Token | felt |
Token ID | 2936 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
ju | +0.020 |
golf | +0.019 |
makeup | +0.018 |
butter | +0.018 |
contest | +0.017 |
Neg logit contributions | |
rum | -0.028 |
ously | -0.027 |
despair | -0.026 |
deeply | -0.026 |
horror | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | +3.50e-04 |
Pos logit contributions | |
ously | +0.063 |
rum | +0.062 |
despair | +0.055 |
aced | +0.054 |
horror | +0.053 |
Neg logit contributions | |
ju | -0.043 |
golf | -0.041 |
makeup | -0.039 |
butter | -0.037 |
gum | -0.036 |
Token | chill |
Token ID | 20493 |
Feature activation | +6.01e-02 |
Pos logit contributions | |
rum | +0.001 |
ously | +0.001 |
uzz | +0.001 |
aced | +0.001 |
ment | +0.001 |
Neg logit contributions | |
ju | -0.000 |
butter | -0.000 |
golf | -0.000 |
gum | -0.000 |
lemon | -0.000 |
Token | down |
Token ID | 866 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
ously | +0.107 |
rum | +0.107 |
despair | +0.100 |
deeply | +0.099 |
horror | +0.096 |
Neg logit contributions | |
ju | -0.090 |
golf | -0.086 |
makeup | -0.084 |
butter | -0.082 |
mons | -0.080 |
Token | his |
Token ID | 465 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
rum | +0.193 |
ously | +0.178 |
despair | +0.177 |
horror | +0.170 |
heartbeat | +0.168 |
Neg logit contributions | |
ju | -0.104 |
golf | -0.102 |
contest | -0.093 |
butter | -0.091 |
makeup | -0.090 |
Token | spine |
Token ID | 19656 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
rum | +0.097 |
ously | +0.095 |
despair | +0.081 |
deeply | +0.081 |
horror | +0.076 |
Neg logit contributions | |
golf | -0.145 |
makeup | -0.144 |
ju | -0.142 |
swing | -0.137 |
file | -0.135 |
Token | . |
Token ID | 13 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
rum | +0.401 |
horror | +0.387 |
despair | +0.386 |
Flash | +0.371 |
Deal | +0.370 |
Neg logit contributions | |
golf | -0.186 |
butter | -0.174 |
makeup | -0.172 |
swing | -0.168 |
ju | -0.164 |
Token | He |
Token ID | 679 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
rum | +0.063 |
ously | +0.061 |
despair | +0.057 |
urbed | +0.056 |
deeply | +0.055 |
Neg logit contributions | |
golf | -0.065 |
ju | -0.065 |
mons | -0.061 |
gum | -0.059 |
makeup | -0.058 |
Token | suddenly |
Token ID | 6451 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
rum | +0.095 |
ously | +0.086 |
uzzle | +0.080 |
aced | +0.077 |
horror | +0.077 |
Neg logit contributions | |
pin | -0.042 |
ju | -0.040 |
butter | -0.039 |
swing | -0.034 |
makeup | -0.034 |
Token | remembered |
Token ID | 12086 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
rum | +0.068 |
ously | +0.061 |
horror | +0.061 |
despair | +0.060 |
deeply | +0.059 |
Neg logit contributions | |
ju | -0.036 |
butter | -0.035 |
golf | -0.035 |
makeup | -0.033 |
gum | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
rum | +0.090 |
ously | +0.088 |
despair | +0.083 |
deeply | +0.083 |
horror | +0.080 |
Neg logit contributions | |
golf | -0.063 |
ju | -0.063 |
makeup | -0.060 |
butter | -0.056 |
grapes | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
rum | +0.101 |
despair | +0.096 |
ously | +0.096 |
horror | +0.093 |
heartbeat | +0.091 |
Neg logit contributions | |
golf | -0.071 |
ju | -0.069 |
contest | -0.065 |
competition | -0.064 |
makeup | -0.064 |
Token | ran |
Token ID | 4966 |
Feature activation | -5.89e-03 |
Pos logit contributions | |
rum | +0.042 |
ously | +0.040 |
deeply | +0.038 |
despair | +0.037 |
horror | +0.037 |
Neg logit contributions | |
ju | -0.028 |
golf | -0.026 |
butter | -0.024 |
makeup | -0.023 |
swing | -0.023 |
Token | off |
Token ID | 572 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
golf | +0.009 |
ju | +0.008 |
makeup | +0.008 |
contest | +0.008 |
competition | +0.007 |
Neg logit contributions | |
rum | -0.011 |
ously | -0.011 |
despair | -0.011 |
horror | -0.010 |
heartbeat | -0.010 |
Token | in |
Token ID | 287 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
rum | +0.440 |
ously | +0.431 |
despair | +0.426 |
aced | +0.421 |
heartbeat | +0.410 |
Neg logit contributions | |
golf | -0.156 |
contest | -0.127 |
swing | -0.124 |
competition | -0.117 |
ju | -0.111 |
Token | the |
Token ID | 262 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
ously | +0.219 |
aned | +0.204 |
rum | +0.204 |
atively | +0.198 |
edly | +0.197 |
Neg logit contributions | |
golf | -0.157 |
makeup | -0.144 |
ju | -0.138 |
gum | -0.137 |
competition | -0.133 |
Token | opposite |
Token ID | 6697 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
ously | +0.049 |
rum | +0.047 |
despair | +0.045 |
deeply | +0.045 |
horror | +0.043 |
Neg logit contributions | |
ju | -0.036 |
golf | -0.034 |
makeup | -0.033 |
mons | -0.033 |
contest | -0.032 |
Token | direction |
Token ID | 4571 |
Feature activation | +9.86e-02 |
Pos logit contributions | |
rum | +0.281 |
ously | +0.277 |
uzz | +0.253 |
heartbeat | +0.250 |
silently | +0.244 |
Neg logit contributions | |
golf | -0.305 |
ju | -0.303 |
swing | -0.292 |
grapes | -0.283 |
competition | -0.282 |
Token | . |
Token ID | 13 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
rum | +0.178 |
ously | +0.172 |
despair | +0.168 |
deeply | +0.160 |
horror | +0.159 |
Neg logit contributions | |
golf | -0.142 |
ju | -0.139 |
makeup | -0.131 |
contest | -0.123 |
gum | -0.122 |
Token | He |
Token ID | 679 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
ously | +0.095 |
rum | +0.093 |
despair | +0.084 |
aced | +0.084 |
deeply | +0.081 |
Neg logit contributions | |
golf | -0.084 |
ju | -0.083 |
mons | -0.072 |
gum | -0.071 |
basketball | -0.070 |
Token | was |
Token ID | 373 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
rum | +0.112 |
ously | +0.105 |
despair | +0.097 |
deeply | +0.096 |
heartbeat | +0.096 |
Neg logit contributions | |
ju | -0.071 |
golf | -0.062 |
swing | -0.060 |
butter | -0.060 |
makeup | -0.058 |
Token | free |
Token ID | 1479 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
ously | +0.063 |
rum | +0.062 |
despair | +0.055 |
urbed | +0.054 |
atively | +0.054 |
Neg logit contributions | |
ju | -0.028 |
golf | -0.025 |
makeup | -0.023 |
butter | -0.023 |
competition | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | -3.29e-03 |
Pos logit contributions | |
ously | +0.064 |
rum | +0.063 |
despair | +0.059 |
deeply | +0.058 |
horror | +0.057 |
Neg logit contributions | |
ju | -0.045 |
golf | -0.043 |
makeup | -0.042 |
mons | -0.040 |
contest | -0.040 |
Token | chill |
Token ID | 20493 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
ju | +0.005 |
butter | +0.004 |
makeup | +0.004 |
golf | +0.004 |
gum | +0.004 |
Neg logit contributions | |
rum | -0.007 |
ously | -0.007 |
uzz | -0.006 |
aced | -0.006 |
despair | -0.006 |
Token | run |
Token ID | 1057 |
Feature activation | -4.26e-03 |
Pos logit contributions | |
ously | +0.096 |
rum | +0.093 |
despair | +0.088 |
deeply | +0.087 |
horror | +0.084 |
Neg logit contributions | |
ju | -0.086 |
golf | -0.082 |
mons | -0.080 |
makeup | -0.080 |
contest | -0.078 |
Token | down |
Token ID | 866 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
ju | +0.008 |
golf | +0.007 |
mons | +0.007 |
makeup | +0.007 |
atmeal | +0.007 |
Neg logit contributions | |
ously | -0.007 |
rum | -0.006 |
despair | -0.006 |
deeply | -0.006 |
horror | -0.006 |
Token | his |
Token ID | 465 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
rum | +0.157 |
ously | +0.149 |
despair | +0.140 |
horror | +0.134 |
deeply | +0.134 |
Neg logit contributions | |
golf | -0.119 |
ju | -0.116 |
makeup | -0.110 |
contest | -0.110 |
competition | -0.105 |
Token | spine |
Token ID | 19656 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
rum | +0.086 |
ously | +0.085 |
despair | +0.072 |
deeply | +0.070 |
horror | +0.068 |
Neg logit contributions | |
makeup | -0.134 |
ju | -0.133 |
golf | -0.133 |
file | -0.127 |
quiz | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
rum | +0.394 |
despair | +0.374 |
horror | +0.373 |
Flash | +0.364 |
ously | +0.360 |
Neg logit contributions | |
golf | -0.187 |
makeup | -0.178 |
butter | -0.167 |
swing | -0.167 |
paints | -0.160 |
Token | He |
Token ID | 679 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
ously | +0.065 |
rum | +0.064 |
despair | +0.058 |
deeply | +0.055 |
urbed | +0.054 |
Neg logit contributions | |
golf | -0.085 |
ju | -0.082 |
makeup | -0.080 |
mons | -0.079 |
gum | -0.078 |
Token | could |
Token ID | 714 |
Feature activation | -2.42e-03 |
Pos logit contributions | |
rum | +0.083 |
ously | +0.079 |
despair | +0.071 |
horror | +0.070 |
deeply | +0.070 |
Neg logit contributions | |
ju | -0.047 |
makeup | -0.044 |
paints | -0.043 |
butter | -0.043 |
golf | -0.042 |
Token | not |
Token ID | 407 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
ju | +0.003 |
golf | +0.003 |
makeup | +0.003 |
swing | +0.003 |
quiz | +0.003 |
Neg logit contributions | |
ously | -0.005 |
rum | -0.005 |
despair | -0.004 |
deeply | -0.004 |
aced | -0.004 |
Token | believe |
Token ID | 1975 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
ously | +0.050 |
rum | +0.049 |
deeply | +0.045 |
despair | +0.045 |
horror | +0.043 |
Neg logit contributions | |
ju | -0.036 |
golf | -0.035 |
makeup | -0.035 |
contest | -0.033 |
quiz | -0.033 |
Token | sh |
Token ID | 427 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
ju | +0.018 |
golf | +0.017 |
makeup | +0.017 |
butter | +0.016 |
quiz | +0.016 |
Neg logit contributions | |
rum | -0.024 |
ously | -0.024 |
despair | -0.021 |
deeply | -0.021 |
aced | -0.021 |
Token | iver |
Token ID | 1428 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
ju | +0.023 |
golf | +0.022 |
makeup | +0.021 |
mons | +0.021 |
grapes | +0.021 |
Neg logit contributions | |
ously | -0.019 |
rum | -0.019 |
deeply | -0.017 |
despair | -0.017 |
horror | -0.016 |
Token | go |
Token ID | 467 |
Feature activation | -9.87e-03 |
Pos logit contributions | |
ously | +0.080 |
rum | +0.078 |
despair | +0.074 |
deeply | +0.073 |
horror | +0.070 |
Neg logit contributions | |
ju | -0.081 |
golf | -0.078 |
mons | -0.076 |
makeup | -0.076 |
contest | -0.073 |
Token | down |
Token ID | 866 |
Feature activation | +7.69e-02 |
Pos logit contributions | |
ju | +0.015 |
mons | +0.015 |
grapes | +0.014 |
atmeal | +0.014 |
golf | +0.013 |
Neg logit contributions | |
ously | -0.017 |
deeply | -0.017 |
rum | -0.016 |
despair | -0.016 |
silently | -0.016 |
Token | her |
Token ID | 607 |
Feature activation | +6.69e-02 |
Pos logit contributions | |
ously | +0.140 |
rum | +0.138 |
despair | +0.129 |
deeply | +0.124 |
horror | +0.123 |
Neg logit contributions | |
ju | -0.117 |
golf | -0.116 |
makeup | -0.112 |
contest | -0.110 |
grapes | -0.107 |
Token | spine |
Token ID | 19656 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
rum | +0.106 |
ously | +0.106 |
despair | +0.096 |
deeply | +0.094 |
horror | +0.092 |
Neg logit contributions | |
ju | -0.119 |
golf | -0.118 |
makeup | -0.116 |
swing | -0.112 |
mons | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
rum | +0.419 |
despair | +0.412 |
horror | +0.408 |
ously | +0.397 |
Flash | +0.396 |
Neg logit contributions | |
makeup | -0.155 |
golf | -0.150 |
butter | -0.146 |
swing | -0.142 |
paints | -0.136 |
Token | It |
Token ID | 632 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
ously | +0.055 |
rum | +0.054 |
despair | +0.050 |
deeply | +0.049 |
horror | +0.047 |
Neg logit contributions | |
ju | -0.075 |
golf | -0.072 |
makeup | -0.071 |
mons | -0.070 |
grapes | -0.068 |
Token | was |
Token ID | 373 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
ju | +0.017 |
makeup | +0.015 |
golf | +0.015 |
butter | +0.014 |
contest | +0.014 |
Neg logit contributions | |
rum | -0.022 |
ously | -0.021 |
despair | -0.019 |
deeply | -0.019 |
horror | -0.019 |
Token | very |
Token ID | 845 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
ously | +0.044 |
rum | +0.044 |
despair | +0.039 |
deeply | +0.038 |
horror | +0.038 |
Neg logit contributions | |
ju | -0.027 |
makeup | -0.026 |
golf | -0.026 |
butter | -0.025 |
grapes | -0.024 |
Token | cold |
Token ID | 4692 |
Feature activation | -6.62e-03 |
Pos logit contributions | |
ously | +0.081 |
rum | +0.077 |
urbed | +0.071 |
despair | +0.069 |
horror | +0.069 |
Neg logit contributions | |
ju | -0.025 |
makeup | -0.023 |
golf | -0.022 |
mean | -0.019 |
butter | -0.018 |
Token | sh |
Token ID | 427 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
ju | +0.016 |
butter | +0.014 |
makeup | +0.014 |
golf | +0.014 |
gum | +0.013 |
Neg logit contributions | |
rum | -0.024 |
ously | -0.023 |
uzz | -0.020 |
despair | -0.020 |
deeply | -0.020 |
Token | iver |
Token ID | 1428 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
ju | +0.019 |
golf | +0.019 |
makeup | +0.018 |
mons | +0.018 |
grapes | +0.018 |
Neg logit contributions | |
ously | -0.017 |
rum | -0.017 |
despair | -0.015 |
deeply | -0.015 |
horror | -0.014 |
Token | go |
Token ID | 467 |
Feature activation | -6.65e-03 |
Pos logit contributions | |
ously | +0.080 |
rum | +0.076 |
despair | +0.073 |
deeply | +0.073 |
horror | +0.069 |
Neg logit contributions | |
ju | -0.083 |
golf | -0.079 |
mons | -0.079 |
makeup | -0.076 |
contest | -0.075 |
Token | down |
Token ID | 866 |
Feature activation | +7.79e-02 |
Pos logit contributions | |
ju | +0.011 |
mons | +0.011 |
atmeal | +0.011 |
grapes | +0.011 |
golf | +0.011 |
Neg logit contributions | |
ously | -0.010 |
rum | -0.010 |
despair | -0.010 |
deeply | -0.010 |
silently | -0.009 |
Token | her |
Token ID | 607 |
Feature activation | +7.16e-02 |
Pos logit contributions | |
rum | +0.146 |
ously | +0.144 |
despair | +0.134 |
deeply | +0.127 |
horror | +0.127 |
Neg logit contributions | |
golf | -0.114 |
ju | -0.113 |
makeup | -0.111 |
contest | -0.109 |
grapes | -0.106 |
Token | spine |
Token ID | 19656 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
ously | +0.121 |
rum | +0.118 |
despair | +0.111 |
deeply | +0.109 |
horror | +0.105 |
Neg logit contributions | |
ju | -0.122 |
golf | -0.119 |
makeup | -0.117 |
mons | -0.114 |
grapes | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
rum | +0.414 |
despair | +0.407 |
horror | +0.401 |
Flash | +0.396 |
Ar | +0.389 |
Neg logit contributions | |
makeup | -0.155 |
golf | -0.155 |
butter | -0.150 |
swing | -0.145 |
mons | -0.136 |
Token | It |
Token ID | 632 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
ously | +0.060 |
rum | +0.058 |
despair | +0.054 |
deeply | +0.053 |
horror | +0.051 |
Neg logit contributions | |
ju | -0.076 |
golf | -0.073 |
makeup | -0.072 |
mons | -0.071 |
contest | -0.069 |
Token | tasted |
Token ID | 29187 |
Feature activation | +5.98e-03 |
Pos logit contributions | |
ju | +0.016 |
makeup | +0.015 |
golf | +0.014 |
butter | +0.014 |
contest | +0.014 |
Neg logit contributions | |
ously | -0.022 |
rum | -0.022 |
despair | -0.020 |
deeply | -0.019 |
horror | -0.019 |
Token | different |
Token ID | 1180 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
ously | +0.013 |
rum | +0.012 |
despair | +0.012 |
deeply | +0.012 |
horror | +0.011 |
Neg logit contributions | |
ju | -0.007 |
golf | -0.007 |
makeup | -0.007 |
mons | -0.007 |
grapes | -0.006 |
Token | than |
Token ID | 621 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
ously | +0.037 |
rum | +0.036 |
despair | +0.034 |
deeply | +0.033 |
horror | +0.032 |
Neg logit contributions | |
ju | -0.042 |
makeup | -0.040 |
golf | -0.040 |
mons | -0.039 |
grapes | -0.038 |
Token | sh |
Token ID | 427 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
ju | +0.012 |
golf | +0.011 |
butter | +0.010 |
makeup | +0.010 |
gum | +0.010 |
Neg logit contributions | |
rum | -0.016 |
ously | -0.016 |
despair | -0.014 |
deeply | -0.014 |
aced | -0.014 |
Token | iver |
Token ID | 1428 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
ju | +0.021 |
golf | +0.021 |
makeup | +0.020 |
mons | +0.020 |
grapes | +0.020 |
Neg logit contributions | |
ously | -0.019 |
rum | -0.019 |
despair | -0.017 |
deeply | -0.017 |
horror | -0.016 |
Token | go |
Token ID | 467 |
Feature activation | -7.83e-03 |
Pos logit contributions | |
ously | +0.080 |
rum | +0.078 |
despair | +0.073 |
deeply | +0.072 |
horror | +0.070 |
Neg logit contributions | |
ju | -0.085 |
golf | -0.082 |
makeup | -0.080 |
mons | -0.079 |
contest | -0.077 |
Token | down |
Token ID | 866 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
ju | +0.013 |
golf | +0.012 |
mons | +0.012 |
makeup | +0.012 |
grapes | +0.011 |
Neg logit contributions | |
ously | -0.013 |
rum | -0.013 |
despair | -0.013 |
deeply | -0.012 |
horror | -0.012 |
Token | his |
Token ID | 465 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
rum | +0.161 |
ously | +0.160 |
despair | +0.148 |
urbed | +0.143 |
horror | +0.143 |
Neg logit contributions | |
golf | -0.103 |
ju | -0.096 |
contest | -0.095 |
makeup | -0.092 |
quiz | -0.088 |
Token | spine |
Token ID | 19656 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
rum | +0.108 |
ously | +0.105 |
despair | +0.095 |
horror | +0.093 |
urbed | +0.091 |
Neg logit contributions | |
swing | -0.090 |
golf | -0.087 |
quiz | -0.083 |
makeup | -0.082 |
coat | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
rum | +0.412 |
horror | +0.406 |
despair | +0.405 |
ously | +0.392 |
Flash | +0.392 |
Neg logit contributions | |
golf | -0.163 |
makeup | -0.151 |
swing | -0.148 |
butter | -0.141 |
which | -0.138 |
Token | He |
Token ID | 679 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
ously | +0.045 |
rum | +0.043 |
despair | +0.039 |
deeply | +0.038 |
horror | +0.036 |
Neg logit contributions | |
ju | -0.080 |
golf | -0.079 |
makeup | -0.076 |
mons | -0.075 |
grapes | -0.073 |
Token | looked |
Token ID | 3114 |
Feature activation | -3.83e-03 |
Pos logit contributions | |
rum | +0.066 |
ously | +0.063 |
despair | +0.058 |
horror | +0.057 |
deeply | +0.057 |
Neg logit contributions | |
ju | -0.031 |
golf | -0.030 |
makeup | -0.030 |
butter | -0.028 |
paints | -0.027 |
Token | around |
Token ID | 1088 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
ju | +0.005 |
golf | +0.004 |
makeup | +0.004 |
paints | +0.004 |
butter | +0.004 |
Neg logit contributions | |
rum | -0.008 |
ously | -0.008 |
despair | -0.008 |
horror | -0.007 |
plea | -0.007 |
Token | but |
Token ID | 475 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
rum | +0.125 |
ously | +0.117 |
despair | +0.110 |
aced | +0.109 |
jaws | +0.107 |
Neg logit contributions | |
golf | -0.067 |
makeup | -0.060 |
ju | -0.059 |
quiz | -0.058 |
contest | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | +5.39e-03 |
Pos logit contributions | |
ously | +0.071 |
rum | +0.070 |
despair | +0.062 |
horror | +0.060 |
urbed | +0.060 |
Neg logit contributions | |
ju | -0.047 |
golf | -0.043 |
makeup | -0.040 |
butter | -0.040 |
contest | -0.039 |
Token | chill |
Token ID | 20493 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
rum | +0.012 |
ously | +0.011 |
uzz | +0.010 |
aced | +0.010 |
sey | +0.010 |
Neg logit contributions | |
ju | -0.007 |
butter | -0.006 |
golf | -0.006 |
contest | -0.006 |
competition | -0.006 |
Token | going |
Token ID | 1016 |
Feature activation | -6.27e-03 |
Pos logit contributions | |
ously | +0.103 |
rum | +0.100 |
despair | +0.094 |
deeply | +0.093 |
horror | +0.090 |
Neg logit contributions | |
ju | -0.101 |
golf | -0.097 |
makeup | -0.095 |
mons | -0.094 |
contest | -0.091 |
Token | down |
Token ID | 866 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
mons | +0.011 |
ju | +0.011 |
makeup | +0.011 |
atmeal | +0.011 |
grapes | +0.011 |
Neg logit contributions | |
ously | -0.009 |
deeply | -0.009 |
rum | -0.009 |
silently | -0.008 |
despair | -0.008 |
Token | their |
Token ID | 511 |
Feature activation | +6.54e-02 |
Pos logit contributions | |
rum | +0.159 |
ously | +0.154 |
despair | +0.149 |
horror | +0.143 |
heartbeat | +0.140 |
Neg logit contributions | |
ju | -0.124 |
golf | -0.123 |
contest | -0.115 |
makeup | -0.110 |
competition | -0.110 |
Token | spine |
Token ID | 19656 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
ously | +0.120 |
rum | +0.118 |
despair | +0.110 |
deeply | +0.109 |
horror | +0.106 |
Neg logit contributions | |
ju | -0.099 |
golf | -0.094 |
makeup | -0.091 |
mons | -0.089 |
contest | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
rum | +0.390 |
horror | +0.369 |
ously | +0.368 |
despair | +0.366 |
Flash | +0.345 |
Neg logit contributions | |
golf | -0.200 |
ju | -0.188 |
butter | -0.177 |
swing | -0.169 |
makeup | -0.168 |
Token | This |
Token ID | 770 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
rum | +0.080 |
ously | +0.079 |
despair | +0.073 |
horror | +0.068 |
jaws | +0.067 |
Neg logit contributions | |
golf | -0.074 |
ju | -0.072 |
contest | -0.068 |
mons | -0.067 |
grapes | -0.066 |
Token | was |
Token ID | 373 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
ju | +0.076 |
golf | +0.073 |
makeup | +0.071 |
mons | +0.071 |
grapes | +0.068 |
Neg logit contributions | |
ously | -0.094 |
rum | -0.092 |
despair | -0.088 |
deeply | -0.087 |
horror | -0.083 |
Token | the |
Token ID | 262 |
Feature activation | -6.40e-03 |
Pos logit contributions | |
ously | +0.066 |
rum | +0.064 |
despair | +0.060 |
deeply | +0.059 |
horror | +0.058 |
Neg logit contributions | |
ju | -0.049 |
golf | -0.047 |
makeup | -0.045 |
mons | -0.044 |
contest | -0.043 |
Token | wisdom |
Token ID | 11501 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
ju | +0.009 |
golf | +0.008 |
makeup | +0.008 |
mons | +0.008 |
contest | +0.007 |
Neg logit contributions | |
ously | -0.013 |
rum | -0.013 |
despair | -0.012 |
deeply | -0.012 |
horror | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
rum | +0.264 |
despair | +0.248 |
ously | +0.244 |
horror | +0.241 |
deeply | +0.230 |
Neg logit contributions | |
ju | -0.126 |
golf | -0.121 |
mons | -0.111 |
makeup | -0.110 |
contest | -0.103 |
Token | Ell |
Token ID | 7122 |
Feature activation | +6.16e-02 |
Pos logit contributions | |
ously | +0.089 |
aced | +0.088 |
rum | +0.083 |
jaws | +0.081 |
aned | +0.076 |
Neg logit contributions | |
makeup | -0.046 |
bake | -0.040 |
store | -0.040 |
that | -0.039 |
ju | -0.039 |
Token | a |
Token ID | 64 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
ously | +0.111 |
rum | +0.110 |
despair | +0.108 |
horror | +0.104 |
deeply | +0.104 |
Neg logit contributions | |
ju | -0.090 |
mons | -0.083 |
golf | -0.083 |
butter | -0.083 |
makeup | -0.083 |
Token | before |
Token ID | 878 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
ously | +0.077 |
rum | +0.076 |
despair | +0.072 |
horror | +0.068 |
deeply | +0.068 |
Neg logit contributions | |
golf | -0.058 |
ju | -0.057 |
makeup | -0.055 |
butter | -0.054 |
quiz | -0.053 |
Token | hopping |
Token ID | 47153 |
Feature activation | -2.10e-03 |
Pos logit contributions | |
rum | +0.114 |
ously | +0.108 |
despair | +0.101 |
aced | +0.099 |
horror | +0.098 |
Neg logit contributions | |
golf | -0.081 |
makeup | -0.074 |
ju | -0.074 |
butter | -0.072 |
gum | -0.071 |
Token | off |
Token ID | 572 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
golf | +0.003 |
ju | +0.003 |
makeup | +0.003 |
competition | +0.003 |
contest | +0.003 |
Neg logit contributions | |
rum | -0.004 |
horror | -0.004 |
despair | -0.004 |
ously | -0.004 |
heartbeat | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
rum | +0.405 |
ously | +0.384 |
despair | +0.370 |
aced | +0.356 |
heartbeat | +0.356 |
Neg logit contributions | |
golf | -0.225 |
contest | -0.212 |
ju | -0.208 |
makeup | -0.207 |
competition | -0.203 |
Token | find |
Token ID | 1064 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ously | +0.117 |
rum | +0.112 |
aced | +0.105 |
aned | +0.104 |
lessly | +0.100 |
Neg logit contributions | |
golf | -0.071 |
swing | -0.067 |
makeup | -0.066 |
quiz | -0.062 |
competition | -0.061 |
Token | his |
Token ID | 465 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
ju | +0.026 |
makeup | +0.024 |
golf | +0.024 |
butter | +0.024 |
grapes | +0.023 |
Neg logit contributions | |
rum | -0.040 |
ously | -0.039 |
despair | -0.036 |
aned | -0.035 |
urbed | -0.035 |
Token | family |
Token ID | 1641 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
ously | +0.045 |
rum | +0.043 |
despair | +0.043 |
deeply | +0.042 |
horror | +0.041 |
Neg logit contributions | |
ju | -0.054 |
mons | -0.051 |
golf | -0.050 |
atmeal | -0.050 |
makeup | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
ously | +0.063 |
rum | +0.059 |
deeply | +0.058 |
despair | +0.057 |
horror | +0.053 |
Neg logit contributions | |
ju | -0.079 |
mons | -0.077 |
golf | -0.075 |
makeup | -0.073 |
atmeal | -0.072 |
Token | time |
Token ID | 640 |
Feature activation | -2.41e-03 |
Pos logit contributions | |
mons | +0.087 |
ju | +0.085 |
golf | +0.084 |
atmeal | +0.082 |
makeup | +0.079 |
Neg logit contributions | |
ously | -0.076 |
despair | -0.072 |
deeply | -0.072 |
horror | -0.067 |
drifted | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | -4.84e-03 |
Pos logit contributions | |
ju | +0.003 |
makeup | +0.003 |
golf | +0.002 |
butter | +0.002 |
contest | +0.002 |
Neg logit contributions | |
rum | -0.005 |
ously | -0.005 |
despair | -0.005 |
deeply | -0.005 |
horror | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -2.28e-03 |
Pos logit contributions | |
ju | +0.005 |
makeup | +0.005 |
golf | +0.005 |
butter | +0.005 |
grapes | +0.004 |
Neg logit contributions | |
rum | -0.011 |
ously | -0.011 |
despair | -0.010 |
jaws | -0.010 |
horror | -0.010 |
Token | mug |
Token ID | 25152 |
Feature activation | -9.69e-03 |
Pos logit contributions | |
atmeal | +0.003 |
mons | +0.003 |
golf | +0.003 |
ju | +0.003 |
paper | +0.003 |
Neg logit contributions | |
despair | -0.004 |
horror | -0.004 |
deeply | -0.004 |
fear | -0.004 |
ously | -0.004 |
Token | flew |
Token ID | 13112 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
ju | +0.015 |
golf | +0.014 |
makeup | +0.014 |
contest | +0.013 |
butter | +0.013 |
Neg logit contributions | |
rum | -0.018 |
ously | -0.017 |
despair | -0.016 |
deeply | -0.016 |
horror | -0.015 |
Token | off |
Token ID | 572 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
rum | +0.042 |
ously | +0.038 |
despair | +0.037 |
horror | +0.036 |
heartbeat | +0.034 |
Neg logit contributions | |
ju | -0.035 |
golf | -0.034 |
makeup | -0.031 |
contest | -0.031 |
competition | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
rum | +0.403 |
ously | +0.369 |
despair | +0.355 |
heartbeat | +0.343 |
jaws | +0.337 |
Neg logit contributions | |
golf | -0.229 |
contest | -0.228 |
ju | -0.222 |
competition | -0.219 |
makeup | -0.218 |
Token | path |
Token ID | 3108 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
ously | +0.039 |
rum | +0.038 |
despair | +0.036 |
deeply | +0.035 |
horror | +0.034 |
Neg logit contributions | |
ju | -0.039 |
golf | -0.037 |
makeup | -0.036 |
mons | -0.036 |
contest | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | +9.10e-03 |
Pos logit contributions | |
rum | +0.078 |
ously | +0.072 |
despair | +0.071 |
heartbeat | +0.068 |
horror | +0.067 |
Neg logit contributions | |
golf | -0.056 |
ju | -0.055 |
contest | -0.054 |
competition | -0.053 |
makeup | -0.052 |
Token | just |
Token ID | 655 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
rum | +0.016 |
ously | +0.016 |
despair | +0.015 |
deeply | +0.014 |
horror | +0.014 |
Neg logit contributions | |
ju | -0.014 |
golf | -0.014 |
makeup | -0.013 |
butter | -0.013 |
mons | -0.013 |
Token | like |
Token ID | 588 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
rum | +0.260 |
ously | +0.257 |
despair | +0.238 |
deeply | +0.235 |
horror | +0.231 |
Neg logit contributions | |
ju | -0.171 |
golf | -0.161 |
makeup | -0.159 |
butter | -0.149 |
mons | -0.148 |
Token | the |
Token ID | 262 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
ously | +0.250 |
rum | +0.248 |
aced | +0.220 |
aned | +0.219 |
deeply | +0.218 |
Neg logit contributions | |
golf | -0.139 |
makeup | -0.129 |
ju | -0.121 |
grapes | -0.121 |
paints | -0.118 |
Token | air |
Token ID | 1633 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
ously | +0.025 |
rum | +0.025 |
despair | +0.024 |
deeply | +0.024 |
horror | +0.023 |
Neg logit contributions | |
ju | -0.019 |
golf | -0.018 |
mons | -0.018 |
makeup | -0.018 |
atmeal | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
rum | +0.096 |
ously | +0.091 |
despair | +0.088 |
horror | +0.084 |
deeply | +0.084 |
Neg logit contributions | |
golf | -0.078 |
ju | -0.078 |
makeup | -0.074 |
contest | -0.072 |
butter | -0.072 |
Token | it |
Token ID | 340 |
Feature activation | +3.54e-03 |
Pos logit contributions | |
rum | +0.021 |
ously | +0.020 |
deeply | +0.018 |
despair | +0.018 |
horror | +0.018 |
Neg logit contributions | |
ju | -0.016 |
golf | -0.016 |
makeup | -0.015 |
butter | -0.014 |
mons | -0.014 |
Token | soared |
Token ID | 35292 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
rum | +0.007 |
ously | +0.007 |
despair | +0.006 |
deeply | +0.006 |
horror | +0.006 |
Neg logit contributions | |
ju | -0.005 |
golf | -0.005 |
makeup | -0.005 |
contest | -0.005 |
butter | -0.005 |
Token | off |
Token ID | 572 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
rum | +0.102 |
ously | +0.095 |
despair | +0.093 |
horror | +0.092 |
heartbeat | +0.086 |
Neg logit contributions | |
ju | -0.067 |
golf | -0.065 |
contest | -0.059 |
competition | -0.058 |
makeup | -0.056 |
Token | under |
Token ID | 739 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
rum | +0.423 |
ously | +0.406 |
despair | +0.376 |
heartbeat | +0.366 |
aced | +0.358 |
Neg logit contributions | |
golf | -0.223 |
contest | -0.219 |
competition | -0.204 |
ju | -0.203 |
grapes | -0.185 |
Token | the |
Token ID | 262 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
rum | +0.253 |
ously | +0.250 |
deeply | +0.219 |
aced | +0.218 |
despair | +0.217 |
Neg logit contributions | |
golf | -0.135 |
makeup | -0.123 |
grapes | -0.118 |
ju | -0.117 |
contest | -0.117 |
Token | rainbow |
Token ID | 27223 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
ously | +0.029 |
rum | +0.028 |
despair | +0.027 |
deeply | +0.026 |
horror | +0.026 |
Neg logit contributions | |
ju | -0.016 |
golf | -0.015 |
makeup | -0.014 |
mons | -0.014 |
contest | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
ously | +0.027 |
rum | +0.027 |
despair | +0.025 |
deeply | +0.025 |
horror | +0.024 |
Neg logit contributions | |
ju | -0.029 |
golf | -0.028 |
makeup | -0.027 |
contest | -0.027 |
mons | -0.027 |
Token | |
Token ID | 220 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
rum | +0.082 |
ously | +0.082 |
despair | +0.075 |
aced | +0.074 |
heartbeat | +0.073 |
Neg logit contributions | |
golf | -0.058 |
ju | -0.054 |
mons | -0.053 |
makeup | -0.051 |
Draw | -0.051 |
Token | breeze |
Token ID | 28633 |
Feature activation | +6.64e-02 |
Pos logit contributions | |
despair | +0.005 |
horror | +0.005 |
heartbeat | +0.005 |
fear | +0.005 |
ously | +0.005 |
Neg logit contributions | |
atmeal | -0.005 |
ju | -0.005 |
mons | -0.005 |
golf | -0.004 |
makeup | -0.004 |
Token | blew |
Token ID | 17948 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
ously | +0.117 |
rum | +0.114 |
despair | +0.108 |
deeply | +0.107 |
horror | +0.103 |
Neg logit contributions | |
ju | -0.103 |
golf | -0.099 |
makeup | -0.097 |
mons | -0.096 |
contest | -0.093 |
Token | her |
Token ID | 607 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
rum | +0.037 |
ously | +0.036 |
despair | +0.032 |
horror | +0.031 |
deeply | +0.031 |
Neg logit contributions | |
ju | -0.030 |
golf | -0.028 |
makeup | -0.028 |
grapes | -0.027 |
butter | -0.027 |
Token | in |
Token ID | 287 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
ously | +0.105 |
rum | +0.103 |
despair | +0.095 |
deeply | +0.095 |
horror | +0.091 |
Neg logit contributions | |
ju | -0.099 |
golf | -0.096 |
makeup | -0.096 |
mons | -0.092 |
butter | -0.091 |
Token | the |
Token ID | 262 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
ously | +0.237 |
rum | +0.237 |
aned | +0.212 |
deeply | +0.211 |
urbed | +0.208 |
Neg logit contributions | |
makeup | -0.137 |
golf | -0.129 |
grapes | -0.124 |
ju | -0.123 |
butter | -0.121 |
Token | opposite |
Token ID | 6697 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
ously | +0.039 |
rum | +0.038 |
despair | +0.036 |
deeply | +0.036 |
horror | +0.034 |
Neg logit contributions | |
ju | -0.030 |
golf | -0.028 |
makeup | -0.028 |
mons | -0.027 |
contest | -0.026 |
Token | direction |
Token ID | 4571 |
Feature activation | +9.13e-02 |
Pos logit contributions | |
ously | +0.288 |
rum | +0.285 |
deeply | +0.252 |
silently | +0.247 |
urbed | +0.239 |
Neg logit contributions | |
golf | -0.304 |
ju | -0.286 |
swing | -0.284 |
contest | -0.282 |
makeup | -0.281 |
Token | of |
Token ID | 286 |
Feature activation | +7.63e-02 |
Pos logit contributions | |
ously | +0.151 |
rum | +0.150 |
despair | +0.140 |
deeply | +0.137 |
horror | +0.133 |
Neg logit contributions | |
ju | -0.149 |
golf | -0.145 |
makeup | -0.142 |
mons | -0.137 |
contest | -0.136 |
Token | the |
Token ID | 262 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
ously | +0.141 |
rum | +0.140 |
deeply | +0.128 |
despair | +0.128 |
horror | +0.123 |
Neg logit contributions | |
ju | -0.113 |
golf | -0.111 |
makeup | -0.109 |
grapes | -0.105 |
mons | -0.104 |
Token | cat |
Token ID | 3797 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
despair | +0.029 |
fear | +0.028 |
horror | +0.028 |
screams | +0.027 |
heartbeat | +0.027 |
Neg logit contributions | |
ju | -0.030 |
atmeal | -0.028 |
mons | -0.028 |
cakes | -0.025 |
redd | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
ously | +0.041 |
rum | +0.040 |
despair | +0.038 |
deeply | +0.038 |
horror | +0.036 |
Neg logit contributions | |
ju | -0.030 |
golf | -0.029 |
makeup | -0.028 |
mons | -0.027 |
contest | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
ju | +0.007 |
golf | +0.006 |
contest | +0.006 |
makeup | +0.006 |
mons | +0.006 |
Neg logit contributions | |
rum | -0.007 |
ously | -0.007 |
despair | -0.007 |
deeply | -0.007 |
horror | -0.007 |
Token | leaving |
Token ID | 4305 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
ju | +0.019 |
golf | +0.018 |
makeup | +0.017 |
butter | +0.017 |
grapes | +0.016 |
Neg logit contributions | |
rum | -0.022 |
ously | -0.022 |
despair | -0.020 |
deeply | -0.020 |
horror | -0.019 |
Token | it |
Token ID | 340 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
rum | +0.040 |
ously | +0.037 |
despair | +0.034 |
urbed | +0.033 |
heartbeat | +0.033 |
Neg logit contributions | |
golf | -0.024 |
ju | -0.023 |
makeup | -0.022 |
butter | -0.022 |
grapes | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
ously | +0.019 |
rum | +0.019 |
despair | +0.017 |
deeply | +0.017 |
horror | +0.017 |
Neg logit contributions | |
ju | -0.016 |
golf | -0.015 |
makeup | -0.014 |
contest | -0.014 |
mons | -0.014 |
Token | its |
Token ID | 663 |
Feature activation | +8.16e-02 |
Pos logit contributions | |
ously | +0.096 |
rum | +0.093 |
aced | +0.083 |
aned | +0.083 |
urbed | +0.082 |
Neg logit contributions | |
golf | -0.073 |
ju | -0.064 |
swing | -0.062 |
contest | -0.059 |
competition | -0.059 |
Token | former |
Token ID | 1966 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
ously | +0.156 |
rum | +0.153 |
aced | +0.128 |
urbed | +0.127 |
lessly | +0.126 |
Neg logit contributions | |
golf | -0.133 |
ju | -0.129 |
makeup | -0.115 |
competition | -0.112 |
swing | -0.111 |
Token | state |
Token ID | 1181 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
ously | +0.309 |
rum | +0.273 |
Flash | +0.258 |
Flash | +0.256 |
despair | +0.248 |
Neg logit contributions | |
golf | -0.298 |
ju | -0.297 |
makeup | -0.281 |
grapes | -0.272 |
paints | -0.267 |
Token | . |
Token ID | 13 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
rum | +0.110 |
ously | +0.108 |
despair | +0.103 |
horror | +0.099 |
deeply | +0.099 |
Neg logit contributions | |
ju | -0.106 |
golf | -0.103 |
makeup | -0.098 |
butter | -0.095 |
mons | -0.094 |
Token | |
Token ID | 220 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
ously | +0.091 |
rum | +0.091 |
heartbeat | +0.084 |
aced | +0.083 |
despair | +0.081 |
Neg logit contributions | |
golf | -0.062 |
ju | -0.059 |
Draw | -0.053 |
mons | -0.053 |
contest | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
ously | +0.053 |
despair | +0.051 |
rum | +0.050 |
horror | +0.048 |
heartbeat | +0.047 |
Neg logit contributions | |
golf | -0.044 |
mons | -0.043 |
ju | -0.042 |
makeup | -0.040 |
paints | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
ously | +0.077 |
rum | +0.074 |
despair | +0.073 |
heartbeat | +0.071 |
jaws | +0.071 |
Neg logit contributions | |
golf | -0.064 |
ju | -0.062 |
Draw | -0.059 |
mons | -0.057 |
contest | -0.057 |
Token | last |
Token ID | 938 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
ously | +0.081 |
rum | +0.078 |
despair | +0.075 |
deeply | +0.075 |
horror | +0.071 |
Neg logit contributions | |
ju | -0.074 |
golf | -0.071 |
mons | -0.070 |
makeup | -0.069 |
grapes | -0.067 |
Token | time |
Token ID | 640 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
rum | +0.209 |
ously | +0.208 |
despair | +0.192 |
deeply | +0.191 |
horror | +0.187 |
Neg logit contributions | |
ju | -0.157 |
golf | -0.153 |
makeup | -0.148 |
contest | -0.145 |
competition | -0.142 |
Token | , |
Token ID | 11 |
Feature activation | +1.78e-03 |
Pos logit contributions | |
rum | +0.032 |
ously | +0.032 |
despair | +0.030 |
horror | +0.029 |
deeply | +0.029 |
Neg logit contributions | |
golf | -0.030 |
ju | -0.030 |
makeup | -0.028 |
contest | -0.027 |
mons | -0.027 |
Token | before |
Token ID | 878 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
rum | +0.003 |
ously | +0.003 |
despair | +0.003 |
deeply | +0.003 |
horror | +0.003 |
Neg logit contributions | |
ju | -0.003 |
golf | -0.003 |
makeup | -0.002 |
butter | -0.002 |
swing | -0.002 |
Token | running |
Token ID | 2491 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
rum | +0.110 |
ously | +0.107 |
despair | +0.100 |
aced | +0.095 |
horror | +0.095 |
Neg logit contributions | |
golf | -0.087 |
ju | -0.083 |
makeup | -0.079 |
butter | -0.078 |
gum | -0.078 |
Token | off |
Token ID | 572 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
rum | +0.035 |
ously | +0.034 |
despair | +0.033 |
horror | +0.032 |
deeply | +0.031 |
Neg logit contributions | |
golf | -0.032 |
ju | -0.030 |
makeup | -0.030 |
contest | -0.029 |
competition | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | +5.44e-02 |
Pos logit contributions | |
rum | +0.384 |
ously | +0.371 |
despair | +0.361 |
aced | +0.344 |
horror | +0.340 |
Neg logit contributions | |
golf | -0.233 |
contest | -0.213 |
ju | -0.210 |
makeup | -0.207 |
competition | -0.203 |
Token | explore |
Token ID | 7301 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
ously | +0.120 |
rum | +0.114 |
aced | +0.107 |
aned | +0.104 |
jaws | +0.104 |
Neg logit contributions | |
swing | -0.065 |
golf | -0.063 |
makeup | -0.062 |
contest | -0.061 |
competition | -0.060 |
Token | something |
Token ID | 1223 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
ously | +0.039 |
rum | +0.039 |
despair | +0.036 |
horror | +0.034 |
deeply | +0.034 |
Neg logit contributions | |
ju | -0.033 |
golf | -0.033 |
makeup | -0.032 |
contest | -0.030 |
butter | -0.030 |
Token | else |
Token ID | 2073 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
ously | +0.036 |
rum | +0.035 |
deeply | +0.034 |
despair | +0.034 |
horror | +0.032 |
Neg logit contributions | |
ju | -0.023 |
mons | -0.022 |
golf | -0.021 |
makeup | -0.021 |
atmeal | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
ously | +0.059 |
rum | +0.056 |
despair | +0.054 |
deeply | +0.054 |
horror | +0.050 |
Neg logit contributions | |
ju | -0.072 |
golf | -0.069 |
mons | -0.069 |
makeup | -0.068 |
grapes | -0.066 |
Token | the |
Token ID | 262 |
Feature activation | +1.89e-03 |
Pos logit contributions | |
ously | +0.060 |
rum | +0.058 |
despair | +0.056 |
deeply | +0.055 |
horror | +0.053 |
Neg logit contributions | |
ju | -0.042 |
golf | -0.040 |
makeup | -0.039 |
mons | -0.038 |
contest | -0.037 |
Token | village |
Token ID | 7404 |
Feature activation | +4.00e-02 |
Pos logit contributions | |
despair | +0.003 |
fear | +0.003 |
horror | +0.003 |
plea | +0.003 |
deeply | +0.003 |
Neg logit contributions | |
atmeal | -0.003 |
mons | -0.003 |
ju | -0.003 |
gloss | -0.003 |
coat | -0.003 |
Token | back |
Token ID | 736 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
ously | +0.069 |
rum | +0.067 |
despair | +0.064 |
deeply | +0.063 |
horror | +0.060 |
Neg logit contributions | |
ju | -0.064 |
golf | -0.061 |
makeup | -0.059 |
mons | -0.059 |
contest | -0.057 |
Token | to |
Token ID | 284 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
ously | +0.198 |
rum | +0.192 |
despair | +0.183 |
deeply | +0.179 |
horror | +0.177 |
Neg logit contributions | |
ju | -0.162 |
golf | -0.156 |
makeup | -0.151 |
mons | -0.147 |
contest | -0.146 |
Token | its |
Token ID | 663 |
Feature activation | +7.64e-02 |
Pos logit contributions | |
ously | +0.078 |
rum | +0.074 |
despair | +0.067 |
deeply | +0.066 |
horror | +0.066 |
Neg logit contributions | |
golf | -0.048 |
ju | -0.048 |
makeup | -0.046 |
contest | -0.045 |
butter | -0.044 |
Token | former |
Token ID | 1966 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
ously | +0.138 |
rum | +0.132 |
despair | +0.114 |
deeply | +0.113 |
horror | +0.112 |
Neg logit contributions | |
ju | -0.128 |
golf | -0.124 |
makeup | -0.118 |
grapes | -0.117 |
contest | -0.116 |
Token | glory |
Token ID | 13476 |
Feature activation | +9.12e-02 |
Pos logit contributions | |
ously | +0.296 |
rum | +0.260 |
horror | +0.240 |
despair | +0.238 |
silently | +0.232 |
Neg logit contributions | |
ju | -0.314 |
golf | -0.295 |
makeup | -0.294 |
grapes | -0.289 |
butter | -0.280 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
ously | +0.143 |
rum | +0.140 |
deeply | +0.132 |
despair | +0.130 |
horror | +0.123 |
Neg logit contributions | |
ju | -0.158 |
golf | -0.153 |
mons | -0.149 |
makeup | -0.149 |
contest | -0.144 |
Token | |
Token ID | 220 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
rum | +0.024 |
ously | +0.024 |
despair | +0.023 |
horror | +0.021 |
heartbeat | +0.021 |
Neg logit contributions | |
ju | -0.027 |
golf | -0.026 |
makeup | -0.025 |
mons | -0.024 |
grapes | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
ously | +0.059 |
despair | +0.057 |
rum | +0.055 |
horror | +0.052 |
heartbeat | +0.052 |
Neg logit contributions | |
golf | -0.049 |
ju | -0.048 |
mons | -0.047 |
contest | -0.043 |
gum | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
ously | +0.066 |
rum | +0.064 |
despair | +0.061 |
horror | +0.058 |
deeply | +0.058 |
Neg logit contributions | |
ju | -0.069 |
golf | -0.069 |
makeup | -0.063 |
contest | -0.063 |
mons | -0.063 |
Token |
|
Token ID | 198 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
ously | +0.060 |
rum | +0.059 |
despair | +0.056 |
deeply | +0.052 |
heartbeat | +0.051 |
Neg logit contributions | |
golf | -0.067 |
ju | -0.066 |
mons | -0.063 |
makeup | -0.063 |
grapes | -0.062 |
Token |
|
Token ID | 198 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
ously | +0.059 |
despair | +0.057 |
rum | +0.056 |
horror | +0.054 |
jaws | +0.053 |
Neg logit contributions | |
golf | -0.061 |
ju | -0.060 |
makeup | -0.057 |
contest | -0.057 |
mons | -0.057 |
Token | The |
Token ID | 464 |
Feature activation | +7.81e-03 |
Pos logit contributions | |
rum | +0.045 |
ously | +0.044 |
deeply | +0.041 |
despair | +0.041 |
silently | +0.038 |
Neg logit contributions | |
ju | -0.070 |
makeup | -0.069 |
mons | -0.068 |
atmeal | -0.068 |
butter | -0.066 |
Token | plane |
Token ID | 6614 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
despair | +0.010 |
ously | +0.010 |
deeply | +0.009 |
horror | +0.009 |
rum | +0.009 |
Neg logit contributions | |
ju | -0.016 |
mons | -0.016 |
atmeal | -0.015 |
golf | -0.015 |
makeup | -0.015 |
Token | flew |
Token ID | 13112 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
rum | +0.025 |
ously | +0.025 |
despair | +0.023 |
deeply | +0.023 |
horror | +0.022 |
Neg logit contributions | |
ju | -0.023 |
golf | -0.022 |
contest | -0.021 |
makeup | -0.021 |
mons | -0.021 |
Token | off |
Token ID | 572 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
rum | +0.037 |
ously | +0.036 |
despair | +0.034 |
horror | +0.034 |
heartbeat | +0.032 |
Neg logit contributions | |
ju | -0.027 |
golf | -0.025 |
contest | -0.023 |
competition | -0.023 |
grapes | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +7.23e-03 |
Pos logit contributions | |
rum | +0.448 |
ously | +0.426 |
despair | +0.411 |
heartbeat | +0.395 |
plea | +0.388 |
Neg logit contributions | |
contest | -0.196 |
golf | -0.187 |
competition | -0.173 |
ju | -0.152 |
grapes | -0.151 |
Token | feeling |
Token ID | 4203 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
rum | +0.013 |
ously | +0.013 |
despair | +0.012 |
deeply | +0.012 |
horror | +0.012 |
Neg logit contributions | |
ju | -0.011 |
golf | -0.011 |
makeup | -0.010 |
contest | -0.010 |
grapes | -0.010 |
Token | very |
Token ID | 845 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
ously | +0.092 |
rum | +0.090 |
despair | +0.085 |
deeply | +0.085 |
horror | +0.081 |
Neg logit contributions | |
ju | -0.091 |
golf | -0.088 |
makeup | -0.086 |
mons | -0.086 |
contest | -0.083 |
Token | proud |
Token ID | 6613 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
ously | +0.127 |
rum | +0.124 |
despair | +0.113 |
horror | +0.112 |
sey | +0.111 |
Neg logit contributions | |
ju | -0.073 |
golf | -0.066 |
butter | -0.058 |
makeup | -0.058 |
contest | -0.057 |
Token | of |
Token ID | 286 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
ously | +0.069 |
rum | +0.067 |
despair | +0.063 |
deeply | +0.062 |
horror | +0.060 |
Neg logit contributions | |
ju | -0.081 |
golf | -0.078 |
makeup | -0.076 |
mons | -0.076 |
contest | -0.074 |
Token | kissed |
Token ID | 27946 |
Feature activation | +8.78e-02 |
Pos logit contributions | |
rum | +0.038 |
ously | +0.037 |
despair | +0.036 |
deeply | +0.035 |
horror | +0.035 |
Neg logit contributions | |
ju | -0.026 |
golf | -0.025 |
makeup | -0.024 |
butter | -0.023 |
mons | -0.023 |
Token | him |
Token ID | 683 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
ously | +0.128 |
rum | +0.125 |
despair | +0.116 |
deeply | +0.115 |
horror | +0.110 |
Neg logit contributions | |
ju | -0.164 |
golf | -0.158 |
makeup | -0.155 |
mons | -0.153 |
contest | -0.150 |
Token | goodbye |
Token ID | 24829 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
ously | +0.076 |
rum | +0.073 |
despair | +0.070 |
horror | +0.067 |
deeply | +0.064 |
Neg logit contributions | |
ju | -0.066 |
golf | -0.065 |
makeup | -0.063 |
butter | -0.060 |
quiz | -0.059 |
Token | before |
Token ID | 878 |
Feature activation | +6.29e-02 |
Pos logit contributions | |
rum | +0.238 |
despair | +0.227 |
ously | +0.227 |
horror | +0.219 |
jaws | +0.204 |
Neg logit contributions | |
ju | -0.163 |
golf | -0.162 |
makeup | -0.148 |
mons | -0.141 |
butter | -0.139 |
Token | running |
Token ID | 2491 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
rum | +0.119 |
ously | +0.116 |
despair | +0.109 |
deeply | +0.107 |
horror | +0.104 |
Neg logit contributions | |
golf | -0.093 |
ju | -0.089 |
makeup | -0.086 |
grapes | -0.084 |
butter | -0.082 |
Token | off |
Token ID | 572 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
rum | +0.036 |
ously | +0.035 |
despair | +0.033 |
horror | +0.032 |
deeply | +0.032 |
Neg logit contributions | |
golf | -0.037 |
makeup | -0.035 |
ju | -0.035 |
contest | -0.033 |
competition | -0.033 |
Token | with |
Token ID | 351 |
Feature activation | +5.69e-02 |
Pos logit contributions | |
rum | +0.377 |
ously | +0.373 |
despair | +0.359 |
aced | +0.349 |
heartbeat | +0.343 |
Neg logit contributions | |
golf | -0.233 |
ju | -0.210 |
makeup | -0.206 |
contest | -0.205 |
competition | -0.196 |
Token | a |
Token ID | 257 |
Feature activation | -6.10e-03 |
Pos logit contributions | |
ously | +0.094 |
rum | +0.093 |
despair | +0.086 |
deeply | +0.085 |
horror | +0.082 |
Neg logit contributions | |
ju | -0.095 |
golf | -0.093 |
makeup | -0.090 |
mons | -0.087 |
grapes | -0.087 |
Token | big |
Token ID | 1263 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
ju | +0.009 |
golf | +0.009 |
makeup | +0.008 |
contest | +0.008 |
butter | +0.008 |
Neg logit contributions | |
ously | -0.011 |
rum | -0.011 |
despair | -0.010 |
deeply | -0.010 |
horror | -0.010 |
Token | smile |
Token ID | 8212 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
ju | +0.031 |
golf | +0.031 |
makeup | +0.030 |
mons | +0.029 |
contest | +0.029 |
Neg logit contributions | |
ously | -0.019 |
rum | -0.019 |
despair | -0.017 |
deeply | -0.017 |
horror | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
ously | +0.196 |
rum | +0.193 |
despair | +0.186 |
deeply | +0.181 |
horror | +0.177 |
Neg logit contributions | |
ju | -0.173 |
golf | -0.168 |
makeup | -0.164 |
mons | -0.159 |
butter | -0.157 |
Token | night |
Token ID | 1755 |
Feature activation | +6.47e-02 |
Pos logit contributions | |
ously | +0.194 |
rum | +0.193 |
jaws | +0.177 |
aned | +0.174 |
heartbeat | +0.172 |
Neg logit contributions | |
ju | -0.113 |
contest | -0.113 |
competition | -0.111 |
golf | -0.108 |
makeup | -0.105 |
Token | . |
Token ID | 13 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
rum | +0.118 |
ously | +0.117 |
despair | +0.108 |
horror | +0.105 |
deeply | +0.103 |
Neg logit contributions | |
ju | -0.097 |
golf | -0.094 |
makeup | -0.090 |
contest | -0.088 |
butter | -0.088 |
Token | As |
Token ID | 1081 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
ously | +0.077 |
rum | +0.077 |
heartbeat | +0.072 |
jaws | +0.070 |
despair | +0.070 |
Neg logit contributions | |
golf | -0.064 |
ju | -0.060 |
basketball | -0.058 |
makeup | -0.057 |
baseball | -0.057 |
Token | she |
Token ID | 673 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
rum | +0.111 |
ously | +0.108 |
aced | +0.098 |
aned | +0.096 |
heartbeat | +0.096 |
Neg logit contributions | |
golf | -0.053 |
ju | -0.048 |
contest | -0.048 |
competition | -0.047 |
gum | -0.046 |
Token | drifted |
Token ID | 38648 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
rum | +0.112 |
ously | +0.103 |
horror | +0.096 |
despair | +0.096 |
lessly | +0.094 |
Neg logit contributions | |
butter | -0.062 |
baked | -0.061 |
atmeal | -0.060 |
ju | -0.060 |
swing | -0.059 |
Token | off |
Token ID | 572 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
rum | +0.209 |
ously | +0.186 |
horror | +0.169 |
despair | +0.167 |
disbelief | +0.161 |
Neg logit contributions | |
makeup | -0.173 |
golf | -0.170 |
competition | -0.168 |
contest | -0.167 |
quiz | -0.156 |
Token | to |
Token ID | 284 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
rum | +0.467 |
ously | +0.427 |
despair | +0.410 |
horror | +0.399 |
aced | +0.395 |
Neg logit contributions | |
golf | -0.160 |
contest | -0.157 |
competition | -0.149 |
ju | -0.146 |
makeup | -0.145 |
Token | sleep |
Token ID | 3993 |
Feature activation | +6.47e-02 |
Pos logit contributions | |
ously | +0.125 |
rum | +0.122 |
aned | +0.118 |
aced | +0.110 |
umbled | +0.109 |
Neg logit contributions | |
swing | -0.076 |
makeup | -0.075 |
quiz | -0.073 |
golf | -0.071 |
party | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
rum | +0.140 |
ously | +0.132 |
despair | +0.123 |
jaws | +0.121 |
horror | +0.121 |
Neg logit contributions | |
contest | -0.075 |
golf | -0.074 |
competition | -0.074 |
makeup | -0.073 |
quiz | -0.072 |
Token | she |
Token ID | 673 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
rum | +0.026 |
ously | +0.025 |
jaws | +0.023 |
aced | +0.023 |
heartbeat | +0.022 |
Neg logit contributions | |
ju | -0.014 |
golf | -0.013 |
makeup | -0.013 |
competition | -0.013 |
contest | -0.013 |
Token | thought |
Token ID | 1807 |
Feature activation | -1.38e-03 |
Pos logit contributions | |
rum | +0.070 |
ously | +0.065 |
jaws | +0.059 |
despair | +0.059 |
horror | +0.058 |
Neg logit contributions | |
ju | -0.043 |
butter | -0.040 |
paints | -0.040 |
makeup | -0.038 |
competition | -0.038 |
Token | kind |
Token ID | 1611 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
ously | +0.013 |
rum | +0.013 |
despair | +0.013 |
deeply | +0.012 |
horror | +0.012 |
Neg logit contributions | |
ju | -0.012 |
mons | -0.012 |
golf | -0.012 |
makeup | -0.012 |
atmeal | -0.011 |
Token | animal |
Token ID | 5044 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
rum | +0.077 |
ously | +0.075 |
despair | +0.071 |
drifted | +0.068 |
deeply | +0.068 |
Neg logit contributions | |
golf | -0.047 |
ju | -0.043 |
contest | -0.043 |
file | -0.042 |
grapes | -0.042 |
Token | before |
Token ID | 878 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
rum | +0.077 |
ously | +0.074 |
despair | +0.070 |
heartbeat | +0.069 |
horror | +0.066 |
Neg logit contributions | |
golf | -0.045 |
ju | -0.041 |
competition | -0.039 |
butter | -0.039 |
swing | -0.039 |
Token | scam |
Token ID | 19126 |
Feature activation | +5.95e-02 |
Pos logit contributions | |
rum | +0.115 |
ously | +0.109 |
despair | +0.100 |
aced | +0.098 |
deeply | +0.097 |
Neg logit contributions | |
golf | -0.088 |
ju | -0.081 |
butter | -0.078 |
contest | -0.077 |
makeup | -0.076 |
Token | pering |
Token ID | 21255 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
ously | +0.075 |
rum | +0.074 |
despair | +0.066 |
heartbeat | +0.064 |
deeply | +0.064 |
Neg logit contributions | |
golf | -0.125 |
contest | -0.116 |
ju | -0.115 |
competition | -0.114 |
makeup | -0.114 |
Token | off |
Token ID | 572 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
rum | +0.040 |
ously | +0.038 |
despair | +0.036 |
deeply | +0.035 |
heartbeat | +0.035 |
Neg logit contributions | |
golf | -0.031 |
competition | -0.029 |
contest | -0.029 |
makeup | -0.027 |
ju | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
rum | +0.421 |
ously | +0.394 |
despair | +0.385 |
heartbeat | +0.374 |
aced | +0.372 |
Neg logit contributions | |
golf | -0.206 |
contest | -0.180 |
competition | -0.172 |
ju | -0.163 |
makeup | -0.157 |
Token | explore |
Token ID | 7301 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
ously | +0.110 |
rum | +0.107 |
aned | +0.101 |
aced | +0.100 |
urbed | +0.097 |
Neg logit contributions | |
golf | -0.062 |
swing | -0.056 |
plant | -0.053 |
makeup | -0.052 |
contest | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | +7.19e-03 |
Pos logit contributions | |
rum | +0.043 |
ously | +0.042 |
despair | +0.039 |
heartbeat | +0.038 |
jaws | +0.038 |
Neg logit contributions | |
golf | -0.030 |
ju | -0.029 |
makeup | -0.027 |
grapes | -0.026 |
contest | -0.026 |
Token | forest |
Token ID | 8222 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
ously | +0.011 |
despair | +0.011 |
rum | +0.011 |
deeply | +0.011 |
horror | +0.010 |
Neg logit contributions | |
ju | -0.012 |
mons | -0.012 |
makeup | -0.012 |
atmeal | -0.011 |
golf | -0.011 |
Token | again |
Token ID | 757 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
rum | +0.054 |
ously | +0.054 |
despair | +0.050 |
horror | +0.048 |
deeply | +0.048 |
Neg logit contributions | |
ju | -0.050 |
golf | -0.049 |
makeup | -0.046 |
contest | -0.046 |
competition | -0.045 |
Token | move |
Token ID | 1445 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
rum | +0.125 |
ously | +0.120 |
jaws | +0.110 |
deeply | +0.109 |
despair | +0.108 |
Neg logit contributions | |
golf | -0.068 |
ju | -0.065 |
swing | -0.065 |
competition | -0.061 |
contest | -0.060 |
Token | again |
Token ID | 757 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
rum | +0.065 |
ously | +0.062 |
despair | +0.059 |
horror | +0.057 |
deeply | +0.056 |
Neg logit contributions | |
ju | -0.055 |
golf | -0.054 |
makeup | -0.052 |
contest | -0.050 |
competition | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
rum | +0.253 |
ously | +0.249 |
despair | +0.233 |
horror | +0.228 |
deeply | +0.225 |
Neg logit contributions | |
golf | -0.200 |
ju | -0.198 |
makeup | -0.188 |
contest | -0.183 |
competition | -0.180 |
Token | she |
Token ID | 673 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
rum | +0.054 |
ously | +0.050 |
deeply | +0.046 |
despair | +0.044 |
jaws | +0.044 |
Neg logit contributions | |
ju | -0.040 |
golf | -0.039 |
contest | -0.037 |
competition | -0.037 |
makeup | -0.036 |
Token | ran |
Token ID | 4966 |
Feature activation | -4.65e-05 |
Pos logit contributions | |
rum | +0.099 |
ously | +0.087 |
despair | +0.086 |
deeply | +0.084 |
heartbeat | +0.084 |
Neg logit contributions | |
ju | -0.059 |
swing | -0.057 |
golf | -0.057 |
competition | -0.055 |
contest | -0.053 |
Token | off |
Token ID | 572 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
golf | +0.000 |
makeup | +0.000 |
competition | +0.000 |
contest | +0.000 |
ju | +0.000 |
Neg logit contributions | |
rum | -0.000 |
ously | -0.000 |
despair | -0.000 |
horror | -0.000 |
heartbeat | -0.000 |
Token | in |
Token ID | 287 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
rum | +0.440 |
ously | +0.406 |
despair | +0.397 |
aced | +0.389 |
jaws | +0.386 |
Neg logit contributions | |
golf | -0.199 |
contest | -0.183 |
competition | -0.176 |
swing | -0.171 |
makeup | -0.162 |
Token | search |
Token ID | 2989 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
rum | +0.246 |
ously | +0.241 |
aned | +0.233 |
aced | +0.222 |
urbed | +0.221 |
Neg logit contributions | |
makeup | -0.146 |
golf | -0.146 |
competition | -0.135 |
contest | -0.133 |
swing | -0.132 |
Token | of |
Token ID | 286 |
Feature activation | +7.04e-02 |
Pos logit contributions | |
ju | +0.026 |
golf | +0.025 |
makeup | +0.025 |
contest | +0.024 |
competition | +0.024 |
Neg logit contributions | |
rum | -0.037 |
ously | -0.034 |
despair | -0.033 |
jaws | -0.032 |
heartbeat | -0.031 |
Token | her |
Token ID | 607 |
Feature activation | +6.45e-02 |
Pos logit contributions | |
ously | +0.136 |
rum | +0.135 |
despair | +0.123 |
deeply | +0.121 |
urbed | +0.117 |
Neg logit contributions | |
ju | -0.101 |
golf | -0.098 |
makeup | -0.096 |
butter | -0.091 |
grapes | -0.091 |
Token | next |
Token ID | 1306 |
Feature activation | +2.19e-03 |
Pos logit contributions | |
ously | +0.119 |
rum | +0.116 |
despair | +0.111 |
deeply | +0.110 |
horror | +0.106 |
Neg logit contributions | |
ju | -0.096 |
golf | -0.090 |
mons | -0.088 |
makeup | -0.088 |
grapes | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | +8.14e-03 |
Pos logit contributions | |
rum | +0.089 |
ously | +0.089 |
despair | +0.081 |
aced | +0.080 |
deeply | +0.079 |
Neg logit contributions | |
ju | -0.054 |
golf | -0.053 |
makeup | -0.053 |
butter | -0.051 |
grapes | -0.050 |
Token | duck |
Token ID | 22045 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
despair | +0.012 |
deeply | +0.012 |
ously | +0.012 |
horror | +0.012 |
rum | +0.012 |
Neg logit contributions | |
ju | -0.013 |
atmeal | -0.013 |
mons | -0.013 |
newspapers | -0.013 |
quiz | -0.013 |
Token | as |
Token ID | 355 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
rum | +0.035 |
ously | +0.035 |
despair | +0.033 |
horror | +0.032 |
deeply | +0.030 |
Neg logit contributions | |
golf | -0.025 |
ju | -0.025 |
butter | -0.023 |
makeup | -0.023 |
contest | -0.023 |
Token | it |
Token ID | 340 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
ously | +0.066 |
rum | +0.065 |
despair | +0.062 |
deeply | +0.060 |
horror | +0.060 |
Neg logit contributions | |
ju | -0.032 |
golf | -0.031 |
makeup | -0.030 |
mons | -0.029 |
contest | -0.028 |
Token | walked |
Token ID | 6807 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
rum | +0.022 |
ously | +0.021 |
despair | +0.020 |
horror | +0.019 |
deeply | +0.019 |
Neg logit contributions | |
ju | -0.017 |
golf | -0.016 |
butter | -0.015 |
contest | -0.015 |
mons | -0.015 |
Token | off |
Token ID | 572 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
rum | +0.035 |
ously | +0.032 |
despair | +0.032 |
horror | +0.031 |
heartbeat | +0.029 |
Neg logit contributions | |
golf | -0.030 |
ju | -0.029 |
makeup | -0.027 |
competition | -0.027 |
contest | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +5.58e-04 |
Pos logit contributions | |
rum | +0.454 |
ously | +0.424 |
despair | +0.416 |
aced | +0.405 |
jaws | +0.397 |
Neg logit contributions | |
golf | -0.177 |
contest | -0.168 |
competition | -0.159 |
makeup | -0.151 |
ju | -0.139 |
Token | and |
Token ID | 290 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
rum | +0.001 |
ously | +0.001 |
despair | +0.001 |
deeply | +0.001 |
jaws | +0.001 |
Neg logit contributions | |
ju | -0.001 |
golf | -0.001 |
makeup | -0.001 |
butter | -0.001 |
swing | -0.001 |
Token | Kate |
Token ID | 16693 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
rum | +0.031 |
ously | +0.029 |
jaws | +0.026 |
deeply | +0.026 |
despair | +0.026 |
Neg logit contributions | |
ju | -0.019 |
golf | -0.017 |
makeup | -0.017 |
competition | -0.016 |
contest | -0.016 |
Token | went |
Token ID | 1816 |
Feature activation | -2.32e-04 |
Pos logit contributions | |
rum | +0.109 |
ously | +0.095 |
despair | +0.095 |
jaws | +0.095 |
horror | +0.093 |
Neg logit contributions | |
ju | -0.061 |
butter | -0.060 |
swing | -0.057 |
baked | -0.056 |
competition | -0.055 |
Token | back |
Token ID | 736 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
golf | +0.000 |
makeup | +0.000 |
ju | +0.000 |
contest | +0.000 |
quiz | +0.000 |
Neg logit contributions | |
rum | -0.000 |
ously | -0.000 |
despair | -0.000 |
horror | -0.000 |
aced | -0.000 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
ju | +0.064 |
makeup | +0.064 |
golf | +0.063 |
butter | +0.061 |
mixer | +0.061 |
Neg logit contributions | |
ously | -0.081 |
rum | -0.080 |
aced | -0.076 |
deeply | -0.072 |
despair | -0.072 |
Token | said |
Token ID | 531 |
Feature activation | +3.42e-03 |
Pos logit contributions | |
ju | +0.046 |
makeup | +0.045 |
golf | +0.043 |
butter | +0.042 |
grapes | +0.041 |
Neg logit contributions | |
rum | -0.072 |
ously | -0.069 |
despair | -0.065 |
deeply | -0.064 |
urbed | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
rum | +0.006 |
despair | +0.006 |
ously | +0.006 |
jaws | +0.006 |
horror | +0.006 |
Neg logit contributions | |
makeup | -0.005 |
golf | -0.005 |
butter | -0.005 |
ju | -0.004 |
gum | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
golf | +0.027 |
makeup | +0.026 |
ju | +0.025 |
grapes | +0.025 |
gum | +0.024 |
Neg logit contributions | |
rum | -0.032 |
ously | -0.032 |
despair | -0.031 |
aced | -0.029 |
deeply | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
golf | +0.018 |
makeup | +0.017 |
gum | +0.016 |
Draw | +0.016 |
paints | +0.015 |
Neg logit contributions | |
ously | -0.022 |
despair | -0.021 |
rum | -0.021 |
urbed | -0.020 |
jaws | -0.020 |
Token | " |
Token ID | 1 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
golf | +0.017 |
ju | +0.016 |
makeup | +0.016 |
mons | +0.015 |
contest | +0.015 |
Neg logit contributions | |
ously | -0.029 |
rum | -0.028 |
despair | -0.027 |
deeply | -0.027 |
horror | -0.026 |
Token | It |
Token ID | 1026 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
ju | +0.258 |
atmeal | +0.254 |
golf | +0.237 |
makeup | +0.230 |
quiz | +0.230 |
Neg logit contributions | |
ously | -0.334 |
despair | -0.308 |
deeply | -0.299 |
rum | -0.299 |
silently | -0.295 |
Token | 's |
Token ID | 338 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
ju | +0.037 |
golf | +0.036 |
mons | +0.034 |
makeup | +0.034 |
grapes | +0.032 |
Neg logit contributions | |
ously | -0.062 |
rum | -0.062 |
despair | -0.059 |
deeply | -0.058 |
horror | -0.056 |
Token | okay |
Token ID | 8788 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
ju | +0.023 |
golf | +0.023 |
makeup | +0.022 |
butter | +0.021 |
grapes | +0.021 |
Neg logit contributions | |
ously | -0.035 |
rum | -0.033 |
despair | -0.032 |
deeply | -0.031 |
horror | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -7.14e-02 |
Pos logit contributions | |
ously | +0.034 |
rum | +0.033 |
despair | +0.031 |
deeply | +0.031 |
horror | +0.030 |
Neg logit contributions | |
ju | -0.034 |
golf | -0.033 |
makeup | -0.032 |
mons | -0.032 |
contest | -0.031 |
Token | my |
Token ID | 616 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
ju | +0.110 |
mons | +0.108 |
atmeal | +0.106 |
golf | +0.105 |
contest | +0.103 |
Neg logit contributions | |
ously | -0.123 |
rum | -0.119 |
despair | -0.118 |
horror | -0.116 |
deeply | -0.116 |
Token | eat |
Token ID | 4483 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
ously | +0.067 |
rum | +0.066 |
deeply | +0.061 |
despair | +0.061 |
horror | +0.061 |
Neg logit contributions | |
ju | -0.048 |
golf | -0.047 |
makeup | -0.044 |
contest | -0.044 |
quiz | -0.043 |
Token | them |
Token ID | 606 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
ju | +0.043 |
golf | +0.041 |
makeup | +0.040 |
mons | +0.040 |
contest | +0.039 |
Neg logit contributions | |
ously | -0.051 |
rum | -0.050 |
despair | -0.048 |
deeply | -0.047 |
horror | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
ously | +0.023 |
rum | +0.022 |
despair | +0.021 |
deeply | +0.020 |
horror | +0.020 |
Neg logit contributions | |
ju | -0.025 |
golf | -0.024 |
makeup | -0.024 |
mons | -0.024 |
contest | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -9.89e-03 |
Pos logit contributions | |
golf | +0.036 |
ju | +0.035 |
makeup | +0.034 |
mons | +0.032 |
grapes | +0.032 |
Neg logit contributions | |
ously | -0.036 |
rum | -0.034 |
despair | -0.031 |
deeply | -0.031 |
aced | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
golf | +0.014 |
makeup | +0.013 |
Draw | +0.013 |
ju | +0.013 |
grapes | +0.013 |
Neg logit contributions | |
ously | -0.019 |
rum | -0.017 |
despair | -0.017 |
urbed | -0.017 |
jaws | -0.016 |
Token | " |
Token ID | 1 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
golf | +0.012 |
ju | +0.011 |
makeup | +0.011 |
mons | +0.010 |
contest | +0.010 |
Neg logit contributions | |
ously | -0.023 |
rum | -0.022 |
deeply | -0.021 |
despair | -0.021 |
horror | -0.021 |
Token | Maybe |
Token ID | 13300 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
ju | +0.244 |
golf | +0.229 |
makeup | +0.225 |
atmeal | +0.221 |
quiz | +0.218 |
Neg logit contributions | |
ously | -0.354 |
rum | -0.340 |
despair | -0.334 |
horror | -0.318 |
deeply | -0.315 |
Token | we |
Token ID | 356 |
Feature activation | -5.84e-02 |
Pos logit contributions | |
ju | +0.034 |
mons | +0.031 |
atmeal | +0.031 |
makeup | +0.031 |
golf | +0.031 |
Neg logit contributions | |
ously | -0.046 |
rum | -0.045 |
despair | -0.043 |
deeply | -0.042 |
horror | -0.042 |
Token | can |
Token ID | 460 |
Feature activation | -5.78e-02 |
Pos logit contributions | |
ju | +0.072 |
golf | +0.068 |
makeup | +0.067 |
mons | +0.066 |
atmeal | +0.063 |
Neg logit contributions | |
ously | -0.121 |
rum | -0.118 |
despair | -0.115 |
deeply | -0.113 |
horror | -0.111 |
Token | fly |
Token ID | 6129 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
ju | +0.082 |
golf | +0.078 |
makeup | +0.076 |
mons | +0.075 |
contest | +0.073 |
Neg logit contributions | |
ously | -0.111 |
rum | -0.108 |
despair | -0.103 |
deeply | -0.102 |
horror | -0.098 |
Token | with |
Token ID | 351 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
ju | +0.016 |
golf | +0.015 |
mons | +0.015 |
makeup | +0.015 |
atmeal | +0.014 |
Neg logit contributions | |
ously | -0.017 |
rum | -0.017 |
despair | -0.016 |
deeply | -0.016 |
horror | -0.015 |
Token | Ben |
Token ID | 3932 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
makeup | +0.010 |
golf | +0.010 |
mons | +0.009 |
ju | +0.009 |
grapes | +0.009 |
Neg logit contributions | |
rum | -0.011 |
ously | -0.011 |
aced | -0.010 |
deeply | -0.010 |
despair | -0.010 |
Token | said |
Token ID | 531 |
Feature activation | -1.86e-03 |
Pos logit contributions | |
ju | +0.060 |
makeup | +0.059 |
golf | +0.057 |
butter | +0.055 |
grapes | +0.055 |
Neg logit contributions | |
rum | -0.087 |
ously | -0.085 |
despair | -0.078 |
deeply | -0.076 |
urbed | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
golf | +0.002 |
makeup | +0.002 |
ju | +0.002 |
contest | +0.002 |
mons | +0.002 |
Neg logit contributions | |
rum | -0.003 |
ously | -0.003 |
despair | -0.003 |
horror | -0.003 |
jaws | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -9.20e-03 |
Pos logit contributions | |
golf | +0.030 |
ju | +0.028 |
gum | +0.027 |
makeup | +0.027 |
baseball | +0.027 |
Neg logit contributions | |
rum | -0.037 |
ously | -0.037 |
despair | -0.036 |
aced | -0.035 |
urbed | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
golf | +0.012 |
makeup | +0.011 |
Draw | +0.011 |
gum | +0.011 |
How | +0.010 |
Neg logit contributions | |
ously | -0.018 |
despair | -0.017 |
rum | -0.017 |
jaws | -0.017 |
iously | -0.017 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
golf | +0.012 |
ju | +0.012 |
makeup | +0.011 |
mons | +0.011 |
contest | +0.011 |
Neg logit contributions | |
ously | -0.023 |
rum | -0.022 |
deeply | -0.022 |
despair | -0.021 |
horror | -0.021 |
Token | No |
Token ID | 2949 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
ju | +0.235 |
golf | +0.224 |
makeup | +0.219 |
atmeal | +0.219 |
quiz | +0.210 |
Neg logit contributions | |
ously | -0.361 |
rum | -0.345 |
despair | -0.337 |
deeply | -0.325 |
horror | -0.320 |
Token | , |
Token ID | 11 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
golf | +0.065 |
makeup | +0.064 |
ju | +0.062 |
butter | +0.061 |
gum | +0.061 |
Neg logit contributions | |
ously | -0.094 |
rum | -0.091 |
despair | -0.089 |
deeply | -0.089 |
silently | -0.085 |
Token | I |
Token ID | 314 |
Feature activation | -8.16e-02 |
Pos logit contributions | |
ju | +0.122 |
mons | +0.119 |
golf | +0.117 |
atmeal | +0.116 |
makeup | +0.111 |
Neg logit contributions | |
ously | -0.152 |
rum | -0.149 |
despair | -0.148 |
horror | -0.147 |
deeply | -0.144 |
Token | 'm |
Token ID | 1101 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
ju | +0.121 |
golf | +0.115 |
makeup | +0.112 |
mons | +0.111 |
grapes | +0.110 |
Neg logit contributions | |
ously | -0.151 |
rum | -0.149 |
despair | -0.141 |
deeply | -0.137 |
horror | -0.135 |
Token | not |
Token ID | 407 |
Feature activation | -6.00e-03 |
Pos logit contributions | |
ju | +0.051 |
golf | +0.049 |
makeup | +0.048 |
mons | +0.046 |
contest | +0.046 |
Neg logit contributions | |
ously | -0.065 |
rum | -0.062 |
despair | -0.059 |
deeply | -0.057 |
horror | -0.056 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
golf | +0.041 |
makeup | +0.040 |
ju | +0.038 |
grapes | +0.036 |
contest | +0.036 |
Neg logit contributions | |
rum | -0.089 |
ously | -0.086 |
aced | -0.081 |
despair | -0.079 |
deeply | -0.078 |
Token | shouts |
Token ID | 34757 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
ju | +0.048 |
makeup | +0.047 |
golf | +0.047 |
paints | +0.046 |
grapes | +0.046 |
Neg logit contributions | |
rum | -0.074 |
ously | -0.074 |
despair | -0.068 |
deeply | -0.068 |
ment | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
ously | +0.034 |
rum | +0.034 |
despair | +0.033 |
deeply | +0.031 |
horror | +0.030 |
Neg logit contributions | |
golf | -0.027 |
makeup | -0.025 |
ju | -0.025 |
contest | -0.024 |
mons | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
golf | +0.026 |
baseball | +0.023 |
gum | +0.023 |
contest | +0.022 |
ju | +0.022 |
Neg logit contributions | |
rum | -0.040 |
despair | -0.038 |
ously | -0.038 |
aced | -0.036 |
urbed | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
golf | +0.014 |
How | +0.011 |
gum | +0.011 |
Draw | +0.011 |
baseball | +0.011 |
Neg logit contributions | |
ously | -0.028 |
despair | -0.026 |
rum | -0.026 |
jaws | -0.026 |
urbed | -0.026 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
golf | +0.016 |
ju | +0.013 |
makeup | +0.013 |
mons | +0.013 |
Draw | +0.013 |
Neg logit contributions | |
ously | -0.032 |
rum | -0.030 |
deeply | -0.030 |
despair | -0.029 |
horror | -0.029 |
Token | No |
Token ID | 2949 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
ju | +0.254 |
makeup | +0.236 |
atmeal | +0.236 |
golf | +0.234 |
mons | +0.228 |
Neg logit contributions | |
ously | -0.340 |
rum | -0.331 |
despair | -0.319 |
deeply | -0.304 |
horror | -0.303 |
Token | , |
Token ID | 11 |
Feature activation | -8.65e-02 |
Pos logit contributions | |
golf | +0.075 |
ju | +0.074 |
makeup | +0.072 |
contest | +0.069 |
butter | +0.069 |
Neg logit contributions | |
ously | -0.091 |
rum | -0.089 |
despair | -0.085 |
deeply | -0.085 |
horror | -0.081 |
Token | it |
Token ID | 340 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
ju | +0.122 |
mons | +0.118 |
atmeal | +0.115 |
golf | +0.112 |
makeup | +0.111 |
Neg logit contributions | |
ously | -0.161 |
rum | -0.159 |
despair | -0.155 |
horror | -0.153 |
deeply | -0.151 |
Token | 's |
Token ID | 338 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
ju | +0.072 |
golf | +0.069 |
mons | +0.067 |
makeup | +0.066 |
grapes | +0.065 |
Neg logit contributions | |
rum | -0.080 |
ously | -0.076 |
deeply | -0.072 |
despair | -0.072 |
horror | -0.070 |
Token | mine |
Token ID | 6164 |
Feature activation | -7.14e-03 |
Pos logit contributions | |
ju | +0.044 |
golf | +0.041 |
mons | +0.040 |
makeup | +0.040 |
atmeal | +0.039 |
Neg logit contributions | |
rum | -0.070 |
ously | -0.069 |
despair | -0.066 |
deeply | -0.066 |
horror | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
ously | +0.022 |
rum | +0.021 |
despair | +0.020 |
deeply | +0.020 |
horror | +0.019 |
Neg logit contributions | |
ju | -0.028 |
golf | -0.027 |
mons | -0.027 |
atmeal | -0.026 |
makeup | -0.026 |
Token | sand |
Token ID | 6450 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
ju | +0.048 |
golf | +0.045 |
mons | +0.044 |
makeup | +0.043 |
atmeal | +0.041 |
Neg logit contributions | |
ously | -0.088 |
rum | -0.086 |
despair | -0.083 |
deeply | -0.082 |
horror | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
ju | +0.081 |
golf | +0.077 |
makeup | +0.075 |
mons | +0.074 |
contest | +0.072 |
Neg logit contributions | |
ously | -0.084 |
rum | -0.080 |
deeply | -0.075 |
despair | -0.074 |
horror | -0.072 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
ju | +0.035 |
golf | +0.035 |
makeup | +0.034 |
mons | +0.033 |
grapes | +0.032 |
Neg logit contributions | |
ously | -0.026 |
rum | -0.025 |
despair | -0.023 |
deeply | -0.023 |
horror | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
golf | +0.018 |
ju | +0.017 |
Draw | +0.016 |
makeup | +0.016 |
mons | +0.016 |
Neg logit contributions | |
ously | -0.019 |
despair | -0.017 |
rum | -0.017 |
jaws | -0.016 |
horror | -0.016 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ju | +0.015 |
golf | +0.014 |
makeup | +0.014 |
mons | +0.013 |
contest | +0.013 |
Neg logit contributions | |
ously | -0.023 |
rum | -0.023 |
despair | -0.022 |
deeply | -0.022 |
horror | -0.021 |
Token | Look |
Token ID | 8567 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
ju | +0.310 |
makeup | +0.302 |
golf | +0.298 |
atmeal | +0.298 |
quiz | +0.293 |
Neg logit contributions | |
ously | -0.258 |
rum | -0.246 |
despair | -0.238 |
horror | -0.229 |
drifted | -0.218 |
Token | , |
Token ID | 11 |
Feature activation | -9.07e-02 |
Pos logit contributions | |
ju | +0.094 |
mons | +0.088 |
golf | +0.086 |
Draw | +0.085 |
atmeal | +0.084 |
Neg logit contributions | |
rum | -0.057 |
ously | -0.057 |
deeply | -0.053 |
sadly | -0.053 |
horror | -0.050 |
Token | Sam |
Token ID | 3409 |
Feature activation | -6.71e-02 |
Pos logit contributions | |
atmeal | +0.165 |
mons | +0.165 |
ju | +0.165 |
organized | +0.162 |
Draw | +0.157 |
Neg logit contributions | |
horror | -0.115 |
despair | -0.112 |
rum | -0.110 |
ously | -0.109 |
deeply | -0.103 |
Token | , |
Token ID | 11 |
Feature activation | -9.45e-02 |
Pos logit contributions | |
ju | +0.125 |
golf | +0.119 |
mons | +0.117 |
makeup | +0.116 |
contest | +0.112 |
Neg logit contributions | |
ously | -0.100 |
rum | -0.098 |
despair | -0.089 |
deeply | -0.088 |
horror | -0.085 |
Token | I |
Token ID | 314 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
mons | +0.168 |
atmeal | +0.166 |
organized | +0.165 |
coat | +0.164 |
ju | +0.156 |
Neg logit contributions | |
his | -0.122 |
its | -0.116 |
fear | -0.116 |
horror | -0.115 |
despair | -0.111 |
Token | Ben |
Token ID | 3932 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
golf | +0.027 |
makeup | +0.026 |
ju | +0.026 |
grapes | +0.025 |
mons | +0.024 |
Neg logit contributions | |
rum | -0.040 |
ously | -0.039 |
despair | -0.037 |
deeply | -0.037 |
aced | -0.036 |
Token | says |
Token ID | 1139 |
Feature activation | +4.21e-03 |
Pos logit contributions | |
ju | +0.062 |
makeup | +0.061 |
butter | +0.060 |
paints | +0.057 |
golf | +0.056 |
Neg logit contributions | |
rum | -0.133 |
ously | -0.125 |
despair | -0.123 |
horror | -0.120 |
deeply | -0.118 |
Token | . |
Token ID | 13 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
rum | +0.008 |
despair | +0.008 |
ously | +0.007 |
horror | +0.007 |
jaws | +0.007 |
Neg logit contributions | |
golf | -0.006 |
makeup | -0.005 |
ju | -0.005 |
contest | -0.005 |
butter | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
golf | +0.029 |
ju | +0.027 |
makeup | +0.025 |
gum | +0.025 |
grapes | +0.025 |
Neg logit contributions | |
rum | -0.031 |
ously | -0.030 |
despair | -0.030 |
aced | -0.028 |
urbed | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
golf | +0.016 |
makeup | +0.014 |
gum | +0.013 |
Draw | +0.013 |
ju | +0.013 |
Neg logit contributions | |
ously | -0.021 |
despair | -0.020 |
rum | -0.020 |
urbed | -0.018 |
horror | -0.018 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
golf | +0.016 |
ju | +0.015 |
makeup | +0.014 |
mons | +0.013 |
contest | +0.013 |
Neg logit contributions | |
ously | -0.029 |
rum | -0.028 |
despair | -0.027 |
deeply | -0.027 |
horror | -0.027 |
Token | You |
Token ID | 1639 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
ju | +0.241 |
atmeal | +0.231 |
makeup | +0.231 |
golf | +0.227 |
butter | +0.220 |
Neg logit contributions | |
ously | -0.345 |
rum | -0.326 |
despair | -0.318 |
deeply | -0.309 |
silently | -0.303 |
Token | are |
Token ID | 389 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
ju | +0.051 |
golf | +0.050 |
mons | +0.048 |
makeup | +0.048 |
contest | +0.046 |
Neg logit contributions | |
ously | -0.063 |
rum | -0.062 |
despair | -0.060 |
deeply | -0.059 |
horror | -0.057 |
Token | very |
Token ID | 845 |
Feature activation | -9.22e-03 |
Pos logit contributions | |
ju | +0.042 |
golf | +0.040 |
makeup | +0.039 |
mons | +0.038 |
contest | +0.037 |
Neg logit contributions | |
ously | -0.057 |
rum | -0.056 |
despair | -0.053 |
deeply | -0.052 |
horror | -0.051 |
Token | wise |
Token ID | 10787 |
Feature activation | +5.12e-04 |
Pos logit contributions | |
ju | +0.012 |
golf | +0.011 |
makeup | +0.011 |
mons | +0.011 |
contest | +0.010 |
Neg logit contributions | |
ously | -0.019 |
rum | -0.019 |
despair | -0.018 |
deeply | -0.018 |
horror | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
ously | +0.001 |
rum | +0.001 |
despair | +0.001 |
deeply | +0.001 |
horror | +0.001 |
Neg logit contributions | |
ju | -0.001 |
golf | -0.001 |
makeup | -0.001 |
mons | -0.001 |
contest | -0.001 |
Token | Tom |
Token ID | 4186 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
golf | +0.013 |
makeup | +0.013 |
ju | +0.012 |
grapes | +0.012 |
gum | +0.012 |
Neg logit contributions | |
ously | -0.016 |
rum | -0.016 |
aced | -0.016 |
despair | -0.015 |
deeply | -0.015 |
Token | says |
Token ID | 1139 |
Feature activation | -5.80e-03 |
Pos logit contributions | |
ju | +0.041 |
makeup | +0.040 |
golf | +0.039 |
paints | +0.038 |
butter | +0.038 |
Neg logit contributions | |
rum | -0.050 |
ously | -0.050 |
despair | -0.046 |
deeply | -0.045 |
horror | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
golf | +0.008 |
makeup | +0.008 |
ju | +0.008 |
contest | +0.008 |
quiz | +0.007 |
Neg logit contributions | |
ously | -0.010 |
rum | -0.010 |
despair | -0.010 |
horror | -0.009 |
jaws | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -7.65e-03 |
Pos logit contributions | |
golf | +0.024 |
ju | +0.023 |
gum | +0.022 |
grapes | +0.022 |
makeup | +0.022 |
Neg logit contributions | |
ously | -0.022 |
rum | -0.021 |
despair | -0.021 |
urbed | -0.020 |
aced | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -9.43e-03 |
Pos logit contributions | |
golf | +0.010 |
Draw | +0.009 |
gum | +0.009 |
ju | +0.009 |
makeup | +0.009 |
Neg logit contributions | |
ously | -0.015 |
despair | -0.014 |
rum | -0.014 |
jaws | -0.014 |
urbed | -0.013 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
golf | +0.010 |
ju | +0.010 |
makeup | +0.009 |
contest | +0.009 |
mons | +0.009 |
Neg logit contributions | |
ously | -0.021 |
rum | -0.020 |
despair | -0.020 |
deeply | -0.020 |
horror | -0.019 |
Token | That |
Token ID | 2504 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
ju | +0.251 |
golf | +0.238 |
makeup | +0.235 |
atmeal | +0.232 |
mons | +0.226 |
Neg logit contributions | |
ously | -0.340 |
rum | -0.326 |
despair | -0.314 |
deeply | -0.304 |
horror | -0.298 |
Token | is |
Token ID | 318 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
ju | +0.084 |
golf | +0.080 |
mons | +0.080 |
makeup | +0.076 |
grapes | +0.075 |
Neg logit contributions | |
ously | -0.096 |
rum | -0.094 |
deeply | -0.090 |
despair | -0.089 |
horror | -0.086 |
Token | a |
Token ID | 257 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
ju | +0.044 |
mons | +0.043 |
atmeal | +0.042 |
golf | +0.041 |
makeup | +0.041 |
Neg logit contributions | |
rum | -0.050 |
deeply | -0.049 |
ously | -0.048 |
despair | -0.047 |
horror | -0.046 |
Token | good |
Token ID | 922 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
mons | +0.078 |
ju | +0.077 |
atmeal | +0.072 |
grapes | +0.067 |
golf | +0.067 |
Neg logit contributions | |
deeply | -0.083 |
rum | -0.082 |
ously | -0.081 |
despair | -0.081 |
horror | -0.080 |
Token | idea |
Token ID | 2126 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
ju | +0.028 |
mons | +0.026 |
golf | +0.026 |
atmeal | +0.025 |
makeup | +0.025 |
Neg logit contributions | |
ously | -0.029 |
rum | -0.028 |
despair | -0.027 |
deeply | -0.027 |
horror | -0.026 |
Token | pretty |
Token ID | 2495 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
ju | +0.089 |
mons | +0.083 |
atmeal | +0.082 |
makeup | +0.079 |
golf | +0.079 |
Neg logit contributions | |
ously | -0.104 |
rum | -0.099 |
despair | -0.096 |
deeply | -0.093 |
horror | -0.091 |
Token | hair |
Token ID | 4190 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
ju | +0.078 |
golf | +0.071 |
mons | +0.070 |
atmeal | +0.069 |
contest | +0.067 |
Neg logit contributions | |
ously | -0.122 |
rum | -0.117 |
deeply | -0.113 |
despair | -0.112 |
horror | -0.107 |
Token | ." |
Token ID | 526 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
golf | +0.032 |
ju | +0.032 |
Draw | +0.031 |
competition | +0.030 |
contest | +0.029 |
Neg logit contributions | |
ously | -0.022 |
deeply | -0.022 |
rum | -0.021 |
sadly | -0.021 |
silently | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
golf | +0.028 |
makeup | +0.027 |
ju | +0.027 |
grapes | +0.026 |
mons | +0.025 |
Neg logit contributions | |
ously | -0.032 |
rum | -0.031 |
despair | -0.030 |
aced | -0.030 |
deeply | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
golf | +0.016 |
makeup | +0.015 |
ju | +0.014 |
Draw | +0.014 |
gum | +0.014 |
Neg logit contributions | |
ously | -0.018 |
despair | -0.017 |
rum | -0.017 |
jaws | -0.017 |
horror | -0.017 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ju | +0.013 |
golf | +0.013 |
makeup | +0.013 |
mons | +0.012 |
contest | +0.012 |
Neg logit contributions | |
ously | -0.023 |
rum | -0.022 |
deeply | -0.022 |
despair | -0.022 |
horror | -0.021 |
Token | Hi |
Token ID | 17250 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
atmeal | +0.249 |
ju | +0.244 |
golf | +0.242 |
quiz | +0.239 |
file | +0.239 |
Neg logit contributions | |
ously | -0.303 |
despair | -0.280 |
rum | -0.270 |
deeply | -0.266 |
silently | -0.266 |
Token | , |
Token ID | 11 |
Feature activation | -8.45e-02 |
Pos logit contributions | |
ju | +0.064 |
mons | +0.063 |
atmeal | +0.062 |
Draw | +0.060 |
golf | +0.059 |
Neg logit contributions | |
rum | -0.039 |
ously | -0.039 |
despair | -0.037 |
horror | -0.036 |
deeply | -0.036 |
Token | Max |
Token ID | 5436 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
ju | +0.127 |
mons | +0.125 |
atmeal | +0.120 |
golf | +0.119 |
makeup | +0.113 |
Neg logit contributions | |
ously | -0.143 |
rum | -0.141 |
despair | -0.136 |
deeply | -0.135 |
horror | -0.133 |
Token | ," |
Token ID | 553 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
ju | +0.042 |
golf | +0.040 |
makeup | +0.039 |
mons | +0.039 |
contest | +0.038 |
Neg logit contributions | |
ously | -0.042 |
rum | -0.041 |
despair | -0.039 |
deeply | -0.038 |
horror | -0.037 |
Token | Tom |
Token ID | 4186 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
ju | +0.046 |
golf | +0.044 |
makeup | +0.043 |
grapes | +0.041 |
mons | +0.041 |
Neg logit contributions | |
ously | -0.064 |
rum | -0.063 |
despair | -0.059 |
deeply | -0.058 |
horror | -0.056 |
Token | were |
Token ID | 547 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
ju | +0.039 |
makeup | +0.036 |
golf | +0.036 |
paints | +0.035 |
butter | +0.034 |
Neg logit contributions | |
rum | -0.066 |
ously | -0.064 |
deeply | -0.059 |
despair | -0.058 |
heartbeat | -0.056 |
Token | bored |
Token ID | 17533 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
ju | +0.024 |
golf | +0.023 |
makeup | +0.023 |
grapes | +0.022 |
mons | +0.022 |
Neg logit contributions | |
ously | -0.031 |
rum | -0.030 |
despair | -0.028 |
deeply | -0.028 |
horror | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
ju | +0.064 |
mons | +0.061 |
atmeal | +0.059 |
golf | +0.059 |
makeup | +0.056 |
Neg logit contributions | |
ously | -0.045 |
rum | -0.042 |
despair | -0.041 |
deeply | -0.041 |
horror | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
golf | +0.037 |
ju | +0.037 |
makeup | +0.036 |
mons | +0.035 |
grapes | +0.034 |
Neg logit contributions | |
ously | -0.027 |
rum | -0.026 |
deeply | -0.023 |
despair | -0.023 |
urbed | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
golf | +0.019 |
makeup | +0.017 |
mons | +0.017 |
ju | +0.017 |
Draw | +0.017 |
Neg logit contributions | |
ously | -0.019 |
rum | -0.017 |
despair | -0.017 |
deeply | -0.016 |
urbed | -0.016 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ju | +0.016 |
golf | +0.016 |
makeup | +0.015 |
mons | +0.015 |
contest | +0.015 |
Neg logit contributions | |
ously | -0.022 |
rum | -0.021 |
despair | -0.020 |
deeply | -0.020 |
horror | -0.019 |
Token | Let |
Token ID | 5756 |
Feature activation | -8.55e-02 |
Pos logit contributions | |
ju | +0.254 |
atmeal | +0.249 |
quiz | +0.230 |
golf | +0.230 |
makeup | +0.223 |
Neg logit contributions | |
ously | -0.334 |
despair | -0.320 |
rum | -0.310 |
horror | -0.309 |
silently | -0.298 |
Token | 's |
Token ID | 338 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
ju | +0.123 |
mons | +0.119 |
golf | +0.112 |
atmeal | +0.110 |
makeup | +0.106 |
Neg logit contributions | |
ously | -0.160 |
rum | -0.157 |
despair | -0.155 |
horror | -0.151 |
sadly | -0.146 |
Token | organize |
Token ID | 16481 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
ju | +0.039 |
mons | +0.037 |
golf | +0.035 |
makeup | +0.034 |
grapes | +0.034 |
Neg logit contributions | |
ously | -0.050 |
rum | -0.050 |
despair | -0.049 |
deeply | -0.047 |
horror | -0.046 |
Token | our |
Token ID | 674 |
Feature activation | -4.18e-03 |
Pos logit contributions | |
ju | +0.044 |
mons | +0.041 |
golf | +0.041 |
atmeal | +0.040 |
makeup | +0.039 |
Neg logit contributions | |
rum | -0.038 |
ously | -0.038 |
deeply | -0.037 |
despair | -0.036 |
horror | -0.034 |
Token | toys |
Token ID | 14958 |
Feature activation | -3.28e-03 |
Pos logit contributions | |
ju | +0.006 |
mons | +0.006 |
atmeal | +0.005 |
golf | +0.005 |
butter | +0.005 |
Neg logit contributions | |
rum | -0.007 |
ously | -0.007 |
despair | -0.007 |
deeply | -0.007 |
horror | -0.007 |
Token | high |
Token ID | 1029 |
Feature activation | -1.61e-03 |
Pos logit contributions | |
ju | +0.130 |
mons | +0.122 |
atmeal | +0.119 |
golf | +0.118 |
contest | +0.112 |
Neg logit contributions | |
ously | -0.127 |
rum | -0.123 |
despair | -0.121 |
deeply | -0.118 |
horror | -0.116 |
Token | things |
Token ID | 1243 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
ju | +0.003 |
golf | +0.002 |
mons | +0.002 |
makeup | +0.002 |
atmeal | +0.002 |
Neg logit contributions | |
ously | -0.003 |
rum | -0.003 |
despair | -0.003 |
deeply | -0.003 |
horror | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
ju | +0.071 |
mons | +0.068 |
golf | +0.068 |
atmeal | +0.067 |
makeup | +0.065 |
Neg logit contributions | |
ously | -0.054 |
rum | -0.051 |
deeply | -0.050 |
despair | -0.048 |
horror | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | -7.40e-03 |
Pos logit contributions | |
ju | +0.027 |
golf | +0.027 |
makeup | +0.027 |
mons | +0.026 |
grapes | +0.025 |
Neg logit contributions | |
ously | -0.028 |
rum | -0.027 |
despair | -0.025 |
deeply | -0.025 |
horror | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -7.99e-03 |
Pos logit contributions | |
golf | +0.012 |
makeup | +0.011 |
Draw | +0.011 |
mons | +0.010 |
ju | +0.010 |
Neg logit contributions | |
ously | -0.013 |
rum | -0.012 |
despair | -0.012 |
jaws | -0.011 |
horror | -0.011 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ju | +0.011 |
golf | +0.010 |
makeup | +0.010 |
mons | +0.010 |
contest | +0.009 |
Neg logit contributions | |
ously | -0.016 |
rum | -0.015 |
despair | -0.015 |
deeply | -0.015 |
horror | -0.014 |
Token | Come |
Token ID | 16773 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
ju | +0.271 |
atmeal | +0.264 |
quiz | +0.251 |
golf | +0.249 |
makeup | +0.238 |
Neg logit contributions | |
ously | -0.312 |
despair | -0.303 |
rum | -0.289 |
horror | -0.281 |
silently | -0.274 |
Token | on |
Token ID | 319 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
ju | +0.151 |
mons | +0.141 |
redd | +0.134 |
grapes | +0.134 |
makeup | +0.131 |
Neg logit contributions | |
rum | -0.110 |
sadly | -0.106 |
ously | -0.106 |
silently | -0.105 |
deeply | -0.103 |
Token | , |
Token ID | 11 |
Feature activation | -8.24e-02 |
Pos logit contributions | |
ously | +0.035 |
rum | +0.035 |
despair | +0.033 |
deeply | +0.032 |
horror | +0.031 |
Neg logit contributions | |
ju | -0.039 |
mons | -0.037 |
golf | -0.037 |
makeup | -0.036 |
atmeal | -0.035 |
Token | Kim |
Token ID | 6502 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
ju | +0.135 |
mons | +0.130 |
golf | +0.128 |
atmeal | +0.125 |
makeup | +0.124 |
Neg logit contributions | |
ously | -0.135 |
rum | -0.131 |
despair | -0.129 |
horror | -0.124 |
deeply | -0.124 |
Token | , |
Token ID | 11 |
Feature activation | -8.59e-02 |
Pos logit contributions | |
ju | +0.074 |
golf | +0.071 |
makeup | +0.069 |
mons | +0.069 |
contest | +0.067 |
Neg logit contributions | |
ously | -0.068 |
rum | -0.066 |
despair | -0.061 |
deeply | -0.061 |
horror | -0.059 |
Token | blamed |
Token ID | 13772 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
makeup | +0.041 |
paints | +0.040 |
ju | +0.036 |
swing | +0.035 |
golf | +0.035 |
Neg logit contributions | |
rum | -0.076 |
ously | -0.074 |
despair | -0.070 |
ment | -0.070 |
heartbeat | -0.069 |
Token | Mia |
Token ID | 32189 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
ously | +0.106 |
rum | +0.103 |
deeply | +0.099 |
despair | +0.096 |
drifted | +0.096 |
Neg logit contributions | |
makeup | -0.085 |
golf | -0.085 |
grapes | -0.077 |
mons | -0.073 |
lipstick | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
golf | +0.022 |
makeup | +0.022 |
ju | +0.021 |
quiz | +0.020 |
butter | +0.020 |
Neg logit contributions | |
ously | -0.028 |
rum | -0.028 |
despair | -0.027 |
deeply | -0.025 |
horror | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -7.08e-03 |
Pos logit contributions | |
golf | +0.023 |
ju | +0.021 |
makeup | +0.021 |
grapes | +0.021 |
contest | +0.020 |
Neg logit contributions | |
rum | -0.031 |
ously | -0.030 |
despair | -0.030 |
urbed | -0.028 |
aced | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -8.91e-03 |
Pos logit contributions | |
golf | +0.010 |
makeup | +0.009 |
Draw | +0.009 |
paints | +0.008 |
contest | +0.008 |
Neg logit contributions | |
ously | -0.013 |
despair | -0.013 |
rum | -0.013 |
jaws | -0.013 |
horror | -0.012 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
golf | +0.010 |
ju | +0.010 |
makeup | +0.010 |
mons | +0.009 |
contest | +0.009 |
Neg logit contributions | |
ously | -0.019 |
rum | -0.018 |
deeply | -0.018 |
despair | -0.018 |
horror | -0.017 |
Token | It |
Token ID | 1026 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
ju | +0.246 |
atmeal | +0.237 |
butter | +0.224 |
quiz | +0.219 |
golf | +0.217 |
Neg logit contributions | |
ously | -0.339 |
despair | -0.319 |
rum | -0.318 |
horror | -0.299 |
deeply | -0.297 |
Token | 's |
Token ID | 338 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
ju | +0.041 |
golf | +0.040 |
mons | +0.039 |
makeup | +0.037 |
atmeal | +0.037 |
Neg logit contributions | |
rum | -0.061 |
ously | -0.060 |
despair | -0.058 |
deeply | -0.057 |
horror | -0.055 |
Token | your |
Token ID | 534 |
Feature activation | -3.27e-03 |
Pos logit contributions | |
ju | +0.027 |
golf | +0.026 |
mons | +0.025 |
makeup | +0.025 |
atmeal | +0.025 |
Neg logit contributions | |
ously | -0.035 |
rum | -0.035 |
despair | -0.033 |
deeply | -0.033 |
horror | -0.032 |
Token | fault |
Token ID | 8046 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
ju | +0.005 |
golf | +0.004 |
makeup | +0.004 |
mons | +0.004 |
contest | +0.004 |
Neg logit contributions | |
ously | -0.006 |
rum | -0.006 |
despair | -0.006 |
deeply | -0.006 |
horror | -0.006 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
ously | +0.056 |
rum | +0.055 |
despair | +0.055 |
horror | +0.053 |
deeply | +0.052 |
Neg logit contributions | |
ju | -0.043 |
golf | -0.042 |
makeup | -0.042 |
mons | -0.041 |
butter | -0.041 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
golf | +0.027 |
makeup | +0.026 |
ju | +0.025 |
grapes | +0.023 |
mons | +0.023 |
Neg logit contributions | |
ously | -0.030 |
rum | -0.030 |
aced | -0.028 |
deeply | -0.028 |
despair | -0.027 |
Token | said |
Token ID | 531 |
Feature activation | -4.08e-04 |
Pos logit contributions | |
ju | +0.047 |
makeup | +0.046 |
golf | +0.045 |
paints | +0.043 |
grapes | +0.042 |
Neg logit contributions | |
rum | -0.064 |
ously | -0.062 |
despair | -0.058 |
deeply | -0.058 |
horror | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
ju | +0.001 |
golf | +0.001 |
makeup | +0.001 |
mons | +0.001 |
contest | +0.001 |
Neg logit contributions | |
rum | -0.001 |
ously | -0.001 |
despair | -0.001 |
horror | -0.001 |
deeply | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -7.18e-03 |
Pos logit contributions | |
golf | +0.039 |
ju | +0.038 |
makeup | +0.037 |
mons | +0.036 |
grapes | +0.035 |
Neg logit contributions | |
ously | -0.038 |
rum | -0.038 |
despair | -0.036 |
deeply | -0.035 |
aced | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | -9.24e-03 |
Pos logit contributions | |
golf | +0.011 |
makeup | +0.010 |
Draw | +0.010 |
ju | +0.010 |
gum | +0.010 |
Neg logit contributions | |
ously | -0.013 |
rum | -0.013 |
despair | -0.013 |
jaws | -0.012 |
deeply | -0.012 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ju | +0.012 |
golf | +0.012 |
makeup | +0.011 |
mons | +0.011 |
contest | +0.011 |
Neg logit contributions | |
ously | -0.019 |
rum | -0.018 |
despair | -0.017 |
deeply | -0.017 |
horror | -0.017 |
Token | Hey |
Token ID | 10814 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
atmeal | +0.271 |
ju | +0.263 |
quiz | +0.253 |
makeup | +0.250 |
golf | +0.247 |
Neg logit contributions | |
ously | -0.318 |
despair | -0.291 |
rum | -0.286 |
silently | -0.274 |
horror | -0.274 |
Token | , |
Token ID | 11 |
Feature activation | -8.92e-02 |
Pos logit contributions | |
ju | +0.036 |
atmeal | +0.035 |
mons | +0.035 |
cakes | +0.033 |
Draw | +0.033 |
Neg logit contributions | |
rum | -0.013 |
horror | -0.013 |
ously | -0.013 |
despair | -0.012 |
sadly | -0.012 |
Token | cat |
Token ID | 3797 |
Feature activation | -5.26e-02 |
Pos logit contributions | |
ju | +0.133 |
mons | +0.126 |
atmeal | +0.125 |
organized | +0.116 |
Draw | +0.114 |
Neg logit contributions | |
ously | -0.159 |
horror | -0.156 |
despair | -0.154 |
rum | -0.153 |
fear | -0.145 |
Token | . |
Token ID | 13 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
ju | +0.113 |
atmeal | +0.103 |
mons | +0.102 |
golf | +0.101 |
makeup | +0.101 |
Neg logit contributions | |
ously | -0.067 |
rum | -0.066 |
deeply | -0.058 |
despair | -0.056 |
horror | -0.056 |
Token | Go |
Token ID | 1514 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
ju | +0.057 |
atmeal | +0.053 |
golf | +0.052 |
butter | +0.051 |
competition | +0.051 |
Neg logit contributions | |
rum | -0.053 |
ously | -0.051 |
horror | -0.050 |
despair | -0.050 |
deeply | -0.047 |
Token | find |
Token ID | 1064 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
ously | +0.029 |
rum | +0.028 |
deeply | +0.026 |
despair | +0.025 |
aced | +0.025 |
Neg logit contributions | |
golf | -0.015 |
ju | -0.014 |
contest | -0.013 |
swing | -0.012 |
competition | -0.012 |
Token | treasure |
Token ID | 14068 |
Feature activation | -6.07e-03 |
Pos logit contributions | |
ju | +0.047 |
golf | +0.046 |
makeup | +0.044 |
butter | +0.043 |
grapes | +0.042 |
Neg logit contributions | |
ously | -0.070 |
rum | -0.069 |
despair | -0.064 |
deeply | -0.063 |
horror | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
ju | +0.011 |
golf | +0.010 |
makeup | +0.010 |
mons | +0.010 |
contest | +0.010 |
Neg logit contributions | |
ously | -0.009 |
rum | -0.009 |
deeply | -0.008 |
despair | -0.008 |
horror | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -6.14e-03 |
Pos logit contributions | |
golf | +0.027 |
ju | +0.027 |
makeup | +0.025 |
mons | +0.025 |
contest | +0.024 |
Neg logit contributions | |
ously | -0.025 |
rum | -0.024 |
despair | -0.022 |
deeply | -0.021 |
aced | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -8.55e-03 |
Pos logit contributions | |
golf | +0.010 |
ju | +0.009 |
makeup | +0.009 |
Draw | +0.009 |
mons | +0.009 |
Neg logit contributions | |
ously | -0.010 |
rum | -0.010 |
despair | -0.010 |
horror | -0.009 |
jaws | -0.009 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ju | +0.011 |
golf | +0.011 |
makeup | +0.010 |
mons | +0.010 |
contest | +0.010 |
Neg logit contributions | |
ously | -0.017 |
rum | -0.017 |
despair | -0.016 |
deeply | -0.016 |
horror | -0.016 |
Token | Look |
Token ID | 8567 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
ju | +0.274 |
golf | +0.262 |
makeup | +0.258 |
mons | +0.252 |
atmeal | +0.248 |
Neg logit contributions | |
ously | -0.320 |
rum | -0.308 |
despair | -0.297 |
deeply | -0.286 |
horror | -0.283 |
Token | , |
Token ID | 11 |
Feature activation | -8.81e-02 |
Pos logit contributions | |
ju | +0.085 |
mons | +0.080 |
golf | +0.078 |
atmeal | +0.076 |
Draw | +0.076 |
Neg logit contributions | |
ously | -0.057 |
rum | -0.057 |
deeply | -0.053 |
sadly | -0.051 |
despair | -0.051 |
Token | Tom |
Token ID | 4186 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
ju | +0.149 |
mons | +0.149 |
atmeal | +0.145 |
Draw | +0.141 |
organized | +0.140 |
Neg logit contributions | |
ously | -0.125 |
rum | -0.124 |
despair | -0.122 |
horror | -0.122 |
deeply | -0.116 |
Token | , |
Token ID | 11 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
ju | +0.090 |
golf | +0.087 |
makeup | +0.085 |
mons | +0.084 |
contest | +0.082 |
Neg logit contributions | |
ously | -0.089 |
rum | -0.087 |
despair | -0.082 |
deeply | -0.081 |
horror | -0.078 |
Token | there |
Token ID | 612 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
mons | +0.150 |
atmeal | +0.144 |
organized | +0.142 |
ju | +0.136 |
coat | +0.136 |
Neg logit contributions | |
Flash | -0.134 |
rum | -0.133 |
despair | -0.133 |
horror | -0.133 |
ously | -0.131 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
ously | +0.012 |
rum | +0.012 |
despair | +0.011 |
deeply | +0.011 |
aced | +0.011 |
Neg logit contributions | |
golf | -0.012 |
makeup | -0.011 |
ju | -0.011 |
grapes | -0.011 |
gum | -0.011 |
Token | said |
Token ID | 531 |
Feature activation | +9.85e-03 |
Pos logit contributions | |
ju | +0.054 |
makeup | +0.054 |
paints | +0.053 |
golf | +0.052 |
butter | +0.051 |
Neg logit contributions | |
ously | -0.074 |
rum | -0.073 |
despair | -0.070 |
horror | -0.066 |
deeply | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
ously | +0.017 |
rum | +0.016 |
despair | +0.016 |
horror | +0.015 |
deeply | +0.015 |
Neg logit contributions | |
ju | -0.015 |
golf | -0.015 |
makeup | -0.014 |
contest | -0.014 |
quiz | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -5.50e-03 |
Pos logit contributions | |
golf | +0.021 |
ju | +0.021 |
makeup | +0.019 |
grapes | +0.019 |
gum | +0.019 |
Neg logit contributions | |
rum | -0.023 |
ously | -0.023 |
despair | -0.022 |
aced | -0.021 |
urbed | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -9.01e-03 |
Pos logit contributions | |
golf | +0.007 |
gum | +0.006 |
makeup | +0.006 |
Draw | +0.006 |
contest | +0.006 |
Neg logit contributions | |
ously | -0.011 |
despair | -0.011 |
rum | -0.011 |
iously | -0.010 |
jaws | -0.010 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
golf | +0.009 |
ju | +0.009 |
makeup | +0.008 |
contest | +0.008 |
Draw | +0.008 |
Neg logit contributions | |
ously | -0.020 |
rum | -0.020 |
despair | -0.019 |
deeply | -0.019 |
horror | -0.019 |
Token | Yes |
Token ID | 5297 |
Feature activation | -6.88e-03 |
Pos logit contributions | |
ju | +0.250 |
atmeal | +0.240 |
makeup | +0.239 |
golf | +0.236 |
butter | +0.231 |
Neg logit contributions | |
ously | -0.330 |
rum | -0.313 |
despair | -0.300 |
deeply | -0.297 |
silently | -0.288 |
Token | , |
Token ID | 11 |
Feature activation | -9.33e-02 |
Pos logit contributions | |
ju | +0.009 |
golf | +0.009 |
makeup | +0.009 |
butter | +0.009 |
contest | +0.009 |
Neg logit contributions | |
ously | -0.013 |
rum | -0.013 |
despair | -0.013 |
deeply | -0.012 |
horror | -0.012 |
Token | I |
Token ID | 314 |
Feature activation | -9.27e-02 |
Pos logit contributions | |
mons | +0.142 |
atmeal | +0.142 |
organized | +0.136 |
ju | +0.135 |
Draw | +0.133 |
Neg logit contributions | |
ously | -0.159 |
rum | -0.150 |
horror | -0.149 |
deeply | -0.149 |
despair | -0.148 |
Token | do |
Token ID | 466 |
Feature activation | -7.05e-02 |
Pos logit contributions | |
ju | +0.138 |
golf | +0.130 |
mons | +0.129 |
makeup | +0.124 |
grapes | +0.124 |
Neg logit contributions | |
ously | -0.170 |
rum | -0.166 |
despair | -0.157 |
deeply | -0.154 |
horror | -0.154 |
Token | . |
Token ID | 13 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
ju | +0.132 |
golf | +0.124 |
mons | +0.122 |
makeup | +0.120 |
atmeal | +0.118 |
Neg logit contributions | |
rum | -0.104 |
ously | -0.103 |
deeply | -0.096 |
despair | -0.093 |
horror | -0.089 |
Token | Ben |
Token ID | 3932 |
Feature activation | -5.58e-02 |
Pos logit contributions | |
golf | +0.039 |
makeup | +0.039 |
ju | +0.037 |
contest | +0.036 |
grapes | +0.036 |
Neg logit contributions | |
rum | -0.065 |
ously | -0.063 |
aced | -0.059 |
deeply | -0.058 |
despair | -0.058 |
Token | said |
Token ID | 531 |
Feature activation | +9.30e-03 |
Pos logit contributions | |
ju | +0.069 |
makeup | +0.068 |
golf | +0.066 |
butter | +0.065 |
quiz | +0.064 |
Neg logit contributions | |
rum | -0.114 |
ously | -0.109 |
despair | -0.102 |
deeply | -0.099 |
urbed | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
despair | +0.018 |
rum | +0.017 |
ously | +0.017 |
horror | +0.017 |
jaws | +0.016 |
Neg logit contributions | |
golf | -0.012 |
makeup | -0.011 |
ju | -0.011 |
contest | -0.011 |
butter | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
golf | +0.030 |
ju | +0.026 |
makeup | +0.025 |
contest | +0.025 |
gum | +0.025 |
Neg logit contributions | |
rum | -0.041 |
ously | -0.040 |
despair | -0.039 |
aced | -0.038 |
urbed | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
golf | +0.018 |
makeup | +0.015 |
gum | +0.015 |
Draw | +0.015 |
baseball | +0.015 |
Neg logit contributions | |
ously | -0.025 |
rum | -0.023 |
despair | -0.023 |
urbed | -0.022 |
iously | -0.022 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
golf | +0.018 |
ju | +0.017 |
makeup | +0.016 |
contest | +0.015 |
mons | +0.015 |
Neg logit contributions | |
ously | -0.031 |
rum | -0.030 |
deeply | -0.030 |
despair | -0.029 |
horror | -0.029 |
Token | Ben |
Token ID | 11696 |
Feature activation | -8.80e-02 |
Pos logit contributions | |
ju | +0.224 |
golf | +0.210 |
makeup | +0.207 |
mons | +0.201 |
atmeal | +0.199 |
Neg logit contributions | |
ously | -0.371 |
rum | -0.359 |
despair | -0.346 |
deeply | -0.341 |
horror | -0.330 |
Token | took |
Token ID | 1718 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
ju | +0.140 |
golf | +0.133 |
mons | +0.129 |
makeup | +0.128 |
atmeal | +0.124 |
Neg logit contributions | |
ously | -0.153 |
rum | -0.148 |
despair | -0.141 |
deeply | -0.139 |
horror | -0.135 |
Token | my |
Token ID | 616 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
ju | +0.071 |
golf | +0.071 |
makeup | +0.069 |
grapes | +0.067 |
butter | +0.065 |
Neg logit contributions | |
ously | -0.093 |
rum | -0.088 |
despair | -0.082 |
deeply | -0.082 |
horror | -0.079 |
Token | car |
Token ID | 1097 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
ju | +0.020 |
golf | +0.019 |
mons | +0.018 |
makeup | +0.018 |
atmeal | +0.018 |
Neg logit contributions | |
ously | -0.029 |
rum | -0.028 |
despair | -0.027 |
deeply | -0.027 |
horror | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
ju | +0.067 |
golf | +0.064 |
makeup | +0.063 |
mons | +0.062 |
contest | +0.061 |
Neg logit contributions | |
ously | -0.066 |
rum | -0.065 |
despair | -0.061 |
deeply | -0.060 |
horror | -0.058 |
Token | a |
Token ID | 257 |
Feature activation | -5.47e-02 |
Pos logit contributions | |
ju | +0.021 |
golf | +0.020 |
mons | +0.020 |
makeup | +0.019 |
contest | +0.019 |
Neg logit contributions | |
ously | -0.027 |
rum | -0.027 |
deeply | -0.025 |
despair | -0.025 |
horror | -0.024 |
Token | feather |
Token ID | 17972 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
ju | +0.073 |
mons | +0.072 |
atmeal | +0.068 |
golf | +0.067 |
makeup | +0.066 |
Neg logit contributions | |
ously | -0.102 |
rum | -0.100 |
deeply | -0.098 |
despair | -0.096 |
horror | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
ju | +0.071 |
golf | +0.068 |
makeup | +0.067 |
mons | +0.066 |
contest | +0.064 |
Neg logit contributions | |
ously | -0.067 |
rum | -0.065 |
deeply | -0.061 |
despair | -0.061 |
horror | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
ju | +0.030 |
golf | +0.029 |
makeup | +0.028 |
mons | +0.028 |
grapes | +0.027 |
Neg logit contributions | |
ously | -0.024 |
rum | -0.023 |
despair | -0.022 |
deeply | -0.021 |
horror | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
golf | +0.017 |
ju | +0.015 |
makeup | +0.015 |
Draw | +0.015 |
grapes | +0.015 |
Neg logit contributions | |
ously | -0.019 |
rum | -0.017 |
despair | -0.017 |
horror | -0.016 |
deeply | -0.016 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
golf | +0.013 |
ju | +0.013 |
makeup | +0.012 |
mons | +0.012 |
contest | +0.011 |
Neg logit contributions | |
ously | -0.022 |
rum | -0.021 |
despair | -0.020 |
deeply | -0.020 |
horror | -0.020 |
Token | Thank |
Token ID | 10449 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
ju | +0.257 |
atmeal | +0.257 |
quiz | +0.250 |
makeup | +0.249 |
golf | +0.247 |
Neg logit contributions | |
ously | -0.307 |
despair | -0.287 |
rum | -0.287 |
deeply | -0.270 |
horror | -0.268 |
Token | you |
Token ID | 345 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
ju | +0.046 |
golf | +0.042 |
makeup | +0.040 |
mons | +0.039 |
atmeal | +0.038 |
Neg logit contributions | |
ously | -0.104 |
rum | -0.103 |
deeply | -0.098 |
despair | -0.098 |
horror | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | -7.97e-02 |
Pos logit contributions | |
ju | +0.043 |
golf | +0.042 |
grapes | +0.041 |
butter | +0.040 |
makeup | +0.039 |
Neg logit contributions | |
ously | -0.082 |
rum | -0.080 |
despair | -0.077 |
drifted | -0.073 |
heartbeat | -0.072 |
Token | bird |
Token ID | 6512 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
ju | +0.129 |
golf | +0.125 |
makeup | +0.121 |
mons | +0.118 |
grapes | +0.117 |
Neg logit contributions | |
ously | -0.138 |
rum | -0.135 |
despair | -0.126 |
deeply | -0.124 |
horror | -0.119 |
Token | , |
Token ID | 11 |
Feature activation | -8.27e-02 |
Pos logit contributions | |
ju | +0.055 |
golf | +0.053 |
makeup | +0.052 |
mons | +0.051 |
contest | +0.050 |
Neg logit contributions | |
ously | -0.059 |
rum | -0.057 |
despair | -0.055 |
deeply | -0.054 |
horror | -0.053 |
Token | Ben |
Token ID | 3932 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
golf | +0.045 |
makeup | +0.044 |
ju | +0.043 |
grapes | +0.041 |
mons | +0.041 |
Neg logit contributions | |
rum | -0.059 |
ously | -0.058 |
despair | -0.054 |
deeply | -0.053 |
aced | -0.053 |
Token | said |
Token ID | 531 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
ju | +0.056 |
makeup | +0.053 |
golf | +0.052 |
butter | +0.051 |
grapes | +0.049 |
Neg logit contributions | |
rum | -0.095 |
ously | -0.092 |
despair | -0.088 |
horror | -0.084 |
deeply | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
despair | +0.025 |
rum | +0.025 |
ously | +0.024 |
horror | +0.024 |
jaws | +0.024 |
Neg logit contributions | |
golf | -0.016 |
makeup | -0.015 |
ju | -0.015 |
butter | -0.014 |
contest | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
golf | +0.026 |
ju | +0.022 |
baseball | +0.022 |
gum | +0.022 |
grapes | +0.022 |
Neg logit contributions | |
rum | -0.035 |
ously | -0.034 |
despair | -0.034 |
aced | -0.033 |
urbed | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
golf | +0.016 |
gum | +0.014 |
Draw | +0.014 |
ju | +0.014 |
makeup | +0.013 |
Neg logit contributions | |
ously | -0.024 |
despair | -0.023 |
rum | -0.023 |
iously | -0.022 |
urbed | -0.022 |
Token | " |
Token ID | 1 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
golf | +0.016 |
ju | +0.015 |
makeup | +0.014 |
contest | +0.014 |
mons | +0.014 |
Neg logit contributions | |
ously | -0.030 |
rum | -0.029 |
despair | -0.028 |
deeply | -0.028 |
horror | -0.027 |
Token | Then |
Token ID | 6423 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
ju | +0.257 |
atmeal | +0.253 |
makeup | +0.246 |
golf | +0.241 |
butter | +0.235 |
Neg logit contributions | |
ously | -0.325 |
rum | -0.298 |
despair | -0.295 |
deeply | -0.289 |
silently | -0.284 |
Token | you |
Token ID | 345 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
ju | +0.036 |
mons | +0.034 |
atmeal | +0.034 |
golf | +0.033 |
Draw | +0.032 |
Neg logit contributions | |
ously | -0.041 |
despair | -0.040 |
rum | -0.040 |
deeply | -0.039 |
horror | -0.038 |
Token | are |
Token ID | 389 |
Feature activation | -7.69e-03 |
Pos logit contributions | |
ju | +0.026 |
mons | +0.025 |
golf | +0.025 |
makeup | +0.024 |
atmeal | +0.024 |
Neg logit contributions | |
ously | -0.034 |
rum | -0.033 |
despair | -0.033 |
deeply | -0.033 |
horror | -0.032 |
Token | very |
Token ID | 845 |
Feature activation | +7.81e-03 |
Pos logit contributions | |
ju | +0.011 |
golf | +0.011 |
makeup | +0.011 |
butter | +0.010 |
contest | +0.010 |
Neg logit contributions | |
ously | -0.015 |
rum | -0.014 |
despair | -0.013 |
deeply | -0.013 |
horror | -0.013 |
Token | lucky |
Token ID | 9670 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
ously | +0.017 |
rum | +0.016 |
despair | +0.015 |
deeply | +0.015 |
horror | +0.015 |
Neg logit contributions | |
ju | -0.010 |
golf | -0.009 |
makeup | -0.009 |
mons | -0.009 |
contest | -0.009 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
golf | +0.017 |
makeup | +0.017 |
gum | +0.015 |
grapes | +0.015 |
ju | +0.015 |
Neg logit contributions | |
rum | -0.023 |
ously | -0.022 |
aced | -0.022 |
despair | -0.021 |
deeply | -0.021 |
Token | said |
Token ID | 531 |
Feature activation | -4.47e-03 |
Pos logit contributions | |
ju | +0.040 |
makeup | +0.038 |
golf | +0.038 |
paints | +0.036 |
grapes | +0.036 |
Neg logit contributions | |
rum | -0.048 |
ously | -0.048 |
despair | -0.045 |
deeply | -0.044 |
horror | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
ju | +0.007 |
golf | +0.007 |
makeup | +0.006 |
mons | +0.006 |
contest | +0.006 |
Neg logit contributions | |
rum | -0.008 |
ously | -0.008 |
despair | -0.008 |
deeply | -0.007 |
horror | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -7.08e-03 |
Pos logit contributions | |
golf | +0.020 |
makeup | +0.020 |
ju | +0.019 |
mons | +0.019 |
grapes | +0.019 |
Neg logit contributions | |
rum | -0.021 |
ously | -0.021 |
despair | -0.020 |
deeply | -0.019 |
aced | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -8.51e-03 |
Pos logit contributions | |
golf | +0.010 |
makeup | +0.009 |
Draw | +0.009 |
gum | +0.009 |
paints | +0.009 |
Neg logit contributions | |
ously | -0.013 |
despair | -0.013 |
rum | -0.013 |
jaws | -0.012 |
horror | -0.012 |
Token | " |
Token ID | 1 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
ju | +0.011 |
golf | +0.011 |
makeup | +0.010 |
mons | +0.010 |
contest | +0.010 |
Neg logit contributions | |
ously | -0.017 |
rum | -0.017 |
despair | -0.016 |
deeply | -0.016 |
horror | -0.015 |
Token | L |
Token ID | 43 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
atmeal | +0.266 |
ju | +0.263 |
golf | +0.248 |
quiz | +0.241 |
grapes | +0.236 |
Neg logit contributions | |
ously | -0.320 |
despair | -0.292 |
rum | -0.284 |
deeply | -0.276 |
silently | -0.276 |
Token | ily |
Token ID | 813 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
ju | +0.095 |
golf | +0.082 |
mons | +0.077 |
makeup | +0.077 |
butter | +0.076 |
Neg logit contributions | |
ously | -0.096 |
rum | -0.095 |
deeply | -0.083 |
despair | -0.082 |
aced | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | -7.78e-02 |
Pos logit contributions | |
ju | +0.093 |
golf | +0.088 |
mons | +0.085 |
atmeal | +0.084 |
makeup | +0.083 |
Neg logit contributions | |
ously | -0.072 |
rum | -0.069 |
despair | -0.064 |
deeply | -0.063 |
horror | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
ju | +0.124 |
mons | +0.122 |
atmeal | +0.118 |
organized | +0.114 |
golf | +0.107 |
Neg logit contributions | |
horror | -0.130 |
ously | -0.128 |
despair | -0.127 |
rum | -0.124 |
deeply | -0.122 |
Token | pictures |
Token ID | 5986 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
atmeal | +0.079 |
ju | +0.079 |
mons | +0.070 |
golf | +0.065 |
organized | +0.064 |
Neg logit contributions | |
ously | -0.097 |
despair | -0.096 |
horror | -0.095 |
rum | -0.090 |
heartbeat | -0.090 |
Token | is |
Token ID | 318 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ju | +0.037 |
golf | +0.036 |
mons | +0.035 |
makeup | +0.034 |
atmeal | +0.033 |
Neg logit contributions | |
ously | -0.047 |
rum | -0.046 |
deeply | -0.044 |
despair | -0.043 |
silently | -0.041 |
Token | pretty |
Token ID | 2495 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
ju | +0.026 |
golf | +0.024 |
mons | +0.024 |
makeup | +0.024 |
atmeal | +0.023 |
Neg logit contributions | |
ously | -0.037 |
rum | -0.037 |
deeply | -0.035 |
despair | -0.035 |
horror | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
ju | +0.083 |
mons | +0.081 |
atmeal | +0.080 |
golf | +0.080 |
Draw | +0.078 |
Neg logit contributions | |
ously | -0.050 |
rum | -0.049 |
deeply | -0.047 |
despair | -0.043 |
sadly | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
ju | +0.033 |
golf | +0.033 |
makeup | +0.032 |
mons | +0.031 |
grapes | +0.030 |
Neg logit contributions | |
ously | -0.032 |
rum | -0.031 |
despair | -0.029 |
deeply | -0.029 |
horror | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
golf | +0.017 |
makeup | +0.015 |
ju | +0.015 |
Draw | +0.015 |
grapes | +0.015 |
Neg logit contributions | |
ously | -0.019 |
rum | -0.018 |
despair | -0.017 |
urbed | -0.017 |
horror | -0.016 |
Token | " |
Token ID | 1 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
golf | +0.013 |
ju | +0.012 |
makeup | +0.012 |
mons | +0.011 |
contest | +0.011 |
Neg logit contributions | |
ously | -0.022 |
rum | -0.021 |
despair | -0.020 |
deeply | -0.020 |
horror | -0.020 |
Token | Can |
Token ID | 6090 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
atmeal | +0.274 |
ju | +0.269 |
quiz | +0.259 |
golf | +0.253 |
makeup | +0.248 |
Neg logit contributions | |
ously | -0.298 |
despair | -0.282 |
rum | -0.275 |
horror | -0.265 |
silently | -0.261 |
Token | I |
Token ID | 314 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
ju | +0.188 |
atmeal | +0.184 |
mons | +0.171 |
newspapers | +0.169 |
golf | +0.166 |
Neg logit contributions | |
despair | -0.222 |
rum | -0.221 |
sadly | -0.219 |
ously | -0.217 |
deeply | -0.213 |
Token | have |
Token ID | 423 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
ju | +0.134 |
mons | +0.130 |
atmeal | +0.124 |
newspapers | +0.123 |
grapes | +0.120 |
Neg logit contributions | |
ously | -0.113 |
rum | -0.113 |
despair | -0.112 |
deeply | -0.108 |
sadly | -0.107 |
Token | your |
Token ID | 534 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
ju | +0.108 |
atmeal | +0.106 |
mons | +0.105 |
coat | +0.095 |
quiz | +0.094 |
Neg logit contributions | |
despair | -0.088 |
rum | -0.088 |
deeply | -0.087 |
ously | -0.085 |
its | -0.083 |
Token | star |
Token ID | 3491 |
Feature activation | +1.52e-03 |
Pos logit contributions | |
ju | +0.032 |
atmeal | +0.029 |
mons | +0.028 |
Draw | +0.027 |
organized | +0.027 |
Neg logit contributions | |
despair | -0.042 |
rum | -0.042 |
jaws | -0.041 |
horror | -0.041 |
deeply | -0.041 |
Token | for |
Token ID | 329 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
ju | +0.044 |
golf | +0.042 |
makeup | +0.041 |
mons | +0.041 |
contest | +0.039 |
Neg logit contributions | |
ously | -0.047 |
rum | -0.046 |
despair | -0.043 |
deeply | -0.043 |
horror | -0.041 |
Token | Ben |
Token ID | 3932 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
ju | +0.019 |
golf | +0.018 |
makeup | +0.018 |
mons | +0.017 |
grapes | +0.017 |
Neg logit contributions | |
ously | -0.028 |
rum | -0.027 |
despair | -0.026 |
deeply | -0.025 |
horror | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
ju | +0.092 |
golf | +0.088 |
mons | +0.086 |
makeup | +0.086 |
contest | +0.083 |
Neg logit contributions | |
ously | -0.080 |
rum | -0.078 |
despair | -0.072 |
deeply | -0.071 |
horror | -0.069 |
Token |
|
Token ID | 198 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
golf | +0.045 |
ju | +0.044 |
makeup | +0.043 |
mons | +0.042 |
grapes | +0.041 |
Neg logit contributions | |
ously | -0.038 |
rum | -0.035 |
despair | -0.033 |
deeply | -0.032 |
urbed | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
golf | +0.022 |
makeup | +0.020 |
Draw | +0.020 |
ju | +0.020 |
gum | +0.020 |
Neg logit contributions | |
ously | -0.026 |
despair | -0.023 |
rum | -0.023 |
urbed | -0.022 |
jaws | -0.022 |
Token | " |
Token ID | 1 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
golf | +0.018 |
ju | +0.018 |
makeup | +0.017 |
mons | +0.017 |
contest | +0.016 |
Neg logit contributions | |
ously | -0.027 |
rum | -0.026 |
despair | -0.025 |
deeply | -0.025 |
horror | -0.024 |
Token | Ben |
Token ID | 11696 |
Feature activation | -8.70e-02 |
Pos logit contributions | |
ju | +0.227 |
golf | +0.213 |
atmeal | +0.213 |
makeup | +0.211 |
quiz | +0.209 |
Neg logit contributions | |
ously | -0.357 |
rum | -0.343 |
despair | -0.337 |
horror | -0.323 |
deeply | -0.318 |
Token | , |
Token ID | 11 |
Feature activation | -8.42e-02 |
Pos logit contributions | |
ju | +0.183 |
golf | +0.172 |
makeup | +0.167 |
atmeal | +0.166 |
mons | +0.166 |
Neg logit contributions | |
ously | -0.106 |
rum | -0.106 |
despair | -0.096 |
deeply | -0.093 |
horror | -0.093 |
Token | you |
Token ID | 345 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
ju | +0.128 |
mons | +0.122 |
golf | +0.118 |
atmeal | +0.114 |
makeup | +0.113 |
Neg logit contributions | |
ously | -0.149 |
rum | -0.146 |
despair | -0.144 |
horror | -0.142 |
deeply | -0.138 |
Token | need |
Token ID | 761 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
ju | +0.062 |
golf | +0.059 |
mons | +0.059 |
atmeal | +0.057 |
makeup | +0.057 |
Neg logit contributions | |
despair | -0.053 |
rum | -0.053 |
ously | -0.052 |
horror | -0.050 |
deeply | -0.048 |
Token | to |
Token ID | 284 |
Feature activation | -5.83e-03 |
Pos logit contributions | |
ju | +0.042 |
golf | +0.040 |
makeup | +0.038 |
mons | +0.038 |
contest | +0.037 |
Neg logit contributions | |
ously | -0.066 |
rum | -0.064 |
despair | -0.061 |
deeply | -0.061 |
horror | -0.059 |
Token | bush |
Token ID | 24484 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
ju | +0.008 |
golf | +0.007 |
makeup | +0.007 |
mons | +0.007 |
contest | +0.007 |
Neg logit contributions | |
ously | -0.010 |
rum | -0.010 |
despair | -0.009 |
deeply | -0.009 |
horror | -0.009 |
Token | nearby |
Token ID | 6716 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
ju | +0.050 |
golf | +0.048 |
makeup | +0.047 |
mons | +0.047 |
atmeal | +0.045 |
Neg logit contributions | |
ously | -0.041 |
rum | -0.039 |
deeply | -0.037 |
despair | -0.037 |
horror | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
ju | +0.022 |
golf | +0.021 |
makeup | +0.021 |
mons | +0.021 |
contest | +0.020 |
Neg logit contributions | |
ously | -0.023 |
rum | -0.022 |
despair | -0.021 |
deeply | -0.021 |
horror | -0.020 |
Token | He |
Token ID | 679 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
ously | +0.033 |
rum | +0.032 |
despair | +0.030 |
deeply | +0.029 |
aced | +0.028 |
Neg logit contributions | |
ju | -0.036 |
golf | -0.036 |
makeup | -0.035 |
grapes | -0.033 |
mons | -0.033 |
Token | flew |
Token ID | 13112 |
Feature activation | -1.13e-03 |
Pos logit contributions | |
rum | +0.037 |
ously | +0.036 |
despair | +0.033 |
deeply | +0.032 |
horror | +0.032 |
Neg logit contributions | |
ju | -0.027 |
makeup | -0.026 |
golf | -0.026 |
paints | -0.026 |
butter | -0.025 |
Token | over |
Token ID | 625 |
Feature activation | +9.79e-02 |
Pos logit contributions | |
ju | +0.002 |
golf | +0.002 |
makeup | +0.001 |
contest | +0.001 |
grapes | +0.001 |
Neg logit contributions | |
rum | -0.002 |
ously | -0.002 |
despair | -0.002 |
horror | -0.002 |
deeply | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
ously | +0.193 |
rum | +0.189 |
despair | +0.180 |
deeply | +0.178 |
horror | +0.173 |
Neg logit contributions | |
ju | -0.133 |
golf | -0.126 |
makeup | -0.122 |
mons | -0.121 |
contest | -0.117 |
Token | called |
Token ID | 1444 |
Feature activation | -7.91e-03 |
Pos logit contributions | |
ously | +0.020 |
rum | +0.020 |
despair | +0.018 |
deeply | +0.018 |
horror | +0.018 |
Neg logit contributions | |
ju | -0.016 |
golf | -0.015 |
makeup | -0.015 |
mons | -0.015 |
contest | -0.014 |
Token | out |
Token ID | 503 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
golf | +0.010 |
ju | +0.010 |
makeup | +0.010 |
quiz | +0.010 |
contest | +0.010 |
Neg logit contributions | |
ously | -0.015 |
rum | -0.015 |
despair | -0.014 |
aced | -0.014 |
horror | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
ously | +0.230 |
rum | +0.227 |
despair | +0.220 |
drifted | +0.208 |
aced | +0.207 |
Neg logit contributions | |
golf | -0.110 |
quiz | -0.106 |
ju | -0.103 |
contest | -0.103 |
makeup | -0.102 |
Token | " |
Token ID | 366 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
ju | +0.025 |
golf | +0.024 |
makeup | +0.023 |
swing | +0.021 |
butter | +0.021 |
Neg logit contributions | |
rum | -0.040 |
ously | -0.039 |
despair | -0.037 |
aced | -0.036 |
horror | -0.035 |
Token | didn |
Token ID | 1422 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
ju | +0.024 |
golf | +0.024 |
makeup | +0.023 |
butter | +0.022 |
mons | +0.022 |
Neg logit contributions | |
rum | -0.037 |
deeply | -0.034 |
ously | -0.034 |
despair | -0.033 |
jaws | -0.033 |
Token | 't |
Token ID | 470 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
ju | +0.036 |
golf | +0.034 |
makeup | +0.033 |
mons | +0.033 |
contest | +0.032 |
Neg logit contributions | |
ously | -0.046 |
rum | -0.045 |
despair | -0.043 |
deeply | -0.043 |
horror | -0.041 |
Token | try |
Token ID | 1949 |
Feature activation | -1.90e-03 |
Pos logit contributions | |
ously | +0.030 |
rum | +0.030 |
despair | +0.028 |
deeply | +0.027 |
horror | +0.026 |
Neg logit contributions | |
ju | -0.028 |
mons | -0.026 |
golf | -0.026 |
makeup | -0.026 |
grapes | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
ju | +0.003 |
mons | +0.003 |
atmeal | +0.003 |
grapes | +0.003 |
contest | +0.003 |
Neg logit contributions | |
ously | -0.003 |
rum | -0.003 |
deeply | -0.003 |
despair | -0.003 |
silently | -0.003 |
Token | hang |
Token ID | 8181 |
Feature activation | -2.23e-03 |
Pos logit contributions | |
ously | +0.040 |
rum | +0.039 |
despair | +0.038 |
deeply | +0.037 |
horror | +0.036 |
Neg logit contributions | |
ju | -0.033 |
golf | -0.031 |
mons | -0.031 |
makeup | -0.030 |
grapes | -0.030 |
Token | from |
Token ID | 422 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
ju | +0.004 |
atmeal | +0.004 |
mons | +0.004 |
Draw | +0.004 |
golf | +0.004 |
Neg logit contributions | |
deeply | -0.003 |
rum | -0.003 |
ously | -0.003 |
despair | -0.003 |
sadly | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
ously | +0.210 |
rum | +0.207 |
deeply | +0.199 |
despair | +0.198 |
horror | +0.192 |
Neg logit contributions | |
ju | -0.217 |
golf | -0.200 |
mons | -0.200 |
atmeal | -0.196 |
makeup | -0.195 |
Token | truck |
Token ID | 7779 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
ju | +0.016 |
golf | +0.015 |
mons | +0.015 |
makeup | +0.015 |
atmeal | +0.014 |
Neg logit contributions | |
ously | -0.021 |
rum | -0.021 |
despair | -0.020 |
deeply | -0.019 |
horror | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
ju | +0.025 |
mons | +0.024 |
makeup | +0.024 |
golf | +0.024 |
grapes | +0.023 |
Neg logit contributions | |
ously | -0.018 |
rum | -0.017 |
despair | -0.016 |
deeply | -0.016 |
horror | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
rum | +0.042 |
ously | +0.041 |
despair | +0.038 |
deeply | +0.037 |
jaws | +0.035 |
Neg logit contributions | |
ju | -0.045 |
makeup | -0.045 |
golf | -0.044 |
mons | -0.044 |
grapes | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
ously | +0.054 |
despair | +0.053 |
rum | +0.053 |
jaws | +0.052 |
horror | +0.051 |
Neg logit contributions | |
golf | -0.044 |
makeup | -0.043 |
Draw | -0.042 |
ju | -0.041 |
mons | -0.040 |
Token | to |
Token ID | 284 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
rum | +0.022 |
ously | +0.021 |
despair | +0.021 |
horror | +0.020 |
aced | +0.020 |
Neg logit contributions | |
golf | -0.006 |
makeup | -0.005 |
ju | -0.005 |
quiz | -0.005 |
contest | -0.004 |
Token | be |
Token ID | 307 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
ously | +0.042 |
rum | +0.039 |
aced | +0.038 |
deeply | +0.038 |
urbed | +0.037 |
Neg logit contributions | |
golf | -0.022 |
makeup | -0.021 |
swing | -0.021 |
contest | -0.020 |
quiz | -0.020 |
Token | with |
Token ID | 351 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
ju | +0.020 |
mons | +0.019 |
golf | +0.018 |
makeup | +0.018 |
grapes | +0.018 |
Neg logit contributions | |
ously | -0.019 |
rum | -0.018 |
despair | -0.017 |
deeply | -0.017 |
horror | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
ously | +0.041 |
rum | +0.037 |
aced | +0.032 |
despair | +0.031 |
deeply | +0.031 |
Neg logit contributions | |
makeup | -0.073 |
golf | -0.073 |
ju | -0.073 |
butter | -0.069 |
grapes | -0.069 |
Token | coat |
Token ID | 13209 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
ously | +0.045 |
rum | +0.045 |
despair | +0.043 |
deeply | +0.042 |
horror | +0.041 |
Neg logit contributions | |
ju | -0.030 |
golf | -0.029 |
makeup | -0.028 |
mons | -0.027 |
butter | -0.026 |
Token | again |
Token ID | 757 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
ju | +0.045 |
golf | +0.043 |
makeup | +0.042 |
mons | +0.042 |
contest | +0.040 |
Neg logit contributions | |
ously | -0.043 |
rum | -0.041 |
despair | -0.039 |
deeply | -0.039 |
horror | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
ously | +0.215 |
rum | +0.211 |
despair | +0.198 |
horror | +0.197 |
deeply | +0.195 |
Neg logit contributions | |
golf | -0.163 |
ju | -0.155 |
makeup | -0.148 |
butter | -0.143 |
gum | -0.142 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
ously | +0.076 |
plea | +0.074 |
despair | +0.073 |
rum | +0.073 |
jaws | +0.073 |
Neg logit contributions | |
golf | -0.035 |
ju | -0.035 |
gum | -0.033 |
Draw | -0.032 |
that | -0.032 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
ju | +0.059 |
golf | +0.059 |
makeup | +0.057 |
contest | +0.056 |
gum | +0.054 |
Neg logit contributions | |
ously | -0.062 |
rum | -0.061 |
despair | -0.058 |
deeply | -0.055 |
horror | -0.055 |
Token | there |
Token ID | 612 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
contest | +0.031 |
ju | +0.031 |
golf | +0.030 |
makeup | +0.030 |
paints | +0.029 |
Neg logit contributions | |
rum | -0.025 |
ously | -0.025 |
despair | -0.024 |
heartbeat | -0.023 |
horror | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | -1.12e-04 |
Pos logit contributions | |
ju | +0.043 |
golf | +0.041 |
makeup | +0.041 |
contest | +0.039 |
mons | +0.038 |
Neg logit contributions | |
ously | -0.061 |
rum | -0.060 |
despair | -0.056 |
deeply | -0.056 |
heartbeat | -0.055 |
Token | sticks |
Token ID | 16461 |
Feature activation | -3.03e-03 |
Pos logit contributions | |
ju | +0.036 |
golf | +0.034 |
makeup | +0.033 |
mons | +0.033 |
grapes | +0.032 |
Neg logit contributions | |
ously | -0.046 |
rum | -0.046 |
despair | -0.042 |
deeply | -0.042 |
horror | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -9.25e-03 |
Pos logit contributions | |
golf | +0.004 |
ju | +0.004 |
contest | +0.004 |
makeup | +0.004 |
butter | +0.004 |
Neg logit contributions | |
rum | -0.005 |
ously | -0.005 |
despair | -0.005 |
horror | -0.005 |
deeply | -0.005 |
Token | The |
Token ID | 383 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
golf | +0.016 |
ju | +0.015 |
makeup | +0.014 |
grapes | +0.014 |
mons | +0.014 |
Neg logit contributions | |
ously | -0.016 |
rum | -0.015 |
despair | -0.014 |
deeply | -0.014 |
aced | -0.013 |
Token | bird |
Token ID | 6512 |
Feature activation | -1.97e-03 |
Pos logit contributions | |
ju | +0.033 |
mons | +0.031 |
atmeal | +0.031 |
makeup | +0.030 |
golf | +0.030 |
Neg logit contributions | |
ously | -0.050 |
despair | -0.048 |
rum | -0.047 |
horror | -0.046 |
deeply | -0.046 |
Token | flew |
Token ID | 13112 |
Feature activation | -3.20e-03 |
Pos logit contributions | |
ju | +0.003 |
golf | +0.003 |
makeup | +0.003 |
contest | +0.003 |
mons | +0.003 |
Neg logit contributions | |
ously | -0.004 |
rum | -0.004 |
despair | -0.003 |
deeply | -0.003 |
horror | -0.003 |
Token | out |
Token ID | 503 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
ju | +0.005 |
golf | +0.004 |
contest | +0.004 |
makeup | +0.004 |
grapes | +0.004 |
Neg logit contributions | |
rum | -0.006 |
ously | -0.006 |
despair | -0.006 |
horror | -0.006 |
heartbeat | -0.005 |
Token | of |
Token ID | 286 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
rum | +0.274 |
ously | +0.266 |
despair | +0.256 |
heartbeat | +0.243 |
plea | +0.237 |
Neg logit contributions | |
golf | -0.157 |
contest | -0.151 |
ju | -0.148 |
grapes | -0.147 |
butter | -0.146 |
Token | the |
Token ID | 262 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
ously | +0.150 |
rum | +0.148 |
deeply | +0.134 |
despair | +0.131 |
aced | +0.129 |
Neg logit contributions | |
golf | -0.100 |
ju | -0.095 |
grapes | -0.095 |
makeup | -0.095 |
butter | -0.092 |
Token | nest |
Token ID | 16343 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
ously | +0.020 |
despair | +0.020 |
deeply | +0.019 |
horror | +0.019 |
jaws | +0.019 |
Neg logit contributions | |
ju | -0.021 |
mons | -0.021 |
atmeal | -0.020 |
makeup | -0.019 |
golf | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | -1.13e-03 |
Pos logit contributions | |
rum | +0.042 |
ously | +0.041 |
despair | +0.041 |
horror | +0.039 |
heartbeat | +0.037 |
Neg logit contributions | |
ju | -0.028 |
golf | -0.027 |
contest | -0.026 |
butter | -0.026 |
mons | -0.026 |
Token | attacked |
Token ID | 7384 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
ju | +0.002 |
golf | +0.001 |
butter | +0.001 |
mons | +0.001 |
makeup | +0.001 |
Neg logit contributions | |
rum | -0.002 |
ously | -0.002 |
despair | -0.002 |
deeply | -0.002 |
horror | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +1.62e-03 |
Pos logit contributions | |
ju | +0.034 |
golf | +0.033 |
makeup | +0.032 |
contest | +0.032 |
butter | +0.032 |
Neg logit contributions | |
ously | -0.038 |
rum | -0.037 |
despair | -0.035 |
deeply | -0.034 |
heartbeat | -0.034 |
Token | The |
Token ID | 383 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
ously | +0.003 |
rum | +0.003 |
despair | +0.003 |
deeply | +0.003 |
aced | +0.002 |
Neg logit contributions | |
ju | -0.002 |
golf | -0.002 |
makeup | -0.002 |
mons | -0.002 |
grapes | -0.002 |
Token | snake |
Token ID | 17522 |
Feature activation | +4.88e-03 |
Pos logit contributions | |
ju | +0.028 |
golf | +0.027 |
makeup | +0.026 |
mons | +0.026 |
atmeal | +0.025 |
Neg logit contributions | |
ously | -0.032 |
rum | -0.032 |
despair | -0.030 |
deeply | -0.030 |
horror | -0.029 |
Token | slit |
Token ID | 40724 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
ously | +0.009 |
rum | +0.009 |
despair | +0.008 |
deeply | +0.008 |
horror | +0.008 |
Neg logit contributions | |
ju | -0.007 |
golf | -0.007 |
makeup | -0.007 |
mons | -0.007 |
butter | -0.006 |
Token | hered |
Token ID | 6083 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
golf | +0.018 |
ju | +0.018 |
mons | +0.017 |
makeup | +0.017 |
contest | +0.017 |
Neg logit contributions | |
ously | -0.021 |
rum | -0.021 |
deeply | -0.020 |
despair | -0.020 |
horror | -0.019 |
Token | away |
Token ID | 1497 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
rum | +0.090 |
ously | +0.088 |
deeply | +0.086 |
despair | +0.086 |
horror | +0.081 |
Neg logit contributions | |
golf | -0.073 |
makeup | -0.071 |
ju | -0.071 |
contest | -0.068 |
grapes | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +4.96e-03 |
Pos logit contributions | |
ously | +0.285 |
despair | +0.280 |
rum | +0.278 |
obby | +0.275 |
tal | +0.275 |
Neg logit contributions | |
contest | -0.095 |
golf | -0.094 |
swing | -0.092 |
that | -0.090 |
makeup | -0.088 |
Token |
|
Token ID | 198 |
Feature activation | +8.99e-03 |
Pos logit contributions | |
ously | +0.010 |
rum | +0.010 |
aced | +0.009 |
despair | +0.009 |
urbed | +0.009 |
Neg logit contributions | |
golf | -0.007 |
makeup | -0.007 |
ju | -0.007 |
gum | -0.006 |
grapes | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +5.66e-03 |
Pos logit contributions | |
ously | +0.019 |
rum | +0.018 |
despair | +0.018 |
urbed | +0.017 |
jaws | +0.017 |
Neg logit contributions | |
golf | -0.011 |
makeup | -0.010 |
Draw | -0.010 |
gum | -0.010 |
mons | -0.009 |
Token | Mom |
Token ID | 29252 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
ously | +0.012 |
rum | +0.012 |
despair | +0.011 |
deeply | +0.011 |
horror | +0.011 |
Neg logit contributions | |
ju | -0.007 |
golf | -0.007 |
makeup | -0.006 |
mons | -0.006 |
contest | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
ju | +0.020 |
golf | +0.019 |
makeup | +0.019 |
butter | +0.018 |
mons | +0.018 |
Neg logit contributions | |
rum | -0.026 |
ously | -0.026 |
despair | -0.025 |
deeply | -0.025 |
horror | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
ously | +0.024 |
rum | +0.023 |
despair | +0.022 |
deeply | +0.022 |
horror | +0.021 |
Neg logit contributions | |
ju | -0.020 |
golf | -0.019 |
makeup | -0.018 |
mons | -0.018 |
contest | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
rum | +0.079 |
ously | +0.075 |
aced | +0.069 |
despair | +0.069 |
urbed | +0.068 |
Neg logit contributions | |
makeup | -0.041 |
ju | -0.041 |
golf | -0.040 |
grapes | -0.039 |
contest | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
ously | +0.057 |
rum | +0.055 |
despair | +0.053 |
horror | +0.051 |
deeply | +0.051 |
Neg logit contributions | |
ju | -0.063 |
golf | -0.062 |
makeup | -0.060 |
contest | -0.059 |
grapes | -0.058 |
Token | " |
Token ID | 1 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
ously | +0.060 |
rum | +0.058 |
despair | +0.056 |
deeply | +0.055 |
silently | +0.052 |
Neg logit contributions | |
ju | -0.058 |
mons | -0.055 |
atmeal | -0.055 |
golf | -0.055 |
makeup | -0.054 |
Token | Da |
Token ID | 26531 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
golf | +0.251 |
mons | +0.247 |
atmeal | +0.246 |
ju | +0.245 |
quiz | +0.240 |
Neg logit contributions | |
ously | -0.214 |
despair | -0.205 |
rum | -0.195 |
horror | -0.181 |
deeply | -0.178 |
Token | isy |
Token ID | 13560 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
ously | +0.062 |
rum | +0.061 |
despair | +0.057 |
deeply | +0.057 |
horror | +0.052 |
Neg logit contributions | |
ju | -0.065 |
golf | -0.064 |
makeup | -0.062 |
mons | -0.061 |
paints | -0.060 |
Token | , |
Token ID | 11 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
ously | +0.049 |
rum | +0.048 |
despair | +0.045 |
deeply | +0.045 |
horror | +0.043 |
Neg logit contributions | |
ju | -0.053 |
golf | -0.051 |
mons | -0.049 |
makeup | -0.049 |
grapes | -0.048 |
Token | come |
Token ID | 1282 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
mons | +0.068 |
ju | +0.066 |
golf | +0.065 |
atmeal | +0.062 |
Draw | +0.059 |
Neg logit contributions | |
ously | -0.072 |
despair | -0.072 |
rum | -0.070 |
horror | -0.068 |
deeply | -0.068 |
Token | inside |
Token ID | 2641 |
Feature activation | +7.08e-02 |
Pos logit contributions | |
ju | +0.035 |
golf | +0.033 |
mons | +0.032 |
makeup | +0.032 |
grapes | +0.031 |
Neg logit contributions | |
ously | -0.042 |
rum | -0.041 |
despair | -0.039 |
deeply | -0.038 |
horror | -0.037 |
Token | now |
Token ID | 783 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
ously | +0.143 |
rum | +0.136 |
despair | +0.129 |
deeply | +0.125 |
horror | +0.124 |
Neg logit contributions | |
ju | -0.093 |
golf | -0.091 |
makeup | -0.088 |
butter | -0.085 |
contest | -0.085 |
Token | ," |
Token ID | 553 |
Feature activation | +2.30e-03 |
Pos logit contributions | |
ously | +0.028 |
rum | +0.027 |
despair | +0.026 |
deeply | +0.025 |
horror | +0.025 |
Neg logit contributions | |
ju | -0.024 |
golf | -0.023 |
makeup | -0.023 |
mons | -0.022 |
contest | -0.022 |
Token | ." |
Token ID | 526 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
ously | +0.034 |
rum | +0.034 |
deeply | +0.031 |
despair | +0.031 |
horror | +0.029 |
Neg logit contributions | |
ju | -0.046 |
golf | -0.044 |
makeup | -0.043 |
mons | -0.042 |
grapes | -0.041 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
rum | +0.040 |
ously | +0.040 |
despair | +0.037 |
jaws | +0.035 |
horror | +0.035 |
Neg logit contributions | |
makeup | -0.038 |
ju | -0.036 |
mons | -0.034 |
grapes | -0.034 |
golf | -0.034 |
Token | nodded |
Token ID | 14464 |
Feature activation | +9.32e-02 |
Pos logit contributions | |
rum | +0.070 |
ously | +0.069 |
despair | +0.065 |
deeply | +0.064 |
horror | +0.063 |
Neg logit contributions | |
ju | -0.048 |
golf | -0.045 |
makeup | -0.044 |
mons | -0.044 |
butter | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
rum | +0.183 |
ously | +0.178 |
despair | +0.168 |
jaws | +0.164 |
horror | +0.163 |
Neg logit contributions | |
makeup | -0.122 |
golf | -0.116 |
ju | -0.113 |
butter | -0.110 |
paints | -0.106 |
Token | went |
Token ID | 1816 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
ju | +0.013 |
golf | +0.012 |
butter | +0.012 |
makeup | +0.012 |
mons | +0.012 |
Neg logit contributions | |
rum | -0.023 |
ously | -0.020 |
deeply | -0.020 |
jaws | -0.020 |
opher | -0.019 |
Token | outside |
Token ID | 2354 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
ju | +0.024 |
golf | +0.024 |
makeup | +0.023 |
mons | +0.022 |
contest | +0.022 |
Neg logit contributions | |
ously | -0.038 |
rum | -0.037 |
despair | -0.036 |
deeply | -0.035 |
horror | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
rum | +0.154 |
ously | +0.152 |
despair | +0.144 |
jaws | +0.144 |
aced | +0.142 |
Neg logit contributions | |
golf | -0.062 |
contest | -0.060 |
makeup | -0.059 |
quiz | -0.058 |
competition | -0.056 |
Token | look |
Token ID | 804 |
Feature activation | +8.07e-03 |
Pos logit contributions | |
ously | +0.059 |
rum | +0.058 |
deeply | +0.055 |
despair | +0.054 |
horror | +0.052 |
Neg logit contributions | |
ju | -0.031 |
golf | -0.030 |
makeup | -0.030 |
contest | -0.028 |
quiz | -0.028 |
Token | for |
Token ID | 329 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
ously | +0.016 |
rum | +0.016 |
despair | +0.015 |
plea | +0.014 |
horror | +0.014 |
Neg logit contributions | |
ju | -0.011 |
golf | -0.011 |
makeup | -0.011 |
grapes | -0.010 |
quiz | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
ously | +0.024 |
rum | +0.024 |
deeply | +0.021 |
aced | +0.021 |
urbed | +0.021 |
Neg logit contributions | |
makeup | -0.013 |
ju | -0.012 |
golf | -0.012 |
butter | -0.012 |
grapes | -0.012 |
Token | mystery |
Token ID | 10715 |
Feature activation | +7.17e-03 |
Pos logit contributions | |
ju | +0.035 |
golf | +0.033 |
makeup | +0.032 |
mons | +0.032 |
contest | +0.031 |
Neg logit contributions | |
ously | -0.045 |
rum | -0.044 |
despair | -0.042 |
deeply | -0.041 |
horror | -0.040 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.97e-04 |
Pos logit contributions | |
ju | +0.069 |
golf | +0.065 |
makeup | +0.064 |
mons | +0.064 |
grapes | +0.062 |
Neg logit contributions | |
ously | -0.068 |
rum | -0.067 |
deeply | -0.063 |
despair | -0.062 |
horror | -0.059 |
Token |
|
Token ID | 198 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
ju | +0.001 |
golf | +0.001 |
mixer | +0.001 |
mons | +0.001 |
makeup | +0.001 |
Neg logit contributions | |
rum | -0.001 |
despair | -0.001 |
ously | -0.001 |
horror | -0.001 |
aced | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
despair | +0.049 |
rum | +0.049 |
ously | +0.049 |
jaws | +0.048 |
horror | +0.048 |
Neg logit contributions | |
ju | -0.043 |
golf | -0.040 |
makeup | -0.038 |
mixer | -0.038 |
contest | -0.038 |
Token | The |
Token ID | 464 |
Feature activation | -7.75e-03 |
Pos logit contributions | |
ously | +0.047 |
rum | +0.046 |
despair | +0.044 |
deeply | +0.043 |
horror | +0.042 |
Neg logit contributions | |
ju | -0.037 |
golf | -0.035 |
makeup | -0.034 |
mons | -0.033 |
contest | -0.032 |
Token | little |
Token ID | 1310 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
ju | +0.012 |
mons | +0.012 |
golf | +0.012 |
atmeal | +0.012 |
makeup | +0.012 |
Neg logit contributions | |
ously | -0.013 |
rum | -0.012 |
despair | -0.012 |
deeply | -0.012 |
horror | -0.012 |
Token | girl |
Token ID | 2576 |
Feature activation | +5.58e-02 |
Pos logit contributions | |
ously | +0.020 |
despair | +0.018 |
horror | +0.017 |
rum | +0.017 |
deeply | +0.017 |
Neg logit contributions | |
ju | -0.021 |
atmeal | -0.021 |
mons | -0.021 |
golf | -0.020 |
contest | -0.020 |
Token | jumped |
Token ID | 11687 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
rum | +0.102 |
ously | +0.101 |
despair | +0.095 |
deeply | +0.093 |
horror | +0.091 |
Neg logit contributions | |
ju | -0.082 |
golf | -0.081 |
makeup | -0.077 |
mons | -0.075 |
paints | -0.075 |
Token | with |
Token ID | 351 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
ju | +0.052 |
golf | +0.052 |
makeup | +0.048 |
contest | +0.047 |
competition | +0.046 |
Neg logit contributions | |
ously | -0.066 |
rum | -0.065 |
despair | -0.061 |
horror | -0.058 |
deeply | -0.058 |
Token | excitement |
Token ID | 14067 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
rum | +0.071 |
ously | +0.071 |
deeply | +0.062 |
despair | +0.061 |
urbed | +0.057 |
Neg logit contributions | |
ju | -0.104 |
golf | -0.101 |
makeup | -0.100 |
grapes | -0.096 |
butter | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
rum | +0.092 |
ously | +0.091 |
despair | +0.088 |
horror | +0.083 |
deeply | +0.082 |
Neg logit contributions | |
ju | -0.088 |
golf | -0.087 |
makeup | -0.085 |
mons | -0.081 |
butter | -0.081 |
Token | She |
Token ID | 1375 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
rum | +0.061 |
ously | +0.059 |
despair | +0.058 |
deeply | +0.056 |
horror | +0.053 |
Neg logit contributions | |
ju | -0.053 |
golf | -0.050 |
mons | -0.049 |
makeup | -0.048 |
grapes | -0.047 |
Token | â„¢ |
Token ID | 8151 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
rum | +0.019 |
ously | +0.019 |
despair | +0.017 |
deeply | +0.017 |
horror | +0.016 |
Neg logit contributions | |
ju | -0.022 |
makeup | -0.021 |
golf | -0.020 |
grapes | -0.020 |
butter | -0.020 |
Token | s |
Token ID | 82 |
Feature activation | +4.11e-03 |
Pos logit contributions | |
ju | +0.172 |
mons | +0.167 |
golf | +0.163 |
contest | +0.160 |
makeup | +0.156 |
Neg logit contributions | |
despair | -0.243 |
ously | -0.241 |
rum | -0.236 |
horror | -0.232 |
deeply | -0.229 |
Token | pick |
Token ID | 2298 |
Feature activation | -5.85e-02 |
Pos logit contributions | |
ously | +0.008 |
rum | +0.008 |
despair | +0.007 |
deeply | +0.007 |
horror | +0.007 |
Neg logit contributions | |
ju | -0.006 |
golf | -0.005 |
makeup | -0.005 |
mons | -0.005 |
contest | -0.005 |
Token | some |
Token ID | 617 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
ju | +0.090 |
atmeal | +0.085 |
contest | +0.085 |
quiz | +0.085 |
mons | +0.084 |
Neg logit contributions | |
ously | -0.096 |
rum | -0.096 |
deeply | -0.089 |
sadly | -0.086 |
silently | -0.085 |
Token | flowers |
Token ID | 12734 |
Feature activation | +9.32e-03 |
Pos logit contributions | |
ju | +0.026 |
atmeal | +0.025 |
mons | +0.024 |
contest | +0.023 |
golf | +0.023 |
Neg logit contributions | |
rum | -0.045 |
ously | -0.044 |
deeply | -0.044 |
despair | -0.042 |
jaws | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
ously | +0.014 |
rum | +0.013 |
deeply | +0.013 |
despair | +0.012 |
silently | +0.012 |
Neg logit contributions | |
ju | -0.016 |
golf | -0.016 |
makeup | -0.015 |
mons | -0.015 |
atmeal | -0.015 |
Token | take |
Token ID | 1011 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
ously | +0.031 |
rum | +0.031 |
despair | +0.030 |
deeply | +0.030 |
silently | +0.029 |
Neg logit contributions | |
ju | -0.023 |
mons | -0.023 |
atmeal | -0.021 |
golf | -0.021 |
makeup | -0.020 |
Token | home |
Token ID | 1363 |
Feature activation | +6.74e-02 |
Pos logit contributions | |
ju | +0.039 |
mons | +0.036 |
golf | +0.036 |
atmeal | +0.036 |
contest | +0.035 |
Neg logit contributions | |
ously | -0.042 |
rum | -0.042 |
deeply | -0.040 |
despair | -0.039 |
silently | -0.037 |
Token | with |
Token ID | 351 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
ously | +0.112 |
rum | +0.110 |
despair | +0.103 |
deeply | +0.102 |
horror | +0.098 |
Neg logit contributions | |
ju | -0.111 |
golf | -0.107 |
makeup | -0.104 |
mons | -0.103 |
contest | -0.101 |
Token | us |
Token ID | 514 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
ously | +0.038 |
rum | +0.038 |
despair | +0.036 |
deeply | +0.035 |
horror | +0.034 |
Neg logit contributions | |
ju | -0.035 |
golf | -0.033 |
makeup | -0.032 |
mons | -0.032 |
contest | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
ously | +0.018 |
rum | +0.018 |
despair | +0.017 |
deeply | +0.016 |
horror | +0.016 |
Neg logit contributions | |
ju | -0.017 |
golf | -0.016 |
makeup | -0.016 |
mons | -0.016 |
contest | -0.015 |
Token | we |
Token ID | 356 |
Feature activation | -6.94e-02 |
Pos logit contributions | |
ju | +0.076 |
golf | +0.071 |
mons | +0.070 |
makeup | +0.067 |
grapes | +0.065 |
Neg logit contributions | |
rum | -0.124 |
ously | -0.122 |
despair | -0.121 |
deeply | -0.119 |
horror | -0.115 |
Token | go |
Token ID | 467 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
ju | +0.097 |
mons | +0.094 |
golf | +0.091 |
makeup | +0.090 |
grapes | +0.089 |
Neg logit contributions | |
ously | -0.125 |
rum | -0.123 |
despair | -0.121 |
deeply | -0.120 |
silently | -0.115 |
Token | inside |
Token ID | 2641 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
ju | +0.075 |
mons | +0.073 |
grapes | +0.069 |
makeup | +0.067 |
golf | +0.066 |
Neg logit contributions | |
ously | -0.067 |
deeply | -0.066 |
rum | -0.065 |
silently | -0.064 |
sadly | -0.063 |
Token | ?" |
Token ID | 1701 |
Feature activation | +5.22e-04 |
Pos logit contributions | |
ously | +0.120 |
rum | +0.117 |
despair | +0.111 |
deeply | +0.109 |
horror | +0.107 |
Neg logit contributions | |
ju | -0.075 |
golf | -0.072 |
makeup | -0.069 |
contest | -0.066 |
mons | -0.066 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.21e-03 |
Pos logit contributions | |
ously | +0.001 |
rum | +0.001 |
deeply | +0.001 |
aced | +0.001 |
despair | +0.001 |
Neg logit contributions | |
golf | -0.001 |
ju | -0.001 |
makeup | -0.001 |
grapes | -0.001 |
paints | -0.001 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
ju | +0.005 |
makeup | +0.004 |
golf | +0.004 |
butter | +0.004 |
paints | +0.004 |
Neg logit contributions | |
ously | -0.006 |
rum | -0.006 |
despair | -0.006 |
deeply | -0.005 |
horror | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +8.64e-03 |
Pos logit contributions | |
rum | +0.058 |
ously | +0.057 |
despair | +0.054 |
jaws | +0.052 |
urbed | +0.051 |
Neg logit contributions | |
makeup | -0.038 |
golf | -0.036 |
butter | -0.035 |
ju | -0.034 |
paints | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
rum | +0.015 |
ously | +0.015 |
despair | +0.014 |
aced | +0.014 |
urbed | +0.014 |
Neg logit contributions | |
golf | -0.014 |
ju | -0.013 |
makeup | -0.013 |
mons | -0.013 |
paints | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +9.09e-03 |
Pos logit contributions | |
ously | +0.026 |
despair | +0.026 |
jaws | +0.025 |
rum | +0.025 |
urbed | +0.024 |
Neg logit contributions | |
golf | -0.014 |
makeup | -0.013 |
paints | -0.013 |
Draw | -0.013 |
mixer | -0.013 |
Token | " |
Token ID | 1 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
ously | +0.018 |
rum | +0.018 |
despair | +0.017 |
deeply | +0.017 |
horror | +0.016 |
Neg logit contributions | |
ju | -0.012 |
golf | -0.012 |
makeup | -0.011 |
mons | -0.011 |
contest | -0.011 |
Token | No |
Token ID | 2949 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
atmeal | +0.231 |
ju | +0.219 |
makeup | +0.218 |
golf | +0.218 |
butter | +0.215 |
Neg logit contributions | |
ously | -0.272 |
despair | -0.242 |
deeply | -0.242 |
rum | -0.240 |
silently | -0.238 |
Token | . |
Token ID | 13 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
rum | +0.177 |
ously | +0.172 |
despair | +0.162 |
horror | +0.159 |
jaws | +0.156 |
Neg logit contributions | |
ju | -0.106 |
golf | -0.104 |
butter | -0.093 |
makeup | -0.093 |
quiz | -0.089 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
ously | +0.083 |
rum | +0.080 |
aced | +0.077 |
plea | +0.075 |
despair | +0.075 |
Neg logit contributions | |
ju | -0.038 |
golf | -0.037 |
gum | -0.032 |
How | -0.032 |
Draw | -0.031 |
Token | Once |
Token ID | 7454 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
golf | +0.060 |
ju | +0.060 |
makeup | +0.058 |
contest | +0.058 |
gum | +0.056 |
Neg logit contributions | |
ously | -0.070 |
rum | -0.069 |
despair | -0.066 |
horror | -0.063 |
deeply | -0.062 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
contest | +0.034 |
golf | +0.034 |
ju | +0.034 |
makeup | +0.034 |
mixer | +0.032 |
Neg logit contributions | |
rum | -0.033 |
ously | -0.031 |
despair | -0.031 |
heartbeat | -0.030 |
edly | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
ju | +0.022 |
golf | +0.021 |
makeup | +0.020 |
mons | +0.019 |
grapes | +0.019 |
Neg logit contributions | |
ously | -0.034 |
rum | -0.033 |
despair | -0.031 |
deeply | -0.031 |
horror | -0.030 |
Token | time |
Token ID | 640 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
ju | +0.076 |
golf | +0.074 |
makeup | +0.071 |
mons | +0.070 |
contest | +0.069 |
Neg logit contributions | |
ously | -0.091 |
rum | -0.089 |
despair | -0.084 |
deeply | -0.083 |
horror | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
golf | +0.055 |
ju | +0.054 |
makeup | +0.053 |
contest | +0.052 |
mons | +0.051 |
Neg logit contributions | |
ously | -0.071 |
rum | -0.070 |
despair | -0.067 |
deeply | -0.065 |
horror | -0.064 |
Token | there |
Token ID | 612 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
ju | +0.036 |
golf | +0.033 |
makeup | +0.031 |
mons | +0.030 |
contest | +0.028 |
Neg logit contributions | |
ously | -0.149 |
rum | -0.147 |
despair | -0.142 |
deeply | -0.140 |
horror | -0.137 |
Token | was |
Token ID | 373 |
Feature activation | -3.10e-03 |
Pos logit contributions | |
ju | +0.040 |
golf | +0.040 |
makeup | +0.039 |
contest | +0.037 |
gum | +0.036 |
Neg logit contributions | |
ously | -0.062 |
rum | -0.061 |
despair | -0.057 |
deeply | -0.057 |
heartbeat | -0.055 |
Token | a |
Token ID | 257 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
ju | +0.005 |
mons | +0.005 |
golf | +0.004 |
makeup | +0.004 |
atmeal | +0.004 |
Neg logit contributions | |
ously | -0.005 |
rum | -0.005 |
deeply | -0.005 |
despair | -0.005 |
horror | -0.005 |
Token | little |
Token ID | 1310 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
mons | +0.077 |
ju | +0.075 |
makeup | +0.072 |
atmeal | +0.072 |
newspapers | +0.070 |
Neg logit contributions | |
deeply | -0.047 |
ously | -0.045 |
horror | -0.045 |
despair | -0.045 |
rum | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
ju | +0.059 |
golf | +0.056 |
makeup | +0.054 |
mons | +0.054 |
contest | +0.052 |
Neg logit contributions | |
ously | -0.073 |
rum | -0.072 |
despair | -0.068 |
deeply | -0.067 |
horror | -0.065 |
Token | sand |
Token ID | 6450 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
ju | +0.026 |
golf | +0.024 |
mons | +0.024 |
makeup | +0.023 |
atmeal | +0.023 |
Neg logit contributions | |
ously | -0.034 |
rum | -0.033 |
despair | -0.032 |
deeply | -0.032 |
horror | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
ju | +0.059 |
golf | +0.056 |
makeup | +0.054 |
mons | +0.053 |
contest | +0.051 |
Neg logit contributions | |
ously | -0.087 |
rum | -0.085 |
despair | -0.080 |
deeply | -0.080 |
horror | -0.077 |
Token | and |
Token ID | 290 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
ju | +0.052 |
mons | +0.051 |
golf | +0.050 |
makeup | +0.049 |
atmeal | +0.049 |
Neg logit contributions | |
ously | -0.059 |
rum | -0.056 |
despair | -0.055 |
deeply | -0.054 |
horror | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
ju | +0.042 |
golf | +0.041 |
makeup | +0.040 |
mons | +0.039 |
contest | +0.038 |
Neg logit contributions | |
ously | -0.049 |
rum | -0.048 |
despair | -0.046 |
deeply | -0.045 |
horror | -0.043 |
Token | water |
Token ID | 1660 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
ju | +0.032 |
golf | +0.030 |
mons | +0.029 |
makeup | +0.029 |
contest | +0.028 |
Neg logit contributions | |
ously | -0.047 |
rum | -0.046 |
despair | -0.044 |
deeply | -0.044 |
horror | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +6.72e-03 |
Pos logit contributions | |
ju | +0.038 |
golf | +0.037 |
makeup | +0.036 |
mons | +0.035 |
contest | +0.034 |
Neg logit contributions | |
ously | -0.045 |
rum | -0.044 |
despair | -0.042 |
deeply | -0.041 |
horror | -0.040 |
Token | She |
Token ID | 1375 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
ously | +0.011 |
rum | +0.010 |
despair | +0.010 |
deeply | +0.009 |
aced | +0.009 |
Neg logit contributions | |
ju | -0.012 |
golf | -0.012 |
makeup | -0.011 |
mons | -0.011 |
grapes | -0.011 |
Token | dreamed |
Token ID | 27947 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
ously | +0.026 |
rum | +0.025 |
despair | +0.024 |
deeply | +0.023 |
horror | +0.023 |
Neg logit contributions | |
ju | -0.019 |
golf | -0.018 |
makeup | -0.018 |
mons | -0.017 |
contest | -0.017 |
Token | of |
Token ID | 286 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
ju | +0.040 |
golf | +0.039 |
makeup | +0.038 |
contest | +0.037 |
grapes | +0.037 |
Neg logit contributions | |
rum | -0.053 |
ously | -0.051 |
despair | -0.050 |
horror | -0.048 |
deeply | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
ously | +0.038 |
rum | +0.037 |
despair | +0.036 |
deeply | +0.035 |
horror | +0.034 |
Neg logit contributions | |
ju | -0.030 |
golf | -0.028 |
mons | -0.027 |
makeup | -0.027 |
atmeal | -0.026 |
Token | on |
Token ID | 319 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
ju | +0.029 |
mons | +0.026 |
atmeal | +0.026 |
grapes | +0.026 |
makeup | +0.025 |
Neg logit contributions | |
ously | -0.027 |
deeply | -0.025 |
rum | -0.024 |
silently | -0.023 |
sadly | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
ously | +0.093 |
rum | +0.090 |
despair | +0.084 |
deeply | +0.083 |
horror | +0.080 |
Neg logit contributions | |
ju | -0.070 |
golf | -0.069 |
makeup | -0.067 |
mons | -0.064 |
grapes | -0.063 |
Token | table |
Token ID | 3084 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
ju | +0.028 |
golf | +0.027 |
mons | +0.026 |
makeup | +0.026 |
atmeal | +0.025 |
Neg logit contributions | |
ously | -0.036 |
rum | -0.036 |
despair | -0.034 |
deeply | -0.034 |
horror | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
ju | +0.081 |
golf | +0.076 |
makeup | +0.075 |
mons | +0.074 |
atmeal | +0.073 |
Neg logit contributions | |
ously | -0.055 |
rum | -0.052 |
deeply | -0.048 |
despair | -0.048 |
horror | -0.046 |
Token | It |
Token ID | 632 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
ously | +0.027 |
rum | +0.025 |
despair | +0.024 |
deeply | +0.024 |
horror | +0.023 |
Neg logit contributions | |
ju | -0.044 |
atmeal | -0.043 |
golf | -0.042 |
makeup | -0.042 |
quiz | -0.042 |
Token | had |
Token ID | 550 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
ju | +0.054 |
golf | +0.052 |
mons | +0.051 |
atmeal | +0.050 |
makeup | +0.050 |
Neg logit contributions | |
ously | -0.041 |
rum | -0.040 |
deeply | -0.038 |
despair | -0.038 |
horror | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
ju | +0.045 |
golf | +0.042 |
makeup | +0.042 |
butter | +0.040 |
grapes | +0.040 |
Neg logit contributions | |
rum | -0.073 |
ously | -0.072 |
despair | -0.068 |
deeply | -0.067 |
horror | -0.065 |
Token | label |
Token ID | 6167 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
ju | +0.045 |
golf | +0.043 |
makeup | +0.041 |
mons | +0.040 |
contest | +0.039 |
Neg logit contributions | |
ously | -0.077 |
rum | -0.075 |
despair | -0.072 |
deeply | -0.070 |
horror | -0.069 |
Token | that |
Token ID | 326 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
ju | +0.058 |
atmeal | +0.054 |
mons | +0.048 |
golf | +0.047 |
grapes | +0.047 |
Neg logit contributions | |
ously | -0.063 |
deeply | -0.060 |
rum | -0.060 |
sadly | -0.056 |
silently | -0.055 |
Token | said |
Token ID | 531 |
Feature activation | -5.29e-03 |
Pos logit contributions | |
ju | +0.090 |
golf | +0.087 |
mons | +0.084 |
makeup | +0.083 |
atmeal | +0.082 |
Neg logit contributions | |
ously | -0.091 |
rum | -0.088 |
deeply | -0.085 |
despair | -0.085 |
horror | -0.081 |
Token | " |
Token ID | 366 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
ju | +0.010 |
atmeal | +0.010 |
golf | +0.009 |
mons | +0.009 |
grapes | +0.009 |
Neg logit contributions | |
ously | -0.008 |
rum | -0.007 |
deeply | -0.007 |
despair | -0.007 |
horror | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
ously | +0.023 |
rum | +0.022 |
despair | +0.020 |
deeply | +0.020 |
horror | +0.019 |
Neg logit contributions | |
ju | -0.031 |
golf | -0.029 |
makeup | -0.029 |
mons | -0.028 |
grapes | -0.028 |
Token | The |
Token ID | 383 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
ously | +0.023 |
rum | +0.022 |
despair | +0.021 |
deeply | +0.020 |
horror | +0.020 |
Neg logit contributions | |
ju | -0.030 |
golf | -0.029 |
makeup | -0.029 |
mons | -0.028 |
contest | -0.028 |
Token | box |
Token ID | 3091 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
atmeal | +0.037 |
mons | +0.033 |
ju | +0.033 |
iable | +0.032 |
please | +0.030 |
Neg logit contributions | |
horror | -0.032 |
despair | -0.032 |
deeply | -0.031 |
ously | -0.030 |
heart | -0.029 |
Token | is |
Token ID | 318 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
ju | +0.018 |
golf | +0.017 |
mons | +0.016 |
makeup | +0.016 |
atmeal | +0.016 |
Neg logit contributions | |
ously | -0.020 |
rum | -0.019 |
deeply | -0.018 |
despair | -0.018 |
horror | -0.018 |
Token | tight |
Token ID | 5381 |
Feature activation | +4.87e-02 |
Pos logit contributions | |
atmeal | +0.019 |
ju | +0.018 |
mons | +0.018 |
golf | +0.017 |
contest | +0.016 |
Neg logit contributions | |
ously | -0.022 |
deeply | -0.022 |
rum | -0.022 |
despair | -0.022 |
horror | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
ously | +0.069 |
rum | +0.066 |
deeply | +0.064 |
despair | +0.060 |
silently | +0.058 |
Neg logit contributions | |
ju | -0.094 |
golf | -0.087 |
mons | -0.087 |
atmeal | -0.086 |
makeup | -0.085 |
Token | hard |
Token ID | 1327 |
Feature activation | -1.25e-03 |
Pos logit contributions | |
ju | +0.033 |
golf | +0.032 |
atmeal | +0.031 |
mons | +0.031 |
makeup | +0.030 |
Neg logit contributions | |
ously | -0.044 |
rum | -0.044 |
deeply | -0.044 |
despair | -0.043 |
horror | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
ju | +0.002 |
golf | +0.002 |
makeup | +0.002 |
mons | +0.002 |
contest | +0.002 |
Neg logit contributions | |
ously | -0.002 |
rum | -0.002 |
despair | -0.002 |
deeply | -0.002 |
horror | -0.002 |
Token | open |
Token ID | 1280 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
ously | +0.029 |
rum | +0.029 |
despair | +0.028 |
deeply | +0.027 |
horror | +0.026 |
Neg logit contributions | |
ju | -0.022 |
golf | -0.021 |
mons | -0.020 |
makeup | -0.020 |
grapes | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
ously | +0.085 |
rum | +0.083 |
despair | +0.079 |
deeply | +0.078 |
horror | +0.075 |
Neg logit contributions | |
ju | -0.058 |
golf | -0.056 |
makeup | -0.054 |
mons | -0.053 |
contest | -0.051 |
Token | They |
Token ID | 1119 |
Feature activation | -4.72e-03 |
Pos logit contributions | |
ously | +0.031 |
rum | +0.030 |
despair | +0.028 |
deeply | +0.028 |
urbed | +0.026 |
Neg logit contributions | |
ju | -0.034 |
golf | -0.032 |
mons | -0.031 |
makeup | -0.031 |
grapes | -0.031 |
Token | proud |
Token ID | 6613 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
ju | +0.018 |
golf | +0.017 |
makeup | +0.016 |
mons | +0.016 |
swing | +0.016 |
Neg logit contributions | |
ously | -0.026 |
rum | -0.025 |
despair | -0.022 |
urbed | -0.022 |
deeply | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +8.32e-03 |
Pos logit contributions | |
ously | +0.065 |
rum | +0.064 |
despair | +0.060 |
deeply | +0.057 |
horror | +0.056 |
Neg logit contributions | |
golf | -0.060 |
ju | -0.058 |
makeup | -0.058 |
grapes | -0.056 |
contest | -0.056 |
Token | It |
Token ID | 632 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
ously | +0.019 |
rum | +0.019 |
urbed | +0.017 |
despair | +0.017 |
aced | +0.017 |
Neg logit contributions | |
golf | -0.009 |
grapes | -0.008 |
ju | -0.008 |
gum | -0.007 |
mons | -0.007 |
Token | thinks |
Token ID | 6834 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
ju | +0.036 |
golf | +0.033 |
makeup | +0.032 |
contest | +0.032 |
quiz | +0.032 |
Neg logit contributions | |
ously | -0.054 |
rum | -0.052 |
despair | -0.048 |
deeply | -0.047 |
aced | -0.045 |
Token | it |
Token ID | 340 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
ju | +0.017 |
golf | +0.016 |
makeup | +0.016 |
contest | +0.015 |
mons | +0.015 |
Neg logit contributions | |
ously | -0.025 |
rum | -0.024 |
despair | -0.023 |
deeply | -0.022 |
horror | -0.022 |
Token | has |
Token ID | 468 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
ju | +0.061 |
golf | +0.061 |
mons | +0.058 |
makeup | +0.057 |
grapes | +0.056 |
Neg logit contributions | |
rum | -0.056 |
ously | -0.055 |
deeply | -0.053 |
despair | -0.052 |
silently | -0.050 |
Token | won |
Token ID | 1839 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
ju | +0.065 |
golf | +0.064 |
makeup | +0.062 |
butter | +0.060 |
grapes | +0.060 |
Neg logit contributions | |
ously | -0.110 |
rum | -0.109 |
deeply | -0.102 |
despair | -0.100 |
horror | -0.097 |
Token | the |
Token ID | 262 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
ously | +0.043 |
rum | +0.040 |
despair | +0.039 |
horror | +0.038 |
deeply | +0.038 |
Neg logit contributions | |
golf | -0.031 |
makeup | -0.030 |
contest | -0.030 |
ju | -0.029 |
mons | -0.029 |
Token | game |
Token ID | 983 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
atmeal | +0.056 |
butter | +0.048 |
ju | +0.046 |
mons | +0.045 |
eating | +0.044 |
Neg logit contributions | |
horror | -0.059 |
despair | -0.058 |
fear | -0.057 |
plea | -0.054 |
pleas | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +3.67e-03 |
Pos logit contributions | |
ju | +0.065 |
golf | +0.062 |
makeup | +0.061 |
mons | +0.061 |
atmeal | +0.059 |
Neg logit contributions | |
ously | -0.051 |
rum | -0.049 |
despair | -0.046 |
deeply | -0.046 |
horror | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +2.66e-03 |
Pos logit contributions | |
ously | +0.008 |
rum | +0.008 |
despair | +0.007 |
urbed | +0.007 |
deeply | +0.007 |
Neg logit contributions | |
golf | -0.005 |
ju | -0.004 |
grapes | -0.004 |
mons | -0.004 |
gum | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
ously | +0.012 |
rum | +0.012 |
despair | +0.011 |
deeply | +0.011 |
horror | +0.011 |
Neg logit contributions | |
ju | -0.008 |
golf | -0.008 |
makeup | -0.008 |
mons | -0.008 |
contest | -0.007 |
Token | fire |
Token ID | 2046 |
Feature activation | -5.52e-03 |
Pos logit contributions | |
ju | +0.046 |
mons | +0.044 |
golf | +0.043 |
makeup | +0.043 |
atmeal | +0.041 |
Neg logit contributions | |
ously | -0.062 |
rum | -0.061 |
despair | -0.059 |
deeply | -0.058 |
horror | -0.057 |
Token | truck |
Token ID | 7779 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
ju | +0.007 |
golf | +0.007 |
mons | +0.007 |
makeup | +0.007 |
contest | +0.007 |
Neg logit contributions | |
ously | -0.011 |
rum | -0.010 |
despair | -0.010 |
deeply | -0.010 |
horror | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
ju | +0.080 |
makeup | +0.073 |
mons | +0.073 |
golf | +0.073 |
grapes | +0.072 |
Neg logit contributions | |
ously | -0.068 |
rum | -0.067 |
deeply | -0.063 |
despair | -0.062 |
horror | -0.060 |
Token | They |
Token ID | 1119 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
ously | +0.028 |
rum | +0.028 |
despair | +0.026 |
deeply | +0.025 |
horror | +0.024 |
Neg logit contributions | |
ju | -0.034 |
golf | -0.033 |
makeup | -0.032 |
mons | -0.032 |
grapes | -0.031 |
Token | wanted |
Token ID | 2227 |
Feature activation | -8.82e-02 |
Pos logit contributions | |
ju | +0.013 |
paints | +0.013 |
golf | +0.013 |
makeup | +0.012 |
swing | +0.012 |
Neg logit contributions | |
rum | -0.024 |
ously | -0.023 |
deeply | -0.022 |
despair | -0.021 |
horror | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -4.57e-03 |
Pos logit contributions | |
ju | +0.126 |
golf | +0.117 |
mons | +0.115 |
makeup | +0.114 |
atmeal | +0.113 |
Neg logit contributions | |
ously | -0.168 |
rum | -0.163 |
despair | -0.157 |
deeply | -0.155 |
horror | -0.151 |
Token | see |
Token ID | 766 |
Feature activation | -6.50e-02 |
Pos logit contributions | |
golf | +0.005 |
ju | +0.005 |
swing | +0.005 |
contest | +0.005 |
makeup | +0.004 |
Neg logit contributions | |
ously | -0.011 |
rum | -0.010 |
deeply | -0.010 |
despair | -0.009 |
urbed | -0.009 |
Token | it |
Token ID | 340 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
ju | +0.100 |
golf | +0.095 |
mons | +0.094 |
makeup | +0.093 |
contest | +0.091 |
Neg logit contributions | |
ously | -0.115 |
rum | -0.113 |
despair | -0.107 |
deeply | -0.106 |
horror | -0.103 |
Token | up |
Token ID | 510 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
ju | +0.024 |
golf | +0.023 |
makeup | +0.022 |
mons | +0.022 |
contest | +0.021 |
Neg logit contributions | |
ously | -0.028 |
rum | -0.027 |
despair | -0.026 |
deeply | -0.026 |
horror | -0.025 |
Token | close |
Token ID | 1969 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
ously | +0.143 |
rum | +0.141 |
despair | +0.128 |
horror | +0.126 |
deeply | +0.126 |
Neg logit contributions | |
golf | -0.116 |
ju | -0.114 |
contest | -0.107 |
makeup | -0.105 |
competition | -0.104 |
Token |
|
Token ID | 198 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
ously | +0.053 |
rum | +0.051 |
despair | +0.050 |
horror | +0.048 |
deeply | +0.048 |
Neg logit contributions | |
golf | -0.055 |
ju | -0.055 |
makeup | -0.054 |
mons | -0.051 |
gum | -0.050 |
Token | L |
Token ID | 43 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
ously | +0.057 |
rum | +0.056 |
despair | +0.053 |
deeply | +0.052 |
horror | +0.051 |
Neg logit contributions | |
ju | -0.040 |
golf | -0.037 |
mons | -0.037 |
makeup | -0.036 |
contest | -0.035 |
Token | ily |
Token ID | 813 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
makeup | +0.031 |
ju | +0.030 |
golf | +0.030 |
mons | +0.029 |
grapes | +0.028 |
Neg logit contributions | |
ously | -0.064 |
rum | -0.062 |
despair | -0.062 |
deeply | -0.060 |
horror | -0.059 |
Token | replied |
Token ID | 8712 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
ju | +0.035 |
golf | +0.033 |
makeup | +0.032 |
mons | +0.032 |
contest | +0.031 |
Neg logit contributions | |
ously | -0.033 |
rum | -0.032 |
despair | -0.030 |
deeply | -0.030 |
horror | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
ously | +0.039 |
rum | +0.039 |
despair | +0.037 |
deeply | +0.036 |
horror | +0.035 |
Neg logit contributions | |
ju | -0.034 |
golf | -0.033 |
makeup | -0.033 |
mons | -0.032 |
contest | -0.031 |
Token | " |
Token ID | 366 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
ju | +0.050 |
golf | +0.048 |
makeup | +0.047 |
mons | +0.045 |
grapes | +0.044 |
Neg logit contributions | |
rum | -0.066 |
ously | -0.065 |
despair | -0.063 |
horror | -0.060 |
deeply | -0.059 |
Token | I |
Token ID | 40 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
atmeal | +0.194 |
contest | +0.190 |
golf | +0.189 |
quiz | +0.188 |
grapes | +0.180 |
Neg logit contributions | |
ously | -0.250 |
despair | -0.231 |
rum | -0.226 |
deeply | -0.215 |
silently | -0.214 |
Token | fell |
Token ID | 3214 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
ju | +0.083 |
golf | +0.080 |
makeup | +0.078 |
mons | +0.078 |
grapes | +0.076 |
Neg logit contributions | |
ously | -0.083 |
rum | -0.081 |
despair | -0.076 |
deeply | -0.075 |
horror | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | -5.36e-03 |
Pos logit contributions | |
ously | +0.025 |
rum | +0.024 |
despair | +0.024 |
horror | +0.023 |
deeply | +0.023 |
Neg logit contributions | |
makeup | -0.020 |
ju | -0.020 |
golf | -0.020 |
contest | -0.019 |
butter | -0.018 |
Token | got |
Token ID | 1392 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
ju | +0.007 |
golf | +0.007 |
mons | +0.007 |
makeup | +0.007 |
grapes | +0.006 |
Neg logit contributions | |
ously | -0.010 |
rum | -0.010 |
despair | -0.010 |
deeply | -0.010 |
horror | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
ju | +0.073 |
golf | +0.070 |
makeup | +0.068 |
mons | +0.067 |
contest | +0.065 |
Neg logit contributions | |
ously | -0.097 |
rum | -0.095 |
despair | -0.090 |
deeply | -0.089 |
horror | -0.086 |
Token | says |
Token ID | 1139 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
ously | +0.006 |
rum | +0.006 |
deeply | +0.005 |
despair | +0.005 |
horror | +0.005 |
Neg logit contributions | |
ju | -0.006 |
golf | -0.006 |
mons | -0.006 |
makeup | -0.005 |
atmeal | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
ju | +0.035 |
mons | +0.032 |
golf | +0.032 |
atmeal | +0.032 |
makeup | +0.031 |
Neg logit contributions | |
ously | -0.029 |
rum | -0.027 |
deeply | -0.026 |
despair | -0.025 |
horror | -0.024 |
Token | Anna |
Token ID | 11735 |
Feature activation | -1.00e-03 |
Pos logit contributions | |
makeup | +0.015 |
golf | +0.015 |
quiz | +0.015 |
contest | +0.014 |
grapes | +0.014 |
Neg logit contributions | |
ously | -0.025 |
rum | -0.024 |
aced | -0.022 |
despair | -0.022 |
deeply | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
ju | +0.002 |
golf | +0.002 |
makeup | +0.002 |
mons | +0.001 |
contest | +0.001 |
Neg logit contributions | |
ously | -0.002 |
rum | -0.002 |
despair | -0.002 |
deeply | -0.002 |
horror | -0.002 |
Token | " |
Token ID | 366 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
ju | +0.089 |
golf | +0.087 |
makeup | +0.084 |
mons | +0.082 |
contest | +0.081 |
Neg logit contributions | |
ously | -0.105 |
rum | -0.103 |
despair | -0.098 |
deeply | -0.096 |
horror | -0.093 |
Token | Why |
Token ID | 5195 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
ju | +0.182 |
golf | +0.176 |
makeup | +0.170 |
grapes | +0.165 |
atmeal | +0.163 |
Neg logit contributions | |
ously | -0.311 |
rum | -0.302 |
despair | -0.291 |
silently | -0.272 |
horror | -0.271 |
Token | don |
Token ID | 836 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
ju | +0.135 |
golf | +0.123 |
mons | +0.121 |
makeup | +0.118 |
grapes | +0.116 |
Neg logit contributions | |
ously | -0.185 |
rum | -0.184 |
deeply | -0.171 |
despair | -0.171 |
horror | -0.165 |
Token | 't |
Token ID | 470 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ju | +0.134 |
mons | +0.124 |
makeup | +0.121 |
grapes | +0.118 |
atmeal | +0.118 |
Neg logit contributions | |
rum | -0.079 |
deeply | -0.077 |
ously | -0.071 |
despair | -0.069 |
horror | -0.066 |
Token | we |
Token ID | 356 |
Feature activation | -5.72e-02 |
Pos logit contributions | |
rum | +0.025 |
ously | +0.025 |
despair | +0.024 |
deeply | +0.023 |
horror | +0.023 |
Neg logit contributions | |
ju | -0.013 |
mons | -0.012 |
golf | -0.011 |
makeup | -0.011 |
grapes | -0.011 |
Token | share |
Token ID | 2648 |
Feature activation | -2.31e-03 |
Pos logit contributions | |
ju | +0.083 |
mons | +0.077 |
golf | +0.076 |
makeup | +0.074 |
atmeal | +0.073 |
Neg logit contributions | |
ously | -0.105 |
rum | -0.103 |
despair | -0.100 |
deeply | -0.098 |
horror | -0.095 |
Token | our |
Token ID | 674 |
Feature activation | -5.99e-03 |
Pos logit contributions | |
ju | +0.004 |
atmeal | +0.004 |
mons | +0.004 |
contest | +0.004 |
golf | +0.004 |
Neg logit contributions | |
rum | -0.003 |
ously | -0.003 |
deeply | -0.003 |
despair | -0.003 |
sadly | -0.003 |
Token | " |
Token ID | 366 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
rum | +0.013 |
ously | +0.013 |
despair | +0.012 |
aced | +0.012 |
deeply | +0.012 |
Neg logit contributions | |
ju | -0.009 |
golf | -0.009 |
makeup | -0.008 |
grapes | -0.008 |
newspapers | -0.008 |
Token | Hello |
Token ID | 15496 |
Feature activation | -3.54e-03 |
Pos logit contributions | |
ju | +0.153 |
golf | +0.149 |
mons | +0.147 |
makeup | +0.144 |
contest | +0.140 |
Neg logit contributions | |
ously | -0.250 |
rum | -0.240 |
despair | -0.233 |
deeply | -0.227 |
horror | -0.219 |
Token | , |
Token ID | 11 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
ju | +0.008 |
mons | +0.008 |
atmeal | +0.007 |
golf | +0.007 |
Draw | +0.007 |
Neg logit contributions | |
rum | -0.003 |
ously | -0.003 |
sadly | -0.003 |
deeply | -0.003 |
despair | -0.003 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
mons | +0.117 |
golf | +0.110 |
ju | +0.110 |
atmeal | +0.110 |
paints | +0.104 |
Neg logit contributions | |
ously | -0.096 |
deeply | -0.095 |
despair | -0.095 |
rum | -0.094 |
horror | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
ju | +0.075 |
golf | +0.072 |
mons | +0.071 |
makeup | +0.070 |
contest | +0.070 |
Neg logit contributions | |
ously | -0.050 |
rum | -0.049 |
deeply | -0.044 |
despair | -0.043 |
horror | -0.041 |
Token | What |
Token ID | 1867 |
Feature activation | -9.43e-02 |
Pos logit contributions | |
ju | +0.023 |
golf | +0.022 |
mons | +0.021 |
makeup | +0.021 |
butter | +0.021 |
Neg logit contributions | |
ously | -0.031 |
rum | -0.031 |
despair | -0.029 |
deeply | -0.028 |
horror | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
ju | +0.172 |
mons | +0.167 |
golf | +0.162 |
atmeal | +0.154 |
coat | +0.150 |
Neg logit contributions | |
rum | -0.128 |
ously | -0.126 |
despair | -0.117 |
aced | -0.113 |
deeply | -0.113 |
Token | nice |
Token ID | 3621 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
mons | +0.060 |
ju | +0.053 |
atmeal | +0.052 |
paints | +0.050 |
pens | +0.048 |
Neg logit contributions | |
horror | -0.051 |
despair | -0.051 |
deeply | -0.051 |
rum | -0.048 |
heartbeat | -0.048 |
Token | surprise |
Token ID | 5975 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
ju | +0.032 |
mons | +0.030 |
golf | +0.030 |
makeup | +0.029 |
atmeal | +0.029 |
Neg logit contributions | |
ously | -0.040 |
rum | -0.039 |
despair | -0.037 |
deeply | -0.037 |
horror | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -4.43e-03 |
Pos logit contributions | |
ju | +0.039 |
golf | +0.038 |
mons | +0.036 |
makeup | +0.036 |
grapes | +0.035 |
Neg logit contributions | |
ously | -0.038 |
rum | -0.038 |
deeply | -0.035 |
despair | -0.035 |
horror | -0.033 |
Token | Are |
Token ID | 4231 |
Feature activation | -6.02e-02 |
Pos logit contributions | |
ju | +0.006 |
golf | +0.006 |
makeup | +0.005 |
mons | +0.005 |
contest | +0.005 |
Neg logit contributions | |
ously | -0.009 |
rum | -0.009 |
despair | -0.008 |
deeply | -0.008 |
horror | -0.008 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
rum | +0.027 |
ously | +0.025 |
despair | +0.025 |
horror | +0.025 |
heartbeat | +0.024 |
Neg logit contributions | |
ju | -0.017 |
golf | -0.017 |
competition | -0.015 |
contest | -0.015 |
swing | -0.014 |
Token | very |
Token ID | 845 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
rum | +0.025 |
ously | +0.024 |
despair | +0.022 |
horror | +0.021 |
heartbeat | +0.021 |
Neg logit contributions | |
ju | -0.013 |
golf | -0.012 |
makeup | -0.011 |
competition | -0.010 |
contest | -0.010 |
Token | excited |
Token ID | 6568 |
Feature activation | +1.85e-03 |
Pos logit contributions | |
ously | +0.109 |
rum | +0.105 |
despair | +0.094 |
aced | +0.093 |
urbed | +0.092 |
Neg logit contributions | |
ju | -0.057 |
golf | -0.057 |
makeup | -0.048 |
contest | -0.046 |
competition | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
ously | +0.003 |
rum | +0.003 |
despair | +0.003 |
deeply | +0.003 |
horror | +0.003 |
Neg logit contributions | |
ju | -0.003 |
golf | -0.003 |
makeup | -0.003 |
mons | -0.003 |
contest | -0.003 |
Token | " |
Token ID | 366 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
rum | +0.070 |
ously | +0.069 |
despair | +0.065 |
horror | +0.062 |
deeply | +0.061 |
Neg logit contributions | |
ju | -0.059 |
golf | -0.058 |
makeup | -0.053 |
mons | -0.052 |
gum | -0.051 |
Token | Can |
Token ID | 6090 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
makeup | +0.139 |
golf | +0.135 |
mons | +0.135 |
ju | +0.134 |
grapes | +0.134 |
Neg logit contributions | |
ously | -0.229 |
rum | -0.219 |
despair | -0.214 |
deeply | -0.202 |
silently | -0.200 |
Token | I |
Token ID | 314 |
Feature activation | -6.50e-02 |
Pos logit contributions | |
ju | +0.133 |
mons | +0.128 |
makeup | +0.128 |
atmeal | +0.124 |
grapes | +0.123 |
Neg logit contributions | |
ously | -0.187 |
rum | -0.182 |
despair | -0.182 |
deeply | -0.178 |
sadly | -0.175 |
Token | come |
Token ID | 1282 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
ju | +0.092 |
golf | +0.087 |
makeup | +0.086 |
mons | +0.086 |
grapes | +0.083 |
Neg logit contributions | |
ously | -0.122 |
rum | -0.120 |
despair | -0.114 |
deeply | -0.113 |
horror | -0.109 |
Token | with |
Token ID | 351 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
ju | +0.044 |
mons | +0.041 |
makeup | +0.040 |
grapes | +0.040 |
golf | +0.039 |
Neg logit contributions | |
rum | -0.044 |
ously | -0.044 |
deeply | -0.041 |
despair | -0.041 |
silently | -0.040 |
Token | you |
Token ID | 345 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
ously | +0.044 |
rum | +0.044 |
despair | +0.042 |
deeply | +0.042 |
urbed | +0.040 |
Neg logit contributions | |
ju | -0.026 |
mons | -0.024 |
golf | -0.023 |
atmeal | -0.023 |
makeup | -0.023 |
Token | ?" |
Token ID | 1701 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
ju | +0.021 |
mons | +0.021 |
makeup | +0.020 |
grapes | +0.019 |
golf | +0.019 |
Neg logit contributions | |
ously | -0.039 |
rum | -0.038 |
deeply | -0.038 |
despair | -0.037 |
silently | -0.037 |