Token | Tom |
Token ID | 4186 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
Offic | +0.007 |
vet | +0.007 |
bandits | +0.007 |
Amen | +0.007 |
knight | +0.007 |
Neg logit contributions | |
ying | -0.010 |
mm | -0.009 |
uch | -0.009 |
ig | -0.009 |
belongs | -0.009 |
Token | cried |
Token ID | 16896 |
Feature activation | +7.66e-02 |
Pos logit contributions | |
mm | +0.069 |
ops | +0.066 |
n | +0.063 |
uch | +0.063 |
happen | +0.062 |
Neg logit contributions | |
Amen | -0.090 |
Ar | -0.077 |
lair | -0.076 |
crew | -0.075 |
generation | -0.073 |
Token | and |
Token ID | 290 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
Amen | +0.168 |
crew | +0.153 |
Ar | +0.153 |
Offic | +0.152 |
known | +0.148 |
Neg logit contributions | |
mm | -0.069 |
happen | -0.063 |
ops | -0.060 |
ying | -0.057 |
tons | -0.056 |
Token | begged |
Token ID | 32946 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
Amen | +0.035 |
Ar | +0.030 |
crew | +0.030 |
lair | +0.029 |
knight | +0.029 |
Neg logit contributions | |
mm | -0.033 |
happen | -0.031 |
tons | -0.030 |
n | -0.030 |
ops | -0.029 |
Token | for |
Token ID | 329 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
Amen | +0.085 |
crew | +0.076 |
Ar | +0.073 |
bandits | +0.072 |
knight | +0.072 |
Neg logit contributions | |
mm | -0.076 |
ops | -0.067 |
uch | -0.067 |
happen | -0.064 |
n | -0.063 |
Token | mercy |
Token ID | 17703 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
Amen | +0.044 |
Ar | +0.040 |
unts | +0.037 |
lair | +0.037 |
crew | +0.037 |
Neg logit contributions | |
mm | -0.041 |
happen | -0.041 |
tons | -0.037 |
n | -0.036 |
next | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -2.97e-03 |
Pos logit contributions | |
Amen | +0.435 |
crew | +0.395 |
Ar | +0.379 |
knight | +0.379 |
lair | +0.375 |
Neg logit contributions | |
mm | -0.289 |
uch | -0.244 |
ops | -0.241 |
€ | -0.233 |
happen | -0.231 |
Token | They |
Token ID | 1119 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
mm | +0.005 |
tons | +0.004 |
ops | +0.004 |
n | +0.004 |
€ | +0.004 |
Neg logit contributions | |
Amen | -0.004 |
knight | -0.004 |
crew | -0.004 |
bandits | -0.004 |
lair | -0.004 |
Token | wished |
Token ID | 16555 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
Amen | +0.114 |
lair | +0.101 |
Ar | +0.100 |
generation | +0.099 |
crew | +0.097 |
Neg logit contributions | |
n | -0.063 |
mm | -0.061 |
liv | -0.059 |
ops | -0.058 |
tons | -0.057 |
Token | they |
Token ID | 484 |
Feature activation | -1.61e-03 |
Pos logit contributions | |
Amen | +0.021 |
crew | +0.019 |
known | +0.018 |
firefighter | +0.017 |
bandits | +0.017 |
Neg logit contributions | |
mm | -0.028 |
uch | -0.028 |
ops | -0.026 |
€ | -0.026 |
belongs | -0.025 |
Token | had |
Token ID | 550 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
mm | +0.002 |
ops | +0.002 |
n | +0.002 |
happen | +0.002 |
uch | +0.002 |
Neg logit contributions | |
Amen | -0.003 |
crew | -0.003 |
Ar | -0.003 |
knight | -0.002 |
lair | -0.002 |
Token | forgot |
Token ID | 16453 |
Feature activation | +6.64e-02 |
Pos logit contributions | |
Amen | +0.059 |
crew | +0.050 |
Ar | +0.050 |
Offic | +0.047 |
bandits | +0.046 |
Neg logit contributions | |
mm | -0.074 |
happen | -0.070 |
ops | -0.068 |
uch | -0.068 |
belongs | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
Amen | +0.106 |
crew | +0.096 |
bandits | +0.092 |
knight | +0.091 |
Ar | +0.091 |
Neg logit contributions | |
mm | -0.104 |
ops | -0.093 |
uch | -0.091 |
happen | -0.088 |
€ | -0.087 |
Token | lion |
Token ID | 18744 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
mm | +0.047 |
happen | +0.044 |
n | +0.042 |
tons | +0.042 |
ops | +0.041 |
Neg logit contributions | |
Amen | -0.053 |
crew | -0.046 |
Ar | -0.045 |
Offic | -0.045 |
bandits | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | -6.60e-03 |
Pos logit contributions | |
Amen | +0.126 |
crew | +0.119 |
knight | +0.115 |
lair | +0.110 |
bandits | +0.109 |
Neg logit contributions | |
mm | -0.131 |
uch | -0.113 |
ops | -0.112 |
€ | -0.112 |
liv | -0.111 |
Token | his |
Token ID | 465 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
mm | +0.009 |
happen | +0.008 |
belongs | +0.008 |
n | +0.008 |
tons | +0.008 |
Neg logit contributions | |
Amen | -0.012 |
Ar | -0.010 |
crew | -0.010 |
Offic | -0.010 |
unts | -0.010 |
Token | mercy |
Token ID | 17703 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
mm | +0.077 |
happen | +0.071 |
ops | +0.069 |
belongs | +0.067 |
n | +0.067 |
Neg logit contributions | |
Amen | -0.076 |
Ar | -0.065 |
Offic | -0.065 |
crew | -0.063 |
bandits | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
Amen | +0.487 |
bandits | +0.458 |
Offic | +0.454 |
Ar | +0.450 |
crew | +0.435 |
Neg logit contributions | |
mm | -0.184 |
happen | -0.176 |
n | -0.174 |
ops | -0.172 |
tons | -0.160 |
Token | He |
Token ID | 679 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
mm | +0.022 |
€ | +0.019 |
n | +0.018 |
tons | +0.018 |
existed | +0.017 |
Neg logit contributions | |
Amen | -0.021 |
crew | -0.020 |
bandits | -0.020 |
messenger | -0.019 |
Offic | -0.018 |
Token | always |
Token ID | 1464 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
Amen | +0.080 |
crew | +0.068 |
bandits | +0.066 |
Ar | +0.065 |
Offic | +0.065 |
Neg logit contributions | |
mm | -0.086 |
belongs | -0.080 |
n | -0.080 |
ops | -0.079 |
happen | -0.077 |
Token | tried |
Token ID | 3088 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
Amen | +0.103 |
crew | +0.089 |
Ar | +0.089 |
bandits | +0.086 |
Offic | +0.085 |
Neg logit contributions | |
mm | -0.123 |
happen | -0.114 |
belongs | -0.114 |
uch | -0.111 |
ops | -0.110 |
Token | to |
Token ID | 284 |
Feature activation | +6.52e-02 |
Pos logit contributions | |
Amen | +0.064 |
crew | +0.055 |
Ar | +0.054 |
Offic | +0.052 |
bandits | +0.052 |
Neg logit contributions | |
mm | -0.077 |
happen | -0.070 |
ops | -0.069 |
n | -0.068 |
uch | -0.067 |
Token | very |
Token ID | 845 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
mm | +0.048 |
happen | +0.044 |
uch | +0.043 |
n | +0.042 |
ops | +0.042 |
Neg logit contributions | |
Amen | -0.046 |
crew | -0.041 |
Ar | -0.040 |
bandits | -0.038 |
Offic | -0.038 |
Token | scared |
Token ID | 12008 |
Feature activation | +4.88e-02 |
Pos logit contributions | |
Amen | +0.023 |
crew | +0.020 |
Ar | +0.019 |
bandits | +0.019 |
knight | +0.019 |
Neg logit contributions | |
mm | -0.031 |
ops | -0.028 |
happen | -0.027 |
uch | -0.027 |
n | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
Amen | +0.075 |
crew | +0.069 |
knight | +0.068 |
bandits | +0.066 |
unts | +0.064 |
Neg logit contributions | |
mm | -0.079 |
uch | -0.071 |
ops | -0.070 |
€ | -0.068 |
liv | -0.067 |
Token | begged |
Token ID | 32946 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
mm | +0.028 |
ops | +0.025 |
happen | +0.025 |
uch | +0.024 |
n | +0.024 |
Neg logit contributions | |
Amen | -0.025 |
crew | -0.022 |
Ar | -0.022 |
bandits | -0.021 |
knight | -0.021 |
Token | for |
Token ID | 329 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
Amen | +0.068 |
crew | +0.061 |
Ar | +0.060 |
bandits | +0.058 |
knight | +0.058 |
Neg logit contributions | |
mm | -0.058 |
ops | -0.051 |
happen | -0.051 |
uch | -0.050 |
n | -0.050 |
Token | mercy |
Token ID | 17703 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
Amen | +0.036 |
Ar | +0.031 |
unts | +0.030 |
Dear | +0.030 |
crew | +0.030 |
Neg logit contributions | |
mm | -0.040 |
happen | -0.039 |
n | -0.036 |
tons | -0.035 |
uch | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
Amen | +0.458 |
crew | +0.420 |
Ar | +0.411 |
bandits | +0.404 |
knight | +0.402 |
Neg logit contributions | |
mm | -0.232 |
ops | -0.195 |
happen | -0.191 |
uch | -0.187 |
n | -0.184 |
Token | He |
Token ID | 679 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
mm | +0.018 |
ops | +0.016 |
n | +0.016 |
happen | +0.016 |
uch | +0.016 |
Neg logit contributions | |
Amen | -0.016 |
crew | -0.014 |
Ar | -0.013 |
bandits | -0.013 |
knight | -0.013 |
Token | said |
Token ID | 531 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
Amen | +0.105 |
crew | +0.088 |
lair | +0.087 |
unts | +0.087 |
bandits | +0.086 |
Neg logit contributions | |
mm | -0.073 |
n | -0.073 |
belongs | -0.066 |
tons | -0.065 |
liv | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
Amen | +0.041 |
crew | +0.037 |
knight | +0.035 |
Ar | +0.035 |
bandits | +0.035 |
Neg logit contributions | |
mm | -0.038 |
ops | -0.034 |
uch | -0.033 |
happen | -0.033 |
€ | -0.032 |
Token | ' |
Token ID | 705 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
mm | +0.021 |
ops | +0.019 |
happen | +0.019 |
uch | +0.019 |
n | +0.019 |
Neg logit contributions | |
Amen | -0.019 |
crew | -0.017 |
Ar | -0.017 |
bandits | -0.016 |
knight | -0.016 |
Token | your |
Token ID | 534 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
mm | +0.034 |
ops | +0.031 |
uch | +0.030 |
liv | +0.029 |
€ | +0.029 |
Neg logit contributions | |
Amen | -0.037 |
crew | -0.033 |
knight | -0.032 |
bandits | -0.032 |
firefighter | -0.031 |
Token | fault |
Token ID | 8046 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
mm | +0.055 |
happen | +0.052 |
ops | +0.050 |
n | +0.050 |
uch | +0.049 |
Neg logit contributions | |
Amen | -0.053 |
Offic | -0.048 |
Ar | -0.047 |
known | -0.045 |
bandits | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
Amen | +0.033 |
crew | +0.030 |
knight | +0.028 |
bandits | +0.028 |
unts | +0.027 |
Neg logit contributions | |
mm | -0.026 |
uch | -0.022 |
€ | -0.022 |
ops | -0.021 |
liv | -0.021 |
Token | ask |
Token ID | 1265 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
mm | +0.019 |
happen | +0.016 |
ops | +0.016 |
uch | +0.016 |
n | +0.015 |
Neg logit contributions | |
Amen | -0.032 |
crew | -0.030 |
Ar | -0.029 |
bandits | -0.028 |
knight | -0.028 |
Token | for |
Token ID | 329 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
Amen | +0.051 |
crew | +0.048 |
bandits | +0.046 |
knight | +0.045 |
Ar | +0.045 |
Neg logit contributions | |
mm | -0.041 |
ops | -0.036 |
uch | -0.035 |
liv | -0.034 |
happen | -0.033 |
Token | mercy |
Token ID | 17703 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
Amen | +0.020 |
Ar | +0.017 |
crew | +0.017 |
knight | +0.016 |
unts | +0.016 |
Neg logit contributions | |
mm | -0.022 |
happen | -0.021 |
n | -0.020 |
uch | -0.019 |
tons | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
Amen | +0.463 |
crew | +0.424 |
Ar | +0.421 |
Offic | +0.415 |
bandits | +0.415 |
Neg logit contributions | |
mm | -0.213 |
ops | -0.181 |
happen | -0.180 |
n | -0.175 |
ying | -0.174 |
Token | I |
Token ID | 314 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
mm | +0.039 |
ops | +0.033 |
happen | +0.033 |
uch | +0.032 |
n | +0.032 |
Neg logit contributions | |
Amen | -0.051 |
crew | -0.046 |
Ar | -0.045 |
bandits | -0.044 |
knight | -0.044 |
Token | will |
Token ID | 481 |
Feature activation | +6.76e-02 |
Pos logit contributions | |
Amen | +0.064 |
crew | +0.058 |
bandits | +0.057 |
Offic | +0.056 |
lair | +0.056 |
Neg logit contributions | |
mm | -0.057 |
n | -0.055 |
uch | -0.053 |
ops | -0.053 |
happen | -0.052 |
Token | spare |
Token ID | 13952 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
Amen | +0.110 |
crew | +0.101 |
Ar | +0.099 |
bandits | +0.098 |
Offic | +0.097 |
Neg logit contributions | |
mm | -0.097 |
happen | -0.095 |
n | -0.092 |
uch | -0.087 |
ops | -0.085 |
Token | you |
Token ID | 345 |
Feature activation | -5.85e-02 |
Pos logit contributions | |
Amen | +0.024 |
crew | +0.022 |
knight | +0.021 |
bandits | +0.021 |
hope | +0.021 |
Neg logit contributions | |
mm | -0.017 |
ops | -0.016 |
uch | -0.016 |
liv | -0.016 |
€ | -0.015 |
Token | were |
Token ID | 547 |
Feature activation | -4.69e-04 |
Pos logit contributions | |
Amen | +0.108 |
Ar | +0.094 |
bandits | +0.092 |
Offic | +0.092 |
lair | +0.090 |
Neg logit contributions | |
mm | -0.091 |
ops | -0.086 |
happen | -0.084 |
n | -0.083 |
ying | -0.082 |
Token | amazed |
Token ID | 24562 |
Feature activation | +7.25e-02 |
Pos logit contributions | |
mm | +0.001 |
ops | +0.001 |
happen | +0.001 |
n | +0.001 |
uch | +0.001 |
Neg logit contributions | |
Amen | -0.001 |
crew | -0.001 |
Ar | -0.001 |
bandits | -0.001 |
knight | -0.001 |
Token | at |
Token ID | 379 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
Amen | +0.128 |
crew | +0.120 |
knight | +0.114 |
bandits | +0.113 |
Ar | +0.112 |
Neg logit contributions | |
mm | -0.100 |
uch | -0.089 |
ops | -0.087 |
€ | -0.084 |
n | -0.083 |
Token | his |
Token ID | 465 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
mm | +0.043 |
happen | +0.039 |
n | +0.037 |
ying | +0.037 |
tons | +0.036 |
Neg logit contributions | |
Amen | -0.080 |
Ar | -0.073 |
crew | -0.071 |
Offic | -0.070 |
Dear | -0.070 |
Token | craft |
Token ID | 5977 |
Feature activation | -9.36e-03 |
Pos logit contributions | |
mm | +0.059 |
uch | +0.052 |
ops | +0.051 |
€ | +0.051 |
liv | +0.050 |
Neg logit contributions | |
Amen | -0.048 |
crew | -0.047 |
knight | -0.041 |
lair | -0.041 |
Ar | -0.041 |
Token | smanship |
Token ID | 49820 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
mm | +0.014 |
ops | +0.012 |
uch | +0.012 |
happen | +0.012 |
n | +0.012 |
Neg logit contributions | |
Amen | -0.016 |
crew | -0.014 |
Ar | -0.014 |
knight | -0.014 |
bandits | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +9.77e-03 |
Pos logit contributions | |
Amen | +0.344 |
crew | +0.321 |
knight | +0.299 |
lair | +0.295 |
Ar | +0.293 |
Neg logit contributions | |
mm | -0.315 |
uch | -0.274 |
€ | -0.267 |
ops | -0.265 |
liv | -0.258 |
Token | cheered |
Token ID | 34767 |
Feature activation | +9.08e-02 |
Pos logit contributions | |
Amen | +0.015 |
Ar | +0.013 |
Offic | +0.012 |
bandits | +0.012 |
Dear | +0.012 |
Neg logit contributions | |
mm | -0.016 |
happen | -0.015 |
n | -0.014 |
ops | -0.014 |
ying | -0.014 |
Token | in |
Token ID | 287 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
Amen | +0.158 |
crew | +0.146 |
knight | +0.135 |
Dear | +0.130 |
alien | +0.129 |
Neg logit contributions | |
mm | -0.141 |
uch | -0.133 |
ops | -0.123 |
liv | -0.117 |
€ | -0.113 |
Token | delight |
Token ID | 10974 |
Feature activation | +6.55e-03 |
Pos logit contributions | |
Amen | +0.069 |
crew | +0.060 |
Ar | +0.058 |
bandits | +0.057 |
knight | +0.056 |
Neg logit contributions | |
mm | -0.091 |
ops | -0.082 |
happen | -0.082 |
uch | -0.080 |
n | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | -1.57e-03 |
Pos logit contributions | |
Amen | +0.011 |
crew | +0.011 |
knight | +0.010 |
subjects | +0.010 |
unts | +0.010 |
Neg logit contributions | |
mm | -0.009 |
uch | -0.008 |
€ | -0.008 |
ops | -0.008 |
liv | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
Amen | +0.069 |
crew | +0.067 |
bandits | +0.064 |
knight | +0.064 |
alien | +0.063 |
Neg logit contributions | |
n | -0.050 |
mm | -0.048 |
ops | -0.043 |
ying | -0.042 |
€ | -0.041 |
Token | The |
Token ID | 464 |
Feature activation | -1.72e-03 |
Pos logit contributions | |
Amen | +0.053 |
crew | +0.047 |
lair | +0.043 |
knight | +0.043 |
bandits | +0.043 |
Neg logit contributions | |
mm | -0.056 |
n | -0.052 |
ops | -0.051 |
uch | -0.050 |
€ | -0.050 |
Token | rabbit |
Token ID | 22746 |
Feature activation | +8.29e-02 |
Pos logit contributions | |
mm | +0.003 |
happen | +0.003 |
ops | +0.003 |
tons | +0.003 |
n | +0.003 |
Neg logit contributions | |
Amen | -0.002 |
Ar | -0.002 |
crew | -0.002 |
known | -0.002 |
lair | -0.002 |
Token | thanked |
Token ID | 26280 |
Feature activation | +7.39e-02 |
Pos logit contributions | |
Amen | +0.160 |
Dear | +0.132 |
Ar | +0.131 |
unts | +0.131 |
crew | +0.131 |
Neg logit contributions | |
n | -0.102 |
ops | -0.097 |
mm | -0.095 |
belongs | -0.090 |
uch | -0.089 |
Token | the |
Token ID | 262 |
Feature activation | -3.47e-04 |
Pos logit contributions | |
Amen | +0.156 |
Ar | +0.144 |
crew | +0.143 |
lair | +0.142 |
Dear | +0.142 |
Neg logit contributions | |
mm | -0.075 |
happen | -0.067 |
n | -0.063 |
tons | -0.061 |
€ | -0.056 |
Token | kindly |
Token ID | 26820 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
mm | +0.001 |
happen | +0.001 |
next | +0.001 |
tons | +0.001 |
ops | +0.001 |
Neg logit contributions | |
Amen | -0.000 |
known | -0.000 |
Ar | -0.000 |
lair | -0.000 |
crew | -0.000 |
Token | bull |
Token ID | 6473 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
Amen | +0.309 |
crew | +0.276 |
Ar | +0.264 |
knight | +0.260 |
bandits | +0.260 |
Neg logit contributions | |
mm | -0.341 |
ops | -0.304 |
uch | -0.298 |
happen | -0.298 |
n | -0.292 |
Token | and |
Token ID | 290 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
Amen | +0.187 |
crew | +0.168 |
Ar | +0.167 |
bandits | +0.162 |
Offic | +0.160 |
Neg logit contributions | |
mm | -0.142 |
ops | -0.126 |
happen | -0.123 |
n | -0.121 |
uch | -0.119 |
Token | said |
Token ID | 531 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
Amen | +0.029 |
unts | +0.025 |
Ar | +0.024 |
lair | +0.024 |
Dear | +0.024 |
Neg logit contributions | |
mm | -0.020 |
n | -0.019 |
happen | -0.019 |
ops | -0.018 |
tons | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -4.65e-03 |
Pos logit contributions | |
Amen | +0.052 |
crew | +0.047 |
knight | +0.044 |
Ar | +0.044 |
bandits | +0.044 |
Neg logit contributions | |
mm | -0.038 |
ops | -0.034 |
uch | -0.033 |
happen | -0.032 |
€ | -0.032 |
Token | " |
Token ID | 366 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
mm | +0.007 |
ops | +0.006 |
happen | +0.006 |
n | +0.006 |
tons | +0.006 |
Neg logit contributions | |
Amen | -0.007 |
crew | -0.006 |
Ar | -0.006 |
bandits | -0.006 |
knight | -0.006 |
Token | make |
Token ID | 787 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
mm | +0.016 |
happen | +0.014 |
ops | +0.013 |
n | +0.013 |
uch | +0.013 |
Neg logit contributions | |
Amen | -0.022 |
crew | -0.019 |
Ar | -0.019 |
bandits | -0.019 |
Offic | -0.018 |
Token | it |
Token ID | 340 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
Amen | +0.039 |
crew | +0.035 |
Ar | +0.034 |
bandits | +0.033 |
knight | +0.033 |
Neg logit contributions | |
mm | -0.038 |
ops | -0.034 |
happen | -0.033 |
uch | -0.033 |
ying | -0.032 |
Token | whole |
Token ID | 2187 |
Feature activation | +6.21e-03 |
Pos logit contributions | |
mm | +0.072 |
ops | +0.065 |
tons | +0.062 |
€ | +0.062 |
ying | +0.062 |
Neg logit contributions | |
Amen | -0.069 |
crew | -0.063 |
knight | -0.059 |
Ar | -0.059 |
lair | -0.058 |
Token | again |
Token ID | 757 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
Amen | +0.011 |
crew | +0.010 |
Ar | +0.010 |
bandits | +0.010 |
Offic | +0.010 |
Neg logit contributions | |
mm | -0.008 |
happen | -0.007 |
ops | -0.007 |
uch | -0.007 |
n | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
Amen | +0.034 |
crew | +0.031 |
Ar | +0.029 |
Dear | +0.029 |
knight | +0.029 |
Neg logit contributions | |
mm | -0.034 |
ops | -0.030 |
liv | -0.029 |
uch | -0.029 |
€ | -0.029 |
Token | Amen |
Token ID | 34717 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
mm | +0.052 |
ops | +0.045 |
happen | +0.044 |
uch | +0.044 |
n | +0.044 |
Neg logit contributions | |
Amen | -0.055 |
crew | -0.050 |
Ar | -0.048 |
bandits | -0.048 |
knight | -0.047 |
Token | ." |
Token ID | 526 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
bandits | +0.381 |
knight | +0.363 |
Amen | +0.362 |
Mic | +0.360 |
Offic | +0.358 |
Neg logit contributions | |
mm | -0.269 |
ops | -0.222 |
n | -0.217 |
uch | -0.211 |
happen | -0.205 |
Token |
|
Token ID | 198 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
Amen | +0.061 |
crew | +0.057 |
bandits | +0.054 |
Ar | +0.054 |
knight | +0.054 |
Neg logit contributions | |
mm | -0.076 |
ops | -0.066 |
n | -0.066 |
tons | -0.065 |
€ | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
crew | +0.073 |
Amen | +0.071 |
knight | +0.070 |
bandits | +0.070 |
lair | +0.068 |
Neg logit contributions | |
mm | -0.071 |
n | -0.066 |
€ | -0.063 |
liv | -0.062 |
ops | -0.060 |
Token | Then |
Token ID | 6423 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
Amen | +0.067 |
crew | +0.062 |
knight | +0.059 |
lair | +0.059 |
bandits | +0.058 |
Neg logit contributions | |
mm | -0.066 |
n | -0.057 |
ops | -0.057 |
uch | -0.057 |
€ | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
mm | +0.054 |
ying | +0.050 |
uch | +0.050 |
ops | +0.050 |
happen | +0.048 |
Neg logit contributions | |
Amen | -0.056 |
crew | -0.052 |
knight | -0.049 |
bandits | -0.048 |
hope | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | +9.76e-02 |
Pos logit contributions | |
Amen | +0.052 |
crew | +0.047 |
Ar | +0.045 |
Offic | +0.045 |
lair | +0.045 |
Neg logit contributions | |
mm | -0.031 |
happen | -0.027 |
n | -0.026 |
tons | -0.026 |
ops | -0.026 |
Token | band |
Token ID | 4097 |
Feature activation | +9.24e-02 |
Pos logit contributions | |
Amen | +0.184 |
crew | +0.163 |
lair | +0.160 |
Offic | +0.159 |
bandits | +0.159 |
Neg logit contributions | |
uch | -0.115 |
mm | -0.112 |
happen | -0.112 |
n | -0.105 |
belongs | -0.103 |
Token | age |
Token ID | 496 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Amen | +0.150 |
crew | +0.138 |
bandits | +0.132 |
Ar | +0.130 |
lair | +0.128 |
Neg logit contributions | |
mm | -0.135 |
n | -0.132 |
ops | -0.130 |
uch | -0.128 |
happen | -0.122 |
Token | around |
Token ID | 1088 |
Feature activation | +7.29e-02 |
Pos logit contributions | |
Amen | +0.221 |
crew | +0.204 |
Ar | +0.197 |
bandits | +0.194 |
knight | +0.192 |
Neg logit contributions | |
mm | -0.130 |
ops | -0.114 |
happen | -0.111 |
uch | -0.111 |
n | -0.109 |
Token | her |
Token ID | 607 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
Amen | +0.148 |
crew | +0.135 |
Ar | +0.134 |
lair | +0.129 |
Dear | +0.127 |
Neg logit contributions | |
mm | -0.088 |
happen | -0.075 |
n | -0.074 |
ops | -0.074 |
tons | -0.071 |
Token | friend |
Token ID | 1545 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
Amen | +0.038 |
crew | +0.034 |
knight | +0.032 |
Ar | +0.031 |
bandits | +0.031 |
Neg logit contributions | |
mm | -0.048 |
ops | -0.042 |
ying | -0.042 |
happen | -0.042 |
uch | -0.041 |
Token | 's |
Token ID | 338 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
Amen | +0.412 |
crew | +0.379 |
knight | +0.365 |
bandits | +0.359 |
hope | +0.356 |
Neg logit contributions | |
mm | -0.238 |
happen | -0.192 |
uch | -0.185 |
ying | -0.184 |
ops | -0.183 |
Token | finger |
Token ID | 7660 |
Feature activation | +6.02e-02 |
Pos logit contributions | |
Amen | +0.050 |
Ar | +0.045 |
Offic | +0.044 |
crew | +0.044 |
bandits | +0.044 |
Neg logit contributions | |
mm | -0.033 |
n | -0.032 |
happen | -0.030 |
tons | -0.029 |
uch | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
Amen | +0.101 |
crew | +0.091 |
Ar | +0.089 |
bandits | +0.087 |
Offic | +0.086 |
Neg logit contributions | |
mm | -0.089 |
ops | -0.080 |
happen | -0.078 |
uch | -0.078 |
n | -0.077 |
Token | told |
Token ID | 1297 |
Feature activation | +6.96e-02 |
Pos logit contributions | |
happen | +0.018 |
mm | +0.018 |
n | +0.017 |
belongs | +0.016 |
ops | +0.015 |
Neg logit contributions | |
Amen | -0.030 |
Ar | -0.028 |
unts | -0.026 |
lair | -0.026 |
knight | -0.026 |
Token | her |
Token ID | 607 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
Amen | +0.112 |
crew | +0.099 |
bandits | +0.095 |
knight | +0.095 |
Ar | +0.093 |
Neg logit contributions | |
mm | -0.109 |
uch | -0.098 |
ops | -0.097 |
tons | -0.093 |
€ | -0.092 |
Token | tail |
Token ID | 7894 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
Amen | +0.020 |
Offic | +0.017 |
bandits | +0.017 |
Dear | +0.016 |
known | +0.016 |
Neg logit contributions | |
mm | -0.024 |
tons | -0.022 |
ops | -0.022 |
uch | -0.022 |
happen | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
Amen | +0.114 |
crew | +0.100 |
Ar | +0.099 |
Offic | +0.097 |
bandits | +0.097 |
Neg logit contributions | |
mm | -0.082 |
ops | -0.077 |
happen | -0.073 |
n | -0.073 |
tons | -0.072 |
Token | l |
Token ID | 300 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
Amen | +0.039 |
crew | +0.034 |
Ar | +0.034 |
bandits | +0.033 |
knight | +0.033 |
Neg logit contributions | |
mm | -0.030 |
ops | -0.026 |
happen | -0.026 |
n | -0.025 |
uch | -0.025 |
Token | icked |
Token ID | 9484 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
mm | +0.059 |
ops | +0.052 |
uch | +0.049 |
ying | +0.049 |
happen | +0.047 |
Neg logit contributions | |
Amen | -0.091 |
crew | -0.082 |
Ar | -0.080 |
bandits | -0.079 |
Offic | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
Amen | +0.090 |
crew | +0.078 |
Ar | +0.078 |
hope | +0.075 |
bandits | +0.075 |
Neg logit contributions | |
mm | -0.155 |
uch | -0.145 |
ops | -0.141 |
€ | -0.137 |
happen | -0.136 |
Token | vet |
Token ID | 20202 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
Amen | +0.033 |
crew | +0.029 |
Ar | +0.028 |
knight | +0.027 |
bandits | +0.027 |
Neg logit contributions | |
mm | -0.043 |
ops | -0.039 |
happen | -0.038 |
ying | -0.038 |
uch | -0.038 |
Token | 's |
Token ID | 338 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
Amen | +0.358 |
crew | +0.332 |
knight | +0.313 |
hope | +0.307 |
reunited | +0.306 |
Neg logit contributions | |
mm | -0.295 |
liv | -0.256 |
uch | -0.254 |
tons | -0.241 |
ying | -0.238 |
Token | hand |
Token ID | 1021 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
Amen | +0.069 |
crew | +0.061 |
Offic | +0.060 |
Ar | +0.060 |
bandits | +0.060 |
Neg logit contributions | |
mm | -0.069 |
ops | -0.062 |
happen | -0.062 |
n | -0.061 |
uch | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
Amen | +0.097 |
crew | +0.088 |
knight | +0.085 |
hope | +0.084 |
lair | +0.081 |
Neg logit contributions | |
mm | -0.096 |
uch | -0.084 |
liv | -0.084 |
€ | -0.083 |
ying | -0.083 |
Token | He |
Token ID | 679 |
Feature activation | +5.08e-02 |
Pos logit contributions | |
mm | +0.022 |
ops | +0.019 |
tons | +0.019 |
n | +0.019 |
€ | +0.019 |
Neg logit contributions | |
Amen | -0.021 |
crew | -0.019 |
knight | -0.018 |
bandits | -0.018 |
messenger | -0.018 |
Token | looked |
Token ID | 3114 |
Feature activation | -2.09e-03 |
Pos logit contributions | |
Amen | +0.088 |
crew | +0.077 |
bandits | +0.075 |
Ar | +0.075 |
knight | +0.073 |
Neg logit contributions | |
mm | -0.074 |
ops | -0.067 |
n | -0.064 |
uch | -0.063 |
happen | -0.063 |
Token | me |
Token ID | 502 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
Amen | +0.078 |
crew | +0.071 |
Ar | +0.070 |
bandits | +0.068 |
Offic | +0.068 |
Neg logit contributions | |
mm | -0.049 |
happen | -0.042 |
ops | -0.041 |
uch | -0.041 |
n | -0.040 |
Token | ?" |
Token ID | 1701 |
Feature activation | +9.59e-02 |
Pos logit contributions | |
mm | +0.027 |
ops | +0.025 |
liv | +0.023 |
happen | +0.023 |
€ | +0.023 |
Neg logit contributions | |
Amen | -0.027 |
crew | -0.024 |
knight | -0.023 |
Dear | -0.022 |
lair | -0.022 |
Token | Her |
Token ID | 2332 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
Amen | +0.120 |
crew | +0.101 |
Ar | +0.095 |
known | +0.094 |
Offic | +0.094 |
Neg logit contributions | |
mm | -0.181 |
happen | -0.168 |
ops | -0.167 |
uch | -0.167 |
ying | -0.164 |
Token | dad |
Token ID | 9955 |
Feature activation | +6.92e-02 |
Pos logit contributions | |
Amen | +0.051 |
crew | +0.045 |
bandits | +0.041 |
knight | +0.041 |
Ar | +0.040 |
Neg logit contributions | |
mm | -0.087 |
ops | -0.080 |
uch | -0.080 |
happen | -0.078 |
n | -0.078 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Amen | +0.093 |
crew | +0.078 |
knight | +0.073 |
Offic | +0.071 |
Dear | +0.070 |
Neg logit contributions | |
mm | -0.132 |
ops | -0.117 |
uch | -0.116 |
happen | -0.115 |
€ | -0.114 |
Token | kindly |
Token ID | 26820 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
Amen | +0.175 |
Ar | +0.160 |
crew | +0.159 |
Offic | +0.157 |
bandits | +0.156 |
Neg logit contributions | |
mm | -0.138 |
happen | -0.121 |
tons | -0.121 |
ops | -0.119 |
ying | -0.119 |
Token | and |
Token ID | 290 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
Amen | +0.311 |
crew | +0.275 |
knight | +0.259 |
Ar | +0.258 |
bandits | +0.255 |
Neg logit contributions | |
mm | -0.328 |
uch | -0.291 |
ops | -0.289 |
happen | -0.286 |
€ | -0.278 |
Token | said |
Token ID | 531 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
Amen | +0.052 |
crew | +0.047 |
Ar | +0.044 |
Offic | +0.043 |
bandits | +0.043 |
Neg logit contributions | |
mm | -0.062 |
ops | -0.056 |
uch | -0.055 |
happen | -0.055 |
ying | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | +3.84e-03 |
Pos logit contributions | |
Amen | +0.068 |
crew | +0.061 |
firefighter | +0.058 |
knight | +0.057 |
messenger | +0.056 |
Neg logit contributions | |
mm | -0.053 |
uch | -0.048 |
ops | -0.048 |
€ | -0.046 |
liv | -0.045 |
Token | " |
Token ID | 366 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
Amen | +0.006 |
crew | +0.005 |
Ar | +0.005 |
bandits | +0.005 |
knight | +0.005 |
Neg logit contributions | |
mm | -0.006 |
ops | -0.005 |
happen | -0.005 |
n | -0.005 |
tons | -0.005 |
Token | The |
Token ID | 464 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
Amen | +0.083 |
crew | +0.074 |
knight | +0.070 |
bandits | +0.070 |
Ar | +0.069 |
Neg logit contributions | |
mm | -0.091 |
ops | -0.081 |
n | -0.081 |
uch | -0.080 |
happen | -0.080 |
Token | very |
Token ID | 845 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
mm | +0.039 |
happen | +0.036 |
n | +0.036 |
uch | +0.036 |
tons | +0.035 |
Neg logit contributions | |
Amen | -0.045 |
crew | -0.040 |
Ar | -0.039 |
Offic | -0.038 |
bandits | -0.038 |
Token | happy |
Token ID | 3772 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
Amen | +0.028 |
crew | +0.025 |
Ar | +0.024 |
bandits | +0.023 |
knight | +0.023 |
Neg logit contributions | |
mm | -0.035 |
ops | -0.032 |
happen | -0.031 |
uch | -0.031 |
n | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
Amen | +0.033 |
crew | +0.030 |
knight | +0.029 |
lair | +0.028 |
hope | +0.028 |
Neg logit contributions | |
mm | -0.044 |
uch | -0.041 |
€ | -0.039 |
ops | -0.039 |
liv | -0.037 |
Token | thanked |
Token ID | 26280 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
mm | +0.027 |
ops | +0.025 |
uch | +0.024 |
happen | +0.024 |
ying | +0.024 |
Neg logit contributions | |
Amen | -0.019 |
crew | -0.017 |
Ar | -0.016 |
bandits | -0.016 |
Offic | -0.016 |
Token | her |
Token ID | 607 |
Feature activation | -3.40e-03 |
Pos logit contributions | |
Amen | +0.117 |
crew | +0.104 |
bandits | +0.099 |
knight | +0.099 |
Ar | +0.098 |
Neg logit contributions | |
mm | -0.111 |
ops | -0.103 |
uch | -0.102 |
ying | -0.097 |
happen | -0.096 |
Token | kindly |
Token ID | 26820 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
mm | +0.005 |
uch | +0.005 |
liv | +0.005 |
ops | +0.005 |
€ | +0.005 |
Neg logit contributions | |
Amen | -0.006 |
crew | -0.005 |
knight | -0.005 |
lair | -0.005 |
hope | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +5.44e-03 |
Pos logit contributions | |
Amen | +0.354 |
crew | +0.320 |
knight | +0.312 |
hope | +0.298 |
lair | +0.296 |
Neg logit contributions | |
mm | -0.298 |
uch | -0.266 |
ops | -0.253 |
liv | -0.251 |
aund | -0.237 |
Token | Sally |
Token ID | 25737 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
Amen | +0.007 |
crew | +0.006 |
Ar | +0.006 |
bandits | +0.006 |
knight | +0.006 |
Neg logit contributions | |
mm | -0.010 |
ops | -0.009 |
n | -0.009 |
happen | -0.009 |
€ | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -1.99e-03 |
Pos logit contributions | |
mm | +0.056 |
ops | +0.050 |
happen | +0.049 |
uch | +0.049 |
n | +0.049 |
Neg logit contributions | |
Amen | -0.050 |
crew | -0.044 |
Ar | -0.042 |
bandits | -0.041 |
knight | -0.041 |
Token | glad |
Token ID | 9675 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
mm | +0.004 |
happen | +0.003 |
ops | +0.003 |
uch | +0.003 |
n | +0.003 |
Neg logit contributions | |
Amen | -0.003 |
crew | -0.002 |
Ar | -0.002 |
bandits | -0.002 |
knight | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
Amen | +0.105 |
crew | +0.096 |
Ar | +0.091 |
bandits | +0.091 |
knight | +0.091 |
Neg logit contributions | |
mm | -0.082 |
ops | -0.072 |
uch | -0.071 |
happen | -0.070 |
€ | -0.068 |
Token | carefully |
Token ID | 7773 |
Feature activation | +9.67e-02 |
Pos logit contributions | |
mm | +0.002 |
happen | +0.002 |
n | +0.002 |
ops | +0.002 |
belongs | +0.002 |
Neg logit contributions | |
Amen | -0.002 |
Ar | -0.002 |
unts | -0.002 |
crew | -0.002 |
bandits | -0.002 |
Token | wrapped |
Token ID | 12908 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
Amen | +0.116 |
crew | +0.105 |
Ar | +0.100 |
known | +0.096 |
Offic | +0.096 |
Neg logit contributions | |
mm | -0.190 |
ops | -0.166 |
uch | -0.165 |
€ | -0.164 |
ying | -0.164 |
Token | it |
Token ID | 340 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
Amen | +0.036 |
crew | +0.033 |
Ar | +0.033 |
bandits | +0.031 |
knight | +0.031 |
Neg logit contributions | |
mm | -0.039 |
uch | -0.034 |
ops | -0.034 |
happen | -0.033 |
ying | -0.032 |
Token | around |
Token ID | 1088 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
mm | +0.020 |
ops | +0.016 |
€ | +0.016 |
liv | +0.016 |
happen | +0.015 |
Neg logit contributions | |
Amen | -0.016 |
hope | -0.015 |
desperately | -0.015 |
lair | -0.015 |
Ar | -0.015 |
Token | his |
Token ID | 465 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
Amen | +0.076 |
crew | +0.070 |
Ar | +0.067 |
knight | +0.066 |
bandits | +0.065 |
Neg logit contributions | |
mm | -0.109 |
uch | -0.101 |
ying | -0.095 |
ops | -0.095 |
belongs | -0.093 |
Token | friend |
Token ID | 1545 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
Amen | +0.016 |
crew | +0.014 |
Ar | +0.014 |
bandits | +0.013 |
knight | +0.013 |
Neg logit contributions | |
mm | -0.018 |
ops | -0.017 |
happen | -0.016 |
uch | -0.016 |
n | -0.016 |
Token | 's |
Token ID | 338 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
Amen | +0.359 |
crew | +0.341 |
knight | +0.327 |
reunited | +0.318 |
mouse | +0.314 |
Neg logit contributions | |
mm | -0.280 |
uch | -0.224 |
happen | -0.221 |
ying | -0.219 |
liv | -0.214 |
Token | paw |
Token ID | 41827 |
Feature activation | +5.92e-02 |
Pos logit contributions | |
Amen | +0.044 |
Offic | +0.040 |
Ar | +0.039 |
bandits | +0.038 |
known | +0.038 |
Neg logit contributions | |
mm | -0.037 |
n | -0.034 |
happen | -0.034 |
ops | -0.033 |
tons | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
Amen | +0.101 |
crew | +0.095 |
lair | +0.091 |
knight | +0.090 |
bandits | +0.088 |
Neg logit contributions | |
mm | -0.086 |
€ | -0.070 |
happen | -0.069 |
uch | -0.069 |
ying | -0.068 |
Token | Original |
Token ID | 13745 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
mm | +0.025 |
ops | +0.023 |
n | +0.022 |
tons | +0.022 |
€ | +0.022 |
Neg logit contributions | |
Amen | -0.018 |
crew | -0.015 |
bandits | -0.014 |
knight | -0.014 |
Ar | -0.014 |
Token | felt |
Token ID | 2936 |
Feature activation | +7.35e-02 |
Pos logit contributions | |
mm | +0.133 |
happen | +0.115 |
ying | +0.114 |
uch | +0.112 |
€ | +0.111 |
Neg logit contributions | |
Amen | -0.093 |
crew | -0.090 |
Ar | -0.084 |
knight | -0.082 |
bandits | -0.082 |
Token | smiled |
Token ID | 13541 |
Feature activation | +9.10e-02 |
Pos logit contributions | |
mm | +0.106 |
happen | +0.090 |
ying | +0.088 |
ops | +0.088 |
uch | +0.087 |
Neg logit contributions | |
Amen | -0.095 |
crew | -0.085 |
knight | -0.081 |
Ar | -0.080 |
known | -0.080 |
Token | and |
Token ID | 290 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
Amen | +0.160 |
crew | +0.146 |
Ar | +0.138 |
knight | +0.138 |
bandits | +0.137 |
Neg logit contributions | |
mm | -0.130 |
ops | -0.114 |
uch | -0.113 |
happen | -0.111 |
n | -0.110 |
Token | talked |
Token ID | 6619 |
Feature activation | +6.80e-02 |
Pos logit contributions | |
Amen | +0.055 |
crew | +0.049 |
Ar | +0.047 |
bandits | +0.046 |
knight | +0.045 |
Neg logit contributions | |
mm | -0.067 |
ops | -0.061 |
happen | -0.060 |
uch | -0.059 |
ying | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | +8.58e-02 |
Pos logit contributions | |
Amen | +0.104 |
crew | +0.095 |
Ar | +0.090 |
knight | +0.089 |
bandits | +0.089 |
Neg logit contributions | |
mm | -0.111 |
uch | -0.098 |
ops | -0.097 |
€ | -0.093 |
n | -0.092 |
Token | them |
Token ID | 606 |
Feature activation | +4.17e-03 |
Pos logit contributions | |
Amen | +0.144 |
Ar | +0.131 |
crew | +0.128 |
Dear | +0.125 |
bandits | +0.125 |
Neg logit contributions | |
mm | -0.123 |
happen | -0.115 |
n | -0.111 |
ops | -0.108 |
tons | -0.106 |
Token | kindly |
Token ID | 26820 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
Amen | +0.006 |
crew | +0.006 |
hope | +0.005 |
understand | +0.005 |
knight | +0.005 |
Neg logit contributions | |
mm | -0.007 |
uch | -0.007 |
€ | -0.006 |
ops | -0.006 |
liv | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +3.12e-03 |
Pos logit contributions | |
Amen | +0.318 |
crew | +0.296 |
hope | +0.281 |
knight | +0.278 |
lair | +0.270 |
Neg logit contributions | |
mm | -0.318 |
uch | -0.280 |
liv | -0.266 |
ops | -0.265 |
€ | -0.254 |
Token | Soon |
Token ID | 15894 |
Feature activation | -8.61e-03 |
Pos logit contributions | |
Amen | +0.004 |
crew | +0.004 |
Ar | +0.003 |
bandits | +0.003 |
knight | +0.003 |
Neg logit contributions | |
mm | -0.006 |
ops | -0.005 |
n | -0.005 |
happen | -0.005 |
tons | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
mm | +0.013 |
uch | +0.013 |
€ | +0.012 |
belongs | +0.012 |
ops | +0.012 |
Neg logit contributions | |
crew | -0.013 |
Amen | -0.012 |
known | -0.012 |
knight | -0.012 |
bandits | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
mm | +0.018 |
uch | +0.016 |
ops | +0.016 |
€ | +0.016 |
happen | +0.016 |
Neg logit contributions | |
Amen | -0.018 |
crew | -0.016 |
Ar | -0.015 |
bandits | -0.015 |
knight | -0.015 |
Token | boys |
Token ID | 6510 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
mm | +0.056 |
ops | +0.050 |
happen | +0.049 |
uch | +0.049 |
n | +0.048 |
Neg logit contributions | |
Amen | -0.052 |
crew | -0.046 |
Ar | -0.044 |
bandits | -0.043 |
knight | -0.043 |
Token | The |
Token ID | 464 |
Feature activation | +3.54e-03 |
Pos logit contributions | |
lair | +0.071 |
knight | +0.071 |
Amen | +0.067 |
crew | +0.066 |
bandits | +0.065 |
Neg logit contributions | |
n | -0.100 |
mm | -0.096 |
� | -0.094 |
next | -0.090 |
happen | -0.086 |
Token | Fairy |
Token ID | 17746 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
Amen | +0.004 |
crew | +0.004 |
Ar | +0.004 |
bandits | +0.004 |
Offic | +0.004 |
Neg logit contributions | |
mm | -0.007 |
happen | -0.006 |
ops | -0.006 |
n | -0.006 |
uch | -0.006 |
Token | God |
Token ID | 1793 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
Amen | +0.114 |
crew | +0.102 |
Ar | +0.095 |
knight | +0.095 |
bandits | +0.093 |
Neg logit contributions | |
mm | -0.147 |
ops | -0.131 |
uch | -0.130 |
happen | -0.129 |
€ | -0.126 |
Token | mother |
Token ID | 13552 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
Amen | +0.059 |
crew | +0.052 |
Ar | +0.050 |
bandits | +0.049 |
knight | +0.048 |
Neg logit contributions | |
mm | -0.074 |
ops | -0.066 |
happen | -0.066 |
uch | -0.065 |
n | -0.064 |
Token | smiled |
Token ID | 13541 |
Feature activation | +9.76e-02 |
Pos logit contributions | |
Amen | +0.154 |
crew | +0.135 |
Ar | +0.131 |
bandits | +0.128 |
knight | +0.127 |
Neg logit contributions | |
mm | -0.178 |
ops | -0.160 |
happen | -0.158 |
uch | -0.156 |
n | -0.155 |
Token | kindly |
Token ID | 26820 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Amen | +0.170 |
crew | +0.154 |
Ar | +0.150 |
bandits | +0.147 |
knight | +0.146 |
Neg logit contributions | |
mm | -0.137 |
ops | -0.120 |
happen | -0.119 |
uch | -0.117 |
n | -0.115 |
Token | and |
Token ID | 290 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
Amen | +0.324 |
crew | +0.291 |
knight | +0.274 |
Dear | +0.273 |
hope | +0.270 |
Neg logit contributions | |
mm | -0.303 |
uch | -0.267 |
ops | -0.258 |
liv | -0.256 |
€ | -0.255 |
Token | pulled |
Token ID | 5954 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
Amen | +0.056 |
Ar | +0.049 |
lair | +0.049 |
crew | +0.048 |
bandits | +0.048 |
Neg logit contributions | |
mm | -0.051 |
happen | -0.047 |
n | -0.046 |
tons | -0.044 |
ops | -0.044 |
Token | Mill |
Token ID | 9212 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
mm | +0.020 |
happen | +0.019 |
n | +0.019 |
ying | +0.018 |
tons | +0.018 |
Neg logit contributions | |
Amen | -0.023 |
crew | -0.020 |
lair | -0.020 |
Ar | -0.020 |
Offic | -0.020 |
Token | ie |
Token ID | 494 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
Amen | +0.069 |
crew | +0.063 |
Ar | +0.060 |
knight | +0.060 |
bandits | +0.060 |
Neg logit contributions | |
mm | -0.052 |
ops | -0.044 |
uch | -0.043 |
happen | -0.043 |
€ | -0.042 |
Token | in |
Token ID | 287 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
hope | +0.081 |
Amen | +0.081 |
crew | +0.078 |
understand | +0.075 |
reunited | +0.074 |
Neg logit contributions | |
mm | -0.134 |
ops | -0.109 |
liv | -0.108 |
uch | -0.106 |
€ | -0.106 |
Token | her |
Token ID | 607 |
Feature activation | -3.55e-03 |
Pos logit contributions | |
mm | +0.011 |
happen | +0.010 |
ops | +0.010 |
n | +0.010 |
tons | +0.010 |
Neg logit contributions | |
Amen | -0.012 |
crew | -0.011 |
Ar | -0.010 |
bandits | -0.010 |
Offic | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -2.63e-03 |
Pos logit contributions | |
mm | +0.006 |
uch | +0.005 |
liv | +0.005 |
€ | +0.005 |
aund | +0.005 |
Neg logit contributions | |
Amen | -0.005 |
crew | -0.005 |
knight | -0.005 |
lair | -0.005 |
hope | -0.005 |
Token | His |
Token ID | 2399 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
mm | +0.003 |
ancing | +0.003 |
ying | +0.003 |
tons | +0.003 |
aund | +0.003 |
Neg logit contributions | |
Amen | -0.004 |
Offic | -0.004 |
Ar | -0.004 |
crew | -0.004 |
traitor | -0.004 |
Token | mom |
Token ID | 1995 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
mm | +0.037 |
ops | +0.033 |
ying | +0.033 |
happen | +0.033 |
uch | +0.033 |
Neg logit contributions | |
Amen | -0.027 |
crew | -0.024 |
knight | -0.022 |
bandits | -0.022 |
Ar | -0.022 |
Token | smiled |
Token ID | 13541 |
Feature activation | +9.37e-02 |
Pos logit contributions | |
Amen | +0.130 |
Ar | +0.116 |
lair | +0.110 |
bandits | +0.110 |
crew | +0.109 |
Neg logit contributions | |
n | -0.098 |
ops | -0.093 |
mm | -0.092 |
happen | -0.089 |
ying | -0.087 |
Token | kindly |
Token ID | 26820 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Amen | +0.156 |
crew | +0.140 |
Ar | +0.136 |
bandits | +0.133 |
knight | +0.132 |
Neg logit contributions | |
mm | -0.140 |
ops | -0.123 |
happen | -0.122 |
uch | -0.120 |
n | -0.119 |
Token | and |
Token ID | 290 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
Amen | +0.319 |
crew | +0.286 |
knight | +0.269 |
hope | +0.266 |
Dear | +0.264 |
Neg logit contributions | |
mm | -0.317 |
uch | -0.278 |
liv | -0.266 |
ops | -0.265 |
€ | -0.264 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
Amen | +0.056 |
Ar | +0.049 |
lair | +0.049 |
bandits | +0.047 |
crew | +0.047 |
Neg logit contributions | |
mm | -0.043 |
n | -0.039 |
happen | -0.038 |
tons | -0.037 |
ops | -0.037 |
Token | over |
Token ID | 625 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
mm | +0.031 |
ops | +0.027 |
uch | +0.027 |
happen | +0.026 |
ying | +0.026 |
Neg logit contributions | |
Amen | -0.020 |
Ar | -0.017 |
crew | -0.017 |
bandits | -0.016 |
knight | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
Amen | +0.042 |
crew | +0.037 |
Ar | +0.035 |
bandits | +0.034 |
knight | +0.034 |
Neg logit contributions | |
mm | -0.059 |
ops | -0.054 |
happen | -0.053 |
uch | -0.053 |
n | -0.052 |
Token | Simon |
Token ID | 11288 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
Amen | +0.132 |
crew | +0.119 |
Ar | +0.116 |
bandits | +0.112 |
lair | +0.111 |
Neg logit contributions | |
mm | -0.130 |
happen | -0.122 |
n | -0.118 |
ops | -0.116 |
tons | -0.116 |
Token | advice |
Token ID | 5608 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
mm | +0.103 |
ops | +0.091 |
happen | +0.091 |
uch | +0.089 |
n | +0.089 |
Neg logit contributions | |
Amen | -0.110 |
crew | -0.098 |
Ar | -0.096 |
bandits | -0.093 |
Offic | -0.093 |
Token | and |
Token ID | 290 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
Amen | +0.036 |
crew | +0.033 |
knight | +0.031 |
known | +0.031 |
unts | +0.031 |
Neg logit contributions | |
mm | -0.037 |
uch | -0.032 |
€ | -0.031 |
ops | -0.030 |
happen | -0.030 |
Token | thanked |
Token ID | 26280 |
Feature activation | +8.34e-02 |
Pos logit contributions | |
Amen | +0.021 |
Ar | +0.018 |
crew | +0.018 |
bandits | +0.017 |
Dear | +0.017 |
Neg logit contributions | |
mm | -0.022 |
happen | -0.020 |
ops | -0.019 |
n | -0.019 |
tons | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
Amen | +0.143 |
crew | +0.128 |
Ar | +0.125 |
bandits | +0.122 |
knight | +0.122 |
Neg logit contributions | |
mm | -0.121 |
ops | -0.106 |
happen | -0.105 |
uch | -0.103 |
n | -0.102 |
Token | bird |
Token ID | 6512 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Amen | +0.016 |
crew | +0.014 |
Ar | +0.013 |
Offic | +0.013 |
bandits | +0.013 |
Neg logit contributions | |
mm | -0.021 |
happen | -0.019 |
ops | -0.019 |
tons | -0.019 |
n | -0.019 |
Token | kindly |
Token ID | 26820 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
Amen | +0.214 |
crew | +0.195 |
Ar | +0.188 |
knight | +0.186 |
bandits | +0.186 |
Neg logit contributions | |
mm | -0.144 |
ops | -0.124 |
happen | -0.121 |
uch | -0.121 |
n | -0.117 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
Amen | +0.343 |
crew | +0.312 |
knight | +0.297 |
Dear | +0.292 |
hope | +0.291 |
Neg logit contributions | |
mm | -0.289 |
uch | -0.256 |
ops | -0.243 |
liv | -0.241 |
€ | -0.234 |
Token | He |
Token ID | 679 |
Feature activation | +6.18e-02 |
Pos logit contributions | |
Amen | +0.016 |
crew | +0.015 |
bandits | +0.014 |
Ar | +0.014 |
knight | +0.014 |
Neg logit contributions | |
mm | -0.021 |
n | -0.018 |
ops | -0.018 |
tons | -0.018 |
€ | -0.018 |
Token | continued |
Token ID | 3767 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
Amen | +0.097 |
bandits | +0.080 |
Ar | +0.079 |
crew | +0.079 |
lair | +0.078 |
Neg logit contributions | |
mm | -0.095 |
n | -0.089 |
ops | -0.089 |
happen | -0.085 |
tons | -0.085 |
Token | on |
Token ID | 319 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
Amen | +0.078 |
crew | +0.069 |
Ar | +0.068 |
Offic | +0.065 |
knight | +0.065 |
Neg logit contributions | |
mm | -0.077 |
happen | -0.070 |
ops | -0.070 |
n | -0.068 |
ying | -0.067 |
Token | his |
Token ID | 465 |
Feature activation | -4.15e-03 |
Pos logit contributions | |
mm | +0.033 |
happen | +0.030 |
ops | +0.029 |
n | +0.029 |
tons | +0.028 |
Neg logit contributions | |
Amen | -0.035 |
crew | -0.031 |
Ar | -0.030 |
bandits | -0.029 |
Offic | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
Amen | +0.060 |
bandits | +0.058 |
crew | +0.058 |
knight | +0.057 |
lair | +0.055 |
Neg logit contributions | |
mm | -0.062 |
n | -0.060 |
ops | -0.055 |
liv | -0.054 |
€ | -0.053 |
Token | One |
Token ID | 3198 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
Amen | +0.056 |
crew | +0.052 |
bandits | +0.051 |
knight | +0.051 |
lair | +0.050 |
Neg logit contributions | |
mm | -0.059 |
n | -0.055 |
ops | -0.052 |
tons | -0.051 |
liv | -0.051 |
Token | day |
Token ID | 1110 |
Feature activation | -8.86e-02 |
Pos logit contributions | |
Amen | +0.031 |
crew | +0.029 |
wonder | +0.029 |
knight | +0.028 |
known | +0.027 |
Neg logit contributions | |
mm | -0.036 |
ops | -0.033 |
ancing | -0.031 |
uch | -0.030 |
happen | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
mm | +0.130 |
uch | +0.115 |
ops | +0.111 |
€ | +0.110 |
happen | +0.108 |
Neg logit contributions | |
crew | -0.145 |
Amen | -0.143 |
knight | -0.136 |
Ar | -0.134 |
bandits | -0.132 |
Token | a |
Token ID | 257 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
mm | +0.023 |
ops | +0.020 |
happen | +0.020 |
n | +0.019 |
uch | +0.019 |
Neg logit contributions | |
Amen | -0.027 |
crew | -0.023 |
Ar | -0.023 |
bandits | -0.023 |
Offic | -0.022 |
Token | kindly |
Token ID | 26820 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
Amen | +0.082 |
Ar | +0.069 |
lair | +0.069 |
Offic | +0.068 |
Dear | +0.067 |
Neg logit contributions | |
mm | -0.085 |
tons | -0.079 |
happen | -0.079 |
uch | -0.077 |
€ | -0.076 |
Token | old |
Token ID | 1468 |
Feature activation | +9.53e-02 |
Pos logit contributions | |
Amen | +0.251 |
crew | +0.219 |
Ar | +0.209 |
bandits | +0.204 |
knight | +0.203 |
Neg logit contributions | |
mm | -0.360 |
ops | -0.326 |
happen | -0.321 |
uch | -0.320 |
n | -0.316 |
Token | man |
Token ID | 582 |
Feature activation | +8.65e-02 |
Pos logit contributions | |
Amen | +0.122 |
crew | +0.104 |
Ar | +0.103 |
bandits | +0.098 |
Offic | +0.097 |
Neg logit contributions | |
mm | -0.175 |
ops | -0.160 |
happen | -0.160 |
tons | -0.156 |
n | -0.156 |
Token | came |
Token ID | 1625 |
Feature activation | +8.96e-03 |
Pos logit contributions | |
Amen | +0.105 |
crew | +0.089 |
Ar | +0.086 |
bandits | +0.083 |
knight | +0.082 |
Neg logit contributions | |
mm | -0.166 |
ops | -0.153 |
happen | -0.151 |
uch | -0.150 |
n | -0.149 |
Token | by |
Token ID | 416 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
Amen | +0.015 |
crew | +0.013 |
Ar | +0.013 |
bandits | +0.013 |
knight | +0.013 |
Neg logit contributions | |
mm | -0.013 |
ops | -0.012 |
happen | -0.012 |
uch | -0.012 |
n | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -8.91e-03 |
Pos logit contributions | |
mm | +0.056 |
ops | +0.049 |
uch | +0.049 |
ying | +0.046 |
happen | +0.046 |
Neg logit contributions | |
Amen | -0.059 |
crew | -0.054 |
Ar | -0.052 |
knight | -0.051 |
bandits | -0.050 |
Token | on |
Token ID | 319 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
mm | +0.014 |
ops | +0.012 |
happen | +0.012 |
uch | +0.012 |
n | +0.012 |
Neg logit contributions | |
Amen | -0.028 |
crew | -0.026 |
Ar | -0.025 |
bandits | -0.024 |
knight | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
mm | +0.067 |
happen | +0.064 |
n | +0.064 |
tons | +0.058 |
ying | +0.056 |
Neg logit contributions | |
Amen | -0.133 |
unts | -0.118 |
crew | -0.118 |
Ar | -0.115 |
Dear | -0.115 |
Token | door |
Token ID | 3420 |
Feature activation | +7.59e-02 |
Pos logit contributions | |
happen | +0.023 |
mm | +0.022 |
belongs | +0.021 |
ying | +0.021 |
n | +0.020 |
Neg logit contributions | |
Amen | -0.033 |
Ar | -0.030 |
known | -0.028 |
Offic | -0.028 |
unts | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +2.48e-04 |
Pos logit contributions | |
Amen | +0.124 |
crew | +0.111 |
knight | +0.107 |
Ar | +0.107 |
Offic | +0.106 |
Neg logit contributions | |
mm | -0.115 |
ops | -0.102 |
uch | -0.101 |
€ | -0.097 |
tons | -0.096 |
Token | a |
Token ID | 257 |
Feature activation | +6.75e-02 |
Pos logit contributions | |
Amen | +0.000 |
unts | +0.000 |
crew | +0.000 |
Ar | +0.000 |
lair | +0.000 |
Neg logit contributions | |
mm | -0.000 |
happen | -0.000 |
n | -0.000 |
belongs | -0.000 |
uch | -0.000 |
Token | kindly |
Token ID | 26820 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
Amen | +0.108 |
known | +0.099 |
lair | +0.097 |
crew | +0.096 |
generation | +0.095 |
Neg logit contributions | |
happen | -0.084 |
mm | -0.075 |
n | -0.073 |
gh | -0.073 |
next | -0.072 |
Token | old |
Token ID | 1468 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Amen | +0.294 |
crew | +0.256 |
Ar | +0.255 |
Offic | +0.247 |
bandits | +0.246 |
Neg logit contributions | |
mm | -0.287 |
happen | -0.275 |
belongs | -0.265 |
tons | -0.264 |
n | -0.262 |
Token | man |
Token ID | 582 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
Amen | +0.157 |
Ar | +0.153 |
crew | +0.145 |
known | +0.143 |
bandits | +0.143 |
Neg logit contributions | |
happen | -0.128 |
mm | -0.122 |
belongs | -0.122 |
tons | -0.116 |
n | -0.114 |
Token | answered |
Token ID | 9373 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
Amen | +0.116 |
crew | +0.100 |
Ar | +0.097 |
bandits | +0.094 |
knight | +0.093 |
Neg logit contributions | |
mm | -0.165 |
ops | -0.150 |
happen | -0.149 |
n | -0.147 |
uch | -0.147 |
Token | . |
Token ID | 13 |
Feature activation | -7.93e-03 |
Pos logit contributions | |
Amen | +0.132 |
crew | +0.120 |
knight | +0.111 |
firefighter | +0.109 |
reunited | +0.107 |
Neg logit contributions | |
mm | -0.138 |
uch | -0.132 |
ops | -0.122 |
liv | -0.119 |
€ | -0.117 |
Token | " |
Token ID | 366 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
mm | +0.011 |
n | +0.010 |
€ | +0.009 |
happen | +0.009 |
belongs | +0.009 |
Neg logit contributions | |
Amen | -0.013 |
crew | -0.012 |
Ar | -0.012 |
bandits | -0.012 |
lair | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
lair | +0.103 |
bandits | +0.102 |
crew | +0.101 |
Amen | +0.099 |
traitor | +0.099 |
Neg logit contributions | |
n | -0.071 |
mm | -0.063 |
€ | -0.061 |
� | -0.057 |
next | -0.057 |
Token | The |
Token ID | 464 |
Feature activation | +9.59e-03 |
Pos logit contributions | |
Amen | +0.081 |
lair | +0.077 |
crew | +0.076 |
bandits | +0.073 |
known | +0.071 |
Neg logit contributions | |
mm | -0.058 |
n | -0.057 |
€ | -0.053 |
uch | -0.051 |
ops | -0.050 |
Token | big |
Token ID | 1263 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
Amen | +0.014 |
Offic | +0.013 |
Mill | +0.012 |
Dear | +0.012 |
Bet | +0.012 |
Neg logit contributions | |
mm | -0.015 |
happen | -0.014 |
next | -0.013 |
uch | -0.013 |
n | -0.013 |
Token | fish |
Token ID | 5916 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
Amen | +0.081 |
crew | +0.067 |
Ar | +0.065 |
Offic | +0.065 |
bandits | +0.063 |
Neg logit contributions | |
mm | -0.145 |
happen | -0.133 |
ops | -0.132 |
uch | -0.130 |
tons | -0.129 |
Token | smiled |
Token ID | 13541 |
Feature activation | +9.63e-02 |
Pos logit contributions | |
Amen | +0.061 |
bandits | +0.054 |
Offic | +0.054 |
Ar | +0.053 |
Dear | +0.051 |
Neg logit contributions | |
mm | -0.048 |
n | -0.047 |
ops | -0.046 |
uch | -0.044 |
happen | -0.042 |
Token | kindly |
Token ID | 26820 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
Amen | +0.179 |
Ar | +0.164 |
Offic | +0.163 |
bandits | +0.159 |
crew | +0.159 |
Neg logit contributions | |
mm | -0.115 |
happen | -0.101 |
tons | -0.101 |
ying | -0.099 |
ancing | -0.097 |
Token | at |
Token ID | 379 |
Feature activation | -2.80e-03 |
Pos logit contributions | |
Amen | +0.290 |
crew | +0.258 |
knight | +0.244 |
Ar | +0.238 |
bandits | +0.236 |
Neg logit contributions | |
mm | -0.327 |
uch | -0.291 |
ops | -0.284 |
liv | -0.283 |
happen | -0.283 |
Token | the |
Token ID | 262 |
Feature activation | -4.29e-03 |
Pos logit contributions | |
mm | +0.004 |
happen | +0.004 |
n | +0.004 |
ops | +0.003 |
tons | +0.003 |
Neg logit contributions | |
Amen | -0.005 |
Ar | -0.004 |
crew | -0.004 |
Dear | -0.004 |
unts | -0.004 |
Token | little |
Token ID | 1310 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
happen | +0.006 |
next | +0.006 |
mm | +0.006 |
tons | +0.005 |
n | +0.005 |
Neg logit contributions | |
Amen | -0.007 |
Offic | -0.006 |
Bet | -0.006 |
Mill | -0.006 |
Dear | -0.006 |
Token | fish |
Token ID | 5916 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
Amen | +0.139 |
crew | +0.115 |
Ar | +0.110 |
bandits | +0.107 |
Offic | +0.106 |
Neg logit contributions | |
mm | -0.270 |
ops | -0.248 |
happen | -0.246 |
uch | -0.243 |
n | -0.241 |
Token | and |
Token ID | 290 |
Feature activation | +9.39e-03 |
Pos logit contributions | |
Amen | +0.119 |
Ar | +0.109 |
crew | +0.106 |
bandits | +0.106 |
Offic | +0.106 |
Neg logit contributions | |
mm | -0.056 |
ops | -0.050 |
happen | -0.047 |
n | -0.047 |
ying | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
mm | +0.091 |
happen | +0.077 |
ops | +0.077 |
n | +0.074 |
uch | +0.073 |
Neg logit contributions | |
Amen | -0.152 |
crew | -0.139 |
Ar | -0.137 |
bandits | -0.133 |
lair | -0.132 |
Token | girl |
Token ID | 2576 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
mm | +0.095 |
ops | +0.087 |
uch | +0.086 |
happen | +0.085 |
n | +0.084 |
Neg logit contributions | |
Amen | -0.056 |
crew | -0.049 |
knight | -0.045 |
Ar | -0.045 |
bandits | -0.045 |
Token | knocked |
Token ID | 13642 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
Amen | +0.111 |
crew | +0.101 |
Ar | +0.097 |
knight | +0.096 |
Offic | +0.095 |
Neg logit contributions | |
mm | -0.134 |
happen | -0.118 |
ops | -0.117 |
ying | -0.116 |
uch | -0.116 |
Token | , |
Token ID | 11 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
mm | +0.013 |
ops | +0.011 |
happen | +0.011 |
uch | +0.010 |
tons | +0.010 |
Neg logit contributions | |
Amen | -0.019 |
crew | -0.017 |
Ar | -0.017 |
knight | -0.017 |
bandits | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
mm | +0.015 |
happen | +0.013 |
ops | +0.013 |
ying | +0.012 |
tons | +0.012 |
Neg logit contributions | |
Amen | -0.027 |
crew | -0.024 |
Ar | -0.024 |
bandits | -0.023 |
lair | -0.023 |
Token | kindly |
Token ID | 26820 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
Amen | +0.091 |
lair | +0.082 |
Ar | +0.082 |
known | +0.082 |
crew | +0.080 |
Neg logit contributions | |
mm | -0.081 |
happen | -0.079 |
tons | -0.071 |
belongs | -0.070 |
uch | -0.068 |
Token | old |
Token ID | 1468 |
Feature activation | +9.92e-02 |
Pos logit contributions | |
Amen | +0.273 |
crew | +0.239 |
Ar | +0.233 |
bandits | +0.225 |
Offic | +0.222 |
Neg logit contributions | |
mm | -0.326 |
happen | -0.298 |
ops | -0.293 |
tons | -0.288 |
n | -0.287 |
Token | woman |
Token ID | 2415 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
Amen | +0.153 |
Ar | +0.148 |
crew | +0.146 |
known | +0.143 |
bandits | +0.138 |
Neg logit contributions | |
mm | -0.129 |
happen | -0.128 |
belongs | -0.127 |
tons | -0.121 |
ops | -0.114 |
Token | answered |
Token ID | 9373 |
Feature activation | +8.38e-02 |
Pos logit contributions | |
Amen | +0.189 |
crew | +0.167 |
Ar | +0.163 |
bandits | +0.157 |
knight | +0.153 |
Neg logit contributions | |
mm | -0.251 |
ops | -0.230 |
happen | -0.230 |
n | -0.227 |
uch | -0.222 |
Token | . |
Token ID | 13 |
Feature activation | -5.24e-03 |
Pos logit contributions | |
Amen | +0.147 |
crew | +0.133 |
knight | +0.128 |
bandits | +0.125 |
messenger | +0.125 |
Neg logit contributions | |
mm | -0.121 |
uch | -0.112 |
ops | -0.104 |
€ | -0.102 |
liv | -0.100 |
Token | She |
Token ID | 1375 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
mm | +0.006 |
n | +0.006 |
€ | +0.006 |
belongs | +0.005 |
happen | +0.005 |
Neg logit contributions | |
Amen | -0.009 |
crew | -0.009 |
Ar | -0.009 |
lair | -0.009 |
bandits | -0.008 |
Token | Tim |
Token ID | 5045 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
mm | +0.032 |
uch | +0.030 |
ops | +0.030 |
happen | +0.029 |
€ | +0.028 |
Neg logit contributions | |
Amen | -0.029 |
crew | -0.027 |
Ar | -0.025 |
bandits | -0.025 |
known | -0.025 |
Token | said |
Token ID | 531 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
mm | +0.102 |
happen | +0.090 |
ops | +0.089 |
uch | +0.089 |
ying | +0.087 |
Neg logit contributions | |
Amen | -0.085 |
crew | -0.078 |
Ar | -0.074 |
knight | -0.074 |
bandits | -0.073 |
Token | , |
Token ID | 11 |
Feature activation | -4.20e-03 |
Pos logit contributions | |
Amen | +0.050 |
crew | +0.046 |
knight | +0.044 |
messenger | +0.043 |
firefighter | +0.043 |
Neg logit contributions | |
mm | -0.038 |
ops | -0.034 |
uch | -0.034 |
liv | -0.032 |
€ | -0.031 |
Token | " |
Token ID | 366 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
mm | +0.008 |
uch | +0.007 |
ops | +0.007 |
happen | +0.007 |
ying | +0.007 |
Neg logit contributions | |
Amen | -0.006 |
crew | -0.005 |
Ar | -0.005 |
knight | -0.005 |
Offic | -0.005 |
Token | Was |
Token ID | 16973 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
Amen | +0.063 |
crew | +0.055 |
Ar | +0.054 |
bandits | +0.052 |
knight | +0.052 |
Neg logit contributions | |
mm | -0.076 |
ops | -0.068 |
happen | -0.068 |
uch | -0.067 |
ying | -0.066 |
Token | n |
Token ID | 77 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
mm | +0.162 |
ops | +0.155 |
ancing | +0.144 |
ying | +0.143 |
uch | +0.143 |
Neg logit contributions | |
Amen | -0.163 |
known | -0.152 |
crew | -0.147 |
Offic | -0.143 |
bandits | -0.143 |
Token | 't |
Token ID | 470 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
mm | +0.485 |
tons | +0.441 |
ops | +0.422 |
€ | +0.408 |
liv | +0.405 |
Neg logit contributions | |
Amen | -0.399 |
crew | -0.358 |
unts | -0.334 |
known | -0.332 |
hear | -0.329 |
Token | that |
Token ID | 326 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
mm | +0.037 |
happen | +0.033 |
ops | +0.032 |
uch | +0.031 |
n | +0.031 |
Neg logit contributions | |
Amen | -0.047 |
crew | -0.041 |
Ar | -0.041 |
bandits | -0.040 |
Offic | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | +7.69e-03 |
Pos logit contributions | |
mm | +0.035 |
ops | +0.034 |
ancing | +0.032 |
liv | +0.032 |
tons | +0.032 |
Neg logit contributions | |
Amen | -0.031 |
knight | -0.028 |
crew | -0.028 |
firefighter | -0.026 |
known | -0.026 |
Token | fair |
Token ID | 3148 |
Feature activation | -4.65e-02 |
Pos logit contributions | |
Amen | +0.013 |
bandits | +0.011 |
Dear | +0.011 |
crew | +0.011 |
known | +0.011 |
Neg logit contributions | |
mm | -0.010 |
happen | -0.010 |
belongs | -0.009 |
uch | -0.009 |
n | -0.009 |
Token | job |
Token ID | 1693 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
mm | +0.069 |
ops | +0.061 |
uch | +0.060 |
tons | +0.058 |
n | +0.058 |
Neg logit contributions | |
Amen | -0.079 |
crew | -0.072 |
knight | -0.068 |
Ar | -0.068 |
bandits | -0.067 |
Token | You |
Token ID | 921 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
happen | +0.073 |
ying | +0.070 |
mm | +0.070 |
ops | +0.069 |
uch | +0.067 |
Neg logit contributions | |
Amen | -0.104 |
knight | -0.089 |
Some | -0.084 |
Dear | -0.083 |
Ar | -0.083 |
Token | are |
Token ID | 389 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
mm | +0.086 |
uch | +0.079 |
ops | +0.079 |
happen | +0.076 |
n | +0.075 |
Neg logit contributions | |
Amen | -0.137 |
crew | -0.119 |
lair | -0.117 |
Offic | -0.116 |
Ar | -0.115 |
Token | brave |
Token ID | 14802 |
Feature activation | -6.77e-03 |
Pos logit contributions | |
ops | +0.099 |
mm | +0.098 |
ying | +0.090 |
ancing | +0.086 |
our | +0.085 |
Neg logit contributions | |
Amen | -0.045 |
knight | -0.044 |
known | -0.043 |
crew | -0.041 |
bandits | -0.038 |
Token | ! |
Token ID | 0 |
Feature activation | -6.40e-02 |
Pos logit contributions | |
mm | +0.009 |
ops | +0.008 |
happen | +0.008 |
tons | +0.008 |
ying | +0.008 |
Neg logit contributions | |
Amen | -0.012 |
crew | -0.011 |
Ar | -0.011 |
Offic | -0.010 |
bandits | -0.010 |
Token | Was |
Token ID | 8920 |
Feature activation | -5.28e-02 |
Pos logit contributions | |
happen | +0.083 |
mm | +0.082 |
ops | +0.080 |
uch | +0.079 |
ying | +0.078 |
Neg logit contributions | |
Amen | -0.122 |
knight | -0.102 |
Ar | -0.100 |
crew | -0.099 |
Dear | -0.099 |
Token | n |
Token ID | 77 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
mm | +0.078 |
ancing | +0.074 |
ying | +0.073 |
inning | +0.072 |
ower | +0.071 |
Neg logit contributions | |
knight | -0.068 |
known | -0.065 |
bandits | -0.065 |
hear | -0.064 |
Amen | -0.064 |
Token | 't |
Token ID | 470 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
mm | +0.470 |
tons | +0.427 |
belongs | +0.401 |
existed | +0.400 |
estival | +0.396 |
Neg logit contributions | |
Amen | -0.357 |
hear | -0.322 |
unts | -0.320 |
understand | -0.303 |
crew | -0.301 |
Token | it |
Token ID | 340 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
mm | +0.020 |
ops | +0.019 |
ancing | +0.017 |
€ | +0.017 |
uch | +0.017 |
Neg logit contributions | |
Amen | -0.016 |
crew | -0.015 |
knight | -0.015 |
known | -0.014 |
bandits | -0.014 |
Token | fun |
Token ID | 1257 |
Feature activation | -8.35e-02 |
Pos logit contributions | |
mm | +0.183 |
tons | +0.174 |
ops | +0.170 |
ern | +0.169 |
ying | +0.163 |
Neg logit contributions | |
knight | -0.125 |
recorded | -0.125 |
hear | -0.121 |
Amen | -0.120 |
known | -0.119 |
Token | ?" |
Token ID | 1701 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
mm | +0.160 |
liv | +0.141 |
ops | +0.137 |
€ | +0.136 |
tons | +0.134 |
Neg logit contributions | |
Amen | -0.095 |
crew | -0.088 |
knight | -0.087 |
lair | -0.086 |
recorded | -0.085 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
Amen | +0.063 |
unts | +0.046 |
Dear | +0.045 |
hope | +0.045 |
wonder | +0.045 |
Neg logit contributions | |
mm | -0.181 |
uch | -0.180 |
happen | -0.178 |
ops | -0.172 |
belongs | -0.171 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
Amen | +0.058 |
crew | +0.053 |
knight | +0.052 |
Ar | +0.051 |
lair | +0.051 |
Neg logit contributions | |
mm | -0.038 |
uch | -0.031 |
ops | -0.031 |
tons | -0.030 |
€ | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
mm | +0.023 |
n | +0.020 |
€ | +0.018 |
ops | +0.018 |
aund | +0.018 |
Neg logit contributions | |
Amen | -0.026 |
crew | -0.025 |
bandits | -0.024 |
lair | -0.024 |
traitor | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +4.04e-02 |
Pos logit contributions | |
crew | +0.078 |
Amen | +0.075 |
bandits | +0.074 |
lair | +0.073 |
knight | +0.073 |
Neg logit contributions | |
n | -0.053 |
mm | -0.050 |
ops | -0.044 |
� | -0.044 |
ying | -0.043 |
Token | " |
Token ID | 1 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
Amen | +0.065 |
crew | +0.061 |
lair | +0.058 |
hope | +0.056 |
bandits | +0.055 |
Neg logit contributions | |
mm | -0.060 |
n | -0.056 |
ops | -0.054 |
uch | -0.052 |
€ | -0.052 |
Token | Was |
Token ID | 16973 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
mm | +0.017 |
n | +0.014 |
uch | +0.014 |
ops | +0.014 |
€ | +0.014 |
Neg logit contributions | |
Amen | -0.015 |
crew | -0.014 |
lair | -0.013 |
bandits | -0.013 |
knight | -0.013 |
Token | n |
Token ID | 77 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
mm | +0.125 |
ops | +0.112 |
ying | +0.104 |
tons | +0.103 |
ancing | +0.102 |
Neg logit contributions | |
Amen | -0.200 |
known | -0.187 |
crew | -0.186 |
bandits | -0.184 |
knight | -0.183 |
Token | 't |
Token ID | 470 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
mm | +0.444 |
tons | +0.390 |
ops | +0.389 |
liv | +0.374 |
€ | +0.374 |
Neg logit contributions | |
Amen | -0.411 |
crew | -0.364 |
knight | -0.341 |
known | -0.336 |
unts | -0.333 |
Token | that |
Token ID | 326 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
happen | +0.029 |
mm | +0.028 |
ying | +0.025 |
n | +0.025 |
existed | +0.024 |
Neg logit contributions | |
Amen | -0.047 |
Ar | -0.043 |
crew | -0.040 |
Offic | -0.040 |
bandits | -0.039 |
Token | fun |
Token ID | 1257 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
mm | +0.020 |
ops | +0.019 |
liv | +0.018 |
tons | +0.017 |
ancing | +0.017 |
Neg logit contributions | |
Amen | -0.018 |
crew | -0.016 |
knight | -0.016 |
bandits | -0.015 |
known | -0.015 |
Token | ?" |
Token ID | 1701 |
Feature activation | +7.90e-02 |
Pos logit contributions | |
mm | +0.111 |
ops | +0.099 |
uch | +0.095 |
liv | +0.093 |
€ | +0.093 |
Neg logit contributions | |
Amen | -0.136 |
crew | -0.123 |
knight | -0.118 |
Ar | -0.115 |
bandits | -0.115 |
Token | the |
Token ID | 262 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
Amen | +0.132 |
crew | +0.119 |
Ar | +0.119 |
lair | +0.118 |
bandits | +0.115 |
Neg logit contributions | |
mm | -0.111 |
n | -0.101 |
happen | -0.100 |
€ | -0.097 |
tons | -0.096 |
Token | spoke |
Token ID | 5158 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
mm | +0.155 |
ying | +0.137 |
happen | +0.136 |
ancing | +0.134 |
tons | +0.133 |
Neg logit contributions | |
Amen | -0.065 |
knight | -0.061 |
bandits | -0.059 |
crew | -0.058 |
heard | -0.057 |
Token | up |
Token ID | 510 |
Feature activation | -1.45e-03 |
Pos logit contributions | |
Amen | +0.056 |
crew | +0.050 |
knight | +0.048 |
bandits | +0.047 |
lair | +0.047 |
Neg logit contributions | |
mm | -0.075 |
ops | -0.066 |
uch | -0.065 |
n | -0.062 |
happen | -0.061 |
Token | : |
Token ID | 25 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
mm | +0.002 |
ops | +0.002 |
uch | +0.002 |
happen | +0.002 |
ying | +0.002 |
Neg logit contributions | |
Amen | -0.002 |
Dear | -0.002 |
crew | -0.002 |
knight | -0.002 |
Ar | -0.002 |
Token | " |
Token ID | 366 |
Feature activation | +4.33e-02 |
Pos logit contributions | |
Amen | +0.037 |
crew | +0.029 |
knight | +0.029 |
Offic | +0.029 |
Ar | +0.028 |
Neg logit contributions | |
mm | -0.057 |
ops | -0.053 |
uch | -0.053 |
ying | -0.052 |
happen | -0.051 |
Token | Should |
Token ID | 19926 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
Amen | +0.058 |
Ar | +0.051 |
Dear | +0.050 |
known | +0.048 |
Offic | +0.047 |
Neg logit contributions | |
mm | -0.078 |
happen | -0.073 |
ying | -0.071 |
ops | -0.071 |
tons | -0.069 |
Token | n |
Token ID | 77 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
mm | +0.111 |
ying | +0.111 |
liv | +0.109 |
ops | +0.108 |
ancing | +0.107 |
Neg logit contributions | |
Amen | -0.121 |
crew | -0.120 |
knight | -0.112 |
hope | -0.111 |
known | -0.109 |
Token | 't |
Token ID | 470 |
Feature activation | -4.97e-03 |
Pos logit contributions | |
mm | +0.521 |
liv | +0.456 |
tons | +0.447 |
belongs | +0.445 |
€ | +0.439 |
Neg logit contributions | |
Amen | -0.293 |
hear | -0.263 |
understand | -0.240 |
known | -0.235 |
unts | -0.231 |
Token | we |
Token ID | 356 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
liv | +0.007 |
mm | +0.007 |
ops | +0.007 |
ying | +0.007 |
ancing | +0.007 |
Neg logit contributions | |
Amen | -0.008 |
crew | -0.007 |
knight | -0.007 |
messenger | -0.007 |
hear | -0.007 |
Token | share |
Token ID | 2648 |
Feature activation | +2.68e-03 |
Pos logit contributions | |
mm | +0.057 |
ops | +0.050 |
ying | +0.049 |
happen | +0.049 |
liv | +0.048 |
Neg logit contributions | |
Amen | -0.056 |
crew | -0.050 |
Ar | -0.048 |
knight | -0.047 |
bandits | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | -5.07e-03 |
Pos logit contributions | |
Amen | +0.004 |
crew | +0.004 |
Ar | +0.003 |
Dear | +0.003 |
knight | +0.003 |
Neg logit contributions | |
mm | -0.004 |
ops | -0.004 |
ying | -0.004 |
€ | -0.004 |
uch | -0.004 |
Token | apples |
Token ID | 22514 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
mm | +0.008 |
ops | +0.007 |
ying | +0.007 |
liv | +0.007 |
tons | +0.007 |
Neg logit contributions | |
Amen | -0.008 |
knight | -0.007 |
crew | -0.007 |
Ar | -0.006 |
bandits | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -6.46e-02 |
Pos logit contributions | |
ying | +0.161 |
mm | +0.159 |
uch | +0.152 |
aund | +0.150 |
ancing | +0.149 |
Neg logit contributions | |
Amen | -0.174 |
crew | -0.164 |
hope | -0.153 |
parent | -0.149 |
understand | -0.147 |
Token | no |
Token ID | 645 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
mm | +0.099 |
ops | +0.096 |
uch | +0.096 |
ying | +0.090 |
€ | +0.088 |
Neg logit contributions | |
Amen | -0.092 |
crew | -0.091 |
knight | -0.090 |
hope | -0.085 |
bandits | -0.082 |
Token | ! |
Token ID | 0 |
Feature activation | -6.79e-02 |
Pos logit contributions | |
Amen | +0.020 |
crew | +0.019 |
knight | +0.018 |
unts | +0.018 |
bandits | +0.018 |
Neg logit contributions | |
mm | -0.022 |
ying | -0.021 |
uch | -0.021 |
ops | -0.020 |
aund | -0.020 |
Token | You |
Token ID | 921 |
Feature activation | -8.79e-02 |
Pos logit contributions | |
happen | +0.100 |
ying | +0.094 |
mm | +0.092 |
ops | +0.090 |
uch | +0.085 |
Neg logit contributions | |
Amen | -0.116 |
Some | -0.099 |
crew | -0.096 |
Offic | -0.094 |
Dear | -0.093 |
Token | must |
Token ID | 1276 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
mm | +0.140 |
ops | +0.124 |
happen | +0.123 |
uch | +0.120 |
ying | +0.120 |
Neg logit contributions | |
Amen | -0.136 |
crew | -0.123 |
Ar | -0.117 |
bandits | -0.116 |
knight | -0.116 |
Token | n |
Token ID | 77 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
Amen | +0.051 |
hope | +0.044 |
crew | +0.043 |
known | +0.042 |
understand | +0.042 |
Neg logit contributions | |
mm | -0.060 |
ops | -0.055 |
ying | -0.053 |
aund | -0.051 |
ries | -0.050 |
Token | 't |
Token ID | 470 |
Feature activation | +1.22e-03 |
Pos logit contributions | |
mm | +0.505 |
tons | +0.446 |
€ | +0.429 |
belongs | +0.428 |
lish | +0.428 |
Neg logit contributions | |
Amen | -0.277 |
unts | -0.241 |
known | -0.234 |
hear | -0.232 |
crew | -0.229 |
Token | touch |
Token ID | 3638 |
Feature activation | -8.68e-02 |
Pos logit contributions | |
Amen | +0.002 |
crew | +0.002 |
Ar | +0.002 |
knight | +0.001 |
Dear | +0.001 |
Neg logit contributions | |
mm | -0.002 |
ops | -0.002 |
ying | -0.002 |
liv | -0.002 |
tons | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
mm | +0.148 |
ops | +0.138 |
uch | +0.130 |
€ | +0.128 |
liv | +0.127 |
Neg logit contributions | |
Amen | -0.108 |
crew | -0.104 |
hope | -0.103 |
bandits | -0.101 |
knight | -0.097 |
Token | cher |
Token ID | 21382 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
mm | +0.073 |
ops | +0.065 |
ying | +0.063 |
uch | +0.062 |
n | +0.062 |
Neg logit contributions | |
Amen | -0.066 |
crew | -0.060 |
knight | -0.058 |
Ar | -0.057 |
bandits | -0.057 |
Token | ries |
Token ID | 1678 |
Feature activation | -7.89e-02 |
Pos logit contributions | |
€ | +0.122 |
happen | +0.111 |
tons | +0.110 |
mm | +0.104 |
gh | +0.104 |
Neg logit contributions | |
Amen | -0.219 |
Mill | -0.207 |
known | -0.203 |
Dear | -0.202 |
Frank | -0.199 |
Token | " |
Token ID | 366 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
mm | +0.035 |
uch | +0.032 |
ops | +0.032 |
happen | +0.031 |
ying | +0.030 |
Neg logit contributions | |
Amen | -0.024 |
crew | -0.021 |
knight | -0.019 |
Ar | -0.019 |
bandits | -0.019 |
Token | No |
Token ID | 2949 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
Amen | +0.051 |
crew | +0.045 |
bandits | +0.042 |
knight | +0.042 |
Ar | +0.042 |
Neg logit contributions | |
mm | -0.062 |
ops | -0.056 |
uch | -0.055 |
n | -0.055 |
happen | -0.055 |
Token | ! |
Token ID | 0 |
Feature activation | -8.02e-02 |
Pos logit contributions | |
mm | +0.172 |
uch | +0.168 |
ying | +0.166 |
aund | +0.157 |
liv | +0.156 |
Neg logit contributions | |
crew | -0.174 |
knight | -0.168 |
Amen | -0.163 |
bandits | -0.157 |
nephew | -0.152 |
Token | You |
Token ID | 921 |
Feature activation | -8.16e-02 |
Pos logit contributions | |
happen | +0.115 |
mm | +0.111 |
ops | +0.107 |
ying | +0.104 |
liv | +0.102 |
Neg logit contributions | |
Amen | -0.134 |
crew | -0.122 |
Some | -0.119 |
knight | -0.115 |
Offic | -0.112 |
Token | must |
Token ID | 1276 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
mm | +0.127 |
uch | +0.117 |
ops | +0.116 |
happen | +0.115 |
n | +0.114 |
Neg logit contributions | |
Amen | -0.131 |
crew | -0.114 |
Ar | -0.113 |
Offic | -0.109 |
bandits | -0.108 |
Token | n |
Token ID | 77 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
Amen | +0.049 |
hope | +0.044 |
understand | +0.043 |
crew | +0.043 |
known | +0.042 |
Neg logit contributions | |
mm | -0.064 |
ops | -0.059 |
ying | -0.057 |
ries | -0.056 |
aund | -0.054 |
Token | 't |
Token ID | 470 |
Feature activation | +4.49e-03 |
Pos logit contributions | |
mm | +0.531 |
belongs | +0.488 |
lish | +0.487 |
tons | +0.485 |
estival | +0.475 |
Neg logit contributions | |
hear | -0.186 |
Amen | -0.177 |
understand | -0.174 |
unts | -0.168 |
known | -0.166 |
Token | play |
Token ID | 711 |
Feature activation | -8.59e-02 |
Pos logit contributions | |
Amen | +0.006 |
crew | +0.005 |
unts | +0.005 |
understand | +0.005 |
Dear | +0.005 |
Neg logit contributions | |
mm | -0.008 |
ops | -0.007 |
ying | -0.007 |
liv | -0.007 |
tons | -0.007 |
Token | with |
Token ID | 351 |
Feature activation | -5.84e-02 |
Pos logit contributions | |
mm | +0.134 |
ops | +0.120 |
liv | +0.118 |
uch | +0.115 |
tons | +0.112 |
Neg logit contributions | |
Amen | -0.132 |
crew | -0.125 |
knight | -0.124 |
bandits | -0.118 |
lair | -0.117 |
Token | it |
Token ID | 340 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
mm | +0.096 |
€ | +0.088 |
liv | +0.087 |
ops | +0.084 |
ancing | +0.083 |
Neg logit contributions | |
crew | -0.078 |
bandits | -0.076 |
Amen | -0.076 |
knight | -0.074 |
lair | -0.071 |
Token | !" |
Token ID | 2474 |
Feature activation | +7.55e-02 |
Pos logit contributions | |
mm | +0.102 |
liv | +0.089 |
€ | +0.089 |
ops | +0.086 |
ancing | +0.085 |
Neg logit contributions | |
Amen | -0.084 |
knight | -0.082 |
crew | -0.080 |
lair | -0.078 |
bandits | -0.076 |
Token | her |
Token ID | 607 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
Amen | +0.101 |
crew | +0.090 |
bandits | +0.086 |
knight | +0.086 |
Ar | +0.085 |
Neg logit contributions | |
mm | -0.114 |
uch | -0.102 |
ops | -0.102 |
happen | -0.099 |
€ | -0.097 |
Token | pet |
Token ID | 4273 |
Feature activation | +6.59e-03 |
Pos logit contributions | |
mm | +0.055 |
ops | +0.049 |
uch | +0.048 |
happen | +0.048 |
n | +0.047 |
Neg logit contributions | |
Amen | -0.050 |
crew | -0.045 |
knight | -0.043 |
bandits | -0.042 |
Ar | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Amen | +0.010 |
crew | +0.009 |
knight | +0.009 |
bandits | +0.009 |
Ar | +0.009 |
Neg logit contributions | |
mm | -0.011 |
ops | -0.009 |
happen | -0.009 |
uch | -0.009 |
n | -0.009 |
Token | " |
Token ID | 366 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
mm | +0.025 |
ops | +0.022 |
uch | +0.022 |
happen | +0.022 |
ying | +0.021 |
Neg logit contributions | |
Amen | -0.021 |
crew | -0.018 |
Ar | -0.017 |
knight | -0.017 |
bandits | -0.017 |
Token | Would |
Token ID | 17353 |
Feature activation | -6.32e-02 |
Pos logit contributions | |
Amen | +0.052 |
crew | +0.044 |
Ar | +0.044 |
bandits | +0.042 |
Offic | +0.042 |
Neg logit contributions | |
mm | -0.077 |
ops | -0.069 |
happen | -0.069 |
uch | -0.068 |
ying | -0.068 |
Token | n |
Token ID | 77 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
ying | +0.087 |
existed | +0.081 |
ancing | +0.081 |
ig | +0.074 |
icking | +0.074 |
Neg logit contributions | |
Amen | -0.118 |
Offic | -0.098 |
Dear | -0.098 |
Frank | -0.096 |
bandits | -0.096 |
Token | 't |
Token ID | 470 |
Feature activation | +8.82e-03 |
Pos logit contributions | |
mm | +0.428 |
liv | +0.364 |
tons | +0.363 |
belongs | +0.360 |
ops | +0.360 |
Neg logit contributions | |
Amen | -0.369 |
crew | -0.335 |
knight | -0.310 |
known | -0.307 |
unts | -0.306 |
Token | it |
Token ID | 340 |
Feature activation | -8.05e-02 |
Pos logit contributions | |
Amen | +0.014 |
crew | +0.013 |
bandits | +0.012 |
knight | +0.012 |
messenger | +0.012 |
Neg logit contributions | |
mm | -0.014 |
ops | -0.013 |
liv | -0.013 |
uch | -0.013 |
ying | -0.012 |
Token | be |
Token ID | 307 |
Feature activation | -7.30e-02 |
Pos logit contributions | |
mm | +0.145 |
ying | +0.140 |
liv | +0.139 |
ops | +0.135 |
tons | +0.135 |
Neg logit contributions | |
Amen | -0.095 |
crew | -0.087 |
knight | -0.085 |
Dear | -0.082 |
known | -0.081 |
Token | fun |
Token ID | 1257 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
mm | +0.118 |
ops | +0.109 |
ying | +0.107 |
tons | +0.102 |
ancing | +0.101 |
Neg logit contributions | |
Amen | -0.111 |
Ar | -0.099 |
crew | -0.098 |
known | -0.096 |
bandits | -0.093 |
Token | to |
Token ID | 284 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
mm | +0.130 |
ops | +0.115 |
uch | +0.112 |
liv | +0.111 |
happen | +0.111 |
Neg logit contributions | |
Amen | -0.132 |
crew | -0.117 |
knight | -0.111 |
Dear | -0.109 |
Ar | -0.108 |
Token | and |
Token ID | 290 |
Feature activation | -4.00e-03 |
Pos logit contributions | |
mm | +0.005 |
uch | +0.004 |
€ | +0.004 |
liv | +0.004 |
aund | +0.004 |
Neg logit contributions | |
crew | -0.003 |
Amen | -0.003 |
lair | -0.003 |
knight | -0.003 |
reunited | -0.003 |
Token | said |
Token ID | 531 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
mm | +0.006 |
ops | +0.005 |
happen | +0.005 |
tons | +0.005 |
n | +0.005 |
Neg logit contributions | |
Amen | -0.007 |
Ar | -0.006 |
lair | -0.006 |
crew | -0.006 |
bandits | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -9.86e-03 |
Pos logit contributions | |
Amen | +0.034 |
crew | +0.031 |
firefighter | +0.029 |
knight | +0.029 |
messenger | +0.029 |
Neg logit contributions | |
mm | -0.029 |
uch | -0.028 |
ops | -0.027 |
liv | -0.026 |
€ | -0.026 |
Token | " |
Token ID | 366 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
mm | +0.020 |
uch | +0.019 |
ops | +0.019 |
happen | +0.018 |
liv | +0.018 |
Neg logit contributions | |
Amen | -0.011 |
crew | -0.009 |
Offic | -0.008 |
Ar | -0.008 |
firefighter | -0.008 |
Token | Would |
Token ID | 17353 |
Feature activation | -6.35e-02 |
Pos logit contributions | |
Amen | +0.056 |
crew | +0.049 |
lair | +0.047 |
knight | +0.046 |
bandits | +0.046 |
Neg logit contributions | |
mm | -0.069 |
ops | -0.061 |
n | -0.061 |
uch | -0.060 |
€ | -0.060 |
Token | n |
Token ID | 77 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
ying | +0.083 |
existed | +0.080 |
ancing | +0.079 |
tons | +0.077 |
belongs | +0.072 |
Neg logit contributions | |
Amen | -0.112 |
Ar | -0.099 |
Frank | -0.097 |
crew | -0.095 |
Da | -0.095 |
Token | 't |
Token ID | 470 |
Feature activation | +6.93e-03 |
Pos logit contributions | |
mm | +0.421 |
liv | +0.364 |
ops | +0.364 |
tons | +0.364 |
belongs | +0.357 |
Neg logit contributions | |
Amen | -0.372 |
crew | -0.345 |
known | -0.313 |
knight | -0.309 |
messenger | -0.308 |
Token | it |
Token ID | 340 |
Feature activation | -8.18e-02 |
Pos logit contributions | |
crew | +0.010 |
Amen | +0.010 |
knight | +0.009 |
parent | +0.009 |
bandits | +0.009 |
Neg logit contributions | |
mm | -0.011 |
ops | -0.011 |
liv | -0.011 |
uch | -0.010 |
€ | -0.010 |
Token | be |
Token ID | 307 |
Feature activation | -7.84e-02 |
Pos logit contributions | |
liv | +0.151 |
ying | +0.147 |
mm | +0.145 |
uct | +0.143 |
ops | +0.142 |
Neg logit contributions | |
Amen | -0.088 |
crew | -0.086 |
knight | -0.082 |
Dear | -0.077 |
recorded | -0.076 |
Token | great |
Token ID | 1049 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
mm | +0.134 |
ops | +0.130 |
ying | +0.122 |
ancing | +0.119 |
tons | +0.118 |
Neg logit contributions | |
Amen | -0.112 |
crew | -0.101 |
Ar | -0.099 |
known | -0.097 |
Dear | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | +9.91e-03 |
Pos logit contributions | |
mm | +0.118 |
uch | +0.114 |
ops | +0.111 |
liv | +0.110 |
€ | +0.107 |
Neg logit contributions | |
Amen | -0.119 |
crew | -0.112 |
knight | -0.105 |
bandits | -0.100 |
subjects | -0.098 |
Token | No |
Token ID | 2949 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
Amen | +0.063 |
crew | +0.056 |
Ar | +0.054 |
bandits | +0.053 |
knight | +0.053 |
Neg logit contributions | |
mm | -0.068 |
ops | -0.061 |
happen | -0.060 |
uch | -0.060 |
ying | -0.059 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
mm | +0.157 |
ying | +0.149 |
uch | +0.145 |
ancing | +0.140 |
ops | +0.138 |
Neg logit contributions | |
Amen | -0.187 |
crew | -0.172 |
hope | -0.169 |
knight | -0.169 |
bandits | -0.163 |
Token | ! |
Token ID | 0 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
mm | +0.084 |
€ | +0.077 |
ying | +0.074 |
uct | +0.073 |
aund | +0.072 |
Neg logit contributions | |
Amen | -0.079 |
hope | -0.077 |
crew | -0.074 |
mouse | -0.074 |
unts | -0.073 |
Token | You |
Token ID | 921 |
Feature activation | -7.95e-02 |
Pos logit contributions | |
happen | +0.094 |
ying | +0.080 |
mm | +0.077 |
ops | +0.077 |
gh | +0.075 |
Neg logit contributions | |
Amen | -0.118 |
Some | -0.109 |
crew | -0.100 |
Dear | -0.099 |
parent | -0.098 |
Token | must |
Token ID | 1276 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
mm | +0.113 |
uch | +0.106 |
ops | +0.104 |
n | +0.103 |
happen | +0.102 |
Neg logit contributions | |
Amen | -0.141 |
crew | -0.122 |
Ar | -0.121 |
lair | -0.119 |
Offic | -0.118 |
Token | n |
Token ID | 77 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
Amen | +0.058 |
crew | +0.051 |
Ar | +0.049 |
known | +0.049 |
knight | +0.048 |
Neg logit contributions | |
mm | -0.062 |
ops | -0.056 |
ying | -0.054 |
tons | -0.053 |
uch | -0.053 |
Token | 't |
Token ID | 470 |
Feature activation | +4.69e-03 |
Pos logit contributions | |
mm | +0.490 |
tons | +0.446 |
belongs | +0.433 |
€ | +0.424 |
lish | +0.422 |
Neg logit contributions | |
Amen | -0.259 |
unts | -0.235 |
known | -0.223 |
hear | -0.218 |
understand | -0.218 |
Token | sign |
Token ID | 1051 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
Amen | +0.007 |
crew | +0.006 |
Ar | +0.006 |
bandits | +0.006 |
knight | +0.006 |
Neg logit contributions | |
mm | -0.008 |
ops | -0.007 |
happen | -0.007 |
uch | -0.007 |
n | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
mm | +0.052 |
ops | +0.048 |
€ | +0.045 |
tons | +0.045 |
uch | +0.045 |
Neg logit contributions | |
Amen | -0.041 |
hope | -0.040 |
Dear | -0.039 |
bandits | -0.038 |
knight | -0.038 |
Token | !" |
Token ID | 2474 |
Feature activation | +7.12e-02 |
Pos logit contributions | |
mm | +0.117 |
ops | +0.102 |
tons | +0.101 |
liv | +0.099 |
€ | +0.097 |
Neg logit contributions | |
Amen | -0.090 |
crew | -0.084 |
Dear | -0.084 |
knight | -0.082 |
hope | -0.081 |
Token |
|
Token ID | 198 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
Amen | +0.101 |
crew | +0.087 |
Ar | +0.084 |
bandits | +0.082 |
knight | +0.081 |
Neg logit contributions | |
mm | -0.125 |
ops | -0.113 |
happen | -0.111 |
uch | -0.111 |
ying | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
Amen | +0.156 |
crew | +0.140 |
knight | +0.132 |
Dear | +0.129 |
Offic | +0.129 |
Neg logit contributions | |
mm | -0.135 |
uch | -0.118 |
happen | -0.114 |
€ | -0.110 |
ops | -0.110 |
Token | " |
Token ID | 366 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
mm | +0.025 |
ops | +0.022 |
tons | +0.021 |
n | +0.021 |
ying | +0.021 |
Neg logit contributions | |
Amen | -0.024 |
crew | -0.022 |
Ar | -0.021 |
bandits | -0.021 |
knight | -0.021 |
Token | But |
Token ID | 1537 |
Feature activation | -7.94e-02 |
Pos logit contributions | |
Amen | +0.077 |
crew | +0.073 |
bandits | +0.069 |
lair | +0.068 |
knight | +0.067 |
Neg logit contributions | |
mm | -0.077 |
n | -0.068 |
ops | -0.067 |
uch | -0.066 |
€ | -0.065 |
Token | you |
Token ID | 345 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
ops | +0.120 |
liv | +0.113 |
€ | +0.112 |
amp | +0.108 |
mm | +0.107 |
Neg logit contributions | |
crew | -0.122 |
Amen | -0.122 |
hope | -0.112 |
knight | -0.108 |
captain | -0.106 |
Token | must |
Token ID | 1276 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
mm | +0.148 |
uch | +0.138 |
n | +0.136 |
happen | +0.135 |
ops | +0.135 |
Neg logit contributions | |
Amen | -0.169 |
Ar | -0.150 |
crew | -0.148 |
Offic | -0.145 |
lair | -0.145 |
Token | n |
Token ID | 77 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
Amen | +0.052 |
crew | +0.046 |
Ar | +0.044 |
knight | +0.043 |
known | +0.043 |
Neg logit contributions | |
mm | -0.053 |
ops | -0.047 |
ying | -0.045 |
uch | -0.045 |
tons | -0.044 |
Token | 't |
Token ID | 470 |
Feature activation | +2.96e-03 |
Pos logit contributions | |
mm | +0.473 |
tons | +0.418 |
€ | +0.406 |
belongs | +0.403 |
liv | +0.389 |
Neg logit contributions | |
Amen | -0.298 |
unts | -0.257 |
known | -0.251 |
crew | -0.246 |
Dear | -0.244 |
Token | go |
Token ID | 467 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
Amen | +0.005 |
crew | +0.004 |
Ar | +0.004 |
knight | +0.004 |
bandits | +0.004 |
Neg logit contributions | |
mm | -0.005 |
ops | -0.004 |
ying | -0.004 |
happen | -0.004 |
tons | -0.004 |
Token | too |
Token ID | 1165 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
mm | +0.188 |
ops | +0.168 |
happen | +0.158 |
uch | +0.158 |
ying | +0.156 |
Neg logit contributions | |
Amen | -0.156 |
crew | -0.140 |
Ar | -0.134 |
bandits | -0.133 |
known | -0.130 |
Token | fast |
Token ID | 3049 |
Feature activation | -8.85e-03 |
Pos logit contributions | |
mm | +0.041 |
uch | +0.039 |
happen | +0.039 |
ying | +0.038 |
tons | +0.038 |
Neg logit contributions | |
Amen | -0.037 |
crew | -0.034 |
knight | -0.031 |
Offic | -0.030 |
bandits | -0.029 |
Token | ." |
Token ID | 526 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
mm | +0.013 |
ops | +0.011 |
uch | +0.011 |
happen | +0.011 |
€ | +0.011 |
Neg logit contributions | |
Amen | -0.015 |
crew | -0.014 |
Ar | -0.013 |
bandits | -0.013 |
knight | -0.013 |
Token | pocket |
Token ID | 10000 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
mm | +0.012 |
ops | +0.011 |
happen | +0.010 |
ying | +0.010 |
uch | +0.010 |
Neg logit contributions | |
Amen | -0.009 |
crew | -0.009 |
knight | -0.008 |
lair | -0.008 |
Ar | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
Amen | +0.086 |
crew | +0.081 |
Ar | +0.076 |
lair | +0.076 |
knight | +0.076 |
Neg logit contributions | |
mm | -0.060 |
ops | -0.050 |
happen | -0.050 |
uch | -0.049 |
€ | -0.049 |
Token | " |
Token ID | 366 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
mm | +0.021 |
uch | +0.019 |
happen | +0.019 |
ops | +0.019 |
ying | +0.019 |
Neg logit contributions | |
Amen | -0.014 |
crew | -0.012 |
Ar | -0.012 |
Offic | -0.011 |
bandits | -0.011 |
Token | I |
Token ID | 40 |
Feature activation | -4.07e-03 |
Pos logit contributions | |
Amen | +0.060 |
crew | +0.055 |
bandits | +0.051 |
lair | +0.051 |
knight | +0.051 |
Neg logit contributions | |
mm | -0.070 |
uch | -0.062 |
ops | -0.062 |
n | -0.061 |
happen | -0.060 |
Token | must |
Token ID | 1276 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
n | +0.005 |
uch | +0.005 |
mm | +0.005 |
belongs | +0.005 |
happen | +0.005 |
Neg logit contributions | |
Amen | -0.007 |
bandits | -0.006 |
Offic | -0.006 |
crew | -0.006 |
lair | -0.006 |
Token | n |
Token ID | 77 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Amen | +0.049 |
crew | +0.043 |
knight | +0.040 |
Ar | +0.040 |
hope | +0.040 |
Neg logit contributions | |
mm | -0.057 |
ops | -0.052 |
ying | -0.051 |
uch | -0.049 |
liv | -0.049 |
Token | 't |
Token ID | 470 |
Feature activation | -1.96e-03 |
Pos logit contributions | |
mm | +0.450 |
tons | +0.389 |
€ | +0.387 |
belongs | +0.387 |
ops | +0.374 |
Neg logit contributions | |
Amen | -0.324 |
crew | -0.272 |
unts | -0.265 |
knight | -0.263 |
Offic | -0.262 |
Token | do |
Token ID | 466 |
Feature activation | -6.32e-02 |
Pos logit contributions | |
mm | +0.003 |
ops | +0.003 |
happen | +0.003 |
uch | +0.003 |
ying | +0.003 |
Neg logit contributions | |
Amen | -0.003 |
crew | -0.003 |
Ar | -0.003 |
bandits | -0.003 |
knight | -0.003 |
Token | that |
Token ID | 326 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
mm | +0.108 |
ying | +0.101 |
€ | +0.099 |
ops | +0.098 |
ancing | +0.098 |
Neg logit contributions | |
Amen | -0.084 |
crew | -0.077 |
knight | -0.074 |
hope | -0.073 |
bandits | -0.072 |
Token | ," |
Token ID | 553 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Amen | +0.021 |
knight | +0.020 |
crew | +0.018 |
pilot | +0.017 |
hunter | +0.017 |
Neg logit contributions | |
mm | -0.033 |
liv | -0.031 |
ancing | -0.030 |
ops | -0.030 |
ying | -0.029 |
Token | her |
Token ID | 607 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Amen | +0.140 |
wonder | +0.112 |
hope | +0.109 |
prayed | +0.105 |
Dear | +0.103 |
Neg logit contributions | |
mm | -0.194 |
ancing | -0.190 |
ops | -0.188 |
uch | -0.186 |
ying | -0.185 |
Token | . |
Token ID | 13 |
Feature activation | -4.85e-03 |
Pos logit contributions | |
Amen | +0.141 |
crew | +0.133 |
hope | +0.127 |
lair | +0.126 |
knight | +0.125 |
Neg logit contributions | |
mm | -0.139 |
uch | -0.119 |
liv | -0.117 |
€ | -0.116 |
ops | -0.115 |
Token | And |
Token ID | 843 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
mm | +0.009 |
ops | +0.008 |
n | +0.008 |
€ | +0.008 |
tons | +0.008 |
Neg logit contributions | |
Amen | -0.006 |
crew | -0.006 |
bandits | -0.005 |
Ar | -0.005 |
knight | -0.005 |
Token | that |
Token ID | 326 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
mm | +0.058 |
uch | +0.053 |
ops | +0.053 |
happen | +0.051 |
€ | +0.050 |
Neg logit contributions | |
Amen | -0.060 |
crew | -0.056 |
Ar | -0.052 |
knight | -0.052 |
known | -0.052 |
Token | 's |
Token ID | 338 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
Amen | +0.063 |
known | +0.057 |
crew | +0.057 |
knight | +0.055 |
hope | +0.055 |
Neg logit contributions | |
mm | -0.047 |
ula | -0.041 |
ancing | -0.041 |
€ | -0.040 |
ops | -0.039 |
Token | why |
Token ID | 1521 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
ops | +0.062 |
mm | +0.061 |
uch | +0.057 |
happen | +0.052 |
€ | +0.052 |
Neg logit contributions | |
Amen | -0.056 |
crew | -0.050 |
known | -0.049 |
hope | -0.047 |
knight | -0.045 |
Token | Tues |
Token ID | 48496 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
mm | +0.110 |
uch | +0.099 |
ops | +0.097 |
€ | +0.097 |
liv | +0.095 |
Neg logit contributions | |
Amen | -0.088 |
crew | -0.087 |
knight | -0.082 |
bandits | -0.080 |
known | -0.080 |
Token | days |
Token ID | 12545 |
Feature activation | -7.98e-02 |
Pos logit contributions | |
mm | +0.407 |
ops | +0.377 |
happen | +0.374 |
€ | +0.355 |
liv | +0.351 |
Neg logit contributions | |
Amen | -0.391 |
crew | -0.331 |
known | -0.311 |
Dear | -0.309 |
Ar | -0.307 |
Token | are |
Token ID | 389 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
mm | +0.153 |
€ | +0.137 |
liv | +0.135 |
ula | +0.133 |
ancing | +0.133 |
Neg logit contributions | |
Amen | -0.083 |
hope | -0.075 |
crew | -0.075 |
known | -0.073 |
received | -0.071 |
Token | so |
Token ID | 523 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
mm | +0.097 |
ops | +0.091 |
uch | +0.086 |
€ | +0.085 |
liv | +0.082 |
Neg logit contributions | |
Amen | -0.068 |
crew | -0.067 |
known | -0.066 |
knight | -0.061 |
firefighter | -0.061 |
Token | important |
Token ID | 1593 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
mm | +0.085 |
ops | +0.079 |
uch | +0.075 |
happen | +0.075 |
€ | +0.074 |
Neg logit contributions | |
Amen | -0.069 |
crew | -0.060 |
Ar | -0.058 |
bandits | -0.056 |
knight | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
uch | +0.068 |
mm | +0.067 |
€ | +0.062 |
liv | +0.060 |
ops | +0.058 |
Neg logit contributions | |
crew | -0.055 |
Amen | -0.055 |
knight | -0.052 |
hope | -0.051 |
bandits | -0.050 |
Token | No |
Token ID | 2949 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
Amen | +0.050 |
crew | +0.045 |
lair | +0.042 |
knight | +0.041 |
Ar | +0.041 |
Neg logit contributions | |
mm | -0.053 |
n | -0.048 |
uch | -0.047 |
ops | -0.047 |
happen | -0.047 |
Token | Freddy |
Token ID | 45437 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
mm | +0.164 |
uch | +0.159 |
ying | +0.148 |
€ | +0.148 |
ops | +0.145 |
Neg logit contributions | |
crew | -0.170 |
Amen | -0.168 |
bandits | -0.168 |
knight | -0.168 |
hope | -0.157 |
Token | , |
Token ID | 11 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
mm | +0.150 |
€ | +0.127 |
uct | +0.124 |
liv | +0.123 |
Whe | +0.122 |
Neg logit contributions | |
Amen | -0.149 |
crew | -0.144 |
unts | -0.142 |
knight | -0.141 |
reunited | -0.139 |
Token | you |
Token ID | 345 |
Feature activation | -7.63e-02 |
Pos logit contributions | |
mm | +0.095 |
ops | +0.091 |
€ | +0.086 |
liv | +0.085 |
uch | +0.085 |
Neg logit contributions | |
Amen | -0.101 |
crew | -0.100 |
bandits | -0.097 |
hope | -0.097 |
knight | -0.095 |
Token | must |
Token ID | 1276 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
mm | +0.111 |
uch | +0.103 |
ops | +0.100 |
happen | +0.100 |
n | +0.099 |
Neg logit contributions | |
Amen | -0.135 |
Ar | -0.118 |
crew | -0.116 |
Offic | -0.116 |
lair | -0.113 |
Token | n |
Token ID | 77 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
Amen | +0.058 |
hope | +0.053 |
known | +0.051 |
understand | +0.050 |
crew | +0.050 |
Neg logit contributions | |
mm | -0.077 |
ops | -0.072 |
ying | -0.069 |
ries | -0.069 |
aund | -0.067 |
Token | 't |
Token ID | 470 |
Feature activation | +7.77e-03 |
Pos logit contributions | |
mm | +0.482 |
tons | +0.418 |
belongs | +0.411 |
€ | +0.399 |
yogurt | +0.398 |
Neg logit contributions | |
Amen | -0.256 |
unts | -0.233 |
understand | -0.220 |
hear | -0.213 |
known | -0.211 |
Token | pinch |
Token ID | 25394 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
Amen | +0.010 |
understand | +0.009 |
crew | +0.009 |
knight | +0.009 |
Ar | +0.009 |
Neg logit contributions | |
mm | -0.015 |
liv | -0.013 |
ops | -0.013 |
tons | -0.012 |
ying | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
mm | +0.062 |
liv | +0.057 |
ops | +0.057 |
uch | +0.052 |
€ | +0.052 |
Neg logit contributions | |
Amen | -0.059 |
bandits | -0.057 |
hope | -0.056 |
Fox | -0.055 |
crew | -0.055 |
Token | bee |
Token ID | 20697 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
mm | +0.062 |
ops | +0.056 |
happen | +0.056 |
uch | +0.055 |
n | +0.054 |
Neg logit contributions | |
Amen | -0.049 |
crew | -0.043 |
Ar | -0.042 |
bandits | -0.041 |
Offic | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
Amen | +0.047 |
crew | +0.042 |
hope | +0.041 |
knight | +0.041 |
mouse | +0.041 |
Neg logit contributions | |
mm | -0.053 |
liv | -0.045 |
€ | -0.043 |
tons | -0.042 |
ops | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
mm | +0.138 |
uch | +0.124 |
ops | +0.122 |
ying | +0.115 |
€ | +0.113 |
Neg logit contributions | |
Amen | -0.192 |
crew | -0.175 |
hope | -0.167 |
knight | -0.167 |
bandits | -0.163 |
Token | Bobby |
Token ID | 17155 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
mm | +0.088 |
ops | +0.085 |
uch | +0.082 |
€ | +0.079 |
liv | +0.075 |
Neg logit contributions | |
Amen | -0.103 |
crew | -0.096 |
hope | -0.090 |
knight | -0.088 |
bandits | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
mm | +0.067 |
uch | +0.055 |
ops | +0.055 |
happen | +0.054 |
ying | +0.053 |
Neg logit contributions | |
Amen | -0.089 |
crew | -0.080 |
knight | -0.076 |
hope | -0.075 |
Dear | -0.074 |
Token | We |
Token ID | 775 |
Feature activation | +3.43e-04 |
Pos logit contributions | |
mm | +0.051 |
happen | +0.048 |
ops | +0.047 |
uch | +0.046 |
ying | +0.044 |
Neg logit contributions | |
Amen | -0.072 |
crew | -0.064 |
Ar | -0.061 |
knight | -0.060 |
Offic | -0.060 |
Token | must |
Token ID | 1276 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
Amen | +0.001 |
Offic | +0.001 |
Ar | +0.001 |
lair | +0.001 |
crew | +0.001 |
Neg logit contributions | |
n | -0.000 |
mm | -0.000 |
ops | -0.000 |
uch | -0.000 |
happen | -0.000 |
Token | n |
Token ID | 77 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
Amen | +0.076 |
crew | +0.066 |
Ar | +0.062 |
known | +0.062 |
hope | +0.062 |
Neg logit contributions | |
mm | -0.083 |
ops | -0.075 |
uch | -0.072 |
ying | -0.071 |
tons | -0.070 |
Token | 't |
Token ID | 470 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
mm | +0.392 |
tons | +0.330 |
belongs | +0.327 |
€ | +0.326 |
ops | +0.321 |
Neg logit contributions | |
Amen | -0.363 |
crew | -0.317 |
unts | -0.310 |
known | -0.307 |
Dear | -0.300 |
Token | borrow |
Token ID | 8804 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
Amen | +0.026 |
crew | +0.023 |
Ar | +0.022 |
knight | +0.022 |
Dear | +0.022 |
Neg logit contributions | |
mm | -0.028 |
ops | -0.025 |
ying | -0.025 |
liv | -0.024 |
uch | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
mm | +0.059 |
ops | +0.054 |
uch | +0.053 |
liv | +0.051 |
€ | +0.051 |
Neg logit contributions | |
Amen | -0.053 |
crew | -0.050 |
bandits | -0.047 |
knight | -0.047 |
Ar | -0.047 |
Token | tray |
Token ID | 26473 |
Feature activation | +2.38e-03 |
Pos logit contributions | |
mm | +0.050 |
ops | +0.044 |
uch | +0.043 |
happen | +0.043 |
ying | +0.042 |
Neg logit contributions | |
Amen | -0.047 |
crew | -0.042 |
Ar | -0.041 |
knight | -0.040 |
bandits | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
Amen | +0.004 |
crew | +0.004 |
knight | +0.004 |
Dear | +0.003 |
lair | +0.003 |
Neg logit contributions | |
mm | -0.003 |
ops | -0.003 |
uch | -0.003 |
happen | -0.003 |
€ | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +6.91e-03 |
Pos logit contributions | |
Amen | +0.004 |
crew | +0.004 |
knight | +0.004 |
unts | +0.004 |
knowing | +0.003 |
Neg logit contributions | |
mm | -0.008 |
uch | -0.008 |
€ | -0.008 |
liv | -0.007 |
n | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
Amen | +0.011 |
crew | +0.009 |
Ar | +0.009 |
knight | +0.009 |
bandits | +0.009 |
Neg logit contributions | |
mm | -0.011 |
ops | -0.010 |
happen | -0.010 |
n | -0.010 |
uch | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -3.62e-03 |
Pos logit contributions | |
Amen | +0.039 |
hope | +0.035 |
crew | +0.035 |
Some | +0.034 |
firefighter | +0.034 |
Neg logit contributions | |
mm | -0.038 |
uch | -0.035 |
ops | -0.034 |
€ | -0.034 |
liv | -0.032 |
Token | " |
Token ID | 366 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
mm | +0.006 |
ops | +0.006 |
uch | +0.006 |
happen | +0.006 |
ying | +0.005 |
Neg logit contributions | |
Amen | -0.005 |
crew | -0.004 |
Ar | -0.004 |
bandits | -0.004 |
knight | -0.004 |
Token | U |
Token ID | 52 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
Amen | +0.070 |
crew | +0.061 |
Ar | +0.060 |
bandits | +0.059 |
Offic | +0.058 |
Neg logit contributions | |
mm | -0.074 |
ops | -0.066 |
happen | -0.065 |
uch | -0.064 |
ying | -0.063 |
Token | gh |
Token ID | 456 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
existed | +0.251 |
fastest | +0.215 |
estival | +0.206 |
halfway | +0.203 |
tons | +0.198 |
Neg logit contributions | |
Amen | -0.399 |
hope | -0.373 |
Karl | -0.342 |
unts | -0.339 |
crew | -0.337 |
Token | , |
Token ID | 11 |
Feature activation | -5.53e-02 |
Pos logit contributions | |
happen | +0.331 |
mm | +0.325 |
€ | +0.296 |
belongs | +0.289 |
tons | +0.287 |
Neg logit contributions | |
Amen | -0.370 |
crew | -0.359 |
parent | -0.319 |
knight | -0.317 |
hope | -0.316 |
Token | what |
Token ID | 644 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
mm | +0.081 |
ops | +0.079 |
uch | +0.073 |
€ | +0.073 |
ying | +0.068 |
Neg logit contributions | |
Amen | -0.085 |
crew | -0.082 |
hope | -0.079 |
knight | -0.078 |
parent | -0.075 |
Token | are |
Token ID | 389 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
mm | +0.061 |
ops | +0.053 |
tons | +0.052 |
€ | +0.051 |
liv | +0.050 |
Neg logit contributions | |
Amen | -0.057 |
knight | -0.054 |
crew | -0.054 |
known | -0.053 |
understand | -0.053 |
Token | you |
Token ID | 345 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
€ | +0.067 |
inning | +0.065 |
ops | +0.065 |
amp | +0.063 |
ement | +0.063 |
Neg logit contributions | |
many | -0.046 |
crew | -0.045 |
bandits | -0.042 |
backup | -0.041 |
firefighters | -0.040 |
Token | doing |
Token ID | 1804 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
mm | +0.136 |
ops | +0.111 |
belongs | +0.103 |
tons | +0.102 |
liv | +0.101 |
Neg logit contributions | |
Amen | -0.200 |
crew | -0.195 |
known | -0.186 |
recorded | -0.179 |
knight | -0.179 |
Token | said |
Token ID | 531 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
Amen | +0.052 |
crew | +0.046 |
Ar | +0.045 |
bandits | +0.044 |
knight | +0.044 |
Neg logit contributions | |
mm | -0.055 |
ops | -0.049 |
happen | -0.049 |
uch | -0.049 |
n | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | +2.14e-03 |
Pos logit contributions | |
Amen | +0.054 |
crew | +0.050 |
knight | +0.047 |
firefighter | +0.047 |
messenger | +0.046 |
Neg logit contributions | |
mm | -0.045 |
uch | -0.041 |
ops | -0.040 |
liv | -0.039 |
€ | -0.038 |
Token | " |
Token ID | 366 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
Amen | +0.003 |
crew | +0.002 |
Ar | +0.002 |
knight | +0.002 |
Offic | +0.002 |
Neg logit contributions | |
mm | -0.004 |
uch | -0.004 |
ops | -0.004 |
happen | -0.004 |
ying | -0.004 |
Token | I |
Token ID | 40 |
Feature activation | -1.16e-03 |
Pos logit contributions | |
Amen | +0.075 |
crew | +0.066 |
Ar | +0.065 |
bandits | +0.062 |
knight | +0.062 |
Neg logit contributions | |
mm | -0.084 |
ops | -0.075 |
happen | -0.074 |
uch | -0.073 |
ying | -0.072 |
Token | like |
Token ID | 588 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
mm | +0.002 |
n | +0.002 |
uch | +0.002 |
ops | +0.002 |
happen | +0.002 |
Neg logit contributions | |
Amen | -0.002 |
crew | -0.002 |
bandits | -0.002 |
lair | -0.001 |
Offic | -0.001 |
Token | Tues |
Token ID | 48496 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
mm | +0.045 |
ops | +0.042 |
uch | +0.040 |
€ | +0.039 |
ying | +0.039 |
Neg logit contributions | |
Amen | -0.045 |
crew | -0.040 |
bandits | -0.038 |
Ar | -0.038 |
knight | -0.037 |
Token | days |
Token ID | 12545 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
mm | +0.415 |
ops | +0.378 |
happen | +0.376 |
€ | +0.360 |
ying | +0.360 |
Neg logit contributions | |
Amen | -0.359 |
crew | -0.304 |
Ar | -0.290 |
Dear | -0.280 |
Offic | -0.278 |
Token | too |
Token ID | 1165 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
mm | +0.106 |
uch | +0.098 |
ops | +0.094 |
liv | +0.094 |
ying | +0.093 |
Neg logit contributions | |
Amen | -0.104 |
crew | -0.087 |
knight | -0.087 |
Dear | -0.086 |
hope | -0.086 |
Token | ! |
Token ID | 0 |
Feature activation | -5.73e-02 |
Pos logit contributions | |
mm | +0.042 |
ops | +0.038 |
uch | +0.037 |
happen | +0.037 |
€ | +0.036 |
Neg logit contributions | |
Amen | -0.050 |
crew | -0.043 |
Ar | -0.043 |
unts | -0.043 |
Dear | -0.042 |
Token | It |
Token ID | 632 |
Feature activation | -2.66e-03 |
Pos logit contributions | |
mm | +0.078 |
happen | +0.074 |
ops | +0.074 |
uch | +0.072 |
ying | +0.071 |
Neg logit contributions | |
Amen | -0.105 |
crew | -0.089 |
Ar | -0.087 |
knight | -0.087 |
Offic | -0.085 |
Token | 's |
Token ID | 338 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
mm | +0.004 |
ops | +0.003 |
ula | +0.003 |
gh | +0.003 |
ying | +0.003 |
Neg logit contributions | |
Amen | -0.005 |
hope | -0.004 |
crew | -0.004 |
knight | -0.004 |
known | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +7.48e-02 |
Pos logit contributions | |
Amen | +0.117 |
crew | +0.104 |
Ar | +0.099 |
bandits | +0.098 |
knight | +0.097 |
Neg logit contributions | |
mm | -0.127 |
ops | -0.114 |
uch | -0.111 |
happen | -0.110 |
n | -0.109 |
Token | new |
Token ID | 649 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
Amen | +0.132 |
crew | +0.118 |
bandits | +0.115 |
Ar | +0.114 |
Offic | +0.113 |
Neg logit contributions | |
mm | -0.102 |
happen | -0.093 |
ops | -0.090 |
n | -0.087 |
uch | -0.087 |
Token | word |
Token ID | 1573 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
Amen | +0.102 |
crew | +0.092 |
Ar | +0.089 |
knight | +0.086 |
bandits | +0.086 |
Neg logit contributions | |
mm | -0.085 |
ops | -0.074 |
uch | -0.073 |
happen | -0.072 |
ying | -0.072 |
Token | : |
Token ID | 25 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
mm | +0.020 |
uch | +0.019 |
€ | +0.018 |
liv | +0.017 |
ops | +0.017 |
Neg logit contributions | |
Amen | -0.021 |
crew | -0.021 |
Dear | -0.019 |
knight | -0.019 |
unts | -0.019 |
Token | " |
Token ID | 366 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
Amen | +0.036 |
crew | +0.032 |
Ar | +0.030 |
knight | +0.030 |
bandits | +0.030 |
Neg logit contributions | |
mm | -0.040 |
ops | -0.036 |
uch | -0.036 |
€ | -0.035 |
liv | -0.035 |
Token | n |
Token ID | 77 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
Amen | +0.056 |
Ar | +0.051 |
Dear | +0.049 |
crew | +0.048 |
Offic | +0.046 |
Neg logit contributions | |
mm | -0.077 |
happen | -0.071 |
ying | -0.070 |
ops | -0.070 |
tons | -0.069 |
Token | est |
Token ID | 395 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
mm | +0.403 |
tons | +0.345 |
existed | +0.334 |
belongs | +0.333 |
yogurt | +0.333 |
Neg logit contributions | |
Amen | -0.346 |
unts | -0.323 |
crew | -0.295 |
Dear | -0.284 |
known | -0.283 |
Token | ". |
Token ID | 1911 |
Feature activation | +1.14e-03 |
Pos logit contributions | |
mm | +0.083 |
ops | +0.075 |
uch | +0.075 |
€ | +0.074 |
liv | +0.074 |
Neg logit contributions | |
Amen | -0.083 |
crew | -0.077 |
Ar | -0.072 |
bandits | -0.071 |
knight | -0.071 |
Token | They |
Token ID | 1119 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
Amen | +0.002 |
crew | +0.002 |
Offic | +0.002 |
messenger | +0.002 |
bandits | +0.002 |
Neg logit contributions | |
mm | -0.002 |
€ | -0.002 |
ops | -0.002 |
n | -0.002 |
tons | -0.002 |
Token | say |
Token ID | 910 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
Amen | +0.077 |
crew | +0.069 |
Ar | +0.066 |
lair | +0.065 |
Offic | +0.065 |
Neg logit contributions | |
mm | -0.070 |
ops | -0.063 |
happen | -0.063 |
n | -0.061 |
uch | -0.060 |
Token | it |
Token ID | 340 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
Amen | +0.020 |
crew | +0.017 |
knight | +0.016 |
Ar | +0.016 |
bandits | +0.016 |
Neg logit contributions | |
mm | -0.026 |
uch | -0.024 |
ops | -0.024 |
liv | -0.023 |
happen | -0.023 |
Token | you |
Token ID | 345 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
mm | +0.062 |
ops | +0.058 |
€ | +0.052 |
aund | +0.051 |
liv | +0.051 |
Neg logit contributions | |
Amen | -0.051 |
crew | -0.046 |
bandits | -0.045 |
known | -0.045 |
hope | -0.044 |
Token | safe |
Token ID | 3338 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
mm | +0.121 |
ops | +0.109 |
liv | +0.106 |
aund | +0.105 |
€ | +0.105 |
Neg logit contributions | |
Amen | -0.056 |
crew | -0.052 |
hope | -0.046 |
Ar | -0.043 |
reunited | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
Amen | +0.083 |
crew | +0.076 |
Ar | +0.071 |
knight | +0.071 |
bandits | +0.070 |
Neg logit contributions | |
mm | -0.077 |
ops | -0.069 |
uch | -0.067 |
liv | -0.066 |
€ | -0.065 |
Token | You |
Token ID | 921 |
Feature activation | -5.84e-02 |
Pos logit contributions | |
mm | +0.061 |
ops | +0.055 |
happen | +0.054 |
uch | +0.053 |
ying | +0.052 |
Neg logit contributions | |
Amen | -0.076 |
crew | -0.068 |
Ar | -0.066 |
knight | -0.064 |
bandits | -0.064 |
Token | must |
Token ID | 1276 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
n | +0.077 |
uch | +0.075 |
mm | +0.074 |
happen | +0.074 |
belongs | +0.074 |
Neg logit contributions | |
Amen | -0.102 |
Offic | -0.095 |
lair | -0.092 |
Ar | -0.092 |
bandits | -0.088 |
Token | n |
Token ID | 77 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
Amen | +0.080 |
crew | +0.069 |
hope | +0.067 |
known | +0.066 |
Ar | +0.065 |
Neg logit contributions | |
mm | -0.090 |
ops | -0.081 |
ries | -0.077 |
uch | -0.077 |
ying | -0.076 |
Token | 't |
Token ID | 470 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
mm | +0.457 |
belongs | +0.404 |
€ | +0.401 |
tons | +0.394 |
happen | +0.387 |
Neg logit contributions | |
Amen | -0.255 |
known | -0.222 |
crew | -0.215 |
unts | -0.215 |
understand | -0.210 |
Token | go |
Token ID | 467 |
Feature activation | -9.82e-02 |
Pos logit contributions | |
Amen | +0.035 |
crew | +0.030 |
Ar | +0.029 |
Dear | +0.029 |
knight | +0.028 |
Neg logit contributions | |
mm | -0.039 |
ops | -0.034 |
ying | -0.033 |
liv | -0.032 |
ries | -0.032 |
Token | running |
Token ID | 2491 |
Feature activation | -7.72e-02 |
Pos logit contributions | |
mm | +0.178 |
ops | +0.157 |
liv | +0.147 |
uch | +0.144 |
€ | +0.144 |
Neg logit contributions | |
Amen | -0.140 |
crew | -0.126 |
bandits | -0.123 |
known | -0.121 |
Ar | -0.121 |
Token | off |
Token ID | 572 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
mm | +0.124 |
ops | +0.107 |
uch | +0.105 |
happen | +0.103 |
liv | +0.102 |
Neg logit contributions | |
Amen | -0.121 |
crew | -0.110 |
Ar | -0.107 |
bandits | -0.106 |
knight | -0.104 |
Token | like |
Token ID | 588 |
Feature activation | +6.15e-03 |
Pos logit contributions | |
mm | +0.025 |
happen | +0.022 |
ops | +0.022 |
n | +0.022 |
uch | +0.022 |
Neg logit contributions | |
Amen | -0.034 |
crew | -0.030 |
Ar | -0.029 |
Offic | -0.029 |
bandits | -0.029 |
Token | ' |
Token ID | 705 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
Amen | +0.002 |
crew | +0.001 |
Ar | +0.001 |
bandits | +0.001 |
knight | +0.001 |
Neg logit contributions | |
mm | -0.002 |
ops | -0.002 |
happen | -0.002 |
n | -0.002 |
tons | -0.002 |
Token | It |
Token ID | 1026 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
mm | +0.046 |
happen | +0.042 |
ops | +0.042 |
ying | +0.041 |
uch | +0.040 |
Neg logit contributions | |
Amen | -0.035 |
Ar | -0.030 |
Dear | -0.029 |
Offic | -0.029 |
crew | -0.029 |
Token | says |
Token ID | 1139 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
mm | +0.037 |
ops | +0.031 |
tons | +0.029 |
uch | +0.029 |
€ | +0.028 |
Neg logit contributions | |
Amen | -0.068 |
crew | -0.062 |
Dear | -0.059 |
knight | -0.059 |
known | -0.059 |
Token | we |
Token ID | 356 |
Feature activation | -2.88e-03 |
Pos logit contributions | |
mm | +0.017 |
ops | +0.016 |
liv | +0.015 |
uch | +0.015 |
ying | +0.015 |
Neg logit contributions | |
Amen | -0.020 |
crew | -0.018 |
messenger | -0.017 |
knight | -0.017 |
Queen | -0.017 |
Token | must |
Token ID | 1276 |
Feature activation | +5.39e-02 |
Pos logit contributions | |
mm | +0.004 |
happen | +0.004 |
n | +0.004 |
ops | +0.004 |
belongs | +0.004 |
Neg logit contributions | |
Amen | -0.005 |
Ar | -0.004 |
crew | -0.004 |
Offic | -0.004 |
lair | -0.004 |
Token | n |
Token ID | 77 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
Amen | +0.082 |
crew | +0.073 |
Ar | +0.070 |
bandits | +0.068 |
knight | +0.068 |
Neg logit contributions | |
mm | -0.088 |
ops | -0.079 |
uch | -0.077 |
happen | -0.076 |
ying | -0.076 |
Token | 't |
Token ID | 470 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
mm | +0.375 |
tons | +0.322 |
€ | +0.306 |
ops | +0.305 |
belongs | +0.303 |
Neg logit contributions | |
Amen | -0.376 |
crew | -0.328 |
unts | -0.320 |
Dear | -0.317 |
known | -0.316 |
Token | borrow |
Token ID | 8804 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
Amen | +0.031 |
crew | +0.028 |
Ar | +0.027 |
bandits | +0.026 |
knight | +0.026 |
Neg logit contributions | |
mm | -0.035 |
ops | -0.031 |
happen | -0.031 |
uch | -0.031 |
n | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
mm | +0.043 |
ops | +0.039 |
uch | +0.038 |
ying | +0.037 |
€ | +0.037 |
Neg logit contributions | |
Amen | -0.038 |
crew | -0.035 |
bandits | -0.033 |
Ar | -0.033 |
knight | -0.033 |
Token | We |
Token ID | 775 |
Feature activation | +7.43e-03 |
Pos logit contributions | |
mm | +0.038 |
ops | +0.033 |
happen | +0.032 |
n | +0.032 |
uch | +0.032 |
Neg logit contributions | |
Amen | -0.047 |
crew | -0.043 |
Ar | -0.042 |
bandits | -0.041 |
knight | -0.041 |
Token | must |
Token ID | 1276 |
Feature activation | +4.88e-02 |
Pos logit contributions | |
Amen | +0.012 |
crew | +0.010 |
Ar | +0.010 |
lair | +0.010 |
Offic | +0.010 |
Neg logit contributions | |
mm | -0.011 |
n | -0.011 |
ops | -0.010 |
happen | -0.010 |
uch | -0.010 |
Token | " |
Token ID | 1 |
Feature activation | -9.69e-03 |
Pos logit contributions | |
Amen | +0.059 |
crew | +0.053 |
Ar | +0.050 |
bandits | +0.050 |
knight | +0.049 |
Neg logit contributions | |
mm | -0.062 |
ops | -0.056 |
n | -0.054 |
uch | -0.054 |
happen | -0.054 |
Token | Rachel |
Token ID | 44045 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
mm | +0.017 |
ops | +0.015 |
uch | +0.015 |
n | +0.015 |
happen | +0.015 |
Neg logit contributions | |
Amen | -0.013 |
crew | -0.012 |
lair | -0.011 |
bandits | -0.011 |
knight | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
mm | +0.035 |
ying | +0.030 |
ops | +0.030 |
€ | +0.030 |
tons | +0.029 |
Neg logit contributions | |
Amen | -0.045 |
crew | -0.041 |
knight | -0.040 |
hope | -0.039 |
messenger | -0.038 |
Token | this |
Token ID | 428 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
mm | +0.051 |
ops | +0.046 |
uch | +0.044 |
€ | +0.043 |
happen | +0.042 |
Neg logit contributions | |
Amen | -0.063 |
crew | -0.057 |
knight | -0.055 |
bandits | -0.054 |
Ar | -0.054 |
Token | is |
Token ID | 318 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
mm | +0.173 |
ops | +0.153 |
happen | +0.146 |
ying | +0.146 |
uch | +0.145 |
Neg logit contributions | |
Amen | -0.212 |
crew | -0.190 |
Ar | -0.184 |
knight | -0.180 |
Dear | -0.178 |
Token | O |
Token ID | 440 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
ops | +0.075 |
mm | +0.073 |
ying | +0.068 |
ancing | +0.064 |
uch | +0.062 |
Neg logit contributions | |
Amen | -0.066 |
known | -0.060 |
crew | -0.059 |
knight | -0.057 |
Dear | -0.055 |
Token | at |
Token ID | 265 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
tons | +0.183 |
ying | +0.179 |
mm | +0.173 |
existed | +0.172 |
happen | +0.164 |
Neg logit contributions | |
Amen | -0.457 |
crew | -0.453 |
knight | -0.428 |
hunter | -0.417 |
Fox | -0.417 |
Token | ," |
Token ID | 553 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
mm | +0.228 |
€ | +0.209 |
tons | +0.207 |
liv | +0.204 |
aund | +0.202 |
Neg logit contributions | |
Amen | -0.208 |
crew | -0.183 |
knight | -0.178 |
firefighter | -0.171 |
Dear | -0.166 |
Token | said |
Token ID | 531 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
Amen | +0.226 |
crew | +0.204 |
Ar | +0.199 |
bandits | +0.195 |
knight | +0.194 |
Neg logit contributions | |
mm | -0.184 |
ops | -0.161 |
happen | -0.158 |
uch | -0.156 |
n | -0.155 |
Token | Polly |
Token ID | 36898 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
Amen | +0.099 |
crew | +0.091 |
knight | +0.086 |
Ar | +0.084 |
known | +0.084 |
Neg logit contributions | |
mm | -0.095 |
ops | -0.084 |
uch | -0.084 |
happen | -0.081 |
ying | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | +7.03e-03 |
Pos logit contributions | |
mm | +0.016 |
uch | +0.013 |
ying | +0.013 |
happen | +0.013 |
ops | +0.013 |
Neg logit contributions | |
Amen | -0.019 |
crew | -0.018 |
knight | -0.017 |
reunited | -0.016 |
lair | -0.016 |
Token | box |
Token ID | 3091 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
mm | +0.042 |
ops | +0.037 |
happen | +0.036 |
uch | +0.036 |
tons | +0.035 |
Neg logit contributions | |
Amen | -0.050 |
crew | -0.044 |
Ar | -0.042 |
bandits | -0.042 |
knight | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
Amen | +0.113 |
bandits | +0.102 |
Ar | +0.101 |
Offic | +0.100 |
crew | +0.099 |
Neg logit contributions | |
mm | -0.050 |
ops | -0.046 |
happen | -0.045 |
ying | -0.044 |
n | -0.044 |
Token | grab |
Token ID | 5552 |
Feature activation | -7.52e-02 |
Pos logit contributions | |
mm | +0.018 |
ops | +0.016 |
happen | +0.016 |
uch | +0.015 |
n | +0.015 |
Neg logit contributions | |
Amen | -0.016 |
crew | -0.014 |
Ar | -0.014 |
bandits | -0.013 |
knight | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +7.49e-02 |
Pos logit contributions | |
mm | +0.149 |
uch | +0.137 |
ops | +0.132 |
liv | +0.130 |
€ | +0.130 |
Neg logit contributions | |
Amen | -0.085 |
Ar | -0.083 |
crew | -0.082 |
hope | -0.080 |
bandits | -0.075 |
Token | bike |
Token ID | 7161 |
Feature activation | +2.60e-03 |
Pos logit contributions | |
Amen | +0.125 |
crew | +0.112 |
Ar | +0.108 |
bandits | +0.106 |
knight | +0.105 |
Neg logit contributions | |
mm | -0.112 |
ops | -0.099 |
happen | -0.098 |
uch | -0.096 |
n | -0.095 |
Token | each |
Token ID | 1123 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Amen | +0.005 |
crew | +0.004 |
Ar | +0.004 |
Dear | +0.004 |
knight | +0.004 |
Neg logit contributions | |
mm | -0.004 |
uch | -0.003 |
liv | -0.003 |
€ | -0.003 |
tons | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
Amen | +0.135 |
Ar | +0.131 |
Dear | +0.120 |
hope | +0.120 |
understand | +0.120 |
Neg logit contributions | |
mm | -0.179 |
liv | -0.154 |
ying | -0.152 |
uch | -0.150 |
happen | -0.149 |
Token | Ben |
Token ID | 3932 |
Feature activation | -6.36e-02 |
Pos logit contributions | |
mm | +0.052 |
ops | +0.046 |
tons | +0.045 |
n | +0.045 |
happen | +0.045 |
Neg logit contributions | |
Amen | -0.037 |
crew | -0.032 |
bandits | -0.031 |
knight | -0.030 |
lair | -0.030 |
Token | takes |
Token ID | 2753 |
Feature activation | -7.58e-03 |
Pos logit contributions | |
mm | +0.105 |
uch | +0.092 |
€ | +0.089 |
liv | +0.088 |
happen | +0.088 |
Neg logit contributions | |
hope | -0.092 |
Amen | -0.090 |
knight | -0.089 |
Ar | -0.089 |
bandits | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
mm | +0.014 |
ops | +0.012 |
uch | +0.012 |
happen | +0.012 |
ying | +0.012 |
Neg logit contributions | |
Amen | -0.010 |
crew | -0.010 |
Ar | -0.010 |
knight | -0.009 |
bandits | -0.009 |
Token | red |
Token ID | 2266 |
Feature activation | -8.24e-03 |
Pos logit contributions | |
mm | +0.024 |
ops | +0.021 |
uch | +0.021 |
ying | +0.020 |
n | +0.020 |
Neg logit contributions | |
Amen | -0.025 |
crew | -0.023 |
Ar | -0.023 |
knight | -0.022 |
bandits | -0.022 |
Token | bird |
Token ID | 6512 |
Feature activation | +6.93e-02 |
Pos logit contributions | |
mm | +0.027 |
ops | +0.025 |
happen | +0.025 |
uch | +0.024 |
ying | +0.024 |
Neg logit contributions | |
Amen | -0.013 |
crew | -0.010 |
Ar | -0.010 |
knight | -0.010 |
bandits | -0.010 |
Token | replied |
Token ID | 8712 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
Amen | +0.104 |
crew | +0.092 |
Ar | +0.088 |
knight | +0.086 |
bandits | +0.086 |
Neg logit contributions | |
mm | -0.116 |
ops | -0.103 |
happen | -0.102 |
uch | -0.100 |
ying | -0.099 |
Token | in |
Token ID | 287 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
Amen | +0.076 |
crew | +0.069 |
firefighter | +0.065 |
messenger | +0.064 |
knight | +0.064 |
Neg logit contributions | |
mm | -0.069 |
liv | -0.063 |
ops | -0.062 |
€ | -0.061 |
happen | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | +9.16e-02 |
Pos logit contributions | |
Amen | +0.035 |
crew | +0.029 |
knight | +0.028 |
Ar | +0.028 |
Offic | +0.027 |
Neg logit contributions | |
mm | -0.052 |
ops | -0.050 |
ying | -0.049 |
uch | -0.048 |
belongs | -0.047 |
Token | gentle |
Token ID | 10296 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Amen | +0.148 |
crew | +0.144 |
generation | +0.142 |
subjects | +0.140 |
lair | +0.140 |
Neg logit contributions | |
happen | -0.110 |
€ | -0.104 |
belongs | -0.102 |
gh | -0.100 |
n | -0.099 |
Token | voice |
Token ID | 3809 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
Amen | +0.159 |
crew | +0.147 |
Ar | +0.146 |
Offic | +0.146 |
Dear | +0.143 |
Neg logit contributions | |
happen | -0.180 |
n | -0.177 |
mm | -0.176 |
€ | -0.170 |
gh | -0.163 |
Token | , |
Token ID | 11 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
Amen | +0.182 |
crew | +0.163 |
knight | +0.151 |
firefighter | +0.150 |
alien | +0.149 |
Neg logit contributions | |
mm | -0.204 |
uch | -0.193 |
liv | -0.185 |
Whe | -0.179 |
€ | -0.177 |
Token | � |
Token ID | 6184 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
mm | +0.026 |
ops | +0.023 |
uch | +0.023 |
happen | +0.023 |
ying | +0.022 |
Neg logit contributions | |
Amen | -0.021 |
crew | -0.018 |
Ar | -0.017 |
knight | -0.017 |
bandits | -0.017 |
Token | � |
Token ID | 95 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
ying | +0.163 |
mm | +0.162 |
ops | +0.153 |
happen | +0.152 |
ries | +0.149 |
Neg logit contributions | |
Amen | -0.190 |
Offic | -0.162 |
hunter | -0.162 |
Fox | -0.156 |
crew | -0.154 |
Token | € |
Token ID | 26391 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
mm | +0.147 |
ying | +0.138 |
ops | +0.134 |
happen | +0.132 |
ries | +0.130 |
Neg logit contributions | |
Amen | -0.160 |
Offic | -0.139 |
Ar | -0.135 |
crew | -0.131 |
Dear | -0.131 |
Token | � |
Token ID | 129 |
Feature activation | -8.16e-02 |
Pos logit contributions | |
mm | +0.220 |
ops | +0.196 |
uch | +0.192 |
n | +0.187 |
happen | +0.186 |
Neg logit contributions | |
Amen | -0.250 |
crew | -0.227 |
Ar | -0.220 |
bandits | -0.216 |
knight | -0.215 |
Token | day |
Token ID | 1110 |
Feature activation | -8.68e-02 |
Pos logit contributions | |
Amen | +0.035 |
crew | +0.032 |
knight | +0.032 |
Ar | +0.031 |
hunter | +0.031 |
Neg logit contributions | |
mm | -0.029 |
ops | -0.026 |
ancing | -0.025 |
happen | -0.024 |
liv | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
mm | +0.122 |
ops | +0.108 |
uch | +0.106 |
€ | +0.105 |
happen | +0.105 |
Neg logit contributions | |
Amen | -0.145 |
crew | -0.139 |
Ar | -0.135 |
firefighter | -0.134 |
knight | -0.133 |
Token | Bob |
Token ID | 5811 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
mm | +0.029 |
ops | +0.025 |
happen | +0.024 |
tons | +0.024 |
uch | +0.024 |
Neg logit contributions | |
Amen | -0.041 |
crew | -0.036 |
Ar | -0.035 |
bandits | -0.035 |
Offic | -0.035 |
Token | asked |
Token ID | 1965 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
mm | +0.081 |
ops | +0.076 |
uch | +0.075 |
n | +0.075 |
belongs | +0.073 |
Neg logit contributions | |
Amen | -0.074 |
crew | -0.061 |
Ar | -0.060 |
Offic | -0.059 |
bandits | -0.058 |
Token | the |
Token ID | 262 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
Amen | +0.127 |
crew | +0.113 |
Ar | +0.112 |
Offic | +0.108 |
Dear | +0.108 |
Neg logit contributions | |
mm | -0.103 |
happen | -0.092 |
n | -0.089 |
ops | -0.089 |
ying | -0.089 |
Token | hawk |
Token ID | 48710 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
mm | +0.040 |
happen | +0.039 |
tons | +0.039 |
next | +0.039 |
uch | +0.038 |
Neg logit contributions | |
Amen | -0.032 |
Offic | -0.027 |
Ar | -0.027 |
crew | -0.027 |
unts | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
Amen | +0.168 |
crew | +0.152 |
Ar | +0.144 |
knight | +0.143 |
bandits | +0.143 |
Neg logit contributions | |
mm | -0.150 |
ops | -0.132 |
uch | -0.129 |
happen | -0.129 |
€ | -0.125 |
Token | lend |
Token ID | 22096 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
Amen | +0.074 |
Ar | +0.066 |
unts | +0.064 |
crew | +0.063 |
Offic | +0.063 |
Neg logit contributions | |
happen | -0.068 |
mm | -0.067 |
n | -0.064 |
uch | -0.060 |
tons | -0.060 |
Token | him |
Token ID | 683 |
Feature activation | -3.09e-03 |
Pos logit contributions | |
Amen | +0.023 |
crew | +0.020 |
Ar | +0.020 |
bandits | +0.019 |
knight | +0.019 |
Neg logit contributions | |
mm | -0.028 |
ops | -0.026 |
uch | -0.025 |
happen | -0.025 |
ying | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | +8.92e-02 |
Pos logit contributions | |
mm | +0.006 |
ops | +0.006 |
uch | +0.005 |
liv | +0.005 |
€ | +0.005 |
Neg logit contributions | |
crew | -0.003 |
Amen | -0.003 |
hope | -0.003 |
bandits | -0.003 |
knight | -0.003 |
Token | wing |
Token ID | 8539 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Amen | +0.150 |
Offic | +0.132 |
Ar | +0.128 |
bandits | +0.127 |
crew | +0.127 |
Neg logit contributions | |
happen | -0.121 |
mm | -0.113 |
uch | -0.113 |
belongs | -0.111 |
n | -0.107 |
Token | from |
Token ID | 422 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
mm | +0.064 |
liv | +0.055 |
€ | +0.055 |
tons | +0.054 |
ops | +0.053 |
Neg logit contributions | |
Amen | -0.059 |
crew | -0.058 |
bandits | -0.055 |
knight | -0.054 |
understand | -0.054 |
Token | breaking |
Token ID | 7163 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
mm | +0.075 |
happen | +0.068 |
ops | +0.065 |
n | +0.065 |
uch | +0.064 |
Neg logit contributions | |
Amen | -0.107 |
crew | -0.096 |
Ar | -0.094 |
Offic | -0.092 |
knight | -0.091 |
Token | and |
Token ID | 290 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
mm | +0.029 |
ops | +0.025 |
uch | +0.025 |
€ | +0.024 |
happen | +0.024 |
Neg logit contributions | |
Amen | -0.033 |
crew | -0.031 |
bandits | -0.030 |
Ar | -0.029 |
knight | -0.029 |
Token | made |
Token ID | 925 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
Amen | +0.021 |
crew | +0.019 |
Ar | +0.019 |
knight | +0.018 |
bandits | +0.018 |
Neg logit contributions | |
mm | -0.016 |
happen | -0.014 |
n | -0.014 |
uch | -0.014 |
ops | -0.014 |
Token | them |
Token ID | 606 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
Amen | +0.051 |
crew | +0.045 |
Ar | +0.044 |
bandits | +0.043 |
knight | +0.043 |
Neg logit contributions | |
mm | -0.052 |
ops | -0.046 |
happen | -0.046 |
uch | -0.045 |
n | -0.045 |
Token | last |
Token ID | 938 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
mm | +0.020 |
ops | +0.018 |
uch | +0.017 |
€ | +0.017 |
tons | +0.017 |
Neg logit contributions | |
Amen | -0.016 |
crew | -0.015 |
knight | -0.014 |
Ar | -0.014 |
bandits | -0.014 |
Token | longer |
Token ID | 2392 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
Amen | +0.195 |
crew | +0.172 |
Ar | +0.168 |
Offic | +0.166 |
bandits | +0.163 |
Neg logit contributions | |
mm | -0.175 |
happen | -0.164 |
ops | -0.157 |
n | -0.155 |
uch | -0.154 |
Token | . |
Token ID | 13 |
Feature activation | +1.85e-03 |
Pos logit contributions | |
Amen | +0.179 |
crew | +0.167 |
knight | +0.163 |
bandits | +0.161 |
hope | +0.161 |
Neg logit contributions | |
mm | -0.157 |
liv | -0.135 |
ops | -0.135 |
uch | -0.134 |
€ | -0.131 |
Token | The |
Token ID | 383 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
Amen | +0.002 |
crew | +0.002 |
Ar | +0.002 |
bandits | +0.002 |
knight | +0.002 |
Neg logit contributions | |
mm | -0.003 |
ops | -0.003 |
n | -0.003 |
€ | -0.003 |
tons | -0.003 |
Token | moral |
Token ID | 6573 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
Amen | +0.031 |
crew | +0.029 |
knight | +0.028 |
bandits | +0.028 |
Ar | +0.027 |
Neg logit contributions | |
mm | -0.030 |
ops | -0.026 |
uch | -0.026 |
happen | -0.026 |
ying | -0.026 |
Token | of |
Token ID | 286 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
Amen | +0.050 |
crew | +0.046 |
knight | +0.043 |
lair | +0.043 |
Ar | +0.042 |
Neg logit contributions | |
mm | -0.044 |
n | -0.039 |
happen | -0.039 |
ops | -0.038 |
uch | -0.038 |
Token | dad |
Token ID | 9955 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
mm | +0.074 |
n | +0.069 |
uch | +0.069 |
ops | +0.068 |
happen | +0.067 |
Neg logit contributions | |
Amen | -0.034 |
crew | -0.031 |
bandits | -0.028 |
knight | -0.028 |
parent | -0.027 |
Token | walked |
Token ID | 6807 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
Amen | +0.091 |
crew | +0.080 |
Ar | +0.077 |
bandits | +0.076 |
knight | +0.075 |
Neg logit contributions | |
mm | -0.097 |
ops | -0.087 |
happen | -0.085 |
uch | -0.085 |
n | -0.084 |
Token | up |
Token ID | 510 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
mm | +0.044 |
ops | +0.039 |
happen | +0.038 |
n | +0.038 |
uch | +0.038 |
Neg logit contributions | |
Amen | -0.052 |
crew | -0.047 |
Ar | -0.045 |
bandits | -0.045 |
knight | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | +7.75e-02 |
Pos logit contributions | |
Amen | +0.019 |
crew | +0.016 |
knight | +0.015 |
Ar | +0.015 |
bandits | +0.015 |
Neg logit contributions | |
mm | -0.018 |
happen | -0.016 |
ops | -0.016 |
n | -0.016 |
tons | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -6.61e-03 |
Pos logit contributions | |
Amen | +0.180 |
lair | +0.168 |
crew | +0.167 |
Ar | +0.164 |
unts | +0.162 |
Neg logit contributions | |
happen | -0.056 |
mm | -0.052 |
n | -0.049 |
tons | -0.049 |
€ | -0.044 |
Token | cop |
Token ID | 2243 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
mm | +0.009 |
happen | +0.009 |
belongs | +0.009 |
ops | +0.009 |
tons | +0.009 |
Neg logit contributions | |
Amen | -0.011 |
crew | -0.010 |
known | -0.009 |
Ar | -0.009 |
knight | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
Amen | +0.190 |
crew | +0.176 |
knight | +0.168 |
lair | +0.165 |
Offic | +0.162 |
Neg logit contributions | |
mm | -0.198 |
uch | -0.174 |
€ | -0.163 |
happen | -0.162 |
liv | -0.162 |
Token | When |
Token ID | 1649 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
mm | +0.020 |
ops | +0.018 |
€ | +0.018 |
happen | +0.017 |
belongs | +0.017 |
Neg logit contributions | |
Amen | -0.018 |
crew | -0.016 |
Ar | -0.015 |
bandits | -0.015 |
knight | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -5.38e-02 |
Pos logit contributions | |
mm | +0.083 |
ops | +0.073 |
uch | +0.072 |
happen | +0.072 |
n | +0.070 |
Neg logit contributions | |
Amen | -0.092 |
crew | -0.083 |
Ar | -0.080 |
bandits | -0.079 |
knight | -0.078 |
Token | cop |
Token ID | 2243 |
Feature activation | +7.23e-02 |
Pos logit contributions | |
mm | +0.123 |
ops | +0.113 |
uch | +0.112 |
happen | +0.112 |
ying | +0.111 |
Neg logit contributions | |
Amen | -0.049 |
crew | -0.039 |
knight | -0.038 |
bandits | -0.037 |
Ar | -0.037 |
Token | saw |
Token ID | 2497 |
Feature activation | +8.58e-02 |
Pos logit contributions | |
Amen | +0.084 |
crew | +0.082 |
knight | +0.079 |
Ar | +0.079 |
Offic | +0.079 |
Neg logit contributions | |
mm | -0.138 |
ying | -0.122 |
happen | -0.121 |
uch | -0.121 |
€ | -0.119 |
Token | my |
Token ID | 616 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
mm | +0.104 |
liv | +0.101 |
ops | +0.101 |
€ | +0.099 |
ying | +0.097 |
Neg logit contributions | |
Amen | -0.115 |
crew | -0.101 |
knight | -0.099 |
bandits | -0.098 |
Ar | -0.098 |
Token | spaghetti |
Token ID | 48241 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
mm | +0.059 |
ying | +0.056 |
liv | +0.054 |
ops | +0.054 |
ancing | +0.054 |
Neg logit contributions | |
Amen | -0.055 |
crew | -0.051 |
knight | -0.049 |
bandits | -0.047 |
lair | -0.046 |
Token | on |
Token ID | 319 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
mm | +0.097 |
liv | +0.084 |
€ | +0.084 |
ying | +0.083 |
uch | +0.083 |
Neg logit contributions | |
Amen | -0.100 |
crew | -0.090 |
knight | -0.089 |
lair | -0.086 |
Ar | -0.085 |
Token | my |
Token ID | 616 |
Feature activation | -7.04e-04 |
Pos logit contributions | |
mm | +0.057 |
ops | +0.050 |
happen | +0.048 |
uch | +0.048 |
ying | +0.047 |
Neg logit contributions | |
Amen | -0.079 |
crew | -0.071 |
Ar | -0.069 |
knight | -0.069 |
bandits | -0.068 |
Token | plate |
Token ID | 7480 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
mm | +0.001 |
ying | +0.001 |
ops | +0.001 |
uch | +0.001 |
n | +0.001 |
Neg logit contributions | |
Amen | -0.001 |
crew | -0.001 |
knight | -0.001 |
lair | -0.001 |
bandits | -0.001 |
Token | ." |
Token ID | 526 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
Amen | +0.037 |
crew | +0.033 |
Ar | +0.033 |
bandits | +0.032 |
knight | +0.032 |
Neg logit contributions | |
mm | -0.027 |
ops | -0.024 |
happen | -0.023 |
uch | -0.023 |
n | -0.023 |
Token | And |
Token ID | 843 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
Amen | +0.073 |
crew | +0.071 |
Ar | +0.069 |
bandits | +0.069 |
lair | +0.068 |
Neg logit contributions | |
mm | -0.083 |
n | -0.072 |
ops | -0.072 |
€ | -0.069 |
tons | -0.068 |
Token | she |
Token ID | 673 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
mm | +0.044 |
ops | +0.039 |
happen | +0.038 |
n | +0.038 |
uch | +0.037 |
Neg logit contributions | |
Amen | -0.057 |
crew | -0.051 |
Ar | -0.050 |
bandits | -0.049 |
knight | -0.049 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
Amen | +0.044 |
crew | +0.039 |
Ar | +0.038 |
bandits | +0.038 |
Offic | +0.037 |
Neg logit contributions | |
mm | -0.047 |
ops | -0.042 |
happen | -0.042 |
n | -0.041 |
uch | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
Amen | +0.083 |
crew | +0.076 |
Ar | +0.075 |
bandits | +0.073 |
knight | +0.072 |
Neg logit contributions | |
mm | -0.065 |
happen | -0.057 |
ops | -0.056 |
n | -0.055 |
tons | -0.054 |
Token | delicious |
Token ID | 12625 |
Feature activation | +5.23e-03 |
Pos logit contributions | |
mm | +0.033 |
ying | +0.032 |
ancing | +0.031 |
ops | +0.030 |
aund | +0.030 |
Neg logit contributions | |
Amen | -0.031 |
crew | -0.027 |
knight | -0.027 |
hope | -0.026 |
bandits | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
Amen | +0.079 |
hope | +0.073 |
known | +0.073 |
wonder | +0.072 |
unts | +0.072 |
Neg logit contributions | |
mm | -0.116 |
ops | -0.101 |
uch | -0.100 |
tons | -0.098 |
€ | -0.094 |
Token | cand |
Token ID | 2658 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
mm | +0.019 |
ying | +0.017 |
ops | +0.017 |
liv | +0.017 |
uch | +0.017 |
Neg logit contributions | |
Amen | -0.020 |
crew | -0.018 |
Ar | -0.018 |
knight | -0.017 |
hope | -0.017 |
Token | ies |
Token ID | 444 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
mm | +0.098 |
€ | +0.094 |
tons | +0.089 |
liv | +0.087 |
uch | +0.085 |
Neg logit contributions | |
crew | -0.078 |
Amen | -0.074 |
bandits | -0.070 |
knight | -0.070 |
Ar | -0.069 |
Token | one |
Token ID | 530 |
Feature activation | +9.16e-04 |
Pos logit contributions | |
hope | +0.038 |
wonder | +0.034 |
known | +0.034 |
reunited | +0.034 |
understand | +0.033 |
Neg logit contributions | |
mm | -0.060 |
€ | -0.054 |
liv | -0.053 |
uch | -0.052 |
ops | -0.050 |
Token | by |
Token ID | 416 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
hope | +0.001 |
crew | +0.001 |
wonder | +0.001 |
known | +0.001 |
knight | +0.001 |
Neg logit contributions | |
mm | -0.002 |
€ | -0.001 |
uch | -0.001 |
ops | -0.001 |
liv | -0.001 |
Token | one |
Token ID | 530 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
mm | +0.024 |
uch | +0.023 |
ops | +0.022 |
ying | +0.022 |
€ | +0.021 |
Neg logit contributions | |
Amen | -0.019 |
crew | -0.018 |
known | -0.018 |
hope | -0.018 |
Ar | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +2.49e-03 |
Pos logit contributions | |
crew | +0.036 |
Amen | +0.036 |
knight | +0.035 |
wonder | +0.035 |
hope | +0.035 |
Neg logit contributions | |
mm | -0.036 |
liv | -0.034 |
€ | -0.033 |
uch | -0.033 |
ops | -0.031 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
Amen | +0.004 |
crew | +0.004 |
Ar | +0.004 |
bandits | +0.004 |
knight | +0.004 |
Neg logit contributions | |
mm | -0.004 |
ops | -0.003 |
happen | -0.003 |
uch | -0.003 |
n | -0.003 |
Token | them |
Token ID | 606 |
Feature activation | -5.08e-03 |
Pos logit contributions | |
Amen | +0.069 |
crew | +0.063 |
known | +0.061 |
hope | +0.060 |
Ar | +0.058 |
Neg logit contributions | |
mm | -0.089 |
uch | -0.086 |
€ | -0.083 |
ops | -0.082 |
ying | -0.082 |
Token | all |
Token ID | 477 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
mm | +0.009 |
€ | +0.009 |
liv | +0.009 |
uch | +0.008 |
aund | +0.008 |
Neg logit contributions | |
Amen | -0.006 |
crew | -0.006 |
reunited | -0.006 |
hope | -0.006 |
knight | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
hope | +0.069 |
wonder | +0.064 |
Amen | +0.064 |
many | +0.061 |
known | +0.059 |
Neg logit contributions | |
mm | -0.128 |
uch | -0.123 |
ops | -0.115 |
aund | -0.113 |
tons | -0.111 |
Token | oll |
Token ID | 692 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
mm | +0.025 |
happen | +0.023 |
€ | +0.022 |
tons | +0.021 |
uch | +0.021 |
Neg logit contributions | |
crew | -0.032 |
Amen | -0.032 |
unts | -0.031 |
knight | -0.031 |
Ar | -0.031 |
Token | ipop |
Token ID | 42800 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
€ | +0.134 |
uct | +0.112 |
happen | +0.110 |
liv | +0.109 |
mm | +0.109 |
Neg logit contributions | |
crew | -0.167 |
subjects | -0.157 |
servants | -0.152 |
knight | -0.152 |
bandits | -0.149 |
Token | into |
Token ID | 656 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
mm | +0.084 |
€ | +0.077 |
uch | +0.074 |
tons | +0.072 |
liv | +0.071 |
Neg logit contributions | |
crew | -0.041 |
lair | -0.038 |
reunited | -0.037 |
Ar | -0.035 |
hope | -0.035 |
Token | different |
Token ID | 1180 |
Feature activation | +6.87e-02 |
Pos logit contributions | |
Amen | +0.065 |
Dear | +0.053 |
Offic | +0.051 |
bandits | +0.050 |
knight | +0.049 |
Neg logit contributions | |
mm | -0.034 |
happen | -0.033 |
tons | -0.033 |
n | -0.032 |
ops | -0.031 |
Token | colors |
Token ID | 7577 |
Feature activation | -8.78e-03 |
Pos logit contributions | |
Amen | +0.092 |
crew | +0.086 |
Ar | +0.083 |
lair | +0.080 |
bandits | +0.080 |
Neg logit contributions | |
mm | -0.123 |
uch | -0.112 |
ops | -0.108 |
€ | -0.105 |
happen | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
mm | +0.015 |
€ | +0.014 |
uch | +0.014 |
liv | +0.013 |
tons | +0.013 |
Neg logit contributions | |
crew | -0.011 |
unts | -0.011 |
Amen | -0.011 |
knight | -0.010 |
lair | -0.010 |
Token | When |
Token ID | 1649 |
Feature activation | -5.75e-02 |
Pos logit contributions | |
mm | +0.035 |
ops | +0.031 |
n | +0.031 |
happen | +0.030 |
tons | +0.030 |
Neg logit contributions | |
Amen | -0.023 |
crew | -0.020 |
Ar | -0.019 |
bandits | -0.019 |
knight | -0.018 |
Token | they |
Token ID | 484 |
Feature activation | -4.12e-03 |
Pos logit contributions | |
mm | +0.091 |
ops | +0.082 |
uch | +0.082 |
ying | +0.081 |
€ | +0.080 |
Neg logit contributions | |
Amen | -0.088 |
crew | -0.081 |
Ar | -0.077 |
bandits | -0.077 |
knight | -0.076 |
Token | were |
Token ID | 547 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
mm | +0.007 |
ops | +0.007 |
uch | +0.007 |
happen | +0.007 |
tons | +0.006 |
Neg logit contributions | |
Amen | -0.005 |
crew | -0.005 |
Ar | -0.005 |
knight | -0.005 |
bandits | -0.005 |
Token | done |
Token ID | 1760 |
Feature activation | -7.11e-02 |
Pos logit contributions | |
mm | +0.056 |
ops | +0.047 |
uch | +0.046 |
happen | +0.045 |
€ | +0.044 |
Neg logit contributions | |
Amen | -0.077 |
crew | -0.074 |
known | -0.072 |
unts | -0.071 |
knight | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
uch | +0.119 |
mm | +0.115 |
€ | +0.108 |
belongs | +0.107 |
liv | +0.103 |
Neg logit contributions | |
Amen | -0.097 |
crew | -0.094 |
known | -0.088 |
reunited | -0.086 |
hope | -0.086 |
Token | and |
Token ID | 290 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Amen | +0.067 |
crew | +0.062 |
knight | +0.060 |
bandits | +0.058 |
Offic | +0.057 |
Neg logit contributions | |
mm | -0.064 |
uch | -0.059 |
ops | -0.056 |
happen | -0.056 |
ying | -0.055 |
Token | she |
Token ID | 673 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
mm | +0.024 |
ops | +0.021 |
uch | +0.021 |
happen | +0.021 |
ying | +0.021 |
Neg logit contributions | |
Amen | -0.022 |
crew | -0.019 |
Ar | -0.018 |
knight | -0.018 |
bandits | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
Amen | +0.056 |
crew | +0.051 |
Ar | +0.049 |
bandits | +0.049 |
Offic | +0.049 |
Neg logit contributions | |
mm | -0.038 |
n | -0.036 |
ops | -0.036 |
happen | -0.035 |
uch | -0.035 |
Token | whist |
Token ID | 32262 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
mm | +0.046 |
happen | +0.042 |
n | +0.042 |
tons | +0.042 |
uch | +0.041 |
Neg logit contributions | |
Amen | -0.050 |
crew | -0.045 |
Ar | -0.045 |
bandits | -0.044 |
Offic | -0.043 |
Token | ling |
Token ID | 1359 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
mm | +0.036 |
happen | +0.030 |
n | +0.029 |
ops | +0.029 |
tons | +0.028 |
Neg logit contributions | |
Amen | -0.029 |
crew | -0.027 |
knight | -0.026 |
bandits | -0.025 |
unts | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
Amen | +0.091 |
crew | +0.082 |
knight | +0.080 |
bandits | +0.078 |
lair | +0.078 |
Neg logit contributions | |
mm | -0.079 |
uch | -0.068 |
ops | -0.067 |
€ | -0.064 |
n | -0.063 |
Token | She |
Token ID | 1375 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
mm | +0.029 |
€ | +0.026 |
ops | +0.026 |
n | +0.026 |
tons | +0.026 |
Neg logit contributions | |
Amen | -0.024 |
crew | -0.022 |
Ar | -0.021 |
bandits | -0.021 |
lair | -0.021 |
Token | had |
Token ID | 550 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
Amen | +0.118 |
crew | +0.103 |
bandits | +0.101 |
Ar | +0.100 |
lair | +0.099 |
Neg logit contributions | |
mm | -0.099 |
ops | -0.091 |
n | -0.091 |
happen | -0.090 |
tons | -0.088 |
Token | a |
Token ID | 257 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
Amen | +0.055 |
crew | +0.045 |
Offic | +0.044 |
Ar | +0.043 |
knight | +0.043 |
Neg logit contributions | |
n | -0.044 |
happen | -0.043 |
mm | -0.043 |
uch | -0.041 |
liv | -0.041 |
Token | big |
Token ID | 1263 |
Feature activation | +6.87e-02 |
Pos logit contributions | |
Amen | +0.091 |
Offic | +0.085 |
aciously | +0.082 |
bandits | +0.082 |
Ar | +0.081 |
Neg logit contributions | |
mm | -0.067 |
uch | -0.066 |
tons | -0.065 |
happen | -0.064 |
gh | -0.063 |
Token | smile |
Token ID | 8212 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
Amen | +0.099 |
Ar | +0.091 |
Offic | +0.089 |
Dear | +0.089 |
crew | +0.088 |
Neg logit contributions | |
mm | -0.108 |
happen | -0.098 |
tons | -0.098 |
uch | -0.097 |
ops | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | -6.73e-03 |
Pos logit contributions | |
Amen | +0.011 |
crew | +0.011 |
Ar | +0.010 |
knight | +0.010 |
bandits | +0.010 |
Neg logit contributions | |
mm | -0.008 |
uch | -0.007 |
ops | -0.007 |
€ | -0.007 |
ying | -0.007 |
Token | |
Token ID | 220 |
Feature activation | +7.29e-03 |
Pos logit contributions | |
mm | +0.011 |
n | +0.010 |
ops | +0.010 |
tons | +0.009 |
€ | +0.009 |
Neg logit contributions | |
Amen | -0.010 |
crew | -0.009 |
Offic | -0.009 |
lair | -0.009 |
bandits | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
Amen | +0.008 |
crew | +0.006 |
Ar | +0.006 |
bandits | +0.006 |
knight | +0.006 |
Neg logit contributions | |
mm | -0.015 |
happen | -0.014 |
ops | -0.014 |
uch | -0.014 |
tons | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
Amen | +0.080 |
crew | +0.074 |
lair | +0.074 |
bandits | +0.072 |
messenger | +0.072 |
Neg logit contributions | |
n | -0.063 |
mm | -0.063 |
ops | -0.058 |
� | -0.054 |
€ | -0.054 |
Token | The |
Token ID | 464 |
Feature activation | +5.32e-03 |
Pos logit contributions | |
Amen | +0.066 |
lair | +0.058 |
crew | +0.058 |
knight | +0.055 |
bandits | +0.055 |
Neg logit contributions | |
mm | -0.068 |
n | -0.065 |
ops | -0.062 |
uch | -0.060 |
tons | -0.059 |
Token | animals |
Token ID | 4695 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
Amen | +0.007 |
crew | +0.007 |
Ar | +0.006 |
knight | +0.006 |
bandits | +0.006 |
Neg logit contributions | |
mm | -0.010 |
ops | -0.009 |
uch | -0.009 |
happen | -0.008 |
ying | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -8.59e-03 |
Pos logit contributions | |
Amen | +0.044 |
Ar | +0.038 |
crew | +0.037 |
lair | +0.036 |
Offic | +0.036 |
Neg logit contributions | |
mm | -0.044 |
ops | -0.042 |
ying | -0.041 |
happen | -0.041 |
n | -0.041 |
Token | Benny |
Token ID | 44275 |
Feature activation | -9.38e-03 |
Pos logit contributions | |
mm | +0.011 |
tons | +0.010 |
n | +0.010 |
happen | +0.010 |
ops | +0.010 |
Neg logit contributions | |
Amen | -0.015 |
lair | -0.013 |
Dear | -0.013 |
Offic | -0.013 |
Ar | -0.013 |
Token | cl |
Token ID | 537 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
mm | +0.015 |
ops | +0.014 |
happen | +0.013 |
n | +0.013 |
uch | +0.013 |
Neg logit contributions | |
Amen | -0.015 |
crew | -0.013 |
Ar | -0.012 |
bandits | -0.012 |
knight | -0.012 |
Token | apped |
Token ID | 6320 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
Amen | +0.021 |
bandits | +0.018 |
Ar | +0.018 |
knight | +0.018 |
Offic | +0.018 |
Neg logit contributions | |
mm | -0.024 |
ops | -0.022 |
uch | -0.021 |
ying | -0.021 |
ancing | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
Amen | +0.224 |
Offic | +0.196 |
Ar | +0.195 |
crew | +0.194 |
bandits | +0.190 |
Neg logit contributions | |
mm | -0.171 |
happen | -0.158 |
ops | -0.150 |
n | -0.149 |
ying | -0.149 |
Token | . |
Token ID | 13 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
Amen | +0.050 |
crew | +0.045 |
knight | +0.044 |
Ar | +0.043 |
bandits | +0.043 |
Neg logit contributions | |
mm | -0.036 |
ops | -0.031 |
happen | -0.030 |
uch | -0.030 |
€ | -0.029 |
Token | I |
Token ID | 314 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
mm | +0.058 |
ops | +0.048 |
n | +0.047 |
tons | +0.047 |
€ | +0.047 |
Neg logit contributions | |
Amen | -0.081 |
crew | -0.076 |
Ar | -0.074 |
bandits | -0.073 |
lair | -0.072 |
Token | was |
Token ID | 373 |
Feature activation | -5.10e-02 |
Pos logit contributions | |
Amen | +0.051 |
bandits | +0.049 |
Offic | +0.048 |
crew | +0.047 |
lair | +0.046 |
Neg logit contributions | |
n | -0.044 |
mm | -0.043 |
uch | -0.043 |
ops | -0.042 |
happen | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
mm | +0.078 |
happen | +0.071 |
uch | +0.070 |
n | +0.070 |
tons | +0.069 |
Neg logit contributions | |
Amen | -0.080 |
crew | -0.072 |
Ar | -0.071 |
Offic | -0.069 |
bandits | -0.068 |
Token | butterfly |
Token ID | 35113 |
Feature activation | +9.20e-03 |
Pos logit contributions | |
Amen | +0.026 |
Offic | +0.024 |
bandits | +0.023 |
Ar | +0.023 |
lair | +0.023 |
Neg logit contributions | |
mm | -0.019 |
tons | -0.018 |
happen | -0.017 |
uch | -0.017 |
€ | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
Amen | +0.017 |
Ar | +0.016 |
crew | +0.016 |
Offic | +0.015 |
bandits | +0.015 |
Neg logit contributions | |
mm | -0.012 |
ops | -0.010 |
happen | -0.010 |
uch | -0.010 |
n | -0.010 |
Token | I |
Token ID | 314 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
mm | +0.055 |
ops | +0.044 |
n | +0.044 |
tons | +0.044 |
€ | +0.044 |
Neg logit contributions | |
Amen | -0.084 |
crew | -0.080 |
Ar | -0.077 |
bandits | -0.077 |
lair | -0.076 |
Token | had |
Token ID | 550 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
Amen | +0.051 |
bandits | +0.047 |
Offic | +0.046 |
crew | +0.046 |
lair | +0.045 |
Neg logit contributions | |
n | -0.048 |
mm | -0.048 |
uch | -0.046 |
ops | -0.046 |
happen | -0.045 |
Token | wings |
Token ID | 12098 |
Feature activation | +8.38e-03 |
Pos logit contributions | |
Amen | +0.027 |
Offic | +0.023 |
crew | +0.023 |
Ar | +0.022 |
knight | +0.021 |
Neg logit contributions | |
mm | -0.024 |
happen | -0.023 |
n | -0.022 |
uch | -0.022 |
ops | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
Amen | +0.015 |
crew | +0.014 |
Offic | +0.014 |
Ar | +0.014 |
bandits | +0.013 |
Neg logit contributions | |
mm | -0.011 |
happen | -0.010 |
ops | -0.010 |
uch | -0.010 |
ying | -0.010 |
Token | legs |
Token ID | 7405 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
mm | +0.037 |
happen | +0.035 |
n | +0.033 |
belongs | +0.032 |
uch | +0.032 |
Neg logit contributions | |
Amen | -0.047 |
Offic | -0.042 |
Ar | -0.041 |
crew | -0.041 |
lair | -0.040 |
Token | little |
Token ID | 1310 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
Amen | +0.082 |
aciously | +0.071 |
ously | +0.070 |
talk | +0.069 |
lair | +0.069 |
Neg logit contributions | |
happen | -0.060 |
tons | -0.058 |
mm | -0.058 |
next | -0.055 |
exercise | -0.055 |
Token | boy |
Token ID | 2933 |
Feature activation | +4.84e-02 |
Pos logit contributions | |
Amen | +0.112 |
Ar | +0.099 |
known | +0.096 |
crew | +0.093 |
Dear | +0.093 |
Neg logit contributions | |
mm | -0.130 |
tons | -0.123 |
belongs | -0.121 |
happen | -0.120 |
ops | -0.117 |
Token | named |
Token ID | 3706 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
Amen | +0.037 |
crew | +0.028 |
Ar | +0.024 |
knight | +0.024 |
bandits | +0.023 |
Neg logit contributions | |
mm | -0.119 |
ops | -0.110 |
happen | -0.109 |
uch | -0.108 |
ying | -0.107 |
Token | Tim |
Token ID | 5045 |
Feature activation | -8.75e-02 |
Pos logit contributions | |
Amen | +0.032 |
crew | +0.029 |
Ar | +0.028 |
knight | +0.028 |
bandits | +0.028 |
Neg logit contributions | |
mm | -0.031 |
ops | -0.028 |
uch | -0.027 |
happen | -0.027 |
ying | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
mm | +0.112 |
ops | +0.094 |
uch | +0.091 |
happen | +0.090 |
ying | +0.087 |
Neg logit contributions | |
Amen | -0.170 |
crew | -0.154 |
knight | -0.148 |
Ar | -0.148 |
bandits | -0.147 |
Token | Tim |
Token ID | 5045 |
Feature activation | -7.56e-02 |
Pos logit contributions | |
mm | +0.038 |
uch | +0.036 |
ops | +0.035 |
happen | +0.035 |
ying | +0.035 |
Neg logit contributions | |
Amen | -0.034 |
Offic | -0.030 |
crew | -0.030 |
Ar | -0.030 |
bandits | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
mm | +0.126 |
ops | +0.113 |
happen | +0.112 |
uch | +0.111 |
n | +0.109 |
Neg logit contributions | |
Amen | -0.113 |
crew | -0.100 |
Ar | -0.096 |
bandits | -0.094 |
knight | -0.093 |
Token | curious |
Token ID | 11040 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
mm | +0.043 |
tons | +0.041 |
n | +0.040 |
uch | +0.039 |
happen | +0.039 |
Neg logit contributions | |
Amen | -0.048 |
Ar | -0.042 |
crew | -0.042 |
bandits | -0.041 |
Offic | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
Amen | +0.064 |
knight | +0.059 |
crew | +0.057 |
bandits | +0.055 |
lair | +0.055 |
Neg logit contributions | |
mm | -0.092 |
uch | -0.086 |
liv | -0.082 |
ops | -0.081 |
€ | -0.080 |
Token | asked |
Token ID | 1965 |
Feature activation | +6.76e-02 |
Pos logit contributions | |
mm | +0.027 |
happen | +0.024 |
ops | +0.024 |
n | +0.024 |
uch | +0.024 |
Neg logit contributions | |
Amen | -0.019 |
crew | -0.017 |
Ar | -0.016 |
bandits | -0.016 |
lair | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
Amen | +0.110 |
crew | +0.101 |
knight | +0.097 |
bandits | +0.097 |
Offic | +0.095 |
Neg logit contributions | |
mm | -0.101 |
uch | -0.093 |
ops | -0.091 |
happen | -0.087 |
liv | -0.086 |
Token |
|
Token ID | 198 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
crew | +0.041 |
bandits | +0.040 |
Amen | +0.039 |
knight | +0.038 |
messenger | +0.038 |
Neg logit contributions | |
n | -0.028 |
mm | -0.028 |
liv | -0.025 |
ying | -0.025 |
ops | -0.024 |
Token | One |
Token ID | 3198 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
Amen | +0.027 |
Ar | +0.022 |
crew | +0.022 |
Offic | +0.021 |
knight | +0.021 |
Neg logit contributions | |
mm | -0.040 |
ops | -0.037 |
happen | -0.037 |
uch | -0.036 |
ying | -0.036 |
Token | day |
Token ID | 1110 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
Amen | +0.020 |
knight | +0.018 |
crew | +0.018 |
Ar | +0.017 |
Dear | +0.017 |
Neg logit contributions | |
mm | -0.016 |
ops | -0.014 |
ancing | -0.014 |
happen | -0.014 |
tons | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
mm | +0.144 |
€ | +0.133 |
ying | +0.133 |
uct | +0.133 |
liv | +0.132 |
Neg logit contributions | |
Amen | -0.132 |
knight | -0.118 |
crew | -0.117 |
heard | -0.117 |
hope | -0.115 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
mm | +0.038 |
ops | +0.032 |
happen | +0.031 |
tons | +0.031 |
n | +0.031 |
Neg logit contributions | |
Amen | -0.063 |
crew | -0.058 |
Ar | -0.056 |
lair | -0.056 |
Offic | -0.055 |
Token | woke |
Token ID | 19092 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
mm | +0.050 |
ops | +0.046 |
uch | +0.044 |
belongs | +0.044 |
n | +0.044 |
Neg logit contributions | |
Amen | -0.092 |
crew | -0.082 |
Ar | -0.080 |
bandits | -0.077 |
Offic | -0.077 |
Token | up |
Token ID | 510 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
mm | +0.051 |
ops | +0.044 |
uch | +0.042 |
aund | +0.040 |
belongs | +0.040 |
Neg logit contributions | |
Amen | -0.093 |
crew | -0.083 |
knight | -0.083 |
bandits | -0.082 |
parent | -0.081 |
Token | early |
Token ID | 1903 |
Feature activation | +2.52e-03 |
Pos logit contributions | |
mm | +0.058 |
ops | +0.051 |
happen | +0.051 |
uch | +0.050 |
n | +0.050 |
Neg logit contributions | |
Amen | -0.061 |
crew | -0.055 |
Ar | -0.053 |
bandits | -0.052 |
knight | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | +3.86e-03 |
Pos logit contributions | |
Amen | +0.004 |
crew | +0.003 |
Ar | +0.003 |
knight | +0.003 |
Dear | +0.003 |
Neg logit contributions | |
mm | -0.004 |
uch | -0.004 |
ops | -0.004 |
ying | -0.004 |
happen | -0.004 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.92e-03 |
Pos logit contributions | |
Amen | +0.005 |
Ar | +0.005 |
crew | +0.005 |
unts | +0.004 |
lair | +0.004 |
Neg logit contributions | |
mm | -0.006 |
happen | -0.006 |
n | -0.006 |
belongs | -0.006 |
€ | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
Amen | +0.003 |
crew | +0.002 |
Ar | +0.002 |
knight | +0.002 |
bandits | +0.002 |
Neg logit contributions | |
mm | -0.003 |
happen | -0.003 |
ops | -0.003 |
n | -0.003 |
uch | -0.003 |
Token | then |
Token ID | 788 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
ops | +0.086 |
mm | +0.086 |
uch | +0.084 |
liv | +0.081 |
ying | +0.079 |
Neg logit contributions | |
Amen | -0.096 |
crew | -0.091 |
hope | -0.086 |
knight | -0.084 |
Ar | -0.080 |
Token | you |
Token ID | 345 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
Amen | +0.032 |
crew | +0.030 |
hope | +0.030 |
knight | +0.029 |
pilot | +0.028 |
Neg logit contributions | |
mm | -0.029 |
liv | -0.029 |
ops | -0.028 |
ying | -0.028 |
uch | -0.027 |
Token | can |
Token ID | 460 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
mm | +0.068 |
n | +0.065 |
uch | +0.064 |
ops | +0.064 |
happen | +0.063 |
Neg logit contributions | |
Amen | -0.102 |
Offic | -0.091 |
lair | -0.090 |
crew | -0.089 |
bandits | -0.088 |
Token | sleep |
Token ID | 3993 |
Feature activation | -6.43e-02 |
Pos logit contributions | |
Amen | +0.049 |
crew | +0.044 |
Ar | +0.043 |
bandits | +0.042 |
knight | +0.041 |
Neg logit contributions | |
mm | -0.051 |
ops | -0.046 |
happen | -0.045 |
uch | -0.045 |
n | -0.044 |
Token | again |
Token ID | 757 |
Feature activation | +1.94e-03 |
Pos logit contributions | |
mm | +0.088 |
ops | +0.077 |
happen | +0.075 |
uch | +0.075 |
n | +0.073 |
Neg logit contributions | |
Amen | -0.115 |
crew | -0.104 |
Ar | -0.101 |
bandits | -0.099 |
knight | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
Amen | +0.003 |
crew | +0.003 |
Ar | +0.003 |
bandits | +0.003 |
knight | +0.003 |
Neg logit contributions | |
mm | -0.003 |
ops | -0.002 |
happen | -0.002 |
uch | -0.002 |
n | -0.002 |
Token | And |
Token ID | 843 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
mm | +0.063 |
ops | +0.054 |
happen | +0.054 |
uch | +0.053 |
n | +0.053 |
Neg logit contributions | |
Amen | -0.070 |
crew | -0.064 |
Ar | -0.062 |
bandits | -0.061 |
knight | -0.060 |
Token | have |
Token ID | 423 |
Feature activation | +6.41e-03 |
Pos logit contributions | |
mm | +0.081 |
ops | +0.080 |
liv | +0.077 |
uch | +0.077 |
ying | +0.074 |
Neg logit contributions | |
Amen | -0.093 |
crew | -0.086 |
hope | -0.083 |
knight | -0.082 |
known | -0.078 |
Token | sweet |
Token ID | 6029 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
Amen | +0.010 |
crew | +0.009 |
Ar | +0.009 |
Offic | +0.009 |
knight | +0.009 |
Neg logit contributions | |
mm | -0.010 |
happen | -0.009 |
ops | -0.009 |
n | -0.009 |
uch | -0.009 |
Token | dreams |
Token ID | 10625 |
Feature activation | -9.79e-02 |
Pos logit contributions | |
mm | +0.019 |
ops | +0.017 |
uch | +0.017 |
happen | +0.017 |
ying | +0.017 |
Neg logit contributions | |
Amen | -0.021 |
crew | -0.018 |
Ar | -0.018 |
knight | -0.017 |
bandits | -0.017 |
Token | ." |
Token ID | 526 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
mm | +0.151 |
ops | +0.135 |
liv | +0.135 |
uch | +0.134 |
tons | +0.134 |
Neg logit contributions | |
Amen | -0.152 |
crew | -0.135 |
knight | -0.131 |
Dear | -0.129 |
Ar | -0.127 |
Token | the |
Token ID | 262 |
Feature activation | +6.16e-03 |
Pos logit contributions | |
Amen | +0.063 |
crew | +0.057 |
Ar | +0.055 |
Dear | +0.055 |
knight | +0.054 |
Neg logit contributions | |
mm | -0.028 |
happen | -0.027 |
n | -0.026 |
tons | -0.025 |
ops | -0.024 |
Token | field |
Token ID | 2214 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
Amen | +0.009 |
crew | +0.008 |
Ar | +0.008 |
bandits | +0.007 |
knight | +0.007 |
Neg logit contributions | |
mm | -0.010 |
uch | -0.009 |
happen | -0.009 |
tons | -0.009 |
belongs | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +4.71e-04 |
Pos logit contributions | |
Amen | +0.086 |
Ar | +0.078 |
bandits | +0.076 |
Offic | +0.075 |
knight | +0.075 |
Neg logit contributions | |
n | -0.025 |
next | -0.024 |
ops | -0.023 |
ying | -0.023 |
happen | -0.022 |
Token | began |
Token ID | 2540 |
Feature activation | +7.14e-02 |
Pos logit contributions | |
Amen | +0.001 |
crew | +0.000 |
Ar | +0.000 |
knight | +0.000 |
lair | +0.000 |
Neg logit contributions | |
mm | -0.001 |
n | -0.001 |
ops | -0.001 |
uch | -0.001 |
happen | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
Amen | +0.089 |
crew | +0.077 |
Ar | +0.074 |
bandits | +0.072 |
knight | +0.071 |
Neg logit contributions | |
mm | -0.137 |
ops | -0.124 |
happen | -0.122 |
uch | -0.122 |
n | -0.120 |
Token | pick |
Token ID | 2298 |
Feature activation | -7.36e-02 |
Pos logit contributions | |
Amen | +0.075 |
crew | +0.066 |
Ar | +0.064 |
bandits | +0.063 |
knight | +0.063 |
Neg logit contributions | |
mm | -0.074 |
ops | -0.066 |
uch | -0.064 |
happen | -0.064 |
ying | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
mm | +0.118 |
uch | +0.106 |
ops | +0.106 |
happen | +0.103 |
€ | +0.101 |
Neg logit contributions | |
Amen | -0.111 |
crew | -0.100 |
Ar | -0.098 |
bandits | -0.096 |
knight | -0.095 |
Token | corn |
Token ID | 11676 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
mm | +0.026 |
happen | +0.024 |
ops | +0.023 |
uch | +0.023 |
ying | +0.023 |
Neg logit contributions | |
Amen | -0.027 |
crew | -0.024 |
Offic | -0.023 |
bandits | -0.023 |
Ar | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
mm | +0.083 |
€ | +0.080 |
uch | +0.078 |
liv | +0.078 |
aund | +0.071 |
Neg logit contributions | |
Amen | -0.058 |
crew | -0.057 |
lair | -0.057 |
hut | -0.057 |
cottage | -0.055 |
Token | She |
Token ID | 1375 |
Feature activation | +6.73e-02 |
Pos logit contributions | |
mm | +0.024 |
ops | +0.021 |
n | +0.021 |
tons | +0.021 |
€ | +0.021 |
Neg logit contributions | |
Amen | -0.022 |
crew | -0.020 |
Ar | -0.019 |
knight | -0.018 |
bandits | -0.018 |
Token | believed |
Token ID | 4762 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
Amen | +0.113 |
crew | +0.094 |
knight | +0.090 |
bandits | +0.090 |
Ar | +0.090 |
Neg logit contributions | |
mm | -0.104 |
ops | -0.099 |
n | -0.098 |
uch | -0.096 |
tons | -0.092 |
Token | out |
Token ID | 503 |
Feature activation | +5.76e-02 |
Pos logit contributions | |
mm | +0.050 |
ops | +0.043 |
liv | +0.043 |
tons | +0.043 |
€ | +0.042 |
Neg logit contributions | |
hope | -0.022 |
Amen | -0.022 |
crew | -0.021 |
bandits | -0.020 |
understand | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
Amen | +0.104 |
crew | +0.094 |
Ar | +0.092 |
bandits | +0.090 |
Offic | +0.089 |
Neg logit contributions | |
mm | -0.078 |
happen | -0.068 |
ops | -0.068 |
n | -0.065 |
uch | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
mm | +0.049 |
ops | +0.045 |
happen | +0.044 |
uch | +0.044 |
n | +0.044 |
Neg logit contributions | |
Amen | -0.027 |
crew | -0.023 |
Ar | -0.022 |
bandits | -0.021 |
knight | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
Amen | +0.049 |
crew | +0.046 |
knight | +0.045 |
bandits | +0.044 |
lair | +0.043 |
Neg logit contributions | |
mm | -0.056 |
n | -0.051 |
ops | -0.048 |
happen | -0.048 |
€ | -0.047 |
Token | " |
Token ID | 1 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
Amen | +0.055 |
crew | +0.049 |
Ar | +0.047 |
bandits | +0.046 |
knight | +0.046 |
Neg logit contributions | |
mm | -0.054 |
ops | -0.048 |
happen | -0.047 |
uch | -0.047 |
n | -0.047 |
Token | Y |
Token ID | 56 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
mm | +0.021 |
ops | +0.019 |
happen | +0.018 |
ying | +0.018 |
uch | +0.018 |
Neg logit contributions | |
Amen | -0.020 |
crew | -0.017 |
Ar | -0.017 |
bandits | -0.017 |
Dear | -0.017 |
Token | uck |
Token ID | 1347 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
tons | +0.192 |
existed | +0.186 |
mm | +0.176 |
ying | +0.175 |
€ | +0.172 |
Neg logit contributions | |
Amen | -0.291 |
unts | -0.278 |
Dear | -0.270 |
Ar | -0.260 |
crew | -0.259 |
Token | ! |
Token ID | 0 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
mm | +0.052 |
€ | +0.046 |
ops | +0.045 |
tons | +0.045 |
existed | +0.045 |
Neg logit contributions | |
Amen | -0.036 |
hope | -0.035 |
crew | -0.034 |
unts | -0.034 |
heard | -0.033 |
Token | These |
Token ID | 2312 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
happen | +0.092 |
mm | +0.091 |
ying | +0.091 |
ops | +0.086 |
tons | +0.084 |
Neg logit contributions | |
Amen | -0.095 |
crew | -0.083 |
Some | -0.082 |
Dear | -0.079 |
unts | -0.078 |
Token | past |
Token ID | 1613 |
Feature activation | -7.53e-02 |
Pos logit contributions | |
mm | +0.169 |
€ | +0.167 |
liv | +0.158 |
uct | +0.157 |
tons | +0.156 |
Neg logit contributions | |
crew | -0.121 |
hope | -0.116 |
bandits | -0.115 |
unts | -0.112 |
traitor | -0.112 |
Token | ries |
Token ID | 1678 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
€ | +0.091 |
uct | +0.081 |
mm | +0.074 |
tons | +0.073 |
liv | +0.073 |
Neg logit contributions | |
crew | -0.123 |
mouse | -0.118 |
bandits | -0.117 |
firefighters | -0.116 |
Some | -0.114 |
Token | called |
Token ID | 1444 |
Feature activation | -6.92e-03 |
Pos logit contributions | |
mm | +0.021 |
uch | +0.020 |
ops | +0.020 |
n | +0.020 |
happen | +0.019 |
Neg logit contributions | |
Amen | -0.011 |
Ar | -0.010 |
crew | -0.010 |
bandits | -0.010 |
Offic | -0.009 |
Token | Sc |
Token ID | 1446 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
mm | +0.012 |
ops | +0.011 |
uch | +0.011 |
ying | +0.011 |
liv | +0.011 |
Neg logit contributions | |
Amen | -0.010 |
crew | -0.009 |
knight | -0.009 |
Ar | -0.008 |
bandits | -0.008 |
Token | amp |
Token ID | 696 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
mm | +0.314 |
happen | +0.282 |
ops | +0.251 |
tons | +0.249 |
uch | +0.244 |
Neg logit contributions | |
Amen | -0.299 |
unts | -0.281 |
Ar | -0.280 |
crew | -0.272 |
Offic | -0.267 |
Token | i |
Token ID | 72 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
mm | +0.198 |
tons | +0.163 |
happen | +0.161 |
uch | +0.160 |
ancing | +0.158 |
Neg logit contributions | |
Amen | -0.230 |
bandits | -0.216 |
Ar | -0.216 |
crew | -0.216 |
Dear | -0.213 |
Token | . |
Token ID | 13 |
Feature activation | -4.65e-02 |
Pos logit contributions | |
mm | +0.056 |
ops | +0.048 |
n | +0.044 |
uch | +0.044 |
ying | +0.044 |
Neg logit contributions | |
Amen | -0.136 |
Ar | -0.124 |
crew | -0.123 |
bandits | -0.122 |
Offic | -0.121 |
Token | Sc |
Token ID | 1446 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
mm | +0.081 |
uch | +0.073 |
ops | +0.073 |
happen | +0.073 |
ying | +0.070 |
Neg logit contributions | |
Amen | -0.065 |
crew | -0.059 |
Ar | -0.058 |
bandits | -0.055 |
Offic | -0.054 |
Token | amp |
Token ID | 696 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
mm | +0.305 |
happen | +0.284 |
uch | +0.240 |
ops | +0.239 |
tons | +0.238 |
Neg logit contributions | |
Ar | -0.290 |
Amen | -0.288 |
unts | -0.285 |
Wal | -0.276 |
crew | -0.270 |
Token | i |
Token ID | 72 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
mm | +0.228 |
ancing | +0.201 |
happen | +0.195 |
tons | +0.195 |
€ | +0.187 |
Neg logit contributions | |
heard | -0.177 |
Ar | -0.176 |
bandits | -0.172 |
saw | -0.167 |
Amen | -0.166 |
Token | was |
Token ID | 373 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
mm | +0.096 |
ops | +0.086 |
happen | +0.086 |
uch | +0.084 |
n | +0.083 |
Neg logit contributions | |
Amen | -0.078 |
crew | -0.069 |
Ar | -0.066 |
knight | -0.064 |
bandits | -0.064 |
Token | unique |
Token ID | 3748 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
mm | +0.071 |
uch | +0.064 |
happen | +0.064 |
n | +0.063 |
tons | +0.063 |
Neg logit contributions | |
Amen | -0.069 |
crew | -0.061 |
Ar | -0.059 |
Offic | -0.059 |
bandits | -0.058 |
Token | because |
Token ID | 780 |
Feature activation | -9.25e-02 |
Pos logit contributions | |
mm | +0.104 |
uch | +0.094 |
ops | +0.093 |
€ | +0.093 |
liv | +0.092 |
Neg logit contributions | |
Amen | -0.093 |
crew | -0.091 |
knight | -0.088 |
bandits | -0.084 |
lair | -0.083 |
Token | to |
Token ID | 284 |
Feature activation | +7.36e-02 |
Pos logit contributions | |
Amen | +0.027 |
crew | +0.023 |
Ar | +0.022 |
bandits | +0.021 |
knight | +0.021 |
Neg logit contributions | |
mm | -0.045 |
ops | -0.041 |
uch | -0.040 |
happen | -0.040 |
€ | -0.040 |
Token | your |
Token ID | 534 |
Feature activation | -6.47e-03 |
Pos logit contributions | |
Amen | +0.137 |
Ar | +0.122 |
crew | +0.122 |
Offic | +0.120 |
knight | +0.118 |
Neg logit contributions | |
mm | -0.095 |
happen | -0.087 |
n | -0.081 |
ops | -0.079 |
ying | -0.078 |
Token | siblings |
Token ID | 20569 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
mm | +0.011 |
ops | +0.010 |
happen | +0.010 |
uch | +0.010 |
n | +0.010 |
Neg logit contributions | |
Amen | -0.009 |
crew | -0.008 |
Ar | -0.008 |
Offic | -0.008 |
bandits | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
Amen | +0.169 |
crew | +0.157 |
hope | +0.148 |
knight | +0.147 |
lair | +0.143 |
Neg logit contributions | |
mm | -0.177 |
liv | -0.151 |
uch | -0.151 |
€ | -0.149 |
ops | -0.145 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
mm | +0.039 |
n | +0.033 |
tons | +0.033 |
happen | +0.033 |
ops | +0.033 |
Neg logit contributions | |
Amen | -0.029 |
crew | -0.027 |
bandits | -0.027 |
knight | -0.026 |
Ar | -0.026 |
Token | S |
Token ID | 50 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
mm | +0.085 |
happen | +0.078 |
liv | +0.076 |
ying | +0.075 |
ancing | +0.075 |
Neg logit contributions | |
Amen | -0.050 |
Ar | -0.045 |
Dear | -0.042 |
knight | -0.040 |
unts | -0.039 |
Token | ara |
Token ID | 3301 |
Feature activation | -8.58e-02 |
Pos logit contributions | |
mm | +0.328 |
happen | +0.305 |
tons | +0.298 |
ops | +0.298 |
ying | +0.296 |
Neg logit contributions | |
Amen | -0.229 |
crew | -0.201 |
Ar | -0.194 |
Dear | -0.190 |
Offic | -0.190 |
Token | and |
Token ID | 290 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
mm | +0.149 |
happen | +0.133 |
ops | +0.132 |
uch | +0.130 |
ying | +0.129 |
Neg logit contributions | |
Amen | -0.121 |
crew | -0.106 |
Ar | -0.103 |
knight | -0.101 |
bandits | -0.100 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
mm | +0.048 |
happen | +0.042 |
n | +0.041 |
€ | +0.040 |
tons | +0.040 |
Neg logit contributions | |
Amen | -0.064 |
Ar | -0.059 |
crew | -0.058 |
lair | -0.058 |
Offic | -0.056 |
Token | are |
Token ID | 389 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
mm | +0.186 |
ops | +0.166 |
happen | +0.164 |
uch | +0.164 |
ying | +0.162 |
Neg logit contributions | |
Amen | -0.154 |
crew | -0.136 |
Ar | -0.132 |
knight | -0.130 |
bandits | -0.129 |
Token | twins |
Token ID | 20345 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
mm | +0.038 |
ops | +0.031 |
ying | +0.031 |
uch | +0.031 |
n | +0.030 |
Neg logit contributions | |
Amen | -0.063 |
Ar | -0.059 |
Offic | -0.058 |
crew | -0.057 |
bandits | -0.056 |
Token | " |
Token ID | 366 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
mm | +0.033 |
uch | +0.031 |
ops | +0.030 |
happen | +0.030 |
€ | +0.029 |
Neg logit contributions | |
Amen | -0.022 |
crew | -0.019 |
knight | -0.018 |
bandits | -0.018 |
Offic | -0.018 |
Token | Tim |
Token ID | 14967 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
Amen | +0.068 |
crew | +0.063 |
lair | +0.059 |
knight | +0.059 |
subjects | +0.057 |
Neg logit contributions | |
mm | -0.058 |
n | -0.050 |
ops | -0.050 |
uch | -0.049 |
liv | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
mm | +0.138 |
ying | +0.118 |
happen | +0.117 |
aund | +0.115 |
tons | +0.115 |
Neg logit contributions | |
Amen | -0.176 |
crew | -0.164 |
knight | -0.159 |
hope | -0.155 |
known | -0.154 |
Token | this |
Token ID | 428 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
mm | +0.087 |
ops | +0.087 |
liv | +0.083 |
€ | +0.083 |
uch | +0.081 |
Neg logit contributions | |
Amen | -0.082 |
crew | -0.080 |
knight | -0.077 |
hope | -0.076 |
bandits | -0.073 |
Token | is |
Token ID | 318 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
mm | +0.184 |
ying | +0.179 |
liv | +0.174 |
uct | +0.171 |
€ | +0.169 |
Neg logit contributions | |
Amen | -0.191 |
Ar | -0.182 |
crew | -0.180 |
knight | -0.173 |
known | -0.165 |
Token | cel |
Token ID | 18725 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
ops | +0.091 |
mm | +0.089 |
ying | +0.084 |
liv | +0.084 |
ancing | +0.081 |
Neg logit contributions | |
Amen | -0.045 |
known | -0.045 |
crew | -0.044 |
knight | -0.038 |
Dear | -0.038 |
Token | ery |
Token ID | 1924 |
Feature activation | -7.06e-02 |
Pos logit contributions | |
mm | +0.270 |
€ | +0.238 |
tons | +0.238 |
happen | +0.236 |
liv | +0.234 |
Neg logit contributions | |
Amen | -0.268 |
crew | -0.246 |
Ar | -0.242 |
knight | -0.236 |
bandits | -0.232 |
Token | . |
Token ID | 13 |
Feature activation | -6.19e-02 |
Pos logit contributions | |
mm | +0.108 |
liv | +0.100 |
€ | +0.100 |
uct | +0.095 |
tons | +0.092 |
Neg logit contributions | |
crew | -0.103 |
Amen | -0.102 |
knight | -0.100 |
lair | -0.096 |
mouse | -0.091 |
Token | It |
Token ID | 632 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
mm | +0.096 |
ops | +0.084 |
happen | +0.082 |
uch | +0.082 |
n | +0.081 |
Neg logit contributions | |
Amen | -0.100 |
crew | -0.090 |
Ar | -0.087 |
bandits | -0.086 |
knight | -0.085 |
Token | is |
Token ID | 318 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
mm | +0.012 |
liv | +0.011 |
ops | +0.011 |
€ | +0.011 |
ying | +0.010 |
Neg logit contributions | |
Amen | -0.020 |
crew | -0.018 |
knight | -0.017 |
Ar | -0.017 |
hope | -0.017 |
Token | tasty |
Token ID | 25103 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
mm | +0.089 |
ops | +0.083 |
uch | +0.079 |
ying | +0.079 |
liv | +0.078 |
Neg logit contributions | |
Amen | -0.073 |
crew | -0.067 |
Ar | -0.064 |
known | -0.063 |
knight | -0.061 |
Token | bigger |
Token ID | 5749 |
Feature activation | -1.90e-03 |
Pos logit contributions | |
mm | +0.140 |
happen | +0.125 |
uch | +0.124 |
n | +0.121 |
ying | +0.121 |
Neg logit contributions | |
Amen | -0.162 |
crew | -0.147 |
Offic | -0.144 |
Ar | -0.144 |
knight | -0.141 |
Token | now |
Token ID | 783 |
Feature activation | -3.53e-03 |
Pos logit contributions | |
mm | +0.003 |
uch | +0.003 |
ops | +0.003 |
happen | +0.003 |
€ | +0.003 |
Neg logit contributions | |
Amen | -0.003 |
crew | -0.002 |
knight | -0.002 |
bandits | -0.002 |
lair | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
mm | +0.006 |
liv | +0.005 |
ops | +0.005 |
uch | +0.005 |
€ | +0.005 |
Neg logit contributions | |
Amen | -0.005 |
crew | -0.004 |
knight | -0.004 |
Dear | -0.004 |
hope | -0.004 |
Token | It |
Token ID | 632 |
Feature activation | -5.81e-03 |
Pos logit contributions | |
mm | +0.079 |
ops | +0.074 |
happen | +0.073 |
uch | +0.072 |
ying | +0.070 |
Neg logit contributions | |
Amen | -0.099 |
crew | -0.086 |
knight | -0.083 |
Ar | -0.083 |
Offic | -0.082 |
Token | 's |
Token ID | 338 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
mm | +0.007 |
ops | +0.006 |
ula | +0.006 |
tons | +0.006 |
ying | +0.006 |
Neg logit contributions | |
Amen | -0.011 |
crew | -0.010 |
knight | -0.010 |
known | -0.010 |
Ar | -0.010 |
Token | time |
Token ID | 640 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
mm | +0.099 |
ops | +0.098 |
ying | +0.086 |
ancing | +0.085 |
uch | +0.084 |
Neg logit contributions | |
Amen | -0.074 |
known | -0.066 |
crew | -0.062 |
knight | -0.061 |
Ar | -0.060 |
Token | for |
Token ID | 329 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
mm | +0.291 |
ops | +0.260 |
happen | +0.255 |
uch | +0.255 |
tons | +0.251 |
Neg logit contributions | |
Amen | -0.236 |
crew | -0.208 |
Ar | -0.199 |
bandits | -0.195 |
knight | -0.194 |
Token | you |
Token ID | 345 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
Amen | +0.019 |
crew | +0.017 |
Ar | +0.017 |
knight | +0.017 |
bandits | +0.017 |
Neg logit contributions | |
mm | -0.016 |
ops | -0.014 |
happen | -0.013 |
uch | -0.013 |
n | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
mm | +0.069 |
ops | +0.063 |
happen | +0.063 |
uch | +0.063 |
n | +0.062 |
Neg logit contributions | |
Amen | -0.056 |
crew | -0.049 |
Ar | -0.048 |
Offic | -0.047 |
bandits | -0.046 |
Token | be |
Token ID | 307 |
Feature activation | -6.91e-02 |
Pos logit contributions | |
Amen | +0.065 |
crew | +0.057 |
Ar | +0.055 |
knight | +0.055 |
hope | +0.055 |
Neg logit contributions | |
mm | -0.061 |
ops | -0.055 |
ying | -0.054 |
uch | -0.052 |
liv | -0.052 |
Token | more |
Token ID | 517 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
mm | +0.110 |
ops | +0.098 |
happen | +0.098 |
uch | +0.097 |
n | +0.096 |
Neg logit contributions | |
Amen | -0.107 |
crew | -0.095 |
Ar | -0.092 |
bandits | -0.090 |
Offic | -0.090 |