Token | he |
Token ID | 339 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
ops | +0.081 |
iest | +0.081 |
affles | +0.078 |
hardly | +0.077 |
gh | +0.075 |
Neg logit contributions | |
nowhere | -0.061 |
who | -0.055 |
crying | -0.048 |
blood | -0.047 |
Blue | -0.047 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
iest | +0.060 |
ops | +0.058 |
affles | +0.057 |
robat | +0.057 |
rd | +0.053 |
Neg logit contributions | |
nowhere | -0.051 |
who | -0.048 |
crying | -0.040 |
Blue | -0.039 |
blood | -0.038 |
Token | 't |
Token ID | 470 |
Feature activation | +7.21e-03 |
Pos logit contributions | |
iest | +0.037 |
ops | +0.037 |
affles | +0.036 |
hardly | +0.034 |
robat | +0.034 |
Neg logit contributions | |
nowhere | -0.035 |
who | -0.032 |
crying | -0.029 |
Blue | -0.029 |
blood | -0.029 |
Token | want |
Token ID | 765 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
ops | +0.013 |
affles | +0.012 |
iest | +0.012 |
hardly | +0.012 |
ves | +0.012 |
Neg logit contributions | |
nowhere | -0.013 |
who | -0.013 |
crying | -0.012 |
Blue | -0.012 |
blood | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
nowhere | +0.014 |
who | +0.011 |
blood | +0.009 |
crying | +0.009 |
Blue | +0.009 |
Neg logit contributions | |
iest | -0.034 |
ops | -0.034 |
hardly | -0.034 |
gh | -0.033 |
affles | -0.033 |
Token | br |
Token ID | 865 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
iest | +0.026 |
ops | +0.026 |
affles | +0.025 |
robat | +0.024 |
zed | +0.024 |
Neg logit contributions | |
who | -0.021 |
nowhere | -0.020 |
Blue | -0.019 |
crying | -0.019 |
blood | -0.017 |
Token | ag |
Token ID | 363 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
iest | +0.220 |
hardly | +0.219 |
ops | +0.212 |
robat | +0.197 |
rd | +0.195 |
Neg logit contributions | |
nowhere | -0.229 |
who | -0.204 |
blood | -0.194 |
crying | -0.181 |
bug | -0.175 |
Token | . |
Token ID | 13 |
Feature activation | +6.78e-04 |
Pos logit contributions | |
ops | +0.075 |
iest | +0.074 |
hardly | +0.072 |
affles | +0.072 |
gh | +0.068 |
Neg logit contributions | |
nowhere | -0.057 |
who | -0.051 |
blood | -0.044 |
crying | -0.043 |
Blue | -0.042 |
Token | |
Token ID | 220 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
ops | +0.001 |
hardly | +0.001 |
iest | +0.001 |
affles | +0.001 |
gh | +0.001 |
Neg logit contributions | |
nowhere | -0.001 |
who | -0.001 |
blood | -0.001 |
crying | -0.001 |
weeping | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +6.65e-03 |
Pos logit contributions | |
ops | +0.111 |
iest | +0.108 |
hardly | +0.108 |
affles | +0.105 |
gh | +0.104 |
Neg logit contributions | |
nowhere | -0.053 |
who | -0.042 |
blood | -0.038 |
crying | -0.037 |
weeping | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -8.39e-03 |
Pos logit contributions | |
ops | +0.017 |
iest | +0.016 |
hardly | +0.016 |
affles | +0.016 |
gh | +0.016 |
Neg logit contributions | |
nowhere | -0.008 |
who | -0.006 |
blood | -0.006 |
crying | -0.005 |
Blue | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
iest | +0.011 |
ops | +0.011 |
hardly | +0.011 |
affles | +0.011 |
gh | +0.011 |
Neg logit contributions | |
nowhere | -0.005 |
who | -0.004 |
crying | -0.003 |
blood | -0.003 |
Blue | -0.003 |
Token | be |
Token ID | 307 |
Feature activation | +1.08e-03 |
Pos logit contributions | |
iest | +0.037 |
ops | +0.037 |
affles | +0.035 |
rd | +0.033 |
zed | +0.032 |
Neg logit contributions | |
who | -0.030 |
nowhere | -0.028 |
blood | -0.025 |
Blue | -0.025 |
crying | -0.025 |
Token | modest |
Token ID | 12949 |
Feature activation | +5.83e-03 |
Pos logit contributions | |
iest | +0.002 |
hardly | +0.002 |
ops | +0.002 |
gh | +0.002 |
affles | +0.002 |
Neg logit contributions | |
nowhere | -0.002 |
who | -0.002 |
blood | -0.002 |
crying | -0.002 |
Blue | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -6.97e-03 |
Pos logit contributions | |
ops | +0.016 |
iest | +0.016 |
hardly | +0.015 |
affles | +0.015 |
gh | +0.015 |
Neg logit contributions | |
nowhere | -0.006 |
who | -0.004 |
blood | -0.004 |
crying | -0.003 |
Blue | -0.003 |
Token | not |
Token ID | 407 |
Feature activation | +1.67e-03 |
Pos logit contributions | |
nowhere | +0.011 |
who | +0.011 |
crying | +0.009 |
blood | +0.009 |
Blue | +0.009 |
Neg logit contributions | |
ops | -0.015 |
iest | -0.015 |
affles | -0.014 |
robat | -0.014 |
gh | -0.013 |
Token | br |
Token ID | 865 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
ops | +0.003 |
iest | +0.003 |
hardly | +0.003 |
affles | +0.003 |
gh | +0.003 |
Neg logit contributions | |
nowhere | -0.003 |
who | -0.002 |
blood | -0.002 |
Blue | -0.002 |
crying | -0.002 |
Token | ag |
Token ID | 363 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
hardly | +0.164 |
iest | +0.164 |
ops | +0.156 |
rd | +0.145 |
robat | +0.143 |
Neg logit contributions | |
nowhere | -0.269 |
who | -0.250 |
blood | -0.240 |
crying | -0.228 |
bug | -0.217 |
Token | about |
Token ID | 546 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
ops | +0.070 |
iest | +0.070 |
hardly | +0.068 |
affles | +0.067 |
gh | +0.065 |
Neg logit contributions | |
nowhere | -0.051 |
who | -0.045 |
blood | -0.039 |
crying | -0.039 |
Blue | -0.038 |
Token | their |
Token ID | 511 |
Feature activation | +6.27e-03 |
Pos logit contributions | |
nowhere | +0.031 |
who | +0.023 |
bug | +0.022 |
Blue | +0.022 |
blood | +0.021 |
Neg logit contributions | |
ops | -0.049 |
hardly | -0.049 |
iest | -0.048 |
affles | -0.047 |
gh | -0.047 |
Token | toys |
Token ID | 14958 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
iest | +0.013 |
ops | +0.013 |
hardly | +0.013 |
affles | +0.013 |
zed | +0.012 |
Neg logit contributions | |
nowhere | -0.010 |
who | -0.009 |
crying | -0.008 |
blood | -0.008 |
Blue | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -8.55e-03 |
Pos logit contributions | |
iest | +0.031 |
ops | +0.031 |
hardly | +0.030 |
affles | +0.030 |
gh | +0.029 |
Neg logit contributions | |
nowhere | -0.018 |
who | -0.015 |
blood | -0.013 |
crying | -0.013 |
Blue | -0.013 |
Token | clown |
Token ID | 25573 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
nowhere | +0.006 |
who | +0.005 |
blood | +0.004 |
crying | +0.004 |
Blue | +0.004 |
Neg logit contributions | |
iest | -0.015 |
ops | -0.015 |
hardly | -0.015 |
affles | -0.015 |
ves | -0.014 |
Token | s |
Token ID | 82 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
ops | +0.034 |
iest | +0.034 |
hardly | +0.033 |
affles | +0.032 |
gh | +0.031 |
Neg logit contributions | |
nowhere | -0.048 |
who | -0.043 |
blood | -0.041 |
crying | -0.040 |
Blue | -0.040 |
Token | juggling |
Token ID | 48097 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
ops | +0.054 |
iest | +0.054 |
hardly | +0.053 |
affles | +0.052 |
gh | +0.051 |
Neg logit contributions | |
nowhere | -0.036 |
who | -0.031 |
blood | -0.027 |
crying | -0.027 |
Blue | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -8.16e-03 |
Pos logit contributions | |
ops | +0.056 |
iest | +0.056 |
hardly | +0.054 |
affles | +0.054 |
gh | +0.052 |
Neg logit contributions | |
nowhere | -0.030 |
who | -0.025 |
blood | -0.022 |
crying | -0.021 |
Blue | -0.021 |
Token | ac |
Token ID | 936 |
Feature activation | -1.19e-03 |
Pos logit contributions | |
nowhere | +0.013 |
who | +0.012 |
crying | +0.011 |
Blue | +0.010 |
blood | +0.010 |
Neg logit contributions | |
iest | -0.017 |
ops | -0.017 |
affles | -0.017 |
hardly | -0.016 |
gh | -0.016 |
Token | rob |
Token ID | 22609 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
nowhere | +0.003 |
who | +0.003 |
blood | +0.003 |
Blue | +0.003 |
crying | +0.003 |
Neg logit contributions | |
hardly | -0.001 |
iest | -0.001 |
affles | -0.001 |
ops | -0.001 |
gh | -0.001 |
Token | ats |
Token ID | 1381 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
iest | +0.119 |
hardly | +0.115 |
vation | +0.099 |
€ | +0.096 |
rated | +0.095 |
Neg logit contributions | |
nowhere | -0.296 |
who | -0.276 |
blood | -0.255 |
crying | -0.250 |
bug | -0.246 |
Token | doing |
Token ID | 1804 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
iest | +0.062 |
ops | +0.062 |
affles | +0.061 |
hardly | +0.060 |
gh | +0.059 |
Neg logit contributions | |
nowhere | -0.031 |
who | -0.027 |
crying | -0.023 |
blood | -0.022 |
Blue | -0.022 |
Token | flips |
Token ID | 45971 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
ops | +0.076 |
iest | +0.076 |
affles | +0.072 |
hardly | +0.072 |
rd | +0.069 |
Neg logit contributions | |
nowhere | -0.067 |
who | -0.062 |
crying | -0.054 |
blood | -0.054 |
Blue | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
iest | +0.039 |
ops | +0.039 |
hardly | +0.038 |
affles | +0.037 |
gh | +0.037 |
Neg logit contributions | |
nowhere | -0.020 |
who | -0.017 |
blood | -0.014 |
crying | -0.014 |
Blue | -0.014 |
Token | But |
Token ID | 887 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
hardly | +0.022 |
ops | +0.022 |
zed | +0.020 |
ves | +0.019 |
affles | +0.019 |
Neg logit contributions | |
nowhere | -0.020 |
blood | -0.017 |
who | -0.016 |
ter | -0.016 |
crying | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +5.41e-03 |
Pos logit contributions | |
iest | +0.009 |
hardly | +0.009 |
particular | +0.009 |
session | +0.008 |
activities | +0.008 |
Neg logit contributions | |
ter | -0.006 |
elt | -0.006 |
who | -0.006 |
nowhere | -0.006 |
Acc | -0.006 |
Token | forest |
Token ID | 8222 |
Feature activation | +7.50e-03 |
Pos logit contributions | |
iest | +0.009 |
ops | +0.008 |
hardly | +0.008 |
rd | +0.008 |
robat | +0.008 |
Neg logit contributions | |
nowhere | -0.010 |
who | -0.010 |
blood | -0.009 |
crying | -0.008 |
bug | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +1.84e-03 |
Pos logit contributions | |
ops | +0.016 |
hardly | +0.015 |
iest | +0.015 |
ves | +0.014 |
zed | +0.014 |
Neg logit contributions | |
nowhere | -0.012 |
blood | -0.009 |
who | -0.009 |
Blue | -0.009 |
crying | -0.008 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
hardly | +0.003 |
ops | +0.003 |
iest | +0.003 |
affles | +0.003 |
zed | +0.003 |
Neg logit contributions | |
nowhere | -0.003 |
who | -0.003 |
blood | -0.003 |
crying | -0.003 |
Blue | -0.002 |
Token | 't |
Token ID | 470 |
Feature activation | +3.31e-03 |
Pos logit contributions | |
ops | +0.031 |
hardly | +0.031 |
iest | +0.030 |
affles | +0.028 |
gh | +0.028 |
Neg logit contributions | |
nowhere | -0.039 |
who | -0.035 |
blood | -0.032 |
crying | -0.032 |
Blue | -0.031 |
Token | br |
Token ID | 865 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
ops | +0.005 |
iest | +0.005 |
affles | +0.005 |
hardly | +0.005 |
gh | +0.005 |
Neg logit contributions | |
nowhere | -0.007 |
who | -0.006 |
crying | -0.006 |
blood | -0.006 |
Blue | -0.006 |
Token | ag |
Token ID | 363 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
iest | +0.185 |
ops | +0.185 |
hardly | +0.182 |
affles | +0.168 |
rd | +0.164 |
Neg logit contributions | |
nowhere | -0.250 |
who | -0.226 |
blood | -0.211 |
crying | -0.205 |
Blue | -0.198 |
Token | about |
Token ID | 546 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
iest | +0.088 |
ops | +0.086 |
hardly | +0.084 |
affles | +0.082 |
gh | +0.082 |
Neg logit contributions | |
nowhere | -0.048 |
who | -0.039 |
blood | -0.036 |
crying | -0.035 |
Blue | -0.035 |
Token | her |
Token ID | 607 |
Feature activation | +7.12e-03 |
Pos logit contributions | |
nowhere | +0.031 |
bug | +0.023 |
late | +0.022 |
elt | +0.022 |
blood | +0.022 |
Neg logit contributions | |
hardly | -0.047 |
iest | -0.044 |
zed | -0.043 |
ops | -0.043 |
gh | -0.043 |
Token | skills |
Token ID | 4678 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
ops | +0.014 |
iest | +0.014 |
hardly | +0.013 |
gh | +0.013 |
affles | +0.013 |
Neg logit contributions | |
nowhere | -0.012 |
who | -0.010 |
blood | -0.009 |
Blue | -0.009 |
crying | -0.009 |
Token | anymore |
Token ID | 7471 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
ops | +0.071 |
iest | +0.071 |
hardly | +0.069 |
gh | +0.066 |
affles | +0.066 |
Neg logit contributions | |
nowhere | -0.049 |
who | -0.042 |
blood | -0.038 |
crying | -0.037 |
Blue | -0.036 |
Token | like |
Token ID | 588 |
Feature activation | -9.07e-03 |
Pos logit contributions | |
ops | +0.005 |
iest | +0.005 |
affles | +0.005 |
rd | +0.005 |
hardly | +0.005 |
Neg logit contributions | |
nowhere | -0.005 |
who | -0.005 |
blood | -0.004 |
crying | -0.004 |
Blue | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
nowhere | +0.011 |
who | +0.009 |
bug | +0.008 |
Blue | +0.008 |
crying | +0.008 |
Neg logit contributions | |
hardly | -0.020 |
iest | -0.020 |
zed | -0.019 |
ops | -0.019 |
affles | -0.019 |
Token | show |
Token ID | 905 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
ops | +0.055 |
iest | +0.055 |
affles | +0.051 |
rd | +0.050 |
hardly | +0.049 |
Neg logit contributions | |
who | -0.047 |
nowhere | -0.047 |
crying | -0.041 |
blood | -0.041 |
Blue | -0.040 |
Token | off |
Token ID | 572 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
iest | +0.068 |
ops | +0.067 |
hardly | +0.066 |
affles | +0.064 |
gh | +0.063 |
Neg logit contributions | |
nowhere | -0.050 |
who | -0.043 |
crying | -0.038 |
blood | -0.038 |
Blue | -0.037 |
Token | or |
Token ID | 393 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
nowhere | +0.051 |
who | +0.043 |
blood | +0.039 |
crying | +0.039 |
Blue | +0.038 |
Neg logit contributions | |
iest | -0.070 |
ops | -0.070 |
hardly | -0.068 |
affles | -0.067 |
zed | -0.066 |
Token | br |
Token ID | 865 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
nowhere | +0.027 |
who | +0.024 |
blood | +0.022 |
crying | +0.021 |
Blue | +0.021 |
Neg logit contributions | |
iest | -0.029 |
ops | -0.029 |
hardly | -0.029 |
affles | -0.028 |
gh | -0.027 |
Token | ag |
Token ID | 363 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
hardly | +0.185 |
iest | +0.182 |
ops | +0.180 |
robat | +0.163 |
rd | +0.160 |
Neg logit contributions | |
nowhere | -0.246 |
who | -0.221 |
blood | -0.211 |
crying | -0.198 |
bug | -0.193 |
Token | . |
Token ID | 13 |
Feature activation | +7.19e-03 |
Pos logit contributions | |
ops | +0.075 |
iest | +0.075 |
hardly | +0.073 |
affles | +0.072 |
gh | +0.070 |
Neg logit contributions | |
nowhere | -0.054 |
who | -0.048 |
blood | -0.042 |
crying | -0.041 |
Blue | -0.040 |
Token | He |
Token ID | 679 |
Feature activation | +9.75e-03 |
Pos logit contributions | |
hardly | +0.015 |
ops | +0.015 |
iest | +0.014 |
rd | +0.014 |
affles | +0.014 |
Neg logit contributions | |
nowhere | -0.011 |
who | -0.008 |
blood | -0.008 |
crying | -0.008 |
weeping | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | +7.60e-03 |
Pos logit contributions | |
iest | +0.019 |
ops | +0.019 |
affles | +0.018 |
hardly | +0.018 |
gh | +0.017 |
Neg logit contributions | |
nowhere | -0.018 |
who | -0.016 |
crying | -0.014 |
blood | -0.014 |
Blue | -0.014 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
hardly | +0.015 |
zed | +0.013 |
iest | +0.013 |
affles | +0.013 |
gh | +0.013 |
Neg logit contributions | |
nowhere | -0.011 |
who | -0.010 |
blood | -0.009 |
bug | -0.009 |
Blue | -0.009 |
Token | elephants |
Token ID | 25368 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
hardly | +0.012 |
ops | +0.012 |
iest | +0.012 |
affles | +0.012 |
ves | +0.012 |
Neg logit contributions | |
nowhere | -0.007 |
who | -0.005 |
blood | -0.005 |
crying | -0.005 |
Blue | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -8.16e-03 |
Pos logit contributions | |
iest | +0.054 |
ops | +0.054 |
hardly | +0.052 |
affles | +0.052 |
gh | +0.050 |
Neg logit contributions | |
nowhere | -0.049 |
who | -0.044 |
blood | -0.039 |
crying | -0.039 |
Blue | -0.038 |
Token | lions |
Token ID | 31420 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
nowhere | +0.010 |
crying | +0.007 |
who | +0.007 |
weeping | +0.007 |
Blue | +0.007 |
Neg logit contributions | |
iest | -0.019 |
ops | -0.019 |
hardly | -0.019 |
affles | -0.018 |
ves | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
ops | +0.029 |
iest | +0.029 |
hardly | +0.028 |
affles | +0.028 |
gh | +0.027 |
Neg logit contributions | |
nowhere | -0.031 |
who | -0.027 |
blood | -0.025 |
crying | -0.025 |
Blue | -0.024 |
Token | ac |
Token ID | 936 |
Feature activation | -9.72e-03 |
Pos logit contributions | |
nowhere | +0.044 |
who | +0.036 |
crying | +0.033 |
blood | +0.033 |
Blue | +0.032 |
Neg logit contributions | |
iest | -0.057 |
ops | -0.056 |
hardly | -0.055 |
affles | -0.054 |
rd | -0.052 |
Token | rob |
Token ID | 22609 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
nowhere | +0.027 |
who | +0.025 |
crying | +0.024 |
blood | +0.024 |
Blue | +0.023 |
Neg logit contributions | |
ops | -0.009 |
iest | -0.008 |
affles | -0.008 |
hardly | -0.008 |
robat | -0.007 |
Token | ats |
Token ID | 1381 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
iest | +0.129 |
hardly | +0.120 |
vation | +0.103 |
session | +0.096 |
€ | +0.095 |
Neg logit contributions | |
who | -0.278 |
nowhere | -0.277 |
blood | -0.252 |
crying | -0.246 |
bug | -0.244 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
iest | +0.053 |
ops | +0.052 |
hardly | +0.051 |
affles | +0.051 |
gh | +0.050 |
Neg logit contributions | |
nowhere | -0.030 |
who | -0.027 |
crying | -0.022 |
blood | -0.022 |
Blue | -0.022 |
Token | They |
Token ID | 1119 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
hardly | +0.021 |
ops | +0.021 |
iest | +0.020 |
ves | +0.020 |
affles | +0.020 |
Neg logit contributions | |
nowhere | -0.017 |
who | -0.014 |
crying | -0.014 |
blood | -0.014 |
weeping | -0.013 |
Token | are |
Token ID | 389 |
Feature activation | +7.04e-03 |
Pos logit contributions | |
iest | +0.047 |
ops | +0.046 |
gh | +0.045 |
affles | +0.044 |
rd | +0.044 |
Neg logit contributions | |
who | -0.044 |
nowhere | -0.041 |
crying | -0.036 |
Blue | -0.036 |
blood | -0.036 |
Token | amazed |
Token ID | 24562 |
Feature activation | +6.66e-03 |
Pos logit contributions | |
ops | +0.017 |
iest | +0.017 |
affles | +0.016 |
hardly | +0.016 |
gh | +0.016 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.008 |
crying | -0.007 |
blood | -0.007 |
Blue | -0.007 |
Token | will |
Token ID | 481 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
who | +0.008 |
nowhere | +0.008 |
bug | +0.007 |
a | +0.007 |
too | +0.006 |
Neg logit contributions | |
affles | -0.009 |
ops | -0.009 |
iest | -0.009 |
gh | -0.009 |
zed | -0.009 |
Token | lead |
Token ID | 1085 |
Feature activation | -6.77e-03 |
Pos logit contributions | |
iest | +0.032 |
ops | +0.031 |
affles | +0.031 |
rd | +0.029 |
robat | +0.029 |
Neg logit contributions | |
who | -0.024 |
nowhere | -0.023 |
a | -0.020 |
Blue | -0.020 |
crying | -0.020 |
Token | you |
Token ID | 345 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
nowhere | +0.012 |
who | +0.010 |
blood | +0.009 |
crying | +0.009 |
weeping | +0.009 |
Neg logit contributions | |
iest | -0.013 |
ops | -0.013 |
hardly | -0.012 |
affles | -0.012 |
gh | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +8.18e-03 |
Pos logit contributions | |
nowhere | +0.024 |
blood | +0.018 |
who | +0.017 |
crying | +0.016 |
Blue | +0.016 |
Neg logit contributions | |
iest | -0.050 |
ops | -0.049 |
hardly | -0.048 |
rd | -0.047 |
gh | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | -1.87e-03 |
Pos logit contributions | |
hardly | +0.018 |
iest | +0.018 |
ops | +0.018 |
gh | +0.018 |
affles | +0.017 |
Neg logit contributions | |
nowhere | -0.011 |
who | -0.009 |
blood | -0.008 |
crying | -0.008 |
ter | -0.007 |
Token | funn |
Token ID | 36090 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
nowhere | +0.003 |
who | +0.003 |
blood | +0.003 |
crying | +0.003 |
Blue | +0.002 |
Neg logit contributions | |
iest | -0.004 |
ops | -0.004 |
hardly | -0.004 |
affles | -0.004 |
gh | -0.003 |
Token | iest |
Token ID | 6386 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
hardly | +0.097 |
rd | +0.075 |
zed | +0.071 |
ops | +0.067 |
vows | +0.067 |
Neg logit contributions | |
who | -0.329 |
nowhere | -0.305 |
ter | -0.295 |
blood | -0.292 |
too | -0.288 |
Token | spot |
Token ID | 4136 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
who | +0.007 |
nowhere | +0.007 |
blood | +0.005 |
crying | +0.005 |
too | +0.005 |
Neg logit contributions | |
ops | -0.009 |
hardly | -0.009 |
affles | -0.009 |
zed | -0.008 |
iest | -0.008 |
Token | in |
Token ID | 287 |
Feature activation | -7.28e-03 |
Pos logit contributions | |
nowhere | +0.049 |
who | +0.042 |
blood | +0.035 |
crying | +0.035 |
Blue | +0.034 |
Neg logit contributions | |
ops | -0.091 |
iest | -0.091 |
hardly | -0.089 |
affles | -0.088 |
gh | -0.086 |
Token | the |
Token ID | 262 |
Feature activation | -8.01e-04 |
Pos logit contributions | |
nowhere | +0.008 |
who | +0.007 |
crying | +0.007 |
blood | +0.006 |
weeping | +0.006 |
Neg logit contributions | |
iest | -0.017 |
hardly | -0.016 |
ops | -0.016 |
gh | -0.016 |
ves | -0.016 |
Token | cage |
Token ID | 17186 |
Feature activation | +9.91e-03 |
Pos logit contributions | |
nowhere | +0.001 |
who | +0.001 |
blood | +0.001 |
crying | +0.001 |
Blue | +0.001 |
Neg logit contributions | |
ops | -0.002 |
iest | -0.002 |
hardly | -0.002 |
affles | -0.002 |
gh | -0.001 |
Token | up |
Token ID | 510 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
hardly | +0.070 |
iest | +0.070 |
ops | +0.069 |
affles | +0.068 |
gh | +0.067 |
Neg logit contributions | |
nowhere | -0.052 |
who | -0.042 |
blood | -0.037 |
crying | -0.037 |
bug | -0.036 |
Token | for |
Token ID | 329 |
Feature activation | -8.64e-03 |
Pos logit contributions | |
nowhere | +0.022 |
who | +0.017 |
blood | +0.016 |
Blue | +0.015 |
crying | +0.015 |
Neg logit contributions | |
iest | -0.034 |
ops | -0.033 |
hardly | -0.033 |
gh | -0.032 |
affles | -0.032 |
Token | it |
Token ID | 340 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
nowhere | +0.014 |
who | +0.011 |
bug | +0.011 |
ety | +0.010 |
blood | +0.010 |
Neg logit contributions | |
hardly | -0.018 |
iest | -0.017 |
ops | -0.017 |
gh | -0.016 |
zed | -0.016 |
Token | by |
Token ID | 416 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
nowhere | +0.029 |
who | +0.024 |
blood | +0.022 |
crying | +0.021 |
Blue | +0.021 |
Neg logit contributions | |
iest | -0.036 |
ops | -0.036 |
hardly | -0.035 |
gh | -0.034 |
affles | -0.033 |
Token | telling |
Token ID | 5149 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
hardly | +0.060 |
iest | +0.059 |
gh | +0.058 |
ops | +0.058 |
zed | +0.057 |
Neg logit contributions | |
nowhere | -0.044 |
who | -0.037 |
bug | -0.034 |
Blue | -0.033 |
blood | -0.033 |
Token | funn |
Token ID | 36090 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
ops | +0.100 |
iest | +0.100 |
hardly | +0.096 |
affles | +0.096 |
gh | +0.093 |
Neg logit contributions | |
nowhere | -0.056 |
who | -0.049 |
blood | -0.042 |
crying | -0.042 |
Blue | -0.041 |
Token | ier |
Token ID | 959 |
Feature activation | -2.47e-03 |
Pos logit contributions | |
hardly | +0.205 |
robat | +0.184 |
rd | +0.183 |
vation | +0.181 |
zed | +0.178 |
Neg logit contributions | |
who | -0.214 |
nowhere | -0.203 |
blood | -0.197 |
too | -0.179 |
crying | -0.174 |
Token | jokes |
Token ID | 14532 |
Feature activation | +4.91e-02 |
Pos logit contributions | |
nowhere | +0.003 |
who | +0.003 |
crying | +0.002 |
blood | +0.002 |
where | +0.002 |
Neg logit contributions | |
ops | -0.006 |
affles | -0.006 |
iest | -0.006 |
hardly | -0.006 |
robat | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
affles | +0.121 |
iest | +0.120 |
ops | +0.119 |
hardly | +0.119 |
rd | +0.115 |
Neg logit contributions | |
nowhere | -0.057 |
who | -0.053 |
crying | -0.042 |
blood | -0.041 |
Blue | -0.040 |
Token | Everyone |
Token ID | 11075 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
hardly | +0.019 |
ops | +0.019 |
zed | +0.017 |
rd | +0.017 |
iest | +0.017 |
Neg logit contributions | |
nowhere | -0.019 |
who | -0.016 |
blood | -0.015 |
crying | -0.015 |
weeping | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -2.02e-03 |
Pos logit contributions | |
ops | +0.033 |
iest | +0.033 |
hardly | +0.032 |
affles | +0.032 |
gh | +0.031 |
Neg logit contributions | |
nowhere | -0.030 |
who | -0.027 |
blood | -0.024 |
crying | -0.024 |
Blue | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
nowhere | +0.022 |
who | +0.018 |
blood | +0.016 |
Blue | +0.016 |
ter | +0.015 |
Neg logit contributions | |
hardly | -0.043 |
ops | -0.042 |
iest | -0.042 |
gh | -0.041 |
zed | -0.041 |
Token | match |
Token ID | 2872 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
nowhere | +0.023 |
who | +0.021 |
crying | +0.020 |
blood | +0.020 |
Blue | +0.019 |
Neg logit contributions | |
ops | -0.020 |
iest | -0.020 |
affles | -0.018 |
hardly | -0.018 |
gh | -0.018 |
Token | things |
Token ID | 1243 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
iest | +0.069 |
ops | +0.069 |
hardly | +0.067 |
affles | +0.066 |
gh | +0.065 |
Neg logit contributions | |
nowhere | -0.034 |
who | -0.027 |
blood | -0.024 |
crying | -0.024 |
Blue | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | +7.88e-03 |
Pos logit contributions | |
iest | +0.061 |
ops | +0.061 |
hardly | +0.059 |
gh | +0.057 |
affles | +0.057 |
Neg logit contributions | |
nowhere | -0.031 |
who | -0.025 |
blood | -0.023 |
crying | -0.022 |
Blue | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
iest | +0.018 |
hardly | +0.017 |
ops | +0.017 |
gh | +0.016 |
rd | +0.016 |
Neg logit contributions | |
nowhere | -0.010 |
who | -0.009 |
blood | -0.008 |
crying | -0.008 |
Blue | -0.008 |
Token | funn |
Token ID | 36090 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
iest | +0.024 |
ops | +0.024 |
hardly | +0.023 |
affles | +0.023 |
gh | +0.022 |
Neg logit contributions | |
nowhere | -0.028 |
who | -0.025 |
blood | -0.024 |
crying | -0.023 |
Blue | -0.023 |
Token | iest |
Token ID | 6386 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
hardly | +0.100 |
robat | +0.087 |
ops | +0.085 |
€ | +0.077 |
liv | +0.075 |
Neg logit contributions | |
nowhere | -0.296 |
who | -0.295 |
blood | -0.270 |
in | -0.263 |
a | -0.262 |
Token | ways |
Token ID | 2842 |
Feature activation | -6.41e-03 |
Pos logit contributions | |
ops | +0.032 |
hardly | +0.031 |
iest | +0.031 |
affles | +0.031 |
gh | +0.029 |
Neg logit contributions | |
nowhere | -0.034 |
who | -0.031 |
blood | -0.028 |
crying | -0.028 |
Blue | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
nowhere | +0.008 |
who | +0.007 |
crying | +0.006 |
blood | +0.006 |
Blue | +0.006 |
Neg logit contributions | |
ops | -0.015 |
iest | -0.015 |
affles | -0.015 |
hardly | -0.015 |
gh | -0.014 |
Token | He |
Token ID | 679 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
hardly | +0.018 |
ops | +0.017 |
iest | +0.016 |
affles | +0.016 |
rd | +0.016 |
Neg logit contributions | |
nowhere | -0.020 |
who | -0.016 |
blood | -0.016 |
crying | -0.015 |
ter | -0.015 |
Token | would |
Token ID | 561 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
iest | +0.020 |
ops | +0.020 |
affles | +0.020 |
hardly | +0.019 |
gh | +0.019 |
Neg logit contributions | |
nowhere | -0.019 |
who | -0.017 |
crying | -0.015 |
blood | -0.015 |
Blue | -0.015 |
Token | thought |
Token ID | 1807 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
iest | +0.012 |
affles | +0.011 |
ops | +0.011 |
gh | +0.011 |
rd | +0.011 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.009 |
crying | -0.008 |
Blue | -0.007 |
blood | -0.007 |
Token | they |
Token ID | 484 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
ops | +0.035 |
iest | +0.034 |
hardly | +0.033 |
affles | +0.033 |
gh | +0.032 |
Neg logit contributions | |
nowhere | -0.019 |
who | -0.016 |
blood | -0.014 |
crying | -0.014 |
Blue | -0.013 |
Token | looked |
Token ID | 3114 |
Feature activation | -2.26e-03 |
Pos logit contributions | |
iest | +0.019 |
ops | +0.019 |
affles | +0.019 |
gh | +0.018 |
hardly | +0.018 |
Neg logit contributions | |
nowhere | -0.021 |
who | -0.019 |
crying | -0.017 |
blood | -0.017 |
Blue | -0.016 |
Token | like |
Token ID | 588 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
nowhere | +0.004 |
who | +0.003 |
blood | +0.003 |
crying | +0.003 |
Blue | +0.003 |
Neg logit contributions | |
iest | -0.005 |
ops | -0.005 |
hardly | -0.005 |
affles | -0.004 |
gh | -0.004 |
Token | ac |
Token ID | 936 |
Feature activation | -8.50e-03 |
Pos logit contributions | |
nowhere | +0.036 |
who | +0.029 |
blood | +0.027 |
crying | +0.026 |
weeping | +0.025 |
Neg logit contributions | |
iest | -0.066 |
hardly | -0.065 |
ops | -0.064 |
affles | -0.062 |
gh | -0.062 |
Token | rob |
Token ID | 22609 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
nowhere | +0.023 |
who | +0.021 |
blood | +0.020 |
Blue | +0.020 |
crying | +0.020 |
Neg logit contributions | |
hardly | -0.008 |
iest | -0.008 |
ops | -0.007 |
affles | -0.007 |
gh | -0.007 |
Token | ats |
Token ID | 1381 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
hardly | +0.115 |
iest | +0.113 |
rated | +0.092 |
vows | +0.092 |
€ | +0.090 |
Neg logit contributions | |
nowhere | -0.280 |
who | -0.262 |
blood | -0.247 |
crying | -0.246 |
bug | -0.239 |
Token | . |
Token ID | 13 |
Feature activation | +7.34e-03 |
Pos logit contributions | |
ops | +0.029 |
iest | +0.029 |
hardly | +0.029 |
affles | +0.029 |
gh | +0.028 |
Neg logit contributions | |
nowhere | -0.012 |
who | -0.010 |
crying | -0.008 |
blood | -0.008 |
Blue | -0.008 |
Token | She |
Token ID | 1375 |
Feature activation | +9.02e-03 |
Pos logit contributions | |
ops | +0.014 |
hardly | +0.014 |
affles | +0.013 |
zed | +0.013 |
rd | +0.013 |
Neg logit contributions | |
nowhere | -0.013 |
blood | -0.010 |
who | -0.010 |
crying | -0.009 |
ter | -0.009 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
iest | +0.019 |
ops | +0.018 |
affles | +0.017 |
gh | +0.017 |
robat | +0.017 |
Neg logit contributions | |
nowhere | -0.016 |
who | -0.014 |
crying | -0.012 |
blood | -0.012 |
Blue | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
nowhere | +0.018 |
who | +0.015 |
blood | +0.014 |
Blue | +0.013 |
bug | +0.013 |
Neg logit contributions | |
hardly | -0.025 |
ops | -0.025 |
iest | -0.024 |
affles | -0.024 |
zed | -0.024 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
nowhere | +0.003 |
who | +0.003 |
blood | +0.002 |
Blue | +0.002 |
crying | +0.002 |
Neg logit contributions | |
ops | -0.003 |
hardly | -0.003 |
iest | -0.003 |
affles | -0.002 |
ves | -0.002 |
Token | clown |
Token ID | 25573 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
nowhere | +0.021 |
who | +0.015 |
weeping | +0.014 |
crying | +0.014 |
bug | +0.014 |
Neg logit contributions | |
iest | -0.033 |
ops | -0.031 |
affles | -0.031 |
ves | -0.031 |
hardly | -0.030 |
Token | s |
Token ID | 82 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
ops | +0.023 |
iest | +0.023 |
hardly | +0.022 |
affles | +0.022 |
gh | +0.021 |
Neg logit contributions | |
nowhere | -0.033 |
who | -0.030 |
blood | -0.028 |
crying | -0.027 |
Blue | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
ops | +0.044 |
iest | +0.043 |
hardly | +0.043 |
gh | +0.041 |
affles | +0.041 |
Neg logit contributions | |
nowhere | -0.021 |
who | -0.016 |
blood | -0.014 |
Blue | -0.014 |
crying | -0.013 |
Token | ac |
Token ID | 936 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
nowhere | +0.026 |
who | +0.023 |
blood | +0.020 |
crying | +0.020 |
Blue | +0.020 |
Neg logit contributions | |
ops | -0.035 |
iest | -0.035 |
hardly | -0.033 |
affles | -0.033 |
gh | -0.032 |
Token | rob |
Token ID | 22609 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
nowhere | +0.038 |
who | +0.036 |
crying | +0.034 |
blood | +0.033 |
Blue | +0.033 |
Neg logit contributions | |
ops | -0.014 |
iest | -0.013 |
affles | -0.012 |
robat | -0.012 |
hardly | -0.011 |
Token | ats |
Token ID | 1381 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
iest | +0.115 |
hardly | +0.108 |
vation | +0.094 |
€ | +0.088 |
session | +0.085 |
Neg logit contributions | |
nowhere | -0.284 |
who | -0.273 |
blood | -0.256 |
crying | -0.251 |
bug | -0.242 |
Token | and |
Token ID | 290 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ops | +0.032 |
iest | +0.032 |
hardly | +0.031 |
affles | +0.031 |
gh | +0.030 |
Neg logit contributions | |
nowhere | -0.014 |
who | -0.011 |
blood | -0.010 |
crying | -0.009 |
Blue | -0.009 |
Token | lions |
Token ID | 31420 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
nowhere | +0.031 |
who | +0.028 |
blood | +0.025 |
crying | +0.024 |
Blue | +0.024 |
Neg logit contributions | |
ops | -0.040 |
iest | -0.040 |
hardly | -0.039 |
affles | -0.038 |
gh | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
ops | +0.046 |
iest | +0.045 |
hardly | +0.044 |
affles | +0.044 |
gh | +0.043 |
Neg logit contributions | |
nowhere | -0.027 |
who | -0.023 |
blood | -0.020 |
crying | -0.020 |
Blue | -0.019 |
Token | elephants |
Token ID | 25368 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
nowhere | +0.039 |
who | +0.034 |
blood | +0.031 |
crying | +0.030 |
Blue | +0.030 |
Neg logit contributions | |
iest | -0.047 |
ops | -0.047 |
hardly | -0.046 |
affles | -0.045 |
gh | -0.043 |
Token | this |
Token ID | 428 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
ops | +0.035 |
iest | +0.035 |
hardly | +0.034 |
affles | +0.033 |
gh | +0.033 |
Neg logit contributions | |
nowhere | -0.026 |
who | -0.023 |
blood | -0.021 |
crying | -0.020 |
Blue | -0.020 |
Token | time |
Token ID | 640 |
Feature activation | +3.06e-03 |
Pos logit contributions | |
ops | +0.079 |
affles | +0.079 |
hardly | +0.074 |
iest | +0.073 |
robat | +0.072 |
Neg logit contributions | |
nowhere | -0.092 |
who | -0.082 |
a | -0.071 |
crying | -0.069 |
too | -0.068 |
Token | she |
Token ID | 673 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
ops | +0.006 |
iest | +0.006 |
hardly | +0.006 |
affles | +0.006 |
gh | +0.006 |
Neg logit contributions | |
nowhere | -0.005 |
who | -0.004 |
blood | -0.004 |
crying | -0.004 |
Blue | -0.004 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
iest | +0.023 |
affles | +0.022 |
rd | +0.021 |
ops | +0.021 |
gh | +0.021 |
Neg logit contributions | |
nowhere | -0.019 |
who | -0.017 |
crying | -0.015 |
blood | -0.014 |
bug | -0.014 |
Token | 't |
Token ID | 470 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
iest | +0.036 |
ops | +0.035 |
affles | +0.034 |
hardly | +0.033 |
gh | +0.032 |
Neg logit contributions | |
nowhere | -0.036 |
who | -0.032 |
crying | -0.029 |
blood | -0.029 |
Blue | -0.029 |
Token | br |
Token ID | 865 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
ops | +0.007 |
iest | +0.006 |
affles | +0.006 |
rd | +0.006 |
gh | +0.006 |
Neg logit contributions | |
nowhere | -0.007 |
who | -0.006 |
crying | -0.006 |
blood | -0.006 |
Blue | -0.006 |
Token | ag |
Token ID | 363 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
iest | +0.170 |
ops | +0.168 |
hardly | +0.168 |
affles | +0.152 |
robat | +0.150 |
Neg logit contributions | |
nowhere | -0.243 |
who | -0.220 |
blood | -0.207 |
crying | -0.201 |
bug | -0.194 |
Token | about |
Token ID | 546 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
ops | +0.076 |
iest | +0.076 |
hardly | +0.073 |
affles | +0.072 |
gh | +0.071 |
Neg logit contributions | |
nowhere | -0.048 |
who | -0.042 |
blood | -0.037 |
crying | -0.036 |
Blue | -0.035 |
Token | her |
Token ID | 607 |
Feature activation | +5.32e-03 |
Pos logit contributions | |
nowhere | +0.031 |
who | +0.024 |
blood | +0.022 |
Blue | +0.021 |
crying | +0.021 |
Neg logit contributions | |
ops | -0.053 |
hardly | -0.052 |
iest | -0.052 |
affles | -0.050 |
gh | -0.050 |
Token | singing |
Token ID | 13777 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
iest | +0.012 |
ops | +0.011 |
hardly | +0.011 |
affles | +0.011 |
gh | +0.011 |
Neg logit contributions | |
nowhere | -0.008 |
who | -0.007 |
crying | -0.006 |
blood | -0.006 |
Blue | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +6.57e-04 |
Pos logit contributions | |
ops | +0.055 |
affles | +0.055 |
iest | +0.055 |
hardly | +0.054 |
rd | +0.052 |
Neg logit contributions | |
nowhere | -0.029 |
who | -0.026 |
crying | -0.022 |
blood | -0.021 |
Blue | -0.021 |
Token | it |
Token ID | 340 |
Feature activation | +3.83e-03 |
Pos logit contributions | |
nowhere | +0.012 |
who | +0.011 |
crying | +0.009 |
blood | +0.009 |
Blue | +0.009 |
Neg logit contributions | |
ops | -0.017 |
iest | -0.016 |
affles | -0.016 |
hardly | -0.016 |
gh | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | +4.34e-03 |
Pos logit contributions | |
ops | +0.007 |
iest | +0.007 |
affles | +0.007 |
hardly | +0.007 |
gh | +0.007 |
Neg logit contributions | |
nowhere | -0.007 |
who | -0.006 |
crying | -0.006 |
blood | -0.006 |
Blue | -0.006 |
Token | best |
Token ID | 1266 |
Feature activation | -2.14e-03 |
Pos logit contributions | |
iest | +0.009 |
hardly | +0.009 |
ops | +0.009 |
gh | +0.009 |
affles | +0.009 |
Neg logit contributions | |
nowhere | -0.006 |
who | -0.005 |
blood | -0.005 |
crying | -0.005 |
Blue | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | -9.42e-03 |
Pos logit contributions | |
nowhere | +0.002 |
who | +0.002 |
crying | +0.001 |
blood | +0.001 |
too | +0.001 |
Neg logit contributions | |
ops | -0.006 |
affles | -0.006 |
robat | -0.006 |
iest | -0.006 |
zed | -0.006 |
Token | not |
Token ID | 407 |
Feature activation | -9.49e-03 |
Pos logit contributions | |
nowhere | +0.015 |
who | +0.014 |
crying | +0.013 |
Blue | +0.012 |
too | +0.012 |
Neg logit contributions | |
ops | -0.020 |
affles | -0.019 |
iest | -0.019 |
ves | -0.017 |
robat | -0.017 |
Token | br |
Token ID | 865 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
nowhere | +0.013 |
who | +0.011 |
blood | +0.010 |
crying | +0.009 |
Blue | +0.009 |
Neg logit contributions | |
iest | -0.022 |
ops | -0.021 |
hardly | -0.021 |
gh | -0.020 |
affles | -0.020 |
Token | ag |
Token ID | 363 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
hardly | +0.184 |
iest | +0.181 |
ops | +0.180 |
robat | +0.163 |
affles | +0.161 |
Neg logit contributions | |
nowhere | -0.225 |
who | -0.202 |
blood | -0.187 |
crying | -0.182 |
bug | -0.174 |
Token | or |
Token ID | 393 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
ops | +0.063 |
iest | +0.063 |
hardly | +0.061 |
affles | +0.061 |
gh | +0.059 |
Neg logit contributions | |
nowhere | -0.043 |
who | -0.037 |
blood | -0.033 |
crying | -0.032 |
Blue | -0.032 |
Token | show |
Token ID | 905 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
nowhere | +0.024 |
who | +0.022 |
crying | +0.019 |
blood | +0.019 |
Blue | +0.018 |
Neg logit contributions | |
ops | -0.028 |
iest | -0.027 |
affles | -0.026 |
hardly | -0.026 |
gh | -0.025 |
Token | off |
Token ID | 572 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
ops | +0.065 |
iest | +0.064 |
affles | +0.063 |
hardly | +0.062 |
rd | +0.060 |
Neg logit contributions | |
nowhere | -0.037 |
who | -0.033 |
crying | -0.028 |
blood | -0.028 |
Blue | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +3.51e-03 |
Pos logit contributions | |
nowhere | +0.040 |
who | +0.035 |
crying | +0.030 |
blood | +0.029 |
Blue | +0.028 |
Neg logit contributions | |
ops | -0.070 |
iest | -0.069 |
affles | -0.067 |
hardly | -0.067 |
gh | -0.065 |
Token | but |
Token ID | 475 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
hardly | +0.002 |
ops | +0.002 |
iest | +0.002 |
ves | +0.002 |
affles | +0.002 |
Neg logit contributions | |
nowhere | -0.002 |
who | -0.001 |
blood | -0.001 |
crying | -0.001 |
weeping | -0.001 |
Token | they |
Token ID | 484 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
hardly | +0.046 |
iest | +0.046 |
ops | +0.044 |
ves | +0.042 |
gh | +0.041 |
Neg logit contributions | |
nowhere | -0.047 |
who | -0.040 |
blood | -0.037 |
Blue | -0.036 |
bug | -0.036 |
Token | chose |
Token ID | 7690 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
iest | +0.060 |
ops | +0.060 |
affles | +0.058 |
gh | +0.057 |
hardly | +0.057 |
Neg logit contributions | |
nowhere | -0.045 |
who | -0.041 |
blood | -0.035 |
crying | -0.035 |
Blue | -0.034 |
Token | safer |
Token ID | 14178 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
nowhere | +0.021 |
who | +0.016 |
crying | +0.015 |
blood | +0.015 |
weeping | +0.014 |
Neg logit contributions | |
iest | -0.041 |
ops | -0.040 |
hardly | -0.040 |
gh | -0.039 |
ves | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | +2.41e-03 |
Pos logit contributions | |
nowhere | +0.031 |
who | +0.026 |
blood | +0.023 |
crying | +0.023 |
Blue | +0.022 |
Neg logit contributions | |
ops | -0.050 |
iest | -0.050 |
hardly | -0.048 |
affles | -0.048 |
gh | -0.046 |
Token | funn |
Token ID | 36090 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
iest | +0.005 |
ops | +0.005 |
hardly | +0.005 |
gh | +0.005 |
affles | +0.005 |
Neg logit contributions | |
nowhere | -0.003 |
who | -0.003 |
blood | -0.003 |
crying | -0.002 |
Blue | -0.002 |
Token | ier |
Token ID | 959 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
hardly | +0.223 |
rd | +0.199 |
zed | +0.193 |
ops | +0.192 |
robat | +0.192 |
Neg logit contributions | |
who | -0.189 |
nowhere | -0.189 |
blood | -0.170 |
ter | -0.152 |
a | -0.149 |
Token | ones |
Token ID | 3392 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
nowhere | +0.023 |
who | +0.023 |
blood | +0.018 |
crying | +0.018 |
where | +0.017 |
Neg logit contributions | |
affles | -0.039 |
ops | -0.039 |
hardly | -0.039 |
iest | -0.038 |
gh | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
ops | +0.067 |
iest | +0.066 |
hardly | +0.065 |
affles | +0.064 |
gh | +0.062 |
Neg logit contributions | |
nowhere | -0.040 |
who | -0.035 |
blood | -0.030 |
crying | -0.030 |
Blue | -0.029 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
hardly | +0.025 |
ops | +0.025 |
rd | +0.024 |
affles | +0.023 |
ves | +0.023 |
Neg logit contributions | |
nowhere | -0.021 |
who | -0.017 |
blood | -0.017 |
ter | -0.017 |
crying | -0.016 |
Token | Tom |
Token ID | 13787 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
nowhere | +0.065 |
who | +0.057 |
crying | +0.054 |
blood | +0.053 |
weeping | +0.052 |
Neg logit contributions | |
ops | -0.075 |
iest | -0.074 |
affles | -0.068 |
robat | -0.068 |
gh | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ops | +0.016 |
iest | +0.016 |
affles | +0.016 |
hardly | +0.016 |
gh | +0.016 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.008 |
crying | -0.007 |
blood | -0.007 |
Blue | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +2.29e-03 |
Pos logit contributions | |
ops | +0.025 |
hardly | +0.025 |
iest | +0.024 |
zed | +0.024 |
rd | +0.023 |
Neg logit contributions | |
nowhere | -0.017 |
blood | -0.013 |
who | -0.013 |
ter | -0.012 |
crying | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
ops | +0.006 |
hardly | +0.006 |
iest | +0.006 |
affles | +0.005 |
ves | +0.005 |
Neg logit contributions | |
nowhere | -0.003 |
who | -0.002 |
blood | -0.002 |
crying | -0.002 |
ter | -0.002 |
Token | Will |
Token ID | 8743 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
nowhere | +0.020 |
who | +0.018 |
crying | +0.016 |
blood | +0.016 |
Blue | +0.016 |
Neg logit contributions | |
ops | -0.023 |
iest | -0.023 |
affles | -0.022 |
hardly | -0.022 |
gh | -0.021 |
Token | 's |
Token ID | 338 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
ops | +0.071 |
iest | +0.071 |
hardly | +0.068 |
affles | +0.068 |
gh | +0.065 |
Neg logit contributions | |
nowhere | -0.075 |
who | -0.068 |
crying | -0.061 |
blood | -0.060 |
Blue | -0.059 |
Token | funn |
Token ID | 36090 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
ops | +0.085 |
iest | +0.083 |
affles | +0.079 |
hardly | +0.078 |
robat | +0.077 |
Neg logit contributions | |
nowhere | -0.046 |
who | -0.041 |
crying | -0.034 |
blood | -0.032 |
Blue | -0.031 |
Token | est |
Token ID | 395 |
Feature activation | -4.21e-03 |
Pos logit contributions | |
hardly | +0.087 |
rd | +0.066 |
ops | +0.060 |
vows | +0.059 |
experiments | +0.057 |
Neg logit contributions | |
nowhere | -0.312 |
who | -0.304 |
blood | -0.293 |
ter | -0.289 |
a | -0.276 |
Token | times |
Token ID | 1661 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
nowhere | +0.007 |
who | +0.007 |
blood | +0.006 |
crying | +0.006 |
a | +0.006 |
Neg logit contributions | |
ops | -0.008 |
affles | -0.008 |
hardly | -0.008 |
gh | -0.008 |
iest | -0.007 |
Token | were |
Token ID | 547 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
ops | +0.047 |
iest | +0.046 |
affles | +0.046 |
hardly | +0.045 |
gh | +0.044 |
Neg logit contributions | |
nowhere | -0.025 |
who | -0.023 |
crying | -0.019 |
blood | -0.019 |
Blue | -0.018 |
Token | when |
Token ID | 618 |
Feature activation | +5.01e-03 |
Pos logit contributions | |
nowhere | +0.046 |
who | +0.040 |
blood | +0.034 |
crying | +0.034 |
Blue | +0.033 |
Neg logit contributions | |
ops | -0.082 |
iest | -0.081 |
hardly | -0.079 |
affles | -0.079 |
gh | -0.076 |
Token | she |
Token ID | 673 |
Feature activation | +5.34e-03 |
Pos logit contributions | |
ops | +0.011 |
iest | +0.011 |
hardly | +0.011 |
affles | +0.011 |
gh | +0.010 |
Neg logit contributions | |
nowhere | -0.007 |
who | -0.006 |
blood | -0.005 |
crying | -0.005 |
Blue | -0.005 |
Token | " |
Token ID | 1 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
nowhere | +0.038 |
who | +0.034 |
blood | +0.030 |
crying | +0.029 |
Blue | +0.029 |
Neg logit contributions | |
iest | -0.047 |
ops | -0.047 |
affles | -0.046 |
hardly | -0.045 |
gh | -0.043 |
Token | Go |
Token ID | 5247 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
nowhere | +0.058 |
who | +0.052 |
crying | +0.047 |
Blue | +0.046 |
blood | +0.046 |
Neg logit contributions | |
iest | -0.066 |
ops | -0.066 |
hardly | -0.064 |
affles | -0.062 |
gh | -0.061 |
Token | away |
Token ID | 1497 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
ops | +0.041 |
iest | +0.041 |
hardly | +0.040 |
affles | +0.040 |
gh | +0.039 |
Neg logit contributions | |
nowhere | -0.026 |
who | -0.022 |
blood | -0.020 |
crying | -0.020 |
Blue | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
nowhere | +0.038 |
who | +0.036 |
crying | +0.032 |
blood | +0.031 |
Blue | +0.031 |
Neg logit contributions | |
ops | -0.042 |
iest | -0.042 |
affles | -0.041 |
hardly | -0.041 |
gh | -0.039 |
Token | you |
Token ID | 345 |
Feature activation | +6.11e-05 |
Pos logit contributions | |
nowhere | +0.022 |
who | +0.019 |
blood | +0.017 |
crying | +0.017 |
Blue | +0.016 |
Neg logit contributions | |
ops | -0.028 |
iest | -0.028 |
hardly | -0.027 |
affles | -0.027 |
gh | -0.026 |
Token | br |
Token ID | 865 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
iest | +0.000 |
ops | +0.000 |
hardly | +0.000 |
affles | +0.000 |
gh | +0.000 |
Neg logit contributions | |
nowhere | -0.000 |
who | -0.000 |
blood | -0.000 |
crying | -0.000 |
Blue | -0.000 |
Token | ats |
Token ID | 1381 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
hardly | +0.125 |
iest | +0.118 |
ops | +0.106 |
vows | +0.098 |
zed | +0.096 |
Neg logit contributions | |
nowhere | -0.267 |
who | -0.251 |
blood | -0.247 |
crying | -0.232 |
half | -0.226 |
Token | ! |
Token ID | 0 |
Feature activation | -2.23e-03 |
Pos logit contributions | |
iest | +0.030 |
ops | +0.030 |
hardly | +0.030 |
affles | +0.029 |
gh | +0.028 |
Neg logit contributions | |
nowhere | -0.033 |
who | -0.031 |
blood | -0.028 |
crying | -0.028 |
Blue | -0.027 |
Token | This |
Token ID | 770 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
nowhere | +0.004 |
who | +0.003 |
blood | +0.003 |
crying | +0.003 |
Blue | +0.003 |
Neg logit contributions | |
ops | -0.005 |
iest | -0.005 |
hardly | -0.004 |
affles | -0.004 |
gh | -0.004 |
Token | is |
Token ID | 318 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
hardly | +0.108 |
zed | +0.105 |
ops | +0.104 |
affles | +0.104 |
iest | +0.104 |
Neg logit contributions | |
who | -0.071 |
nowhere | -0.068 |
blood | -0.061 |
a | -0.058 |
where | -0.057 |
Token | none |
Token ID | 4844 |
Feature activation | +1.95e-03 |
Pos logit contributions | |
hardly | +0.022 |
iest | +0.022 |
ops | +0.022 |
affles | +0.021 |
gh | +0.021 |
Neg logit contributions | |
nowhere | -0.017 |
who | -0.015 |
crying | -0.014 |
blood | -0.014 |
weeping | -0.013 |
Token | she |
Token ID | 673 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
ops | +0.037 |
iest | +0.037 |
affles | +0.036 |
hardly | +0.035 |
gh | +0.035 |
Neg logit contributions | |
nowhere | -0.019 |
who | -0.017 |
crying | -0.014 |
Blue | -0.014 |
blood | -0.013 |
Token | didn |
Token ID | 1422 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
iest | +0.036 |
ops | +0.036 |
affles | +0.035 |
robat | +0.033 |
rd | +0.033 |
Neg logit contributions | |
nowhere | -0.036 |
who | -0.033 |
crying | -0.028 |
blood | -0.028 |
Blue | -0.027 |
Token | 't |
Token ID | 470 |
Feature activation | +5.73e-03 |
Pos logit contributions | |
iest | +0.040 |
ops | +0.040 |
hardly | +0.038 |
affles | +0.038 |
gh | +0.037 |
Neg logit contributions | |
nowhere | -0.043 |
who | -0.038 |
blood | -0.035 |
crying | -0.035 |
Blue | -0.034 |
Token | want |
Token ID | 765 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
ops | +0.010 |
affles | +0.010 |
iest | +0.009 |
gh | +0.009 |
hardly | +0.009 |
Neg logit contributions | |
nowhere | -0.011 |
who | -0.011 |
crying | -0.010 |
blood | -0.010 |
Blue | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
nowhere | +0.018 |
who | +0.014 |
crying | +0.012 |
blood | +0.012 |
Blue | +0.011 |
Neg logit contributions | |
iest | -0.043 |
hardly | -0.042 |
ops | -0.042 |
gh | -0.040 |
affles | -0.040 |
Token | sp |
Token ID | 599 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
ops | +0.031 |
iest | +0.031 |
affles | +0.031 |
robat | +0.029 |
rd | +0.029 |
Neg logit contributions | |
who | -0.024 |
nowhere | -0.023 |
Blue | -0.021 |
crying | -0.021 |
blood | -0.020 |
Token | ook |
Token ID | 566 |
Feature activation | -3.00e-03 |
Pos logit contributions | |
ops | +0.159 |
iest | +0.154 |
hardly | +0.150 |
robat | +0.146 |
rd | +0.145 |
Neg logit contributions | |
nowhere | -0.235 |
who | -0.214 |
blood | -0.201 |
crying | -0.195 |
Blue | -0.185 |
Token | it |
Token ID | 340 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
nowhere | +0.004 |
who | +0.003 |
crying | +0.003 |
blood | +0.003 |
Blue | +0.002 |
Neg logit contributions | |
iest | -0.007 |
ops | -0.007 |
hardly | -0.007 |
affles | -0.007 |
gh | -0.007 |
Token | away |
Token ID | 1497 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
nowhere | +0.040 |
who | +0.033 |
blood | +0.032 |
Blue | +0.030 |
crying | +0.029 |
Neg logit contributions | |
iest | -0.066 |
ops | -0.066 |
gh | -0.063 |
hardly | -0.063 |
zed | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +3.84e-03 |
Pos logit contributions | |
nowhere | +0.027 |
who | +0.024 |
blood | +0.021 |
crying | +0.021 |
Blue | +0.020 |
Neg logit contributions | |
iest | -0.043 |
ops | -0.043 |
hardly | -0.042 |
affles | -0.041 |
gh | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | +7.61e-04 |
Pos logit contributions | |
hardly | +0.008 |
ops | +0.008 |
iest | +0.008 |
zed | +0.008 |
rd | +0.008 |
Neg logit contributions | |
nowhere | -0.006 |
blood | -0.005 |
who | -0.005 |
crying | -0.004 |
weeping | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
iest | +0.046 |
ops | +0.046 |
hardly | +0.044 |
affles | +0.044 |
gh | +0.042 |
Neg logit contributions | |
nowhere | -0.043 |
who | -0.040 |
crying | -0.034 |
blood | -0.034 |
Blue | -0.034 |
Token | very |
Token ID | 845 |
Feature activation | +9.51e-03 |
Pos logit contributions | |
nowhere | +0.018 |
bug | +0.014 |
blood | +0.014 |
who | +0.014 |
Blue | +0.013 |
Neg logit contributions | |
hardly | -0.025 |
iest | -0.023 |
gh | -0.022 |
working | -0.021 |
zed | -0.021 |
Token | modest |
Token ID | 12949 |
Feature activation | -4.04e-03 |
Pos logit contributions | |
ops | +0.023 |
iest | +0.022 |
affles | +0.021 |
robat | +0.021 |
hardly | +0.021 |
Neg logit contributions | |
nowhere | -0.012 |
who | -0.011 |
crying | -0.010 |
blood | -0.009 |
Blue | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -8.94e-03 |
Pos logit contributions | |
nowhere | +0.005 |
who | +0.004 |
blood | +0.004 |
Blue | +0.004 |
crying | +0.004 |
Neg logit contributions | |
ops | -0.009 |
iest | -0.009 |
hardly | -0.009 |
affles | -0.009 |
rd | -0.009 |
Token | never |
Token ID | 1239 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
who | +0.013 |
nowhere | +0.012 |
a | +0.011 |
Blue | +0.010 |
where | +0.010 |
Neg logit contributions | |
ops | -0.021 |
affles | -0.020 |
iest | -0.020 |
robat | -0.019 |
rd | -0.019 |
Token | br |
Token ID | 865 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
nowhere | +0.040 |
who | +0.036 |
blood | +0.033 |
crying | +0.032 |
Blue | +0.032 |
Neg logit contributions | |
ops | -0.039 |
iest | -0.039 |
hardly | -0.037 |
affles | -0.037 |
gh | -0.036 |
Token | agged |
Token ID | 14655 |
Feature activation | +6.52e-03 |
Pos logit contributions | |
iest | +0.174 |
hardly | +0.174 |
ops | +0.173 |
rd | +0.158 |
robat | +0.157 |
Neg logit contributions | |
nowhere | -0.220 |
who | -0.197 |
blood | -0.190 |
crying | -0.179 |
weeping | -0.169 |
Token | about |
Token ID | 546 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
ops | +0.014 |
iest | +0.014 |
affles | +0.014 |
hardly | +0.014 |
gh | +0.013 |
Neg logit contributions | |
nowhere | -0.010 |
who | -0.009 |
blood | -0.008 |
crying | -0.007 |
Blue | -0.007 |
Token | his |
Token ID | 465 |
Feature activation | +5.32e-03 |
Pos logit contributions | |
nowhere | +0.027 |
bug | +0.023 |
who | +0.020 |
Blue | +0.020 |
blood | +0.019 |
Neg logit contributions | |
hardly | -0.037 |
iest | -0.036 |
ops | -0.036 |
gh | -0.035 |
affles | -0.035 |
Token | work |
Token ID | 670 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
ops | +0.010 |
iest | +0.010 |
hardly | +0.010 |
affles | +0.010 |
gh | +0.010 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.008 |
bug | -0.007 |
blood | -0.007 |
Blue | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
ops | +0.042 |
iest | +0.042 |
affles | +0.041 |
hardly | +0.041 |
gh | +0.039 |
Neg logit contributions | |
nowhere | -0.024 |
who | -0.021 |
crying | -0.018 |
blood | -0.017 |
Blue | -0.017 |
Token | looked |
Token ID | 3114 |
Feature activation | +6.74e-03 |
Pos logit contributions | |
hardly | +0.011 |
ops | +0.011 |
iest | +0.011 |
affles | +0.010 |
zed | +0.010 |
Neg logit contributions | |
nowhere | -0.011 |
who | -0.010 |
blood | -0.010 |
crying | -0.009 |
Blue | -0.009 |
Token | for |
Token ID | 329 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
iest | +0.016 |
ops | +0.016 |
hardly | +0.016 |
affles | +0.015 |
gh | +0.015 |
Neg logit contributions | |
nowhere | -0.008 |
who | -0.007 |
blood | -0.006 |
crying | -0.006 |
Blue | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
nowhere | +0.035 |
who | +0.029 |
crying | +0.025 |
blood | +0.024 |
Blue | +0.024 |
Neg logit contributions | |
iest | -0.079 |
hardly | -0.076 |
ops | -0.076 |
affles | -0.073 |
gh | -0.073 |
Token | biggest |
Token ID | 4094 |
Feature activation | -7.36e-03 |
Pos logit contributions | |
nowhere | +0.038 |
who | +0.034 |
blood | +0.030 |
crying | +0.030 |
Blue | +0.029 |
Neg logit contributions | |
iest | -0.048 |
ops | -0.045 |
hardly | -0.045 |
gh | -0.043 |
affles | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -3.67e-04 |
Pos logit contributions | |
nowhere | +0.012 |
who | +0.010 |
blood | +0.009 |
crying | +0.009 |
Blue | +0.009 |
Neg logit contributions | |
iest | -0.015 |
ops | -0.015 |
hardly | -0.015 |
affles | -0.014 |
gh | -0.014 |
Token | funn |
Token ID | 36090 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
nowhere | +0.001 |
who | +0.001 |
blood | +0.000 |
crying | +0.000 |
Blue | +0.000 |
Neg logit contributions | |
iest | -0.001 |
ops | -0.001 |
hardly | -0.001 |
affles | -0.001 |
gh | -0.001 |
Token | iest |
Token ID | 6386 |
Feature activation | -6.84e-04 |
Pos logit contributions | |
hardly | +0.081 |
vows | +0.056 |
rd | +0.055 |
robat | +0.054 |
ops | +0.052 |
Neg logit contributions | |
who | -0.300 |
nowhere | -0.296 |
blood | -0.277 |
a | -0.275 |
ter | -0.272 |
Token | pink |
Token ID | 11398 |
Feature activation | -6.96e-03 |
Pos logit contributions | |
nowhere | +0.001 |
who | +0.001 |
blood | +0.001 |
crying | +0.001 |
Blue | +0.001 |
Neg logit contributions | |
ops | -0.001 |
iest | -0.001 |
hardly | -0.001 |
affles | -0.001 |
gh | -0.001 |
Token | balloon |
Token ID | 21190 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
nowhere | +0.011 |
who | +0.009 |
blood | +0.008 |
crying | +0.008 |
Blue | +0.008 |
Neg logit contributions | |
ops | -0.015 |
iest | -0.015 |
hardly | -0.015 |
affles | -0.015 |
gh | -0.014 |
Token | that |
Token ID | 326 |
Feature activation | +7.51e-03 |
Pos logit contributions | |
nowhere | +0.035 |
who | +0.032 |
Blue | +0.027 |
crying | +0.026 |
blood | +0.026 |
Neg logit contributions | |
affles | -0.058 |
ops | -0.057 |
iest | -0.057 |
hardly | -0.057 |
zed | -0.054 |
Token | he |
Token ID | 339 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
affles | +0.014 |
ops | +0.014 |
iest | +0.014 |
hardly | +0.013 |
rd | +0.013 |
Neg logit contributions | |
nowhere | -0.014 |
who | -0.012 |
Blue | -0.011 |
crying | -0.011 |
blood | -0.011 |
Token | with |
Token ID | 351 |
Feature activation | -1.43e-03 |
Pos logit contributions | |
ops | +0.052 |
iest | +0.052 |
hardly | +0.051 |
affles | +0.051 |
gh | +0.050 |
Neg logit contributions | |
nowhere | -0.021 |
who | -0.018 |
crying | -0.014 |
blood | -0.014 |
Blue | -0.013 |
Token | fries |
Token ID | 31757 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
nowhere | +0.002 |
who | +0.002 |
crying | +0.001 |
blood | +0.001 |
Blue | +0.001 |
Neg logit contributions | |
iest | -0.003 |
ops | -0.003 |
hardly | -0.003 |
affles | -0.003 |
gh | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +4.47e-04 |
Pos logit contributions | |
iest | +0.133 |
affles | +0.131 |
ops | +0.130 |
hardly | +0.130 |
rd | +0.130 |
Neg logit contributions | |
nowhere | -0.047 |
who | -0.046 |
crying | -0.032 |
blood | -0.032 |
Blue | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
iest | +0.001 |
ops | +0.001 |
robat | +0.001 |
gh | +0.001 |
rd | +0.001 |
Neg logit contributions | |
nowhere | -0.001 |
who | -0.001 |
a | -0.001 |
blood | -0.001 |
Blue | -0.001 |
Token | mil |
Token ID | 1465 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
ops | +0.032 |
iest | +0.031 |
affles | +0.030 |
zed | +0.030 |
robat | +0.030 |
Neg logit contributions | |
nowhere | -0.021 |
who | -0.021 |
blood | -0.018 |
crying | -0.018 |
a | -0.016 |
Token | ksh |
Token ID | 50133 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
ops | +0.022 |
iest | +0.021 |
affles | +0.021 |
hardly | +0.021 |
gh | +0.020 |
Neg logit contributions | |
nowhere | -0.021 |
who | -0.018 |
blood | -0.017 |
crying | -0.017 |
Blue | -0.016 |
Token | ake |
Token ID | 539 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
iest | +0.200 |
hardly | +0.197 |
ves | +0.182 |
zed | +0.179 |
ops | +0.179 |
Neg logit contributions | |
nowhere | -0.181 |
who | -0.165 |
blood | -0.146 |
crying | -0.144 |
Blue | -0.144 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ops | +0.063 |
iest | +0.063 |
hardly | +0.062 |
affles | +0.061 |
gh | +0.059 |
Neg logit contributions | |
nowhere | -0.036 |
who | -0.032 |
crying | -0.027 |
blood | -0.027 |
Blue | -0.026 |
Token | But |
Token ID | 887 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
ops | +0.021 |
hardly | +0.021 |
affles | +0.019 |
zed | +0.018 |
vows | +0.018 |
Neg logit contributions | |
nowhere | -0.018 |
who | -0.014 |
blood | -0.013 |
crying | -0.013 |
weeping | -0.012 |
Token | after |
Token ID | 706 |
Feature activation | +5.49e-03 |
Pos logit contributions | |
ops | +0.056 |
iest | +0.056 |
hardly | +0.055 |
affles | +0.053 |
gh | +0.052 |
Neg logit contributions | |
nowhere | -0.035 |
who | -0.030 |
blood | -0.027 |
crying | -0.026 |
Blue | -0.025 |
Token | eating |
Token ID | 6600 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
ops | +0.013 |
hardly | +0.013 |
iest | +0.013 |
affles | +0.012 |
gh | +0.012 |
Neg logit contributions | |
nowhere | -0.007 |
who | -0.006 |
blood | -0.005 |
crying | -0.005 |
Blue | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
nowhere | +0.001 |
blood | +0.001 |
who | +0.001 |
Blue | +0.001 |
crying | +0.001 |
Neg logit contributions | |
ops | -0.001 |
iest | -0.001 |
hardly | -0.001 |
affles | -0.001 |
rd | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | +6.35e-03 |
Pos logit contributions | |
nowhere | +0.021 |
who | +0.018 |
Blue | +0.016 |
crying | +0.016 |
blood | +0.015 |
Neg logit contributions | |
iest | -0.036 |
ops | -0.035 |
affles | -0.034 |
hardly | -0.034 |
gh | -0.033 |
Token | was |
Token ID | 373 |
Feature activation | -7.86e-03 |
Pos logit contributions | |
iest | +0.015 |
ops | +0.014 |
affles | +0.014 |
robat | +0.013 |
vation | +0.013 |
Neg logit contributions | |
nowhere | -0.010 |
who | -0.009 |
crying | -0.008 |
Blue | -0.007 |
a | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
nowhere | +0.012 |
blood | +0.010 |
tighter | +0.009 |
who | +0.009 |
itself | +0.009 |
Neg logit contributions | |
hardly | -0.017 |
iest | -0.014 |
affles | -0.014 |
barely | -0.014 |
zed | -0.013 |
Token | happiest |
Token ID | 49414 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
nowhere | +0.106 |
who | +0.093 |
blood | +0.086 |
crying | +0.084 |
Blue | +0.083 |
Neg logit contributions | |
iest | -0.099 |
ops | -0.097 |
hardly | -0.095 |
affles | -0.093 |
gh | -0.090 |
Token | little |
Token ID | 1310 |
Feature activation | -8.24e-02 |
Pos logit contributions | |
nowhere | +0.066 |
who | +0.056 |
blood | +0.051 |
crying | +0.050 |
Blue | +0.049 |
Neg logit contributions | |
iest | -0.094 |
ops | -0.092 |
hardly | -0.090 |
affles | -0.088 |
gh | -0.086 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
nowhere | +0.106 |
who | +0.084 |
blood | +0.075 |
Blue | +0.070 |
crying | +0.070 |
Neg logit contributions | |
iest | -0.201 |
ops | -0.192 |
hardly | -0.186 |
affles | -0.182 |
robat | -0.179 |
Token | around |
Token ID | 1088 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
nowhere | +0.039 |
who | +0.033 |
crying | +0.024 |
blood | +0.023 |
Blue | +0.023 |
Neg logit contributions | |
ops | -0.111 |
iest | -0.110 |
hardly | -0.108 |
affles | -0.108 |
gh | -0.105 |
Token | ! |
Token ID | 0 |
Feature activation | -7.80e-03 |
Pos logit contributions | |
nowhere | +0.033 |
who | +0.027 |
blood | +0.027 |
Blue | +0.024 |
bug | +0.024 |
Neg logit contributions | |
iest | -0.069 |
ops | -0.068 |
hardly | -0.068 |
gh | -0.065 |
zed | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -5.32e-03 |
Pos logit contributions | |
nowhere | +0.011 |
who | +0.010 |
crying | +0.008 |
blood | +0.008 |
Blue | +0.008 |
Neg logit contributions | |
iest | -0.017 |
ops | -0.017 |
hardly | -0.017 |
affles | -0.017 |
gh | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
nowhere | +0.006 |
who | +0.005 |
blood | +0.004 |
crying | +0.004 |
Blue | +0.004 |
Neg logit contributions | |
ops | -0.013 |
iest | -0.013 |
hardly | -0.013 |
affles | -0.013 |
gh | -0.013 |
Token | ! |
Token ID | 0 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
nowhere | +0.012 |
who | +0.010 |
blood | +0.009 |
crying | +0.008 |
Blue | +0.008 |
Neg logit contributions | |
iest | -0.039 |
ops | -0.039 |
hardly | -0.038 |
gh | -0.037 |
zed | -0.036 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
nowhere | +0.015 |
who | +0.013 |
blood | +0.013 |
crying | +0.012 |
ter | +0.012 |
Neg logit contributions | |
hardly | -0.022 |
ops | -0.021 |
zed | -0.020 |
iest | -0.020 |
rd | -0.019 |
Token | John |
Token ID | 7554 |
Feature activation | +5.09e-03 |
Pos logit contributions | |
nowhere | +0.063 |
who | +0.056 |
blood | +0.056 |
crying | +0.053 |
weeping | +0.052 |
Neg logit contributions | |
ops | -0.096 |
iest | -0.093 |
gh | -0.087 |
hardly | -0.086 |
robat | -0.086 |
Token | was |
Token ID | 373 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
iest | +0.012 |
ops | +0.012 |
hardly | +0.012 |
affles | +0.012 |
gh | +0.011 |
Neg logit contributions | |
nowhere | -0.007 |
who | -0.006 |
crying | -0.005 |
blood | -0.005 |
Blue | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
nowhere | +0.039 |
blood | +0.037 |
who | +0.034 |
ety | +0.034 |
Blue | +0.033 |
Neg logit contributions | |
hardly | -0.060 |
zed | -0.054 |
gh | -0.053 |
iest | -0.052 |
affles | -0.051 |
Token | little |
Token ID | 1310 |
Feature activation | -8.24e-02 |
Pos logit contributions | |
nowhere | +0.082 |
blood | +0.070 |
who | +0.064 |
crying | +0.061 |
Blue | +0.060 |
Neg logit contributions | |
hardly | -0.115 |
iest | -0.111 |
ops | -0.108 |
affles | -0.105 |
gh | -0.104 |
Token | boy |
Token ID | 2933 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
nowhere | +0.135 |
blood | +0.111 |
itself | +0.099 |
half | +0.090 |
Blue | +0.088 |
Neg logit contributions | |
iest | -0.157 |
working | -0.148 |
ops | -0.141 |
hardly | -0.140 |
affles | -0.139 |
Token | and |
Token ID | 290 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
nowhere | +0.035 |
blood | +0.030 |
Blue | +0.024 |
crying | +0.023 |
weeping | +0.022 |
Neg logit contributions | |
ops | -0.057 |
iest | -0.057 |
hardly | -0.055 |
rd | -0.054 |
affles | -0.053 |
Token | he |
Token ID | 339 |
Feature activation | -3.56e-03 |
Pos logit contributions | |
who | +0.046 |
nowhere | +0.045 |
a | +0.039 |
crying | +0.038 |
Blue | +0.037 |
Neg logit contributions | |
ops | -0.060 |
affles | -0.055 |
robat | -0.055 |
iest | -0.055 |
gh | -0.052 |
Token | was |
Token ID | 373 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
nowhere | +0.006 |
who | +0.005 |
crying | +0.005 |
Blue | +0.004 |
a | +0.004 |
Neg logit contributions | |
iest | -0.008 |
ops | -0.008 |
affles | -0.007 |
robat | -0.007 |
ves | -0.007 |
Token | worried |
Token ID | 7960 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
nowhere | +0.024 |
blood | +0.021 |
who | +0.018 |
tighter | +0.017 |
ter | +0.017 |
Neg logit contributions | |
hardly | -0.034 |
zed | -0.031 |
iest | -0.031 |
gh | -0.031 |
rd | -0.030 |
Token | ever |
Token ID | 1683 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
nowhere | +0.049 |
who | +0.045 |
crying | +0.040 |
blood | +0.039 |
Blue | +0.039 |
Neg logit contributions | |
ops | -0.050 |
iest | -0.049 |
hardly | -0.048 |
affles | -0.048 |
gh | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -7.98e-03 |
Pos logit contributions | |
nowhere | +0.063 |
who | +0.057 |
crying | +0.049 |
Blue | +0.047 |
blood | +0.046 |
Neg logit contributions | |
iest | -0.092 |
affles | -0.091 |
ops | -0.091 |
hardly | -0.090 |
gh | -0.088 |
Token | was |
Token ID | 373 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
nowhere | +0.012 |
who | +0.011 |
Blue | +0.010 |
crying | +0.010 |
a | +0.009 |
Neg logit contributions | |
iest | -0.017 |
affles | -0.017 |
ops | -0.017 |
robat | -0.017 |
gh | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
nowhere | +0.014 |
who | +0.012 |
blood | +0.011 |
crying | +0.011 |
Blue | +0.010 |
Neg logit contributions | |
iest | -0.025 |
ops | -0.025 |
hardly | -0.024 |
affles | -0.024 |
gh | -0.023 |
Token | happiest |
Token ID | 49414 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
nowhere | +0.104 |
who | +0.094 |
blood | +0.086 |
crying | +0.086 |
Blue | +0.084 |
Neg logit contributions | |
iest | -0.084 |
ops | -0.083 |
hardly | -0.080 |
affles | -0.079 |
gh | -0.076 |
Token | little |
Token ID | 1310 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
nowhere | +0.060 |
who | +0.052 |
blood | +0.046 |
crying | +0.046 |
Blue | +0.045 |
Neg logit contributions | |
iest | -0.087 |
ops | -0.086 |
hardly | -0.084 |
affles | -0.083 |
gh | -0.080 |
Token | girl |
Token ID | 2576 |
Feature activation | -5.38e-02 |
Pos logit contributions | |
nowhere | +0.124 |
who | +0.109 |
blood | +0.098 |
crying | +0.095 |
Blue | +0.093 |
Neg logit contributions | |
iest | -0.173 |
ops | -0.167 |
hardly | -0.164 |
affles | -0.158 |
gh | -0.156 |
Token | . |
Token ID | 13 |
Feature activation | -6.80e-03 |
Pos logit contributions | |
nowhere | +0.078 |
who | +0.069 |
crying | +0.058 |
blood | +0.057 |
Blue | +0.057 |
Neg logit contributions | |
ops | -0.120 |
iest | -0.119 |
affles | -0.116 |
hardly | -0.116 |
gh | -0.112 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
nowhere | +0.009 |
blood | +0.008 |
who | +0.007 |
crying | +0.007 |
ter | +0.007 |
Neg logit contributions | |
hardly | -0.015 |
ops | -0.015 |
rd | -0.014 |
iest | -0.014 |
zed | -0.014 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
nowhere | +0.066 |
who | +0.058 |
blood | +0.057 |
crying | +0.054 |
weeping | +0.053 |
Neg logit contributions | |
ops | -0.086 |
iest | -0.085 |
robat | -0.077 |
hardly | -0.077 |
rd | -0.076 |
Token | upon |
Token ID | 2402 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
ops | +0.081 |
iest | +0.080 |
hardly | +0.078 |
affles | +0.078 |
gh | +0.076 |
Neg logit contributions | |
nowhere | -0.047 |
who | -0.040 |
crying | -0.035 |
blood | -0.034 |
Blue | -0.033 |
Token | my |
Token ID | 616 |
Feature activation | -1.55e-03 |
Pos logit contributions | |
nowhere | +0.039 |
who | +0.032 |
crying | +0.029 |
blood | +0.029 |
Blue | +0.029 |
Neg logit contributions | |
iest | -0.054 |
ops | -0.052 |
hardly | -0.051 |
affles | -0.051 |
rd | -0.048 |
Token | loves |
Token ID | 10408 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
nowhere | +0.003 |
who | +0.003 |
blood | +0.002 |
crying | +0.002 |
Blue | +0.002 |
Neg logit contributions | |
ops | -0.003 |
iest | -0.003 |
hardly | -0.003 |
affles | -0.003 |
gh | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -7.81e-03 |
Pos logit contributions | |
nowhere | +0.035 |
who | +0.030 |
blood | +0.028 |
crying | +0.027 |
Blue | +0.027 |
Neg logit contributions | |
iest | -0.046 |
ops | -0.046 |
affles | -0.044 |
hardly | -0.043 |
gh | -0.043 |
Token | I |
Token ID | 314 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
nowhere | +0.012 |
who | +0.010 |
blood | +0.010 |
crying | +0.009 |
Blue | +0.009 |
Neg logit contributions | |
iest | -0.016 |
ops | -0.016 |
hardly | -0.016 |
affles | -0.015 |
gh | -0.015 |
Token | am |
Token ID | 716 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
nowhere | +0.035 |
who | +0.033 |
crying | +0.029 |
blood | +0.029 |
Blue | +0.028 |
Neg logit contributions | |
ops | -0.029 |
iest | -0.029 |
affles | -0.028 |
hardly | -0.027 |
gh | -0.027 |
Token | back |
Token ID | 736 |
Feature activation | -8.03e-02 |
Pos logit contributions | |
hardly | +0.040 |
iest | +0.037 |
affles | +0.037 |
gh | +0.037 |
zed | +0.036 |
Neg logit contributions | |
nowhere | -0.021 |
blood | -0.017 |
who | -0.016 |
tighter | -0.014 |
ety | -0.014 |
Token | from |
Token ID | 422 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
nowhere | +0.100 |
who | +0.080 |
blood | +0.074 |
crying | +0.069 |
Blue | +0.067 |
Neg logit contributions | |
iest | -0.193 |
ops | -0.190 |
hardly | -0.187 |
affles | -0.184 |
gh | -0.180 |
Token | the |
Token ID | 262 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
nowhere | +0.042 |
who | +0.036 |
blood | +0.033 |
bug | +0.032 |
Blue | +0.031 |
Neg logit contributions | |
hardly | -0.062 |
iest | -0.061 |
affles | -0.060 |
zed | -0.059 |
gh | -0.058 |
Token | store |
Token ID | 3650 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
nowhere | +0.030 |
who | +0.027 |
bug | +0.024 |
Blue | +0.024 |
blood | +0.023 |
Neg logit contributions | |
iest | -0.030 |
ops | -0.030 |
hardly | -0.029 |
affles | -0.028 |
gh | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -4.05e-03 |
Pos logit contributions | |
nowhere | +0.051 |
who | +0.041 |
blood | +0.041 |
crying | +0.039 |
Blue | +0.038 |
Neg logit contributions | |
iest | -0.063 |
ops | -0.063 |
hardly | -0.059 |
rd | -0.058 |
affles | -0.058 |
Token | I |
Token ID | 314 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
nowhere | +0.006 |
who | +0.005 |
blood | +0.005 |
crying | +0.005 |
Blue | +0.005 |
Neg logit contributions | |
ops | -0.009 |
iest | -0.009 |
hardly | -0.008 |
affles | -0.008 |
gh | -0.008 |
Token | The |
Token ID | 383 |
Feature activation | -5.29e-03 |
Pos logit contributions | |
ops | +0.015 |
hardly | +0.015 |
iest | +0.014 |
zed | +0.014 |
affles | +0.014 |
Neg logit contributions | |
nowhere | -0.015 |
who | -0.013 |
blood | -0.012 |
crying | -0.012 |
bug | -0.011 |
Token | bug |
Token ID | 5434 |
Feature activation | -8.68e-03 |
Pos logit contributions | |
nowhere | +0.010 |
who | +0.008 |
blood | +0.008 |
crying | +0.008 |
Blue | +0.008 |
Neg logit contributions | |
iest | -0.010 |
ops | -0.010 |
hardly | -0.010 |
affles | -0.009 |
ves | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -1.44e-03 |
Pos logit contributions | |
nowhere | +0.015 |
who | +0.013 |
crying | +0.011 |
Blue | +0.011 |
bug | +0.010 |
Neg logit contributions | |
iest | -0.018 |
affles | -0.018 |
ops | -0.018 |
gh | -0.017 |
hardly | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
nowhere | +0.002 |
who | +0.002 |
blood | +0.002 |
Blue | +0.001 |
crying | +0.001 |
Neg logit contributions | |
hardly | -0.003 |
iest | -0.003 |
gh | -0.003 |
zed | -0.003 |
ops | -0.003 |
Token | happiest |
Token ID | 49414 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
nowhere | +0.100 |
who | +0.090 |
blood | +0.083 |
crying | +0.083 |
Blue | +0.081 |
Neg logit contributions | |
iest | -0.074 |
ops | -0.073 |
hardly | -0.072 |
affles | -0.068 |
gh | -0.066 |
Token | little |
Token ID | 1310 |
Feature activation | -7.99e-02 |
Pos logit contributions | |
nowhere | +0.064 |
who | +0.056 |
blood | +0.050 |
crying | +0.050 |
Blue | +0.049 |
Neg logit contributions | |
iest | -0.085 |
ops | -0.083 |
hardly | -0.081 |
affles | -0.078 |
gh | -0.077 |
Token | bug |
Token ID | 5434 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
nowhere | +0.128 |
who | +0.114 |
blood | +0.101 |
crying | +0.100 |
Blue | +0.097 |
Neg logit contributions | |
iest | -0.165 |
ops | -0.155 |
hardly | -0.152 |
gh | -0.143 |
robat | -0.143 |
Token | who |
Token ID | 508 |
Feature activation | -7.03e-04 |
Pos logit contributions | |
nowhere | +0.056 |
who | +0.049 |
crying | +0.040 |
Blue | +0.040 |
blood | +0.039 |
Neg logit contributions | |
ops | -0.099 |
iest | -0.098 |
affles | -0.097 |
hardly | -0.096 |
gh | -0.093 |
Token | ever |
Token ID | 1683 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
nowhere | +0.001 |
who | +0.001 |
crying | +0.001 |
Blue | +0.001 |
blood | +0.001 |
Neg logit contributions | |
ops | -0.001 |
iest | -0.001 |
affles | -0.001 |
hardly | -0.001 |
gh | -0.001 |
Token | lived |
Token ID | 5615 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
nowhere | +0.066 |
who | +0.059 |
crying | +0.051 |
Blue | +0.050 |
blood | +0.049 |
Neg logit contributions | |
affles | -0.076 |
hardly | -0.074 |
iest | -0.074 |
ops | -0.074 |
gh | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
nowhere | +0.066 |
blood | +0.058 |
who | +0.056 |
crying | +0.050 |
Blue | +0.050 |
Neg logit contributions | |
iest | -0.152 |
gh | -0.144 |
ops | -0.144 |
hardly | -0.144 |
zed | -0.139 |
Token | . |
Token ID | 13 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
nowhere | +0.009 |
blood | +0.007 |
who | +0.007 |
Blue | +0.007 |
crying | +0.007 |
Neg logit contributions | |
iest | -0.011 |
ops | -0.011 |
hardly | -0.010 |
gh | -0.010 |
rd | -0.010 |
Token | She |
Token ID | 1375 |
Feature activation | +5.60e-03 |
Pos logit contributions | |
nowhere | +0.023 |
who | +0.020 |
blood | +0.019 |
crying | +0.017 |
ter | +0.017 |
Neg logit contributions | |
ops | -0.033 |
hardly | -0.033 |
iest | -0.033 |
rd | -0.032 |
affles | -0.031 |
Token | was |
Token ID | 373 |
Feature activation | -8.86e-03 |
Pos logit contributions | |
iest | +0.012 |
ops | +0.012 |
affles | +0.012 |
robat | +0.011 |
hardly | +0.011 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.008 |
crying | -0.007 |
Blue | -0.007 |
a | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
nowhere | +0.014 |
blood | +0.012 |
who | +0.011 |
again | +0.011 |
tighter | +0.010 |
Neg logit contributions | |
hardly | -0.018 |
iest | -0.015 |
affles | -0.015 |
barely | -0.015 |
zed | -0.015 |
Token | happiest |
Token ID | 49414 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
nowhere | +0.102 |
who | +0.089 |
blood | +0.082 |
crying | +0.080 |
Blue | +0.079 |
Neg logit contributions | |
iest | -0.092 |
ops | -0.089 |
hardly | -0.089 |
affles | -0.085 |
gh | -0.083 |
Token | little |
Token ID | 1310 |
Feature activation | -7.97e-02 |
Pos logit contributions | |
nowhere | +0.058 |
who | +0.046 |
blood | +0.043 |
crying | +0.040 |
Blue | +0.040 |
Neg logit contributions | |
iest | -0.099 |
ops | -0.090 |
hardly | -0.089 |
affles | -0.084 |
rd | -0.084 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
nowhere | +0.111 |
who | +0.089 |
blood | +0.080 |
Blue | +0.076 |
crying | +0.073 |
Neg logit contributions | |
iest | -0.186 |
ops | -0.175 |
hardly | -0.166 |
robat | -0.163 |
affles | -0.163 |
Token | in |
Token ID | 287 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
nowhere | +0.043 |
who | +0.034 |
blood | +0.029 |
crying | +0.028 |
Blue | +0.027 |
Neg logit contributions | |
iest | -0.111 |
ops | -0.110 |
hardly | -0.107 |
affles | -0.106 |
gh | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
nowhere | +0.047 |
who | +0.041 |
blood | +0.037 |
bug | +0.036 |
tighter | +0.036 |
Neg logit contributions | |
hardly | -0.082 |
iest | -0.081 |
ops | -0.079 |
zed | -0.077 |
affles | -0.076 |
Token | world |
Token ID | 995 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
nowhere | +0.026 |
who | +0.023 |
blood | +0.022 |
tighter | +0.020 |
Blue | +0.020 |
Neg logit contributions | |
iest | -0.020 |
ops | -0.019 |
rd | -0.019 |
gh | -0.019 |
hardly | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
nowhere | +0.022 |
who | +0.019 |
crying | +0.016 |
blood | +0.016 |
Blue | +0.016 |
Neg logit contributions | |
ops | -0.036 |
iest | -0.036 |
hardly | -0.035 |
affles | -0.035 |
gh | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +1.12e-04 |
Pos logit contributions | |
nowhere | +0.023 |
who | +0.021 |
crying | +0.018 |
blood | +0.018 |
Blue | +0.018 |
Neg logit contributions | |
ops | -0.026 |
iest | -0.026 |
hardly | -0.025 |
affles | -0.025 |
gh | -0.024 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
hardly | +0.000 |
ops | +0.000 |
zed | +0.000 |
iest | +0.000 |
rd | +0.000 |
Neg logit contributions | |
nowhere | -0.000 |
blood | -0.000 |
crying | -0.000 |
who | -0.000 |
weeping | -0.000 |
Token | Jimmy |
Token ID | 40335 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
nowhere | +0.069 |
who | +0.061 |
blood | +0.058 |
crying | +0.056 |
weeping | +0.054 |
Neg logit contributions | |
ops | -0.093 |
iest | -0.092 |
robat | -0.085 |
hardly | -0.085 |
gh | -0.084 |
Token | was |
Token ID | 373 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
ops | +0.042 |
iest | +0.042 |
hardly | +0.041 |
affles | +0.040 |
gh | +0.039 |
Neg logit contributions | |
nowhere | -0.030 |
who | -0.026 |
blood | -0.023 |
crying | -0.023 |
Blue | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
nowhere | +0.034 |
blood | +0.030 |
who | +0.029 |
Blue | +0.028 |
ety | +0.026 |
Neg logit contributions | |
hardly | -0.053 |
iest | -0.049 |
gh | -0.049 |
zed | -0.049 |
affles | -0.048 |
Token | little |
Token ID | 1310 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
nowhere | +0.076 |
who | +0.061 |
blood | +0.061 |
crying | +0.055 |
Blue | +0.055 |
Neg logit contributions | |
iest | -0.115 |
hardly | -0.114 |
ops | -0.112 |
affles | -0.109 |
gh | -0.107 |
Token | boy |
Token ID | 2933 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
nowhere | +0.121 |
blood | +0.095 |
who | +0.081 |
Blue | +0.081 |
itself | +0.077 |
Neg logit contributions | |
iest | -0.168 |
ops | -0.151 |
affles | -0.149 |
hardly | -0.148 |
working | -0.146 |
Token | who |
Token ID | 508 |
Feature activation | +3.69e-03 |
Pos logit contributions | |
nowhere | +0.030 |
blood | +0.024 |
Blue | +0.021 |
crying | +0.020 |
who | +0.019 |
Neg logit contributions | |
iest | -0.048 |
ops | -0.048 |
hardly | -0.046 |
rd | -0.046 |
affles | -0.045 |
Token | was |
Token ID | 373 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
iest | +0.009 |
ops | +0.009 |
affles | +0.008 |
ves | +0.008 |
robat | +0.008 |
Neg logit contributions | |
who | -0.005 |
nowhere | -0.005 |
crying | -0.004 |
a | -0.004 |
weeping | -0.004 |
Token | three |
Token ID | 1115 |
Feature activation | -3.92e-03 |
Pos logit contributions | |
nowhere | +0.032 |
blood | +0.027 |
Blue | +0.024 |
who | +0.023 |
tighter | +0.023 |
Neg logit contributions | |
hardly | -0.045 |
iest | -0.041 |
gh | -0.041 |
zed | -0.040 |
rd | -0.040 |
Token | years |
Token ID | 812 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
who | +0.006 |
nowhere | +0.006 |
crying | +0.005 |
blood | +0.004 |
too | +0.004 |
Neg logit contributions | |
ops | -0.008 |
hardly | -0.008 |
iest | -0.008 |
gh | -0.008 |
robat | -0.008 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
hardly | +0.008 |
ops | +0.008 |
iest | +0.008 |
zed | +0.008 |
rd | +0.008 |
Neg logit contributions | |
nowhere | -0.005 |
who | -0.004 |
blood | -0.004 |
crying | -0.003 |
ter | -0.003 |
Token | Tom |
Token ID | 13787 |
Feature activation | +8.94e-03 |
Pos logit contributions | |
nowhere | +0.074 |
who | +0.067 |
blood | +0.065 |
crying | +0.064 |
weeping | +0.063 |
Neg logit contributions | |
ops | -0.082 |
iest | -0.080 |
robat | -0.073 |
gh | -0.072 |
affles | -0.071 |
Token | my |
Token ID | 1820 |
Feature activation | +6.36e-03 |
Pos logit contributions | |
iest | +0.024 |
ops | +0.024 |
hardly | +0.023 |
affles | +0.023 |
gh | +0.023 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.008 |
crying | -0.006 |
blood | -0.006 |
Blue | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
iest | +0.014 |
ops | +0.013 |
affles | +0.013 |
hardly | +0.013 |
gh | +0.012 |
Neg logit contributions | |
nowhere | -0.010 |
who | -0.010 |
crying | -0.008 |
Blue | -0.008 |
blood | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
nowhere | +0.033 |
who | +0.028 |
blood | +0.028 |
Blue | +0.026 |
ter | +0.025 |
Neg logit contributions | |
hardly | -0.053 |
affles | -0.051 |
iest | -0.051 |
zed | -0.051 |
gh | -0.050 |
Token | little |
Token ID | 1310 |
Feature activation | -7.94e-02 |
Pos logit contributions | |
nowhere | +0.075 |
who | +0.062 |
blood | +0.059 |
crying | +0.055 |
Blue | +0.054 |
Neg logit contributions | |
iest | -0.112 |
ops | -0.111 |
hardly | -0.110 |
affles | -0.107 |
gh | -0.104 |
Token | boy |
Token ID | 2933 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
nowhere | +0.133 |
blood | +0.102 |
itself | +0.098 |
Blue | +0.091 |
half | +0.085 |
Neg logit contributions | |
iest | -0.151 |
working | -0.143 |
affles | -0.140 |
ops | -0.139 |
hardly | -0.132 |
Token | who |
Token ID | 508 |
Feature activation | +4.72e-04 |
Pos logit contributions | |
nowhere | +0.046 |
blood | +0.039 |
Blue | +0.033 |
crying | +0.032 |
weeping | +0.031 |
Neg logit contributions | |
ops | -0.064 |
iest | -0.063 |
hardly | -0.061 |
rd | -0.061 |
affles | -0.061 |
Token | liked |
Token ID | 8288 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
iest | +0.001 |
ops | +0.001 |
gh | +0.001 |
affles | +0.001 |
ves | +0.001 |
Neg logit contributions | |
who | -0.001 |
nowhere | -0.001 |
crying | -0.001 |
Blue | -0.001 |
a | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -2.91e-03 |
Pos logit contributions | |
nowhere | +0.039 |
who | +0.031 |
bug | +0.029 |
blood | +0.029 |
ter | +0.028 |
Neg logit contributions | |
hardly | -0.069 |
affles | -0.069 |
ops | -0.068 |
iest | -0.067 |
gh | -0.066 |
Token | play |
Token ID | 711 |
Feature activation | -4.63e-03 |
Pos logit contributions | |
who | +0.004 |
a | +0.004 |
nowhere | +0.004 |
crying | +0.004 |
Blue | +0.003 |
Neg logit contributions | |
ops | -0.006 |
iest | -0.006 |
robat | -0.006 |
affles | -0.006 |
ves | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +6.93e-03 |
Pos logit contributions | |
nowhere | +0.008 |
who | +0.007 |
crying | +0.006 |
blood | +0.006 |
Blue | +0.006 |
Neg logit contributions | |
affles | -0.015 |
ops | -0.015 |
iest | -0.015 |
hardly | -0.015 |
rd | -0.014 |
Token | She |
Token ID | 1375 |
Feature activation | +2.00e-03 |
Pos logit contributions | |
hardly | +0.013 |
ops | +0.013 |
zed | +0.012 |
iest | +0.012 |
rd | +0.012 |
Neg logit contributions | |
nowhere | -0.013 |
blood | -0.011 |
who | -0.010 |
crying | -0.010 |
ter | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | -2.23e-03 |
Pos logit contributions | |
iest | +0.004 |
ops | +0.004 |
affles | +0.004 |
rd | +0.004 |
gh | +0.004 |
Neg logit contributions | |
nowhere | -0.003 |
who | -0.003 |
crying | -0.003 |
bug | -0.003 |
Blue | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
nowhere | +0.003 |
who | +0.002 |
blood | +0.002 |
crying | +0.002 |
Blue | +0.002 |
Neg logit contributions | |
ops | -0.006 |
iest | -0.006 |
hardly | -0.005 |
affles | -0.005 |
gh | -0.005 |
Token | happiest |
Token ID | 49414 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
nowhere | +0.093 |
who | +0.083 |
blood | +0.076 |
crying | +0.076 |
Blue | +0.074 |
Neg logit contributions | |
ops | -0.086 |
iest | -0.086 |
hardly | -0.082 |
affles | -0.081 |
gh | -0.078 |
Token | little |
Token ID | 1310 |
Feature activation | -7.90e-02 |
Pos logit contributions | |
nowhere | +0.058 |
who | +0.050 |
blood | +0.044 |
crying | +0.044 |
Blue | +0.043 |
Neg logit contributions | |
ops | -0.085 |
iest | -0.084 |
hardly | -0.082 |
affles | -0.082 |
gh | -0.079 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
nowhere | +0.119 |
who | +0.104 |
blood | +0.092 |
crying | +0.091 |
Blue | +0.089 |
Neg logit contributions | |
iest | -0.171 |
ops | -0.168 |
hardly | -0.163 |
affles | -0.159 |
gh | -0.157 |
Token | in |
Token ID | 287 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
nowhere | +0.059 |
who | +0.054 |
crying | +0.042 |
Blue | +0.041 |
blood | +0.038 |
Neg logit contributions | |
ops | -0.112 |
affles | -0.111 |
iest | -0.110 |
hardly | -0.108 |
gh | -0.105 |
Token | the |
Token ID | 262 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
nowhere | +0.033 |
who | +0.029 |
blood | +0.027 |
Blue | +0.025 |
crying | +0.024 |
Neg logit contributions | |
iest | -0.077 |
hardly | -0.076 |
ops | -0.074 |
zed | -0.072 |
affles | -0.071 |
Token | whole |
Token ID | 2187 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
nowhere | +0.022 |
who | +0.020 |
blood | +0.019 |
Blue | +0.018 |
crying | +0.017 |
Neg logit contributions | |
iest | -0.023 |
ops | -0.022 |
hardly | -0.021 |
gh | -0.021 |
rd | -0.021 |
Token | world |
Token ID | 995 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
nowhere | +0.034 |
who | +0.030 |
blood | +0.027 |
crying | +0.027 |
Blue | +0.026 |
Neg logit contributions | |
ops | -0.047 |
iest | -0.046 |
hardly | -0.045 |
affles | -0.044 |
gh | -0.043 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
hardly | +0.015 |
ops | +0.015 |
iest | +0.014 |
ves | +0.014 |
rd | +0.014 |
Neg logit contributions | |
nowhere | -0.011 |
blood | -0.008 |
who | -0.008 |
crying | -0.008 |
ter | -0.008 |
Token | S |
Token ID | 50 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
nowhere | +0.077 |
who | +0.070 |
crying | +0.067 |
blood | +0.067 |
tighter | +0.066 |
Neg logit contributions | |
ops | -0.076 |
iest | -0.074 |
robat | -0.068 |
gh | -0.066 |
affles | -0.066 |
Token | ara |
Token ID | 3301 |
Feature activation | +2.59e-04 |
Pos logit contributions | |
ops | +0.038 |
iest | +0.038 |
hardly | +0.037 |
affles | +0.036 |
gh | +0.035 |
Neg logit contributions | |
nowhere | -0.047 |
who | -0.042 |
blood | -0.038 |
crying | -0.038 |
Blue | -0.038 |
Token | was |
Token ID | 373 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
ops | +0.001 |
iest | +0.001 |
hardly | +0.001 |
affles | +0.001 |
gh | +0.001 |
Neg logit contributions | |
nowhere | -0.000 |
who | -0.000 |
blood | -0.000 |
crying | -0.000 |
Blue | -0.000 |
Token | a |
Token ID | 257 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
nowhere | +0.035 |
blood | +0.032 |
who | +0.030 |
Blue | +0.029 |
ter | +0.028 |
Neg logit contributions | |
hardly | -0.049 |
affles | -0.045 |
zed | -0.045 |
gh | -0.043 |
irl | -0.043 |
Token | little |
Token ID | 1310 |
Feature activation | -7.90e-02 |
Pos logit contributions | |
nowhere | +0.078 |
blood | +0.064 |
who | +0.062 |
crying | +0.058 |
Blue | +0.056 |
Neg logit contributions | |
hardly | -0.108 |
iest | -0.107 |
ops | -0.105 |
affles | -0.103 |
gh | -0.099 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.37e-02 |
Pos logit contributions | |
nowhere | +0.140 |
blood | +0.106 |
itself | +0.105 |
Blue | +0.093 |
unconscious | +0.089 |
Neg logit contributions | |
iest | -0.140 |
working | -0.136 |
affles | -0.132 |
ops | -0.129 |
robat | -0.119 |
Token | who |
Token ID | 508 |
Feature activation | +3.49e-03 |
Pos logit contributions | |
nowhere | +0.048 |
blood | +0.041 |
crying | +0.032 |
Blue | +0.032 |
weeping | +0.030 |
Neg logit contributions | |
ops | -0.104 |
iest | -0.103 |
hardly | -0.100 |
rd | -0.100 |
gh | -0.098 |
Token | liked |
Token ID | 8288 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
iest | +0.008 |
ops | +0.007 |
affles | +0.007 |
gh | +0.007 |
hardly | +0.007 |
Neg logit contributions | |
who | -0.005 |
nowhere | -0.005 |
crying | -0.004 |
Blue | -0.004 |
a | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -1.30e-03 |
Pos logit contributions | |
nowhere | +0.040 |
who | +0.031 |
ter | +0.030 |
bug | +0.029 |
blood | +0.029 |
Neg logit contributions | |
hardly | -0.059 |
affles | -0.058 |
ops | -0.056 |
iest | -0.055 |
gh | -0.054 |
Token | play |
Token ID | 711 |
Feature activation | -1.38e-03 |
Pos logit contributions | |
who | +0.002 |
a | +0.002 |
nowhere | +0.002 |
crying | +0.002 |
Blue | +0.002 |
Neg logit contributions | |
ops | -0.003 |
iest | -0.003 |
robat | -0.003 |
affles | -0.003 |
ves | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +3.02e-03 |
Pos logit contributions | |
nowhere | +0.033 |
blood | +0.028 |
who | +0.026 |
crying | +0.026 |
weeping | +0.025 |
Neg logit contributions | |
iest | -0.046 |
ops | -0.045 |
hardly | -0.044 |
rd | -0.043 |
gh | -0.042 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
ops | +0.008 |
iest | +0.007 |
hardly | +0.007 |
affles | +0.007 |
gh | +0.007 |
Neg logit contributions | |
nowhere | -0.003 |
who | -0.003 |
blood | -0.002 |
crying | -0.002 |
Blue | -0.002 |
Token | Anna |
Token ID | 31160 |
Feature activation | +9.76e-03 |
Pos logit contributions | |
nowhere | +0.075 |
who | +0.069 |
tighter | +0.068 |
weeping | +0.066 |
crying | +0.066 |
Neg logit contributions | |
ops | -0.075 |
iest | -0.072 |
robat | -0.067 |
gh | -0.065 |
affles | -0.063 |
Token | was |
Token ID | 373 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
iest | +0.023 |
ops | +0.022 |
affles | +0.022 |
hardly | +0.022 |
gh | +0.021 |
Neg logit contributions | |
nowhere | -0.014 |
who | -0.012 |
crying | -0.010 |
Blue | -0.010 |
blood | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
nowhere | +0.035 |
blood | +0.033 |
who | +0.030 |
Blue | +0.029 |
ter | +0.029 |
Neg logit contributions | |
hardly | -0.052 |
zed | -0.048 |
affles | -0.048 |
gh | -0.046 |
irl | -0.046 |
Token | little |
Token ID | 1310 |
Feature activation | -7.89e-02 |
Pos logit contributions | |
nowhere | +0.078 |
blood | +0.063 |
who | +0.062 |
crying | +0.058 |
Blue | +0.056 |
Neg logit contributions | |
hardly | -0.108 |
iest | -0.107 |
ops | -0.106 |
affles | -0.103 |
gh | -0.100 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
nowhere | +0.138 |
blood | +0.107 |
itself | +0.105 |
Blue | +0.092 |
half | +0.089 |
Neg logit contributions | |
iest | -0.142 |
working | -0.137 |
affles | -0.132 |
ops | -0.130 |
hardly | -0.120 |
Token | who |
Token ID | 508 |
Feature activation | +1.12e-03 |
Pos logit contributions | |
nowhere | +0.062 |
blood | +0.055 |
Blue | +0.045 |
crying | +0.045 |
weeping | +0.043 |
Neg logit contributions | |
ops | -0.095 |
iest | -0.093 |
hardly | -0.091 |
rd | -0.090 |
gh | -0.088 |
Token | liked |
Token ID | 8288 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
iest | +0.003 |
ops | +0.003 |
affles | +0.002 |
gh | +0.002 |
ves | +0.002 |
Neg logit contributions | |
who | -0.002 |
nowhere | -0.001 |
crying | -0.001 |
a | -0.001 |
Blue | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -3.29e-03 |
Pos logit contributions | |
nowhere | +0.043 |
who | +0.034 |
ter | +0.032 |
blood | +0.032 |
bug | +0.031 |
Neg logit contributions | |
hardly | -0.063 |
affles | -0.062 |
ops | -0.060 |
iest | -0.059 |
gh | -0.058 |
Token | dream |
Token ID | 4320 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
who | +0.005 |
a | +0.005 |
nowhere | +0.004 |
crying | +0.004 |
Blue | +0.004 |
Neg logit contributions | |
ops | -0.007 |
iest | -0.007 |
robat | -0.007 |
affles | -0.007 |
ves | -0.006 |
Token | big |
Token ID | 1263 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
nowhere | +0.019 |
blood | +0.014 |
who | +0.013 |
crying | +0.013 |
weeping | +0.013 |
Neg logit contributions | |
iest | -0.024 |
ops | -0.023 |
hardly | -0.023 |
gh | -0.023 |
affles | -0.023 |
Token | gorilla |
Token ID | 45314 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
nowhere | +0.046 |
who | +0.039 |
blood | +0.036 |
crying | +0.034 |
bug | +0.033 |
Neg logit contributions | |
iest | -0.042 |
ops | -0.040 |
affles | -0.040 |
robat | -0.039 |
affle | -0.038 |
Token | named |
Token ID | 3706 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
nowhere | +0.018 |
who | +0.016 |
crying | +0.014 |
blood | +0.014 |
Blue | +0.014 |
Neg logit contributions | |
ops | -0.025 |
iest | -0.025 |
hardly | -0.024 |
affles | -0.024 |
gh | -0.023 |
Token | Bob |
Token ID | 5811 |
Feature activation | -4.02e-03 |
Pos logit contributions | |
nowhere | +0.027 |
who | +0.022 |
blood | +0.019 |
crying | +0.019 |
Blue | +0.017 |
Neg logit contributions | |
ops | -0.053 |
iest | -0.053 |
hardly | -0.052 |
affles | -0.051 |
gh | -0.050 |
Token | o |
Token ID | 78 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
nowhere | +0.007 |
who | +0.006 |
blood | +0.006 |
crying | +0.006 |
Blue | +0.005 |
Neg logit contributions | |
ops | -0.008 |
iest | -0.008 |
hardly | -0.007 |
affles | -0.007 |
gh | -0.007 |
Token | lived |
Token ID | 5615 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
iest | +0.032 |
ops | +0.032 |
hardly | +0.031 |
affles | +0.031 |
gh | +0.029 |
Neg logit contributions | |
nowhere | -0.033 |
who | -0.030 |
crying | -0.027 |
Blue | -0.027 |
blood | -0.026 |
Token | in |
Token ID | 287 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
nowhere | +0.101 |
blood | +0.099 |
who | +0.090 |
accidents | +0.087 |
chick | +0.082 |
Neg logit contributions | |
gh | -0.152 |
hardly | -0.139 |
irl | -0.139 |
iest | -0.136 |
affles | -0.136 |
Token | the |
Token ID | 262 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
nowhere | +0.036 |
who | +0.031 |
weeping | +0.030 |
ter | +0.029 |
crying | +0.029 |
Neg logit contributions | |
hardly | -0.079 |
zed | -0.075 |
iest | -0.074 |
ops | -0.072 |
rd | -0.072 |
Token | forest |
Token ID | 8222 |
Feature activation | -3.44e-03 |
Pos logit contributions | |
nowhere | +0.043 |
who | +0.039 |
blood | +0.036 |
crying | +0.035 |
ter | +0.034 |
Neg logit contributions | |
ops | -0.033 |
iest | -0.033 |
hardly | -0.033 |
gh | -0.031 |
rd | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
nowhere | +0.005 |
who | +0.004 |
blood | +0.004 |
Blue | +0.004 |
crying | +0.004 |
Neg logit contributions | |
ops | -0.007 |
iest | -0.007 |
hardly | -0.007 |
affles | -0.007 |
gh | -0.007 |
Token | Bob |
Token ID | 5811 |
Feature activation | -6.07e-04 |
Pos logit contributions | |
nowhere | +0.020 |
who | +0.018 |
blood | +0.015 |
crying | +0.015 |
Blue | +0.015 |
Neg logit contributions | |
iest | -0.029 |
ops | -0.029 |
hardly | -0.028 |
affles | -0.028 |
gh | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -3.43e-03 |
Pos logit contributions | |
iest | +0.018 |
ops | +0.017 |
hardly | +0.016 |
gh | +0.016 |
rd | +0.016 |
Neg logit contributions | |
nowhere | -0.015 |
who | -0.012 |
blood | -0.011 |
Blue | -0.011 |
crying | -0.011 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
nowhere | +0.005 |
blood | +0.004 |
who | +0.004 |
crying | +0.004 |
ter | +0.004 |
Neg logit contributions | |
ops | -0.008 |
hardly | -0.008 |
zed | -0.007 |
iest | -0.007 |
rd | -0.007 |
Token | Jimmy |
Token ID | 40335 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
nowhere | +0.055 |
who | +0.048 |
blood | +0.048 |
crying | +0.046 |
weeping | +0.044 |
Neg logit contributions | |
ops | -0.087 |
iest | -0.085 |
hardly | -0.079 |
robat | -0.079 |
gh | -0.078 |
Token | was |
Token ID | 373 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
iest | +0.044 |
ops | +0.043 |
hardly | +0.042 |
affles | +0.042 |
gh | +0.041 |
Neg logit contributions | |
nowhere | -0.032 |
who | -0.029 |
crying | -0.025 |
blood | -0.024 |
Blue | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
nowhere | +0.034 |
blood | +0.031 |
who | +0.029 |
Blue | +0.028 |
ety | +0.027 |
Neg logit contributions | |
hardly | -0.053 |
zed | -0.048 |
gh | -0.047 |
iest | -0.047 |
affles | -0.046 |
Token | little |
Token ID | 1310 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
nowhere | +0.074 |
blood | +0.060 |
who | +0.059 |
crying | +0.054 |
Blue | +0.053 |
Neg logit contributions | |
hardly | -0.112 |
iest | -0.112 |
ops | -0.110 |
affles | -0.107 |
gh | -0.104 |
Token | boy |
Token ID | 2933 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
nowhere | +0.124 |
blood | +0.098 |
itself | +0.083 |
Blue | +0.082 |
who | +0.080 |
Neg logit contributions | |
iest | -0.159 |
ops | -0.146 |
hardly | -0.144 |
working | -0.144 |
affles | -0.144 |
Token | who |
Token ID | 508 |
Feature activation | +4.49e-03 |
Pos logit contributions | |
nowhere | +0.028 |
blood | +0.023 |
Blue | +0.018 |
crying | +0.017 |
weeping | +0.017 |
Neg logit contributions | |
ops | -0.053 |
iest | -0.053 |
hardly | -0.051 |
rd | -0.050 |
affles | -0.050 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
iest | +0.010 |
ops | +0.010 |
ves | +0.010 |
affles | +0.010 |
gh | +0.010 |
Neg logit contributions | |
who | -0.006 |
nowhere | -0.006 |
crying | -0.005 |
a | -0.005 |
Blue | -0.005 |
Token | 't |
Token ID | 470 |
Feature activation | +5.15e-03 |
Pos logit contributions | |
ops | +0.033 |
iest | +0.033 |
hardly | +0.032 |
affles | +0.031 |
gh | +0.030 |
Neg logit contributions | |
nowhere | -0.036 |
who | -0.032 |
blood | -0.030 |
crying | -0.029 |
Blue | -0.029 |
Token | like |
Token ID | 588 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
ops | +0.011 |
iest | +0.010 |
affles | +0.010 |
gh | +0.010 |
hardly | +0.010 |
Neg logit contributions | |
who | -0.008 |
nowhere | -0.008 |
crying | -0.007 |
blood | -0.007 |
Blue | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +9.47e-04 |
Pos logit contributions | |
nowhere | +0.050 |
who | +0.043 |
blood | +0.041 |
crying | +0.039 |
Blue | +0.039 |
Neg logit contributions | |
iest | -0.044 |
ops | -0.043 |
hardly | -0.041 |
affles | -0.039 |
gh | -0.039 |
Token | she |
Token ID | 673 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
iest | +0.002 |
ops | +0.002 |
hardly | +0.002 |
affles | +0.002 |
gh | +0.002 |
Neg logit contributions | |
nowhere | -0.002 |
who | -0.001 |
blood | -0.001 |
crying | -0.001 |
Blue | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -6.78e-03 |
Pos logit contributions | |
nowhere | +0.023 |
who | +0.022 |
crying | +0.018 |
Blue | +0.018 |
blood | +0.018 |
Neg logit contributions | |
iest | -0.024 |
affles | -0.024 |
ops | -0.023 |
rd | -0.022 |
gh | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
nowhere | +0.010 |
who | +0.009 |
blood | +0.008 |
crying | +0.008 |
Blue | +0.007 |
Neg logit contributions | |
iest | -0.015 |
ops | -0.015 |
hardly | -0.014 |
affles | -0.014 |
gh | -0.014 |
Token | happiest |
Token ID | 49414 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
nowhere | +0.091 |
who | +0.082 |
blood | +0.076 |
crying | +0.075 |
Blue | +0.074 |
Neg logit contributions | |
iest | -0.075 |
ops | -0.074 |
hardly | -0.073 |
affles | -0.069 |
gh | -0.068 |
Token | little |
Token ID | 1310 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
nowhere | +0.056 |
who | +0.048 |
blood | +0.043 |
crying | +0.043 |
Blue | +0.042 |
Neg logit contributions | |
iest | -0.084 |
ops | -0.083 |
hardly | -0.081 |
affles | -0.079 |
gh | -0.077 |
Token | girl |
Token ID | 2576 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
nowhere | +0.122 |
who | +0.103 |
blood | +0.097 |
crying | +0.093 |
Blue | +0.091 |
Neg logit contributions | |
iest | -0.167 |
ops | -0.157 |
hardly | -0.154 |
rd | -0.146 |
gh | -0.146 |
Token | in |
Token ID | 287 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
nowhere | +0.067 |
who | +0.060 |
crying | +0.049 |
Blue | +0.048 |
blood | +0.046 |
Neg logit contributions | |
ops | -0.120 |
iest | -0.118 |
affles | -0.118 |
hardly | -0.115 |
gh | -0.112 |
Token | the |
Token ID | 262 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
nowhere | +0.036 |
who | +0.031 |
blood | +0.029 |
crying | +0.026 |
Blue | +0.026 |
Neg logit contributions | |
iest | -0.083 |
hardly | -0.081 |
ops | -0.080 |
zed | -0.078 |
affles | -0.076 |
Token | world |
Token ID | 995 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
nowhere | +0.023 |
who | +0.020 |
blood | +0.019 |
crying | +0.018 |
Blue | +0.018 |
Neg logit contributions | |
ops | -0.025 |
iest | -0.025 |
hardly | -0.024 |
rd | -0.023 |
gh | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -3.64e-03 |
Pos logit contributions | |
nowhere | +0.035 |
who | +0.030 |
crying | +0.026 |
blood | +0.026 |
Blue | +0.026 |
Neg logit contributions | |
ops | -0.053 |
iest | -0.053 |
hardly | -0.052 |
affles | -0.052 |
gh | -0.050 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
nowhere | +0.009 |
who | +0.007 |
ter | +0.007 |
blood | +0.007 |
crying | +0.007 |
Neg logit contributions | |
hardly | -0.014 |
ops | -0.013 |
rd | -0.013 |
iest | -0.013 |
zed | -0.013 |
Token | Le |
Token ID | 3123 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
nowhere | +0.088 |
who | +0.079 |
blood | +0.073 |
crying | +0.073 |
weeping | +0.072 |
Neg logit contributions | |
ops | -0.090 |
iest | -0.089 |
affles | -0.082 |
robat | -0.081 |
gh | -0.078 |
Token | o |
Token ID | 78 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
ops | +0.042 |
iest | +0.042 |
hardly | +0.040 |
affles | +0.040 |
gh | +0.039 |
Neg logit contributions | |
nowhere | -0.023 |
who | -0.019 |
blood | -0.016 |
crying | -0.016 |
Blue | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
iest | +0.041 |
ops | +0.041 |
hardly | +0.040 |
affles | +0.039 |
gh | +0.038 |
Neg logit contributions | |
nowhere | -0.038 |
who | -0.034 |
crying | -0.030 |
blood | -0.030 |
Blue | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
nowhere | +0.034 |
who | +0.031 |
blood | +0.030 |
ter | +0.027 |
Blue | +0.027 |
Neg logit contributions | |
hardly | -0.057 |
zed | -0.055 |
gh | -0.054 |
affles | -0.054 |
iest | -0.053 |
Token | little |
Token ID | 1310 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
nowhere | +0.074 |
blood | +0.059 |
who | +0.059 |
crying | +0.054 |
Blue | +0.053 |
Neg logit contributions | |
hardly | -0.105 |
iest | -0.104 |
ops | -0.102 |
affles | -0.099 |
gh | -0.098 |
Token | lion |
Token ID | 18744 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
nowhere | +0.133 |
blood | +0.100 |
itself | +0.097 |
Blue | +0.089 |
who | +0.085 |
Neg logit contributions | |
iest | -0.146 |
working | -0.139 |
affles | -0.133 |
ops | -0.133 |
affle | -0.123 |
Token | who |
Token ID | 508 |
Feature activation | +3.22e-04 |
Pos logit contributions | |
nowhere | +0.047 |
blood | +0.038 |
who | +0.037 |
crying | +0.036 |
Blue | +0.034 |
Neg logit contributions | |
ops | -0.064 |
iest | -0.063 |
hardly | -0.060 |
affles | -0.059 |
gh | -0.059 |
Token | loved |
Token ID | 6151 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
iest | +0.001 |
ops | +0.001 |
affles | +0.001 |
gh | +0.001 |
hardly | +0.001 |
Neg logit contributions | |
nowhere | -0.001 |
who | -0.001 |
Blue | -0.001 |
crying | -0.000 |
blood | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | -8.42e-04 |
Pos logit contributions | |
nowhere | +0.063 |
who | +0.049 |
bug | +0.049 |
blood | +0.047 |
ter | +0.047 |
Neg logit contributions | |
hardly | -0.094 |
affles | -0.091 |
iest | -0.088 |
gh | -0.087 |
ops | -0.087 |
Token | roar |
Token ID | 34532 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
nowhere | +0.001 |
who | +0.001 |
Blue | +0.001 |
crying | +0.001 |
blood | +0.001 |
Neg logit contributions | |
iest | -0.002 |
ops | -0.002 |
affles | -0.001 |
rd | -0.001 |
hardly | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +3.05e-03 |
Pos logit contributions | |
nowhere | +0.021 |
who | +0.019 |
crying | +0.016 |
Blue | +0.014 |
blood | +0.014 |
Neg logit contributions | |
iest | -0.038 |
affles | -0.037 |
ops | -0.037 |
hardly | -0.036 |
robat | -0.035 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
iest | +0.008 |
ops | +0.007 |
affles | +0.007 |
hardly | +0.007 |
gh | +0.007 |
Neg logit contributions | |
nowhere | -0.004 |
who | -0.003 |
blood | -0.003 |
Blue | -0.003 |
crying | -0.003 |
Token | John |
Token ID | 7554 |
Feature activation | +8.55e-03 |
Pos logit contributions | |
nowhere | +0.066 |
who | +0.059 |
blood | +0.057 |
crying | +0.055 |
weeping | +0.053 |
Neg logit contributions | |
ops | -0.087 |
iest | -0.085 |
hardly | -0.078 |
robat | -0.078 |
gh | -0.077 |
Token | was |
Token ID | 373 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
iest | +0.019 |
ops | +0.019 |
hardly | +0.018 |
affles | +0.018 |
gh | +0.018 |
Neg logit contributions | |
nowhere | -0.012 |
who | -0.011 |
crying | -0.009 |
blood | -0.009 |
Blue | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
nowhere | +0.035 |
blood | +0.032 |
who | +0.029 |
Blue | +0.029 |
ety | +0.028 |
Neg logit contributions | |
hardly | -0.053 |
zed | -0.048 |
gh | -0.047 |
iest | -0.047 |
affles | -0.046 |
Token | young |
Token ID | 1862 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
nowhere | +0.076 |
blood | +0.062 |
who | +0.061 |
crying | +0.056 |
Blue | +0.056 |
Neg logit contributions | |
hardly | -0.115 |
iest | -0.114 |
ops | -0.112 |
affles | -0.109 |
gh | -0.107 |
Token | boy |
Token ID | 2933 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
nowhere | +0.123 |
blood | +0.104 |
itself | +0.084 |
Blue | +0.082 |
who | +0.081 |
Neg logit contributions | |
iest | -0.159 |
ops | -0.141 |
hardly | -0.139 |
working | -0.137 |
robat | -0.134 |
Token | of |
Token ID | 286 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
nowhere | +0.031 |
blood | +0.025 |
Blue | +0.021 |
crying | +0.021 |
who | +0.020 |
Neg logit contributions | |
ops | -0.058 |
iest | -0.057 |
hardly | -0.056 |
rd | -0.054 |
affles | -0.054 |
Token | three |
Token ID | 1115 |
Feature activation | +4.28e-03 |
Pos logit contributions | |
nowhere | +0.064 |
bug | +0.050 |
blood | +0.047 |
crying | +0.045 |
Blue | +0.044 |
Neg logit contributions | |
iest | -0.124 |
hardly | -0.123 |
affles | -0.118 |
working | -0.116 |
session | -0.114 |
Token | who |
Token ID | 508 |
Feature activation | +8.75e-04 |
Pos logit contributions | |
ops | +0.009 |
hardly | +0.009 |
iest | +0.009 |
gh | +0.009 |
affles | +0.009 |
Neg logit contributions | |
who | -0.006 |
nowhere | -0.006 |
crying | -0.005 |
blood | -0.005 |
Blue | -0.004 |
Token | loved |
Token ID | 6151 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
iest | +0.002 |
ops | +0.002 |
affles | +0.002 |
gh | +0.002 |
robat | +0.002 |
Neg logit contributions | |
who | -0.001 |
nowhere | -0.001 |
crying | -0.001 |
a | -0.001 |
Blue | -0.001 |
Token | ! |
Token ID | 0 |
Feature activation | -8.12e-03 |
Pos logit contributions | |
iest | +0.054 |
ops | +0.054 |
hardly | +0.053 |
affles | +0.052 |
gh | +0.051 |
Neg logit contributions | |
nowhere | -0.038 |
who | -0.033 |
blood | -0.029 |
crying | -0.029 |
Blue | -0.029 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
nowhere | +0.008 |
who | +0.006 |
blood | +0.005 |
crying | +0.005 |
Blue | +0.005 |
Neg logit contributions | |
ops | -0.022 |
iest | -0.022 |
hardly | -0.021 |
affles | -0.021 |
gh | -0.020 |
Token | Bob |
Token ID | 18861 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
nowhere | +0.065 |
who | +0.057 |
blood | +0.053 |
crying | +0.052 |
weeping | +0.049 |
Neg logit contributions | |
ops | -0.091 |
iest | -0.090 |
hardly | -0.083 |
robat | -0.083 |
affles | -0.082 |
Token | was |
Token ID | 373 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
iest | +0.040 |
ops | +0.039 |
hardly | +0.039 |
affles | +0.038 |
gh | +0.037 |
Neg logit contributions | |
nowhere | -0.020 |
who | -0.018 |
crying | -0.015 |
blood | -0.014 |
Blue | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
nowhere | +0.036 |
who | +0.030 |
blood | +0.030 |
Blue | +0.027 |
ter | +0.025 |
Neg logit contributions | |
hardly | -0.067 |
iest | -0.065 |
affles | -0.063 |
gh | -0.063 |
zed | -0.063 |
Token | little |
Token ID | 1310 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
nowhere | +0.078 |
who | +0.064 |
blood | +0.063 |
crying | +0.058 |
Blue | +0.057 |
Neg logit contributions | |
iest | -0.110 |
hardly | -0.110 |
ops | -0.108 |
affles | -0.105 |
gh | -0.102 |
Token | boy |
Token ID | 2933 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
nowhere | +0.122 |
blood | +0.094 |
who | +0.080 |
Blue | +0.079 |
itself | +0.078 |
Neg logit contributions | |
iest | -0.162 |
ops | -0.148 |
affles | -0.148 |
hardly | -0.144 |
working | -0.143 |
Token | who |
Token ID | 508 |
Feature activation | +2.84e-03 |
Pos logit contributions | |
nowhere | +0.035 |
blood | +0.029 |
Blue | +0.025 |
crying | +0.025 |
weeping | +0.024 |
Neg logit contributions | |
iest | -0.048 |
ops | -0.048 |
hardly | -0.046 |
affles | -0.045 |
rd | -0.045 |
Token | had |
Token ID | 550 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
iest | +0.006 |
ops | +0.006 |
affles | +0.006 |
ves | +0.006 |
robat | +0.006 |
Neg logit contributions | |
who | -0.004 |
nowhere | -0.004 |
crying | -0.003 |
a | -0.003 |
Blue | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -4.22e-03 |
Pos logit contributions | |
nowhere | +0.030 |
who | +0.021 |
blood | +0.014 |
crying | +0.014 |
Blue | +0.013 |
Neg logit contributions | |
ops | -0.133 |
iest | -0.132 |
hardly | -0.129 |
affles | -0.128 |
gh | -0.126 |
Token | lot |
Token ID | 1256 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
nowhere | +0.007 |
who | +0.006 |
blood | +0.006 |
crying | +0.006 |
Blue | +0.006 |
Neg logit contributions | |
ops | -0.008 |
iest | -0.008 |
hardly | -0.008 |
affles | -0.008 |
gh | -0.008 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
hardly | +0.006 |
ops | +0.005 |
iest | +0.005 |
ves | +0.005 |
rd | +0.005 |
Neg logit contributions | |
nowhere | -0.005 |
tighter | -0.005 |
who | -0.004 |
weeping | -0.004 |
crying | -0.004 |
Token | L |
Token ID | 43 |
Feature activation | +2.22e-03 |
Pos logit contributions | |
nowhere | +0.074 |
who | +0.068 |
crying | +0.063 |
tighter | +0.063 |
weeping | +0.063 |
Neg logit contributions | |
ops | -0.081 |
iest | -0.080 |
gh | -0.073 |
robat | -0.073 |
affles | -0.072 |
Token | ily |
Token ID | 813 |
Feature activation | -3.16e-03 |
Pos logit contributions | |
ops | +0.004 |
iest | +0.004 |
affles | +0.004 |
hardly | +0.004 |
ves | +0.004 |
Neg logit contributions | |
nowhere | -0.004 |
who | -0.003 |
Blue | -0.003 |
crying | -0.003 |
weeping | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
nowhere | +0.004 |
who | +0.003 |
blood | +0.002 |
crying | +0.002 |
Blue | +0.002 |
Neg logit contributions | |
ops | -0.008 |
iest | -0.008 |
hardly | -0.008 |
affles | -0.008 |
gh | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
nowhere | +0.034 |
blood | +0.031 |
who | +0.030 |
ter | +0.029 |
Blue | +0.028 |
Neg logit contributions | |
hardly | -0.051 |
affles | -0.048 |
zed | -0.047 |
gh | -0.046 |
irl | -0.045 |
Token | young |
Token ID | 1862 |
Feature activation | -7.81e-02 |
Pos logit contributions | |
nowhere | +0.077 |
blood | +0.064 |
who | +0.061 |
crying | +0.057 |
Blue | +0.056 |
Neg logit contributions | |
hardly | -0.100 |
iest | -0.098 |
ops | -0.097 |
affles | -0.095 |
gh | -0.091 |
Token | cat |
Token ID | 3797 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
nowhere | +0.131 |
blood | +0.104 |
itself | +0.090 |
tighter | +0.088 |
unconscious | +0.087 |
Neg logit contributions | |
iest | -0.152 |
ops | -0.135 |
working | -0.131 |
affles | -0.129 |
hardly | -0.126 |
Token | who |
Token ID | 508 |
Feature activation | -4.46e-04 |
Pos logit contributions | |
nowhere | +0.054 |
blood | +0.047 |
crying | +0.041 |
weeping | +0.040 |
who | +0.038 |
Neg logit contributions | |
ops | -0.067 |
iest | -0.065 |
hardly | -0.063 |
ves | -0.062 |
affles | -0.061 |
Token | liked |
Token ID | 8288 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
nowhere | +0.001 |
who | +0.001 |
Blue | +0.001 |
crying | +0.001 |
blood | +0.001 |
Neg logit contributions | |
iest | -0.001 |
ops | -0.001 |
affles | -0.001 |
gh | -0.001 |
hardly | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -3.55e-06 |
Pos logit contributions | |
nowhere | +0.044 |
who | +0.035 |
bug | +0.034 |
ter | +0.033 |
blood | +0.032 |
Neg logit contributions | |
hardly | -0.058 |
affles | -0.056 |
gh | -0.054 |
iest | -0.054 |
ops | -0.054 |
Token | play |
Token ID | 711 |
Feature activation | -3.07e-03 |
Token | . |
Token ID | 13 |
Feature activation | +2.48e-03 |
Pos logit contributions | |
iest | +0.010 |
ops | +0.010 |
hardly | +0.010 |
affles | +0.009 |
gh | +0.009 |
Neg logit contributions | |
nowhere | -0.007 |
who | -0.006 |
blood | -0.005 |
crying | -0.005 |
Blue | -0.005 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
ops | +0.006 |
hardly | +0.006 |
iest | +0.006 |
zed | +0.005 |
rd | +0.005 |
Neg logit contributions | |
nowhere | -0.003 |
who | -0.003 |
blood | -0.003 |
crying | -0.002 |
weeping | -0.002 |
Token | John |
Token ID | 7554 |
Feature activation | +9.34e-03 |
Pos logit contributions | |
nowhere | +0.061 |
who | +0.054 |
blood | +0.054 |
crying | +0.051 |
weeping | +0.049 |
Neg logit contributions | |
ops | -0.087 |
iest | -0.085 |
robat | -0.078 |
hardly | -0.078 |
gh | -0.078 |
Token | was |
Token ID | 373 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
iest | +0.021 |
ops | +0.021 |
hardly | +0.020 |
affles | +0.020 |
gh | +0.020 |
Neg logit contributions | |
nowhere | -0.013 |
who | -0.012 |
crying | -0.010 |
blood | -0.010 |
Blue | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
nowhere | +0.037 |
blood | +0.035 |
who | +0.032 |
Blue | +0.032 |
ety | +0.031 |
Neg logit contributions | |
hardly | -0.055 |
zed | -0.050 |
gh | -0.049 |
iest | -0.049 |
affles | -0.048 |
Token | young |
Token ID | 1862 |
Feature activation | -7.81e-02 |
Pos logit contributions | |
nowhere | +0.075 |
blood | +0.063 |
who | +0.058 |
crying | +0.055 |
Blue | +0.055 |
Neg logit contributions | |
hardly | -0.108 |
iest | -0.106 |
ops | -0.103 |
affles | -0.100 |
gh | -0.099 |
Token | boy |
Token ID | 2933 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
nowhere | +0.124 |
blood | +0.106 |
itself | +0.084 |
bleeding | +0.082 |
crying | +0.082 |
Neg logit contributions | |
iest | -0.156 |
ops | -0.138 |
hardly | -0.135 |
working | -0.135 |
robat | -0.132 |
Token | about |
Token ID | 546 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
nowhere | +0.028 |
blood | +0.023 |
Blue | +0.018 |
crying | +0.018 |
weeping | +0.017 |
Neg logit contributions | |
ops | -0.056 |
iest | -0.055 |
hardly | -0.053 |
rd | -0.052 |
affles | -0.052 |
Token | three |
Token ID | 1115 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
nowhere | +0.046 |
bug | +0.040 |
who | +0.037 |
blood | +0.036 |
ety | +0.035 |
Neg logit contributions | |
hardly | -0.116 |
iest | -0.114 |
gh | -0.112 |
zed | -0.108 |
session | -0.108 |
Token | years |
Token ID | 812 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
ops | +0.022 |
hardly | +0.022 |
iest | +0.022 |
gh | +0.021 |
affles | +0.021 |
Neg logit contributions | |
nowhere | -0.017 |
who | -0.016 |
crying | -0.013 |
blood | -0.013 |
Blue | -0.013 |
Token | old |
Token ID | 1468 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
nowhere | +0.028 |
who | +0.023 |
blood | +0.020 |
crying | +0.019 |
Blue | +0.018 |
Neg logit contributions | |
iest | -0.062 |
ops | -0.062 |
hardly | -0.061 |
affles | -0.059 |
gh | -0.058 |
Token | old |
Token ID | 1468 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
nowhere | +0.024 |
who | +0.021 |
blood | +0.020 |
Blue | +0.019 |
crying | +0.019 |
Neg logit contributions | |
iest | -0.024 |
ops | -0.024 |
hardly | -0.023 |
affles | -0.022 |
rd | -0.022 |
Token | man |
Token ID | 582 |
Feature activation | +2.91e-03 |
Pos logit contributions | |
nowhere | +0.076 |
who | +0.069 |
crying | +0.063 |
blood | +0.062 |
Blue | +0.061 |
Neg logit contributions | |
iest | -0.072 |
ops | -0.072 |
affles | -0.069 |
hardly | -0.069 |
gh | -0.066 |
Token | 's |
Token ID | 338 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
ops | +0.006 |
iest | +0.006 |
hardly | +0.005 |
affles | +0.005 |
gh | +0.005 |
Neg logit contributions | |
nowhere | -0.005 |
who | -0.004 |
blood | -0.004 |
crying | -0.004 |
Blue | -0.004 |
Token | gate |
Token ID | 8946 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
ops | +0.026 |
iest | +0.026 |
hardly | +0.026 |
affles | +0.025 |
gh | +0.025 |
Neg logit contributions | |
nowhere | -0.021 |
who | -0.018 |
Blue | -0.017 |
blood | -0.017 |
ter | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
nowhere | +0.046 |
who | +0.041 |
blood | +0.037 |
crying | +0.036 |
Blue | +0.036 |
Neg logit contributions | |
ops | -0.048 |
iest | -0.048 |
affles | -0.046 |
hardly | -0.046 |
gh | -0.044 |
Token | back |
Token ID | 736 |
Feature activation | -7.81e-02 |
Pos logit contributions | |
nowhere | +0.016 |
who | +0.014 |
blood | +0.012 |
crying | +0.012 |
Blue | +0.012 |
Neg logit contributions | |
iest | -0.025 |
ops | -0.025 |
hardly | -0.025 |
affles | -0.024 |
gh | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
nowhere | +0.092 |
blood | +0.073 |
who | +0.072 |
Blue | +0.070 |
bug | +0.061 |
Neg logit contributions | |
iest | -0.185 |
hardly | -0.179 |
ops | -0.178 |
gh | -0.171 |
rd | -0.167 |
Token | and |
Token ID | 290 |
Feature activation | -8.52e-03 |
Pos logit contributions | |
nowhere | +0.019 |
blood | +0.015 |
who | +0.015 |
bug | +0.014 |
Blue | +0.014 |
Neg logit contributions | |
hardly | -0.034 |
iest | -0.032 |
ops | -0.031 |
zed | -0.031 |
gh | -0.030 |
Token | he |
Token ID | 339 |
Feature activation | -3.08e-03 |
Pos logit contributions | |
nowhere | +0.015 |
who | +0.013 |
blood | +0.012 |
crying | +0.012 |
Blue | +0.012 |
Neg logit contributions | |
ops | -0.016 |
iest | -0.016 |
hardly | -0.016 |
affles | -0.015 |
gh | -0.015 |
Token | never |
Token ID | 1239 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
nowhere | +0.006 |
who | +0.005 |
crying | +0.004 |
blood | +0.004 |
Blue | +0.004 |
Neg logit contributions | |
iest | -0.006 |
ops | -0.006 |
affles | -0.006 |
hardly | -0.005 |
gh | -0.005 |
Token | forgot |
Token ID | 16453 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
nowhere | +0.061 |
who | +0.055 |
crying | +0.051 |
blood | +0.050 |
Blue | +0.049 |
Neg logit contributions | |
ops | -0.047 |
iest | -0.047 |
affles | -0.045 |
hardly | -0.045 |
gh | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +4.82e-03 |
Pos logit contributions | |
ops | +0.002 |
iest | +0.002 |
hardly | +0.002 |
affles | +0.002 |
gh | +0.002 |
Neg logit contributions | |
nowhere | -0.001 |
who | -0.001 |
blood | -0.001 |
crying | -0.001 |
Blue | -0.001 |
Token | They |
Token ID | 1119 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
hardly | +0.010 |
ops | +0.010 |
affles | +0.009 |
zed | +0.009 |
ves | +0.009 |
Neg logit contributions | |
nowhere | -0.008 |
who | -0.006 |
blood | -0.006 |
crying | -0.006 |
weeping | -0.006 |
Token | put |
Token ID | 1234 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
iest | +0.081 |
ops | +0.079 |
gh | +0.078 |
robat | +0.076 |
affles | +0.075 |
Neg logit contributions | |
nowhere | -0.058 |
who | -0.057 |
bug | -0.045 |
crying | -0.044 |
a | -0.043 |
Token | on |
Token ID | 319 |
Feature activation | +7.19e-03 |
Pos logit contributions | |
nowhere | +0.015 |
who | +0.013 |
crying | +0.012 |
weeping | +0.012 |
blood | +0.011 |
Neg logit contributions | |
iest | -0.022 |
ops | -0.022 |
affles | -0.022 |
ves | -0.021 |
hardly | -0.021 |
Token | their |
Token ID | 511 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
ops | +0.014 |
hardly | +0.014 |
iest | +0.014 |
affles | +0.014 |
gh | +0.013 |
Neg logit contributions | |
nowhere | -0.012 |
who | -0.010 |
blood | -0.010 |
crying | -0.010 |
Blue | -0.009 |
Token | prett |
Token ID | 46442 |
Feature activation | +9.63e-02 |
Pos logit contributions | |
iest | +0.084 |
ops | +0.084 |
hardly | +0.079 |
robat | +0.078 |
gh | +0.074 |
Neg logit contributions | |
nowhere | -0.071 |
who | -0.064 |
crying | -0.057 |
a | -0.056 |
blood | -0.056 |
Token | iest |
Token ID | 6386 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
hardly | +0.106 |
ops | +0.097 |
gh | +0.091 |
zed | +0.090 |
rd | +0.089 |
Neg logit contributions | |
nowhere | -0.238 |
who | -0.230 |
blood | -0.215 |
where | -0.205 |
too | -0.203 |
Token | dresses |
Token ID | 27309 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
ops | +0.046 |
hardly | +0.045 |
iest | +0.044 |
zed | +0.043 |
gh | +0.042 |
Neg logit contributions | |
nowhere | -0.040 |
who | -0.037 |
blood | -0.033 |
crying | -0.032 |
a | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +5.13e-03 |
Pos logit contributions | |
ops | +0.065 |
iest | +0.065 |
affles | +0.065 |
hardly | +0.064 |
gh | +0.063 |
Neg logit contributions | |
nowhere | -0.022 |
who | -0.019 |
crying | -0.014 |
Blue | -0.013 |
blood | -0.013 |
Token | their |
Token ID | 511 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
iest | +0.011 |
ops | +0.011 |
robat | +0.010 |
gh | +0.010 |
hardly | +0.010 |
Neg logit contributions | |
nowhere | -0.008 |
who | -0.008 |
a | -0.007 |
Blue | -0.006 |
blood | -0.006 |
Token | most |
Token ID | 749 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
iest | +0.029 |
ops | +0.028 |
hardly | +0.027 |
affles | +0.026 |
gh | +0.026 |
Neg logit contributions | |
nowhere | -0.025 |
who | -0.023 |
crying | -0.021 |
blood | -0.020 |
Blue | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e-06 |
Pos logit contributions | |
ops | +0.019 |
iest | +0.019 |
hardly | +0.019 |
affles | +0.019 |
gh | +0.018 |
Neg logit contributions | |
nowhere | -0.011 |
who | -0.010 |
crying | -0.008 |
Blue | -0.008 |
blood | -0.007 |
Token | They |
Token ID | 1119 |
Feature activation | +2.23e-02 |
Token | then |
Token ID | 788 |
Feature activation | +7.19e-03 |
Pos logit contributions | |
iest | +0.048 |
ops | +0.047 |
affles | +0.046 |
hardly | +0.046 |
gh | +0.045 |
Neg logit contributions | |
nowhere | -0.034 |
who | -0.031 |
Blue | -0.026 |
crying | -0.026 |
blood | -0.025 |
Token | ate |
Token ID | 15063 |
Feature activation | +5.83e-03 |
Pos logit contributions | |
iest | +0.014 |
ops | +0.014 |
affles | +0.014 |
hardly | +0.014 |
gh | +0.013 |
Neg logit contributions | |
nowhere | -0.012 |
who | -0.011 |
crying | -0.010 |
Blue | -0.010 |
blood | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
ops | +0.012 |
iest | +0.012 |
affles | +0.012 |
hardly | +0.012 |
gh | +0.011 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.007 |
blood | -0.007 |
crying | -0.007 |
weeping | -0.006 |
Token | y |
Token ID | 331 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
ops | +0.040 |
iest | +0.040 |
hardly | +0.039 |
affles | +0.038 |
gh | +0.037 |
Neg logit contributions | |
nowhere | -0.041 |
who | -0.037 |
blood | -0.034 |
crying | -0.033 |
Blue | -0.033 |
Token | ummy |
Token ID | 13513 |
Feature activation | +9.55e-03 |
Pos logit contributions | |
ops | +0.150 |
iest | +0.150 |
hardly | +0.146 |
affles | +0.143 |
gh | +0.139 |
Neg logit contributions | |
nowhere | -0.134 |
who | -0.119 |
blood | -0.108 |
crying | -0.107 |
Blue | -0.105 |
Token | dinner |
Token ID | 8073 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
ops | +0.015 |
iest | +0.015 |
hardly | +0.015 |
affles | +0.015 |
gh | +0.014 |
Neg logit contributions | |
nowhere | -0.019 |
who | -0.017 |
crying | -0.016 |
blood | -0.016 |
Blue | -0.015 |
Token | together |
Token ID | 1978 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
affles | +0.035 |
iest | +0.035 |
ops | +0.035 |
hardly | +0.034 |
gh | +0.033 |
Neg logit contributions | |
nowhere | -0.013 |
who | -0.011 |
Blue | -0.008 |
crying | -0.008 |
blood | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +9.40e-03 |
Pos logit contributions | |
iest | +0.057 |
ops | +0.057 |
affles | +0.055 |
hardly | +0.055 |
gh | +0.053 |
Neg logit contributions | |
nowhere | -0.025 |
who | -0.022 |
crying | -0.017 |
blood | -0.017 |
Blue | -0.017 |
Token | |
Token ID | 220 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
ops | +0.020 |
hardly | +0.020 |
zed | +0.019 |
iest | +0.018 |
affles | +0.018 |
Neg logit contributions | |
nowhere | -0.014 |
blood | -0.011 |
who | -0.011 |
crying | -0.011 |
ter | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
nowhere | +0.013 |
blood | +0.011 |
Blue | +0.011 |
crying | +0.010 |
who | +0.010 |
Neg logit contributions | |
ops | -0.016 |
iest | -0.015 |
hardly | -0.015 |
zed | -0.014 |
gh | -0.014 |
Token | She |
Token ID | 1375 |
Feature activation | +9.75e-03 |
Pos logit contributions | |
nowhere | +0.023 |
blood | +0.021 |
who | +0.019 |
ter | +0.018 |
crying | +0.017 |
Neg logit contributions | |
ops | -0.021 |
hardly | -0.020 |
rd | -0.020 |
affles | -0.019 |
Today | -0.018 |
Token | loved |
Token ID | 6151 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
iest | +0.020 |
ops | +0.020 |
affles | +0.019 |
robat | +0.018 |
hardly | +0.018 |
Neg logit contributions | |
nowhere | -0.017 |
who | -0.016 |
crying | -0.013 |
Blue | -0.013 |
blood | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +3.91e-03 |
Pos logit contributions | |
nowhere | +0.045 |
who | +0.036 |
blood | +0.032 |
bug | +0.031 |
ter | +0.030 |
Neg logit contributions | |
hardly | -0.082 |
affles | -0.081 |
ops | -0.079 |
iest | -0.078 |
zed | -0.076 |
Token | eat |
Token ID | 4483 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
iest | +0.009 |
ops | +0.009 |
robat | +0.008 |
rd | +0.008 |
vation | +0.008 |
Neg logit contributions | |
who | -0.006 |
nowhere | -0.005 |
a | -0.005 |
Blue | -0.005 |
crying | -0.005 |
Token | y |
Token ID | 331 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
ops | +0.038 |
iest | +0.038 |
affles | +0.038 |
hardly | +0.036 |
gh | +0.035 |
Neg logit contributions | |
nowhere | -0.036 |
who | -0.031 |
crying | -0.028 |
blood | -0.028 |
Blue | -0.027 |
Token | ummy |
Token ID | 13513 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
iest | +0.159 |
ops | +0.155 |
hardly | +0.155 |
zed | +0.147 |
affles | +0.146 |
Neg logit contributions | |
nowhere | -0.130 |
who | -0.115 |
blood | -0.105 |
crying | -0.104 |
Blue | -0.103 |
Token | cookies |
Token ID | 14746 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
hardly | +0.028 |
ops | +0.026 |
gh | +0.026 |
zed | +0.026 |
iest | +0.026 |
Neg logit contributions | |
who | -0.017 |
nowhere | -0.017 |
a | -0.014 |
blood | -0.014 |
Blue | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
iest | +0.099 |
ops | +0.098 |
hardly | +0.096 |
affles | +0.095 |
gh | +0.093 |
Neg logit contributions | |
nowhere | -0.051 |
who | -0.044 |
Blue | -0.036 |
blood | -0.036 |
crying | -0.036 |
Token | She |
Token ID | 1375 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
ops | +0.015 |
hardly | +0.014 |
affles | +0.013 |
rd | +0.012 |
robat | +0.012 |
Neg logit contributions | |
nowhere | -0.025 |
blood | -0.022 |
who | -0.022 |
crying | -0.021 |
ter | -0.021 |
Token | asked |
Token ID | 1965 |
Feature activation | -3.47e-03 |
Pos logit contributions | |
iest | +0.028 |
ops | +0.026 |
robat | +0.025 |
affles | +0.024 |
rd | +0.024 |
Neg logit contributions | |
nowhere | -0.024 |
who | -0.024 |
Blue | -0.020 |
crying | -0.019 |
bug | -0.019 |
Token | They |
Token ID | 1119 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
nowhere | +0.002 |
who | +0.002 |
blood | +0.002 |
crying | +0.002 |
weeping | +0.001 |
Neg logit contributions | |
ops | -0.003 |
hardly | -0.003 |
iest | -0.003 |
affles | -0.003 |
rd | -0.003 |
Token | put |
Token ID | 1234 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
iest | +0.038 |
ops | +0.036 |
gh | +0.035 |
rd | +0.034 |
affles | +0.034 |
Neg logit contributions | |
who | -0.034 |
nowhere | -0.034 |
Blue | -0.029 |
blood | -0.029 |
crying | -0.029 |
Token | on |
Token ID | 319 |
Feature activation | +6.24e-03 |
Pos logit contributions | |
nowhere | +0.018 |
who | +0.015 |
crying | +0.014 |
weeping | +0.013 |
blood | +0.013 |
Neg logit contributions | |
iest | -0.025 |
ops | -0.024 |
affles | -0.024 |
robat | -0.023 |
hardly | -0.023 |
Token | their |
Token ID | 511 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
ops | +0.013 |
hardly | +0.013 |
iest | +0.013 |
affles | +0.013 |
gh | +0.013 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.008 |
blood | -0.007 |
crying | -0.007 |
Blue | -0.007 |
Token | rain |
Token ID | 6290 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
iest | +0.071 |
ops | +0.071 |
hardly | +0.066 |
affles | +0.065 |
robat | +0.062 |
Neg logit contributions | |
nowhere | -0.078 |
who | -0.070 |
crying | -0.064 |
blood | -0.064 |
Blue | -0.062 |
Token | co |
Token ID | 1073 |
Feature activation | +7.86e-02 |
Pos logit contributions | |
iest | +0.027 |
ops | +0.027 |
hardly | +0.026 |
affles | +0.026 |
gh | +0.025 |
Neg logit contributions | |
nowhere | -0.015 |
who | -0.013 |
blood | -0.011 |
crying | -0.011 |
Blue | -0.011 |
Token | ats |
Token ID | 1381 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
iest | +0.056 |
hardly | +0.050 |
ops | +0.043 |
affles | +0.041 |
vows | +0.039 |
Neg logit contributions | |
nowhere | -0.227 |
who | -0.223 |
crying | -0.203 |
blood | -0.201 |
Blue | -0.198 |
Token | and |
Token ID | 290 |
Feature activation | -6.44e-03 |
Pos logit contributions | |
ops | +0.062 |
iest | +0.061 |
hardly | +0.060 |
gh | +0.057 |
ves | +0.057 |
Neg logit contributions | |
nowhere | -0.028 |
who | -0.021 |
blood | -0.020 |
crying | -0.018 |
Blue | -0.017 |
Token | boots |
Token ID | 14412 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
nowhere | +0.010 |
who | +0.010 |
a | +0.009 |
blood | +0.009 |
Blue | +0.009 |
Neg logit contributions | |
iest | -0.014 |
ops | -0.013 |
affles | -0.013 |
gh | -0.013 |
robat | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -7.68e-03 |
Pos logit contributions | |
ops | +0.083 |
iest | +0.082 |
hardly | +0.080 |
affles | +0.077 |
gh | +0.077 |
Neg logit contributions | |
nowhere | -0.041 |
who | -0.032 |
blood | -0.028 |
crying | -0.028 |
Blue | -0.027 |
Token | go |
Token ID | 467 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
nowhere | +0.013 |
who | +0.013 |
blood | +0.011 |
a | +0.011 |
Blue | +0.011 |
Neg logit contributions | |
iest | -0.016 |
ops | -0.016 |
affles | -0.015 |
gh | -0.015 |
hardly | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | -7.78e-03 |
Pos logit contributions | |
nowhere | +0.022 |
blood | +0.019 |
who | +0.016 |
Blue | +0.016 |
crying | +0.016 |
Neg logit contributions | |
iest | -0.046 |
ops | -0.044 |
hardly | -0.043 |
zed | -0.042 |
gh | -0.042 |
Token | find |
Token ID | 1064 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
nowhere | +0.015 |
who | +0.014 |
blood | +0.012 |
crying | +0.012 |
Blue | +0.012 |
Neg logit contributions | |
ops | -0.013 |
iest | -0.013 |
hardly | -0.013 |
affles | -0.012 |
gh | -0.012 |
Token | food |
Token ID | 2057 |
Feature activation | +7.52e-03 |
Pos logit contributions | |
nowhere | +0.033 |
crying | +0.026 |
who | +0.026 |
ety | +0.025 |
weeping | +0.025 |
Neg logit contributions | |
iest | -0.054 |
hardly | -0.046 |
ves | -0.046 |
zed | -0.044 |
affles | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
iest | +0.018 |
ops | +0.018 |
hardly | +0.018 |
affles | +0.017 |
gh | +0.017 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.007 |
blood | -0.007 |
crying | -0.006 |
Blue | -0.006 |
Token | On |
Token ID | 1550 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
hardly | +0.017 |
ops | +0.015 |
zed | +0.013 |
ves | +0.013 |
rd | +0.013 |
Neg logit contributions | |
nowhere | -0.022 |
ter | -0.019 |
crying | -0.019 |
blood | -0.019 |
weeping | -0.018 |
Token | one |
Token ID | 530 |
Feature activation | +7.58e-02 |
Pos logit contributions | |
iest | +0.057 |
ops | +0.056 |
hardly | +0.056 |
gh | +0.054 |
zed | +0.054 |
Neg logit contributions | |
nowhere | -0.027 |
who | -0.024 |
blood | -0.021 |
crying | -0.021 |
Blue | -0.019 |
Token | day |
Token ID | 1110 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
iest | +0.138 |
ops | +0.125 |
hardly | +0.123 |
rd | +0.117 |
gh | +0.117 |
Neg logit contributions | |
nowhere | -0.142 |
who | -0.120 |
blood | -0.117 |
crying | -0.115 |
weeping | -0.111 |
Token | , |
Token ID | 11 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
ops | +0.063 |
iest | +0.062 |
affles | +0.061 |
hardly | +0.060 |
robat | +0.058 |
Neg logit contributions | |
nowhere | -0.055 |
who | -0.049 |
crying | -0.044 |
Blue | -0.044 |
blood | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | -1.77e-03 |
Pos logit contributions | |
hardly | +0.064 |
iest | +0.063 |
ops | +0.060 |
ves | +0.058 |
zed | +0.057 |
Neg logit contributions | |
nowhere | -0.037 |
who | -0.031 |
crying | -0.030 |
blood | -0.029 |
weeping | -0.029 |
Token | bird |
Token ID | 6512 |
Feature activation | -3.13e-03 |
Pos logit contributions | |
nowhere | +0.003 |
who | +0.002 |
bug | +0.002 |
crying | +0.002 |
Being | +0.002 |
Neg logit contributions | |
iest | -0.004 |
hardly | -0.003 |
ops | -0.003 |
ves | -0.003 |
dream | -0.003 |
Token | found |
Token ID | 1043 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
nowhere | +0.005 |
who | +0.005 |
crying | +0.004 |
blood | +0.004 |
Blue | +0.004 |
Neg logit contributions | |
ops | -0.006 |
iest | -0.006 |
hardly | -0.006 |
affles | -0.006 |
gh | -0.006 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
nowhere | +0.048 |
who | +0.042 |
blood | +0.040 |
crying | +0.038 |
weeping | +0.038 |
Neg logit contributions | |
ops | -0.082 |
iest | -0.079 |
hardly | -0.077 |
gh | -0.075 |
robat | -0.074 |
Token | upon |
Token ID | 2402 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
ops | +0.070 |
iest | +0.070 |
hardly | +0.068 |
affles | +0.067 |
gh | +0.066 |
Neg logit contributions | |
nowhere | -0.044 |
who | -0.038 |
blood | -0.033 |
crying | -0.033 |
Blue | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
nowhere | +0.023 |
weeping | +0.017 |
who | +0.017 |
crying | +0.016 |
blood | +0.016 |
Neg logit contributions | |
hardly | -0.068 |
iest | -0.066 |
ops | -0.064 |
gh | -0.064 |
zed | -0.064 |
Token | time |
Token ID | 640 |
Feature activation | +8.90e-03 |
Pos logit contributions | |
iest | +0.036 |
ops | +0.036 |
hardly | +0.034 |
affles | +0.034 |
gh | +0.032 |
Neg logit contributions | |
nowhere | -0.066 |
who | -0.061 |
blood | -0.056 |
crying | -0.056 |
Blue | -0.055 |
Token | , |
Token ID | 11 |
Feature activation | +2.58e-03 |
Pos logit contributions | |
ops | +0.015 |
iest | +0.015 |
hardly | +0.014 |
gh | +0.014 |
affles | +0.014 |
Neg logit contributions | |
nowhere | -0.018 |
who | -0.016 |
blood | -0.015 |
crying | -0.014 |
Blue | -0.014 |
Token | there |
Token ID | 612 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
iest | +0.006 |
gh | +0.005 |
hardly | +0.005 |
ops | +0.005 |
zed | +0.005 |
Neg logit contributions | |
nowhere | -0.003 |
blood | -0.003 |
crying | -0.002 |
weeping | -0.002 |
mice | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
affles | +0.043 |
ops | +0.043 |
iest | +0.043 |
hardly | +0.042 |
robat | +0.040 |
Neg logit contributions | |
nowhere | -0.033 |
who | -0.029 |
crying | -0.025 |
blood | -0.025 |
Blue | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
nowhere | +0.028 |
who | +0.024 |
blood | +0.024 |
crying | +0.023 |
weeping | +0.023 |
Neg logit contributions | |
hardly | -0.066 |
iest | -0.064 |
affles | -0.064 |
ops | -0.063 |
gh | -0.062 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
nowhere | +0.031 |
blood | +0.024 |
who | +0.022 |
crying | +0.022 |
Blue | +0.021 |
Neg logit contributions | |
hardly | -0.041 |
iest | -0.041 |
ops | -0.040 |
affles | -0.040 |
gh | -0.039 |
Token | named |
Token ID | 3706 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
nowhere | +0.022 |
blood | +0.018 |
Blue | +0.017 |
who | +0.017 |
crying | +0.016 |
Neg logit contributions | |
ops | -0.024 |
iest | -0.024 |
affles | -0.023 |
hardly | -0.023 |
gh | -0.023 |
Token | Tim |
Token ID | 5045 |
Feature activation | -3.06e-03 |
Pos logit contributions | |
nowhere | +0.025 |
who | +0.020 |
blood | +0.018 |
crying | +0.017 |
weeping | +0.016 |
Neg logit contributions | |
ops | -0.058 |
iest | -0.057 |
hardly | -0.057 |
affles | -0.056 |
gh | -0.054 |
Token | words |
Token ID | 2456 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
nowhere | +0.026 |
who | +0.023 |
blood | +0.021 |
crying | +0.021 |
Blue | +0.020 |
Neg logit contributions | |
iest | -0.023 |
ops | -0.023 |
hardly | -0.022 |
affles | -0.022 |
gh | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | +9.01e-03 |
Pos logit contributions | |
iest | +0.093 |
ops | +0.093 |
hardly | +0.091 |
affles | +0.089 |
gh | +0.088 |
Neg logit contributions | |
nowhere | -0.039 |
who | -0.031 |
blood | -0.027 |
crying | -0.026 |
Blue | -0.025 |
Token | like |
Token ID | 588 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
hardly | +0.021 |
ops | +0.020 |
iest | +0.020 |
affles | +0.020 |
ves | +0.019 |
Neg logit contributions | |
nowhere | -0.012 |
who | -0.009 |
crying | -0.008 |
blood | -0.008 |
Blue | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
nowhere | +0.032 |
who | +0.024 |
blood | +0.023 |
crying | +0.022 |
bug | +0.022 |
Neg logit contributions | |
ops | -0.046 |
hardly | -0.045 |
iest | -0.045 |
affles | -0.044 |
gh | -0.043 |
Token | ban |
Token ID | 3820 |
Feature activation | +7.04e-02 |
Pos logit contributions | |
nowhere | +0.043 |
who | +0.037 |
crying | +0.036 |
blood | +0.035 |
weeping | +0.034 |
Neg logit contributions | |
ops | -0.052 |
iest | -0.051 |
hardly | -0.047 |
robat | -0.047 |
affles | -0.047 |
Token | ana |
Token ID | 2271 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
iest | +0.122 |
hardly | +0.117 |
ops | +0.114 |
affles | +0.113 |
zed | +0.111 |
Neg logit contributions | |
nowhere | -0.129 |
who | -0.127 |
Blue | -0.110 |
crying | -0.110 |
blood | -0.107 |
Token | " |
Token ID | 1 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
iest | +0.039 |
ops | +0.039 |
hardly | +0.038 |
affles | +0.037 |
gh | +0.036 |
Neg logit contributions | |
nowhere | -0.037 |
who | -0.035 |
Blue | -0.030 |
crying | -0.030 |
blood | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
nowhere | +0.023 |
who | +0.019 |
blood | +0.017 |
crying | +0.016 |
Blue | +0.016 |
Neg logit contributions | |
ops | -0.045 |
iest | -0.045 |
hardly | -0.043 |
affles | -0.042 |
gh | -0.042 |
Token | " |
Token ID | 366 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
nowhere | +0.018 |
who | +0.015 |
blood | +0.013 |
crying | +0.012 |
Blue | +0.012 |
Neg logit contributions | |
ops | -0.034 |
iest | -0.033 |
hardly | -0.033 |
affles | -0.032 |
gh | -0.032 |
Token | flower |
Token ID | 25547 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
nowhere | +0.030 |
who | +0.026 |
blood | +0.024 |
crying | +0.024 |
Blue | +0.023 |
Neg logit contributions | |
ops | -0.036 |
iest | -0.036 |
hardly | -0.034 |
affles | -0.034 |
gh | -0.033 |
Token | ". |
Token ID | 1911 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
hardly | +0.044 |
iest | +0.044 |
ops | +0.044 |
affles | +0.042 |
zed | +0.041 |
Neg logit contributions | |
nowhere | -0.034 |
who | -0.034 |
Blue | -0.028 |
crying | -0.028 |
blood | -0.027 |
Token | books |
Token ID | 3835 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
ops | +0.053 |
iest | +0.053 |
hardly | +0.052 |
affles | +0.051 |
gh | +0.050 |
Neg logit contributions | |
nowhere | -0.035 |
who | -0.030 |
blood | -0.026 |
crying | -0.026 |
Blue | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -3.77e-03 |
Pos logit contributions | |
ops | +0.070 |
iest | +0.070 |
affles | +0.069 |
hardly | +0.069 |
gh | +0.067 |
Neg logit contributions | |
nowhere | -0.030 |
who | -0.026 |
Blue | -0.021 |
crying | -0.019 |
blood | -0.019 |
Token | learn |
Token ID | 2193 |
Feature activation | +9.45e-03 |
Pos logit contributions | |
nowhere | +0.007 |
who | +0.007 |
Blue | +0.006 |
crying | +0.006 |
blood | +0.006 |
Neg logit contributions | |
iest | -0.007 |
ops | -0.007 |
affles | -0.006 |
gh | -0.006 |
hardly | -0.006 |
Token | new |
Token ID | 649 |
Feature activation | -5.66e-03 |
Pos logit contributions | |
iest | +0.021 |
ops | +0.021 |
hardly | +0.020 |
affles | +0.020 |
gh | +0.019 |
Neg logit contributions | |
nowhere | -0.014 |
who | -0.011 |
crying | -0.010 |
blood | -0.010 |
Blue | -0.010 |
Token | things |
Token ID | 1243 |
Feature activation | +8.61e-03 |
Pos logit contributions | |
who | +0.008 |
nowhere | +0.008 |
Blue | +0.007 |
a | +0.007 |
too | +0.007 |
Neg logit contributions | |
iest | -0.012 |
ops | -0.012 |
hardly | -0.012 |
affles | -0.011 |
zed | -0.011 |
Token | ." |
Token ID | 526 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
ops | +0.019 |
iest | +0.018 |
hardly | +0.018 |
affles | +0.018 |
zed | +0.018 |
Neg logit contributions | |
nowhere | -0.012 |
who | -0.012 |
Blue | -0.010 |
crying | -0.009 |
blood | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +1.33e-03 |
Pos logit contributions | |
ops | +0.048 |
hardly | +0.046 |
affles | +0.044 |
iest | +0.044 |
gh | +0.043 |
Neg logit contributions | |
nowhere | -0.027 |
who | -0.021 |
blood | -0.020 |
bug | -0.019 |
crying | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
ops | +0.003 |
iest | +0.003 |
hardly | +0.003 |
affles | +0.003 |
ves | +0.003 |
Neg logit contributions | |
nowhere | -0.002 |
who | -0.001 |
blood | -0.001 |
crying | -0.001 |
Blue | -0.001 |
Token | After |
Token ID | 3260 |
Feature activation | -1.15e-03 |
Pos logit contributions | |
nowhere | +0.018 |
who | +0.016 |
blood | +0.014 |
Blue | +0.014 |
crying | +0.014 |
Neg logit contributions | |
iest | -0.022 |
ops | -0.022 |
affles | -0.021 |
hardly | -0.021 |
gh | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
nowhere | +0.001 |
who | +0.001 |
bug | +0.001 |
blood | +0.001 |
crying | +0.001 |
Neg logit contributions | |
ops | -0.003 |
hardly | -0.003 |
iest | -0.003 |
gh | -0.003 |
zed | -0.003 |
Token | while |
Token ID | 981 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
ops | +0.114 |
iest | +0.114 |
affles | +0.109 |
hardly | +0.109 |
gh | +0.105 |
Neg logit contributions | |
nowhere | -0.098 |
who | -0.087 |
crying | -0.077 |
blood | -0.077 |
Blue | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +6.51e-03 |
Pos logit contributions | |
ops | +0.000 |
iest | +0.000 |
rd | +0.000 |
hardly | +0.000 |
vows | +0.000 |
Neg logit contributions | |
nowhere | -0.000 |
blood | -0.000 |
ter | -0.000 |
who | -0.000 |
Blue | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -7.62e-03 |
Pos logit contributions | |
ops | +0.015 |
iest | +0.015 |
hardly | +0.015 |
affles | +0.015 |
gh | +0.014 |
Neg logit contributions | |
nowhere | -0.008 |
who | -0.007 |
blood | -0.006 |
crying | -0.006 |
Blue | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
nowhere | +0.010 |
who | +0.008 |
ter | +0.007 |
blood | +0.007 |
bug | +0.007 |
Neg logit contributions | |
ops | -0.017 |
hardly | -0.017 |
affles | -0.017 |
ves | -0.017 |
iest | -0.017 |
Token | " |
Token ID | 1 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
nowhere | +0.033 |
who | +0.031 |
crying | +0.026 |
blood | +0.026 |
Blue | +0.026 |
Neg logit contributions | |
iest | -0.042 |
ops | -0.042 |
hardly | -0.040 |
affles | -0.040 |
gh | -0.039 |
Token | I |
Token ID | 40 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
nowhere | +0.051 |
who | +0.041 |
crying | +0.040 |
blood | +0.039 |
weeping | +0.039 |
Neg logit contributions | |
iest | -0.064 |
ops | -0.063 |
hardly | -0.062 |
ves | -0.058 |
gh | -0.058 |
Token | am |
Token ID | 716 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
who | +0.029 |
nowhere | +0.029 |
crying | +0.027 |
blood | +0.026 |
a | +0.024 |
Neg logit contributions | |
iest | -0.027 |
affles | -0.026 |
ops | -0.026 |
zed | -0.025 |
hardly | -0.024 |
Token | sorry |
Token ID | 7926 |
Feature activation | +9.32e-03 |
Pos logit contributions | |
hardly | +0.038 |
affles | +0.032 |
gh | +0.031 |
zed | +0.029 |
iest | +0.029 |
Neg logit contributions | |
nowhere | -0.047 |
tighter | -0.039 |
blood | -0.038 |
ety | -0.037 |
ipt | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
iest | +0.016 |
ops | +0.016 |
hardly | +0.015 |
affles | +0.015 |
gh | +0.015 |
Neg logit contributions | |
nowhere | -0.018 |
who | -0.016 |
blood | -0.015 |
crying | -0.014 |
Blue | -0.014 |
Token | Anna |
Token ID | 11735 |
Feature activation | -9.69e-03 |
Pos logit contributions | |
nowhere | +0.042 |
who | +0.036 |
blood | +0.031 |
crying | +0.031 |
Blue | +0.030 |
Neg logit contributions | |
ops | -0.069 |
iest | -0.069 |
hardly | -0.067 |
affles | -0.066 |
gh | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
nowhere | +0.013 |
who | +0.011 |
blood | +0.009 |
crying | +0.009 |
Blue | +0.009 |
Neg logit contributions | |
ops | -0.023 |
iest | -0.023 |
hardly | -0.022 |
affles | -0.022 |
gh | -0.021 |
Token | You |
Token ID | 921 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
nowhere | +0.020 |
who | +0.017 |
blood | +0.015 |
crying | +0.015 |
Blue | +0.014 |
Neg logit contributions | |
ops | -0.027 |
iest | -0.026 |
hardly | -0.026 |
affles | -0.025 |
gh | -0.024 |
Token | ice |
Token ID | 4771 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ops | +0.036 |
hardly | +0.036 |
iest | +0.036 |
affles | +0.035 |
gh | +0.034 |
Neg logit contributions | |
nowhere | -0.019 |
who | -0.017 |
blood | -0.014 |
crying | -0.014 |
Blue | -0.013 |
Token | cream |
Token ID | 8566 |
Feature activation | +4.51e-03 |
Pos logit contributions | |
hardly | +0.020 |
ops | +0.020 |
iest | +0.020 |
affles | +0.019 |
gh | +0.019 |
Neg logit contributions | |
nowhere | -0.016 |
who | -0.015 |
blood | -0.013 |
crying | -0.013 |
Blue | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +9.45e-03 |
Pos logit contributions | |
iest | +0.011 |
ops | +0.011 |
hardly | +0.010 |
affles | +0.010 |
gh | +0.010 |
Neg logit contributions | |
nowhere | -0.006 |
who | -0.005 |
blood | -0.004 |
crying | -0.004 |
Blue | -0.004 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
ops | +0.020 |
hardly | +0.020 |
iest | +0.019 |
affles | +0.019 |
zed | +0.018 |
Neg logit contributions | |
nowhere | -0.015 |
who | -0.013 |
blood | -0.012 |
crying | -0.011 |
Blue | -0.011 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
nowhere | +0.051 |
who | +0.044 |
blood | +0.042 |
crying | +0.041 |
weeping | +0.039 |
Neg logit contributions | |
ops | -0.074 |
iest | -0.073 |
hardly | -0.067 |
robat | -0.067 |
gh | -0.066 |
Token | there |
Token ID | 612 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
ops | +0.078 |
iest | +0.078 |
hardly | +0.076 |
affles | +0.074 |
gh | +0.073 |
Neg logit contributions | |
nowhere | -0.048 |
who | -0.042 |
blood | -0.036 |
crying | -0.036 |
Blue | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
ops | +0.077 |
iest | +0.076 |
affles | +0.075 |
hardly | +0.074 |
gh | +0.071 |
Neg logit contributions | |
nowhere | -0.068 |
who | -0.061 |
crying | -0.054 |
blood | -0.053 |
Blue | -0.053 |
Token | a |
Token ID | 257 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
nowhere | +0.022 |
blood | +0.016 |
who | +0.016 |
crying | +0.015 |
weeping | +0.015 |
Neg logit contributions | |
iest | -0.059 |
hardly | -0.057 |
affles | -0.057 |
ops | -0.057 |
rd | -0.057 |
Token | cup |
Token ID | 6508 |
Feature activation | +7.71e-03 |
Pos logit contributions | |
nowhere | +0.022 |
blood | +0.017 |
crying | +0.015 |
who | +0.014 |
Blue | +0.014 |
Neg logit contributions | |
hardly | -0.028 |
iest | -0.028 |
ops | -0.028 |
affles | -0.028 |
gh | -0.027 |
Token | that |
Token ID | 326 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ops | +0.017 |
iest | +0.017 |
hardly | +0.016 |
affles | +0.016 |
gh | +0.016 |
Neg logit contributions | |
nowhere | -0.011 |
who | -0.010 |
blood | -0.009 |
crying | -0.009 |
Blue | -0.008 |
Token | wanted |
Token ID | 2227 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
ops | +0.020 |
iest | +0.020 |
affles | +0.019 |
hardly | +0.019 |
gh | +0.018 |
Neg logit contributions | |
nowhere | -0.018 |
who | -0.016 |
crying | -0.014 |
blood | -0.014 |
Blue | -0.014 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
nowhere | +0.038 |
who | +0.034 |
blood | +0.031 |
crying | +0.031 |
Blue | +0.030 |
Neg logit contributions | |
ops | -0.039 |
iest | -0.039 |
hardly | -0.037 |
affles | -0.037 |
gh | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
ops | +0.025 |
iest | +0.025 |
hardly | +0.024 |
affles | +0.024 |
gh | +0.023 |
Neg logit contributions | |
nowhere | -0.024 |
who | -0.021 |
blood | -0.019 |
crying | -0.019 |
Blue | -0.019 |
Token | Jane |
Token ID | 12091 |
Feature activation | +4.84e-03 |
Pos logit contributions | |
iest | +0.035 |
hardly | +0.034 |
ops | +0.034 |
affles | +0.032 |
zed | +0.032 |
Neg logit contributions | |
nowhere | -0.024 |
who | -0.021 |
blood | -0.020 |
crying | -0.019 |
weeping | -0.018 |
Token | heard |
Token ID | 2982 |
Feature activation | +3.85e-03 |
Pos logit contributions | |
iest | +0.009 |
ops | +0.009 |
affles | +0.009 |
hardly | +0.008 |
gh | +0.008 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.008 |
crying | -0.007 |
blood | -0.007 |
Blue | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +4.11e-03 |
Pos logit contributions | |
iest | +0.010 |
hardly | +0.010 |
ops | +0.010 |
zed | +0.010 |
ves | +0.010 |
Neg logit contributions | |
nowhere | -0.004 |
who | -0.003 |
blood | -0.002 |
crying | -0.002 |
Blue | -0.002 |
Token | strange |
Token ID | 6283 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ops | +0.008 |
iest | +0.008 |
hardly | +0.007 |
affles | +0.007 |
gh | +0.007 |
Neg logit contributions | |
nowhere | -0.007 |
who | -0.006 |
blood | -0.006 |
crying | -0.006 |
Blue | -0.006 |
Token | noise |
Token ID | 7838 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
iest | +0.018 |
ops | +0.017 |
hardly | +0.017 |
gh | +0.016 |
affles | +0.016 |
Neg logit contributions | |
nowhere | -0.019 |
who | -0.017 |
blood | -0.016 |
Blue | -0.016 |
crying | -0.015 |
Token | behind |
Token ID | 2157 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
ops | +0.037 |
iest | +0.037 |
hardly | +0.036 |
affles | +0.035 |
gh | +0.035 |
Neg logit contributions | |
nowhere | -0.018 |
who | -0.015 |
blood | -0.013 |
crying | -0.013 |
Blue | -0.012 |
Token | her |
Token ID | 607 |
Feature activation | +4.60e-03 |
Pos logit contributions | |
nowhere | +0.025 |
who | +0.023 |
ety | +0.022 |
blood | +0.021 |
bug | +0.021 |
Neg logit contributions | |
hardly | -0.038 |
iest | -0.038 |
zed | -0.037 |
ops | -0.036 |
gh | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -8.05e-03 |
Pos logit contributions | |
ops | +0.010 |
iest | +0.010 |
hardly | +0.010 |
affles | +0.010 |
gh | +0.010 |
Neg logit contributions | |
nowhere | -0.007 |
who | -0.006 |
blood | -0.005 |
crying | -0.005 |
Blue | -0.005 |
Token | She |
Token ID | 1375 |
Feature activation | +2.66e-03 |
Pos logit contributions | |
nowhere | +0.013 |
blood | +0.011 |
who | +0.011 |
ter | +0.010 |
tighter | +0.010 |
Neg logit contributions | |
hardly | -0.016 |
zed | -0.014 |
rd | -0.014 |
ops | -0.014 |
iest | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
ops | +0.003 |
iest | +0.003 |
hardly | +0.003 |
affles | +0.002 |
gh | +0.002 |
Neg logit contributions | |
nowhere | -0.002 |
who | -0.001 |
blood | -0.001 |
crying | -0.001 |
Blue | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | -2.29e-03 |
Pos logit contributions | |
nowhere | +0.017 |
who | +0.014 |
blood | +0.013 |
crying | +0.012 |
bug | +0.012 |
Neg logit contributions | |
hardly | -0.022 |
ops | -0.021 |
zed | -0.020 |
rd | -0.019 |
affles | -0.019 |
Token | softened |
Token ID | 41534 |
Feature activation | +8.93e-03 |
Pos logit contributions | |
nowhere | +0.004 |
who | +0.004 |
blood | +0.004 |
crying | +0.004 |
Blue | +0.003 |
Neg logit contributions | |
iest | -0.004 |
ops | -0.004 |
hardly | -0.004 |
affles | -0.004 |
gh | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -7.42e-03 |
Pos logit contributions | |
hardly | +0.019 |
iest | +0.019 |
zed | +0.019 |
gh | +0.018 |
ops | +0.018 |
Neg logit contributions | |
nowhere | -0.012 |
who | -0.010 |
blood | -0.009 |
bug | -0.009 |
Blue | -0.008 |
Token | so |
Token ID | 523 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
nowhere | +0.010 |
ter | +0.009 |
Blue | +0.008 |
bug | +0.008 |
who | +0.008 |
Neg logit contributions | |
hardly | -0.016 |
zed | -0.014 |
ops | -0.013 |
iest | -0.013 |
barely | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
hardly | +0.027 |
ops | +0.026 |
iest | +0.026 |
zed | +0.026 |
ves | +0.025 |
Neg logit contributions | |
nowhere | -0.034 |
who | -0.029 |
blood | -0.028 |
Blue | -0.027 |
bug | -0.026 |
Token | animal |
Token ID | 5044 |
Feature activation | +3.48e-04 |
Pos logit contributions | |
nowhere | +0.044 |
who | +0.037 |
bug | +0.034 |
Blue | +0.033 |
blood | +0.033 |
Neg logit contributions | |
iest | -0.064 |
hardly | -0.060 |
affles | -0.059 |
ops | -0.059 |
ves | -0.058 |
Token | knew |
Token ID | 2993 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
iest | +0.001 |
ops | +0.001 |
hardly | +0.001 |
zed | +0.001 |
affles | +0.001 |
Neg logit contributions | |
nowhere | -0.001 |
who | -0.001 |
blood | -0.001 |
Blue | -0.001 |
crying | -0.001 |
Token | she |
Token ID | 673 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
nowhere | +0.034 |
who | +0.028 |
crying | +0.027 |
blood | +0.026 |
Blue | +0.026 |
Neg logit contributions | |
iest | -0.043 |
hardly | -0.041 |
ops | -0.040 |
zed | -0.040 |
ves | -0.040 |
Token | wasn |
Token ID | 2492 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
nowhere | +0.057 |
who | +0.051 |
blood | +0.048 |
crying | +0.047 |
Blue | +0.046 |
Neg logit contributions | |
ops | -0.042 |
iest | -0.041 |
hardly | -0.040 |
affles | -0.039 |
gh | -0.037 |
Token | â |
Token ID | 22940 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
nowhere | +0.024 |
who | +0.022 |
blood | +0.020 |
crying | +0.020 |
Blue | +0.020 |
Neg logit contributions | |
iest | -0.023 |
ops | -0.023 |
affles | -0.022 |
hardly | -0.022 |
gh | -0.021 |
Token | something |
Token ID | 1223 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
nowhere | +0.016 |
who | +0.013 |
blood | +0.011 |
crying | +0.011 |
Blue | +0.010 |
Neg logit contributions | |
ops | -0.035 |
iest | -0.034 |
hardly | -0.034 |
affles | -0.033 |
gh | -0.032 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
ops | +0.054 |
affles | +0.050 |
iest | +0.050 |
hardly | +0.049 |
robat | +0.049 |
Neg logit contributions | |
nowhere | -0.029 |
who | -0.029 |
crying | -0.024 |
Blue | -0.023 |
blood | -0.023 |
Token | Ben |
Token ID | 3932 |
Feature activation | +6.18e-03 |
Pos logit contributions | |
hardly | +0.029 |
ops | +0.028 |
iest | +0.027 |
affles | +0.026 |
ves | +0.026 |
Neg logit contributions | |
nowhere | -0.020 |
who | -0.016 |
blood | -0.015 |
crying | -0.014 |
bug | -0.014 |
Token | said |
Token ID | 531 |
Feature activation | -1.40e-03 |
Pos logit contributions | |
iest | +0.012 |
ops | +0.011 |
affles | +0.011 |
gh | +0.010 |
hardly | +0.010 |
Neg logit contributions | |
nowhere | -0.012 |
who | -0.011 |
crying | -0.010 |
blood | -0.010 |
Blue | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
nowhere | +0.002 |
who | +0.002 |
blood | +0.002 |
Blue | +0.002 |
bug | +0.002 |
Neg logit contributions | |
iest | -0.003 |
hardly | -0.003 |
ops | -0.003 |
gh | -0.003 |
zed | -0.002 |
Token | excited |
Token ID | 6568 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
nowhere | +0.035 |
blood | +0.031 |
ety | +0.028 |
Blue | +0.027 |
who | +0.027 |
Neg logit contributions | |
hardly | -0.036 |
zed | -0.030 |
barely | -0.030 |
gh | -0.029 |
ops | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -1.79e-03 |
Pos logit contributions | |
ops | +0.024 |
iest | +0.024 |
hardly | +0.023 |
affles | +0.023 |
gh | +0.022 |
Neg logit contributions | |
nowhere | -0.025 |
who | -0.022 |
blood | -0.020 |
crying | -0.020 |
Blue | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +6.93e-03 |
Pos logit contributions | |
nowhere | +0.003 |
ter | +0.002 |
blood | +0.002 |
bug | +0.002 |
who | +0.002 |
Neg logit contributions | |
hardly | -0.004 |
ops | -0.003 |
ves | -0.003 |
iest | -0.003 |
zed | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -5.23e-03 |
Pos logit contributions | |
hardly | +0.015 |
ves | +0.015 |
ops | +0.015 |
gh | +0.014 |
iest | +0.014 |
Neg logit contributions | |
nowhere | -0.010 |
ter | -0.008 |
who | -0.008 |
blood | -0.007 |
crying | -0.007 |
Token | " |
Token ID | 1 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
nowhere | +0.009 |
who | +0.008 |
crying | +0.007 |
blood | +0.007 |
Blue | +0.007 |
Neg logit contributions | |
iest | -0.011 |
ops | -0.011 |
affles | -0.010 |
hardly | -0.010 |
gh | -0.010 |
Token | Maybe |
Token ID | 13300 |
Feature activation | -1.18e-03 |
Pos logit contributions | |
nowhere | +0.039 |
who | +0.033 |
blood | +0.030 |
crying | +0.030 |
Blue | +0.029 |
Neg logit contributions | |
iest | -0.058 |
ops | -0.058 |
hardly | -0.057 |
affles | -0.055 |
gh | -0.054 |
Token | her |
Token ID | 607 |
Feature activation | -4.85e-03 |
Pos logit contributions | |
nowhere | +0.005 |
who | +0.004 |
blood | +0.004 |
Blue | +0.004 |
crying | +0.003 |
Neg logit contributions | |
ops | -0.013 |
iest | -0.013 |
hardly | -0.012 |
affles | -0.012 |
gh | -0.012 |
Token | violin |
Token ID | 38283 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
nowhere | +0.008 |
who | +0.008 |
blood | +0.007 |
crying | +0.007 |
Blue | +0.007 |
Neg logit contributions | |
ops | -0.009 |
iest | -0.009 |
hardly | -0.009 |
affles | -0.009 |
gh | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -2.40e-05 |
Pos logit contributions | |
nowhere | +0.017 |
who | +0.014 |
blood | +0.012 |
crying | +0.012 |
Blue | +0.012 |
Neg logit contributions | |
ops | -0.033 |
iest | -0.033 |
hardly | -0.032 |
affles | -0.031 |
gh | -0.031 |
Token | She |
Token ID | 1375 |
Feature activation | -1.84e-03 |
Pos logit contributions | |
nowhere | +0.000 |
who | +0.000 |
blood | +0.000 |
ter | +0.000 |
crying | +0.000 |
Neg logit contributions | |
hardly | -0.000 |
ops | -0.000 |
zed | -0.000 |
iest | -0.000 |
affles | -0.000 |
Token | knew |
Token ID | 2993 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
nowhere | +0.003 |
who | +0.003 |
crying | +0.002 |
blood | +0.002 |
a | +0.002 |
Neg logit contributions | |
iest | -0.004 |
affles | -0.004 |
ops | -0.004 |
rd | -0.004 |
gh | -0.004 |
Token | that |
Token ID | 326 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
nowhere | +0.019 |
who | +0.015 |
bug | +0.015 |
blood | +0.015 |
crying | +0.014 |
Neg logit contributions | |
iest | -0.025 |
hardly | -0.025 |
ops | -0.024 |
gh | -0.023 |
zed | -0.023 |
Token | her |
Token ID | 607 |
Feature activation | +6.28e-03 |
Pos logit contributions | |
nowhere | +0.018 |
who | +0.016 |
blood | +0.014 |
Blue | +0.014 |
bug | +0.014 |
Neg logit contributions | |
ops | -0.025 |
iest | -0.025 |
hardly | -0.025 |
gh | -0.024 |
affles | -0.024 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
ops | +0.014 |
iest | +0.013 |
hardly | +0.013 |
affles | +0.013 |
gh | +0.013 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.008 |
blood | -0.007 |
crying | -0.007 |
Blue | -0.007 |
Token | loved |
Token ID | 6151 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
nowhere | +0.024 |
who | +0.021 |
blood | +0.020 |
crying | +0.020 |
Blue | +0.019 |
Neg logit contributions | |
ops | -0.017 |
iest | -0.017 |
hardly | -0.016 |
affles | -0.016 |
gh | -0.015 |
Token | her |
Token ID | 607 |
Feature activation | -9.12e-03 |
Pos logit contributions | |
nowhere | +0.048 |
who | +0.040 |
blood | +0.035 |
Blue | +0.034 |
crying | +0.034 |
Neg logit contributions | |
ops | -0.088 |
iest | -0.087 |
hardly | -0.086 |
affles | -0.085 |
gh | -0.084 |
Token | no |
Token ID | 645 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
nowhere | +0.009 |
who | +0.007 |
blood | +0.006 |
crying | +0.006 |
Blue | +0.006 |
Neg logit contributions | |
ops | -0.024 |
iest | -0.024 |
hardly | -0.024 |
affles | -0.024 |
gh | -0.023 |
Token | day |
Token ID | 1110 |
Feature activation | -2.97e-03 |
Pos logit contributions | |
ops | +0.014 |
iest | +0.013 |
hardly | +0.013 |
affles | +0.013 |
gh | +0.012 |
Neg logit contributions | |
nowhere | -0.022 |
who | -0.020 |
crying | -0.018 |
Blue | -0.018 |
blood | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
nowhere | +0.005 |
who | +0.005 |
crying | +0.004 |
Blue | +0.004 |
blood | +0.004 |
Neg logit contributions | |
ops | -0.006 |
affles | -0.006 |
iest | -0.006 |
hardly | -0.006 |
gh | -0.005 |
Token | Jack |
Token ID | 3619 |
Feature activation | +1.93e-03 |
Pos logit contributions | |
iest | +0.030 |
ops | +0.030 |
hardly | +0.029 |
affles | +0.028 |
zed | +0.028 |
Neg logit contributions | |
nowhere | -0.015 |
who | -0.012 |
blood | -0.011 |
crying | -0.011 |
weeping | -0.010 |
Token | carefully |
Token ID | 7773 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
iest | +0.005 |
ops | +0.005 |
affles | +0.005 |
hardly | +0.005 |
gh | +0.005 |
Neg logit contributions | |
nowhere | -0.002 |
who | -0.002 |
crying | -0.001 |
Blue | -0.001 |
blood | -0.001 |
Token | unt |
Token ID | 1418 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
iest | +0.022 |
ops | +0.021 |
affles | +0.021 |
hardly | +0.020 |
robat | +0.020 |
Neg logit contributions | |
nowhere | -0.019 |
who | -0.017 |
crying | -0.015 |
blood | -0.015 |
Blue | -0.015 |
Token | ied |
Token ID | 798 |
Feature activation | -9.30e-03 |
Pos logit contributions | |
iest | +0.084 |
ops | +0.084 |
hardly | +0.079 |
affles | +0.078 |
gh | +0.075 |
Neg logit contributions | |
nowhere | -0.138 |
who | -0.126 |
blood | -0.116 |
crying | -0.116 |
Blue | -0.115 |
Token | the |
Token ID | 262 |
Feature activation | +7.65e-03 |
Pos logit contributions | |
nowhere | +0.012 |
who | +0.010 |
blood | +0.009 |
crying | +0.009 |
bug | +0.008 |
Neg logit contributions | |
iest | -0.021 |
hardly | -0.021 |
ops | -0.021 |
zed | -0.020 |
gh | -0.020 |
Token | string |
Token ID | 4731 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
iest | +0.014 |
ops | +0.013 |
hardly | +0.013 |
gh | +0.012 |
vows | +0.012 |
Neg logit contributions | |
nowhere | -0.014 |
who | -0.013 |
blood | -0.011 |
Blue | -0.011 |
crying | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +4.56e-03 |
Pos logit contributions | |
nowhere | +0.021 |
who | +0.016 |
blood | +0.015 |
crying | +0.014 |
Blue | +0.014 |
Neg logit contributions | |
iest | -0.041 |
ops | -0.041 |
hardly | -0.040 |
gh | -0.038 |
affles | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
iest | +0.009 |
ops | +0.009 |
affles | +0.009 |
hardly | +0.009 |
gh | +0.009 |
Neg logit contributions | |
nowhere | -0.008 |
who | -0.007 |
crying | -0.006 |
Blue | -0.006 |
blood | -0.006 |
Token | symbol |
Token ID | 6194 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
nowhere | +0.021 |
who | +0.018 |
blood | +0.017 |
late | +0.017 |
crying | +0.017 |
Neg logit contributions | |
iest | -0.023 |
ops | -0.022 |
hardly | -0.021 |
vows | -0.020 |
gh | -0.020 |
Token | came |
Token ID | 1625 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
nowhere | +0.010 |
Blue | +0.009 |
blood | +0.009 |
who | +0.009 |
ter | +0.008 |
Neg logit contributions | |
ops | -0.009 |
hardly | -0.009 |
iest | -0.009 |
gh | -0.009 |
zed | -0.008 |
Token | in |
Token ID | 287 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
Blue | +0.072 |
blood | +0.064 |
nowhere | +0.063 |
Tweet | +0.061 |
Twe | +0.061 |
Neg logit contributions | |
hardly | -0.061 |
gh | -0.054 |
activities | -0.054 |
ines | -0.054 |
rd | -0.054 |
Token | and |
Token ID | 290 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
nowhere | +0.048 |
ter | +0.042 |
Blue | +0.042 |
bug | +0.039 |
ety | +0.038 |
Neg logit contributions | |
hardly | -0.087 |
rd | -0.083 |
iest | -0.083 |
ops | -0.081 |
zed | -0.081 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
nowhere | +0.042 |
blood | +0.039 |
Blue | +0.037 |
ter | +0.036 |
itself | +0.036 |
Neg logit contributions | |
hardly | -0.025 |
ops | -0.023 |
iest | -0.023 |
zed | -0.023 |
ves | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
nowhere | +0.050 |
blood | +0.040 |
who | +0.039 |
Blue | +0.038 |
ter | +0.037 |
Neg logit contributions | |
ops | -0.050 |
iest | -0.049 |
gh | -0.049 |
affles | -0.049 |
zed | -0.048 |
Token | " |
Token ID | 366 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
nowhere | +0.051 |
blood | +0.043 |
bug | +0.043 |
ety | +0.041 |
ipt | +0.040 |
Neg logit contributions | |
hardly | -0.037 |
zed | -0.029 |
barely | -0.028 |
gh | -0.028 |
iest | -0.027 |
Token | What |
Token ID | 2061 |
Feature activation | +5.55e-03 |
Pos logit contributions | |
Blue | +0.084 |
weeping | +0.077 |
nowhere | +0.073 |
Tweet | +0.072 |
crying | +0.071 |
Neg logit contributions | |
iest | -0.100 |
robat | -0.094 |
gh | -0.092 |
ede | -0.091 |
ops | -0.091 |
Token | 's |
Token ID | 338 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
ops | +0.010 |
iest | +0.010 |
hardly | +0.010 |
affles | +0.009 |
gh | +0.009 |
Neg logit contributions | |
nowhere | -0.010 |
who | -0.009 |
blood | -0.009 |
crying | -0.009 |
Blue | -0.008 |
Token | wrong |
Token ID | 2642 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
gh | +0.057 |
hardly | +0.056 |
iest | +0.056 |
affles | +0.055 |
ops | +0.055 |
Neg logit contributions | |
nowhere | -0.050 |
who | -0.044 |
bug | -0.042 |
blood | -0.041 |
crying | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
ops | +0.021 |
iest | +0.021 |
hardly | +0.020 |
affles | +0.020 |
gh | +0.019 |
Neg logit contributions | |
nowhere | -0.020 |
who | -0.018 |
crying | -0.016 |
blood | -0.016 |
Blue | -0.016 |
Token | sweet |
Token ID | 6029 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
nowhere | +0.015 |
who | +0.011 |
crying | +0.010 |
blood | +0.010 |
weeping | +0.009 |
Neg logit contributions | |
iest | -0.027 |
hardly | -0.027 |
affles | -0.026 |
ops | -0.026 |
ves | -0.025 |
Token | nod |
Token ID | 18666 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
iest | +0.056 |
ops | +0.055 |
hardly | +0.053 |
affles | +0.052 |
gh | +0.051 |
Neg logit contributions | |
nowhere | -0.076 |
who | -0.070 |
blood | -0.064 |
crying | -0.063 |
Blue | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -6.11e-03 |
Pos logit contributions | |
ops | +0.023 |
iest | +0.022 |
hardly | +0.022 |
gh | +0.022 |
rd | +0.021 |
Neg logit contributions | |
nowhere | -0.017 |
who | -0.013 |
blood | -0.013 |
Blue | -0.012 |
crying | -0.012 |
Token | They |
Token ID | 1119 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
nowhere | +0.009 |
who | +0.007 |
blood | +0.007 |
crying | +0.007 |
weeping | +0.006 |
Neg logit contributions | |
ops | -0.013 |
hardly | -0.013 |
iest | -0.013 |
affles | -0.013 |
rd | -0.012 |
Token | say |
Token ID | 910 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
iest | +0.032 |
ops | +0.031 |
gh | +0.030 |
hardly | +0.029 |
rd | +0.029 |
Neg logit contributions | |
nowhere | -0.033 |
who | -0.033 |
blood | -0.029 |
Blue | -0.028 |
crying | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
ops | +0.030 |
iest | +0.030 |
hardly | +0.029 |
affles | +0.029 |
gh | +0.028 |
Neg logit contributions | |
nowhere | -0.031 |
who | -0.028 |
blood | -0.025 |
crying | -0.025 |
Blue | -0.025 |
Token | " |
Token ID | 366 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
nowhere | +0.023 |
blood | +0.017 |
who | +0.017 |
bug | +0.015 |
crying | +0.015 |
Neg logit contributions | |
hardly | -0.034 |
ops | -0.033 |
zed | -0.032 |
affles | -0.031 |
ves | -0.031 |
Token | Deal |
Token ID | 45776 |
Feature activation | +5.59e-03 |
Pos logit contributions | |
nowhere | +0.059 |
who | +0.056 |
Blue | +0.048 |
blood | +0.047 |
crying | +0.046 |
Neg logit contributions | |
iest | -0.063 |
ops | -0.062 |
affles | -0.061 |
hardly | -0.060 |
zed | -0.058 |
Token | ." |
Token ID | 526 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
ops | +0.011 |
affles | +0.011 |
iest | +0.011 |
hardly | +0.010 |
robat | +0.010 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.009 |
where | -0.008 |
too | -0.008 |
crying | -0.008 |
Token | They |
Token ID | 1119 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
ops | +0.043 |
hardly | +0.042 |
rd | +0.040 |
iest | +0.039 |
ves | +0.039 |
Neg logit contributions | |
nowhere | -0.033 |
blood | -0.025 |
ter | -0.025 |
crying | -0.024 |
weeping | -0.024 |
Token | go |
Token ID | 467 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
iest | +0.026 |
ops | +0.025 |
hardly | +0.024 |
gh | +0.024 |
rd | +0.024 |
Neg logit contributions | |
nowhere | -0.024 |
who | -0.023 |
blood | -0.020 |
Blue | -0.020 |
crying | -0.020 |
Token | with |
Token ID | 351 |
Feature activation | -3.53e-03 |
Pos logit contributions | |
nowhere | +0.028 |
who | +0.020 |
blood | +0.019 |
Blue | +0.018 |
crying | +0.017 |
Neg logit contributions | |
ops | -0.062 |
iest | -0.062 |
hardly | -0.060 |
affles | -0.060 |
rd | -0.058 |
Token | something |
Token ID | 1223 |
Feature activation | +9.82e-03 |
Pos logit contributions | |
nowhere | +0.029 |
who | +0.025 |
bug | +0.022 |
crying | +0.021 |
blood | +0.021 |
Neg logit contributions | |
iest | -0.042 |
ops | -0.040 |
hardly | -0.039 |
gh | -0.039 |
affles | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +9.05e-03 |
Pos logit contributions | |
iest | +0.022 |
ops | +0.021 |
hardly | +0.021 |
affles | +0.021 |
gh | +0.020 |
Neg logit contributions | |
nowhere | -0.014 |
who | -0.012 |
blood | -0.010 |
crying | -0.010 |
Blue | -0.010 |
Token | The |
Token ID | 383 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
ops | +0.019 |
hardly | +0.019 |
iest | +0.018 |
zed | +0.018 |
affles | +0.017 |
Neg logit contributions | |
nowhere | -0.014 |
blood | -0.012 |
crying | -0.011 |
who | -0.011 |
weeping | -0.010 |
Token | end |
Token ID | 886 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
nowhere | +0.023 |
who | +0.020 |
bug | +0.018 |
blood | +0.018 |
Blue | +0.018 |
Neg logit contributions | |
ops | -0.024 |
iest | -0.023 |
affles | -0.023 |
hardly | -0.022 |
robat | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +7.15e-03 |
Pos logit contributions | |
nowhere | +0.041 |
who | +0.037 |
blood | +0.030 |
crying | +0.030 |
Blue | +0.030 |
Neg logit contributions | |
ops | -0.073 |
iest | -0.072 |
hardly | -0.071 |
affles | -0.070 |
gh | -0.068 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
ops | +0.014 |
hardly | +0.014 |
iest | +0.013 |
zed | +0.013 |
robat | +0.013 |
Neg logit contributions | |
nowhere | -0.011 |
blood | -0.009 |
crying | -0.009 |
weeping | -0.009 |
ter | -0.009 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
nowhere | +0.052 |
who | +0.046 |
blood | +0.045 |
crying | +0.044 |
weeping | +0.043 |
Neg logit contributions | |
ops | -0.081 |
iest | -0.078 |
hardly | -0.075 |
gh | -0.074 |
ves | -0.073 |
Token | upon |
Token ID | 2402 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
iest | +0.078 |
ops | +0.077 |
hardly | +0.075 |
affles | +0.074 |
gh | +0.072 |
Neg logit contributions | |
nowhere | -0.050 |
who | -0.043 |
blood | -0.038 |
crying | -0.037 |
Blue | -0.037 |
Token | a |
Token ID | 257 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
nowhere | +0.025 |
blood | +0.018 |
weeping | +0.018 |
crying | +0.018 |
who | +0.018 |
Neg logit contributions | |
hardly | -0.071 |
iest | -0.068 |
ops | -0.067 |
zed | -0.067 |
gh | -0.066 |
Token | time |
Token ID | 640 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
ops | +0.036 |
iest | +0.036 |
hardly | +0.034 |
affles | +0.033 |
gh | +0.031 |
Neg logit contributions | |
nowhere | -0.067 |
who | -0.062 |
blood | -0.057 |
crying | -0.057 |
Blue | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | +9.99e-03 |
Pos logit contributions | |
ops | +0.021 |
iest | +0.021 |
hardly | +0.020 |
gh | +0.019 |
affles | +0.019 |
Neg logit contributions | |
nowhere | -0.025 |
who | -0.022 |
blood | -0.021 |
crying | -0.020 |
Blue | -0.020 |
Token | shop |
Token ID | 6128 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
nowhere | +0.027 |
who | +0.023 |
blood | +0.022 |
bug | +0.021 |
crying | +0.021 |
Neg logit contributions | |
iest | -0.032 |
ops | -0.031 |
hardly | -0.030 |
affles | -0.029 |
rd | -0.029 |
Token | keeper |
Token ID | 13884 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
nowhere | +0.014 |
who | +0.012 |
blood | +0.011 |
crying | +0.010 |
Blue | +0.010 |
Neg logit contributions | |
iest | -0.029 |
ops | -0.029 |
hardly | -0.028 |
affles | -0.027 |
gh | -0.027 |
Token | every |
Token ID | 790 |
Feature activation | +2.68e-03 |
Pos logit contributions | |
nowhere | +0.035 |
who | +0.030 |
crying | +0.025 |
blood | +0.025 |
Blue | +0.025 |
Neg logit contributions | |
ops | -0.062 |
iest | -0.062 |
hardly | -0.061 |
affles | -0.060 |
gh | -0.059 |
Token | week |
Token ID | 1285 |
Feature activation | -4.11e-03 |
Pos logit contributions | |
iest | +0.003 |
ops | +0.003 |
hardly | +0.003 |
affles | +0.003 |
gh | +0.003 |
Neg logit contributions | |
nowhere | -0.007 |
who | -0.006 |
blood | -0.006 |
Blue | -0.006 |
crying | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +6.58e-03 |
Pos logit contributions | |
nowhere | +0.006 |
who | +0.005 |
crying | +0.004 |
Blue | +0.004 |
blood | +0.004 |
Neg logit contributions | |
iest | -0.009 |
ops | -0.009 |
affles | -0.009 |
hardly | -0.009 |
robat | -0.009 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
ops | +0.015 |
hardly | +0.015 |
iest | +0.014 |
zed | +0.014 |
affles | +0.014 |
Neg logit contributions | |
nowhere | -0.009 |
who | -0.008 |
blood | -0.007 |
crying | -0.007 |
ter | -0.007 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
nowhere | +0.056 |
blood | +0.050 |
who | +0.049 |
crying | +0.047 |
weeping | +0.046 |
Neg logit contributions | |
ops | -0.077 |
iest | -0.075 |
hardly | -0.069 |
robat | -0.069 |
gh | -0.068 |
Token | upon |
Token ID | 2402 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
iest | +0.087 |
ops | +0.087 |
hardly | +0.084 |
affles | +0.083 |
gh | +0.081 |
Neg logit contributions | |
nowhere | -0.054 |
who | -0.047 |
blood | -0.041 |
crying | -0.040 |
Blue | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
nowhere | +0.022 |
blood | +0.015 |
crying | +0.015 |
weeping | +0.015 |
who | +0.015 |
Neg logit contributions | |
hardly | -0.063 |
iest | -0.062 |
ops | -0.060 |
zed | -0.060 |
gh | -0.060 |
Token | time |
Token ID | 640 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
ops | +0.037 |
iest | +0.036 |
hardly | +0.035 |
affles | +0.034 |
gh | +0.033 |
Neg logit contributions | |
nowhere | -0.074 |
who | -0.068 |
blood | -0.065 |
crying | -0.064 |
Blue | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
ops | +0.026 |
iest | +0.026 |
hardly | +0.025 |
gh | +0.024 |
affles | +0.024 |
Neg logit contributions | |
nowhere | -0.031 |
who | -0.028 |
blood | -0.027 |
crying | -0.026 |
Blue | -0.025 |
Token | lived |
Token ID | 5615 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
who | +0.001 |
nowhere | +0.001 |
Blue | +0.001 |
crying | +0.001 |
bug | +0.001 |
Neg logit contributions | |
ops | -0.001 |
iest | -0.001 |
gh | -0.001 |
affles | -0.001 |
hardly | -0.001 |
Token | happily |
Token ID | 18177 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
nowhere | +0.072 |
blood | +0.060 |
who | +0.058 |
crying | +0.056 |
weeping | +0.054 |
Neg logit contributions | |
iest | -0.089 |
hardly | -0.088 |
zed | -0.087 |
gh | -0.086 |
ops | -0.085 |
Token | ever |
Token ID | 1683 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
nowhere | +0.081 |
who | +0.072 |
blood | +0.068 |
crying | +0.066 |
Blue | +0.065 |
Neg logit contributions | |
ops | -0.073 |
iest | -0.072 |
hardly | -0.069 |
affles | -0.067 |
gh | -0.066 |
Token | after |
Token ID | 706 |
Feature activation | -4.75e-03 |
Pos logit contributions | |
nowhere | +0.063 |
who | +0.057 |
blood | +0.053 |
crying | +0.052 |
Blue | +0.051 |
Neg logit contributions | |
ops | -0.056 |
iest | -0.055 |
hardly | -0.053 |
affles | -0.052 |
gh | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +6.96e-03 |
Pos logit contributions | |
nowhere | +0.007 |
who | +0.006 |
crying | +0.005 |
blood | +0.005 |
Blue | +0.005 |
Neg logit contributions | |
iest | -0.011 |
ops | -0.011 |
hardly | -0.010 |
affles | -0.010 |
gh | -0.010 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
hardly | +0.014 |
ops | +0.014 |
rd | +0.013 |
zed | +0.013 |
iest | +0.012 |
Neg logit contributions | |
nowhere | -0.010 |
blood | -0.009 |
crying | -0.008 |
ter | -0.008 |
weeping | -0.008 |
Token | One |
Token ID | 3198 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
nowhere | +0.051 |
who | +0.046 |
blood | +0.045 |
crying | +0.044 |
tighter | +0.043 |
Neg logit contributions | |
ops | -0.075 |
iest | -0.073 |
gh | -0.069 |
robat | -0.068 |
hardly | -0.068 |
Token | day |
Token ID | 1110 |
Feature activation | +9.41e-03 |
Pos logit contributions | |
ops | +0.083 |
affles | +0.077 |
hardly | +0.076 |
iest | +0.075 |
robat | +0.074 |
Neg logit contributions | |
nowhere | -0.101 |
who | -0.093 |
a | -0.085 |
crying | -0.083 |
too | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | +6.97e-03 |
Pos logit contributions | |
affles | +0.020 |
ops | +0.020 |
robat | +0.020 |
iest | +0.020 |
hardly | +0.020 |
Neg logit contributions | |
nowhere | -0.014 |
who | -0.012 |
crying | -0.011 |
Blue | -0.011 |
a | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -1.62e-03 |
Pos logit contributions | |
hardly | +0.016 |
iest | +0.015 |
gh | +0.015 |
ops | +0.015 |
zed | +0.014 |
Neg logit contributions | |
blood | -0.007 |
nowhere | -0.007 |
weeping | -0.006 |
crying | -0.006 |
who | -0.006 |
Token | little |
Token ID | 1310 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
nowhere | +0.002 |
blood | +0.002 |
who | +0.002 |
crying | +0.002 |
weeping | +0.002 |
Neg logit contributions | |
iest | -0.003 |
ops | -0.003 |
hardly | -0.003 |
affles | -0.003 |
gh | -0.003 |