Token | closer |
Token ID | 5699 |
Feature activation | +7.71e-01 |
Pos logit contributions | |
Lu | +0.073 |
Ann | +0.067 |
Lu | +0.065 |
angels | +0.065 |
pastor | +0.063 |
Neg logit contributions | |
ret | -0.044 |
orse | -0.040 |
ets | -0.040 |
peanut | -0.038 |
sel | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
Lu | +0.042 |
Ann | +0.035 |
Lu | +0.035 |
firefighters | +0.035 |
angels | +0.035 |
Neg logit contributions | |
ret | -0.025 |
orse | -0.025 |
orses | -0.022 |
ets | -0.021 |
peanut | -0.020 |
Token | closer |
Token ID | 5699 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
ret | +0.026 |
orse | +0.026 |
orses | +0.025 |
peanut | +0.023 |
ets | +0.022 |
Neg logit contributions | |
Lu | -0.025 |
angels | -0.019 |
Ann | -0.019 |
Lu | -0.019 |
firefighters | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
Lu | +0.013 |
angels | +0.010 |
Ann | +0.010 |
Lu | +0.010 |
firefighters | +0.010 |
Neg logit contributions | |
ret | -0.009 |
orse | -0.009 |
orses | -0.008 |
peanut | -0.008 |
ets | -0.007 |
Token | but |
Token ID | 475 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
ret | +0.020 |
orse | +0.020 |
orses | +0.019 |
peanut | +0.018 |
ets | +0.017 |
Neg logit contributions | |
Lu | -0.021 |
angels | -0.017 |
Ann | -0.016 |
firefighters | -0.016 |
Lu | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
Lu | +0.013 |
angels | +0.011 |
Ann | +0.011 |
Lu | +0.011 |
firefighters | +0.011 |
Neg logit contributions | |
ret | -0.009 |
orse | -0.008 |
orses | -0.008 |
peanut | -0.007 |
ets | -0.007 |
Token | fog |
Token ID | 19558 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
ussy | +0.175 |
whe | +0.175 |
ators | +0.165 |
jack | +0.160 |
beh | +0.160 |
Neg logit contributions | |
holes | -0.099 |
middle | -0.086 |
long | -0.086 |
it | -0.083 |
shell | -0.083 |
Token | was |
Token ID | 373 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
Lu | +0.035 |
angels | +0.029 |
youths | +0.029 |
Lu | +0.028 |
Ann | +0.028 |
Neg logit contributions | |
ret | -0.025 |
orse | -0.025 |
orses | -0.022 |
ork | -0.021 |
peanut | -0.021 |
Token | too |
Token ID | 1165 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
Lu | +0.108 |
youths | +0.096 |
Ann | +0.094 |
angels | +0.094 |
Taxi | +0.092 |
Neg logit contributions | |
orse | -0.066 |
peanut | -0.065 |
ret | -0.062 |
ifying | -0.060 |
accidentally | -0.058 |
Token | thick |
Token ID | 6546 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
Lu | +0.062 |
angels | +0.055 |
Ann | +0.052 |
youths | +0.051 |
ators | +0.050 |
Neg logit contributions | |
ret | -0.035 |
orse | -0.034 |
peanut | -0.031 |
orses | -0.030 |
ork | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
Lu | +0.015 |
angels | +0.012 |
Ann | +0.012 |
Lu | +0.012 |
firefighters | +0.012 |
Neg logit contributions | |
ret | -0.011 |
orse | -0.011 |
orses | -0.010 |
ets | -0.009 |
peanut | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
ret | +0.005 |
orse | +0.004 |
orses | +0.004 |
peanut | +0.004 |
ets | +0.004 |
Neg logit contributions | |
Lu | -0.005 |
angels | -0.004 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
Lu | +0.009 |
angels | +0.007 |
Ann | +0.007 |
Lu | +0.007 |
firefighters | +0.007 |
Neg logit contributions | |
ret | -0.006 |
orse | -0.006 |
orses | -0.006 |
peanut | -0.005 |
ets | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
Lu | +0.035 |
angels | +0.031 |
firefighters | +0.030 |
youths | +0.029 |
Ann | +0.028 |
Neg logit contributions | |
ret | -0.023 |
orse | -0.023 |
orses | -0.021 |
ork | -0.019 |
peanut | -0.018 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
Lu | +0.012 |
angels | +0.010 |
firefighters | +0.009 |
Ann | +0.009 |
youths | +0.009 |
Neg logit contributions | |
ret | -0.008 |
orse | -0.008 |
orses | -0.007 |
peanut | -0.007 |
ets | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
Lu | +0.045 |
Ann | +0.039 |
Lu | +0.038 |
angels | +0.038 |
co | +0.038 |
Neg logit contributions | |
ret | -0.022 |
orse | -0.021 |
orses | -0.019 |
peanut | -0.017 |
ork | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
Lu | +0.046 |
Ann | +0.040 |
co | +0.040 |
angels | +0.039 |
Lu | +0.039 |
Neg logit contributions | |
ret | -0.026 |
orse | -0.025 |
orses | -0.022 |
peanut | -0.022 |
sel | -0.020 |
Token | fog |
Token ID | 19558 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
ussy | +0.178 |
whe | +0.174 |
beh | +0.165 |
Lu | +0.162 |
ators | +0.160 |
Neg logit contributions | |
holes | -0.085 |
long | -0.080 |
sudden | -0.078 |
middle | -0.078 |
shell | -0.077 |
Token | began |
Token ID | 2540 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
Lu | +0.016 |
angels | +0.013 |
Ann | +0.013 |
Lu | +0.013 |
youths | +0.013 |
Neg logit contributions | |
orse | -0.011 |
ret | -0.011 |
orses | -0.010 |
peanut | -0.009 |
ets | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
Lu | +0.029 |
Ann | +0.024 |
Lu | +0.024 |
angels | +0.024 |
youths | +0.023 |
Neg logit contributions | |
orse | -0.022 |
ret | -0.021 |
orses | -0.020 |
peanut | -0.019 |
ork | -0.018 |
Token | move |
Token ID | 1445 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Lu | +0.083 |
Ann | +0.072 |
youths | +0.072 |
Lu | +0.072 |
Barber | +0.069 |
Neg logit contributions | |
orse | -0.052 |
ret | -0.047 |
peanut | -0.046 |
orses | -0.045 |
sel | -0.040 |
Token | away |
Token ID | 1497 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Lu | +0.055 |
angels | +0.047 |
Ann | +0.046 |
Lu | +0.045 |
Barber | +0.045 |
Neg logit contributions | |
ret | -0.033 |
orse | -0.030 |
ets | -0.028 |
orses | -0.028 |
ork | -0.027 |
Token | thought |
Token ID | 1807 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
ret | +0.002 |
orse | +0.002 |
orses | +0.001 |
peanut | +0.001 |
ets | +0.001 |
Neg logit contributions | |
Lu | -0.002 |
angels | -0.002 |
Ann | -0.001 |
Lu | -0.001 |
firefighters | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
ret | +0.024 |
orse | +0.023 |
orses | +0.022 |
peanut | +0.021 |
ets | +0.020 |
Neg logit contributions | |
Lu | -0.026 |
angels | -0.021 |
firefighters | -0.020 |
Lu | -0.019 |
Ann | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
ret | +0.025 |
orse | +0.025 |
orses | +0.023 |
peanut | +0.022 |
ocket | +0.022 |
Neg logit contributions | |
Lu | -0.026 |
angels | -0.021 |
firefighters | -0.020 |
Ann | -0.020 |
Lu | -0.020 |
Token | then |
Token ID | 788 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
ret | +0.013 |
orse | +0.012 |
orses | +0.012 |
ets | +0.011 |
peanut | +0.011 |
Neg logit contributions | |
Lu | -0.013 |
angels | -0.011 |
firefighters | -0.010 |
Ann | -0.010 |
Lu | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
Lu | +0.010 |
angels | +0.008 |
Ann | +0.008 |
Lu | +0.008 |
firefighters | +0.008 |
Neg logit contributions | |
ret | -0.006 |
orse | -0.006 |
orses | -0.006 |
peanut | -0.005 |
ets | -0.005 |
Token | organized |
Token ID | 8389 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
Lu | +0.140 |
ators | +0.119 |
Ann | +0.116 |
Stev | +0.115 |
lo | +0.113 |
Neg logit contributions | |
peanut | -0.088 |
tomato | -0.079 |
orse | -0.076 |
pen | -0.074 |
mouth | -0.072 |
Token | friend |
Token ID | 1545 |
Feature activation | -6.25e-01 |
Pos logit contributions | |
Lu | +0.153 |
Michael | +0.153 |
Mark | +0.152 |
mercy | +0.150 |
Ann | +0.150 |
Neg logit contributions | |
's | -0.102 |
tomato | -0.097 |
mouth | -0.092 |
peanut | -0.090 |
picture | -0.089 |
Token | had |
Token ID | 550 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
ret | +0.025 |
orse | +0.024 |
orses | +0.023 |
ets | +0.021 |
peanut | +0.020 |
Neg logit contributions | |
Lu | -0.031 |
angels | -0.025 |
Lu | -0.025 |
Ann | -0.025 |
firefighters | -0.025 |
Token | an |
Token ID | 281 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
ret | +0.003 |
orse | +0.003 |
orses | +0.003 |
peanut | +0.003 |
ets | +0.003 |
Neg logit contributions | |
Lu | -0.004 |
angels | -0.003 |
Ann | -0.003 |
Lu | -0.003 |
firefighters | -0.003 |
Token | idea |
Token ID | 2126 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
Lu | +0.080 |
Ann | +0.072 |
angels | +0.069 |
Lu | +0.068 |
Michael | +0.063 |
Neg logit contributions | |
ret | -0.076 |
orse | -0.075 |
peanut | -0.071 |
ets | -0.067 |
ocket | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
Lu | +0.012 |
angels | +0.010 |
Ann | +0.009 |
Lu | +0.009 |
firefighters | +0.009 |
Neg logit contributions | |
ret | -0.009 |
orse | -0.009 |
orses | -0.008 |
peanut | -0.008 |
ets | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
ret | +0.034 |
orse | +0.033 |
orses | +0.031 |
ets | +0.031 |
peanut | +0.031 |
Neg logit contributions | |
Lu | -0.031 |
Ann | -0.025 |
Lu | -0.025 |
angels | -0.024 |
youths | -0.023 |
Token | Ben |
Token ID | 11696 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
ret | +0.045 |
orse | +0.042 |
ocket | +0.038 |
orses | +0.038 |
peanut | +0.037 |
Neg logit contributions | |
Lu | -0.045 |
Lu | -0.039 |
Ann | -0.038 |
youths | -0.036 |
angels | -0.035 |
Token | agreed |
Token ID | 4987 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
Lu | +0.086 |
angels | +0.079 |
firefighters | +0.075 |
Taxi | +0.074 |
taxis | +0.073 |
Neg logit contributions | |
orse | -0.060 |
ret | -0.056 |
orses | -0.051 |
peanut | -0.050 |
's | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
ret | +0.001 |
orse | +0.001 |
orses | +0.001 |
peanut | +0.001 |
ets | +0.001 |
Neg logit contributions | |
Lu | -0.001 |
angels | -0.001 |
Ann | -0.001 |
Lu | -0.001 |
firefighters | -0.001 |
Token | soon |
Token ID | 2582 |
Feature activation | -7.15e-01 |
Pos logit contributions | |
Lu | +0.006 |
angels | +0.005 |
Ann | +0.005 |
Lu | +0.005 |
firefighters | +0.005 |
Neg logit contributions | |
ret | -0.003 |
orse | -0.003 |
orses | -0.003 |
peanut | -0.003 |
ets | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
ret | +0.027 |
orse | +0.026 |
orses | +0.025 |
ets | +0.022 |
ocket | +0.022 |
Neg logit contributions | |
Lu | -0.037 |
angels | -0.030 |
Ann | -0.029 |
Lu | -0.029 |
youths | -0.028 |
Token | two |
Token ID | 734 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
Lu | +0.146 |
Ann | +0.142 |
Kitty | +0.136 |
Tit | +0.134 |
Lu | +0.134 |
Neg logit contributions | |
picture | -0.106 |
drawing | -0.103 |
's | -0.099 |
ink | -0.097 |
face | -0.096 |
Token | of |
Token ID | 286 |
Feature activation | +9.13e-01 |
Pos logit contributions | |
Lu | +0.144 |
iger | +0.142 |
Amen | +0.133 |
Johnson | +0.132 |
jack | +0.132 |
Neg logit contributions | |
's | -0.097 |
s | -0.095 |
orse | -0.091 |
orses | -0.090 |
bs | -0.089 |
Token | them |
Token ID | 606 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
Lu | +0.050 |
jack | +0.043 |
angels | +0.043 |
attendant | +0.043 |
Taxi | +0.042 |
Neg logit contributions | |
orse | -0.028 |
ret | -0.028 |
orses | -0.026 |
peanut | -0.024 |
notebooks | -0.022 |
Token | were |
Token ID | 547 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Lu | +0.067 |
angels | +0.058 |
Ann | +0.057 |
Lu | +0.057 |
youths | +0.056 |
Neg logit contributions | |
orse | -0.042 |
ret | -0.040 |
orses | -0.039 |
ets | -0.037 |
peanut | -0.034 |
Token | walking |
Token ID | 6155 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
Lu | +0.075 |
jack | +0.066 |
Ann | +0.066 |
Lu | +0.066 |
attendant | +0.063 |
Neg logit contributions | |
orse | -0.056 |
orses | -0.053 |
ret | -0.047 |
peanut | -0.045 |
ets | -0.045 |
Token | ! |
Token ID | 0 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
ret | +0.011 |
orse | +0.011 |
orses | +0.010 |
peanut | +0.009 |
ets | +0.009 |
Neg logit contributions | |
Lu | -0.013 |
angels | -0.011 |
Ann | -0.010 |
Lu | -0.010 |
firefighters | -0.010 |
Token | He |
Token ID | 679 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ret | +0.009 |
orse | +0.009 |
orses | +0.009 |
peanut | +0.008 |
ets | +0.008 |
Neg logit contributions | |
Lu | -0.010 |
angels | -0.008 |
Ann | -0.008 |
Lu | -0.008 |
firefighters | -0.008 |
Token | even |
Token ID | 772 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
Lu | +0.058 |
angels | +0.051 |
firefighters | +0.049 |
Taxi | +0.049 |
attendant | +0.048 |
Neg logit contributions | |
orse | -0.037 |
ret | -0.034 |
orses | -0.034 |
peanut | -0.031 |
ets | -0.030 |
Token | showed |
Token ID | 3751 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
Lu | +0.014 |
angels | +0.012 |
Ann | +0.011 |
Lu | +0.011 |
firefighters | +0.011 |
Neg logit contributions | |
orse | -0.010 |
ret | -0.010 |
orses | -0.009 |
peanut | -0.009 |
ets | -0.008 |
Token | his |
Token ID | 465 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
orse | +0.023 |
ret | +0.023 |
orses | +0.022 |
peanut | +0.020 |
ets | +0.019 |
Neg logit contributions | |
Lu | -0.022 |
angels | -0.017 |
Ann | -0.017 |
firefighters | -0.017 |
Lu | -0.016 |
Token | new |
Token ID | 649 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
Lu | +0.107 |
Lu | +0.093 |
Ann | +0.088 |
ators | +0.088 |
guard | +0.085 |
Neg logit contributions | |
orse | -0.079 |
mouth | -0.077 |
orses | -0.076 |
peanut | -0.073 |
ret | -0.072 |
Token | friends |
Token ID | 2460 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
Lu | +0.167 |
Ann | +0.154 |
Lu | +0.150 |
K | +0.144 |
guard | +0.142 |
Neg logit contributions | |
drawing | -0.112 |
mouth | -0.111 |
tomato | -0.110 |
's | -0.109 |
orse | -0.103 |
Token | how |
Token ID | 703 |
Feature activation | +7.30e-01 |
Pos logit contributions | |
ret | +0.027 |
orse | +0.027 |
orses | +0.025 |
peanut | +0.023 |
ets | +0.023 |
Neg logit contributions | |
Lu | -0.025 |
angels | -0.019 |
Ann | -0.018 |
Lu | -0.018 |
youths | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
Lu | +0.037 |
Ann | +0.031 |
angels | +0.030 |
attendant | +0.030 |
Lu | +0.030 |
Neg logit contributions | |
orse | -0.027 |
ret | -0.027 |
orses | -0.025 |
peanut | -0.022 |
ets | -0.022 |
Token | do |
Token ID | 466 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
Lu | +0.139 |
firefighters | +0.127 |
angels | +0.126 |
Taxi | +0.124 |
attendant | +0.124 |
Neg logit contributions | |
orse | -0.078 |
orses | -0.069 |
ret | -0.063 |
ets | -0.063 |
scored | -0.054 |
Token | them |
Token ID | 606 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
Lu | +0.021 |
angels | +0.017 |
Ann | +0.017 |
Lu | +0.017 |
firefighters | +0.017 |
Neg logit contributions | |
ret | -0.013 |
orse | -0.013 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Lu | +0.041 |
angels | +0.036 |
youths | +0.034 |
Ann | +0.034 |
firefighters | +0.034 |
Neg logit contributions | |
orse | -0.028 |
ret | -0.028 |
orses | -0.025 |
peanut | -0.024 |
sel | -0.022 |
Token | surprised |
Token ID | 6655 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
Lu | +0.082 |
angels | +0.074 |
ators | +0.073 |
Ann | +0.072 |
lo | +0.072 |
Neg logit contributions | |
orse | -0.047 |
ret | -0.039 |
orses | -0.038 |
peanut | -0.038 |
's | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
ret | +0.009 |
orse | +0.009 |
orses | +0.009 |
peanut | +0.008 |
ets | +0.008 |
Neg logit contributions | |
Lu | -0.010 |
angels | -0.008 |
Lu | -0.008 |
Ann | -0.008 |
firefighters | -0.008 |
Token | wondered |
Token ID | 14028 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ret | +0.009 |
orse | +0.008 |
orses | +0.008 |
peanut | +0.007 |
ets | +0.007 |
Neg logit contributions | |
Lu | -0.009 |
angels | -0.007 |
Ann | -0.007 |
Lu | -0.007 |
firefighters | -0.007 |
Token | how |
Token ID | 703 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
Lu | +0.057 |
Ann | +0.051 |
Lu | +0.049 |
youths | +0.048 |
angels | +0.048 |
Neg logit contributions | |
ret | -0.033 |
orse | -0.032 |
orses | -0.028 |
peanut | -0.025 |
ocket | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
Lu | +0.038 |
Ann | +0.032 |
angels | +0.032 |
co | +0.031 |
attendant | +0.030 |
Neg logit contributions | |
ret | -0.028 |
orse | -0.027 |
orses | -0.026 |
peanut | -0.024 |
ets | -0.023 |
Token | all |
Token ID | 477 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
Lu | +0.184 |
Ann | +0.183 |
co | +0.173 |
Michael | +0.171 |
Kim | +0.166 |
Neg logit contributions | |
peanut | -0.082 |
photos | -0.078 |
shrimp | -0.077 |
bean | -0.068 |
photo | -0.068 |
Token | igator |
Token ID | 23823 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
Lu | +0.101 |
co | +0.096 |
Michael | +0.087 |
Barber | +0.085 |
angels | +0.084 |
Neg logit contributions | |
's | -0.109 |
peanut | -0.108 |
ets | -0.105 |
ret | -0.105 |
s | -0.101 |
Token | could |
Token ID | 714 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
Lu | +0.030 |
angels | +0.025 |
Ann | +0.025 |
Lu | +0.025 |
firefighters | +0.024 |
Neg logit contributions | |
ret | -0.022 |
orse | -0.022 |
orses | -0.020 |
peanut | -0.019 |
ets | -0.019 |
Token | talk |
Token ID | 1561 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
Lu | +0.083 |
Ann | +0.068 |
Lu | +0.068 |
youths | +0.068 |
angels | +0.067 |
Neg logit contributions | |
orse | -0.053 |
ret | -0.049 |
orses | -0.046 |
peanut | -0.043 |
ork | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
Lu | +0.035 |
Ann | +0.029 |
angels | +0.028 |
Lu | +0.028 |
firefighters | +0.027 |
Neg logit contributions | |
ret | -0.025 |
orse | -0.024 |
orses | -0.022 |
ets | -0.022 |
peanut | -0.021 |
Token | of |
Token ID | 286 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
Lu | +0.058 |
angels | +0.048 |
Lu | +0.047 |
Ann | +0.047 |
Stev | +0.045 |
Neg logit contributions | |
orse | -0.048 |
ret | -0.047 |
orses | -0.045 |
ets | -0.044 |
peanut | -0.042 |
Token | them |
Token ID | 606 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
Lu | +0.020 |
Ann | +0.016 |
angels | +0.016 |
Lu | +0.016 |
youths | +0.016 |
Neg logit contributions | |
ret | -0.016 |
orse | -0.015 |
orses | -0.014 |
peanut | -0.013 |
ets | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
ret | +0.006 |
orse | +0.006 |
orses | +0.006 |
peanut | +0.005 |
ets | +0.005 |
Neg logit contributions | |
Lu | -0.007 |
angels | -0.006 |
Ann | -0.006 |
Lu | -0.006 |
firefighters | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +7.30e-01 |
Pos logit contributions | |
ret | +0.002 |
orse | +0.002 |
orses | +0.001 |
peanut | +0.001 |
ets | +0.001 |
Neg logit contributions | |
Lu | -0.002 |
angels | -0.002 |
Ann | -0.002 |
Lu | -0.002 |
firefighters | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
Lu | +0.038 |
angels | +0.034 |
firefighters | +0.032 |
youths | +0.032 |
Ann | +0.031 |
Neg logit contributions | |
orse | -0.025 |
ret | -0.024 |
orses | -0.023 |
ork | -0.020 |
peanut | -0.019 |
Token | The |
Token ID | 464 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
Lu | +0.021 |
angels | +0.017 |
firefighters | +0.016 |
youths | +0.016 |
Ann | +0.016 |
Neg logit contributions | |
orse | -0.015 |
ret | -0.015 |
orses | -0.014 |
ork | -0.012 |
ets | -0.012 |
Token | fog |
Token ID | 19558 |
Feature activation | +9.94e-01 |
Pos logit contributions | |
whe | +0.156 |
ussy | +0.153 |
Lu | +0.149 |
wood | +0.144 |
Ann | +0.142 |
Neg logit contributions | |
wishes | -0.086 |
wish | -0.084 |
bad | -0.080 |
dream | -0.079 |
holes | -0.079 |
Token | was |
Token ID | 373 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
Lu | +0.051 |
Lu | +0.042 |
youths | +0.042 |
Ann | +0.041 |
angels | +0.041 |
Neg logit contributions | |
orse | -0.036 |
ret | -0.035 |
orses | -0.033 |
ork | -0.030 |
peanut | -0.030 |
Token | gone |
Token ID | 3750 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
Lu | +0.071 |
Ann | +0.062 |
youths | +0.061 |
angels | +0.061 |
Lu | +0.059 |
Neg logit contributions | |
orse | -0.043 |
ret | -0.040 |
peanut | -0.039 |
orses | -0.038 |
ets | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
Lu | +0.009 |
angels | +0.007 |
Ann | +0.007 |
Lu | +0.007 |
firefighters | +0.007 |
Neg logit contributions | |
ret | -0.006 |
orse | -0.006 |
orses | -0.006 |
peanut | -0.005 |
ets | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
ret | +0.010 |
orse | +0.010 |
orses | +0.009 |
peanut | +0.008 |
ets | +0.008 |
Neg logit contributions | |
Lu | -0.011 |
angels | -0.009 |
Ann | -0.009 |
Lu | -0.009 |
firefighters | -0.009 |
Token | ! |
Token ID | 0 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
ret | +0.013 |
orse | +0.013 |
orses | +0.012 |
peanut | +0.011 |
ets | +0.011 |
Neg logit contributions | |
Lu | -0.013 |
angels | -0.011 |
Ann | -0.011 |
Lu | -0.010 |
firefighters | -0.010 |
Token | |
Token ID | 220 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
ret | +0.031 |
orse | +0.031 |
orses | +0.029 |
ets | +0.027 |
peanut | +0.026 |
Neg logit contributions | |
Lu | -0.041 |
Lu | -0.032 |
Lucy | -0.032 |
Ann | -0.032 |
attendant | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
Lu | +0.007 |
angels | +0.006 |
Ann | +0.005 |
Lu | +0.005 |
firefighters | +0.005 |
Neg logit contributions | |
ret | -0.005 |
orse | -0.005 |
orses | -0.005 |
peanut | -0.004 |
ets | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
ret | +0.000 |
orse | +0.000 |
orses | +0.000 |
peanut | +0.000 |
ets | +0.000 |
Neg logit contributions | |
Lu | -0.001 |
angels | -0.000 |
Ann | -0.000 |
Lu | -0.000 |
firefighters | -0.000 |
Token | The |
Token ID | 464 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
ret | +0.013 |
orse | +0.013 |
orses | +0.012 |
peanut | +0.011 |
ets | +0.011 |
Neg logit contributions | |
Lu | -0.014 |
Lu | -0.012 |
Ann | -0.012 |
angels | -0.011 |
firefighters | -0.011 |
Token | new |
Token ID | 649 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
Lu | +0.154 |
Ann | +0.151 |
Michael | +0.136 |
co | +0.133 |
Kim | +0.133 |
Neg logit contributions | |
peanut | -0.066 |
shrimp | -0.064 |
photos | -0.057 |
bean | -0.056 |
tomato | -0.052 |
Token | all |
Token ID | 477 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
Ann | +0.190 |
Lu | +0.187 |
Michael | +0.178 |
Kim | +0.174 |
co | +0.171 |
Neg logit contributions | |
shrimp | -0.078 |
shell | -0.073 |
peanut | -0.072 |
s | -0.070 |
net | -0.070 |
Token | igator |
Token ID | 23823 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Lu | +0.009 |
angels | +0.006 |
Ann | +0.006 |
Lu | +0.006 |
firefighters | +0.005 |
Neg logit contributions | |
orse | -0.023 |
ret | -0.023 |
orses | -0.022 |
peanut | -0.021 |
ets | -0.021 |
Token | said |
Token ID | 531 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
ret | +0.048 |
orse | +0.044 |
orses | +0.043 |
ets | +0.042 |
notebooks | +0.040 |
Neg logit contributions | |
Lu | -0.048 |
angels | -0.038 |
attendant | -0.038 |
firefighters | -0.037 |
Ann | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
Lu | +0.001 |
angels | +0.001 |
Ann | +0.001 |
Lu | +0.001 |
firefighters | +0.001 |
Neg logit contributions | |
ret | -0.001 |
orse | -0.001 |
orses | -0.001 |
peanut | -0.001 |
ets | -0.001 |
Token | " |
Token ID | 366 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
ret | +0.011 |
orse | +0.010 |
orses | +0.010 |
peanut | +0.009 |
ets | +0.009 |
Neg logit contributions | |
Lu | -0.010 |
angels | -0.008 |
Ann | -0.008 |
Lu | -0.008 |
firefighters | -0.008 |
Token | nodded |
Token ID | 14464 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
Lu | +0.025 |
angels | +0.022 |
firefighters | +0.021 |
Ann | +0.021 |
Lu | +0.021 |
Neg logit contributions | |
orse | -0.019 |
ret | -0.019 |
orses | -0.017 |
peanut | -0.016 |
ets | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
Lu | +0.009 |
angels | +0.007 |
Ann | +0.007 |
Lu | +0.007 |
firefighters | +0.007 |
Neg logit contributions | |
ret | -0.007 |
orse | -0.007 |
orses | -0.006 |
peanut | -0.006 |
ets | -0.006 |
Token | watched |
Token ID | 7342 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
Lu | +0.022 |
angels | +0.019 |
Ann | +0.019 |
Lu | +0.018 |
youths | +0.018 |
Neg logit contributions | |
orse | -0.018 |
ret | -0.018 |
orses | -0.016 |
peanut | -0.015 |
ets | -0.014 |
Token | carefully |
Token ID | 7773 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
ret | +0.008 |
orse | +0.008 |
orses | +0.008 |
peanut | +0.007 |
ets | +0.007 |
Neg logit contributions | |
Lu | -0.011 |
angels | -0.009 |
Ann | -0.008 |
Lu | -0.008 |
firefighters | -0.008 |
Token | as |
Token ID | 355 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
ret | +0.011 |
orse | +0.011 |
orses | +0.010 |
peanut | +0.009 |
ets | +0.009 |
Neg logit contributions | |
Lu | -0.013 |
angels | -0.010 |
Ann | -0.010 |
Lu | -0.010 |
firefighters | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
Lu | +0.026 |
Ann | +0.022 |
angels | +0.022 |
Lu | +0.022 |
co | +0.021 |
Neg logit contributions | |
ret | -0.018 |
orse | -0.018 |
orses | -0.016 |
peanut | -0.015 |
ork | -0.014 |
Token | grown |
Token ID | 7334 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Johnson | +0.145 |
Jenkins | +0.140 |
Tit | +0.140 |
girls | +0.139 |
department | +0.137 |
Neg logit contributions | |
picture | -0.096 |
's | -0.093 |
drawing | -0.092 |
photo | -0.088 |
it | -0.087 |
Token | up |
Token ID | 510 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Lu | +0.066 |
Ann | +0.056 |
Taxi | +0.055 |
Lu | +0.055 |
youths | +0.055 |
Neg logit contributions | |
ret | -0.047 |
orse | -0.044 |
orses | -0.041 |
sel | -0.041 |
ets | -0.039 |
Token | stretched |
Token ID | 19110 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
Lu | +0.005 |
angels | +0.004 |
Ann | +0.004 |
Lu | +0.004 |
firefighters | +0.004 |
Neg logit contributions | |
ret | -0.004 |
orse | -0.004 |
orses | -0.004 |
peanut | -0.004 |
ets | -0.004 |
Token | her |
Token ID | 607 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
orse | +0.013 |
ret | +0.013 |
orses | +0.012 |
peanut | +0.011 |
ets | +0.011 |
Neg logit contributions | |
Lu | -0.014 |
angels | -0.012 |
Ann | -0.011 |
Lu | -0.011 |
firefighters | -0.011 |
Token | arms |
Token ID | 5101 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
Lu | +0.056 |
Ann | +0.050 |
firefighters | +0.049 |
Lu | +0.048 |
youths | +0.048 |
Neg logit contributions | |
ret | -0.035 |
orse | -0.033 |
orses | -0.030 |
ork | -0.029 |
ets | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
ret | +0.008 |
orse | +0.008 |
orses | +0.008 |
peanut | +0.007 |
ets | +0.007 |
Neg logit contributions | |
Lu | -0.009 |
angels | -0.007 |
Ann | -0.007 |
Lu | -0.007 |
firefighters | -0.007 |
Token | little |
Token ID | 1310 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
ret | +0.009 |
orse | +0.009 |
orses | +0.009 |
ets | +0.008 |
peanut | +0.008 |
Neg logit contributions | |
Lu | -0.013 |
angels | -0.011 |
Lu | -0.011 |
Ann | -0.010 |
firefighters | -0.010 |
Token | girl |
Token ID | 2576 |
Feature activation | -9.24e-01 |
Pos logit contributions | |
Lu | +0.026 |
Ann | +0.021 |
angels | +0.020 |
Lu | +0.020 |
firefighters | +0.019 |
Neg logit contributions | |
orse | -0.034 |
ret | -0.033 |
peanut | -0.032 |
orses | -0.032 |
ets | -0.029 |
Token | named |
Token ID | 3706 |
Feature activation | -5.17e-01 |
Pos logit contributions | |
ret | +0.040 |
orse | +0.037 |
orses | +0.036 |
ets | +0.035 |
sel | +0.034 |
Neg logit contributions | |
Lu | -0.044 |
Lu | -0.035 |
Ann | -0.034 |
angels | -0.034 |
co | -0.033 |
Token | Lu |
Token ID | 6026 |
Feature activation | +9.13e-01 |
Pos logit contributions | |
orse | +0.024 |
ret | +0.024 |
orses | +0.023 |
peanut | +0.021 |
sel | +0.021 |
Neg logit contributions | |
Lu | -0.023 |
Lu | -0.018 |
Ann | -0.018 |
angels | -0.018 |
Lucy | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
Lu | +0.038 |
angels | +0.034 |
firefighters | +0.034 |
youths | +0.033 |
attendant | +0.031 |
Neg logit contributions | |
orse | -0.040 |
ret | -0.039 |
orses | -0.036 |
peanut | -0.033 |
ets | -0.032 |
Token | She |
Token ID | 1375 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
guard | +0.204 |
helpers | +0.197 |
mercy | +0.196 |
firefighters | +0.196 |
fighters | +0.194 |
Neg logit contributions | |
This | -0.081 |
Its | -0.067 |
It | -0.067 |
s | -0.066 |
's | -0.059 |
Token | was |
Token ID | 373 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
ret | +0.024 |
orse | +0.023 |
orses | +0.022 |
ets | +0.020 |
peanut | +0.020 |
Neg logit contributions | |
Lu | -0.022 |
Lu | -0.017 |
Ann | -0.017 |
angels | -0.017 |
firefighters | -0.016 |
Token | three |
Token ID | 1115 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Lu | +0.036 |
Ann | +0.031 |
Lu | +0.030 |
angels | +0.030 |
attendant | +0.029 |
Neg logit contributions | |
orse | -0.024 |
ret | -0.024 |
peanut | -0.022 |
orses | -0.022 |
ocket | -0.019 |
Token | years |
Token ID | 812 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Lu | +0.057 |
attendant | +0.049 |
angels | +0.048 |
Ann | +0.048 |
Taxi | +0.047 |
Neg logit contributions | |
orse | -0.032 |
ret | -0.030 |
orses | -0.029 |
peanut | -0.028 |
ets | -0.026 |
Token | old |
Token ID | 1468 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Lu | +0.048 |
Ann | +0.042 |
angels | +0.040 |
Lu | +0.039 |
Michael | +0.038 |
Neg logit contributions | |
orse | -0.040 |
ret | -0.040 |
orses | -0.037 |
peanut | -0.037 |
ork | -0.034 |
Token | Can |
Token ID | 6090 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Lu | +0.080 |
angels | +0.071 |
firefighters | +0.067 |
youths | +0.066 |
attendant | +0.066 |
Neg logit contributions | |
peanut | -0.046 |
orse | -0.045 |
ret | -0.044 |
orses | -0.042 |
scored | -0.041 |
Token | I |
Token ID | 314 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Lu | +0.067 |
firefighters | +0.059 |
angels | +0.059 |
youths | +0.058 |
Ann | +0.058 |
Neg logit contributions | |
ret | -0.021 |
orse | -0.021 |
orses | -0.017 |
peanut | -0.016 |
ork | -0.014 |
Token | use |
Token ID | 779 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
Lu | +0.055 |
angels | +0.049 |
Ann | +0.046 |
Lu | +0.046 |
youths | +0.046 |
Neg logit contributions | |
orse | -0.035 |
ret | -0.034 |
orses | -0.031 |
peanut | -0.030 |
ets | -0.027 |
Token | it |
Token ID | 340 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
orse | +0.012 |
ret | +0.012 |
orses | +0.011 |
peanut | +0.010 |
ets | +0.010 |
Neg logit contributions | |
Lu | -0.020 |
angels | -0.017 |
Ann | -0.016 |
Lu | -0.016 |
firefighters | -0.016 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
Lu | +0.052 |
Ann | +0.043 |
angels | +0.043 |
Lu | +0.042 |
youths | +0.042 |
Neg logit contributions | |
ret | -0.036 |
orse | -0.035 |
peanut | -0.032 |
orses | -0.032 |
ets | -0.030 |
Token | asked |
Token ID | 1965 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
ators | +0.060 |
angels | +0.059 |
Lu | +0.057 |
co | +0.056 |
department | +0.053 |
Neg logit contributions | |
orse | -0.096 |
peanut | -0.090 |
's | -0.088 |
bs | -0.087 |
This | -0.087 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +6.44e-01 |
Pos logit contributions | |
co | +0.134 |
iddle | +0.132 |
ators | +0.130 |
fighters | +0.128 |
outage | +0.125 |
Neg logit contributions | |
's | -0.119 |
it | -0.110 |
bs | -0.108 |
peanut | -0.108 |
accidentally | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Lu | +0.033 |
angels | +0.028 |
Ann | +0.028 |
firefighters | +0.028 |
Lu | +0.028 |
Neg logit contributions | |
orse | -0.023 |
ret | -0.023 |
orses | -0.021 |
peanut | -0.019 |
ets | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
Lu | +0.074 |
angels | +0.071 |
firefighters | +0.068 |
ators | +0.067 |
youths | +0.067 |
Neg logit contributions | |
orse | -0.034 |
ret | -0.031 |
orses | -0.028 |
This | -0.028 |
ork | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
Lu | +0.062 |
angels | +0.058 |
firefighters | +0.057 |
youths | +0.055 |
ators | +0.054 |
Neg logit contributions | |
ret | -0.033 |
orse | -0.033 |
orses | -0.029 |
ork | -0.028 |
peanut | -0.024 |
Token | " |
Token ID | 1 |
Feature activation | +8.24e-01 |
Pos logit contributions | |
Lu | +0.088 |
angels | +0.083 |
firefighters | +0.081 |
youths | +0.080 |
ators | +0.079 |
Neg logit contributions | |
orse | -0.029 |
ork | -0.028 |
's | -0.027 |
ret | -0.026 |
orses | -0.024 |
Token | bright |
Token ID | 6016 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Lu | +0.036 |
angels | +0.030 |
Ann | +0.030 |
youths | +0.029 |
Lu | +0.029 |
Neg logit contributions | |
orse | -0.027 |
ret | -0.026 |
orses | -0.024 |
peanut | -0.024 |
ets | -0.023 |
Token | fan |
Token ID | 4336 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
Lu | +0.075 |
Ann | +0.067 |
youths | +0.064 |
angels | +0.064 |
Lu | +0.063 |
Neg logit contributions | |
orse | -0.049 |
peanut | -0.047 |
ret | -0.046 |
orses | -0.042 |
ork | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
Lu | +0.014 |
angels | +0.011 |
Ann | +0.011 |
Lu | +0.011 |
youths | +0.011 |
Neg logit contributions | |
orse | -0.011 |
ret | -0.011 |
orses | -0.010 |
peanut | -0.009 |
ets | -0.009 |
Token | Joe |
Token ID | 5689 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
Lu | +0.008 |
angels | +0.007 |
Ann | +0.006 |
Lu | +0.006 |
firefighters | +0.006 |
Neg logit contributions | |
ret | -0.007 |
orse | -0.007 |
orses | -0.007 |
peanut | -0.006 |
ets | -0.006 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
Lu | +0.020 |
angels | +0.017 |
Ann | +0.016 |
Lu | +0.016 |
firefighters | +0.016 |
Neg logit contributions | |
orse | -0.013 |
ret | -0.013 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
Lu | +0.079 |
youths | +0.070 |
attendant | +0.069 |
angels | +0.068 |
Ann | +0.068 |
Neg logit contributions | |
orse | -0.039 |
peanut | -0.039 |
ret | -0.037 |
orses | -0.034 |
ork | -0.031 |
Token | fan |
Token ID | 4336 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
Lu | +0.138 |
Ann | +0.135 |
jack | +0.131 |
Jenkins | +0.130 |
Lu | +0.128 |
Neg logit contributions | |
's | -0.117 |
picture | -0.112 |
drawing | -0.104 |
may | -0.101 |
ret | -0.100 |
Token | too |
Token ID | 1165 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
Lu | +0.078 |
Ann | +0.068 |
Lu | +0.066 |
youths | +0.064 |
angels | +0.064 |
Neg logit contributions | |
ret | -0.046 |
orse | -0.044 |
peanut | -0.042 |
orses | -0.040 |
ets | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
Lu | +0.027 |
angels | +0.023 |
Ann | +0.023 |
Lu | +0.023 |
firefighters | +0.023 |
Neg logit contributions | |
ret | -0.016 |
orse | -0.015 |
orses | -0.014 |
peanut | -0.013 |
ets | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +8.60e-01 |
Pos logit contributions | |
ret | +0.010 |
orse | +0.010 |
orses | +0.009 |
peanut | +0.008 |
ets | +0.008 |
Neg logit contributions | |
Lu | -0.011 |
angels | -0.009 |
Ann | -0.009 |
Lu | -0.009 |
firefighters | -0.008 |
Token | so |
Token ID | 523 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
Lu | +0.044 |
angels | +0.037 |
Ann | +0.036 |
Lu | +0.036 |
Taxi | +0.036 |
Neg logit contributions | |
orse | -0.031 |
ret | -0.028 |
orses | -0.028 |
peanut | -0.026 |
ets | -0.024 |
Token | today |
Token ID | 1909 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
ret | +0.026 |
orse | +0.025 |
orses | +0.024 |
ets | +0.022 |
peanut | +0.022 |
Neg logit contributions | |
Lu | -0.020 |
angels | -0.016 |
firefighters | -0.015 |
Ann | -0.015 |
Lu | -0.015 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
Lu | +0.024 |
angels | +0.019 |
Ann | +0.019 |
Lu | +0.019 |
firefighters | +0.019 |
Neg logit contributions | |
orse | -0.023 |
ret | -0.022 |
orses | -0.021 |
peanut | -0.020 |
ets | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
Lu | +0.011 |
angels | +0.010 |
Ann | +0.009 |
firefighters | +0.009 |
Lu | +0.009 |
Neg logit contributions | |
orse | -0.008 |
ret | -0.008 |
orses | -0.007 |
peanut | -0.007 |
ets | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +5.08e-02 |
Pos logit contributions | |
Lu | +0.015 |
angels | +0.013 |
Ann | +0.012 |
firefighters | +0.012 |
Lu | +0.012 |
Neg logit contributions | |
ret | -0.011 |
orse | -0.011 |
orses | -0.010 |
peanut | -0.010 |
ork | -0.009 |
Token | Mr |
Token ID | 5246 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
Lu | +0.003 |
angels | +0.002 |
Ann | +0.002 |
Lu | +0.002 |
firefighters | +0.002 |
Neg logit contributions | |
ret | -0.002 |
orse | -0.002 |
orses | -0.002 |
peanut | -0.002 |
ets | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
Lu | +0.140 |
youths | +0.123 |
Ann | +0.123 |
Michael | +0.122 |
Lu | +0.122 |
Neg logit contributions | |
orse | -0.058 |
ret | -0.055 |
peanut | -0.049 |
orses | -0.049 |
ork | -0.046 |
Token | Sun |
Token ID | 3825 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
aid | +0.116 |
guard | +0.116 |
youths | +0.115 |
co | +0.112 |
mercy | +0.109 |
Neg logit contributions | |
orse | -0.142 |
T | -0.140 |
This | -0.138 |
orses | -0.129 |
ret | -0.122 |
Token | smiled |
Token ID | 13541 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
Lu | +0.095 |
youths | +0.084 |
firefighters | +0.082 |
angels | +0.082 |
Ann | +0.082 |
Neg logit contributions | |
orse | -0.066 |
ret | -0.058 |
orses | -0.058 |
peanut | -0.056 |
's | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
Lu | +0.031 |
Ann | +0.027 |
Lu | +0.026 |
angels | +0.026 |
firefighters | +0.025 |
Neg logit contributions | |
orse | -0.018 |
ret | -0.018 |
orses | -0.016 |
peanut | -0.016 |
ets | -0.015 |
Token | said |
Token ID | 531 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
Lu | +0.029 |
angels | +0.025 |
Ann | +0.025 |
Lu | +0.024 |
firefighters | +0.024 |
Neg logit contributions | |
orse | -0.019 |
ret | -0.018 |
orses | -0.017 |
peanut | -0.016 |
ets | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
Lu | +0.021 |
Ann | +0.018 |
angels | +0.017 |
Lu | +0.017 |
firefighters | +0.017 |
Neg logit contributions | |
ret | -0.012 |
orse | -0.012 |
orses | -0.011 |
peanut | -0.010 |
ets | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Lu | +0.012 |
angels | +0.010 |
Ann | +0.010 |
firefighters | +0.010 |
Lu | +0.010 |
Neg logit contributions | |
orse | -0.008 |
ret | -0.008 |
orses | -0.007 |
peanut | -0.007 |
ets | -0.006 |
Token | excited |
Token ID | 6568 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
Lu | +0.089 |
Ann | +0.082 |
lo | +0.079 |
angels | +0.079 |
co | +0.078 |
Neg logit contributions | |
orse | -0.049 |
ret | -0.041 |
peanut | -0.038 |
orses | -0.038 |
ocket | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
ret | +0.018 |
orse | +0.018 |
orses | +0.017 |
peanut | +0.016 |
ocket | +0.015 |
Neg logit contributions | |
Lu | -0.017 |
angels | -0.013 |
Lu | -0.013 |
firefighters | -0.013 |
Ann | -0.013 |
Token | have |
Token ID | 423 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
Lu | +0.035 |
angels | +0.030 |
Ann | +0.030 |
Lu | +0.030 |
youths | +0.029 |
Neg logit contributions | |
orse | -0.022 |
ret | -0.020 |
orses | -0.019 |
peanut | -0.017 |
ets | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Lu | +0.009 |
angels | +0.008 |
Ann | +0.007 |
Lu | +0.007 |
youths | +0.007 |
Neg logit contributions | |
orse | -0.008 |
ret | -0.008 |
orses | -0.007 |
peanut | -0.007 |
ets | -0.006 |
Token | new |
Token ID | 649 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
Lu | +0.087 |
Ann | +0.081 |
lo | +0.077 |
co | +0.077 |
ators | +0.076 |
Neg logit contributions | |
orse | -0.042 |
peanut | -0.042 |
ret | -0.041 |
ork | -0.038 |
orses | -0.035 |
Token | one |
Token ID | 530 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
Ann | +0.179 |
lo | +0.175 |
Andy | +0.167 |
Lu | +0.167 |
co | +0.164 |
Neg logit contributions | |
's | -0.094 |
folder | -0.093 |
pen | -0.092 |
notebook | -0.091 |
peanut | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
Lu | +0.033 |
Ann | +0.028 |
angels | +0.028 |
Lu | +0.027 |
firefighters | +0.027 |
Neg logit contributions | |
orse | -0.021 |
ret | -0.021 |
orses | -0.019 |
peanut | -0.019 |
ets | -0.018 |
Token | " |
Token ID | 366 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
ret | +0.002 |
orse | +0.002 |
orses | +0.002 |
peanut | +0.002 |
ets | +0.002 |
Neg logit contributions | |
Lu | -0.002 |
angels | -0.002 |
Ann | -0.002 |
Lu | -0.002 |
firefighters | -0.002 |
Token | I |
Token ID | 40 |
Feature activation | -8.61e-02 |
Pos logit contributions | |
Lu | +0.008 |
angels | +0.007 |
Ann | +0.006 |
firefighters | +0.006 |
youths | +0.006 |
Neg logit contributions | |
ret | -0.010 |
orse | -0.010 |
orses | -0.009 |
peanut | -0.009 |
ets | -0.009 |
Token | hope |
Token ID | 2911 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
ret | +0.003 |
orse | +0.003 |
orses | +0.003 |
peanut | +0.003 |
ets | +0.003 |
Neg logit contributions | |
Lu | -0.004 |
angels | -0.003 |
Ann | -0.003 |
Lu | -0.003 |
firefighters | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Lu | +0.022 |
Ann | +0.018 |
Lu | +0.018 |
angels | +0.018 |
firefighters | +0.018 |
Neg logit contributions | |
ret | -0.016 |
orse | -0.015 |
orses | -0.014 |
ets | -0.013 |
peanut | -0.013 |
Token | " |
Token ID | 366 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
Lu | +0.053 |
angels | +0.046 |
ators | +0.046 |
Ann | +0.045 |
co | +0.044 |
Neg logit contributions | |
orse | -0.032 |
ret | -0.030 |
orses | -0.028 |
peanut | -0.027 |
ork | -0.027 |
Token | Do |
Token ID | 5211 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ret | +0.006 |
orse | +0.006 |
orses | +0.006 |
peanut | +0.005 |
ets | +0.005 |
Neg logit contributions | |
Lu | -0.006 |
angels | -0.005 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token | you |
Token ID | 345 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
ret | +0.035 |
orse | +0.035 |
orses | +0.030 |
sel | +0.030 |
peanut | +0.029 |
Neg logit contributions | |
Lu | -0.067 |
angels | -0.060 |
Ann | -0.059 |
Lu | -0.057 |
Lucy | -0.055 |
Token | like |
Token ID | 588 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ret | +0.009 |
orse | +0.009 |
orses | +0.008 |
peanut | +0.008 |
ork | +0.007 |
Neg logit contributions | |
Lu | -0.012 |
angels | -0.010 |
Ann | -0.010 |
Lu | -0.009 |
firefighters | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
Lu | +0.094 |
jack | +0.091 |
attendant | +0.086 |
Ann | +0.082 |
firefighters | +0.081 |
Neg logit contributions | |
ret | -0.032 |
peanut | -0.029 |
orse | -0.029 |
orses | -0.028 |
drawing | -0.027 |
Token | stairs |
Token ID | 16046 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
Andy | +0.177 |
Lu | +0.169 |
jack | +0.166 |
Michael | +0.164 |
co | +0.159 |
Neg logit contributions | |
folder | -0.098 |
notebook | -0.094 |
picture | -0.090 |
drawing | -0.086 |
photos | -0.082 |
Token | ?" |
Token ID | 1701 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
Lu | +0.026 |
angels | +0.021 |
Lu | +0.021 |
Ann | +0.021 |
firefighters | +0.020 |
Neg logit contributions | |
ret | -0.024 |
orse | -0.023 |
orses | -0.022 |
ets | -0.021 |
peanut | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
Lu | +0.026 |
angels | +0.022 |
firefighters | +0.022 |
Lu | +0.022 |
Ann | +0.021 |
Neg logit contributions | |
orse | -0.017 |
ret | -0.017 |
orses | -0.016 |
peanut | -0.014 |
ets | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
Lu | +0.024 |
angels | +0.020 |
firefighters | +0.019 |
youths | +0.019 |
Ann | +0.019 |
Neg logit contributions | |
ret | -0.016 |
orse | -0.016 |
orses | -0.015 |
ork | -0.013 |
peanut | -0.013 |
Token | " |
Token ID | 1 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
Lu | +0.013 |
angels | +0.011 |
firefighters | +0.010 |
youths | +0.010 |
Ann | +0.010 |
Neg logit contributions | |
ret | -0.011 |
orse | -0.011 |
orses | -0.011 |
peanut | -0.010 |
ets | -0.010 |
Token | Sam |
Token ID | 3409 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
Lu | +0.036 |
angels | +0.031 |
Ann | +0.030 |
attendant | +0.029 |
co | +0.029 |
Neg logit contributions | |
orse | -0.027 |
ret | -0.025 |
orses | -0.024 |
peanut | -0.023 |
sel | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
ret | +0.019 |
orse | +0.018 |
orses | +0.017 |
peanut | +0.016 |
ets | +0.016 |
Neg logit contributions | |
Lu | -0.020 |
angels | -0.016 |
Lu | -0.015 |
Ann | -0.015 |
firefighters | -0.015 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
Lu | +0.070 |
attendant | +0.062 |
ators | +0.061 |
angels | +0.061 |
co | +0.060 |
Neg logit contributions | |
orse | -0.051 |
peanut | -0.045 |
orses | -0.045 |
ret | -0.045 |
T | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +8.86e-01 |
Pos logit contributions | |
Lu | +0.024 |
Ann | +0.020 |
angels | +0.020 |
Lu | +0.019 |
firefighters | +0.019 |
Neg logit contributions | |
orse | -0.017 |
ret | -0.017 |
orses | -0.015 |
peanut | -0.014 |
ets | -0.014 |
Token | Sam |
Token ID | 3409 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
Lu | +0.051 |
angels | +0.044 |
attendant | +0.043 |
guard | +0.043 |
Lu | +0.042 |
Neg logit contributions | |
orse | -0.024 |
orses | -0.023 |
ret | -0.023 |
peanut | -0.021 |
ets | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
ret | +0.012 |
orse | +0.011 |
orses | +0.011 |
peanut | +0.010 |
ets | +0.010 |
Neg logit contributions | |
Lu | -0.015 |
angels | -0.012 |
Ann | -0.012 |
Lu | -0.012 |
firefighters | -0.012 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
attendant | +0.133 |
iddle | +0.133 |
ambulance | +0.128 |
pr | +0.124 |
ators | +0.122 |
Neg logit contributions | |
This | -0.126 |
T | -0.121 |
's | -0.121 |
Po | -0.115 |
peanut | -0.112 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Lu | +0.017 |
angels | +0.014 |
Ann | +0.014 |
Lu | +0.014 |
firefighters | +0.014 |
Neg logit contributions | |
orse | -0.013 |
ret | -0.013 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
Lu | +0.074 |
Ann | +0.067 |
attendant | +0.067 |
Lu | +0.064 |
Michael | +0.062 |
Neg logit contributions | |
peanut | -0.041 |
orse | -0.041 |
orses | -0.038 |
notebooks | -0.037 |
ol | -0.036 |
Token | play |
Token ID | 711 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
Lu | +0.048 |
Ann | +0.041 |
Lu | +0.040 |
attendant | +0.040 |
angels | +0.039 |
Neg logit contributions | |
orse | -0.029 |
orses | -0.026 |
ret | -0.025 |
peanut | -0.024 |
notebooks | -0.021 |
Token | together |
Token ID | 1978 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
Lu | +0.040 |
Ann | +0.033 |
Taxi | +0.032 |
attendant | +0.032 |
Lu | +0.031 |
Neg logit contributions | |
orse | -0.026 |
ret | -0.025 |
orses | -0.024 |
peanut | -0.023 |
ets | -0.022 |
Token | able |
Token ID | 540 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
orse | +0.026 |
ret | +0.026 |
orses | +0.025 |
peanut | +0.023 |
ocket | +0.022 |
Neg logit contributions | |
Lu | -0.022 |
Ann | -0.018 |
Lu | -0.017 |
angels | -0.017 |
youths | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
Lu | +0.043 |
Ann | +0.036 |
Lu | +0.036 |
angels | +0.035 |
firefighters | +0.034 |
Neg logit contributions | |
ret | -0.027 |
orse | -0.026 |
orses | -0.024 |
ets | -0.023 |
peanut | -0.023 |
Token | And |
Token ID | 843 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
ret | +0.013 |
orse | +0.013 |
orses | +0.012 |
peanut | +0.011 |
ets | +0.011 |
Neg logit contributions | |
Lu | -0.015 |
angels | -0.012 |
Ann | -0.012 |
Lu | -0.012 |
firefighters | -0.012 |
Token | they |
Token ID | 484 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
ret | +0.028 |
orse | +0.026 |
orses | +0.024 |
ork | +0.023 |
ets | +0.022 |
Neg logit contributions | |
Lu | -0.037 |
angels | -0.031 |
firefighters | -0.029 |
youths | -0.029 |
Lucy | -0.029 |
Token | liked |
Token ID | 8288 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
Lu | +0.077 |
angels | +0.067 |
Ann | +0.065 |
firefighters | +0.064 |
taxis | +0.064 |
Neg logit contributions | |
orse | -0.043 |
ret | -0.043 |
ets | -0.038 |
peanut | -0.038 |
orses | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
Lu | +0.026 |
Ann | +0.022 |
Lu | +0.021 |
angels | +0.021 |
attendant | +0.021 |
Neg logit contributions | |
orse | -0.021 |
ret | -0.021 |
orses | -0.019 |
peanut | -0.019 |
notebooks | -0.017 |
Token | fog |
Token ID | 19558 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Andy | +0.153 |
Lu | +0.150 |
Ann | +0.147 |
jack | +0.145 |
Lu | +0.145 |
Neg logit contributions | |
seas | -0.102 |
shell | -0.101 |
picture | -0.093 |
photo | -0.092 |
photos | -0.090 |
Token | . |
Token ID | 13 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
Lu | +0.070 |
Lu | +0.059 |
Ann | +0.059 |
Michael | +0.056 |
youths | +0.056 |
Neg logit contributions | |
ret | -0.045 |
orse | -0.044 |
orses | -0.041 |
ets | -0.040 |
peanut | -0.038 |
Token | It |
Token ID | 632 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
ret | +0.035 |
orse | +0.033 |
orses | +0.031 |
ets | +0.029 |
peanut | +0.029 |
Neg logit contributions | |
Lu | -0.035 |
firefighters | -0.027 |
youths | -0.027 |
angels | -0.026 |
Ann | -0.026 |
Token | made |
Token ID | 925 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
Lu | +0.101 |
youths | +0.091 |
Taxi | +0.091 |
taxis | +0.089 |
angels | +0.088 |
Neg logit contributions | |
orse | -0.051 |
ret | -0.041 |
peanut | -0.041 |
orses | -0.041 |
's | -0.040 |
Token | their |
Token ID | 511 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
orse | +0.034 |
ret | +0.033 |
orses | +0.031 |
ocket | +0.029 |
peanut | +0.029 |
Neg logit contributions | |
Lu | -0.031 |
angels | -0.026 |
firefighters | -0.025 |
Ann | -0.023 |
youths | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +9.94e-01 |
Pos logit contributions | |
Lu | +0.075 |
angels | +0.066 |
firefighters | +0.065 |
youths | +0.063 |
Taxi | +0.061 |
Neg logit contributions | |
orse | -0.043 |
ret | -0.041 |
orses | -0.038 |
ork | -0.034 |
peanut | -0.031 |
Token | One |
Token ID | 3198 |
Feature activation | +7.08e-02 |
Pos logit contributions | |
Lu | +0.049 |
angels | +0.041 |
youths | +0.040 |
firefighters | +0.040 |
attendant | +0.039 |
Neg logit contributions | |
orse | -0.038 |
orses | -0.036 |
ret | -0.035 |
ork | -0.031 |
peanut | -0.031 |
Token | day |
Token ID | 1110 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
Lu | +0.003 |
angels | +0.003 |
Ann | +0.003 |
Lu | +0.003 |
firefighters | +0.003 |
Neg logit contributions | |
ret | -0.003 |
orse | -0.003 |
orses | -0.003 |
peanut | -0.003 |
ets | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +6.53e-01 |
Pos logit contributions | |
Lu | +0.003 |
angels | +0.002 |
Ann | +0.002 |
Lu | +0.002 |
firefighters | +0.002 |
Neg logit contributions | |
ret | -0.002 |
orse | -0.002 |
orses | -0.002 |
peanut | -0.002 |
ets | -0.002 |
Token | Sam |
Token ID | 3409 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
Lu | +0.036 |
angels | +0.031 |
co | +0.030 |
Ann | +0.030 |
attendant | +0.030 |
Neg logit contributions | |
orse | -0.019 |
ret | -0.019 |
orses | -0.017 |
peanut | -0.016 |
ets | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
ret | +0.027 |
orse | +0.026 |
orses | +0.024 |
ocket | +0.022 |
peanut | +0.022 |
Neg logit contributions | |
Lu | -0.031 |
Lu | -0.025 |
Ann | -0.025 |
angels | -0.024 |
firefighters | -0.024 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
attendant | +0.126 |
iddle | +0.124 |
iren | +0.120 |
ambulance | +0.118 |
outage | +0.118 |
Neg logit contributions | |
's | -0.116 |
Po | -0.114 |
This | -0.114 |
T | -0.112 |
peanut | -0.111 |
Token | went |
Token ID | 1816 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ret | +0.004 |
orse | +0.004 |
orses | +0.004 |
peanut | +0.003 |
ets | +0.003 |
Neg logit contributions | |
Lu | -0.005 |
angels | -0.004 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +8.03e-01 |
Pos logit contributions | |
Lu | +0.064 |
Ann | +0.055 |
Lu | +0.054 |
Michael | +0.052 |
Stev | +0.052 |
Neg logit contributions | |
ret | -0.035 |
orse | -0.034 |
ets | -0.032 |
orses | -0.032 |
peanut | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
Lu | +0.040 |
angels | +0.036 |
Ann | +0.035 |
Lu | +0.034 |
co | +0.034 |
Neg logit contributions | |
orse | -0.027 |
ret | -0.027 |
orses | -0.024 |
peanut | -0.024 |
ork | -0.021 |
Token | park |
Token ID | 3952 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
Lu | +0.014 |
angels | +0.011 |
Ann | +0.011 |
Lu | +0.011 |
firefighters | +0.011 |
Neg logit contributions | |
ret | -0.015 |
orse | -0.015 |
orses | -0.014 |
peanut | -0.014 |
ork | -0.013 |
Token | curious |
Token ID | 11040 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
ret | +0.010 |
orse | +0.010 |
orses | +0.009 |
ets | +0.009 |
peanut | +0.009 |
Neg logit contributions | |
Lu | -0.012 |
angels | -0.010 |
Lu | -0.009 |
Ann | -0.009 |
firefighters | -0.009 |
Token | little |
Token ID | 1310 |
Feature activation | +9.95e-01 |
Pos logit contributions | |
Lu | +0.029 |
Ann | +0.024 |
angels | +0.024 |
Lu | +0.023 |
firefighters | +0.022 |
Neg logit contributions | |
orse | -0.023 |
ret | -0.022 |
orses | -0.021 |
peanut | -0.020 |
ets | -0.019 |
Token | cat |
Token ID | 3797 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
Lu | +0.043 |
Ann | +0.036 |
angels | +0.033 |
Lu | +0.033 |
ators | +0.033 |
Neg logit contributions | |
orse | -0.044 |
peanut | -0.043 |
ret | -0.042 |
orses | -0.040 |
notebooks | -0.037 |
Token | named |
Token ID | 3706 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
ret | +0.017 |
orse | +0.016 |
orses | +0.015 |
ets | +0.014 |
peanut | +0.014 |
Neg logit contributions | |
Lu | -0.022 |
angels | -0.018 |
Lu | -0.018 |
Ann | -0.018 |
firefighters | -0.018 |
Token | Kitty |
Token ID | 21168 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
Lu | +0.006 |
angels | +0.005 |
Ann | +0.005 |
firefighters | +0.005 |
Lu | +0.005 |
Neg logit contributions | |
ret | -0.008 |
orse | -0.008 |
orses | -0.007 |
peanut | -0.007 |
ets | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
Lu | +0.018 |
angels | +0.015 |
Ann | +0.014 |
firefighters | +0.014 |
youths | +0.014 |
Neg logit contributions | |
orse | -0.013 |
ret | -0.013 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.011 |
Token | Kitty |
Token ID | 21168 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
guard | +0.151 |
ators | +0.142 |
pastor | +0.135 |
attendant | +0.132 |
helpers | +0.123 |
Neg logit contributions | |
This | -0.148 |
Its | -0.123 |
It | -0.120 |
The | -0.115 |
Find | -0.113 |
Token | loved |
Token ID | 6151 |
Feature activation | +8.75e-01 |
Pos logit contributions | |
Lu | +0.044 |
angels | +0.039 |
attendant | +0.038 |
firefighters | +0.037 |
youths | +0.037 |
Neg logit contributions | |
orse | -0.028 |
ret | -0.027 |
orses | -0.025 |
peanut | -0.024 |
ets | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
Lu | +0.045 |
Ann | +0.039 |
attendant | +0.038 |
Lu | +0.037 |
angels | +0.036 |
Neg logit contributions | |
orse | -0.030 |
peanut | -0.029 |
ret | -0.029 |
orses | -0.028 |
notebooks | -0.026 |
Token | play |
Token ID | 711 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
Lu | +0.013 |
angels | +0.011 |
Ann | +0.011 |
Lu | +0.011 |
firefighters | +0.011 |
Neg logit contributions | |
orse | -0.009 |
ret | -0.009 |
orses | -0.009 |
peanut | -0.008 |
ets | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
Lu | +0.016 |
Ann | +0.013 |
angels | +0.013 |
Lu | +0.013 |
attendant | +0.013 |
Neg logit contributions | |
orse | -0.013 |
ret | -0.013 |
orses | -0.012 |
peanut | -0.012 |
ets | -0.011 |
Token | happy |
Token ID | 3772 |
Feature activation | -7.74e-02 |
Pos logit contributions | |
Lu | +0.017 |
angels | +0.015 |
Ann | +0.015 |
Lu | +0.014 |
youths | +0.014 |
Neg logit contributions | |
orse | -0.012 |
ret | -0.012 |
orses | -0.011 |
peanut | -0.010 |
ocket | -0.009 |
Token | that |
Token ID | 326 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
ret | +0.003 |
orse | +0.003 |
orses | +0.003 |
peanut | +0.003 |
ets | +0.003 |
Neg logit contributions | |
Lu | -0.004 |
angels | -0.003 |
Ann | -0.003 |
Lu | -0.003 |
firefighters | -0.003 |
Token | he |
Token ID | 339 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Lu | +0.050 |
co | +0.044 |
Ann | +0.044 |
youths | +0.043 |
angels | +0.043 |
Neg logit contributions | |
orse | -0.025 |
ret | -0.023 |
peanut | -0.022 |
orses | -0.021 |
notebooks | -0.018 |
Token | had |
Token ID | 550 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Lu | +0.006 |
angels | +0.005 |
Ann | +0.005 |
Lu | +0.005 |
firefighters | +0.005 |
Neg logit contributions | |
ret | -0.003 |
orse | -0.003 |
orses | -0.003 |
peanut | -0.003 |
ets | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ret | +0.006 |
orse | +0.006 |
orses | +0.005 |
peanut | +0.005 |
ets | +0.005 |
Neg logit contributions | |
Lu | -0.006 |
angels | -0.005 |
Ann | -0.005 |
Lu | -0.005 |
firefighters | -0.005 |
Token | new |
Token ID | 649 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
Lu | +0.055 |
Ann | +0.048 |
co | +0.045 |
angels | +0.045 |
youths | +0.045 |
Neg logit contributions | |
ret | -0.036 |
orse | -0.034 |
peanut | -0.033 |
ork | -0.031 |
orses | -0.030 |
Token | bed |
Token ID | 3996 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
Ann | +0.172 |
lo | +0.170 |
Lu | +0.161 |
co | +0.159 |
Andy | +0.158 |
Neg logit contributions | |
pen | -0.102 |
folder | -0.100 |
notebook | -0.100 |
mouth | -0.098 |
ork | -0.096 |
Token | to |
Token ID | 284 |
Feature activation | +5.22e-01 |
Pos logit contributions | |
Lu | +0.009 |
angels | +0.008 |
Ann | +0.008 |
Lu | +0.008 |
firefighters | +0.007 |
Neg logit contributions | |
ret | -0.008 |
orse | -0.008 |
orses | -0.007 |
peanut | -0.007 |
ets | -0.006 |
Token | sleep |
Token ID | 3993 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
Lu | +0.026 |
angels | +0.021 |
youths | +0.021 |
Ann | +0.021 |
firefighters | +0.021 |
Neg logit contributions | |
orse | -0.020 |
ret | -0.020 |
orses | -0.018 |
peanut | -0.017 |
ets | -0.017 |
Token | in |
Token ID | 287 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
Lu | +0.003 |
angels | +0.003 |
Ann | +0.003 |
Lu | +0.003 |
firefighters | +0.003 |
Neg logit contributions | |
ret | -0.003 |
orse | -0.003 |
orses | -0.003 |
peanut | -0.002 |
ets | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
ret | +0.013 |
orse | +0.013 |
orses | +0.012 |
peanut | +0.011 |
ets | +0.011 |
Neg logit contributions | |
Lu | -0.014 |
angels | -0.011 |
Ann | -0.011 |
Lu | -0.011 |
firefighters | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
ret | +0.004 |
orse | +0.004 |
orses | +0.004 |
peanut | +0.004 |
ets | +0.004 |
Neg logit contributions | |
Lu | -0.005 |
angels | -0.004 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token | nap |
Token ID | 25422 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ret | +0.021 |
orse | +0.020 |
orses | +0.019 |
ets | +0.018 |
sel | +0.018 |
Neg logit contributions | |
Lu | -0.023 |
angels | -0.019 |
Lu | -0.018 |
Ann | -0.018 |
firefighters | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
Lu | +0.008 |
angels | +0.006 |
Ann | +0.006 |
Lu | +0.006 |
firefighters | +0.006 |
Neg logit contributions | |
ret | -0.006 |
orse | -0.006 |
orses | -0.006 |
peanut | -0.005 |
ets | -0.005 |
Token | One |
Token ID | 1881 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
ret | +0.089 |
orse | +0.088 |
ocket | +0.084 |
orses | +0.084 |
ocus | +0.083 |
Neg logit contributions | |
Lu | -0.060 |
Taxi | -0.042 |
Barber | -0.042 |
Luis | -0.039 |
taxis | -0.039 |
Token | day |
Token ID | 1110 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
Lu | +0.002 |
angels | +0.002 |
Ann | +0.002 |
Lu | +0.002 |
firefighters | +0.002 |
Neg logit contributions | |
ret | -0.002 |
orse | -0.002 |
orses | -0.002 |
peanut | -0.002 |
ets | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
ret | +0.179 |
iche | +0.176 |
ets | +0.174 |
orse | +0.173 |
orses | +0.172 |
Neg logit contributions | |
Lu | -0.082 |
angels | -0.052 |
Dog | -0.048 |
Taxi | -0.048 |
youths | -0.047 |
Token | Gro |
Token ID | 10299 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
iche | +0.223 |
pler | +0.219 |
ocus | +0.219 |
ets | +0.217 |
ulp | +0.215 |
Neg logit contributions | |
Lu | -0.098 |
Squirrel | -0.085 |
angels | -0.080 |
Dog | -0.080 |
Bird | -0.079 |
Token | ver |
Token ID | 332 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
orse | +0.085 |
peanut | +0.085 |
orses | +0.080 |
ret | +0.078 |
notebooks | +0.076 |
Neg logit contributions | |
Lu | -0.052 |
Ann | -0.040 |
angels | -0.039 |
Lu | -0.038 |
Taxi | -0.037 |
Token | decided |
Token ID | 3066 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
iche | +0.107 |
ret | +0.104 |
ork | +0.100 |
ets | +0.100 |
ocket | +0.100 |
Neg logit contributions | |
Lu | -0.076 |
Karen | -0.059 |
Dog | -0.057 |
angels | -0.056 |
Taxi | -0.055 |
Token | he |
Token ID | 339 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
ret | +0.054 |
orse | +0.053 |
orses | +0.051 |
sel | +0.048 |
ets | +0.047 |
Neg logit contributions | |
Lu | -0.037 |
angels | -0.029 |
firefighters | -0.027 |
Lu | -0.026 |
Lucy | -0.026 |
Token | wanted |
Token ID | 2227 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
Lu | +0.006 |
angels | +0.005 |
Ann | +0.005 |
Lu | +0.005 |
firefighters | +0.005 |
Neg logit contributions | |
ret | -0.004 |
orse | -0.004 |
orses | -0.004 |
peanut | -0.004 |
ets | -0.004 |
Token | found |
Token ID | 1043 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
ret | +0.048 |
orse | +0.047 |
orses | +0.045 |
ocus | +0.041 |
ets | +0.040 |
Neg logit contributions | |
Lu | -0.045 |
Lu | -0.035 |
firefighters | -0.034 |
Ann | -0.034 |
youths | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
orse | +0.019 |
ret | +0.018 |
orses | +0.018 |
ets | +0.016 |
peanut | +0.016 |
Neg logit contributions | |
Lu | -0.017 |
angels | -0.014 |
firefighters | -0.013 |
Ann | -0.013 |
Lu | -0.013 |
Token | brilliant |
Token ID | 10457 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
orse | +0.061 |
ret | +0.060 |
orses | +0.058 |
ets | +0.055 |
sel | +0.050 |
Neg logit contributions | |
Lu | -0.056 |
angels | -0.046 |
Lu | -0.044 |
attendant | -0.044 |
firefighters | -0.043 |
Token | skirt |
Token ID | 23967 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
orse | +0.031 |
ret | +0.031 |
orses | +0.030 |
ets | +0.027 |
sel | +0.025 |
Neg logit contributions | |
Lu | -0.035 |
angels | -0.029 |
Lu | -0.027 |
firefighters | -0.027 |
attendant | -0.027 |
Token | in |
Token ID | 287 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
ret | +0.037 |
orse | +0.036 |
orses | +0.034 |
peanut | +0.031 |
ets | +0.030 |
Neg logit contributions | |
Lu | -0.031 |
angels | -0.026 |
Lu | -0.024 |
firefighters | -0.024 |
Ann | -0.024 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
Lu | +0.022 |
angels | +0.018 |
Ann | +0.018 |
Lu | +0.017 |
youths | +0.017 |
Neg logit contributions | |
ret | -0.016 |
orse | -0.016 |
orses | -0.015 |
peanut | -0.014 |
notebooks | -0.013 |
Token | 's |
Token ID | 338 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ret | +0.188 |
AW | +0.158 |
orse | +0.157 |
alks | +0.149 |
eg | +0.147 |
Neg logit contributions | |
Lu | -0.154 |
co | -0.144 |
Lu | -0.130 |
Pen | -0.127 |
my | -0.125 |
Token | closet |
Token ID | 21615 |
Feature activation | -9.29e-01 |
Pos logit contributions | |
orse | +0.096 |
ret | +0.092 |
orses | +0.092 |
ocus | +0.081 |
sel | +0.080 |
Neg logit contributions | |
Lu | -0.057 |
angels | -0.043 |
attendant | -0.040 |
Lu | -0.040 |
Lucy | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
ret | +0.044 |
orse | +0.043 |
orses | +0.041 |
peanut | +0.037 |
ets | +0.036 |
Neg logit contributions | |
Lu | -0.038 |
angels | -0.031 |
attendant | -0.030 |
firefighters | -0.030 |
Ann | -0.029 |
Token | It |
Token ID | 632 |
Feature activation | -8.19e-03 |
Pos logit contributions | |
ret | +0.011 |
orse | +0.010 |
orses | +0.010 |
peanut | +0.009 |
ets | +0.009 |
Neg logit contributions | |
Lu | -0.011 |
Ann | -0.009 |
angels | -0.009 |
Lu | -0.009 |
firefighters | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
ret | +0.000 |
orse | +0.000 |
orses | +0.000 |
peanut | +0.000 |
ets | +0.000 |
Neg logit contributions | |
Lu | -0.000 |
angels | -0.000 |
Ann | -0.000 |
Lu | -0.000 |
firefighters | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | -9.05e-01 |
Pos logit contributions | |
ret | +0.010 |
orse | +0.010 |
orses | +0.009 |
ets | +0.008 |
peanut | +0.008 |
Neg logit contributions | |
Lu | -0.010 |
angels | -0.008 |
Ann | -0.008 |
Lu | -0.008 |
firefighters | -0.008 |
Token | beach |
Token ID | 10481 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ret | +0.047 |
orse | +0.047 |
orses | +0.043 |
ets | +0.040 |
ocket | +0.039 |
Neg logit contributions | |
Lu | -0.035 |
angels | -0.028 |
firefighters | -0.027 |
Lu | -0.026 |
attendant | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
ret | +0.048 |
orse | +0.046 |
orses | +0.042 |
peanut | +0.041 |
ocket | +0.040 |
Neg logit contributions | |
Lu | -0.044 |
angels | -0.037 |
Ann | -0.034 |
Lu | -0.034 |
firefighters | -0.034 |
Token | But |
Token ID | 887 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
orse | +0.163 |
ret | +0.160 |
arding | +0.156 |
orses | +0.156 |
ocus | +0.155 |
Neg logit contributions | |
Lu | -0.085 |
Barber | -0.062 |
youths | -0.059 |
Pen | -0.056 |
Bob | -0.053 |
Token | first |
Token ID | 717 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
ets | +0.161 |
ret | +0.155 |
ork | +0.154 |
iche | +0.154 |
orses | +0.153 |
Neg logit contributions | |
Lu | -0.105 |
Barber | -0.083 |
Pen | -0.076 |
Bob | -0.074 |
tons | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
iche | +0.193 |
pler | +0.187 |
orse | +0.185 |
orses | +0.182 |
ork | +0.181 |
Neg logit contributions | |
Lu | -0.083 |
helpers | -0.062 |
Barber | -0.062 |
youths | -0.062 |
guard | -0.060 |
Token | he |
Token ID | 339 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
ork | +0.191 |
orse | +0.190 |
ocus | +0.189 |
orses | +0.188 |
pler | +0.188 |
Neg logit contributions | |
Lu | -0.117 |
youths | -0.099 |
Barber | -0.098 |
helpers | -0.096 |
answer | -0.092 |
Token | had |
Token ID | 550 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
Lu | +0.006 |
angels | +0.005 |
Ann | +0.005 |
Lu | +0.005 |
firefighters | +0.005 |
Neg logit contributions | |
orse | -0.004 |
ret | -0.004 |
orses | -0.004 |
peanut | -0.004 |
ets | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
Lu | +0.006 |
angels | +0.005 |
Ann | +0.005 |
Lu | +0.005 |
firefighters | +0.005 |
Neg logit contributions | |
ret | -0.005 |
orse | -0.005 |
orses | -0.005 |
peanut | -0.005 |
ets | -0.004 |
Token | pack |
Token ID | 2353 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
Lu | +0.001 |
angels | +0.001 |
Ann | +0.001 |
Lu | +0.001 |
firefighters | +0.001 |
Neg logit contributions | |
ret | -0.001 |
orse | -0.001 |
orses | -0.001 |
peanut | -0.001 |
ets | -0.001 |
Token | ! |
Token ID | 0 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
orse | +0.054 |
ret | +0.052 |
orses | +0.050 |
pler | +0.045 |
ocus | +0.045 |
Neg logit contributions | |
Lu | -0.036 |
angels | -0.028 |
firefighters | -0.026 |
Lu | -0.026 |
Ann | -0.025 |
Token | what |
Token ID | 644 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
ret | +0.014 |
orse | +0.014 |
orses | +0.013 |
ets | +0.012 |
peanut | +0.012 |
Neg logit contributions | |
Lu | -0.017 |
angels | -0.014 |
firefighters | -0.013 |
Ann | -0.013 |
Lu | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | +9.04e-01 |
Pos logit contributions | |
Lu | +0.028 |
angels | +0.023 |
Ann | +0.023 |
Lu | +0.022 |
attendant | +0.022 |
Neg logit contributions | |
orse | -0.021 |
ret | -0.020 |
orses | -0.019 |
peanut | -0.018 |
ork | -0.017 |
Token | do |
Token ID | 466 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
Lu | +0.050 |
Lu | +0.042 |
angels | +0.042 |
firefighters | +0.042 |
Ann | +0.041 |
Neg logit contributions | |
orse | -0.028 |
ret | -0.027 |
orses | -0.026 |
peanut | -0.025 |
ets | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
Lu | +0.006 |
angels | +0.005 |
Ann | +0.005 |
Lu | +0.005 |
firefighters | +0.005 |
Neg logit contributions | |
ret | -0.005 |
orse | -0.005 |
orses | -0.004 |
peanut | -0.004 |
ets | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ifying | +0.202 |
ocket | +0.197 |
ret | +0.191 |
ocus | +0.190 |
ets | +0.190 |
Neg logit contributions | |
Lu | -0.072 |
youths | -0.059 |
Bob | -0.057 |
attendant | -0.055 |
Bird | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | -4.12e+00 |
Pos logit contributions | |
ret | +0.075 |
orse | +0.074 |
ets | +0.073 |
peanut | +0.071 |
orses | +0.070 |
Neg logit contributions | |
Lu | -0.058 |
Ann | -0.046 |
Lu | -0.044 |
angels | -0.043 |
attendant | -0.041 |
Token | But |
Token ID | 1537 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
ocket | +0.199 |
ret | +0.195 |
icol | +0.193 |
ifying | +0.191 |
ulp | +0.189 |
Neg logit contributions | |
Ann | -0.087 |
youths | -0.084 |
Mark | -0.074 |
Fine | -0.073 |
attendant | -0.072 |
Token | then |
Token ID | 788 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
ret | +0.191 |
ork | +0.188 |
ets | +0.178 |
iche | +0.177 |
ables | +0.172 |
Neg logit contributions | |
Lu | -0.088 |
Bob | -0.073 |
Bird | -0.071 |
Dog | -0.071 |
helpers | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
ret | +0.106 |
ets | +0.103 |
orse | +0.097 |
orses | +0.096 |
ork | +0.095 |
Neg logit contributions | |
Lu | -0.069 |
angels | -0.056 |
Bird | -0.053 |
firefighters | -0.052 |
Lucy | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
ets | +0.218 |
ret | +0.209 |
ocus | +0.206 |
ard | +0.205 |
orses | +0.204 |
Neg logit contributions | |
Lu | -0.067 |
helpers | -0.063 |
angels | -0.059 |
firefighters | -0.059 |
Bird | -0.058 |
Token | kind |
Token ID | 1611 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
Lu | +0.002 |
angels | +0.002 |
Ann | +0.002 |
Lu | +0.002 |
firefighters | +0.002 |
Neg logit contributions | |
ret | -0.002 |
orse | -0.002 |
orses | -0.002 |
peanut | -0.002 |
ets | -0.002 |
Token | pumpkin |
Token ID | 30089 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
Lu | +0.007 |
Ann | +0.005 |
angels | +0.005 |
Lu | +0.005 |
firefighters | +0.005 |
Neg logit contributions | |
orse | -0.010 |
ret | -0.010 |
orses | -0.010 |
peanut | -0.009 |
ets | -0.009 |
Token | friends |
Token ID | 2460 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
ret | +0.019 |
orse | +0.019 |
orses | +0.018 |
ets | +0.017 |
peanut | +0.017 |
Neg logit contributions | |
Lu | -0.011 |
angels | -0.009 |
firefighters | -0.008 |
Lu | -0.008 |
Ann | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
Lu | +0.018 |
Ann | +0.015 |
angels | +0.015 |
Lu | +0.015 |
firefighters | +0.015 |
Neg logit contributions | |
orse | -0.013 |
ret | -0.013 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.010 |
Token | One |
Token ID | 1881 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
ret | +0.145 |
ocket | +0.141 |
orse | +0.139 |
ifying | +0.139 |
pler | +0.136 |
Neg logit contributions | |
Lu | -0.074 |
youths | -0.059 |
attendant | -0.053 |
Ann | -0.053 |
Lucy | -0.052 |
Token | day |
Token ID | 1110 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
ret | +0.025 |
orse | +0.023 |
orses | +0.023 |
ets | +0.021 |
sel | +0.021 |
Neg logit contributions | |
Lu | -0.020 |
Lu | -0.016 |
angels | -0.016 |
Ann | -0.015 |
firefighters | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -4.07e+00 |
Pos logit contributions | |
ret | +0.164 |
ets | +0.155 |
orse | +0.155 |
ocket | +0.155 |
generation | +0.152 |
Neg logit contributions | |
Lu | -0.072 |
angels | -0.060 |
youths | -0.058 |
helpers | -0.051 |
Lucy | -0.046 |
Token | a |
Token ID | 257 |
Feature activation | -9.90e-01 |
Pos logit contributions | |
ret | +0.180 |
ets | +0.175 |
pler | +0.173 |
ork | +0.173 |
ocus | +0.171 |
Neg logit contributions | |
helpers | -0.118 |
youths | -0.117 |
Bob | -0.112 |
angels | -0.110 |
Lu | -0.105 |
Token | big |
Token ID | 1263 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
ret | +0.044 |
orse | +0.042 |
orses | +0.040 |
ets | +0.039 |
ocus | +0.037 |
Neg logit contributions | |
Lu | -0.044 |
angels | -0.035 |
firefighters | -0.035 |
Lu | -0.035 |
youths | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
ret | +0.013 |
orse | +0.012 |
orses | +0.012 |
ets | +0.011 |
peanut | +0.011 |
Neg logit contributions | |
Lu | -0.011 |
angels | -0.008 |
firefighters | -0.008 |
Ann | -0.008 |
Lu | -0.008 |
Token | mean |
Token ID | 1612 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
ret | +0.002 |
orse | +0.002 |
orses | +0.002 |
peanut | +0.002 |
ets | +0.001 |
Neg logit contributions | |
Lu | -0.002 |
angels | -0.002 |
Ann | -0.002 |
Lu | -0.002 |
firefighters | -0.002 |
Token | squash |
Token ID | 34613 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ret | +0.032 |
orse | +0.032 |
orses | +0.031 |
ets | +0.029 |
ocus | +0.027 |
Neg logit contributions | |
Lu | -0.027 |
angels | -0.022 |
firefighters | -0.021 |
youths | -0.020 |
Lu | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
ret | +0.014 |
orse | +0.013 |
orses | +0.013 |
ets | +0.012 |
peanut | +0.011 |
Neg logit contributions | |
Lu | -0.013 |
angels | -0.010 |
Lu | -0.010 |
Ann | -0.010 |
firefighters | -0.010 |
Token | fair |
Token ID | 3148 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
ret | +0.014 |
orse | +0.014 |
orses | +0.013 |
ets | +0.012 |
peanut | +0.012 |
Neg logit contributions | |
Lu | -0.017 |
angels | -0.014 |
Lu | -0.014 |
Ann | -0.014 |
firefighters | -0.014 |
Token | peanut |
Token ID | 26636 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
ret | +0.002 |
orse | +0.002 |
orses | +0.002 |
peanut | +0.002 |
ets | +0.002 |
Neg logit contributions | |
Lu | -0.002 |
angels | -0.002 |
Ann | -0.002 |
Lu | -0.002 |
firefighters | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
ret | +0.014 |
orse | +0.013 |
orses | +0.013 |
ets | +0.012 |
peanut | +0.012 |
Neg logit contributions | |
Lu | -0.015 |
angels | -0.012 |
Lu | -0.012 |
Ann | -0.011 |
firefighters | -0.011 |
Token | It |
Token ID | 632 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
ret | +0.086 |
orses | +0.084 |
ets | +0.083 |
orse | +0.083 |
pler | +0.070 |
Neg logit contributions | |
Lu | -0.094 |
Lucy | -0.084 |
Lu | -0.075 |
Stev | -0.073 |
Kitty | -0.073 |
Token | looked |
Token ID | 3114 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
ret | +0.136 |
ets | +0.129 |
orses | +0.128 |
arding | +0.127 |
pieces | +0.127 |
Neg logit contributions | |
Lu | -0.055 |
Lu | -0.041 |
Lucy | -0.041 |
jack | -0.038 |
Bird | -0.037 |
Token | so |
Token ID | 523 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
orse | +0.286 |
orses | +0.275 |
ork | +0.259 |
pler | +0.258 |
stand | +0.253 |
Neg logit contributions | |
Lucy | -0.045 |
Lu | -0.044 |
too | -0.033 |
all | -0.032 |
along | -0.027 |
Token | tasty |
Token ID | 25103 |
Feature activation | -9.12e-02 |
Pos logit contributions | |
pler | +0.131 |
ret | +0.126 |
orse | +0.125 |
ork | +0.123 |
ets | +0.123 |
Neg logit contributions | |
Lu | -0.077 |
Lucy | -0.067 |
angels | -0.062 |
youths | -0.052 |
Lu | -0.051 |
Token | that |
Token ID | 326 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ret | +0.003 |
orse | +0.003 |
orses | +0.003 |
peanut | +0.003 |
ets | +0.003 |
Neg logit contributions | |
Lu | -0.005 |
angels | -0.004 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
ret | +0.039 |
orse | +0.038 |
orses | +0.036 |
ets | +0.035 |
ocus | +0.033 |
Neg logit contributions | |
Lu | -0.055 |
angels | -0.047 |
Lucy | -0.044 |
firefighters | -0.044 |
Lu | -0.043 |
Token | lion |
Token ID | 18744 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ret | +0.096 |
ets | +0.091 |
ocus | +0.086 |
orse | +0.085 |
orses | +0.085 |
Neg logit contributions | |
Lu | -0.078 |
angels | -0.067 |
Lu | -0.064 |
firefighters | -0.064 |
firefighter | -0.061 |
Token | big |
Token ID | 1263 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ret | +0.099 |
ets | +0.092 |
orse | +0.090 |
sel | +0.087 |
orses | +0.087 |
Neg logit contributions | |
Lu | -0.060 |
Lu | -0.046 |
angels | -0.044 |
youths | -0.043 |
firefighters | -0.043 |
Token | ocean |
Token ID | 9151 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ret | +0.066 |
orse | +0.061 |
orses | +0.058 |
ets | +0.057 |
sel | +0.055 |
Neg logit contributions | |
Lu | -0.052 |
angels | -0.039 |
youths | -0.039 |
Taxi | -0.038 |
firefighters | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
ret | +0.067 |
orse | +0.061 |
orses | +0.058 |
ocket | +0.055 |
ets | +0.055 |
Neg logit contributions | |
Lu | -0.049 |
angels | -0.041 |
firefighters | -0.038 |
youths | -0.038 |
Lu | -0.038 |
Token | One |
Token ID | 1881 |
Feature activation | -8.74e-01 |
Pos logit contributions | |
ret | +0.146 |
orse | +0.142 |
pler | +0.139 |
orses | +0.139 |
ocket | +0.138 |
Neg logit contributions | |
Lu | -0.072 |
Lucy | -0.051 |
Jenkins | -0.051 |
youths | -0.049 |
Barber | -0.046 |
Token | day |
Token ID | 1110 |
Feature activation | -3.37e+00 |
Pos logit contributions | |
ret | +0.046 |
orse | +0.042 |
orses | +0.041 |
sel | +0.040 |
ets | +0.040 |
Neg logit contributions | |
Lu | -0.032 |
Lu | -0.024 |
angels | -0.023 |
Ann | -0.023 |
firefighters | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
ret | +0.200 |
generation | +0.193 |
ocket | +0.189 |
ets | +0.188 |
pieces | +0.186 |
Neg logit contributions | |
Lu | -0.071 |
youths | -0.045 |
Lucy | -0.043 |
angels | -0.040 |
bud | -0.039 |
Token | P |
Token ID | 350 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
pler | +0.207 |
ret | +0.203 |
ets | +0.202 |
ocket | +0.195 |
ox | +0.195 |
Neg logit contributions | |
Lu | -0.097 |
Bob | -0.085 |
Squirrel | -0.083 |
Lucy | -0.083 |
Beaver | -0.082 |
Token | inch |
Token ID | 8589 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
orse | +0.077 |
orses | +0.069 |
pler | +0.063 |
bs | +0.057 |
ret | +0.056 |
Neg logit contributions | |
Lu | -0.118 |
youths | -0.102 |
angels | -0.102 |
ators | -0.097 |
Lu | -0.097 |
Token | y |
Token ID | 88 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
ret | +0.092 |
orse | +0.091 |
orses | +0.088 |
ocket | +0.085 |
peanut | +0.083 |
Neg logit contributions | |
Lu | -0.059 |
angels | -0.048 |
guard | -0.045 |
Lu | -0.045 |
Ann | -0.044 |
Token | met |
Token ID | 1138 |
Feature activation | +9.11e-02 |
Pos logit contributions | |
ret | +0.093 |
orse | +0.082 |
orses | +0.080 |
ocket | +0.079 |
ifying | +0.078 |
Neg logit contributions | |
Lu | -0.081 |
youths | -0.061 |
Lu | -0.061 |
firefighters | -0.061 |
Ann | -0.060 |
Token | a |
Token ID | 257 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
Lu | +0.004 |
angels | +0.003 |
Ann | +0.003 |
Lu | +0.003 |
firefighters | +0.003 |
Neg logit contributions | |
ret | -0.004 |
orse | -0.004 |
orses | -0.003 |
peanut | -0.003 |
ets | -0.003 |
Token | eating |
Token ID | 6600 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Lu | +0.023 |
Lu | +0.019 |
Ann | +0.019 |
firefighters | +0.018 |
angels | +0.018 |
Neg logit contributions | |
orse | -0.020 |
ret | -0.020 |
orses | -0.019 |
peanut | -0.018 |
ets | -0.017 |
Token | bamboo |
Token ID | 37252 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
Lu | +0.075 |
attendant | +0.066 |
Ann | +0.065 |
Michael | +0.063 |
ators | +0.061 |
Neg logit contributions | |
peanut | -0.058 |
ol | -0.054 |
ret | -0.052 |
orse | -0.050 |
orses | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
Lu | +0.044 |
Ann | +0.036 |
firefighters | +0.035 |
Lu | +0.035 |
Michael | +0.035 |
Neg logit contributions | |
ret | -0.027 |
orse | -0.025 |
orses | -0.024 |
ets | -0.024 |
peanut | -0.023 |
Token | One |
Token ID | 1881 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
ret | +0.091 |
orse | +0.089 |
orses | +0.086 |
ocket | +0.083 |
ets | +0.083 |
Neg logit contributions | |
Lu | -0.060 |
Lucy | -0.043 |
Lu | -0.041 |
Jenkins | -0.040 |
Ann | -0.040 |
Token | day |
Token ID | 1110 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
Lu | +0.022 |
angels | +0.018 |
firefighters | +0.018 |
Ann | +0.017 |
youths | +0.017 |
Neg logit contributions | |
orse | -0.016 |
ret | -0.016 |
orses | -0.015 |
peanut | -0.014 |
ets | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
ret | +0.200 |
generation | +0.194 |
ocket | +0.190 |
ets | +0.188 |
pieces | +0.186 |
Neg logit contributions | |
Lu | -0.072 |
Lucy | -0.045 |
Olivia | -0.043 |
Lu | -0.035 |
bud | -0.034 |
Token | P |
Token ID | 350 |
Feature activation | -9.92e-01 |
Pos logit contributions | |
ret | +0.213 |
ets | +0.210 |
pler | +0.204 |
orses | +0.198 |
ience | +0.197 |
Neg logit contributions | |
Lu | -0.101 |
Lucy | -0.094 |
Bunny | -0.090 |
Olivia | -0.089 |
Squirrel | -0.084 |
Token | andy |
Token ID | 10757 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
orse | +0.040 |
ret | +0.038 |
orses | +0.037 |
peanut | +0.034 |
ocket | +0.032 |
Neg logit contributions | |
Lu | -0.047 |
angels | -0.039 |
Ann | -0.038 |
Lu | -0.037 |
co | -0.036 |
Token | met |
Token ID | 1138 |
Feature activation | +8.24e-01 |
Pos logit contributions | |
ret | +0.027 |
orse | +0.026 |
orses | +0.024 |
ocket | +0.022 |
peanut | +0.022 |
Neg logit contributions | |
Lu | -0.038 |
angels | -0.031 |
Ann | -0.031 |
Lu | -0.031 |
firefighters | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
Lu | +0.044 |
Ann | +0.037 |
angels | +0.037 |
attendant | +0.036 |
firefighters | +0.036 |
Neg logit contributions | |
ret | -0.028 |
orse | -0.028 |
peanut | -0.026 |
orses | -0.025 |
ocket | -0.022 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
Lu | +0.008 |
angels | +0.006 |
Ann | +0.006 |
Lu | +0.006 |
firefighters | +0.006 |
Neg logit contributions | |
orse | -0.007 |
ret | -0.007 |
orses | -0.007 |
peanut | -0.006 |
ets | -0.006 |
Token | every |
Token ID | 790 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
Lu | +0.015 |
Ann | +0.012 |
angels | +0.012 |
firefighters | +0.012 |
Lu | +0.012 |
Neg logit contributions | |
ret | -0.011 |
orse | -0.011 |
orses | -0.010 |
peanut | -0.010 |
ets | -0.010 |
Token | day |
Token ID | 1110 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Lu | +0.016 |
angels | +0.013 |
Ann | +0.013 |
Lu | +0.013 |
firefighters | +0.012 |
Neg logit contributions | |
orse | -0.013 |
ret | -0.012 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
ret | +0.064 |
orse | +0.060 |
orses | +0.057 |
peanut | +0.054 |
sel | +0.053 |
Neg logit contributions | |
Lu | -0.046 |
angels | -0.035 |
Lu | -0.034 |
Lucy | -0.033 |
youths | -0.032 |
Token | One |
Token ID | 1881 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
orse | +0.074 |
orses | +0.069 |
ret | +0.069 |
ets | +0.063 |
ocket | +0.063 |
Neg logit contributions | |
Lu | -0.057 |
angels | -0.042 |
Lucy | -0.042 |
youths | -0.042 |
Lu | -0.041 |
Token | day |
Token ID | 1110 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
ret | +0.010 |
orse | +0.010 |
orses | +0.010 |
ets | +0.009 |
peanut | +0.009 |
Neg logit contributions | |
Lu | -0.010 |
angels | -0.008 |
Lu | -0.008 |
Ann | -0.008 |
firefighters | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
orse | +0.158 |
orses | +0.153 |
ret | +0.152 |
battlefield | +0.148 |
ets | +0.148 |
Neg logit contributions | |
Lu | -0.079 |
Lucy | -0.057 |
angels | -0.055 |
youths | -0.048 |
Pen | -0.048 |
Token | while |
Token ID | 981 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
orse | +0.198 |
orses | +0.193 |
oc | +0.187 |
ables | +0.174 |
ets | +0.174 |
Neg logit contributions | |
Lu | -0.131 |
angels | -0.123 |
Lucy | -0.123 |
Zoe | -0.118 |
Megan | -0.111 |
Token | Av |
Token ID | 5184 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
ret | +0.081 |
ets | +0.080 |
orse | +0.072 |
ocado | +0.070 |
sel | +0.070 |
Neg logit contributions | |
Lu | -0.068 |
angels | -0.057 |
youths | -0.052 |
Lucy | -0.050 |
Barber | -0.049 |
Token | a |
Token ID | 64 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
orse | +0.082 |
ret | +0.080 |
peanut | +0.078 |
orses | +0.075 |
lens | +0.073 |
Neg logit contributions | |
Lu | -0.074 |
Lu | -0.061 |
Ann | -0.060 |
youths | -0.059 |
angels | -0.059 |
Token | was |
Token ID | 373 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
ret | +0.014 |
orse | +0.013 |
orses | +0.012 |
peanut | +0.012 |
ets | +0.011 |
Neg logit contributions | |
Lu | -0.014 |
angels | -0.011 |
Ann | -0.011 |
Lu | -0.011 |
firefighters | -0.011 |
Token | playing |
Token ID | 2712 |
Feature activation | +8.97e-01 |
Pos logit contributions | |
ret | +0.001 |
orse | +0.001 |
orses | +0.001 |
peanut | +0.001 |
ets | +0.001 |
Neg logit contributions | |
Lu | -0.001 |
angels | -0.001 |
Ann | -0.001 |
Lu | -0.001 |
firefighters | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
ret | +0.043 |
orse | +0.041 |
orses | +0.038 |
ocket | +0.036 |
ork | +0.036 |
Neg logit contributions | |
Lu | -0.036 |
angels | -0.031 |
youths | -0.029 |
firefighters | -0.028 |
Lu | -0.026 |
Token | eat |
Token ID | 4483 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
ret | +0.015 |
orse | +0.014 |
orses | +0.013 |
ets | +0.012 |
peanut | +0.012 |
Neg logit contributions | |
Lu | -0.016 |
angels | -0.013 |
Ann | -0.012 |
firefighters | -0.012 |
Lu | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ret | +0.021 |
orse | +0.020 |
orses | +0.019 |
ets | +0.017 |
ocket | +0.017 |
Neg logit contributions | |
Lu | -0.020 |
angels | -0.017 |
Lu | -0.016 |
youths | -0.015 |
firefighters | -0.015 |
Token | One |
Token ID | 1881 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
orse | +0.070 |
ret | +0.069 |
orses | +0.066 |
ocket | +0.065 |
ocus | +0.063 |
Neg logit contributions | |
Lu | -0.049 |
Barber | -0.035 |
Ann | -0.034 |
Lucy | -0.033 |
angels | -0.033 |
Token | day |
Token ID | 1110 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
ret | +0.012 |
orse | +0.011 |
orses | +0.011 |
ets | +0.010 |
peanut | +0.010 |
Neg logit contributions | |
Lu | -0.011 |
angels | -0.009 |
Ann | -0.009 |
Lu | -0.009 |
firefighters | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
ret | +0.193 |
generation | +0.187 |
ocket | +0.185 |
orse | +0.183 |
orses | +0.182 |
Neg logit contributions | |
Lu | -0.077 |
youths | -0.045 |
angels | -0.044 |
Lucy | -0.044 |
Olivia | -0.043 |
Token | P |
Token ID | 350 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
ret | +0.204 |
pler | +0.202 |
ets | +0.200 |
orses | +0.197 |
ocket | +0.196 |
Neg logit contributions | |
Lu | -0.095 |
whe | -0.076 |
Lucy | -0.073 |
Olivia | -0.072 |
Squirrel | -0.071 |
Token | ingu |
Token ID | 6680 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
orse | +0.080 |
orses | +0.069 |
ret | +0.061 |
pler | +0.061 |
peanut | +0.059 |
Neg logit contributions | |
Lu | -0.104 |
angels | -0.092 |
youths | -0.088 |
Barber | -0.088 |
ators | -0.084 |
Token | was |
Token ID | 373 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
ret | +0.057 |
orse | +0.052 |
orses | +0.049 |
ocket | +0.048 |
ets | +0.047 |
Neg logit contributions | |
Lu | -0.063 |
angels | -0.051 |
firefighters | -0.051 |
Lu | -0.050 |
youths | -0.050 |
Token | feeling |
Token ID | 4203 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
ret | +0.006 |
orse | +0.006 |
orses | +0.005 |
peanut | +0.005 |
ets | +0.005 |
Neg logit contributions | |
Lu | -0.006 |
angels | -0.005 |
Ann | -0.005 |
Lu | -0.005 |
firefighters | -0.005 |
Token | tired |
Token ID | 10032 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
ret | +0.020 |
orse | +0.019 |
orses | +0.018 |
ets | +0.016 |
ork | +0.016 |
Neg logit contributions | |
Lu | -0.022 |
angels | -0.018 |
firefighters | -0.017 |
youths | -0.017 |
Lu | -0.017 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
ret | +0.053 |
orse | +0.049 |
orses | +0.046 |
ocket | +0.043 |
ifying | +0.042 |
Neg logit contributions | |
Lu | -0.060 |
Ann | -0.049 |
Lu | -0.048 |
firefighters | -0.048 |
youths | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
Lu | +0.011 |
angels | +0.009 |
Ann | +0.009 |
Lu | +0.009 |
youths | +0.008 |
Neg logit contributions | |
orse | -0.008 |
ret | -0.008 |
orses | -0.008 |
peanut | -0.008 |
ets | -0.007 |
Token | new |
Token ID | 649 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ret | +0.019 |
orse | +0.019 |
orses | +0.018 |
ets | +0.016 |
ocket | +0.016 |
Neg logit contributions | |
Lu | -0.019 |
angels | -0.015 |
Lu | -0.015 |
firefighters | -0.015 |
Ann | -0.015 |
Token | toy |
Token ID | 13373 |
Feature activation | -7.51e-01 |
Pos logit contributions | |
Lu | +0.054 |
Ann | +0.045 |
Lu | +0.043 |
Michael | +0.043 |
angels | +0.042 |
Neg logit contributions | |
orse | -0.046 |
ret | -0.045 |
peanut | -0.044 |
ork | -0.041 |
orses | -0.040 |
Token | in |
Token ID | 287 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
ret | +0.034 |
orse | +0.033 |
orses | +0.032 |
ets | +0.029 |
peanut | +0.028 |
Neg logit contributions | |
Lu | -0.032 |
angels | -0.027 |
firefighters | -0.026 |
Ann | -0.025 |
youths | -0.025 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
Lu | +0.026 |
angels | +0.022 |
Ann | +0.021 |
Lu | +0.021 |
co | +0.021 |
Neg logit contributions | |
ret | -0.020 |
orse | -0.020 |
peanut | -0.018 |
orses | -0.018 |
notebooks | -0.016 |
Token | 's |
Token ID | 338 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ret | +0.167 |
orse | +0.134 |
AW | +0.132 |
sel | +0.127 |
orses | +0.126 |
Neg logit contributions | |
Lu | -0.154 |
co | -0.140 |
Lucy | -0.126 |
angels | -0.125 |
Pen | -0.125 |
Token | part |
Token ID | 636 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
orse | +0.073 |
ret | +0.073 |
orses | +0.071 |
sel | +0.065 |
ets | +0.064 |
Neg logit contributions | |
Lu | -0.035 |
angels | -0.025 |
firefighters | -0.024 |
attendant | -0.024 |
taxis | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
Lu | +0.006 |
angels | +0.005 |
Ann | +0.005 |
Lu | +0.005 |
firefighters | +0.005 |
Neg logit contributions | |
ret | -0.007 |
orse | -0.007 |
orses | -0.007 |
peanut | -0.006 |
ets | -0.006 |
Token | It |
Token ID | 632 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
ret | +0.005 |
orse | +0.005 |
orses | +0.005 |
peanut | +0.004 |
ets | +0.004 |
Neg logit contributions | |
Lu | -0.006 |
angels | -0.005 |
Ann | -0.005 |
Lu | -0.005 |
firefighters | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | -6.85e-01 |
Pos logit contributions | |
Lu | +0.006 |
angels | +0.005 |
Ann | +0.004 |
Lu | +0.004 |
firefighters | +0.004 |
Neg logit contributions | |
ret | -0.005 |
orse | -0.005 |
orses | -0.004 |
peanut | -0.004 |
ets | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
ret | +0.077 |
orse | +0.075 |
ets | +0.074 |
peanut | +0.073 |
sel | +0.072 |
Neg logit contributions | |
Lu | -0.058 |
Ann | -0.046 |
Lu | -0.045 |
attendant | -0.041 |
angels | -0.041 |
Token | From |
Token ID | 4863 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
ret | +0.167 |
orse | +0.163 |
ocket | +0.162 |
mouths | +0.153 |
ulp | +0.152 |
Neg logit contributions | |
Ann | -0.076 |
Lu | -0.074 |
Lu | -0.071 |
attendant | -0.070 |
youths | -0.067 |
Token | that |
Token ID | 326 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
ret | +0.006 |
orse | +0.006 |
orses | +0.006 |
peanut | +0.005 |
ets | +0.005 |
Neg logit contributions | |
Lu | -0.012 |
angels | -0.011 |
Ann | -0.010 |
Lu | -0.010 |
firefighters | -0.010 |
Token | day |
Token ID | 1110 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
ret | +0.022 |
orse | +0.022 |
orses | +0.021 |
ets | +0.020 |
peanut | +0.019 |
Neg logit contributions | |
Lu | -0.019 |
angels | -0.015 |
Lu | -0.015 |
Ann | -0.015 |
firefighters | -0.014 |
Token | on |
Token ID | 319 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ret | +0.060 |
orse | +0.060 |
orses | +0.057 |
peanut | +0.054 |
ets | +0.053 |
Neg logit contributions | |
Lu | -0.039 |
angels | -0.030 |
Ann | -0.028 |
Lu | -0.027 |
youths | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
ret | +0.051 |
orse | +0.050 |
orses | +0.048 |
ets | +0.047 |
sel | +0.046 |
Neg logit contributions | |
Lu | -0.042 |
angels | -0.033 |
Lucy | -0.032 |
Luis | -0.031 |
firefighters | -0.031 |
Token | Jolly |
Token ID | 40276 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
ifying | +0.154 |
orse | +0.154 |
ork | +0.153 |
ocket | +0.152 |
ox | +0.151 |
Neg logit contributions | |
youths | -0.127 |
helpers | -0.121 |
angels | -0.116 |
Lu | -0.114 |
Wise | -0.111 |
Token | and |
Token ID | 290 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
ret | +0.028 |
orse | +0.028 |
orses | +0.026 |
peanut | +0.024 |
ets | +0.023 |
Neg logit contributions | |
Lu | -0.031 |
angels | -0.026 |
Lu | -0.025 |
Ann | -0.025 |
firefighters | -0.025 |
Token | his |
Token ID | 465 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
ret | +0.081 |
orse | +0.081 |
orses | +0.078 |
ocus | +0.077 |
ifying | +0.076 |
Neg logit contributions | |
Lu | -0.042 |
youths | -0.034 |
firefighters | -0.033 |
angels | -0.033 |
Lucy | -0.030 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Lu | +0.023 |
Ann | +0.018 |
Lu | +0.016 |
Michael | +0.015 |
Taxi | +0.015 |
Neg logit contributions | |
ret | -0.030 |
orse | -0.030 |
peanut | -0.028 |
orses | -0.028 |
notebooks | -0.026 |
Token | were |
Token ID | 547 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Lu | +0.056 |
Ann | +0.047 |
angels | +0.047 |
Lu | +0.046 |
attendant | +0.046 |
Neg logit contributions | |
ret | -0.033 |
orse | -0.033 |
orses | -0.030 |
ets | -0.028 |
peanut | -0.028 |
Token | make |
Token ID | 787 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
Lu | +0.004 |
angels | +0.003 |
Ann | +0.003 |
Lu | +0.003 |
firefighters | +0.003 |
Neg logit contributions | |
ret | -0.003 |
orse | -0.003 |
orses | -0.003 |
peanut | -0.003 |
ets | -0.003 |
Token | funny |
Token ID | 8258 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
orse | +0.057 |
ret | +0.057 |
orses | +0.053 |
ocket | +0.050 |
ets | +0.049 |
Neg logit contributions | |
Lu | -0.055 |
angels | -0.047 |
firefighters | -0.042 |
Ann | -0.040 |
Lucy | -0.039 |
Token | noises |
Token ID | 26782 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
ret | +0.012 |
orse | +0.012 |
orses | +0.011 |
peanut | +0.010 |
ets | +0.010 |
Neg logit contributions | |
Lu | -0.012 |
angels | -0.010 |
Ann | -0.009 |
Lu | -0.009 |
firefighters | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
ret | +0.026 |
orse | +0.025 |
orses | +0.024 |
peanut | +0.022 |
ork | +0.021 |
Neg logit contributions | |
Lu | -0.026 |
angels | -0.021 |
firefighters | -0.020 |
Ann | -0.020 |
Lu | -0.020 |
Token | So |
Token ID | 1406 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
orse | +0.125 |
orses | +0.119 |
ret | +0.117 |
peanut | +0.112 |
ocket | +0.112 |
Neg logit contributions | |
Lu | -0.074 |
attendant | -0.052 |
firefighters | -0.049 |
Barber | -0.048 |
Ann | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
iche | +0.175 |
orse | +0.167 |
ocado | +0.166 |
orses | +0.166 |
ret | +0.162 |
Neg logit contributions | |
Lu | -0.084 |
attendant | -0.061 |
helpers | -0.061 |
Luis | -0.058 |
Ann | -0.055 |
Token | he |
Token ID | 339 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
iche | +0.203 |
ocado | +0.201 |
ork | +0.194 |
peanut | +0.192 |
orse | +0.190 |
Neg logit contributions | |
Lu | -0.100 |
attendant | -0.075 |
Luis | -0.075 |
helpers | -0.074 |
angels | -0.069 |
Token | kept |
Token ID | 4030 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
Lu | +0.028 |
angels | +0.023 |
Ann | +0.023 |
youths | +0.022 |
firefighters | +0.022 |
Neg logit contributions | |
orse | -0.019 |
ret | -0.019 |
orses | -0.017 |
peanut | -0.017 |
ets | -0.016 |
Token | rubbing |
Token ID | 27714 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
Lu | +0.022 |
Ann | +0.018 |
Lu | +0.018 |
angels | +0.018 |
firefighters | +0.017 |
Neg logit contributions | |
orse | -0.019 |
ret | -0.018 |
orses | -0.017 |
peanut | -0.016 |
ets | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
ret | +0.014 |
orse | +0.014 |
orses | +0.013 |
peanut | +0.012 |
ets | +0.012 |
Neg logit contributions | |
Lu | -0.015 |
angels | -0.012 |
Ann | -0.011 |
Lu | -0.011 |
firefighters | -0.011 |
Token | rubbing |
Token ID | 27714 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
ret | +0.017 |
orse | +0.017 |
orses | +0.016 |
peanut | +0.015 |
ets | +0.015 |
Neg logit contributions | |
Lu | -0.017 |
angels | -0.014 |
Ann | -0.014 |
Lu | -0.013 |
firefighters | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
Lu | +0.060 |
Ann | +0.053 |
Lu | +0.051 |
youths | +0.050 |
attendant | +0.050 |
Neg logit contributions | |
orse | -0.036 |
peanut | -0.035 |
ret | -0.035 |
orses | -0.032 |
ork | -0.029 |
Token | break |
Token ID | 2270 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Lu | +0.021 |
Ann | +0.018 |
angels | +0.017 |
Lu | +0.017 |
youths | +0.017 |
Neg logit contributions | |
orse | -0.015 |
ret | -0.015 |
orses | -0.014 |
peanut | -0.013 |
ork | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
Lu | +0.055 |
Ann | +0.048 |
Lu | +0.047 |
angels | +0.046 |
attendant | +0.045 |
Neg logit contributions | |
ret | -0.033 |
orse | -0.031 |
ets | -0.029 |
orses | -0.029 |
peanut | -0.028 |
Token | One |
Token ID | 1881 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
orse | +0.109 |
ets | +0.108 |
ocket | +0.107 |
ret | +0.107 |
orses | +0.106 |
Neg logit contributions | |
Lu | -0.067 |
attendant | -0.046 |
Jenkins | -0.046 |
Barber | -0.046 |
Lucy | -0.045 |
Token | day |
Token ID | 1110 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
Lu | +0.020 |
angels | +0.017 |
Ann | +0.017 |
youths | +0.016 |
firefighters | +0.016 |
Neg logit contributions | |
orse | -0.016 |
ret | -0.015 |
orses | -0.014 |
peanut | -0.014 |
ork | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
ret | +0.184 |
ets | +0.180 |
ocket | +0.179 |
generation | +0.176 |
ocado | +0.175 |
Neg logit contributions | |
Lu | -0.070 |
Olivia | -0.040 |
Lucy | -0.040 |
Pupp | -0.037 |
Dog | -0.036 |
Token | his |
Token ID | 465 |
Feature activation | +9.11e-01 |
Pos logit contributions | |
ets | +0.221 |
pler | +0.208 |
ocus | +0.208 |
ocket | +0.201 |
ocado | +0.201 |
Neg logit contributions | |
Lu | -0.085 |
Squirrel | -0.082 |
Bunny | -0.074 |
his | -0.073 |
Olivia | -0.071 |
Token | friend |
Token ID | 1545 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
Lu | +0.037 |
Lu | +0.031 |
Ann | +0.030 |
angels | +0.027 |
Karen | +0.027 |
Neg logit contributions | |
orse | -0.043 |
ret | -0.043 |
peanut | -0.040 |
orses | -0.040 |
ork | -0.036 |
Token | Sammy |
Token ID | 44067 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
ret | +0.023 |
orse | +0.022 |
orses | +0.021 |
ets | +0.020 |
ocket | +0.019 |
Neg logit contributions | |
Lu | -0.026 |
Lu | -0.021 |
Ann | -0.021 |
angels | -0.021 |
firefighters | -0.020 |
Token | came |
Token ID | 1625 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ret | +0.020 |
orse | +0.019 |
orses | +0.018 |
ets | +0.017 |
peanut | +0.016 |
Neg logit contributions | |
Lu | -0.026 |
Lu | -0.021 |
angels | -0.021 |
Ann | -0.021 |
firefighters | -0.020 |
Token | over |
Token ID | 625 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
Lu | +0.074 |
Ann | +0.062 |
Lu | +0.060 |
angels | +0.060 |
pastor | +0.059 |
Neg logit contributions | |
ret | -0.052 |
orse | -0.050 |
orses | -0.046 |
's | -0.045 |
peanut | -0.045 |
Token | was |
Token ID | 373 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ret | +0.048 |
orse | +0.047 |
orses | +0.046 |
ets | +0.044 |
ocket | +0.041 |
Neg logit contributions | |
Lu | -0.038 |
Ann | -0.029 |
Lu | -0.029 |
angels | -0.028 |
attendant | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
ret | +0.084 |
orse | +0.084 |
orses | +0.080 |
ets | +0.075 |
oc | +0.071 |
Neg logit contributions | |
Lu | -0.057 |
firefighters | -0.047 |
angels | -0.046 |
attendant | -0.043 |
Lucy | -0.042 |
Token | big |
Token ID | 1263 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
ret | +0.138 |
ets | +0.131 |
orses | +0.128 |
orse | +0.128 |
ocus | +0.116 |
Neg logit contributions | |
Lu | -0.108 |
attendant | -0.099 |
angels | -0.098 |
firefighter | -0.098 |
firefighters | -0.097 |
Token | card |
Token ID | 2657 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ret | +0.098 |
orse | +0.094 |
orses | +0.092 |
ets | +0.083 |
earing | +0.077 |
Neg logit contributions | |
Lu | -0.082 |
angels | -0.072 |
attendant | -0.070 |
angel | -0.070 |
firefighters | -0.070 |
Token | with |
Token ID | 351 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
ret | +0.081 |
orse | +0.074 |
orses | +0.073 |
ets | +0.066 |
ocus | +0.064 |
Neg logit contributions | |
Lu | -0.064 |
angels | -0.054 |
attendant | -0.052 |
firefighters | -0.051 |
Lu | -0.051 |
Token | a |
Token ID | 257 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
orse | +0.197 |
ret | +0.186 |
orses | +0.179 |
ossom | +0.172 |
ork | +0.171 |
Neg logit contributions | |
Lu | -0.085 |
angels | -0.084 |
firefighters | -0.082 |
Lucy | -0.067 |
youths | -0.065 |
Token | picture |
Token ID | 4286 |
Feature activation | -8.54e-01 |
Pos logit contributions | |
ret | +0.163 |
ets | +0.147 |
orse | +0.145 |
orses | +0.140 |
earing | +0.136 |
Neg logit contributions | |
Lu | -0.146 |
angels | -0.140 |
attendant | -0.135 |
firefighters | -0.132 |
firefighter | -0.130 |
Token | of |
Token ID | 286 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ret | +0.046 |
orse | +0.045 |
orses | +0.043 |
peanut | +0.040 |
ets | +0.039 |
Neg logit contributions | |
Lu | -0.029 |
angels | -0.023 |
firefighters | -0.021 |
Ann | -0.021 |
Lu | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
ret | +0.081 |
orse | +0.080 |
orses | +0.075 |
ets | +0.073 |
ork | +0.070 |
Neg logit contributions | |
Lu | -0.056 |
angels | -0.048 |
firefighters | -0.046 |
Lucy | -0.042 |
youths | -0.041 |
Token | rainbow |
Token ID | 27223 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
ret | +0.139 |
orse | +0.123 |
ets | +0.121 |
orses | +0.117 |
earing | +0.116 |
Neg logit contributions | |
Lu | -0.125 |
angels | -0.116 |
firefighters | -0.113 |
attendant | -0.109 |
firefighter | -0.108 |
Token | . |
Token ID | 13 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
ret | +0.118 |
orse | +0.112 |
orses | +0.107 |
fortable | +0.095 |
ocket | +0.094 |
Neg logit contributions | |
Lu | -0.078 |
angels | -0.069 |
co | -0.060 |
attendant | -0.060 |
firefighters | -0.060 |
Token | chase |
Token ID | 15505 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
Lu | +0.009 |
angels | +0.008 |
Ann | +0.008 |
Lu | +0.008 |
firefighters | +0.007 |
Neg logit contributions | |
orse | -0.007 |
ret | -0.007 |
orses | -0.007 |
peanut | -0.006 |
ets | -0.006 |
Token | Jolly |
Token ID | 40276 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
orse | +0.162 |
ret | +0.152 |
inkle | +0.151 |
ocket | +0.149 |
orses | +0.149 |
Neg logit contributions | |
Lu | -0.084 |
youths | -0.078 |
firefighters | -0.056 |
lo | -0.054 |
Bob | -0.054 |
Token | and |
Token ID | 290 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
orse | +0.060 |
ret | +0.059 |
orses | +0.055 |
peanut | +0.052 |
ocket | +0.051 |
Neg logit contributions | |
Lu | -0.058 |
angels | -0.050 |
youths | -0.046 |
Lu | -0.045 |
co | -0.045 |
Token | his |
Token ID | 465 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
ret | +0.086 |
orse | +0.076 |
orses | +0.073 |
sel | +0.072 |
ocket | +0.071 |
Neg logit contributions | |
Lu | -0.070 |
firefighters | -0.060 |
angels | -0.058 |
Lucy | -0.056 |
jack | -0.053 |
Token | friends |
Token ID | 2460 |
Feature activation | -5.78e-03 |
Pos logit contributions | |
ret | +0.016 |
orse | +0.016 |
orses | +0.015 |
ets | +0.015 |
peanut | +0.014 |
Neg logit contributions | |
Lu | -0.008 |
angels | -0.006 |
firefighters | -0.006 |
Lu | -0.006 |
youths | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
ret | +0.000 |
orse | +0.000 |
orses | +0.000 |
peanut | +0.000 |
ets | +0.000 |
Neg logit contributions | |
Lu | -0.000 |
angels | -0.000 |
Ann | -0.000 |
Lu | -0.000 |
firefighters | -0.000 |
Token | They |
Token ID | 1119 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
ifying | +0.186 |
ocket | +0.184 |
ret | +0.182 |
orse | +0.176 |
ork | +0.170 |
Neg logit contributions | |
Lu | -0.102 |
youths | -0.101 |
Bob | -0.091 |
attendant | -0.087 |
Beaver | -0.082 |
Token | were |
Token ID | 547 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
Lu | +0.039 |
angels | +0.033 |
Ann | +0.033 |
Lu | +0.033 |
attendant | +0.031 |
Neg logit contributions | |
orse | -0.026 |
ret | -0.026 |
orses | -0.024 |
peanut | -0.023 |
ets | -0.022 |
Token | scared |
Token ID | 12008 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
Lu | +0.025 |
Lu | +0.020 |
Ann | +0.020 |
angels | +0.020 |
attendant | +0.019 |
Neg logit contributions | |
orse | -0.019 |
ret | -0.019 |
orses | -0.018 |
peanut | -0.017 |
ets | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
ret | +0.006 |
orse | +0.006 |
orses | +0.005 |
peanut | +0.005 |
ets | +0.005 |
Neg logit contributions | |
Lu | -0.007 |
angels | -0.005 |
Ann | -0.005 |
Lu | -0.005 |
firefighters | -0.005 |
Token | didn |
Token ID | 1422 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
ret | +0.035 |
orse | +0.033 |
orses | +0.032 |
ork | +0.030 |
peanut | +0.029 |
Neg logit contributions | |
Lu | -0.037 |
firefighters | -0.030 |
angels | -0.030 |
youths | -0.029 |
Ann | -0.028 |
Token | 's |
Token ID | 338 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
ret | +0.106 |
orse | +0.092 |
ork | +0.088 |
orses | +0.087 |
AW | +0.085 |
Neg logit contributions | |
Lu | -0.129 |
co | -0.116 |
Lucy | -0.113 |
ma | -0.112 |
Ann | -0.111 |
Token | purse |
Token ID | 27517 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
orse | +0.145 |
ret | +0.140 |
orses | +0.139 |
plets | +0.127 |
arding | +0.123 |
Neg logit contributions | |
Lu | -0.059 |
firefighters | -0.051 |
angels | -0.051 |
helpers | -0.050 |
ambulance | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ret | +0.049 |
orse | +0.048 |
orses | +0.046 |
peanut | +0.041 |
ets | +0.041 |
Neg logit contributions | |
Lu | -0.039 |
angels | -0.031 |
Ann | -0.030 |
Lu | -0.030 |
attendant | -0.030 |
Token | It |
Token ID | 632 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ret | +0.060 |
orse | +0.060 |
orses | +0.059 |
ets | +0.053 |
ocket | +0.051 |
Neg logit contributions | |
Lu | -0.047 |
youths | -0.033 |
Lucy | -0.033 |
Pen | -0.032 |
firefighters | -0.032 |
Token | has |
Token ID | 468 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
orse | +0.084 |
ret | +0.081 |
orses | +0.080 |
ets | +0.072 |
alks | +0.071 |
Neg logit contributions | |
Lu | -0.059 |
Ann | -0.046 |
Lucy | -0.045 |
angels | -0.045 |
Ann | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
orse | +0.129 |
ret | +0.124 |
orses | +0.119 |
inkle | +0.106 |
ossom | +0.106 |
Neg logit contributions | |
Lu | -0.072 |
angels | -0.064 |
firefighters | -0.062 |
taxis | -0.054 |
Ann | -0.053 |
Token | picture |
Token ID | 4286 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
ret | +0.165 |
orse | +0.155 |
ets | +0.154 |
orses | +0.154 |
iece | +0.139 |
Neg logit contributions | |
Lu | -0.131 |
angels | -0.128 |
angel | -0.124 |
attendant | -0.124 |
firefighter | -0.123 |
Token | of |
Token ID | 286 |
Feature activation | -9.49e-01 |
Pos logit contributions | |
ret | +0.024 |
orse | +0.024 |
orses | +0.022 |
peanut | +0.021 |
ets | +0.020 |
Neg logit contributions | |
Lu | -0.016 |
angels | -0.012 |
Ann | -0.012 |
Lu | -0.011 |
firefighters | -0.011 |
Token | his |
Token ID | 465 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
ret | +0.046 |
orse | +0.045 |
orses | +0.042 |
ets | +0.040 |
ork | +0.039 |
Neg logit contributions | |
Lu | -0.038 |
angels | -0.032 |
firefighters | -0.030 |
youths | -0.028 |
Ann | -0.028 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
ret | +0.122 |
orse | +0.117 |
orses | +0.111 |
sel | +0.106 |
ets | +0.105 |
Neg logit contributions | |
Lu | -0.080 |
angels | -0.075 |
firefighters | -0.065 |
helpers | -0.064 |
youths | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ret | +0.096 |
orse | +0.087 |
orses | +0.081 |
ork | +0.080 |
ulp | +0.076 |
Neg logit contributions | |
Lu | -0.127 |
Lucy | -0.108 |
Ann | -0.105 |
Ann | -0.105 |
co | -0.104 |
Token | card |
Token ID | 2657 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ret | +0.041 |
orse | +0.040 |
orses | +0.039 |
ets | +0.037 |
ocus | +0.034 |
Neg logit contributions | |
Lu | -0.039 |
angels | -0.031 |
Lu | -0.030 |
firefighters | -0.030 |
attendant | -0.029 |
Token | on |
Token ID | 319 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
ret | +0.068 |
orse | +0.063 |
orses | +0.062 |
ets | +0.057 |
ocus | +0.055 |
Neg logit contributions | |
Lu | -0.049 |
angels | -0.041 |
firefighters | -0.040 |
Lu | -0.039 |
attendant | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
orse | +0.027 |
ret | +0.027 |
orses | +0.026 |
ets | +0.024 |
ocket | +0.023 |
Neg logit contributions | |
Lu | -0.023 |
Lu | -0.018 |
Ann | -0.018 |
angels | -0.018 |
firefighters | -0.017 |
Token | floor |
Token ID | 4314 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
orse | +0.140 |
ret | +0.138 |
orses | +0.133 |
bs | +0.129 |
ocus | +0.119 |
Neg logit contributions | |
Lu | -0.090 |
angels | -0.087 |
attendant | -0.087 |
firefighters | -0.082 |
angel | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
ret | +0.050 |
orse | +0.048 |
orses | +0.046 |
peanut | +0.043 |
ets | +0.042 |
Neg logit contributions | |
Lu | -0.040 |
angels | -0.032 |
Lu | -0.031 |
Ann | -0.031 |
firefighters | -0.030 |
Token | The |
Token ID | 383 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
Lu | +0.024 |
angels | +0.021 |
Ann | +0.020 |
Lu | +0.020 |
youths | +0.020 |
Neg logit contributions | |
orse | -0.013 |
ret | -0.013 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.010 |
Token | card |
Token ID | 2657 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
orse | +0.190 |
ret | +0.185 |
orses | +0.183 |
ience | +0.181 |
ifying | +0.176 |
Neg logit contributions | |
lady | -0.109 |
angels | -0.106 |
attendant | -0.103 |
firefighters | -0.100 |
helpers | -0.092 |
Token | had |
Token ID | 550 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
ret | +0.071 |
orse | +0.068 |
orses | +0.067 |
ets | +0.064 |
ifying | +0.059 |
Neg logit contributions | |
Lu | -0.055 |
angels | -0.046 |
attendant | -0.043 |
Lu | -0.043 |
Ann | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
orse | +0.156 |
orses | +0.147 |
ret | +0.145 |
oppy | +0.137 |
oc | +0.136 |
Neg logit contributions | |
firefighters | -0.061 |
Lu | -0.060 |
angels | -0.060 |
taxis | -0.044 |
Ann | -0.043 |
Token | big |
Token ID | 1263 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
ret | +0.162 |
ets | +0.160 |
orses | +0.159 |
orse | +0.158 |
iece | +0.143 |
Neg logit contributions | |
Lu | -0.093 |
angels | -0.092 |
bud | -0.087 |
attendant | -0.085 |
firefighters | -0.085 |
Token | , |
Token ID | 11 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
ret | +0.117 |
orses | +0.115 |
orse | +0.113 |
ets | +0.106 |
pieces | +0.098 |
Neg logit contributions | |
Lu | -0.086 |
bud | -0.085 |
angels | -0.083 |
angel | -0.079 |
firefighters | -0.075 |
Token | a |
Token ID | 257 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
Lu | +0.034 |
Lu | +0.028 |
Ann | +0.028 |
angels | +0.027 |
youths | +0.026 |
Neg logit contributions | |
ret | -0.024 |
orse | -0.024 |
orses | -0.022 |
peanut | -0.021 |
ets | -0.020 |
Token | big |
Token ID | 1263 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
ret | +0.010 |
orse | +0.010 |
orses | +0.009 |
ets | +0.008 |
peanut | +0.008 |
Neg logit contributions | |
Lu | -0.011 |
angels | -0.009 |
Ann | -0.009 |
Lu | -0.009 |
firefighters | -0.009 |
Token | ice |
Token ID | 4771 |
Feature activation | +7.45e-02 |
Pos logit contributions | |
Lu | +0.002 |
angels | +0.002 |
Ann | +0.002 |
Lu | +0.002 |
firefighters | +0.002 |
Neg logit contributions | |
ret | -0.002 |
orse | -0.002 |
orses | -0.002 |
peanut | -0.002 |
ets | -0.002 |
Token | cream |
Token ID | 8566 |
Feature activation | -7.93e-01 |
Pos logit contributions | |
Lu | +0.003 |
angels | +0.002 |
Ann | +0.002 |
Lu | +0.002 |
firefighters | +0.002 |
Neg logit contributions | |
ret | -0.004 |
orse | -0.004 |
orses | -0.004 |
peanut | -0.004 |
ets | -0.003 |
Token | in |
Token ID | 287 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
ret | +0.040 |
orse | +0.039 |
orses | +0.037 |
ets | +0.034 |
peanut | +0.032 |
Neg logit contributions | |
Lu | -0.031 |
angels | -0.025 |
firefighters | -0.024 |
Ann | -0.023 |
attendant | -0.023 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
ret | +0.018 |
orse | +0.018 |
orses | +0.017 |
ets | +0.016 |
peanut | +0.015 |
Neg logit contributions | |
Lu | -0.018 |
Ann | -0.014 |
angels | -0.014 |
Lu | -0.014 |
firefighters | -0.013 |
Token | 's |
Token ID | 338 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
ret | +0.160 |
AW | +0.137 |
orse | +0.124 |
alks | +0.123 |
ocus | +0.122 |
Neg logit contributions | |
Lu | -0.149 |
Pen | -0.137 |
youths | -0.127 |
Dog | -0.126 |
co | -0.123 |
Token | hand |
Token ID | 1021 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
ret | +0.100 |
orse | +0.096 |
orses | +0.093 |
arding | +0.084 |
ets | +0.083 |
Neg logit contributions | |
Lu | -0.063 |
firefighters | -0.051 |
angels | -0.050 |
attendant | -0.049 |
Taxi | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | -8.35e-01 |
Pos logit contributions | |
ret | +0.049 |
orse | +0.047 |
orses | +0.045 |
peanut | +0.041 |
ets | +0.039 |
Neg logit contributions | |
Lu | -0.042 |
angels | -0.034 |
Ann | -0.032 |
firefighters | -0.032 |
co | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
ret | +0.039 |
orse | +0.036 |
orses | +0.035 |
ets | +0.032 |
ork | +0.032 |
Neg logit contributions | |
Lu | -0.037 |
angels | -0.030 |
firefighters | -0.030 |
Ann | -0.028 |
youths | -0.027 |
Token | new |
Token ID | 649 |
Feature activation | +8.57e-01 |
Pos logit contributions | |
ret | +0.039 |
orse | +0.037 |
orses | +0.035 |
ets | +0.034 |
sel | +0.031 |
Neg logit contributions | |
Lu | -0.040 |
angels | -0.034 |
firefighters | -0.032 |
Lu | -0.032 |
Ann | -0.032 |
Token | puts |
Token ID | 7584 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
Lu | +0.017 |
angels | +0.014 |
Lu | +0.014 |
Ann | +0.013 |
firefighters | +0.013 |
Neg logit contributions | |
ret | -0.012 |
orse | -0.011 |
orses | -0.011 |
peanut | -0.010 |
ets | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
orse | +0.091 |
orses | +0.086 |
ret | +0.079 |
ets | +0.076 |
ocus | +0.073 |
Neg logit contributions | |
Lu | -0.068 |
angels | -0.056 |
Lucy | -0.052 |
firefighters | -0.051 |
Ann | -0.050 |
Token | shiny |
Token ID | 22441 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
orses | +0.123 |
ret | +0.123 |
orse | +0.122 |
ets | +0.120 |
ocus | +0.107 |
Neg logit contributions | |
Lu | -0.093 |
firefighter | -0.088 |
angels | -0.085 |
angel | -0.081 |
bud | -0.080 |
Token | sticker |
Token ID | 27735 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
orse | +0.104 |
orses | +0.104 |
ret | +0.098 |
cious | +0.088 |
ets | +0.085 |
Neg logit contributions | |
Lu | -0.086 |
angels | -0.074 |
firefighter | -0.071 |
angel | -0.070 |
firefighters | -0.069 |
Token | on |
Token ID | 319 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
orse | +0.076 |
ret | +0.074 |
orses | +0.073 |
peanut | +0.065 |
ets | +0.063 |
Neg logit contributions | |
Lu | -0.042 |
angels | -0.035 |
attendant | -0.031 |
Ann | -0.030 |
co | -0.030 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
orse | +0.144 |
orses | +0.141 |
ret | +0.134 |
ets | +0.131 |
ocus | +0.128 |
Neg logit contributions | |
Lu | -0.087 |
Lucy | -0.071 |
Ann | -0.064 |
Ann | -0.063 |
Lu | -0.063 |
Token | 's |
Token ID | 338 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
ret | +0.139 |
orse | +0.126 |
orses | +0.125 |
seas | +0.121 |
AW | +0.118 |
Neg logit contributions | |
Lu | -0.158 |
co | -0.139 |
Pen | -0.136 |
angels | -0.133 |
Lucy | -0.130 |
Token | folder |
Token ID | 9483 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ret | +0.115 |
orse | +0.115 |
orses | +0.114 |
ocus | +0.104 |
iche | +0.098 |
Neg logit contributions | |
angels | -0.096 |
Lu | -0.090 |
angel | -0.089 |
attendant | -0.082 |
firefighters | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ret | +0.056 |
orse | +0.056 |
orses | +0.053 |
peanut | +0.048 |
ocus | +0.045 |
Neg logit contributions | |
Lu | -0.050 |
angels | -0.040 |
Lu | -0.038 |
attendant | -0.038 |
Ann | -0.038 |
Token | Ben |
Token ID | 3932 |
Feature activation | -6.36e-01 |
Pos logit contributions | |
ret | +0.048 |
orse | +0.047 |
orses | +0.044 |
ets | +0.042 |
ocket | +0.041 |
Neg logit contributions | |
Lu | -0.043 |
Ann | -0.033 |
Lu | -0.032 |
Lucy | -0.032 |
youths | -0.032 |
Token | feels |
Token ID | 5300 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
orse | +0.026 |
ret | +0.026 |
orses | +0.024 |
peanut | +0.022 |
ork | +0.022 |
Neg logit contributions | |
Lu | -0.030 |
angels | -0.024 |
Ann | -0.024 |
youths | -0.023 |
firefighters | -0.023 |
Token | itty |
Token ID | 9760 |
Feature activation | -8.23e-02 |
Pos logit contributions | |
Lu | +0.041 |
angels | +0.031 |
firefighters | +0.031 |
Ann | +0.030 |
Lu | +0.029 |
Neg logit contributions | |
orse | -0.047 |
ret | -0.047 |
orses | -0.044 |
peanut | -0.041 |
ocus | -0.039 |
Token | never |
Token ID | 1239 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
ret | +0.003 |
orse | +0.003 |
orses | +0.003 |
peanut | +0.003 |
ets | +0.003 |
Neg logit contributions | |
Lu | -0.004 |
angels | -0.003 |
Ann | -0.003 |
Lu | -0.003 |
firefighters | -0.003 |
Token | delayed |
Token ID | 11038 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
ret | +0.015 |
orse | +0.015 |
orses | +0.014 |
peanut | +0.012 |
ork | +0.012 |
Neg logit contributions | |
Lu | -0.022 |
angels | -0.018 |
Ann | -0.018 |
Lu | -0.018 |
firefighters | -0.017 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
Lu | +0.005 |
angels | +0.004 |
Ann | +0.004 |
Lu | +0.004 |
firefighters | +0.004 |
Neg logit contributions | |
ret | -0.004 |
orse | -0.004 |
orses | -0.004 |
peanut | -0.004 |
ets | -0.004 |
Token | with |
Token ID | 351 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
Lu | +0.077 |
Ann | +0.067 |
Lu | +0.064 |
angels | +0.064 |
Taxi | +0.064 |
Neg logit contributions | |
orse | -0.039 |
ret | -0.038 |
peanut | -0.034 |
orses | -0.034 |
ork | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Lu | +0.012 |
angels | +0.010 |
Ann | +0.009 |
Lu | +0.009 |
firefighters | +0.009 |
Neg logit contributions | |
ret | -0.011 |
orse | -0.011 |
orses | -0.010 |
peanut | -0.010 |
ets | -0.009 |
Token | jar |
Token ID | 17379 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
Lu | +0.086 |
Ann | +0.073 |
Lu | +0.067 |
ators | +0.066 |
Michael | +0.064 |
Neg logit contributions | |
peanut | -0.081 |
ret | -0.077 |
orse | -0.075 |
orses | -0.072 |
ink | -0.069 |
Token | again |
Token ID | 757 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
Lu | +0.001 |
angels | +0.001 |
Ann | +0.001 |
Lu | +0.001 |
firefighters | +0.001 |
Neg logit contributions | |
ret | -0.001 |
orse | -0.001 |
orses | -0.001 |
peanut | -0.001 |
ets | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
ret | +0.014 |
orse | +0.014 |
orses | +0.013 |
peanut | +0.012 |
ets | +0.012 |
Neg logit contributions | |
Lu | -0.015 |
angels | -0.012 |
Ann | -0.012 |
Lu | -0.012 |
firefighters | -0.012 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.92e-01 |
Pos logit contributions | |
ret | +0.008 |
orse | +0.008 |
orses | +0.007 |
peanut | +0.007 |
ets | +0.007 |
Neg logit contributions | |
Lu | -0.011 |
angels | -0.009 |
Ann | -0.009 |
Lu | -0.009 |
firefighters | -0.009 |
Token | Once |
Token ID | 7454 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
Lu | +0.049 |
angels | +0.042 |
firefighters | +0.042 |
youths | +0.040 |
attendant | +0.039 |
Neg logit contributions | |
ret | -0.037 |
orse | -0.036 |
orses | -0.033 |
peanut | -0.031 |
ork | -0.030 |
Token | big |
Token ID | 1263 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
Lu | +0.004 |
angels | +0.003 |
Ann | +0.003 |
Lu | +0.003 |
firefighters | +0.003 |
Neg logit contributions | |
ret | -0.003 |
orse | -0.003 |
orses | -0.003 |
peanut | -0.003 |
ets | -0.003 |
Token | bathroom |
Token ID | 12436 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
Lu | +0.021 |
Ann | +0.017 |
angels | +0.017 |
Lu | +0.017 |
youths | +0.016 |
Neg logit contributions | |
orse | -0.018 |
ret | -0.017 |
orses | -0.016 |
peanut | -0.016 |
ets | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
ret | +0.023 |
orse | +0.023 |
orses | +0.021 |
peanut | +0.019 |
ets | +0.019 |
Neg logit contributions | |
Lu | -0.024 |
angels | -0.019 |
Lu | -0.019 |
firefighters | -0.019 |
Ann | -0.019 |
Token | A |
Token ID | 317 |
Feature activation | +7.00e-01 |
Pos logit contributions | |
Lu | +0.020 |
angels | +0.018 |
Ann | +0.017 |
Lu | +0.017 |
firefighters | +0.017 |
Neg logit contributions | |
orse | -0.010 |
ret | -0.010 |
orses | -0.009 |
peanut | -0.008 |
ets | -0.007 |
Token | family |
Token ID | 1641 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
Lu | +0.038 |
Ann | +0.032 |
angels | +0.032 |
Lu | +0.031 |
co | +0.031 |
Neg logit contributions | |
orse | -0.023 |
ret | -0.022 |
orses | -0.020 |
peanut | -0.020 |
ork | -0.018 |
Token | of |
Token ID | 286 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
Lu | +0.007 |
angels | +0.005 |
Ann | +0.005 |
Lu | +0.005 |
firefighters | +0.005 |
Neg logit contributions | |
ret | -0.006 |
orse | -0.006 |
orses | -0.005 |
peanut | -0.005 |
ets | -0.005 |
Token | ducks |
Token ID | 39694 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
Lu | +0.092 |
Ann | +0.086 |
attendant | +0.082 |
Lu | +0.079 |
K | +0.079 |
Neg logit contributions | |
peanut | -0.072 |
notebooks | -0.068 |
ol | -0.067 |
ink | -0.067 |
ret | -0.065 |
Token | lived |
Token ID | 5615 |
Feature activation | +9.54e-01 |
Pos logit contributions | |
ret | +0.016 |
orse | +0.016 |
orses | +0.015 |
peanut | +0.014 |
ork | +0.013 |
Neg logit contributions | |
Lu | -0.019 |
angels | -0.015 |
Ann | -0.015 |
Lu | -0.015 |
firefighters | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | +5.68e-01 |
Pos logit contributions | |
Lu | +0.049 |
Ann | +0.042 |
Lu | +0.040 |
Taxi | +0.039 |
Michael | +0.039 |
Neg logit contributions | |
ret | -0.034 |
orse | -0.033 |
orses | -0.031 |
peanut | -0.029 |
ets | -0.027 |
Token | this |
Token ID | 428 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
Lu | +0.029 |
angels | +0.024 |
Ann | +0.024 |
Lu | +0.024 |
co | +0.023 |
Neg logit contributions | |
ret | -0.020 |
orse | -0.020 |
orses | -0.018 |
peanut | -0.017 |
ork | -0.016 |
Token | house |
Token ID | 2156 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
ret | +0.025 |
orse | +0.025 |
orses | +0.023 |
ets | +0.022 |
ocket | +0.021 |
Neg logit contributions | |
Lu | -0.019 |
angels | -0.015 |
firefighters | -0.014 |
Ann | -0.014 |
Lu | -0.014 |
Token | steep |
Token ID | 14559 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
Lu | +0.106 |
Ann | +0.093 |
Lu | +0.093 |
youths | +0.093 |
angels | +0.089 |
Neg logit contributions | |
orse | -0.054 |
peanut | -0.051 |
ret | -0.049 |
orses | -0.047 |
mouth | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
Lu | +0.085 |
Lu | +0.072 |
Ann | +0.072 |
youths | +0.069 |
angels | +0.069 |
Neg logit contributions | |
ret | -0.035 |
orse | -0.034 |
ets | -0.033 |
orses | -0.031 |
peanut | -0.030 |
Token | They |
Token ID | 1119 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
ret | +0.026 |
orse | +0.026 |
orses | +0.024 |
ets | +0.023 |
peanut | +0.022 |
Neg logit contributions | |
Lu | -0.035 |
Ann | -0.028 |
angels | -0.027 |
firefighters | -0.027 |
Lu | -0.027 |
Token | try |
Token ID | 1949 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
Lu | +0.018 |
angels | +0.015 |
Lu | +0.015 |
Ann | +0.015 |
firefighters | +0.015 |
Neg logit contributions | |
ret | -0.012 |
orse | -0.012 |
orses | -0.011 |
ets | -0.010 |
peanut | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
Lu | +0.062 |
Lu | +0.054 |
Ann | +0.054 |
firefighters | +0.053 |
youths | +0.052 |
Neg logit contributions | |
ret | -0.035 |
orse | -0.033 |
orses | -0.030 |
ork | -0.029 |
peanut | -0.027 |
Token | go |
Token ID | 467 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
Lu | +0.084 |
Lu | +0.075 |
Ann | +0.074 |
angels | +0.074 |
youths | +0.072 |
Neg logit contributions | |
orse | -0.044 |
ret | -0.040 |
orses | -0.037 |
peanut | -0.033 |
carve | -0.033 |
Token | up |
Token ID | 510 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Lu | +0.096 |
Lu | +0.085 |
pastor | +0.084 |
youths | +0.083 |
Ann | +0.083 |
Neg logit contributions | |
ret | -0.057 |
s | -0.053 |
orse | -0.050 |
ets | -0.048 |
ork | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
Lu | +0.082 |
Ann | +0.074 |
Lu | +0.073 |
youths | +0.073 |
angels | +0.072 |
Neg logit contributions | |
ret | -0.035 |
orse | -0.035 |
ork | -0.032 |
orses | -0.030 |
peanut | -0.029 |
Token | hill |
Token ID | 12788 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
Lu | +0.071 |
Ann | +0.061 |
Lu | +0.058 |
angels | +0.056 |
pr | +0.056 |
Neg logit contributions | |
peanut | -0.057 |
ret | -0.055 |
orse | -0.054 |
orses | -0.052 |
ork | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
Lu | +0.033 |
angels | +0.027 |
Ann | +0.027 |
Lu | +0.027 |
firefighters | +0.026 |
Neg logit contributions | |
ret | -0.022 |
orse | -0.021 |
orses | -0.019 |
ets | -0.018 |
peanut | -0.018 |
Token | But |
Token ID | 887 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
ret | +0.012 |
orse | +0.012 |
orses | +0.011 |
peanut | +0.011 |
ets | +0.010 |
Neg logit contributions | |
Lu | -0.016 |
Ann | -0.013 |
angels | -0.013 |
Lu | -0.013 |
firefighters | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ret | +0.062 |
orse | +0.061 |
ets | +0.058 |
sel | +0.057 |
orses | +0.057 |
Neg logit contributions | |
Lu | -0.049 |
Ann | -0.041 |
Lu | -0.039 |
angels | -0.037 |
Pen | -0.036 |
Token | So |
Token ID | 2396 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
ret | +0.076 |
orse | +0.076 |
orses | +0.069 |
ets | +0.068 |
sel | +0.067 |
Neg logit contributions | |
Lu | -0.044 |
Ann | -0.039 |
Lu | -0.037 |
angels | -0.032 |
youths | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
orse | +0.084 |
orses | +0.078 |
ret | +0.078 |
ocket | +0.070 |
ets | +0.068 |
Neg logit contributions | |
Lu | -0.074 |
Lucy | -0.058 |
angels | -0.058 |
Ann | -0.057 |
Megan | -0.057 |
Token | Sophia |
Token ID | 39953 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
orse | +0.072 |
ret | +0.071 |
orses | +0.068 |
ets | +0.062 |
sel | +0.061 |
Neg logit contributions | |
Lu | -0.062 |
angels | -0.049 |
Lucy | -0.048 |
Ann | -0.047 |
Megan | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +8.05e-01 |
Pos logit contributions | |
Lu | +0.036 |
angels | +0.031 |
Ann | +0.030 |
firefighters | +0.030 |
youths | +0.030 |
Neg logit contributions | |
orse | -0.024 |
ret | -0.024 |
orses | -0.021 |
peanut | -0.020 |
ets | -0.020 |
Token | Mom |
Token ID | 11254 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
Lu | +0.036 |
Ann | +0.031 |
firefighters | +0.031 |
co | +0.031 |
Lu | +0.031 |
Neg logit contributions | |
ret | -0.030 |
orse | -0.029 |
peanut | -0.029 |
orses | -0.027 |
notebooks | -0.026 |
Token | my |
Token ID | 1820 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
Lu | +0.114 |
firefighters | +0.101 |
youths | +0.101 |
pastor | +0.096 |
firefighter | +0.096 |
Neg logit contributions | |
ret | -0.077 |
ets | -0.074 |
's | -0.074 |
peanut | -0.070 |
orse | -0.070 |
Token | went |
Token ID | 1816 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
Lu | +0.103 |
youths | +0.094 |
angels | +0.092 |
firefighters | +0.092 |
attendant | +0.089 |
Neg logit contributions | |
ret | -0.058 |
orse | -0.055 |
ets | -0.054 |
peanut | -0.050 |
sel | -0.050 |
Token | home |
Token ID | 1363 |
Feature activation | +9.96e-01 |
Pos logit contributions | |
Lu | +0.063 |
Lu | +0.054 |
Ann | +0.054 |
angels | +0.052 |
youths | +0.052 |
Neg logit contributions | |
ret | -0.039 |
orse | -0.038 |
ets | -0.035 |
orses | -0.035 |
ork | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
Lu | +0.054 |
Ann | +0.046 |
Lu | +0.046 |
Taxi | +0.045 |
youths | +0.044 |
Neg logit contributions | |
orse | -0.031 |
ret | -0.031 |
orses | -0.028 |
ork | -0.025 |
peanut | -0.025 |
Token | The |
Token ID | 383 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
orse | +0.082 |
ret | +0.078 |
orses | +0.075 |
peanut | +0.068 |
ocket | +0.068 |
Neg logit contributions | |
Lu | -0.065 |
Ann | -0.049 |
Lucy | -0.049 |
Lu | -0.047 |
Pen | -0.047 |
Token | him |
Token ID | 683 |
Feature activation | +6.89e-02 |
Pos logit contributions | |
ret | +0.007 |
orse | +0.006 |
orses | +0.006 |
peanut | +0.006 |
ets | +0.006 |
Neg logit contributions | |
Lu | -0.007 |
angels | -0.006 |
Ann | -0.005 |
Lu | -0.005 |
firefighters | -0.005 |
Token | scared |
Token ID | 12008 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
Lu | +0.003 |
angels | +0.003 |
Ann | +0.003 |
Lu | +0.003 |
firefighters | +0.002 |
Neg logit contributions | |
ret | -0.003 |
orse | -0.003 |
orses | -0.003 |
peanut | -0.002 |
ets | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
Lu | +0.027 |
Ann | +0.023 |
Lu | +0.023 |
angels | +0.022 |
attendant | +0.022 |
Neg logit contributions | |
ret | -0.018 |
orse | -0.018 |
orses | -0.017 |
peanut | -0.016 |
ets | -0.016 |
Token | He |
Token ID | 679 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
Lu | +0.001 |
angels | +0.001 |
Ann | +0.001 |
Lu | +0.001 |
firefighters | +0.001 |
Neg logit contributions | |
ret | -0.001 |
orse | -0.001 |
orses | -0.000 |
peanut | -0.000 |
ets | -0.000 |
Token | quickly |
Token ID | 2952 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
Lu | +0.022 |
angels | +0.019 |
Ann | +0.019 |
Lu | +0.019 |
firefighters | +0.018 |
Neg logit contributions | |
orse | -0.015 |
ret | -0.015 |
orses | -0.014 |
peanut | -0.013 |
ets | -0.013 |
Token | went |
Token ID | 1816 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
ret | +0.015 |
orse | +0.015 |
orses | +0.014 |
peanut | +0.013 |
ets | +0.012 |
Neg logit contributions | |
Lu | -0.015 |
angels | -0.012 |
Ann | -0.012 |
Lu | -0.012 |
firefighters | -0.012 |
Token | back |
Token ID | 736 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
Lu | +0.110 |
Ann | +0.100 |
Lu | +0.100 |
ators | +0.098 |
Stev | +0.097 |
Neg logit contributions | |
ets | -0.041 |
drawing | -0.040 |
ret | -0.040 |
orse | -0.039 |
ork | -0.039 |
Token | inside |
Token ID | 2641 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Lu | +0.113 |
Ann | +0.109 |
Lu | +0.108 |
youths | +0.107 |
angels | +0.105 |
Neg logit contributions | |
ret | -0.042 |
orse | -0.041 |
peanut | -0.037 |
's | -0.037 |
drawing | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +9.47e-02 |
Pos logit contributions | |
Lu | +0.008 |
angels | +0.007 |
Ann | +0.007 |
Lu | +0.007 |
firefighters | +0.007 |
Neg logit contributions | |
ret | -0.007 |
orse | -0.007 |
orses | -0.006 |
peanut | -0.006 |
ets | -0.005 |
Token | wrapped |
Token ID | 12908 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
Lu | +0.005 |
angels | +0.004 |
Ann | +0.004 |
Lu | +0.004 |
firefighters | +0.004 |
Neg logit contributions | |
ret | -0.004 |
orse | -0.004 |
orses | -0.003 |
peanut | -0.003 |
ets | -0.003 |
Token | himself |
Token ID | 2241 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
Lu | +0.036 |
Ann | +0.030 |
angels | +0.030 |
Lu | +0.030 |
youths | +0.029 |
Neg logit contributions | |
ret | -0.033 |
orse | -0.031 |
peanut | -0.029 |
orses | -0.028 |
ets | -0.027 |
Token | for |
Token ID | 329 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
ret | +0.026 |
orse | +0.026 |
orses | +0.024 |
peanut | +0.023 |
ets | +0.022 |
Neg logit contributions | |
Lu | -0.030 |
angels | -0.024 |
Ann | -0.023 |
Lu | -0.023 |
attendant | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
Lu | +0.016 |
Ann | +0.013 |
angels | +0.013 |
Lu | +0.013 |
youths | +0.012 |
Neg logit contributions | |
orse | -0.015 |
ret | -0.015 |
orses | -0.013 |
peanut | -0.013 |
ets | -0.012 |
Token | fun |
Token ID | 1257 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
Lu | +0.010 |
Ann | +0.008 |
angels | +0.008 |
Lu | +0.008 |
firefighters | +0.008 |
Neg logit contributions | |
ret | -0.009 |
orse | -0.009 |
orses | -0.008 |
peanut | -0.008 |
ets | -0.008 |
Token | afternoon |
Token ID | 6672 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
Lu | +0.016 |
Ann | +0.013 |
angels | +0.013 |
Lu | +0.013 |
firefighters | +0.013 |
Neg logit contributions | |
orse | -0.013 |
ret | -0.013 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
ret | +0.028 |
orse | +0.027 |
orses | +0.025 |
peanut | +0.024 |
ets | +0.023 |
Neg logit contributions | |
Lu | -0.028 |
angels | -0.023 |
Ann | -0.021 |
firefighters | -0.021 |
Lu | -0.021 |
Token | He |
Token ID | 679 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
ret | +0.013 |
orse | +0.013 |
orses | +0.012 |
peanut | +0.011 |
ets | +0.011 |
Neg logit contributions | |
Lu | -0.018 |
Ann | -0.015 |
angels | -0.015 |
Lu | -0.015 |
youths | -0.014 |
Token | then |
Token ID | 788 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
Lu | +0.017 |
angels | +0.014 |
Ann | +0.013 |
Lu | +0.013 |
firefighters | +0.013 |
Neg logit contributions | |
orse | -0.013 |
ret | -0.013 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.011 |
Token | bid |
Token ID | 8406 |
Feature activation | -8.62e-01 |
Pos logit contributions | |
ret | +0.004 |
orse | +0.004 |
orses | +0.003 |
peanut | +0.003 |
ets | +0.003 |
Neg logit contributions | |
Lu | -0.004 |
angels | -0.003 |
Ann | -0.003 |
Lu | -0.003 |
firefighters | -0.003 |
Token | farewell |
Token ID | 33458 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
orse | +0.036 |
ret | +0.034 |
orses | +0.034 |
peanut | +0.032 |
ets | +0.031 |
Neg logit contributions | |
Lu | -0.041 |
angels | -0.032 |
Ann | -0.031 |
attendant | -0.031 |
Lu | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
ret | +0.066 |
orse | +0.065 |
orses | +0.063 |
peanut | +0.058 |
notebooks | +0.057 |
Neg logit contributions | |
Lu | -0.056 |
attendant | -0.045 |
angels | -0.043 |
co | -0.041 |
Lu | -0.040 |
Token | all |
Token ID | 477 |
Feature activation | +7.58e-01 |
Pos logit contributions | |
ret | +0.001 |
orse | +0.001 |
orses | +0.001 |
peanut | +0.001 |
ets | +0.001 |
Neg logit contributions | |
Lu | -0.001 |
angels | -0.001 |
Ann | -0.001 |
Lu | -0.001 |
firefighters | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
ret | +0.004 |
orse | +0.004 |
orses | +0.004 |
peanut | +0.003 |
ets | +0.003 |
Neg logit contributions | |
Lu | -0.005 |
angels | -0.004 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
Lu | +0.024 |
angels | +0.021 |
firefighters | +0.020 |
youths | +0.020 |
Ann | +0.019 |
Neg logit contributions | |
orse | -0.019 |
ret | -0.018 |
orses | -0.017 |
peanut | -0.016 |
ork | -0.015 |
Token | Mom |
Token ID | 29252 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
Lu | +0.019 |
angels | +0.016 |
firefighters | +0.015 |
Ann | +0.015 |
youths | +0.015 |
Neg logit contributions | |
orse | -0.016 |
ret | -0.016 |
orses | -0.015 |
peanut | -0.014 |
ork | -0.013 |
Token | my |
Token ID | 1820 |
Feature activation | +9.56e-01 |
Pos logit contributions | |
Lu | +0.078 |
angels | +0.069 |
youths | +0.067 |
firefighters | +0.066 |
taxis | +0.063 |
Neg logit contributions | |
ret | -0.071 |
orse | -0.071 |
orses | -0.065 |
's | -0.064 |
peanut | -0.064 |
Token | said |
Token ID | 531 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
Lu | +0.049 |
angels | +0.043 |
youths | +0.041 |
firefighters | +0.041 |
Taxi | +0.040 |
Neg logit contributions | |
orse | -0.035 |
ret | -0.034 |
orses | -0.031 |
peanut | -0.029 |
sel | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
ret | +0.009 |
orse | +0.009 |
orses | +0.008 |
ets | +0.007 |
peanut | +0.007 |
Neg logit contributions | |
Lu | -0.012 |
angels | -0.010 |
Lu | -0.010 |
Ann | -0.010 |
firefighters | -0.010 |
Token | " |
Token ID | 366 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
Lu | +0.013 |
angels | +0.010 |
Ann | +0.010 |
Lu | +0.010 |
firefighters | +0.010 |
Neg logit contributions | |
orse | -0.011 |
ret | -0.011 |
orses | -0.010 |
peanut | -0.009 |
ets | -0.009 |
Token | No |
Token ID | 2949 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
Lu | +0.030 |
angels | +0.024 |
firefighters | +0.023 |
youths | +0.022 |
Taxi | +0.021 |
Neg logit contributions | |
orse | -0.034 |
ret | -0.034 |
orses | -0.032 |
peanut | -0.030 |
ets | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
Lu | +0.016 |
angels | +0.013 |
Ann | +0.013 |
Lu | +0.013 |
firefighters | +0.013 |
Neg logit contributions | |
ret | -0.009 |
orse | -0.009 |
orses | -0.008 |
peanut | -0.008 |
ets | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
ret | +0.010 |
orse | +0.010 |
orses | +0.009 |
peanut | +0.008 |
ets | +0.008 |
Neg logit contributions | |
Lu | -0.017 |
angels | -0.014 |
Ann | -0.014 |
Lu | -0.013 |
firefighters | -0.013 |
Token | 's |
Token ID | 338 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
Lu | +0.036 |
angels | +0.032 |
Ann | +0.032 |
Lu | +0.031 |
firefighters | +0.031 |
Neg logit contributions | |
ret | -0.008 |
orse | -0.007 |
orses | -0.006 |
ets | -0.005 |
peanut | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -7.20e-01 |
Pos logit contributions | |
orse | +0.063 |
ret | +0.060 |
orses | +0.059 |
sel | +0.053 |
peanut | +0.053 |
Neg logit contributions | |
Lu | -0.063 |
angels | -0.050 |
Lucy | -0.047 |
Lu | -0.045 |
firefighters | -0.045 |
Token | I |
Token ID | 314 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
ret | +0.031 |
orse | +0.029 |
orses | +0.028 |
ets | +0.026 |
ork | +0.025 |
Neg logit contributions | |
Lu | -0.034 |
angels | -0.028 |
firefighters | -0.028 |
Ann | -0.026 |
Lu | -0.026 |
Token | can |
Token ID | 460 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
ret | +0.024 |
orse | +0.023 |
orses | +0.022 |
peanut | +0.021 |
notebooks | +0.020 |
Neg logit contributions | |
Lu | -0.025 |
Ann | -0.020 |
angels | -0.020 |
Lu | -0.019 |
firefighters | -0.019 |
Token | see |
Token ID | 766 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
Lu | +0.014 |
Ann | +0.011 |
angels | +0.011 |
Lu | +0.011 |
youths | +0.011 |
Neg logit contributions | |
orse | -0.012 |
ret | -0.012 |
orses | -0.011 |
peanut | -0.010 |
ets | -0.010 |
Token | something |
Token ID | 1223 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
ret | +0.021 |
orse | +0.021 |
orses | +0.019 |
ets | +0.018 |
peanut | +0.017 |
Neg logit contributions | |
Lu | -0.024 |
angels | -0.020 |
firefighters | -0.019 |
Lu | -0.018 |
Ann | -0.018 |
Token | shiny |
Token ID | 22441 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
ret | +0.017 |
orse | +0.016 |
orses | +0.015 |
peanut | +0.014 |
ets | +0.014 |
Neg logit contributions | |
Lu | -0.019 |
angels | -0.016 |
firefighters | -0.015 |
Lu | -0.015 |
Ann | -0.015 |
Token | on |
Token ID | 319 |
Feature activation | -9.54e-01 |
Pos logit contributions | |
ret | +0.009 |
orse | +0.009 |
orses | +0.009 |
peanut | +0.008 |
ets | +0.008 |
Neg logit contributions | |
Lu | -0.010 |
angels | -0.008 |
Ann | -0.008 |
firefighters | -0.008 |
Lu | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -7.43e-01 |
Pos logit contributions | |
orse | +0.047 |
ret | +0.047 |
orses | +0.045 |
ets | +0.042 |
ocket | +0.041 |
Neg logit contributions | |
Lu | -0.038 |
angels | -0.030 |
firefighters | -0.030 |
Lu | -0.029 |
Ann | -0.028 |
Token | other |
Token ID | 584 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
ret | +0.034 |
orse | +0.033 |
orses | +0.031 |
peanut | +0.029 |
ocket | +0.028 |
Neg logit contributions | |
Lu | -0.032 |
angels | -0.027 |
firefighters | -0.026 |
attendant | -0.025 |
Ann | -0.025 |
Token | side |
Token ID | 1735 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
ret | +0.031 |
orse | +0.030 |
orses | +0.028 |
peanut | +0.027 |
ocket | +0.026 |
Neg logit contributions | |
Lu | -0.025 |
angels | -0.020 |
Ann | -0.019 |
firefighters | -0.019 |
Lu | -0.018 |
Token | ." |
Token ID | 526 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
ret | +0.004 |
orse | +0.004 |
orses | +0.004 |
peanut | +0.004 |
ets | +0.004 |
Neg logit contributions | |
Lu | -0.005 |
angels | -0.004 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token | The |
Token ID | 383 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
ret | +0.001 |
orse | +0.001 |
orses | +0.001 |
peanut | +0.001 |
ets | +0.001 |
Neg logit contributions | |
Lu | -0.002 |
angels | -0.001 |
Ann | -0.001 |
Lu | -0.001 |
firefighters | -0.001 |
Token | laser |
Token ID | 12855 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
Lu | +0.003 |
angels | +0.003 |
Ann | +0.003 |
Lu | +0.003 |
firefighters | +0.003 |
Neg logit contributions | |
ret | -0.004 |
orse | -0.004 |
orses | -0.004 |
peanut | -0.003 |
ets | -0.003 |
Token | hits |
Token ID | 7127 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
ret | +0.017 |
orse | +0.017 |
orses | +0.016 |
peanut | +0.015 |
ets | +0.014 |
Neg logit contributions | |
Lu | -0.020 |
angels | -0.017 |
Ann | -0.016 |
firefighters | -0.016 |
Lu | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -9.20e-01 |
Pos logit contributions | |
Lu | +0.011 |
angels | +0.009 |
Lu | +0.008 |
Ann | +0.008 |
youths | +0.008 |
Neg logit contributions | |
ret | -0.009 |
orse | -0.009 |
orses | -0.008 |
peanut | -0.008 |
ets | -0.007 |
Token | cat |
Token ID | 3797 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ret | +0.042 |
orse | +0.041 |
orses | +0.039 |
ets | +0.036 |
ocket | +0.035 |
Neg logit contributions | |
Lu | -0.039 |
angels | -0.033 |
firefighters | -0.032 |
attendant | -0.031 |
youths | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
ret | +0.057 |
orse | +0.057 |
orses | +0.054 |
peanut | +0.051 |
ocket | +0.050 |
Neg logit contributions | |
Lu | -0.052 |
angels | -0.042 |
Ann | -0.041 |
Lu | -0.038 |
co | -0.038 |
Token | The |
Token ID | 383 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
ret | +0.006 |
orse | +0.006 |
orses | +0.005 |
peanut | +0.005 |
ets | +0.005 |
Neg logit contributions | |
Lu | -0.011 |
angels | -0.009 |
Ann | -0.009 |
Lu | -0.009 |
firefighters | -0.009 |
Token | cat |
Token ID | 3797 |
Feature activation | -7.93e-01 |
Pos logit contributions | |
ret | +0.027 |
orse | +0.026 |
orses | +0.025 |
ets | +0.023 |
ocket | +0.022 |
Neg logit contributions | |
Lu | -0.026 |
angels | -0.021 |
firefighters | -0.021 |
attendant | -0.020 |
youths | -0.020 |
Token | makes |
Token ID | 1838 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
ret | +0.033 |
orse | +0.033 |
orses | +0.030 |
peanut | +0.028 |
ocket | +0.028 |
Neg logit contributions | |
Lu | -0.037 |
angels | -0.031 |
Ann | -0.030 |
firefighters | -0.029 |
Lu | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ret | +0.042 |
orse | +0.042 |
orses | +0.039 |
ets | +0.037 |
ocket | +0.037 |
Neg logit contributions | |
Lu | -0.037 |
angels | -0.030 |
Ann | -0.028 |
firefighters | -0.028 |
Lu | -0.026 |
Token | loud |
Token ID | 7812 |
Feature activation | -9.89e-01 |
Pos logit contributions | |
ret | +0.083 |
orse | +0.079 |
orses | +0.075 |
ocket | +0.074 |
ets | +0.073 |
Neg logit contributions | |
Lu | -0.058 |
angels | -0.051 |
firefighters | -0.047 |
Ann | -0.044 |
attendant | -0.044 |
Token | as |
Token ID | 355 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
ret | +0.002 |
orse | +0.002 |
orses | +0.002 |
peanut | +0.002 |
ets | +0.002 |
Neg logit contributions | |
Lu | -0.002 |
angels | -0.002 |
Ann | -0.002 |
Lu | -0.002 |
firefighters | -0.002 |
Token | fun |
Token ID | 1257 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
Lu | +0.013 |
Ann | +0.011 |
angels | +0.011 |
Lu | +0.011 |
youths | +0.010 |
Neg logit contributions | |
ret | -0.011 |
orse | -0.011 |
orses | -0.010 |
peanut | -0.009 |
ets | -0.009 |
Token | as |
Token ID | 355 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
ret | +0.036 |
orse | +0.034 |
orses | +0.032 |
peanut | +0.030 |
ocket | +0.029 |
Neg logit contributions | |
Lu | -0.036 |
angels | -0.030 |
firefighters | -0.028 |
youths | -0.028 |
Lu | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +8.55e-01 |
Pos logit contributions | |
ret | +0.019 |
orse | +0.018 |
orses | +0.017 |
peanut | +0.016 |
ets | +0.016 |
Neg logit contributions | |
Lu | -0.026 |
angels | -0.022 |
firefighters | -0.021 |
Lu | -0.020 |
Ann | -0.020 |
Token | old |
Token ID | 1468 |
Feature activation | +8.02e-01 |
Pos logit contributions | |
Lu | +0.041 |
Ann | +0.034 |
Lu | +0.033 |
co | +0.032 |
angels | +0.031 |
Neg logit contributions | |
orse | -0.033 |
ret | -0.033 |
orses | -0.031 |
peanut | -0.031 |
ets | -0.028 |
Token | one |
Token ID | 530 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
Lu | +0.036 |
Ann | +0.030 |
Lu | +0.029 |
angels | +0.029 |
co | +0.028 |
Neg logit contributions | |
orse | -0.033 |
ret | -0.032 |
orses | -0.031 |
peanut | -0.030 |
ets | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
Lu | +0.001 |
angels | +0.001 |
Ann | +0.001 |
Lu | +0.001 |
firefighters | +0.001 |
Neg logit contributions | |
ret | -0.001 |
orse | -0.001 |
orses | -0.001 |
peanut | -0.001 |
ets | -0.001 |
Token | They |
Token ID | 1119 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
ret | +0.028 |
orse | +0.027 |
orses | +0.025 |
peanut | +0.024 |
ocket | +0.024 |
Neg logit contributions | |
Lu | -0.028 |
Ann | -0.022 |
Lu | -0.021 |
angels | -0.021 |
Taxi | -0.021 |
Token | learned |
Token ID | 4499 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
Lu | +0.018 |
angels | +0.015 |
Ann | +0.015 |
Lu | +0.015 |
firefighters | +0.015 |
Neg logit contributions | |
ret | -0.013 |
orse | -0.013 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.011 |
Token | that |
Token ID | 326 |
Feature activation | +4.77e-01 |
Pos logit contributions | |
Lu | +0.038 |
Ann | +0.033 |
Lu | +0.032 |
angels | +0.031 |
co | +0.031 |
Neg logit contributions | |
ret | -0.023 |
orse | -0.021 |
peanut | -0.019 |
orses | -0.019 |
ets | -0.018 |
Token | sometimes |
Token ID | 3360 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
Lu | +0.025 |
Ann | +0.021 |
Lu | +0.021 |
angels | +0.020 |
co | +0.020 |
Neg logit contributions | |
orse | -0.017 |
ret | -0.017 |
orses | -0.016 |
peanut | -0.015 |
ets | -0.014 |
Token | cup |
Token ID | 25244 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
ret | +0.020 |
orse | +0.020 |
orses | +0.018 |
peanut | +0.017 |
ets | +0.017 |
Neg logit contributions | |
Lu | -0.029 |
Ann | -0.024 |
Lu | -0.024 |
angels | -0.024 |
co | -0.023 |
Token | ine |
Token ID | 500 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
ret | +0.016 |
orse | +0.016 |
orses | +0.015 |
peanut | +0.014 |
ets | +0.014 |
Neg logit contributions | |
Lu | -0.023 |
angels | -0.019 |
Ann | -0.019 |
Lu | -0.018 |
firefighters | -0.018 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ret | +0.014 |
orse | +0.014 |
orses | +0.013 |
peanut | +0.012 |
ets | +0.011 |
Neg logit contributions | |
Lu | -0.018 |
angels | -0.015 |
Ann | -0.014 |
Lu | -0.014 |
firefighters | -0.014 |
Token | said |
Token ID | 531 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
ret | +0.062 |
orse | +0.058 |
orses | +0.056 |
ets | +0.052 |
ocus | +0.051 |
Neg logit contributions | |
Lu | -0.053 |
Lucy | -0.042 |
Lu | -0.040 |
Pen | -0.039 |
firefighters | -0.038 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
Lu | +0.026 |
angels | +0.023 |
Ann | +0.022 |
firefighters | +0.021 |
youths | +0.021 |
Neg logit contributions | |
ret | -0.020 |
orse | -0.019 |
peanut | -0.018 |
orses | -0.017 |
ork | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
ret | +0.005 |
orse | +0.005 |
orses | +0.004 |
peanut | +0.004 |
ets | +0.004 |
Neg logit contributions | |
Lu | -0.005 |
angels | -0.004 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token | The |
Token ID | 383 |
Feature activation | +6.48e-01 |
Pos logit contributions | |
ret | +0.032 |
orse | +0.031 |
orses | +0.030 |
ets | +0.028 |
ocket | +0.027 |
Neg logit contributions | |
Lu | -0.030 |
Lu | -0.023 |
Lucy | -0.022 |
angels | -0.022 |
attendant | -0.022 |
Token | por |
Token ID | 16964 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
Lu | +0.031 |
angels | +0.025 |
Ann | +0.025 |
Lu | +0.024 |
ators | +0.024 |
Neg logit contributions | |
orse | -0.026 |
ret | -0.026 |
peanut | -0.025 |
orses | -0.024 |
ocket | -0.022 |
Token | cup |
Token ID | 25244 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
Lu | +0.020 |
angels | +0.016 |
firefighters | +0.016 |
youths | +0.015 |
Lu | +0.015 |
Neg logit contributions | |
orse | -0.026 |
ret | -0.025 |
orses | -0.024 |
peanut | -0.023 |
ets | -0.022 |
Token | ine |
Token ID | 500 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
ret | +0.010 |
orse | +0.010 |
orses | +0.009 |
peanut | +0.009 |
ets | +0.009 |
Neg logit contributions | |
Lu | -0.017 |
angels | -0.014 |
Ann | -0.014 |
Lu | -0.014 |
firefighters | -0.013 |
Token | said |
Token ID | 531 |
Feature activation | +7.62e-02 |
Pos logit contributions | |
Lu | +0.008 |
angels | +0.007 |
Ann | +0.007 |
Lu | +0.007 |
firefighters | +0.007 |
Neg logit contributions | |
orse | -0.007 |
ret | -0.007 |
orses | -0.006 |
peanut | -0.006 |
ets | -0.006 |
Token | of |
Token ID | 286 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
Lu | +0.030 |
angels | +0.024 |
Lu | +0.024 |
Ann | +0.024 |
youths | +0.024 |
Neg logit contributions | |
ret | -0.028 |
orse | -0.028 |
orses | -0.026 |
peanut | -0.024 |
ork | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
Lu | +0.017 |
angels | +0.014 |
Ann | +0.014 |
Lu | +0.014 |
youths | +0.013 |
Neg logit contributions | |
ret | -0.013 |
orse | -0.013 |
orses | -0.012 |
peanut | -0.012 |
ork | -0.011 |
Token | visit |
Token ID | 3187 |
Feature activation | -8.14e-01 |
Pos logit contributions | |
ret | +0.010 |
orse | +0.010 |
orses | +0.009 |
peanut | +0.009 |
ets | +0.009 |
Neg logit contributions | |
Lu | -0.010 |
angels | -0.008 |
Ann | -0.008 |
Lu | -0.008 |
firefighters | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
orse | +0.033 |
ret | +0.033 |
orses | +0.030 |
peanut | +0.028 |
ets | +0.027 |
Neg logit contributions | |
Lu | -0.040 |
angels | -0.032 |
Ann | -0.031 |
Lu | -0.031 |
firefighters | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
ret | +0.011 |
orse | +0.011 |
orses | +0.011 |
ets | +0.010 |
peanut | +0.010 |
Neg logit contributions | |
Lu | -0.011 |
angels | -0.009 |
firefighters | -0.009 |
Lu | -0.009 |
Ann | -0.009 |
Token | doctor |
Token ID | 6253 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
Lu | +0.039 |
Ann | +0.031 |
Lu | +0.030 |
Karen | +0.029 |
co | +0.028 |
Neg logit contributions | |
orse | -0.047 |
ret | -0.047 |
peanut | -0.045 |
orses | -0.044 |
ork | -0.042 |
Token | opened |
Token ID | 4721 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
ret | +0.013 |
orse | +0.012 |
orses | +0.012 |
peanut | +0.011 |
ets | +0.011 |
Neg logit contributions | |
Lu | -0.015 |
angels | -0.012 |
Ann | -0.012 |
Lu | -0.012 |
firefighters | -0.012 |
Token | two |
Token ID | 734 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
ret | +0.004 |
orse | +0.004 |
orses | +0.004 |
peanut | +0.004 |
ets | +0.004 |
Neg logit contributions | |
Lu | -0.005 |
angels | -0.004 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token | pieces |
Token ID | 5207 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
ret | +0.049 |
orse | +0.048 |
orses | +0.044 |
peanut | +0.040 |
ork | +0.039 |
Neg logit contributions | |
Lu | -0.043 |
angels | -0.036 |
firefighters | -0.034 |
youths | -0.034 |
Ann | -0.032 |
Token | of |
Token ID | 286 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
ret | +0.038 |
orse | +0.038 |
orses | +0.036 |
peanut | +0.033 |
ork | +0.032 |
Neg logit contributions | |
Lu | -0.031 |
angels | -0.025 |
firefighters | -0.024 |
Ann | -0.023 |
Lu | -0.023 |
Token | paper |
Token ID | 3348 |
Feature activation | -9.83e-01 |
Pos logit contributions | |
orse | +0.029 |
ret | +0.029 |
orses | +0.027 |
ets | +0.025 |
ork | +0.024 |
Neg logit contributions | |
Lu | -0.030 |
angels | -0.025 |
firefighters | -0.024 |
Ann | -0.023 |
Lu | -0.023 |
Token | touch |
Token ID | 3638 |
Feature activation | -6.19e-01 |
Pos logit contributions | |
ret | +0.029 |
orse | +0.027 |
orses | +0.026 |
sel | +0.025 |
ets | +0.024 |
Neg logit contributions | |
Lu | -0.030 |
angels | -0.024 |
Ann | -0.023 |
youths | -0.023 |
firefighters | -0.023 |
Token | it |
Token ID | 340 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
orse | +0.023 |
ret | +0.022 |
orses | +0.021 |
peanut | +0.019 |
ets | +0.018 |
Neg logit contributions | |
Lu | -0.032 |
angels | -0.027 |
Ann | -0.026 |
firefighters | -0.026 |
Lu | -0.026 |
Token | or |
Token ID | 393 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ret | +0.024 |
orse | +0.024 |
orses | +0.023 |
peanut | +0.021 |
sel | +0.020 |
Neg logit contributions | |
Lu | -0.024 |
angels | -0.020 |
Ann | -0.018 |
Lu | -0.018 |
firefighters | -0.018 |
Token | it |
Token ID | 340 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
ret | +0.053 |
orse | +0.050 |
orses | +0.047 |
sel | +0.043 |
ork | +0.043 |
Neg logit contributions | |
Lu | -0.059 |
angels | -0.049 |
firefighters | -0.047 |
youths | -0.046 |
Ann | -0.045 |
Token | will |
Token ID | 481 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
orse | +0.021 |
ret | +0.021 |
orses | +0.019 |
peanut | +0.018 |
ocket | +0.017 |
Neg logit contributions | |
Lu | -0.021 |
angels | -0.017 |
Ann | -0.016 |
Lu | -0.016 |
firefighters | -0.016 |
Token | break |
Token ID | 2270 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
ret | +0.017 |
orse | +0.016 |
orses | +0.015 |
peanut | +0.014 |
ets | +0.014 |
Neg logit contributions | |
Lu | -0.019 |
angels | -0.016 |
Ann | -0.015 |
firefighters | -0.015 |
Lu | -0.015 |
Token | ." |
Token ID | 526 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
orse | +0.014 |
ret | +0.014 |
orses | +0.013 |
peanut | +0.012 |
ets | +0.012 |
Neg logit contributions | |
Lu | -0.016 |
angels | -0.014 |
Ann | -0.013 |
Lu | -0.013 |
firefighters | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
ret | +0.030 |
orse | +0.029 |
orses | +0.028 |
ets | +0.027 |
sel | +0.026 |
Neg logit contributions | |
Lu | -0.028 |
Ann | -0.022 |
Lu | -0.022 |
angels | -0.021 |
Lucy | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
ret | +0.015 |
orse | +0.015 |
orses | +0.014 |
peanut | +0.013 |
ets | +0.013 |
Neg logit contributions | |
Lu | -0.016 |
Ann | -0.013 |
Lu | -0.013 |
angels | -0.013 |
firefighters | -0.012 |
Token | M |
Token ID | 44 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
ret | +0.015 |
orse | +0.015 |
orses | +0.014 |
peanut | +0.013 |
ets | +0.013 |
Neg logit contributions | |
Lu | -0.016 |
Ann | -0.013 |
Lu | -0.013 |
angels | -0.013 |
firefighters | -0.012 |
Token | ia |
Token ID | 544 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
orse | +0.041 |
ret | +0.040 |
orses | +0.039 |
peanut | +0.038 |
ets | +0.037 |
Neg logit contributions | |
Lu | -0.032 |
Ann | -0.026 |
angels | -0.026 |
firefighters | -0.024 |
Lu | -0.024 |
Token | adventure |
Token ID | 8855 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
ret | +0.016 |
orse | +0.015 |
orses | +0.015 |
peanut | +0.013 |
ets | +0.013 |
Neg logit contributions | |
Lu | -0.013 |
angels | -0.010 |
Lu | -0.010 |
Ann | -0.010 |
firefighters | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
ret | +0.004 |
orse | +0.004 |
orses | +0.004 |
peanut | +0.004 |
ets | +0.004 |
Neg logit contributions | |
Lu | -0.005 |
angels | -0.004 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token | He |
Token ID | 679 |
Feature activation | -8.66e-02 |
Pos logit contributions | |
Lu | +0.031 |
angels | +0.028 |
Ann | +0.026 |
Lu | +0.026 |
co | +0.026 |
Neg logit contributions | |
orse | -0.024 |
ret | -0.024 |
orses | -0.022 |
peanut | -0.020 |
ork | -0.019 |
Token | found |
Token ID | 1043 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
ret | +0.003 |
orse | +0.003 |
orses | +0.003 |
peanut | +0.002 |
ets | +0.002 |
Neg logit contributions | |
Lu | -0.005 |
angels | -0.004 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
Lu | +0.006 |
angels | +0.005 |
Ann | +0.005 |
Lu | +0.004 |
firefighters | +0.004 |
Neg logit contributions | |
ret | -0.005 |
orse | -0.005 |
orses | -0.004 |
peanut | -0.004 |
ets | -0.004 |
Token | peaceful |
Token ID | 12309 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
ret | +0.025 |
orse | +0.025 |
orses | +0.023 |
ets | +0.022 |
sel | +0.021 |
Neg logit contributions | |
Lu | -0.029 |
angels | -0.023 |
Lu | -0.023 |
firefighters | -0.022 |
Ann | -0.022 |
Token | case |
Token ID | 1339 |
Feature activation | -6.91e-01 |
Pos logit contributions | |
ret | +0.025 |
orse | +0.024 |
orses | +0.023 |
ets | +0.021 |
peanut | +0.020 |
Neg logit contributions | |
Lu | -0.023 |
angels | -0.018 |
Lu | -0.018 |
Ann | -0.017 |
firefighters | -0.017 |
Token | deep |
Token ID | 2769 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
ret | +0.029 |
orse | +0.028 |
orses | +0.027 |
peanut | +0.025 |
ets | +0.024 |
Neg logit contributions | |
Lu | -0.032 |
angels | -0.026 |
attendant | -0.025 |
Lu | -0.025 |
Ann | -0.025 |
Token | in |
Token ID | 287 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
Lu | +0.011 |
angels | +0.009 |
Ann | +0.009 |
Lu | +0.009 |
firefighters | +0.009 |
Neg logit contributions | |
ret | -0.010 |
orse | -0.010 |
orses | -0.009 |
peanut | -0.009 |
ets | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
ret | +0.011 |
orse | +0.010 |
orses | +0.010 |
ets | +0.009 |
peanut | +0.009 |
Neg logit contributions | |
Lu | -0.010 |
angels | -0.008 |
Ann | -0.008 |
Lu | -0.008 |
firefighters | -0.008 |
Token | woods |
Token ID | 16479 |
Feature activation | -8.32e-01 |
Pos logit contributions | |
orse | +0.042 |
ret | +0.042 |
orses | +0.039 |
ets | +0.035 |
ocus | +0.035 |
Neg logit contributions | |
Lu | -0.030 |
Lu | -0.022 |
Ann | -0.022 |
angels | -0.022 |
Taxi | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -6.49e-01 |
Pos logit contributions | |
ret | +0.017 |
orse | +0.017 |
orses | +0.016 |
peanut | +0.015 |
ets | +0.014 |
Neg logit contributions | |
Lu | -0.019 |
angels | -0.015 |
Lu | -0.015 |
Ann | -0.015 |
firefighters | -0.014 |
Token | He |
Token ID | 679 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
ret | +0.032 |
orse | +0.032 |
orses | +0.031 |
ocket | +0.027 |
ets | +0.027 |
Neg logit contributions | |
Lu | -0.027 |
Lu | -0.020 |
firefighters | -0.020 |
Ann | -0.020 |
Lucy | -0.020 |
Token | asked |
Token ID | 1965 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
Lu | +0.008 |
angels | +0.006 |
Ann | +0.006 |
Lu | +0.006 |
firefighters | +0.006 |
Neg logit contributions | |
ret | -0.006 |
orse | -0.006 |
orses | -0.005 |
peanut | -0.005 |
ets | -0.005 |
Token | what |
Token ID | 644 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
ret | +0.002 |
orse | +0.002 |
orses | +0.002 |
peanut | +0.002 |
ets | +0.002 |
Neg logit contributions | |
Lu | -0.002 |
angels | -0.001 |
Ann | -0.001 |
Lu | -0.001 |
firefighters | -0.001 |
Token | they |
Token ID | 484 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
Lu | +0.024 |
angels | +0.020 |
Ann | +0.019 |
Taxi | +0.019 |
attendant | +0.019 |
Neg logit contributions | |
orse | -0.016 |
ret | -0.016 |
orses | -0.015 |
peanut | -0.015 |
ork | -0.013 |
Token | were |
Token ID | 547 |
Feature activation | -6.42e-01 |
Pos logit contributions | |
ret | +0.021 |
orse | +0.021 |
orses | +0.020 |
peanut | +0.018 |
ork | +0.018 |
Neg logit contributions | |
Lu | -0.025 |
angels | -0.020 |
Lu | -0.020 |
Ann | -0.019 |
firefighters | -0.019 |
Token | having |
Token ID | 1719 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
ret | +0.031 |
orse | +0.030 |
orses | +0.028 |
ork | +0.026 |
peanut | +0.026 |
Neg logit contributions | |
Lu | -0.026 |
angels | -0.021 |
firefighters | -0.020 |
youths | -0.019 |
Lu | -0.019 |
Token | for |
Token ID | 329 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
ret | +0.018 |
orse | +0.017 |
orses | +0.016 |
peanut | +0.015 |
ets | +0.015 |
Neg logit contributions | |
Lu | -0.019 |
angels | -0.016 |
firefighters | -0.015 |
Lu | -0.015 |
Ann | -0.015 |
Token | dinner |
Token ID | 8073 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
ret | +0.043 |
orse | +0.041 |
orses | +0.039 |
peanut | +0.036 |
ets | +0.036 |
Neg logit contributions | |
Lu | -0.044 |
angels | -0.035 |
Lu | -0.034 |
firefighters | -0.033 |
youths | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +6.63e-02 |
Pos logit contributions | |
ret | +0.018 |
orse | +0.018 |
orses | +0.016 |
peanut | +0.015 |
ets | +0.015 |
Neg logit contributions | |
Lu | -0.021 |
angels | -0.017 |
Lu | -0.016 |
Ann | -0.016 |
firefighters | -0.016 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
Lu | +0.003 |
angels | +0.003 |
Ann | +0.003 |
Lu | +0.003 |
firefighters | +0.003 |
Neg logit contributions | |
ret | -0.003 |
orse | -0.003 |
orses | -0.002 |
peanut | -0.002 |
ets | -0.002 |
Token | up |
Token ID | 510 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
ret | +0.008 |
orse | +0.008 |
orses | +0.008 |
peanut | +0.007 |
ets | +0.007 |
Neg logit contributions | |
Lu | -0.008 |
angels | -0.006 |
Ann | -0.006 |
Lu | -0.006 |
firefighters | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
orse | +0.037 |
ret | +0.037 |
orses | +0.035 |
ocket | +0.032 |
peanut | +0.032 |
Neg logit contributions | |
Lu | -0.034 |
angels | -0.027 |
Lu | -0.026 |
firefighters | -0.026 |
Ann | -0.026 |
Token | mess |
Token ID | 2085 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
Lu | +0.087 |
Ann | +0.075 |
Pen | +0.073 |
Luis | +0.073 |
K | +0.072 |
Neg logit contributions | |
ret | -0.027 |
peanut | -0.026 |
notebooks | -0.025 |
ork | -0.024 |
drawing | -0.023 |
Token | they |
Token ID | 484 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
ret | +0.010 |
orse | +0.010 |
orses | +0.009 |
peanut | +0.009 |
ets | +0.008 |
Neg logit contributions | |
Lu | -0.011 |
angels | -0.009 |
Ann | -0.009 |
Lu | -0.009 |
firefighters | -0.009 |
Token | made |
Token ID | 925 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
ret | +0.024 |
orse | +0.023 |
orses | +0.022 |
peanut | +0.020 |
ork | +0.019 |
Neg logit contributions | |
Lu | -0.030 |
angels | -0.025 |
firefighters | -0.024 |
Ann | -0.024 |
Lu | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
orse | +0.020 |
ret | +0.020 |
orses | +0.019 |
peanut | +0.018 |
ocket | +0.017 |
Neg logit contributions | |
Lu | -0.020 |
angels | -0.017 |
Lu | -0.016 |
Ann | -0.016 |
firefighters | -0.016 |
Token | |
Token ID | 220 |
Feature activation | -9.90e-02 |
Pos logit contributions | |
ret | +0.132 |
orse | +0.129 |
ocket | +0.122 |
orses | +0.122 |
ocus | +0.119 |
Neg logit contributions | |
Lu | -0.090 |
Lucy | -0.078 |
Jenkins | -0.068 |
Ann | -0.067 |
youths | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
ret | +0.004 |
orse | +0.004 |
orses | +0.004 |
peanut | +0.003 |
ets | +0.003 |
Neg logit contributions | |
Lu | -0.005 |
angels | -0.004 |
Ann | -0.004 |
Lu | -0.004 |
firefighters | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
ret | +0.084 |
orse | +0.078 |
ets | +0.078 |
ocket | +0.077 |
sel | +0.077 |
Neg logit contributions | |
Lu | -0.055 |
Lu | -0.044 |
Ann | -0.044 |
attendant | -0.038 |
Lucy | -0.038 |
Token | The |
Token ID | 464 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
ret | +0.107 |
orse | +0.097 |
ocket | +0.096 |
alks | +0.095 |
peanut | +0.092 |
Neg logit contributions | |
Lu | -0.064 |
Lu | -0.061 |
Ann | -0.059 |
Michael | -0.051 |
attendant | -0.049 |
Token | boys |
Token ID | 6510 |
Feature activation | +9.87e-01 |
Pos logit contributions | |
Lu | +0.072 |
Luis | +0.058 |
Stev | +0.058 |
Pen | +0.056 |
Ann | +0.056 |
Neg logit contributions | |
peanut | -0.041 |
orse | -0.041 |
ret | -0.040 |
orses | -0.039 |
notebooks | -0.038 |
Token | was |
Token ID | 373 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
ret | +0.018 |
orse | +0.018 |
orses | +0.016 |
peanut | +0.015 |
ets | +0.015 |
Neg logit contributions | |
Lu | -0.020 |
angels | -0.016 |
firefighters | -0.015 |
Ann | -0.015 |
Lu | -0.015 |
Token | going |
Token ID | 1016 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
ret | +0.001 |
orse | +0.001 |
orses | +0.001 |
peanut | +0.001 |
ets | +0.001 |
Neg logit contributions | |
Lu | -0.001 |
angels | -0.001 |
Ann | -0.001 |
Lu | -0.001 |
firefighters | -0.001 |
Token | away |
Token ID | 1497 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
orse | +0.047 |
ret | +0.045 |
orses | +0.043 |
ocket | +0.039 |
peanut | +0.038 |
Neg logit contributions | |
Lu | -0.038 |
angels | -0.031 |
firefighters | -0.029 |
attendant | -0.029 |
Taxi | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
Lu | +0.007 |
angels | +0.006 |
Ann | +0.005 |
Lu | +0.005 |
firefighters | +0.005 |
Neg logit contributions | |
ret | -0.006 |
orse | -0.006 |
orses | -0.005 |
peanut | -0.005 |
ets | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
ret | +0.126 |
ocket | +0.122 |
orse | +0.120 |
orses | +0.114 |
ocus | +0.112 |
Neg logit contributions | |
Lu | -0.069 |
youths | -0.051 |
Ann | -0.047 |
Lucy | -0.047 |
Jenkins | -0.046 |
Token |
|
Token ID | 198 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
ret | +0.038 |
orse | +0.038 |
orses | +0.035 |
peanut | +0.035 |
ets | +0.035 |
Neg logit contributions | |
Lu | -0.036 |
Ann | -0.028 |
Lu | -0.028 |
angels | -0.027 |
attendant | -0.026 |
Token | At |
Token ID | 2953 |
Feature activation | +8.21e-02 |
Pos logit contributions | |
ret | +0.139 |
ocket | +0.138 |
orse | +0.128 |
peanut | +0.123 |
alks | +0.121 |
Neg logit contributions | |
Ann | -0.071 |
Lu | -0.068 |
Lu | -0.068 |
youths | -0.062 |
Michael | -0.058 |
Token | first |
Token ID | 717 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
Lu | +0.004 |
angels | +0.003 |
Ann | +0.003 |
Lu | +0.003 |
firefighters | +0.003 |
Neg logit contributions | |
ret | -0.003 |
orse | -0.003 |
orses | -0.003 |
peanut | -0.003 |
ets | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
ret | +0.095 |
orse | +0.092 |
ifying | +0.090 |
ocket | +0.090 |
orses | +0.090 |
Neg logit contributions | |
Lu | -0.074 |
Lucy | -0.060 |
Ann | -0.054 |
angels | -0.053 |
Pen | -0.052 |
Token | Ben |
Token ID | 3932 |
Feature activation | -8.58e-01 |
Pos logit contributions | |
ret | +0.110 |
ocket | +0.101 |
orse | +0.098 |
ocus | +0.097 |
orses | +0.095 |
Neg logit contributions | |
Lu | -0.074 |
youths | -0.064 |
Luis | -0.057 |
angels | -0.056 |
firefighters | -0.056 |
Token | was |
Token ID | 373 |
Feature activation | +8.95e-01 |
Pos logit contributions | |
ret | +0.036 |
orse | +0.034 |
orses | +0.032 |
peanut | +0.029 |
ets | +0.029 |
Neg logit contributions | |
Lu | -0.041 |
Ann | -0.032 |
angels | -0.032 |
Lu | -0.032 |
firefighters | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
ret | +0.032 |
orse | +0.032 |
orses | +0.030 |
peanut | +0.027 |
ets | +0.026 |
Neg logit contributions | |
Lu | -0.033 |
angels | -0.027 |
Lu | -0.026 |
Ann | -0.025 |
firefighters | -0.025 |
Token | It |
Token ID | 632 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
Lu | +0.043 |
angels | +0.038 |
guard | +0.036 |
co | +0.036 |
Ann | +0.035 |
Neg logit contributions | |
ret | -0.053 |
orse | -0.053 |
peanut | -0.051 |
This | -0.051 |
orses | -0.050 |
Token | made |
Token ID | 925 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
Lu | +0.018 |
angels | +0.015 |
Ann | +0.015 |
youths | +0.014 |
firefighters | +0.014 |
Neg logit contributions | |
ret | -0.013 |
orse | -0.013 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Lu | +0.020 |
angels | +0.016 |
Ann | +0.016 |
youths | +0.015 |
Lu | +0.015 |
Neg logit contributions | |
ret | -0.024 |
orse | -0.023 |
orses | -0.022 |
peanut | -0.022 |
ets | -0.021 |
Token | salad |
Token ID | 20698 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
ret | +0.043 |
orse | +0.042 |
orses | +0.038 |
ocus | +0.035 |
ork | +0.035 |
Neg logit contributions | |
Lu | -0.049 |
angels | -0.045 |
firefighters | -0.042 |
youths | -0.041 |
Lu | -0.040 |
Token | look |
Token ID | 804 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
ret | +0.033 |
orse | +0.033 |
orses | +0.030 |
peanut | +0.027 |
ocket | +0.027 |
Neg logit contributions | |
Lu | -0.036 |
angels | -0.030 |
Lu | -0.028 |
firefighters | -0.028 |
Ann | -0.028 |
Token | wet |
Token ID | 9583 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
orse | +0.148 |
ret | +0.141 |
orses | +0.137 |
ocket | +0.128 |
alks | +0.123 |
Neg logit contributions | |
Lu | -0.076 |
Lucy | -0.063 |
angels | -0.061 |
firefighters | -0.059 |
Lu | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | -9.26e-01 |
Pos logit contributions | |
Lu | +0.008 |
angels | +0.007 |
Ann | +0.006 |
Lu | +0.006 |
firefighters | +0.006 |
Neg logit contributions | |
ret | -0.006 |
orse | -0.006 |
orses | -0.006 |
peanut | -0.005 |
ets | -0.005 |
Token | but |
Token ID | 475 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ret | +0.044 |
orse | +0.041 |
orses | +0.039 |
ork | +0.036 |
ocket | +0.036 |
Neg logit contributions | |
Lu | -0.040 |
angels | -0.033 |
firefighters | -0.032 |
Lu | -0.031 |
Lucy | -0.030 |
Token | it |
Token ID | 340 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
Lu | +0.049 |
co | +0.042 |
youths | +0.041 |
Taxi | +0.041 |
Ann | +0.040 |
Neg logit contributions | |
ret | -0.054 |
orse | -0.053 |
peanut | -0.050 |
orses | -0.049 |
ets | -0.045 |
Token | tasted |
Token ID | 29187 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
Lu | +0.028 |
angels | +0.024 |
Ann | +0.023 |
youths | +0.023 |
Taxi | +0.023 |
Neg logit contributions | |
orse | -0.018 |
ret | -0.018 |
orses | -0.017 |
peanut | -0.016 |
ets | -0.015 |
Token | She |
Token ID | 1375 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
Lu | +0.013 |
angels | +0.010 |
Ann | +0.010 |
Lu | +0.010 |
firefighters | +0.010 |
Neg logit contributions | |
ret | -0.013 |
orse | -0.012 |
orses | -0.012 |
peanut | -0.011 |
ets | -0.011 |
Token | picks |
Token ID | 11103 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Lu | +0.021 |
angels | +0.018 |
Lu | +0.017 |
Ann | +0.017 |
firefighters | +0.017 |
Neg logit contributions | |
ret | -0.015 |
orse | -0.014 |
orses | -0.013 |
ets | -0.012 |
peanut | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
orse | +0.056 |
orses | +0.052 |
ret | +0.052 |
ets | +0.046 |
ocket | +0.046 |
Neg logit contributions | |
Lu | -0.036 |
angels | -0.029 |
youths | -0.027 |
firefighters | -0.027 |
Ann | -0.027 |
Token | card |
Token ID | 2657 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
orse | +0.103 |
ret | +0.103 |
orses | +0.098 |
ets | +0.086 |
ocket | +0.086 |
Neg logit contributions | |
Lu | -0.088 |
angels | -0.082 |
firefighter | -0.080 |
attendant | -0.079 |
firefighters | -0.078 |
Token | with |
Token ID | 351 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
ret | +0.017 |
orse | +0.016 |
orses | +0.015 |
peanut | +0.014 |
ets | +0.014 |
Neg logit contributions | |
Lu | -0.019 |
angels | -0.016 |
Ann | -0.015 |
Lu | -0.015 |
firefighters | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
orse | +0.091 |
ret | +0.087 |
orses | +0.084 |
ork | +0.077 |
ets | +0.076 |
Neg logit contributions | |
Lu | -0.052 |
angels | -0.041 |
Lucy | -0.038 |
firefighters | -0.038 |
taxis | -0.036 |
Token | cat |
Token ID | 3797 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
ret | +0.136 |
orse | +0.127 |
orses | +0.124 |
ets | +0.119 |
earing | +0.113 |
Neg logit contributions | |
Lu | -0.102 |
firefighter | -0.096 |
angels | -0.096 |
attendant | -0.092 |
firefighters | -0.090 |
Token | on |
Token ID | 319 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ret | +0.050 |
orse | +0.048 |
orses | +0.045 |
peanut | +0.041 |
ets | +0.040 |
Neg logit contributions | |
Lu | -0.047 |
angels | -0.037 |
Ann | -0.037 |
Lu | -0.036 |
firefighters | -0.035 |
Token | it |
Token ID | 340 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
orse | +0.063 |
ret | +0.061 |
orses | +0.059 |
ets | +0.053 |
ocket | +0.053 |
Neg logit contributions | |
Lu | -0.057 |
Ann | -0.046 |
angels | -0.045 |
Lucy | -0.044 |
Lu | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
ret | +0.023 |
orse | +0.023 |
orses | +0.021 |
peanut | +0.020 |
ets | +0.019 |
Neg logit contributions | |
Lu | -0.024 |
angels | -0.019 |
Ann | -0.018 |
Lu | -0.018 |
firefighters | -0.018 |
Token | She |
Token ID | 1375 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
Lu | +0.002 |
angels | +0.002 |
Ann | +0.002 |
Lu | +0.002 |
firefighters | +0.001 |
Neg logit contributions | |
ret | -0.002 |
orse | -0.002 |
orses | -0.002 |
peanut | -0.002 |
ets | -0.002 |
Token | of |
Token ID | 286 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Lu | +0.022 |
angels | +0.017 |
Ann | +0.017 |
firefighters | +0.016 |
attendant | +0.016 |
Neg logit contributions | |
orse | -0.022 |
ret | -0.022 |
orses | -0.021 |
ets | -0.020 |
peanut | -0.020 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Lu | +0.088 |
attendant | +0.078 |
department | +0.077 |
jack | +0.076 |
Williams | +0.074 |
Neg logit contributions | |
ets | -0.043 |
orse | -0.043 |
orses | -0.042 |
ret | -0.040 |
orm | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
Lu | +0.056 |
attendant | +0.048 |
Ann | +0.047 |
Taxi | +0.046 |
Lu | +0.046 |
Neg logit contributions | |
orse | -0.033 |
orses | -0.031 |
ret | -0.029 |
ets | -0.028 |
ork | -0.024 |
Token | |
Token ID | 220 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
ret | +0.134 |
orse | +0.131 |
peanut | +0.129 |
ork | +0.122 |
ocus | +0.121 |
Neg logit contributions | |
Lu | -0.104 |
Lucy | -0.088 |
Ann | -0.087 |
Ann | -0.084 |
angels | -0.079 |
Token |
|
Token ID | 198 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
Lu | +0.006 |
angels | +0.005 |
Ann | +0.005 |
Lu | +0.005 |
firefighters | +0.004 |
Neg logit contributions | |
ret | -0.005 |
orse | -0.004 |
orses | -0.004 |
peanut | -0.004 |
ets | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
ret | +0.098 |
peanut | +0.096 |
orse | +0.096 |
sel | +0.091 |
ifying | +0.091 |
Neg logit contributions | |
Lu | -0.066 |
Ann | -0.056 |
Lu | -0.053 |
angels | -0.050 |
Pen | -0.046 |
Token | Rachel |
Token ID | 44045 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ret | +0.144 |
orse | +0.140 |
peanut | +0.139 |
ocket | +0.128 |
eg | +0.126 |
Neg logit contributions | |
Ann | -0.084 |
Lu | -0.083 |
Lu | -0.080 |
Mark | -0.073 |
Andy | -0.068 |
Token | was |
Token ID | 373 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
ret | +0.051 |
orse | +0.048 |
orses | +0.045 |
peanut | +0.043 |
ocket | +0.043 |
Neg logit contributions | |
Lu | -0.059 |
Ann | -0.047 |
angels | -0.046 |
Lu | -0.046 |
co | -0.045 |
Token | so |
Token ID | 523 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
Lu | +0.041 |
Lu | +0.035 |
angels | +0.035 |
Ann | +0.035 |
firefighters | +0.033 |
Neg logit contributions | |
orse | -0.027 |
ret | -0.026 |
orses | -0.024 |
peanut | -0.022 |
ets | -0.021 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
ret | +0.009 |
orse | +0.009 |
orses | +0.008 |
peanut | +0.008 |
ets | +0.008 |
Neg logit contributions | |
Lu | -0.010 |
angels | -0.008 |
Ann | -0.008 |
Lu | -0.008 |
firefighters | -0.007 |
Token | that |
Token ID | 326 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
ret | +0.007 |
orse | +0.007 |
orses | +0.006 |
peanut | +0.006 |
ets | +0.006 |
Neg logit contributions | |
Lu | -0.010 |
angels | -0.008 |
Ann | -0.008 |
Lu | -0.008 |
firefighters | -0.008 |