Token | wealthy |
Token ID | 11574 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
owner | +0.002 |
tr | +0.001 |
flower | +0.001 |
hits | +0.001 |
hurts | +0.001 |
Neg logit contributions | |
ocol | -0.003 |
irs | -0.002 |
sie | -0.002 |
ait | -0.002 |
andel | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
owner | +0.013 |
not | +0.012 |
still | +0.011 |
was | +0.011 |
only | +0.010 |
Neg logit contributions | |
ocol | -0.034 |
irs | -0.030 |
ait | -0.027 |
andel | -0.027 |
sie | -0.026 |
Token | This |
Token ID | 770 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
owner | +0.012 |
not | +0.011 |
still | +0.010 |
hurts | +0.010 |
hits | +0.009 |
Neg logit contributions | |
ocol | -0.039 |
irs | -0.033 |
ait | -0.031 |
andel | -0.030 |
sie | -0.030 |
Token | made |
Token ID | 925 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
Fighting | +0.057 |
neither | +0.056 |
tr | +0.054 |
flower | +0.054 |
gy | +0.053 |
Neg logit contributions | |
ocol | -0.064 |
seas | -0.052 |
sie | -0.048 |
ages | -0.046 |
andel | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
owner | +0.022 |
not | +0.021 |
still | +0.020 |
was | +0.019 |
hurts | +0.019 |
Neg logit contributions | |
ocol | -0.032 |
irs | -0.026 |
ait | -0.024 |
andel | -0.023 |
sie | -0.023 |
Token | other |
Token ID | 584 |
Feature activation | +6.06e+00 |
Pos logit contributions | |
tr | +0.042 |
zag | +0.041 |
not | +0.041 |
hurts | +0.040 |
owner | +0.039 |
Neg logit contributions | |
ocol | -0.076 |
irs | -0.057 |
andel | -0.056 |
sie | -0.055 |
ages | -0.053 |
Token | rock |
Token ID | 3881 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
not | +0.112 |
owner | +0.108 |
tr | +0.104 |
zag | +0.104 |
hurts | +0.102 |
Neg logit contributions | |
ocol | -0.239 |
irs | -0.189 |
andel | -0.180 |
ait | -0.178 |
sie | -0.175 |
Token | er |
Token ID | 263 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
owner | +0.052 |
not | +0.045 |
hurts | +0.045 |
hits | +0.045 |
still | +0.044 |
Neg logit contributions | |
ocol | -0.102 |
irs | -0.087 |
ait | -0.078 |
andel | -0.077 |
sie | -0.076 |
Token | angry |
Token ID | 7954 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
owner | +0.043 |
hits | +0.038 |
hurts | +0.038 |
not | +0.036 |
remembered | +0.036 |
Neg logit contributions | |
ocol | -0.102 |
irs | -0.087 |
ait | -0.079 |
andel | -0.077 |
sie | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
owner | +0.033 |
not | +0.029 |
still | +0.028 |
zag | +0.028 |
flower | +0.028 |
Neg logit contributions | |
ocol | -0.096 |
irs | -0.082 |
ait | -0.075 |
andel | -0.074 |
sie | -0.073 |
Token | He |
Token ID | 679 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
owner | +0.011 |
not | +0.010 |
still | +0.010 |
hurts | +0.009 |
hits | +0.009 |
Neg logit contributions | |
ocol | -0.037 |
irs | -0.032 |
ait | -0.030 |
andel | -0.029 |
sie | -0.029 |
Token | than |
Token ID | 621 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
owner | +0.034 |
not | +0.031 |
hurts | +0.030 |
remembered | +0.030 |
still | +0.029 |
Neg logit contributions | |
ocol | -0.095 |
irs | -0.080 |
ait | -0.074 |
andel | -0.074 |
sie | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
ocol | +0.047 |
irs | +0.040 |
ait | +0.037 |
andel | +0.037 |
prise | +0.035 |
Neg logit contributions | |
not | -0.021 |
owner | -0.020 |
still | -0.020 |
only | -0.019 |
was | -0.018 |
Token | others |
Token ID | 1854 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
owner | +0.014 |
not | +0.013 |
hurts | +0.013 |
zag | +0.013 |
hits | +0.013 |
Neg logit contributions | |
ocol | -0.029 |
irs | -0.023 |
ait | -0.022 |
andel | -0.021 |
sie | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
owner | +0.036 |
not | +0.034 |
still | +0.032 |
hurts | +0.031 |
was | +0.030 |
Neg logit contributions | |
ocol | -0.086 |
irs | -0.074 |
ait | -0.067 |
andel | -0.066 |
sie | -0.065 |
Token | The |
Token ID | 383 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
owner | +0.012 |
not | +0.010 |
still | +0.010 |
was | +0.009 |
hurts | +0.009 |
Neg logit contributions | |
ocol | -0.040 |
irs | -0.034 |
ait | -0.032 |
andel | -0.031 |
sie | -0.031 |
Token | other |
Token ID | 584 |
Feature activation | +5.71e+00 |
Pos logit contributions | |
zag | +0.065 |
hurts | +0.063 |
tr | +0.063 |
owner | +0.061 |
not | +0.061 |
Neg logit contributions | |
ocol | -0.097 |
irs | -0.073 |
andel | -0.071 |
sie | -0.069 |
ait | -0.069 |
Token | ducks |
Token ID | 39694 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
zag | +0.150 |
hurts | +0.149 |
tr | +0.147 |
not | +0.141 |
hits | +0.140 |
Neg logit contributions | |
ocol | -0.167 |
ait | -0.119 |
sie | -0.115 |
andel | -0.115 |
irs | -0.113 |
Token | didn |
Token ID | 1422 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
owner | +0.044 |
tr | +0.042 |
hurts | +0.041 |
hits | +0.040 |
zag | +0.039 |
Neg logit contributions | |
ocol | -0.087 |
irs | -0.074 |
sie | -0.071 |
ait | -0.067 |
year | -0.066 |
Token | 't |
Token ID | 470 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
ocol | +0.026 |
irs | +0.024 |
andel | +0.020 |
ait | +0.020 |
Annie | +0.019 |
Neg logit contributions | |
owner | -0.014 |
remembered | -0.014 |
flower | -0.013 |
hurts | -0.012 |
hates | -0.012 |
Token | mind |
Token ID | 2000 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
owner | +0.026 |
not | +0.025 |
still | +0.023 |
also | +0.021 |
hurts | +0.021 |
Neg logit contributions | |
ocol | -0.069 |
irs | -0.060 |
ait | -0.055 |
andel | -0.053 |
sie | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
owner | +0.032 |
not | +0.031 |
still | +0.029 |
hurts | +0.027 |
was | +0.027 |
Neg logit contributions | |
ocol | -0.087 |
irs | -0.075 |
ait | -0.069 |
andel | -0.067 |
sie | -0.066 |
Token | starting |
Token ID | 3599 |
Feature activation | +3.81e+00 |
Pos logit contributions | |
ocol | +0.049 |
irs | +0.041 |
ait | +0.037 |
andel | +0.036 |
sie | +0.036 |
Neg logit contributions | |
owner | -0.025 |
not | -0.023 |
still | -0.022 |
hurts | -0.022 |
zag | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
owner | +0.053 |
not | +0.053 |
still | +0.044 |
was | +0.041 |
also | +0.039 |
Neg logit contributions | |
ocol | -0.187 |
irs | -0.160 |
ait | -0.149 |
andel | -0.148 |
sie | -0.147 |
Token | rot |
Token ID | 5724 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ocol | +0.009 |
irs | +0.007 |
ait | +0.007 |
andel | +0.007 |
sie | +0.006 |
Neg logit contributions | |
owner | -0.004 |
not | -0.004 |
still | -0.003 |
also | -0.003 |
hurts | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
owner | +0.021 |
still | +0.020 |
not | +0.020 |
was | +0.019 |
only | +0.018 |
Neg logit contributions | |
ocol | -0.070 |
irs | -0.062 |
ait | -0.057 |
andel | -0.056 |
prise | -0.055 |
Token | The |
Token ID | 383 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
owner | +0.005 |
not | +0.005 |
was | +0.004 |
still | +0.004 |
only | +0.004 |
Neg logit contributions | |
ocol | -0.023 |
irs | -0.020 |
ait | -0.019 |
sie | -0.019 |
andel | -0.019 |
Token | other |
Token ID | 584 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
owner | +0.053 |
not | +0.047 |
still | +0.046 |
only | +0.042 |
hurts | +0.042 |
Neg logit contributions | |
ocol | -0.106 |
irs | -0.093 |
ait | -0.082 |
sie | -0.080 |
andel | -0.080 |
Token | side |
Token ID | 1735 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
owner | +0.122 |
not | +0.113 |
still | +0.110 |
hurts | +0.106 |
was | +0.104 |
Neg logit contributions | |
ocol | -0.215 |
irs | -0.181 |
ait | -0.162 |
andel | -0.158 |
sie | -0.156 |
Token | was |
Token ID | 373 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
owner | +0.050 |
not | +0.047 |
hurts | +0.045 |
zag | +0.044 |
still | +0.042 |
Neg logit contributions | |
ocol | -0.162 |
irs | -0.135 |
ait | -0.132 |
sie | -0.125 |
year | -0.124 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
owner | +0.022 |
hurts | +0.021 |
hits | +0.021 |
flower | +0.020 |
Because | +0.020 |
Neg logit contributions | |
ocol | -0.045 |
irs | -0.036 |
ait | -0.034 |
andel | -0.033 |
rubble | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
owner | +0.006 |
not | +0.006 |
hurts | +0.005 |
still | +0.005 |
hits | +0.005 |
Neg logit contributions | |
ocol | -0.019 |
irs | -0.016 |
ait | -0.015 |
andel | -0.015 |
sie | -0.015 |
Token | creative |
Token ID | 7325 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
ocol | +0.023 |
irs | +0.019 |
ait | +0.017 |
andel | +0.017 |
sie | +0.017 |
Neg logit contributions | |
owner | -0.012 |
not | -0.011 |
zag | -0.010 |
hurts | -0.010 |
still | -0.010 |
Token | all |
Token ID | 477 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
owner | +0.014 |
not | +0.013 |
still | +0.013 |
was | +0.012 |
also | +0.012 |
Neg logit contributions | |
ocol | -0.033 |
irs | -0.028 |
ait | -0.026 |
andel | -0.025 |
sie | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -7.57e-01 |
Pos logit contributions | |
owner | +0.006 |
not | +0.005 |
still | +0.005 |
was | +0.005 |
hits | +0.005 |
Neg logit contributions | |
ocol | -0.014 |
irs | -0.012 |
ait | -0.011 |
andel | -0.011 |
sie | -0.011 |
Token | sights |
Token ID | 21343 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ocol | +0.035 |
irs | +0.030 |
sie | +0.028 |
year | +0.027 |
andel | +0.027 |
Neg logit contributions | |
owner | -0.014 |
still | -0.012 |
only | -0.012 |
not | -0.012 |
best | -0.012 |
Token | ! |
Token ID | 0 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
ocol | +0.053 |
irs | +0.045 |
ait | +0.042 |
andel | +0.041 |
sie | +0.041 |
Neg logit contributions | |
owner | -0.018 |
not | -0.017 |
still | -0.017 |
was | -0.016 |
hits | -0.015 |
Token | The |
Token ID | 383 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
ocol | +0.020 |
irs | +0.017 |
ait | +0.016 |
andel | +0.016 |
sie | +0.016 |
Neg logit contributions | |
owner | -0.005 |
not | -0.004 |
still | -0.004 |
hurts | -0.004 |
was | -0.004 |
Token | other |
Token ID | 584 |
Feature activation | +5.46e+00 |
Pos logit contributions | |
owner | +0.033 |
not | +0.029 |
still | +0.028 |
was | +0.026 |
hits | +0.026 |
Neg logit contributions | |
ocol | -0.095 |
irs | -0.083 |
ait | -0.076 |
andel | -0.074 |
sie | -0.073 |
Token | passengers |
Token ID | 10405 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
not | +0.122 |
hurts | +0.121 |
owner | +0.118 |
still | +0.116 |
hits | +0.115 |
Neg logit contributions | |
ocol | -0.193 |
irs | -0.156 |
ait | -0.144 |
andel | -0.140 |
ages | -0.137 |
Token | were |
Token ID | 547 |
Feature activation | +9.92e-02 |
Pos logit contributions | |
owner | +0.030 |
hurts | +0.028 |
not | +0.028 |
hits | +0.026 |
tr | +0.025 |
Neg logit contributions | |
ocol | -0.062 |
irs | -0.053 |
ait | -0.050 |
sie | -0.048 |
year | -0.047 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
hurts | +0.003 |
tr | +0.002 |
hits | +0.002 |
owner | +0.002 |
hates | +0.002 |
Neg logit contributions | |
ocol | -0.003 |
irs | -0.002 |
ait | -0.002 |
Annie | -0.002 |
ages | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -5.17e-01 |
Pos logit contributions | |
ocol | +0.001 |
irs | +0.001 |
ait | +0.001 |
Annie | +0.001 |
andel | +0.001 |
Neg logit contributions | |
owner | -0.000 |
hurts | -0.000 |
tr | -0.000 |
hits | -0.000 |
zag | -0.000 |
Token | see |
Token ID | 766 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
ocol | +0.018 |
irs | +0.015 |
ait | +0.013 |
andel | +0.012 |
ages | +0.012 |
Neg logit contributions | |
owner | -0.012 |
hurts | -0.012 |
tr | -0.012 |
hates | -0.011 |
flower | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
burn | +0.017 |
uck | +0.017 |
flower | +0.017 |
owner | +0.017 |
zag | +0.016 |
Neg logit contributions | |
ocol | -0.036 |
rubble | -0.031 |
Annie | -0.031 |
ages | -0.030 |
sie | -0.030 |
Token | all |
Token ID | 477 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
owner | +0.030 |
flower | +0.029 |
hurts | +0.029 |
zag | +0.029 |
Because | +0.028 |
Neg logit contributions | |
ocol | -0.064 |
irs | -0.053 |
ait | -0.049 |
andel | -0.048 |
ages | -0.046 |
Token | alone |
Token ID | 3436 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
owner | +0.012 |
not | +0.012 |
still | +0.011 |
hurts | +0.011 |
hits | +0.011 |
Neg logit contributions | |
ocol | -0.024 |
irs | -0.020 |
ait | -0.019 |
andel | -0.018 |
sie | -0.018 |
Token | because |
Token ID | 780 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
owner | +0.044 |
not | +0.041 |
still | +0.039 |
hurts | +0.038 |
hits | +0.037 |
Neg logit contributions | |
ocol | -0.108 |
irs | -0.092 |
ait | -0.085 |
andel | -0.083 |
sie | -0.081 |
Token | the |
Token ID | 262 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
owner | +0.013 |
not | +0.012 |
still | +0.012 |
hurts | +0.012 |
hits | +0.012 |
Neg logit contributions | |
ocol | -0.023 |
irs | -0.019 |
ait | -0.017 |
andel | -0.017 |
sie | -0.017 |
Token | other |
Token ID | 584 |
Feature activation | +5.43e+00 |
Pos logit contributions | |
owner | +0.039 |
not | +0.036 |
still | +0.034 |
hurts | +0.032 |
was | +0.032 |
Neg logit contributions | |
ocol | -0.075 |
irs | -0.064 |
ait | -0.058 |
andel | -0.056 |
sie | -0.055 |
Token | cars |
Token ID | 5006 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
owner | +0.116 |
not | +0.114 |
hurts | +0.113 |
zag | +0.111 |
still | +0.110 |
Neg logit contributions | |
ocol | -0.198 |
irs | -0.165 |
ait | -0.151 |
andel | -0.145 |
sie | -0.141 |
Token | were |
Token ID | 547 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
owner | +0.057 |
zag | +0.051 |
flower | +0.050 |
hurts | +0.050 |
not | +0.050 |
Neg logit contributions | |
ocol | -0.117 |
irs | -0.101 |
ait | -0.093 |
sie | -0.090 |
rubble | -0.090 |
Token | far |
Token ID | 1290 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
owner | +0.005 |
hurts | +0.005 |
not | +0.005 |
zag | +0.005 |
hits | +0.005 |
Neg logit contributions | |
ocol | -0.013 |
irs | -0.011 |
ait | -0.011 |
andel | -0.010 |
sie | -0.010 |
Token | away |
Token ID | 1497 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
ocol | +0.031 |
irs | +0.027 |
ait | +0.024 |
andel | +0.024 |
sie | +0.023 |
Neg logit contributions | |
owner | -0.014 |
not | -0.013 |
still | -0.013 |
hurts | -0.012 |
hits | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
owner | +0.007 |
not | +0.006 |
still | +0.006 |
hurts | +0.006 |
was | +0.005 |
Neg logit contributions | |
ocol | -0.015 |
irs | -0.013 |
ait | -0.012 |
andel | -0.012 |
sie | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
owner | +0.032 |
hurts | +0.029 |
not | +0.028 |
hits | +0.028 |
still | +0.028 |
Neg logit contributions | |
ocol | -0.086 |
irs | -0.073 |
ait | -0.067 |
andel | -0.066 |
sie | -0.065 |
Token | so |
Token ID | 523 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
owner | +0.008 |
not | +0.007 |
still | +0.007 |
hurts | +0.007 |
hits | +0.007 |
Neg logit contributions | |
ocol | -0.016 |
irs | -0.013 |
ait | -0.012 |
andel | -0.012 |
sie | -0.012 |
Token | it |
Token ID | 340 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
owner | +0.043 |
zag | +0.040 |
hurts | +0.040 |
tr | +0.039 |
flower | +0.039 |
Neg logit contributions | |
ocol | -0.078 |
irs | -0.065 |
ait | -0.058 |
sie | -0.057 |
andel | -0.057 |
Token | made |
Token ID | 925 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
owner | +0.009 |
zag | +0.008 |
flower | +0.008 |
hurts | +0.008 |
not | +0.008 |
Neg logit contributions | |
ocol | -0.024 |
sie | -0.020 |
irs | -0.020 |
ait | -0.019 |
year | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
owner | +0.002 |
not | +0.002 |
still | +0.002 |
was | +0.002 |
only | +0.002 |
Neg logit contributions | |
ocol | -0.005 |
irs | -0.004 |
ait | -0.004 |
sie | -0.004 |
andel | -0.004 |
Token | other |
Token ID | 584 |
Feature activation | +5.42e+00 |
Pos logit contributions | |
not | +0.034 |
owner | +0.033 |
Because | +0.033 |
zag | +0.032 |
still | +0.032 |
Neg logit contributions | |
ocol | -0.084 |
irs | -0.069 |
andel | -0.067 |
ait | -0.066 |
sie | -0.064 |
Token | cat |
Token ID | 3797 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
not | +0.083 |
still | +0.079 |
owner | +0.078 |
zag | +0.074 |
hits | +0.074 |
Neg logit contributions | |
ocol | -0.223 |
irs | -0.184 |
ait | -0.180 |
andel | -0.179 |
sie | -0.167 |
Token | even |
Token ID | 772 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
owner | +0.024 |
not | +0.022 |
still | +0.021 |
hurts | +0.021 |
hits | +0.021 |
Neg logit contributions | |
ocol | -0.059 |
irs | -0.051 |
ait | -0.046 |
andel | -0.045 |
sie | -0.045 |
Token | more |
Token ID | 517 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
ocol | +0.001 |
irs | +0.001 |
sie | +0.001 |
ait | +0.001 |
andel | +0.001 |
Neg logit contributions | |
owner | -0.001 |
zag | -0.001 |
flower | -0.001 |
not | -0.001 |
still | -0.001 |
Token | mad |
Token ID | 8805 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
owner | +0.026 |
not | +0.025 |
still | +0.024 |
hurts | +0.023 |
was | +0.022 |
Neg logit contributions | |
ocol | -0.053 |
irs | -0.046 |
ait | -0.042 |
andel | -0.040 |
sie | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
owner | +0.029 |
not | +0.027 |
still | +0.026 |
zag | +0.026 |
hurts | +0.026 |
Neg logit contributions | |
ocol | -0.051 |
irs | -0.042 |
ait | -0.038 |
andel | -0.038 |
sie | -0.036 |
Token | chase |
Token ID | 15505 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
owner | +0.015 |
not | +0.014 |
still | +0.013 |
hurts | +0.013 |
hits | +0.012 |
Neg logit contributions | |
ocol | -0.041 |
irs | -0.035 |
ait | -0.033 |
andel | -0.032 |
sie | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
owner | +0.004 |
not | +0.004 |
hurts | +0.004 |
was | +0.004 |
hits | +0.003 |
Neg logit contributions | |
ocol | -0.010 |
irs | -0.008 |
ait | -0.008 |
inations | -0.008 |
prise | -0.008 |
Token | chickens |
Token ID | 25972 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
owner | +0.017 |
not | +0.015 |
still | +0.014 |
hurts | +0.014 |
hits | +0.013 |
Neg logit contributions | |
ocol | -0.060 |
irs | -0.053 |
ait | -0.048 |
andel | -0.047 |
sie | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
owner | +0.021 |
not | +0.020 |
still | +0.019 |
hurts | +0.018 |
also | +0.017 |
Neg logit contributions | |
ocol | -0.054 |
irs | -0.046 |
ait | -0.042 |
andel | -0.041 |
sie | -0.040 |
Token | The |
Token ID | 383 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
owner | +0.010 |
not | +0.009 |
still | +0.009 |
hurts | +0.008 |
was | +0.008 |
Neg logit contributions | |
ocol | -0.040 |
irs | -0.035 |
ait | -0.032 |
andel | -0.031 |
sie | -0.031 |
Token | other |
Token ID | 584 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
zag | +0.039 |
flower | +0.036 |
not | +0.035 |
tr | +0.035 |
hurts | +0.035 |
Neg logit contributions | |
ocol | -0.105 |
irs | -0.086 |
ait | -0.085 |
andel | -0.084 |
sie | -0.081 |
Token | chickens |
Token ID | 25972 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
zag | +0.119 |
not | +0.115 |
hits | +0.114 |
hurts | +0.114 |
tr | +0.111 |
Neg logit contributions | |
ocol | -0.177 |
ait | -0.147 |
irs | -0.138 |
andel | -0.135 |
sie | -0.128 |
Token | were |
Token ID | 547 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
owner | +0.031 |
tr | +0.029 |
hurts | +0.029 |
hits | +0.028 |
zag | +0.027 |
Neg logit contributions | |
ocol | -0.073 |
irs | -0.063 |
sie | -0.059 |
ait | -0.058 |
andel | -0.055 |
Token | scared |
Token ID | 12008 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
flower | +0.018 |
tr | +0.018 |
hurts | +0.018 |
Because | +0.018 |
zag | +0.018 |
Neg logit contributions | |
ocol | -0.041 |
irs | -0.034 |
ait | -0.031 |
sie | -0.030 |
andel | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +6.36e-01 |
Pos logit contributions | |
owner | +0.021 |
not | +0.019 |
still | +0.018 |
was | +0.016 |
hurts | +0.015 |
Neg logit contributions | |
ocol | -0.088 |
irs | -0.078 |
ait | -0.071 |
sie | -0.070 |
andel | -0.070 |
Token | did |
Token ID | 750 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
owner | +0.008 |
not | +0.007 |
still | +0.007 |
hurts | +0.006 |
remembered | +0.006 |
Neg logit contributions | |
ocol | -0.031 |
irs | -0.027 |
ait | -0.025 |
andel | -0.025 |
sie | -0.024 |
Token | taller |
Token ID | 25242 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
owner | +0.004 |
zag | +0.003 |
not | +0.003 |
hurts | +0.003 |
Because | +0.003 |
Neg logit contributions | |
ocol | -0.007 |
irs | -0.005 |
andel | -0.005 |
sie | -0.005 |
ait | -0.005 |
Token | than |
Token ID | 621 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
owner | +0.024 |
hurts | +0.022 |
tr | +0.021 |
remembered | +0.021 |
zag | +0.021 |
Neg logit contributions | |
ocol | -0.052 |
irs | -0.041 |
andel | -0.039 |
pounds | -0.039 |
ait | -0.038 |
Token | before |
Token ID | 878 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
ocol | +0.056 |
irs | +0.048 |
ait | +0.045 |
andel | +0.044 |
sie | +0.043 |
Neg logit contributions | |
not | -0.026 |
owner | -0.025 |
still | -0.024 |
only | -0.023 |
was | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
ocol | +0.096 |
irs | +0.082 |
ait | +0.075 |
andel | +0.073 |
sie | +0.072 |
Neg logit contributions | |
owner | -0.045 |
not | -0.042 |
still | -0.040 |
hurts | -0.039 |
hits | -0.038 |
Token | The |
Token ID | 383 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
ocol | +0.025 |
irs | +0.021 |
ait | +0.020 |
sie | +0.019 |
andel | +0.019 |
Neg logit contributions | |
owner | -0.005 |
not | -0.004 |
still | -0.004 |
was | -0.004 |
only | -0.004 |
Token | other |
Token ID | 584 |
Feature activation | +5.36e+00 |
Pos logit contributions | |
owner | +0.050 |
not | +0.048 |
zag | +0.047 |
hurts | +0.046 |
hits | +0.046 |
Neg logit contributions | |
ocol | -0.080 |
irs | -0.065 |
andel | -0.060 |
ait | -0.059 |
sie | -0.057 |
Token | trees |
Token ID | 7150 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
owner | +0.116 |
not | +0.115 |
zag | +0.111 |
hurts | +0.110 |
hits | +0.110 |
Neg logit contributions | |
ocol | -0.190 |
irs | -0.153 |
ait | -0.145 |
andel | -0.143 |
sie | -0.136 |
Token | in |
Token ID | 287 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
owner | +0.019 |
hurts | +0.018 |
tr | +0.017 |
hits | +0.017 |
zag | +0.017 |
Neg logit contributions | |
ocol | -0.039 |
irs | -0.033 |
sie | -0.031 |
ait | -0.030 |
andel | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
owner | +0.027 |
not | +0.025 |
still | +0.024 |
was | +0.023 |
only | +0.023 |
Neg logit contributions | |
ocol | -0.072 |
irs | -0.062 |
ait | -0.058 |
andel | -0.055 |
sie | -0.055 |
Token | forest |
Token ID | 8222 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
owner | +0.011 |
not | +0.011 |
hurts | +0.010 |
zag | +0.010 |
hits | +0.010 |
Neg logit contributions | |
ocol | -0.020 |
irs | -0.017 |
andel | -0.015 |
ait | -0.015 |
sie | -0.015 |
Token | were |
Token ID | 547 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
owner | +0.017 |
hurts | +0.016 |
burn | +0.015 |
tr | +0.015 |
hits | +0.015 |
Neg logit contributions | |
ocol | -0.050 |
irs | -0.043 |
ages | -0.040 |
ait | -0.040 |
sie | -0.040 |
Token | was |
Token ID | 373 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
owner | +0.018 |
ich | +0.018 |
flower | +0.017 |
uck | +0.017 |
burn | +0.017 |
Neg logit contributions | |
ages | -0.031 |
Annie | -0.031 |
ocol | -0.030 |
ts | -0.030 |
irs | -0.030 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
not | +0.017 |
owner | +0.017 |
still | +0.016 |
only | +0.014 |
also | +0.014 |
Neg logit contributions | |
ocol | -0.038 |
irs | -0.033 |
ait | -0.031 |
andel | -0.030 |
sie | -0.030 |
Token | to |
Token ID | 284 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
owner | +0.023 |
not | +0.021 |
still | +0.020 |
hurts | +0.020 |
flower | +0.019 |
Neg logit contributions | |
ocol | -0.046 |
irs | -0.039 |
ait | -0.035 |
andel | -0.034 |
sie | -0.034 |
Token | make |
Token ID | 787 |
Feature activation | +6.24e-01 |
Pos logit contributions | |
flower | +0.001 |
owner | +0.001 |
hurts | +0.001 |
remembered | +0.001 |
Because | +0.001 |
Neg logit contributions | |
ocol | -0.002 |
irs | -0.001 |
ait | -0.001 |
rubble | -0.001 |
ages | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
owner | +0.011 |
not | +0.011 |
still | +0.010 |
was | +0.010 |
hurts | +0.009 |
Neg logit contributions | |
ocol | -0.028 |
irs | -0.023 |
ait | -0.022 |
sie | -0.022 |
andel | -0.021 |
Token | other |
Token ID | 584 |
Feature activation | +5.35e+00 |
Pos logit contributions | |
owner | +0.051 |
not | +0.048 |
still | +0.046 |
hurts | +0.045 |
hits | +0.044 |
Neg logit contributions | |
ocol | -0.088 |
irs | -0.074 |
ait | -0.067 |
andel | -0.065 |
sie | -0.064 |
Token | kids |
Token ID | 3988 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
owner | +0.120 |
not | +0.110 |
still | +0.107 |
hurts | +0.105 |
was | +0.103 |
Neg logit contributions | |
ocol | -0.210 |
irs | -0.178 |
ait | -0.162 |
andel | -0.156 |
sie | -0.152 |
Token | afraid |
Token ID | 7787 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
owner | +0.034 |
not | +0.033 |
still | +0.031 |
hurts | +0.030 |
hits | +0.029 |
Neg logit contributions | |
ocol | -0.083 |
irs | -0.071 |
ait | -0.065 |
andel | -0.064 |
sie | -0.062 |
Token | or |
Token ID | 393 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
owner | +0.054 |
not | +0.050 |
still | +0.047 |
was | +0.046 |
hurts | +0.045 |
Neg logit contributions | |
ocol | -0.125 |
irs | -0.107 |
ait | -0.098 |
andel | -0.096 |
sie | -0.096 |
Token | mad |
Token ID | 8805 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
owner | +0.010 |
zag | +0.009 |
flower | +0.009 |
still | +0.008 |
hurts | +0.008 |
Neg logit contributions | |
ocol | -0.024 |
irs | -0.020 |
ait | -0.019 |
sie | -0.018 |
andel | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
owner | +0.041 |
not | +0.039 |
zag | +0.038 |
still | +0.038 |
flower | +0.038 |
Neg logit contributions | |
ocol | -0.069 |
irs | -0.058 |
ait | -0.052 |
andel | -0.051 |
sie | -0.049 |
Token | you |
Token ID | 345 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
ocol | +0.004 |
ait | +0.003 |
irs | +0.003 |
uster | +0.003 |
sie | +0.003 |
Neg logit contributions | |
not | -0.002 |
was | -0.001 |
owner | -0.001 |
still | -0.001 |
hurts | -0.001 |
Token | very |
Token ID | 845 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
ocol | +0.002 |
irs | +0.001 |
ait | +0.001 |
andel | +0.001 |
sie | +0.001 |
Neg logit contributions | |
owner | -0.001 |
not | -0.001 |
still | -0.001 |
hurts | -0.001 |
hits | -0.001 |
Token | sick |
Token ID | 6639 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
owner | +0.033 |
not | +0.030 |
still | +0.029 |
hurts | +0.028 |
flower | +0.028 |
Neg logit contributions | |
ocol | -0.084 |
irs | -0.071 |
ait | -0.065 |
andel | -0.064 |
sie | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
owner | +0.030 |
not | +0.029 |
still | +0.025 |
was | +0.025 |
hurts | +0.024 |
Neg logit contributions | |
ocol | -0.099 |
irs | -0.088 |
ait | -0.082 |
andel | -0.079 |
sie | -0.078 |
Token | The |
Token ID | 383 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
owner | +0.010 |
not | +0.009 |
still | +0.008 |
was | +0.008 |
also | +0.007 |
Neg logit contributions | |
ocol | -0.057 |
irs | -0.050 |
ait | -0.047 |
andel | -0.047 |
sie | -0.047 |
Token | other |
Token ID | 584 |
Feature activation | +5.34e+00 |
Pos logit contributions | |
owner | +0.057 |
not | +0.054 |
still | +0.052 |
hurts | +0.052 |
hits | +0.051 |
Neg logit contributions | |
ocol | -0.078 |
irs | -0.063 |
ait | -0.056 |
andel | -0.055 |
sie | -0.054 |
Token | mushrooms |
Token ID | 23452 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
owner | +0.128 |
not | +0.123 |
still | +0.119 |
hurts | +0.117 |
hits | +0.116 |
Neg logit contributions | |
ocol | -0.191 |
irs | -0.157 |
ait | -0.140 |
andel | -0.138 |
sie | -0.136 |
Token | were |
Token ID | 547 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
owner | +0.000 |
not | +0.000 |
zag | +0.000 |
hurts | +0.000 |
hits | +0.000 |
Neg logit contributions | |
ocol | -0.001 |
irs | -0.001 |
ait | -0.001 |
sie | -0.001 |
andel | -0.001 |
Token | not |
Token ID | 407 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
owner | +0.005 |
not | +0.005 |
still | +0.004 |
hurts | +0.004 |
hits | +0.004 |
Neg logit contributions | |
ocol | -0.016 |
irs | -0.014 |
ait | -0.013 |
andel | -0.012 |
sie | -0.012 |
Token | dangerous |
Token ID | 4923 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
not | +0.012 |
owner | +0.012 |
still | +0.011 |
also | +0.010 |
only | +0.010 |
Neg logit contributions | |
ocol | -0.026 |
irs | -0.023 |
ait | -0.021 |
andel | -0.021 |
sie | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
owner | +0.018 |
not | +0.018 |
hurts | +0.016 |
still | +0.016 |
was | +0.016 |
Neg logit contributions | |
ocol | -0.058 |
irs | -0.051 |
ait | -0.047 |
andel | -0.046 |
sie | -0.045 |
Token | train |
Token ID | 4512 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
hurts | +0.040 |
hits | +0.040 |
not | +0.038 |
remembered | +0.037 |
neither | +0.036 |
Neg logit contributions | |
ocol | -0.074 |
ait | -0.064 |
irs | -0.061 |
andel | -0.061 |
year | -0.060 |
Token | was |
Token ID | 373 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
hurts | +0.017 |
owner | +0.016 |
not | +0.015 |
still | +0.015 |
hits | +0.015 |
Neg logit contributions | |
ocol | -0.059 |
irs | -0.052 |
ait | -0.050 |
ages | -0.048 |
Eve | -0.048 |
Token | red |
Token ID | 2266 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
owner | +0.032 |
hurts | +0.031 |
not | +0.030 |
still | +0.030 |
hits | +0.029 |
Neg logit contributions | |
ocol | -0.064 |
irs | -0.053 |
ait | -0.050 |
andel | -0.048 |
sie | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
owner | +0.019 |
not | +0.019 |
remembered | +0.018 |
still | +0.018 |
hurts | +0.017 |
Neg logit contributions | |
ocol | -0.075 |
irs | -0.064 |
ait | -0.061 |
andel | -0.060 |
sie | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
ocol | +0.026 |
irs | +0.022 |
ait | +0.020 |
andel | +0.020 |
sie | +0.020 |
Neg logit contributions | |
owner | -0.011 |
not | -0.010 |
still | -0.010 |
hurts | -0.010 |
hits | -0.010 |
Token | other |
Token ID | 584 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
owner | +0.037 |
not | +0.035 |
still | +0.033 |
hurts | +0.032 |
hits | +0.031 |
Neg logit contributions | |
ocol | -0.076 |
irs | -0.064 |
ait | -0.059 |
andel | -0.057 |
sie | -0.056 |
Token | train |
Token ID | 4512 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
owner | +0.108 |
not | +0.107 |
hurts | +0.104 |
still | +0.101 |
hits | +0.100 |
Neg logit contributions | |
ocol | -0.196 |
irs | -0.167 |
ait | -0.155 |
andel | -0.149 |
ages | -0.143 |
Token | was |
Token ID | 373 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
owner | +0.011 |
hurts | +0.010 |
not | +0.010 |
hits | +0.010 |
flower | +0.009 |
Neg logit contributions | |
ocol | -0.058 |
irs | -0.051 |
ait | -0.048 |
sie | -0.047 |
ages | -0.046 |
Token | blue |
Token ID | 4171 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
owner | +0.025 |
hurts | +0.024 |
hits | +0.023 |
not | +0.023 |
flower | +0.022 |
Neg logit contributions | |
ocol | -0.048 |
irs | -0.040 |
ait | -0.037 |
andel | -0.035 |
sie | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
owner | +0.030 |
not | +0.029 |
remembered | +0.029 |
hurts | +0.028 |
hits | +0.028 |
Neg logit contributions | |
ocol | -0.057 |
irs | -0.047 |
ait | -0.044 |
andel | -0.043 |
sie | -0.041 |
Token | They |
Token ID | 1119 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
owner | +0.008 |
not | +0.007 |
still | +0.007 |
was | +0.006 |
only | +0.006 |
Neg logit contributions | |
ocol | -0.035 |
irs | -0.030 |
sie | -0.028 |
ait | -0.028 |
andel | -0.028 |
Token | 't |
Token ID | 470 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ocol | +0.040 |
irs | +0.034 |
ait | +0.031 |
andel | +0.030 |
sie | +0.029 |
Neg logit contributions | |
owner | -0.018 |
not | -0.015 |
flower | -0.015 |
remembered | -0.015 |
still | -0.015 |
Token | find |
Token ID | 1064 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
owner | +0.016 |
not | +0.015 |
still | +0.015 |
also | +0.013 |
was | +0.013 |
Neg logit contributions | |
ocol | -0.047 |
irs | -0.040 |
ait | -0.037 |
andel | -0.037 |
sie | -0.035 |
Token | her |
Token ID | 607 |
Feature activation | +8.15e-01 |
Pos logit contributions | |
ocol | +0.014 |
irs | +0.012 |
ait | +0.011 |
andel | +0.011 |
inations | +0.011 |
Neg logit contributions | |
owner | -0.006 |
not | -0.006 |
still | -0.006 |
only | -0.006 |
was | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
owner | +0.016 |
not | +0.015 |
still | +0.014 |
hurts | +0.014 |
hits | +0.014 |
Neg logit contributions | |
ocol | -0.033 |
irs | -0.028 |
ait | -0.026 |
andel | -0.025 |
sie | -0.025 |
Token | The |
Token ID | 383 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
owner | +0.010 |
not | +0.009 |
still | +0.008 |
was | +0.008 |
only | +0.007 |
Neg logit contributions | |
ocol | -0.040 |
irs | -0.035 |
ait | -0.032 |
andel | -0.031 |
sie | -0.031 |
Token | other |
Token ID | 584 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
owner | +0.041 |
not | +0.039 |
still | +0.037 |
hurts | +0.037 |
hits | +0.036 |
Neg logit contributions | |
ocol | -0.075 |
irs | -0.063 |
ait | -0.057 |
andel | -0.056 |
sie | -0.055 |
Token | sheep |
Token ID | 15900 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
hurts | +0.142 |
hits | +0.141 |
not | +0.140 |
zag | +0.138 |
owner | +0.134 |
Neg logit contributions | |
ocol | -0.163 |
ait | -0.126 |
irs | -0.125 |
andel | -0.117 |
sie | -0.117 |
Token | were |
Token ID | 547 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
owner | +0.016 |
hits | +0.014 |
hurts | +0.014 |
zag | +0.014 |
not | +0.014 |
Neg logit contributions | |
ocol | -0.050 |
irs | -0.044 |
sie | -0.041 |
ait | -0.041 |
rubble | -0.039 |
Token | sad |
Token ID | 6507 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
owner | +0.009 |
not | +0.009 |
still | +0.008 |
hurts | +0.008 |
also | +0.008 |
Neg logit contributions | |
ocol | -0.025 |
irs | -0.021 |
ait | -0.020 |
andel | -0.019 |
sie | -0.019 |
Token | because |
Token ID | 780 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
owner | +0.038 |
not | +0.032 |
zag | +0.032 |
hurts | +0.031 |
hits | +0.031 |
Neg logit contributions | |
ocol | -0.120 |
irs | -0.103 |
ait | -0.097 |
andel | -0.095 |
sie | -0.093 |
Token | their |
Token ID | 511 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
owner | +0.004 |
not | +0.004 |
still | +0.003 |
was | +0.003 |
also | +0.003 |
Neg logit contributions | |
ocol | -0.010 |
irs | -0.009 |
ait | -0.008 |
andel | -0.008 |
sie | -0.008 |
Token | place |
Token ID | 1295 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
owner | +0.012 |
not | +0.010 |
still | +0.010 |
hurts | +0.010 |
hits | +0.010 |
Neg logit contributions | |
ocol | -0.028 |
irs | -0.024 |
ait | -0.022 |
andel | -0.021 |
sie | -0.021 |
Token | all |
Token ID | 477 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
owner | +0.009 |
not | +0.009 |
still | +0.008 |
was | +0.007 |
only | +0.007 |
Neg logit contributions | |
ocol | -0.022 |
irs | -0.019 |
andel | -0.017 |
ait | -0.017 |
sie | -0.016 |
Token | day |
Token ID | 1110 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
ocol | +0.014 |
irs | +0.012 |
ait | +0.011 |
sie | +0.011 |
andel | +0.011 |
Neg logit contributions | |
owner | -0.006 |
not | -0.005 |
hurts | -0.005 |
still | -0.005 |
zag | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
ocol | +0.097 |
irs | +0.084 |
ait | +0.076 |
andel | +0.074 |
prise | +0.074 |
Neg logit contributions | |
not | -0.032 |
owner | -0.029 |
still | -0.027 |
only | -0.023 |
next | -0.022 |
Token | The |
Token ID | 383 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
owner | +0.007 |
not | +0.007 |
still | +0.006 |
was | +0.006 |
only | +0.005 |
Neg logit contributions | |
ocol | -0.029 |
irs | -0.025 |
ait | -0.023 |
andel | -0.023 |
sie | -0.023 |
Token | other |
Token ID | 584 |
Feature activation | +5.28e+00 |
Pos logit contributions | |
owner | +0.059 |
not | +0.056 |
still | +0.054 |
hurts | +0.053 |
hits | +0.052 |
Neg logit contributions | |
ocol | -0.086 |
irs | -0.070 |
ait | -0.063 |
andel | -0.061 |
sie | -0.060 |
Token | cube |
Token ID | 23441 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
owner | +0.174 |
not | +0.171 |
hurts | +0.166 |
still | +0.166 |
zag | +0.164 |
Neg logit contributions | |
ocol | -0.139 |
irs | -0.104 |
ait | -0.088 |
andel | -0.085 |
sie | -0.081 |
Token | was |
Token ID | 373 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
owner | +0.023 |
burn | +0.023 |
flower | +0.021 |
Fighting | +0.021 |
zag | +0.020 |
Neg logit contributions | |
ocol | -0.078 |
sie | -0.067 |
inations | -0.066 |
ait | -0.065 |
irs | -0.065 |
Token | much |
Token ID | 881 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
hurts | +0.031 |
owner | +0.031 |
zag | +0.030 |
flower | +0.030 |
hits | +0.030 |
Neg logit contributions | |
ocol | -0.058 |
irs | -0.047 |
ait | -0.044 |
andel | -0.042 |
sie | -0.041 |
Token | more |
Token ID | 517 |
Feature activation | +9.57e-01 |
Pos logit contributions | |
ocol | +0.029 |
irs | +0.024 |
ait | +0.022 |
andel | +0.021 |
sie | +0.021 |
Neg logit contributions | |
owner | -0.016 |
not | -0.015 |
zag | -0.015 |
hurts | -0.015 |
still | -0.014 |
Token | adventurous |
Token ID | 39770 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
owner | +0.022 |
flower | +0.021 |
hurts | +0.021 |
zag | +0.021 |
not | +0.021 |
Neg logit contributions | |
ocol | -0.034 |
irs | -0.028 |
ait | -0.025 |
andel | -0.025 |
inations | -0.024 |
Token | looked |
Token ID | 3114 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
ocol | +0.001 |
irs | +0.001 |
ait | +0.001 |
sie | +0.001 |
year | +0.001 |
Neg logit contributions | |
hurts | -0.000 |
owner | -0.000 |
hits | -0.000 |
not | -0.000 |
remembered | -0.000 |
Token | very |
Token ID | 845 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
ocol | +0.125 |
irs | +0.115 |
ait | +0.107 |
andel | +0.101 |
sie | +0.098 |
Neg logit contributions | |
still | -0.044 |
not | -0.043 |
owner | -0.038 |
was | -0.038 |
only | -0.037 |
Token | useful |
Token ID | 4465 |
Feature activation | -9.88e-01 |
Pos logit contributions | |
ocol | +0.083 |
irs | +0.074 |
ait | +0.067 |
andel | +0.067 |
uster | +0.065 |
Neg logit contributions | |
still | -0.038 |
not | -0.037 |
owner | -0.034 |
also | -0.033 |
only | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
ocol | +0.046 |
irs | +0.040 |
andel | +0.037 |
ait | +0.036 |
sie | +0.035 |
Neg logit contributions | |
owner | -0.015 |
not | -0.015 |
still | -0.014 |
was | -0.013 |
only | -0.012 |
Token | The |
Token ID | 383 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
ocol | +0.013 |
irs | +0.011 |
ait | +0.010 |
andel | +0.010 |
sie | +0.010 |
Neg logit contributions | |
owner | -0.004 |
not | -0.003 |
still | -0.003 |
was | -0.003 |
only | -0.003 |
Token | other |
Token ID | 584 |
Feature activation | +5.25e+00 |
Pos logit contributions | |
owner | +0.034 |
not | +0.034 |
hurts | +0.033 |
still | +0.032 |
hits | +0.032 |
Neg logit contributions | |
ocol | -0.058 |
irs | -0.048 |
ait | -0.044 |
andel | -0.043 |
sie | -0.042 |
Token | agreed |
Token ID | 4987 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
not | +0.109 |
hurts | +0.108 |
zag | +0.104 |
owner | +0.103 |
hits | +0.103 |
Neg logit contributions | |
ocol | -0.193 |
irs | -0.156 |
ait | -0.153 |
andel | -0.143 |
sie | -0.140 |
Token | and |
Token ID | 290 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
owner | +0.001 |
not | +0.001 |
still | +0.001 |
was | +0.001 |
hurts | +0.001 |
Neg logit contributions | |
ocol | -0.005 |
irs | -0.004 |
ait | -0.004 |
andel | -0.004 |
sie | -0.004 |
Token | said |
Token ID | 531 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
owner | +0.002 |
not | +0.002 |
still | +0.002 |
hurts | +0.002 |
hits | +0.002 |
Neg logit contributions | |
ocol | -0.005 |
irs | -0.005 |
ait | -0.004 |
andel | -0.004 |
sie | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
ocol | +0.065 |
irs | +0.058 |
ait | +0.054 |
andel | +0.053 |
sie | +0.051 |
Neg logit contributions | |
owner | -0.015 |
not | -0.014 |
still | -0.012 |
was | -0.011 |
hurts | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
owner | +0.003 |
hits | +0.003 |
hurts | +0.003 |
flower | +0.002 |
not | +0.002 |
Neg logit contributions | |
ocol | -0.023 |
irs | -0.020 |
ait | -0.019 |
sie | -0.019 |
Annie | -0.019 |
Token | we |
Token ID | 356 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
owner | +0.014 |
not | +0.014 |
still | +0.013 |
hurts | +0.013 |
also | +0.012 |
Neg logit contributions | |
ocol | -0.019 |
irs | -0.015 |
ait | -0.014 |
andel | -0.013 |
year | -0.013 |
Token | stay |
Token ID | 2652 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
ocol | +0.056 |
irs | +0.047 |
andel | +0.045 |
year | +0.043 |
ait | +0.043 |
Neg logit contributions | |
still | -0.018 |
owner | -0.018 |
not | -0.018 |
only | -0.016 |
also | -0.016 |
Token | here |
Token ID | 994 |
Feature activation | -7.86e-01 |
Pos logit contributions | |
owner | +0.005 |
not | +0.004 |
still | +0.004 |
hurts | +0.004 |
remembered | +0.004 |
Neg logit contributions | |
ocol | -0.009 |
irs | -0.008 |
ait | -0.007 |
andel | -0.007 |
sie | -0.007 |
Token | ?" |
Token ID | 1701 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
ocol | +0.034 |
irs | +0.029 |
ait | +0.027 |
andel | +0.026 |
sie | +0.026 |
Neg logit contributions | |
owner | -0.013 |
not | -0.012 |
hurts | -0.011 |
hits | -0.011 |
still | -0.011 |
Token | The |
Token ID | 383 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
owner | +0.008 |
not | +0.008 |
hurts | +0.007 |
still | +0.007 |
hits | +0.007 |
Neg logit contributions | |
ocol | -0.028 |
irs | -0.024 |
ait | -0.022 |
andel | -0.022 |
sie | -0.022 |
Token | other |
Token ID | 584 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
owner | +0.033 |
not | +0.032 |
still | +0.031 |
hurts | +0.030 |
remembered | +0.030 |
Neg logit contributions | |
ocol | -0.075 |
irs | -0.063 |
ait | -0.058 |
andel | -0.058 |
sie | -0.056 |
Token | cat |
Token ID | 3797 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
owner | +0.122 |
not | +0.122 |
still | +0.118 |
hurts | +0.117 |
hits | +0.115 |
Neg logit contributions | |
ocol | -0.173 |
irs | -0.140 |
ait | -0.133 |
andel | -0.128 |
sie | -0.124 |
Token | said |
Token ID | 531 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
owner | +0.007 |
hits | +0.006 |
zag | +0.006 |
hurts | +0.006 |
not | +0.006 |
Neg logit contributions | |
ocol | -0.017 |
irs | -0.014 |
sie | -0.014 |
Annie | -0.014 |
ait | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
ocol | +0.072 |
irs | +0.061 |
ait | +0.056 |
sie | +0.055 |
andel | +0.055 |
Neg logit contributions | |
owner | -0.023 |
hurts | -0.020 |
hits | -0.020 |
not | -0.020 |
zag | -0.020 |
Token | " |
Token ID | 366 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
ocol | +0.012 |
irs | +0.010 |
ait | +0.009 |
andel | +0.009 |
sie | +0.009 |
Neg logit contributions | |
owner | -0.004 |
hurts | -0.004 |
not | -0.004 |
hits | -0.004 |
still | -0.004 |
Token | Maybe |
Token ID | 13300 |
Feature activation | -9.90e-01 |
Pos logit contributions | |
hits | +0.038 |
remembered | +0.035 |
head | +0.035 |
hurts | +0.034 |
owner | +0.032 |
Neg logit contributions | |
ocol | -0.065 |
irs | -0.057 |
ait | -0.056 |
chie | -0.055 |
sie | -0.055 |
Token | was |
Token ID | 373 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
owner | +0.037 |
not | +0.036 |
hurts | +0.036 |
still | +0.035 |
zag | +0.034 |
Neg logit contributions | |
ocol | -0.093 |
irs | -0.080 |
ait | -0.073 |
sie | -0.072 |
ages | -0.071 |
Token | very |
Token ID | 845 |
Feature activation | +8.17e-01 |
Pos logit contributions | |
hurts | +0.037 |
owner | +0.037 |
flower | +0.036 |
remembered | +0.034 |
zag | +0.034 |
Neg logit contributions | |
ocol | -0.068 |
irs | -0.058 |
ait | -0.053 |
andel | -0.051 |
sie | -0.050 |
Token | wealthy |
Token ID | 11574 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
owner | +0.016 |
not | +0.015 |
still | +0.015 |
hurts | +0.014 |
hits | +0.014 |
Neg logit contributions | |
ocol | -0.033 |
irs | -0.028 |
ait | -0.026 |
andel | -0.025 |
sie | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
owner | +0.015 |
not | +0.013 |
still | +0.012 |
was | +0.011 |
hurts | +0.011 |
Neg logit contributions | |
ocol | -0.053 |
irs | -0.046 |
ait | -0.042 |
andel | -0.042 |
sie | -0.041 |
Token | The |
Token ID | 383 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
owner | +0.004 |
not | +0.003 |
still | +0.003 |
was | +0.003 |
hits | +0.003 |
Neg logit contributions | |
ocol | -0.012 |
irs | -0.010 |
ait | -0.010 |
andel | -0.010 |
sie | -0.009 |
Token | other |
Token ID | 584 |
Feature activation | +5.20e+00 |
Pos logit contributions | |
owner | +0.051 |
not | +0.049 |
hurts | +0.048 |
zag | +0.048 |
flower | +0.048 |
Neg logit contributions | |
ocol | -0.083 |
irs | -0.070 |
ait | -0.063 |
andel | -0.062 |
sie | -0.061 |
Token | team |
Token ID | 1074 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
not | +0.119 |
owner | +0.118 |
hurts | +0.117 |
zag | +0.115 |
tr | +0.114 |
Neg logit contributions | |
ocol | -0.169 |
irs | -0.140 |
ait | -0.127 |
ages | -0.124 |
sie | -0.123 |
Token | had |
Token ID | 550 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
owner | +0.029 |
hurts | +0.028 |
not | +0.027 |
zag | +0.026 |
flower | +0.025 |
Neg logit contributions | |
ocol | -0.119 |
irs | -0.105 |
sie | -0.098 |
ait | -0.098 |
ages | -0.096 |
Token | hardly |
Token ID | 8941 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
owner | +0.022 |
not | +0.020 |
still | +0.020 |
hurts | +0.019 |
hits | +0.018 |
Neg logit contributions | |
ocol | -0.055 |
irs | -0.047 |
ait | -0.043 |
andel | -0.042 |
sie | -0.041 |
Token | anything |
Token ID | 1997 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
ocol | +0.007 |
irs | +0.006 |
ait | +0.006 |
andel | +0.006 |
sie | +0.006 |
Neg logit contributions | |
owner | -0.003 |
not | -0.003 |
still | -0.002 |
hurts | -0.002 |
hits | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
ocol | +0.068 |
irs | +0.058 |
ait | +0.054 |
andel | +0.053 |
sie | +0.051 |
Neg logit contributions | |
owner | -0.028 |
not | -0.026 |
still | -0.025 |
hurts | -0.024 |
was | -0.023 |
Token | felt |
Token ID | 2936 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
owner | +0.013 |
flower | +0.012 |
hurts | +0.012 |
hits | +0.012 |
tr | +0.011 |
Neg logit contributions | |
ocol | -0.047 |
irs | -0.042 |
ait | -0.041 |
Annie | -0.040 |
sie | -0.040 |
Token | en |
Token ID | 551 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
ocol | +0.018 |
irs | +0.015 |
ait | +0.014 |
andel | +0.014 |
sie | +0.013 |
Neg logit contributions | |
owner | -0.009 |
zag | -0.008 |
flower | -0.008 |
not | -0.008 |
hurts | -0.008 |
Token | vious |
Token ID | 1442 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
ocol | +0.018 |
irs | +0.014 |
andel | +0.013 |
ait | +0.013 |
sie | +0.012 |
Neg logit contributions | |
not | -0.013 |
owner | -0.013 |
zag | -0.013 |
flower | -0.012 |
tr | -0.012 |
Token | because |
Token ID | 780 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
owner | +0.050 |
not | +0.047 |
flower | +0.045 |
hurts | +0.045 |
zag | +0.045 |
Neg logit contributions | |
ocol | -0.092 |
irs | -0.077 |
ait | -0.071 |
andel | -0.069 |
sie | -0.066 |
Token | the |
Token ID | 262 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
owner | +0.022 |
not | +0.020 |
still | +0.019 |
hurts | +0.018 |
hits | +0.018 |
Neg logit contributions | |
ocol | -0.054 |
irs | -0.047 |
ait | -0.043 |
andel | -0.042 |
sie | -0.041 |
Token | other |
Token ID | 584 |
Feature activation | +5.19e+00 |
Pos logit contributions | |
owner | +0.046 |
not | +0.042 |
still | +0.039 |
hurts | +0.038 |
was | +0.037 |
Neg logit contributions | |
ocol | -0.097 |
irs | -0.084 |
ait | -0.075 |
sie | -0.074 |
andel | -0.073 |
Token | dogs |
Token ID | 6844 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
owner | +0.111 |
not | +0.107 |
still | +0.105 |
hurts | +0.103 |
hits | +0.102 |
Neg logit contributions | |
ocol | -0.194 |
irs | -0.161 |
ait | -0.151 |
andel | -0.143 |
sie | -0.137 |
Token | had |
Token ID | 550 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
owner | +0.019 |
not | +0.017 |
hits | +0.016 |
hurts | +0.016 |
zag | +0.016 |
Neg logit contributions | |
ocol | -0.043 |
irs | -0.037 |
ait | -0.033 |
sie | -0.032 |
andel | -0.032 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
owner | +0.013 |
not | +0.012 |
still | +0.011 |
hurts | +0.011 |
hits | +0.010 |
Neg logit contributions | |
ocol | -0.030 |
irs | -0.026 |
ait | -0.024 |
andel | -0.023 |
sie | -0.023 |
Token | toys |
Token ID | 14958 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
owner | +0.018 |
not | +0.018 |
was | +0.016 |
still | +0.016 |
hurts | +0.015 |
Neg logit contributions | |
ocol | -0.055 |
irs | -0.047 |
ait | -0.045 |
andel | -0.044 |
sie | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
owner | +0.013 |
not | +0.012 |
still | +0.012 |
hurts | +0.011 |
hits | +0.011 |
Neg logit contributions | |
ocol | -0.035 |
irs | -0.030 |
ait | -0.028 |
andel | -0.027 |
sie | -0.027 |
Token | car |
Token ID | 1097 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
flower | +0.042 |
hurts | +0.041 |
burn | +0.039 |
not | +0.039 |
remembered | +0.038 |
Neg logit contributions | |
ocol | -0.058 |
ait | -0.049 |
ages | -0.049 |
year | -0.048 |
andel | -0.047 |
Token | was |
Token ID | 373 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
flower | +0.039 |
burn | +0.039 |
hurts | +0.035 |
owner | +0.032 |
still | +0.031 |
Neg logit contributions | |
ocol | -0.106 |
ages | -0.093 |
irs | -0.092 |
Eve | -0.090 |
ait | -0.089 |
Token | red |
Token ID | 2266 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
owner | +0.030 |
hurts | +0.029 |
flower | +0.028 |
not | +0.027 |
still | +0.027 |
Neg logit contributions | |
ocol | -0.055 |
irs | -0.046 |
ait | -0.043 |
andel | -0.041 |
ages | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
remembered | +0.019 |
not | +0.017 |
still | +0.016 |
missed | +0.016 |
owner | +0.016 |
Neg logit contributions | |
ocol | -0.051 |
andel | -0.042 |
irs | -0.041 |
ait | -0.041 |
uster | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
ocol | +0.023 |
irs | +0.020 |
ait | +0.018 |
sie | +0.017 |
andel | +0.017 |
Neg logit contributions | |
owner | -0.011 |
hurts | -0.010 |
not | -0.010 |
still | -0.010 |
flower | -0.010 |
Token | other |
Token ID | 584 |
Feature activation | +5.17e+00 |
Pos logit contributions | |
owner | +0.032 |
not | +0.030 |
still | +0.029 |
hurts | +0.028 |
hits | +0.027 |
Neg logit contributions | |
ocol | -0.075 |
irs | -0.064 |
ait | -0.059 |
andel | -0.057 |
sie | -0.056 |
Token | car |
Token ID | 1097 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
not | +0.062 |
owner | +0.062 |
hurts | +0.059 |
still | +0.056 |
flower | +0.055 |
Neg logit contributions | |
ocol | -0.230 |
irs | -0.200 |
ait | -0.189 |
andel | -0.181 |
ages | -0.180 |
Token | was |
Token ID | 373 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
flower | +0.029 |
burn | +0.026 |
owner | +0.025 |
hurts | +0.024 |
not | +0.021 |
Neg logit contributions | |
ocol | -0.110 |
irs | -0.097 |
ages | -0.094 |
ait | -0.093 |
inations | -0.090 |
Token | blue |
Token ID | 4171 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
owner | +0.029 |
hurts | +0.029 |
flower | +0.028 |
hits | +0.027 |
not | +0.026 |
Neg logit contributions | |
ocol | -0.053 |
irs | -0.045 |
ait | -0.041 |
andel | -0.039 |
ages | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
remembered | +0.031 |
not | +0.029 |
owner | +0.029 |
hurts | +0.029 |
hits | +0.029 |
Neg logit contributions | |
ocol | -0.048 |
ait | -0.038 |
irs | -0.038 |
andel | -0.037 |
Annie | -0.034 |
Token | They |
Token ID | 1119 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
owner | +0.007 |
not | +0.006 |
still | +0.006 |
was | +0.005 |
only | +0.005 |
Neg logit contributions | |
ocol | -0.032 |
irs | -0.028 |
ait | -0.025 |
sie | -0.025 |
prise | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
owner | +0.006 |
not | +0.006 |
still | +0.005 |
only | +0.005 |
also | +0.005 |
Neg logit contributions | |
ocol | -0.017 |
irs | -0.015 |
ait | -0.014 |
inations | -0.014 |
sie | -0.014 |
Token | apples |
Token ID | 22514 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
ocol | +0.029 |
irs | +0.024 |
ait | +0.022 |
andel | +0.021 |
sie | +0.021 |
Neg logit contributions | |
owner | -0.018 |
not | -0.017 |
hurts | -0.016 |
zag | -0.016 |
still | -0.016 |
Token | easily |
Token ID | 3538 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ocol | +0.007 |
irs | +0.006 |
ait | +0.005 |
andel | +0.005 |
sie | +0.005 |
Neg logit contributions | |
owner | -0.004 |
hurts | -0.004 |
remembered | -0.003 |
hits | -0.003 |
hates | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +7.58e-01 |
Pos logit contributions | |
ocol | +0.044 |
irs | +0.038 |
ait | +0.034 |
andel | +0.033 |
sie | +0.033 |
Neg logit contributions | |
owner | -0.022 |
not | -0.020 |
hurts | -0.020 |
still | -0.019 |
hits | -0.019 |
Token | The |
Token ID | 383 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
owner | +0.013 |
hurts | +0.013 |
not | +0.013 |
remembered | +0.012 |
flower | +0.012 |
Neg logit contributions | |
ocol | -0.030 |
irs | -0.026 |
ait | -0.024 |
Annie | -0.023 |
andel | -0.023 |
Token | other |
Token ID | 584 |
Feature activation | +5.15e+00 |
Pos logit contributions | |
hurts | +0.022 |
hits | +0.020 |
Because | +0.020 |
zag | +0.019 |
not | +0.018 |
Neg logit contributions | |
ocol | -0.082 |
irs | -0.069 |
ait | -0.068 |
sie | -0.066 |
andel | -0.066 |
Token | friend |
Token ID | 1545 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
hurts | +0.085 |
hits | +0.082 |
misses | +0.080 |
zag | +0.079 |
bike | +0.078 |
Neg logit contributions | |
ocol | -0.177 |
ait | -0.150 |
sie | -0.147 |
ages | -0.143 |
andel | -0.140 |
Token | was |
Token ID | 373 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
owner | +0.013 |
not | +0.012 |
hurts | +0.012 |
zag | +0.011 |
hits | +0.011 |
Neg logit contributions | |
ocol | -0.058 |
irs | -0.050 |
ait | -0.048 |
sie | -0.047 |
andel | -0.045 |
Token | short |
Token ID | 1790 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
hurts | +0.039 |
tr | +0.038 |
flower | +0.036 |
hits | +0.036 |
zag | +0.036 |
Neg logit contributions | |
ocol | -0.050 |
year | -0.039 |
irs | -0.038 |
ait | -0.037 |
sie | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
owner | +0.015 |
not | +0.013 |
still | +0.013 |
hurts | +0.012 |
hits | +0.012 |
Neg logit contributions | |
ocol | -0.064 |
irs | -0.054 |
ait | -0.051 |
andel | -0.050 |
sie | -0.050 |
Token | couldn |
Token ID | 3521 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
owner | +0.000 |
hurts | +0.000 |
not | +0.000 |
zag | +0.000 |
hits | +0.000 |
Neg logit contributions | |
ocol | -0.002 |
irs | -0.002 |
ait | -0.001 |
sie | -0.001 |
andel | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
ocol | +0.041 |
irs | +0.036 |
ait | +0.035 |
sie | +0.035 |
andel | +0.034 |
Neg logit contributions | |
not | -0.014 |
owner | -0.013 |
still | -0.012 |
was | -0.012 |
only | -0.011 |
Token | other |
Token ID | 584 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
owner | +0.011 |
not | +0.010 |
still | +0.010 |
hurts | +0.010 |
only | +0.009 |
Neg logit contributions | |
ocol | -0.023 |
irs | -0.020 |
ait | -0.018 |
andel | -0.017 |
sie | -0.017 |
Token | clouds |
Token ID | 15114 |
Feature activation | +8.63e-01 |
Pos logit contributions | |
owner | +0.093 |
not | +0.087 |
still | +0.084 |
hurts | +0.082 |
hits | +0.080 |
Neg logit contributions | |
ocol | -0.171 |
irs | -0.143 |
ait | -0.130 |
andel | -0.126 |
sie | -0.124 |
Token | . |
Token ID | 13 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
owner | +0.016 |
not | +0.015 |
still | +0.014 |
hurts | +0.014 |
hits | +0.014 |
Neg logit contributions | |
ocol | -0.036 |
irs | -0.031 |
ait | -0.028 |
andel | -0.027 |
sie | -0.027 |
Token | The |
Token ID | 383 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
owner | +0.012 |
not | +0.011 |
still | +0.010 |
hurts | +0.010 |
hits | +0.010 |
Neg logit contributions | |
ocol | -0.044 |
irs | -0.039 |
ait | -0.036 |
andel | -0.035 |
sie | -0.034 |
Token | other |
Token ID | 584 |
Feature activation | +5.14e+00 |
Pos logit contributions | |
owner | +0.044 |
not | +0.042 |
still | +0.040 |
hurts | +0.039 |
hits | +0.038 |
Neg logit contributions | |
ocol | -0.082 |
irs | -0.069 |
ait | -0.063 |
andel | -0.061 |
sie | -0.060 |
Token | clouds |
Token ID | 15114 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
owner | +0.113 |
not | +0.108 |
still | +0.104 |
hurts | +0.102 |
hits | +0.100 |
Neg logit contributions | |
ocol | -0.193 |
irs | -0.161 |
ait | -0.147 |
andel | -0.143 |
sie | -0.140 |
Token | were |
Token ID | 547 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
owner | +0.029 |
not | +0.026 |
zag | +0.026 |
hurts | +0.024 |
tr | +0.024 |
Neg logit contributions | |
ocol | -0.066 |
irs | -0.055 |
sie | -0.052 |
ait | -0.052 |
year | -0.049 |
Token | en |
Token ID | 551 |
Feature activation | -7.18e-01 |
Pos logit contributions | |
ocol | +0.006 |
irs | +0.005 |
rubble | +0.004 |
ait | +0.004 |
andel | +0.004 |
Neg logit contributions | |
Because | -0.003 |
hurts | -0.003 |
owner | -0.003 |
zag | -0.003 |
tr | -0.003 |
Token | vious |
Token ID | 1442 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ocol | +0.028 |
irs | +0.023 |
ait | +0.021 |
andel | +0.021 |
sie | +0.020 |
Neg logit contributions | |
owner | -0.016 |
not | -0.015 |
still | -0.014 |
zag | -0.014 |
hurts | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
owner | +0.057 |
not | +0.054 |
zag | +0.052 |
flower | +0.052 |
hurts | +0.052 |
Neg logit contributions | |
ocol | -0.090 |
irs | -0.075 |
ait | -0.069 |
andel | -0.067 |
sie | -0.063 |
Token | a |
Token ID | 257 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
ocol | +0.002 |
irs | +0.001 |
ait | +0.001 |
andel | +0.001 |
sie | +0.001 |
Neg logit contributions | |
owner | -0.001 |
not | -0.001 |
still | -0.001 |
hurts | -0.001 |
hits | -0.001 |
Token | strange |
Token ID | 6283 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
ocol | +0.063 |
irs | +0.056 |
ait | +0.051 |
sie | +0.047 |
andel | +0.046 |
Neg logit contributions | |
owner | -0.028 |
not | -0.024 |
still | -0.024 |
was | -0.024 |
hurts | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
ocol | +0.011 |
irs | +0.009 |
ait | +0.009 |
andel | +0.008 |
sie | +0.008 |
Neg logit contributions | |
owner | -0.005 |
not | -0.005 |
still | -0.004 |
hurts | -0.004 |
was | -0.004 |
Token | raspberry |
Token ID | 38973 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
ocol | +0.123 |
irs | +0.109 |
ait | +0.101 |
sie | +0.094 |
andel | +0.094 |
Neg logit contributions | |
owner | -0.059 |
not | -0.058 |
still | -0.056 |
hurts | -0.054 |
was | -0.053 |
Token | - |
Token ID | 12 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
ocol | +0.022 |
irs | +0.019 |
ait | +0.017 |
andel | +0.017 |
sie | +0.016 |
Neg logit contributions | |
owner | -0.011 |
not | -0.011 |
still | -0.010 |
hurts | -0.010 |
hits | -0.010 |
Token | col |
Token ID | 4033 |
Feature activation | -8.66e+00 |
Pos logit contributions | |
ocol | +0.117 |
irs | +0.096 |
ait | +0.086 |
andel | +0.081 |
sie | +0.080 |
Neg logit contributions | |
owner | -0.099 |
not | -0.095 |
still | -0.091 |
hurts | -0.090 |
was | -0.088 |
Token | oured |
Token ID | 8167 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
ocol | +0.251 |
irs | +0.208 |
sie | +0.187 |
ages | +0.184 |
inations | +0.183 |
Neg logit contributions | |
hits | -0.226 |
owner | -0.224 |
flower | -0.220 |
hurts | -0.219 |
still | -0.218 |
Token | object |
Token ID | 2134 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
ocol | +0.014 |
irs | +0.012 |
ait | +0.011 |
sie | +0.010 |
andel | +0.010 |
Neg logit contributions | |
owner | -0.008 |
not | -0.007 |
still | -0.007 |
was | -0.007 |
hurts | -0.007 |
Token | hiding |
Token ID | 11816 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
ocol | +0.021 |
irs | +0.018 |
ait | +0.017 |
andel | +0.016 |
sie | +0.016 |
Neg logit contributions | |
owner | -0.009 |
not | -0.008 |
still | -0.008 |
hurts | -0.008 |
hits | -0.007 |
Token | under |
Token ID | 739 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ocol | +0.017 |
irs | +0.014 |
ait | +0.013 |
andel | +0.012 |
sie | +0.012 |
Neg logit contributions | |
owner | -0.009 |
not | -0.008 |
still | -0.008 |
was | -0.008 |
remembered | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
ocol | +0.062 |
irs | +0.054 |
ait | +0.054 |
inations | +0.053 |
sie | +0.051 |
Neg logit contributions | |
not | -0.015 |
owner | -0.014 |
was | -0.013 |
only | -0.013 |
still | -0.012 |
Token | get |
Token ID | 651 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
ocol | +0.062 |
irs | +0.052 |
ait | +0.048 |
sie | +0.046 |
andel | +0.045 |
Neg logit contributions | |
owner | -0.030 |
not | -0.027 |
hurts | -0.026 |
zag | -0.026 |
hits | -0.026 |
Token | something |
Token ID | 1223 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
ocol | +0.009 |
irs | +0.008 |
ait | +0.008 |
andel | +0.007 |
uster | +0.007 |
Neg logit contributions | |
not | -0.004 |
still | -0.004 |
owner | -0.004 |
only | -0.003 |
also | -0.003 |
Token | sweet |
Token ID | 6029 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
ocol | +0.133 |
irs | +0.114 |
ait | +0.105 |
andel | +0.102 |
sie | +0.098 |
Neg logit contributions | |
owner | -0.059 |
not | -0.059 |
still | -0.057 |
hurts | -0.052 |
was | -0.052 |
Token | or |
Token ID | 393 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
ocol | +0.088 |
irs | +0.073 |
ait | +0.067 |
andel | +0.066 |
sie | +0.066 |
Neg logit contributions | |
owner | -0.038 |
not | -0.034 |
zag | -0.034 |
still | -0.032 |
hurts | -0.032 |
Token | something |
Token ID | 1223 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
ocol | +0.082 |
irs | +0.068 |
ait | +0.062 |
andel | +0.060 |
sie | +0.059 |
Neg logit contributions | |
owner | -0.047 |
not | -0.044 |
still | -0.042 |
hurts | -0.041 |
hits | -0.041 |
Token | sav |
Token ID | 6799 |
Feature activation | -8.45e+00 |
Pos logit contributions | |
ocol | +0.143 |
irs | +0.123 |
ait | +0.115 |
andel | +0.112 |
uster | +0.108 |
Neg logit contributions | |
not | -0.068 |
owner | -0.065 |
still | -0.064 |
was | -0.061 |
hurts | -0.060 |
Token | our |
Token ID | 454 |
Feature activation | -5.23e-01 |
Pos logit contributions | |
ocol | +0.287 |
irs | +0.246 |
ait | +0.216 |
rubble | +0.207 |
andel | +0.206 |
Neg logit contributions | |
owner | -0.205 |
still | -0.197 |
not | -0.197 |
hits | -0.193 |
hurts | -0.189 |
Token | y |
Token ID | 88 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ocol | +0.021 |
irs | +0.018 |
ait | +0.016 |
andel | +0.016 |
sie | +0.016 |
Neg logit contributions | |
owner | -0.010 |
not | -0.009 |
still | -0.009 |
hurts | -0.009 |
hits | -0.009 |
Token | ? |
Token ID | 30 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ocol | +0.054 |
irs | +0.046 |
ait | +0.042 |
andel | +0.041 |
sie | +0.040 |
Neg logit contributions | |
owner | -0.025 |
not | -0.023 |
still | -0.022 |
hurts | -0.022 |
hits | -0.021 |
Token | In |
Token ID | 554 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
ocol | +0.060 |
irs | +0.052 |
ait | +0.047 |
andel | +0.047 |
sie | +0.046 |
Neg logit contributions | |
owner | -0.015 |
still | -0.013 |
not | -0.012 |
was | -0.012 |
only | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
owner | +0.005 |
not | +0.005 |
still | +0.005 |
hurts | +0.005 |
hits | +0.005 |
Neg logit contributions | |
ocol | -0.014 |
irs | -0.012 |
ait | -0.011 |
andel | -0.011 |
sie | -0.010 |
Token | big |
Token ID | 1263 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ocol | +0.081 |
irs | +0.073 |
ait | +0.066 |
andel | +0.061 |
sie | +0.060 |
Neg logit contributions | |
owner | -0.039 |
still | -0.035 |
was | -0.035 |
not | -0.034 |
also | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
ocol | +0.053 |
irs | +0.048 |
ait | +0.043 |
andel | +0.041 |
inations | +0.040 |
Neg logit contributions | |
owner | -0.025 |
not | -0.023 |
still | -0.023 |
hits | -0.022 |
hurts | -0.022 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
ocol | +0.112 |
irs | +0.103 |
ait | +0.093 |
andel | +0.087 |
sie | +0.086 |
Neg logit contributions | |
owner | -0.047 |
still | -0.044 |
not | -0.043 |
also | -0.040 |
was | -0.039 |
Token | rainbow |
Token ID | 27223 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
ocol | +0.021 |
irs | +0.019 |
ait | +0.017 |
inations | +0.016 |
year | +0.016 |
Neg logit contributions | |
owner | -0.010 |
was | -0.009 |
flower | -0.009 |
still | -0.009 |
not | -0.009 |
Token | - |
Token ID | 12 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
ocol | +0.006 |
irs | +0.005 |
ait | +0.005 |
andel | +0.005 |
sie | +0.005 |
Neg logit contributions | |
owner | -0.003 |
not | -0.003 |
was | -0.002 |
still | -0.002 |
flower | -0.002 |
Token | col |
Token ID | 4033 |
Feature activation | -8.37e+00 |
Pos logit contributions | |
ocol | +0.110 |
irs | +0.086 |
ait | +0.079 |
andel | +0.073 |
sie | +0.073 |
Neg logit contributions | |
owner | -0.102 |
not | -0.096 |
was | -0.094 |
still | -0.093 |
zag | -0.093 |
Token | oured |
Token ID | 8167 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ocol | +0.254 |
irs | +0.211 |
ages | +0.194 |
inations | +0.192 |
rubble | +0.192 |
Neg logit contributions | |
hits | -0.203 |
flower | -0.201 |
owner | -0.197 |
hurts | -0.195 |
uck | -0.195 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
ocol | +0.048 |
irs | +0.044 |
ait | +0.040 |
inations | +0.038 |
uster | +0.037 |
Neg logit contributions | |
owner | -0.021 |
was | -0.019 |
flower | -0.018 |
not | -0.017 |
still | -0.017 |
Token | ! |
Token ID | 0 |
Feature activation | -8.51e-01 |
Pos logit contributions | |
ocol | +0.040 |
irs | +0.034 |
ait | +0.031 |
andel | +0.030 |
sie | +0.030 |
Neg logit contributions | |
owner | -0.018 |
not | -0.017 |
still | -0.016 |
hurts | -0.016 |
hits | -0.015 |
Token | The |
Token ID | 383 |
Feature activation | +9.92e-01 |
Pos logit contributions | |
ocol | +0.041 |
irs | +0.034 |
sie | +0.032 |
inations | +0.032 |
ait | +0.032 |
Neg logit contributions | |
owner | -0.012 |
was | -0.010 |
not | -0.010 |
still | -0.010 |
Because | -0.009 |
Token | little |
Token ID | 1310 |
Feature activation | +6.79e-01 |
Pos logit contributions | |
owner | +0.021 |
still | +0.018 |
not | +0.018 |
was | +0.017 |
flower | +0.017 |
Neg logit contributions | |
ocol | -0.042 |
irs | -0.036 |
ait | -0.033 |
andel | -0.031 |
sie | -0.030 |
Token | date |
Token ID | 3128 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
owner | +0.006 |
not | +0.006 |
still | +0.006 |
hurts | +0.006 |
was | +0.006 |
Neg logit contributions | |
ocol | -0.007 |
irs | -0.005 |
ait | -0.005 |
uster | -0.004 |
sie | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
ocol | +0.053 |
irs | +0.046 |
ait | +0.042 |
andel | +0.041 |
sie | +0.041 |
Neg logit contributions | |
owner | -0.016 |
not | -0.014 |
hurts | -0.014 |
still | -0.013 |
hits | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
owner | +0.008 |
not | +0.008 |
still | +0.007 |
was | +0.007 |
only | +0.006 |
Neg logit contributions | |
ocol | -0.020 |
irs | -0.018 |
ait | -0.016 |
andel | -0.016 |
sie | -0.015 |
Token | rainbow |
Token ID | 27223 |
Feature activation | +6.11e-02 |
Pos logit contributions | |
ocol | +0.057 |
irs | +0.051 |
ait | +0.045 |
andel | +0.041 |
sie | +0.041 |
Neg logit contributions | |
owner | -0.028 |
not | -0.024 |
still | -0.024 |
was | -0.024 |
remembered | -0.022 |
Token | - |
Token ID | 12 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
owner | +0.001 |
not | +0.001 |
still | +0.001 |
hurts | +0.001 |
hits | +0.001 |
Neg logit contributions | |
ocol | -0.003 |
irs | -0.002 |
ait | -0.002 |
andel | -0.002 |
sie | -0.002 |
Token | col |
Token ID | 4033 |
Feature activation | -8.35e+00 |
Pos logit contributions | |
ocol | +0.110 |
irs | +0.087 |
ait | +0.079 |
andel | +0.074 |
sie | +0.074 |
Neg logit contributions | |
owner | -0.093 |
not | -0.088 |
still | -0.085 |
was | -0.085 |
zag | -0.085 |
Token | oured |
Token ID | 8167 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
ocol | +0.258 |
irs | +0.208 |
sie | +0.195 |
inations | +0.194 |
rubble | +0.193 |
Neg logit contributions | |
hits | -0.199 |
owner | -0.197 |
tr | -0.194 |
uck | -0.194 |
still | -0.194 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
ocol | +0.002 |
irs | +0.002 |
ait | +0.002 |
uster | +0.002 |
inations | +0.001 |
Neg logit contributions | |
owner | -0.001 |
was | -0.001 |
flower | -0.001 |
not | -0.001 |
still | -0.001 |
Token | flew |
Token ID | 13112 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
ocol | +0.030 |
irs | +0.026 |
ait | +0.023 |
andel | +0.023 |
sie | +0.023 |
Neg logit contributions | |
owner | -0.011 |
not | -0.010 |
still | -0.010 |
hurts | -0.009 |
hits | -0.009 |
Token | in |
Token ID | 287 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
ocol | +0.079 |
irs | +0.068 |
ait | +0.064 |
inations | +0.062 |
sie | +0.060 |
Neg logit contributions | |
owner | -0.025 |
still | -0.024 |
not | -0.023 |
was | -0.021 |
only | -0.020 |
Token | through |
Token ID | 832 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
ocol | +0.045 |
irs | +0.038 |
inations | +0.037 |
ait | +0.037 |
sie | +0.037 |
Neg logit contributions | |
was | -0.010 |
still | -0.010 |
only | -0.009 |
not | -0.009 |
and | -0.009 |
Token | day |
Token ID | 1110 |
Feature activation | +6.85e-02 |
Pos logit contributions | |
owner | +0.005 |
not | +0.005 |
still | +0.005 |
hurts | +0.005 |
hits | +0.005 |
Neg logit contributions | |
ocol | -0.012 |
irs | -0.010 |
sie | -0.010 |
andel | -0.010 |
ait | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
owner | +0.001 |
not | +0.001 |
still | +0.001 |
hurts | +0.001 |
hits | +0.001 |
Neg logit contributions | |
ocol | -0.003 |
irs | -0.003 |
ait | -0.003 |
andel | -0.003 |
sie | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
owner | +0.025 |
not | +0.025 |
still | +0.024 |
was | +0.023 |
hurts | +0.022 |
Neg logit contributions | |
ocol | -0.048 |
irs | -0.041 |
ait | -0.037 |
andel | -0.036 |
sie | -0.035 |
Token | chocolate |
Token ID | 11311 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
ocol | +0.062 |
irs | +0.054 |
ait | +0.048 |
andel | +0.045 |
sie | +0.045 |
Neg logit contributions | |
owner | -0.029 |
not | -0.026 |
still | -0.026 |
was | -0.024 |
also | -0.023 |
Token | - |
Token ID | 12 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
not | +0.007 |
owner | +0.007 |
still | +0.006 |
remembered | +0.006 |
hurts | +0.006 |
Neg logit contributions | |
ocol | -0.014 |
irs | -0.011 |
ait | -0.011 |
andel | -0.011 |
sie | -0.010 |
Token | col |
Token ID | 4033 |
Feature activation | -8.34e+00 |
Pos logit contributions | |
ocol | +0.125 |
irs | +0.103 |
ait | +0.093 |
andel | +0.090 |
sie | +0.088 |
Neg logit contributions | |
owner | -0.085 |
not | -0.080 |
still | -0.077 |
hurts | -0.076 |
hits | -0.074 |
Token | oured |
Token ID | 8167 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
ocol | +0.250 |
irs | +0.209 |
inations | +0.191 |
ages | +0.187 |
sie | +0.187 |
Neg logit contributions | |
owner | -0.213 |
hits | -0.210 |
hurts | -0.206 |
flower | -0.205 |
still | -0.204 |
Token | whip |
Token ID | 22972 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
owner | +0.005 |
not | +0.004 |
still | +0.004 |
hurts | +0.004 |
was | +0.004 |
Neg logit contributions | |
ocol | -0.010 |
irs | -0.009 |
ait | -0.008 |
sie | -0.007 |
andel | -0.007 |
Token | named |
Token ID | 3706 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
owner | +0.020 |
not | +0.019 |
still | +0.018 |
hurts | +0.018 |
zag | +0.018 |
Neg logit contributions | |
ocol | -0.043 |
irs | -0.036 |
ait | -0.032 |
andel | -0.032 |
ages | -0.031 |
Token | C |
Token ID | 327 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
ocol | +0.162 |
irs | +0.135 |
inations | +0.122 |
ait | +0.121 |
sie | +0.120 |
Neg logit contributions | |
owner | -0.081 |
not | -0.079 |
still | -0.071 |
only | -0.070 |
hurts | -0.069 |
Token | heddar |
Token ID | 44937 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
ocol | +0.168 |
irs | +0.140 |
ait | +0.125 |
andel | +0.124 |
sie | +0.122 |
Neg logit contributions | |
owner | -0.095 |
not | -0.087 |
still | -0.086 |
hurts | -0.082 |
zag | -0.082 |
Token | suddenly |
Token ID | 6451 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
ocol | +0.018 |
irs | +0.016 |
ait | +0.015 |
andel | +0.014 |
uster | +0.014 |
Neg logit contributions | |
owner | -0.006 |
not | -0.006 |
still | -0.006 |
was | -0.005 |
also | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +4.22e-01 |
Pos logit contributions | |
not | +0.017 |
owner | +0.016 |
was | +0.015 |
remembered | +0.014 |
still | +0.014 |
Neg logit contributions | |
ocol | -0.063 |
irs | -0.056 |
ait | -0.052 |
inations | -0.050 |
uster | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
not | +0.007 |
owner | +0.006 |
still | +0.006 |
was | +0.006 |
only | +0.006 |
Neg logit contributions | |
ocol | -0.020 |
irs | -0.017 |
ait | -0.016 |
uster | -0.015 |
andel | -0.015 |
Token | brightly |
Token ID | 35254 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ocol | +0.101 |
irs | +0.095 |
ait | +0.085 |
inations | +0.079 |
sie | +0.078 |
Neg logit contributions | |
owner | -0.039 |
not | -0.032 |
was | -0.032 |
still | -0.031 |
hurts | -0.030 |
Token | - |
Token ID | 12 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
ocol | +0.044 |
irs | +0.033 |
andel | +0.032 |
year | +0.030 |
ages | +0.029 |
Neg logit contributions | |
owner | -0.031 |
not | -0.028 |
neither | -0.028 |
still | -0.028 |
remembered | -0.026 |
Token | col |
Token ID | 4033 |
Feature activation | -8.28e+00 |
Pos logit contributions | |
ocol | +0.102 |
irs | +0.082 |
ait | +0.071 |
sie | +0.067 |
andel | +0.065 |
Neg logit contributions | |
owner | -0.106 |
tr | -0.096 |
not | -0.096 |
still | -0.094 |
was | -0.093 |
Token | oured |
Token ID | 8167 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
ocol | +0.247 |
irs | +0.213 |
sie | +0.187 |
ages | +0.185 |
inations | +0.185 |
Neg logit contributions | |
hits | -0.211 |
owner | -0.210 |
hurts | -0.205 |
tr | -0.203 |
flower | -0.202 |
Token | belt |
Token ID | 10999 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
ocol | +0.002 |
irs | +0.002 |
ait | +0.002 |
inations | +0.002 |
year | +0.002 |
Neg logit contributions | |
owner | -0.001 |
flower | -0.001 |
was | -0.001 |
head | -0.001 |
not | -0.001 |
Token | caught |
Token ID | 4978 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
ocol | +0.028 |
irs | +0.024 |
ait | +0.022 |
andel | +0.022 |
rubble | +0.021 |
Neg logit contributions | |
owner | -0.013 |
not | -0.012 |
still | -0.012 |
hurts | -0.011 |
flower | -0.011 |
Token | her |
Token ID | 607 |
Feature activation | -7.74e-01 |
Pos logit contributions | |
ocol | +0.094 |
inations | +0.090 |
irs | +0.086 |
ait | +0.080 |
rubble | +0.078 |
Neg logit contributions | |
owner | -0.021 |
was | -0.021 |
next | -0.020 |
still | -0.019 |
not | -0.019 |
Token | eye |
Token ID | 4151 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
ocol | +0.033 |
irs | +0.028 |
ait | +0.026 |
andel | +0.026 |
sie | +0.025 |
Neg logit contributions | |
owner | -0.012 |
not | -0.012 |
hurts | -0.011 |
hits | -0.011 |
still | -0.011 |
Token | went |
Token ID | 1816 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
ocol | +0.011 |
irs | +0.009 |
ait | +0.009 |
andel | +0.008 |
sie | +0.008 |
Neg logit contributions | |
owner | -0.004 |
not | -0.003 |
still | -0.003 |
hurts | -0.003 |
zag | -0.003 |
Token | outside |
Token ID | 2354 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ocol | +0.029 |
irs | +0.025 |
ait | +0.023 |
andel | +0.022 |
sie | +0.022 |
Neg logit contributions | |
owner | -0.006 |
not | -0.006 |
still | -0.005 |
zag | -0.005 |
hits | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
owner | +0.016 |
not | +0.016 |
still | +0.014 |
was | +0.014 |
also | +0.012 |
Neg logit contributions | |
ocol | -0.068 |
irs | -0.060 |
ait | -0.055 |
andel | -0.054 |
sie | -0.054 |
Token | find |
Token ID | 1064 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
ocol | +0.031 |
irs | +0.026 |
ait | +0.024 |
andel | +0.024 |
sie | +0.023 |
Neg logit contributions | |
owner | -0.013 |
not | -0.013 |
still | -0.012 |
hurts | -0.012 |
hits | -0.012 |
Token | something |
Token ID | 1223 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
ocol | +0.012 |
irs | +0.010 |
ait | +0.010 |
sie | +0.010 |
andel | +0.010 |
Neg logit contributions | |
owner | -0.003 |
not | -0.003 |
was | -0.003 |
still | -0.003 |
only | -0.002 |
Token | su |
Token ID | 424 |
Feature activation | -7.94e+00 |
Pos logit contributions | |
ocol | +0.122 |
irs | +0.103 |
ait | +0.095 |
andel | +0.092 |
sie | +0.089 |
Neg logit contributions | |
owner | -0.072 |
not | -0.070 |
still | -0.067 |
was | -0.064 |
hurts | -0.063 |
Token | gary |
Token ID | 14849 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
ocol | +0.191 |
Annie | +0.170 |
rubble | +0.146 |
hotels | +0.145 |
irs | +0.145 |
Neg logit contributions | |
uck | -0.245 |
uch | -0.233 |
still | -0.229 |
owner | -0.226 |
flower | -0.223 |
Token | . |
Token ID | 13 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
ocol | +0.074 |
irs | +0.066 |
ait | +0.061 |
year | +0.057 |
andel | +0.057 |
Neg logit contributions | |
owner | -0.030 |
was | -0.027 |
not | -0.026 |
hurts | -0.025 |
still | -0.025 |
Token | But |
Token ID | 887 |
Feature activation | +8.23e-01 |
Pos logit contributions | |
owner | +0.006 |
not | +0.006 |
hurts | +0.006 |
still | +0.006 |
hits | +0.005 |
Neg logit contributions | |
ocol | -0.018 |
irs | -0.015 |
ait | -0.014 |
andel | -0.014 |
sie | -0.014 |
Token | what |
Token ID | 644 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
owner | +0.015 |
hurts | +0.015 |
zag | +0.014 |
hits | +0.014 |
hates | +0.014 |
Neg logit contributions | |
ocol | -0.033 |
irs | -0.027 |
ait | -0.025 |
sie | -0.024 |
andel | -0.024 |
Token | he |
Token ID | 339 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
flower | +0.001 |
owner | +0.001 |
zag | +0.001 |
hits | +0.001 |
hurts | +0.001 |
Neg logit contributions | |
ocol | -0.003 |
irs | -0.002 |
ait | -0.002 |
Annie | -0.002 |
pounds | -0.002 |
Token | ipop |
Token ID | 42800 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
ocol | +0.119 |
irs | +0.106 |
ait | +0.093 |
andel | +0.092 |
rubble | +0.087 |
Neg logit contributions | |
owner | -0.071 |
not | -0.067 |
hurts | -0.065 |
still | -0.064 |
remembered | -0.062 |
Token | , |
Token ID | 11 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ocol | +0.024 |
irs | +0.021 |
andel | +0.019 |
ait | +0.018 |
sie | +0.017 |
Neg logit contributions | |
owner | -0.012 |
not | -0.012 |
still | -0.011 |
was | -0.010 |
hurts | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
ocol | +0.061 |
irs | +0.054 |
andel | +0.048 |
ait | +0.048 |
sie | +0.048 |
Neg logit contributions | |
owner | -0.021 |
not | -0.021 |
still | -0.020 |
was | -0.017 |
also | -0.017 |
Token | 's |
Token ID | 338 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
ocol | +0.008 |
irs | +0.007 |
ait | +0.007 |
andel | +0.007 |
sie | +0.006 |
Neg logit contributions | |
owner | -0.002 |
not | -0.002 |
still | -0.002 |
hurts | -0.002 |
hits | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
not | +0.014 |
owner | +0.013 |
still | +0.013 |
also | +0.012 |
only | +0.012 |
Neg logit contributions | |
ocol | -0.027 |
irs | -0.023 |
ait | -0.022 |
andel | -0.021 |
sie | -0.021 |
Token | sav |
Token ID | 6799 |
Feature activation | -7.94e+00 |
Pos logit contributions | |
ocol | +0.063 |
irs | +0.056 |
ait | +0.050 |
andel | +0.047 |
sie | +0.045 |
Neg logit contributions | |
owner | -0.038 |
not | -0.035 |
still | -0.034 |
was | -0.033 |
only | -0.032 |
Token | our |
Token ID | 454 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
ocol | +0.223 |
irs | +0.217 |
rubble | +0.180 |
Annie | +0.173 |
ait | +0.169 |
Neg logit contributions | |
owner | -0.207 |
hits | -0.206 |
zag | -0.201 |
head | -0.200 |
hurts | -0.200 |
Token | y |
Token ID | 88 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
owner | +0.011 |
not | +0.010 |
still | +0.010 |
hurts | +0.010 |
hits | +0.010 |
Neg logit contributions | |
ocol | -0.017 |
irs | -0.014 |
ait | -0.013 |
andel | -0.012 |
sie | -0.012 |
Token | l |
Token ID | 300 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
ocol | +0.024 |
irs | +0.021 |
ait | +0.019 |
andel | +0.018 |
sie | +0.017 |
Neg logit contributions | |
owner | -0.013 |
not | -0.012 |
still | -0.012 |
was | -0.012 |
hurts | -0.012 |
Token | oll |
Token ID | 692 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
ocol | +0.134 |
irs | +0.126 |
ait | +0.108 |
hotels | +0.104 |
rubble | +0.104 |
Neg logit contributions | |
owner | -0.086 |
still | -0.082 |
not | -0.079 |
also | -0.074 |
only | -0.074 |
Token | ipop |
Token ID | 42800 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ocol | +0.144 |
irs | +0.129 |
ait | +0.113 |
andel | +0.112 |
rubble | +0.106 |
Neg logit contributions | |
owner | -0.084 |
not | -0.080 |
hurts | -0.078 |
still | -0.077 |
remembered | -0.074 |
Token | - |
Token ID | 532 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
ocol | +0.040 |
irs | +0.035 |
ait | +0.032 |
andel | +0.032 |
sie | +0.031 |
Neg logit contributions | |
owner | -0.014 |
not | -0.013 |
was | -0.013 |
still | -0.012 |
only | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
owner | +0.019 |
not | +0.018 |
still | +0.017 |
hurts | +0.016 |
was | +0.016 |
Neg logit contributions | |
ocol | -0.043 |
irs | -0.036 |
ait | -0.033 |
andel | -0.033 |
sie | -0.032 |
Token | box |
Token ID | 3091 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ocol | +0.028 |
irs | +0.024 |
ait | +0.022 |
andel | +0.021 |
sie | +0.021 |
Neg logit contributions | |
owner | -0.015 |
not | -0.014 |
still | -0.014 |
was | -0.013 |
hurts | -0.013 |
Token | of |
Token ID | 286 |
Feature activation | -9.74e-01 |
Pos logit contributions | |
ocol | +0.059 |
irs | +0.050 |
ait | +0.045 |
andel | +0.044 |
sie | +0.043 |
Neg logit contributions | |
owner | -0.032 |
not | -0.031 |
still | -0.030 |
was | -0.029 |
hurts | -0.028 |
Token | mult |
Token ID | 1963 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
ocol | +0.037 |
irs | +0.034 |
ait | +0.034 |
sie | +0.032 |
andel | +0.031 |
Neg logit contributions | |
owner | -0.020 |
was | -0.019 |
still | -0.019 |
not | -0.018 |
only | -0.018 |
Token | icol |
Token ID | 27045 |
Feature activation | -7.89e+00 |
Pos logit contributions | |
ocol | +0.147 |
irs | +0.128 |
ait | +0.113 |
sie | +0.112 |
year | +0.112 |
Neg logit contributions | |
owner | -0.114 |
hits | -0.105 |
head | -0.104 |
still | -0.103 |
flower | -0.101 |
Token | ored |
Token ID | 1850 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
ocol | +0.264 |
irs | +0.224 |
ait | +0.193 |
andel | +0.192 |
rubble | +0.191 |
Neg logit contributions | |
owner | -0.183 |
still | -0.183 |
hits | -0.180 |
not | -0.176 |
hurts | -0.174 |
Token | bubbles |
Token ID | 25037 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
ocol | +0.044 |
irs | +0.039 |
sie | +0.036 |
ait | +0.036 |
inations | +0.034 |
Neg logit contributions | |
owner | -0.025 |
was | -0.024 |
still | -0.021 |
not | -0.021 |
's | -0.021 |
Token | ! |
Token ID | 0 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ocol | +0.034 |
irs | +0.028 |
ait | +0.026 |
andel | +0.026 |
sie | +0.025 |
Neg logit contributions | |
owner | -0.014 |
not | -0.013 |
still | -0.013 |
was | -0.012 |
hurts | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
ocol | +0.067 |
irs | +0.057 |
sie | +0.053 |
inations | +0.053 |
ait | +0.053 |
Neg logit contributions | |
owner | -0.012 |
was | -0.010 |
still | -0.008 |
not | -0.008 |
's | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
ocol | +0.037 |
irs | +0.031 |
ait | +0.030 |
andel | +0.029 |
sie | +0.029 |
Neg logit contributions | |
owner | -0.005 |
not | -0.004 |
was | -0.004 |
still | -0.004 |
also | -0.003 |
Token | delicious |
Token ID | 12625 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ocol | +0.066 |
irs | +0.059 |
ait | +0.057 |
uster | +0.055 |
sie | +0.054 |
Neg logit contributions | |
owner | -0.030 |
was | -0.026 |
still | -0.025 |
not | -0.025 |
only | -0.024 |
Token | vegetables |
Token ID | 13701 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ocol | +0.068 |
irs | +0.065 |
ait | +0.062 |
uster | +0.061 |
chie | +0.059 |
Neg logit contributions | |
was | -0.028 |
owner | -0.028 |
and | -0.026 |
's | -0.024 |
is | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
ocol | +0.057 |
irs | +0.049 |
ait | +0.045 |
andel | +0.044 |
sie | +0.044 |
Neg logit contributions | |
owner | -0.019 |
not | -0.017 |
still | -0.016 |
hurts | -0.015 |
hits | -0.015 |
Token | bright |
Token ID | 6016 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ocol | +0.032 |
irs | +0.028 |
ait | +0.025 |
andel | +0.024 |
uster | +0.024 |
Neg logit contributions | |
owner | -0.022 |
not | -0.021 |
was | -0.021 |
still | -0.020 |
also | -0.020 |
Token | - |
Token ID | 12 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
ocol | +0.055 |
irs | +0.051 |
ait | +0.050 |
prise | +0.047 |
sie | +0.047 |
Neg logit contributions | |
owner | -0.021 |
was | -0.021 |
flower | -0.018 |
and | -0.018 |
still | -0.017 |
Token | col |
Token ID | 4033 |
Feature activation | -7.75e+00 |
Pos logit contributions | |
ocol | +0.097 |
irs | +0.079 |
ait | +0.071 |
uster | +0.068 |
sie | +0.067 |
Neg logit contributions | |
owner | -0.082 |
not | -0.073 |
was | -0.070 |
flower | -0.069 |
's | -0.068 |
Token | oured |
Token ID | 8167 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
ocol | +0.224 |
irs | +0.183 |
ages | +0.180 |
sie | +0.174 |
Annie | +0.174 |
Neg logit contributions | |
flower | -0.185 |
hits | -0.184 |
erved | -0.183 |
owner | -0.183 |
swung | -0.181 |
Token | flowers |
Token ID | 12734 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ocol | +0.010 |
irs | +0.010 |
ait | +0.009 |
uster | +0.009 |
prise | +0.009 |
Neg logit contributions | |
owner | -0.005 |
was | -0.005 |
and | -0.004 |
not | -0.004 |
liked | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
ocol | +0.060 |
irs | +0.051 |
ait | +0.047 |
andel | +0.046 |
sie | +0.045 |
Neg logit contributions | |
owner | -0.023 |
not | -0.021 |
still | -0.020 |
was | -0.019 |
flower | -0.018 |
Token | |
Token ID | 220 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
owner | +0.007 |
not | +0.006 |
still | +0.006 |
was | +0.006 |
also | +0.005 |
Neg logit contributions | |
ocol | -0.026 |
irs | -0.022 |
ait | -0.020 |
andel | -0.020 |
sie | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
owner | +0.014 |
not | +0.013 |
still | +0.012 |
hurts | +0.010 |
was | +0.010 |
Neg logit contributions | |
ocol | -0.067 |
irs | -0.058 |
ait | -0.054 |
andel | -0.053 |
sie | -0.053 |
Token | things |
Token ID | 1243 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
ocol | +0.014 |
ait | +0.013 |
irs | +0.013 |
sie | +0.012 |
uster | +0.012 |
Neg logit contributions | |
was | -0.006 |
and | -0.005 |
is | -0.005 |
owner | -0.005 |
's | -0.005 |
Token | â |
Token ID | 22940 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
ocol | +0.138 |
irs | +0.118 |
andel | +0.109 |
ait | +0.109 |
sie | +0.104 |
Neg logit contributions | |
owner | -0.047 |
not | -0.046 |
still | -0.044 |
was | -0.040 |
hurts | -0.039 |
Token | € |
Token ID | 26391 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
ocol | +0.122 |
irs | +0.102 |
ait | +0.088 |
andel | +0.087 |
sie | +0.084 |
Neg logit contributions | |
owner | -0.083 |
not | -0.079 |
still | -0.078 |
hurts | -0.076 |
zag | -0.075 |
Token | � |
Token ID | 447 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
ocol | +0.023 |
irs | +0.019 |
andel | +0.017 |
sie | +0.017 |
ait | +0.017 |
Neg logit contributions | |
owner | -0.010 |
not | -0.010 |
was | -0.009 |
still | -0.009 |
zag | -0.008 |
Token | � |
Token ID | 251 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
ocol | +0.166 |
irs | +0.138 |
ait | +0.127 |
andel | +0.126 |
sie | +0.121 |
Neg logit contributions | |
owner | -0.084 |
still | -0.080 |
not | -0.078 |
was | -0.076 |
hurts | -0.075 |
Token | col |
Token ID | 4033 |
Feature activation | -7.57e+00 |
Pos logit contributions | |
ocol | +0.038 |
irs | +0.033 |
ait | +0.031 |
andel | +0.029 |
sie | +0.029 |
Neg logit contributions | |
was | -0.013 |
still | -0.012 |
owner | -0.012 |
not | -0.012 |
felt | -0.011 |
Token | ouring |
Token ID | 21823 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
ocol | +0.236 |
rubble | +0.188 |
irs | +0.187 |
ages | +0.178 |
hotels | +0.174 |
Neg logit contributions | |
zag | -0.172 |
uck | -0.169 |
still | -0.168 |
flower | -0.167 |
swung | -0.165 |
Token | books |
Token ID | 3835 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
not | +0.007 |
owner | +0.007 |
was | +0.007 |
still | +0.007 |
also | +0.006 |
Neg logit contributions | |
ocol | -0.019 |
irs | -0.016 |
ait | -0.015 |
andel | -0.014 |
prise | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ocol | +0.099 |
irs | +0.086 |
ait | +0.079 |
andel | +0.078 |
sie | +0.077 |
Neg logit contributions | |
owner | -0.027 |
not | -0.025 |
still | -0.023 |
hurts | -0.021 |
was | -0.021 |
Token | stickers |
Token ID | 28568 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
ocol | +0.049 |
irs | +0.042 |
ait | +0.038 |
andel | +0.038 |
sie | +0.037 |
Neg logit contributions | |
owner | -0.019 |
not | -0.017 |
still | -0.016 |
hurts | -0.016 |
hits | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
ocol | +0.026 |
irs | +0.023 |
ait | +0.021 |
andel | +0.021 |
sie | +0.021 |
Neg logit contributions | |
owner | -0.006 |
not | -0.005 |
still | -0.005 |
hurts | -0.005 |
hits | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
ocol | +0.023 |
sie | +0.020 |
irs | +0.019 |
ait | +0.019 |
Annie | +0.019 |
Neg logit contributions | |
hurts | -0.007 |
owner | -0.007 |
hits | -0.007 |
not | -0.006 |
ses | -0.006 |
Token | known |
Token ID | 1900 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
owner | +0.012 |
hurts | +0.011 |
hits | +0.011 |
not | +0.010 |
zag | +0.010 |
Neg logit contributions | |
ocol | -0.025 |
irs | -0.021 |
ait | -0.019 |
sie | -0.019 |
andel | -0.019 |
Token | as |
Token ID | 355 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
ocol | +0.082 |
irs | +0.074 |
ait | +0.067 |
ann | +0.064 |
inations | +0.064 |
Neg logit contributions | |
not | -0.030 |
still | -0.027 |
owner | -0.027 |
only | -0.024 |
never | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ocol | +0.147 |
irs | +0.130 |
ait | +0.120 |
andel | +0.117 |
uster | +0.114 |
Neg logit contributions | |
not | -0.051 |
owner | -0.046 |
still | -0.045 |
was | -0.043 |
only | -0.043 |
Token | powerful |
Token ID | 3665 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
ocol | +0.045 |
irs | +0.041 |
ait | +0.036 |
andel | +0.034 |
inations | +0.033 |
Neg logit contributions | |
owner | -0.021 |
still | -0.019 |
only | -0.018 |
not | -0.018 |
also | -0.017 |
Token | sav |
Token ID | 6799 |
Feature activation | -7.57e+00 |
Pos logit contributions | |
owner | +0.021 |
not | +0.019 |
still | +0.019 |
was | +0.018 |
hurts | +0.018 |
Neg logit contributions | |
ocol | -0.036 |
irs | -0.032 |
ait | -0.028 |
andel | -0.027 |
sie | -0.026 |
Token | iour |
Token ID | 49439 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ocol | +0.233 |
irs | +0.212 |
ait | +0.177 |
rubble | +0.175 |
Annie | +0.173 |
Neg logit contributions | |
owner | -0.187 |
not | -0.184 |
hits | -0.183 |
still | -0.183 |
zag | -0.179 |
Token | of |
Token ID | 286 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
ocol | +0.040 |
irs | +0.034 |
ait | +0.030 |
andel | +0.029 |
sie | +0.029 |
Neg logit contributions | |
owner | -0.025 |
not | -0.023 |
still | -0.023 |
was | -0.022 |
zag | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
ocol | +0.043 |
irs | +0.038 |
ait | +0.036 |
andel | +0.034 |
uster | +0.034 |
Neg logit contributions | |
not | -0.018 |
owner | -0.018 |
still | -0.017 |
only | -0.016 |
was | -0.016 |
Token | forest |
Token ID | 8222 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
ocol | +0.053 |
irs | +0.047 |
ait | +0.043 |
andel | +0.040 |
uster | +0.039 |
Neg logit contributions | |
owner | -0.031 |
still | -0.029 |
not | -0.028 |
was | -0.027 |
only | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -8.77e-01 |
Pos logit contributions | |
ocol | +0.026 |
irs | +0.023 |
ait | +0.021 |
andel | +0.020 |
sie | +0.020 |
Neg logit contributions | |
owner | -0.010 |
not | -0.009 |
still | -0.009 |
was | -0.009 |
hurts | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -7.39e-01 |
Pos logit contributions | |
ocol | +0.074 |
irs | +0.064 |
ait | +0.059 |
andel | +0.058 |
sie | +0.057 |
Neg logit contributions | |
owner | -0.021 |
hurts | -0.019 |
not | -0.019 |
hits | -0.018 |
zag | -0.017 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
ocol | +0.035 |
irs | +0.030 |
ait | +0.028 |
andel | +0.028 |
sie | +0.027 |
Neg logit contributions | |
owner | -0.010 |
not | -0.009 |
still | -0.009 |
hurts | -0.008 |
remembered | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ocol | +0.051 |
irs | +0.043 |
ait | +0.040 |
andel | +0.039 |
sie | +0.039 |
Neg logit contributions | |
owner | -0.019 |
not | -0.018 |
still | -0.017 |
hurts | -0.016 |
was | -0.016 |
Token | magical |
Token ID | 10883 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
ocol | +0.061 |
irs | +0.053 |
ait | +0.048 |
andel | +0.046 |
sie | +0.045 |
Neg logit contributions | |
owner | -0.029 |
not | -0.026 |
still | -0.025 |
was | -0.024 |
remembered | -0.023 |
Token | fair |
Token ID | 3148 |
Feature activation | -7.39e-01 |
Pos logit contributions | |
ocol | +0.061 |
irs | +0.055 |
ait | +0.049 |
andel | +0.045 |
sie | +0.044 |
Neg logit contributions | |
owner | -0.034 |
still | -0.030 |
not | -0.030 |
was | -0.030 |
flower | -0.028 |
Token | yt |
Token ID | 20760 |
Feature activation | -7.51e+00 |
Pos logit contributions | |
ocol | +0.032 |
irs | +0.028 |
ait | +0.025 |
andel | +0.023 |
year | +0.023 |
Neg logit contributions | |
owner | -0.015 |
not | -0.013 |
still | -0.013 |
was | -0.012 |
flower | -0.012 |
Token | ale |
Token ID | 1000 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
ocol | +0.241 |
irs | +0.202 |
sie | +0.171 |
ait | +0.171 |
Annie | +0.170 |
Neg logit contributions | |
owner | -0.201 |
still | -0.182 |
not | -0.182 |
hits | -0.182 |
zag | -0.181 |
Token | creature |
Token ID | 7185 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
ocol | +0.029 |
irs | +0.026 |
ait | +0.024 |
andel | +0.022 |
sie | +0.022 |
Neg logit contributions | |
owner | -0.013 |
not | -0.011 |
still | -0.011 |
was | -0.011 |
also | -0.010 |
Token | called |
Token ID | 1444 |
Feature activation | -5.00e+00 |
Pos logit contributions | |
ocol | +0.079 |
irs | +0.068 |
ait | +0.063 |
andel | +0.061 |
sie | +0.060 |
Neg logit contributions | |
owner | -0.032 |
not | -0.030 |
still | -0.029 |
was | -0.028 |
hurts | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
ocol | +0.210 |
irs | +0.183 |
inations | +0.173 |
ait | +0.170 |
sie | +0.161 |
Neg logit contributions | |
owner | -0.095 |
not | -0.092 |
was | -0.083 |
only | -0.079 |
still | -0.079 |
Token | n |
Token ID | 299 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
ocol | +0.135 |
irs | +0.120 |
ait | +0.111 |
andel | +0.102 |
sie | +0.100 |
Neg logit contributions | |
owner | -0.064 |
was | -0.055 |
not | -0.054 |
still | -0.053 |
also | -0.050 |
Token | for |
Token ID | 329 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
ocol | +0.115 |
irs | +0.101 |
andel | +0.090 |
ait | +0.089 |
sie | +0.087 |
Neg logit contributions | |
owner | -0.046 |
not | -0.042 |
still | -0.041 |
was | -0.039 |
only | -0.036 |
Token | him |
Token ID | 683 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
ocol | +0.097 |
irs | +0.085 |
ait | +0.082 |
andel | +0.082 |
sie | +0.082 |
Neg logit contributions | |
not | -0.038 |
owner | -0.036 |
still | -0.035 |
only | -0.032 |
was | -0.031 |
Token | - |
Token ID | 532 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ocol | +0.031 |
irs | +0.027 |
ait | +0.024 |
andel | +0.024 |
sie | +0.024 |
Neg logit contributions | |
owner | -0.011 |
not | -0.010 |
still | -0.010 |
was | -0.009 |
hurts | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
owner | +0.021 |
not | +0.020 |
still | +0.019 |
was | +0.018 |
also | +0.017 |
Neg logit contributions | |
ocol | -0.051 |
irs | -0.044 |
ait | -0.040 |
andel | -0.040 |
sie | -0.039 |
Token | little |
Token ID | 1310 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
ocol | +0.026 |
irs | +0.024 |
ait | +0.021 |
andel | +0.020 |
sie | +0.019 |
Neg logit contributions | |
owner | -0.015 |
still | -0.014 |
not | -0.014 |
was | -0.013 |
remembered | -0.012 |
Token | su |
Token ID | 424 |
Feature activation | -7.51e+00 |
Pos logit contributions | |
ocol | +0.079 |
irs | +0.072 |
ait | +0.062 |
andel | +0.060 |
sie | +0.059 |
Neg logit contributions | |
owner | -0.034 |
still | -0.030 |
not | -0.029 |
was | -0.028 |
hurts | -0.028 |
Token | gary |
Token ID | 14849 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
ocol | +0.174 |
Annie | +0.141 |
irs | +0.137 |
hotels | +0.131 |
props | +0.129 |
Neg logit contributions | |
uck | -0.242 |
uch | -0.219 |
owner | -0.219 |
Because | -0.213 |
still | -0.212 |
Token | treat |
Token ID | 2190 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
ocol | +0.025 |
irs | +0.024 |
ait | +0.021 |
sie | +0.020 |
inations | +0.019 |
Neg logit contributions | |
owner | -0.012 |
was | -0.011 |
still | -0.010 |
not | -0.010 |
hurts | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -8.11e-01 |
Pos logit contributions | |
ocol | +0.046 |
irs | +0.040 |
andel | +0.036 |
ait | +0.035 |
sie | +0.035 |
Neg logit contributions | |
owner | -0.021 |
not | -0.020 |
was | -0.019 |
still | -0.019 |
only | -0.017 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
ocol | +0.040 |
irs | +0.035 |
ait | +0.033 |
sie | +0.032 |
andel | +0.032 |
Neg logit contributions | |
owner | -0.009 |
not | -0.008 |
still | -0.008 |
was | -0.007 |
hurts | -0.007 |
Token | Once |
Token ID | 7454 |
Feature activation | +8.86e-02 |
Pos logit contributions | |
ocol | +0.096 |
irs | +0.081 |
ait | +0.077 |
andel | +0.076 |
sie | +0.074 |
Neg logit contributions | |
was | -0.029 |
owner | -0.027 |
still | -0.025 |
not | -0.023 |
hurts | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
ocol | +0.065 |
irs | +0.056 |
ait | +0.052 |
andel | +0.051 |
sie | +0.050 |
Neg logit contributions | |
owner | -0.025 |
not | -0.025 |
still | -0.024 |
was | -0.022 |
only | -0.021 |
Token | paper |
Token ID | 3348 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
ocol | +0.017 |
irs | +0.015 |
ait | +0.014 |
andel | +0.013 |
year | +0.013 |
Neg logit contributions | |
owner | -0.009 |
not | -0.008 |
still | -0.008 |
only | -0.007 |
head | -0.007 |
Token | became |
Token ID | 2627 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
ocol | +0.007 |
irs | +0.006 |
ait | +0.006 |
sie | +0.005 |
andel | +0.005 |
Neg logit contributions | |
owner | -0.002 |
not | -0.002 |
hurts | -0.002 |
hits | -0.002 |
still | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
ocol | +0.026 |
irs | +0.023 |
ait | +0.021 |
andel | +0.020 |
sie | +0.020 |
Neg logit contributions | |
still | -0.011 |
owner | -0.011 |
not | -0.011 |
remembered | -0.010 |
also | -0.010 |
Token | 3 |
Token ID | 513 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
ocol | +0.102 |
irs | +0.090 |
ait | +0.081 |
sie | +0.075 |
andel | +0.074 |
Neg logit contributions | |
owner | -0.054 |
still | -0.050 |
not | -0.049 |
was | -0.047 |
only | -0.045 |
Token | D |
Token ID | 35 |
Feature activation | -7.43e+00 |
Pos logit contributions | |
ocol | +0.016 |
irs | +0.015 |
ait | +0.013 |
uster | +0.013 |
andel | +0.013 |
Neg logit contributions | |
owner | -0.006 |
was | -0.006 |
not | -0.005 |
's | -0.005 |
still | -0.005 |
Token | cube |
Token ID | 23441 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
ocol | +0.250 |
irs | +0.211 |
rubble | +0.200 |
anticipation | +0.192 |
inations | +0.187 |
Neg logit contributions | |
owner | -0.172 |
zag | -0.170 |
uck | -0.165 |
flower | -0.161 |
still | -0.158 |
Token | ! |
Token ID | 0 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ocol | +0.025 |
irs | +0.021 |
ait | +0.019 |
andel | +0.019 |
sie | +0.018 |
Neg logit contributions | |
owner | -0.012 |
not | -0.011 |
still | -0.011 |
was | -0.010 |
hurts | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
ocol | +0.086 |
irs | +0.074 |
sie | +0.071 |
ait | +0.069 |
prise | +0.068 |
Neg logit contributions | |
owner | -0.017 |
was | -0.012 |
still | -0.012 |
not | -0.012 |
hits | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
ocol | +0.042 |
irs | +0.035 |
ait | +0.034 |
sie | +0.034 |
andel | +0.033 |
Neg logit contributions | |
owner | -0.005 |
not | -0.004 |
was | -0.004 |
still | -0.004 |
also | -0.003 |
Token | The |
Token ID | 464 |
Feature activation | +8.51e-01 |
Pos logit contributions | |
ocol | +0.073 |
irs | +0.062 |
ait | +0.059 |
sie | +0.057 |
andel | +0.056 |
Neg logit contributions | |
owner | -0.021 |
not | -0.018 |
still | -0.018 |
was | -0.017 |
also | -0.016 |
Token | with |
Token ID | 351 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
owner | +0.030 |
not | +0.028 |
still | +0.027 |
remembered | +0.024 |
also | +0.023 |
Neg logit contributions | |
ocol | -0.068 |
irs | -0.059 |
ait | -0.054 |
andel | -0.052 |
sie | -0.051 |
Token | her |
Token ID | 607 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ocol | +0.008 |
irs | +0.007 |
ait | +0.006 |
andel | +0.006 |
sie | +0.006 |
Neg logit contributions | |
owner | -0.003 |
not | -0.003 |
still | -0.003 |
only | -0.003 |
was | -0.003 |
Token | friends |
Token ID | 2460 |
Feature activation | +9.24e-01 |
Pos logit contributions | |
owner | +0.018 |
not | +0.017 |
hurts | +0.016 |
still | +0.016 |
hits | +0.016 |
Neg logit contributions | |
ocol | -0.056 |
irs | -0.048 |
ait | -0.045 |
andel | -0.044 |
sie | -0.043 |
Token | in |
Token ID | 287 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
owner | +0.016 |
not | +0.014 |
still | +0.014 |
was | +0.013 |
hurts | +0.013 |
Neg logit contributions | |
ocol | -0.041 |
irs | -0.035 |
ait | -0.032 |
andel | -0.032 |
sie | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
owner | +0.020 |
not | +0.019 |
still | +0.018 |
also | +0.017 |
only | +0.017 |
Neg logit contributions | |
ocol | -0.056 |
irs | -0.049 |
ait | -0.046 |
andel | -0.045 |
sie | -0.044 |
Token | sav |
Token ID | 6799 |
Feature activation | -7.40e+00 |
Pos logit contributions | |
ocol | +0.024 |
irs | +0.021 |
ait | +0.019 |
andel | +0.018 |
sie | +0.018 |
Neg logit contributions | |
owner | -0.010 |
not | -0.009 |
still | -0.009 |
hurts | -0.009 |
hits | -0.008 |
Token | anna |
Token ID | 7697 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
ocol | +0.190 |
irs | +0.157 |
ait | +0.127 |
Annie | +0.127 |
rubble | +0.126 |
Neg logit contributions | |
owner | -0.235 |
not | -0.226 |
hits | -0.225 |
still | -0.223 |
zag | -0.222 |
Token | . |
Token ID | 13 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
ocol | +0.013 |
irs | +0.011 |
ait | +0.010 |
andel | +0.010 |
sie | +0.010 |
Neg logit contributions | |
owner | -0.006 |
not | -0.006 |
still | -0.005 |
hurts | -0.005 |
hits | -0.005 |
Token | They |
Token ID | 1119 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
owner | +0.009 |
not | +0.008 |
still | +0.008 |
was | +0.007 |
hurts | +0.007 |
Neg logit contributions | |
ocol | -0.029 |
irs | -0.025 |
ait | -0.023 |
andel | -0.023 |
sie | -0.023 |
Token | are |
Token ID | 389 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
owner | +0.000 |
not | +0.000 |
still | +0.000 |
also | +0.000 |
only | +0.000 |
Neg logit contributions | |
ocol | -0.001 |
irs | -0.001 |
ait | -0.001 |
andel | -0.001 |
sie | -0.001 |
Token | all |
Token ID | 477 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
owner | +0.016 |
not | +0.015 |
still | +0.014 |
hurts | +0.014 |
hits | +0.014 |
Neg logit contributions | |
ocol | -0.036 |
irs | -0.030 |
ait | -0.028 |
andel | -0.027 |
sie | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ocol | +0.044 |
irs | +0.037 |
ait | +0.035 |
sie | +0.034 |
andel | +0.034 |
Neg logit contributions | |
owner | -0.014 |
hurts | -0.013 |
not | -0.012 |
hits | -0.012 |
zag | -0.012 |
Token | with |
Token ID | 351 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
ocol | +0.056 |
irs | +0.045 |
andel | +0.041 |
ages | +0.040 |
Annie | +0.040 |
Neg logit contributions | |
owner | -0.036 |
zag | -0.036 |
hurts | -0.033 |
flower | -0.033 |
hits | -0.033 |
Token | all |
Token ID | 477 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
ocol | +0.008 |
irs | +0.007 |
ait | +0.006 |
sie | +0.006 |
andel | +0.006 |
Neg logit contributions | |
owner | -0.005 |
not | -0.005 |
still | -0.005 |
was | -0.004 |
only | -0.004 |
Token | its |
Token ID | 663 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
ocol | +0.008 |
irs | +0.007 |
ait | +0.006 |
andel | +0.006 |
sie | +0.006 |
Neg logit contributions | |
owner | -0.004 |
not | -0.004 |
still | -0.004 |
was | -0.004 |
also | -0.004 |
Token | different |
Token ID | 1180 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
ocol | +0.005 |
irs | +0.005 |
ait | +0.004 |
sie | +0.004 |
andel | +0.004 |
Neg logit contributions | |
owner | -0.003 |
not | -0.003 |
still | -0.003 |
was | -0.003 |
hurts | -0.003 |
Token | col |
Token ID | 4033 |
Feature activation | -7.24e+00 |
Pos logit contributions | |
owner | +0.007 |
was | +0.007 |
not | +0.007 |
still | +0.006 |
hurts | +0.006 |
Neg logit contributions | |
ocol | -0.015 |
irs | -0.013 |
ait | -0.012 |
sie | -0.011 |
andel | -0.011 |
Token | ours |
Token ID | 4662 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
ocol | +0.207 |
irs | +0.172 |
sie | +0.170 |
rubble | +0.169 |
Annie | +0.169 |
Neg logit contributions | |
hits | -0.168 |
owner | -0.166 |
flower | -0.166 |
still | -0.163 |
erved | -0.163 |
Token | , |
Token ID | 11 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
ocol | +0.014 |
irs | +0.012 |
ait | +0.012 |
andel | +0.011 |
sie | +0.011 |
Neg logit contributions | |
owner | -0.004 |
not | -0.004 |
still | -0.004 |
was | -0.004 |
hurts | -0.004 |
Token | shapes |
Token ID | 15268 |
Feature activation | +9.24e-01 |
Pos logit contributions | |
ocol | +0.060 |
irs | +0.049 |
ait | +0.044 |
andel | +0.044 |
sie | +0.042 |
Neg logit contributions | |
owner | -0.034 |
zag | -0.032 |
hurts | -0.031 |
not | -0.030 |
hits | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
owner | +0.014 |
not | +0.013 |
hurts | +0.012 |
still | +0.012 |
hits | +0.012 |
Neg logit contributions | |
ocol | -0.041 |
irs | -0.035 |
ait | -0.032 |
andel | -0.032 |
sie | -0.031 |
Token | textures |
Token ID | 20028 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
ocol | +0.019 |
irs | +0.015 |
ait | +0.014 |
andel | +0.014 |
sie | +0.013 |
Neg logit contributions | |
owner | -0.012 |
not | -0.012 |
still | -0.011 |
hurts | -0.011 |
was | -0.011 |
Token | children |
Token ID | 1751 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
owner | +0.009 |
not | +0.009 |
still | +0.008 |
was | +0.008 |
also | +0.008 |
Neg logit contributions | |
ocol | -0.017 |
irs | -0.014 |
ait | -0.013 |
andel | -0.013 |
sie | -0.012 |
Token | running |
Token ID | 2491 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
owner | +0.008 |
not | +0.007 |
still | +0.007 |
hurts | +0.007 |
was | +0.007 |
Neg logit contributions | |
ocol | -0.023 |
irs | -0.019 |
ait | -0.018 |
andel | -0.017 |
sie | -0.017 |
Token | around |
Token ID | 1088 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
owner | +0.008 |
not | +0.007 |
was | +0.007 |
still | +0.007 |
hits | +0.006 |
Neg logit contributions | |
ocol | -0.023 |
irs | -0.019 |
ait | -0.018 |
andel | -0.017 |
prise | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -9.07e-01 |
Pos logit contributions | |
was | +0.005 |
owner | +0.005 |
not | +0.005 |
still | +0.005 |
only | +0.004 |
Neg logit contributions | |
ocol | -0.031 |
irs | -0.028 |
ait | -0.026 |
andel | -0.026 |
sie | -0.025 |
Token | families |
Token ID | 4172 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
ocol | +0.034 |
irs | +0.029 |
ait | +0.026 |
Annie | +0.025 |
sie | +0.025 |
Neg logit contributions | |
owner | -0.018 |
hurts | -0.017 |
hits | -0.016 |
zag | -0.016 |
not | -0.016 |
Token | pic |
Token ID | 8301 |
Feature activation | -7.23e+00 |
Pos logit contributions | |
ocol | +0.006 |
irs | +0.005 |
ait | +0.005 |
andel | +0.005 |
sie | +0.005 |
Neg logit contributions | |
owner | -0.002 |
not | -0.001 |
still | -0.001 |
hurts | -0.001 |
hits | -0.001 |
Token | n |
Token ID | 77 |
Feature activation | -8.50e-01 |
Pos logit contributions | |
ocol | +0.278 |
irs | +0.223 |
ait | +0.209 |
andel | +0.206 |
sie | +0.198 |
Neg logit contributions | |
not | -0.160 |
owner | -0.158 |
still | -0.149 |
zag | -0.146 |
was | -0.141 |
Token | icking |
Token ID | 7958 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
ocol | +0.029 |
irs | +0.027 |
ait | +0.024 |
andel | +0.022 |
inations | +0.021 |
Neg logit contributions | |
owner | -0.020 |
hurts | -0.019 |
hits | -0.018 |
tr | -0.017 |
not | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
ocol | +0.036 |
irs | +0.031 |
ait | +0.029 |
andel | +0.028 |
sie | +0.028 |
Neg logit contributions | |
owner | -0.011 |
not | -0.010 |
still | -0.010 |
was | -0.009 |
hurts | -0.009 |
Token | couples |
Token ID | 11886 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
ocol | +0.083 |
irs | +0.071 |
ait | +0.066 |
andel | +0.064 |
sie | +0.063 |
Neg logit contributions | |
owner | -0.029 |
not | -0.026 |
hurts | -0.025 |
still | -0.025 |
hits | -0.024 |
Token | st |
Token ID | 336 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
ocol | +0.102 |
irs | +0.088 |
ait | +0.081 |
andel | +0.080 |
sie | +0.078 |
Neg logit contributions | |
owner | -0.033 |
not | -0.030 |
still | -0.029 |
was | -0.027 |
hurts | -0.026 |
Token | was |
Token ID | 373 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
ocol | +0.071 |
irs | +0.061 |
ait | +0.058 |
sie | +0.057 |
andel | +0.056 |
Neg logit contributions | |
owner | -0.015 |
not | -0.012 |
hurts | -0.011 |
hits | -0.011 |
zag | -0.011 |
Token | walking |
Token ID | 6155 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
owner | +0.003 |
not | +0.003 |
still | +0.003 |
hurts | +0.003 |
remembered | +0.003 |
Neg logit contributions | |
ocol | -0.007 |
irs | -0.006 |
ait | -0.005 |
andel | -0.005 |
sie | -0.005 |
Token | through |
Token ID | 832 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
owner | +0.001 |
still | +0.001 |
not | +0.001 |
was | +0.001 |
only | +0.000 |
Neg logit contributions | |
ocol | -0.002 |
irs | -0.002 |
ait | -0.001 |
andel | -0.001 |
inations | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
ocol | +0.006 |
irs | +0.005 |
ait | +0.005 |
uster | +0.005 |
andel | +0.004 |
Neg logit contributions | |
was | -0.002 |
still | -0.001 |
not | -0.001 |
owner | -0.001 |
only | -0.001 |
Token | fair |
Token ID | 3148 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
ocol | +0.089 |
irs | +0.076 |
ait | +0.069 |
uster | +0.064 |
andel | +0.064 |
Neg logit contributions | |
owner | -0.038 |
still | -0.037 |
was | -0.035 |
's | -0.034 |
also | -0.032 |
Token | yt |
Token ID | 20760 |
Feature activation | -7.20e+00 |
Pos logit contributions | |
ocol | +0.029 |
irs | +0.025 |
ait | +0.022 |
year | +0.021 |
andel | +0.020 |
Neg logit contributions | |
owner | -0.013 |
not | -0.012 |
still | -0.012 |
was | -0.011 |
zag | -0.011 |
Token | ale |
Token ID | 1000 |
Feature activation | -9.53e-01 |
Pos logit contributions | |
ocol | +0.230 |
irs | +0.191 |
Annie | +0.160 |
ait | +0.159 |
sie | +0.158 |
Neg logit contributions | |
owner | -0.196 |
zag | -0.177 |
still | -0.176 |
not | -0.176 |
flower | -0.174 |
Token | . |
Token ID | 13 |
Feature activation | -7.86e-01 |
Pos logit contributions | |
ocol | +0.040 |
irs | +0.035 |
ait | +0.031 |
andel | +0.031 |
year | +0.030 |
Neg logit contributions | |
owner | -0.018 |
still | -0.017 |
not | -0.016 |
was | -0.016 |
's | -0.015 |
Token | She |
Token ID | 1375 |
Feature activation | -8.82e-01 |
Pos logit contributions | |
ocol | +0.036 |
irs | +0.031 |
ait | +0.028 |
sie | +0.028 |
andel | +0.027 |
Neg logit contributions | |
owner | -0.013 |
not | -0.011 |
still | -0.011 |
was | -0.011 |
only | -0.010 |
Token | fell |
Token ID | 3214 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
ocol | +0.036 |
irs | +0.032 |
ait | +0.031 |
sie | +0.031 |
Annie | +0.030 |
Neg logit contributions | |
owner | -0.012 |
hurts | -0.012 |
hits | -0.011 |
zag | -0.011 |
tr | -0.010 |
Token | asleep |
Token ID | 16039 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
owner | +0.056 |
not | +0.053 |
hurts | +0.052 |
still | +0.051 |
hits | +0.051 |
Neg logit contributions | |
ocol | -0.039 |
irs | -0.030 |
ait | -0.026 |
andel | -0.025 |
sie | -0.024 |
Token | found |
Token ID | 1043 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
ocol | +0.027 |
irs | +0.024 |
ait | +0.022 |
sie | +0.021 |
andel | +0.021 |
Neg logit contributions | |
owner | -0.009 |
not | -0.009 |
hurts | -0.008 |
zag | -0.008 |
still | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
ocol | +0.003 |
irs | +0.003 |
ait | +0.003 |
andel | +0.002 |
sie | +0.002 |
Neg logit contributions | |
owner | -0.001 |
not | -0.001 |
still | -0.001 |
was | -0.001 |
only | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | +9.01e-02 |
Pos logit contributions | |
ocol | +0.017 |
irs | +0.015 |
ait | +0.013 |
sie | +0.013 |
Annie | +0.013 |
Neg logit contributions | |
owner | -0.006 |
hurts | -0.005 |
hits | -0.005 |
not | -0.005 |
flower | -0.005 |
Token | all |
Token ID | 477 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
owner | +0.002 |
not | +0.002 |
still | +0.002 |
hurts | +0.002 |
hits | +0.002 |
Neg logit contributions | |
ocol | -0.004 |
irs | -0.003 |
ait | -0.003 |
andel | -0.003 |
sie | -0.003 |
Token | strawberry |
Token ID | 41236 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ocol | +0.020 |
irs | +0.017 |
ait | +0.016 |
andel | +0.015 |
sie | +0.015 |
Neg logit contributions | |
owner | -0.008 |
not | -0.008 |
still | -0.007 |
hurts | -0.007 |
hits | -0.007 |
Token | flav |
Token ID | 10525 |
Feature activation | -7.18e+00 |
Pos logit contributions | |
ocol | +0.059 |
irs | +0.052 |
ait | +0.045 |
andel | +0.045 |
uster | +0.044 |
Neg logit contributions | |
owner | -0.042 |
was | -0.040 |
still | -0.039 |
not | -0.039 |
flower | -0.038 |
Token | oured |
Token ID | 8167 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ocol | +0.223 |
irs | +0.183 |
ait | +0.157 |
andel | +0.155 |
sie | +0.153 |
Neg logit contributions | |
still | -0.198 |
owner | -0.196 |
not | -0.187 |
flower | -0.181 |
hits | -0.181 |
Token | , |
Token ID | 11 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
ocol | +0.058 |
irs | +0.053 |
ait | +0.049 |
andel | +0.047 |
sie | +0.046 |
Neg logit contributions | |
owner | -0.020 |
not | -0.019 |
was | -0.019 |
still | -0.019 |
only | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | +6.99e-01 |
Pos logit contributions | |
ocol | +0.080 |
irs | +0.066 |
ait | +0.060 |
andel | +0.059 |
Annie | +0.058 |
Neg logit contributions | |
owner | -0.034 |
zag | -0.031 |
hurts | -0.031 |
hits | -0.030 |
not | -0.029 |
Token | favourite |
Token ID | 12507 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
owner | +0.013 |
not | +0.012 |
still | +0.012 |
hurts | +0.011 |
was | +0.011 |
Neg logit contributions | |
ocol | -0.029 |
irs | -0.026 |
ait | -0.023 |
andel | -0.023 |
sie | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
owner | +0.012 |
not | +0.011 |
still | +0.011 |
was | +0.011 |
hurts | +0.011 |
Neg logit contributions | |
ocol | -0.026 |
irs | -0.022 |
ait | -0.020 |
andel | -0.020 |
sie | -0.019 |
Token | took |
Token ID | 1718 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
owner | +0.006 |
not | +0.005 |
zag | +0.005 |
hurts | +0.005 |
flower | +0.005 |
Neg logit contributions | |
ocol | -0.017 |
irs | -0.015 |
ait | -0.014 |
sie | -0.014 |
andel | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
ocol | +0.045 |
irs | +0.038 |
inations | +0.037 |
ait | +0.037 |
sie | +0.036 |
Neg logit contributions | |
owner | -0.013 |
not | -0.013 |
only | -0.012 |
still | -0.012 |
was | -0.011 |
Token | bike |
Token ID | 7161 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
owner | +0.021 |
not | +0.020 |
still | +0.019 |
hurts | +0.018 |
hits | +0.017 |
Neg logit contributions | |
ocol | -0.059 |
irs | -0.051 |
ait | -0.047 |
andel | -0.046 |
sie | -0.045 |
Token | out |
Token ID | 503 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
owner | +0.021 |
not | +0.020 |
hurts | +0.019 |
remembered | +0.019 |
still | +0.019 |
Neg logit contributions | |
ocol | -0.060 |
irs | -0.052 |
ait | -0.049 |
andel | -0.047 |
sie | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
ocol | +0.058 |
irs | +0.049 |
ait | +0.044 |
sie | +0.044 |
andel | +0.043 |
Neg logit contributions | |
owner | -0.026 |
not | -0.025 |
still | -0.024 |
was | -0.023 |
hurts | -0.022 |
Token | tried |
Token ID | 3088 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
owner | +0.004 |
zag | +0.004 |
not | +0.004 |
hurts | +0.004 |
flower | +0.004 |
Neg logit contributions | |
ocol | -0.011 |
irs | -0.010 |
ait | -0.009 |
sie | -0.009 |
Annie | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
owner | +0.054 |
not | +0.048 |
hurts | +0.048 |
still | +0.047 |
hits | +0.047 |
Neg logit contributions | |
ocol | -0.136 |
irs | -0.118 |
ait | -0.107 |
andel | -0.104 |
sie | -0.102 |
Token | pedal |
Token ID | 26667 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
owner | +0.012 |
hurts | +0.011 |
flower | +0.011 |
not | +0.011 |
hits | +0.011 |
Neg logit contributions | |
ocol | -0.023 |
irs | -0.020 |
ait | -0.018 |
andel | -0.017 |
sie | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
owner | +0.048 |
not | +0.045 |
still | +0.044 |
hurts | +0.042 |
hits | +0.041 |
Neg logit contributions | |
ocol | -0.100 |
irs | -0.085 |
ait | -0.077 |
andel | -0.075 |
sie | -0.074 |
Token | But |
Token ID | 887 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
owner | +0.003 |
not | +0.003 |
hurts | +0.002 |
flower | +0.002 |
still | +0.002 |
Neg logit contributions | |
ocol | -0.008 |
irs | -0.007 |
ait | -0.006 |
andel | -0.006 |
sie | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
owner | +0.013 |
hurts | +0.012 |
zag | +0.012 |
flower | +0.012 |
hits | +0.012 |
Neg logit contributions | |
ocol | -0.023 |
irs | -0.019 |
ait | -0.017 |
sie | -0.016 |
Annie | -0.016 |
Token | 's |
Token ID | 338 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
ocol | +0.008 |
irs | +0.007 |
ait | +0.006 |
andel | +0.006 |
sie | +0.006 |
Neg logit contributions | |
owner | -0.003 |
not | -0.003 |
still | -0.003 |
hurts | -0.003 |
hits | -0.003 |
Token | feathers |
Token ID | 27400 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
ocol | +0.060 |
irs | +0.054 |
ait | +0.048 |
andel | +0.046 |
sie | +0.046 |
Neg logit contributions | |
owner | -0.030 |
not | -0.027 |
still | -0.026 |
hurts | -0.024 |
was | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
ocol | +0.023 |
irs | +0.019 |
ait | +0.017 |
andel | +0.017 |
sie | +0.017 |
Neg logit contributions | |
owner | -0.012 |
not | -0.011 |
hits | -0.011 |
hurts | -0.011 |
still | -0.010 |
Token | He |
Token ID | 679 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
ocol | +0.032 |
irs | +0.028 |
ait | +0.025 |
Annie | +0.024 |
andel | +0.024 |
Neg logit contributions | |
owner | -0.015 |
hurts | -0.014 |
not | -0.014 |
zag | -0.014 |
flower | -0.014 |
Token | realized |
Token ID | 6939 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
owner | +0.002 |
zag | +0.001 |
hurts | +0.001 |
tr | +0.001 |
hits | +0.001 |
Neg logit contributions | |
ocol | -0.005 |
irs | -0.004 |
sie | -0.004 |
ait | -0.004 |
Annie | -0.004 |
Token | that |
Token ID | 326 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
owner | +0.007 |
not | +0.006 |
still | +0.006 |
hurts | +0.005 |
was | +0.005 |
Neg logit contributions | |
ocol | -0.025 |
irs | -0.022 |
ait | -0.021 |
andel | -0.020 |
sie | -0.020 |
Token | sometimes |
Token ID | 3360 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
owner | +0.060 |
zag | +0.058 |
hits | +0.054 |
hurts | +0.054 |
tr | +0.054 |
Neg logit contributions | |
ocol | -0.110 |
irs | -0.090 |
ait | -0.080 |
Annie | -0.079 |
sie | -0.078 |
Token | it |
Token ID | 340 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
owner | +0.046 |
not | +0.042 |
still | +0.041 |
hurts | +0.040 |
zag | +0.040 |
Neg logit contributions | |
ocol | -0.091 |
irs | -0.077 |
ait | -0.070 |
andel | -0.068 |
sie | -0.067 |
Token | takes |
Token ID | 2753 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
owner | +0.004 |
tr | +0.003 |
zag | +0.003 |
flower | +0.003 |
burn | +0.003 |
Neg logit contributions | |
ocol | -0.012 |
irs | -0.011 |
sie | -0.010 |
ait | -0.010 |
Annie | -0.010 |
Token | only |
Token ID | 691 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
ocol | +0.041 |
irs | +0.035 |
ait | +0.033 |
andel | +0.031 |
sie | +0.030 |
Neg logit contributions | |
owner | -0.021 |
not | -0.020 |
still | -0.020 |
was | -0.019 |
hurts | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
owner | +0.032 |
not | +0.029 |
still | +0.028 |
hurts | +0.028 |
hits | +0.027 |
Neg logit contributions | |
ocol | -0.064 |
irs | -0.054 |
ait | -0.049 |
andel | -0.048 |
sie | -0.047 |
Token | do |
Token ID | 466 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
ocol | +0.090 |
irs | +0.077 |
ait | +0.071 |
andel | +0.070 |
sie | +0.068 |
Neg logit contributions | |
owner | -0.036 |
not | -0.034 |
still | -0.033 |
hurts | -0.031 |
was | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +8.57e-01 |
Pos logit contributions | |
ocol | +0.032 |
irs | +0.028 |
ait | +0.026 |
andel | +0.025 |
sie | +0.025 |
Neg logit contributions | |
owner | -0.012 |
not | -0.012 |
still | -0.011 |
was | -0.011 |
hurts | -0.010 |
Token | He |
Token ID | 679 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
owner | +0.012 |
not | +0.011 |
still | +0.010 |
hurts | +0.009 |
was | +0.009 |
Neg logit contributions | |
ocol | -0.041 |
irs | -0.035 |
ait | -0.032 |
andel | -0.032 |
sie | -0.032 |
Token | was |
Token ID | 373 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
owner | +0.006 |
not | +0.005 |
hurts | +0.005 |
hits | +0.005 |
zag | +0.005 |
Neg logit contributions | |
ocol | -0.018 |
irs | -0.015 |
ait | -0.014 |
sie | -0.014 |
andel | -0.014 |
Token | so |
Token ID | 523 |
Feature activation | +6.55e-01 |
Pos logit contributions | |
owner | +0.034 |
hurts | +0.031 |
flower | +0.031 |
zag | +0.030 |
hits | +0.030 |
Neg logit contributions | |
ocol | -0.068 |
irs | -0.056 |
ait | -0.052 |
andel | -0.051 |
sie | -0.050 |
Token | bored |
Token ID | 17533 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
owner | +0.013 |
not | +0.012 |
hurts | +0.012 |
flower | +0.012 |
hits | +0.012 |
Neg logit contributions | |
ocol | -0.026 |
irs | -0.022 |
ait | -0.020 |
andel | -0.019 |
sie | -0.019 |
Token | that |
Token ID | 326 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
owner | +0.029 |
not | +0.027 |
still | +0.025 |
was | +0.023 |
hurts | +0.023 |
Neg logit contributions | |
ocol | -0.121 |
irs | -0.105 |
ait | -0.098 |
andel | -0.096 |
sie | -0.095 |
Token | he |
Token ID | 339 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
owner | +0.031 |
not | +0.028 |
still | +0.027 |
hurts | +0.027 |
zag | +0.027 |
Neg logit contributions | |
ocol | -0.075 |
irs | -0.064 |
ait | -0.058 |
andel | -0.057 |
sie | -0.056 |
Token | decided |
Token ID | 3066 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
ocol | +0.043 |
irs | +0.037 |
ait | +0.035 |
andel | +0.034 |
sie | +0.033 |
Neg logit contributions | |
owner | -0.010 |
not | -0.009 |
still | -0.009 |
hurts | -0.008 |
hits | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
ocol | +0.008 |
irs | +0.007 |
ait | +0.006 |
sie | +0.006 |
andel | +0.006 |
Neg logit contributions | |
owner | -0.004 |
flower | -0.003 |
zag | -0.003 |
hurts | -0.003 |
hits | -0.003 |
Token | go |
Token ID | 467 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ocol | +0.013 |
irs | +0.011 |
ait | +0.010 |
andel | +0.010 |
sie | +0.010 |
Neg logit contributions | |
owner | -0.006 |
not | -0.005 |
still | -0.005 |
hurts | -0.005 |
hits | -0.005 |
Token | gone |
Token ID | 3750 |
Feature activation | +9.87e-02 |
Pos logit contributions | |
owner | +0.006 |
not | +0.005 |
hurts | +0.005 |
hits | +0.005 |
zag | +0.005 |
Neg logit contributions | |
ocol | -0.014 |
irs | -0.012 |
ait | -0.011 |
andel | -0.011 |
sie | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
owner | +0.002 |
not | +0.002 |
still | +0.002 |
hurts | +0.002 |
was | +0.001 |
Neg logit contributions | |
ocol | -0.004 |
irs | -0.004 |
ait | -0.003 |
andel | -0.003 |
sie | -0.003 |
Token | He |
Token ID | 679 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
ocol | +0.012 |
irs | +0.010 |
ait | +0.009 |
andel | +0.009 |
sie | +0.009 |
Neg logit contributions | |
owner | -0.005 |
not | -0.004 |
still | -0.004 |
hurts | -0.004 |
was | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
owner | +0.006 |
zag | +0.005 |
not | +0.005 |
flower | +0.005 |
hits | +0.005 |
Neg logit contributions | |
ocol | -0.021 |
irs | -0.018 |
ait | -0.017 |
sie | -0.017 |
Annie | -0.017 |
Token | so |
Token ID | 523 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
owner | +0.027 |
not | +0.024 |
hurts | +0.024 |
hits | +0.023 |
still | +0.023 |
Neg logit contributions | |
ocol | -0.058 |
irs | -0.049 |
ait | -0.045 |
andel | -0.044 |
sie | -0.043 |
Token | sad |
Token ID | 6507 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
owner | +0.013 |
not | +0.012 |
hits | +0.012 |
zag | +0.012 |
hurts | +0.012 |
Neg logit contributions | |
ocol | -0.026 |
irs | -0.022 |
ait | -0.020 |
andel | -0.019 |
sie | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
owner | +0.041 |
not | +0.036 |
still | +0.035 |
hurts | +0.034 |
hits | +0.033 |
Neg logit contributions | |
ocol | -0.133 |
irs | -0.115 |
ait | -0.107 |
andel | -0.104 |
sie | -0.102 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
owner | +0.004 |
not | +0.003 |
zag | +0.003 |
still | +0.003 |
hits | +0.003 |
Neg logit contributions | |
ocol | -0.009 |
irs | -0.008 |
ait | -0.007 |
sie | -0.007 |
andel | -0.007 |
Token | everywhere |
Token ID | 8347 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
ocol | +0.061 |
irs | +0.053 |
ait | +0.050 |
andel | +0.047 |
sie | +0.045 |
Neg logit contributions | |
still | -0.024 |
not | -0.024 |
owner | -0.023 |
hurts | -0.021 |
was | -0.021 |
Token | for |
Token ID | 329 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
ocol | +0.053 |
irs | +0.047 |
ait | +0.044 |
andel | +0.042 |
sie | +0.042 |
Neg logit contributions | |
not | -0.017 |
still | -0.015 |
owner | -0.014 |
was | -0.014 |
hurts | -0.012 |
Token | his |
Token ID | 465 |
Feature activation | -8.58e-02 |
Pos logit contributions | |
ocol | +0.087 |
irs | +0.075 |
ait | +0.072 |
year | +0.072 |
andel | +0.070 |
Neg logit contributions | |
not | -0.048 |
still | -0.044 |
only | -0.042 |
was | -0.041 |
owner | -0.041 |
Token | " |
Token ID | 366 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
ocol | +0.038 |
irs | +0.034 |
ait | +0.031 |
andel | +0.031 |
sie | +0.031 |
Neg logit contributions | |
not | -0.007 |
owner | -0.007 |
still | -0.005 |
was | -0.005 |
also | -0.004 |
Token | Wow |
Token ID | 22017 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
owner | +0.023 |
not | +0.021 |
still | +0.020 |
hurts | +0.020 |
hits | +0.019 |
Neg logit contributions | |
ocol | -0.057 |
irs | -0.049 |
ait | -0.045 |
andel | -0.044 |
sie | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | +8.45e-01 |
Pos logit contributions | |
ocol | +0.013 |
irs | +0.011 |
andel | +0.011 |
year | +0.011 |
ait | +0.011 |
Neg logit contributions | |
not | -0.001 |
still | -0.001 |
was | -0.001 |
's | -0.001 |
owner | -0.001 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
not | +0.017 |
owner | +0.016 |
still | +0.015 |
was | +0.015 |
only | +0.015 |
Neg logit contributions | |
ocol | -0.036 |
irs | -0.031 |
ait | -0.029 |
andel | -0.029 |
uster | -0.027 |
Token | ! |
Token ID | 0 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
owner | +0.019 |
not | +0.018 |
still | +0.017 |
hurts | +0.017 |
hits | +0.016 |
Neg logit contributions | |
ocol | -0.048 |
irs | -0.041 |
ait | -0.037 |
andel | -0.037 |
sie | -0.036 |
Token | Your |
Token ID | 3406 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
ocol | +0.063 |
irs | +0.052 |
ait | +0.047 |
andel | +0.046 |
sie | +0.044 |
Neg logit contributions | |
owner | -0.038 |
not | -0.035 |
still | -0.034 |
hurts | -0.034 |
hits | -0.033 |
Token | sand |
Token ID | 6450 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
owner | +0.066 |
not | +0.062 |
still | +0.060 |
was | +0.059 |
hits | +0.056 |
Neg logit contributions | |
ocol | -0.106 |
irs | -0.091 |
ait | -0.082 |
andel | -0.079 |
sie | -0.078 |
Token | castle |
Token ID | 18676 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
owner | +0.049 |
not | +0.043 |
still | +0.042 |
hurts | +0.042 |
remembered | +0.041 |
Neg logit contributions | |
ocol | -0.100 |
irs | -0.082 |
ait | -0.074 |
andel | -0.072 |
sie | -0.072 |
Token | is |
Token ID | 318 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
owner | +0.016 |
not | +0.012 |
zag | +0.011 |
hurts | +0.011 |
flower | +0.010 |
Neg logit contributions | |
ocol | -0.068 |
irs | -0.059 |
ait | -0.054 |
inations | -0.052 |
sie | -0.052 |
Token | so |
Token ID | 523 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
ocol | +0.001 |
irs | +0.001 |
ait | +0.001 |
andel | +0.001 |
sie | +0.001 |
Neg logit contributions | |
owner | -0.001 |
not | -0.001 |
still | -0.001 |
was | -0.001 |
also | -0.001 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ocol | +0.008 |
irs | +0.007 |
ait | +0.007 |
andel | +0.006 |
sie | +0.006 |
Neg logit contributions | |
not | -0.004 |
still | -0.004 |
owner | -0.003 |
was | -0.003 |
also | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
owner | +0.008 |
not | +0.008 |
still | +0.007 |
hurts | +0.007 |
zag | +0.007 |
Neg logit contributions | |
ocol | -0.022 |
irs | -0.019 |
ait | -0.018 |
andel | -0.017 |
sie | -0.017 |
Token | roll |
Token ID | 4836 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
owner | +0.007 |
not | +0.006 |
still | +0.006 |
hurts | +0.006 |
zag | +0.006 |
Neg logit contributions | |
ocol | -0.012 |
irs | -0.010 |
ait | -0.009 |
andel | -0.009 |
sie | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
owner | +0.028 |
not | +0.026 |
still | +0.025 |
hurts | +0.025 |
hits | +0.024 |
Neg logit contributions | |
ocol | -0.069 |
irs | -0.059 |
ait | -0.054 |
andel | -0.053 |
sie | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
owner | +0.007 |
not | +0.006 |
still | +0.006 |
was | +0.005 |
hurts | +0.005 |
Neg logit contributions | |
ocol | -0.027 |
irs | -0.024 |
ait | -0.022 |
andel | -0.022 |
sie | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +9.78e-02 |
Pos logit contributions | |
owner | +0.008 |
not | +0.007 |
still | +0.006 |
was | +0.006 |
hurts | +0.005 |
Neg logit contributions | |
ocol | -0.041 |
irs | -0.035 |
ait | -0.033 |
andel | -0.033 |
sie | -0.032 |
Token | So |
Token ID | 2396 |
Feature activation | +7.58e-01 |
Pos logit contributions | |
owner | +0.001 |
not | +0.001 |
still | +0.001 |
was | +0.001 |
also | +0.001 |
Neg logit contributions | |
ocol | -0.005 |
irs | -0.004 |
ait | -0.004 |
andel | -0.004 |
sie | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
owner | +0.011 |
not | +0.010 |
still | +0.009 |
hurts | +0.009 |
hits | +0.008 |
Neg logit contributions | |
ocol | -0.035 |
irs | -0.031 |
ait | -0.028 |
andel | -0.028 |
sie | -0.027 |
Token | rolled |
Token ID | 11686 |
Feature activation | +9.30e-01 |
Pos logit contributions | |
owner | +0.007 |
not | +0.007 |
hurts | +0.006 |
flower | +0.006 |
zag | +0.006 |
Neg logit contributions | |
ocol | -0.020 |
irs | -0.017 |
sie | -0.017 |
ait | -0.016 |
andel | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
owner | +0.016 |
not | +0.015 |
still | +0.015 |
hurts | +0.014 |
hits | +0.014 |
Neg logit contributions | |
ocol | -0.041 |
irs | -0.035 |
ait | -0.033 |
andel | -0.031 |
sie | -0.031 |
Token | rolled |
Token ID | 11686 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ocol | +0.027 |
irs | +0.024 |
ait | +0.022 |
sie | +0.021 |
andel | +0.021 |
Neg logit contributions | |
owner | -0.007 |
not | -0.007 |
still | -0.006 |
hurts | -0.006 |
zag | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
owner | +0.019 |
not | +0.018 |
still | +0.017 |
was | +0.016 |
hits | +0.015 |
Neg logit contributions | |
ocol | -0.057 |
irs | -0.049 |
ait | -0.047 |
prise | -0.044 |
andel | -0.044 |
Token | picture |
Token ID | 4286 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
ocol | +0.058 |
irs | +0.049 |
ait | +0.044 |
sie | +0.043 |
andel | +0.043 |
Neg logit contributions | |
owner | -0.030 |
not | -0.025 |
still | -0.024 |
was | -0.024 |
tr | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
owner | +0.000 |
not | +0.000 |
was | +0.000 |
still | +0.000 |
hurts | +0.000 |
Neg logit contributions | |
ocol | -0.001 |
irs | -0.000 |
ait | -0.000 |
andel | -0.000 |
sie | -0.000 |
Token | grandma |
Token ID | 49890 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
ocol | +0.076 |
irs | +0.063 |
andel | +0.061 |
sie | +0.060 |
ait | +0.059 |
Neg logit contributions | |
not | -0.037 |
owner | -0.036 |
still | -0.034 |
also | -0.032 |
only | -0.031 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
owner | +0.000 |
not | +0.000 |
still | +0.000 |
was | +0.000 |
's | +0.000 |
Neg logit contributions | |
ocol | -0.001 |
irs | -0.001 |
andel | -0.001 |
ait | -0.001 |
sie | -0.001 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
owner | +0.002 |
not | +0.002 |
still | +0.002 |
was | +0.002 |
hurts | +0.002 |
Neg logit contributions | |
ocol | -0.006 |
irs | -0.005 |
ait | -0.005 |
andel | -0.005 |
sie | -0.005 |
Token | shouts |
Token ID | 34757 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
owner | +0.003 |
not | +0.003 |
still | +0.002 |
hurts | +0.002 |
zag | +0.002 |
Neg logit contributions | |
ocol | -0.008 |
irs | -0.007 |
ait | -0.007 |
andel | -0.007 |
sie | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
ocol | +0.051 |
irs | +0.045 |
ait | +0.041 |
andel | +0.039 |
sie | +0.039 |
Neg logit contributions | |
owner | -0.017 |
not | -0.016 |
still | -0.015 |
was | -0.014 |
hurts | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +9.67e-01 |
Pos logit contributions | |
owner | +0.003 |
not | +0.003 |
still | +0.003 |
hurts | +0.003 |
zag | +0.002 |
Neg logit contributions | |
ocol | -0.011 |
irs | -0.010 |
ait | -0.009 |
andel | -0.009 |
sie | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +5.40e-01 |
Pos logit contributions | |
owner | +0.010 |
not | +0.009 |
still | +0.008 |
hurts | +0.007 |
was | +0.007 |
Neg logit contributions | |
ocol | -0.050 |
irs | -0.043 |
ait | -0.040 |
andel | -0.040 |
sie | -0.039 |
Token | " |
Token ID | 1 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
owner | +0.008 |
not | +0.008 |
still | +0.007 |
hurts | +0.007 |
hits | +0.007 |
Neg logit contributions | |
ocol | -0.024 |
irs | -0.021 |
ait | -0.019 |
andel | -0.019 |
sie | -0.018 |
Token | Hello |
Token ID | 15496 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
owner | +0.014 |
not | +0.014 |
hurts | +0.013 |
hits | +0.013 |
still | +0.013 |
Neg logit contributions | |
ocol | -0.029 |
irs | -0.024 |
ait | -0.022 |
sie | -0.021 |
andel | -0.021 |
Token | baby |
Token ID | 5156 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
ocol | +0.027 |
irs | +0.025 |
ait | +0.021 |
andel | +0.021 |
inations | +0.020 |
Neg logit contributions | |
owner | -0.012 |
still | -0.010 |
not | -0.010 |
was | -0.009 |
friend | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
ocol | +0.008 |
irs | +0.008 |
andel | +0.007 |
ait | +0.007 |
inations | +0.006 |
Neg logit contributions | |
owner | -0.003 |
still | -0.003 |
was | -0.003 |
not | -0.003 |
flower | -0.003 |
Token | The |
Token ID | 383 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
ocol | +0.020 |
irs | +0.017 |
inations | +0.016 |
andel | +0.016 |
sie | +0.016 |
Neg logit contributions | |
owner | -0.004 |
still | -0.003 |
not | -0.003 |
was | -0.003 |
also | -0.003 |
Token | baby |
Token ID | 5156 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
owner | +0.032 |
not | +0.029 |
still | +0.028 |
hurts | +0.027 |
was | +0.026 |
Neg logit contributions | |
ocol | -0.074 |
irs | -0.064 |
ait | -0.058 |
andel | -0.057 |
sie | -0.056 |
Token | 's |
Token ID | 338 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
owner | +0.016 |
not | +0.016 |
hits | +0.016 |
zag | +0.015 |
hurts | +0.013 |
Neg logit contributions | |
ocol | -0.098 |
ait | -0.083 |
irs | -0.082 |
sie | -0.082 |
Annie | -0.081 |
Token | name |
Token ID | 1438 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
owner | +0.026 |
not | +0.025 |
hits | +0.024 |
hurts | +0.024 |
still | +0.024 |
Neg logit contributions | |
ocol | -0.081 |
irs | -0.069 |
ait | -0.065 |
andel | -0.063 |
sie | -0.062 |
Token | was |
Token ID | 373 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
ocol | +0.031 |
irs | +0.027 |
ait | +0.026 |
Annie | +0.025 |
sie | +0.025 |
Neg logit contributions | |
owner | -0.004 |
not | -0.003 |
zag | -0.003 |
hits | -0.003 |
hurts | -0.003 |
Token | Mary |
Token ID | 5335 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
owner | +0.007 |
not | +0.007 |
still | +0.006 |
was | +0.006 |
hurts | +0.006 |
Neg logit contributions | |
ocol | -0.010 |
irs | -0.008 |
ait | -0.007 |
andel | -0.007 |
sie | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
ocol | +0.033 |
irs | +0.029 |
ait | +0.026 |
andel | +0.026 |
sie | +0.025 |
Neg logit contributions | |
owner | -0.009 |
not | -0.009 |
still | -0.008 |
hurts | -0.007 |
was | -0.007 |
Token | Mary |
Token ID | 5335 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
owner | +0.007 |
not | +0.007 |
still | +0.006 |
hurts | +0.006 |
hits | +0.006 |
Neg logit contributions | |
ocol | -0.012 |
irs | -0.010 |
ait | -0.009 |
andel | -0.009 |
Annie | -0.008 |
Token | loved |
Token ID | 6151 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
owner | +0.004 |
not | +0.004 |
hits | +0.003 |
zag | +0.003 |
hurts | +0.003 |
Neg logit contributions | |
ocol | -0.023 |
irs | -0.020 |
ait | -0.019 |
sie | -0.019 |
Annie | -0.018 |
Token | their |
Token ID | 511 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
ocol | +0.041 |
irs | +0.034 |
andel | +0.033 |
inations | +0.032 |
sie | +0.032 |
Neg logit contributions | |
was | -0.020 |
owner | -0.020 |
not | -0.019 |
still | -0.019 |
's | -0.018 |
Token | room |
Token ID | 2119 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
owner | +0.001 |
still | +0.001 |
not | +0.001 |
was | +0.001 |
next | +0.001 |
Neg logit contributions | |
ocol | -0.003 |
irs | -0.003 |
ait | -0.002 |
andel | -0.002 |
sie | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
owner | +0.006 |
not | +0.006 |
still | +0.006 |
was | +0.005 |
also | +0.005 |
Neg logit contributions | |
ocol | -0.018 |
irs | -0.016 |
andel | -0.014 |
ait | -0.014 |
sie | -0.014 |
Token | They |
Token ID | 1119 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
owner | +0.003 |
not | +0.002 |
still | +0.002 |
hurts | +0.002 |
was | +0.002 |
Neg logit contributions | |
ocol | -0.008 |
irs | -0.007 |
ait | -0.007 |
andel | -0.007 |
sie | -0.006 |
Token | are |
Token ID | 389 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
still | +0.003 |
not | +0.003 |
also | +0.003 |
owner | +0.003 |
only | +0.003 |
Neg logit contributions | |
ocol | -0.009 |
irs | -0.008 |
andel | -0.007 |
ait | -0.007 |
sie | -0.007 |
Token | proud |
Token ID | 6613 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
not | +0.008 |
owner | +0.008 |
still | +0.008 |
also | +0.007 |
only | +0.006 |
Neg logit contributions | |
ocol | -0.022 |
irs | -0.019 |
andel | -0.017 |
ait | -0.017 |
sie | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
owner | +0.001 |
not | +0.001 |
still | +0.001 |
was | +0.001 |
also | +0.001 |
Neg logit contributions | |
ocol | -0.002 |
irs | -0.002 |
ait | -0.001 |
sie | -0.001 |
andel | -0.001 |
Token | They |
Token ID | 1119 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
owner | +0.009 |
not | +0.008 |
still | +0.007 |
was | +0.007 |
hurts | +0.007 |
Neg logit contributions | |
ocol | -0.028 |
irs | -0.024 |
ait | -0.022 |
andel | -0.022 |
sie | -0.021 |
Token | say |
Token ID | 910 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ocol | +0.006 |
irs | +0.005 |
andel | +0.005 |
ait | +0.005 |
sie | +0.005 |
Neg logit contributions | |
still | -0.002 |
not | -0.002 |
owner | -0.002 |
also | -0.002 |
was | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
ocol | +0.065 |
irs | +0.058 |
ait | +0.054 |
andel | +0.054 |
sie | +0.052 |
Neg logit contributions | |
not | -0.013 |
owner | -0.011 |
was | -0.009 |
still | -0.009 |
hurts | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | +9.60e-01 |
Pos logit contributions | |
ocol | +0.013 |
irs | +0.011 |
ait | +0.010 |
andel | +0.010 |
sie | +0.010 |
Neg logit contributions | |
owner | -0.003 |
not | -0.003 |
still | -0.003 |
hurts | -0.003 |
was | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
ocol | +0.089 |
irs | +0.077 |
inations | +0.074 |
ait | +0.073 |
andel | +0.073 |
Neg logit contributions | |
owner | -0.024 |
hits | -0.022 |
not | -0.022 |
hit | -0.021 |
still | -0.020 |
Token | room |
Token ID | 2119 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
ocol | +0.063 |
irs | +0.059 |
ait | +0.052 |
year | +0.050 |
sie | +0.050 |
Neg logit contributions | |
owner | -0.035 |
hurts | -0.031 |
hits | -0.031 |
head | -0.029 |
not | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
owner | +0.002 |
not | +0.002 |
still | +0.002 |
was | +0.001 |
hurts | +0.001 |
Neg logit contributions | |
ocol | -0.005 |
irs | -0.004 |
andel | -0.004 |
ait | -0.004 |
sie | -0.004 |
Token | He |
Token ID | 679 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
owner | +0.006 |
not | +0.006 |
still | +0.005 |
hurts | +0.005 |
was | +0.005 |
Neg logit contributions | |
ocol | -0.019 |
irs | -0.016 |
ait | -0.015 |
andel | -0.015 |
sie | -0.015 |
Token | said |
Token ID | 531 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
owner | +0.003 |
flower | +0.003 |
zag | +0.003 |
not | +0.003 |
hurts | +0.003 |
Neg logit contributions | |
ocol | -0.008 |
irs | -0.007 |
ait | -0.006 |
sie | -0.006 |
rubble | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
ocol | +0.054 |
irs | +0.047 |
ait | +0.043 |
andel | +0.042 |
sie | +0.042 |
Neg logit contributions | |
owner | -0.014 |
not | -0.012 |
still | -0.012 |
hurts | -0.011 |
hits | -0.011 |
Token | " |
Token ID | 366 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
ocol | +0.007 |
irs | +0.006 |
ait | +0.006 |
andel | +0.006 |
sie | +0.006 |
Neg logit contributions | |
owner | -0.002 |
not | -0.002 |
still | -0.002 |
hurts | -0.002 |
hits | -0.002 |
Token | It |
Token ID | 1026 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
owner | +0.032 |
not | +0.032 |
zag | +0.030 |
still | +0.029 |
hits | +0.029 |
Neg logit contributions | |
ocol | -0.053 |
irs | -0.044 |
ait | -0.040 |
sie | -0.040 |
andel | -0.039 |
Token | doesn |
Token ID | 1595 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
owner | +0.007 |
flower | +0.005 |
not | +0.004 |
still | +0.004 |
hits | +0.004 |
Neg logit contributions | |
ocol | -0.032 |
irs | -0.029 |
ait | -0.027 |
Annie | -0.027 |
sie | -0.027 |
Token | 't |
Token ID | 470 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
ocol | +0.139 |
irs | +0.118 |
ait | +0.111 |
sie | +0.108 |
andel | +0.108 |
Neg logit contributions | |
owner | -0.041 |
not | -0.040 |
still | -0.036 |
hurts | -0.035 |
hits | -0.033 |
Token | matter |
Token ID | 2300 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
not | +0.013 |
owner | +0.011 |
still | +0.011 |
hurt | +0.011 |
hurts | +0.010 |
Neg logit contributions | |
ocol | -0.033 |
irs | -0.028 |
ait | -0.026 |
andel | -0.026 |
year | -0.026 |
Token | more |
Token ID | 517 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ocol | +0.052 |
irs | +0.045 |
ait | +0.042 |
sie | +0.041 |
year | +0.041 |
Neg logit contributions | |
owner | -0.025 |
not | -0.024 |
hurts | -0.024 |
still | -0.024 |
was | -0.023 |
Token | ," |
Token ID | 553 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
not | +0.028 |
owner | +0.028 |
still | +0.026 |
hurts | +0.025 |
was | +0.025 |
Neg logit contributions | |
ocol | -0.056 |
irs | -0.048 |
ait | -0.044 |
andel | -0.043 |
sie | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
ocol | +0.027 |
irs | +0.024 |
ait | +0.022 |
andel | +0.022 |
inations | +0.021 |
Neg logit contributions | |
owner | -0.009 |
not | -0.009 |
still | -0.009 |
was | -0.009 |
only | -0.008 |
Token | he |
Token ID | 339 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
ocol | +0.005 |
irs | +0.004 |
ait | +0.004 |
andel | +0.004 |
sie | +0.004 |
Neg logit contributions | |
owner | -0.002 |
not | -0.002 |
still | -0.002 |
hurts | -0.002 |
hits | -0.002 |
Token | located |
Token ID | 5140 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
ocol | +0.045 |
irs | +0.039 |
ait | +0.037 |
sie | +0.036 |
andel | +0.035 |
Neg logit contributions | |
owner | -0.016 |
hurts | -0.016 |
not | -0.015 |
hates | -0.014 |
hits | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
ocol | +0.060 |
irs | +0.052 |
ait | +0.048 |
inations | +0.047 |
andel | +0.047 |
Neg logit contributions | |
owner | -0.022 |
not | -0.020 |
still | -0.019 |
was | -0.018 |
only | -0.018 |
Token | nearby |
Token ID | 6716 |
Feature activation | -8.13e-01 |
Pos logit contributions | |
ocol | +0.067 |
irs | +0.058 |
ait | +0.052 |
andel | +0.050 |
sie | +0.049 |
Neg logit contributions | |
owner | -0.034 |
not | -0.030 |
still | -0.030 |
was | -0.028 |
hurts | -0.027 |
Token | pond |
Token ID | 16723 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
ocol | +0.038 |
irs | +0.036 |
inations | +0.031 |
ait | +0.031 |
year | +0.030 |
Neg logit contributions | |
owner | -0.013 |
was | -0.011 |
still | -0.011 |
be | -0.010 |
head | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
ocol | +0.015 |
irs | +0.013 |
ait | +0.012 |
andel | +0.012 |
sie | +0.012 |
Neg logit contributions | |
owner | -0.005 |
not | -0.005 |
still | -0.005 |
was | -0.004 |
hurts | -0.004 |
Token | soak |
Token ID | 33182 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
ocol | +0.067 |
irs | +0.058 |
ait | +0.053 |
andel | +0.053 |
sie | +0.050 |
Neg logit contributions | |
not | -0.026 |
still | -0.026 |
owner | -0.026 |
also | -0.024 |
was | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
ocol | +0.009 |
irs | +0.008 |
ait | +0.007 |
andel | +0.007 |
inations | +0.007 |
Neg logit contributions | |
owner | -0.003 |
still | -0.003 |
was | -0.003 |
not | -0.003 |
hurts | -0.002 |
Token | showed |
Token ID | 3751 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
hurts | +0.009 |
tr | +0.009 |
burn | +0.008 |
ained | +0.008 |
flower | +0.008 |
Neg logit contributions | |
sie | -0.013 |
ocol | -0.012 |
ait | -0.012 |
irs | -0.012 |
Annie | -0.012 |
Token | him |
Token ID | 683 |
Feature activation | +8.31e-01 |
Pos logit contributions | |
ocol | +0.017 |
irs | +0.014 |
ait | +0.013 |
andel | +0.013 |
sie | +0.012 |
Neg logit contributions | |
owner | -0.010 |
not | -0.009 |
still | -0.009 |
hurts | -0.009 |
zag | -0.009 |
Token | how |
Token ID | 703 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
owner | +0.019 |
flower | +0.018 |
tr | +0.018 |
zag | +0.017 |
still | +0.017 |
Neg logit contributions | |
ocol | -0.027 |
irs | -0.024 |
sie | -0.020 |
Annie | -0.020 |
andel | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
ocol | +0.083 |
irs | +0.071 |
ait | +0.065 |
andel | +0.064 |
sie | +0.063 |
Neg logit contributions | |
owner | -0.031 |
not | -0.028 |
still | -0.027 |
hurts | -0.026 |
zag | -0.026 |
Token | move |
Token ID | 1445 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
owner | +0.004 |
tr | +0.004 |
flower | +0.004 |
hurts | +0.003 |
burn | +0.003 |
Neg logit contributions | |
ocol | -0.007 |
irs | -0.006 |
inations | -0.005 |
ait | -0.005 |
sie | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
owner | +0.029 |
not | +0.027 |
still | +0.026 |
hurts | +0.025 |
hits | +0.025 |
Neg logit contributions | |
ocol | -0.073 |
irs | -0.062 |
ait | -0.057 |
andel | -0.056 |
sie | -0.055 |
Token | pieces |
Token ID | 5207 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
ocol | +0.071 |
irs | +0.059 |
ait | +0.053 |
andel | +0.051 |
sie | +0.050 |
Neg logit contributions | |
owner | -0.048 |
not | -0.046 |
still | -0.044 |
hurts | -0.044 |
hits | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
ocol | +0.049 |
irs | +0.042 |
ait | +0.039 |
andel | +0.038 |
sie | +0.037 |
Neg logit contributions | |
owner | -0.019 |
not | -0.018 |
still | -0.017 |
hurts | -0.017 |
hits | -0.016 |
Token | they |
Token ID | 484 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
owner | +0.006 |
hurts | +0.006 |
flower | +0.006 |
hates | +0.006 |
Because | +0.006 |
Neg logit contributions | |
ocol | -0.011 |
irs | -0.010 |
Annie | -0.009 |
ages | -0.009 |
inations | -0.008 |
Token | started |
Token ID | 2067 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
ocol | +0.055 |
irs | +0.050 |
ait | +0.045 |
sie | +0.045 |
andel | +0.043 |
Neg logit contributions | |
owner | -0.022 |
hurts | -0.022 |
not | -0.020 |
hits | -0.019 |
zag | -0.019 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
owner | +0.034 |
tr | +0.031 |
flower | +0.030 |
hurts | +0.030 |
still | +0.030 |
Neg logit contributions | |
ocol | -0.065 |
irs | -0.057 |
ait | -0.048 |
andel | -0.048 |
Annie | -0.047 |
Token | brave |
Token ID | 14802 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
owner | +0.031 |
not | +0.029 |
still | +0.028 |
hurts | +0.027 |
hits | +0.026 |
Neg logit contributions | |
ocol | -0.059 |
irs | -0.051 |
ait | -0.046 |
andel | -0.044 |
sie | -0.044 |
Token | dinosaur |
Token ID | 28257 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
owner | +0.024 |
was | +0.021 |
not | +0.020 |
still | +0.020 |
hits | +0.020 |
Neg logit contributions | |
ocol | -0.054 |
irs | -0.049 |
ait | -0.044 |
sie | -0.041 |
andel | -0.041 |
Token | bell |
Token ID | 8966 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
burn | +0.011 |
hurts | +0.011 |
hits | +0.010 |
flower | +0.010 |
owner | +0.009 |
Neg logit contributions | |
ocol | -0.043 |
irs | -0.038 |
sie | -0.037 |
ait | -0.036 |
ages | -0.036 |
Token | owed |
Token ID | 6972 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
hurts | +0.010 |
owner | +0.010 |
hates | +0.010 |
not | +0.010 |
Fighting | +0.009 |
Neg logit contributions | |
ocol | -0.027 |
irs | -0.023 |
ait | -0.021 |
year | -0.021 |
ages | -0.021 |
Token | in |
Token ID | 287 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
ocol | +0.030 |
irs | +0.027 |
ait | +0.025 |
andel | +0.023 |
sie | +0.023 |
Neg logit contributions | |
owner | -0.009 |
not | -0.008 |
still | -0.008 |
was | -0.007 |
only | -0.007 |
Token | surprise |
Token ID | 5975 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
owner | +0.006 |
not | +0.006 |
was | +0.006 |
still | +0.005 |
only | +0.005 |
Neg logit contributions | |
ocol | -0.017 |
irs | -0.015 |
ait | -0.015 |
sie | -0.014 |
inations | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
ocol | +0.122 |
irs | +0.111 |
sie | +0.103 |
andel | +0.102 |
chie | +0.100 |
Neg logit contributions | |
owner | -0.024 |
was | -0.020 |
not | -0.019 |
still | -0.017 |
and | -0.014 |
Token | " |
Token ID | 366 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
ocol | +0.054 |
irs | +0.047 |
ait | +0.044 |
sie | +0.043 |
andel | +0.042 |
Neg logit contributions | |
owner | -0.014 |
not | -0.012 |
still | -0.011 |
was | -0.011 |
hits | -0.010 |
Token | How |
Token ID | 2437 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
owner | +0.039 |
not | +0.038 |
hurts | +0.036 |
zag | +0.035 |
hits | +0.035 |
Neg logit contributions | |
ocol | -0.072 |
irs | -0.060 |
ait | -0.054 |
andel | -0.053 |
sie | -0.053 |
Token | can |
Token ID | 460 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
ocol | +0.151 |
irs | +0.129 |
ait | +0.118 |
andel | +0.118 |
sie | +0.115 |
Neg logit contributions | |
owner | -0.057 |
not | -0.057 |
still | -0.053 |
was | -0.050 |
also | -0.049 |
Token | I |
Token ID | 314 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ocol | +0.112 |
irs | +0.094 |
ait | +0.086 |
andel | +0.083 |
sie | +0.081 |
Neg logit contributions | |
owner | -0.064 |
not | -0.062 |
still | -0.058 |
hurts | -0.058 |
hits | -0.056 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
not | +0.022 |
owner | +0.021 |
still | +0.021 |
also | +0.019 |
remembered | +0.018 |
Neg logit contributions | |
ocol | -0.051 |
irs | -0.045 |
andel | -0.040 |
ait | -0.040 |
uster | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +7.40e-03 |
Pos logit contributions | |
owner | +0.015 |
not | +0.014 |
still | +0.013 |
hurts | +0.012 |
hits | +0.012 |
Neg logit contributions | |
ocol | -0.046 |
irs | -0.040 |
ait | -0.037 |
andel | -0.036 |
sie | -0.035 |
Token | too |
Token ID | 1165 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
owner | +0.000 |
not | +0.000 |
still | +0.000 |
hurts | +0.000 |
hits | +0.000 |
Neg logit contributions | |
ocol | -0.000 |
irs | -0.000 |
ait | -0.000 |
andel | -0.000 |
sie | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +9.99e-01 |
Pos logit contributions | |
owner | +0.028 |
not | +0.027 |
still | +0.026 |
also | +0.023 |
was | +0.023 |
Neg logit contributions | |
ocol | -0.074 |
irs | -0.065 |
ait | -0.058 |
andel | -0.057 |
sie | -0.056 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
owner | +0.013 |
not | +0.012 |
still | +0.011 |
hurts | +0.011 |
hits | +0.011 |
Neg logit contributions | |
ocol | -0.046 |
irs | -0.040 |
ait | -0.037 |
andel | -0.036 |
sie | -0.036 |
Token | L |
Token ID | 43 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
ocol | +0.073 |
irs | +0.061 |
ait | +0.058 |
andel | +0.057 |
sie | +0.055 |
Neg logit contributions | |
owner | -0.029 |
was | -0.028 |
still | -0.027 |
not | -0.026 |
hits | -0.025 |
Token | ily |
Token ID | 813 |
Feature activation | -8.47e-02 |
Pos logit contributions | |
ocol | +0.093 |
irs | +0.072 |
andel | +0.071 |
chie | +0.066 |
rubble | +0.065 |
Neg logit contributions | |
not | -0.045 |
was | -0.044 |
also | -0.044 |
owner | -0.043 |
's | -0.042 |
Token | liked |
Token ID | 8288 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
ocol | +0.004 |
irs | +0.003 |
ait | +0.003 |
Annie | +0.003 |
sie | +0.003 |
Neg logit contributions | |
owner | -0.001 |
zag | -0.001 |
neither | -0.001 |
not | -0.001 |
hurts | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
ocol | +0.031 |
irs | +0.028 |
sie | +0.026 |
ait | +0.026 |
andel | +0.026 |
Neg logit contributions | |
owner | -0.006 |
not | -0.005 |
still | -0.003 |
was | -0.003 |
next | -0.003 |
Token | help |
Token ID | 1037 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
ocol | +0.002 |
irs | +0.002 |
ait | +0.001 |
andel | +0.001 |
sie | +0.001 |
Neg logit contributions | |
not | -0.001 |
owner | -0.001 |
still | -0.001 |
was | -0.001 |
hurts | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | +8.95e-02 |
Pos logit contributions | |
ocol | +0.004 |
irs | +0.004 |
ait | +0.003 |
andel | +0.003 |
sie | +0.003 |
Neg logit contributions | |
owner | -0.002 |
not | -0.002 |
still | -0.002 |
hurts | -0.002 |
was | -0.002 |
Token | sing |
Token ID | 1702 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ocol | +0.032 |
irs | +0.027 |
ait | +0.025 |
andel | +0.024 |
sie | +0.024 |
Neg logit contributions | |
owner | -0.013 |
not | -0.012 |
still | -0.012 |
hurts | -0.011 |
hits | -0.011 |
Token | her |
Token ID | 607 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
owner | +0.017 |
not | +0.016 |
still | +0.015 |
was | +0.014 |
only | +0.013 |
Neg logit contributions | |
ocol | -0.060 |
irs | -0.051 |
ait | -0.048 |
andel | -0.046 |
sie | -0.046 |
Token | most |
Token ID | 749 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
owner | +0.002 |
not | +0.002 |
hits | +0.002 |
still | +0.002 |
hurts | +0.002 |
Neg logit contributions | |
ocol | -0.005 |
irs | -0.004 |
ait | -0.004 |
andel | -0.004 |
sie | -0.004 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
owner | +0.003 |
not | +0.003 |
still | +0.003 |
was | +0.002 |
hurts | +0.002 |
Neg logit contributions | |
ocol | -0.004 |
irs | -0.004 |
ait | -0.003 |
andel | -0.003 |
sie | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ocol | +0.021 |
irs | +0.018 |
ait | +0.016 |
andel | +0.015 |
sie | +0.015 |
Neg logit contributions | |
owner | -0.010 |
not | -0.009 |
still | -0.009 |
hurts | -0.009 |
was | -0.008 |
Token | upl |
Token ID | 28996 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
ocol | +0.046 |
irs | +0.040 |
Annie | +0.036 |
ages | +0.035 |
sie | +0.035 |
Neg logit contributions | |
owner | -0.025 |
zag | -0.023 |
flower | -0.023 |
hurts | -0.022 |
hits | -0.022 |
Token | ifting |
Token ID | 13309 |
Feature activation | +5.52e-01 |
Pos logit contributions | |
ocol | +0.133 |
irs | +0.104 |
ait | +0.098 |
sie | +0.094 |
andel | +0.091 |
Neg logit contributions | |
not | -0.109 |
owner | -0.106 |
still | -0.105 |
zag | -0.102 |
also | -0.101 |
Token | song |
Token ID | 3496 |
Feature activation | +8.46e-01 |
Pos logit contributions | |
owner | +0.010 |
not | +0.010 |
still | +0.009 |
was | +0.009 |
hurts | +0.009 |
Neg logit contributions | |
ocol | -0.024 |
irs | -0.021 |
ait | -0.019 |
andel | -0.018 |
sie | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
owner | +0.014 |
not | +0.013 |
still | +0.012 |
was | +0.012 |
zag | +0.011 |
Neg logit contributions | |
ocol | -0.039 |
irs | -0.033 |
ait | -0.031 |
andel | -0.030 |
sie | -0.030 |
Token | Slowly |
Token ID | 46484 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
ocol | +0.059 |
irs | +0.051 |
ait | +0.047 |
sie | +0.046 |
inations | +0.046 |
Neg logit contributions | |
owner | -0.012 |
not | -0.010 |
still | -0.010 |
was | -0.010 |
only | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
not | +0.003 |
owner | +0.003 |
still | +0.003 |
was | +0.003 |
only | +0.003 |
Neg logit contributions | |
ocol | -0.022 |
irs | -0.019 |
ait | -0.018 |
year | -0.018 |
andel | -0.018 |
Token | they |
Token ID | 484 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
owner | +0.024 |
not | +0.022 |
still | +0.022 |
hurts | +0.022 |
hits | +0.021 |
Neg logit contributions | |
ocol | -0.051 |
irs | -0.044 |
ait | -0.040 |
andel | -0.039 |
sie | -0.038 |
Token | built |
Token ID | 3170 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
ocol | +0.048 |
irs | +0.042 |
ait | +0.038 |
sie | +0.036 |
andel | +0.036 |
Neg logit contributions | |
owner | -0.024 |
hurts | -0.023 |
not | -0.022 |
hits | -0.021 |
still | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
not | +0.011 |
owner | +0.011 |
still | +0.010 |
was | +0.010 |
hits | +0.009 |
Neg logit contributions | |
ocol | -0.033 |
irs | -0.028 |
ait | -0.026 |
sie | -0.026 |
andel | -0.026 |
Token | biggest |
Token ID | 4094 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
owner | +0.023 |
not | +0.021 |
hurts | +0.021 |
remembered | +0.020 |
flower | +0.020 |
Neg logit contributions | |
ocol | -0.044 |
irs | -0.037 |
ait | -0.034 |
andel | -0.033 |
rubble | -0.032 |
Token | tower |
Token ID | 10580 |
Feature activation | +5.23e-01 |
Pos logit contributions | |
owner | +0.045 |
not | +0.043 |
hurts | +0.041 |
still | +0.041 |
zag | +0.039 |
Neg logit contributions | |
ocol | -0.091 |
irs | -0.076 |
ait | -0.069 |
andel | -0.068 |
sie | -0.065 |
Token | they |
Token ID | 484 |
Feature activation | -5.44e+00 |
Pos logit contributions | |
owner | +0.009 |
not | +0.008 |
hurts | +0.008 |
still | +0.008 |
remembered | +0.008 |
Neg logit contributions | |
ocol | -0.022 |
irs | -0.018 |
ait | -0.017 |
andel | -0.016 |
sie | -0.016 |
Token | had |
Token ID | 550 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
ocol | +0.268 |
irs | +0.231 |
ait | +0.214 |
andel | +0.214 |
sie | +0.204 |
Neg logit contributions | |
owner | -0.068 |
not | -0.066 |
still | -0.064 |
was | -0.057 |
remembered | -0.057 |
Token | ever |
Token ID | 1683 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
ocol | +0.081 |
irs | +0.068 |
ait | +0.061 |
andel | +0.058 |
sie | +0.057 |
Neg logit contributions | |
owner | -0.051 |
hurts | -0.047 |
not | -0.047 |
still | -0.046 |
hits | -0.045 |
Token | made |
Token ID | 925 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
ocol | +0.069 |
irs | +0.057 |
ait | +0.052 |
andel | +0.051 |
sie | +0.050 |
Neg logit contributions | |
owner | -0.041 |
not | -0.039 |
still | -0.038 |
hurts | -0.037 |
was | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
ocol | +0.105 |
irs | +0.089 |
ait | +0.082 |
sie | +0.082 |
andel | +0.081 |
Neg logit contributions | |
owner | -0.037 |
not | -0.035 |
still | -0.032 |
was | -0.031 |
zag | -0.029 |
Token | They |
Token ID | 1119 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
ocol | +0.002 |
irs | +0.001 |
ait | +0.001 |
andel | +0.001 |
sie | +0.001 |
Neg logit contributions | |
owner | -0.001 |
not | -0.001 |
still | -0.001 |
hurts | -0.001 |
hits | -0.001 |
Token | park |
Token ID | 3952 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
ocol | +0.024 |
irs | +0.020 |
andel | +0.018 |
ait | +0.018 |
sie | +0.018 |
Neg logit contributions | |
not | -0.015 |
zag | -0.015 |
owner | -0.015 |
hurts | -0.014 |
hits | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
hurts | +0.014 |
hits | +0.013 |
tr | +0.012 |
owner | +0.012 |
burn | +0.012 |
Neg logit contributions | |
ocol | -0.024 |
irs | -0.020 |
Annie | -0.019 |
ait | -0.019 |
ages | -0.019 |
Token | look |
Token ID | 804 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
owner | +0.027 |
Because | +0.024 |
hurts | +0.024 |
hates | +0.024 |
ung | +0.023 |
Neg logit contributions | |
ocol | -0.045 |
irs | -0.037 |
Annie | -0.035 |
sie | -0.034 |
ait | -0.032 |
Token | for |
Token ID | 329 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
ocol | +0.082 |
irs | +0.075 |
ait | +0.071 |
andel | +0.066 |
prise | +0.065 |
Neg logit contributions | |
still | -0.018 |
owner | -0.017 |
not | -0.017 |
was | -0.014 |
's | -0.014 |
Token | something |
Token ID | 1223 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
ocol | +0.021 |
irs | +0.018 |
ait | +0.017 |
sie | +0.016 |
andel | +0.016 |
Neg logit contributions | |
owner | -0.010 |
not | -0.009 |
was | -0.008 |
still | -0.008 |
only | -0.008 |
Token | y |
Token ID | 331 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
ocol | +0.144 |
irs | +0.123 |
ait | +0.113 |
andel | +0.111 |
sie | +0.107 |
Neg logit contributions | |
owner | -0.067 |
not | -0.064 |
still | -0.060 |
was | -0.056 |
hurts | -0.056 |
Token | ummy |
Token ID | 13513 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
ocol | +0.168 |
irs | +0.138 |
ait | +0.122 |
sie | +0.118 |
andel | +0.115 |
Neg logit contributions | |
owner | -0.144 |
not | -0.135 |
still | -0.132 |
hurts | -0.130 |
hits | -0.126 |
Token | to |
Token ID | 284 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
ocol | +0.096 |
irs | +0.082 |
ait | +0.076 |
andel | +0.074 |
sie | +0.073 |
Neg logit contributions | |
owner | -0.037 |
not | -0.034 |
still | -0.032 |
hurts | -0.031 |
hits | -0.030 |
Token | eat |
Token ID | 4483 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
ocol | +0.115 |
irs | +0.097 |
ait | +0.089 |
andel | +0.087 |
sie | +0.085 |
Neg logit contributions | |
owner | -0.049 |
not | -0.045 |
still | -0.043 |
hurts | -0.042 |
hits | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
ocol | +0.088 |
irs | +0.078 |
ait | +0.074 |
andel | +0.073 |
uster | +0.071 |
Neg logit contributions | |
owner | -0.024 |
not | -0.023 |
was | -0.022 |
still | -0.022 |
next | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
flower | +0.015 |
remembered | +0.015 |
owner | +0.015 |
hurts | +0.014 |
not | +0.014 |
Neg logit contributions | |
ocol | -0.031 |
irs | -0.028 |
ait | -0.024 |
andel | -0.024 |
Annie | -0.023 |
Token | of |
Token ID | 286 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
ocol | +0.040 |
irs | +0.034 |
ait | +0.030 |
andel | +0.029 |
sie | +0.028 |
Neg logit contributions | |
owner | -0.025 |
not | -0.025 |
still | -0.024 |
was | -0.023 |
hurts | -0.022 |
Token | all |
Token ID | 477 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
owner | +0.005 |
not | +0.004 |
still | +0.004 |
hurts | +0.004 |
was | +0.004 |
Neg logit contributions | |
ocol | -0.009 |
irs | -0.008 |
ait | -0.007 |
andel | -0.007 |
sie | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
ocol | +0.024 |
irs | +0.020 |
ait | +0.018 |
andel | +0.018 |
sie | +0.017 |
Neg logit contributions | |
owner | -0.009 |
not | -0.008 |
still | -0.008 |
was | -0.008 |
also | -0.008 |
Token | tiny |
Token ID | 7009 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
ocol | +0.056 |
irs | +0.048 |
ait | +0.047 |
uster | +0.044 |
andel | +0.044 |
Neg logit contributions | |
owner | -0.028 |
not | -0.027 |
still | -0.027 |
was | -0.026 |
only | -0.025 |
Token | things |
Token ID | 1243 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
ocol | +0.009 |
irs | +0.007 |
ait | +0.007 |
andel | +0.007 |
uster | +0.006 |
Neg logit contributions | |
owner | -0.005 |
was | -0.005 |
still | -0.005 |
not | -0.005 |
hurts | -0.005 |
Token | she |
Token ID | 673 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
ocol | +0.159 |
irs | +0.136 |
andel | +0.127 |
ait | +0.126 |
sie | +0.120 |
Neg logit contributions | |
not | -0.051 |
owner | -0.050 |
still | -0.049 |
was | -0.045 |
also | -0.042 |
Token | would |
Token ID | 561 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
ocol | +0.243 |
irs | +0.209 |
ait | +0.193 |
sie | +0.189 |
andel | +0.184 |
Neg logit contributions | |
owner | -0.081 |
hurts | -0.071 |
not | -0.071 |
zag | -0.069 |
hits | -0.069 |
Token | see |
Token ID | 766 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
ocol | +0.131 |
irs | +0.112 |
ait | +0.101 |
andel | +0.097 |
sie | +0.097 |
Neg logit contributions | |
owner | -0.063 |
hurts | -0.056 |
not | -0.055 |
hits | -0.054 |
zag | -0.052 |
Token | with |
Token ID | 351 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
ocol | +0.120 |
irs | +0.103 |
ait | +0.095 |
andel | +0.095 |
sie | +0.091 |
Neg logit contributions | |
owner | -0.040 |
not | -0.038 |
still | -0.037 |
was | -0.035 |
hurts | -0.032 |
Token | her |
Token ID | 607 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
ocol | +0.080 |
irs | +0.068 |
ait | +0.065 |
andel | +0.064 |
sie | +0.062 |
Neg logit contributions | |
owner | -0.035 |
not | -0.034 |
still | -0.034 |
was | -0.031 |
only | -0.030 |
Token | microscope |
Token ID | 36396 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ocol | +0.010 |
irs | +0.008 |
ait | +0.007 |
andel | +0.007 |
sie | +0.007 |
Neg logit contributions | |
owner | -0.005 |
not | -0.005 |
still | -0.005 |
hurts | -0.004 |
was | -0.004 |
Token | Annie |
Token ID | 25860 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
owner | +0.012 |
not | +0.012 |
still | +0.011 |
was | +0.011 |
remembered | +0.011 |
Neg logit contributions | |
ocol | -0.033 |
irs | -0.028 |
andel | -0.025 |
ait | -0.025 |
sie | -0.024 |
Token | said |
Token ID | 531 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
owner | +0.003 |
hits | +0.003 |
hurts | +0.003 |
zag | +0.003 |
tr | +0.003 |
Neg logit contributions | |
ocol | -0.008 |
irs | -0.007 |
sie | -0.006 |
ait | -0.006 |
Annie | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | -6.39e-02 |
Pos logit contributions | |
ocol | +0.075 |
irs | +0.063 |
ait | +0.057 |
andel | +0.055 |
sie | +0.055 |
Neg logit contributions | |
owner | -0.036 |
not | -0.033 |
hurts | -0.032 |
hits | -0.032 |
zag | -0.032 |
Token | was |
Token ID | 373 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
ocol | +0.003 |
irs | +0.002 |
Annie | +0.002 |
sie | +0.002 |
Nancy | +0.002 |
Neg logit contributions | |
owner | -0.001 |
hits | -0.001 |
zag | -0.001 |
erved | -0.000 |
flower | -0.000 |
Token | called |
Token ID | 1444 |
Feature activation | -4.72e+00 |
Pos logit contributions | |
owner | +0.009 |
not | +0.009 |
still | +0.008 |
was | +0.008 |
only | +0.008 |
Neg logit contributions | |
ocol | -0.021 |
irs | -0.018 |
ait | -0.017 |
andel | -0.016 |
sie | -0.016 |
Token | � |
Token ID | 6184 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
ocol | +0.194 |
irs | +0.174 |
inations | +0.166 |
ait | +0.159 |
rubble | +0.155 |
Neg logit contributions | |
not | -0.090 |
owner | -0.087 |
only | -0.077 |
still | -0.075 |
was | -0.075 |
Token | � |
Token ID | 95 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
ocol | +0.189 |
irs | +0.157 |
andel | +0.132 |
ait | +0.130 |
rubble | +0.128 |
Neg logit contributions | |
still | -0.130 |
Because | -0.128 |
not | -0.125 |
zag | -0.123 |
owner | -0.122 |
Token | € |
Token ID | 26391 |
Feature activation | +6.52e-01 |
Pos logit contributions | |
ocol | +0.092 |
irs | +0.074 |
andel | +0.065 |
ait | +0.065 |
sie | +0.062 |
Neg logit contributions | |
owner | -0.061 |
not | -0.060 |
still | -0.059 |
zag | -0.056 |
was | -0.056 |
Token | ˜ |
Token ID | 41185 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
owner | +0.013 |
not | +0.012 |
hurts | +0.011 |
still | +0.011 |
hits | +0.011 |
Neg logit contributions | |
ocol | -0.025 |
irs | -0.022 |
ait | -0.020 |
sie | -0.019 |
andel | -0.019 |
Token | Pretty |
Token ID | 35700 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
owner | +0.024 |
not | +0.023 |
still | +0.021 |
was | +0.021 |
flower | +0.020 |
Neg logit contributions | |
ocol | -0.057 |
irs | -0.049 |
andel | -0.044 |
ait | -0.044 |
rubble | -0.042 |
Token | Flowers |
Token ID | 30036 |
Feature activation | +5.86e-01 |
Pos logit contributions | |
owner | +0.007 |
not | +0.007 |
still | +0.007 |
was | +0.006 |
also | +0.006 |
Neg logit contributions | |
ocol | -0.014 |
irs | -0.012 |
ait | -0.011 |
andel | -0.011 |
sie | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
ocol | +0.069 |
irs | +0.059 |
andel | +0.055 |
ait | +0.055 |
uster | +0.052 |
Neg logit contributions | |
not | -0.024 |
still | -0.023 |
owner | -0.023 |
was | -0.020 |
's | -0.020 |
Token | appreciate |
Token ID | 9144 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
owner | +0.010 |
not | +0.008 |
hurts | +0.008 |
zag | +0.008 |
hits | +0.008 |
Neg logit contributions | |
ocol | -0.022 |
irs | -0.018 |
ait | -0.017 |
sie | -0.016 |
andel | -0.016 |
Token | them |
Token ID | 606 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
ocol | +0.018 |
ait | +0.015 |
andel | +0.015 |
irs | +0.015 |
uster | +0.014 |
Neg logit contributions | |
not | -0.009 |
still | -0.008 |
was | -0.008 |
owner | -0.008 |
also | -0.008 |
Token | for |
Token ID | 329 |
Feature activation | -8.98e-01 |
Pos logit contributions | |
ocol | +0.126 |
irs | +0.108 |
andel | +0.102 |
ait | +0.101 |
uster | +0.095 |
Neg logit contributions | |
not | -0.045 |
still | -0.043 |
owner | -0.041 |
was | -0.037 |
also | -0.036 |
Token | what |
Token ID | 644 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
ocol | +0.038 |
andel | +0.032 |
irs | +0.031 |
ait | +0.031 |
uster | +0.030 |
Neg logit contributions | |
not | -0.019 |
still | -0.016 |
never | -0.016 |
owner | -0.016 |
also | -0.015 |
Token | they |
Token ID | 484 |
Feature activation | -5.76e+00 |
Pos logit contributions | |
ocol | +0.086 |
irs | +0.073 |
andel | +0.068 |
ait | +0.068 |
sie | +0.064 |
Neg logit contributions | |
owner | -0.038 |
not | -0.037 |
still | -0.034 |
was | -0.034 |
hurts | -0.033 |
Token | are |
Token ID | 389 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
ocol | +0.260 |
irs | +0.224 |
ait | +0.206 |
andel | +0.202 |
sie | +0.198 |
Neg logit contributions | |
owner | -0.088 |
not | -0.081 |
still | -0.077 |
hurts | -0.074 |
hits | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -7.15e-01 |
Pos logit contributions | |
ocol | +0.156 |
irs | +0.134 |
ait | +0.123 |
andel | +0.122 |
sie | +0.117 |
Neg logit contributions | |
not | -0.065 |
owner | -0.065 |
still | -0.061 |
also | -0.056 |
was | -0.055 |
Token | She |
Token ID | 1375 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
ocol | +0.035 |
irs | +0.029 |
andel | +0.028 |
sie | +0.028 |
ait | +0.027 |
Neg logit contributions | |
owner | -0.009 |
not | -0.008 |
still | -0.008 |
was | -0.008 |
also | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
ocol | +0.040 |
irs | +0.035 |
ait | +0.034 |
sie | +0.033 |
Annie | +0.032 |
Neg logit contributions | |
owner | -0.014 |
hurts | -0.014 |
hits | -0.013 |
zag | -0.013 |
not | -0.013 |
Token | that |
Token ID | 326 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
ocol | +0.113 |
irs | +0.100 |
ait | +0.092 |
andel | +0.090 |
sie | +0.088 |
Neg logit contributions | |
not | -0.028 |
owner | -0.027 |
still | -0.024 |
was | -0.022 |
only | -0.021 |