Token | . |
Token ID | 13 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
Help | +0.015 |
tearing | +0.014 |
What | +0.013 |
smashing | +0.013 |
free | +0.013 |
Neg logit contributions | |
rey | -0.041 |
asha | -0.039 |
aleb | -0.039 |
onia | -0.038 |
iri | -0.037 |
Token | " |
Token ID | 366 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
rey | +0.059 |
aleb | +0.058 |
iper | +0.057 |
asha | +0.057 |
i | +0.057 |
Neg logit contributions | |
What | -0.020 |
Help | -0.020 |
Whe | -0.018 |
Where | -0.018 |
Was | -0.017 |
Token | Come |
Token ID | 16773 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
iper | +0.002 |
rey | +0.002 |
asha | +0.002 |
i | +0.002 |
ishment | +0.002 |
Neg logit contributions | |
What | -0.001 |
Help | -0.001 |
Was | -0.001 |
Do | -0.001 |
Where | -0.001 |
Token | on |
Token ID | 319 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
rey | +0.013 |
aleb | +0.012 |
iper | +0.012 |
asha | +0.012 |
i | +0.012 |
Neg logit contributions | |
Help | -0.007 |
Was | -0.007 |
smashing | -0.007 |
tearing | -0.007 |
What | -0.006 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
Help | +0.018 |
What | +0.017 |
Was | +0.017 |
Do | +0.017 |
smashing | +0.016 |
Neg logit contributions | |
rey | -0.040 |
aleb | -0.039 |
asha | -0.038 |
iper | -0.038 |
aman | -0.038 |
Token | ", |
Token ID | 1600 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
Help | +0.003 |
Do | +0.003 |
What | +0.003 |
Was | +0.003 |
Where | +0.003 |
Neg logit contributions | |
rey | -0.008 |
aleb | -0.008 |
asha | -0.008 |
iper | -0.008 |
onia | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
otton | +0.182 |
aleb | +0.179 |
ust | +0.178 |
wy | +0.177 |
athy | +0.167 |
Neg logit contributions | |
" | -0.070 |
pulling | -0.056 |
pushing | -0.055 |
to | -0.054 |
The | -0.054 |
Token | Brad |
Token ID | 8114 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
aman | +0.110 |
asha | +0.109 |
ust | +0.109 |
aleb | +0.108 |
otton | +0.108 |
Neg logit contributions | |
Help | -0.059 |
What | -0.054 |
Where | -0.054 |
Whe | -0.052 |
" | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
rey | +0.021 |
aleb | +0.020 |
asha | +0.020 |
aman | +0.020 |
iper | +0.019 |
Neg logit contributions | |
Help | -0.010 |
What | -0.009 |
Was | -0.009 |
Where | -0.009 |
Do | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
aleb | +0.074 |
rey | +0.074 |
iper | +0.074 |
asha | +0.073 |
aman | +0.072 |
Neg logit contributions | |
Help | -0.026 |
What | -0.025 |
Was | -0.023 |
Where | -0.023 |
Do | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
rey | +0.049 |
aleb | +0.047 |
asha | +0.046 |
iper | +0.046 |
aman | +0.045 |
Neg logit contributions | |
Help | -0.023 |
What | -0.022 |
Was | -0.022 |
Do | -0.022 |
smashing | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
rey | +0.010 |
asha | +0.009 |
aleb | +0.009 |
iper | +0.009 |
iri | +0.009 |
Neg logit contributions | |
Help | -0.004 |
tearing | -0.004 |
What | -0.004 |
knocking | -0.004 |
destroying | -0.004 |
Token | " |
Token ID | 366 |
Feature activation | +9.94e-01 |
Pos logit contributions | |
rey | +0.107 |
aleb | +0.106 |
asha | +0.104 |
iper | +0.103 |
aman | +0.103 |
Neg logit contributions | |
What | -0.024 |
Help | -0.022 |
Where | -0.020 |
Was | -0.020 |
Do | -0.019 |
Token | Wait |
Token ID | 21321 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
iper | +0.037 |
ishment | +0.035 |
rey | +0.034 |
i | +0.034 |
icity | +0.034 |
Neg logit contributions | |
What | -0.018 |
Help | -0.017 |
Are | -0.017 |
Was | -0.017 |
Do | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
rey | +0.001 |
aleb | +0.001 |
iper | +0.001 |
asha | +0.001 |
aman | +0.001 |
Neg logit contributions | |
Help | -0.000 |
What | -0.000 |
Was | -0.000 |
Do | -0.000 |
smashing | -0.000 |
Token | Brad |
Token ID | 8114 |
Feature activation | +9.71e-02 |
Pos logit contributions | |
ishment | +0.041 |
onia | +0.041 |
asha | +0.041 |
aman | +0.040 |
andra | +0.040 |
Neg logit contributions | |
Help | -0.027 |
What | -0.026 |
Do | -0.025 |
Where | -0.024 |
Whe | -0.023 |
Token | ," |
Token ID | 553 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
rey | +0.004 |
aleb | +0.004 |
asha | +0.004 |
aman | +0.004 |
iper | +0.004 |
Neg logit contributions | |
Help | -0.001 |
What | -0.001 |
Was | -0.001 |
Do | -0.001 |
Where | -0.001 |
Token | she |
Token ID | 673 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
iri | +0.149 |
otton | +0.149 |
aleb | +0.147 |
ust | +0.147 |
wy | +0.144 |
Neg logit contributions | |
What | -0.083 |
Help | -0.079 |
" | -0.078 |
Are | -0.077 |
Where | -0.074 |
Token | said |
Token ID | 531 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
aleb | +0.075 |
ishment | +0.072 |
aman | +0.070 |
rey | +0.070 |
Him | +0.070 |
Neg logit contributions | |
bumped | -0.038 |
Help | -0.038 |
Was | -0.036 |
" | -0.036 |
v | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
rey | +0.064 |
aleb | +0.063 |
asha | +0.061 |
iper | +0.061 |
aman | +0.060 |
Neg logit contributions | |
Help | -0.035 |
What | -0.033 |
Was | -0.032 |
tearing | -0.031 |
smashing | -0.031 |
Token | " |
Token ID | 366 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
rey | +0.130 |
aleb | +0.129 |
asha | +0.127 |
iper | +0.126 |
iri | +0.124 |
Neg logit contributions | |
Help | -0.024 |
What | -0.024 |
Where | -0.020 |
Was | -0.020 |
Do | -0.020 |
Token | Let |
Token ID | 5756 |
Feature activation | +9.29e-01 |
Pos logit contributions | |
iper | +0.026 |
rey | +0.026 |
aman | +0.025 |
asha | +0.025 |
aleb | +0.025 |
Neg logit contributions | |
Help | -0.015 |
What | -0.015 |
Do | -0.014 |
Was | -0.014 |
Whe | -0.014 |
Token | an |
Token ID | 281 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
rey | +0.005 |
aleb | +0.005 |
asha | +0.005 |
iper | +0.005 |
aman | +0.005 |
Neg logit contributions | |
Help | -0.003 |
What | -0.003 |
Do | -0.003 |
Was | -0.003 |
Where | -0.003 |
Token | oy |
Token ID | 35104 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
rey | +0.021 |
asha | +0.020 |
aleb | +0.020 |
iper | +0.020 |
aman | +0.019 |
Neg logit contributions | |
Help | -0.021 |
What | -0.021 |
Do | -0.020 |
tearing | -0.020 |
Where | -0.020 |
Token | ster |
Token ID | 1706 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
rey | +0.069 |
aleb | +0.067 |
iper | +0.066 |
aman | +0.066 |
asha | +0.065 |
Neg logit contributions | |
Help | -0.067 |
What | -0.065 |
Was | -0.065 |
Do | -0.064 |
Where | -0.063 |
Token | inside |
Token ID | 2641 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
rey | +0.013 |
aleb | +0.012 |
asha | +0.012 |
iper | +0.012 |
aman | +0.012 |
Neg logit contributions | |
Help | -0.009 |
What | -0.009 |
Do | -0.008 |
Was | -0.008 |
Where | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
Help | +0.017 |
What | +0.017 |
Do | +0.016 |
Was | +0.016 |
Where | +0.016 |
Neg logit contributions | |
rey | -0.033 |
aleb | -0.032 |
iper | -0.031 |
asha | -0.031 |
aman | -0.031 |
Token | Oy |
Token ID | 39447 |
Feature activation | +4.21e+00 |
Pos logit contributions | |
Help | +0.001 |
What | +0.001 |
smashing | +0.001 |
Was | +0.001 |
tearing | +0.001 |
Neg logit contributions | |
rey | -0.002 |
aleb | -0.002 |
iper | -0.002 |
asha | -0.002 |
aman | -0.002 |
Token | sters |
Token ID | 5937 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
rey | +0.105 |
iper | +0.104 |
aleb | +0.102 |
aman | +0.101 |
asha | +0.097 |
Neg logit contributions | |
Help | -0.116 |
What | -0.115 |
Do | -0.112 |
Was | -0.111 |
Where | -0.110 |
Token | are |
Token ID | 389 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
rey | +0.054 |
aleb | +0.054 |
asha | +0.053 |
aman | +0.052 |
iper | +0.051 |
Neg logit contributions | |
Help | -0.025 |
What | -0.025 |
Do | -0.025 |
Was | -0.024 |
tearing | -0.023 |
Token | y |
Token ID | 331 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
aleb | +0.030 |
rey | +0.030 |
asha | +0.030 |
aman | +0.030 |
iper | +0.029 |
Neg logit contributions | |
Help | -0.017 |
tearing | -0.016 |
Do | -0.016 |
What | -0.016 |
smashing | -0.016 |
Token | ummy |
Token ID | 13513 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
rey | +0.033 |
aleb | +0.032 |
asha | +0.031 |
iper | +0.031 |
aman | +0.030 |
Neg logit contributions | |
Help | -0.037 |
What | -0.036 |
Was | -0.035 |
Do | -0.035 |
smashing | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
rey | +0.020 |
aleb | +0.020 |
aman | +0.020 |
iper | +0.020 |
asha | +0.019 |
Neg logit contributions | |
Help | -0.009 |
Do | -0.009 |
What | -0.009 |
Was | -0.008 |
Where | -0.008 |
Token | finally |
Token ID | 3443 |
Feature activation | -5.74e-01 |
Pos logit contributions | |
rey | +0.021 |
iper | +0.021 |
ishment | +0.021 |
aleb | +0.021 |
ita | +0.021 |
Neg logit contributions | |
Help | -0.009 |
Was | -0.009 |
What | -0.009 |
Where | -0.008 |
tearing | -0.007 |
Token | left |
Token ID | 1364 |
Feature activation | -9.15e-03 |
Pos logit contributions | |
Help | +0.008 |
Was | +0.008 |
tearing | +0.008 |
What | +0.007 |
smashing | +0.007 |
Neg logit contributions | |
aleb | -0.021 |
rey | -0.021 |
ishment | -0.021 |
iper | -0.021 |
Him | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +5.95e-01 |
Pos logit contributions | |
Help | +0.000 |
What | +0.000 |
Was | +0.000 |
smashing | +0.000 |
tearing | +0.000 |
Neg logit contributions | |
rey | -0.000 |
aleb | -0.000 |
asha | -0.000 |
iper | -0.000 |
isi | -0.000 |
Token | shop |
Token ID | 6128 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
rey | +0.017 |
aleb | +0.016 |
asha | +0.016 |
iper | +0.016 |
onia | +0.015 |
Neg logit contributions | |
Help | -0.015 |
What | -0.015 |
Was | -0.014 |
smashing | -0.014 |
tearing | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
rey | +0.019 |
aleb | +0.019 |
iper | +0.018 |
iri | +0.018 |
asha | +0.018 |
Neg logit contributions | |
Help | -0.006 |
Was | -0.006 |
What | -0.006 |
Where | -0.005 |
tearing | -0.005 |
Token | Cornel |
Token ID | 48214 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
aleb | +0.046 |
iper | +0.046 |
isi | +0.046 |
asha | +0.045 |
onia | +0.045 |
Neg logit contributions | |
Help | -0.030 |
What | -0.030 |
Where | -0.027 |
Was | -0.027 |
Whe | -0.027 |
Token | ius |
Token ID | 3754 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
iper | +0.077 |
isi | +0.076 |
asha | +0.071 |
ita | +0.070 |
hel | +0.068 |
Neg logit contributions | |
Help | -0.134 |
What | -0.131 |
Was | -0.129 |
smashing | -0.126 |
Where | -0.124 |
Token | was |
Token ID | 373 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
Help | +0.010 |
Was | +0.010 |
What | +0.009 |
tearing | +0.009 |
smashing | +0.009 |
Neg logit contributions | |
rey | -0.018 |
aleb | -0.018 |
aman | -0.017 |
asha | -0.017 |
isi | -0.017 |
Token | feeling |
Token ID | 4203 |
Feature activation | -7.42e-01 |
Pos logit contributions | |
Help | +0.001 |
What | +0.001 |
smashing | +0.001 |
tearing | +0.001 |
Was | +0.001 |
Neg logit contributions | |
rey | -0.001 |
aleb | -0.001 |
iper | -0.001 |
asha | -0.001 |
aman | -0.001 |
Token | much |
Token ID | 881 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Help | +0.017 |
What | +0.017 |
Do | +0.017 |
Was | +0.016 |
smashing | +0.016 |
Neg logit contributions | |
rey | -0.021 |
aleb | -0.021 |
iper | -0.020 |
asha | -0.020 |
aman | -0.020 |
Token | better |
Token ID | 1365 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
rey | +0.031 |
aleb | +0.031 |
asha | +0.030 |
iper | +0.030 |
aman | +0.029 |
Neg logit contributions | |
Help | -0.032 |
What | -0.031 |
Do | -0.030 |
Was | -0.030 |
Where | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | +9.83e-01 |
Pos logit contributions | |
rey | +0.037 |
asha | +0.036 |
aman | +0.036 |
iper | +0.035 |
icity | +0.035 |
Neg logit contributions | |
Help | -0.013 |
What | -0.012 |
to | -0.012 |
Do | -0.011 |
v | -0.011 |
Token | passport |
Token ID | 17981 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
rey | +0.028 |
aleb | +0.027 |
asha | +0.026 |
iper | +0.026 |
aman | +0.025 |
Neg logit contributions | |
Help | -0.025 |
What | -0.023 |
Was | -0.023 |
Do | -0.023 |
smashing | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
asha | +0.119 |
aleb | +0.118 |
rey | +0.117 |
iri | +0.116 |
iper | +0.114 |
Neg logit contributions | |
to | -0.024 |
Help | -0.020 |
v | -0.017 |
Do | -0.016 |
fixing | -0.015 |
Token | His |
Token ID | 2399 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
aleb | +0.097 |
asha | +0.096 |
i | +0.095 |
aman | +0.095 |
iper | +0.095 |
Neg logit contributions | |
Help | -0.038 |
What | -0.037 |
Was | -0.033 |
Where | -0.033 |
This | -0.032 |
Token | mom |
Token ID | 1995 |
Feature activation | +9.09e-01 |
Pos logit contributions | |
asha | +0.085 |
aman | +0.084 |
hel | +0.084 |
onia | +0.081 |
isi | +0.080 |
Neg logit contributions | |
Help | -0.064 |
What | -0.056 |
Where | -0.055 |
drawing | -0.054 |
Was | -0.054 |
Token | gave |
Token ID | 2921 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
ishment | +0.029 |
icity | +0.029 |
rey | +0.029 |
avail | +0.028 |
aleb | +0.028 |
Neg logit contributions | |
Was | -0.017 |
Help | -0.017 |
to | -0.017 |
What | -0.016 |
poking | -0.016 |
Token | him |
Token ID | 683 |
Feature activation | +5.25e-01 |
Pos logit contributions | |
ust | +0.133 |
aman | +0.126 |
asha | +0.124 |
iri | +0.123 |
ishment | +0.121 |
Neg logit contributions | |
What | -0.089 |
Do | -0.084 |
Help | -0.082 |
Where | -0.079 |
to | -0.077 |
Token | a |
Token ID | 257 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
rey | +0.020 |
aman | +0.020 |
asha | +0.019 |
onia | +0.019 |
ishment | +0.019 |
Neg logit contributions | |
What | -0.008 |
Help | -0.008 |
Was | -0.007 |
Where | -0.007 |
This | -0.006 |
Token | big |
Token ID | 1263 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
rey | +0.010 |
aleb | +0.010 |
asha | +0.010 |
iper | +0.009 |
aman | +0.009 |
Neg logit contributions | |
Help | -0.008 |
What | -0.008 |
Was | -0.008 |
Do | -0.008 |
Where | -0.008 |
Token | smile |
Token ID | 8212 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
rey | +0.041 |
aleb | +0.039 |
asha | +0.039 |
iper | +0.038 |
aman | +0.038 |
Neg logit contributions | |
Help | -0.033 |
What | -0.031 |
Was | -0.031 |
Do | -0.031 |
smashing | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +9.28e-01 |
Pos logit contributions | |
Help | +0.020 |
What | +0.020 |
free | +0.019 |
Was | +0.019 |
Do | +0.019 |
Neg logit contributions | |
rey | -0.042 |
aleb | -0.041 |
asha | -0.040 |
iper | -0.039 |
aman | -0.039 |
Token | She |
Token ID | 1375 |
Feature activation | +8.06e-01 |
Pos logit contributions | |
rey | +0.073 |
aleb | +0.072 |
iper | +0.070 |
aman | +0.069 |
asha | +0.069 |
Neg logit contributions | |
Help | -0.038 |
What | -0.036 |
Do | -0.035 |
Where | -0.034 |
Was | -0.034 |
Token | shows |
Token ID | 2523 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
rey | +0.027 |
aleb | +0.027 |
asha | +0.026 |
iper | +0.026 |
aman | +0.026 |
Neg logit contributions | |
What | -0.016 |
Help | -0.016 |
Was | -0.015 |
Do | -0.015 |
tearing | -0.015 |
Token | her |
Token ID | 607 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
aleb | +0.059 |
aman | +0.057 |
rey | +0.056 |
iper | +0.056 |
ishment | +0.055 |
Neg logit contributions | |
Help | -0.043 |
What | -0.041 |
Where | -0.039 |
Do | -0.038 |
Was | -0.038 |
Token | paper |
Token ID | 3348 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
rey | +0.004 |
aleb | +0.004 |
iper | +0.004 |
asha | +0.004 |
aman | +0.004 |
Neg logit contributions | |
Help | -0.003 |
What | -0.003 |
Was | -0.003 |
Do | -0.003 |
smashing | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
aleb | +0.067 |
rey | +0.065 |
aman | +0.065 |
asha | +0.063 |
iper | +0.063 |
Neg logit contributions | |
Help | -0.025 |
What | -0.025 |
Was | -0.023 |
tearing | -0.022 |
Do | -0.022 |
Token | R |
Token ID | 371 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
aleb | +0.036 |
iper | +0.035 |
ust | +0.033 |
aman | +0.033 |
asha | +0.033 |
Neg logit contributions | |
Help | -0.021 |
What | -0.019 |
Do | -0.018 |
Whe | -0.018 |
Where | -0.018 |
Token | avi |
Token ID | 15820 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
aleb | +0.064 |
iper | +0.062 |
rey | +0.062 |
aman | +0.061 |
ishment | +0.060 |
Neg logit contributions | |
Help | -0.144 |
What | -0.137 |
Was | -0.132 |
Do | -0.131 |
Where | -0.129 |
Token | . |
Token ID | 13 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
aleb | +0.016 |
rey | +0.015 |
aman | +0.015 |
asha | +0.015 |
iper | +0.015 |
Neg logit contributions | |
Help | -0.006 |
What | -0.005 |
Do | -0.005 |
tearing | -0.005 |
smart | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
rey | +0.033 |
aleb | +0.033 |
iper | +0.032 |
asha | +0.032 |
aman | +0.032 |
Neg logit contributions | |
Help | -0.019 |
What | -0.019 |
Do | -0.018 |
Was | -0.018 |
Where | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
rey | +0.029 |
aleb | +0.028 |
asha | +0.027 |
iper | +0.027 |
aman | +0.027 |
Neg logit contributions | |
Help | -0.014 |
What | -0.014 |
smashing | -0.013 |
Was | -0.013 |
Do | -0.013 |
Token | " |
Token ID | 1 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
aleb | +0.017 |
rey | +0.017 |
iper | +0.017 |
aman | +0.017 |
asha | +0.017 |
Neg logit contributions | |
Help | -0.012 |
What | -0.011 |
Do | -0.011 |
Was | -0.011 |
Where | -0.010 |
Token | for |
Token ID | 329 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
aleb | +0.094 |
rey | +0.091 |
asha | +0.088 |
aman | +0.086 |
iper | +0.084 |
Neg logit contributions | |
free | -0.021 |
Help | -0.021 |
v | -0.021 |
poking | -0.019 |
destroying | -0.019 |
Token | your |
Token ID | 534 |
Feature activation | +9.88e-01 |
Pos logit contributions | |
aleb | +0.007 |
rey | +0.007 |
asha | +0.006 |
aman | +0.006 |
onia | +0.006 |
Neg logit contributions | |
Help | -0.005 |
tearing | -0.005 |
What | -0.004 |
Do | -0.004 |
destroying | -0.004 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
rey | +0.030 |
aleb | +0.029 |
asha | +0.029 |
aman | +0.028 |
onia | +0.028 |
Neg logit contributions | |
Help | -0.024 |
What | -0.023 |
Was | -0.022 |
Do | -0.022 |
tearing | -0.022 |
Token | 's |
Token ID | 338 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
rey | +0.006 |
aleb | +0.006 |
asha | +0.006 |
icity | +0.005 |
aman | +0.005 |
Neg logit contributions | |
Do | -0.003 |
Help | -0.003 |
What | -0.003 |
cutting | -0.002 |
tearing | -0.002 |
Token | birthday |
Token ID | 10955 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
rey | +0.038 |
aleb | +0.037 |
asha | +0.036 |
onia | +0.035 |
aman | +0.035 |
Neg logit contributions | |
Help | -0.028 |
What | -0.027 |
Do | -0.026 |
Was | -0.026 |
v | -0.026 |
Token | ," |
Token ID | 553 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
rey | +0.058 |
aleb | +0.057 |
asha | +0.056 |
iri | +0.054 |
iper | +0.053 |
Neg logit contributions | |
Help | -0.011 |
v | -0.010 |
Was | -0.009 |
What | -0.009 |
Do | -0.009 |
Token | said |
Token ID | 531 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
aleb | +0.133 |
ust | +0.129 |
ishment | +0.129 |
iper | +0.129 |
iet | +0.124 |
Neg logit contributions | |
Help | -0.093 |
What | -0.090 |
Are | -0.086 |
Was | -0.084 |
Where | -0.082 |
Token | grandma |
Token ID | 49890 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
aleb | +0.091 |
iper | +0.086 |
iet | +0.085 |
asha | +0.085 |
ishment | +0.083 |
Neg logit contributions | |
Help | -0.091 |
What | -0.083 |
Where | -0.082 |
" | -0.078 |
Was | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
rey | +0.001 |
aleb | +0.001 |
asha | +0.001 |
iper | +0.001 |
aman | +0.001 |
Neg logit contributions | |
Help | -0.001 |
What | -0.001 |
Was | -0.001 |
Do | -0.001 |
tearing | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
aleb | +0.118 |
iper | +0.114 |
rey | +0.112 |
ishment | +0.111 |
asha | +0.111 |
Neg logit contributions | |
Help | -0.029 |
What | -0.027 |
Do | -0.024 |
Was | -0.023 |
Where | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
aleb | +0.100 |
ishment | +0.100 |
iper | +0.098 |
i | +0.097 |
rey | +0.096 |
Neg logit contributions | |
Help | -0.032 |
What | -0.028 |
Was | -0.026 |
Do | -0.025 |
Where | -0.024 |
Token | them |
Token ID | 606 |
Feature activation | +7.64e-01 |
Pos logit contributions | |
rey | +0.022 |
iper | +0.022 |
aleb | +0.022 |
asha | +0.021 |
aman | +0.021 |
Neg logit contributions | |
What | -0.013 |
Help | -0.012 |
Do | -0.012 |
Was | -0.012 |
Where | -0.011 |
Token | at |
Token ID | 379 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
rey | +0.033 |
asha | +0.033 |
aman | +0.032 |
aleb | +0.032 |
onia | +0.031 |
Neg logit contributions | |
Help | -0.010 |
What | -0.009 |
Do | -0.008 |
Was | -0.007 |
v | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
aleb | +0.070 |
rey | +0.070 |
iper | +0.069 |
asha | +0.069 |
aman | +0.067 |
Neg logit contributions | |
Help | -0.033 |
What | -0.031 |
Do | -0.031 |
Was | -0.029 |
Where | -0.029 |
Token | market |
Token ID | 1910 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
iper | +0.028 |
rey | +0.028 |
aleb | +0.027 |
asha | +0.026 |
aman | +0.026 |
Neg logit contributions | |
Help | -0.028 |
What | -0.027 |
Do | -0.027 |
Was | -0.027 |
Where | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
rey | +0.009 |
aleb | +0.009 |
iper | +0.009 |
asha | +0.009 |
aman | +0.009 |
Neg logit contributions | |
Help | -0.003 |
What | -0.003 |
Do | -0.003 |
Was | -0.003 |
Where | -0.003 |
Token | Oy |
Token ID | 39447 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
rey | +0.007 |
aleb | +0.006 |
asha | +0.006 |
aman | +0.006 |
iper | +0.006 |
Neg logit contributions | |
What | -0.002 |
Do | -0.002 |
Help | -0.002 |
Where | -0.002 |
Was | -0.002 |
Token | sters |
Token ID | 5937 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
iper | +0.112 |
rey | +0.112 |
aman | +0.108 |
aleb | +0.106 |
asha | +0.103 |
Neg logit contributions | |
What | -0.103 |
Help | -0.102 |
Do | -0.098 |
Was | -0.098 |
Where | -0.097 |
Token | are |
Token ID | 389 |
Feature activation | +6.58e-01 |
Pos logit contributions | |
aleb | +0.050 |
rey | +0.049 |
asha | +0.048 |
aman | +0.048 |
iper | +0.047 |
Neg logit contributions | |
Help | -0.022 |
What | -0.021 |
Was | -0.020 |
Do | -0.020 |
tearing | -0.019 |
Token | very |
Token ID | 845 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
aman | +0.024 |
aleb | +0.024 |
asha | +0.024 |
rey | +0.024 |
iper | +0.023 |
Neg logit contributions | |
Do | -0.011 |
Help | -0.011 |
What | -0.011 |
tearing | -0.011 |
destroying | -0.010 |
Token | valuable |
Token ID | 8119 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
rey | +0.000 |
asha | +0.000 |
aman | +0.000 |
aleb | +0.000 |
iper | +0.000 |
Neg logit contributions | |
Help | -0.000 |
Do | -0.000 |
Was | -0.000 |
What | -0.000 |
tearing | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
rey | +0.052 |
aleb | +0.050 |
asha | +0.050 |
aman | +0.050 |
iper | +0.049 |
Neg logit contributions | |
Help | -0.022 |
What | -0.021 |
Do | -0.021 |
Was | -0.020 |
Where | -0.019 |
Token | idea |
Token ID | 2126 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
rey | +0.029 |
aleb | +0.028 |
asha | +0.027 |
iper | +0.027 |
aman | +0.026 |
Neg logit contributions | |
Help | -0.042 |
What | -0.041 |
Was | -0.040 |
Do | -0.040 |
smashing | -0.040 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
rey | +0.015 |
aleb | +0.014 |
asha | +0.014 |
iper | +0.014 |
aman | +0.014 |
Neg logit contributions | |
Help | -0.007 |
What | -0.007 |
Was | -0.007 |
Do | -0.007 |
smashing | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +8.87e-01 |
Pos logit contributions | |
aleb | +0.074 |
rey | +0.072 |
asha | +0.070 |
iper | +0.068 |
ust | +0.068 |
Neg logit contributions | |
What | -0.023 |
Help | -0.022 |
Was | -0.021 |
Do | -0.021 |
Where | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
rey | +0.030 |
aleb | +0.028 |
asha | +0.028 |
iper | +0.028 |
aman | +0.027 |
Neg logit contributions | |
smashing | -0.017 |
Help | -0.017 |
What | -0.017 |
destroying | -0.017 |
tearing | -0.016 |
Token | But |
Token ID | 1537 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
rey | +0.013 |
aleb | +0.013 |
asha | +0.012 |
iper | +0.012 |
aman | +0.012 |
Neg logit contributions | |
Help | -0.015 |
What | -0.015 |
Was | -0.015 |
Do | -0.015 |
smashing | -0.014 |
Token | T |
Token ID | 309 |
Feature activation | +4.10e+00 |
Pos logit contributions | |
aleb | +0.041 |
asha | +0.040 |
rey | +0.039 |
iper | +0.039 |
aman | +0.038 |
Neg logit contributions | |
Help | -0.023 |
What | -0.023 |
Do | -0.022 |
tearing | -0.021 |
Was | -0.021 |
Token | erence |
Token ID | 1945 |
Feature activation | +5.95e-01 |
Pos logit contributions | |
aleb | +0.065 |
aman | +0.064 |
ishment | +0.063 |
rey | +0.062 |
asha | +0.060 |
Neg logit contributions | |
Help | -0.135 |
What | -0.134 |
Was | -0.129 |
smart | -0.126 |
tearing | -0.126 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
aleb | +0.020 |
rey | +0.019 |
aman | +0.018 |
iper | +0.018 |
oto | +0.018 |
Neg logit contributions | |
Help | -0.011 |
Was | -0.011 |
tearing | -0.011 |
smashing | -0.011 |
What | -0.010 |
Token | 't |
Token ID | 470 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
rey | +0.072 |
aleb | +0.071 |
iper | +0.070 |
asha | +0.070 |
aman | +0.069 |
Neg logit contributions | |
Help | -0.025 |
What | -0.023 |
Do | -0.022 |
Was | -0.022 |
Where | -0.022 |
Token | want |
Token ID | 765 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
Help | +0.004 |
What | +0.004 |
Was | +0.004 |
Do | +0.004 |
smashing | +0.004 |
Neg logit contributions | |
rey | -0.004 |
aleb | -0.004 |
asha | -0.004 |
iper | -0.004 |
i | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
Help | +0.010 |
Was | +0.009 |
smashing | +0.009 |
What | +0.009 |
Do | +0.009 |
Neg logit contributions | |
rey | -0.027 |
aleb | -0.027 |
asha | -0.026 |
iper | -0.026 |
i | -0.025 |
Token | then |
Token ID | 788 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
aleb | +0.012 |
rey | +0.012 |
asha | +0.012 |
iper | +0.011 |
aman | +0.011 |
Neg logit contributions | |
What | -0.007 |
Help | -0.007 |
Was | -0.007 |
Do | -0.007 |
Where | -0.007 |
Token | walking |
Token ID | 6155 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
Help | +0.017 |
Was | +0.017 |
What | +0.017 |
smashing | +0.016 |
tearing | +0.016 |
Neg logit contributions | |
rey | -0.025 |
aleb | -0.024 |
iper | -0.023 |
asha | -0.023 |
iri | -0.023 |
Token | away |
Token ID | 1497 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Help | +0.027 |
What | +0.026 |
Was | +0.025 |
free | +0.025 |
Where | +0.024 |
Neg logit contributions | |
aleb | -0.053 |
rey | -0.053 |
iper | -0.051 |
asha | -0.051 |
aman | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
Help | +0.019 |
What | +0.018 |
Was | +0.017 |
smashing | +0.017 |
Where | +0.017 |
Neg logit contributions | |
aleb | -0.040 |
rey | -0.040 |
iper | -0.039 |
asha | -0.038 |
aman | -0.037 |
Token | Little |
Token ID | 7703 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
rey | +0.025 |
aleb | +0.025 |
asha | +0.024 |
iper | +0.024 |
aman | +0.024 |
Neg logit contributions | |
Help | -0.011 |
What | -0.011 |
Was | -0.011 |
Where | -0.010 |
Do | -0.010 |
Token | Cornel |
Token ID | 48214 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
rey | +0.045 |
iper | +0.044 |
aleb | +0.042 |
aman | +0.041 |
icity | +0.041 |
Neg logit contributions | |
What | -0.047 |
Was | -0.045 |
Help | -0.045 |
Where | -0.044 |
Do | -0.044 |
Token | ius |
Token ID | 3754 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
isi | +0.077 |
iper | +0.076 |
asha | +0.072 |
hel | +0.070 |
ita | +0.069 |
Neg logit contributions | |
Help | -0.130 |
What | -0.127 |
Was | -0.126 |
smashing | -0.122 |
bulldo | -0.120 |
Token | was |
Token ID | 373 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
Help | +0.007 |
Was | +0.007 |
What | +0.006 |
tearing | +0.006 |
smashing | +0.006 |
Neg logit contributions | |
rey | -0.013 |
aleb | -0.013 |
aman | -0.012 |
asha | -0.012 |
oto | -0.012 |
Token | so |
Token ID | 523 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
Help | +0.001 |
What | +0.001 |
tearing | +0.001 |
smashing | +0.001 |
Was | +0.001 |
Neg logit contributions | |
rey | -0.002 |
aleb | -0.002 |
iper | -0.002 |
asha | -0.002 |
aman | -0.002 |
Token | proud |
Token ID | 6613 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
Help | +0.002 |
What | +0.002 |
Was | +0.002 |
tearing | +0.002 |
Where | +0.002 |
Neg logit contributions | |
rey | -0.004 |
asha | -0.004 |
aleb | -0.004 |
iper | -0.004 |
aman | -0.004 |
Token | that |
Token ID | 326 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
rey | +0.041 |
aleb | +0.039 |
iper | +0.039 |
onia | +0.038 |
icity | +0.038 |
Neg logit contributions | |
What | -0.012 |
Help | -0.011 |
Was | -0.010 |
Do | -0.010 |
tearing | -0.010 |
Token | so |
Token ID | 523 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
Help | +0.002 |
What | +0.002 |
Do | +0.001 |
tearing | +0.001 |
Was | +0.001 |
Neg logit contributions | |
aleb | -0.003 |
rey | -0.003 |
iper | -0.003 |
asha | -0.003 |
aman | -0.003 |
Token | high |
Token ID | 1029 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
rey | +0.007 |
aleb | +0.007 |
iper | +0.007 |
asha | +0.007 |
aman | +0.007 |
Neg logit contributions | |
Help | -0.006 |
What | -0.006 |
Do | -0.005 |
smashing | -0.005 |
Where | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
aleb | +0.029 |
rey | +0.029 |
iper | +0.029 |
asha | +0.028 |
aman | +0.028 |
Neg logit contributions | |
What | -0.010 |
destroying | -0.010 |
free | -0.010 |
tearing | -0.010 |
v | -0.010 |
Token | However |
Token ID | 2102 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
rey | +0.018 |
aleb | +0.018 |
asha | +0.017 |
iper | +0.017 |
aman | +0.017 |
Neg logit contributions | |
Help | -0.009 |
What | -0.009 |
Was | -0.009 |
Do | -0.009 |
Where | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
Help | +0.001 |
What | +0.001 |
Do | +0.001 |
Where | +0.001 |
Was | +0.001 |
Neg logit contributions | |
aleb | -0.006 |
rey | -0.006 |
asha | -0.006 |
iper | -0.005 |
isi | -0.005 |
Token | T |
Token ID | 309 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
aleb | +0.021 |
rey | +0.020 |
asha | +0.020 |
iper | +0.020 |
aman | +0.019 |
Neg logit contributions | |
Help | -0.016 |
What | -0.015 |
Do | -0.015 |
Was | -0.015 |
Where | -0.015 |
Token | erence |
Token ID | 1945 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
aleb | +0.062 |
aman | +0.061 |
rey | +0.060 |
ishment | +0.058 |
asha | +0.058 |
Neg logit contributions | |
What | -0.136 |
Help | -0.135 |
Was | -0.129 |
Do | -0.127 |
tearing | -0.127 |
Token | really |
Token ID | 1107 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
aleb | +0.023 |
aman | +0.022 |
rey | +0.022 |
oto | +0.021 |
ishment | +0.021 |
Neg logit contributions | |
Help | -0.011 |
Was | -0.011 |
tearing | -0.010 |
smashing | -0.010 |
What | -0.010 |
Token | wanted |
Token ID | 2227 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
Help | +0.002 |
Was | +0.002 |
What | +0.002 |
tearing | +0.002 |
smashing | +0.001 |
Neg logit contributions | |
aleb | -0.004 |
rey | -0.004 |
aman | -0.004 |
onia | -0.004 |
asha | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
rey | +0.014 |
aleb | +0.013 |
asha | +0.013 |
iper | +0.013 |
aman | +0.013 |
Neg logit contributions | |
Help | -0.003 |
What | -0.003 |
Was | -0.003 |
Do | -0.003 |
smashing | -0.003 |
Token | explore |
Token ID | 7301 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
aleb | +0.026 |
rey | +0.025 |
iper | +0.025 |
aman | +0.025 |
asha | +0.025 |
Neg logit contributions | |
Help | -0.015 |
tearing | -0.013 |
What | -0.013 |
crash | -0.013 |
Where | -0.012 |
Token | so |
Token ID | 523 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
rey | +0.024 |
aleb | +0.023 |
iper | +0.023 |
asha | +0.023 |
aman | +0.023 |
Neg logit contributions | |
What | -0.014 |
Do | -0.014 |
Help | -0.014 |
Was | -0.014 |
tearing | -0.014 |
Token | big |
Token ID | 1263 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
rey | +0.020 |
aman | +0.019 |
asha | +0.019 |
aleb | +0.018 |
isi | +0.018 |
Neg logit contributions | |
What | -0.011 |
Help | -0.010 |
Where | -0.010 |
Was | -0.010 |
Do | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
rey | +0.083 |
aleb | +0.082 |
iri | +0.081 |
asha | +0.080 |
onia | +0.080 |
Neg logit contributions | |
What | -0.031 |
Do | -0.031 |
Was | -0.028 |
Where | -0.028 |
Help | -0.027 |
Token | special |
Token ID | 2041 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
rey | +0.013 |
aleb | +0.013 |
asha | +0.013 |
iper | +0.013 |
aman | +0.013 |
Neg logit contributions | |
Do | -0.009 |
What | -0.009 |
Was | -0.009 |
tearing | -0.009 |
Help | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
rey | +0.052 |
aleb | +0.051 |
iper | +0.050 |
asha | +0.050 |
aman | +0.050 |
Neg logit contributions | |
What | -0.022 |
Help | -0.021 |
Was | -0.020 |
Do | -0.020 |
Where | -0.020 |
Token | Fel |
Token ID | 13937 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
rey | +0.032 |
aleb | +0.032 |
asha | +0.031 |
iper | +0.031 |
aman | +0.030 |
Neg logit contributions | |
Help | -0.022 |
What | -0.021 |
Was | -0.021 |
Do | -0.021 |
Where | -0.021 |
Token | icity |
Token ID | 8467 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
asha | +0.042 |
aman | +0.040 |
aleb | +0.039 |
isi | +0.037 |
rey | +0.036 |
Neg logit contributions | |
Help | -0.154 |
What | -0.154 |
Was | -0.151 |
Where | -0.150 |
tearing | -0.149 |
Token | was |
Token ID | 373 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
Help | +0.004 |
Was | +0.004 |
What | +0.003 |
tearing | +0.003 |
smashing | +0.003 |
Neg logit contributions | |
aleb | -0.007 |
aman | -0.007 |
rey | -0.007 |
asha | -0.007 |
isi | -0.006 |
Token | so |
Token ID | 523 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
rey | +0.007 |
aleb | +0.007 |
iper | +0.006 |
asha | +0.006 |
aman | +0.006 |
Neg logit contributions | |
Help | -0.005 |
What | -0.005 |
Was | -0.005 |
Do | -0.005 |
smashing | -0.005 |
Token | proud |
Token ID | 6613 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
rey | +0.019 |
aleb | +0.018 |
asha | +0.018 |
iper | +0.018 |
aman | +0.018 |
Neg logit contributions | |
Help | -0.014 |
What | -0.013 |
Was | -0.013 |
Do | -0.013 |
Where | -0.013 |
Token | of |
Token ID | 286 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
rey | +0.053 |
aleb | +0.051 |
iper | +0.050 |
onia | +0.050 |
asha | +0.050 |
Neg logit contributions | |
What | -0.015 |
Help | -0.014 |
Do | -0.013 |
Was | -0.013 |
Where | -0.013 |
Token | an |
Token ID | 281 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
rey | +0.072 |
aleb | +0.072 |
iper | +0.070 |
asha | +0.070 |
aman | +0.069 |
Neg logit contributions | |
Help | -0.046 |
What | -0.044 |
tearing | -0.042 |
Do | -0.042 |
smashing | -0.042 |
Token | old |
Token ID | 1468 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
rey | +0.042 |
asha | +0.041 |
aleb | +0.040 |
aman | +0.040 |
iper | +0.039 |
Neg logit contributions | |
Help | -0.035 |
tearing | -0.034 |
What | -0.034 |
destroying | -0.034 |
Where | -0.033 |
Token | tower |
Token ID | 10580 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
aleb | +0.050 |
rey | +0.050 |
iper | +0.048 |
asha | +0.047 |
aman | +0.046 |
Neg logit contributions | |
Help | -0.054 |
What | -0.051 |
Was | -0.051 |
tearing | -0.051 |
smashing | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
rey | +0.002 |
aleb | +0.002 |
iper | +0.002 |
asha | +0.002 |
aman | +0.002 |
Neg logit contributions | |
Help | -0.001 |
What | -0.001 |
Was | -0.001 |
smashing | -0.001 |
tearing | -0.001 |
Token | Sir |
Token ID | 7361 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
rey | +0.013 |
aleb | +0.013 |
iper | +0.013 |
asha | +0.013 |
aman | +0.012 |
Neg logit contributions | |
Help | -0.008 |
smashing | -0.007 |
tearing | -0.007 |
What | -0.007 |
Was | -0.007 |
Token | T |
Token ID | 309 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
rey | +0.044 |
aleb | +0.043 |
asha | +0.041 |
iper | +0.041 |
aman | +0.040 |
Neg logit contributions | |
Help | -0.053 |
What | -0.052 |
Was | -0.051 |
Do | -0.051 |
Where | -0.050 |
Token | erence |
Token ID | 1945 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
aleb | +0.058 |
rey | +0.058 |
aman | +0.055 |
ishment | +0.053 |
iper | +0.053 |
Neg logit contributions | |
What | -0.140 |
Help | -0.138 |
Was | -0.135 |
Do | -0.133 |
tearing | -0.133 |
Token | was |
Token ID | 373 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
aleb | +0.008 |
rey | +0.008 |
iper | +0.007 |
aman | +0.007 |
asha | +0.007 |
Neg logit contributions | |
Help | -0.004 |
Was | -0.004 |
tearing | -0.004 |
smashing | -0.004 |
What | -0.004 |
Token | brave |
Token ID | 14802 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
rey | +0.019 |
aleb | +0.018 |
asha | +0.018 |
iper | +0.018 |
aman | +0.017 |
Neg logit contributions | |
Help | -0.012 |
What | -0.011 |
Was | -0.011 |
Do | -0.011 |
Where | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
rey | +0.045 |
aleb | +0.044 |
asha | +0.043 |
iper | +0.043 |
aman | +0.043 |
Neg logit contributions | |
What | -0.031 |
Help | -0.031 |
Do | -0.030 |
tearing | -0.030 |
smashing | -0.030 |
Token | curious |
Token ID | 11040 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
rey | +0.009 |
aleb | +0.009 |
aman | +0.009 |
asha | +0.009 |
iper | +0.009 |
Neg logit contributions | |
Help | -0.007 |
What | -0.007 |
Do | -0.007 |
Was | -0.007 |
smashing | -0.006 |
Token | rockets |
Token ID | 20293 |
Feature activation | +7.81e-01 |
Pos logit contributions | |
Help | +0.009 |
What | +0.008 |
Do | +0.008 |
destroying | +0.008 |
tearing | +0.008 |
Neg logit contributions | |
aleb | -0.016 |
rey | -0.016 |
asha | -0.016 |
aman | -0.016 |
iper | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
aleb | +0.030 |
rey | +0.030 |
asha | +0.029 |
iper | +0.029 |
aman | +0.028 |
Neg logit contributions | |
Help | -0.012 |
tearing | -0.012 |
smashing | -0.012 |
destroying | -0.012 |
What | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
rey | +0.040 |
aleb | +0.040 |
asha | +0.039 |
aman | +0.038 |
i | +0.038 |
Neg logit contributions | |
What | -0.021 |
Help | -0.021 |
Was | -0.020 |
Where | -0.020 |
Do | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
rey | +0.048 |
aleb | +0.048 |
asha | +0.047 |
iper | +0.047 |
aman | +0.046 |
Neg logit contributions | |
Help | -0.020 |
What | -0.019 |
Was | -0.018 |
Do | -0.018 |
Where | -0.018 |
Token | Laura |
Token ID | 43687 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
i | +0.049 |
ishment | +0.046 |
aman | +0.045 |
aleb | +0.045 |
iper | +0.044 |
Neg logit contributions | |
Help | -0.032 |
What | -0.032 |
Was | -0.032 |
Where | -0.030 |
Do | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
Help | +0.008 |
Was | +0.007 |
What | +0.007 |
tearing | +0.007 |
smashing | +0.007 |
Neg logit contributions | |
aleb | -0.012 |
aman | -0.012 |
rey | -0.012 |
asha | -0.012 |
iper | -0.012 |
Token | Connor |
Token ID | 27599 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
ust | +0.107 |
lights | +0.096 |
argo | +0.095 |
rig | +0.093 |
rily | +0.091 |
Neg logit contributions | |
What | -0.109 |
Do | -0.102 |
Help | -0.097 |
" | -0.097 |
This | -0.096 |
Token | were |
Token ID | 547 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
Help | +0.009 |
What | +0.009 |
Was | +0.009 |
tearing | +0.008 |
smashing | +0.008 |
Neg logit contributions | |
rey | -0.013 |
aman | -0.013 |
asha | -0.013 |
aleb | -0.013 |
iper | -0.013 |
Token | so |
Token ID | 523 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
Help | +0.006 |
What | +0.006 |
smashing | +0.006 |
tearing | +0.006 |
Do | +0.006 |
Neg logit contributions | |
rey | -0.010 |
aleb | -0.010 |
asha | -0.009 |
iper | -0.009 |
aman | -0.009 |
Token | proud |
Token ID | 6613 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
Help | +0.008 |
What | +0.007 |
Was | +0.007 |
Do | +0.007 |
smashing | +0.007 |
Neg logit contributions | |
rey | -0.011 |
aleb | -0.010 |
asha | -0.010 |
iper | -0.010 |
aman | -0.010 |
Token | of |
Token ID | 286 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
rey | +0.031 |
aleb | +0.030 |
asha | +0.030 |
iper | +0.030 |
aman | +0.029 |
Neg logit contributions | |
Help | -0.008 |
What | -0.008 |
Do | -0.007 |
Was | -0.007 |
Where | -0.007 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
aleb | +0.041 |
rey | +0.040 |
aman | +0.040 |
asha | +0.039 |
iper | +0.039 |
Neg logit contributions | |
Help | -0.035 |
Was | -0.033 |
What | -0.033 |
tearing | -0.033 |
smashing | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
aleb | +0.063 |
rey | +0.062 |
iper | +0.061 |
asha | +0.061 |
icity | +0.061 |
Neg logit contributions | |
Help | -0.025 |
What | -0.025 |
" | -0.022 |
Was | -0.022 |
joy | -0.022 |
Token | " |
Token ID | 366 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
rey | +0.134 |
aleb | +0.132 |
i | +0.130 |
iper | +0.130 |
iri | +0.129 |
Neg logit contributions | |
What | -0.019 |
Help | -0.017 |
Was | -0.015 |
Do | -0.014 |
Where | -0.014 |
Token | Sure |
Token ID | 19457 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
rey | +0.022 |
aleb | +0.021 |
iper | +0.021 |
asha | +0.021 |
aman | +0.021 |
Neg logit contributions | |
Help | -0.016 |
What | -0.016 |
Do | -0.015 |
Was | -0.015 |
Where | -0.015 |
Token | thing |
Token ID | 1517 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
rey | +0.020 |
aleb | +0.019 |
asha | +0.019 |
iper | +0.019 |
aman | +0.019 |
Neg logit contributions | |
Help | -0.008 |
What | -0.007 |
Do | -0.007 |
Was | -0.007 |
Where | -0.007 |
Token | ," |
Token ID | 553 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
rey | +0.052 |
aleb | +0.050 |
ishment | +0.049 |
asha | +0.049 |
aman | +0.049 |
Neg logit contributions | |
Help | -0.016 |
What | -0.015 |
Do | -0.015 |
Was | -0.013 |
Where | -0.013 |
Token | she |
Token ID | 673 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
aleb | +0.142 |
rey | +0.141 |
iri | +0.141 |
ust | +0.138 |
icity | +0.136 |
Neg logit contributions | |
What | -0.078 |
" | -0.077 |
Help | -0.076 |
Was | -0.071 |
Do | -0.068 |
Token | said |
Token ID | 531 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
aleb | +0.046 |
aman | +0.046 |
rey | +0.045 |
iri | +0.044 |
asha | +0.043 |
Neg logit contributions | |
Help | -0.045 |
Was | -0.045 |
What | -0.043 |
Where | -0.041 |
v | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
rey | +0.037 |
iper | +0.036 |
aleb | +0.036 |
i | +0.035 |
asha | +0.034 |
Neg logit contributions | |
What | -0.032 |
Was | -0.032 |
Do | -0.032 |
Help | -0.031 |
Where | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
rey | +0.090 |
aleb | +0.089 |
asha | +0.087 |
iper | +0.086 |
i | +0.085 |
Neg logit contributions | |
Help | -0.020 |
What | -0.020 |
Do | -0.018 |
Was | -0.017 |
Where | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
aleb | +0.080 |
rey | +0.080 |
i | +0.078 |
iper | +0.077 |
ishment | +0.077 |
Neg logit contributions | |
Help | -0.027 |
What | -0.025 |
Was | -0.024 |
Do | -0.023 |
Where | -0.022 |
Token | it |
Token ID | 340 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Help | +0.023 |
What | +0.022 |
Was | +0.021 |
Do | +0.021 |
smashing | +0.021 |
Neg logit contributions | |
rey | -0.035 |
aleb | -0.034 |
asha | -0.033 |
iper | -0.033 |
aman | -0.033 |
Token | on |
Token ID | 319 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
Help | +0.003 |
What | +0.003 |
Was | +0.003 |
Do | +0.003 |
smashing | +0.003 |
Neg logit contributions | |
rey | -0.004 |
aleb | -0.004 |
asha | -0.004 |
iper | -0.004 |
aman | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
Help | +0.003 |
What | +0.003 |
Was | +0.003 |
Do | +0.003 |
smashing | +0.003 |
Neg logit contributions | |
rey | -0.002 |
aleb | -0.002 |
asha | -0.002 |
iper | -0.002 |
aman | -0.002 |
Token | chickens |
Token ID | 25972 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
rey | +0.006 |
aleb | +0.006 |
asha | +0.005 |
iper | +0.005 |
aman | +0.005 |
Neg logit contributions | |
Help | -0.007 |
What | -0.007 |
Was | -0.006 |
smashing | -0.006 |
Do | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
Help | +0.004 |
What | +0.004 |
tearing | +0.004 |
smashing | +0.004 |
Was | +0.004 |
Neg logit contributions | |
rey | -0.008 |
aleb | -0.008 |
asha | -0.008 |
iper | -0.008 |
aman | -0.008 |
Token | Fel |
Token ID | 13937 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
iper | +0.008 |
rey | +0.008 |
asha | +0.007 |
aleb | +0.007 |
ita | +0.007 |
Neg logit contributions | |
Help | -0.007 |
destroying | -0.007 |
smashing | -0.007 |
tearing | -0.007 |
Do | -0.007 |
Token | icity |
Token ID | 8467 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
asha | +0.034 |
rey | +0.033 |
aleb | +0.033 |
aman | +0.033 |
onia | +0.029 |
Neg logit contributions | |
Help | -0.160 |
What | -0.159 |
tearing | -0.157 |
Was | -0.156 |
Where | -0.155 |
Token | blinked |
Token ID | 40360 |
Feature activation | -8.44e-01 |
Pos logit contributions | |
Help | +0.042 |
What | +0.041 |
Do | +0.040 |
Was | +0.039 |
Where | +0.039 |
Neg logit contributions | |
rey | -0.054 |
aleb | -0.052 |
iper | -0.051 |
asha | -0.051 |
aman | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
Help | +0.020 |
What | +0.020 |
Was | +0.019 |
smashing | +0.018 |
tearing | +0.018 |
Neg logit contributions | |
rey | -0.025 |
aleb | -0.025 |
iper | -0.024 |
asha | -0.024 |
aman | -0.024 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
aleb | +0.025 |
rey | +0.025 |
iper | +0.024 |
asha | +0.024 |
aman | +0.023 |
Neg logit contributions | |
Help | -0.019 |
What | -0.019 |
Was | -0.018 |
Do | -0.018 |
tearing | -0.018 |
Token | that |
Token ID | 326 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
Help | +0.006 |
What | +0.006 |
Was | +0.006 |
Do | +0.006 |
Where | +0.006 |
Neg logit contributions | |
rey | -0.009 |
aleb | -0.009 |
iper | -0.009 |
asha | -0.009 |
aman | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
rey | +0.085 |
aleb | +0.083 |
aman | +0.082 |
i | +0.082 |
iper | +0.082 |
Neg logit contributions | |
What | -0.048 |
Help | -0.047 |
Where | -0.045 |
Was | -0.044 |
Do | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
rey | +0.066 |
aleb | +0.065 |
iper | +0.064 |
asha | +0.064 |
aman | +0.063 |
Neg logit contributions | |
Help | -0.027 |
What | -0.025 |
Was | -0.024 |
Do | -0.024 |
Where | -0.024 |
Token | " |
Token ID | 1 |
Feature activation | +8.00e-01 |
Pos logit contributions | |
i | +0.066 |
ishment | +0.064 |
rey | +0.062 |
aman | +0.062 |
iper | +0.061 |
Neg logit contributions | |
Help | -0.065 |
What | -0.062 |
Was | -0.059 |
Where | -0.059 |
Do | -0.058 |
Token | Mom |
Token ID | 29252 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
iper | +0.031 |
i | +0.030 |
aleb | +0.029 |
ishment | +0.029 |
rey | +0.029 |
Neg logit contributions | |
Help | -0.014 |
What | -0.014 |
Was | -0.014 |
Where | -0.013 |
Do | -0.012 |
Token | my |
Token ID | 1820 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
Help | +0.015 |
What | +0.015 |
smashing | +0.015 |
Was | +0.015 |
Where | +0.015 |
Neg logit contributions | |
rey | -0.014 |
aleb | -0.013 |
iper | -0.013 |
aman | -0.013 |
asha | -0.013 |
Token | ," |
Token ID | 553 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
rey | +0.000 |
aleb | +0.000 |
iper | +0.000 |
asha | +0.000 |
aman | +0.000 |
Neg logit contributions | |
Help | -0.000 |
What | -0.000 |
Was | -0.000 |
Do | -0.000 |
smashing | -0.000 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
ind | +0.144 |
ust | +0.142 |
iet | +0.142 |
la | +0.140 |
iri | +0.139 |
Neg logit contributions | |
Are | -0.075 |
What | -0.075 |
Help | -0.074 |
Do | -0.070 |
" | -0.070 |
Token | Charlie |
Token ID | 11526 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
iri | +0.061 |
ust | +0.061 |
aman | +0.061 |
asha | +0.060 |
iet | +0.059 |
Neg logit contributions | |
Help | -0.054 |
What | -0.051 |
Where | -0.049 |
" | -0.048 |
The | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | -5.84e-01 |
Pos logit contributions | |
rey | +0.014 |
aleb | +0.014 |
asha | +0.013 |
iper | +0.013 |
aman | +0.013 |
Neg logit contributions | |
Help | -0.008 |
What | -0.008 |
Was | -0.008 |
Do | -0.008 |
smashing | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
Help | +0.008 |
What | +0.007 |
Was | +0.007 |
Do | +0.007 |
smashing | +0.007 |
Neg logit contributions | |
rey | -0.023 |
aleb | -0.023 |
asha | -0.022 |
iper | -0.022 |
aman | -0.022 |
Token | Can |
Token ID | 6090 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
rey | +0.013 |
aleb | +0.012 |
iper | +0.012 |
asha | +0.012 |
aman | +0.012 |
Neg logit contributions | |
Help | -0.008 |
What | -0.007 |
Was | -0.007 |
Do | -0.007 |
Where | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
rey | +0.074 |
aleb | +0.073 |
iper | +0.072 |
asha | +0.071 |
aman | +0.070 |
Neg logit contributions | |
Help | -0.013 |
What | -0.012 |
Do | -0.011 |
Was | -0.010 |
Where | -0.010 |
Token | explore |
Token ID | 7301 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
aleb | +0.027 |
aman | +0.027 |
ishment | +0.027 |
asha | +0.027 |
iper | +0.026 |
Neg logit contributions | |
Help | -0.013 |
This | -0.011 |
Where | -0.011 |
What | -0.010 |
have | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
aleb | +0.047 |
rey | +0.047 |
asha | +0.046 |
iper | +0.046 |
aman | +0.046 |
Neg logit contributions | |
Help | -0.020 |
What | -0.020 |
Do | -0.019 |
tearing | -0.018 |
destroying | -0.018 |
Token | woods |
Token ID | 16479 |
Feature activation | +6.52e-01 |
Pos logit contributions | |
iper | +0.062 |
aleb | +0.062 |
asha | +0.061 |
rey | +0.061 |
iri | +0.059 |
Neg logit contributions | |
Help | -0.038 |
What | -0.037 |
v | -0.037 |
Was | -0.036 |
building | -0.035 |
Token | near |
Token ID | 1474 |
Feature activation | +8.66e-01 |
Pos logit contributions | |
aleb | +0.024 |
rey | +0.024 |
asha | +0.023 |
iper | +0.023 |
aman | +0.023 |
Neg logit contributions | |
Help | -0.011 |
What | -0.011 |
Do | -0.011 |
tearing | -0.010 |
smashing | -0.010 |
Token | their |
Token ID | 511 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
aleb | +0.030 |
rey | +0.030 |
iper | +0.029 |
asha | +0.029 |
aman | +0.029 |
Neg logit contributions | |
Help | -0.017 |
What | -0.016 |
Was | -0.015 |
Do | -0.015 |
tearing | -0.015 |
Token | home |
Token ID | 1363 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
asha | +0.134 |
onia | +0.132 |
isi | +0.131 |
iper | +0.131 |
rey | +0.129 |
Neg logit contributions | |
Help | -0.081 |
What | -0.074 |
Was | -0.069 |
Where | -0.068 |
building | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
Help | +0.006 |
What | +0.005 |
Was | +0.005 |
tearing | +0.005 |
Do | +0.005 |
Neg logit contributions | |
aleb | -0.018 |
rey | -0.018 |
iper | -0.017 |
asha | -0.017 |
onia | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
rey | +0.008 |
aleb | +0.008 |
asha | +0.008 |
iper | +0.008 |
i | +0.007 |
Neg logit contributions | |
What | -0.004 |
Help | -0.004 |
Where | -0.003 |
Was | -0.003 |
Do | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
rey | +0.009 |
aleb | +0.009 |
asha | +0.008 |
iper | +0.008 |
aman | +0.008 |
Neg logit contributions | |
Help | -0.004 |
What | -0.004 |
smashing | -0.004 |
Do | -0.004 |
Was | -0.004 |
Token | As |
Token ID | 1722 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
rey | +0.009 |
aleb | +0.009 |
iper | +0.008 |
asha | +0.008 |
i | +0.008 |
Neg logit contributions | |
Help | -0.009 |
What | -0.008 |
Was | -0.008 |
Do | -0.008 |
Where | -0.008 |
Token | could |
Token ID | 714 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
tearing | +0.003 |
Was | +0.003 |
What | +0.002 |
smashing | +0.002 |
destroying | +0.002 |
Neg logit contributions | |
aleb | -0.009 |
iper | -0.008 |
rey | -0.008 |
aman | -0.008 |
iri | -0.008 |
Token | lift |
Token ID | 10303 |
Feature activation | +7.23e-02 |
Pos logit contributions | |
Help | +0.011 |
Was | +0.010 |
What | +0.010 |
tearing | +0.010 |
Do | +0.010 |
Neg logit contributions | |
aleb | -0.020 |
rey | -0.019 |
iper | -0.019 |
aman | -0.019 |
asha | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
aleb | +0.002 |
rey | +0.002 |
asha | +0.002 |
aman | +0.002 |
iper | +0.002 |
Neg logit contributions | |
What | -0.001 |
Help | -0.001 |
Do | -0.001 |
Was | -0.001 |
smashing | -0.001 |
Token | car |
Token ID | 1097 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
Help | +0.015 |
smart | +0.015 |
What | +0.014 |
Do | +0.014 |
Where | +0.014 |
Neg logit contributions | |
rey | -0.015 |
aleb | -0.014 |
asha | -0.014 |
aman | -0.014 |
iper | -0.013 |
Token | ! |
Token ID | 0 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
aleb | +0.003 |
asha | +0.003 |
rey | +0.003 |
iper | +0.003 |
iri | +0.003 |
Neg logit contributions | |
Help | -0.001 |
tearing | -0.001 |
What | -0.001 |
smashing | -0.001 |
knocking | -0.001 |
Token | Fel |
Token ID | 13937 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
rey | +0.029 |
aleb | +0.029 |
asha | +0.028 |
iper | +0.027 |
aman | +0.027 |
Neg logit contributions | |
Help | -0.021 |
What | -0.021 |
Was | -0.020 |
Do | -0.020 |
Where | -0.020 |
Token | icity |
Token ID | 8467 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
asha | +0.038 |
aman | +0.036 |
aleb | +0.036 |
rey | +0.035 |
isi | +0.033 |
Neg logit contributions | |
Help | -0.154 |
What | -0.154 |
Was | -0.152 |
tearing | -0.151 |
Where | -0.150 |
Token | 's |
Token ID | 338 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
Help | +0.008 |
Was | +0.008 |
What | +0.008 |
tearing | +0.008 |
smashing | +0.008 |
Neg logit contributions | |
rey | -0.013 |
aleb | -0.013 |
asha | -0.013 |
aman | -0.013 |
iper | -0.013 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
rey | +0.035 |
asha | +0.034 |
aleb | +0.033 |
aman | +0.033 |
iper | +0.032 |
Neg logit contributions | |
Help | -0.045 |
What | -0.043 |
Was | -0.043 |
Where | -0.042 |
tearing | -0.042 |
Token | my |
Token ID | 1820 |
Feature activation | +8.46e-01 |
Pos logit contributions | |
rey | +0.006 |
aleb | +0.006 |
asha | +0.005 |
icity | +0.005 |
iri | +0.005 |
Neg logit contributions | |
tearing | -0.005 |
Help | -0.005 |
What | -0.004 |
Was | -0.004 |
smashing | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
aleb | +0.027 |
rey | +0.027 |
asha | +0.026 |
iri | +0.026 |
icity | +0.026 |
Neg logit contributions | |
Help | -0.016 |
tearing | -0.016 |
Was | -0.016 |
smashing | -0.016 |
What | -0.016 |
Token | fork |
Token ID | 15563 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
rey | +0.004 |
aleb | +0.004 |
asha | +0.004 |
onia | +0.003 |
iper | +0.003 |
Neg logit contributions | |
smart | -0.008 |
Help | -0.008 |
What | -0.008 |
Do | -0.008 |
Those | -0.008 |
Token | that |
Token ID | 326 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
rey | +0.004 |
aleb | +0.004 |
asha | +0.004 |
i | +0.004 |
aman | +0.004 |
Neg logit contributions | |
Help | -0.003 |
What | -0.003 |
Was | -0.003 |
Do | -0.003 |
Whe | -0.003 |
Token | big |
Token ID | 1263 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
What | +0.009 |
Was | +0.009 |
Help | +0.009 |
Where | +0.008 |
Do | +0.008 |
Neg logit contributions | |
aleb | -0.015 |
rey | -0.015 |
iper | -0.015 |
asha | -0.015 |
aman | -0.015 |
Token | before |
Token ID | 878 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
rey | +0.037 |
aleb | +0.036 |
asha | +0.035 |
iper | +0.035 |
aman | +0.034 |
Neg logit contributions | |
Help | -0.032 |
What | -0.031 |
Was | -0.031 |
Do | -0.031 |
Where | -0.030 |
Token | ! |
Token ID | 0 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
aleb | +0.079 |
rey | +0.077 |
iper | +0.074 |
asha | +0.073 |
onia | +0.071 |
Neg logit contributions | |
Help | -0.033 |
What | -0.031 |
tearing | -0.031 |
Do | -0.030 |
Was | -0.029 |
Token | Fel |
Token ID | 13937 |
Feature activation | +3.93e+00 |
Pos logit contributions | |
rey | +0.031 |
aleb | +0.030 |
asha | +0.029 |
iper | +0.029 |
aman | +0.029 |
Neg logit contributions | |
Help | -0.019 |
What | -0.019 |
Was | -0.018 |
Do | -0.018 |
Where | -0.018 |
Token | icity |
Token ID | 8467 |
Feature activation | -9.63e-01 |
Pos logit contributions | |
asha | +0.035 |
aleb | +0.034 |
aman | +0.034 |
rey | +0.033 |
isi | +0.031 |
Neg logit contributions | |
Help | -0.158 |
What | -0.157 |
tearing | -0.155 |
Was | -0.155 |
Where | -0.153 |
Token | was |
Token ID | 373 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
Help | +0.020 |
Was | +0.019 |
What | +0.018 |
tearing | +0.018 |
smashing | +0.017 |
Neg logit contributions | |
rey | -0.032 |
aleb | -0.032 |
aman | -0.031 |
asha | -0.031 |
iper | -0.030 |
Token | so |
Token ID | 523 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
rey | +0.015 |
aleb | +0.014 |
iper | +0.014 |
asha | +0.014 |
aman | +0.014 |
Neg logit contributions | |
Help | -0.013 |
What | -0.012 |
Do | -0.012 |
Was | -0.012 |
Where | -0.012 |
Token | excited |
Token ID | 6568 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
rey | +0.029 |
aleb | +0.028 |
asha | +0.028 |
iper | +0.028 |
aman | +0.027 |
Neg logit contributions | |
Help | -0.025 |
What | -0.024 |
Was | -0.024 |
Do | -0.023 |
Where | -0.023 |
Token | ! |
Token ID | 0 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
rey | +0.085 |
aleb | +0.084 |
iper | +0.080 |
otton | +0.080 |
iri | +0.078 |
Neg logit contributions | |
What | -0.020 |
to | -0.018 |
Do | -0.015 |
tearing | -0.015 |
Help | -0.015 |
Token | not |
Token ID | 407 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Was | +0.007 |
Help | +0.007 |
Where | +0.007 |
bulldo | +0.007 |
Whe | +0.007 |
Neg logit contributions | |
rey | -0.004 |
oto | -0.003 |
aleb | -0.003 |
hel | -0.003 |
pa | -0.003 |
Token | want |
Token ID | 765 |
Feature activation | -8.45e-01 |
Pos logit contributions | |
Help | +0.046 |
Was | +0.045 |
What | +0.044 |
Want | +0.044 |
Where | +0.043 |
Neg logit contributions | |
aleb | -0.030 |
rey | -0.029 |
oto | -0.029 |
isi | -0.028 |
i | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
Help | +0.013 |
What | +0.012 |
Was | +0.012 |
Do | +0.012 |
smashing | +0.012 |
Neg logit contributions | |
rey | -0.032 |
aleb | -0.031 |
iper | -0.030 |
asha | -0.030 |
aman | -0.029 |
Token | play |
Token ID | 711 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
Was | +0.055 |
joy | +0.054 |
What | +0.054 |
Want | +0.053 |
Do | +0.052 |
Neg logit contributions | |
rey | -0.049 |
aleb | -0.046 |
pa | -0.046 |
i | -0.044 |
oto | -0.043 |
Token | with |
Token ID | 351 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
Want | +0.050 |
Help | +0.049 |
Was | +0.049 |
What | +0.049 |
Do | +0.048 |
Neg logit contributions | |
rey | -0.057 |
hel | -0.055 |
aleb | -0.054 |
aman | -0.053 |
jackets | -0.052 |
Token | each |
Token ID | 1123 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
rey | +0.009 |
asha | +0.009 |
aman | +0.009 |
iper | +0.009 |
aleb | +0.009 |
Neg logit contributions | |
Help | -0.004 |
What | -0.004 |
tearing | -0.004 |
smashing | -0.004 |
Do | -0.004 |
Token | other |
Token ID | 584 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
onna | +0.151 |
Was | +0.149 |
Want | +0.146 |
joy | +0.142 |
What | +0.142 |
Neg logit contributions | |
pa | -0.067 |
rey | -0.065 |
uncle | -0.064 |
aleb | -0.063 |
i | -0.059 |
Token | anymore |
Token ID | 7471 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Help | +0.025 |
What | +0.025 |
Do | +0.024 |
Was | +0.024 |
smashing | +0.023 |
Neg logit contributions | |
rey | -0.051 |
asha | -0.050 |
aleb | -0.050 |
iper | -0.049 |
aman | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -5.42e-02 |
Pos logit contributions | |
Help | +0.024 |
What | +0.023 |
Was | +0.023 |
Do | +0.023 |
smashing | +0.023 |
Neg logit contributions | |
rey | -0.030 |
aleb | -0.028 |
asha | -0.028 |
iper | -0.027 |
i | -0.027 |
Token | They |
Token ID | 1119 |
Feature activation | +8.68e-01 |
Pos logit contributions | |
smashing | +0.001 |
Help | +0.001 |
tearing | +0.001 |
destroying | +0.001 |
free | +0.001 |
Neg logit contributions | |
rey | -0.001 |
aleb | -0.001 |
asha | -0.001 |
iper | -0.001 |
aman | -0.001 |
Token | sit |
Token ID | 1650 |
Feature activation | -9.16e-01 |
Pos logit contributions | |
rey | +0.030 |
asha | +0.029 |
aman | +0.029 |
aleb | +0.028 |
iper | +0.028 |
Neg logit contributions | |
What | -0.017 |
Help | -0.017 |
Was | -0.017 |
tearing | -0.016 |
Do | -0.016 |
Token | sister |
Token ID | 6621 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
rey | +0.013 |
asha | +0.013 |
aleb | +0.012 |
iper | +0.012 |
aman | +0.012 |
Neg logit contributions | |
Help | -0.028 |
What | -0.027 |
Was | -0.026 |
smashing | -0.026 |
tearing | -0.026 |
Token | yelled |
Token ID | 22760 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
Help | +0.022 |
What | +0.021 |
Was | +0.021 |
Do | +0.020 |
smashing | +0.020 |
Neg logit contributions | |
rey | -0.043 |
aleb | -0.042 |
asha | -0.041 |
iper | -0.041 |
aman | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Help | +0.021 |
Was | +0.020 |
What | +0.020 |
Do | +0.020 |
Where | +0.019 |
Neg logit contributions | |
rey | -0.027 |
aleb | -0.026 |
iper | -0.025 |
asha | -0.025 |
aman | -0.025 |
Token | " |
Token ID | 366 |
Feature activation | -9.60e-01 |
Pos logit contributions | |
Help | +0.013 |
Was | +0.012 |
What | +0.011 |
Do | +0.011 |
Where | +0.011 |
Neg logit contributions | |
rey | -0.051 |
aleb | -0.049 |
aman | -0.049 |
asha | -0.049 |
iper | -0.049 |
Token | Look |
Token ID | 8567 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
What | +0.018 |
Help | +0.018 |
Was | +0.017 |
Do | +0.017 |
smashing | +0.017 |
Neg logit contributions | |
rey | -0.033 |
iper | -0.033 |
aleb | -0.032 |
asha | -0.032 |
icity | -0.031 |
Token | out |
Token ID | 503 |
Feature activation | -4.33e+00 |
Pos logit contributions | |
Help | +0.024 |
What | +0.024 |
Was | +0.022 |
Do | +0.022 |
Where | +0.022 |
Neg logit contributions | |
rey | -0.038 |
aleb | -0.037 |
asha | -0.036 |
iper | -0.036 |
aman | -0.035 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
Whe | +0.110 |
Was | +0.109 |
Do | +0.106 |
Blood | +0.103 |
Did | +0.102 |
Neg logit contributions | |
rey | -0.104 |
aleb | -0.100 |
i | -0.096 |
aman | -0.094 |
iper | -0.093 |
Token | A |
Token ID | 317 |
Feature activation | -8.35e-01 |
Pos logit contributions | |
Help | +0.006 |
Do | +0.006 |
smashing | +0.006 |
What | +0.006 |
Was | +0.006 |
Neg logit contributions | |
rey | -0.009 |
aman | -0.009 |
aleb | -0.009 |
asha | -0.009 |
iper | -0.009 |
Token | big |
Token ID | 1263 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
What | +0.021 |
Help | +0.021 |
Do | +0.021 |
Where | +0.021 |
Was | +0.020 |
Neg logit contributions | |
rey | -0.023 |
aleb | -0.021 |
asha | -0.021 |
iper | -0.020 |
onia | -0.019 |
Token | fish |
Token ID | 5916 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
rey | +0.015 |
aleb | +0.014 |
asha | +0.014 |
iper | +0.013 |
aman | +0.013 |
Neg logit contributions | |
Help | -0.013 |
What | -0.013 |
Was | -0.013 |
Do | -0.012 |
Where | -0.012 |
Token | jumped |
Token ID | 11687 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
rey | +0.009 |
aleb | +0.009 |
iper | +0.009 |
asha | +0.009 |
aman | +0.009 |
Neg logit contributions | |
Help | -0.005 |
What | -0.005 |
Was | -0.005 |
Do | -0.005 |
smashing | -0.005 |
Token | had |
Token ID | 550 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Help | +0.027 |
What | +0.026 |
Was | +0.025 |
Do | +0.025 |
Where | +0.025 |
Neg logit contributions | |
rey | -0.040 |
aleb | -0.038 |
iper | -0.037 |
asha | -0.037 |
aman | -0.036 |
Token | fun |
Token ID | 1257 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
Help | +0.031 |
What | +0.031 |
Do | +0.030 |
Was | +0.030 |
smashing | +0.029 |
Neg logit contributions | |
rey | -0.039 |
aleb | -0.038 |
asha | -0.037 |
iper | -0.037 |
aman | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
Help | +0.040 |
Was | +0.040 |
Where | +0.040 |
What | +0.038 |
Do | +0.038 |
Neg logit contributions | |
rey | -0.029 |
i | -0.026 |
aleb | -0.026 |
pa | -0.025 |
ita | -0.025 |
Token | They |
Token ID | 1119 |
Feature activation | -9.70e-01 |
Pos logit contributions | |
Help | +0.006 |
smashing | +0.005 |
tearing | +0.005 |
Was | +0.005 |
destroying | +0.005 |
Neg logit contributions | |
rey | -0.008 |
aleb | -0.008 |
asha | -0.007 |
iper | -0.007 |
aman | -0.007 |
Token | made |
Token ID | 925 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
Was | +0.014 |
Help | +0.014 |
tearing | +0.014 |
What | +0.014 |
Do | +0.013 |
Neg logit contributions | |
aman | -0.035 |
aleb | -0.035 |
rey | -0.035 |
iper | -0.035 |
asha | -0.035 |
Token | each |
Token ID | 1123 |
Feature activation | -4.30e+00 |
Pos logit contributions | |
Help | +0.004 |
What | +0.004 |
smashing | +0.004 |
tearing | +0.003 |
Do | +0.003 |
Neg logit contributions | |
rey | -0.006 |
aleb | -0.006 |
asha | -0.006 |
iper | -0.005 |
aman | -0.005 |
Token | other |
Token ID | 584 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Was | +0.129 |
Help | +0.129 |
Where | +0.129 |
Do | +0.126 |
What | +0.126 |
Neg logit contributions | |
rey | -0.094 |
aleb | -0.084 |
ita | -0.079 |
iper | -0.079 |
pa | -0.077 |
Token | happy |
Token ID | 3772 |
Feature activation | -6.31e-01 |
Pos logit contributions | |
Help | +0.025 |
What | +0.024 |
Do | +0.023 |
Was | +0.023 |
smashing | +0.023 |
Neg logit contributions | |
rey | -0.039 |
aleb | -0.039 |
asha | -0.038 |
iper | -0.038 |
aman | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -6.24e-02 |
Pos logit contributions | |
Help | +0.015 |
Was | +0.015 |
Do | +0.015 |
Where | +0.015 |
What | +0.015 |
Neg logit contributions | |
rey | -0.018 |
aleb | -0.017 |
asha | -0.016 |
i | -0.016 |
iper | -0.016 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
Help | +0.001 |
smashing | +0.001 |
tearing | +0.001 |
Was | +0.001 |
destroying | +0.001 |
Neg logit contributions | |
rey | -0.002 |
aleb | -0.002 |
asha | -0.002 |
iper | -0.002 |
aman | -0.002 |
Token | L |
Token ID | 43 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
smashing | +0.021 |
Help | +0.020 |
tearing | +0.019 |
destroying | +0.019 |
knocking | +0.019 |
Neg logit contributions | |
rey | -0.024 |
aleb | -0.022 |
hel | -0.021 |
asha | -0.021 |
onia | -0.020 |
Token | my |
Token ID | 1820 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
What | +0.016 |
tearing | +0.015 |
Help | +0.015 |
Was | +0.015 |
Do | +0.015 |
Neg logit contributions | |
asha | -0.020 |
rey | -0.019 |
iper | -0.019 |
aleb | -0.019 |
ishment | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | -8.14e-02 |
Pos logit contributions | |
Help | +0.016 |
What | +0.016 |
Was | +0.016 |
tearing | +0.015 |
Do | +0.015 |
Neg logit contributions | |
rey | -0.034 |
aleb | -0.034 |
asha | -0.034 |
aman | -0.034 |
ishment | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
Help | +0.002 |
What | +0.002 |
Do | +0.002 |
Whe | +0.002 |
tearing | +0.002 |
Neg logit contributions | |
rey | -0.002 |
asha | -0.002 |
iper | -0.002 |
onia | -0.002 |
aman | -0.002 |
Token | kid |
Token ID | 5141 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
Help | +0.046 |
Do | +0.045 |
What | +0.045 |
smashing | +0.044 |
Was | +0.044 |
Neg logit contributions | |
rey | -0.051 |
aleb | -0.050 |
asha | -0.047 |
iper | -0.047 |
aman | -0.046 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
What | +0.019 |
Help | +0.019 |
Was | +0.019 |
Do | +0.018 |
tearing | +0.017 |
Neg logit contributions | |
rey | -0.032 |
asha | -0.031 |
aleb | -0.031 |
aman | -0.030 |
iper | -0.030 |
Token | each |
Token ID | 1123 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
Help | +0.002 |
What | +0.002 |
Was | +0.002 |
Do | +0.002 |
Where | +0.002 |
Neg logit contributions | |
rey | -0.008 |
aleb | -0.007 |
iper | -0.007 |
asha | -0.007 |
aman | -0.007 |
Token | other |
Token ID | 584 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
Was | +0.126 |
Help | +0.125 |
Do | +0.123 |
smart | +0.121 |
smashing | +0.119 |
Neg logit contributions | |
aleb | -0.094 |
rey | -0.093 |
ita | -0.087 |
iper | -0.085 |
asha | -0.084 |
Token | very |
Token ID | 845 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
Help | +0.034 |
What | +0.032 |
Was | +0.032 |
Do | +0.032 |
smashing | +0.031 |
Neg logit contributions | |
rey | -0.050 |
aleb | -0.049 |
iper | -0.047 |
asha | -0.047 |
aman | -0.046 |
Token | much |
Token ID | 881 |
Feature activation | +7.91e-01 |
Pos logit contributions | |
Help | +0.007 |
What | +0.007 |
Was | +0.006 |
Do | +0.006 |
tearing | +0.006 |
Neg logit contributions | |
rey | -0.024 |
asha | -0.023 |
aleb | -0.023 |
iper | -0.023 |
aman | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
rey | +0.032 |
asha | +0.031 |
iper | +0.031 |
aleb | +0.031 |
aman | +0.030 |
Neg logit contributions | |
What | -0.010 |
Help | -0.010 |
Was | -0.009 |
Do | -0.009 |
free | -0.009 |
Token | They |
Token ID | 1119 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
What | +0.004 |
Help | +0.004 |
Do | +0.003 |
Was | +0.003 |
Where | +0.003 |
Neg logit contributions | |
rey | -0.008 |
iper | -0.008 |
aleb | -0.008 |
asha | -0.008 |
i | -0.008 |
Token | I |
Token ID | 40 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
rey | +0.006 |
aleb | +0.006 |
asha | +0.006 |
iper | +0.005 |
aman | +0.005 |
Neg logit contributions | |
Help | -0.005 |
What | -0.005 |
Do | -0.005 |
Was | -0.005 |
smashing | -0.005 |
Token | 'm |
Token ID | 1101 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
smashing | +0.021 |
tearing | +0.021 |
Blood | +0.021 |
Glass | +0.021 |
smart | +0.020 |
Neg logit contributions | |
aleb | -0.023 |
rey | -0.023 |
ust | -0.022 |
hel | -0.022 |
i | -0.022 |
Token | sorry |
Token ID | 7926 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
Was | +0.007 |
Help | +0.007 |
What | +0.007 |
Where | +0.007 |
Do | +0.007 |
Neg logit contributions | |
rey | -0.014 |
aleb | -0.013 |
iper | -0.013 |
aman | -0.013 |
asha | -0.013 |
Token | " |
Token ID | 1 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
Was | +0.021 |
smashing | +0.021 |
Help | +0.021 |
tearing | +0.020 |
destroying | +0.020 |
Neg logit contributions | |
rey | -0.029 |
aleb | -0.028 |
i | -0.028 |
iper | -0.027 |
oto | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
iper | +0.023 |
aleb | +0.023 |
rey | +0.023 |
aman | +0.023 |
asha | +0.022 |
Neg logit contributions | |
What | -0.012 |
Help | -0.011 |
Do | -0.011 |
Was | -0.011 |
Where | -0.011 |
Token | each |
Token ID | 1123 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
Was | +0.030 |
Help | +0.030 |
What | +0.029 |
Where | +0.029 |
free | +0.028 |
Neg logit contributions | |
rey | -0.036 |
aleb | -0.033 |
iper | -0.032 |
aman | -0.032 |
asha | -0.032 |
Token | other |
Token ID | 584 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
Was | +0.126 |
What | +0.122 |
onna | +0.121 |
Where | +0.120 |
Do | +0.120 |
Neg logit contributions | |
rey | -0.088 |
aleb | -0.086 |
pa | -0.075 |
aman | -0.075 |
isi | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
Help | +0.044 |
Was | +0.043 |
What | +0.042 |
Do | +0.042 |
Where | +0.041 |
Neg logit contributions | |
rey | -0.053 |
aleb | -0.052 |
iper | -0.050 |
asha | -0.049 |
aman | -0.048 |
Token | They |
Token ID | 1119 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
rey | +0.004 |
aman | +0.004 |
aleb | +0.004 |
asha | +0.003 |
Charles | +0.003 |
Neg logit contributions | |
destroying | -0.006 |
smashing | -0.006 |
bulldo | -0.006 |
free | -0.006 |
tearing | -0.005 |
Token | hug |
Token ID | 16225 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
Help | +0.004 |
What | +0.004 |
Was | +0.004 |
tearing | +0.004 |
smashing | +0.004 |
Neg logit contributions | |
rey | -0.006 |
asha | -0.005 |
aleb | -0.005 |
aman | -0.005 |
iper | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -5.07e-01 |
Pos logit contributions | |
Help | +0.014 |
What | +0.014 |
Do | +0.014 |
Was | +0.013 |
smashing | +0.013 |
Neg logit contributions | |
rey | -0.017 |
aleb | -0.017 |
iper | -0.017 |
asha | -0.017 |
aman | -0.016 |
Token | They |
Token ID | 1119 |
Feature activation | -9.68e-01 |
Pos logit contributions | |
Help | +0.019 |
smashing | +0.019 |
tearing | +0.018 |
destroying | +0.018 |
Was | +0.018 |
Neg logit contributions | |
rey | -0.019 |
aleb | -0.019 |
asha | -0.018 |
aman | -0.018 |
iper | -0.018 |
Token | had |
Token ID | 550 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
Help | +0.018 |
Was | +0.018 |
tearing | +0.018 |
smashing | +0.017 |
What | +0.017 |
Neg logit contributions | |
rey | -0.033 |
aman | -0.032 |
aleb | -0.032 |
asha | -0.032 |
iper | -0.032 |
Token | fun |
Token ID | 1257 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Help | +0.019 |
What | +0.018 |
Was | +0.018 |
Where | +0.018 |
Do | +0.018 |
Neg logit contributions | |
rey | -0.015 |
aleb | -0.015 |
asha | -0.014 |
iper | -0.014 |
aman | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Where | +0.040 |
Was | +0.039 |
Help | +0.038 |
Want | +0.037 |
What | +0.037 |
Neg logit contributions | |
rey | -0.014 |
hel | -0.013 |
pa | -0.011 |
ishment | -0.011 |
i | -0.011 |
Token | made |
Token ID | 925 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
Help | +0.025 |
What | +0.024 |
Was | +0.024 |
Do | +0.023 |
Where | +0.023 |
Neg logit contributions | |
rey | -0.031 |
aleb | -0.030 |
iper | -0.030 |
asha | -0.029 |
aman | -0.029 |
Token | each |
Token ID | 1123 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
What | +0.007 |
Help | +0.007 |
tearing | +0.007 |
smashing | +0.007 |
Do | +0.007 |
Neg logit contributions | |
rey | -0.011 |
aleb | -0.011 |
asha | -0.011 |
iper | -0.011 |
aman | -0.011 |
Token | other |
Token ID | 584 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
Help | +0.130 |
Was | +0.128 |
onna | +0.125 |
Where | +0.125 |
smart | +0.124 |
Neg logit contributions | |
rey | -0.082 |
aleb | -0.075 |
ita | -0.070 |
iper | -0.067 |
asha | -0.065 |
Token | laugh |
Token ID | 6487 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
Help | +0.031 |
What | +0.029 |
Was | +0.029 |
Do | +0.029 |
Where | +0.028 |
Neg logit contributions | |
rey | -0.044 |
aleb | -0.043 |
iper | -0.042 |
asha | -0.042 |
aman | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
Help | +0.017 |
What | +0.017 |
Do | +0.016 |
Was | +0.016 |
smashing | +0.016 |
Neg logit contributions | |
rey | -0.030 |
aleb | -0.030 |
asha | -0.030 |
iper | -0.030 |
aman | -0.029 |
Token | Mom |
Token ID | 11254 |
Feature activation | -7.95e-01 |
Pos logit contributions | |
smashing | +0.011 |
Help | +0.010 |
tearing | +0.010 |
destroying | +0.010 |
smart | +0.010 |
Neg logit contributions | |
rey | -0.012 |
aleb | -0.012 |
asha | -0.012 |
aman | -0.011 |
iper | -0.011 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Help | +0.015 |
What | +0.015 |
Was | +0.015 |
tearing | +0.015 |
smashing | +0.015 |
Neg logit contributions | |
rey | -0.026 |
aleb | -0.026 |
iper | -0.025 |
asha | -0.025 |
aman | -0.025 |
Token | 'm |
Token ID | 1101 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
Help | +0.032 |
What | +0.031 |
Do | +0.030 |
smashing | +0.030 |
tearing | +0.030 |
Neg logit contributions | |
rey | -0.046 |
aleb | -0.045 |
asha | -0.043 |
iper | -0.043 |
aman | -0.042 |
Token | sorry |
Token ID | 7926 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Help | +0.009 |
What | +0.009 |
Do | +0.009 |
tearing | +0.008 |
Was | +0.008 |
Neg logit contributions | |
rey | -0.027 |
aleb | -0.027 |
asha | -0.026 |
iper | -0.026 |
aman | -0.026 |
Token | " |
Token ID | 1 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
Help | +0.024 |
Was | +0.023 |
smashing | +0.023 |
tearing | +0.023 |
What | +0.023 |
Neg logit contributions | |
rey | -0.039 |
aleb | -0.038 |
iper | -0.037 |
asha | -0.037 |
i | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -9.25e-01 |
Pos logit contributions | |
What | +0.002 |
Do | +0.002 |
Was | +0.002 |
Help | +0.002 |
Where | +0.002 |
Neg logit contributions | |
aleb | -0.004 |
rey | -0.004 |
iper | -0.004 |
aman | -0.004 |
asha | -0.004 |
Token | hugged |
Token ID | 41130 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
Help | +0.014 |
smashing | +0.014 |
Was | +0.014 |
destroying | +0.013 |
free | +0.013 |
Neg logit contributions | |
rey | -0.029 |
aleb | -0.029 |
i | -0.028 |
iper | -0.027 |
aman | -0.027 |
Token | each |
Token ID | 1123 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
Help | +0.025 |
What | +0.025 |
Do | +0.024 |
Was | +0.023 |
Where | +0.023 |
Neg logit contributions | |
rey | -0.047 |
aleb | -0.047 |
aman | -0.046 |
iper | -0.046 |
asha | -0.046 |
Token | other |
Token ID | 584 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
tearing | +0.117 |
Help | +0.117 |
destroying | +0.116 |
smashing | +0.116 |
Was | +0.116 |
Neg logit contributions | |
rey | -0.093 |
aleb | -0.091 |
iper | -0.084 |
asha | -0.083 |
oto | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
Help | +0.037 |
What | +0.036 |
Do | +0.034 |
Was | +0.034 |
smashing | +0.033 |
Neg logit contributions | |
rey | -0.069 |
aleb | -0.067 |
asha | -0.067 |
iper | -0.066 |
aman | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
smashing | +0.009 |
destroying | +0.009 |
tearing | +0.009 |
bulldo | +0.009 |
knocking | +0.009 |
Neg logit contributions | |
rey | -0.010 |
aleb | -0.009 |
aman | -0.009 |
asha | -0.009 |
onia | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -4.88e-01 |
Pos logit contributions | |
smashing | +0.016 |
Help | +0.016 |
tearing | +0.015 |
destroying | +0.015 |
Want | +0.015 |
Neg logit contributions | |
rey | -0.024 |
aleb | -0.022 |
iper | -0.022 |
oto | -0.022 |
asha | -0.022 |
Token | They |
Token ID | 2990 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
Help | +0.015 |
What | +0.014 |
Was | +0.014 |
Do | +0.014 |
smashing | +0.014 |
Neg logit contributions | |
rey | -0.011 |
aleb | -0.011 |
iper | -0.011 |
asha | -0.011 |
aman | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
Help | +0.036 |
Was | +0.035 |
Do | +0.035 |
What | +0.035 |
Where | +0.034 |
Neg logit contributions | |
rey | -0.032 |
aleb | -0.031 |
asha | -0.030 |
iper | -0.030 |
i | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
smashing | +0.014 |
destroying | +0.014 |
free | +0.014 |
tearing | +0.014 |
knocking | +0.013 |
Neg logit contributions | |
rey | -0.013 |
asha | -0.013 |
aleb | -0.013 |
Stanley | -0.013 |
Hugo | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
smashing | +0.004 |
destroying | +0.004 |
tearing | +0.003 |
free | +0.003 |
knocking | +0.003 |
Neg logit contributions | |
rey | -0.004 |
aleb | -0.004 |
asha | -0.004 |
oto | -0.003 |
iper | -0.003 |
Token | " |
Token ID | 1 |
Feature activation | -7.34e-01 |
Pos logit contributions | |
Help | +0.010 |
tearing | +0.010 |
smashing | +0.010 |
What | +0.010 |
destroying | +0.010 |
Neg logit contributions | |
rey | -0.013 |
aleb | -0.013 |
asha | -0.012 |
iper | -0.012 |
aman | -0.012 |
Token | Look |
Token ID | 8567 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
Help | +0.018 |
What | +0.017 |
Was | +0.017 |
Do | +0.017 |
smashing | +0.017 |
Neg logit contributions | |
rey | -0.021 |
aleb | -0.021 |
asha | -0.020 |
iper | -0.020 |
aman | -0.020 |
Token | out |
Token ID | 503 |
Feature activation | -4.11e+00 |
Pos logit contributions | |
Help | +0.023 |
Was | +0.022 |
smashing | +0.022 |
Do | +0.022 |
What | +0.021 |
Neg logit contributions | |
rey | -0.033 |
aleb | -0.033 |
iper | -0.032 |
i | -0.031 |
asha | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -9.88e-01 |
Pos logit contributions | |
Whe | +0.112 |
Was | +0.109 |
Do | +0.105 |
Are | +0.103 |
Did | +0.103 |
Neg logit contributions | |
rey | -0.089 |
i | -0.086 |
aleb | -0.083 |
pa | -0.081 |
iet | -0.081 |
Token | dog |
Token ID | 3290 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
Help | +0.021 |
Was | +0.020 |
What | +0.020 |
smashing | +0.020 |
Do | +0.020 |
Neg logit contributions | |
rey | -0.031 |
aleb | -0.030 |
iper | -0.029 |
asha | -0.029 |
aman | -0.029 |
Token | !" |
Token ID | 2474 |
Feature activation | +6.53e-01 |
Pos logit contributions | |
Help | +0.005 |
What | +0.005 |
Was | +0.005 |
Do | +0.005 |
Where | +0.005 |
Neg logit contributions | |
rey | -0.008 |
aleb | -0.008 |
asha | -0.008 |
iper | -0.008 |
aman | -0.008 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
rey | +0.021 |
aleb | +0.021 |
iper | +0.020 |
asha | +0.020 |
aman | +0.020 |
Neg logit contributions | |
Help | -0.014 |
What | -0.014 |
Do | -0.013 |
Was | -0.013 |
Where | -0.013 |
Token | shouts |
Token ID | 34757 |
Feature activation | -8.88e-01 |
Pos logit contributions | |
rey | +0.042 |
aleb | +0.041 |
aman | +0.041 |
asha | +0.040 |
iper | +0.040 |
Neg logit contributions | |
Help | -0.034 |
What | -0.033 |
Do | -0.032 |
Was | -0.032 |
tearing | -0.032 |
Token | share |
Token ID | 2648 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Help | +0.024 |
What | +0.022 |
Do | +0.022 |
Was | +0.022 |
tearing | +0.022 |
Neg logit contributions | |
rey | -0.035 |
aleb | -0.035 |
asha | -0.034 |
iper | -0.034 |
aman | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
Help | +0.028 |
What | +0.027 |
Was | +0.026 |
Where | +0.026 |
Do | +0.026 |
Neg logit contributions | |
rey | -0.036 |
aleb | -0.035 |
asha | -0.034 |
iper | -0.034 |
aman | -0.033 |
Token | be |
Token ID | 307 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
Help | +0.012 |
What | +0.011 |
Do | +0.011 |
Was | +0.011 |
tearing | +0.010 |
Neg logit contributions | |
rey | -0.023 |
aleb | -0.023 |
asha | -0.022 |
iper | -0.022 |
aman | -0.022 |
Token | nice |
Token ID | 3621 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
Help | +0.057 |
What | +0.055 |
Was | +0.054 |
Where | +0.053 |
smashing | +0.053 |
Neg logit contributions | |
rey | -0.075 |
aleb | -0.073 |
iper | -0.070 |
asha | -0.070 |
aman | -0.068 |
Token | to |
Token ID | 284 |
Feature activation | -8.60e-01 |
Pos logit contributions | |
Help | +0.004 |
What | +0.004 |
Do | +0.004 |
Was | +0.004 |
Where | +0.004 |
Neg logit contributions | |
rey | -0.008 |
aleb | -0.008 |
asha | -0.008 |
iper | -0.007 |
aman | -0.007 |
Token | each |
Token ID | 1123 |
Feature activation | -4.09e+00 |
Pos logit contributions | |
smashing | +0.030 |
Was | +0.030 |
Help | +0.029 |
bulldo | +0.029 |
Where | +0.029 |
Neg logit contributions | |
rey | -0.015 |
aleb | -0.013 |
i | -0.013 |
Him | -0.012 |
iper | -0.012 |
Token | other |
Token ID | 584 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
Was | +0.132 |
smashing | +0.130 |
Where | +0.130 |
Help | +0.129 |
tearing | +0.128 |
Neg logit contributions | |
rey | -0.079 |
aleb | -0.077 |
oto | -0.072 |
asha | -0.072 |
i | -0.072 |
Token | ." |
Token ID | 526 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
Help | +0.002 |
free | +0.002 |
What | +0.002 |
Do | +0.002 |
tearing | +0.002 |
Neg logit contributions | |
rey | -0.006 |
asha | -0.006 |
aman | -0.006 |
aleb | -0.006 |
iper | -0.006 |
Token | Lily |
Token ID | 20037 |
Feature activation | +9.88e-01 |
Pos logit contributions | |
Help | +0.003 |
smashing | +0.003 |
Was | +0.003 |
What | +0.003 |
tearing | +0.003 |
Neg logit contributions | |
rey | -0.006 |
aleb | -0.006 |
iper | -0.006 |
asha | -0.006 |
aman | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +4.43e-01 |
Pos logit contributions | |
rey | +0.029 |
aleb | +0.029 |
aman | +0.028 |
iper | +0.028 |
asha | +0.027 |
Neg logit contributions | |
Help | -0.022 |
What | -0.021 |
tearing | -0.021 |
Was | -0.020 |
v | -0.020 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
rey | +0.016 |
aman | +0.015 |
asha | +0.015 |
onia | +0.015 |
aleb | +0.015 |
Neg logit contributions | |
Help | -0.008 |
What | -0.008 |
Whe | -0.008 |
tearing | -0.008 |
Do | -0.008 |
Token | She |
Token ID | 1375 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
rey | +0.004 |
asha | +0.004 |
aleb | +0.004 |
iper | +0.004 |
aman | +0.004 |
Neg logit contributions | |
smashing | -0.006 |
destroying | -0.006 |
tearing | -0.006 |
biting | -0.005 |
knocking | -0.005 |
Token | said |
Token ID | 531 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
Help | +0.001 |
Was | +0.001 |
smashing | +0.001 |
tearing | +0.001 |
What | +0.001 |
Neg logit contributions | |
aleb | -0.002 |
aman | -0.002 |
iper | -0.002 |
asha | -0.002 |
rey | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -9.91e-01 |
Pos logit contributions | |
rey | +0.001 |
asha | +0.001 |
iper | +0.001 |
i | +0.001 |
aleb | +0.001 |
Neg logit contributions | |
smashing | -0.001 |
Help | -0.001 |
Was | -0.001 |
tearing | -0.001 |
What | -0.001 |
Token | " |
Token ID | 366 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
Help | +0.014 |
smashing | +0.014 |
smart | +0.013 |
Was | +0.013 |
matched | +0.013 |
Neg logit contributions | |
rey | -0.032 |
Him | -0.030 |
Hugo | -0.030 |
asha | -0.030 |
aman | -0.029 |
Token | Come |
Token ID | 16773 |
Feature activation | +5.35e-01 |
Pos logit contributions | |
rey | +0.005 |
aleb | +0.005 |
iper | +0.005 |
asha | +0.005 |
aman | +0.005 |
Neg logit contributions | |
Help | -0.003 |
What | -0.003 |
Was | -0.003 |
Do | -0.003 |
Where | -0.003 |
Token | out |
Token ID | 503 |
Feature activation | -4.08e+00 |
Pos logit contributions | |
rey | +0.016 |
aleb | +0.016 |
asha | +0.015 |
i | +0.015 |
iper | +0.015 |
Neg logit contributions | |
smashing | -0.011 |
Was | -0.011 |
Help | -0.011 |
tearing | -0.011 |
Blood | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
Whe | +0.094 |
Was | +0.093 |
smashing | +0.089 |
Where | +0.088 |
What | +0.087 |
Neg logit contributions | |
rey | -0.109 |
i | -0.103 |
iper | -0.099 |
ita | -0.099 |
aman | -0.098 |
Token | crab |
Token ID | 32202 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
Help | +0.010 |
Do | +0.009 |
What | +0.009 |
Where | +0.009 |
poking | +0.008 |
Neg logit contributions | |
rey | -0.018 |
iper | -0.017 |
aleb | -0.017 |
onia | -0.017 |
asha | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
rey | +0.015 |
aleb | +0.015 |
iper | +0.015 |
asha | +0.015 |
aman | +0.014 |
Neg logit contributions | |
Help | -0.007 |
What | -0.007 |
Do | -0.006 |
Was | -0.006 |
smashing | -0.006 |
Token | come |
Token ID | 1282 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
Help | +0.008 |
What | +0.008 |
Do | +0.007 |
Was | +0.007 |
Where | +0.007 |
Neg logit contributions | |
rey | -0.011 |
aleb | -0.011 |
iper | -0.011 |
asha | -0.011 |
aman | -0.010 |
Token | out |
Token ID | 503 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
Help | +0.015 |
What | +0.015 |
Was | +0.014 |
Do | +0.014 |
smashing | +0.014 |
Neg logit contributions | |
rey | -0.020 |
aleb | -0.020 |
asha | -0.019 |
iper | -0.019 |
aman | -0.019 |
Token | Anna |
Token ID | 11735 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
Help | +0.006 |
What | +0.006 |
Do | +0.005 |
Was | +0.005 |
tearing | +0.005 |
Neg logit contributions | |
rey | -0.009 |
aleb | -0.009 |
iper | -0.009 |
asha | -0.009 |
aman | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
Help | +0.015 |
What | +0.014 |
Do | +0.014 |
tearing | +0.014 |
Was | +0.013 |
Neg logit contributions | |
rey | -0.030 |
aman | -0.029 |
aleb | -0.029 |
asha | -0.029 |
iper | -0.028 |
Token | They |
Token ID | 1119 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
rey | +0.011 |
aleb | +0.011 |
iper | +0.010 |
asha | +0.010 |
i | +0.010 |
Neg logit contributions | |
What | -0.005 |
Do | -0.005 |
Help | -0.005 |
Where | -0.005 |
Was | -0.005 |
Token | also |
Token ID | 635 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
Was | +0.022 |
Help | +0.021 |
What | +0.021 |
tearing | +0.020 |
Do | +0.020 |
Neg logit contributions | |
rey | -0.044 |
iper | -0.044 |
aman | -0.044 |
aleb | -0.043 |
asha | -0.043 |
Token | liked |
Token ID | 8288 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
Help | +0.065 |
What | +0.063 |
Do | +0.062 |
Was | +0.062 |
Where | +0.061 |
Neg logit contributions | |
rey | -0.075 |
aleb | -0.072 |
asha | -0.070 |
iper | -0.070 |
aman | -0.068 |
Token | each |
Token ID | 1123 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
Help | +0.011 |
What | +0.010 |
Do | +0.010 |
Was | +0.010 |
tearing | +0.010 |
Neg logit contributions | |
rey | -0.022 |
aleb | -0.021 |
iper | -0.021 |
asha | -0.021 |
aman | -0.021 |
Token | other |
Token ID | 584 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
Help | +0.121 |
Was | +0.120 |
smashing | +0.119 |
Where | +0.118 |
Do | +0.118 |
Neg logit contributions | |
rey | -0.084 |
aleb | -0.082 |
asha | -0.076 |
pa | -0.075 |
i | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
What | +0.013 |
Help | +0.013 |
Was | +0.012 |
Do | +0.012 |
tearing | +0.012 |
Neg logit contributions | |
rey | -0.024 |
aleb | -0.023 |
asha | -0.023 |
aman | -0.023 |
iper | -0.023 |
Token | They |
Token ID | 1119 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
rey | +0.022 |
aleb | +0.022 |
iper | +0.022 |
asha | +0.022 |
aman | +0.021 |
Neg logit contributions | |
What | -0.010 |
Do | -0.010 |
Help | -0.010 |
Where | -0.010 |
Was | -0.010 |
Token | had |
Token ID | 550 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
Was | +0.018 |
Help | +0.017 |
What | +0.016 |
tearing | +0.016 |
Do | +0.016 |
Neg logit contributions | |
aman | -0.042 |
iper | -0.041 |
rey | -0.041 |
aleb | -0.040 |
asha | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
Help | +0.027 |
What | +0.026 |
Do | +0.026 |
Was | +0.025 |
tearing | +0.025 |
Neg logit contributions | |
rey | -0.043 |
aleb | -0.041 |
iper | -0.041 |
aman | -0.040 |
asha | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
Help | +0.012 |
What | +0.012 |
Was | +0.012 |
smashing | +0.012 |
Do | +0.012 |
Neg logit contributions | |
rey | -0.024 |
aleb | -0.024 |
iper | -0.023 |
asha | -0.023 |
aman | -0.023 |
Token | Now |
Token ID | 2735 |
Feature activation | -8.94e-01 |
Pos logit contributions | |
Help | +0.030 |
smashing | +0.030 |
Was | +0.029 |
tearing | +0.029 |
destroying | +0.029 |
Neg logit contributions | |
rey | -0.055 |
aleb | -0.055 |
iper | -0.053 |
asha | -0.052 |
aman | -0.051 |
Token | say |
Token ID | 910 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
Help | +0.013 |
What | +0.013 |
Do | +0.012 |
Was | +0.012 |
Where | +0.012 |
Neg logit contributions | |
rey | -0.035 |
aleb | -0.034 |
asha | -0.034 |
iper | -0.034 |
aman | -0.033 |
Token | sorry |
Token ID | 7926 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
Was | +0.041 |
smashing | +0.039 |
bulldo | +0.039 |
Help | +0.038 |
Whe | +0.038 |
Neg logit contributions | |
rey | -0.046 |
iper | -0.045 |
Him | -0.044 |
icity | -0.043 |
asha | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
smashing | +0.032 |
Help | +0.031 |
tearing | +0.031 |
destroying | +0.030 |
Was | +0.030 |
Neg logit contributions | |
rey | -0.058 |
iper | -0.056 |
aleb | -0.056 |
asha | -0.055 |
i | -0.054 |
Token | each |
Token ID | 1123 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
Help | +0.033 |
Was | +0.032 |
smashing | +0.032 |
What | +0.032 |
Do | +0.031 |
Neg logit contributions | |
rey | -0.047 |
aleb | -0.045 |
iper | -0.044 |
asha | -0.044 |
aman | -0.043 |
Token | other |
Token ID | 584 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
Was | +0.132 |
smashing | +0.130 |
onna | +0.127 |
destroying | +0.127 |
Help | +0.126 |
Neg logit contributions | |
rey | -0.072 |
pa | -0.070 |
aleb | -0.069 |
isi | -0.067 |
iper | -0.066 |
Token | ." |
Token ID | 526 |
Feature activation | -6.73e-02 |
Pos logit contributions | |
Help | +0.030 |
What | +0.029 |
Was | +0.029 |
Do | +0.029 |
Where | +0.029 |
Neg logit contributions | |
rey | -0.043 |
aleb | -0.042 |
iper | -0.040 |
asha | -0.040 |
aman | -0.040 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
smashing | +0.002 |
destroying | +0.002 |
smart | +0.002 |
Help | +0.002 |
tearing | +0.002 |
Neg logit contributions | |
rey | -0.002 |
aleb | -0.002 |
iper | -0.002 |
aman | -0.002 |
asha | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
rey | +0.034 |
aleb | +0.033 |
iper | +0.033 |
asha | +0.033 |
aman | +0.032 |
Neg logit contributions | |
Help | -0.019 |
What | -0.019 |
Was | -0.018 |
tearing | -0.018 |
smashing | -0.018 |
Token | Anna |
Token ID | 11735 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
rey | +0.038 |
asha | +0.037 |
aleb | +0.037 |
iper | +0.037 |
aman | +0.036 |
Neg logit contributions | |
Help | -0.025 |
What | -0.024 |
tearing | -0.023 |
Do | -0.023 |
Whe | -0.023 |
Token | Daisy |
Token ID | 40355 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
aleb | +0.020 |
iper | +0.020 |
rey | +0.020 |
asha | +0.019 |
aman | +0.019 |
Neg logit contributions | |
Help | -0.010 |
What | -0.010 |
Whe | -0.009 |
Do | -0.009 |
Where | -0.009 |
Token | accidentally |
Token ID | 14716 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
aleb | +0.001 |
rey | +0.001 |
aman | +0.001 |
Him | +0.001 |
ishment | +0.001 |
Neg logit contributions | |
Help | -0.000 |
Was | -0.000 |
What | -0.000 |
tearing | -0.000 |
Where | -0.000 |
Token | broke |
Token ID | 6265 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
Help | +0.013 |
smashing | +0.012 |
tearing | +0.012 |
What | +0.012 |
Was | +0.011 |
Neg logit contributions | |
rey | -0.033 |
aleb | -0.033 |
asha | -0.032 |
iper | -0.032 |
aman | -0.031 |
Token | Ph |
Token ID | 1380 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
tearing | +0.011 |
destroying | +0.011 |
smashing | +0.011 |
Help | +0.011 |
What | +0.011 |
Neg logit contributions | |
iper | -0.018 |
aleb | -0.018 |
rey | -0.018 |
asha | -0.018 |
iri | -0.017 |
Token | oe |
Token ID | 2577 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
Help | +0.001 |
What | +0.001 |
Was | +0.001 |
Do | +0.001 |
Where | +0.001 |
Neg logit contributions | |
rey | -0.002 |
aleb | -0.002 |
iper | -0.002 |
asha | -0.002 |
aman | -0.002 |
Token | be |
Token ID | 1350 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
Do | +0.087 |
What | +0.085 |
Where | +0.084 |
Was | +0.083 |
Help | +0.081 |
Neg logit contributions | |
rey | -0.115 |
i | -0.113 |
aleb | -0.113 |
iper | -0.109 |
asha | -0.108 |
Token | 's |
Token ID | 338 |
Feature activation | -9.87e-01 |
Pos logit contributions | |
Do | +0.115 |
Want | +0.111 |
Whe | +0.108 |
Where | +0.107 |
What | +0.107 |
Neg logit contributions | |
i | -0.097 |
pa | -0.091 |
rey | -0.091 |
ita | -0.091 |
iper | -0.089 |
Token | favourite |
Token ID | 12507 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
Help | +0.021 |
tearing | +0.020 |
smashing | +0.020 |
What | +0.019 |
destroying | +0.019 |
Neg logit contributions | |
rey | -0.032 |
asha | -0.031 |
aleb | -0.031 |
iper | -0.030 |
onia | -0.030 |
Token | toy |
Token ID | 13373 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
Help | +0.042 |
What | +0.040 |
Was | +0.040 |
Do | +0.040 |
Where | +0.039 |
Neg logit contributions | |
rey | -0.058 |
aleb | -0.056 |
asha | -0.055 |
iper | -0.055 |
aman | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | -8.17e-01 |
Pos logit contributions | |
Help | +0.001 |
tearing | +0.001 |
smashing | +0.001 |
What | +0.001 |
destroying | +0.001 |
Neg logit contributions | |
rey | -0.002 |
asha | -0.002 |
aleb | -0.002 |
iper | -0.002 |
aman | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
tearing | +0.016 |
smashing | +0.016 |
Help | +0.016 |
What | +0.016 |
destroying | +0.015 |
Neg logit contributions | |
rey | -0.029 |
aleb | -0.029 |
asha | -0.028 |
iper | -0.028 |
onia | -0.027 |
Token | very |
Token ID | 845 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
rey | +0.019 |
aleb | +0.018 |
asha | +0.018 |
iper | +0.018 |
aman | +0.018 |
Neg logit contributions | |
Help | -0.010 |
What | -0.010 |
smashing | -0.009 |
tearing | -0.009 |
Do | -0.009 |
Token | stubborn |
Token ID | 24444 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
Help | +0.006 |
What | +0.006 |
Was | +0.006 |
Do | +0.006 |
Where | +0.006 |
Neg logit contributions | |
rey | -0.012 |
aleb | -0.011 |
iper | -0.011 |
asha | -0.011 |
aman | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
rey | +0.003 |
aleb | +0.003 |
asha | +0.003 |
iper | +0.003 |
aman | +0.003 |
Neg logit contributions | |
Help | -0.002 |
What | -0.002 |
tearing | -0.002 |
smashing | -0.002 |
Was | -0.002 |
Token | They |
Token ID | 1119 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
smashing | +0.010 |
Help | +0.010 |
tearing | +0.010 |
smart | +0.010 |
destroying | +0.010 |
Neg logit contributions | |
rey | -0.011 |
iper | -0.011 |
aleb | -0.010 |
asha | -0.010 |
aman | -0.010 |
Token | made |
Token ID | 925 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
rey | +0.002 |
asha | +0.002 |
aman | +0.002 |
aleb | +0.002 |
iper | +0.002 |
Neg logit contributions | |
Was | -0.001 |
Help | -0.001 |
What | -0.001 |
tearing | -0.001 |
smashing | -0.001 |
Token | each |
Token ID | 1123 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
rey | +0.005 |
aleb | +0.005 |
asha | +0.005 |
iper | +0.004 |
aman | +0.004 |
Neg logit contributions | |
Help | -0.003 |
What | -0.003 |
tearing | -0.003 |
smashing | -0.003 |
Was | -0.003 |
Token | other |
Token ID | 584 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Was | +0.112 |
Help | +0.111 |
Do | +0.111 |
Where | +0.109 |
What | +0.109 |
Neg logit contributions | |
rey | -0.089 |
aleb | -0.086 |
iper | -0.078 |
ita | -0.078 |
asha | -0.075 |
Token | sad |
Token ID | 6507 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
Help | +0.028 |
What | +0.028 |
Was | +0.028 |
Do | +0.028 |
Where | +0.027 |
Neg logit contributions | |
rey | -0.036 |
aleb | -0.035 |
iper | -0.034 |
asha | -0.033 |
aman | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
Help | +0.011 |
What | +0.010 |
Was | +0.010 |
Do | +0.010 |
smashing | +0.010 |
Neg logit contributions | |
rey | -0.017 |
aleb | -0.017 |
asha | -0.016 |
iper | -0.016 |
aman | -0.016 |
Token | angry |
Token ID | 7954 |
Feature activation | -5.36e-01 |
Pos logit contributions | |
rey | +0.018 |
asha | +0.018 |
aleb | +0.018 |
aman | +0.017 |
iper | +0.017 |
Neg logit contributions | |
tearing | -0.009 |
Help | -0.009 |
smashing | -0.009 |
Do | -0.009 |
destroying | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
Help | +0.011 |
What | +0.011 |
Was | +0.010 |
Do | +0.010 |
Where | +0.010 |
Neg logit contributions | |
rey | -0.018 |
aleb | -0.017 |
asha | -0.017 |
iper | -0.017 |
aman | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
i | +0.014 |
iper | +0.013 |
aleb | +0.013 |
hel | +0.013 |
Him | +0.013 |
Neg logit contributions | |
Want | -0.021 |
Was | -0.021 |
Where | -0.021 |
Do | -0.021 |
Whe | -0.021 |
Token | " |
Token ID | 366 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
rey | +0.006 |
asha | +0.006 |
aman | +0.006 |
ita | +0.006 |
aleb | +0.006 |
Neg logit contributions | |
free | -0.005 |
bulldo | -0.005 |
smashing | -0.005 |
destroying | -0.005 |
Want | -0.005 |
Token | You |
Token ID | 1639 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
Help | +0.028 |
Was | +0.027 |
What | +0.026 |
Where | +0.026 |
Do | +0.026 |
Neg logit contributions | |
rey | -0.024 |
aleb | -0.024 |
asha | -0.022 |
aman | -0.021 |
iper | -0.021 |
Token | are |
Token ID | 389 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
Help | +0.012 |
What | +0.012 |
Was | +0.011 |
Do | +0.011 |
Where | +0.011 |
Neg logit contributions | |
rey | -0.016 |
aleb | -0.016 |
iper | -0.016 |
asha | -0.015 |
aman | -0.015 |
Token | hurting |
Token ID | 21530 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
Was | +0.006 |
Help | +0.006 |
Where | +0.006 |
What | +0.006 |
Want | +0.006 |
Neg logit contributions | |
rey | -0.008 |
aleb | -0.008 |
iper | -0.007 |
icity | -0.007 |
onia | -0.007 |
Token | each |
Token ID | 1123 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
Help | +0.030 |
What | +0.029 |
Do | +0.028 |
Was | +0.028 |
smashing | +0.028 |
Neg logit contributions | |
rey | -0.041 |
aleb | -0.040 |
iper | -0.040 |
asha | -0.040 |
aman | -0.039 |
Token | other |
Token ID | 584 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
onna | +0.138 |
Was | +0.137 |
Help | +0.130 |
Where | +0.130 |
Do | +0.129 |
Neg logit contributions | |
rey | -0.062 |
aleb | -0.062 |
pa | -0.057 |
oto | -0.053 |
isi | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
Help | +0.036 |
What | +0.034 |
smashing | +0.033 |
Was | +0.033 |
Do | +0.033 |
Neg logit contributions | |
rey | -0.056 |
aleb | -0.054 |
asha | -0.053 |
iper | -0.053 |
aman | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Help | +0.006 |
What | +0.006 |
Do | +0.006 |
Was | +0.006 |
smashing | +0.006 |
Neg logit contributions | |
rey | -0.009 |
aleb | -0.008 |
asha | -0.008 |
iper | -0.008 |
aman | -0.008 |
Token | car |
Token ID | 1097 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
Help | +0.027 |
What | +0.026 |
Was | +0.026 |
Do | +0.025 |
Where | +0.025 |
Neg logit contributions | |
rey | -0.033 |
aleb | -0.032 |
iper | -0.030 |
asha | -0.030 |
aman | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
Help | +0.020 |
What | +0.019 |
Do | +0.018 |
Was | +0.018 |
smashing | +0.018 |
Neg logit contributions | |
rey | -0.038 |
aleb | -0.038 |
asha | -0.037 |
iper | -0.037 |
aman | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
Help | +0.034 |
What | +0.033 |
Do | +0.032 |
Was | +0.032 |
tearing | +0.032 |
Neg logit contributions | |
rey | -0.043 |
aleb | -0.042 |
asha | -0.041 |
iper | -0.041 |
aman | -0.040 |
Token | daddy |
Token ID | 35822 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
Help | +0.008 |
What | +0.008 |
Do | +0.008 |
tearing | +0.008 |
Where | +0.007 |
Neg logit contributions | |
rey | -0.011 |
asha | -0.010 |
aleb | -0.010 |
iper | -0.010 |
aman | -0.010 |
Token | who |
Token ID | 508 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Help | +0.033 |
What | +0.032 |
Was | +0.031 |
Do | +0.031 |
smashing | +0.031 |
Neg logit contributions | |
rey | -0.042 |
aleb | -0.041 |
iper | -0.039 |
asha | -0.039 |
aman | -0.039 |
Token | both |
Token ID | 1111 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
What | +0.023 |
Help | +0.023 |
Was | +0.023 |
Do | +0.022 |
Where | +0.022 |
Neg logit contributions | |
rey | -0.035 |
asha | -0.034 |
iper | -0.034 |
aleb | -0.033 |
aman | -0.033 |
Token | loved |
Token ID | 6151 |
Feature activation | -8.50e-02 |
Pos logit contributions | |
Help | +0.058 |
Do | +0.055 |
smashing | +0.055 |
tearing | +0.054 |
Was | +0.054 |
Neg logit contributions | |
rey | -0.049 |
aleb | -0.049 |
iper | -0.046 |
i | -0.046 |
asha | -0.045 |
Token | each |
Token ID | 1123 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
Help | +0.001 |
What | +0.001 |
Do | +0.001 |
Was | +0.001 |
smashing | +0.001 |
Neg logit contributions | |
rey | -0.003 |
aleb | -0.003 |
asha | -0.003 |
iper | -0.003 |
aman | -0.003 |
Token | other |
Token ID | 584 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Help | +0.105 |
Was | +0.104 |
Do | +0.102 |
What | +0.101 |
smashing | +0.099 |
Neg logit contributions | |
rey | -0.102 |
aleb | -0.101 |
aman | -0.095 |
asha | -0.095 |
iper | -0.095 |
Token | very |
Token ID | 845 |
Feature activation | -8.69e-01 |
Pos logit contributions | |
Help | +0.028 |
Do | +0.026 |
What | +0.026 |
Was | +0.026 |
Where | +0.025 |
Neg logit contributions | |
rey | -0.037 |
aleb | -0.036 |
iper | -0.034 |
asha | -0.034 |
aman | -0.034 |
Token | much |
Token ID | 881 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
Help | +0.014 |
What | +0.013 |
Was | +0.012 |
Do | +0.012 |
Where | +0.012 |
Neg logit contributions | |
rey | -0.033 |
aleb | -0.032 |
asha | -0.031 |
iper | -0.031 |
aman | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
rey | +0.034 |
asha | +0.033 |
iper | +0.033 |
aleb | +0.033 |
aman | +0.032 |
Neg logit contributions | |
What | -0.011 |
Help | -0.011 |
Do | -0.010 |
Was | -0.010 |
free | -0.010 |
Token | Every |
Token ID | 3887 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
What | +0.003 |
Help | +0.003 |
Do | +0.003 |
Was | +0.003 |
Where | +0.003 |
Neg logit contributions | |
rey | -0.006 |
iper | -0.006 |
aleb | -0.006 |
asha | -0.006 |
i | -0.005 |
Token | Mitt |
Token ID | 16627 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Help | +0.023 |
What | +0.022 |
Do | +0.021 |
tearing | +0.021 |
smashing | +0.021 |
Neg logit contributions | |
aleb | -0.041 |
rey | -0.041 |
iper | -0.039 |
asha | -0.039 |
onia | -0.037 |
Token | ens |
Token ID | 641 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
rey | +0.033 |
aleb | +0.031 |
aman | +0.030 |
iper | +0.029 |
asha | +0.029 |
Neg logit contributions | |
What | -0.037 |
Do | -0.037 |
Help | -0.036 |
Was | -0.035 |
Where | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
rey | +0.016 |
aleb | +0.016 |
asha | +0.015 |
iper | +0.015 |
aman | +0.015 |
Neg logit contributions | |
Help | -0.009 |
What | -0.008 |
Was | -0.008 |
Do | -0.008 |
tearing | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
smart | +0.028 |
smashing | +0.027 |
Help | +0.026 |
destroying | +0.026 |
tearing | +0.026 |
Neg logit contributions | |
aman | -0.047 |
iper | -0.047 |
aleb | -0.047 |
asha | -0.047 |
rey | -0.046 |
Token | Look |
Token ID | 8567 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
Help | +0.023 |
What | +0.022 |
Was | +0.022 |
Do | +0.022 |
smashing | +0.021 |
Neg logit contributions | |
rey | -0.035 |
aleb | -0.034 |
asha | -0.033 |
iper | -0.033 |
aman | -0.032 |
Token | out |
Token ID | 503 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
Help | +0.031 |
Was | +0.030 |
Do | +0.030 |
What | +0.030 |
smashing | +0.029 |
Neg logit contributions | |
rey | -0.040 |
aleb | -0.040 |
iper | -0.039 |
i | -0.039 |
asha | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
Whe | +0.111 |
Was | +0.108 |
Are | +0.102 |
Do | +0.101 |
Blood | +0.100 |
Neg logit contributions | |
i | -0.081 |
rey | -0.080 |
uncle | -0.078 |
pa | -0.077 |
iper | -0.077 |
Token | Tim |
Token ID | 5045 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
Help | +0.046 |
Was | +0.044 |
What | +0.044 |
smashing | +0.044 |
tearing | +0.043 |
Neg logit contributions | |
rey | -0.076 |
aleb | -0.075 |
iper | -0.073 |
asha | -0.073 |
aman | -0.072 |
Token | my |
Token ID | 1820 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
Help | +0.010 |
What | +0.010 |
Was | +0.010 |
Do | +0.010 |
Where | +0.010 |
Neg logit contributions | |
rey | -0.006 |
aleb | -0.006 |
iper | -0.005 |
asha | -0.005 |
aman | -0.005 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
Help | +0.001 |
What | +0.001 |
Was | +0.001 |
Do | +0.001 |
Where | +0.001 |
Neg logit contributions | |
rey | -0.002 |
aleb | -0.002 |
asha | -0.002 |
iper | -0.002 |
aman | -0.002 |
Token | yelled |
Token ID | 22760 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
rey | +0.007 |
aleb | +0.007 |
asha | +0.007 |
iper | +0.007 |
aman | +0.007 |
Neg logit contributions | |
Help | -0.005 |
What | -0.005 |
Do | -0.005 |
Was | -0.005 |
tearing | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
Help | +0.023 |
What | +0.022 |
Was | +0.022 |
Do | +0.022 |
Where | +0.021 |
Neg logit contributions | |
rey | -0.032 |
aleb | -0.031 |
asha | -0.031 |
iper | -0.030 |
aman | -0.030 |
Token | He |
Token ID | 679 |
Feature activation | -8.25e-01 |
Pos logit contributions | |
Help | +0.005 |
smashing | +0.004 |
Was | +0.004 |
What | +0.004 |
tearing | +0.004 |
Neg logit contributions | |
rey | -0.012 |
aleb | -0.012 |
asha | -0.011 |
iper | -0.011 |
aman | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
Help | +0.020 |
What | +0.019 |
Was | +0.019 |
Do | +0.019 |
smashing | +0.019 |
Neg logit contributions | |
rey | -0.024 |
aleb | -0.022 |
iper | -0.022 |
asha | -0.022 |
i | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
Help | +0.011 |
What | +0.011 |
Was | +0.010 |
Do | +0.010 |
smashing | +0.010 |
Neg logit contributions | |
rey | -0.015 |
aleb | -0.014 |
iper | -0.014 |
asha | -0.014 |
aman | -0.014 |
Token | mis |
Token ID | 2984 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
What | +0.026 |
Help | +0.026 |
Was | +0.026 |
Do | +0.026 |
smashing | +0.025 |
Neg logit contributions | |
rey | -0.030 |
aleb | -0.029 |
iper | -0.027 |
asha | -0.027 |
aman | -0.026 |
Token | chie |
Token ID | 3043 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
What | +0.020 |
Help | +0.020 |
Do | +0.020 |
Was | +0.020 |
Where | +0.019 |
Neg logit contributions | |
aleb | -0.035 |
rey | -0.035 |
iper | -0.034 |
asha | -0.033 |
ita | -0.032 |
Token | vous |
Token ID | 31222 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
Help | +0.086 |
smashing | +0.084 |
destroying | +0.083 |
tearing | +0.083 |
Was | +0.082 |
Neg logit contributions | |
rey | -0.118 |
iper | -0.114 |
aleb | -0.114 |
ita | -0.114 |
asha | -0.111 |
Token | little |
Token ID | 1310 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
Help | +0.003 |
What | +0.003 |
Was | +0.003 |
Do | +0.003 |
smashing | +0.003 |
Neg logit contributions | |
rey | -0.003 |
aleb | -0.003 |
asha | -0.003 |
iper | -0.003 |
aman | -0.003 |
Token | baby |
Token ID | 5156 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
rey | +0.012 |
aleb | +0.012 |
asha | +0.012 |
iper | +0.012 |
aman | +0.011 |
Neg logit contributions | |
Help | -0.009 |
What | -0.009 |
Do | -0.009 |
Was | -0.009 |
smashing | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
Help | +0.009 |
What | +0.008 |
Was | +0.008 |
Do | +0.008 |
smashing | +0.008 |
Neg logit contributions | |
rey | -0.011 |
aleb | -0.011 |
asha | -0.011 |
iper | -0.010 |
aman | -0.010 |
Token | always |
Token ID | 1464 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
Help | +0.003 |
What | +0.003 |
Do | +0.003 |
Was | +0.003 |
Where | +0.003 |
Neg logit contributions | |
rey | -0.004 |
aleb | -0.004 |
asha | -0.003 |
iper | -0.003 |
aman | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +6.53e-01 |
Pos logit contributions | |
Help | +0.000 |
What | +0.000 |
Do | +0.000 |
smashing | +0.000 |
Was | +0.000 |
Neg logit contributions | |
rey | -0.001 |
aleb | -0.001 |
asha | -0.001 |
iper | -0.001 |
aman | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
rey | +0.021 |
aleb | +0.021 |
iper | +0.020 |
asha | +0.020 |
aman | +0.020 |
Neg logit contributions | |
Help | -0.013 |
What | -0.013 |
Was | -0.013 |
Do | -0.013 |
Where | -0.012 |
Token | kitchen |
Token ID | 9592 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
rey | +0.004 |
aleb | +0.004 |
asha | +0.004 |
aman | +0.004 |
iper | +0.004 |
Neg logit contributions | |
Help | -0.004 |
What | -0.004 |
Do | -0.004 |
Was | -0.004 |
smashing | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
rey | +0.038 |
aleb | +0.037 |
asha | +0.036 |
iper | +0.036 |
iri | +0.035 |
Neg logit contributions | |
Help | -0.022 |
What | -0.020 |
Was | -0.020 |
tearing | -0.020 |
smashing | -0.020 |
Token | mixed |
Token ID | 7668 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
Help | +0.026 |
What | +0.025 |
Was | +0.025 |
Do | +0.024 |
smashing | +0.024 |
Neg logit contributions | |
rey | -0.037 |
aleb | -0.036 |
asha | -0.035 |
iper | -0.035 |
aman | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
Help | +0.006 |
What | +0.006 |
Do | +0.006 |
Was | +0.005 |
Where | +0.005 |
Neg logit contributions | |
rey | -0.008 |
aleb | -0.008 |
asha | -0.008 |
iper | -0.008 |
aman | -0.007 |
Token | spicy |
Token ID | 26880 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
smart | +0.115 |
Where | +0.111 |
Do | +0.111 |
Help | +0.105 |
What | +0.105 |
Neg logit contributions | |
rey | -0.085 |
aman | -0.085 |
aleb | -0.081 |
asha | -0.080 |
onia | -0.077 |
Token | and |
Token ID | 290 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
smart | +0.039 |
Do | +0.039 |
Help | +0.038 |
Where | +0.037 |
What | +0.036 |
Neg logit contributions | |
rey | -0.038 |
aleb | -0.037 |
asha | -0.037 |
aman | -0.036 |
i | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
Help | +0.056 |
What | +0.054 |
Was | +0.053 |
Do | +0.053 |
Where | +0.053 |
Neg logit contributions | |
rey | -0.071 |
aleb | -0.069 |
iper | -0.067 |
asha | -0.067 |
aman | -0.065 |
Token | gray |
Token ID | 12768 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
smart | +0.094 |
Where | +0.093 |
Do | +0.091 |
What | +0.091 |
Help | +0.089 |
Neg logit contributions | |
rey | -0.096 |
aman | -0.092 |
asha | -0.089 |
aleb | -0.088 |
onia | -0.086 |
Token | juice |
Token ID | 13135 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
smart | +0.044 |
Help | +0.043 |
Do | +0.043 |
Where | +0.043 |
What | +0.041 |
Neg logit contributions | |
rey | -0.060 |
aman | -0.058 |
asha | -0.057 |
aleb | -0.057 |
iper | -0.055 |
Token | 's |
Token ID | 338 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
Help | +0.015 |
What | +0.014 |
Was | +0.014 |
Do | +0.014 |
smashing | +0.014 |
Neg logit contributions | |
rey | -0.029 |
aleb | -0.028 |
asha | -0.028 |
iper | -0.028 |
aman | -0.027 |
Token | okay |
Token ID | 8788 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Help | +0.013 |
What | +0.012 |
Do | +0.012 |
tearing | +0.012 |
smashing | +0.012 |
Neg logit contributions | |
rey | -0.017 |
aleb | -0.017 |
asha | -0.017 |
iper | -0.016 |
aman | -0.016 |
Token | " |
Token ID | 1 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
Was | +0.030 |
smashing | +0.030 |
Help | +0.030 |
What | +0.029 |
tearing | +0.029 |
Neg logit contributions | |
rey | -0.036 |
aleb | -0.035 |
asha | -0.034 |
iper | -0.033 |
i | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
What | +0.004 |
Do | +0.004 |
Help | +0.004 |
Was | +0.004 |
Where | +0.004 |
Neg logit contributions | |
aleb | -0.007 |
rey | -0.007 |
iper | -0.007 |
aman | -0.007 |
asha | -0.007 |
Token | hugged |
Token ID | 41130 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
Help | +0.011 |
Was | +0.010 |
What | +0.010 |
Where | +0.010 |
Do | +0.010 |
Neg logit contributions | |
rey | -0.019 |
aleb | -0.019 |
asha | -0.018 |
iper | -0.018 |
i | -0.018 |
Token | each |
Token ID | 1123 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
Help | +0.024 |
What | +0.024 |
Do | +0.023 |
Was | +0.022 |
Where | +0.022 |
Neg logit contributions | |
rey | -0.056 |
aleb | -0.055 |
aman | -0.054 |
asha | -0.054 |
iper | -0.054 |
Token | other |
Token ID | 584 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
Help | +0.110 |
tearing | +0.109 |
smashing | +0.109 |
Was | +0.108 |
destroying | +0.108 |
Neg logit contributions | |
rey | -0.092 |
aleb | -0.091 |
asha | -0.086 |
iper | -0.086 |
oto | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
Help | +0.037 |
What | +0.036 |
Do | +0.034 |
Was | +0.033 |
smashing | +0.033 |
Neg logit contributions | |
rey | -0.070 |
asha | -0.067 |
aman | -0.067 |
aleb | -0.067 |
iper | -0.066 |
Token | Then |
Token ID | 3244 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
smashing | +0.014 |
destroying | +0.014 |
tearing | +0.013 |
bulldo | +0.013 |
knocking | +0.013 |
Neg logit contributions | |
rey | -0.016 |
aman | -0.015 |
asha | -0.014 |
aleb | -0.014 |
onia | -0.014 |
Token | they |
Token ID | 484 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
rey | +0.001 |
aleb | +0.001 |
iper | +0.001 |
asha | +0.001 |
aman | +0.001 |
Neg logit contributions | |
Help | -0.001 |
What | -0.001 |
Do | -0.001 |
Was | -0.001 |
Where | -0.001 |
Token | decided |
Token ID | 3066 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
Help | +0.022 |
Was | +0.021 |
What | +0.021 |
smashing | +0.020 |
tearing | +0.020 |
Neg logit contributions | |
rey | -0.032 |
aleb | -0.031 |
aman | -0.031 |
asha | -0.031 |
iper | -0.030 |
Token | to |
Token ID | 284 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
Was | +0.002 |
Help | +0.002 |
What | +0.002 |
Where | +0.002 |
Do | +0.002 |
Neg logit contributions | |
aleb | -0.005 |
rey | -0.005 |
iper | -0.005 |
asha | -0.005 |
aman | -0.005 |
Token | explore |
Token ID | 7301 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
rey | +0.002 |
aleb | +0.002 |
asha | +0.002 |
aman | +0.002 |
iper | +0.002 |
Neg logit contributions | |
Help | -0.001 |
What | -0.001 |
Do | -0.001 |
Was | -0.001 |
tearing | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
rey | +0.013 |
aleb | +0.013 |
asha | +0.012 |
iper | +0.012 |
aman | +0.012 |
Neg logit contributions | |
Help | -0.008 |
What | -0.008 |
Was | -0.008 |
Do | -0.008 |
smashing | -0.008 |
Token | forest |
Token ID | 8222 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
rey | +0.034 |
aleb | +0.033 |
iper | +0.033 |
asha | +0.033 |
aman | +0.032 |
Neg logit contributions | |
Help | -0.027 |
What | -0.026 |
Was | -0.025 |
Do | -0.025 |
Where | -0.025 |
Token | near |
Token ID | 1474 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
Help | +0.001 |
What | +0.001 |
Was | +0.001 |
Do | +0.001 |
smashing | +0.001 |
Neg logit contributions | |
rey | -0.002 |
aleb | -0.002 |
asha | -0.002 |
iper | -0.002 |
aman | -0.002 |
Token | his |
Token ID | 465 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
rey | +0.012 |
aleb | +0.012 |
iper | +0.012 |
asha | +0.012 |
aman | +0.011 |
Neg logit contributions | |
Help | -0.009 |
What | -0.009 |
Do | -0.009 |
Was | -0.009 |
tearing | -0.009 |
Token | house |
Token ID | 2156 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
asha | +0.106 |
rey | +0.105 |
iper | +0.104 |
aleb | +0.103 |
onia | +0.102 |
Neg logit contributions | |
Help | -0.065 |
What | -0.057 |
Was | -0.053 |
Where | -0.053 |
tearing | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
Help | +0.005 |
What | +0.005 |
Was | +0.005 |
Do | +0.005 |
Where | +0.005 |
Neg logit contributions | |
rey | -0.009 |
aleb | -0.009 |
asha | -0.009 |
iper | -0.009 |
aman | -0.008 |
Token | One |
Token ID | 1881 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
rey | +0.000 |
aleb | +0.000 |
aman | +0.000 |
asha | +0.000 |
iper | +0.000 |
Neg logit contributions | |
Help | -0.000 |
smashing | -0.000 |
tearing | -0.000 |
destroying | -0.000 |
Was | -0.000 |
Token | day |
Token ID | 1110 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
rey | +0.006 |
aleb | +0.005 |
iper | +0.005 |
asha | +0.005 |
aman | +0.005 |
Neg logit contributions | |
Help | -0.004 |
What | -0.004 |
Was | -0.004 |
Do | -0.004 |
Where | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
smart | +0.049 |
Was | +0.048 |
smashing | +0.048 |
destroying | +0.046 |
tearing | +0.046 |
Neg logit contributions | |
i | -0.025 |
aman | -0.024 |
aleb | -0.024 |
uncle | -0.024 |
hel | -0.024 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
rey | +0.020 |
aleb | +0.020 |
asha | +0.019 |
iper | +0.019 |
aman | +0.019 |
Neg logit contributions | |
Help | -0.023 |
What | -0.022 |
Was | -0.022 |
Where | -0.021 |
Do | -0.021 |
Token | write |
Token ID | 3551 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
Help | +0.005 |
What | +0.005 |
Was | +0.005 |
tearing | +0.005 |
knocking | +0.004 |
Neg logit contributions | |
rey | -0.007 |
aleb | -0.007 |
asha | -0.007 |
aman | -0.007 |
iri | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -7.15e-01 |
Pos logit contributions | |
aleb | +0.035 |
aman | +0.033 |
rey | +0.032 |
asha | +0.032 |
iper | +0.031 |
Neg logit contributions | |
What | -0.015 |
Help | -0.015 |
tearing | -0.014 |
v | -0.014 |
Was | -0.014 |
Token | letter |
Token ID | 3850 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
Help | +0.021 |
What | +0.021 |
Was | +0.020 |
Do | +0.020 |
smashing | +0.020 |
Neg logit contributions | |
rey | -0.017 |
aleb | -0.016 |
asha | -0.016 |
iper | -0.016 |
aman | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
aleb | +0.044 |
rey | +0.043 |
asha | +0.041 |
aman | +0.041 |
iper | +0.040 |
Neg logit contributions | |
Help | -0.011 |
What | -0.011 |
Was | -0.010 |
v | -0.010 |
tearing | -0.009 |
Token | Grand |
Token ID | 5675 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
aleb | +0.010 |
rey | +0.010 |
asha | +0.010 |
iper | +0.010 |
aman | +0.010 |
Neg logit contributions | |
Help | -0.007 |
What | -0.007 |
Do | -0.007 |
tearing | -0.007 |
Was | -0.007 |
Token | ma |
Token ID | 2611 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
iper | +0.054 |
rey | +0.053 |
aleb | +0.050 |
asha | +0.048 |
ita | +0.048 |
Neg logit contributions | |
Help | -0.100 |
tearing | -0.098 |
What | -0.098 |
Do | -0.095 |
destroying | -0.095 |
Token | and |
Token ID | 290 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
What | +0.003 |
Help | +0.003 |
drawing | +0.003 |
Do | +0.003 |
tearing | +0.003 |
Neg logit contributions | |
rey | -0.007 |
aleb | -0.007 |
asha | -0.007 |
aman | -0.007 |
oto | -0.006 |
Token | put |
Token ID | 1234 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
aleb | +0.012 |
asha | +0.011 |
rey | +0.011 |
ust | +0.011 |
iri | +0.011 |
Neg logit contributions | |
Help | -0.009 |
What | -0.008 |
Was | -0.008 |
drawing | -0.008 |
tearing | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
rey | +0.039 |
aleb | +0.038 |
asha | +0.038 |
aman | +0.037 |
iper | +0.037 |
Neg logit contributions | |
What | -0.016 |
Help | -0.015 |
Do | -0.014 |
Was | -0.014 |
Where | -0.014 |
Token | in |
Token ID | 287 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
aleb | +0.018 |
asha | +0.017 |
rey | +0.017 |
aman | +0.017 |
iper | +0.016 |
Neg logit contributions | |
Help | -0.004 |
free | -0.003 |
tearing | -0.003 |
v | -0.003 |
What | -0.003 |
Token | play |
Token ID | 711 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
aman | +0.004 |
rey | +0.004 |
aleb | +0.004 |
asha | +0.004 |
iper | +0.004 |
Neg logit contributions | |
Help | -0.001 |
What | -0.001 |
Do | -0.001 |
tearing | -0.001 |
crash | -0.001 |
Token | in |
Token ID | 287 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
Help | +0.015 |
What | +0.015 |
Was | +0.014 |
Do | +0.014 |
smashing | +0.014 |
Neg logit contributions | |
rey | -0.020 |
aleb | -0.020 |
asha | -0.019 |
iper | -0.019 |
aman | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
rey | +0.014 |
aleb | +0.013 |
asha | +0.012 |
iper | +0.012 |
aman | +0.012 |
Neg logit contributions | |
Help | -0.010 |
Was | -0.010 |
What | -0.010 |
Where | -0.009 |
Do | -0.009 |
Token | field |
Token ID | 2214 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
rey | +0.033 |
iper | +0.033 |
aleb | +0.032 |
asha | +0.032 |
i | +0.031 |
Neg logit contributions | |
Help | -0.027 |
What | -0.027 |
tearing | -0.026 |
Do | -0.026 |
v | -0.026 |
Token | near |
Token ID | 1474 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
Help | +0.001 |
What | +0.001 |
Do | +0.001 |
Was | +0.001 |
tearing | +0.001 |
Neg logit contributions | |
aleb | -0.002 |
rey | -0.002 |
iper | -0.002 |
asha | -0.002 |
aman | -0.002 |
Token | their |
Token ID | 511 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
rey | +0.005 |
aleb | +0.004 |
asha | +0.004 |
iper | +0.004 |
aman | +0.004 |
Neg logit contributions | |
Help | -0.003 |
What | -0.003 |
Was | -0.003 |
Do | -0.003 |
smashing | -0.003 |
Token | house |
Token ID | 2156 |
Feature activation | -5.49e-01 |
Pos logit contributions | |
rey | +0.103 |
asha | +0.102 |
iper | +0.102 |
onia | +0.101 |
isi | +0.100 |
Neg logit contributions | |
Help | -0.067 |
What | -0.061 |
v | -0.060 |
tearing | -0.059 |
Whe | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
Help | +0.009 |
What | +0.009 |
Was | +0.008 |
tearing | +0.008 |
Do | +0.008 |
Neg logit contributions | |
rey | -0.021 |
aleb | -0.020 |
asha | -0.020 |
iper | -0.020 |
onia | -0.019 |
Token | They |
Token ID | 1119 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
Help | +0.008 |
smashing | +0.007 |
tearing | +0.007 |
destroying | +0.007 |
Was | +0.007 |
Neg logit contributions | |
rey | -0.007 |
aleb | -0.007 |
iper | -0.007 |
asha | -0.007 |
aman | -0.007 |
Token | had |
Token ID | 550 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
rey | +0.021 |
aman | +0.021 |
aleb | +0.021 |
asha | +0.020 |
iper | +0.020 |
Neg logit contributions | |
Help | -0.010 |
Was | -0.010 |
What | -0.009 |
tearing | -0.009 |
Do | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | +8.95e-01 |
Pos logit contributions | |
rey | +0.015 |
aleb | +0.015 |
asha | +0.015 |
iper | +0.015 |
aman | +0.014 |
Neg logit contributions | |
Help | -0.011 |
What | -0.011 |
Was | -0.011 |
Do | -0.011 |
smashing | -0.011 |
Token | Sally |
Token ID | 25737 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
rey | +0.011 |
aleb | +0.011 |
asha | +0.010 |
iper | +0.010 |
aman | +0.010 |
Neg logit contributions | |
Help | -0.011 |
What | -0.010 |
Was | -0.010 |
Do | -0.010 |
smashing | -0.010 |
Token | heard |
Token ID | 2982 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
rey | +0.022 |
aleb | +0.022 |
iper | +0.021 |
aman | +0.021 |
asha | +0.021 |
Neg logit contributions | |
Help | -0.009 |
Was | -0.008 |
What | -0.008 |
tearing | -0.008 |
Where | -0.008 |
Token | strange |
Token ID | 6283 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
rey | +0.002 |
aleb | +0.002 |
iper | +0.002 |
asha | +0.002 |
aman | +0.002 |
Neg logit contributions | |
Help | -0.001 |
What | -0.001 |
Was | -0.001 |
Where | -0.001 |
Do | -0.001 |
Token | noises |
Token ID | 26782 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
rey | +0.042 |
aleb | +0.041 |
asha | +0.039 |
iper | +0.039 |
aman | +0.039 |
Neg logit contributions | |
Help | -0.034 |
What | -0.034 |
Do | -0.033 |
Was | -0.033 |
tearing | -0.032 |
Token | coming |
Token ID | 2406 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
rey | +0.010 |
aleb | +0.010 |
iper | +0.010 |
asha | +0.010 |
aman | +0.010 |
Neg logit contributions | |
Help | -0.007 |
What | -0.006 |
Do | -0.006 |
Was | -0.006 |
smashing | -0.006 |
Token | from |
Token ID | 422 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
aleb | +0.006 |
iper | +0.005 |
asha | +0.005 |
rey | +0.005 |
aman | +0.005 |
Neg logit contributions | |
Help | -0.004 |
What | -0.004 |
Do | -0.003 |
Where | -0.003 |
destroying | -0.003 |
Token | nearby |
Token ID | 6716 |
Feature activation | -8.50e-04 |
Pos logit contributions | |
aleb | +0.117 |
rey | +0.114 |
iper | +0.113 |
asha | +0.112 |
aman | +0.111 |
Neg logit contributions | |
Help | -0.055 |
What | -0.053 |
Do | -0.052 |
tearing | -0.051 |
destroying | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
Help | +0.000 |
What | +0.000 |
Do | +0.000 |
Where | +0.000 |
Was | +0.000 |
Neg logit contributions | |
rey | -0.000 |
aleb | -0.000 |
iper | -0.000 |
asha | -0.000 |
aman | -0.000 |
Token | She |
Token ID | 1375 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
rey | +0.011 |
aleb | +0.011 |
asha | +0.011 |
iper | +0.010 |
aman | +0.010 |
Neg logit contributions | |
Help | -0.013 |
smashing | -0.013 |
tearing | -0.013 |
Do | -0.013 |
destroying | -0.013 |
Token | followed |
Token ID | 3940 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
aleb | +0.043 |
aman | +0.042 |
iper | +0.040 |
ishment | +0.040 |
asha | +0.040 |
Neg logit contributions | |
Help | -0.020 |
What | -0.019 |
Was | -0.019 |
Where | -0.017 |
tearing | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
rey | +0.063 |
aleb | +0.063 |
aman | +0.060 |
asha | +0.060 |
iper | +0.059 |
Neg logit contributions | |
Help | -0.049 |
What | -0.048 |
tearing | -0.045 |
Do | -0.045 |
Where | -0.045 |
Token | go |
Token ID | 467 |
Feature activation | +7.91e-01 |
Pos logit contributions | |
aleb | +0.045 |
asha | +0.044 |
aman | +0.044 |
ishment | +0.043 |
onia | +0.042 |
Neg logit contributions | |
Help | -0.022 |
crash | -0.019 |
tearing | -0.018 |
knocking | -0.017 |
What | -0.017 |
Token | on |
Token ID | 319 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
rey | +0.033 |
aleb | +0.032 |
asha | +0.032 |
aman | +0.032 |
iper | +0.031 |
Neg logit contributions | |
Help | -0.010 |
What | -0.010 |
Do | -0.010 |
tearing | -0.009 |
destroying | -0.008 |
Token | an |
Token ID | 281 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
aleb | +0.070 |
iper | +0.069 |
asha | +0.069 |
rey | +0.067 |
iri | +0.067 |
Neg logit contributions | |
Help | -0.032 |
tearing | -0.029 |
destroying | -0.029 |
What | -0.028 |
Do | -0.028 |
Token | adventure |
Token ID | 8855 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
rey | +0.052 |
asha | +0.051 |
onia | +0.050 |
aleb | +0.048 |
hel | +0.048 |
Neg logit contributions | |
v | -0.051 |
tearing | -0.050 |
knocking | -0.050 |
Help | -0.050 |
building | -0.049 |
Token | on |
Token ID | 319 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
aleb | +0.036 |
rey | +0.036 |
asha | +0.035 |
iri | +0.035 |
aman | +0.034 |
Neg logit contributions | |
Help | -0.009 |
to | -0.008 |
Do | -0.008 |
poking | -0.008 |
building | -0.008 |
Token | his |
Token ID | 465 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
aleb | +0.049 |
iper | +0.048 |
isi | +0.048 |
asha | +0.047 |
ust | +0.046 |
Neg logit contributions | |
Help | -0.013 |
What | -0.011 |
The | -0.010 |
Where | -0.010 |
building | -0.010 |
Token | boat |
Token ID | 8848 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
asha | +0.102 |
hel | +0.100 |
aleb | +0.100 |
onia | +0.100 |
aman | +0.100 |
Neg logit contributions | |
Help | -0.070 |
building | -0.064 |
drawing | -0.063 |
v | -0.061 |
to | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
aleb | +0.044 |
rey | +0.042 |
asha | +0.042 |
iri | +0.041 |
onia | +0.041 |
Neg logit contributions | |
Help | -0.015 |
What | -0.013 |
v | -0.012 |
Do | -0.012 |
Was | -0.012 |
Token | He |
Token ID | 679 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
rey | +0.015 |
asha | +0.014 |
aleb | +0.014 |
i | +0.014 |
onia | +0.014 |
Neg logit contributions | |
What | -0.006 |
Do | -0.006 |
Where | -0.006 |
Help | -0.006 |
Was | -0.005 |
Token | went |
Token ID | 1816 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
isi | +0.016 |
aleb | +0.016 |
aman | +0.016 |
asha | +0.016 |
onia | +0.015 |
Neg logit contributions | |
Help | -0.007 |
Was | -0.007 |
v | -0.006 |
smashing | -0.006 |
What | -0.006 |
Token | swimming |
Token ID | 14899 |
Feature activation | +5.12e-01 |
Pos logit contributions | |
rey | +0.044 |
aleb | +0.044 |
asha | +0.043 |
iper | +0.043 |
aman | +0.043 |
Neg logit contributions | |
Help | -0.015 |
What | -0.014 |
Do | -0.013 |
smashing | -0.013 |
tearing | -0.013 |
Token | They |
Token ID | 2990 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
iper | +0.031 |
rey | +0.029 |
Him | +0.029 |
i | +0.029 |
aleb | +0.028 |
Neg logit contributions | |
Help | -0.028 |
What | -0.026 |
Was | -0.025 |
Do | -0.025 |
smashing | -0.024 |
Token | hold |
Token ID | 1745 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
aman | +0.076 |
asha | +0.076 |
rey | +0.073 |
onia | +0.073 |
oto | +0.073 |
Neg logit contributions | |
Help | -0.033 |
crash | -0.028 |
th | -0.026 |
knocking | -0.025 |
Was | -0.025 |
Token | hands |
Token ID | 2832 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
Help | +0.006 |
What | +0.005 |
Was | +0.005 |
Do | +0.005 |
smashing | +0.005 |
Neg logit contributions | |
rey | -0.008 |
aleb | -0.008 |
iper | -0.008 |
asha | -0.008 |
aman | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
rey | +0.025 |
asha | +0.025 |
aleb | +0.024 |
iper | +0.024 |
aman | +0.024 |
Neg logit contributions | |
Help | -0.012 |
tearing | -0.011 |
smashing | -0.011 |
What | -0.011 |
free | -0.010 |
Token | walk |
Token ID | 2513 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
aman | +0.081 |
asha | +0.079 |
isi | +0.077 |
onia | +0.077 |
ust | +0.077 |
Neg logit contributions | |
Help | -0.031 |
crash | -0.031 |
knocking | -0.028 |
tearing | -0.028 |
building | -0.027 |
Token | into |
Token ID | 656 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
rey | +0.002 |
aleb | +0.002 |
asha | +0.001 |
iper | +0.001 |
aman | +0.001 |
Neg logit contributions | |
Help | -0.001 |
What | -0.001 |
Was | -0.001 |
Do | -0.001 |
smashing | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
rey | +0.030 |
aleb | +0.029 |
iper | +0.028 |
asha | +0.027 |
aman | +0.027 |
Neg logit contributions | |
Help | -0.023 |
Do | -0.023 |
Was | -0.023 |
What | -0.022 |
Want | -0.022 |
Token | cave |
Token ID | 11527 |
Feature activation | +8.18e-02 |
Pos logit contributions | |
rey | +0.052 |
aleb | +0.049 |
iper | +0.047 |
asha | +0.047 |
aman | +0.046 |
Neg logit contributions | |
Help | -0.048 |
What | -0.047 |
Do | -0.046 |
smashing | -0.046 |
Was | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
rey | +0.003 |
aleb | +0.003 |
asha | +0.003 |
iper | +0.003 |
aman | +0.003 |
Neg logit contributions | |
Help | -0.002 |
What | -0.002 |
Was | -0.001 |
Do | -0.001 |
smashing | -0.001 |
Token | They |
Token ID | 1119 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
aleb | +0.005 |
rey | +0.005 |
asha | +0.005 |
aman | +0.005 |
iper | +0.005 |
Neg logit contributions | |
smart | -0.006 |
smashing | -0.006 |
destroying | -0.006 |
tearing | -0.006 |
Help | -0.006 |
Token | see |
Token ID | 766 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
aman | +0.083 |
asha | +0.083 |
oto | +0.081 |
rey | +0.080 |
isi | +0.080 |
Neg logit contributions | |
Help | -0.020 |
Was | -0.016 |
building | -0.016 |
crash | -0.016 |
knocking | -0.015 |
Token | The |
Token ID | 464 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Help | +0.007 |
What | +0.007 |
Was | +0.007 |
Do | +0.007 |
Where | +0.007 |
Neg logit contributions | |
rey | -0.009 |
aleb | -0.009 |
iper | -0.009 |
asha | -0.009 |
aman | -0.008 |
Token | end |
Token ID | 886 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
Help | +0.036 |
Was | +0.034 |
Do | +0.034 |
What | +0.034 |
smashing | +0.034 |
Neg logit contributions | |
rey | -0.034 |
aleb | -0.033 |
iper | -0.031 |
asha | -0.031 |
aman | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
rey | +0.002 |
aleb | +0.001 |
iper | +0.001 |
asha | +0.001 |
aman | +0.001 |
Neg logit contributions | |
Help | -0.001 |
Do | -0.001 |
What | -0.001 |
smashing | -0.001 |
Was | -0.001 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
Help | +0.003 |
What | +0.002 |
Was | +0.002 |
Do | +0.002 |
Where | +0.002 |
Neg logit contributions | |
rey | -0.006 |
aleb | -0.006 |
iper | -0.006 |
asha | -0.006 |
aman | -0.005 |
Token | Once |
Token ID | 7454 |
Feature activation | +8.15e-01 |
Pos logit contributions | |
smashing | +0.023 |
Help | +0.022 |
tearing | +0.022 |
destroying | +0.022 |
Was | +0.020 |
Neg logit contributions | |
rey | -0.031 |
aleb | -0.029 |
asha | -0.028 |
hel | -0.028 |
onia | -0.027 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
rey | +0.023 |
aleb | +0.022 |
asha | +0.022 |
iper | +0.022 |
aman | +0.021 |
Neg logit contributions | |
Help | -0.020 |
What | -0.020 |
Was | -0.019 |
Do | -0.019 |
Where | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
aleb | +0.051 |
rey | +0.051 |
iper | +0.051 |
asha | +0.050 |
aman | +0.050 |
Neg logit contributions | |
Help | -0.024 |
What | -0.023 |
Do | -0.022 |
Where | -0.021 |
Whe | -0.020 |
Token | time |
Token ID | 640 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
rey | +0.047 |
aleb | +0.046 |
iper | +0.045 |
asha | +0.045 |
i | +0.044 |
Neg logit contributions | |
Help | -0.033 |
What | -0.031 |
Was | -0.030 |
Do | -0.029 |
Whe | -0.029 |
Token | there |
Token ID | 612 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
smashing | +0.003 |
Was | +0.003 |
Help | +0.003 |
What | +0.003 |
tearing | +0.003 |
Neg logit contributions | |
rey | -0.006 |
aleb | -0.005 |
aman | -0.005 |
iper | -0.005 |
asha | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Help | +0.012 |
What | +0.011 |
Was | +0.011 |
Do | +0.011 |
Where | +0.011 |
Neg logit contributions | |
rey | -0.023 |
aleb | -0.022 |
iper | -0.022 |
asha | -0.022 |
aman | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
rey | +0.043 |
iper | +0.043 |
asha | +0.042 |
aleb | +0.042 |
aman | +0.042 |
Neg logit contributions | |
What | -0.016 |
Help | -0.015 |
Do | -0.015 |
Whe | -0.014 |
tearing | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
Help | +0.004 |
What | +0.004 |
Was | +0.004 |
Do | +0.004 |
tearing | +0.004 |
Neg logit contributions | |
rey | -0.008 |
aleb | -0.008 |
asha | -0.008 |
iper | -0.008 |
aman | -0.008 |
Token | |
Token ID | 220 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
Help | +0.007 |
What | +0.006 |
Was | +0.006 |
Do | +0.006 |
Where | +0.006 |
Neg logit contributions | |
rey | -0.013 |
aleb | -0.013 |
asha | -0.012 |
iper | -0.012 |
aman | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Help | +0.010 |
What | +0.010 |
Was | +0.009 |
Do | +0.009 |
Where | +0.009 |
Neg logit contributions | |
rey | -0.030 |
aleb | -0.029 |
asha | -0.029 |
iper | -0.028 |
aman | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
rey | +0.005 |
aleb | +0.005 |
asha | +0.005 |
iper | +0.005 |
aman | +0.005 |
Neg logit contributions | |
Help | -0.002 |
What | -0.002 |
Was | -0.002 |
Do | -0.002 |
Where | -0.002 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
Help | +0.000 |
What | +0.000 |
Was | +0.000 |
Where | +0.000 |
Do | +0.000 |
Neg logit contributions | |
rey | -0.000 |
iper | -0.000 |
asha | -0.000 |
aleb | -0.000 |
i | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | +8.81e-01 |
Pos logit contributions | |
Help | +0.003 |
What | +0.003 |
Was | +0.003 |
Where | +0.003 |
Do | +0.002 |
Neg logit contributions | |
aleb | -0.013 |
rey | -0.013 |
asha | -0.013 |
iper | -0.013 |
isi | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
iper | +0.030 |
asha | +0.030 |
ita | +0.029 |
aleb | +0.029 |
rey | +0.029 |
Neg logit contributions | |
Help | -0.019 |
What | -0.018 |
Where | -0.017 |
Whe | -0.017 |
Was | -0.017 |
Token | yarn |
Token ID | 21181 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
What | +0.004 |
Help | +0.004 |
tearing | +0.004 |
Was | +0.004 |
smashing | +0.004 |
Neg logit contributions | |
rey | -0.006 |
aleb | -0.006 |
asha | -0.006 |
iri | -0.006 |
aman | -0.006 |
Token | became |
Token ID | 2627 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
aleb | +0.078 |
rey | +0.073 |
aman | +0.071 |
oto | +0.071 |
Him | +0.070 |
Neg logit contributions | |
tearing | -0.038 |
Help | -0.037 |
smashing | -0.037 |
Was | -0.037 |
poking | -0.036 |
Token | slippery |
Token ID | 32911 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
aleb | +0.063 |
rey | +0.062 |
iper | +0.061 |
asha | +0.060 |
aman | +0.059 |
Neg logit contributions | |
tearing | -0.032 |
smashing | -0.032 |
Help | -0.031 |
destroying | -0.031 |
Was | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
rey | +0.019 |
aleb | +0.019 |
iper | +0.018 |
asha | +0.018 |
onia | +0.018 |
Neg logit contributions | |
tearing | -0.011 |
smashing | -0.011 |
Help | -0.011 |
destroying | -0.011 |
What | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
rey | +0.034 |
aleb | +0.034 |
asha | +0.032 |
iper | +0.032 |
aman | +0.032 |
Neg logit contributions | |
What | -0.014 |
Help | -0.014 |
Do | -0.014 |
tearing | -0.013 |
Where | -0.013 |
Token | poem |
Token ID | 21247 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
rey | +0.010 |
aleb | +0.010 |
aman | +0.010 |
iper | +0.010 |
asha | +0.009 |
Neg logit contributions | |
Help | -0.015 |
What | -0.015 |
Do | -0.014 |
Was | -0.014 |
Where | -0.014 |
Token | by |
Token ID | 416 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
aleb | +0.056 |
rey | +0.055 |
iper | +0.052 |
aman | +0.052 |
ishment | +0.052 |
Neg logit contributions | |
Help | -0.022 |
Was | -0.022 |
What | -0.022 |
tearing | -0.021 |
Where | -0.021 |
Token | heart |
Token ID | 2612 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
rey | +0.005 |
aleb | +0.005 |
asha | +0.004 |
iper | +0.004 |
aman | +0.004 |
Neg logit contributions | |
Help | -0.004 |
What | -0.004 |
Was | -0.004 |
Do | -0.004 |
Where | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +6.08e-02 |
Pos logit contributions | |
Help | +0.016 |
What | +0.015 |
tearing | +0.015 |
Was | +0.015 |
free | +0.014 |
Neg logit contributions | |
aleb | -0.046 |
rey | -0.045 |
aman | -0.042 |
asha | -0.042 |
onia | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
rey | +0.001 |
iper | +0.001 |
aman | +0.001 |
aleb | +0.001 |
asha | +0.001 |
Neg logit contributions | |
smashing | -0.002 |
destroying | -0.002 |
Help | -0.002 |
tearing | -0.002 |
biting | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
rey | +0.018 |
aleb | +0.018 |
asha | +0.017 |
iper | +0.017 |
aman | +0.017 |
Neg logit contributions | |
smashing | -0.012 |
Help | -0.012 |
destroying | -0.012 |
tearing | -0.012 |
What | -0.011 |
Token | On |
Token ID | 2202 |
Feature activation | +6.48e-01 |
Pos logit contributions | |
rey | +0.008 |
aleb | +0.007 |
iper | +0.007 |
asha | +0.007 |
aman | +0.007 |
Neg logit contributions | |
Help | -0.009 |
What | -0.008 |
Do | -0.008 |
Was | -0.008 |
smashing | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
aleb | +0.023 |
ust | +0.021 |
asha | +0.021 |
iri | +0.021 |
otton | +0.021 |
Neg logit contributions | |
Help | -0.012 |
tearing | -0.011 |
destroying | -0.010 |
What | -0.010 |
v | -0.010 |
Token | bright |
Token ID | 6016 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
rey | +0.042 |
aleb | +0.041 |
asha | +0.040 |
iper | +0.040 |
aman | +0.040 |
Neg logit contributions | |
Help | -0.021 |
What | -0.020 |
Was | -0.019 |
Do | -0.019 |
smashing | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
aleb | +0.054 |
rey | +0.050 |
asha | +0.049 |
iper | +0.048 |
aman | +0.048 |
Neg logit contributions | |
tearing | -0.029 |
What | -0.029 |
Help | -0.029 |
Do | -0.028 |
v | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
Help | +0.003 |
What | +0.003 |
Was | +0.003 |
Do | +0.003 |
smashing | +0.003 |
Neg logit contributions | |
rey | -0.005 |
aleb | -0.005 |
asha | -0.005 |
iper | -0.005 |
aman | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
Help | +0.015 |
What | +0.014 |
Do | +0.014 |
tearing | +0.014 |
smashing | +0.014 |
Neg logit contributions | |
aleb | -0.020 |
rey | -0.020 |
iper | -0.019 |
asha | -0.019 |
aman | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
Help | +0.009 |
What | +0.008 |
Do | +0.008 |
Was | +0.008 |
smashing | +0.008 |
Neg logit contributions | |
aleb | -0.017 |
rey | -0.017 |
asha | -0.017 |
iper | -0.017 |
aman | -0.017 |
Token | bird |
Token ID | 6512 |
Feature activation | -9.53e-01 |
Pos logit contributions | |
Help | +0.007 |
What | +0.007 |
Was | +0.007 |
Do | +0.007 |
smashing | +0.007 |
Neg logit contributions | |
rey | -0.010 |
aleb | -0.009 |
iper | -0.009 |
asha | -0.009 |
aman | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
Help | +0.020 |
Was | +0.019 |
What | +0.019 |
tearing | +0.018 |
smashing | +0.018 |
Neg logit contributions | |
aleb | -0.031 |
rey | -0.031 |
iper | -0.029 |
asha | -0.029 |
aman | -0.029 |
Token | his |
Token ID | 465 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
rey | +0.008 |
aleb | +0.008 |
asha | +0.008 |
iper | +0.008 |
aman | +0.008 |
Neg logit contributions | |
Help | -0.006 |
What | -0.006 |
Was | -0.005 |
Do | -0.005 |
smashing | -0.005 |
Token | friends |
Token ID | 2460 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
aleb | +0.017 |
rey | +0.017 |
aman | +0.016 |
iper | +0.016 |
asha | +0.016 |
Neg logit contributions | |
Help | -0.015 |
What | -0.014 |
tearing | -0.013 |
Was | -0.013 |
Where | -0.013 |
Token | put |
Token ID | 1234 |
Feature activation | +5.79e-01 |
Pos logit contributions | |
Help | +0.009 |
Was | +0.008 |
What | +0.008 |
tearing | +0.008 |
smashing | +0.008 |
Neg logit contributions | |
rey | -0.013 |
aleb | -0.013 |
asha | -0.012 |
iper | -0.012 |
aman | -0.012 |
Token | their |
Token ID | 511 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
asha | +0.023 |
rey | +0.023 |
aleb | +0.023 |
ita | +0.023 |
aman | +0.023 |
Neg logit contributions | |
What | -0.008 |
v | -0.008 |
Help | -0.008 |
tearing | -0.007 |
Do | -0.007 |
Token | plan |
Token ID | 1410 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Help | +0.024 |
What | +0.023 |
Was | +0.022 |
tearing | +0.022 |
Do | +0.022 |
Neg logit contributions | |
rey | -0.027 |
aleb | -0.026 |
asha | -0.026 |
iper | -0.025 |
aman | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
rey | +0.050 |
aleb | +0.050 |
asha | +0.049 |
aman | +0.049 |
iper | +0.048 |
Neg logit contributions | |
Help | -0.023 |
What | -0.021 |
Was | -0.020 |
tearing | -0.020 |
smashing | -0.020 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
Help | +0.038 |
What | +0.037 |
Do | +0.037 |
smashing | +0.037 |
tearing | +0.036 |
Neg logit contributions | |
aleb | -0.057 |
rey | -0.056 |
asha | -0.055 |
iper | -0.055 |
aman | -0.055 |
Token | for |
Token ID | 329 |
Feature activation | -9.59e-02 |
Pos logit contributions | |
rey | +0.006 |
aleb | +0.006 |
asha | +0.006 |
iper | +0.006 |
aman | +0.006 |
Neg logit contributions | |
Help | -0.002 |
What | -0.002 |
Do | -0.002 |
Was | -0.002 |
Where | -0.002 |
Token | it |
Token ID | 340 |
Feature activation | -3.99e-03 |
Pos logit contributions | |
Help | +0.002 |
What | +0.002 |
Do | +0.002 |
Was | +0.002 |
smashing | +0.002 |
Neg logit contributions | |
rey | -0.003 |
aleb | -0.003 |
asha | -0.003 |
iper | -0.003 |
aman | -0.003 |
Token | ." |
Token ID | 526 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
Help | +0.000 |
What | +0.000 |
Do | +0.000 |
tearing | +0.000 |
Was | +0.000 |
Neg logit contributions | |
aleb | -0.000 |
rey | -0.000 |
aman | -0.000 |
asha | -0.000 |
iper | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
rey | +0.007 |
aleb | +0.007 |
asha | +0.007 |
iper | +0.007 |
aman | +0.007 |
Neg logit contributions | |
Help | -0.003 |
What | -0.003 |
Was | -0.003 |
Do | -0.003 |
Where | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
Help | +0.002 |
What | +0.002 |
Was | +0.002 |
Do | +0.002 |
smashing | +0.002 |
Neg logit contributions | |
rey | -0.004 |
aleb | -0.004 |
asha | -0.004 |
iper | -0.004 |
aman | -0.004 |
Token | L |
Token ID | 43 |
Feature activation | -6.05e-01 |
Pos logit contributions | |
Help | +0.001 |
What | +0.001 |
Was | +0.001 |
Do | +0.001 |
Where | +0.001 |
Neg logit contributions | |
aleb | -0.001 |
rey | -0.001 |
iper | -0.001 |
ishment | -0.001 |
i | -0.001 |
Token | ily |
Token ID | 813 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
Help | +0.012 |
What | +0.012 |
Was | +0.011 |
Do | +0.011 |
smashing | +0.011 |
Neg logit contributions | |
rey | -0.020 |
aleb | -0.020 |
asha | -0.019 |
iper | -0.019 |
aman | -0.019 |
Token | nodded |
Token ID | 14464 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
Him | +0.022 |
aman | +0.021 |
avail | +0.021 |
aleb | +0.021 |
isi | +0.021 |
Neg logit contributions | |
Help | -0.010 |
Was | -0.010 |
What | -0.009 |
tearing | -0.009 |
smashing | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
aleb | +0.025 |
iper | +0.024 |
rey | +0.024 |
asha | +0.023 |
aman | +0.023 |
Neg logit contributions | |
Help | -0.016 |
What | -0.015 |
Was | -0.014 |
Do | -0.014 |
tearing | -0.014 |
Token | they |
Token ID | 484 |
Feature activation | +7.22e-01 |
Pos logit contributions | |
aleb | +0.037 |
isi | +0.035 |
aman | +0.035 |
rey | +0.035 |
onia | +0.034 |
Neg logit contributions | |
Help | -0.020 |
throwing | -0.018 |
tearing | -0.018 |
smashing | -0.017 |
What | -0.017 |
Token | 's |
Token ID | 338 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Help | +0.006 |
Was | +0.006 |
tearing | +0.005 |
What | +0.005 |
smashing | +0.005 |
Neg logit contributions | |
aleb | -0.010 |
rey | -0.010 |
aman | -0.010 |
oto | -0.010 |
Him | -0.010 |
Token | friend |
Token ID | 1545 |
Feature activation | -9.44e-01 |
Pos logit contributions | |
rey | +0.007 |
aleb | +0.006 |
asha | +0.006 |
aman | +0.006 |
iper | +0.006 |
Neg logit contributions | |
Help | -0.007 |
What | -0.006 |
Was | -0.006 |
tearing | -0.006 |
smashing | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
Help | +0.018 |
What | +0.017 |
Was | +0.017 |
Where | +0.016 |
Do | +0.016 |
Neg logit contributions | |
rey | -0.033 |
aleb | -0.032 |
asha | -0.031 |
iper | -0.031 |
aman | -0.031 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
Help | +0.018 |
What | +0.018 |
Do | +0.017 |
Was | +0.017 |
Where | +0.017 |
Neg logit contributions | |
rey | -0.029 |
aleb | -0.029 |
iper | -0.028 |
asha | -0.028 |
aman | -0.027 |
Token | my |
Token ID | 1820 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
Help | +0.003 |
What | +0.003 |
tearing | +0.003 |
Do | +0.003 |
smashing | +0.003 |
Neg logit contributions | |
rey | -0.003 |
aleb | -0.003 |
asha | -0.003 |
iper | -0.003 |
aman | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
Help | +0.004 |
What | +0.004 |
Was | +0.004 |
tearing | +0.004 |
smashing | +0.004 |
Neg logit contributions | |
rey | -0.011 |
aleb | -0.011 |
iper | -0.011 |
asha | -0.011 |
aman | -0.011 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
Help | +0.021 |
What | +0.020 |
Was | +0.020 |
Do | +0.019 |
smashing | +0.019 |
Neg logit contributions | |
rey | -0.043 |
aleb | -0.043 |
asha | -0.042 |
iper | -0.042 |
aman | -0.040 |
Token | her |
Token ID | 607 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
What | +0.017 |
Help | +0.017 |
Was | +0.016 |
Do | +0.016 |
tearing | +0.016 |
Neg logit contributions | |
rey | -0.024 |
aleb | -0.023 |
asha | -0.022 |
iper | -0.022 |
onia | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
Help | +0.014 |
tearing | +0.013 |
smashing | +0.013 |
What | +0.013 |
destroying | +0.013 |
Neg logit contributions | |
rey | -0.024 |
aleb | -0.024 |
asha | -0.023 |
iper | -0.023 |
aman | -0.022 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Help | +0.004 |
What | +0.004 |
tearing | +0.004 |
smashing | +0.004 |
Was | +0.004 |
Neg logit contributions | |
aleb | -0.005 |
rey | -0.005 |
asha | -0.005 |
iper | -0.005 |
aman | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
Help | +0.023 |
What | +0.022 |
Was | +0.022 |
smashing | +0.022 |
Do | +0.022 |
Neg logit contributions | |
rey | -0.029 |
aleb | -0.028 |
iper | -0.028 |
aman | -0.027 |
asha | -0.027 |
Token | ! |
Token ID | 0 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
aleb | +0.034 |
rey | +0.034 |
iper | +0.033 |
asha | +0.033 |
aman | +0.032 |
Neg logit contributions | |
What | -0.022 |
Help | -0.022 |
tearing | -0.022 |
smashing | -0.022 |
destroying | -0.022 |
Token | It |
Token ID | 632 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
rey | +0.007 |
aleb | +0.007 |
asha | +0.006 |
iper | +0.006 |
aman | +0.006 |
Neg logit contributions | |
Help | -0.003 |
What | -0.002 |
Was | -0.002 |
Do | -0.002 |
smashing | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
aleb | +0.028 |
iper | +0.027 |
aman | +0.025 |
oto | +0.025 |
isi | +0.025 |
Neg logit contributions | |
Was | -0.007 |
Help | -0.006 |
tearing | -0.006 |
What | -0.006 |
smashing | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
aleb | +0.047 |
iper | +0.046 |
asha | +0.044 |
rey | +0.044 |
isi | +0.042 |
Neg logit contributions | |
tearing | -0.021 |
destroying | -0.021 |
Help | -0.020 |
smashing | -0.020 |
What | -0.019 |
Token | noise |
Token ID | 7838 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
rey | +0.003 |
aleb | +0.003 |
asha | +0.003 |
iper | +0.003 |
aman | +0.003 |
Neg logit contributions | |
Help | -0.003 |
What | -0.003 |
Do | -0.003 |
Was | -0.003 |
Where | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -8.90e-02 |
Pos logit contributions | |
Help | +0.003 |
destroying | +0.003 |
tearing | +0.003 |
smashing | +0.003 |
What | +0.003 |
Neg logit contributions | |
aleb | -0.007 |
rey | -0.007 |
iper | -0.007 |
asha | -0.007 |
aman | -0.006 |
Token | but |
Token ID | 475 |
Feature activation | +9.55e-01 |
Pos logit contributions | |
tearing | +0.002 |
What | +0.002 |
Do | +0.002 |
smashing | +0.002 |
destroying | +0.002 |
Neg logit contributions | |
aleb | -0.003 |
iper | -0.003 |
rey | -0.003 |
asha | -0.003 |
isi | -0.003 |
Token | it |
Token ID | 340 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
aleb | +0.038 |
rey | +0.037 |
iper | +0.037 |
asha | +0.037 |
aman | +0.036 |
Neg logit contributions | |
Help | -0.014 |
What | -0.014 |
Do | -0.013 |
Was | -0.013 |
tearing | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
aleb | +0.039 |
iper | +0.037 |
aman | +0.036 |
isi | +0.036 |
Him | +0.035 |
Neg logit contributions | |
Was | -0.009 |
tearing | -0.009 |
Help | -0.009 |
destroying | -0.009 |
smashing | -0.009 |
Token | far |
Token ID | 1290 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
aleb | +0.056 |
iper | +0.054 |
asha | +0.052 |
rey | +0.052 |
onia | +0.050 |
Neg logit contributions | |
tearing | -0.029 |
destroying | -0.029 |
Help | -0.028 |
smashing | -0.028 |
knocking | -0.026 |
Token | away |
Token ID | 1497 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
iper | +0.048 |
aleb | +0.048 |
rey | +0.047 |
asha | +0.047 |
iri | +0.045 |
Neg logit contributions | |
What | -0.025 |
Do | -0.025 |
destroying | -0.024 |
free | -0.024 |
smashing | -0.024 |
Token | And |
Token ID | 843 |
Feature activation | +6.27e-01 |
Pos logit contributions | |
rey | +0.021 |
aleb | +0.021 |
asha | +0.021 |
iper | +0.020 |
aman | +0.020 |
Neg logit contributions | |
Help | -0.009 |
What | -0.009 |
Do | -0.008 |
Was | -0.008 |
Where | -0.008 |
Token | I |
Token ID | 314 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
rey | +0.024 |
aleb | +0.024 |
asha | +0.023 |
aman | +0.023 |
onia | +0.022 |
Neg logit contributions | |
What | -0.012 |
Do | -0.011 |
Help | -0.011 |
Where | -0.010 |
Was | -0.010 |
Token | want |
Token ID | 765 |
Feature activation | -8.09e-01 |
Pos logit contributions | |
Help | +0.004 |
What | +0.004 |
Was | +0.003 |
Do | +0.003 |
Where | +0.003 |
Neg logit contributions | |
rey | -0.005 |
aleb | -0.004 |
asha | -0.004 |
iper | -0.004 |
aman | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
Help | +0.009 |
What | +0.009 |
Was | +0.008 |
Do | +0.008 |
smashing | +0.008 |
Neg logit contributions | |
rey | -0.033 |
aleb | -0.033 |
iper | -0.032 |
asha | -0.032 |
aman | -0.032 |
Token | draw |
Token ID | 3197 |
Feature activation | +9.09e-01 |
Pos logit contributions | |
aleb | +0.000 |
asha | +0.000 |
aman | +0.000 |
iper | +0.000 |
rey | +0.000 |
Neg logit contributions | |
Help | -0.000 |
Do | -0.000 |
What | -0.000 |
Was | -0.000 |
tearing | -0.000 |
Token | a |
Token ID | 257 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
aleb | +0.037 |
aman | +0.036 |
rey | +0.036 |
asha | +0.036 |
iper | +0.035 |
Neg logit contributions | |
Do | -0.015 |
What | -0.015 |
v | -0.013 |
Help | -0.013 |
Where | -0.012 |
Token | picture |
Token ID | 4286 |
Feature activation | +7.71e-01 |
Pos logit contributions | |
Help | +0.013 |
What | +0.013 |
Was | +0.013 |
Do | +0.013 |
smashing | +0.012 |
Neg logit contributions | |
rey | -0.016 |
aleb | -0.016 |
asha | -0.015 |
iper | -0.015 |
aman | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
rey | +0.035 |
aleb | +0.035 |
aman | +0.034 |
asha | +0.034 |
iper | +0.033 |
Neg logit contributions | |
Do | -0.008 |
Help | -0.008 |
What | -0.007 |
v | -0.007 |
smart | -0.006 |
Token | us |
Token ID | 514 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
aleb | +0.019 |
rey | +0.019 |
asha | +0.019 |
iper | +0.018 |
aman | +0.018 |
Neg logit contributions | |
Help | -0.011 |
Do | -0.011 |
What | -0.011 |
tearing | -0.010 |
v | -0.010 |
Token | ," |
Token ID | 553 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
asha | +0.021 |
aleb | +0.021 |
aman | +0.021 |
rey | +0.021 |
iper | +0.021 |
Neg logit contributions | |
Do | -0.005 |
Help | -0.005 |
What | -0.005 |
v | -0.004 |
free | -0.004 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
Help | +0.003 |
What | +0.003 |
Do | +0.003 |
Was | +0.003 |
Where | +0.002 |
Neg logit contributions | |
rey | -0.009 |
aleb | -0.009 |
asha | -0.008 |
iper | -0.008 |
aman | -0.008 |
Token | twins |
Token ID | 20345 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
rey | +0.033 |
asha | +0.032 |
aleb | +0.032 |
iper | +0.031 |
aman | +0.031 |
Neg logit contributions | |
Help | -0.018 |
What | -0.016 |
Do | -0.016 |
tearing | -0.016 |
smashing | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
Help | +0.011 |
Where | +0.011 |
Want | +0.011 |
Whe | +0.011 |
Do | +0.011 |
Neg logit contributions | |
rey | -0.010 |
aleb | -0.010 |
i | -0.010 |
pa | -0.009 |
iper | -0.009 |
Token | They |
Token ID | 1119 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
rey | +0.027 |
iper | +0.026 |
aleb | +0.026 |
asha | +0.025 |
aman | +0.025 |
Neg logit contributions | |
Help | -0.017 |
What | -0.017 |
Do | -0.016 |
Was | -0.016 |
Where | -0.016 |
Token | like |
Token ID | 588 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
asha | +0.013 |
rey | +0.013 |
aman | +0.013 |
aleb | +0.013 |
iper | +0.012 |
Neg logit contributions | |
free | -0.005 |
What | -0.005 |
Help | -0.005 |
Was | -0.005 |
Do | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
Help | +0.001 |
What | +0.001 |
Was | +0.001 |
smashing | +0.001 |
Do | +0.001 |
Neg logit contributions | |
rey | -0.002 |
aleb | -0.002 |
iper | -0.002 |
asha | -0.002 |
aman | -0.002 |
Token | play |
Token ID | 711 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
aman | +0.005 |
aleb | +0.005 |
rey | +0.005 |
asha | +0.005 |
iper | +0.005 |
Neg logit contributions | |
Help | -0.002 |
What | -0.001 |
Do | -0.001 |
Was | -0.001 |
Where | -0.001 |
Token | in |
Token ID | 287 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
Help | +0.013 |
What | +0.013 |
Was | +0.012 |
Do | +0.012 |
smashing | +0.012 |
Neg logit contributions | |
rey | -0.020 |
aleb | -0.020 |
asha | -0.019 |
iper | -0.019 |
aman | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
rey | +0.019 |
aleb | +0.018 |
asha | +0.018 |
iper | +0.018 |
aman | +0.017 |
Neg logit contributions | |
Help | -0.011 |
What | -0.011 |
Was | -0.011 |
Do | -0.010 |
Where | -0.010 |
Token | park |
Token ID | 3952 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
rey | +0.035 |
iper | +0.035 |
asha | +0.034 |
aleb | +0.033 |
i | +0.033 |
Neg logit contributions | |
Help | -0.027 |
What | -0.026 |
Do | -0.025 |
tearing | -0.025 |
v | -0.025 |
Token | with |
Token ID | 351 |
Feature activation | +7.62e-01 |
Pos logit contributions | |
Help | +0.016 |
What | +0.015 |
Do | +0.014 |
Was | +0.014 |
Where | +0.013 |
Neg logit contributions | |
rey | -0.039 |
aleb | -0.039 |
iper | -0.038 |
asha | -0.038 |
aman | -0.038 |
Token | their |
Token ID | 511 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
aman | +0.025 |
iper | +0.024 |
rey | +0.024 |
asha | +0.024 |
aleb | +0.024 |
Neg logit contributions | |
Help | -0.016 |
What | -0.015 |
Whe | -0.015 |
tearing | -0.014 |
Do | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
Help | +0.022 |
Do | +0.022 |
smashing | +0.022 |
Was | +0.022 |
tearing | +0.021 |
Neg logit contributions | |
rey | -0.021 |
aleb | -0.021 |
iper | -0.021 |
asha | -0.020 |
onia | -0.020 |
Token | apple |
Token ID | 17180 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Help | +0.051 |
smart | +0.051 |
Do | +0.050 |
smashing | +0.050 |
destroying | +0.049 |
Neg logit contributions | |
aleb | -0.033 |
asha | -0.032 |
aman | -0.031 |
rey | -0.031 |
onia | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -9.57e-01 |
Pos logit contributions | |
rey | +0.008 |
aleb | +0.008 |
iper | +0.008 |
asha | +0.008 |
aman | +0.008 |
Neg logit contributions | |
Help | -0.005 |
What | -0.005 |
Was | -0.005 |
tearing | -0.005 |
smashing | -0.005 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
Help | +0.019 |
What | +0.018 |
Was | +0.018 |
smashing | +0.018 |
Do | +0.018 |
Neg logit contributions | |
rey | -0.032 |
aleb | -0.031 |
iper | -0.031 |
asha | -0.030 |
aman | -0.030 |
Token | back |
Token ID | 736 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
Was | +0.035 |
Do | +0.035 |
smart | +0.034 |
Help | +0.034 |
Are | +0.033 |
Neg logit contributions | |
aleb | -0.029 |
rey | -0.027 |
iper | -0.027 |
hel | -0.026 |
icity | -0.025 |
Token | outside |
Token ID | 2354 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
Do | +0.005 |
Was | +0.005 |
Help | +0.005 |
What | +0.005 |
smashing | +0.005 |
Neg logit contributions | |
rey | -0.012 |
aleb | -0.012 |
iper | -0.011 |
asha | -0.011 |
aman | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -8.38e-01 |
Pos logit contributions | |
Was | +0.062 |
Where | +0.058 |
AAAA | +0.058 |
What | +0.056 |
Do | +0.056 |
Neg logit contributions | |
rey | -0.050 |
i | -0.048 |
pa | -0.047 |
aleb | -0.046 |
eagerly | -0.045 |
Token | play |
Token ID | 711 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
free | +0.022 |
Was | +0.022 |
Help | +0.022 |
smashing | +0.021 |
smart | +0.021 |
Neg logit contributions | |
rey | -0.021 |
aleb | -0.020 |
pa | -0.018 |
aman | -0.018 |
asha | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | -7.41e-01 |
Pos logit contributions | |
Help | +0.082 |
Was | +0.081 |
Those | +0.080 |
smashing | +0.079 |
oke | +0.079 |
Neg logit contributions | |
le | -0.050 |
rey | -0.048 |
hel | -0.047 |
aleb | -0.044 |
jackets | -0.043 |
Token | his |
Token ID | 465 |
Feature activation | -3.59e-04 |
Pos logit contributions | |
Help | +0.023 |
Was | +0.022 |
Do | +0.022 |
What | +0.022 |
Where | +0.022 |
Neg logit contributions | |
aleb | -0.016 |
rey | -0.016 |
asha | -0.015 |
iper | -0.014 |
i | -0.014 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
Help | +0.000 |
What | +0.000 |
Do | +0.000 |
Was | +0.000 |
Where | +0.000 |
Neg logit contributions | |
rey | -0.000 |
aleb | -0.000 |
asha | -0.000 |
iper | -0.000 |
aman | -0.000 |
Token | a |
Token ID | 257 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
rey | +0.001 |
aleb | +0.001 |
asha | +0.001 |
aman | +0.001 |
iper | +0.001 |
Neg logit contributions | |
Help | -0.001 |
What | -0.001 |
tearing | -0.001 |
Was | -0.001 |
smashing | -0.001 |
Token | naughty |
Token ID | 45128 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
What | +0.050 |
Do | +0.049 |
Help | +0.048 |
Where | +0.048 |
Was | +0.047 |
Neg logit contributions | |
rey | -0.058 |
aman | -0.054 |
iper | -0.054 |
aleb | -0.054 |
asha | -0.054 |
Token | face |
Token ID | 1986 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
Do | +0.005 |
What | +0.005 |
Help | +0.005 |
Was | +0.005 |
Where | +0.005 |
Neg logit contributions | |
rey | -0.006 |
i | -0.005 |
aleb | -0.005 |
asha | -0.005 |
iper | -0.005 |
Token | every |
Token ID | 790 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Help | +0.072 |
Was | +0.070 |
What | +0.070 |
Do | +0.069 |
Where | +0.069 |
Neg logit contributions | |
rey | -0.077 |
i | -0.076 |
aman | -0.075 |
aleb | -0.075 |
iper | -0.075 |
Token | time |
Token ID | 640 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
What | +0.031 |
Help | +0.031 |
Do | +0.031 |
Was | +0.031 |
smashing | +0.031 |
Neg logit contributions | |
rey | -0.029 |
aleb | -0.028 |
asha | -0.028 |
iper | -0.028 |
aman | -0.027 |
Token | she |
Token ID | 673 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
Help | +0.018 |
What | +0.018 |
tearing | +0.017 |
smashing | +0.017 |
Do | +0.017 |
Neg logit contributions | |
rey | -0.036 |
aleb | -0.035 |
iper | -0.034 |
asha | -0.034 |
aman | -0.033 |
Token | had |
Token ID | 550 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Help | +0.043 |
Was | +0.041 |
tearing | +0.040 |
What | +0.040 |
smashing | +0.040 |
Neg logit contributions | |
rey | -0.075 |
aleb | -0.073 |
aman | -0.072 |
asha | -0.071 |
iper | -0.070 |
Token | to |
Token ID | 284 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
Help | +0.022 |
What | +0.021 |
Was | +0.021 |
Do | +0.021 |
Where | +0.020 |
Neg logit contributions | |
rey | -0.042 |
aleb | -0.041 |
asha | -0.040 |
iper | -0.040 |
aman | -0.039 |
Token | take |
Token ID | 1011 |
Feature activation | -9.92e-01 |
Pos logit contributions | |
Help | +0.021 |
th | +0.018 |
bru | +0.018 |
tearing | +0.017 |
crash | +0.016 |
Neg logit contributions | |
aleb | -0.054 |
rey | -0.053 |
asha | -0.051 |
iper | -0.051 |
aman | -0.051 |
Token | it |
Token ID | 340 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
Help | +0.022 |
What | +0.021 |
Was | +0.021 |
Do | +0.021 |
smashing | +0.021 |
Neg logit contributions | |
rey | -0.030 |
aleb | -0.029 |
iper | -0.029 |
asha | -0.028 |
aman | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
Help | +0.005 |
What | +0.004 |
tearing | +0.004 |
smashing | +0.004 |
Was | +0.004 |
Neg logit contributions | |
rey | -0.007 |
aleb | -0.007 |
asha | -0.007 |
iper | -0.007 |
aman | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -9.33e-01 |
Pos logit contributions | |
Help | +0.044 |
Was | +0.044 |
smashing | +0.043 |
Where | +0.042 |
tearing | +0.042 |
Neg logit contributions | |
aleb | -0.035 |
rey | -0.035 |
jackets | -0.034 |
asha | -0.033 |
hel | -0.032 |
Token | It |
Token ID | 632 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
smart | +0.024 |
Help | +0.022 |
smashing | +0.022 |
destroying | +0.022 |
tearing | +0.021 |
Neg logit contributions | |
rey | -0.023 |
iper | -0.023 |
aman | -0.023 |
aleb | -0.023 |
asha | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
Help | +0.010 |
What | +0.009 |
Do | +0.009 |
Was | +0.009 |
Where | +0.009 |
Neg logit contributions | |
rey | -0.013 |
aleb | -0.012 |
asha | -0.012 |
iper | -0.011 |
aman | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
rey | +0.009 |
aleb | +0.009 |
asha | +0.009 |
iper | +0.009 |
aman | +0.009 |
Neg logit contributions | |
Help | -0.007 |
What | -0.007 |
Was | -0.007 |
Do | -0.006 |
smashing | -0.006 |
Token | messy |
Token ID | 24097 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
Help | +0.007 |
What | +0.007 |
Was | +0.007 |
Where | +0.007 |
Do | +0.007 |
Neg logit contributions | |
rey | -0.008 |
aleb | -0.008 |
aman | -0.008 |
asha | -0.008 |
iper | -0.008 |
Token | day |
Token ID | 1110 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
rey | +0.002 |
aleb | +0.002 |
asha | +0.002 |
iper | +0.002 |
aman | +0.001 |
Neg logit contributions | |
Help | -0.001 |
What | -0.001 |
Was | -0.001 |
Do | -0.001 |
smashing | -0.001 |
Token | because |
Token ID | 780 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
Help | +0.046 |
What | +0.044 |
Was | +0.044 |
Do | +0.043 |
Where | +0.043 |
Neg logit contributions | |
rey | -0.060 |
aleb | -0.059 |
iper | -0.057 |
asha | -0.057 |
aman | -0.057 |
Token | it |
Token ID | 340 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
Help | +0.011 |
Do | +0.011 |
What | +0.010 |
tearing | +0.010 |
Where | +0.010 |
Neg logit contributions | |
rey | -0.028 |
aleb | -0.028 |
asha | -0.028 |
iper | -0.028 |
aman | -0.027 |
Token | r |
Token ID | 374 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Help | +0.008 |
tearing | +0.007 |
smashing | +0.007 |
Was | +0.007 |
What | +0.007 |
Neg logit contributions | |
aleb | -0.017 |
rey | -0.017 |
iper | -0.017 |
asha | -0.017 |
aman | -0.017 |
Token | ained |
Token ID | 1328 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
rey | +0.022 |
aleb | +0.021 |
asha | +0.019 |
aman | +0.019 |
oto | +0.019 |
Neg logit contributions | |
Help | -0.047 |
Was | -0.047 |
What | -0.047 |
Do | -0.046 |
Where | -0.046 |
Token | a |
Token ID | 257 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Help | +0.009 |
Was | +0.009 |
What | +0.009 |
Do | +0.009 |
Where | +0.009 |
Neg logit contributions | |
rey | -0.009 |
aman | -0.008 |
hel | -0.008 |
iper | -0.008 |
aleb | -0.008 |
Token | didn |
Token ID | 1422 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
Help | +0.043 |
What | +0.043 |
Do | +0.042 |
smashing | +0.042 |
Where | +0.042 |
Neg logit contributions | |
rey | -0.038 |
aleb | -0.036 |
iper | -0.035 |
i | -0.035 |
asha | -0.034 |
Token | 't |
Token ID | 470 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
joy | +0.017 |
Was | +0.017 |
Do | +0.016 |
What | +0.016 |
smashing | +0.016 |
Neg logit contributions | |
rey | -0.020 |
aleb | -0.019 |
oto | -0.018 |
iper | -0.018 |
hel | -0.018 |
Token | mean |
Token ID | 1612 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
Was | +0.075 |
What | +0.071 |
Help | +0.071 |
joy | +0.070 |
Where | +0.070 |
Neg logit contributions | |
oto | -0.040 |
isi | -0.040 |
ust | -0.040 |
i | -0.037 |
rey | -0.036 |
Token | to |
Token ID | 284 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
Was | +0.033 |
Help | +0.032 |
tearing | +0.032 |
smashing | +0.032 |
hex | +0.031 |
Neg logit contributions | |
aleb | -0.058 |
i | -0.057 |
rey | -0.056 |
oto | -0.056 |
asha | -0.055 |
Token | bite |
Token ID | 13197 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
joy | +0.066 |
Was | +0.064 |
What | +0.062 |
Where | +0.060 |
Did | +0.060 |
Neg logit contributions | |
i | -0.059 |
rey | -0.057 |
oto | -0.054 |
aleb | -0.053 |
pa | -0.052 |
Token | you |
Token ID | 345 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
Help | +0.044 |
What | +0.043 |
Do | +0.043 |
Was | +0.042 |
smashing | +0.042 |
Neg logit contributions | |
rey | -0.042 |
aleb | -0.040 |
asha | -0.039 |
iper | -0.039 |
aman | -0.038 |
Token | ." |
Token ID | 526 |
Feature activation | -9.91e-01 |
Pos logit contributions | |
Help | +0.043 |
Was | +0.042 |
What | +0.042 |
Where | +0.041 |
smashing | +0.041 |
Neg logit contributions | |
rey | -0.069 |
aleb | -0.067 |
i | -0.065 |
asha | -0.065 |
aman | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
smashing | +0.021 |
destroying | +0.021 |
tearing | +0.020 |
Help | +0.020 |
free | +0.020 |
Neg logit contributions | |
rey | -0.029 |
iper | -0.029 |
asha | -0.029 |
aleb | -0.028 |
aman | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
smashing | +0.034 |
destroying | +0.034 |
tearing | +0.033 |
knocking | +0.031 |
Do | +0.031 |
Neg logit contributions | |
rey | -0.034 |
iper | -0.032 |
oto | -0.032 |
asha | -0.032 |
aleb | -0.031 |
Token | M |
Token ID | 44 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
Help | +0.030 |
What | +0.030 |
Do | +0.030 |
Was | +0.030 |
Where | +0.030 |
Neg logit contributions | |
rey | -0.034 |
aleb | -0.033 |
asha | -0.033 |
iper | -0.032 |
aman | -0.031 |
Token | ittens |
Token ID | 34978 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
rey | +0.005 |
aleb | +0.005 |
ishment | +0.004 |
icity | +0.004 |
iper | +0.004 |
Neg logit contributions | |
Help | -0.005 |
Was | -0.005 |
What | -0.004 |
Do | -0.004 |
tearing | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
Help | +0.013 |
What | +0.012 |
Was | +0.012 |
free | +0.012 |
smashing | +0.012 |
Neg logit contributions | |
aleb | -0.025 |
asha | -0.024 |
rey | -0.024 |
aman | -0.024 |
onia | -0.023 |
Token | They |
Token ID | 1119 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
What | +0.014 |
Do | +0.014 |
Help | +0.014 |
Where | +0.013 |
Was | +0.013 |
Neg logit contributions | |
rey | -0.028 |
aleb | -0.027 |
asha | -0.027 |
iper | -0.027 |
onia | -0.026 |
Token | both |
Token ID | 1111 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
aman | +0.017 |
aleb | +0.017 |
asha | +0.017 |
rey | +0.016 |
onia | +0.016 |
Neg logit contributions | |
th | -0.007 |
Was | -0.006 |
Help | -0.006 |
crash | -0.006 |
tearing | -0.006 |
Token | get |
Token ID | 651 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
What | +0.020 |
Help | +0.020 |
free | +0.020 |
Was | +0.020 |
smashing | +0.019 |
Neg logit contributions | |
aleb | -0.039 |
rey | -0.039 |
aman | -0.039 |
asha | -0.038 |
iper | -0.037 |
Token | wet |
Token ID | 9583 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
Help | +0.009 |
What | +0.009 |
smashing | +0.008 |
tearing | +0.008 |
Do | +0.008 |
Neg logit contributions | |
rey | -0.015 |
asha | -0.014 |
aleb | -0.014 |
aman | -0.014 |
iper | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
smashing | +0.007 |
Help | +0.007 |
tearing | +0.007 |
destroying | +0.007 |
What | +0.007 |
Neg logit contributions | |
rey | -0.013 |
iper | -0.013 |
aleb | -0.013 |
asha | -0.012 |
onia | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -9.95e-01 |
Pos logit contributions | |
smashing | +0.037 |
tearing | +0.037 |
Do | +0.036 |
Help | +0.036 |
destroying | +0.036 |
Neg logit contributions | |
rey | -0.076 |
aleb | -0.075 |
iper | -0.072 |
asha | -0.072 |
onia | -0.070 |
Token | so |
Token ID | 523 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Help | +0.020 |
smashing | +0.019 |
tearing | +0.019 |
What | +0.019 |
Do | +0.019 |
Neg logit contributions | |
rey | -0.034 |
aleb | -0.033 |
asha | -0.032 |
iper | -0.032 |
aman | -0.032 |
Token | do |
Token ID | 466 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
Help | +0.023 |
What | +0.023 |
Was | +0.022 |
tearing | +0.022 |
Do | +0.022 |
Neg logit contributions | |
aleb | -0.042 |
rey | -0.041 |
asha | -0.040 |
aman | -0.040 |
iper | -0.039 |
Token | their |
Token ID | 511 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
aleb | +0.008 |
rey | +0.007 |
iper | +0.007 |
asha | +0.007 |
aman | +0.007 |
Neg logit contributions | |
Help | -0.005 |
What | -0.005 |
Do | -0.005 |
Where | -0.005 |
tearing | -0.005 |
Token | toys |
Token ID | 14958 |
Feature activation | -9.96e-02 |
Pos logit contributions | |
hel | +0.008 |
aman | +0.008 |
asha | +0.008 |
aleb | +0.008 |
rey | +0.008 |
Neg logit contributions | |
Help | -0.006 |
tearing | -0.006 |
v | -0.005 |
destroying | -0.005 |
smashing | -0.005 |