056 |
photos | +0.055 |
Neg logit contributions | |
Proced | -0.072 |
yip | -0.069 |
Address | -0.064 |
aning | -0.064 |
Fn | -0.064 |
Token | imated |
Token ID | 15655 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
Proced | +0.079 |
yip | +0.076 |
aning | +0.069 |
Address | +0.068 |
Agg | +0.067 |
Neg logit contributions | |
eyed | -0.106 |
CBS | -0.106 |
didn | -0.105 |
wasn | -0.103 |
hadn | -0.103 |
Token | the |
Token ID | 262 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
Proced | +0.145 |
yip | +0.140 |
Fn | +0.130 |
aning | +0.129 |
Agg | +0.129 |
Neg logit contributions | |
eyed | -0.115 |
didn | -0.114 |
CBS | -0.113 |
atra | -0.111 |
photos | -0.109 |
Token | country |
Token ID | 1499 |
Feature activation | -2.10e-03 |
Pos logit contributions | |
eyed | +0.052 |
didn | +0.052 |
CBS | +0.052 |
atra | +0.051 |
wasn | +0.050 |
Neg logit contributions | |
Proced | -0.047 |
yip | -0.045 |
Fn | -0.041 |
Address | -0.040 |
aning | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -7.10e-03 |
Pos logit contributions | |
eyed | +0.001 |
didn | +0.001 |
CBS | +0.001 |
atra | +0.001 |
photos | +0.001 |
Neg logit contributions | |
Proced | -0.002 |
yip | -0.002 |
Address | -0.001 |
aning | -0.001 |
Fn | -0.001 |
Token | The |
Token ID | 383 |
Feature activation | -9.89e-02 |
Pos logit contributions | |
eyed | +0.005 |
didn | +0.005 |
CBS | +0.005 |
atra | +0.005 |
wasn | +0.004 |
Neg logit contributions | |
Proced | -0.005 |
yip | -0.005 |
Fn | -0.004 |
Address | -0.004 |
Agg | -0.004 |
Token | subsequent |
Token ID | 8840 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
eyed | +0.068 |
CBS | +0.067 |
didn | +0.067 |
atra | +0.066 |
wasn | +0.065 |
Neg logit contributions | |
Proced | -0.065 |
yip | -0.062 |
Address | -0.057 |
Fn | -0.057 |
aning | -0.057 |
Token | change |
Token ID | 1487 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
eyed | +0.091 |
didn | +0.090 |
CBS | +0.089 |
atra | +0.087 |
photos | +0.086 |
Neg logit contributions | |
Proced | -0.097 |
yip | -0.092 |
Fn | -0.085 |
aning | -0.085 |
Address | -0.084 |
Token | of |
Token ID | 286 |
Feature activation | +8.09e-02 |
Pos logit contributions | |
eyed | +0.067 |
CBS | +0.066 |
didn | +0.065 |
photos | +0.063 |
atra | +0.063 |
Neg logit contributions | |
Proced | -0.083 |
yip | -0.081 |
Address | -0.074 |
Fn | -0.074 |
aning | -0.074 |
Token | velocity |
Token ID | 15432 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
Proced | +0.056 |
yip | +0.053 |
Fn | +0.050 |
aning | +0.049 |
Address | +0.049 |
Neg logit contributions | |
eyed | -0.053 |
didn | -0.052 |
CBS | -0.052 |
atra | -0.050 |
photos | -0.050 |
Token | placed |
Token ID | 4624 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
Proced | +0.112 |
yip | +0.105 |
Fn | +0.100 |
Address | +0.099 |
aning | +0.099 |
Neg logit contributions | |
eyed | -0.079 |
CBS | -0.078 |
didn | -0.076 |
atra | -0.075 |
photos | -0.075 |
Token | Ranger |
Token ID | 21888 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
Proced | +0.023 |
yip | +0.021 |
Fn | +0.020 |
Address | +0.020 |
Agg | +0.019 |
Neg logit contributions | |
eyed | -0.023 |
didn | -0.023 |
CBS | -0.023 |
wasn | -0.022 |
atra | -0.022 |
Token | 6 |
Token ID | 718 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
Proced | +0.025 |
yip | +0.023 |
Fn | +0.021 |
Address | +0.021 |
aning | +0.021 |
Neg logit contributions | |
eyed | -0.028 |
didn | -0.027 |
CBS | -0.027 |
photos | -0.026 |
atra | -0.026 |
Token | on |
Token ID | 319 |
Feature activation | +9.27e-02 |
Pos logit contributions | |
Proced | +0.029 |
yip | +0.027 |
Fn | +0.025 |
Address | +0.025 |
aning | +0.025 |
Neg logit contributions | |
eyed | -0.028 |
didn | -0.028 |
CBS | -0.028 |
photos | -0.027 |
atra | -0.027 |
Token | course |
Token ID | 1781 |
Feature activation | +6.18e-02 |
Pos logit contributions | |
Proced | +0.070 |
yip | +0.066 |
Fn | +0.062 |
aning | +0.062 |
Address | +0.062 |
Neg logit contributions | |
eyed | -0.055 |
didn | -0.055 |
CBS | -0.054 |
photos | -0.052 |
wasn | -0.052 |
Token | for |
Token ID | 329 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
Proced | +0.034 |
yip | +0.032 |
Fn | +0.029 |
Address | +0.029 |
aning | +0.029 |
Neg logit contributions | |
eyed | -0.049 |
didn | -0.048 |
CBS | -0.048 |
atra | -0.047 |
photos | -0.047 |
Token | an |
Token ID | 281 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
Proced | +0.119 |
yip | +0.113 |
aning | +0.106 |
Fn | +0.105 |
Address | +0.104 |
Neg logit contributions | |
eyed | -0.106 |
didn | -0.104 |
CBS | -0.103 |
wasn | -0.100 |
photos | -0.100 |
Token | impact |
Token ID | 2928 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
Proced | +0.144 |
yip | +0.135 |
Fn | +0.126 |
Address | +0.125 |
aning | +0.124 |
Neg logit contributions | |
eyed | -0.155 |
CBS | -0.154 |
didn | -0.154 |
atra | -0.149 |
photos | -0.149 |
Token | One |
Token ID | 1881 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
eyed | +0.008 |
CBS | +0.008 |
didn | +0.008 |
atra | +0.007 |
wasn | +0.007 |
Neg logit contributions | |
Proced | -0.010 |
yip | -0.010 |
Address | -0.009 |
Fn | -0.009 |
aning | -0.009 |
Token | … |
Token ID | 1399 |
Feature activation | -1.08e-03 |
Pos logit contributions | |
eyed | +0.118 |
CBS | +0.117 |
didn | +0.116 |
atra | +0.115 |
wasn | +0.113 |
Neg logit contributions | |
Proced | -0.112 |
yip | -0.106 |
Address | -0.097 |
Fn | -0.096 |
aning | -0.096 |
Token | based |
Token ID | 1912 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
eyed | +0.001 |
CBS | +0.001 |
atra | +0.001 |
didn | +0.001 |
wasn | +0.001 |
Neg logit contributions | |
Proced | -0.001 |
yip | -0.001 |
aning | -0.001 |
Agg | -0.001 |
Address | -0.001 |
Token | on |
Token ID | 319 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
Proced | +0.018 |
yip | +0.018 |
Address | +0.016 |
Fn | +0.016 |
aning | +0.016 |
Neg logit contributions | |
eyed | -0.018 |
didn | -0.018 |
CBS | -0.018 |
atra | -0.017 |
wasn | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | +8.65e-03 |
Pos logit contributions | |
eyed | +0.038 |
CBS | +0.037 |
didn | +0.037 |
atra | +0.037 |
wasn | +0.036 |
Neg logit contributions | |
Proced | -0.050 |
yip | -0.047 |
Fn | -0.044 |
aning | -0.044 |
Address | -0.044 |
Token | but |
Token ID | 475 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
Proced | +0.006 |
yip | +0.006 |
Fn | +0.006 |
aning | +0.006 |
Agg | +0.006 |
Neg logit contributions | |
eyed | -0.005 |
CBS | -0.005 |
didn | -0.005 |
atra | -0.005 |
photos | -0.005 |
Token | having |
Token ID | 1719 |
Feature activation | +7.25e-02 |
Pos logit contributions | |
eyed | +0.077 |
CBS | +0.076 |
atra | +0.075 |
didn | +0.075 |
photos | +0.073 |
Neg logit contributions | |
Proced | -0.123 |
yip | -0.118 |
Fn | -0.110 |
aning | -0.110 |
Address | -0.110 |
Token | absolutely |
Token ID | 5543 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
Proced | +0.054 |
yip | +0.051 |
Fn | +0.048 |
Agg | +0.047 |
Address | +0.047 |
Neg logit contributions | |
eyed | -0.044 |
CBS | -0.043 |
didn | -0.043 |
atra | -0.042 |
photos | -0.041 |
Token | nothing |
Token ID | 2147 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
Proced | +0.224 |
yip | +0.214 |
Fn | +0.204 |
Agg | +0.201 |
aning | +0.200 |
Neg logit contributions | |
CBS | -0.118 |
eyed | -0.118 |
atra | -0.116 |
didn | -0.116 |
photos | -0.111 |
Token | to |
Token ID | 284 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
eyed | +0.028 |
didn | +0.028 |
CBS | +0.027 |
atra | +0.027 |
wasn | +0.027 |
Neg logit contributions | |
Proced | -0.021 |
yip | -0.019 |
Fn | -0.017 |
Address | -0.017 |
aning | -0.017 |
Token | do |
Token ID | 466 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
Proced | +0.079 |
yip | +0.076 |
Address | +0.071 |
aning | +0.071 |
Fn | +0.071 |
Neg logit contributions | |
eyed | -0.058 |
didn | -0.058 |
CBS | -0.057 |
atra | -0.056 |
wasn | -0.055 |
Token | ," |
Token ID | 553 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
Proced | +0.020 |
yip | +0.020 |
Address | +0.018 |
aning | +0.018 |
Fn | +0.018 |
Neg logit contributions | |
eyed | -0.017 |
didn | -0.017 |
CBS | -0.017 |
atra | -0.017 |
wasn | -0.017 |
Token | said |
Token ID | 531 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
eyed | +0.109 |
CBS | +0.108 |
didn | +0.107 |
atra | +0.105 |
photos | +0.104 |
Neg logit contributions | |
Proced | -0.122 |
yip | -0.119 |
Fn | -0.108 |
Address | -0.108 |
aning | -0.107 |
Token | Lord |
Token ID | 4453 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
Proced | +0.018 |
yip | +0.018 |
Fn | +0.017 |
Address | +0.017 |
aning | +0.017 |
Neg logit contributions | |
eyed | -0.011 |
CBS | -0.011 |
didn | -0.011 |
atra | -0.011 |
wasn | -0.011 |
Token | West |
Token ID | 2688 |
Feature activation | -8.31e-02 |
Pos logit contributions | |
eyed | +0.028 |
didn | +0.027 |
CBS | +0.027 |
wasn | +0.027 |
atra | +0.026 |
Neg logit contributions | |
Proced | -0.046 |
yip | -0.045 |
Fn | -0.041 |
Address | -0.041 |
aning | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
eyed | +0.058 |
didn | +0.057 |
CBS | +0.057 |
wasn | +0.056 |
atra | +0.056 |
Neg logit contributions | |
Proced | -0.054 |
yip | -0.052 |
Address | -0.047 |
Fn | -0.047 |
Agg | -0.047 |
Token | " |
Token ID | 366 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
Proced | +0.023 |
yip | +0.022 |
Address | +0.021 |
Fn | +0.021 |
aning | +0.020 |
Neg logit contributions | |
eyed | -0.016 |
didn | -0.016 |
CBS | -0.016 |
wasn | -0.015 |
atra | -0.015 |
Token | Regular |
Token ID | 40164 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
eyed | +0.102 |
didn | +0.100 |
hadn | +0.099 |
wasn | +0.099 |
atra | +0.099 |
Neg logit contributions | |
Proced | -0.096 |
yip | -0.089 |
Address | -0.088 |
Agg | -0.087 |
Finding | -0.085 |
Token | software |
Token ID | 3788 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
Proced | +0.098 |
yip | +0.095 |
Fn | +0.086 |
Address | +0.086 |
aning | +0.086 |
Neg logit contributions | |
didn | -0.097 |
CBS | -0.096 |
eyed | -0.096 |
photos | -0.094 |
atra | -0.094 |
Token | patch |
Token ID | 8529 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
eyed | +0.076 |
didn | +0.076 |
CBS | +0.076 |
atra | +0.074 |
photos | +0.073 |
Neg logit contributions | |
Proced | -0.079 |
yip | -0.077 |
aning | -0.070 |
Address | -0.070 |
Agg | -0.069 |
Token | ing |
Token ID | 278 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
Proced | +0.026 |
yip | +0.023 |
aning | +0.019 |
Address | +0.019 |
Fn | +0.019 |
Neg logit contributions | |
eyed | -0.087 |
didn | -0.086 |
CBS | -0.086 |
atra | -0.085 |
photos | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | -7.17e-03 |
Pos logit contributions | |
Proced | +0.296 |
yip | +0.280 |
aning | +0.259 |
Address | +0.259 |
Fn | +0.258 |
Neg logit contributions | |
eyed | -0.273 |
didn | -0.267 |
CBS | -0.267 |
atra | -0.260 |
photos | -0.256 |
Token | the |
Token ID | 262 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
Proced | +0.019 |
yip | +0.018 |
Fn | +0.017 |
aning | +0.017 |
Address | +0.017 |
Neg logit contributions | |
eyed | -0.013 |
didn | -0.013 |
CBS | -0.012 |
atra | -0.012 |
photos | -0.012 |
Token | diss |
Token ID | 6249 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
eyed | +0.009 |
didn | +0.009 |
CBS | +0.009 |
atra | +0.009 |
photos | +0.009 |
Neg logit contributions | |
Proced | -0.010 |
yip | -0.010 |
Fn | -0.009 |
aning | -0.009 |
Address | -0.009 |
Token | ident |
Token ID | 738 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
Proced | +0.151 |
yip | +0.142 |
Address | +0.121 |
aning | +0.120 |
Agg | +0.119 |
Neg logit contributions | |
eyed | -0.320 |
didn | -0.317 |
CBS | -0.316 |
wasn | -0.312 |
photos | -0.311 |
Token | Real |
Token ID | 6416 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
Proced | +0.207 |
yip | +0.197 |
Fn | +0.184 |
aning | +0.181 |
Address | +0.181 |
Neg logit contributions | |
eyed | -0.178 |
didn | -0.177 |
CBS | -0.176 |
atra | -0.173 |
photos | -0.171 |
Token | News |
Token ID | 3000 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
Proced | +0.019 |
yip | +0.018 |
Address | +0.017 |
Fn | +0.017 |
Agg | +0.017 |
Neg logit contributions | |
eyed | -0.014 |
CBS | -0.014 |
didn | -0.014 |
atra | -0.014 |
wasn | -0.013 |
Token | Network |
Token ID | 7311 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
Proced | +0.013 |
yip | +0.012 |
Address | +0.011 |
Fn | +0.011 |
Agg | +0.011 |
Neg logit contributions | |
eyed | -0.012 |
didn | -0.012 |
CBS | -0.012 |
atra | -0.012 |
wasn | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
eyed | +0.040 |
didn | +0.040 |
CBS | +0.039 |
atra | +0.039 |
wasn | +0.039 |
Neg logit contributions | |
Proced | -0.037 |
yip | -0.035 |
Address | -0.032 |
Fn | -0.032 |
Agg | -0.032 |
Token | hosted |
Token ID | 12007 |
Feature activation | -8.08e-02 |
Pos logit contributions | |
Proced | +0.074 |
yip | +0.071 |
Fn | +0.065 |
Address | +0.065 |
aning | +0.065 |
Neg logit contributions | |
eyed | -0.072 |
didn | -0.071 |
CBS | -0.071 |
atra | -0.071 |
wasn | -0.069 |
Token | by |
Token ID | 416 |
Feature activation | +7.82e-02 |
Pos logit contributions | |
eyed | +0.056 |
didn | +0.055 |
CBS | +0.055 |
atra | +0.054 |
wasn | +0.054 |
Neg logit contributions | |
Proced | -0.053 |
yip | -0.051 |
Address | -0.046 |
Fn | -0.046 |
aning | -0.046 |
Token | Paul |
Token ID | 3362 |
Feature activation | -5.72e-02 |
Pos logit contributions | |
Proced | +0.059 |
yip | +0.057 |
Fn | +0.052 |
Address | +0.052 |
aning | +0.051 |
Neg logit contributions | |
eyed | -0.047 |
didn | -0.046 |
CBS | -0.046 |
atra | -0.045 |
wasn | -0.044 |
Token | Jay |
Token ID | 9180 |
Feature activation | +9.58e-02 |
Pos logit contributions | |
eyed | +0.034 |
CBS | +0.034 |
didn | +0.034 |
atra | +0.033 |
wasn | +0.033 |
Neg logit contributions | |
Proced | -0.043 |
yip | -0.041 |
Fn | -0.038 |
Address | -0.038 |
aning | -0.037 |
Token | related |
Token ID | 3519 |
Feature activation | -8.31e-02 |
Pos logit contributions | |
eyed | +0.120 |
didn | +0.118 |
CBS | +0.117 |
wasn | +0.115 |
atra | +0.114 |
Neg logit contributions | |
Proced | -0.217 |
yip | -0.211 |
Address | -0.196 |
Fn | -0.195 |
Agg | -0.195 |
Token | areas |
Token ID | 3006 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
eyed | +0.051 |
didn | +0.051 |
CBS | +0.051 |
atra | +0.050 |
wasn | +0.049 |
Neg logit contributions | |
Proced | -0.060 |
yip | -0.057 |
Fn | -0.053 |
aning | -0.053 |
Address | -0.053 |
Token | this |
Token ID | 428 |
Feature activation | -6.91e-03 |
Pos logit contributions | |
Proced | +0.042 |
yip | +0.040 |
Fn | +0.037 |
aning | +0.037 |
Address | +0.037 |
Neg logit contributions | |
eyed | -0.036 |
didn | -0.036 |
CBS | -0.036 |
atra | -0.035 |
wasn | -0.034 |
Token | screen |
Token ID | 3159 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
eyed | +0.004 |
CBS | +0.004 |
didn | +0.004 |
atra | +0.004 |
photos | +0.004 |
Neg logit contributions | |
Proced | -0.005 |
yip | -0.005 |
Fn | -0.005 |
aning | -0.004 |
Agg | -0.004 |
Token | also |
Token ID | 635 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
eyed | +0.160 |
CBS | +0.159 |
didn | +0.155 |
atra | +0.155 |
photos | +0.152 |
Neg logit contributions | |
Proced | -0.205 |
yip | -0.196 |
Fn | -0.184 |
aning | -0.182 |
Address | -0.182 |
Token | supports |
Token ID | 6971 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
Proced | +0.024 |
yip | +0.023 |
Fn | +0.022 |
aning | +0.022 |
Address | +0.022 |
Neg logit contributions | |
eyed | -0.019 |
CBS | -0.019 |
atra | -0.018 |
didn | -0.018 |
photos | -0.018 |
Token | NVIDIA |
Token ID | 15127 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
eyed | +0.117 |
didn | +0.116 |
CBS | +0.116 |
atra | +0.114 |
wasn | +0.111 |
Neg logit contributions | |
Proced | -0.128 |
yip | -0.121 |
Fn | -0.113 |
Address | -0.110 |
aning | -0.110 |
Token | 's |
Token ID | 338 |
Feature activation | -6.82e-04 |
Pos logit contributions | |
Proced | +0.117 |
yip | +0.110 |
Fn | +0.103 |
aning | +0.101 |
Address | +0.101 |
Neg logit contributions | |
CBS | -0.108 |
eyed | -0.107 |
didn | -0.107 |
atra | -0.106 |
photos | -0.102 |
Token | G |
Token ID | 402 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
eyed | +0.000 |
CBS | +0.000 |
didn | +0.000 |
atra | +0.000 |
wasn | +0.000 |
Neg logit contributions | |
Proced | -0.000 |
yip | -0.000 |
Fn | -0.000 |
aning | -0.000 |
Agg | -0.000 |
Token | - |
Token ID | 12 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
Proced | +0.080 |
yip | +0.077 |
Address | +0.072 |
aning | +0.072 |
Agg | +0.072 |
Neg logit contributions | |
eyed | -0.059 |
didn | -0.059 |
CBS | -0.058 |
atra | -0.057 |
wasn | -0.056 |
Token | sync |
Token ID | 27261 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
eyed | +0.022 |
didn | +0.022 |
CBS | +0.022 |
wasn | +0.021 |
atra | +0.021 |
Neg logit contributions | |
Proced | -0.031 |
yip | -0.030 |
Address | -0.029 |
Agg | -0.029 |
Fn | -0.028 |
Token | Romney |
Token ID | 8431 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
eyed | +0.017 |
didn | +0.016 |
CBS | +0.016 |
atra | +0.016 |
photos | +0.016 |
Neg logit contributions | |
Proced | -0.021 |
yip | -0.020 |
aning | -0.019 |
Address | -0.018 |
Fn | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
Proced | +0.112 |
yip | +0.107 |
Address | +0.098 |
aning | +0.098 |
Agg | +0.097 |
Neg logit contributions | |
eyed | -0.118 |
didn | -0.115 |
CBS | -0.115 |
photos | -0.113 |
atra | -0.113 |
Token | the |
Token ID | 262 |
Feature activation | -1.48e-03 |
Pos logit contributions | |
Proced | +0.105 |
yip | +0.099 |
Fn | +0.093 |
aning | +0.093 |
Agg | +0.092 |
Neg logit contributions | |
eyed | -0.095 |
didn | -0.091 |
CBS | -0.091 |
atra | -0.091 |
photos | -0.089 |
Token | presumptive |
Token ID | 39715 |
Feature activation | -8.43e-02 |
Pos logit contributions | |
eyed | +0.001 |
didn | +0.001 |
atra | +0.001 |
CBS | +0.001 |
photos | +0.001 |
Neg logit contributions | |
Proced | -0.001 |
yip | -0.001 |
Address | -0.001 |
Fn | -0.001 |
aning | -0.001 |
Token | Republican |
Token ID | 3415 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
eyed | +0.051 |
didn | +0.050 |
CBS | +0.050 |
atra | +0.049 |
photos | +0.049 |
Neg logit contributions | |
Proced | -0.063 |
yip | -0.059 |
aning | -0.056 |
Address | -0.055 |
Fn | -0.055 |
Token | presidential |
Token ID | 4787 |
Feature activation | +9.10e-04 |
Pos logit contributions | |
Proced | +0.013 |
yip | +0.012 |
Address | +0.011 |
Agg | +0.011 |
aning | +0.011 |
Neg logit contributions | |
eyed | -0.009 |
didn | -0.009 |
CBS | -0.009 |
atra | -0.008 |
photos | -0.008 |
Token | nominee |
Token ID | 10429 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
Proced | +0.001 |
yip | +0.001 |
Address | +0.001 |
aning | +0.001 |
Fn | +0.001 |
Neg logit contributions | |
eyed | -0.000 |
didn | -0.000 |
CBS | -0.000 |
atra | -0.000 |
photos | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
eyed | +0.109 |
didn | +0.107 |
CBS | +0.107 |
atra | +0.106 |
photos | +0.104 |
Neg logit contributions | |
Proced | -0.095 |
yip | -0.090 |
Address | -0.083 |
aning | -0.082 |
Fn | -0.082 |
Token | In |
Token ID | 554 |
Feature activation | -9.03e-02 |
Pos logit contributions | |
Proced | +0.019 |
yip | +0.017 |
Address | +0.016 |
Fn | +0.016 |
Agg | +0.016 |
Neg logit contributions | |
eyed | -0.022 |
didn | -0.021 |
atra | -0.021 |
CBS | -0.021 |
wasn | -0.020 |
Token | both |
Token ID | 1111 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
eyed | +0.055 |
didn | +0.054 |
CBS | +0.054 |
atra | +0.053 |
wasn | +0.052 |
Neg logit contributions | |
Proced | -0.067 |
yip | -0.063 |
Address | -0.059 |
aning | -0.059 |
Agg | -0.058 |
Token | Florida |
Token ID | 4744 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
Proced | +0.038 |
yip | +0.036 |
Address | +0.033 |
aning | +0.033 |
Fn | +0.033 |
Neg logit contributions | |
eyed | -0.041 |
didn | -0.041 |
CBS | -0.040 |
atra | -0.040 |
wasn | -0.039 |
Token | we |
Token ID | 356 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
Proced | +0.022 |
yip | +0.021 |
Fn | +0.020 |
Address | +0.019 |
aning | +0.019 |
Neg logit contributions | |
eyed | -0.019 |
didn | -0.019 |
CBS | -0.019 |
atra | -0.019 |
hadn | -0.018 |
Token | dated |
Token ID | 14567 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
Proced | +0.050 |
yip | +0.047 |
aning | +0.045 |
Address | +0.044 |
Fn | +0.044 |
Neg logit contributions | |
eyed | -0.036 |
CBS | -0.036 |
didn | -0.035 |
atra | -0.035 |
photos | -0.034 |
Token | we |
Token ID | 356 |
Feature activation | +6.35e-02 |
Pos logit contributions | |
Proced | +0.088 |
yip | +0.085 |
Fn | +0.077 |
aning | +0.077 |
Address | +0.077 |
Neg logit contributions | |
eyed | -0.087 |
CBS | -0.086 |
didn | -0.086 |
atra | -0.085 |
wasn | -0.083 |
Token | shared |
Token ID | 4888 |
Feature activation | -7.06e-02 |
Pos logit contributions | |
Proced | +0.049 |
yip | +0.046 |
aning | +0.044 |
Address | +0.043 |
Fn | +0.043 |
Neg logit contributions | |
eyed | -0.036 |
CBS | -0.036 |
didn | -0.035 |
atra | -0.035 |
photos | -0.035 |
Token | our |
Token ID | 674 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
eyed | +0.044 |
didn | +0.043 |
CBS | +0.043 |
atra | +0.043 |
turn | +0.041 |
Neg logit contributions | |
Proced | -0.051 |
yip | -0.048 |
Fn | -0.045 |
Address | -0.045 |
aning | -0.045 |
Token | vision |
Token ID | 5761 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
Proced | +0.039 |
yip | +0.036 |
aning | +0.034 |
Address | +0.034 |
Fn | +0.034 |
Neg logit contributions | |
eyed | -0.030 |
didn | -0.030 |
CBS | -0.030 |
atra | -0.029 |
wasn | -0.028 |
Token | for |
Token ID | 329 |
Feature activation | +9.26e-02 |
Pos logit contributions | |
eyed | +0.295 |
didn | +0.293 |
CBS | +0.292 |
atra | +0.286 |
wasn | +0.286 |
Neg logit contributions | |
Proced | -0.288 |
yip | -0.276 |
Address | -0.251 |
Fn | -0.250 |
aning | -0.249 |
Token | life |
Token ID | 1204 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
Proced | +0.066 |
yip | +0.064 |
Address | +0.059 |
Fn | +0.058 |
aning | +0.058 |
Neg logit contributions | |
eyed | -0.058 |
didn | -0.058 |
CBS | -0.057 |
atra | -0.056 |
wasn | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
eyed | +0.147 |
didn | +0.146 |
CBS | +0.145 |
atra | +0.143 |
wasn | +0.141 |
Neg logit contributions | |
Proced | -0.117 |
yip | -0.111 |
Address | -0.101 |
Fn | -0.100 |
aning | -0.099 |
Token | our |
Token ID | 674 |
Feature activation | +9.79e-02 |
Pos logit contributions | |
Proced | +0.040 |
yip | +0.038 |
Address | +0.035 |
aning | +0.035 |
Fn | +0.035 |
Neg logit contributions | |
eyed | -0.035 |
CBS | -0.035 |
didn | -0.035 |
atra | -0.034 |
wasn | -0.033 |
Token | personal |
Token ID | 2614 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
Proced | +0.072 |
yip | +0.068 |
Address | +0.064 |
aning | +0.064 |
Fn | +0.063 |
Neg logit contributions | |
eyed | -0.060 |
didn | -0.059 |
CBS | -0.059 |
atra | -0.058 |
wasn | -0.057 |
Token | s |
Token ID | 82 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
eyed | +0.062 |
CBS | +0.061 |
didn | +0.061 |
atra | +0.061 |
wasn | +0.060 |
Neg logit contributions | |
Proced | -0.071 |
yip | -0.068 |
Address | -0.063 |
Agg | -0.062 |
aning | -0.062 |
Token | The |
Token ID | 383 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
eyed | +0.014 |
CBS | +0.014 |
didn | +0.014 |
atra | +0.014 |
wasn | +0.014 |
Neg logit contributions | |
Proced | -0.015 |
yip | -0.014 |
Address | -0.013 |
Fn | -0.013 |
aning | -0.013 |
Token | _ |
Token ID | 62 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
eyed | +0.069 |
CBS | +0.069 |
didn | +0.069 |
wasn | +0.067 |
atra | +0.067 |
Neg logit contributions | |
Proced | -0.070 |
yip | -0.066 |
Address | -0.061 |
Fn | -0.060 |
Agg | -0.060 |
Token | Donald |
Token ID | 7371 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
eyed | +0.009 |
didn | +0.009 |
CBS | +0.009 |
wasn | +0.009 |
atra | +0.009 |
Neg logit contributions | |
Proced | -0.010 |
yip | -0.009 |
Address | -0.009 |
Agg | -0.009 |
aning | -0.009 |
Token | forum |
Token ID | 10041 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
eyed | +0.016 |
CBS | +0.016 |
didn | +0.016 |
atra | +0.016 |
wasn | +0.016 |
Neg logit contributions | |
Proced | -0.013 |
yip | -0.013 |
Address | -0.012 |
Agg | -0.012 |
Fn | -0.011 |
Token | ― |
Token ID | 21096 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
Proced | +0.236 |
yip | +0.225 |
Address | +0.206 |
Fn | +0.204 |
Agg | +0.203 |
Neg logit contributions | |
eyed | -0.240 |
CBS | -0.237 |
didn | -0.237 |
atra | -0.236 |
wasn | -0.229 |
Token | where |
Token ID | 810 |
Feature activation | +4.92e-04 |
Pos logit contributions | |
eyed | +0.262 |
didn | +0.259 |
CBS | +0.259 |
atra | +0.255 |
wasn | +0.252 |
Neg logit contributions | |
Proced | -0.282 |
yip | -0.273 |
Address | -0.248 |
aning | -0.246 |
Fn | -0.246 |
Token | Donald |
Token ID | 3759 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
Proced | +0.000 |
yip | +0.000 |
Fn | +0.000 |
Address | +0.000 |
aning | +0.000 |
Neg logit contributions | |
eyed | -0.000 |
CBS | -0.000 |
didn | -0.000 |
atra | -0.000 |
wasn | -0.000 |
Token | Trump |
Token ID | 1301 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
eyed | +0.052 |
didn | +0.051 |
CBS | +0.050 |
atra | +0.048 |
photos | +0.048 |
Neg logit contributions | |
Proced | -0.127 |
yip | -0.124 |
Fn | -0.116 |
Address | -0.116 |
aning | -0.116 |
Token | this |
Token ID | 428 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
eyed | +0.124 |
didn | +0.121 |
CBS | +0.121 |
atra | +0.118 |
photos | +0.117 |
Neg logit contributions | |
Proced | -0.158 |
yip | -0.154 |
Address | -0.142 |
Fn | -0.142 |
aning | -0.141 |
Token | week |
Token ID | 1285 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
eyed | +0.033 |
CBS | +0.033 |
didn | +0.032 |
atra | +0.032 |
photos | +0.032 |
Neg logit contributions | |
Proced | -0.027 |
yip | -0.026 |
Address | -0.024 |
Fn | -0.023 |
aning | -0.023 |
Token | Day |
Token ID | 3596 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
eyed | +0.052 |
didn | +0.052 |
CBS | +0.051 |
atra | +0.051 |
wasn | +0.050 |
Neg logit contributions | |
Proced | -0.045 |
yip | -0.042 |
Address | -0.039 |
Fn | -0.039 |
aning | -0.039 |
Token | fun |
Token ID | 1257 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
Proced | +0.068 |
yip | +0.064 |
Fn | +0.059 |
Address | +0.059 |
aning | +0.058 |
Neg logit contributions | |
eyed | -0.073 |
didn | -0.072 |
CBS | -0.071 |
atra | -0.070 |
wasn | -0.069 |
Token | for |
Token ID | 329 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
Proced | +0.198 |
yip | +0.190 |
Address | +0.173 |
Fn | +0.173 |
aning | +0.173 |
Neg logit contributions | |
eyed | -0.182 |
didn | -0.180 |
CBS | -0.179 |
atra | -0.176 |
wasn | -0.174 |
Token | all |
Token ID | 477 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
Proced | +0.141 |
yip | +0.133 |
aning | +0.125 |
Fn | +0.125 |
Address | +0.124 |
Neg logit contributions | |
eyed | -0.114 |
didn | -0.111 |
atra | -0.109 |
CBS | -0.109 |
wasn | -0.106 |
Token | of |
Token ID | 286 |
Feature activation | +7.93e-02 |
Pos logit contributions | |
Proced | +0.119 |
yip | +0.112 |
aning | +0.104 |
Fn | +0.104 |
Address | +0.104 |
Neg logit contributions | |
eyed | -0.108 |
didn | -0.106 |
CBS | -0.105 |
atra | -0.103 |
hadn | -0.102 |
Token | Dub |
Token ID | 10322 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
Proced | +0.050 |
yip | +0.047 |
Fn | +0.043 |
Address | +0.043 |
Agg | +0.042 |
Neg logit contributions | |
eyed | -0.057 |
didn | -0.057 |
CBS | -0.056 |
atra | -0.056 |
hadn | -0.055 |
Token | Nation |
Token ID | 8741 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
eyed | +0.213 |
didn | +0.210 |
CBS | +0.209 |
atra | +0.206 |
wasn | +0.202 |
Neg logit contributions | |
Proced | -0.340 |
yip | -0.330 |
Address | -0.306 |
aning | -0.305 |
Fn | -0.305 |
Token | is |
Token ID | 318 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
eyed | +0.077 |
didn | +0.076 |
CBS | +0.075 |
atra | +0.074 |
hadn | +0.073 |
Neg logit contributions | |
Proced | -0.093 |
yip | -0.090 |
Fn | -0.082 |
Address | -0.082 |
aning | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
Proced | +0.047 |
yip | +0.045 |
Fn | +0.042 |
Address | +0.042 |
aning | +0.041 |
Neg logit contributions | |
eyed | -0.028 |
didn | -0.027 |
CBS | -0.027 |
atra | -0.027 |
hadn | -0.026 |
Token | players |
Token ID | 1938 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
eyed | +0.026 |
didn | +0.026 |
CBS | +0.026 |
atra | +0.025 |
wasn | +0.025 |
Neg logit contributions | |
Proced | -0.029 |
yip | -0.028 |
Fn | -0.026 |
Address | -0.025 |
aning | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +6.29e-02 |
Pos logit contributions | |
eyed | +0.077 |
didn | +0.076 |
CBS | +0.076 |
atra | +0.075 |
wasn | +0.074 |
Neg logit contributions | |
Proced | -0.080 |
yip | -0.076 |
Address | -0.069 |
Fn | -0.069 |
aning | -0.069 |
Token | , |
Token ID | 11 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
Proced | +0.044 |
yip | +0.042 |
Fn | +0.039 |
Address | +0.039 |
aning | +0.038 |
Neg logit contributions | |
eyed | -0.041 |
didn | -0.041 |
CBS | -0.040 |
atra | -0.040 |
wasn | -0.039 |
Token | May |
Token ID | 1737 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Proced | +0.023 |
yip | +0.022 |
Fn | +0.020 |
Agg | +0.020 |
Address | +0.020 |
Neg logit contributions | |
eyed | -0.020 |
didn | -0.020 |
CBS | -0.019 |
atra | -0.019 |
hadn | -0.019 |
Token | 11 |
Token ID | 1367 |
Feature activation | +8.83e-02 |
Pos logit contributions | |
Proced | +0.065 |
yip | +0.059 |
Fn | +0.055 |
Address | +0.055 |
aning | +0.055 |
Neg logit contributions | |
eyed | -0.083 |
didn | -0.082 |
CBS | -0.082 |
atra | -0.080 |
wasn | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
Proced | +0.067 |
yip | +0.065 |
Address | +0.060 |
Fn | +0.059 |
Agg | +0.059 |
Neg logit contributions | |
eyed | -0.052 |
didn | -0.051 |
CBS | -0.051 |
wasn | -0.050 |
atra | -0.049 |
Token | 2016 |
Token ID | 1584 |
Feature activation | +4.10e-02 |
Pos logit contributions | |
Proced | +0.042 |
yip | +0.040 |
Fn | +0.037 |
Agg | +0.037 |
Address | +0.037 |
Neg logit contributions | |
eyed | -0.037 |
didn | -0.036 |
CBS | -0.036 |
atra | -0.035 |
hadn | -0.035 |
Token | Bronx |
Token ID | 32486 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
Proced | +0.028 |
yip | +0.027 |
Fn | +0.025 |
Address | +0.025 |
Agg | +0.025 |
Neg logit contributions | |
eyed | -0.027 |
didn | -0.027 |
CBS | -0.026 |
atra | -0.026 |
wasn | -0.026 |
Token | Defenders |
Token ID | 26983 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
eyed | +0.222 |
didn | +0.221 |
CBS | +0.216 |
atra | +0.213 |
wasn | +0.212 |
Neg logit contributions | |
Proced | -0.261 |
yip | -0.243 |
Address | -0.227 |
Agg | -0.226 |
Fn | -0.225 |
Token | – |
Token ID | 784 |
Feature activation | -9.94e-02 |
Pos logit contributions | |
Proced | +0.012 |
yip | +0.011 |
Address | +0.010 |
Fn | +0.010 |
Agg | +0.010 |
Neg logit contributions | |
eyed | -0.011 |
didn | -0.011 |
CBS | -0.011 |
atra | -0.010 |
wasn | -0.010 |
Token | advocacy |
Token ID | 15364 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
eyed | +0.060 |
didn | +0.059 |
CBS | +0.058 |
atra | +0.058 |
wasn | +0.057 |
Neg logit contributions | |
Proced | -0.075 |
yip | -0.071 |
Address | -0.066 |
Agg | -0.066 |
Fn | -0.065 |
Token | civil |
Token ID | 3026 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
Proced | +0.054 |
yip | +0.051 |
Address | +0.046 |
Fn | +0.046 |
aning | +0.046 |
Neg logit contributions | |
eyed | -0.062 |
CBS | -0.061 |
didn | -0.061 |
atra | -0.060 |
photos | -0.060 |
Token | rights |
Token ID | 2489 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
Proced | +0.019 |
yip | +0.017 |
Address | +0.016 |
Fn | +0.016 |
aning | +0.016 |
Neg logit contributions | |
eyed | -0.017 |
CBS | -0.017 |
didn | -0.017 |
atra | -0.017 |
photos | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -5.52e-02 |
Pos logit contributions | |
Proced | +0.025 |
yip | +0.023 |
Fn | +0.021 |
aning | +0.021 |
Address | +0.021 |
Neg logit contributions | |
eyed | -0.029 |
CBS | -0.028 |
didn | -0.028 |
atra | -0.028 |
photos | -0.028 |
Token | near |
Token ID | 1474 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
eyed | +0.037 |
didn | +0.037 |
CBS | +0.037 |
atra | +0.036 |
photos | +0.036 |
Neg logit contributions | |
Proced | -0.037 |
yip | -0.035 |
Fn | -0.032 |
aning | -0.032 |
Address | -0.032 |
Token | - |
Token ID | 12 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
eyed | +0.028 |
didn | +0.028 |
CBS | +0.028 |
wasn | +0.027 |
atra | +0.027 |
Neg logit contributions | |
Proced | -0.047 |
yip | -0.046 |
Address | -0.042 |
Fn | -0.042 |
Agg | -0.042 |
Token | term |
Token ID | 4354 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
eyed | +0.033 |
CBS | +0.032 |
didn | +0.032 |
wasn | +0.032 |
atra | +0.032 |
Neg logit contributions | |
Proced | -0.024 |
yip | -0.023 |
Address | -0.021 |
Agg | -0.020 |
aning | -0.020 |
Token | economic |
Token ID | 3034 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
Proced | +0.154 |
yip | +0.146 |
Fn | +0.135 |
aning | +0.135 |
Address | +0.134 |
Neg logit contributions | |
eyed | -0.135 |
CBS | -0.133 |
atra | -0.133 |
didn | -0.132 |
photos | -0.129 |
Token | picture |
Token ID | 4286 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
Proced | +0.151 |
yip | +0.143 |
aning | +0.133 |
Address | +0.133 |
Fn | +0.132 |
Neg logit contributions | |
eyed | -0.123 |
CBS | -0.121 |
didn | -0.121 |
atra | -0.120 |
photos | -0.116 |
Token | was |
Token ID | 373 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
eyed | +0.180 |
CBS | +0.176 |
atra | +0.174 |
didn | +0.173 |
photos | +0.168 |
Neg logit contributions | |
Proced | -0.321 |
yip | -0.307 |
aning | -0.289 |
Fn | -0.288 |
Address | -0.288 |
Token | one |
Token ID | 530 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
eyed | +0.069 |
didn | +0.068 |
atra | +0.068 |
CBS | +0.067 |
photos | +0.065 |
Neg logit contributions | |
Proced | -0.085 |
yip | -0.080 |
aning | -0.076 |
Fn | -0.075 |
Address | -0.075 |
Token | of |
Token ID | 286 |
Feature activation | -8.34e-03 |
Pos logit contributions | |
didn | +0.036 |
eyed | +0.036 |
CBS | +0.036 |
atra | +0.035 |
photos | +0.035 |
Neg logit contributions | |
Proced | -0.038 |
yip | -0.036 |
Address | -0.033 |
Agg | -0.033 |
Fn | -0.033 |
Token | slowdown |
Token ID | 35621 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
eyed | +0.005 |
CBS | +0.005 |
didn | +0.005 |
atra | +0.005 |
photos | +0.005 |
Neg logit contributions | |
Proced | -0.006 |
yip | -0.006 |
aning | -0.005 |
Fn | -0.005 |
Address | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +8.02e-02 |
Pos logit contributions | |
Proced | +0.159 |
yip | +0.151 |
aning | +0.138 |
Address | +0.138 |
Fn | +0.137 |
Neg logit contributions | |
eyed | -0.167 |
didn | -0.165 |
CBS | -0.165 |
atra | -0.163 |
photos | -0.161 |