Token | " |
Token ID | 366 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
pillar | +0.025 |
squirrel | +0.024 |
villains | +0.024 |
bandits | +0.023 |
weaknesses | +0.023 |
Neg logit contributions | |
he | -0.024 |
ows | -0.024 |
like | -0.023 |
beyond | -0.021 |
inkle | -0.020 |
Token | Please |
Token ID | 5492 |
Feature activation | +7.13e-01 |
Pos logit contributions | |
he | +0.007 |
like | +0.006 |
ows | +0.006 |
beyond | +0.006 |
inkle | +0.006 |
Neg logit contributions | |
villains | -0.005 |
squirrel | -0.005 |
pillar | -0.005 |
bandits | -0.005 |
appearances | -0.005 |
Token | keep |
Token ID | 1394 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
he | +0.023 |
ows | +0.022 |
like | +0.021 |
inkle | +0.019 |
beyond | +0.018 |
Neg logit contributions | |
squirrel | -0.028 |
pillar | -0.028 |
villains | -0.027 |
bandits | -0.027 |
monster | -0.026 |
Token | your |
Token ID | 534 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
he | +0.005 |
ows | +0.005 |
like | +0.005 |
inkle | +0.004 |
beyond | +0.004 |
Neg logit contributions | |
pillar | -0.007 |
squirrel | -0.007 |
villains | -0.007 |
bandits | -0.006 |
appearances | -0.006 |
Token | seat |
Token ID | 5852 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
squirrel | +0.006 |
pillar | +0.006 |
villains | +0.005 |
bandits | +0.005 |
monster | +0.005 |
Neg logit contributions | |
he | -0.009 |
ows | -0.008 |
like | -0.008 |
inkle | -0.008 |
beyond | -0.007 |
Token | bel |
Token ID | 6667 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
he | +0.023 |
ows | +0.022 |
aron | +0.021 |
rig | +0.021 |
inkle | +0.020 |
Neg logit contributions | |
squirrel | -0.027 |
bear | -0.027 |
bird | -0.026 |
hero | -0.026 |
monster | -0.026 |
Token | ts |
Token ID | 912 |
Feature activation | +8.11e-02 |
Pos logit contributions | |
ows | +0.170 |
rig | +0.163 |
like | +0.153 |
adow | +0.149 |
overlooking | +0.145 |
Neg logit contributions | |
villains | -0.143 |
Check | -0.126 |
appearances | -0.125 |
arrest | -0.125 |
ha | -0.124 |
Token | on |
Token ID | 319 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
he | +0.002 |
ows | +0.002 |
inkle | +0.002 |
like | +0.002 |
rig | +0.002 |
Neg logit contributions | |
squirrel | -0.004 |
pillar | -0.003 |
villains | -0.003 |
turtle | -0.003 |
monster | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
squirrel | +0.022 |
pillar | +0.021 |
villains | +0.021 |
bandits | +0.020 |
monster | +0.020 |
Neg logit contributions | |
ows | -0.017 |
he | -0.017 |
like | -0.015 |
inkle | -0.014 |
immer | -0.014 |
Token | We |
Token ID | 775 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
he | +0.015 |
ows | +0.015 |
like | +0.014 |
inkle | +0.013 |
beyond | +0.013 |
Neg logit contributions | |
squirrel | -0.012 |
pillar | -0.012 |
villains | -0.011 |
bandits | -0.011 |
robbers | -0.011 |
Token | will |
Token ID | 481 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
squirrel | +0.014 |
pillar | +0.014 |
villains | +0.014 |
bandits | +0.014 |
appearances | +0.013 |
Neg logit contributions | |
he | -0.015 |
ows | -0.014 |
like | -0.014 |
inkle | -0.012 |
beyond | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
pillar | +0.027 |
bandits | +0.027 |
villains | +0.027 |
Check | +0.027 |
rook | +0.026 |
Neg logit contributions | |
like | -0.012 |
ows | -0.012 |
he | -0.011 |
behind | -0.009 |
about | -0.009 |
Token | jump |
Token ID | 4391 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
pillar | +0.031 |
villains | +0.029 |
bandits | +0.029 |
squirrel | +0.029 |
rook | +0.029 |
Neg logit contributions | |
he | -0.028 |
like | -0.027 |
ows | -0.026 |
inkle | -0.023 |
beyond | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
he | +0.009 |
ows | +0.009 |
like | +0.008 |
inkle | +0.007 |
beyond | +0.007 |
Neg logit contributions | |
squirrel | -0.017 |
pillar | -0.017 |
villains | -0.017 |
bandits | -0.016 |
monster | -0.016 |
Token | His |
Token ID | 2399 |
Feature activation | -8.42e-02 |
Pos logit contributions | |
pillar | +0.009 |
bandits | +0.009 |
villains | +0.008 |
appearances | +0.008 |
squirrel | +0.008 |
Neg logit contributions | |
ows | -0.006 |
like | -0.006 |
he | -0.006 |
about | -0.005 |
early | -0.005 |
Token | rain |
Token ID | 6290 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
pillar | +0.004 |
pill | +0.004 |
Check | +0.004 |
rook | +0.004 |
appearances | +0.004 |
Neg logit contributions | |
like | -0.002 |
ows | -0.002 |
early | -0.002 |
but | -0.002 |
behind | -0.002 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.45e+00 |
Pos logit contributions | |
ows | +0.049 |
he | +0.049 |
like | +0.048 |
beyond | +0.041 |
inkle | +0.041 |
Neg logit contributions | |
pillar | -0.044 |
villains | -0.044 |
appearances | -0.044 |
squirrel | -0.044 |
rook | -0.044 |
Token | didn |
Token ID | 1422 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
he | +0.148 |
ows | +0.142 |
like | +0.139 |
inkle | +0.121 |
beyond | +0.121 |
Neg logit contributions | |
pillar | -0.168 |
squirrel | -0.168 |
villains | -0.166 |
bandits | -0.161 |
appearances | -0.160 |
Token | 't |
Token ID | 470 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
he | +0.022 |
ows | +0.022 |
like | +0.021 |
inkle | +0.019 |
beyond | +0.019 |
Neg logit contributions | |
squirrel | -0.023 |
pillar | -0.022 |
villains | -0.022 |
bandits | -0.021 |
monster | -0.021 |
Token | fit |
Token ID | 4197 |
Feature activation | +7.13e-01 |
Pos logit contributions | |
he | +0.009 |
ows | +0.009 |
inkle | +0.008 |
like | +0.008 |
beyond | +0.008 |
Neg logit contributions | |
squirrel | -0.008 |
villains | -0.008 |
pillar | -0.008 |
robbers | -0.008 |
bandits | -0.008 |
Token | him |
Token ID | 683 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
he | +0.017 |
ows | +0.017 |
like | +0.016 |
inkle | +0.013 |
beyond | +0.013 |
Neg logit contributions | |
squirrel | -0.034 |
pillar | -0.034 |
villains | -0.033 |
bandits | -0.032 |
appearances | -0.032 |
Token | well |
Token ID | 880 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
pillar | +0.002 |
squirrel | +0.002 |
villains | +0.002 |
bandits | +0.002 |
rook | +0.001 |
Neg logit contributions | |
he | -0.001 |
ows | -0.001 |
like | -0.001 |
beyond | -0.001 |
inkle | -0.001 |
Token | on |
Token ID | 319 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
ows | +0.044 |
rig | +0.043 |
he | +0.039 |
inkle | +0.038 |
adow | +0.037 |
Neg logit contributions | |
squirrel | -0.046 |
rabbit | -0.041 |
mice | -0.040 |
bear | -0.040 |
monster | -0.040 |
Token | her |
Token ID | 607 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
pillar | +0.016 |
villains | +0.015 |
squirrel | +0.015 |
bandits | +0.015 |
rook | +0.015 |
Neg logit contributions | |
he | -0.008 |
like | -0.008 |
ows | -0.008 |
early | -0.007 |
beyond | -0.007 |
Token | new |
Token ID | 649 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
squirrel | +0.004 |
monster | +0.004 |
pillar | +0.004 |
villains | +0.004 |
rabbit | +0.004 |
Neg logit contributions | |
he | -0.006 |
ows | -0.006 |
like | -0.005 |
immer | -0.005 |
beyond | -0.005 |
Token | velvet |
Token ID | 47750 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
pillar | +0.009 |
squirrel | +0.009 |
villains | +0.008 |
bandits | +0.008 |
appearances | +0.008 |
Neg logit contributions | |
he | -0.010 |
ows | -0.010 |
like | -0.010 |
inkle | -0.009 |
beyond | -0.009 |
Token | rain |
Token ID | 6290 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ows | +0.016 |
he | +0.016 |
like | +0.015 |
inkle | +0.014 |
early | +0.014 |
Neg logit contributions | |
pillar | -0.016 |
villains | -0.016 |
squirrel | -0.016 |
Check | -0.015 |
bandits | -0.015 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.44e+00 |
Pos logit contributions | |
he | +0.025 |
ows | +0.023 |
like | +0.021 |
inkle | +0.020 |
beyond | +0.018 |
Neg logit contributions | |
squirrel | -0.056 |
pillar | -0.056 |
villains | -0.055 |
bandits | -0.053 |
robbers | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
he | +0.105 |
ows | +0.104 |
inkle | +0.100 |
immer | +0.092 |
rig | +0.092 |
Neg logit contributions | |
squirrel | -0.204 |
pillar | -0.199 |
rabbit | -0.189 |
monster | -0.188 |
turtle | -0.186 |
Token | her |
Token ID | 607 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
squirrel | +0.004 |
pillar | +0.004 |
villains | +0.004 |
bandits | +0.004 |
robbers | +0.004 |
Neg logit contributions | |
he | -0.006 |
ows | -0.006 |
like | -0.006 |
inkle | -0.006 |
beyond | -0.005 |
Token | shiny |
Token ID | 22441 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
pillar | +0.006 |
villains | +0.006 |
squirrel | +0.006 |
bandits | +0.006 |
appearances | +0.006 |
Neg logit contributions | |
like | -0.007 |
ows | -0.007 |
he | -0.007 |
inkle | -0.006 |
early | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -6.67e-02 |
Pos logit contributions | |
he | +0.015 |
ows | +0.015 |
like | +0.014 |
inkle | +0.013 |
beyond | +0.013 |
Neg logit contributions | |
squirrel | -0.011 |
pillar | -0.011 |
villains | -0.010 |
monster | -0.010 |
bandits | -0.010 |
Token | red |
Token ID | 2266 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
squirrel | +0.003 |
pillar | +0.002 |
rabbit | +0.002 |
bandits | +0.002 |
robbers | +0.002 |
Neg logit contributions | |
he | -0.002 |
ows | -0.002 |
inkle | -0.002 |
immer | -0.002 |
like | -0.002 |
Token | got |
Token ID | 1392 |
Feature activation | -8.53e-02 |
Pos logit contributions | |
he | +0.024 |
like | +0.023 |
ows | +0.023 |
beyond | +0.020 |
early | +0.020 |
Neg logit contributions | |
pillar | -0.031 |
villains | -0.030 |
appearances | -0.030 |
weaknesses | -0.030 |
squirrel | -0.029 |
Token | out |
Token ID | 503 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
pillar | +0.004 |
squirrel | +0.004 |
villains | +0.004 |
appearances | +0.004 |
bandits | +0.004 |
Neg logit contributions | |
he | -0.002 |
like | -0.002 |
ows | -0.002 |
beyond | -0.002 |
early | -0.002 |
Token | his |
Token ID | 465 |
Feature activation | -8.25e-01 |
Pos logit contributions | |
squirrel | +0.021 |
pillar | +0.019 |
villains | +0.019 |
monster | +0.019 |
bandits | +0.019 |
Neg logit contributions | |
ows | -0.018 |
he | -0.018 |
like | -0.016 |
inkle | -0.016 |
immer | -0.015 |
Token | biggest |
Token ID | 4094 |
Feature activation | +7.52e-01 |
Pos logit contributions | |
pillar | +0.027 |
squirrel | +0.027 |
villains | +0.026 |
bandits | +0.025 |
appearances | +0.025 |
Neg logit contributions | |
he | -0.032 |
ows | -0.031 |
like | -0.030 |
inkle | -0.028 |
beyond | -0.027 |
Token | rain |
Token ID | 6290 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
he | +0.028 |
like | +0.028 |
ows | +0.027 |
inkle | +0.025 |
early | +0.024 |
Neg logit contributions | |
pillar | -0.026 |
villains | -0.025 |
squirrel | -0.024 |
appearances | -0.024 |
bandits | -0.024 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
he | +0.026 |
ows | +0.025 |
like | +0.024 |
inkle | +0.021 |
beyond | +0.021 |
Neg logit contributions | |
squirrel | -0.037 |
pillar | -0.037 |
villains | -0.036 |
bandits | -0.035 |
appearances | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
he | +0.127 |
ows | +0.123 |
like | +0.116 |
inkle | +0.107 |
beyond | +0.102 |
Neg logit contributions | |
squirrel | -0.184 |
pillar | -0.182 |
villains | -0.178 |
bandits | -0.172 |
monster | -0.171 |
Token | Sue |
Token ID | 26089 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
he | +0.003 |
ows | +0.003 |
like | +0.003 |
beyond | +0.002 |
early | +0.002 |
Neg logit contributions | |
pillar | -0.004 |
squirrel | -0.004 |
villains | -0.004 |
bandits | -0.004 |
appearances | -0.004 |
Token | put |
Token ID | 1234 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
he | +0.047 |
ows | +0.046 |
like | +0.045 |
inkle | +0.038 |
early | +0.038 |
Neg logit contributions | |
appearances | -0.053 |
villains | -0.053 |
pillar | -0.052 |
bandits | -0.051 |
squirrel | -0.051 |
Token | it |
Token ID | 340 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
ows | +0.034 |
he | +0.033 |
like | +0.030 |
rig | +0.030 |
inkle | +0.029 |
Neg logit contributions | |
squirrel | -0.051 |
pillar | -0.048 |
monster | -0.047 |
bandits | -0.046 |
rabbit | -0.046 |
Token | on |
Token ID | 319 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
squirrel | +0.031 |
pillar | +0.030 |
villains | +0.030 |
bandits | +0.029 |
robbers | +0.029 |
Neg logit contributions | |
he | -0.015 |
ows | -0.014 |
like | -0.013 |
inkle | -0.012 |
beyond | -0.011 |
Token | ashing |
Token ID | 2140 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
pillar | +0.023 |
squirrel | +0.023 |
villains | +0.022 |
weaknesses | +0.021 |
bandits | +0.021 |
Neg logit contributions | |
he | -0.039 |
ows | -0.037 |
like | -0.037 |
inkle | -0.035 |
beyond | -0.034 |
Token | around |
Token ID | 1088 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
like | +0.012 |
he | +0.012 |
ows | +0.011 |
beyond | +0.009 |
behind | +0.009 |
Neg logit contributions | |
pillar | -0.031 |
appearances | -0.030 |
villains | -0.030 |
rook | -0.030 |
squirrel | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
pillar | +0.010 |
weaknesses | +0.010 |
appearances | +0.010 |
bandits | +0.009 |
villains | +0.009 |
Neg logit contributions | |
like | -0.004 |
he | -0.003 |
behind | -0.003 |
beyond | -0.003 |
ows | -0.003 |
Token | His |
Token ID | 2399 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
squirrel | +0.023 |
appearances | +0.023 |
villains | +0.023 |
pillar | +0.023 |
bandits | +0.022 |
Neg logit contributions | |
he | -0.016 |
like | -0.015 |
ows | -0.015 |
about | -0.014 |
early | -0.014 |
Token | rain |
Token ID | 6290 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
pillar | +0.008 |
pill | +0.007 |
appearances | +0.007 |
Le | +0.007 |
Carl | +0.007 |
Neg logit contributions | |
like | -0.005 |
early | -0.005 |
ows | -0.005 |
he | -0.004 |
inkle | -0.004 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
ows | +0.039 |
he | +0.039 |
like | +0.037 |
inkle | +0.033 |
beyond | +0.032 |
Neg logit contributions | |
pillar | -0.041 |
squirrel | -0.041 |
villains | -0.040 |
bandits | -0.040 |
appearances | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
he | +0.141 |
ows | +0.137 |
like | +0.136 |
beyond | +0.114 |
inkle | +0.113 |
Neg logit contributions | |
pillar | -0.164 |
squirrel | -0.162 |
villains | -0.162 |
bandits | -0.160 |
appearances | -0.159 |
Token | a |
Token ID | 257 |
Feature activation | -7.31e-02 |
Pos logit contributions | |
squirrel | +0.018 |
pillar | +0.017 |
villains | +0.017 |
monster | +0.016 |
bandits | +0.016 |
Neg logit contributions | |
he | -0.012 |
ows | -0.012 |
like | -0.010 |
inkle | -0.010 |
beyond | -0.009 |
Token | bit |
Token ID | 1643 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
appearances | +0.003 |
pillar | +0.003 |
villains | +0.003 |
weaknesses | +0.003 |
bandits | +0.003 |
Neg logit contributions | |
like | -0.002 |
he | -0.002 |
ows | -0.002 |
early | -0.002 |
inkle | -0.002 |
Token | too |
Token ID | 1165 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
squirrel | +0.047 |
pillar | +0.046 |
villains | +0.045 |
bandits | +0.044 |
monster | +0.043 |
Neg logit contributions | |
he | -0.039 |
ows | -0.038 |
like | -0.035 |
inkle | -0.033 |
beyond | -0.032 |
Token | big |
Token ID | 1263 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
squirrel | +0.005 |
pillar | +0.005 |
villains | +0.004 |
monster | +0.004 |
bandits | +0.004 |
Neg logit contributions | |
he | -0.004 |
ows | -0.004 |
like | -0.004 |
inkle | -0.004 |
immer | -0.004 |
Token | He |
Token ID | 679 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
pillar | +0.012 |
villains | +0.012 |
squirrel | +0.012 |
appearances | +0.012 |
bandits | +0.012 |
Neg logit contributions | |
he | -0.010 |
ows | -0.010 |
like | -0.010 |
beyond | -0.009 |
early | -0.009 |
Token | put |
Token ID | 1234 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
appearances | +0.009 |
pillar | +0.009 |
rook | +0.008 |
weaknesses | +0.008 |
villains | +0.008 |
Neg logit contributions | |
like | -0.010 |
ows | -0.010 |
he | -0.009 |
beyond | -0.009 |
almost | -0.009 |
Token | on |
Token ID | 319 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
ows | +0.042 |
he | +0.041 |
rig | +0.037 |
like | +0.037 |
inkle | +0.036 |
Neg logit contributions | |
squirrel | -0.051 |
monster | -0.047 |
pillar | -0.046 |
rabbit | -0.046 |
bandits | -0.046 |
Token | his |
Token ID | 465 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
pillar | +0.008 |
villains | +0.008 |
squirrel | +0.008 |
appearances | +0.008 |
weaknesses | +0.007 |
Neg logit contributions | |
like | -0.006 |
he | -0.006 |
ows | -0.006 |
early | -0.006 |
beyond | -0.005 |
Token | rain |
Token ID | 6290 |
Feature activation | +9.65e-01 |
Pos logit contributions | |
squirrel | +0.001 |
pillar | +0.001 |
villains | +0.001 |
bandits | +0.001 |
robbers | +0.001 |
Neg logit contributions | |
he | -0.001 |
ows | -0.001 |
like | -0.001 |
inkle | -0.001 |
beyond | -0.001 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
he | +0.024 |
ows | +0.022 |
like | +0.021 |
inkle | +0.020 |
beyond | +0.018 |
Neg logit contributions | |
squirrel | -0.045 |
pillar | -0.044 |
villains | -0.043 |
bandits | -0.042 |
robbers | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
he | +0.101 |
ows | +0.101 |
inkle | +0.094 |
rig | +0.087 |
immer | +0.087 |
Neg logit contributions | |
squirrel | -0.194 |
pillar | -0.187 |
monster | -0.179 |
rabbit | -0.178 |
turtle | -0.178 |
Token | boots |
Token ID | 14412 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
pillar | +0.007 |
villains | +0.007 |
squirrel | +0.007 |
appearances | +0.006 |
weaknesses | +0.006 |
Neg logit contributions | |
he | -0.009 |
ows | -0.009 |
like | -0.009 |
early | -0.008 |
beyond | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
he | +0.056 |
ows | +0.056 |
inkle | +0.055 |
rig | +0.051 |
like | +0.049 |
Neg logit contributions | |
squirrel | -0.104 |
pillar | -0.100 |
villains | -0.095 |
rabbit | -0.095 |
turtle | -0.095 |
Token | spun |
Token ID | 26843 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
pillar | +0.021 |
squirrel | +0.020 |
villains | +0.020 |
appearances | +0.020 |
bandits | +0.019 |
Neg logit contributions | |
he | -0.030 |
ows | -0.029 |
like | -0.029 |
beyond | -0.026 |
early | -0.026 |
Token | in |
Token ID | 287 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
squirrel | +0.024 |
pillar | +0.023 |
villains | +0.022 |
monster | +0.022 |
bandits | +0.022 |
Neg logit contributions | |
he | -0.015 |
ows | -0.015 |
inkle | -0.013 |
immer | -0.012 |
adow | -0.012 |
Token | He |
Token ID | 679 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
pillar | +0.014 |
squirrel | +0.013 |
villains | +0.013 |
bandits | +0.013 |
appearances | +0.013 |
Neg logit contributions | |
ows | -0.014 |
he | -0.014 |
like | -0.014 |
about | -0.013 |
early | -0.013 |
Token | put |
Token ID | 1234 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
pillar | +0.011 |
squirrel | +0.011 |
appearances | +0.011 |
villains | +0.011 |
rook | +0.011 |
Neg logit contributions | |
he | -0.014 |
ows | -0.013 |
like | -0.013 |
inkle | -0.012 |
early | -0.012 |
Token | on |
Token ID | 319 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
ows | +0.050 |
he | +0.047 |
rig | +0.046 |
inkle | +0.044 |
adow | +0.044 |
Neg logit contributions | |
squirrel | -0.043 |
monster | -0.039 |
bear | -0.038 |
rabbit | -0.038 |
robbers | -0.037 |
Token | his |
Token ID | 465 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
pillar | +0.006 |
squirrel | +0.006 |
villains | +0.006 |
bandits | +0.006 |
appearances | +0.006 |
Neg logit contributions | |
he | -0.006 |
ows | -0.006 |
like | -0.006 |
inkle | -0.005 |
early | -0.005 |
Token | rain |
Token ID | 6290 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
squirrel | +0.001 |
pillar | +0.001 |
villains | +0.001 |
bandits | +0.001 |
monster | +0.001 |
Neg logit contributions | |
he | -0.002 |
ows | -0.002 |
like | -0.002 |
inkle | -0.002 |
beyond | -0.002 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.21e+00 |
Pos logit contributions | |
he | +0.021 |
ows | +0.020 |
like | +0.018 |
inkle | +0.018 |
beyond | +0.016 |
Neg logit contributions | |
squirrel | -0.045 |
pillar | -0.044 |
villains | -0.043 |
bandits | -0.042 |
robbers | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
ows | +0.107 |
inkle | +0.104 |
he | +0.103 |
rig | +0.096 |
immer | +0.096 |
Neg logit contributions | |
squirrel | -0.187 |
pillar | -0.179 |
monster | -0.171 |
rabbit | -0.170 |
turtle | -0.169 |
Token | boots |
Token ID | 14412 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
pillar | +0.007 |
squirrel | +0.007 |
villains | +0.006 |
appearances | +0.006 |
bandits | +0.006 |
Neg logit contributions | |
he | -0.010 |
ows | -0.010 |
like | -0.010 |
inkle | -0.009 |
beyond | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
inkle | +0.062 |
ows | +0.060 |
he | +0.059 |
rig | +0.057 |
immer | +0.053 |
Neg logit contributions | |
squirrel | -0.099 |
pillar | -0.095 |
villains | -0.091 |
rabbit | -0.089 |
monster | -0.089 |
Token | ran |
Token ID | 4966 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
pillar | +0.020 |
squirrel | +0.019 |
villains | +0.018 |
appearances | +0.018 |
bandits | +0.017 |
Neg logit contributions | |
he | -0.029 |
ows | -0.028 |
like | -0.028 |
beyond | -0.025 |
early | -0.025 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
squirrel | +0.048 |
pillar | +0.048 |
villains | +0.047 |
bandits | +0.046 |
appearances | +0.045 |
Neg logit contributions | |
he | -0.021 |
ows | -0.020 |
like | -0.019 |
inkle | -0.016 |
beyond | -0.015 |
Token | She |
Token ID | 1375 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
he | +0.001 |
ows | +0.001 |
like | +0.001 |
beyond | +0.001 |
early | +0.001 |
Neg logit contributions | |
pillar | -0.001 |
squirrel | -0.001 |
villains | -0.001 |
bandits | -0.001 |
weaknesses | -0.001 |
Token | took |
Token ID | 1718 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
pillar | +0.005 |
squirrel | +0.005 |
appearances | +0.005 |
villains | +0.005 |
weaknesses | +0.005 |
Neg logit contributions | |
ows | -0.005 |
he | -0.005 |
like | -0.005 |
inkle | -0.004 |
?" | -0.004 |
Token | off |
Token ID | 572 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
like | +0.007 |
he | +0.007 |
ows | +0.005 |
behind | +0.005 |
beyond | +0.005 |
Neg logit contributions | |
pillar | -0.017 |
villains | -0.016 |
weaknesses | -0.016 |
appearances | -0.015 |
colony | -0.015 |
Token | her |
Token ID | 607 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
ows | +0.016 |
he | +0.016 |
like | +0.015 |
inkle | +0.014 |
immer | +0.014 |
Neg logit contributions | |
squirrel | -0.020 |
pillar | -0.018 |
villains | -0.018 |
bandits | -0.018 |
monster | -0.018 |
Token | rain |
Token ID | 6290 |
Feature activation | +9.66e-01 |
Pos logit contributions | |
he | +0.002 |
ows | +0.002 |
like | +0.002 |
inkle | +0.002 |
beyond | +0.002 |
Neg logit contributions | |
squirrel | -0.001 |
pillar | -0.001 |
villains | -0.001 |
monster | -0.001 |
bandits | -0.001 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.20e+00 |
Pos logit contributions | |
he | +0.023 |
ows | +0.021 |
like | +0.020 |
inkle | +0.020 |
immer | +0.018 |
Neg logit contributions | |
squirrel | -0.045 |
pillar | -0.043 |
villains | -0.043 |
monster | -0.042 |
bandits | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
ows | +0.097 |
he | +0.096 |
inkle | +0.093 |
immer | +0.085 |
rig | +0.084 |
Neg logit contributions | |
squirrel | -0.196 |
pillar | -0.188 |
villains | -0.182 |
monster | -0.182 |
robbers | -0.179 |
Token | boots |
Token ID | 14412 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
pillar | +0.003 |
villains | +0.003 |
squirrel | +0.003 |
appearances | +0.003 |
weaknesses | +0.003 |
Neg logit contributions | |
ows | -0.003 |
like | -0.003 |
he | -0.003 |
early | -0.003 |
?" | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -2.69e-03 |
Pos logit contributions | |
inkle | +0.055 |
he | +0.055 |
ows | +0.053 |
rig | +0.051 |
like | +0.048 |
Neg logit contributions | |
squirrel | -0.095 |
pillar | -0.091 |
villains | -0.088 |
monster | -0.086 |
rabbit | -0.086 |
Token | She |
Token ID | 1375 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
pillar | +0.000 |
squirrel | +0.000 |
villains | +0.000 |
bandits | +0.000 |
monster | +0.000 |
Neg logit contributions | |
ows | -0.000 |
he | -0.000 |
like | -0.000 |
beyond | -0.000 |
early | -0.000 |
Token | ran |
Token ID | 4966 |
Feature activation | -5.25e-02 |
Pos logit contributions | |
pillar | +0.004 |
squirrel | +0.004 |
villains | +0.004 |
appearances | +0.004 |
weaknesses | +0.004 |
Neg logit contributions | |
ows | -0.004 |
he | -0.004 |
like | -0.004 |
inkle | -0.003 |
beyond | -0.003 |
Token | put |
Token ID | 1234 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
he | +0.042 |
ows | +0.041 |
like | +0.038 |
inkle | +0.037 |
immer | +0.036 |
Neg logit contributions | |
squirrel | -0.031 |
pillar | -0.029 |
villains | -0.029 |
bandits | -0.028 |
robbers | -0.028 |
Token | on |
Token ID | 319 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
ows | +0.043 |
he | +0.042 |
rig | +0.041 |
immer | +0.038 |
inkle | +0.038 |
Neg logit contributions | |
squirrel | -0.055 |
mice | -0.049 |
rabbit | -0.049 |
monster | -0.048 |
rabbits | -0.048 |
Token | her |
Token ID | 607 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
pillar | +0.017 |
villains | +0.017 |
squirrel | +0.016 |
rook | +0.016 |
warrior | +0.016 |
Neg logit contributions | |
like | -0.008 |
he | -0.008 |
ows | -0.008 |
early | -0.007 |
?" | -0.006 |
Token | colorful |
Token ID | 20239 |
Feature activation | +7.51e-01 |
Pos logit contributions | |
squirrel | +0.007 |
rabbit | +0.006 |
monster | +0.006 |
turtle | +0.006 |
pillar | +0.006 |
Neg logit contributions | |
he | -0.012 |
ows | -0.011 |
like | -0.010 |
immer | -0.010 |
beyond | -0.010 |
Token | rain |
Token ID | 6290 |
Feature activation | +8.17e-01 |
Pos logit contributions | |
ows | +0.027 |
he | +0.026 |
like | +0.026 |
inkle | +0.024 |
?" | +0.023 |
Neg logit contributions | |
pillar | -0.026 |
villains | -0.025 |
squirrel | -0.025 |
appearances | -0.024 |
bandits | -0.024 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.17e+00 |
Pos logit contributions | |
he | +0.027 |
inkle | +0.025 |
immer | +0.023 |
ours | +0.022 |
ows | +0.022 |
Neg logit contributions | |
squirrel | -0.032 |
pillar | -0.029 |
villains | -0.029 |
rabbit | -0.029 |
turtle | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
rig | +0.145 |
inkle | +0.144 |
immer | +0.135 |
ows | +0.132 |
he | +0.130 |
Neg logit contributions | |
squirrel | -0.152 |
rabbit | -0.136 |
bear | -0.135 |
pillar | -0.135 |
turtle | -0.132 |
Token | jump |
Token ID | 4391 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
squirrel | +0.007 |
pillar | +0.007 |
villains | +0.007 |
bandits | +0.006 |
robbers | +0.006 |
Neg logit contributions | |
he | -0.011 |
ows | -0.010 |
like | -0.010 |
inkle | -0.009 |
immer | -0.009 |
Token | in |
Token ID | 287 |
Feature activation | -5.49e-01 |
Pos logit contributions | |
squirrel | +0.001 |
pillar | +0.001 |
rabbit | +0.001 |
monster | +0.001 |
villains | +0.001 |
Neg logit contributions | |
he | -0.001 |
ows | -0.000 |
inkle | -0.000 |
adow | -0.000 |
rig | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
squirrel | +0.020 |
pillar | +0.018 |
monster | +0.018 |
rabbit | +0.017 |
villains | +0.017 |
Neg logit contributions | |
ows | -0.021 |
he | -0.020 |
inkle | -0.018 |
like | -0.017 |
immer | -0.017 |
Token | p |
Token ID | 279 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
villains | +0.024 |
pillar | +0.023 |
squirrel | +0.022 |
appearances | +0.022 |
rook | +0.022 |
Neg logit contributions | |
he | -0.042 |
ows | -0.041 |
like | -0.040 |
inkle | -0.037 |
beyond | -0.037 |
Token | I |
Token ID | 314 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
pillar | +0.003 |
squirrel | +0.003 |
villains | +0.003 |
bandits | +0.003 |
robbers | +0.003 |
Neg logit contributions | |
he | -0.003 |
ows | -0.003 |
like | -0.003 |
inkle | -0.003 |
beyond | -0.003 |
Token | can |
Token ID | 460 |
Feature activation | -9.18e-01 |
Pos logit contributions | |
squirrel | +0.008 |
pillar | +0.008 |
villains | +0.007 |
monster | +0.007 |
weaknesses | +0.007 |
Neg logit contributions | |
he | -0.007 |
ows | -0.006 |
immer | -0.006 |
beyond | -0.006 |
inkle | -0.006 |
Token | wear |
Token ID | 5806 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
squirrel | +0.033 |
pillar | +0.032 |
villains | +0.031 |
monster | +0.030 |
bandits | +0.030 |
Neg logit contributions | |
he | -0.032 |
ows | -0.032 |
like | -0.030 |
inkle | -0.028 |
beyond | -0.028 |
Token | my |
Token ID | 616 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
he | +0.029 |
ows | +0.028 |
like | +0.024 |
rig | +0.024 |
immer | +0.024 |
Neg logit contributions | |
squirrel | -0.032 |
pillar | -0.031 |
villains | -0.030 |
monster | -0.030 |
bandits | -0.030 |
Token | rain |
Token ID | 6290 |
Feature activation | +8.70e-01 |
Pos logit contributions | |
he | +0.032 |
ows | +0.031 |
like | +0.028 |
beyond | +0.027 |
inkle | +0.027 |
Neg logit contributions | |
squirrel | -0.023 |
pillar | -0.022 |
villains | -0.021 |
monster | -0.021 |
rabbit | -0.021 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
he | +0.021 |
ows | +0.019 |
inkle | +0.019 |
like | +0.017 |
immer | +0.016 |
Neg logit contributions | |
squirrel | -0.040 |
pillar | -0.039 |
villains | -0.039 |
monster | -0.038 |
robbers | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
inkle | +0.118 |
ows | +0.118 |
he | +0.114 |
rig | +0.111 |
immer | +0.110 |
Neg logit contributions | |
squirrel | -0.167 |
pillar | -0.162 |
monster | -0.158 |
rabbit | -0.153 |
turtle | -0.152 |
Token | boots |
Token ID | 14412 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
pillar | +0.010 |
squirrel | +0.010 |
villains | +0.010 |
bandits | +0.009 |
appearances | +0.009 |
Neg logit contributions | |
he | -0.012 |
ows | -0.012 |
like | -0.011 |
inkle | -0.010 |
beyond | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +7.32e-02 |
Pos logit contributions | |
ows | +0.058 |
inkle | +0.058 |
he | +0.058 |
rig | +0.055 |
immer | +0.053 |
Neg logit contributions | |
squirrel | -0.090 |
pillar | -0.089 |
villains | -0.084 |
monster | -0.084 |
robbers | -0.082 |
Token | Please |
Token ID | 4222 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
he | +0.003 |
ows | +0.003 |
like | +0.003 |
inkle | +0.002 |
immer | +0.002 |
Neg logit contributions | |
squirrel | -0.002 |
pillar | -0.002 |
villains | -0.002 |
bandits | -0.002 |
robbers | -0.002 |
Token | mom |
Token ID | 1995 |
Feature activation | +9.49e-01 |
Pos logit contributions | |
he | +0.012 |
ows | +0.011 |
beyond | +0.010 |
rig | +0.010 |
early | +0.010 |
Neg logit contributions | |
squirrel | -0.012 |
pillar | -0.011 |
rabbit | -0.011 |
monster | -0.011 |
villains | -0.011 |
Token | She |
Token ID | 1375 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
pillar | +0.005 |
squirrel | +0.005 |
villains | +0.005 |
appearances | +0.005 |
bandits | +0.005 |
Neg logit contributions | |
he | -0.005 |
ows | -0.005 |
like | -0.005 |
early | -0.004 |
about | -0.004 |
Token | put |
Token ID | 1234 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
he | +0.001 |
like | +0.001 |
ows | +0.001 |
?" | +0.001 |
beyond | +0.001 |
Neg logit contributions | |
pillar | -0.001 |
appearances | -0.001 |
weaknesses | -0.001 |
rook | -0.001 |
squirrel | -0.001 |
Token | on |
Token ID | 319 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
ows | +0.039 |
he | +0.038 |
rig | +0.037 |
inkle | +0.034 |
immer | +0.033 |
Neg logit contributions | |
squirrel | -0.056 |
mice | -0.051 |
pillar | -0.051 |
monster | -0.051 |
rabbit | -0.051 |
Token | her |
Token ID | 607 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
pillar | +0.006 |
villains | +0.006 |
squirrel | +0.006 |
bandits | +0.006 |
weaknesses | +0.006 |
Neg logit contributions | |
he | -0.004 |
ows | -0.003 |
like | -0.003 |
early | -0.003 |
inkle | -0.003 |
Token | rain |
Token ID | 6290 |
Feature activation | +8.99e-01 |
Pos logit contributions | |
squirrel | +0.001 |
monster | +0.001 |
turtle | +0.001 |
rabbit | +0.001 |
pillar | +0.001 |
Neg logit contributions | |
he | -0.002 |
ows | -0.002 |
like | -0.002 |
immer | -0.002 |
beyond | -0.001 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
he | +0.023 |
inkle | +0.022 |
ows | +0.020 |
immer | +0.020 |
like | +0.019 |
Neg logit contributions | |
squirrel | -0.040 |
pillar | -0.038 |
villains | -0.037 |
monster | -0.037 |
robbers | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
inkle | +0.111 |
ows | +0.109 |
he | +0.109 |
rig | +0.103 |
immer | +0.102 |
Neg logit contributions | |
squirrel | -0.180 |
pillar | -0.169 |
rabbit | -0.163 |
monster | -0.162 |
turtle | -0.162 |
Token | boots |
Token ID | 14412 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
pillar | +0.008 |
squirrel | +0.008 |
villains | +0.008 |
appearances | +0.007 |
weaknesses | +0.007 |
Neg logit contributions | |
he | -0.010 |
ows | -0.010 |
like | -0.010 |
beyond | -0.009 |
inkle | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
inkle | +0.061 |
he | +0.059 |
ows | +0.058 |
rig | +0.057 |
immer | +0.053 |
Neg logit contributions | |
squirrel | -0.096 |
pillar | -0.092 |
villains | -0.086 |
rabbit | -0.086 |
turtle | -0.086 |
Token | She |
Token ID | 1375 |
Feature activation | +6.06e-02 |
Pos logit contributions | |
pillar | +0.005 |
squirrel | +0.005 |
villains | +0.005 |
appearances | +0.005 |
bandits | +0.005 |
Neg logit contributions | |
ows | -0.004 |
he | -0.004 |
like | -0.004 |
early | -0.003 |
beyond | -0.003 |
Token | did |
Token ID | 750 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
ows | +0.002 |
he | +0.002 |
like | +0.002 |
beyond | +0.002 |
inkle | +0.002 |
Neg logit contributions | |
pillar | -0.002 |
appearances | -0.002 |
squirrel | -0.002 |
villains | -0.002 |
weaknesses | -0.002 |
Token | on |
Token ID | 319 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
he | +0.037 |
ows | +0.035 |
like | +0.035 |
inkle | +0.030 |
beyond | +0.030 |
Neg logit contributions | |
pillar | -0.050 |
villains | -0.050 |
squirrel | -0.050 |
appearances | -0.047 |
bandits | -0.047 |
Token | his |
Token ID | 465 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
pillar | +0.020 |
villains | +0.020 |
squirrel | +0.019 |
weaknesses | +0.018 |
bandits | +0.018 |
Neg logit contributions | |
like | -0.011 |
he | -0.010 |
ows | -0.010 |
early | -0.009 |
inkle | -0.009 |
Token | bright |
Token ID | 6016 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
squirrel | +0.006 |
pillar | +0.006 |
villains | +0.006 |
bandits | +0.005 |
appearances | +0.005 |
Neg logit contributions | |
he | -0.008 |
ows | -0.008 |
like | -0.008 |
inkle | -0.007 |
beyond | -0.007 |
Token | yellow |
Token ID | 7872 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
he | +0.009 |
ows | +0.009 |
like | +0.009 |
inkle | +0.008 |
beyond | +0.008 |
Neg logit contributions | |
squirrel | -0.007 |
pillar | -0.007 |
villains | -0.007 |
bandits | -0.006 |
robbers | -0.006 |
Token | rain |
Token ID | 6290 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
like | +0.038 |
ows | +0.037 |
he | +0.036 |
inkle | +0.035 |
?" | +0.033 |
Neg logit contributions | |
pillar | -0.031 |
villains | -0.031 |
squirrel | -0.029 |
appearances | -0.029 |
bandits | -0.029 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
ows | +0.020 |
he | +0.019 |
like | +0.019 |
beyond | +0.015 |
early | +0.015 |
Neg logit contributions | |
pillar | -0.040 |
squirrel | -0.039 |
villains | -0.039 |
bandits | -0.038 |
appearances | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
he | +0.090 |
ows | +0.087 |
like | +0.077 |
inkle | +0.076 |
rig | +0.069 |
Neg logit contributions | |
squirrel | -0.201 |
pillar | -0.198 |
villains | -0.190 |
monster | -0.187 |
bandits | -0.186 |
Token | went |
Token ID | 1816 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
pillar | +0.015 |
villains | +0.015 |
appearances | +0.015 |
squirrel | +0.014 |
weaknesses | +0.014 |
Neg logit contributions | |
like | -0.015 |
ows | -0.015 |
he | -0.014 |
early | -0.013 |
?" | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
squirrel | +0.030 |
pillar | +0.030 |
villains | +0.029 |
bandits | +0.028 |
robbers | +0.028 |
Neg logit contributions | |
he | -0.019 |
ows | -0.018 |
like | -0.017 |
inkle | -0.015 |
beyond | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
pillar | +0.029 |
squirrel | +0.029 |
villains | +0.028 |
bandits | +0.027 |
appearances | +0.027 |
Neg logit contributions | |
he | -0.025 |
ows | -0.024 |
like | -0.023 |
inkle | -0.021 |
beyond | -0.021 |
Token | river |
Token ID | 7850 |
Feature activation | -8.95e-01 |
Pos logit contributions | |
pillar | +0.005 |
weaknesses | +0.005 |
appearances | +0.004 |
villains | +0.004 |
emy | +0.004 |
Neg logit contributions | |
like | -0.006 |
beyond | -0.006 |
he | -0.006 |
behind | -0.006 |
early | -0.006 |
Token | She |
Token ID | 1375 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
villains | +0.008 |
pillar | +0.008 |
squirrel | +0.008 |
appearances | +0.007 |
rook | +0.007 |
Neg logit contributions | |
he | -0.006 |
ows | -0.006 |
like | -0.006 |
beyond | -0.005 |
early | -0.005 |
Token | put |
Token ID | 1234 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
squirrel | +0.005 |
pillar | +0.005 |
villains | +0.005 |
bandits | +0.005 |
robbers | +0.005 |
Neg logit contributions | |
he | -0.007 |
ows | -0.007 |
like | -0.007 |
inkle | -0.006 |
beyond | -0.006 |
Token | on |
Token ID | 319 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
ows | +0.037 |
rig | +0.036 |
he | +0.034 |
inkle | +0.032 |
immer | +0.031 |
Neg logit contributions | |
squirrel | -0.041 |
rabbit | -0.036 |
bear | -0.036 |
rabbits | -0.035 |
monster | -0.035 |
Token | her |
Token ID | 607 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
squirrel | +0.026 |
pillar | +0.026 |
villains | +0.025 |
bandits | +0.025 |
appearances | +0.024 |
Neg logit contributions | |
he | -0.020 |
ows | -0.019 |
like | -0.018 |
inkle | -0.016 |
beyond | -0.016 |
Token | rain |
Token ID | 6290 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
squirrel | +0.007 |
pillar | +0.007 |
monster | +0.007 |
rabbit | +0.006 |
villains | +0.006 |
Neg logit contributions | |
he | -0.009 |
ows | -0.008 |
like | -0.008 |
immer | -0.007 |
beyond | -0.007 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.10e+00 |
Pos logit contributions | |
he | +0.026 |
inkle | +0.024 |
ows | +0.022 |
like | +0.022 |
ours | +0.022 |
Neg logit contributions | |
squirrel | -0.030 |
pillar | -0.027 |
rabbit | -0.027 |
robbers | -0.026 |
villains | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
inkle | +0.120 |
rig | +0.118 |
he | +0.118 |
ows | +0.115 |
immer | +0.109 |
Neg logit contributions | |
squirrel | -0.164 |
rabbit | -0.149 |
pillar | -0.148 |
turtle | -0.146 |
bear | -0.146 |
Token | went |
Token ID | 1816 |
Feature activation | -6.99e-01 |
Pos logit contributions | |
squirrel | +0.012 |
pillar | +0.011 |
villains | +0.011 |
bandits | +0.011 |
robbers | +0.011 |
Neg logit contributions | |
he | -0.018 |
ows | -0.018 |
like | -0.017 |
inkle | -0.016 |
beyond | -0.016 |
Token | outside |
Token ID | 2354 |
Feature activation | -9.36e-01 |
Pos logit contributions | |
squirrel | +0.030 |
robbers | +0.027 |
turtle | +0.027 |
rabbit | +0.027 |
bandits | +0.027 |
Neg logit contributions | |
he | -0.021 |
ows | -0.019 |
inkle | -0.018 |
ieved | -0.017 |
immer | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
squirrel | +0.033 |
monster | +0.030 |
rabbit | +0.030 |
pillar | +0.030 |
villains | +0.030 |
Neg logit contributions | |
ows | -0.033 |
he | -0.033 |
inkle | -0.030 |
immer | -0.029 |
like | -0.028 |
Token | play |
Token ID | 711 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
squirrel | +0.019 |
pillar | +0.018 |
villains | +0.017 |
monster | +0.017 |
rabbit | +0.017 |
Neg logit contributions | |
he | -0.026 |
ows | -0.026 |
like | -0.024 |
inkle | -0.023 |
immer | -0.023 |
Token | Rose |
Token ID | 31087 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
squirrel | +0.003 |
pillar | +0.003 |
appearances | +0.003 |
rook | +0.003 |
villains | +0.003 |
Neg logit contributions | |
he | -0.004 |
ows | -0.004 |
like | -0.004 |
beyond | -0.003 |
inkle | -0.003 |
Token | grabbed |
Token ID | 13176 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
he | +0.005 |
ows | +0.005 |
like | +0.005 |
inkle | +0.004 |
beyond | +0.004 |
Neg logit contributions | |
squirrel | -0.005 |
villains | -0.005 |
pillar | -0.005 |
appearances | -0.005 |
rook | -0.005 |
Token | her |
Token ID | 607 |
Feature activation | -6.19e-01 |
Pos logit contributions | |
he | +0.025 |
ows | +0.024 |
like | +0.024 |
inkle | +0.021 |
beyond | +0.021 |
Neg logit contributions | |
pillar | -0.035 |
squirrel | -0.035 |
villains | -0.035 |
bandits | -0.033 |
appearances | -0.033 |
Token | blue |
Token ID | 4171 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
squirrel | +0.018 |
pillar | +0.017 |
monster | +0.016 |
rabbit | +0.016 |
villains | +0.016 |
Neg logit contributions | |
he | -0.027 |
ows | -0.026 |
like | -0.025 |
inkle | -0.023 |
immer | -0.023 |
Token | rain |
Token ID | 6290 |
Feature activation | +7.91e-01 |
Pos logit contributions | |
he | +0.019 |
ows | +0.019 |
like | +0.018 |
inkle | +0.017 |
beyond | +0.016 |
Neg logit contributions | |
pillar | -0.016 |
squirrel | -0.016 |
villains | -0.016 |
bandits | -0.015 |
weaknesses | -0.015 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
he | +0.018 |
ows | +0.016 |
like | +0.015 |
inkle | +0.015 |
immer | +0.014 |
Neg logit contributions | |
squirrel | -0.038 |
pillar | -0.037 |
villains | -0.036 |
monster | -0.036 |
bandits | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
inkle | +0.105 |
ows | +0.104 |
he | +0.104 |
rig | +0.096 |
immer | +0.095 |
Neg logit contributions | |
squirrel | -0.179 |
pillar | -0.170 |
rabbit | -0.164 |
monster | -0.163 |
turtle | -0.161 |
Token | ran |
Token ID | 4966 |
Feature activation | -8.09e-01 |
Pos logit contributions | |
pillar | +0.007 |
squirrel | +0.007 |
villains | +0.007 |
bandits | +0.006 |
appearances | +0.006 |
Neg logit contributions | |
he | -0.011 |
ows | -0.011 |
like | -0.010 |
inkle | -0.009 |
beyond | -0.009 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
squirrel | +0.038 |
pillar | +0.037 |
villains | +0.037 |
robbers | +0.036 |
bandits | +0.036 |
Neg logit contributions | |
he | -0.020 |
ows | -0.019 |
inkle | -0.017 |
like | -0.016 |
immer | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
squirrel | +0.033 |
monster | +0.030 |
villains | +0.030 |
rabbit | +0.030 |
turtle | +0.029 |
Neg logit contributions | |
ows | -0.039 |
he | -0.038 |
inkle | -0.037 |
immer | -0.036 |
like | -0.034 |
Token | She |
Token ID | 1375 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
pillar | +0.019 |
squirrel | +0.019 |
villains | +0.019 |
appearances | +0.019 |
bandits | +0.018 |
Neg logit contributions | |
ows | -0.012 |
like | -0.012 |
he | -0.011 |
about | -0.010 |
beyond | -0.010 |
Token | She |
Token ID | 1375 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
he | +0.021 |
like | +0.020 |
ows | +0.019 |
?" | +0.018 |
beyond | +0.017 |
Neg logit contributions | |
squirrel | -0.027 |
villains | -0.027 |
pillar | -0.027 |
bandits | -0.026 |
weaknesses | -0.026 |
Token | put |
Token ID | 1234 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
he | +0.017 |
like | +0.017 |
ows | +0.017 |
inkle | +0.015 |
?" | +0.015 |
Neg logit contributions | |
squirrel | -0.018 |
pillar | -0.018 |
villains | -0.017 |
appearances | -0.017 |
weaknesses | -0.017 |
Token | on |
Token ID | 319 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
he | +0.039 |
ows | +0.038 |
like | +0.036 |
inkle | +0.032 |
beyond | +0.031 |
Neg logit contributions | |
squirrel | -0.055 |
pillar | -0.054 |
villains | -0.053 |
bandits | -0.052 |
appearances | -0.051 |
Token | her |
Token ID | 607 |
Feature activation | +5.10e-02 |
Pos logit contributions | |
pillar | +0.050 |
villains | +0.050 |
squirrel | +0.048 |
weaknesses | +0.047 |
bandits | +0.047 |
Neg logit contributions | |
he | -0.023 |
like | -0.022 |
ows | -0.021 |
early | -0.021 |
inkle | -0.019 |
Token | rain |
Token ID | 6290 |
Feature activation | +8.45e-01 |
Pos logit contributions | |
he | +0.002 |
ows | +0.002 |
like | +0.002 |
inkle | +0.002 |
immer | +0.002 |
Neg logit contributions | |
squirrel | -0.002 |
pillar | -0.002 |
villains | -0.001 |
monster | -0.001 |
bandits | -0.001 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
he | +0.021 |
ows | +0.019 |
like | +0.018 |
inkle | +0.018 |
immer | +0.016 |
Neg logit contributions | |
squirrel | -0.039 |
pillar | -0.038 |
villains | -0.037 |
robbers | -0.036 |
bandits | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
he | +0.106 |
ows | +0.105 |
inkle | +0.104 |
rig | +0.102 |
immer | +0.096 |
Neg logit contributions | |
squirrel | -0.175 |
pillar | -0.168 |
rabbit | -0.160 |
turtle | -0.157 |
monster | -0.157 |
Token | ran |
Token ID | 4966 |
Feature activation | -8.90e-01 |
Pos logit contributions | |
pillar | +0.016 |
villains | +0.015 |
appearances | +0.015 |
squirrel | +0.015 |
weaknesses | +0.014 |
Neg logit contributions | |
ows | -0.018 |
like | -0.018 |
he | -0.018 |
early | -0.016 |
inkle | -0.016 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
pillar | +0.041 |
squirrel | +0.041 |
villains | +0.040 |
bandits | +0.039 |
appearances | +0.039 |
Neg logit contributions | |
he | -0.023 |
ows | -0.022 |
like | -0.021 |
inkle | -0.018 |
beyond | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | -8.56e-01 |
Pos logit contributions | |
squirrel | +0.038 |
monster | +0.035 |
rabbit | +0.035 |
pillar | +0.034 |
villains | +0.034 |
Neg logit contributions | |
ows | -0.038 |
he | -0.037 |
inkle | -0.035 |
immer | -0.034 |
rig | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
pillar | +0.045 |
villains | +0.044 |
appearances | +0.043 |
weaknesses | +0.043 |
rook | +0.042 |
Neg logit contributions | |
like | -0.019 |
he | -0.019 |
ows | -0.015 |
behind | -0.014 |
beyond | -0.013 |
Token | She |
Token ID | 1375 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
pillar | +0.008 |
villains | +0.008 |
squirrel | +0.007 |
bandits | +0.007 |
appearances | +0.007 |
Neg logit contributions | |
ows | -0.006 |
he | -0.006 |
like | -0.006 |
beyond | -0.005 |
early | -0.005 |
Token | put |
Token ID | 1234 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
appearances | +0.001 |
pillar | +0.001 |
villains | +0.001 |
rook | +0.001 |
squirrel | +0.001 |
Neg logit contributions | |
ows | -0.001 |
he | -0.001 |
like | -0.001 |
beyond | -0.001 |
?" | -0.001 |
Token | on |
Token ID | 319 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
ows | +0.045 |
rig | +0.045 |
he | +0.043 |
inkle | +0.041 |
adow | +0.039 |
Neg logit contributions | |
squirrel | -0.048 |
mice | -0.043 |
rabbit | -0.042 |
rabbits | -0.041 |
bear | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
pillar | +0.013 |
squirrel | +0.013 |
villains | +0.013 |
bandits | +0.013 |
appearances | +0.012 |
Neg logit contributions | |
he | -0.009 |
ows | -0.008 |
like | -0.008 |
inkle | -0.007 |
beyond | -0.007 |
Token | rain |
Token ID | 6290 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
squirrel | +0.003 |
pillar | +0.003 |
monster | +0.003 |
villains | +0.003 |
rabbit | +0.003 |
Neg logit contributions | |
he | -0.005 |
ows | -0.005 |
like | -0.004 |
immer | -0.004 |
beyond | -0.004 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
he | +0.022 |
inkle | +0.020 |
ows | +0.018 |
like | +0.017 |
immer | +0.017 |
Neg logit contributions | |
squirrel | -0.038 |
pillar | -0.036 |
villains | -0.035 |
robbers | -0.035 |
monster | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
inkle | +0.116 |
he | +0.110 |
ows | +0.110 |
rig | +0.107 |
immer | +0.103 |
Neg logit contributions | |
squirrel | -0.168 |
pillar | -0.158 |
rabbit | -0.153 |
monster | -0.151 |
turtle | -0.150 |
Token | her |
Token ID | 607 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
squirrel | +0.010 |
pillar | +0.010 |
villains | +0.010 |
bandits | +0.009 |
robbers | +0.009 |
Neg logit contributions | |
he | -0.015 |
ows | -0.014 |
like | -0.014 |
inkle | -0.013 |
beyond | -0.013 |
Token | boots |
Token ID | 14412 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
pillar | +0.016 |
villains | +0.015 |
squirrel | +0.014 |
appearances | +0.014 |
bandits | +0.014 |
Neg logit contributions | |
ows | -0.017 |
he | -0.017 |
like | -0.017 |
inkle | -0.015 |
early | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
he | +0.068 |
inkle | +0.065 |
ows | +0.065 |
rig | +0.062 |
like | +0.060 |
Neg logit contributions | |
squirrel | -0.094 |
pillar | -0.091 |
villains | -0.086 |
rabbit | -0.086 |
monster | -0.085 |
Token | went |
Token ID | 1816 |
Feature activation | -7.97e-01 |
Pos logit contributions | |
squirrel | +0.020 |
pillar | +0.019 |
villains | +0.018 |
bandits | +0.018 |
robbers | +0.018 |
Neg logit contributions | |
he | -0.030 |
ows | -0.030 |
like | -0.028 |
inkle | -0.027 |
beyond | -0.026 |
Token | She |
Token ID | 1375 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
squirrel | +0.000 |
pillar | +0.000 |
villains | +0.000 |
rabbit | +0.000 |
mice | +0.000 |
Neg logit contributions | |
he | -0.000 |
ows | -0.000 |
inkle | -0.000 |
like | -0.000 |
immer | -0.000 |
Token | put |
Token ID | 1234 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
he | +0.005 |
ows | +0.005 |
like | +0.005 |
inkle | +0.004 |
beyond | +0.004 |
Neg logit contributions | |
pillar | -0.004 |
squirrel | -0.004 |
villains | -0.004 |
bandits | -0.004 |
appearances | -0.004 |
Token | on |
Token ID | 319 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
rig | +0.057 |
ows | +0.053 |
adow | +0.048 |
he | +0.047 |
immer | +0.047 |
Neg logit contributions | |
squirrel | -0.046 |
bear | -0.042 |
monster | -0.041 |
mice | -0.041 |
rabbit | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
squirrel | +0.011 |
pillar | +0.010 |
villains | +0.010 |
bandits | +0.010 |
monster | +0.010 |
Neg logit contributions | |
he | -0.009 |
ows | -0.009 |
like | -0.008 |
inkle | -0.008 |
immer | -0.007 |
Token | rain |
Token ID | 6290 |
Feature activation | +8.01e-01 |
Pos logit contributions | |
squirrel | +0.005 |
monster | +0.004 |
fle | +0.004 |
bur | +0.004 |
rabbit | +0.004 |
Neg logit contributions | |
he | -0.007 |
ows | -0.007 |
immer | -0.006 |
like | -0.006 |
beyond | -0.006 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
he | +0.024 |
inkle | +0.023 |
ours | +0.021 |
immer | +0.021 |
ieved | +0.020 |
Neg logit contributions | |
squirrel | -0.032 |
wandered | -0.030 |
pillar | -0.029 |
rabbit | -0.029 |
monster | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
inkle | +0.136 |
rig | +0.132 |
immer | +0.125 |
ows | +0.125 |
he | +0.124 |
Neg logit contributions | |
squirrel | -0.156 |
bear | -0.140 |
rabbit | -0.139 |
pillar | -0.137 |
bur | -0.134 |
Token | went |
Token ID | 1816 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
squirrel | +0.007 |
pillar | +0.006 |
bandits | +0.006 |
villains | +0.006 |
rabbit | +0.006 |
Neg logit contributions | |
he | -0.013 |
ows | -0.013 |
like | -0.012 |
inkle | -0.012 |
rig | -0.012 |
Token | outside |
Token ID | 2354 |
Feature activation | -8.95e-01 |
Pos logit contributions | |
squirrel | +0.026 |
robbers | +0.024 |
monster | +0.024 |
villains | +0.024 |
bandits | +0.024 |
Neg logit contributions | |
he | -0.018 |
ows | -0.017 |
inkle | -0.016 |
immer | -0.015 |
rig | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
squirrel | +0.027 |
monster | +0.025 |
rabbit | +0.024 |
bur | +0.024 |
villains | +0.024 |
Neg logit contributions | |
ows | -0.037 |
he | -0.036 |
inkle | -0.035 |
immer | -0.035 |
rig | -0.033 |
Token | play |
Token ID | 711 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
squirrel | +0.012 |
monster | +0.011 |
rabbit | +0.011 |
bur | +0.010 |
pillar | +0.010 |
Neg logit contributions | |
ows | -0.021 |
he | -0.021 |
aron | -0.019 |
like | -0.019 |
immer | -0.019 |
Token | He |
Token ID | 679 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
he | +0.003 |
ows | +0.003 |
like | +0.003 |
inkle | +0.003 |
immer | +0.003 |
Neg logit contributions | |
squirrel | -0.003 |
pillar | -0.002 |
villains | -0.002 |
robbers | -0.002 |
bandits | -0.002 |
Token | put |
Token ID | 1234 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
pillar | +0.006 |
squirrel | +0.006 |
appearances | +0.006 |
villains | +0.006 |
rook | +0.005 |
Neg logit contributions | |
ows | -0.006 |
he | -0.006 |
like | -0.006 |
beyond | -0.005 |
inkle | -0.005 |
Token | on |
Token ID | 319 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
rig | +0.044 |
ows | +0.043 |
he | +0.042 |
immer | +0.039 |
inkle | +0.039 |
Neg logit contributions | |
squirrel | -0.047 |
monster | -0.044 |
bear | -0.042 |
rabbit | -0.042 |
mice | -0.041 |
Token | his |
Token ID | 465 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
pillar | +0.011 |
squirrel | +0.010 |
villains | +0.010 |
bandits | +0.010 |
weaknesses | +0.010 |
Neg logit contributions | |
he | -0.009 |
like | -0.009 |
ows | -0.009 |
early | -0.008 |
beyond | -0.008 |
Token | rain |
Token ID | 6290 |
Feature activation | +8.51e-01 |
Pos logit contributions | |
squirrel | +0.006 |
villains | +0.005 |
monster | +0.005 |
pillar | +0.005 |
bandits | +0.005 |
Neg logit contributions | |
he | -0.008 |
ows | -0.007 |
like | -0.007 |
immer | -0.007 |
beyond | -0.006 |
Token | coat |
Token ID | 31434 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
he | +0.028 |
inkle | +0.026 |
ows | +0.025 |
immer | +0.024 |
like | +0.024 |
Neg logit contributions | |
squirrel | -0.032 |
villains | -0.030 |
pillar | -0.030 |
monster | -0.029 |
robbers | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
inkle | +0.124 |
rig | +0.117 |
he | +0.117 |
immer | +0.112 |
ows | +0.111 |
Neg logit contributions | |
squirrel | -0.167 |
rabbit | -0.150 |
monster | -0.148 |
bear | -0.148 |
turtle | -0.147 |
Token | boots |
Token ID | 14412 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
squirrel | +0.010 |
pillar | +0.010 |
villains | +0.010 |
bandits | +0.009 |
appearances | +0.009 |
Neg logit contributions | |
he | -0.016 |
ows | -0.015 |
like | -0.015 |
inkle | -0.014 |
beyond | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
inkle | +0.069 |
rig | +0.064 |
he | +0.062 |
?", | +0.060 |
ows | +0.059 |
Neg logit contributions | |
squirrel | -0.084 |
pillar | -0.076 |
rabbit | -0.075 |
villains | -0.075 |
turtle | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
squirrel | +0.002 |
villains | +0.002 |
pillar | +0.002 |
robbers | +0.002 |
bandits | +0.002 |
Neg logit contributions | |
he | -0.003 |
ows | -0.003 |
inkle | -0.003 |
like | -0.003 |
immer | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
ows | +0.016 |
inkle | +0.016 |
ours | +0.016 |
immer | +0.016 |
ist | +0.015 |
Neg logit contributions | |
squirrel | -0.008 |
Le | -0.007 |
Carl | -0.007 |
pillar | -0.007 |
villains | -0.007 |
Token | No |
Token ID | 2949 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
he | +0.009 |
like | +0.009 |
ows | +0.009 |
beyond | +0.008 |
but | +0.008 |
Neg logit contributions | |
squirrel | -0.010 |
pillar | -0.009 |
villains | -0.009 |
rook | -0.009 |
bandits | -0.009 |
Token | way |
Token ID | 835 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
but | +0.010 |
like | +0.009 |
?" | +0.009 |
about | +0.008 |
ows | +0.007 |
Neg logit contributions | |
pillar | -0.037 |
Check | -0.035 |
Le | -0.035 |
rook | -0.034 |
growth | -0.034 |
Token | ! |
Token ID | 0 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
he | +0.014 |
ows | +0.014 |
like | +0.014 |
beyond | +0.011 |
?" | +0.011 |
Neg logit contributions | |
pillar | -0.028 |
villains | -0.027 |
squirrel | -0.027 |
bandits | -0.026 |
weaknesses | -0.026 |
Token | That |
Token ID | 1320 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
pillar | +0.002 |
squirrel | +0.002 |
bandits | +0.002 |
villains | +0.002 |
rook | +0.002 |
Neg logit contributions | |
ows | -0.002 |
like | -0.001 |
he | -0.001 |
?" | -0.001 |
beyond | -0.001 |
Token | all |
Token ID | 477 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
rook | +0.015 |
pawn | +0.015 |
villains | +0.015 |
pillar | +0.014 |
appearances | +0.014 |
Neg logit contributions | |
like | -0.017 |
he | -0.017 |
?" | -0.016 |
beyond | -0.016 |
?", | -0.016 |
Token | igator |
Token ID | 23823 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
he | +0.041 |
like | +0.039 |
ows | +0.038 |
beyond | +0.035 |
?" | +0.034 |
Neg logit contributions | |
pillar | -0.048 |
rook | -0.048 |
squirrel | -0.047 |
villains | -0.047 |
bandits | -0.047 |
Token | is |
Token ID | 318 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
he | +0.129 |
like | +0.128 |
ows | +0.128 |
?" | +0.112 |
inkle | +0.111 |
Neg logit contributions | |
villains | -0.155 |
squirrel | -0.154 |
pillar | -0.154 |
bandits | -0.151 |
rook | -0.149 |
Token | too |
Token ID | 1165 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
he | +0.002 |
ows | +0.002 |
like | +0.002 |
inkle | +0.002 |
immer | +0.002 |
Neg logit contributions | |
squirrel | -0.002 |
pillar | -0.002 |
villains | -0.002 |
bandits | -0.002 |
robbers | -0.002 |
Token | deep |
Token ID | 2769 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
he | +0.013 |
ows | +0.013 |
like | +0.012 |
inkle | +0.011 |
beyond | +0.011 |
Neg logit contributions | |
pillar | -0.018 |
squirrel | -0.017 |
villains | -0.017 |
rook | -0.017 |
bandits | -0.017 |
Token | in |
Token ID | 287 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
he | +0.013 |
ows | +0.012 |
inkle | +0.012 |
immer | +0.011 |
rig | +0.011 |
Neg logit contributions | |
squirrel | -0.022 |
pillar | -0.021 |
monster | -0.021 |
mice | -0.021 |
robbers | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
like | +0.005 |
he | +0.005 |
?" | +0.005 |
beyond | +0.004 |
?", | +0.004 |
Neg logit contributions | |
pillar | -0.012 |
villains | -0.011 |
squirrel | -0.011 |
weaknesses | -0.011 |
Boo | -0.011 |
Token | He |
Token ID | 679 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
he | +0.004 |
like | +0.004 |
ows | +0.004 |
?" | +0.004 |
beyond | +0.004 |
Neg logit contributions | |
pillar | -0.006 |
appearances | -0.006 |
villains | -0.006 |
squirrel | -0.006 |
weaknesses | -0.006 |
Token | would |
Token ID | 561 |
Feature activation | +9.78e-01 |
Pos logit contributions | |
pillar | +0.002 |
appearances | +0.002 |
squirrel | +0.002 |
bandits | +0.002 |
villains | +0.002 |
Neg logit contributions | |
like | -0.002 |
ows | -0.001 |
he | -0.001 |
?" | -0.001 |
beyond | -0.001 |
Token | race |
Token ID | 3234 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
like | +0.034 |
he | +0.033 |
ows | +0.031 |
beyond | +0.028 |
inkle | +0.028 |
Neg logit contributions | |
pillar | -0.038 |
squirrel | -0.035 |
villains | -0.035 |
appearances | -0.034 |
bandits | -0.034 |
Token | make |
Token ID | 787 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
he | +0.029 |
ows | +0.028 |
like | +0.026 |
inkle | +0.024 |
beyond | +0.023 |
Neg logit contributions | |
squirrel | -0.043 |
pillar | -0.042 |
villains | -0.041 |
bandits | -0.040 |
robbers | -0.040 |
Token | - |
Token ID | 12 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
like | +0.017 |
ows | +0.017 |
he | +0.016 |
about | +0.015 |
beyond | +0.014 |
Neg logit contributions | |
pillar | -0.022 |
squirrel | -0.020 |
rook | -0.020 |
appearances | -0.020 |
weaknesses | -0.020 |
Token | bel |
Token ID | 6667 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
he | +0.001 |
ows | +0.001 |
like | +0.001 |
bel | +0.001 |
but | +0.000 |
Neg logit contributions | |
appearances | -0.001 |
pillar | -0.001 |
squirrel | -0.001 |
rook | -0.001 |
robbers | -0.001 |
Token | ieve |
Token ID | 12311 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
ows | +0.139 |
he | +0.132 |
like | +0.128 |
rig | +0.123 |
adow | +0.122 |
Neg logit contributions | |
villains | -0.147 |
squirrel | -0.136 |
bandits | -0.134 |
pillar | -0.134 |
weaknesses | -0.133 |
Token | cars |
Token ID | 5006 |
Feature activation | +8.46e-01 |
Pos logit contributions | |
pillar | +0.032 |
squirrel | +0.031 |
appearances | +0.030 |
villains | +0.030 |
weaknesses | +0.030 |
Neg logit contributions | |
he | -0.016 |
like | -0.016 |
ows | -0.016 |
beyond | -0.013 |
early | -0.012 |
Token | around |
Token ID | 1088 |
Feature activation | +8.57e-01 |
Pos logit contributions | |
he | +0.020 |
ows | +0.019 |
like | +0.018 |
inkle | +0.016 |
beyond | +0.015 |
Neg logit contributions | |
squirrel | -0.040 |
pillar | -0.040 |
villains | -0.039 |
bandits | -0.038 |
appearances | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
he | +0.024 |
like | +0.023 |
ows | +0.022 |
beyond | +0.020 |
inkle | +0.019 |
Neg logit contributions | |
pillar | -0.038 |
squirrel | -0.037 |
villains | -0.036 |
weaknesses | -0.035 |
bandits | -0.035 |
Token | track |
Token ID | 2610 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
appearances | +0.002 |
pillar | +0.002 |
weaknesses | +0.002 |
villains | +0.002 |
Shark | +0.002 |
Neg logit contributions | |
like | -0.002 |
he | -0.002 |
ows | -0.002 |
beyond | -0.002 |
behind | -0.002 |
Token | box |
Token ID | 3091 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
villains | +0.000 |
pillar | +0.000 |
squirrel | +0.000 |
bandits | +0.000 |
weaknesses | +0.000 |
Neg logit contributions | |
he | -0.000 |
ows | -0.000 |
like | -0.000 |
beyond | -0.000 |
inkle | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
squirrel | +0.009 |
pillar | +0.009 |
rabbit | +0.009 |
monster | +0.008 |
rook | +0.008 |
Neg logit contributions | |
he | -0.005 |
ows | -0.005 |
like | -0.005 |
inkle | -0.004 |
rig | -0.004 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
he | +0.015 |
ows | +0.014 |
like | +0.014 |
inkle | +0.012 |
beyond | +0.012 |
Neg logit contributions | |
squirrel | -0.016 |
pillar | -0.016 |
villains | -0.016 |
bandits | -0.015 |
appearances | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -7.95e-02 |
Pos logit contributions | |
like | +0.007 |
he | +0.006 |
ows | +0.006 |
about | +0.006 |
beyond | +0.006 |
Neg logit contributions | |
pillar | -0.013 |
villains | -0.013 |
weaknesses | -0.012 |
robbers | -0.012 |
bandits | -0.012 |
Token | j |
Token ID | 474 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
weaknesses | +0.003 |
pillar | +0.003 |
villains | +0.003 |
bandits | +0.003 |
appearances | +0.003 |
Neg logit contributions | |
he | -0.003 |
like | -0.003 |
ows | -0.003 |
inkle | -0.002 |
but | -0.002 |
Token | igsaw |
Token ID | 45636 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
squirrel | +0.074 |
villains | +0.072 |
appearances | +0.071 |
bandits | +0.070 |
mice | +0.069 |
Neg logit contributions | |
he | -0.076 |
ows | -0.069 |
inkle | -0.068 |
like | -0.063 |
rig | -0.061 |
Token | puzzle |
Token ID | 15027 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
squirrel | +0.129 |
pillar | +0.128 |
villains | +0.124 |
bandits | +0.120 |
appearances | +0.117 |
Neg logit contributions | |
he | -0.191 |
ows | -0.186 |
like | -0.180 |
inkle | -0.168 |
beyond | -0.165 |
Token | inside |
Token ID | 2641 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
pillar | +0.077 |
squirrel | +0.076 |
villains | +0.076 |
weaknesses | +0.075 |
bandits | +0.075 |
Neg logit contributions | |
he | -0.042 |
like | -0.040 |
ows | -0.039 |
beyond | -0.033 |
behind | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
squirrel | +0.025 |
pillar | +0.025 |
monster | +0.024 |
villains | +0.024 |
bandits | +0.023 |
Neg logit contributions | |
ows | -0.020 |
he | -0.020 |
like | -0.018 |
inkle | -0.018 |
rig | -0.017 |
Token | He |
Token ID | 679 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
squirrel | +0.005 |
pillar | +0.005 |
villains | +0.005 |
bandits | +0.005 |
robbers | +0.005 |
Neg logit contributions | |
he | -0.005 |
ows | -0.005 |
like | -0.005 |
inkle | -0.004 |
beyond | -0.004 |
Token | paused |
Token ID | 24487 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
appearances | +0.001 |
villains | +0.000 |
pillar | +0.000 |
weaknesses | +0.000 |
bandits | +0.000 |
Neg logit contributions | |
like | -0.000 |
he | -0.000 |
ows | -0.000 |
almost | -0.000 |
about | -0.000 |
Token | would |
Token ID | 561 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
he | +0.014 |
like | +0.013 |
ows | +0.013 |
beyond | +0.011 |
?" | +0.011 |
Neg logit contributions | |
pillar | -0.017 |
squirrel | -0.016 |
appearances | -0.016 |
villains | -0.016 |
bandits | -0.016 |
Token | take |
Token ID | 1011 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
he | +0.031 |
like | +0.030 |
ows | +0.028 |
beyond | +0.024 |
early | +0.024 |
Neg logit contributions | |
pillar | -0.043 |
squirrel | -0.041 |
villains | -0.040 |
appearances | -0.040 |
bandits | -0.039 |
Token | out |
Token ID | 503 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
he | +0.009 |
like | +0.009 |
ows | +0.008 |
beyond | +0.007 |
early | +0.007 |
Neg logit contributions | |
pillar | -0.012 |
squirrel | -0.012 |
appearances | -0.011 |
villains | -0.011 |
weaknesses | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
like | +0.009 |
he | +0.009 |
ows | +0.008 |
early | +0.008 |
about | +0.008 |
Neg logit contributions | |
pillar | -0.013 |
villains | -0.012 |
squirrel | -0.012 |
Carl | -0.011 |
drowned | -0.011 |
Token | j |
Token ID | 474 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
he | +0.021 |
like | +0.020 |
ows | +0.020 |
inkle | +0.018 |
beyond | +0.018 |
Neg logit contributions | |
pillar | -0.016 |
villains | -0.016 |
squirrel | -0.016 |
appearances | -0.016 |
bandits | -0.015 |
Token | igsaw |
Token ID | 45636 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
squirrel | +0.043 |
villains | +0.041 |
pillar | +0.040 |
appearances | +0.039 |
bandits | +0.039 |
Neg logit contributions | |
he | -0.057 |
ows | -0.054 |
like | -0.051 |
inkle | -0.049 |
early | -0.048 |
Token | puzzle |
Token ID | 15027 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
squirrel | +0.095 |
pillar | +0.093 |
villains | +0.090 |
bandits | +0.085 |
appearances | +0.083 |
Neg logit contributions | |
he | -0.219 |
ows | -0.214 |
like | -0.207 |
inkle | -0.195 |
beyond | -0.193 |
Token | and |
Token ID | 290 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
squirrel | +0.049 |
pillar | +0.049 |
villains | +0.048 |
bandits | +0.047 |
robbers | +0.046 |
Neg logit contributions | |
he | -0.026 |
ows | -0.025 |
like | -0.024 |
inkle | -0.021 |
beyond | -0.020 |
Token | try |
Token ID | 1949 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
pillar | +0.001 |
squirrel | +0.001 |
villains | +0.001 |
bandits | +0.001 |
appearances | +0.001 |
Neg logit contributions | |
he | -0.001 |
ows | -0.001 |
like | -0.001 |
inkle | -0.001 |
beyond | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
squirrel | +0.023 |
pillar | +0.022 |
villains | +0.021 |
bandits | +0.021 |
monster | +0.021 |
Neg logit contributions | |
he | -0.024 |
ows | -0.022 |
like | -0.021 |
inkle | -0.020 |
immer | -0.020 |
Token | solve |
Token ID | 8494 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
squirrel | +0.011 |
villains | +0.011 |
pillar | +0.010 |
rook | +0.010 |
robbers | +0.010 |
Neg logit contributions | |
ows | -0.010 |
he | -0.010 |
like | -0.009 |
inkle | -0.009 |
rig | -0.008 |
Token | better |
Token ID | 1365 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
he | +0.005 |
ows | +0.005 |
like | +0.004 |
inkle | +0.004 |
beyond | +0.004 |
Neg logit contributions | |
squirrel | -0.005 |
pillar | -0.005 |
villains | -0.005 |
bandits | -0.005 |
appearances | -0.005 |
Token | than |
Token ID | 621 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
pillar | +0.073 |
squirrel | +0.072 |
villains | +0.070 |
bandits | +0.070 |
rook | +0.069 |
Neg logit contributions | |
he | -0.015 |
ows | -0.014 |
like | -0.013 |
inkle | -0.009 |
beyond | -0.009 |
Token | selfish |
Token ID | 20363 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
like | +0.059 |
beyond | +0.049 |
about | +0.049 |
ows | +0.048 |
?" | +0.047 |
Neg logit contributions | |
pillar | -0.072 |
Carl | -0.062 |
Boo | -0.061 |
squirrel | -0.060 |
Check | -0.060 |
Token | ness |
Token ID | 1108 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
Carl | +0.082 |
pillar | +0.080 |
Le | +0.079 |
Ro | +0.078 |
colony | +0.078 |
Neg logit contributions | |
like | -0.037 |
ows | -0.033 |
he | -0.028 |
about | -0.027 |
but | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +8.18e-02 |
Pos logit contributions | |
squirrel | +0.037 |
pillar | +0.037 |
bandits | +0.037 |
villains | +0.037 |
Check | +0.036 |
Neg logit contributions | |
like | -0.017 |
ows | -0.017 |
he | -0.016 |
about | -0.013 |
beyond | -0.013 |
Token | rud |
Token ID | 28906 |
Feature activation | -4.23e+00 |
Pos logit contributions | |
like | +0.003 |
ows | +0.002 |
beyond | +0.002 |
about | +0.002 |
he | +0.002 |
Neg logit contributions | |
pillar | -0.004 |
Carl | -0.003 |
squirrel | -0.003 |
Le | -0.003 |
bandits | -0.003 |
Token | eness |
Token ID | 9449 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
squirrel | +0.175 |
colony | +0.169 |
turtle | +0.166 |
lizard | +0.164 |
robbers | +0.164 |
Neg logit contributions | |
ows | -0.140 |
like | -0.128 |
he | -0.127 |
aron | -0.117 |
inkle | -0.108 |
Token | . |
Token ID | 13 |
Feature activation | +6.18e-02 |
Pos logit contributions | |
pillar | +0.011 |
squirrel | +0.011 |
bandits | +0.011 |
villains | +0.011 |
rook | +0.011 |
Neg logit contributions | |
ows | -0.006 |
like | -0.006 |
he | -0.006 |
about | -0.005 |
beyond | -0.005 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
he | +0.002 |
ows | +0.002 |
like | +0.002 |
inkle | +0.002 |
beyond | +0.002 |
Neg logit contributions | |
squirrel | -0.002 |
pillar | -0.002 |
villains | -0.002 |
appearances | -0.002 |
bandits | -0.002 |
Token | Tom |
Token ID | 13787 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
villains | +0.052 |
squirrel | +0.052 |
weaknesses | +0.051 |
pillar | +0.051 |
bandits | +0.051 |
Neg logit contributions | |
he | -0.046 |
ows | -0.041 |
like | -0.041 |
?" | -0.037 |
beyond | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
squirrel | +0.009 |
pillar | +0.009 |
villains | +0.008 |
bandits | +0.008 |
monster | +0.008 |
Neg logit contributions | |
he | -0.007 |
ows | -0.007 |
like | -0.007 |
inkle | -0.006 |
beyond | -0.006 |
Token | get |
Token ID | 651 |
Feature activation | +9.99e-01 |
Pos logit contributions | |
pillar | +0.010 |
squirrel | +0.010 |
villains | +0.009 |
bandits | +0.009 |
robbers | +0.009 |
Neg logit contributions | |
he | -0.007 |
ows | -0.007 |
like | -0.007 |
inkle | -0.006 |
beyond | -0.006 |
Token | whatever |
Token ID | 4232 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
like | +0.030 |
he | +0.027 |
ows | +0.025 |
beyond | +0.025 |
?" | +0.024 |
Neg logit contributions | |
pillar | -0.044 |
villains | -0.043 |
squirrel | -0.041 |
appearances | -0.040 |
bandits | -0.040 |
Token | he |
Token ID | 339 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
like | +0.004 |
about | +0.004 |
beyond | +0.004 |
?" | +0.004 |
early | +0.003 |
Neg logit contributions | |
pillar | -0.005 |
bandits | -0.005 |
rook | -0.005 |
appearances | -0.005 |
OTS | -0.005 |
Token | wanted |
Token ID | 2227 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
he | +0.037 |
ows | +0.036 |
like | +0.034 |
inkle | +0.032 |
beyond | +0.032 |
Neg logit contributions | |
squirrel | -0.026 |
pillar | -0.026 |
villains | -0.025 |
bandits | -0.024 |
robbers | -0.023 |
Token | with |
Token ID | 351 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
squirrel | +0.017 |
pillar | +0.017 |
villains | +0.017 |
bandits | +0.016 |
robbers | +0.016 |
Neg logit contributions | |
he | -0.011 |
ows | -0.011 |
like | -0.010 |
inkle | -0.009 |
beyond | -0.009 |
Token | rud |
Token ID | 28906 |
Feature activation | -4.06e+00 |
Pos logit contributions | |
pillar | +0.016 |
squirrel | +0.015 |
villains | +0.015 |
bandits | +0.015 |
Check | +0.015 |
Neg logit contributions | |
he | -0.013 |
like | -0.013 |
ows | -0.012 |
inkle | -0.011 |
beyond | -0.011 |
Token | eness |
Token ID | 9449 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
squirrel | +0.188 |
Ro | +0.183 |
villains | +0.179 |
Le | +0.179 |
lizard | +0.178 |
Neg logit contributions | |
ows | -0.112 |
aron | -0.103 |
he | -0.103 |
like | -0.102 |
inkle | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
he | +0.005 |
ows | +0.005 |
like | +0.005 |
beyond | +0.004 |
about | +0.004 |
Neg logit contributions | |
squirrel | -0.009 |
pillar | -0.009 |
villains | -0.009 |
bandits | -0.009 |
appearances | -0.009 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
squirrel | +0.005 |
villains | +0.005 |
bandits | +0.005 |
pillar | +0.005 |
appearances | +0.005 |
Neg logit contributions | |
he | -0.003 |
like | -0.003 |
ows | -0.003 |
about | -0.002 |
early | -0.002 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
squirrel | +0.036 |
pillar | +0.036 |
villains | +0.035 |
bandits | +0.034 |
robbers | +0.033 |
Neg logit contributions | |
he | -0.035 |
ows | -0.034 |
like | -0.033 |
inkle | -0.030 |
beyond | -0.029 |
Token | upon |
Token ID | 2402 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
ows | +0.005 |
he | +0.005 |
like | +0.005 |
aron | +0.004 |
inkle | +0.004 |
Neg logit contributions | |
squirrel | -0.010 |
pillar | -0.010 |
monster | -0.010 |
villains | -0.009 |
rabbit | -0.009 |
Token | It |
Token ID | 632 |
Feature activation | -9.24e-01 |
Pos logit contributions | |
villains | +0.005 |
squirrel | +0.005 |
appearances | +0.005 |
bandits | +0.005 |
pillar | +0.004 |
Neg logit contributions | |
he | -0.004 |
like | -0.004 |
ows | -0.004 |
about | -0.004 |
early | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | -7.26e-02 |
Pos logit contributions | |
villains | +0.035 |
pillar | +0.035 |
bandits | +0.034 |
squirrel | +0.034 |
appearances | +0.034 |
Neg logit contributions | |
he | -0.031 |
like | -0.030 |
ows | -0.030 |
beyond | -0.026 |
inkle | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
squirrel | +0.003 |
pillar | +0.003 |
villains | +0.003 |
bandits | +0.003 |
appearances | +0.003 |
Neg logit contributions | |
he | -0.002 |
ows | -0.002 |
like | -0.002 |
inkle | -0.002 |
beyond | -0.002 |
Token | big |
Token ID | 1263 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
like | +0.003 |
he | +0.003 |
inkle | +0.003 |
?" | +0.003 |
early | +0.003 |
Neg logit contributions | |
weaknesses | -0.004 |
villains | -0.004 |
pillar | -0.004 |
bandits | -0.004 |
appearances | -0.004 |
Token | j |
Token ID | 474 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
like | +0.016 |
he | +0.015 |
ows | +0.014 |
?" | +0.013 |
inkle | +0.013 |
Neg logit contributions | |
villains | -0.020 |
pillar | -0.019 |
weaknesses | -0.019 |
bandits | -0.019 |
Le | -0.019 |
Token | igsaw |
Token ID | 45636 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
squirrel | +0.054 |
villains | +0.052 |
appearances | +0.051 |
bandits | +0.051 |
weaknesses | +0.050 |
Neg logit contributions | |
he | -0.070 |
ows | -0.064 |
like | -0.062 |
inkle | -0.059 |
early | -0.058 |
Token | puzzle |
Token ID | 15027 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
squirrel | +0.111 |
pillar | +0.108 |
villains | +0.107 |
bandits | +0.104 |
appearances | +0.102 |
Neg logit contributions | |
he | -0.174 |
ows | -0.167 |
like | -0.163 |
inkle | -0.151 |
beyond | -0.149 |
Token | and |
Token ID | 290 |
Feature activation | -8.41e-02 |
Pos logit contributions | |
squirrel | +0.034 |
pillar | +0.034 |
villains | +0.033 |
bandits | +0.033 |
weaknesses | +0.032 |
Neg logit contributions | |
he | -0.019 |
ows | -0.018 |
like | -0.018 |
beyond | -0.015 |
inkle | -0.015 |
Token | she |
Token ID | 673 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
pillar | +0.003 |
squirrel | +0.003 |
villains | +0.003 |
bandits | +0.003 |
appearances | +0.003 |
Neg logit contributions | |
he | -0.003 |
like | -0.003 |
ows | -0.003 |
beyond | -0.002 |
inkle | -0.002 |
Token | wanted |
Token ID | 2227 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
he | +0.014 |
like | +0.014 |
ows | +0.014 |
inkle | +0.012 |
beyond | +0.012 |
Neg logit contributions | |
squirrel | -0.017 |
pillar | -0.017 |
appearances | -0.017 |
bandits | -0.016 |
robbers | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
pillar | +0.039 |
squirrel | +0.037 |
villains | +0.037 |
Carl | +0.037 |
bandits | +0.037 |
Neg logit contributions | |
like | -0.021 |
about | -0.018 |
he | -0.018 |
beyond | -0.017 |
ows | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
pillar | +0.087 |
Carl | +0.081 |
squirrel | +0.081 |
Le | +0.080 |
villains | +0.079 |
Neg logit contributions | |
like | -0.040 |
ows | -0.034 |
about | -0.032 |
he | -0.030 |
beyond | -0.029 |
Token | being |
Token ID | 852 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
like | +0.004 |
ows | +0.004 |
he | +0.004 |
beyond | +0.003 |
about | +0.003 |
Neg logit contributions | |
pillar | -0.005 |
squirrel | -0.005 |
bandits | -0.005 |
rook | -0.005 |
villains | -0.005 |
Token | kind |
Token ID | 1611 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
pillar | +0.041 |
squirrel | +0.039 |
villains | +0.038 |
bandits | +0.037 |
rook | +0.037 |
Neg logit contributions | |
like | -0.038 |
ows | -0.036 |
he | -0.035 |
beyond | -0.032 |
about | -0.032 |
Token | is |
Token ID | 318 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
pillar | +0.101 |
bandits | +0.101 |
Carl | +0.101 |
rook | +0.099 |
squirrel | +0.099 |
Neg logit contributions | |
like | -0.043 |
ows | -0.039 |
about | -0.035 |
he | -0.033 |
beyond | -0.030 |
Token | what |
Token ID | 644 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
squirrel | +0.013 |
pillar | +0.013 |
villains | +0.013 |
bandits | +0.012 |
robbers | +0.012 |
Neg logit contributions | |
he | -0.009 |
ows | -0.009 |
like | -0.008 |
inkle | -0.008 |
beyond | -0.007 |
Token | matters |
Token ID | 6067 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
pillar | +0.051 |
bur | +0.049 |
growth | +0.047 |
fury | +0.046 |
rook | +0.045 |
Neg logit contributions | |
like | -0.028 |
about | -0.026 |
?" | -0.025 |
beyond | -0.023 |
but | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
squirrel | +0.179 |
pillar | +0.179 |
villains | +0.172 |
bandits | +0.171 |
Check | +0.167 |
Neg logit contributions | |
ows | -0.103 |
he | -0.101 |
like | -0.098 |
inkle | -0.084 |
beyond | -0.084 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
he | +0.001 |
ows | +0.001 |
like | +0.001 |
inkle | +0.001 |
immer | +0.001 |
Neg logit contributions | |
squirrel | -0.001 |
pillar | -0.001 |
villains | -0.001 |
bandits | -0.001 |
appearances | -0.001 |
Token | One |
Token ID | 3198 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
squirrel | +0.046 |
pillar | +0.046 |
villains | +0.044 |
bandits | +0.043 |
monster | +0.043 |
Neg logit contributions | |
he | -0.040 |
ows | -0.040 |
like | -0.038 |
inkle | -0.035 |
immer | -0.034 |
Token | day |
Token ID | 1110 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
pillar | +0.015 |
squirrel | +0.015 |
villains | +0.015 |
bandits | +0.015 |
appearances | +0.015 |
Neg logit contributions | |
he | -0.015 |
like | -0.015 |
ows | -0.014 |
inkle | -0.013 |
early | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +6.45e-02 |
Pos logit contributions | |
pillar | +0.023 |
villains | +0.023 |
squirrel | +0.023 |
bandits | +0.022 |
weaknesses | +0.022 |
Neg logit contributions | |
he | -0.010 |
like | -0.010 |
ows | -0.010 |
?" | -0.009 |
inkle | -0.009 |
Token | angry |
Token ID | 7954 |
Feature activation | +9.67e-01 |
Pos logit contributions | |
pillar | +0.021 |
squirrel | +0.021 |
villains | +0.021 |
appearances | +0.020 |
bandits | +0.020 |
Neg logit contributions | |
he | -0.013 |
like | -0.013 |
ows | -0.012 |
beyond | -0.011 |
inkle | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
he | +0.023 |
ows | +0.022 |
like | +0.021 |
inkle | +0.018 |
beyond | +0.018 |
Neg logit contributions | |
squirrel | -0.046 |
pillar | -0.046 |
villains | -0.045 |
bandits | -0.044 |
appearances | -0.043 |
Token | decided |
Token ID | 3066 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
like | +0.000 |
he | +0.000 |
ows | +0.000 |
beyond | +0.000 |
?" | +0.000 |
Neg logit contributions | |
pillar | -0.000 |
villains | -0.000 |
weaknesses | -0.000 |
squirrel | -0.000 |
appearances | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | -9.92e-01 |
Pos logit contributions | |
squirrel | +0.015 |
pillar | +0.015 |
villains | +0.014 |
bandits | +0.014 |
robbers | +0.014 |
Neg logit contributions | |
he | -0.014 |
ows | -0.013 |
like | -0.013 |
inkle | -0.012 |
beyond | -0.011 |
Token | take |
Token ID | 1011 |
Feature activation | -8.50e-01 |
Pos logit contributions | |
pillar | +0.039 |
squirrel | +0.037 |
villains | +0.037 |
bandits | +0.036 |
weaknesses | +0.036 |
Neg logit contributions | |
he | -0.033 |
like | -0.032 |
ows | -0.031 |
?" | -0.027 |
inkle | -0.027 |
Token | matters |
Token ID | 6067 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
pillar | +0.040 |
villains | +0.038 |
weaknesses | +0.038 |
appearances | +0.038 |
squirrel | +0.038 |
Neg logit contributions | |
he | -0.022 |
like | -0.022 |
ows | -0.020 |
beyond | -0.018 |
behind | -0.017 |
Token | into |
Token ID | 656 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
squirrel | +0.172 |
pillar | +0.171 |
villains | +0.166 |
bandits | +0.162 |
robbers | +0.160 |
Neg logit contributions | |
he | -0.108 |
ows | -0.106 |
like | -0.100 |
inkle | -0.088 |
beyond | -0.087 |
Token | their |
Token ID | 511 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
like | +0.032 |
he | +0.032 |
ows | +0.030 |
early | +0.027 |
beyond | +0.026 |
Neg logit contributions | |
pillar | -0.048 |
squirrel | -0.045 |
villains | -0.044 |
bandits | -0.044 |
Check | -0.043 |
Token | own |
Token ID | 898 |
Feature activation | +5.71e-01 |
Pos logit contributions | |
pillar | +0.006 |
squirrel | +0.005 |
Carl | +0.005 |
Eagle | +0.005 |
Le | +0.005 |
Neg logit contributions | |
like | -0.004 |
early | -0.004 |
he | -0.004 |
inkle | -0.004 |
ows | -0.004 |
Token | hands |
Token ID | 2832 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
he | +0.021 |
ows | +0.021 |
like | +0.020 |
inkle | +0.018 |
early | +0.018 |
Neg logit contributions | |
pillar | -0.020 |
squirrel | -0.020 |
villains | -0.019 |
bandits | -0.019 |
Carl | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
pillar | +0.001 |
squirrel | +0.001 |
villains | +0.001 |
monster | +0.001 |
rabbit | +0.001 |
Neg logit contributions | |
ows | -0.001 |
he | -0.001 |
inkle | -0.001 |
rig | -0.001 |
like | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -7.61e-01 |
Pos logit contributions | |
pillar | +0.028 |
villains | +0.027 |
squirrel | +0.027 |
bandits | +0.027 |
rook | +0.027 |
Neg logit contributions | |
he | -0.020 |
like | -0.019 |
ows | -0.019 |
beyond | -0.017 |
about | -0.016 |
Token | scared |
Token ID | 12008 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
squirrel | +0.034 |
pillar | +0.034 |
villains | +0.033 |
bandits | +0.032 |
appearances | +0.032 |
Neg logit contributions | |
he | -0.020 |
ows | -0.020 |
like | -0.019 |
inkle | -0.017 |
beyond | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -5.36e-01 |
Pos logit contributions | |
rook | +0.053 |
villains | +0.053 |
pawn | +0.052 |
Check | +0.052 |
robbers | +0.051 |
Neg logit contributions | |
like | -0.028 |
he | -0.024 |
about | -0.023 |
ows | -0.022 |
beyond | -0.022 |
Token | To |
Token ID | 1675 |
Feature activation | +8.23e-01 |
Pos logit contributions | |
villains | +0.023 |
squirrel | +0.023 |
bandits | +0.023 |
robbers | +0.023 |
rook | +0.022 |
Neg logit contributions | |
ows | -0.015 |
he | -0.014 |
like | -0.014 |
about | -0.012 |
beyond | -0.012 |
Token | make |
Token ID | 787 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
he | +0.024 |
ows | +0.023 |
like | +0.023 |
inkle | +0.020 |
beyond | +0.019 |
Neg logit contributions | |
pillar | -0.035 |
squirrel | -0.035 |
villains | -0.034 |
bandits | -0.033 |
robbers | -0.033 |
Token | matters |
Token ID | 6067 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
ows | +0.003 |
he | +0.003 |
like | +0.003 |
inkle | +0.003 |
beyond | +0.003 |
Neg logit contributions | |
pillar | -0.007 |
squirrel | -0.007 |
villains | -0.007 |
rook | -0.007 |
robbers | -0.007 |
Token | worse |
Token ID | 4785 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
pillar | +0.178 |
squirrel | +0.177 |
villains | +0.173 |
Check | +0.169 |
bandits | +0.169 |
Neg logit contributions | |
ows | -0.105 |
he | -0.101 |
like | -0.098 |
beyond | -0.086 |
early | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
like | +0.017 |
ows | +0.016 |
he | +0.016 |
beyond | +0.014 |
early | +0.014 |
Neg logit contributions | |
pillar | -0.024 |
squirrel | -0.023 |
villains | -0.023 |
rook | -0.023 |
bandits | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +5.44e-01 |
Pos logit contributions | |
he | +0.021 |
ows | +0.020 |
like | +0.019 |
inkle | +0.018 |
beyond | +0.017 |
Neg logit contributions | |
squirrel | -0.024 |
pillar | -0.024 |
villains | -0.023 |
bandits | -0.023 |
appearances | -0.022 |
Token | sun |
Token ID | 4252 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
like | +0.023 |
behind | +0.022 |
he | +0.021 |
beyond | +0.020 |
inkle | +0.020 |
Neg logit contributions | |
villains | -0.014 |
rook | -0.014 |
Check | -0.014 |
weaknesses | -0.014 |
pill | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
he | +0.001 |
ows | +0.001 |
like | +0.001 |
inkle | +0.001 |
beyond | +0.001 |
Neg logit contributions | |
squirrel | -0.001 |
pillar | -0.001 |
villains | -0.001 |
bandits | -0.001 |
robbers | -0.001 |
Token | thin |
Token ID | 7888 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
he | +0.016 |
ows | +0.016 |
like | +0.015 |
inkle | +0.014 |
adow | +0.014 |
Neg logit contributions | |
pillar | -0.014 |
villains | -0.014 |
bandits | -0.013 |
squirrel | -0.013 |
robbers | -0.013 |
Token | stem |
Token ID | 10717 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
like | +0.011 |
ows | +0.010 |
he | +0.010 |
adow | +0.010 |
?" | +0.010 |
Neg logit contributions | |
villains | -0.009 |
appearances | -0.009 |
Check | -0.009 |
robbers | -0.009 |
bandits | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
squirrel | +0.005 |
pillar | +0.005 |
villains | +0.005 |
bandits | +0.005 |
robbers | +0.005 |
Neg logit contributions | |
he | -0.003 |
ows | -0.003 |
like | -0.003 |
inkle | -0.003 |
beyond | -0.002 |
Token | To |
Token ID | 1675 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
squirrel | +0.004 |
pillar | +0.004 |
villains | +0.004 |
bandits | +0.004 |
robbers | +0.004 |
Neg logit contributions | |
he | -0.005 |
ows | -0.005 |
like | -0.005 |
inkle | -0.005 |
beyond | -0.005 |
Token | make |
Token ID | 787 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ows | +0.023 |
he | +0.023 |
like | +0.021 |
beyond | +0.020 |
rig | +0.019 |
Neg logit contributions | |
squirrel | -0.029 |
pillar | -0.029 |
villains | -0.028 |
appearances | -0.027 |
bandits | -0.027 |
Token | matters |
Token ID | 6067 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
ows | +0.003 |
he | +0.003 |
like | +0.003 |
inkle | +0.002 |
beyond | +0.002 |
Neg logit contributions | |
pillar | -0.005 |
villains | -0.004 |
squirrel | -0.004 |
robbers | -0.004 |
rook | -0.004 |
Token | worse |
Token ID | 4785 |
Feature activation | +6.11e-01 |
Pos logit contributions | |
squirrel | +0.163 |
pillar | +0.162 |
villains | +0.158 |
bandits | +0.154 |
appearances | +0.153 |
Neg logit contributions | |
he | -0.108 |
ows | -0.104 |
like | -0.098 |
inkle | -0.088 |
beyond | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
he | +0.017 |
ows | +0.017 |
like | +0.017 |
inkle | +0.014 |
beyond | +0.014 |
Neg logit contributions | |
pillar | -0.026 |
squirrel | -0.026 |
villains | -0.025 |
bandits | -0.024 |
rook | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
he | +0.022 |
ows | +0.022 |
like | +0.020 |
immer | +0.019 |
inkle | +0.019 |
Neg logit contributions | |
squirrel | -0.020 |
pillar | -0.019 |
villains | -0.018 |
monster | -0.018 |
bandits | -0.018 |
Token | wind |
Token ID | 2344 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
like | +0.013 |
he | +0.012 |
ows | +0.011 |
early | +0.011 |
?" | +0.011 |
Neg logit contributions | |
villains | -0.010 |
pillar | -0.010 |
appearances | -0.010 |
robbers | -0.010 |
rook | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
he | +0.042 |
ows | +0.041 |
like | +0.040 |
inkle | +0.037 |
beyond | +0.036 |
Neg logit contributions | |
squirrel | -0.035 |
pillar | -0.035 |
villains | -0.034 |
bandits | -0.033 |
robbers | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -8.97e-01 |
Pos logit contributions | |
squirrel | +0.106 |
villains | +0.103 |
pillar | +0.102 |
monster | +0.100 |
weaknesses | +0.099 |
Neg logit contributions | |
he | -0.065 |
ows | -0.058 |
inkle | -0.054 |
immer | -0.052 |
like | -0.052 |
Token | that |
Token ID | 326 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
pillar | +0.035 |
squirrel | +0.034 |
villains | +0.033 |
bandits | +0.032 |
rook | +0.032 |
Neg logit contributions | |
he | -0.029 |
ows | -0.029 |
like | -0.028 |
beyond | -0.025 |
inkle | -0.025 |
Token | 's |
Token ID | 338 |
Feature activation | -7.92e-01 |
Pos logit contributions | |
pillar | +0.013 |
squirrel | +0.012 |
villains | +0.012 |
bandits | +0.012 |
rook | +0.012 |
Neg logit contributions | |
like | -0.012 |
he | -0.012 |
ows | -0.012 |
beyond | -0.011 |
?" | -0.011 |
Token | all |
Token ID | 477 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
squirrel | +0.031 |
appearances | +0.029 |
monster | +0.029 |
mice | +0.029 |
fle | +0.029 |
Neg logit contributions | |
he | -0.028 |
ows | -0.025 |
but | -0.022 |
immer | -0.021 |
inkle | -0.021 |
Token | that |
Token ID | 326 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
squirrel | +0.002 |
pillar | +0.002 |
villains | +0.002 |
monster | +0.002 |
bandits | +0.002 |
Neg logit contributions | |
he | -0.002 |
ows | -0.002 |
like | -0.002 |
inkle | -0.002 |
immer | -0.002 |
Token | matters |
Token ID | 6067 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
like | +0.002 |
he | +0.002 |
ows | +0.002 |
beyond | +0.002 |
?" | +0.002 |
Neg logit contributions | |
pillar | -0.002 |
squirrel | -0.002 |
bandits | -0.002 |
villains | -0.002 |
rook | -0.002 |
Token | ." |
Token ID | 526 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
squirrel | +0.169 |
pillar | +0.167 |
villains | +0.163 |
bandits | +0.160 |
robbers | +0.158 |
Neg logit contributions | |
he | -0.096 |
ows | -0.096 |
like | -0.090 |
inkle | -0.079 |
beyond | -0.077 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
he | +0.005 |
ows | +0.005 |
like | +0.005 |
inkle | +0.004 |
beyond | +0.004 |
Neg logit contributions | |
squirrel | -0.005 |
pillar | -0.005 |
villains | -0.005 |
rabbit | -0.005 |
mice | -0.005 |
Token | Once |
Token ID | 7454 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
squirrel | +0.069 |
pillar | +0.068 |
villains | +0.064 |
monster | +0.063 |
mice | +0.062 |
Neg logit contributions | |
ows | -0.066 |
he | -0.063 |
like | -0.062 |
inkle | -0.059 |
immer | -0.058 |
Token | upon |
Token ID | 2402 |
Feature activation | +7.13e-01 |
Pos logit contributions | |
squirrel | +0.018 |
pillar | +0.017 |
monster | +0.017 |
villains | +0.017 |
rabbit | +0.016 |
Neg logit contributions | |
ows | -0.009 |
he | -0.009 |
like | -0.008 |
aron | -0.007 |
beyond | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ows | +0.029 |
he | +0.027 |
inkle | +0.027 |
immer | +0.026 |
rig | +0.025 |
Neg logit contributions | |
squirrel | -0.024 |
villains | -0.022 |
monster | -0.021 |
pillar | -0.021 |
mice | -0.020 |
Token | about |
Token ID | 546 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
squirrel | +0.038 |
villains | +0.038 |
weaknesses | +0.037 |
pillar | +0.037 |
monster | +0.037 |
Neg logit contributions | |
he | -0.022 |
ows | -0.021 |
inkle | -0.020 |
immer | -0.019 |
adow | -0.018 |
Token | you |
Token ID | 345 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
like | +0.041 |
ows | +0.035 |
he | +0.035 |
?" | +0.034 |
beyond | +0.034 |
Neg logit contributions | |
pillar | -0.055 |
villains | -0.050 |
rook | -0.049 |
bandits | -0.049 |
Check | -0.049 |
Token | is |
Token ID | 318 |
Feature activation | -1.87e-03 |
Pos logit contributions | |
squirrel | +0.018 |
pillar | +0.017 |
villains | +0.017 |
bandits | +0.017 |
robbers | +0.016 |
Neg logit contributions | |
he | -0.013 |
ows | -0.012 |
like | -0.011 |
inkle | -0.011 |
immer | -0.010 |
Token | what |
Token ID | 644 |
Feature activation | -8.85e-01 |
Pos logit contributions | |
squirrel | +0.000 |
pillar | +0.000 |
villains | +0.000 |
bandits | +0.000 |
appearances | +0.000 |
Neg logit contributions | |
he | -0.000 |
ows | -0.000 |
like | -0.000 |
inkle | -0.000 |
beyond | -0.000 |
Token | really |
Token ID | 1107 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
growth | +0.042 |
pillar | +0.041 |
fury | +0.040 |
bur | +0.038 |
OTS | +0.038 |
Neg logit contributions | |
like | -0.024 |
?" | -0.022 |
about | -0.022 |
but | -0.019 |
beyond | -0.018 |
Token | matters |
Token ID | 6067 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
he | +0.017 |
ows | +0.017 |
like | +0.015 |
inkle | +0.015 |
immer | +0.014 |
Neg logit contributions | |
squirrel | -0.017 |
pillar | -0.016 |
villains | -0.016 |
robbers | -0.015 |
bandits | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
squirrel | +0.169 |
pillar | +0.167 |
villains | +0.163 |
bandits | +0.162 |
Check | +0.158 |
Neg logit contributions | |
he | -0.096 |
ows | -0.095 |
like | -0.094 |
inkle | -0.079 |
?" | -0.078 |
Token | From |
Token ID | 3574 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
he | +0.003 |
ows | +0.002 |
about | +0.002 |
like | +0.002 |
?" | +0.002 |
Neg logit contributions | |
bandits | -0.006 |
robbers | -0.006 |
weaknesses | -0.006 |
brains | -0.006 |
rook | -0.006 |
Token | that |
Token ID | 326 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
ows | +0.026 |
he | +0.026 |
like | +0.024 |
inkle | +0.023 |
immer | +0.023 |
Neg logit contributions | |
squirrel | -0.022 |
pillar | -0.021 |
villains | -0.020 |
bandits | -0.020 |
monster | -0.020 |
Token | day |
Token ID | 1110 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
he | +0.022 |
like | +0.021 |
?" | +0.018 |
inkle | +0.018 |
ows | +0.018 |
Neg logit contributions | |
villains | -0.027 |
bandits | -0.027 |
squirrel | -0.026 |
pillar | -0.026 |
appearances | -0.025 |
Token | on |
Token ID | 319 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
he | +0.007 |
ows | +0.007 |
like | +0.006 |
inkle | +0.005 |
beyond | +0.005 |
Neg logit contributions | |
squirrel | -0.016 |
pillar | -0.016 |
villains | -0.015 |
bandits | -0.015 |
robbers | -0.015 |
Token | young |
Token ID | 1862 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
appearances | +0.014 |
weaknesses | +0.014 |
Carl | +0.014 |
Le | +0.014 |
pillar | +0.014 |
Neg logit contributions | |
like | -0.013 |
he | -0.013 |
about | -0.012 |
early | -0.011 |
beyond | -0.011 |
Token | boy |
Token ID | 2933 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
like | +0.046 |
he | +0.040 |
but | +0.040 |
beyond | +0.038 |
behind | +0.037 |
Neg logit contributions | |
pillar | -0.074 |
weaknesses | -0.072 |
appearances | -0.071 |
growth | -0.071 |
emy | -0.068 |
Token | decided |
Token ID | 3066 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
like | +0.020 |
ows | +0.018 |
he | +0.018 |
beyond | +0.016 |
behind | +0.016 |
Neg logit contributions | |
appearances | -0.025 |
villains | -0.025 |
pillar | -0.024 |
weaknesses | -0.024 |
rook | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
like | +0.009 |
ows | +0.009 |
he | +0.008 |
beyond | +0.008 |
about | +0.008 |
Neg logit contributions | |
pillar | -0.015 |
villains | -0.014 |
squirrel | -0.014 |
Carl | -0.014 |
Check | -0.014 |
Token | take |
Token ID | 1011 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
pillar | +0.024 |
villains | +0.022 |
squirrel | +0.022 |
bandits | +0.022 |
Le | +0.022 |
Neg logit contributions | |
like | -0.017 |
he | -0.017 |
ows | -0.015 |
?" | -0.014 |
inkle | -0.014 |
Token | matters |
Token ID | 6067 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
pillar | +0.007 |
villains | +0.007 |
appearances | +0.006 |
squirrel | +0.006 |
weaknesses | +0.006 |
Neg logit contributions | |
like | -0.004 |
he | -0.004 |
ows | -0.003 |
beyond | -0.003 |
inkle | -0.003 |
Token | into |
Token ID | 656 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
pillar | +0.177 |
squirrel | +0.177 |
villains | +0.171 |
bandits | +0.167 |
Check | +0.165 |
Neg logit contributions | |
ows | -0.087 |
he | -0.086 |
like | -0.081 |
early | -0.070 |
beyond | -0.069 |
Token | his |
Token ID | 465 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
like | +0.041 |
he | +0.039 |
ows | +0.037 |
beyond | +0.035 |
early | +0.034 |
Neg logit contributions | |
pillar | -0.064 |
squirrel | -0.058 |
villains | -0.056 |
Boo | -0.056 |
Carl | -0.055 |
Token | own |
Token ID | 898 |
Feature activation | +7.64e-01 |
Pos logit contributions | |
like | +0.008 |
inkle | +0.008 |
he | +0.008 |
ows | +0.008 |
early | +0.007 |
Neg logit contributions | |
pillar | -0.012 |
Le | -0.011 |
squirrel | -0.011 |
villains | -0.010 |
Ro | -0.010 |
Token | hands |
Token ID | 2832 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
he | +0.026 |
like | +0.025 |
ows | +0.025 |
inkle | +0.023 |
early | +0.022 |
Neg logit contributions | |
pillar | -0.029 |
squirrel | -0.028 |
villains | -0.028 |
bandits | -0.027 |
Carl | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
he | +0.011 |
ows | +0.010 |
inkle | +0.009 |
like | +0.009 |
rig | +0.009 |
Neg logit contributions | |
squirrel | -0.016 |
pillar | -0.016 |
villains | -0.016 |
rook | -0.015 |
rabbit | -0.015 |
Token | not |
Token ID | 407 |
Feature activation | +4.62e-02 |
Pos logit contributions | |
squirrel | +0.020 |
villains | +0.019 |
pillar | +0.018 |
bandits | +0.018 |
robbers | +0.018 |
Neg logit contributions | |
he | -0.017 |
ows | -0.017 |
inkle | -0.014 |
like | -0.014 |
rig | -0.014 |
Token | as |
Token ID | 355 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
he | +0.001 |
ows | +0.001 |
like | +0.001 |
inkle | +0.001 |
immer | +0.001 |
Neg logit contributions | |
squirrel | -0.002 |
pillar | -0.002 |
villains | -0.002 |
bandits | -0.002 |
robbers | -0.002 |
Token | important |
Token ID | 1593 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
squirrel | +0.011 |
pillar | +0.011 |
villains | +0.011 |
bandits | +0.011 |
appearances | +0.010 |
Neg logit contributions | |
he | -0.009 |
ows | -0.009 |
like | -0.009 |
inkle | -0.008 |
beyond | -0.008 |
Token | as |
Token ID | 355 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
pillar | +0.166 |
squirrel | +0.164 |
villains | +0.159 |
bandits | +0.157 |
appearances | +0.155 |
Neg logit contributions | |
like | -0.046 |
he | -0.046 |
ows | -0.046 |
beyond | -0.034 |
early | -0.031 |
Token | having |
Token ID | 1719 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
pillar | +0.007 |
squirrel | +0.006 |
Carl | +0.006 |
Check | +0.006 |
villains | +0.006 |
Neg logit contributions | |
like | -0.005 |
ows | -0.005 |
beyond | -0.005 |
about | -0.004 |
early | -0.004 |
Token | kind |
Token ID | 1611 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
pillar | +0.068 |
Carl | +0.062 |
Le | +0.061 |
rook | +0.060 |
squirrel | +0.060 |
Neg logit contributions | |
like | -0.045 |
about | -0.036 |
ows | -0.036 |
beyond | -0.036 |
he | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
pillar | +0.179 |
Carl | +0.173 |
Le | +0.167 |
Ro | +0.165 |
villains | +0.164 |
Neg logit contributions | |
like | -0.099 |
ows | -0.083 |
beyond | -0.076 |
about | -0.074 |
he | -0.072 |
Token | helpful |
Token ID | 7613 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
pillar | +0.050 |
squirrel | +0.045 |
Le | +0.044 |
Carl | +0.044 |
villains | +0.044 |
Neg logit contributions | |
like | -0.029 |
ows | -0.026 |
he | -0.025 |
beyond | -0.024 |
early | -0.023 |
Token | friends |
Token ID | 2460 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
pillar | +0.138 |
Carl | +0.132 |
Le | +0.131 |
squirrel | +0.126 |
Ro | +0.126 |
Neg logit contributions | |
like | -0.075 |
ows | -0.062 |
beyond | -0.058 |
about | -0.056 |
he | -0.053 |
Token | like |
Token ID | 588 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
squirrel | +0.025 |
pillar | +0.024 |
villains | +0.024 |
bandits | +0.023 |
appearances | +0.023 |
Neg logit contributions | |
he | -0.016 |
ows | -0.015 |
like | -0.014 |
inkle | -0.013 |
beyond | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
like | +0.012 |
beyond | +0.010 |
about | +0.009 |
he | +0.009 |
?" | +0.009 |
Neg logit contributions | |
pillar | -0.016 |
Carl | -0.015 |
Boo | -0.014 |
weaknesses | -0.014 |
Check | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
squirrel | +0.033 |
pillar | +0.033 |
villains | +0.032 |
bandits | +0.031 |
monster | +0.031 |
Neg logit contributions | |
he | -0.017 |
ows | -0.016 |
like | -0.014 |
inkle | -0.014 |
immer | -0.013 |
Token | treat |
Token ID | 2190 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
pillar | +0.005 |
squirrel | +0.004 |
Carl | +0.004 |
rook | +0.004 |
bandits | +0.004 |
Neg logit contributions | |
like | -0.003 |
ows | -0.003 |
he | -0.003 |
beyond | -0.003 |
?" | -0.003 |
Token | others |
Token ID | 1854 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
squirrel | +0.005 |
pillar | +0.005 |
villains | +0.005 |
bandits | +0.005 |
appearances | +0.005 |
Neg logit contributions | |
he | -0.004 |
ows | -0.004 |
like | -0.004 |
inkle | -0.003 |
immer | -0.003 |
Token | that |
Token ID | 326 |
Feature activation | +4.98e-01 |
Pos logit contributions | |
squirrel | +0.048 |
pillar | +0.047 |
villains | +0.046 |
bandits | +0.045 |
monster | +0.045 |
Neg logit contributions | |
he | -0.025 |
ows | -0.023 |
like | -0.021 |
inkle | -0.020 |
immer | -0.019 |
Token | really |
Token ID | 1107 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
like | +0.016 |
he | +0.016 |
ows | +0.015 |
?" | +0.014 |
beyond | +0.014 |
Neg logit contributions | |
pillar | -0.019 |
squirrel | -0.019 |
villains | -0.019 |
bandits | -0.018 |
rook | -0.018 |
Token | matters |
Token ID | 6067 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
he | +0.018 |
ows | +0.018 |
like | +0.017 |
inkle | +0.015 |
beyond | +0.015 |
Neg logit contributions | |
pillar | -0.024 |
squirrel | -0.024 |
villains | -0.023 |
bandits | -0.023 |
appearances | -0.023 |
Token | ! |
Token ID | 0 |
Feature activation | +6.92e-02 |
Pos logit contributions | |
squirrel | +0.170 |
pillar | +0.169 |
villains | +0.165 |
bandits | +0.162 |
robbers | +0.160 |
Neg logit contributions | |
he | -0.089 |
ows | -0.089 |
like | -0.084 |
inkle | -0.073 |
beyond | -0.071 |
Token | Being |
Token ID | 11204 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
he | +0.002 |
ows | +0.002 |
like | +0.002 |
beyond | +0.002 |
?" | +0.002 |
Neg logit contributions | |
squirrel | -0.003 |
pillar | -0.002 |
villains | -0.002 |
bandits | -0.002 |
monster | -0.002 |
Token | kind |
Token ID | 1611 |
Feature activation | -9.93e-01 |
Pos logit contributions | |
pillar | +0.052 |
squirrel | +0.049 |
villains | +0.049 |
bandits | +0.048 |
appearances | +0.048 |
Neg logit contributions | |
like | -0.033 |
ows | -0.030 |
beyond | -0.028 |
he | -0.027 |
about | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
squirrel | +0.044 |
villains | +0.043 |
pillar | +0.043 |
bandits | +0.043 |
rook | +0.042 |
Neg logit contributions | |
like | -0.026 |
he | -0.026 |
ows | -0.026 |
?" | -0.021 |
beyond | -0.021 |
Token | caring |
Token ID | 18088 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
like | +0.010 |
ows | +0.010 |
he | +0.009 |
beyond | +0.009 |
about | +0.008 |
Neg logit contributions | |
pillar | -0.014 |
squirrel | -0.013 |
villains | -0.013 |
bandits | -0.013 |
rook | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
squirrel | +0.016 |
pillar | +0.016 |
villains | +0.016 |
robbers | +0.015 |
weaknesses | +0.015 |
Neg logit contributions | |
he | -0.010 |
ows | -0.009 |
inkle | -0.008 |
immer | -0.008 |
like | -0.008 |
Token | that |
Token ID | 326 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
pillar | +0.005 |
villains | +0.004 |
squirrel | +0.004 |
Carl | +0.004 |
Le | +0.004 |
Neg logit contributions | |
like | -0.004 |
ows | -0.003 |
he | -0.003 |
about | -0.003 |
?" | -0.003 |
Token | 's |
Token ID | 338 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
like | +0.004 |
?" | +0.004 |
he | +0.004 |
ows | +0.004 |
?", | +0.004 |
Neg logit contributions | |
villains | -0.005 |
pillar | -0.005 |
rook | -0.005 |
bandits | -0.005 |
Le | -0.005 |
Token | all |
Token ID | 477 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
squirrel | +0.016 |
pillar | +0.015 |
villains | +0.015 |
appearances | +0.015 |
monster | +0.014 |
Neg logit contributions | |
he | -0.011 |
ows | -0.011 |
like | -0.009 |
inkle | -0.009 |
immer | -0.008 |
Token | that |
Token ID | 326 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
he | +0.010 |
ows | +0.010 |
like | +0.010 |
inkle | +0.009 |
beyond | +0.009 |
Neg logit contributions | |
pillar | -0.014 |
squirrel | -0.014 |
villains | -0.013 |
bandits | -0.013 |
rook | -0.013 |
Token | matters |
Token ID | 6067 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
like | +0.012 |
he | +0.011 |
ows | +0.011 |
?" | +0.011 |
beyond | +0.010 |
Neg logit contributions | |
pillar | -0.015 |
villains | -0.015 |
rook | -0.015 |
bandits | -0.015 |
squirrel | -0.014 |
Token | ." |
Token ID | 526 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
squirrel | +0.163 |
pillar | +0.162 |
villains | +0.159 |
bandits | +0.155 |
Check | +0.153 |
Neg logit contributions | |
ows | -0.097 |
he | -0.095 |
like | -0.092 |
inkle | -0.081 |
beyond | -0.078 |
Token | Benny |
Token ID | 44275 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
he | +0.015 |
ows | +0.015 |
like | +0.015 |
inkle | +0.014 |
beyond | +0.013 |
Neg logit contributions | |
squirrel | -0.009 |
pillar | -0.009 |
villains | -0.009 |
bandits | -0.008 |
appearances | -0.008 |
Token | felt |
Token ID | 2936 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
villains | +0.010 |
appearances | +0.010 |
weaknesses | +0.010 |
bandits | +0.010 |
brains | +0.009 |
Neg logit contributions | |
he | -0.007 |
like | -0.006 |
ows | -0.006 |
early | -0.006 |
inkle | -0.006 |
Token | better |
Token ID | 1365 |
Feature activation | -9.69e-01 |
Pos logit contributions | |
ows | +0.016 |
he | +0.016 |
inkle | +0.014 |
immer | +0.013 |
rig | +0.013 |
Neg logit contributions | |
squirrel | -0.024 |
pillar | -0.023 |
villains | -0.023 |
monster | -0.022 |
turtle | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
squirrel | +0.048 |
pillar | +0.047 |
villains | +0.046 |
bandits | +0.044 |
monster | +0.044 |
Neg logit contributions | |
he | -0.022 |
ows | -0.022 |
like | -0.018 |
inkle | -0.018 |
beyond | -0.017 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
like | +0.003 |
he | +0.003 |
ows | +0.002 |
?" | +0.002 |
early | +0.002 |
Neg logit contributions | |
pillar | -0.005 |
weaknesses | -0.005 |
appearances | -0.005 |
squirrel | -0.005 |
rook | -0.005 |
Token | playing |
Token ID | 2712 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
pillar | +0.008 |
appearances | +0.007 |
weaknesses | +0.007 |
Carl | +0.007 |
villains | +0.007 |
Neg logit contributions | |
like | -0.004 |
early | -0.003 |
about | -0.003 |
beyond | -0.003 |
he | -0.003 |
Token | outside |
Token ID | 2354 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
ows | +0.008 |
immer | +0.007 |
inkle | +0.007 |
he | +0.007 |
rig | +0.006 |
Neg logit contributions | |
squirrel | -0.012 |
monster | -0.012 |
villains | -0.011 |
robbers | -0.011 |
mice | -0.011 |
Token | until |
Token ID | 1566 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
ows | +0.007 |
he | +0.007 |
like | +0.006 |
inkle | +0.005 |
immer | +0.005 |
Neg logit contributions | |
squirrel | -0.016 |
pillar | -0.015 |
villains | -0.015 |
monster | -0.015 |
bandits | -0.015 |
Token | no |
Token ID | 645 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
like | +0.003 |
he | +0.003 |
early | +0.003 |
?" | +0.003 |
beyond | +0.003 |
Neg logit contributions | |
pillar | -0.004 |
villains | -0.004 |
weaknesses | -0.004 |
appearances | -0.004 |
squirrel | -0.004 |
Token | ont |
Token ID | 756 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
pillar | +0.004 |
Carl | +0.004 |
weaknesses | +0.004 |
pill | +0.004 |
rook | +0.004 |
Neg logit contributions | |
like | -0.005 |
?" | -0.004 |
but | -0.004 |
early | -0.004 |
he | -0.004 |
Token | ime |
Token ID | 524 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
squirrel | +0.141 |
villains | +0.139 |
robbers | +0.137 |
pillar | +0.136 |
bandits | +0.135 |
Neg logit contributions | |
ows | -0.129 |
he | -0.113 |
like | -0.108 |
owed | -0.106 |
adow | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
pillar | +0.043 |
weaknesses | +0.043 |
appearances | +0.042 |
squirrel | +0.042 |
bandits | +0.042 |
Neg logit contributions | |
he | -0.049 |
like | -0.047 |
ows | -0.045 |
beyond | -0.040 |
?" | -0.039 |
Token | He |
Token ID | 679 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
pillar | +0.013 |
squirrel | +0.012 |
bandits | +0.012 |
villains | +0.012 |
appearances | +0.012 |
Neg logit contributions | |
like | -0.013 |
he | -0.013 |
ows | -0.013 |
about | -0.012 |
beyond | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
appearances | +0.013 |
pillar | +0.013 |
squirrel | +0.012 |
weaknesses | +0.012 |
rook | +0.012 |
Neg logit contributions | |
like | -0.012 |
he | -0.011 |
ows | -0.011 |
early | -0.010 |
?" | -0.010 |
Token | always |
Token ID | 1464 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
squirrel | +0.020 |
villains | +0.019 |
pillar | +0.018 |
bandits | +0.018 |
robbers | +0.018 |
Neg logit contributions | |
he | -0.015 |
ows | -0.015 |
like | -0.013 |
inkle | -0.013 |
immer | -0.013 |
Token | friendly |
Token ID | 8030 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
he | +0.008 |
ows | +0.008 |
like | +0.007 |
inkle | +0.006 |
beyond | +0.006 |
Neg logit contributions | |
squirrel | -0.016 |
pillar | -0.015 |
villains | -0.015 |
bandits | -0.015 |
robbers | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
weaknesses | +0.026 |
villains | +0.026 |
pillar | +0.026 |
Check | +0.026 |
appearances | +0.026 |
Neg logit contributions | |
like | -0.009 |
ows | -0.007 |
about | -0.006 |
he | -0.006 |
beyond | -0.006 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
pillar | +0.005 |
appearances | +0.005 |
villains | +0.005 |
rook | +0.005 |
weaknesses | +0.005 |
Neg logit contributions | |
like | -0.005 |
ows | -0.004 |
he | -0.004 |
beyond | -0.004 |
early | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
pillar | +0.058 |
villains | +0.055 |
weaknesses | +0.055 |
Check | +0.054 |
squirrel | +0.054 |
Neg logit contributions | |
like | -0.028 |
ows | -0.025 |
he | -0.024 |
about | -0.023 |
beyond | -0.023 |
Token | l |
Token ID | 300 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
growth | +0.026 |
pillar | +0.026 |
appearances | +0.026 |
arrest | +0.026 |
ez | +0.025 |
Neg logit contributions | |
like | -0.024 |
happy | -0.019 |
proud | -0.019 |
but | -0.018 |
he | -0.017 |
Token | itt |
Token ID | 715 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
robbers | +0.047 |
bandits | +0.046 |
squirrel | +0.046 |
appearances | +0.046 |
arrest | +0.045 |
Neg logit contributions | |
ows | -0.035 |
he | -0.033 |
like | -0.032 |
adow | -0.030 |
about | -0.028 |
Token | ler |
Token ID | 1754 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
Check | +0.151 |
villains | +0.146 |
bandits | +0.142 |
robbers | +0.142 |
pillar | +0.140 |
Neg logit contributions | |
like | -0.109 |
ows | -0.108 |
he | -0.104 |
inkle | -0.093 |
about | -0.092 |
Token | frog |
Token ID | 21264 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
Check | +0.037 |
pillar | +0.036 |
villains | +0.036 |
arrest | +0.035 |
drowned | +0.035 |
Neg logit contributions | |
like | -0.023 |
ows | -0.021 |
about | -0.019 |
?" | -0.017 |
adow | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
pillar | +0.084 |
villains | +0.083 |
Check | +0.082 |
bandits | +0.081 |
weaknesses | +0.080 |
Neg logit contributions | |
like | -0.037 |
ows | -0.035 |
he | -0.030 |
about | -0.030 |
beyond | -0.029 |
Token | join |
Token ID | 4654 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
pillar | +0.039 |
villains | +0.037 |
weaknesses | +0.036 |
bandits | +0.035 |
rook | +0.035 |
Neg logit contributions | |
like | -0.031 |
he | -0.030 |
ows | -0.027 |
beyond | -0.025 |
?" | -0.025 |
Token | them |
Token ID | 606 |
Feature activation | +7.92e-01 |
Pos logit contributions | |
pillar | +0.024 |
villains | +0.023 |
weaknesses | +0.023 |
squirrel | +0.023 |
bandits | +0.022 |
Neg logit contributions | |
like | -0.014 |
he | -0.014 |
ows | -0.013 |
beyond | -0.012 |
behind | -0.012 |
Token | getting |
Token ID | 1972 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
squirrel | +0.022 |
villains | +0.021 |
pillar | +0.021 |
bandits | +0.021 |
monster | +0.021 |
Neg logit contributions | |
he | -0.013 |
ows | -0.012 |
like | -0.011 |
inkle | -0.011 |
immer | -0.010 |
Token | very |
Token ID | 845 |
Feature activation | -8.28e-02 |
Pos logit contributions | |
pillar | +0.006 |
squirrel | +0.006 |
villains | +0.006 |
appearances | +0.006 |
bandits | +0.006 |
Neg logit contributions | |
he | -0.004 |
ows | -0.004 |
like | -0.004 |
beyond | -0.003 |
inkle | -0.003 |
Token | annoyed |
Token ID | 25602 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
pillar | +0.003 |
squirrel | +0.003 |
villains | +0.003 |
bandits | +0.003 |
appearances | +0.003 |
Neg logit contributions | |
he | -0.003 |
ows | -0.002 |
like | -0.002 |
inkle | -0.002 |
beyond | -0.002 |
Token | with |
Token ID | 351 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
he | +0.019 |
ows | +0.018 |
inkle | +0.016 |
like | +0.015 |
immer | +0.015 |
Neg logit contributions | |
squirrel | -0.034 |
pillar | -0.034 |
villains | -0.033 |
monster | -0.033 |
bandits | -0.032 |
Token | his |
Token ID | 465 |
Feature activation | -8.55e-01 |
Pos logit contributions | |
pillar | +0.029 |
villains | +0.026 |
Check | +0.026 |
squirrel | +0.026 |
rook | +0.026 |
Neg logit contributions | |
like | -0.017 |
he | -0.017 |
ows | -0.015 |
about | -0.014 |
beyond | -0.014 |
Token | rud |
Token ID | 28906 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
pillar | +0.028 |
Check | +0.025 |
villains | +0.025 |
weaknesses | +0.024 |
Carl | +0.024 |
Neg logit contributions | |
like | -0.034 |
ows | -0.033 |
he | -0.032 |
inkle | -0.031 |
early | -0.030 |
Token | eness |
Token ID | 9449 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
squirrel | +0.137 |
villains | +0.133 |
Carl | +0.132 |
robbers | +0.131 |
Ro | +0.131 |
Neg logit contributions | |
ows | -0.120 |
he | -0.114 |
aron | -0.108 |
like | -0.106 |
inkle | -0.103 |
Token | and |
Token ID | 290 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
ows | +0.010 |
he | +0.010 |
like | +0.010 |
about | +0.008 |
beyond | +0.008 |
Neg logit contributions | |
pillar | -0.021 |
squirrel | -0.021 |
villains | -0.021 |
bandits | -0.021 |
Check | -0.021 |
Token | complained |
Token ID | 13832 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
pillar | +0.013 |
rook | +0.011 |
Carl | +0.011 |
Check | +0.011 |
Le | +0.011 |
Neg logit contributions | |
like | -0.009 |
ows | -0.008 |
about | -0.008 |
but | -0.008 |
beyond | -0.007 |
Token | that |
Token ID | 326 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
pillar | +0.056 |
squirrel | +0.053 |
Check | +0.053 |
rook | +0.053 |
bandits | +0.053 |
Neg logit contributions | |
like | -0.026 |
ows | -0.025 |
he | -0.022 |
about | -0.022 |
beyond | -0.020 |
Token | he |
Token ID | 339 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
pillar | +0.016 |
Carl | +0.014 |
OTS | +0.014 |
emy | +0.014 |
Check | +0.014 |
Neg logit contributions | |
like | -0.009 |
about | -0.008 |
early | -0.008 |
beyond | -0.008 |
but | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
he | +0.022 |
ows | +0.022 |
like | +0.020 |
inkle | +0.017 |
beyond | +0.017 |
Neg logit contributions | |
squirrel | -0.051 |
pillar | -0.050 |
villains | -0.049 |
bandits | -0.048 |
appearances | -0.048 |
Token | that |
Token ID | 326 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
pillar | +0.018 |
villains | +0.017 |
squirrel | +0.017 |
bandits | +0.017 |
rook | +0.017 |
Neg logit contributions | |
like | -0.015 |
he | -0.014 |
ows | -0.014 |
?" | -0.013 |
beyond | -0.013 |
Token | 's |
Token ID | 338 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
pillar | +0.009 |
rook | +0.008 |
bandits | +0.008 |
growth | +0.008 |
pawn | +0.008 |
Neg logit contributions | |
like | -0.006 |
?" | -0.006 |
?", | -0.005 |
about | -0.005 |
beyond | -0.005 |
Token | all |
Token ID | 477 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
squirrel | +0.026 |
monster | +0.024 |
appearances | +0.024 |
villains | +0.023 |
mice | +0.023 |
Neg logit contributions | |
he | -0.021 |
ows | -0.020 |
inkle | -0.016 |
but | -0.016 |
immer | -0.016 |
Token | that |
Token ID | 326 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
he | +0.005 |
ows | +0.005 |
like | +0.005 |
inkle | +0.004 |
beyond | +0.004 |
Neg logit contributions | |
squirrel | -0.006 |
pillar | -0.006 |
villains | -0.006 |
bandits | -0.006 |
monster | -0.006 |
Token | matters |
Token ID | 6067 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
like | +0.003 |
?" | +0.003 |
beyond | +0.003 |
about | +0.003 |
he | +0.003 |
Neg logit contributions | |
pillar | -0.005 |
rook | -0.005 |
bandits | -0.005 |
pawn | -0.005 |
villains | -0.005 |
Token | ." |
Token ID | 526 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
squirrel | +0.161 |
pillar | +0.158 |
villains | +0.156 |
bandits | +0.153 |
robbers | +0.151 |
Neg logit contributions | |
ows | -0.092 |
he | -0.089 |
like | -0.086 |
inkle | -0.075 |
beyond | -0.073 |
Token | From |
Token ID | 3574 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
he | +0.007 |
ows | +0.007 |
like | +0.006 |
?" | +0.006 |
early | +0.006 |
Neg logit contributions | |
bandits | -0.011 |
weaknesses | -0.011 |
villains | -0.011 |
brains | -0.011 |
appearances | -0.011 |
Token | that |
Token ID | 326 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
ows | +0.023 |
he | +0.023 |
like | +0.021 |
inkle | +0.020 |
immer | +0.020 |
Neg logit contributions | |
squirrel | -0.019 |
pillar | -0.018 |
monster | -0.017 |
bandits | -0.017 |
villains | -0.017 |
Token | day |
Token ID | 1110 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
he | +0.020 |
like | +0.019 |
?" | +0.017 |
ows | +0.016 |
inkle | +0.016 |
Neg logit contributions | |
villains | -0.024 |
bandits | -0.023 |
squirrel | -0.023 |
pillar | -0.023 |
appearances | -0.023 |
Token | on |
Token ID | 319 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
he | +0.002 |
like | +0.002 |
ows | +0.002 |
inkle | +0.001 |
?" | +0.001 |
Neg logit contributions | |
villains | -0.008 |
pillar | -0.008 |
appearances | -0.008 |
weaknesses | -0.008 |
squirrel | -0.008 |
Token | all |
Token ID | 477 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
he | +0.007 |
ows | +0.007 |
like | +0.007 |
inkle | +0.006 |
beyond | +0.006 |
Neg logit contributions | |
squirrel | -0.008 |
pillar | -0.008 |
villains | -0.008 |
bandits | -0.008 |
appearances | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -8.88e-01 |
Pos logit contributions | |
squirrel | +0.032 |
pillar | +0.032 |
villains | +0.031 |
monster | +0.031 |
bandits | +0.030 |
Neg logit contributions | |
ows | -0.028 |
he | -0.028 |
like | -0.026 |
inkle | -0.024 |
immer | -0.024 |
Token | n |
Token ID | 299 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
squirrel | +0.031 |
pillar | +0.031 |
villains | +0.030 |
bandits | +0.029 |
appearances | +0.028 |
Neg logit contributions | |
he | -0.033 |
ows | -0.032 |
like | -0.031 |
inkle | -0.028 |
beyond | -0.028 |
Token | ooks |
Token ID | 31085 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
pillar | +0.083 |
villains | +0.081 |
squirrel | +0.080 |
weaknesses | +0.078 |
appearances | +0.078 |
Neg logit contributions | |
he | -0.062 |
like | -0.059 |
ows | -0.059 |
beyond | -0.052 |
inkle | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
squirrel | +0.085 |
pillar | +0.084 |
villains | +0.082 |
bandits | +0.080 |
robbers | +0.080 |
Neg logit contributions | |
he | -0.037 |
ows | -0.035 |
like | -0.032 |
inkle | -0.028 |
beyond | -0.027 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
pillar | +0.063 |
villains | +0.057 |
squirrel | +0.055 |
appearances | +0.054 |
colony | +0.053 |
Neg logit contributions | |
like | -0.058 |
ows | -0.052 |
he | -0.051 |
but | -0.049 |
behind | -0.049 |
Token | ann |
Token ID | 1236 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
squirrel | +0.129 |
villains | +0.123 |
turtle | +0.121 |
monster | +0.120 |
pillar | +0.117 |
Neg logit contributions | |
ows | -0.139 |
he | -0.131 |
like | -0.117 |
inkle | -0.115 |
immer | -0.110 |
Token | ies |
Token ID | 444 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
squirrel | +0.098 |
pillar | +0.097 |
villains | +0.097 |
Check | +0.095 |
bandits | +0.094 |
Neg logit contributions | |
ows | -0.057 |
he | -0.050 |
like | -0.048 |
inkle | -0.044 |
adow | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
villains | +0.028 |
pillar | +0.028 |
squirrel | +0.028 |
appearances | +0.027 |
Check | +0.027 |
Neg logit contributions | |
ows | -0.014 |
like | -0.013 |
he | -0.013 |
beyond | -0.011 |
behind | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
squirrel | +0.022 |
appearances | +0.021 |
pillar | +0.021 |
bandits | +0.021 |
villains | +0.021 |
Neg logit contributions | |
he | -0.010 |
ows | -0.010 |
like | -0.009 |
about | -0.008 |
early | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
he | +0.003 |
ows | +0.003 |
like | +0.003 |
inkle | +0.003 |
beyond | +0.003 |
Neg logit contributions | |
squirrel | -0.004 |
pillar | -0.004 |
villains | -0.004 |
bandits | -0.004 |
robbers | -0.004 |
Token | ball |
Token ID | 2613 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
he | +0.024 |
like | +0.023 |
ows | +0.023 |
?" | +0.021 |
beyond | +0.021 |
Neg logit contributions | |
villains | -0.016 |
pillar | -0.016 |
weaknesses | -0.015 |
bandits | -0.015 |
squirrel | -0.015 |
Token | ! |
Token ID | 0 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
he | +0.030 |
like | +0.030 |
ows | +0.027 |
?" | +0.025 |
beyond | +0.022 |
Neg logit contributions | |
weaknesses | -0.069 |
villains | -0.069 |
bandits | -0.067 |
rook | -0.067 |
robbers | -0.067 |
Token | Can |
Token ID | 1680 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
like | +0.007 |
he | +0.007 |
ows | +0.006 |
beyond | +0.006 |
?" | +0.006 |
Neg logit contributions | |
villains | -0.008 |
squirrel | -0.008 |
bandits | -0.008 |
pillar | -0.008 |
weaknesses | -0.008 |
Token | you |
Token ID | 345 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
like | +0.001 |
he | +0.001 |
?" | +0.001 |
ows | +0.001 |
beyond | +0.001 |
Neg logit contributions | |
pillar | -0.002 |
villains | -0.001 |
bandits | -0.001 |
squirrel | -0.001 |
weaknesses | -0.001 |
Token | teach |
Token ID | 4545 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
ows | +0.017 |
he | +0.016 |
like | +0.014 |
inkle | +0.014 |
beyond | +0.014 |
Neg logit contributions | |
squirrel | -0.012 |
pillar | -0.011 |
villains | -0.011 |
monster | -0.010 |
bandits | -0.010 |
Token | me |
Token ID | 502 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
pillar | +0.008 |
villains | +0.008 |
rook | +0.008 |
robbers | +0.008 |
bandits | +0.008 |
Neg logit contributions | |
like | -0.009 |
?" | -0.009 |
he | -0.008 |
about | -0.008 |
beyond | -0.008 |
Token | how |
Token ID | 703 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
he | +0.088 |
ows | +0.085 |
like | +0.076 |
inkle | +0.072 |
immer | +0.069 |
Neg logit contributions | |
squirrel | -0.111 |
pillar | -0.108 |
villains | -0.105 |
monster | -0.103 |
appearances | -0.103 |
Token | to |
Token ID | 284 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
pillar | +0.027 |
squirrel | +0.026 |
bandits | +0.026 |
villains | +0.026 |
rook | +0.026 |
Neg logit contributions | |
like | -0.016 |
?" | -0.014 |
he | -0.014 |
beyond | -0.013 |
about | -0.013 |
Token | fly |
Token ID | 6129 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
like | +0.004 |
he | +0.004 |
ows | +0.004 |
?" | +0.003 |
inkle | +0.003 |
Neg logit contributions | |
pillar | -0.004 |
villains | -0.004 |
squirrel | -0.004 |
bandits | -0.004 |
weaknesses | -0.004 |
Token | ?" |
Token ID | 1701 |
Feature activation | +5.71e-01 |
Pos logit contributions | |
squirrel | +0.016 |
pillar | +0.015 |
monster | +0.015 |
rabbit | +0.015 |
turtle | +0.015 |
Neg logit contributions | |
he | -0.013 |
ows | -0.012 |
immer | -0.010 |
aron | -0.010 |
adow | -0.010 |
Token | The |
Token ID | 383 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
he | +0.020 |
ows | +0.019 |
like | +0.018 |
inkle | +0.017 |
beyond | +0.017 |
Neg logit contributions | |
squirrel | -0.021 |
pillar | -0.021 |
villains | -0.020 |
bandits | -0.020 |
monster | -0.019 |
Token | on |
Token ID | 319 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
he | +0.033 |
ows | +0.033 |
like | +0.030 |
rig | +0.030 |
inkle | +0.030 |
Neg logit contributions | |
squirrel | -0.046 |
mice | -0.042 |
rabbit | -0.042 |
monster | -0.042 |
bandits | -0.041 |
Token | their |
Token ID | 511 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
pillar | +0.008 |
villains | +0.008 |
squirrel | +0.008 |
weaknesses | +0.008 |
bandits | +0.008 |
Neg logit contributions | |
he | -0.004 |
ows | -0.003 |
like | -0.003 |
early | -0.003 |
inkle | -0.003 |
Token | boots |
Token ID | 14412 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
pillar | +0.016 |
squirrel | +0.015 |
villains | +0.015 |
appearances | +0.014 |
rook | +0.014 |
Neg logit contributions | |
ows | -0.015 |
he | -0.014 |
like | -0.014 |
inkle | -0.013 |
early | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +8.91e-02 |
Pos logit contributions | |
inkle | +0.080 |
rig | +0.073 |
?", | +0.071 |
ows | +0.070 |
he | +0.069 |
Neg logit contributions | |
squirrel | -0.120 |
pillar | -0.113 |
rabbit | -0.109 |
robbers | -0.108 |
rook | -0.107 |
Token | their |
Token ID | 511 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
he | +0.003 |
ows | +0.003 |
like | +0.003 |
inkle | +0.003 |
beyond | +0.003 |
Neg logit contributions | |
pillar | -0.003 |
squirrel | -0.003 |
villains | -0.003 |
bandits | -0.003 |
appearances | -0.003 |
Token | coats |
Token ID | 30720 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
pillar | +0.032 |
appearances | +0.027 |
warrior | +0.027 |
Eagle | +0.027 |
Check | +0.027 |
Neg logit contributions | |
ows | -0.020 |
like | -0.020 |
behind | -0.017 |
he | -0.017 |
early | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
inkle | +0.114 |
mmm | +0.111 |
rig | +0.111 |
ieved | +0.104 |
?", | +0.104 |
Neg logit contributions | |
squirrel | -0.103 |
bear | -0.092 |
wandered | -0.091 |
rabbit | -0.090 |
bird | -0.090 |
Token | They |
Token ID | 1119 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
he | +0.007 |
ows | +0.006 |
like | +0.006 |
inkle | +0.006 |
beyond | +0.006 |
Neg logit contributions | |
squirrel | -0.007 |
pillar | -0.007 |
villains | -0.007 |
bandits | -0.006 |
appearances | -0.006 |
Token | take |
Token ID | 1011 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
he | +0.003 |
ows | +0.003 |
like | +0.003 |
inkle | +0.003 |
beyond | +0.003 |
Neg logit contributions | |
squirrel | -0.003 |
pillar | -0.003 |
villains | -0.003 |
bandits | -0.002 |
robbers | -0.002 |
Token | Spot |
Token ID | 15899 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
pillar | +0.014 |
appearances | +0.013 |
squirrel | +0.013 |
colony | +0.013 |
villains | +0.013 |
Neg logit contributions | |
like | -0.007 |
ows | -0.007 |
he | -0.006 |
beyond | -0.005 |
behind | -0.005 |
Token | with |
Token ID | 351 |
Feature activation | -8.41e-02 |
Pos logit contributions | |
ows | +0.004 |
he | +0.004 |
like | +0.004 |
beyond | +0.003 |
early | +0.003 |
Neg logit contributions | |
pillar | -0.007 |
villains | -0.007 |
squirrel | -0.007 |
weaknesses | -0.007 |
bandits | -0.007 |
Token | followed |
Token ID | 3940 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
he | +0.005 |
ows | +0.005 |
like | +0.005 |
beyond | +0.005 |
early | +0.005 |
Neg logit contributions | |
pillar | -0.006 |
squirrel | -0.006 |
villains | -0.006 |
appearances | -0.006 |
weaknesses | -0.006 |
Token | his |
Token ID | 465 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
squirrel | +0.016 |
bandits | +0.015 |
villains | +0.015 |
pillar | +0.015 |
robbers | +0.015 |
Neg logit contributions | |
ows | -0.019 |
he | -0.018 |
like | -0.017 |
inkle | -0.016 |
adow | -0.016 |
Token | uncle |
Token ID | 7711 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
pillar | +0.021 |
villains | +0.020 |
squirrel | +0.020 |
Carl | +0.019 |
rook | +0.019 |
Neg logit contributions | |
he | -0.016 |
ows | -0.016 |
like | -0.015 |
beyond | -0.014 |
early | -0.014 |
Token | around |
Token ID | 1088 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
pillar | +0.014 |
villains | +0.013 |
weaknesses | +0.013 |
appearances | +0.013 |
Check | +0.013 |
Neg logit contributions | |
ows | -0.009 |
like | -0.008 |
he | -0.008 |
behind | -0.007 |
beyond | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
he | +0.010 |
like | +0.010 |
ows | +0.009 |
beyond | +0.008 |
early | +0.008 |
Neg logit contributions | |
pillar | -0.017 |
squirrel | -0.016 |
villains | -0.016 |
weaknesses | -0.015 |
bandits | -0.015 |
Token | Everywhere |
Token ID | 48201 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
squirrel | +0.010 |
pillar | +0.010 |
villains | +0.010 |
bandits | +0.009 |
appearances | +0.009 |
Neg logit contributions | |
he | -0.012 |
ows | -0.012 |
like | -0.011 |
inkle | -0.010 |
beyond | -0.010 |
Token | they |
Token ID | 484 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
ows | +0.138 |
he | +0.136 |
like | +0.127 |
inkle | +0.122 |
immer | +0.122 |
Neg logit contributions | |
squirrel | -0.082 |
pillar | -0.079 |
villains | -0.075 |
bandits | -0.073 |
monster | -0.073 |
Token | went |
Token ID | 1816 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
he | +0.051 |
ows | +0.049 |
like | +0.048 |
inkle | +0.045 |
beyond | +0.044 |
Neg logit contributions | |
squirrel | -0.027 |
pillar | -0.027 |
villains | -0.026 |
bandits | -0.025 |
appearances | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +8.97e-01 |
Pos logit contributions | |
he | +0.011 |
ows | +0.011 |
like | +0.011 |
beyond | +0.009 |
inkle | +0.009 |
Neg logit contributions | |
pillar | -0.019 |
squirrel | -0.019 |
villains | -0.019 |
weaknesses | -0.018 |
appearances | -0.018 |
Token | his |
Token ID | 465 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
he | +0.036 |
ows | +0.036 |
like | +0.034 |
inkle | +0.032 |
immer | +0.031 |
Neg logit contributions | |
squirrel | -0.028 |
pillar | -0.027 |
villains | -0.027 |
bandits | -0.026 |
robbers | -0.025 |
Token | uncle |
Token ID | 7711 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
he | +0.004 |
ows | +0.004 |
like | +0.004 |
inkle | +0.004 |
beyond | +0.003 |
Neg logit contributions | |
squirrel | -0.004 |
pillar | -0.004 |
villains | -0.004 |
bandits | -0.004 |
robbers | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
he | +0.010 |
ows | +0.010 |
like | +0.009 |
inkle | +0.008 |
immer | +0.008 |
Neg logit contributions | |
squirrel | -0.016 |
pillar | -0.015 |
villains | -0.015 |
bandits | -0.015 |
robbers | -0.015 |
Token | began |
Token ID | 2540 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
pillar | +0.007 |
squirrel | +0.007 |
appearances | +0.007 |
villains | +0.007 |
weaknesses | +0.007 |
Neg logit contributions | |
he | -0.007 |
like | -0.007 |
ows | -0.007 |
beyond | -0.006 |
early | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
pillar | +0.062 |
villains | +0.057 |
squirrel | +0.056 |
weaknesses | +0.055 |
appearances | +0.055 |
Neg logit contributions | |
like | -0.038 |
ows | -0.036 |
he | -0.036 |
early | -0.035 |
beyond | -0.033 |
Token | explore |
Token ID | 7301 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
squirrel | +0.051 |
pillar | +0.051 |
villains | +0.049 |
bandits | +0.048 |
appearances | +0.047 |
Neg logit contributions | |
he | -0.050 |
ows | -0.048 |
like | -0.047 |
inkle | -0.043 |
beyond | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
he | +0.013 |
ows | +0.013 |
inkle | +0.012 |
immer | +0.012 |
like | +0.011 |
Neg logit contributions | |
squirrel | -0.013 |
villains | -0.012 |
monster | -0.012 |
bandits | -0.012 |
rabbit | -0.012 |
Token | Everywhere |
Token ID | 48201 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
squirrel | +0.009 |
pillar | +0.009 |
villains | +0.009 |
appearances | +0.008 |
monster | +0.008 |
Neg logit contributions | |
ows | -0.006 |
like | -0.006 |
he | -0.006 |
about | -0.005 |
beyond | -0.005 |
Token | they |
Token ID | 484 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
he | +0.074 |
like | +0.074 |
ows | +0.068 |
beyond | +0.063 |
early | +0.061 |
Neg logit contributions | |
pillar | -0.121 |
squirrel | -0.118 |
villains | -0.118 |
appearances | -0.117 |
weaknesses | -0.116 |
Token | went |
Token ID | 1816 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
he | +0.044 |
ows | +0.042 |
like | +0.042 |
beyond | +0.038 |
inkle | +0.038 |
Neg logit contributions | |
pillar | -0.028 |
squirrel | -0.028 |
villains | -0.027 |
appearances | -0.026 |
weaknesses | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
he | +0.013 |
ows | +0.013 |
like | +0.012 |
inkle | +0.011 |
beyond | +0.011 |
Neg logit contributions | |
squirrel | -0.021 |
pillar | -0.021 |
villains | -0.021 |
bandits | -0.020 |
appearances | -0.020 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
he | +0.026 |
ows | +0.025 |
like | +0.025 |
beyond | +0.022 |
early | +0.022 |
Neg logit contributions | |
pillar | -0.033 |
squirrel | -0.032 |
villains | -0.032 |
appearances | -0.031 |
bandits | -0.031 |
Token | bounced |
Token ID | 31318 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
ows | +0.006 |
like | +0.006 |
he | +0.005 |
beyond | +0.005 |
adow | +0.005 |
Neg logit contributions | |
appearances | -0.006 |
villains | -0.006 |
pillar | -0.006 |
squirrel | -0.006 |
weaknesses | -0.006 |
Token | there |
Token ID | 612 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
ows | +0.012 |
he | +0.011 |
rig | +0.011 |
inkle | +0.011 |
aron | +0.010 |
Neg logit contributions | |
squirrel | -0.015 |
monster | -0.014 |
pillar | -0.014 |
rabbit | -0.013 |
Carl | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
squirrel | +0.002 |
pillar | +0.002 |
monster | +0.002 |
villains | +0.002 |
rabbit | +0.002 |
Neg logit contributions | |
ows | -0.003 |
he | -0.003 |
like | -0.002 |
immer | -0.002 |
inkle | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
ows | +0.013 |
he | +0.012 |
inkle | +0.011 |
immer | +0.011 |
rig | +0.010 |
Neg logit contributions | |
squirrel | -0.012 |
pillar | -0.011 |
villains | -0.011 |
mice | -0.011 |
monster | -0.011 |
Token | big |
Token ID | 1263 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
he | +0.009 |
like | +0.009 |
ows | +0.008 |
inkle | +0.008 |
early | +0.008 |
Neg logit contributions | |
weaknesses | -0.006 |
pillar | -0.006 |
villains | -0.006 |
appearances | -0.006 |
brains | -0.006 |
Token | all |
Token ID | 477 |
Feature activation | +9.06e-02 |
Pos logit contributions | |
he | +0.013 |
like | +0.013 |
?" | +0.012 |
inkle | +0.012 |
ows | +0.012 |
Neg logit contributions | |
weaknesses | -0.008 |
villains | -0.008 |
pillar | -0.008 |
appearances | -0.008 |
Carl | -0.008 |
Token | igator |
Token ID | 23823 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
ows | +0.004 |
he | +0.003 |
inkle | +0.003 |
immer | +0.003 |
like | +0.003 |
Neg logit contributions | |
squirrel | -0.003 |
pillar | -0.003 |
monster | -0.003 |
villains | -0.003 |
turtle | -0.003 |
Token | who |
Token ID | 508 |
Feature activation | -9.48e-01 |
Pos logit contributions | |
he | +0.084 |
ows | +0.080 |
like | +0.079 |
inkle | +0.068 |
beyond | +0.068 |
Neg logit contributions | |
pillar | -0.136 |
squirrel | -0.135 |
villains | -0.133 |
bandits | -0.131 |
weaknesses | -0.130 |
Token | lived |
Token ID | 5615 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
pillar | +0.044 |
appearances | +0.044 |
rook | +0.043 |
weaknesses | +0.043 |
squirrel | +0.042 |
Neg logit contributions | |
like | -0.024 |
he | -0.024 |
ows | -0.023 |
beyond | -0.019 |
adow | -0.018 |
Token | in |
Token ID | 287 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
pillar | +0.024 |
squirrel | +0.023 |
villains | +0.023 |
weaknesses | +0.023 |
appearances | +0.022 |
Neg logit contributions | |
he | -0.008 |
ows | -0.007 |
like | -0.007 |
beyond | -0.005 |
early | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
pillar | +0.016 |
weaknesses | +0.015 |
villains | +0.014 |
rook | +0.014 |
bandits | +0.014 |
Neg logit contributions | |
like | -0.005 |
he | -0.005 |
beyond | -0.003 |
?" | -0.003 |
behind | -0.003 |
Token | river |
Token ID | 7850 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
he | +0.012 |
?" | +0.012 |
like | +0.011 |
early | +0.011 |
proud | +0.010 |
Neg logit contributions | |
appearances | -0.024 |
weaknesses | -0.024 |
endings | -0.023 |
brains | -0.023 |
reflex | -0.022 |
Token | with |
Token ID | 351 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
he | +0.025 |
ows | +0.024 |
inkle | +0.023 |
rig | +0.023 |
?", | +0.021 |
Neg logit contributions | |
squirrel | -0.043 |
pillar | -0.041 |
villains | -0.041 |
rook | -0.040 |
rabbit | -0.040 |
Token | it |
Token ID | 340 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
pillar | +0.016 |
villains | +0.016 |
weaknesses | +0.016 |
colony | +0.015 |
appearances | +0.015 |
Neg logit contributions | |
he | -0.010 |
like | -0.010 |
ows | -0.009 |
behind | -0.008 |
early | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +9.89e-02 |
Pos logit contributions | |
pillar | +0.022 |
squirrel | +0.022 |
villains | +0.021 |
bandits | +0.021 |
robbers | +0.021 |
Neg logit contributions | |
he | -0.013 |
ows | -0.013 |
like | -0.012 |
inkle | -0.011 |
beyond | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
he | +0.004 |
ows | +0.004 |
inkle | +0.004 |
like | +0.004 |
immer | +0.003 |
Neg logit contributions | |
squirrel | -0.003 |
pillar | -0.003 |
robbers | -0.003 |
villains | -0.003 |
bandits | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
ows | +0.021 |
inkle | +0.020 |
immer | +0.020 |
like | +0.019 |
adow | +0.019 |
Neg logit contributions | |
Le | -0.010 |
squirrel | -0.010 |
pillar | -0.009 |
Carl | -0.009 |
B | -0.008 |
Token | " |
Token ID | 1 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
ows | +0.009 |
inkle | +0.009 |
he | +0.009 |
like | +0.009 |
beyond | +0.008 |
Neg logit contributions | |
squirrel | -0.005 |
pillar | -0.005 |
robbers | -0.005 |
villains | -0.005 |
Check | -0.005 |
Token | No |
Token ID | 2949 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
he | +0.028 |
ows | +0.027 |
like | +0.026 |
inkle | +0.024 |
beyond | +0.024 |
Neg logit contributions | |
squirrel | -0.023 |
pillar | -0.023 |
villains | -0.022 |
bandits | -0.022 |
robbers | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -8.28e-02 |
Pos logit contributions | |
he | +0.015 |
ows | +0.015 |
like | +0.014 |
inkle | +0.012 |
?" | +0.012 |
Neg logit contributions | |
pillar | -0.027 |
squirrel | -0.027 |
villains | -0.027 |
bandits | -0.026 |
rook | -0.026 |
Token | it |
Token ID | 340 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
squirrel | +0.003 |
pillar | +0.003 |
villains | +0.003 |
robbers | +0.002 |
monster | +0.002 |
Neg logit contributions | |
he | -0.003 |
ows | -0.003 |
like | -0.003 |
inkle | -0.003 |
beyond | -0.003 |
Token | is |
Token ID | 318 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
ows | +0.022 |
he | +0.021 |
like | +0.021 |
inkle | +0.019 |
?" | +0.019 |
Neg logit contributions | |
squirrel | -0.018 |
pillar | -0.018 |
villains | -0.018 |
bandits | -0.018 |
robbers | -0.017 |
Token | mine |
Token ID | 6164 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
squirrel | +0.007 |
robbers | +0.007 |
villains | +0.006 |
monster | +0.006 |
fle | +0.006 |
Neg logit contributions | |
he | -0.008 |
ows | -0.007 |
but | -0.006 |
inkle | -0.006 |
rig | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +9.75e-02 |
Pos logit contributions | |
villains | +0.039 |
pillar | +0.039 |
squirrel | +0.039 |
bandits | +0.038 |
weaknesses | +0.037 |
Neg logit contributions | |
he | -0.019 |
ows | -0.019 |
like | -0.018 |
inkle | -0.016 |
?" | -0.015 |
Token | look |
Token ID | 804 |
Feature activation | -7.51e-02 |
Pos logit contributions | |
like | +0.003 |
?" | +0.002 |
he | +0.002 |
ours | +0.002 |
but | +0.002 |
Neg logit contributions | |
colony | -0.005 |
pillar | -0.005 |
weaknesses | -0.005 |
villains | -0.005 |
OTS | -0.004 |
Token | at |
Token ID | 379 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
squirrel | +0.003 |
monster | +0.003 |
pillar | +0.002 |
bug | +0.002 |
lizard | +0.002 |
Neg logit contributions | |
he | -0.003 |
ows | -0.003 |
but | -0.002 |
immer | -0.002 |
rig | -0.002 |
Token | this |
Token ID | 428 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
like | +0.010 |
?" | +0.010 |
he | +0.010 |
beyond | +0.009 |
behind | +0.009 |
Neg logit contributions | |
villains | -0.018 |
pillar | -0.018 |
weaknesses | -0.017 |
rook | -0.017 |
Check | -0.017 |
Token | mushroom |
Token ID | 28520 |
Feature activation | -8.15e-02 |
Pos logit contributions | |
he | +0.026 |
ows | +0.025 |
like | +0.025 |
inkle | +0.023 |
beyond | +0.023 |
Neg logit contributions | |
squirrel | -0.012 |
pillar | -0.012 |
villains | -0.012 |
bandits | -0.011 |
appearances | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
villains | +0.004 |
squirrel | +0.004 |
bandits | +0.004 |
pillar | +0.004 |
weaknesses | +0.004 |
Neg logit contributions | |
he | -0.002 |
ows | -0.002 |
like | -0.002 |
?" | -0.001 |
beyond | -0.001 |
Token | It |
Token ID | 632 |
Feature activation | -8.36e-01 |
Pos logit contributions | |
like | +0.012 |
ows | +0.012 |
he | +0.011 |
beyond | +0.010 |
?" | +0.010 |
Neg logit contributions | |
pillar | -0.014 |
squirrel | -0.014 |
villains | -0.014 |
bandits | -0.013 |
weaknesses | -0.013 |
Token | 's |
Token ID | 338 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
villains | +0.032 |
pillar | +0.031 |
bandits | +0.031 |
squirrel | +0.031 |
robbers | +0.031 |
Neg logit contributions | |
like | -0.028 |
ows | -0.028 |
he | -0.026 |
inkle | -0.024 |
?" | -0.024 |
Token | so |
Token ID | 523 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
squirrel | +0.016 |
rabbit | +0.014 |
mice | +0.014 |
monster | +0.014 |
rabbits | +0.014 |
Neg logit contributions | |
he | -0.011 |
ows | -0.010 |
but | -0.009 |
immer | -0.008 |
inkle | -0.008 |
Token | big |
Token ID | 1263 |
Feature activation | +6.55e-01 |
Pos logit contributions | |
like | +0.035 |
ows | +0.035 |
he | +0.035 |
inkle | +0.029 |
beyond | +0.028 |
Neg logit contributions | |
pillar | -0.060 |
squirrel | -0.058 |
villains | -0.058 |
bandits | -0.057 |
appearances | -0.057 |
Token | and |
Token ID | 290 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
he | +0.016 |
ows | +0.015 |
inkle | +0.012 |
like | +0.012 |
immer | +0.012 |
Neg logit contributions | |
squirrel | -0.032 |
pillar | -0.031 |
villains | -0.030 |
monster | -0.030 |
bandits | -0.030 |
Token | trem |
Token ID | 10023 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
pillar | +0.012 |
squirrel | +0.012 |
bandits | +0.012 |
villains | +0.012 |
appearances | +0.012 |
Neg logit contributions | |
like | -0.007 |
he | -0.007 |
ows | -0.007 |
inkle | -0.006 |
?" | -0.005 |
Token | ble |
Token ID | 903 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
he | +0.039 |
ows | +0.036 |
like | +0.036 |
rig | +0.031 |
inkle | +0.031 |
Neg logit contributions | |
squirrel | -0.054 |
villains | -0.052 |
monster | -0.052 |
pillar | -0.052 |
appearances | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
like | +0.012 |
ows | +0.012 |
he | +0.012 |
early | +0.009 |
about | +0.009 |
Neg logit contributions | |
pillar | -0.027 |
squirrel | -0.026 |
bandits | -0.026 |
robbers | -0.026 |
villains | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
he | +0.012 |
like | +0.012 |
ows | +0.011 |
?" | +0.010 |
beyond | +0.009 |
Neg logit contributions | |
pillar | -0.021 |
squirrel | -0.020 |
bandits | -0.020 |
villains | -0.020 |
monster | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
he | +0.003 |
ows | +0.003 |
like | +0.003 |
inkle | +0.002 |
?" | +0.002 |
Neg logit contributions | |
pillar | -0.005 |
squirrel | -0.005 |
villains | -0.005 |
bandits | -0.005 |
rook | -0.005 |
Token | " |
Token ID | 1 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
pillar | +0.006 |
squirrel | +0.006 |
monster | +0.006 |
rook | +0.006 |
brains | +0.006 |
Neg logit contributions | |
he | -0.004 |
like | -0.004 |
ows | -0.003 |
?" | -0.003 |
early | -0.003 |
Token | O |
Token ID | 46 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
he | +0.008 |
like | +0.007 |
ows | +0.007 |
but | +0.006 |
behind | +0.006 |
Neg logit contributions | |
rook | -0.006 |
pillar | -0.006 |
bandits | -0.006 |
weaknesses | -0.006 |
villains | -0.006 |
Token | uch |
Token ID | 794 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
pillar | +0.004 |
squirrel | +0.004 |
villains | +0.004 |
appearances | +0.004 |
weaknesses | +0.004 |
Neg logit contributions | |
he | -0.004 |
ows | -0.004 |
like | -0.004 |
beyond | -0.003 |
immer | -0.003 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
he | +0.016 |
ows | +0.015 |
like | +0.014 |
inkle | +0.012 |
beyond | +0.012 |
Neg logit contributions | |
squirrel | -0.032 |
pillar | -0.032 |
villains | -0.031 |
bandits | -0.030 |
monster | -0.030 |
Token | Sam |
Token ID | 3409 |
Feature activation | -6.63e-01 |
Pos logit contributions | |
he | +0.009 |
ows | +0.009 |
like | +0.009 |
inkle | +0.008 |
beyond | +0.008 |
Neg logit contributions | |
pillar | -0.007 |
squirrel | -0.007 |
villains | -0.007 |
bandits | -0.007 |
weaknesses | -0.007 |
Token | says |
Token ID | 1139 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
squirrel | +0.025 |
pillar | +0.025 |
villains | +0.025 |
bandits | +0.024 |
appearances | +0.024 |
Neg logit contributions | |
he | -0.022 |
ows | -0.021 |
like | -0.020 |
inkle | -0.018 |
beyond | -0.018 |
Token | very |
Token ID | 845 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
squirrel | +0.001 |
villains | +0.001 |
pillar | +0.001 |
robbers | +0.001 |
bandits | +0.001 |
Neg logit contributions | |
he | -0.001 |
ows | -0.001 |
like | -0.001 |
inkle | -0.001 |
rig | -0.001 |
Token | dangerous |
Token ID | 4923 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
he | +0.005 |
ows | +0.005 |
like | +0.005 |
inkle | +0.004 |
beyond | +0.004 |
Neg logit contributions | |
squirrel | -0.006 |
villains | -0.006 |
pillar | -0.005 |
bandits | -0.005 |
robbers | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
he | +0.011 |
ows | +0.011 |
like | +0.010 |
inkle | +0.009 |
beyond | +0.009 |
Neg logit contributions | |
squirrel | -0.019 |
pillar | -0.019 |
villains | -0.019 |
bandits | -0.018 |
robbers | -0.018 |
Token | he |
Token ID | 339 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
pillar | +0.007 |
squirrel | +0.006 |
rook | +0.006 |
villains | +0.006 |
appearances | +0.006 |
Neg logit contributions | |
like | -0.006 |
he | -0.006 |
ows | -0.006 |
beyond | -0.006 |
?" | -0.006 |
Token | should |
Token ID | 815 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
like | +0.000 |
he | +0.000 |
ows | +0.000 |
?" | +0.000 |
inkle | +0.000 |
Neg logit contributions | |
pillar | -0.001 |
appearances | -0.001 |
rook | -0.001 |
squirrel | -0.001 |
pawn | -0.001 |
Token | stay |
Token ID | 2652 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
pillar | +0.010 |
rook | +0.009 |
pawn | +0.009 |
bandits | +0.009 |
villains | +0.009 |
Neg logit contributions | |
like | -0.006 |
he | -0.005 |
ows | -0.005 |
?" | -0.005 |
inkle | -0.004 |
Token | away |
Token ID | 1497 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
he | +0.045 |
ows | +0.041 |
book | +0.037 |
mmm | +0.036 |
rig | +0.035 |
Neg logit contributions | |
squirrel | -0.041 |
drowned | -0.038 |
robbers | -0.037 |
bur | -0.037 |
fox | -0.037 |
Token | from |
Token ID | 422 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
he | +0.016 |
ows | +0.015 |
inkle | +0.014 |
immer | +0.013 |
ieved | +0.012 |
Neg logit contributions | |
squirrel | -0.030 |
villains | -0.029 |
monster | -0.028 |
rabbit | -0.028 |
turtle | -0.028 |
Token | it |
Token ID | 340 |
Feature activation | -6.85e-02 |
Pos logit contributions | |
like | +0.051 |
he | +0.047 |
beyond | +0.045 |
behind | +0.045 |
?" | +0.042 |
Neg logit contributions | |
pillar | -0.081 |
rook | -0.074 |
Carl | -0.073 |
weaknesses | -0.072 |
pawn | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
pillar | +0.003 |
squirrel | +0.003 |
villains | +0.003 |
bandits | +0.003 |
appearances | +0.003 |
Neg logit contributions | |
he | -0.002 |
ows | -0.002 |
like | -0.002 |
beyond | -0.001 |
inkle | -0.001 |
Token | Tim |
Token ID | 5045 |
Feature activation | +8.53e-01 |
Pos logit contributions | |
he | +0.011 |
ows | +0.011 |
like | +0.010 |
inkle | +0.010 |
beyond | +0.009 |
Neg logit contributions | |
squirrel | -0.011 |
pillar | -0.011 |
villains | -0.010 |
bandits | -0.010 |
robbers | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
he | +0.031 |
ows | +0.030 |
like | +0.028 |
inkle | +0.025 |
beyond | +0.024 |
Neg logit contributions | |
squirrel | -0.048 |
pillar | -0.047 |
villains | -0.046 |
bandits | -0.045 |
appearances | -0.045 |
Token | |
Token ID | 220 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
villains | +0.002 |
squirrel | +0.001 |
pillar | +0.001 |
bandits | +0.001 |
rook | +0.001 |
Neg logit contributions | |
he | -0.001 |
ows | -0.001 |
like | -0.001 |
beyond | -0.001 |
inkle | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
pillar | +0.001 |
squirrel | +0.001 |
Carl | +0.000 |
Le | +0.000 |
mice | +0.000 |
Neg logit contributions | |
inkle | -0.001 |
like | -0.001 |
immer | -0.001 |
he | -0.001 |
appet | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
ows | +0.014 |
like | +0.014 |
inkle | +0.014 |
ours | +0.013 |
immer | +0.013 |
Neg logit contributions | |
pillar | -0.009 |
squirrel | -0.008 |
Le | -0.008 |
Carl | -0.008 |
emy | -0.007 |
Token | But |
Token ID | 1537 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
he | +0.004 |
ows | +0.004 |
like | +0.004 |
inkle | +0.003 |
beyond | +0.003 |
Neg logit contributions | |
squirrel | -0.004 |
pillar | -0.004 |
villains | -0.004 |
appearances | -0.004 |
bandits | -0.004 |
Token | something |
Token ID | 1223 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
he | +0.001 |
ows | +0.001 |
like | +0.001 |
rig | +0.001 |
inkle | +0.001 |
Neg logit contributions | |
squirrel | -0.002 |
pillar | -0.002 |
rabbit | -0.002 |
monster | -0.002 |
villains | -0.002 |
Token | went |
Token ID | 1816 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
he | +0.028 |
ows | +0.028 |
like | +0.027 |
inkle | +0.025 |
beyond | +0.024 |
Neg logit contributions | |
squirrel | -0.025 |
pillar | -0.025 |
villains | -0.024 |
bandits | -0.024 |
appearances | -0.023 |
Token | wrong |
Token ID | 2642 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
he | +0.002 |
like | +0.002 |
ows | +0.002 |
beyond | +0.001 |
inkle | +0.001 |
Neg logit contributions | |
pillar | -0.003 |
villains | -0.003 |
squirrel | -0.003 |
rook | -0.003 |
weaknesses | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +3.40e-03 |
Pos logit contributions | |
ows | +0.010 |
he | +0.010 |
like | +0.010 |
beyond | +0.008 |
early | +0.008 |
Neg logit contributions | |
pillar | -0.019 |
squirrel | -0.019 |
villains | -0.018 |
bandits | -0.018 |
rook | -0.018 |
Token | The |
Token ID | 383 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
like | +0.000 |
ows | +0.000 |
he | +0.000 |
beyond | +0.000 |
about | +0.000 |
Neg logit contributions | |
villains | -0.000 |
pillar | -0.000 |
appearances | -0.000 |
rook | -0.000 |
Check | -0.000 |
Token | fire |
Token ID | 2046 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
squirrel | +0.003 |
pillar | +0.003 |
villains | +0.003 |
bandits | +0.003 |
monster | +0.003 |
Neg logit contributions | |
he | -0.005 |
ows | -0.005 |
like | -0.005 |
inkle | -0.005 |
beyond | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
squirrel | +0.048 |
pillar | +0.048 |
villains | +0.046 |
bandits | +0.045 |
appearances | +0.044 |
Neg logit contributions | |
he | -0.054 |
ows | -0.053 |
like | -0.051 |
inkle | -0.047 |
beyond | -0.046 |
Token | a |
Token ID | 257 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
squirrel | +0.011 |
monster | +0.010 |
villains | +0.010 |
bandits | +0.010 |
mice | +0.010 |
Neg logit contributions | |
he | -0.011 |
ows | -0.011 |
inkle | -0.009 |
immer | -0.009 |
like | -0.009 |
Token | hunter |
Token ID | 19177 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
like | +0.004 |
he | +0.004 |
early | +0.003 |
?" | +0.003 |
inkle | +0.003 |
Neg logit contributions | |
weaknesses | -0.004 |
pillar | -0.004 |
brains | -0.004 |
appearances | -0.004 |
villains | -0.004 |
Token | who |
Token ID | 508 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
like | +0.012 |
he | +0.011 |
ows | +0.011 |
beyond | +0.010 |
behind | +0.009 |
Neg logit contributions | |
weaknesses | -0.022 |
pillar | -0.022 |
brains | -0.021 |
villains | -0.021 |
bandits | -0.021 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
pillar | +0.039 |
weaknesses | +0.038 |
squirrel | +0.038 |
villains | +0.037 |
appearances | +0.037 |
Neg logit contributions | |
he | -0.039 |
ows | -0.038 |
like | -0.037 |
beyond | -0.034 |
inkle | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | -6.31e-01 |
Pos logit contributions | |
pillar | +0.064 |
Check | +0.059 |
villains | +0.058 |
weaknesses | +0.058 |
rook | +0.057 |
Neg logit contributions | |
like | -0.034 |
ows | -0.029 |
beyond | -0.029 |
about | -0.027 |
behind | -0.026 |
Token | catch |
Token ID | 4929 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
pillar | +0.022 |
squirrel | +0.022 |
villains | +0.021 |
bandits | +0.021 |
appearances | +0.021 |
Neg logit contributions | |
he | -0.023 |
ows | -0.022 |
like | -0.021 |
inkle | -0.020 |
beyond | -0.019 |
Token | Brown |
Token ID | 4373 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
like | +0.008 |
he | +0.008 |
ows | +0.007 |
beyond | +0.007 |
behind | +0.007 |
Neg logit contributions | |
pillar | -0.013 |
villains | -0.013 |
rook | -0.013 |
weaknesses | -0.013 |
appearances | -0.012 |
Token | ie |
Token ID | 494 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
squirrel | +0.038 |
pillar | +0.037 |
villains | +0.036 |
bandits | +0.035 |
appearances | +0.035 |
Neg logit contributions | |
he | -0.039 |
ows | -0.038 |
like | -0.037 |
inkle | -0.034 |
beyond | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
he | +0.020 |
ows | +0.020 |
like | +0.019 |
inkle | +0.016 |
beyond | +0.016 |
Neg logit contributions | |
pillar | -0.034 |
squirrel | -0.034 |
villains | -0.034 |
bandits | -0.033 |
weaknesses | -0.032 |
Token | Brown |
Token ID | 4373 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
squirrel | +0.007 |
villains | +0.006 |
pillar | +0.006 |
robbers | +0.006 |
rabbit | +0.006 |
Neg logit contributions | |
he | -0.010 |
ows | -0.010 |
like | -0.009 |
inkle | -0.009 |
immer | -0.009 |
Token | aid |
Token ID | 1698 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
he | +0.008 |
like | +0.008 |
ows | +0.007 |
immer | +0.006 |
but | +0.006 |
Neg logit contributions | |
villains | -0.012 |
squirrel | -0.012 |
robbers | -0.012 |
bandits | -0.012 |
pillar | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
he | +0.001 |
ows | +0.001 |
inkle | +0.001 |
like | +0.001 |
rig | +0.001 |
Neg logit contributions | |
squirrel | -0.001 |
pillar | -0.001 |
villains | -0.001 |
monster | -0.001 |
rabbit | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
ows | +0.002 |
he | +0.002 |
like | +0.002 |
about | +0.001 |
beyond | +0.001 |
Neg logit contributions | |
pillar | -0.002 |
squirrel | -0.002 |
villains | -0.002 |
weaknesses | -0.002 |
appearances | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
squirrel | +0.001 |
pillar | +0.001 |
villains | +0.001 |
bandits | +0.001 |
robbers | +0.001 |
Neg logit contributions | |
he | -0.001 |
ows | -0.001 |
like | -0.001 |
inkle | -0.001 |
beyond | -0.001 |
Token | L |
Token ID | 43 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
squirrel | +0.010 |
brains | +0.010 |
pillar | +0.009 |
weaknesses | +0.009 |
rook | +0.009 |
Neg logit contributions | |
he | -0.009 |
ows | -0.008 |
like | -0.008 |
early | -0.007 |
but | -0.007 |
Token | ily |
Token ID | 813 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
pillar | +0.033 |
villains | +0.032 |
squirrel | +0.031 |
monster | +0.029 |
bandits | +0.028 |
Neg logit contributions | |
he | -0.057 |
like | -0.055 |
ows | -0.055 |
inkle | -0.054 |
early | -0.053 |
Token | smiled |
Token ID | 13541 |
Feature activation | -3.66e-03 |
Pos logit contributions | |
villains | +0.053 |
pillar | +0.053 |
squirrel | +0.053 |
weaknesses | +0.052 |
appearances | +0.052 |
Neg logit contributions | |
he | -0.023 |
ows | -0.023 |
like | -0.021 |
inkle | -0.017 |
beyond | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
squirrel | +0.000 |
pillar | +0.000 |
rabbit | +0.000 |
villains | +0.000 |
monster | +0.000 |
Neg logit contributions | |
he | -0.000 |
ows | -0.000 |
inkle | -0.000 |
like | -0.000 |
immer | -0.000 |
Token | thanked |
Token ID | 26280 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
he | +0.012 |
ows | +0.012 |
like | +0.012 |
inkle | +0.010 |
beyond | +0.010 |
Neg logit contributions | |
squirrel | -0.016 |
pillar | -0.016 |
villains | -0.016 |
bandits | -0.015 |
appearances | -0.015 |
Token | her |
Token ID | 607 |
Feature activation | -6.90e-02 |
Pos logit contributions | |
he | +0.030 |
like | +0.030 |
ows | +0.028 |
but | +0.026 |
beyond | +0.026 |
Neg logit contributions | |
pillar | -0.033 |
weaknesses | -0.033 |
colony | -0.031 |
villains | -0.031 |
rook | -0.030 |
Token | dad |
Token ID | 9955 |
Feature activation | +8.49e-02 |
Pos logit contributions | |
pillar | +0.002 |
villains | +0.002 |
squirrel | +0.002 |
weaknesses | +0.002 |
bandits | +0.002 |
Neg logit contributions | |
ows | -0.003 |
he | -0.003 |
like | -0.003 |
inkle | -0.002 |
beyond | -0.002 |
Token | ingredients |
Token ID | 9391 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
mice | +0.009 |
pillar | +0.009 |
brains | +0.009 |
squirrel | +0.009 |
fle | +0.008 |
Neg logit contributions | |
ows | -0.013 |
immer | -0.013 |
he | -0.013 |
beyond | -0.013 |
rig | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
rig | +0.035 |
ows | +0.034 |
he | +0.034 |
inkle | +0.034 |
?", | +0.033 |
Neg logit contributions | |
squirrel | -0.039 |
halves | -0.038 |
pawn | -0.038 |
rook | -0.038 |
rabbit | -0.038 |
Token | watch |
Token ID | 2342 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
squirrel | +0.008 |
villains | +0.008 |
pillar | +0.007 |
robbers | +0.007 |
mice | +0.007 |
Neg logit contributions | |
he | -0.007 |
ows | -0.007 |
inkle | -0.007 |
immer | -0.006 |
like | -0.006 |
Token | as |
Token ID | 355 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
mice | +0.037 |
squirrel | +0.036 |
monster | +0.034 |
rabbits | +0.034 |
dinosaur | +0.033 |
Neg logit contributions | |
ows | -0.050 |
he | -0.050 |
adow | -0.047 |
rig | -0.047 |
immer | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
he | +0.001 |
like | +0.001 |
ows | +0.001 |
early | +0.001 |
beyond | +0.001 |
Neg logit contributions | |
pillar | -0.001 |
squirrel | -0.001 |
villains | -0.001 |
weaknesses | -0.001 |
bandits | -0.001 |
Token | dough |
Token ID | 15756 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
pillar | +0.033 |
villains | +0.033 |
squirrel | +0.033 |
appearances | +0.032 |
weaknesses | +0.031 |
Neg logit contributions | |
he | -0.025 |
like | -0.024 |
ows | -0.024 |
early | -0.021 |
inkle | -0.021 |
Token | stretched |
Token ID | 19110 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
pillar | +0.029 |
squirrel | +0.029 |
villains | +0.029 |
bandits | +0.028 |
appearances | +0.028 |
Neg logit contributions | |
he | -0.024 |
ows | -0.024 |
like | -0.023 |
inkle | -0.021 |
beyond | -0.020 |
Token | out |
Token ID | 503 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
squirrel | +0.016 |
pillar | +0.016 |
rabbit | +0.015 |
monster | +0.015 |
mice | +0.015 |
Neg logit contributions | |
he | -0.013 |
ows | -0.013 |
inkle | -0.011 |
rig | -0.011 |
like | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -8.23e-02 |
Pos logit contributions | |
pillar | +0.047 |
squirrel | +0.047 |
villains | +0.046 |
bandits | +0.044 |
robbers | +0.044 |
Neg logit contributions | |
he | -0.031 |
ows | -0.030 |
like | -0.029 |
inkle | -0.026 |
beyond | -0.025 |
Token | When |
Token ID | 1649 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
squirrel | +0.003 |
pillar | +0.003 |
villains | +0.003 |
bandits | +0.003 |
robbers | +0.003 |
Neg logit contributions | |
he | -0.003 |
ows | -0.002 |
like | -0.002 |
inkle | -0.002 |
beyond | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
he | +0.019 |
like | +0.017 |
early | +0.016 |
?" | +0.015 |
yet | +0.014 |
Neg logit contributions | |
pillar | -0.036 |
villains | -0.034 |
appearances | -0.033 |
weaknesses | -0.033 |
bandits | -0.032 |
Token | said |
Token ID | 531 |
Feature activation | +1.15e-03 |
Pos logit contributions | |
like | +0.014 |
he | +0.013 |
ows | +0.013 |
early | +0.011 |
beyond | +0.011 |
Neg logit contributions | |
pillar | -0.025 |
appearances | -0.024 |
villains | -0.024 |
weaknesses | -0.024 |
brains | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
like | +0.000 |
ows | +0.000 |
about | +0.000 |
he | +0.000 |
?" | +0.000 |
Neg logit contributions | |
pillar | -0.000 |
Check | -0.000 |
villains | -0.000 |
weaknesses | -0.000 |
bandits | -0.000 |
Token | " |
Token ID | 366 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
villains | +0.016 |
pillar | +0.016 |
squirrel | +0.016 |
bandits | +0.015 |
weaknesses | +0.015 |
Neg logit contributions | |
he | -0.013 |
like | -0.013 |
ows | -0.013 |
early | -0.011 |
beyond | -0.011 |
Token | Wow |
Token ID | 22017 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
he | +0.010 |
like | +0.009 |
but | +0.008 |
aron | +0.008 |
ows | +0.007 |
Neg logit contributions | |
appearances | -0.010 |
brains | -0.009 |
villains | -0.009 |
rook | -0.009 |
monster | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
villains | +0.023 |
pillar | +0.023 |
squirrel | +0.023 |
bandits | +0.022 |
weaknesses | +0.022 |
Neg logit contributions | |
ows | -0.009 |
he | -0.009 |
like | -0.009 |
?" | -0.007 |
inkle | -0.007 |
Token | what |
Token ID | 644 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
like | +0.006 |
he | +0.006 |
ows | +0.005 |
about | +0.005 |
?" | +0.005 |
Neg logit contributions | |
colony | -0.009 |
pillar | -0.009 |
weaknesses | -0.008 |
rook | -0.008 |
bandits | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
pillar | +0.020 |
Carl | +0.019 |
bur | +0.019 |
bandits | +0.019 |
drowned | +0.019 |
Neg logit contributions | |
like | -0.012 |
he | -0.010 |
about | -0.010 |
ows | -0.010 |
?" | -0.010 |
Token | noisy |
Token ID | 31210 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
like | +0.030 |
he | +0.028 |
?" | +0.022 |
proud | +0.021 |
beyond | +0.021 |
Neg logit contributions | |
appearances | -0.061 |
pillar | -0.060 |
mice | -0.058 |
bandits | -0.058 |
weaknesses | -0.058 |
Token | penny |
Token ID | 26369 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
he | +0.048 |
like | +0.047 |
?" | +0.041 |
ows | +0.041 |
inkle | +0.041 |
Neg logit contributions | |
villains | -0.037 |
bandits | -0.036 |
Le | -0.036 |
Carl | -0.036 |
weaknesses | -0.035 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
he | +0.022 |
ows | +0.021 |
like | +0.020 |
inkle | +0.018 |
beyond | +0.018 |
Neg logit contributions | |
squirrel | -0.029 |
pillar | -0.028 |
villains | -0.028 |
bandits | -0.027 |
appearances | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
he | +0.015 |
ows | +0.015 |
like | +0.015 |
inkle | +0.014 |
immer | +0.013 |
Neg logit contributions | |
squirrel | -0.013 |
pillar | -0.013 |
villains | -0.012 |
bandits | -0.012 |
robbers | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
villains | +0.036 |
pillar | +0.035 |
squirrel | +0.035 |
appearances | +0.035 |
robbers | +0.034 |
Neg logit contributions | |
ows | -0.030 |
like | -0.029 |
he | -0.028 |
beyond | -0.025 |
behind | -0.024 |
Token | All |
Token ID | 1439 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
he | +0.003 |
ows | +0.003 |
like | +0.003 |
inkle | +0.002 |
beyond | +0.002 |
Neg logit contributions | |
squirrel | -0.003 |
pillar | -0.003 |
villains | -0.003 |
bandits | -0.003 |
appearances | -0.003 |
Token | ie |
Token ID | 494 |
Feature activation | +7.89e-01 |
Pos logit contributions | |
villains | +0.087 |
robbers | +0.084 |
rook | +0.084 |
squirrel | +0.084 |
pillar | +0.083 |
Neg logit contributions | |
he | -0.044 |
like | -0.043 |
ows | -0.039 |
beyond | -0.037 |
about | -0.034 |
Token | thought |
Token ID | 1807 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
ows | +0.027 |
he | +0.026 |
like | +0.026 |
inkle | +0.023 |
beyond | +0.023 |
Neg logit contributions | |
appearances | -0.029 |
villains | -0.029 |
squirrel | -0.029 |
robbers | -0.029 |
rook | -0.028 |
Token | about |
Token ID | 546 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
squirrel | +0.014 |
pillar | +0.014 |
villains | +0.014 |
bandits | +0.013 |
robbers | +0.013 |
Neg logit contributions | |
he | -0.011 |
ows | -0.011 |
like | -0.010 |
inkle | -0.009 |
beyond | -0.009 |
Token | how |
Token ID | 703 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
like | +0.035 |
about | +0.029 |
ows | +0.029 |
beyond | +0.028 |
he | +0.027 |
Neg logit contributions | |
pillar | -0.060 |
villains | -0.057 |
bandits | -0.057 |
Check | -0.056 |
Carl | -0.056 |
Token | perfect |
Token ID | 2818 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
pillar | +0.067 |
robbers | +0.066 |
appearances | +0.066 |
rook | +0.066 |
villains | +0.065 |
Neg logit contributions | |
like | -0.035 |
ows | -0.031 |
he | -0.030 |
early | -0.029 |
about | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
he | +0.002 |
ows | +0.002 |
like | +0.002 |
beyond | +0.002 |
inkle | +0.002 |
Neg logit contributions | |
pillar | -0.003 |
squirrel | -0.003 |
villains | -0.003 |
bandits | -0.003 |
robbers | -0.003 |
Token | gift |
Token ID | 6979 |
Feature activation | -8.69e-01 |
Pos logit contributions | |
like | +0.024 |
he | +0.023 |
beyond | +0.021 |
early | +0.021 |
ows | +0.020 |
Neg logit contributions | |
villains | -0.020 |
pillar | -0.019 |
Carl | -0.019 |
bandits | -0.019 |
Le | -0.019 |
Token | would |
Token ID | 561 |
Feature activation | +6.93e-01 |
Pos logit contributions | |
squirrel | +0.030 |
pillar | +0.030 |
villains | +0.029 |
monster | +0.028 |
bandits | +0.028 |
Neg logit contributions | |
he | -0.031 |
ows | -0.030 |
like | -0.029 |
inkle | -0.027 |
immer | -0.026 |
Token | be |
Token ID | 307 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
he | +0.024 |
ows | +0.023 |
like | +0.023 |
inkle | +0.021 |
beyond | +0.020 |
Neg logit contributions | |
pillar | -0.025 |
squirrel | -0.025 |
villains | -0.024 |
bandits | -0.024 |
robbers | -0.023 |
Token | ." |
Token ID | 526 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
he | +0.024 |
ows | +0.023 |
like | +0.022 |
inkle | +0.019 |
beyond | +0.019 |
Neg logit contributions | |
squirrel | -0.041 |
pillar | -0.041 |
villains | -0.040 |
bandits | -0.039 |
appearances | -0.038 |
Token | S |
Token ID | 311 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
he | +0.005 |
ows | +0.005 |
like | +0.005 |
inkle | +0.004 |
beyond | +0.004 |
Neg logit contributions | |
pillar | -0.006 |
squirrel | -0.006 |
villains | -0.006 |
bandits | -0.006 |
appearances | -0.006 |
Token | issy |
Token ID | 36419 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
squirrel | +0.009 |
pillar | +0.009 |
villains | +0.008 |
bandits | +0.008 |
rabbit | +0.007 |
Neg logit contributions | |
he | -0.016 |
ows | -0.016 |
like | -0.016 |
inkle | -0.015 |
beyond | -0.015 |
Token | didn |
Token ID | 1422 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
villains | +0.013 |
pillar | +0.013 |
weaknesses | +0.013 |
robbers | +0.013 |
squirrel | +0.012 |
Neg logit contributions | |
ows | -0.009 |
like | -0.008 |
he | -0.008 |
?" | -0.007 |
about | -0.007 |
Token | 't |
Token ID | 470 |
Feature activation | +3.71e-03 |
Pos logit contributions | |
ows | +0.016 |
adow | +0.015 |
he | +0.015 |
like | +0.014 |
immer | +0.014 |
Neg logit contributions | |
pillar | -0.010 |
squirrel | -0.010 |
villains | -0.009 |
monster | -0.009 |
Monster | -0.009 |
Token | listen |
Token ID | 6004 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
he | +0.000 |
ows | +0.000 |
inkle | +0.000 |
rig | +0.000 |
beyond | +0.000 |
Neg logit contributions | |
squirrel | -0.000 |
villains | -0.000 |
pillar | -0.000 |
robbers | -0.000 |
turtle | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
squirrel | +0.089 |
pillar | +0.086 |
villains | +0.084 |
monster | +0.082 |
bandits | +0.082 |
Neg logit contributions | |
he | -0.068 |
ows | -0.065 |
like | -0.060 |
inkle | -0.058 |
beyond | -0.055 |
Token | kept |
Token ID | 4030 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
like | +0.002 |
ows | +0.002 |
he | +0.002 |
early | +0.002 |
beyond | +0.001 |
Neg logit contributions | |
pillar | -0.002 |
villains | -0.002 |
appearances | -0.002 |
weaknesses | -0.002 |
bandits | -0.002 |
Token | chewing |
Token ID | 36615 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
pillar | +0.018 |
villains | +0.017 |
weaknesses | +0.016 |
appearances | +0.015 |
bandits | +0.015 |
Neg logit contributions | |
like | -0.014 |
he | -0.013 |
early | -0.012 |
behind | -0.012 |
beyond | -0.012 |
Token | on |
Token ID | 319 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
squirrel | +0.072 |
pillar | +0.070 |
villains | +0.068 |
monster | +0.067 |
mice | +0.067 |
Neg logit contributions | |
he | -0.042 |
ows | -0.040 |
like | -0.036 |
rig | -0.035 |
inkle | -0.034 |
Token | her |
Token ID | 607 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
villains | +0.033 |
pillar | +0.033 |
bandits | +0.032 |
Check | +0.032 |
weaknesses | +0.031 |
Neg logit contributions | |
like | -0.017 |
ows | -0.016 |
he | -0.015 |
behind | -0.013 |
early | -0.013 |
Token | have |
Token ID | 423 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
pillar | +0.006 |
bandits | +0.006 |
squirrel | +0.006 |
villains | +0.006 |
Le | +0.006 |
Neg logit contributions | |
like | -0.003 |
he | -0.003 |
?" | -0.003 |
ows | -0.003 |
inkle | -0.002 |
Token | been |
Token ID | 587 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
pillar | +0.016 |
villains | +0.015 |
squirrel | +0.015 |
bandits | +0.014 |
appearances | +0.014 |
Neg logit contributions | |
like | -0.012 |
he | -0.012 |
ows | -0.011 |
?" | -0.011 |
?", | -0.010 |
Token | polite |
Token ID | 23507 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
he | +0.004 |
ows | +0.004 |
like | +0.004 |
inkle | +0.004 |
beyond | +0.004 |
Neg logit contributions | |
pillar | -0.005 |
squirrel | -0.004 |
villains | -0.004 |
bandits | -0.004 |
appearances | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
squirrel | +0.054 |
pillar | +0.054 |
villains | +0.053 |
bandits | +0.052 |
robbers | +0.051 |
Neg logit contributions | |
he | -0.025 |
ows | -0.024 |
like | -0.023 |
inkle | -0.020 |
beyond | -0.019 |
Token | asked |
Token ID | 1965 |
Feature activation | -9.25e-01 |
Pos logit contributions | |
like | +0.001 |
ows | +0.001 |
he | +0.001 |
early | +0.000 |
beyond | +0.000 |
Neg logit contributions | |
pillar | -0.001 |
squirrel | -0.001 |
villains | -0.001 |
bandits | -0.001 |
appearances | -0.001 |
Token | nicely |
Token ID | 16576 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
squirrel | +0.038 |
pillar | +0.037 |
villains | +0.036 |
bandits | +0.035 |
robbers | +0.035 |
Neg logit contributions | |
he | -0.029 |
ows | -0.028 |
like | -0.026 |
inkle | -0.024 |
beyond | -0.023 |
Token | ." |
Token ID | 526 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
squirrel | +0.110 |
villains | +0.106 |
pillar | +0.106 |
mice | +0.105 |
robbers | +0.105 |
Neg logit contributions | |
he | -0.094 |
ows | -0.083 |
inkle | -0.082 |
immer | -0.079 |
like | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
he | +0.005 |
ows | +0.004 |
like | +0.004 |
inkle | +0.004 |
immer | +0.004 |
Neg logit contributions | |
squirrel | -0.005 |
pillar | -0.004 |
villains | -0.004 |
rook | -0.004 |
robbers | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
squirrel | +0.009 |
pillar | +0.009 |
villains | +0.009 |
bandits | +0.008 |
appearances | +0.008 |
Neg logit contributions | |
he | -0.007 |
ows | -0.007 |
like | -0.006 |
inkle | -0.006 |
beyond | -0.006 |
Token | They |
Token ID | 2990 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
pillar | +0.014 |
brains | +0.014 |
squirrel | +0.014 |
weaknesses | +0.014 |
monster | +0.013 |
Neg logit contributions | |
he | -0.007 |
like | -0.006 |
ows | -0.006 |
early | -0.006 |
?" | -0.005 |
Token | smiled |
Token ID | 13541 |
Feature activation | -8.90e-02 |
Pos logit contributions | |
pillar | +0.009 |
villains | +0.009 |
appearances | +0.008 |
weaknesses | +0.008 |
squirrel | +0.008 |
Neg logit contributions | |
he | -0.008 |
like | -0.008 |
ows | -0.007 |
behind | -0.006 |
early | -0.006 |
Token | mouse |
Token ID | 10211 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
Boo | +0.006 |
Check | +0.005 |
weaknesses | +0.005 |
appearances | +0.005 |
Le | +0.005 |
Neg logit contributions | |
like | -0.012 |
he | -0.011 |
early | -0.011 |
ows | -0.011 |
but | -0.011 |
Token | took |
Token ID | 1718 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
appearances | +0.015 |
villains | +0.014 |
weaknesses | +0.014 |
Check | +0.014 |
robbers | +0.014 |
Neg logit contributions | |
like | -0.010 |
ows | -0.009 |
he | -0.008 |
about | -0.008 |
?" | -0.008 |
Token | some |
Token ID | 617 |
Feature activation | +8.52e-02 |
Pos logit contributions | |
like | +0.014 |
he | +0.014 |
ows | +0.012 |
beyond | +0.011 |
behind | +0.010 |
Neg logit contributions | |
villains | -0.029 |
pillar | -0.029 |
appearances | -0.029 |
weaknesses | -0.028 |
squirrel | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
he | +0.003 |
like | +0.003 |
ows | +0.003 |
inkle | +0.002 |
beyond | +0.002 |
Neg logit contributions | |
villains | -0.003 |
pillar | -0.003 |
squirrel | -0.003 |
appearances | -0.003 |
bandits | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
appearances | +0.011 |
pillar | +0.011 |
villains | +0.010 |
weaknesses | +0.010 |
squirrel | +0.009 |
Neg logit contributions | |
like | -0.008 |
early | -0.007 |
he | -0.007 |
ows | -0.007 |
behind | -0.007 |
Token | other |
Token ID | 584 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
pillar | +0.025 |
villains | +0.024 |
Check | +0.024 |
Boo | +0.023 |
appearances | +0.023 |
Neg logit contributions | |
like | -0.029 |
ows | -0.026 |
he | -0.025 |
but | -0.025 |
early | -0.025 |
Token | mouse |
Token ID | 10211 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
pillar | +0.071 |
villains | +0.069 |
appearances | +0.069 |
Check | +0.067 |
rook | +0.065 |
Neg logit contributions | |
like | -0.107 |
ows | -0.100 |
he | -0.095 |
beyond | -0.092 |
but | -0.091 |
Token | took |
Token ID | 1718 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
appearances | +0.073 |
villains | +0.072 |
pillar | +0.070 |
Check | +0.069 |
Ro | +0.069 |
Neg logit contributions | |
like | -0.041 |
ows | -0.036 |
about | -0.032 |
he | -0.032 |
behind | -0.031 |
Token | some |
Token ID | 617 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
pillar | +0.002 |
villains | +0.002 |
appearances | +0.001 |
squirrel | +0.001 |
weaknesses | +0.001 |
Neg logit contributions | |
like | -0.001 |
he | -0.001 |
ows | -0.001 |
beyond | -0.001 |
behind | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
villains | +0.015 |
pillar | +0.014 |
squirrel | +0.014 |
appearances | +0.014 |
bandits | +0.014 |
Neg logit contributions | |
he | -0.011 |
like | -0.011 |
ows | -0.010 |
inkle | -0.009 |
beyond | -0.009 |
Token | too |
Token ID | 1165 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
pillar | +0.007 |
villains | +0.007 |
squirrel | +0.007 |
appearances | +0.007 |
bandits | +0.007 |
Neg logit contributions | |
he | -0.006 |
ows | -0.006 |
like | -0.006 |
beyond | -0.006 |
early | -0.005 |
Token | his |
Token ID | 465 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
he | +0.002 |
ows | +0.002 |
like | +0.002 |
inkle | +0.002 |
immer | +0.002 |
Neg logit contributions | |
squirrel | -0.003 |
pillar | -0.003 |
villains | -0.003 |
bandits | -0.003 |
monster | -0.003 |
Token | wall |
Token ID | 3355 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ows | +0.017 |
he | +0.017 |
like | +0.016 |
inkle | +0.015 |
early | +0.015 |
Neg logit contributions | |
pillar | -0.017 |
Check | -0.016 |
squirrel | -0.016 |
villains | -0.016 |
bandits | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
squirrel | +0.026 |
pillar | +0.026 |
villains | +0.025 |
bandits | +0.025 |
robbers | +0.024 |
Neg logit contributions | |
he | -0.014 |
ows | -0.013 |
like | -0.012 |
inkle | -0.011 |
beyond | -0.010 |
Token | put |
Token ID | 1234 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
pillar | +0.007 |
squirrel | +0.006 |
villains | +0.006 |
appearances | +0.006 |
weaknesses | +0.006 |
Neg logit contributions | |
he | -0.006 |
ows | -0.006 |
like | -0.006 |
beyond | -0.005 |
early | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
he | +0.011 |
ows | +0.010 |
like | +0.010 |
inkle | +0.009 |
rig | +0.009 |
Neg logit contributions | |
squirrel | -0.012 |
pillar | -0.011 |
villains | -0.011 |
monster | -0.011 |
bandits | -0.011 |
Token | picture |
Token ID | 4286 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
squirrel | +0.017 |
pillar | +0.017 |
villains | +0.016 |
bandits | +0.016 |
robbers | +0.015 |
Neg logit contributions | |
he | -0.019 |
ows | -0.018 |
like | -0.017 |
inkle | -0.016 |
beyond | -0.016 |
Token | of |
Token ID | 286 |
Feature activation | +7.70e-01 |
Pos logit contributions | |
pillar | +0.110 |
appearances | +0.109 |
squirrel | +0.108 |
weaknesses | +0.108 |
villains | +0.108 |
Neg logit contributions | |
he | -0.049 |
ows | -0.048 |
like | -0.046 |
beyond | -0.037 |
about | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -6.67e-02 |
Pos logit contributions | |
like | +0.021 |
he | +0.019 |
ows | +0.018 |
beyond | +0.017 |
about | +0.016 |
Neg logit contributions | |
pillar | -0.036 |
weaknesses | -0.033 |
villains | -0.033 |
squirrel | -0.032 |
bandits | -0.032 |
Token | two |
Token ID | 734 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
pillar | +0.002 |
villains | +0.002 |
weaknesses | +0.002 |
Carl | +0.002 |
bandits | +0.002 |
Neg logit contributions | |
he | -0.003 |
like | -0.003 |
ows | -0.003 |
beyond | -0.002 |
inkle | -0.002 |
Token | friends |
Token ID | 2460 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
appearances | +0.059 |
Carl | +0.059 |
Shark | +0.058 |
Ro | +0.058 |
pillar | +0.058 |
Neg logit contributions | |
like | -0.062 |
he | -0.059 |
ows | -0.058 |
behind | -0.055 |
beyond | -0.052 |
Token | in |
Token ID | 287 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
pillar | +0.016 |
villains | +0.016 |
squirrel | +0.016 |
appearances | +0.016 |
bandits | +0.016 |
Neg logit contributions | |
ows | -0.007 |
like | -0.006 |
he | -0.006 |
beyond | -0.005 |
adow | -0.005 |
Token | time |
Token ID | 640 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
like | +0.010 |
he | +0.010 |
early | +0.009 |
beyond | +0.008 |
but | +0.008 |
Neg logit contributions | |
pillar | -0.017 |
weaknesses | -0.016 |
villains | -0.016 |
appearances | -0.016 |
Check | -0.016 |
Token | passed |
Token ID | 3804 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
he | +0.001 |
like | +0.001 |
ows | +0.001 |
about | +0.001 |
beyond | +0.001 |
Neg logit contributions | |
appearances | -0.001 |
weaknesses | -0.001 |
pillar | -0.001 |
villains | -0.001 |
emy | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -6.35e-04 |
Pos logit contributions | |
squirrel | +0.011 |
pillar | +0.010 |
villains | +0.010 |
bandits | +0.010 |
monster | +0.010 |
Neg logit contributions | |
he | -0.008 |
ows | -0.007 |
like | -0.007 |
inkle | -0.006 |
immer | -0.006 |
Token | Sarah |
Token ID | 10490 |
Feature activation | +9.96e-02 |
Pos logit contributions | |
pillar | +0.000 |
weaknesses | +0.000 |
villains | +0.000 |
appearances | +0.000 |
squirrel | +0.000 |
Neg logit contributions | |
he | -0.000 |
like | -0.000 |
ows | -0.000 |
early | -0.000 |
beyond | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
like | +0.003 |
he | +0.003 |
ows | +0.003 |
?" | +0.003 |
beyond | +0.003 |
Neg logit contributions | |
villains | -0.004 |
appearances | -0.004 |
weaknesses | -0.004 |
squirrel | -0.004 |
brains | -0.004 |
Token | Gray |
Token ID | 12723 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
pillar | +0.056 |
villains | +0.054 |
weaknesses | +0.054 |
Carl | +0.052 |
appearances | +0.052 |
Neg logit contributions | |
he | -0.028 |
like | -0.027 |
early | -0.024 |
beyond | -0.023 |
about | -0.022 |
Token | became |
Token ID | 2627 |
Feature activation | -5.75e-01 |
Pos logit contributions | |
appearances | +0.088 |
pillar | +0.086 |
villains | +0.085 |
weaknesses | +0.083 |
emy | +0.082 |
Neg logit contributions | |
like | -0.079 |
ows | -0.071 |
he | -0.070 |
beyond | -0.060 |
adow | -0.059 |
Token | good |
Token ID | 922 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
pillar | +0.024 |
villains | +0.022 |
appearances | +0.022 |
weaknesses | +0.022 |
squirrel | +0.021 |
Neg logit contributions | |
like | -0.019 |
he | -0.017 |
ows | -0.016 |
early | -0.016 |
beyond | -0.016 |
Token | friends |
Token ID | 2460 |
Feature activation | -9.14e-03 |
Pos logit contributions | |
pillar | +0.049 |
appearances | +0.048 |
Carl | +0.045 |
rook | +0.045 |
Le | +0.044 |
Neg logit contributions | |
like | -0.072 |
he | -0.067 |
early | -0.066 |
ows | -0.064 |
about | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
squirrel | +0.000 |
pillar | +0.000 |
villains | +0.000 |
bandits | +0.000 |
appearances | +0.000 |
Neg logit contributions | |
he | -0.000 |
ows | -0.000 |
like | -0.000 |
inkle | -0.000 |
beyond | -0.000 |
Token | Sarah |
Token ID | 10490 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
squirrel | +0.007 |
pillar | +0.007 |
villains | +0.007 |
bandits | +0.007 |
monster | +0.007 |
Neg logit contributions | |
he | -0.005 |
like | -0.005 |
ows | -0.005 |
about | -0.004 |
early | -0.004 |