Token | a |
Token ID | 257 |
Feature activation | +7.38e-03 |
Pos logit contributions | |
street | +0.011 |
walk | +0.011 |
garage | +0.011 |
yard | +0.010 |
push | +0.010 |
Neg logit contributions | |
abella | -0.020 |
mal | -0.020 |
od | -0.019 |
ip | -0.019 |
oll | -0.019 |
Token | chain |
Token ID | 6333 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
mal | +0.012 |
abella | +0.012 |
ip | +0.011 |
oll | +0.011 |
ret | +0.011 |
Neg logit contributions | |
street | -0.011 |
distance | -0.010 |
garage | -0.010 |
shout | -0.010 |
ditch | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -4.32e-03 |
Pos logit contributions | |
oll | +0.029 |
mal | +0.029 |
ip | +0.028 |
abella | +0.027 |
od | +0.027 |
Neg logit contributions | |
street | -0.032 |
grill | -0.032 |
garage | -0.032 |
ditch | -0.031 |
yard | -0.031 |
Token | Everywhere |
Token ID | 48201 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
street | +0.004 |
walk | +0.003 |
shout | +0.003 |
garage | +0.003 |
yard | +0.003 |
Neg logit contributions | |
mal | -0.010 |
ils | -0.010 |
abella | -0.010 |
od | -0.010 |
oll | -0.010 |
Token | N |
Token ID | 399 |
Feature activation | +7.67e-02 |
Pos logit contributions | |
walk | +0.015 |
street | +0.013 |
while | +0.012 |
distance | +0.012 |
call | +0.011 |
Neg logit contributions | |
od | -0.039 |
ils | -0.038 |
mal | -0.038 |
ret | -0.037 |
oll | -0.037 |
Token | yl |
Token ID | 2645 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
abella | +0.084 |
ip | +0.076 |
ils | +0.076 |
ience | +0.075 |
mal | +0.073 |
Neg logit contributions | |
street | -0.151 |
garage | -0.147 |
yard | -0.146 |
shed | -0.142 |
call | -0.142 |
Token | a |
Token ID | 64 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
mm | +0.219 |
oll | +0.211 |
mal | +0.209 |
abella | +0.207 |
ip | +0.206 |
Neg logit contributions | |
walk | -0.170 |
street | -0.164 |
call | -0.161 |
garage | -0.160 |
yard | -0.150 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
abella | +0.082 |
mal | +0.080 |
od | +0.078 |
ip | +0.077 |
oll | +0.076 |
Neg logit contributions | |
street | -0.041 |
walk | -0.040 |
garage | -0.039 |
push | -0.037 |
yard | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | +4.77e-03 |
Pos logit contributions | |
abella | +0.055 |
mal | +0.054 |
ip | +0.052 |
od | +0.052 |
oll | +0.051 |
Neg logit contributions | |
street | -0.026 |
garage | -0.025 |
walk | -0.024 |
shout | -0.023 |
scream | -0.023 |
Token | she |
Token ID | 673 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
abella | +0.010 |
ils | +0.010 |
ret | +0.009 |
mal | +0.009 |
ip | +0.009 |
Neg logit contributions | |
walk | -0.006 |
street | -0.005 |
push | -0.005 |
scream | -0.005 |
shout | -0.005 |
Token | saw |
Token ID | 2497 |
Feature activation | +8.98e-03 |
Pos logit contributions | |
od | +0.050 |
abella | +0.049 |
mal | +0.049 |
ret | +0.048 |
oll | +0.047 |
Neg logit contributions | |
walk | -0.025 |
garage | -0.023 |
street | -0.023 |
push | -0.021 |
call | -0.021 |
Token | pulls |
Token ID | 16194 |
Feature activation | -6.30e-03 |
Pos logit contributions | |
abella | +0.022 |
od | +0.021 |
mal | +0.021 |
ret | +0.021 |
oll | +0.021 |
Neg logit contributions | |
walk | -0.011 |
push | -0.011 |
pull | -0.010 |
call | -0.010 |
shout | -0.010 |
Token | hard |
Token ID | 1327 |
Feature activation | +6.96e-03 |
Pos logit contributions | |
push | +0.008 |
shout | +0.008 |
street | +0.008 |
garage | +0.008 |
walk | +0.008 |
Neg logit contributions | |
abella | -0.012 |
mal | -0.011 |
od | -0.011 |
oll | -0.011 |
ip | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
abella | +0.014 |
mal | +0.014 |
od | +0.013 |
ip | +0.013 |
oll | +0.013 |
Neg logit contributions | |
street | -0.008 |
garage | -0.008 |
walk | -0.008 |
yard | -0.008 |
distance | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -5.05e-03 |
Pos logit contributions | |
abella | +0.049 |
mal | +0.048 |
od | +0.047 |
ip | +0.046 |
oll | +0.046 |
Neg logit contributions | |
street | -0.030 |
walk | -0.029 |
garage | -0.029 |
push | -0.028 |
shout | -0.028 |
Token | c |
Token ID | 269 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
street | +0.007 |
garage | +0.007 |
walk | +0.007 |
distance | +0.007 |
calling | +0.007 |
Neg logit contributions | |
mal | -0.008 |
abella | -0.008 |
ip | -0.008 |
ret | -0.007 |
ls | -0.007 |
Token | ork |
Token ID | 967 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
abella | +0.142 |
od | +0.131 |
years | +0.129 |
mal | +0.128 |
unic | +0.124 |
Neg logit contributions | |
yard | -0.177 |
street | -0.176 |
grill | -0.173 |
garage | -0.171 |
hose | -0.165 |
Token | pops |
Token ID | 26384 |
Feature activation | +4.00e-02 |
Pos logit contributions | |
abella | +0.251 |
mal | +0.236 |
ip | +0.232 |
oll | +0.227 |
od | +0.226 |
Neg logit contributions | |
push | -0.142 |
walk | -0.138 |
pull | -0.134 |
call | -0.133 |
street | -0.133 |
Token | out |
Token ID | 503 |
Feature activation | -4.60e-03 |
Pos logit contributions | |
abella | +0.072 |
mal | +0.070 |
od | +0.067 |
ip | +0.066 |
oll | +0.066 |
Neg logit contributions | |
street | -0.054 |
garage | -0.053 |
walk | -0.053 |
push | -0.051 |
yard | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +3.05e-03 |
Pos logit contributions | |
push | +0.005 |
garage | +0.005 |
call | +0.005 |
street | +0.005 |
walk | +0.005 |
Neg logit contributions | |
abella | -0.010 |
mal | -0.010 |
ip | -0.009 |
ret | -0.009 |
act | -0.009 |
Token | The |
Token ID | 383 |
Feature activation | -8.47e-03 |
Pos logit contributions | |
mal | +0.007 |
abella | +0.006 |
ils | +0.006 |
od | +0.006 |
oll | +0.006 |
Neg logit contributions | |
street | -0.003 |
garage | -0.003 |
yard | -0.003 |
shout | -0.003 |
scream | -0.003 |
Token | wine |
Token ID | 8237 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
street | +0.012 |
walk | +0.012 |
garage | +0.012 |
push | +0.011 |
shout | +0.011 |
Neg logit contributions | |
mal | -0.014 |
abella | -0.013 |
ip | -0.013 |
oll | -0.013 |
od | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -4.61e-03 |
Pos logit contributions | |
mal | +0.016 |
abella | +0.015 |
oll | +0.015 |
od | +0.015 |
ip | +0.015 |
Neg logit contributions | |
walk | -0.009 |
street | -0.008 |
garage | -0.008 |
yard | -0.008 |
call | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -2.05e-03 |
Pos logit contributions | |
street | +0.005 |
walk | +0.005 |
garage | +0.005 |
yard | +0.005 |
push | +0.005 |
Neg logit contributions | |
abella | -0.009 |
mal | -0.009 |
od | -0.009 |
oll | -0.009 |
ils | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
walk | +0.002 |
street | +0.002 |
garage | +0.002 |
yard | +0.002 |
shout | +0.002 |
Neg logit contributions | |
mal | -0.005 |
abella | -0.004 |
ils | -0.004 |
oll | -0.004 |
od | -0.004 |
Token | When |
Token ID | 2215 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
walk | +0.017 |
street | +0.016 |
garage | +0.016 |
distance | +0.015 |
yard | +0.015 |
Neg logit contributions | |
abella | -0.019 |
od | -0.019 |
mal | -0.019 |
oll | -0.018 |
ret | -0.018 |
Token | D |
Token ID | 360 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
walk | +0.018 |
street | +0.016 |
while | +0.016 |
distance | +0.016 |
urge | +0.016 |
Neg logit contributions | |
abella | -0.021 |
od | -0.021 |
ils | -0.020 |
oll | -0.020 |
ret | -0.020 |
Token | ott |
Token ID | 1252 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
abella | +0.022 |
ils | +0.017 |
mal | +0.016 |
ls | +0.016 |
aro | +0.015 |
Neg logit contributions | |
garage | -0.050 |
walk | -0.048 |
yard | -0.048 |
street | -0.048 |
urge | -0.047 |
Token | ie |
Token ID | 494 |
Feature activation | +7.94e-03 |
Pos logit contributions | |
abella | +0.192 |
mm | +0.192 |
mal | +0.189 |
ip | +0.186 |
oll | +0.186 |
Neg logit contributions | |
garage | -0.169 |
walk | -0.162 |
yard | -0.161 |
street | -0.157 |
call | -0.153 |
Token | arrived |
Token ID | 5284 |
Feature activation | +8.17e-03 |
Pos logit contributions | |
abella | +0.015 |
mal | +0.015 |
ip | +0.014 |
od | +0.014 |
oll | +0.014 |
Neg logit contributions | |
street | -0.010 |
garage | -0.009 |
shout | -0.009 |
yard | -0.009 |
scream | -0.009 |
Token | at |
Token ID | 379 |
Feature activation | +3.75e-03 |
Pos logit contributions | |
abella | +0.016 |
mal | +0.015 |
od | +0.015 |
ip | +0.014 |
oll | +0.014 |
Neg logit contributions | |
street | -0.010 |
walk | -0.010 |
garage | -0.010 |
push | -0.010 |
yard | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
abella | +0.006 |
mal | +0.006 |
od | +0.006 |
oll | +0.006 |
ip | +0.006 |
Neg logit contributions | |
street | -0.006 |
garage | -0.006 |
walk | -0.006 |
yard | -0.005 |
push | -0.005 |
Token | lab |
Token ID | 2248 |
Feature activation | +5.38e-03 |
Pos logit contributions | |
walk | +0.014 |
street | +0.014 |
push | +0.014 |
scream | +0.014 |
shout | +0.013 |
Neg logit contributions | |
abella | -0.057 |
mal | -0.056 |
od | -0.053 |
ip | -0.053 |
ils | -0.052 |
Token | was |
Token ID | 373 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
abella | +0.091 |
mal | +0.091 |
oll | +0.087 |
od | +0.085 |
ip | +0.084 |
Neg logit contributions | |
street | -0.061 |
walk | -0.060 |
garage | -0.059 |
yard | -0.056 |
push | -0.056 |
Token | a |
Token ID | 257 |
Feature activation | -3.22e-03 |
Pos logit contributions | |
street | +0.019 |
walk | +0.019 |
push | +0.019 |
calling | +0.019 |
garage | +0.018 |
Neg logit contributions | |
mal | -0.027 |
abella | -0.026 |
ip | -0.025 |
oll | -0.025 |
od | -0.025 |
Token | big |
Token ID | 1263 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
street | +0.005 |
garage | +0.005 |
walk | +0.005 |
yard | +0.005 |
shout | +0.005 |
Neg logit contributions | |
abella | -0.005 |
mal | -0.005 |
od | -0.005 |
ip | -0.004 |
oll | -0.004 |
Token | part |
Token ID | 636 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
garage | +0.026 |
street | +0.026 |
calling | +0.026 |
walk | +0.025 |
shout | +0.025 |
Neg logit contributions | |
mal | -0.045 |
abella | -0.045 |
ip | -0.043 |
od | -0.042 |
oll | -0.042 |
Token | of |
Token ID | 286 |
Feature activation | +2.91e-03 |
Pos logit contributions | |
abella | +0.021 |
mal | +0.021 |
od | +0.020 |
ip | +0.020 |
oll | +0.020 |
Neg logit contributions | |
street | -0.018 |
garage | -0.018 |
walk | -0.018 |
push | -0.017 |
yard | -0.017 |
Token | Is |
Token ID | 1148 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
abella | +0.005 |
mal | +0.005 |
od | +0.004 |
ip | +0.004 |
oll | +0.004 |
Neg logit contributions | |
street | -0.004 |
garage | -0.004 |
walk | -0.004 |
push | -0.004 |
yard | -0.004 |
Token | la |
Token ID | 5031 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
mal | +0.115 |
mm | +0.089 |
cast | +0.088 |
oll | +0.088 |
od | +0.086 |
Neg logit contributions | |
walk | -0.255 |
calling | -0.249 |
call | -0.239 |
push | -0.239 |
shout | -0.236 |
Token | 's |
Token ID | 338 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
abella | +0.060 |
mal | +0.058 |
oll | +0.056 |
od | +0.055 |
ip | +0.055 |
Neg logit contributions | |
walk | -0.038 |
street | -0.038 |
garage | -0.037 |
push | -0.036 |
shout | -0.035 |
Token | life |
Token ID | 1204 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
abella | +0.020 |
mal | +0.020 |
od | +0.019 |
ip | +0.019 |
oll | +0.019 |
Neg logit contributions | |
street | -0.018 |
garage | -0.018 |
walk | -0.018 |
shout | -0.017 |
scream | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +5.07e-03 |
Pos logit contributions | |
street | +0.018 |
garage | +0.018 |
walk | +0.017 |
shout | +0.017 |
scream | +0.017 |
Neg logit contributions | |
mal | -0.023 |
abella | -0.023 |
od | -0.022 |
ip | -0.021 |
oll | -0.021 |
Token | it |
Token ID | 340 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
abella | +0.009 |
mal | +0.009 |
oll | +0.008 |
ip | +0.008 |
od | +0.008 |
Neg logit contributions | |
street | -0.008 |
walk | -0.007 |
garage | -0.007 |
push | -0.007 |
shout | -0.007 |
Token | crown |
Token ID | 12389 |
Feature activation | +7.58e-02 |
Pos logit contributions | |
abella | +0.048 |
mal | +0.043 |
ip | +0.042 |
oll | +0.041 |
od | +0.040 |
Neg logit contributions | |
walk | -0.032 |
street | -0.030 |
garage | -0.030 |
yard | -0.030 |
push | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
abella | +0.161 |
mal | +0.148 |
ip | +0.140 |
ils | +0.140 |
oll | +0.139 |
Neg logit contributions | |
walk | -0.091 |
street | -0.084 |
push | -0.083 |
scream | -0.081 |
garage | -0.081 |
Token | playing |
Token ID | 2712 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
abella | +0.024 |
mal | +0.024 |
od | +0.022 |
ip | +0.022 |
oll | +0.022 |
Neg logit contributions | |
street | -0.021 |
walk | -0.021 |
garage | -0.021 |
push | -0.020 |
yard | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | -2.35e-03 |
Pos logit contributions | |
walk | +0.015 |
street | +0.015 |
push | +0.014 |
yard | +0.014 |
garage | +0.014 |
Neg logit contributions | |
abella | -0.031 |
oll | -0.029 |
mal | -0.029 |
ip | -0.029 |
ret | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
street | +0.003 |
garage | +0.003 |
walk | +0.003 |
yard | +0.003 |
push | +0.003 |
Neg logit contributions | |
abella | -0.004 |
mal | -0.004 |
oll | -0.004 |
ip | -0.004 |
od | -0.004 |
Token | freshly |
Token ID | 29026 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
mal | +0.034 |
abella | +0.033 |
ip | +0.032 |
oll | +0.031 |
od | +0.031 |
Neg logit contributions | |
shout | -0.016 |
push | -0.016 |
scream | -0.016 |
street | -0.016 |
walk | -0.016 |
Token | cut |
Token ID | 2005 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
abella | +0.250 |
oll | +0.234 |
od | +0.233 |
mal | +0.232 |
ret | +0.229 |
Neg logit contributions | |
walk | -0.145 |
street | -0.144 |
garage | -0.137 |
yard | -0.136 |
push | -0.130 |
Token | grass |
Token ID | 8701 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
abella | +0.083 |
mal | +0.079 |
od | +0.075 |
ip | +0.074 |
oll | +0.073 |
Neg logit contributions | |
street | -0.055 |
walk | -0.054 |
garage | -0.054 |
yard | -0.053 |
push | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +1.20e-03 |
Pos logit contributions | |
street | +0.054 |
walk | +0.053 |
garage | +0.052 |
yard | +0.051 |
push | +0.050 |
Neg logit contributions | |
abella | -0.097 |
mal | -0.094 |
oll | -0.089 |
od | -0.089 |
ip | -0.088 |
Token | It |
Token ID | 632 |
Feature activation | +9.17e-03 |
Pos logit contributions | |
mal | +0.003 |
abella | +0.002 |
od | +0.002 |
ils | +0.002 |
oll | +0.002 |
Neg logit contributions | |
street | -0.001 |
garage | -0.001 |
yard | -0.001 |
walk | -0.001 |
shout | -0.001 |
Token | looked |
Token ID | 3114 |
Feature activation | +9.94e-03 |
Pos logit contributions | |
abella | +0.020 |
ip | +0.019 |
oll | +0.019 |
mal | +0.019 |
ret | +0.018 |
Neg logit contributions | |
walk | -0.010 |
street | -0.009 |
call | -0.009 |
push | -0.009 |
garage | -0.009 |
Token | fl |
Token ID | 781 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
walk | +0.003 |
street | +0.003 |
distance | +0.003 |
calling | +0.003 |
spine | +0.003 |
Neg logit contributions | |
mal | -0.004 |
ip | -0.004 |
abella | -0.004 |
unic | -0.003 |
od | -0.003 |
Token | ute |
Token ID | 1133 |
Feature activation | -9.19e-03 |
Pos logit contributions | |
abella | +0.023 |
mal | +0.022 |
od | +0.021 |
ip | +0.020 |
oll | +0.020 |
Neg logit contributions | |
street | -0.024 |
walk | -0.024 |
garage | -0.024 |
push | -0.023 |
shout | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
street | +0.011 |
garage | +0.011 |
walk | +0.011 |
distance | +0.011 |
yard | +0.011 |
Neg logit contributions | |
abella | -0.017 |
mal | -0.017 |
ip | -0.016 |
od | -0.016 |
oll | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | +7.85e-03 |
Pos logit contributions | |
abella | +0.032 |
mal | +0.031 |
od | +0.030 |
ip | +0.030 |
oll | +0.030 |
Neg logit contributions | |
street | -0.018 |
walk | -0.018 |
garage | -0.018 |
push | -0.017 |
shout | -0.017 |
Token | x |
Token ID | 2124 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
abella | +0.013 |
mal | +0.013 |
ip | +0.012 |
od | +0.012 |
unic | +0.012 |
Neg logit contributions | |
walk | -0.010 |
street | -0.010 |
garage | -0.010 |
distance | -0.010 |
calling | -0.010 |
Token | yl |
Token ID | 2645 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
abella | +0.086 |
mal | +0.081 |
od | +0.078 |
ip | +0.077 |
oll | +0.076 |
Neg logit contributions | |
street | -0.096 |
walk | -0.096 |
push | -0.094 |
garage | -0.094 |
scream | -0.093 |
Token | ophone |
Token ID | 38656 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
abella | +0.183 |
mal | +0.177 |
oll | +0.166 |
ip | +0.163 |
mm | +0.159 |
Neg logit contributions | |
walk | -0.193 |
call | -0.191 |
push | -0.186 |
street | -0.186 |
garage | -0.184 |
Token | . |
Token ID | 13 |
Feature activation | -3.90e-03 |
Pos logit contributions | |
abella | +0.021 |
mal | +0.021 |
od | +0.020 |
ip | +0.020 |
oll | +0.020 |
Neg logit contributions | |
street | -0.019 |
garage | -0.018 |
distance | -0.018 |
walk | -0.018 |
yard | -0.017 |
Token | He |
Token ID | 679 |
Feature activation | +5.38e-03 |
Pos logit contributions | |
street | +0.006 |
garage | +0.005 |
shout | +0.005 |
call | +0.005 |
push | +0.005 |
Neg logit contributions | |
mal | -0.007 |
abella | -0.007 |
od | -0.007 |
ils | -0.007 |
ret | -0.007 |
Token | plays |
Token ID | 5341 |
Feature activation | +5.19e-03 |
Pos logit contributions | |
abella | +0.010 |
mal | +0.010 |
od | +0.010 |
oll | +0.010 |
ip | +0.010 |
Neg logit contributions | |
walk | -0.007 |
garage | -0.007 |
street | -0.007 |
push | -0.007 |
call | -0.006 |
Token | them |
Token ID | 606 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
abella | +0.010 |
mal | +0.010 |
oll | +0.009 |
ip | +0.009 |
od | +0.009 |
Neg logit contributions | |
street | -0.007 |
walk | -0.007 |
garage | -0.007 |
push | -0.006 |
shout | -0.006 |
Token | After |
Token ID | 2293 |
Feature activation | +2.13e-03 |
Pos logit contributions | |
street | +0.004 |
garage | +0.004 |
walk | +0.004 |
yard | +0.004 |
shout | +0.004 |
Neg logit contributions | |
abella | -0.009 |
mal | -0.008 |
od | -0.008 |
ils | -0.008 |
oll | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
abella | +0.004 |
mal | +0.004 |
od | +0.004 |
ip | +0.004 |
oll | +0.004 |
Neg logit contributions | |
street | -0.003 |
garage | -0.003 |
walk | -0.003 |
push | -0.003 |
yard | -0.003 |
Token | while |
Token ID | 981 |
Feature activation | -3.56e-03 |
Pos logit contributions | |
abella | +0.040 |
mal | +0.038 |
ip | +0.037 |
od | +0.037 |
ret | +0.036 |
Neg logit contributions | |
garage | -0.024 |
yard | -0.023 |
street | -0.023 |
shed | -0.021 |
scream | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
street | +0.003 |
garage | +0.003 |
walk | +0.003 |
yard | +0.003 |
grill | +0.003 |
Neg logit contributions | |
abella | -0.008 |
mal | -0.008 |
ip | -0.008 |
od | -0.008 |
oll | -0.008 |
Token | D |
Token ID | 360 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
walk | +0.015 |
while | +0.014 |
distance | +0.013 |
push | +0.013 |
shout | +0.013 |
Neg logit contributions | |
ils | -0.028 |
abella | -0.027 |
od | -0.027 |
oll | -0.027 |
mal | -0.026 |
Token | ott |
Token ID | 1252 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
abella | +0.021 |
ils | +0.017 |
mal | +0.015 |
ls | +0.015 |
aro | +0.015 |
Neg logit contributions | |
garage | -0.049 |
walk | -0.047 |
street | -0.047 |
yard | -0.047 |
shed | -0.046 |
Token | ie |
Token ID | 494 |
Feature activation | +2.14e-03 |
Pos logit contributions | |
abella | +0.185 |
mal | +0.178 |
mm | +0.173 |
Qu | +0.172 |
ip | +0.172 |
Neg logit contributions | |
garage | -0.169 |
street | -0.165 |
call | -0.164 |
yard | -0.164 |
calling | -0.162 |
Token | was |
Token ID | 373 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
abella | +0.004 |
mal | +0.004 |
od | +0.004 |
ip | +0.004 |
oll | +0.004 |
Neg logit contributions | |
street | -0.003 |
garage | -0.003 |
walk | -0.003 |
push | -0.002 |
yard | -0.002 |
Token | no |
Token ID | 645 |
Feature activation | +8.50e-03 |
Pos logit contributions | |
street | +0.031 |
garage | +0.030 |
walk | +0.030 |
grill | +0.029 |
push | +0.028 |
Neg logit contributions | |
abella | -0.037 |
mal | -0.034 |
od | -0.033 |
oll | -0.033 |
ip | -0.031 |
Token | longer |
Token ID | 2392 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
abella | +0.011 |
mal | +0.011 |
od | +0.011 |
oll | +0.011 |
ip | +0.010 |
Neg logit contributions | |
street | -0.016 |
garage | -0.015 |
yard | -0.015 |
grill | -0.014 |
walk | -0.014 |
Token | scared |
Token ID | 12008 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
garage | +0.034 |
street | +0.034 |
walk | +0.034 |
grill | +0.032 |
push | +0.031 |
Neg logit contributions | |
abella | -0.049 |
mal | -0.048 |
od | -0.047 |
oll | -0.045 |
ip | -0.045 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
street | +0.023 |
garage | +0.023 |
walk | +0.022 |
shout | +0.022 |
scream | +0.022 |
Neg logit contributions | |
abella | -0.022 |
mal | -0.021 |
od | -0.020 |
ip | -0.020 |
oll | -0.020 |
Token | |
Token ID | 220 |
Feature activation | +7.61e-02 |
Pos logit contributions | |
ils | +0.028 |
abella | +0.028 |
mal | +0.027 |
oll | +0.026 |
od | +0.026 |
Neg logit contributions | |
shout | -0.010 |
walk | -0.010 |
street | -0.010 |
call | -0.010 |
push | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
abella | +0.149 |
mal | +0.147 |
od | +0.139 |
oll | +0.138 |
ip | +0.138 |
Neg logit contributions | |
street | -0.090 |
garage | -0.089 |
walk | -0.088 |
yard | -0.084 |
shout | -0.084 |
Token |
|
Token ID | 198 |
Feature activation | +2.97e-03 |
Pos logit contributions | |
ils | +0.030 |
abella | +0.029 |
oll | +0.028 |
mal | +0.028 |
od | +0.028 |
Neg logit contributions | |
walk | -0.008 |
shout | -0.007 |
push | -0.007 |
rescue | -0.007 |
call | -0.007 |
Token | D |
Token ID | 35 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
abella | +0.005 |
oll | +0.005 |
ils | +0.005 |
od | +0.005 |
mal | +0.005 |
Neg logit contributions | |
walk | -0.005 |
push | -0.004 |
street | -0.004 |
shout | -0.004 |
pull | -0.004 |
Token | ott |
Token ID | 1252 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
abella | +0.058 |
ils | +0.048 |
aro | +0.045 |
unic | +0.045 |
ls | +0.042 |
Neg logit contributions | |
garage | -0.112 |
shed | -0.107 |
yard | -0.104 |
street | -0.102 |
push | -0.102 |
Token | ie |
Token ID | 494 |
Feature activation | +1.99e-03 |
Pos logit contributions | |
abella | +0.189 |
mal | +0.183 |
Qu | +0.180 |
mm | +0.180 |
ip | +0.178 |
Neg logit contributions | |
garage | -0.159 |
yard | -0.155 |
street | -0.154 |
call | -0.148 |
calling | -0.148 |
Token | per |
Token ID | 583 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
abella | +0.004 |
mal | +0.004 |
od | +0.003 |
ip | +0.003 |
oll | +0.003 |
Neg logit contributions | |
street | -0.003 |
garage | -0.003 |
walk | -0.003 |
yard | -0.002 |
push | -0.002 |
Token | ked |
Token ID | 9091 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
abella | +0.040 |
mal | +0.039 |
od | +0.037 |
ip | +0.037 |
oll | +0.037 |
Neg logit contributions | |
street | -0.031 |
garage | -0.030 |
walk | -0.030 |
yard | -0.029 |
push | -0.029 |
Token | up |
Token ID | 510 |
Feature activation | +8.64e-03 |
Pos logit contributions | |
abella | +0.026 |
mal | +0.026 |
od | +0.025 |
ip | +0.024 |
oll | +0.024 |
Neg logit contributions | |
street | -0.020 |
garage | -0.020 |
walk | -0.020 |
yard | -0.019 |
push | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
abella | +0.018 |
mal | +0.018 |
od | +0.017 |
ip | +0.017 |
oll | +0.017 |
Neg logit contributions | |
street | -0.009 |
garage | -0.009 |
walk | -0.008 |
yard | -0.008 |
scream | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
street | +0.018 |
garage | +0.017 |
push | +0.017 |
walk | +0.017 |
yard | +0.017 |
Neg logit contributions | |
abella | -0.021 |
oll | -0.020 |
mal | -0.019 |
ip | -0.019 |
od | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
abella | +0.038 |
mal | +0.038 |
ip | +0.036 |
oll | +0.036 |
od | +0.036 |
Neg logit contributions | |
street | -0.035 |
garage | -0.034 |
walk | -0.034 |
shout | -0.033 |
scream | -0.032 |
Token | your |
Token ID | 534 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
abella | +0.036 |
mal | +0.034 |
od | +0.033 |
ip | +0.033 |
oll | +0.033 |
Neg logit contributions | |
street | -0.031 |
walk | -0.030 |
garage | -0.030 |
push | -0.029 |
scream | -0.029 |
Token | friendship |
Token ID | 14738 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
abella | +0.079 |
mal | +0.075 |
od | +0.072 |
ip | +0.071 |
oll | +0.071 |
Neg logit contributions | |
street | -0.072 |
walk | -0.071 |
garage | -0.071 |
yard | -0.068 |
push | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +1.79e-03 |
Pos logit contributions | |
mal | +0.029 |
abella | +0.029 |
od | +0.028 |
ip | +0.027 |
oll | +0.027 |
Neg logit contributions | |
street | -0.029 |
garage | -0.029 |
walk | -0.028 |
shout | -0.028 |
push | -0.027 |
Token | Is |
Token ID | 1148 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
abella | +0.003 |
mal | +0.003 |
ip | +0.003 |
od | +0.003 |
ils | +0.003 |
Neg logit contributions | |
street | -0.002 |
garage | -0.002 |
shout | -0.002 |
call | -0.002 |
push | -0.002 |
Token | that |
Token ID | 326 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
mal | +0.203 |
cast | +0.192 |
ink | +0.189 |
act | +0.187 |
icer | +0.182 |
Neg logit contributions | |
calling | -0.138 |
Charlie | -0.138 |
walk | -0.136 |
run | -0.131 |
shout | -0.128 |
Token | what |
Token ID | 644 |
Feature activation | +5.74e-02 |
Pos logit contributions | |
abella | +0.054 |
mal | +0.053 |
od | +0.050 |
ip | +0.049 |
oll | +0.049 |
Neg logit contributions | |
street | -0.057 |
walk | -0.056 |
garage | -0.056 |
push | -0.054 |
yard | -0.053 |
Token | you |
Token ID | 345 |
Feature activation | +6.03e-04 |
Pos logit contributions | |
abella | +0.079 |
mal | +0.077 |
od | +0.073 |
ip | +0.072 |
oll | +0.071 |
Neg logit contributions | |
street | -0.102 |
walk | -0.100 |
garage | -0.100 |
push | -0.097 |
yard | -0.096 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.14e-03 |
Pos logit contributions | |
abella | +0.001 |
mal | +0.001 |
od | +0.001 |
ip | +0.001 |
oll | +0.001 |
Neg logit contributions | |
street | -0.001 |
walk | -0.001 |
garage | -0.001 |
push | -0.001 |
shout | -0.001 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
street | +0.005 |
garage | +0.005 |
walk | +0.004 |
scream | +0.004 |
distance | +0.004 |
Neg logit contributions | |
abella | -0.005 |
mal | -0.005 |
od | -0.005 |
ip | -0.005 |
ils | -0.005 |
Token | He |
Token ID | 679 |
Feature activation | +4.41e-03 |
Pos logit contributions | |
street | +0.004 |
garage | +0.004 |
call | +0.004 |
yard | +0.004 |
shout | +0.004 |
Neg logit contributions | |
mal | -0.008 |
ils | -0.008 |
ls | -0.008 |
abella | -0.008 |
ip | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
abella | +0.008 |
mal | +0.008 |
od | +0.008 |
oll | +0.008 |
ip | +0.008 |
Neg logit contributions | |
street | -0.006 |
walk | -0.006 |
garage | -0.006 |
push | -0.005 |
shout | -0.005 |
Token | no |
Token ID | 645 |
Feature activation | +9.87e-03 |
Pos logit contributions | |
street | +0.026 |
walk | +0.026 |
garage | +0.026 |
push | +0.025 |
shout | +0.025 |
Neg logit contributions | |
abella | -0.038 |
mal | -0.038 |
od | -0.036 |
ip | -0.036 |
oll | -0.036 |
Token | match |
Token ID | 2872 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
abella | +0.016 |
mal | +0.015 |
oll | +0.015 |
od | +0.015 |
ip | +0.015 |
Neg logit contributions | |
street | -0.015 |
garage | -0.015 |
walk | -0.014 |
yard | -0.014 |
grill | -0.014 |
Token | for |
Token ID | 329 |
Feature activation | +3.91e-03 |
Pos logit contributions | |
abella | +0.034 |
mal | +0.033 |
ip | +0.031 |
od | +0.031 |
oll | +0.031 |
Neg logit contributions | |
street | -0.022 |
walk | -0.021 |
garage | -0.021 |
push | -0.020 |
shout | -0.020 |
Token | Is |
Token ID | 1148 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
mal | +0.007 |
abella | +0.007 |
ip | +0.007 |
oll | +0.006 |
ret | +0.006 |
Neg logit contributions | |
walk | -0.006 |
push | -0.005 |
calling | -0.005 |
call | -0.005 |
street | -0.005 |
Token | la |
Token ID | 5031 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
mal | +0.130 |
mm | +0.112 |
od | +0.109 |
act | +0.109 |
cast | +0.109 |
Neg logit contributions | |
walk | -0.228 |
calling | -0.218 |
call | -0.216 |
push | -0.215 |
shout | -0.209 |
Token | when |
Token ID | 618 |
Feature activation | +1.37e-03 |
Pos logit contributions | |
abella | +0.056 |
mal | +0.053 |
oll | +0.051 |
ip | +0.050 |
od | +0.050 |
Neg logit contributions | |
walk | -0.036 |
street | -0.036 |
garage | -0.034 |
push | -0.034 |
call | -0.033 |
Token | it |
Token ID | 340 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
mal | +0.003 |
oll | +0.003 |
abella | +0.002 |
ip | +0.002 |
ils | +0.002 |
Neg logit contributions | |
walk | -0.002 |
street | -0.002 |
garage | -0.002 |
calling | -0.002 |
shout | -0.002 |
Token | came |
Token ID | 1625 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
abella | +0.055 |
mal | +0.054 |
od | +0.052 |
ip | +0.051 |
oll | +0.051 |
Neg logit contributions | |
street | -0.042 |
garage | -0.042 |
walk | -0.041 |
yard | -0.039 |
shout | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | +9.50e-03 |
Pos logit contributions | |
abella | +0.029 |
mal | +0.028 |
od | +0.028 |
ip | +0.026 |
oll | +0.026 |
Neg logit contributions | |
street | -0.022 |
garage | -0.021 |
walk | -0.021 |
scream | -0.020 |
shout | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
abella | +0.020 |
mal | +0.019 |
od | +0.018 |
ip | +0.018 |
oll | +0.018 |
Neg logit contributions | |
street | -0.019 |
garage | -0.018 |
walk | -0.018 |
yard | -0.017 |
distance | -0.017 |
Token | book |
Token ID | 1492 |
Feature activation | +6.37e-02 |
Pos logit contributions | |
ip | +0.043 |
mal | +0.041 |
l | +0.039 |
ls | +0.038 |
abella | +0.038 |
Neg logit contributions | |
street | -0.036 |
garage | -0.036 |
spine | -0.035 |
walk | -0.034 |
urge | -0.034 |
Token | as |
Token ID | 355 |
Feature activation | -6.14e-03 |
Pos logit contributions | |
abella | +0.116 |
mal | +0.113 |
oll | +0.112 |
od | +0.112 |
ip | +0.112 |
Neg logit contributions | |
street | -0.080 |
garage | -0.079 |
walk | -0.077 |
urge | -0.073 |
grill | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | -8.60e-03 |
Pos logit contributions | |
street | +0.008 |
garage | +0.008 |
walk | +0.008 |
yard | +0.007 |
push | +0.007 |
Neg logit contributions | |
abella | -0.011 |
mal | -0.011 |
od | -0.011 |
ip | -0.011 |
oll | -0.011 |
Token | ga |
Token ID | 31986 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
garage | +0.009 |
street | +0.009 |
walk | +0.009 |
distance | +0.009 |
yard | +0.009 |
Neg logit contributions | |
mal | -0.016 |
abella | -0.016 |
ip | -0.015 |
od | -0.015 |
oll | -0.015 |
Token | vel |
Token ID | 626 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
abella | +0.068 |
mal | +0.065 |
od | +0.061 |
ip | +0.060 |
oll | +0.059 |
Neg logit contributions | |
street | -0.116 |
walk | -0.114 |
garage | -0.114 |
push | -0.111 |
shout | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | +7.50e-04 |
Pos logit contributions | |
abella | +0.180 |
mal | +0.172 |
od | +0.158 |
oll | +0.157 |
ip | +0.157 |
Neg logit contributions | |
street | -0.193 |
garage | -0.191 |
walk | -0.189 |
push | -0.185 |
yard | -0.183 |
Token | They |
Token ID | 1119 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
abella | +0.002 |
mal | +0.002 |
od | +0.001 |
oll | +0.001 |
ip | +0.001 |
Neg logit contributions | |
street | -0.001 |
garage | -0.001 |
walk | -0.001 |
yard | -0.001 |
shout | -0.001 |
Token | used |
Token ID | 973 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
walk | +0.019 |
street | +0.019 |
garage | +0.018 |
push | +0.018 |
shout | +0.018 |
Neg logit contributions | |
abella | -0.035 |
mal | -0.034 |
od | -0.033 |
oll | -0.033 |
ip | -0.033 |
Token | a |
Token ID | 257 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
abella | +0.021 |
mal | +0.021 |
od | +0.020 |
ip | +0.020 |
oll | +0.020 |
Neg logit contributions | |
street | -0.019 |
garage | -0.019 |
walk | -0.019 |
yard | -0.018 |
scream | -0.018 |
Token | hat |
Token ID | 6877 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
ip | +0.043 |
mal | +0.042 |
abella | +0.040 |
ret | +0.039 |
od | +0.039 |
Neg logit contributions | |
street | -0.037 |
garage | -0.037 |
walk | -0.036 |
urge | -0.035 |
calling | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +2.31e-03 |
Pos logit contributions | |
abella | +0.025 |
mal | +0.023 |
od | +0.022 |
ip | +0.021 |
ils | +0.021 |
Neg logit contributions | |
garage | -0.028 |
street | -0.028 |
walk | -0.028 |
shout | -0.027 |
scream | -0.027 |
Token | bush |
Token ID | 24484 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
mal | +0.005 |
abella | +0.005 |
ip | +0.004 |
od | +0.004 |
ret | +0.004 |
Neg logit contributions | |
walk | -0.002 |
shout | -0.002 |
nod | -0.002 |
scream | -0.002 |
street | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
od | +0.058 |
ip | +0.058 |
abella | +0.057 |
mal | +0.057 |
oll | +0.055 |
Neg logit contributions | |
street | -0.040 |
garage | -0.040 |
walk | -0.038 |
brake | -0.038 |
shout | -0.037 |
Token | watched |
Token ID | 7342 |
Feature activation | +3.39e-04 |
Pos logit contributions | |
abella | +0.050 |
mal | +0.049 |
od | +0.046 |
ip | +0.046 |
oll | +0.046 |
Neg logit contributions | |
street | -0.041 |
walk | -0.040 |
garage | -0.040 |
push | -0.038 |
yard | -0.038 |
Token | in |
Token ID | 287 |
Feature activation | -9.96e-03 |
Pos logit contributions | |
abella | +0.000 |
mal | +0.000 |
oll | +0.000 |
ip | +0.000 |
od | +0.000 |
Neg logit contributions | |
walk | -0.001 |
street | -0.001 |
push | -0.001 |
shout | -0.001 |
scream | -0.001 |
Token | amaz |
Token ID | 40642 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
street | +0.015 |
garage | +0.014 |
walk | +0.014 |
grill | +0.013 |
yard | +0.013 |
Neg logit contributions | |
abella | -0.016 |
mal | -0.015 |
od | -0.015 |
ret | -0.015 |
ip | -0.014 |
Token | ement |
Token ID | 972 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
mal | +0.191 |
abella | +0.179 |
ip | +0.167 |
ret | +0.159 |
od | +0.158 |
Neg logit contributions | |
walk | -0.174 |
garage | -0.171 |
push | -0.168 |
yard | -0.168 |
street | -0.168 |
Token | as |
Token ID | 355 |
Feature activation | +8.59e-05 |
Pos logit contributions | |
abella | +0.101 |
mal | +0.099 |
od | +0.094 |
ip | +0.094 |
oll | +0.093 |
Neg logit contributions | |
street | -0.059 |
walk | -0.058 |
garage | -0.057 |
push | -0.054 |
shout | -0.054 |
Token | it |
Token ID | 340 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
mal | +0.000 |
abella | +0.000 |
ip | +0.000 |
oll | +0.000 |
od | +0.000 |
Neg logit contributions | |
walk | -0.000 |
street | -0.000 |
garage | -0.000 |
push | -0.000 |
scream | -0.000 |
Token | flew |
Token ID | 13112 |
Feature activation | -6.16e-03 |
Pos logit contributions | |
abella | +0.073 |
mal | +0.071 |
od | +0.068 |
oll | +0.068 |
ip | +0.068 |
Neg logit contributions | |
walk | -0.047 |
street | -0.047 |
garage | -0.046 |
push | -0.045 |
call | -0.044 |
Token | away |
Token ID | 1497 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
street | +0.008 |
garage | +0.008 |
walk | +0.008 |
scream | +0.008 |
yard | +0.008 |
Neg logit contributions | |
abella | -0.011 |
mal | -0.011 |
od | -0.010 |
ip | -0.010 |
ret | -0.010 |
Token | cheer |
Token ID | 14042 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
abella | +0.035 |
mal | +0.034 |
od | +0.033 |
ip | +0.033 |
oll | +0.033 |
Neg logit contributions | |
street | -0.016 |
garage | -0.016 |
walk | -0.016 |
yard | -0.015 |
push | -0.015 |
Token | each |
Token ID | 1123 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
abella | +0.034 |
mal | +0.034 |
oll | +0.032 |
ip | +0.032 |
od | +0.030 |
Neg logit contributions | |
walk | -0.021 |
push | -0.021 |
shout | -0.021 |
street | -0.021 |
call | -0.021 |
Token | other |
Token ID | 584 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
street | +0.024 |
garage | +0.024 |
walk | +0.024 |
yard | +0.023 |
push | +0.022 |
Neg logit contributions | |
abella | -0.023 |
mal | -0.023 |
od | -0.022 |
ip | -0.021 |
oll | -0.021 |
Token | up |
Token ID | 510 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
street | +0.052 |
walk | +0.052 |
garage | +0.051 |
push | +0.049 |
shout | +0.048 |
Neg logit contributions | |
abella | -0.080 |
mal | -0.078 |
ip | -0.074 |
oll | -0.074 |
od | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +2.21e-04 |
Pos logit contributions | |
street | +0.019 |
walk | +0.018 |
garage | +0.018 |
push | +0.018 |
yard | +0.017 |
Neg logit contributions | |
abella | -0.033 |
mal | -0.033 |
oll | -0.031 |
od | -0.031 |
ip | -0.031 |
Token | Is |
Token ID | 1148 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
abella | +0.000 |
mal | +0.000 |
od | +0.000 |
ip | +0.000 |
oll | +0.000 |
Neg logit contributions | |
street | -0.000 |
garage | -0.000 |
walk | -0.000 |
push | -0.000 |
yard | -0.000 |
Token | that |
Token ID | 326 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
mal | +0.198 |
oll | +0.183 |
act | +0.179 |
cast | +0.179 |
ubb | +0.177 |
Neg logit contributions | |
walk | -0.152 |
calling | -0.148 |
push | -0.148 |
run | -0.141 |
Charlie | -0.141 |
Token | okay |
Token ID | 8788 |
Feature activation | +7.44e-03 |
Pos logit contributions | |
abella | +0.055 |
mal | +0.053 |
od | +0.051 |
ip | +0.050 |
oll | +0.050 |
Neg logit contributions | |
street | -0.049 |
garage | -0.048 |
walk | -0.048 |
push | -0.046 |
yard | -0.046 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
abella | +0.012 |
mal | +0.012 |
od | +0.011 |
ip | +0.011 |
oll | +0.011 |
Neg logit contributions | |
street | -0.011 |
walk | -0.011 |
garage | -0.011 |
distance | -0.011 |
shout | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
abella | +0.030 |
od | +0.029 |
ip | +0.028 |
mal | +0.028 |
oll | +0.026 |
Neg logit contributions | |
street | -0.036 |
walk | -0.035 |
garage | -0.035 |
spine | -0.035 |
grill | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
abella | +0.034 |
mal | +0.033 |
od | +0.032 |
ip | +0.032 |
oll | +0.031 |
Neg logit contributions | |
street | -0.031 |
walk | -0.030 |
garage | -0.030 |
push | -0.029 |
urge | -0.029 |
Token | just |
Token ID | 655 |
Feature activation | -3.54e-03 |
Pos logit contributions | |
mal | +0.052 |
abella | +0.051 |
oll | +0.049 |
ip | +0.049 |
od | +0.049 |
Neg logit contributions | |
walk | -0.035 |
street | -0.034 |
garage | -0.033 |
push | -0.033 |
call | -0.033 |
Token | stared |
Token ID | 18484 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
street | +0.004 |
walk | +0.004 |
garage | +0.004 |
push | +0.004 |
shout | +0.004 |
Neg logit contributions | |
abella | -0.007 |
mal | -0.007 |
od | -0.006 |
ip | -0.006 |
oll | -0.006 |
Token | at |
Token ID | 379 |
Feature activation | +6.34e-03 |
Pos logit contributions | |
street | +0.020 |
walk | +0.020 |
garage | +0.020 |
push | +0.019 |
yard | +0.019 |
Neg logit contributions | |
abella | -0.022 |
mal | -0.021 |
ip | -0.020 |
od | -0.020 |
oll | -0.020 |
Token | him |
Token ID | 683 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
mal | +0.011 |
abella | +0.011 |
oll | +0.011 |
ip | +0.011 |
od | +0.010 |
Neg logit contributions | |
walk | -0.009 |
street | -0.009 |
garage | -0.009 |
push | -0.008 |
call | -0.008 |
Token | in |
Token ID | 287 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
walk | +0.012 |
street | +0.012 |
garage | +0.012 |
push | +0.011 |
shout | +0.011 |
Neg logit contributions | |
abella | -0.022 |
mal | -0.021 |
od | -0.020 |
oll | -0.020 |
ip | -0.020 |
Token | amaz |
Token ID | 40642 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
street | +0.014 |
garage | +0.013 |
walk | +0.013 |
yard | +0.012 |
push | +0.012 |
Neg logit contributions | |
abella | -0.019 |
mal | -0.019 |
od | -0.018 |
oll | -0.018 |
ip | -0.018 |
Token | ement |
Token ID | 972 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
mal | +0.180 |
abella | +0.176 |
ip | +0.163 |
ret | +0.153 |
od | +0.150 |
Neg logit contributions | |
garage | -0.181 |
walk | -0.178 |
street | -0.178 |
yard | -0.175 |
push | -0.172 |
Token | . |
Token ID | 13 |
Feature activation | -4.05e-03 |
Pos logit contributions | |
abella | +0.091 |
mal | +0.089 |
ip | +0.084 |
oll | +0.084 |
od | +0.083 |
Neg logit contributions | |
street | -0.049 |
walk | -0.048 |
garage | -0.046 |
push | -0.045 |
shout | -0.045 |
Token | So |
Token ID | 1406 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
walk | +0.006 |
street | +0.005 |
garage | +0.005 |
push | +0.005 |
distance | +0.005 |
Neg logit contributions | |
abella | -0.007 |
mal | -0.007 |
od | -0.007 |
ip | -0.007 |
oll | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
walk | +0.025 |
street | +0.025 |
push | +0.024 |
garage | +0.024 |
shout | +0.023 |
Neg logit contributions | |
mal | -0.052 |
abella | -0.052 |
oll | -0.050 |
od | -0.049 |
ip | -0.049 |
Token | he |
Token ID | 339 |
Feature activation | +8.25e-03 |
Pos logit contributions | |
walk | +0.030 |
street | +0.030 |
push | +0.029 |
shout | +0.029 |
garage | +0.028 |
Neg logit contributions | |
abella | -0.040 |
mal | -0.040 |
oll | -0.038 |
od | -0.037 |
ip | -0.037 |
Token | his |
Token ID | 465 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
mal | +0.068 |
abella | +0.067 |
oll | +0.067 |
ils | +0.066 |
od | +0.065 |
Neg logit contributions | |
walk | -0.041 |
push | -0.041 |
garage | -0.040 |
street | -0.040 |
shout | -0.039 |
Token | shirt |
Token ID | 10147 |
Feature activation | +6.53e-02 |
Pos logit contributions | |
abella | +0.099 |
mal | +0.096 |
od | +0.093 |
ip | +0.092 |
oll | +0.091 |
Neg logit contributions | |
street | -0.051 |
walk | -0.050 |
garage | -0.050 |
push | -0.047 |
yard | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
abella | +0.151 |
mal | +0.149 |
od | +0.144 |
ip | +0.144 |
oll | +0.143 |
Neg logit contributions | |
street | -0.054 |
walk | -0.052 |
garage | -0.052 |
push | -0.048 |
shout | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | +6.86e-03 |
Pos logit contributions | |
abella | +0.046 |
mal | +0.045 |
od | +0.045 |
ret | +0.044 |
oll | +0.043 |
Neg logit contributions | |
walk | -0.030 |
street | -0.029 |
push | -0.028 |
garage | -0.028 |
shout | -0.027 |
Token | l |
Token ID | 300 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
abella | +0.012 |
mal | +0.012 |
od | +0.011 |
ip | +0.011 |
oll | +0.011 |
Neg logit contributions | |
street | -0.010 |
walk | -0.010 |
garage | -0.010 |
push | -0.009 |
scream | -0.009 |
Token | oll |
Token ID | 692 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
abella | +0.061 |
ils | +0.056 |
od | +0.055 |
mal | +0.055 |
unic | +0.055 |
Neg logit contributions | |
push | -0.193 |
call | -0.184 |
scream | -0.184 |
garage | -0.183 |
pull | -0.181 |
Token | ipop |
Token ID | 42800 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
od | +0.179 |
abella | +0.177 |
unic | +0.160 |
years | +0.156 |
Jul | +0.155 |
Neg logit contributions | |
push | -0.170 |
walk | -0.167 |
pull | -0.157 |
stick | -0.157 |
while | -0.156 |
Token | in |
Token ID | 287 |
Feature activation | +7.43e-03 |
Pos logit contributions | |
abella | +0.114 |
mal | +0.110 |
od | +0.106 |
ip | +0.103 |
oll | +0.103 |
Neg logit contributions | |
street | -0.088 |
walk | -0.087 |
garage | -0.086 |
push | -0.083 |
yard | -0.082 |
Token | his |
Token ID | 465 |
Feature activation | +5.44e-02 |
Pos logit contributions | |
abella | +0.015 |
mal | +0.015 |
oll | +0.015 |
ip | +0.015 |
ils | +0.014 |
Neg logit contributions | |
push | -0.009 |
garage | -0.009 |
distance | -0.009 |
walk | -0.009 |
shout | -0.009 |
Token | mouth |
Token ID | 5422 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
abella | +0.116 |
mal | +0.112 |
od | +0.109 |
ip | +0.107 |
oll | +0.107 |
Neg logit contributions | |
street | -0.058 |
garage | -0.057 |
walk | -0.057 |
yard | -0.053 |
push | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -8.77e-03 |
Pos logit contributions | |
abella | +0.097 |
mal | +0.095 |
od | +0.091 |
oll | +0.091 |
ip | +0.091 |
Neg logit contributions | |
street | -0.071 |
garage | -0.069 |
walk | -0.069 |
yard | -0.065 |
push | -0.065 |
Token | win |
Token ID | 1592 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
street | +0.015 |
garage | +0.015 |
distance | +0.013 |
walk | +0.013 |
yard | +0.013 |
Neg logit contributions | |
abella | -0.032 |
mal | -0.031 |
od | -0.031 |
ip | -0.029 |
oll | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +2.22e-03 |
Pos logit contributions | |
mal | +0.066 |
abella | +0.065 |
oll | +0.064 |
ip | +0.063 |
ret | +0.060 |
Neg logit contributions | |
street | -0.035 |
walk | -0.035 |
push | -0.034 |
shout | -0.033 |
calling | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | +2.13e-03 |
Pos logit contributions | |
mal | +0.005 |
abella | +0.005 |
ils | +0.005 |
od | +0.005 |
oll | +0.005 |
Neg logit contributions | |
street | -0.002 |
garage | -0.002 |
yard | -0.002 |
walk | -0.002 |
shout | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -4.50e-03 |
Pos logit contributions | |
mal | +0.005 |
ils | +0.005 |
oll | +0.005 |
abella | +0.004 |
od | +0.004 |
Neg logit contributions | |
street | -0.002 |
walk | -0.002 |
garage | -0.002 |
yard | -0.002 |
shout | -0.002 |
Token | Sometimes |
Token ID | 15468 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
street | +0.006 |
walk | +0.006 |
garage | +0.006 |
yard | +0.006 |
push | +0.006 |
Neg logit contributions | |
mal | -0.008 |
abella | -0.008 |
oll | -0.008 |
od | -0.008 |
ip | -0.007 |
Token | Is |
Token ID | 1148 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
street | +0.016 |
walk | +0.015 |
garage | +0.015 |
push | +0.014 |
shout | +0.014 |
Neg logit contributions | |
abella | -0.042 |
mal | -0.042 |
oll | -0.040 |
ip | -0.040 |
od | -0.040 |
Token | la |
Token ID | 5031 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
mal | +0.165 |
mm | +0.149 |
oll | +0.138 |
cast | +0.138 |
ink | +0.137 |
Neg logit contributions | |
walk | -0.194 |
Charlie | -0.176 |
calling | -0.176 |
street | -0.173 |
push | -0.173 |
Token | 's |
Token ID | 338 |
Feature activation | +5.71e-03 |
Pos logit contributions | |
abella | +0.068 |
mal | +0.066 |
oll | +0.064 |
od | +0.061 |
ip | +0.061 |
Neg logit contributions | |
walk | -0.045 |
street | -0.043 |
push | -0.042 |
garage | -0.042 |
call | -0.040 |
Token | dad |
Token ID | 9955 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
abella | +0.010 |
mal | +0.010 |
od | +0.009 |
ip | +0.009 |
oll | +0.009 |
Neg logit contributions | |
street | -0.008 |
garage | -0.008 |
walk | -0.007 |
shout | -0.007 |
scream | -0.007 |
Token | would |
Token ID | 561 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
street | +0.031 |
garage | +0.030 |
walk | +0.030 |
yard | +0.029 |
push | +0.029 |
Neg logit contributions | |
abella | -0.046 |
mal | -0.045 |
od | -0.044 |
ip | -0.043 |
oll | -0.043 |
Token | play |
Token ID | 711 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
garage | +0.015 |
street | +0.015 |
yard | +0.015 |
spine | +0.014 |
grill | +0.014 |
Neg logit contributions | |
abella | -0.027 |
mal | -0.025 |
od | -0.025 |
ret | -0.024 |
oll | -0.023 |
Token | Is |
Token ID | 3792 |
Feature activation | +9.47e-02 |
Pos logit contributions | |
abella | +0.057 |
ip | +0.055 |
oll | +0.052 |
ils | +0.052 |
mal | +0.051 |
Neg logit contributions | |
street | -0.044 |
garage | -0.042 |
scream | -0.042 |
push | -0.042 |
walk | -0.042 |
Token | that |
Token ID | 326 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
former | +0.169 |
ink | +0.169 |
oll | +0.167 |
ouses | +0.164 |
ubb | +0.164 |
Neg logit contributions | |
walk | -0.117 |
push | -0.111 |
run | -0.107 |
fight | -0.106 |
Charlie | -0.105 |
Token | another |
Token ID | 1194 |
Feature activation | -8.69e-03 |
Pos logit contributions | |
abella | +0.030 |
mal | +0.029 |
ip | +0.028 |
oll | +0.027 |
od | +0.027 |
Neg logit contributions | |
street | -0.024 |
garage | -0.023 |
walk | -0.023 |
push | -0.022 |
shout | -0.022 |
Token | giant |
Token ID | 6175 |
Feature activation | +5.84e-04 |
Pos logit contributions | |
street | +0.012 |
garage | +0.011 |
walk | +0.011 |
yard | +0.011 |
push | +0.011 |
Neg logit contributions | |
abella | -0.016 |
mal | -0.016 |
ip | -0.015 |
od | -0.015 |
oll | -0.015 |
Token | ? |
Token ID | 30 |
Feature activation | -2.54e-03 |
Pos logit contributions | |
abella | +0.001 |
mal | +0.001 |
od | +0.001 |
ip | +0.001 |
oll | +0.001 |
Neg logit contributions | |
street | -0.001 |
walk | -0.001 |
garage | -0.001 |
push | -0.001 |
yard | -0.001 |
Token | Is |
Token ID | 1148 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
garage | +0.003 |
street | +0.003 |
push | +0.003 |
call | +0.003 |
shout | +0.003 |
Neg logit contributions | |
abella | -0.005 |
mal | -0.004 |
ils | -0.004 |
ip | -0.004 |
ls | -0.004 |
Token | he |
Token ID | 339 |
Feature activation | +9.34e-03 |
Pos logit contributions | |
mal | +0.181 |
act | +0.179 |
ip | +0.167 |
ind | +0.165 |
unic | +0.164 |
Neg logit contributions | |
push | -0.160 |
walk | -0.158 |
Charlie | -0.156 |
calling | -0.151 |
shout | -0.151 |
Token | stronger |
Token ID | 7387 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
abella | +0.018 |
mal | +0.017 |
od | +0.016 |
ip | +0.016 |
oll | +0.016 |
Neg logit contributions | |
walk | -0.012 |
street | -0.012 |
garage | -0.012 |
push | -0.011 |
call | -0.011 |
Token | than |
Token ID | 621 |
Feature activation | +8.88e-03 |
Pos logit contributions | |
abella | +0.075 |
mal | +0.073 |
od | +0.070 |
ip | +0.069 |
oll | +0.069 |
Neg logit contributions | |
street | -0.048 |
garage | -0.046 |
walk | -0.046 |
push | -0.044 |
yard | -0.044 |
Token | me |
Token ID | 502 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
abella | +0.014 |
mal | +0.014 |
od | +0.013 |
ip | +0.013 |
oll | +0.013 |
Neg logit contributions | |
street | -0.014 |
walk | -0.014 |
garage | -0.014 |
push | -0.014 |
shout | -0.013 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
street | +0.017 |
garage | +0.017 |
walk | +0.017 |
shout | +0.016 |
push | +0.016 |
Neg logit contributions | |
abella | -0.019 |
mal | -0.019 |
od | -0.018 |
ip | -0.018 |
oll | -0.018 |
Token | I |
Token ID | 314 |
Feature activation | -9.90e-03 |
Pos logit contributions | |
street | +0.018 |
garage | +0.018 |
walk | +0.018 |
push | +0.017 |
shout | +0.017 |
Neg logit contributions | |
abella | -0.025 |
mal | -0.024 |
od | -0.023 |
ip | -0.023 |
oll | -0.022 |
Token | dropped |
Token ID | 5710 |
Feature activation | +6.77e-03 |
Pos logit contributions | |
walk | +0.012 |
street | +0.012 |
garage | +0.012 |
push | +0.012 |
scream | +0.012 |
Neg logit contributions | |
abella | -0.019 |
mal | -0.018 |
od | -0.018 |
ip | -0.017 |
oll | -0.017 |
Token | your |
Token ID | 534 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
abella | +0.012 |
mal | +0.012 |
oll | +0.012 |
od | +0.012 |
ils | +0.011 |
Neg logit contributions | |
push | -0.010 |
street | -0.009 |
walk | -0.009 |
shout | -0.009 |
garage | -0.009 |
Token | violin |
Token ID | 38283 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
mal | +0.088 |
abella | +0.087 |
ip | +0.084 |
od | +0.082 |
ret | +0.082 |
Neg logit contributions | |
street | -0.052 |
garage | -0.051 |
walk | -0.050 |
yard | -0.048 |
shout | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -7.73e-03 |
Pos logit contributions | |
mal | +0.030 |
ip | +0.029 |
oll | +0.029 |
abella | +0.029 |
od | +0.028 |
Neg logit contributions | |
street | -0.025 |
garage | -0.025 |
walk | -0.025 |
shout | -0.024 |
distance | -0.024 |
Token | Is |
Token ID | 1148 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
street | +0.010 |
garage | +0.010 |
push | +0.010 |
shout | +0.009 |
scream | +0.009 |
Neg logit contributions | |
abella | -0.015 |
mal | -0.014 |
od | -0.014 |
ip | -0.014 |
ils | -0.014 |
Token | it |
Token ID | 340 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
ind | +0.203 |
former | +0.203 |
act | +0.202 |
mal | +0.197 |
ink | +0.197 |
Neg logit contributions | |
walk | -0.135 |
push | -0.131 |
run | -0.130 |
Charlie | -0.129 |
tried | -0.124 |
Token | broken |
Token ID | 5445 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
abella | +0.028 |
mal | +0.027 |
ip | +0.026 |
od | +0.026 |
oll | +0.026 |
Neg logit contributions | |
street | -0.024 |
garage | -0.023 |
walk | -0.023 |
yard | -0.022 |
shout | -0.022 |
Token | ?" |
Token ID | 1701 |
Feature activation | +9.85e-03 |
Pos logit contributions | |
abella | +0.023 |
mal | +0.023 |
ip | +0.021 |
oll | +0.021 |
od | +0.021 |
Neg logit contributions | |
street | -0.023 |
walk | -0.023 |
garage | -0.023 |
shout | -0.022 |
scream | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
abella | +0.020 |
mal | +0.019 |
od | +0.019 |
oll | +0.018 |
ip | +0.018 |
Neg logit contributions | |
street | -0.012 |
garage | -0.011 |
walk | -0.011 |
shout | -0.011 |
push | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +8.38e-03 |
Pos logit contributions | |
mal | +0.038 |
abella | +0.037 |
ils | +0.036 |
od | +0.036 |
oll | +0.036 |
Neg logit contributions | |
street | -0.013 |
garage | -0.013 |
walk | -0.013 |
yard | -0.012 |
shout | -0.012 |
Token | not |
Token ID | 407 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
abella | +0.012 |
mal | +0.012 |
od | +0.012 |
ip | +0.011 |
oll | +0.011 |
Neg logit contributions | |
street | -0.007 |
garage | -0.007 |
walk | -0.006 |
yard | -0.006 |
push | -0.006 |
Token | see |
Token ID | 766 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
mal | +0.032 |
abella | +0.031 |
od | +0.031 |
ip | +0.030 |
oll | +0.030 |
Neg logit contributions | |
garage | -0.020 |
street | -0.019 |
yard | -0.018 |
spine | -0.018 |
distance | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
abella | +0.024 |
mal | +0.024 |
od | +0.022 |
oll | +0.022 |
ip | +0.022 |
Neg logit contributions | |
street | -0.024 |
walk | -0.024 |
garage | -0.024 |
push | -0.023 |
calling | -0.023 |
Token | surprise |
Token ID | 5975 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
street | +0.017 |
walk | +0.017 |
garage | +0.016 |
push | +0.016 |
shout | +0.016 |
Neg logit contributions | |
abella | -0.023 |
mal | -0.023 |
od | -0.021 |
oll | -0.021 |
ip | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -2.13e-03 |
Pos logit contributions | |
abella | +0.072 |
mal | +0.070 |
od | +0.067 |
ip | +0.066 |
oll | +0.065 |
Neg logit contributions | |
street | -0.063 |
walk | -0.062 |
garage | -0.062 |
push | -0.059 |
yard | -0.059 |
Token | Is |
Token ID | 1148 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
street | +0.003 |
walk | +0.003 |
garage | +0.003 |
push | +0.003 |
yard | +0.003 |
Neg logit contributions | |
abella | -0.004 |
mal | -0.004 |
od | -0.003 |
oll | -0.003 |
ip | -0.003 |
Token | that |
Token ID | 326 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
mal | +0.187 |
act | +0.171 |
cast | +0.169 |
od | +0.169 |
ink | +0.169 |
Neg logit contributions | |
walk | -0.160 |
push | -0.154 |
Charlie | -0.154 |
calling | -0.152 |
run | -0.151 |
Token | what |
Token ID | 644 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
abella | +0.046 |
mal | +0.044 |
od | +0.042 |
ip | +0.042 |
oll | +0.041 |
Neg logit contributions | |
street | -0.053 |
garage | -0.051 |
walk | -0.051 |
shout | -0.050 |
distance | -0.049 |
Token | you |
Token ID | 345 |
Feature activation | +5.12e-03 |
Pos logit contributions | |
abella | +0.073 |
mal | +0.070 |
od | +0.067 |
ip | +0.066 |
oll | +0.066 |
Neg logit contributions | |
street | -0.096 |
garage | -0.093 |
distance | -0.092 |
walk | -0.092 |
shout | -0.090 |
Token | want |
Token ID | 765 |
Feature activation | -8.12e-03 |
Pos logit contributions | |
mal | +0.010 |
abella | +0.010 |
od | +0.009 |
oll | +0.009 |
ip | +0.009 |
Neg logit contributions | |
street | -0.007 |
garage | -0.006 |
walk | -0.006 |
distance | -0.006 |
yard | -0.006 |
Token | ?" |
Token ID | 1701 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
street | +0.014 |
garage | +0.013 |
distance | +0.013 |
walk | +0.013 |
spine | +0.013 |
Neg logit contributions | |
abella | -0.012 |
mal | -0.011 |
ip | -0.011 |
ils | -0.011 |
oll | -0.011 |
Token | € |
Token ID | 26391 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
abella | +0.042 |
mal | +0.041 |
od | +0.040 |
ip | +0.039 |
oll | +0.038 |
Neg logit contributions | |
street | -0.036 |
walk | -0.035 |
garage | -0.035 |
push | -0.034 |
scream | -0.034 |
Token | � |
Token ID | 129 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
abella | +0.062 |
mal | +0.060 |
oll | +0.057 |
od | +0.057 |
ip | +0.056 |
Neg logit contributions | |
street | -0.044 |
walk | -0.043 |
garage | -0.043 |
yard | -0.041 |
push | -0.041 |
Token | � |
Token ID | 241 |
Feature activation | +3.10e-03 |
Pos logit contributions | |
mal | +0.074 |
abella | +0.074 |
oll | +0.071 |
ip | +0.071 |
od | +0.070 |
Neg logit contributions | |
street | -0.061 |
walk | -0.058 |
shout | -0.057 |
garage | -0.057 |
scream | -0.056 |
Token | Hello |
Token ID | 15496 |
Feature activation | -8.52e-03 |
Pos logit contributions | |
ils | +0.005 |
mal | +0.005 |
abella | +0.005 |
oll | +0.005 |
od | +0.005 |
Neg logit contributions | |
street | -0.004 |
walk | -0.004 |
shout | -0.004 |
push | -0.004 |
garage | -0.004 |
Token | ? |
Token ID | 30 |
Feature activation | +6.72e-03 |
Pos logit contributions | |
walk | +0.008 |
garage | +0.008 |
street | +0.007 |
push | +0.007 |
call | +0.007 |
Neg logit contributions | |
abella | -0.018 |
mal | -0.018 |
ret | -0.017 |
oll | -0.017 |
ils | -0.017 |
Token | Is |
Token ID | 1148 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
abella | +0.011 |
mal | +0.011 |
od | +0.010 |
ip | +0.010 |
oll | +0.010 |
Neg logit contributions | |
street | -0.010 |
walk | -0.010 |
garage | -0.010 |
yard | -0.010 |
push | -0.010 |
Token | anyone |
Token ID | 2687 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
mal | +0.171 |
act | +0.158 |
cast | +0.154 |
mm | +0.150 |
ip | +0.150 |
Neg logit contributions | |
walk | -0.168 |
Charlie | -0.164 |
push | -0.154 |
run | -0.149 |
follow | -0.146 |
Token | there |
Token ID | 612 |
Feature activation | -3.22e-03 |
Pos logit contributions | |
street | +0.025 |
garage | +0.025 |
walk | +0.025 |
yard | +0.024 |
push | +0.023 |
Neg logit contributions | |
abella | -0.034 |
mal | -0.033 |
od | -0.032 |
ip | -0.031 |
oll | -0.031 |
Token | ? |
Token ID | 30 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
street | +0.005 |
garage | +0.004 |
walk | +0.004 |
push | +0.004 |
yard | +0.004 |
Neg logit contributions | |
abella | -0.005 |
mal | -0.005 |
od | -0.005 |
ip | -0.005 |
oll | -0.005 |
Token | â |
Token ID | 22940 |
Feature activation | +6.82e-03 |
Pos logit contributions | |
abella | +0.020 |
mal | +0.020 |
od | +0.019 |
ip | +0.019 |
oll | +0.018 |
Neg logit contributions | |
street | -0.017 |
walk | -0.017 |
garage | -0.016 |
yard | -0.016 |
scream | -0.016 |
Token | € |
Token ID | 26391 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
abella | +0.011 |
mal | +0.011 |
od | +0.010 |
oll | +0.010 |
ip | +0.010 |
Neg logit contributions | |
garage | -0.010 |
street | -0.010 |
walk | -0.010 |
push | -0.010 |
yard | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -8.21e-03 |
Pos logit contributions | |
garage | +0.051 |
street | +0.049 |
yard | +0.048 |
grill | +0.047 |
nod | +0.047 |
Neg logit contributions | |
abella | -0.068 |
mal | -0.066 |
ip | -0.061 |
ls | -0.060 |
od | -0.059 |
Token | He |
Token ID | 679 |
Feature activation | -2.36e-03 |
Pos logit contributions | |
calling | +0.015 |
distance | +0.015 |
console | +0.014 |
walk | +0.014 |
coat | +0.014 |
Neg logit contributions | |
abella | -0.010 |
od | -0.009 |
mal | -0.009 |
ip | -0.008 |
oll | -0.008 |
Token | shouted |
Token ID | 17293 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
street | +0.002 |
walk | +0.002 |
garage | +0.002 |
push | +0.002 |
yard | +0.002 |
Neg logit contributions | |
abella | -0.005 |
mal | -0.005 |
od | -0.005 |
ip | -0.005 |
oll | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
street | +0.063 |
garage | +0.061 |
distance | +0.058 |
walk | +0.057 |
yard | +0.057 |
Neg logit contributions | |
abella | -0.117 |
mal | -0.112 |
od | -0.111 |
ip | -0.109 |
oll | -0.106 |
Token | " |
Token ID | 366 |
Feature activation | +4.57e-04 |
Pos logit contributions | |
garage | +0.061 |
street | +0.059 |
distance | +0.058 |
yard | +0.056 |
walk | +0.055 |
Neg logit contributions | |
abella | -0.065 |
mal | -0.061 |
ip | -0.060 |
od | -0.059 |
Qu | -0.055 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
abella | +0.001 |
mal | +0.001 |
od | +0.001 |
oll | +0.001 |
ip | +0.001 |
Neg logit contributions | |
street | -0.001 |
garage | -0.001 |
walk | -0.001 |
yard | -0.001 |
scream | -0.001 |
Token | ! |
Token ID | 0 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
garage | +0.196 |
shed | +0.179 |
yard | +0.178 |
street | +0.177 |
workshop | +0.175 |
Neg logit contributions | |
abella | -0.149 |
mal | -0.147 |
ip | -0.144 |
ls | -0.137 |
Qu | -0.132 |
Token | Run |
Token ID | 5660 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
street | +0.041 |
walk | +0.041 |
garage | +0.040 |
distance | +0.040 |
yard | +0.038 |
Neg logit contributions | |
abella | -0.043 |
mal | -0.042 |
od | -0.039 |
ip | -0.039 |
oll | -0.039 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.55e-03 |
Pos logit contributions | |
garage | +0.099 |
yard | +0.092 |
street | +0.092 |
shed | +0.092 |
distance | +0.088 |
Neg logit contributions | |
abella | -0.081 |
mal | -0.078 |
ip | -0.074 |
ret | -0.070 |
od | -0.069 |
Token | His |
Token ID | 2399 |
Feature activation | -1.06e-03 |
Pos logit contributions | |
garage | +0.003 |
calling | +0.003 |
coat | +0.003 |
shed | +0.002 |
walk | +0.002 |
Neg logit contributions | |
abella | -0.002 |
ip | -0.002 |
od | -0.002 |
mal | -0.002 |
Qu | -0.002 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
street | +0.002 |
garage | +0.002 |
walk | +0.002 |
distance | +0.002 |
push | +0.002 |
Neg logit contributions | |
mal | -0.001 |
abella | -0.001 |
ip | -0.001 |
od | -0.001 |
oll | -0.001 |
Token | strange |
Token ID | 6283 |
Feature activation | -8.34e-03 |
Pos logit contributions | |
abella | +0.045 |
mal | +0.043 |
od | +0.041 |
ip | +0.041 |
oll | +0.040 |
Neg logit contributions | |
street | -0.048 |
walk | -0.047 |
garage | -0.047 |
push | -0.046 |
shout | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -5.18e-03 |
Pos logit contributions | |
street | +0.013 |
garage | +0.013 |
yard | +0.012 |
walk | +0.012 |
distance | +0.012 |
Neg logit contributions | |
abella | -0.013 |
mal | -0.013 |
od | -0.013 |
oll | -0.012 |
ils | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -1.23e-03 |
Pos logit contributions | |
street | +0.008 |
walk | +0.008 |
garage | +0.008 |
push | +0.008 |
call | +0.007 |
Neg logit contributions | |
abella | -0.008 |
mal | -0.008 |
od | -0.007 |
ip | -0.007 |
oll | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -8.63e-03 |
Pos logit contributions | |
street | +0.002 |
garage | +0.002 |
walk | +0.002 |
push | +0.002 |
scream | +0.002 |
Neg logit contributions | |
abella | -0.002 |
mal | -0.002 |
od | -0.002 |
ip | -0.002 |
oll | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
street | +0.013 |
garage | +0.013 |
walk | +0.013 |
scream | +0.012 |
yard | +0.012 |
Neg logit contributions | |
abella | -0.014 |
mal | -0.014 |
od | -0.013 |
ip | -0.013 |
oll | -0.013 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
abella | +0.019 |
mal | +0.019 |
ip | +0.018 |
od | +0.018 |
oll | +0.018 |
Neg logit contributions | |
street | -0.019 |
walk | -0.019 |
garage | -0.019 |
push | -0.018 |
shout | -0.018 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.59e-03 |
Pos logit contributions | |
garage | +0.188 |
street | +0.178 |
yard | +0.174 |
walk | +0.164 |
distance | +0.163 |
Neg logit contributions | |
abella | -0.161 |
mal | -0.159 |
ip | -0.152 |
ls | -0.149 |
od | -0.147 |
Token | warned |
Token ID | 7728 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
abella | +0.003 |
mal | +0.003 |
od | +0.003 |
oll | +0.002 |
ip | +0.002 |
Neg logit contributions | |
garage | -0.002 |
street | -0.002 |
yard | -0.002 |
walk | -0.002 |
distance | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
street | +0.019 |
garage | +0.019 |
walk | +0.018 |
yard | +0.018 |
push | +0.017 |
Neg logit contributions | |
abella | -0.026 |
mal | -0.025 |
od | -0.025 |
ip | -0.024 |
oll | -0.024 |
Token | voice |
Token ID | 3809 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
garage | +0.014 |
street | +0.013 |
yard | +0.013 |
walk | +0.012 |
grill | +0.011 |
Neg logit contributions | |
abella | -0.023 |
mal | -0.023 |
od | -0.022 |
ip | -0.021 |
ret | -0.021 |
Token | from |
Token ID | 422 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
garage | +0.032 |
street | +0.032 |
yard | +0.031 |
driveway | +0.029 |
distance | +0.028 |
Neg logit contributions | |
od | -0.029 |
abella | -0.028 |
mal | -0.027 |
oll | -0.027 |
ls | -0.027 |
Token | : |
Token ID | 25 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
abella | +0.002 |
od | +0.002 |
oll | +0.002 |
mal | +0.002 |
ret | +0.002 |
Neg logit contributions | |
street | -0.001 |
push | -0.001 |
walk | -0.001 |
grill | -0.001 |
shout | -0.001 |
Token | " |
Token ID | 366 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
street | +0.039 |
walk | +0.038 |
garage | +0.038 |
yard | +0.036 |
push | +0.036 |
Neg logit contributions | |
abella | -0.070 |
mal | -0.069 |
od | -0.066 |
ip | -0.066 |
oll | -0.065 |
Token | These |
Token ID | 4711 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
street | +0.016 |
walk | +0.015 |
garage | +0.015 |
push | +0.015 |
shout | +0.014 |
Neg logit contributions | |
abella | -0.019 |
mal | -0.019 |
od | -0.018 |
oll | -0.018 |
ip | -0.018 |
Token | are |
Token ID | 389 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
street | +0.015 |
garage | +0.015 |
walk | +0.015 |
push | +0.014 |
shout | +0.014 |
Neg logit contributions | |
abella | -0.021 |
mal | -0.020 |
oll | -0.019 |
od | -0.019 |
ip | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -6.19e-02 |
Pos logit contributions | |
call | +0.077 |
push | +0.076 |
calling | +0.075 |
shout | +0.074 |
street | +0.074 |
Neg logit contributions | |
abella | -0.097 |
mal | -0.096 |
ip | -0.093 |
oll | -0.092 |
od | -0.091 |
Token | dogs |
Token ID | 6844 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
walk | +0.094 |
street | +0.093 |
garage | +0.091 |
scream | +0.091 |
shout | +0.090 |
Neg logit contributions | |
mal | -0.092 |
abella | -0.090 |
od | -0.087 |
oll | -0.084 |
ip | -0.084 |
Token | from |
Token ID | 422 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
garage | +0.156 |
street | +0.154 |
walk | +0.148 |
distance | +0.144 |
yard | +0.144 |
Neg logit contributions | |
mal | -0.181 |
abella | -0.179 |
od | -0.175 |
ip | -0.174 |
ils | -0.171 |
Token | the |
Token ID | 262 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
street | +0.024 |
walk | +0.023 |
garage | +0.023 |
push | +0.022 |
yard | +0.022 |
Neg logit contributions | |
abella | -0.026 |
mal | -0.025 |
od | -0.024 |
oll | -0.024 |
ip | -0.024 |
Token | shelter |
Token ID | 11772 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
garage | +0.019 |
walk | +0.019 |
street | +0.019 |
push | +0.018 |
scream | +0.018 |
Neg logit contributions | |
abella | -0.035 |
mal | -0.035 |
ip | -0.033 |
od | -0.033 |
oll | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
garage | +0.029 |
walk | +0.028 |
street | +0.028 |
push | +0.027 |
shout | +0.027 |
Neg logit contributions | |
abella | -0.039 |
mal | -0.038 |
od | -0.037 |
ip | -0.037 |
ils | -0.036 |
Token | They |
Token ID | 1119 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
street | +0.030 |
garage | +0.030 |
walk | +0.029 |
shout | +0.028 |
call | +0.028 |
Neg logit contributions | |
abella | -0.051 |
mal | -0.050 |
od | -0.048 |
ip | -0.047 |
oll | -0.047 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
abella | +0.011 |
mal | +0.011 |
ip | +0.010 |
od | +0.010 |
oll | +0.010 |
Neg logit contributions | |
street | -0.012 |
garage | -0.011 |
walk | -0.011 |
yard | -0.011 |
distance | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -1.34e-03 |
Pos logit contributions | |
mal | +0.038 |
od | +0.036 |
ils | +0.036 |
ip | +0.036 |
abella | +0.035 |
Neg logit contributions | |
street | -0.045 |
garage | -0.044 |
yard | -0.041 |
grill | -0.041 |
driveway | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | +2.78e-03 |
Pos logit contributions | |
street | +0.001 |
walk | +0.001 |
shout | +0.001 |
distance | +0.001 |
nod | +0.001 |
Neg logit contributions | |
ils | -0.003 |
od | -0.003 |
mal | -0.003 |
unic | -0.003 |
oll | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -5.97e-03 |
Pos logit contributions | |
mal | +0.006 |
od | +0.006 |
ils | +0.006 |
oll | +0.006 |
abella | +0.006 |
Neg logit contributions | |
street | -0.003 |
walk | -0.003 |
shout | -0.002 |
garage | -0.002 |
distance | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
street | +0.011 |
walk | +0.010 |
garage | +0.010 |
push | +0.010 |
distance | +0.010 |
Neg logit contributions | |
abella | -0.008 |
mal | -0.008 |
od | -0.008 |
oll | -0.008 |
ret | -0.007 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
abella | +0.023 |
mal | +0.023 |
od | +0.022 |
ip | +0.022 |
oll | +0.021 |
Neg logit contributions | |
street | -0.017 |
walk | -0.017 |
garage | -0.017 |
push | -0.016 |
yard | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
garage | +0.190 |
street | +0.172 |
yard | +0.170 |
workshop | +0.168 |
shed | +0.167 |
Neg logit contributions | |
ip | -0.149 |
mal | -0.148 |
abella | -0.147 |
ls | -0.146 |
ils | -0.133 |
Token | run |
Token ID | 1057 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
walk | +0.051 |
push | +0.050 |
scream | +0.050 |
shout | +0.050 |
call | +0.049 |
Neg logit contributions | |
abella | -0.046 |
od | -0.045 |
mal | -0.045 |
ret | -0.043 |
ip | -0.043 |
Token | !" |
Token ID | 2474 |
Feature activation | +9.73e-03 |
Pos logit contributions | |
garage | +0.076 |
street | +0.073 |
yard | +0.071 |
walk | +0.071 |
nod | +0.069 |
Neg logit contributions | |
abella | -0.063 |
mal | -0.062 |
ip | -0.060 |
ls | -0.058 |
od | -0.055 |
Token | Sara |
Token ID | 24799 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
abella | +0.014 |
mal | +0.014 |
od | +0.013 |
oll | +0.012 |
ip | +0.012 |
Neg logit contributions | |
street | -0.017 |
walk | -0.017 |
garage | -0.016 |
shout | -0.016 |
scream | -0.016 |
Token | says |
Token ID | 1139 |
Feature activation | -5.18e-03 |
Pos logit contributions | |
ls | +0.033 |
abella | +0.031 |
ip | +0.030 |
od | +0.030 |
mal | +0.030 |
Neg logit contributions | |
garage | -0.041 |
distance | -0.039 |
street | -0.038 |
driveway | -0.038 |
spine | -0.037 |
Token | teeth |
Token ID | 9941 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
abella | +0.030 |
mal | +0.029 |
od | +0.028 |
ip | +0.027 |
oll | +0.027 |
Neg logit contributions | |
street | -0.025 |
garage | -0.024 |
walk | -0.024 |
push | -0.023 |
yard | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +1.71e-03 |
Pos logit contributions | |
abella | +0.027 |
mal | +0.027 |
od | +0.026 |
ip | +0.025 |
oll | +0.025 |
Neg logit contributions | |
street | -0.019 |
garage | -0.018 |
walk | -0.018 |
yard | -0.017 |
shout | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +3.50e-03 |
Pos logit contributions | |
abella | +0.003 |
mal | +0.003 |
od | +0.003 |
oll | +0.003 |
ip | +0.003 |
Neg logit contributions | |
street | -0.003 |
garage | -0.002 |
walk | -0.002 |
shout | -0.002 |
push | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -3.74e-03 |
Pos logit contributions | |
abella | +0.007 |
mal | +0.006 |
od | +0.006 |
ip | +0.006 |
oll | +0.006 |
Neg logit contributions | |
street | -0.005 |
walk | -0.004 |
garage | -0.004 |
push | -0.004 |
yard | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
street | +0.006 |
garage | +0.005 |
walk | +0.005 |
scream | +0.005 |
yard | +0.005 |
Neg logit contributions | |
abella | -0.006 |
mal | -0.006 |
ip | -0.006 |
od | -0.006 |
oll | -0.006 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
abella | +0.037 |
mal | +0.036 |
od | +0.035 |
oll | +0.035 |
ip | +0.035 |
Neg logit contributions | |
street | -0.021 |
walk | -0.020 |
garage | -0.020 |
push | -0.019 |
shout | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
garage | +0.170 |
street | +0.156 |
yard | +0.152 |
driveway | +0.147 |
workshop | +0.146 |
Neg logit contributions | |
mal | -0.165 |
abella | -0.165 |
ip | -0.162 |
ls | -0.160 |
ils | -0.153 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.66e-03 |
Pos logit contributions | |
street | +0.049 |
garage | +0.048 |
walk | +0.047 |
yard | +0.046 |
distance | +0.045 |
Neg logit contributions | |
abella | -0.050 |
mal | -0.049 |
od | -0.046 |
ip | -0.046 |
oll | -0.045 |
Token | , |
Token ID | 11 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
od | +0.006 |
ip | +0.006 |
abella | +0.006 |
mal | +0.005 |
ls | +0.005 |
Neg logit contributions | |
street | -0.008 |
garage | -0.008 |
walk | -0.008 |
distance | -0.008 |
spine | -0.008 |
Token | run |
Token ID | 1057 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
street | +0.028 |
garage | +0.027 |
distance | +0.025 |
yard | +0.024 |
driveway | +0.023 |
Neg logit contributions | |
mal | -0.070 |
abella | -0.069 |
ip | -0.066 |
oll | -0.065 |
od | -0.065 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
garage | +0.071 |
street | +0.067 |
yard | +0.066 |
nod | +0.064 |
walk | +0.064 |
Neg logit contributions | |
mal | -0.056 |
abella | -0.055 |
ip | -0.054 |
ls | -0.053 |
ret | -0.050 |
Token | dangerous |
Token ID | 4923 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
abella | +0.062 |
mal | +0.060 |
od | +0.059 |
oll | +0.057 |
ip | +0.057 |
Neg logit contributions | |
walk | -0.037 |
street | -0.037 |
shout | -0.036 |
push | -0.035 |
scream | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +3.49e-04 |
Pos logit contributions | |
garage | +0.082 |
street | +0.079 |
walk | +0.078 |
distance | +0.077 |
nod | +0.075 |
Neg logit contributions | |
oll | -0.060 |
mal | -0.060 |
ls | -0.058 |
ils | -0.058 |
ip | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | +5.38e-03 |
Pos logit contributions | |
abella | +0.001 |
mal | +0.001 |
od | +0.001 |
ip | +0.001 |
oll | +0.001 |
Neg logit contributions | |
street | -0.000 |
garage | -0.000 |
walk | -0.000 |
yard | -0.000 |
shout | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -2.18e-03 |
Pos logit contributions | |
abella | +0.011 |
mal | +0.010 |
od | +0.010 |
oll | +0.010 |
ip | +0.010 |
Neg logit contributions | |
street | -0.006 |
garage | -0.006 |
walk | -0.006 |
yard | -0.006 |
push | -0.006 |
Token | " |
Token ID | 1 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
garage | +0.003 |
street | +0.003 |
walk | +0.003 |
scream | +0.003 |
shout | +0.003 |
Neg logit contributions | |
abella | -0.004 |
mal | -0.003 |
ip | -0.003 |
od | -0.003 |
oll | -0.003 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
abella | +0.028 |
mal | +0.027 |
od | +0.025 |
ip | +0.025 |
oll | +0.025 |
Neg logit contributions | |
street | -0.020 |
garage | -0.020 |
walk | -0.020 |
yard | -0.019 |
push | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
garage | +0.189 |
street | +0.176 |
yard | +0.171 |
workshop | +0.167 |
shed | +0.167 |
Neg logit contributions | |
mal | -0.144 |
abella | -0.142 |
ip | -0.141 |
ls | -0.139 |
ils | -0.130 |
Token | Sue |
Token ID | 26089 |
Feature activation | -3.89e-03 |
Pos logit contributions | |
street | +0.032 |
garage | +0.031 |
walk | +0.030 |
yard | +0.029 |
distance | +0.028 |
Neg logit contributions | |
abella | -0.053 |
mal | -0.052 |
od | -0.049 |
ip | -0.049 |
oll | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
street | +0.006 |
garage | +0.006 |
distance | +0.006 |
walk | +0.006 |
spine | +0.005 |
Neg logit contributions | |
ip | -0.006 |
mal | -0.006 |
od | -0.006 |
abella | -0.006 |
ls | -0.006 |
Token | run |
Token ID | 1057 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
street | +0.023 |
garage | +0.021 |
yard | +0.020 |
distance | +0.020 |
spine | +0.018 |
Neg logit contributions | |
mal | -0.064 |
abella | -0.062 |
ip | -0.061 |
oll | -0.059 |
od | -0.057 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
garage | +0.071 |
street | +0.068 |
yard | +0.066 |
walk | +0.066 |
nod | +0.065 |
Neg logit contributions | |
mal | -0.055 |
abella | -0.054 |
ip | -0.052 |
ls | -0.051 |
ret | -0.049 |
Token | bear |
Token ID | 6842 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
street | +0.014 |
walk | +0.014 |
garage | +0.014 |
push | +0.013 |
shout | +0.013 |
Neg logit contributions | |
mal | -0.014 |
abella | -0.014 |
od | -0.013 |
ip | -0.013 |
unic | -0.013 |
Token | ! |
Token ID | 0 |
Feature activation | -8.90e-03 |
Pos logit contributions | |
street | +0.028 |
garage | +0.027 |
walk | +0.026 |
yard | +0.025 |
distance | +0.025 |
Neg logit contributions | |
mal | -0.033 |
abella | -0.032 |
od | -0.032 |
ip | -0.031 |
oll | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +4.52e-03 |
Pos logit contributions | |
street | +0.011 |
garage | +0.011 |
walk | +0.010 |
yard | +0.010 |
shout | +0.010 |
Neg logit contributions | |
mal | -0.017 |
abella | -0.017 |
od | -0.016 |
oll | -0.016 |
ils | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -1.50e-03 |
Pos logit contributions | |
mal | +0.009 |
abella | +0.009 |
oll | +0.009 |
od | +0.009 |
ip | +0.009 |
Neg logit contributions | |
street | -0.005 |
garage | -0.005 |
walk | -0.005 |
yard | -0.005 |
shout | -0.005 |
Token | " |
Token ID | 1 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
street | +0.002 |
garage | +0.002 |
walk | +0.002 |
push | +0.002 |
yard | +0.002 |
Neg logit contributions | |
abella | -0.002 |
mal | -0.002 |
od | -0.002 |
oll | -0.002 |
ip | -0.002 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
abella | +0.036 |
mal | +0.035 |
ip | +0.034 |
oll | +0.034 |
od | +0.034 |
Neg logit contributions | |
street | -0.021 |
walk | -0.021 |
garage | -0.020 |
push | -0.019 |
shout | -0.019 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
garage | +0.167 |
street | +0.158 |
yard | +0.152 |
distance | +0.144 |
walk | +0.144 |
Neg logit contributions | |
abella | -0.162 |
mal | -0.162 |
ls | -0.157 |
ip | -0.156 |
od | -0.153 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
abella | +0.020 |
mal | +0.020 |
oll | +0.019 |
od | +0.019 |
ip | +0.018 |
Neg logit contributions | |
street | -0.020 |
walk | -0.020 |
garage | -0.020 |
shout | -0.019 |
scream | -0.019 |
Token | screamed |
Token ID | 25421 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
abella | +0.030 |
od | +0.028 |
mal | +0.028 |
ip | +0.027 |
ls | +0.027 |
Neg logit contributions | |
garage | -0.029 |
street | -0.028 |
distance | -0.027 |
walk | -0.027 |
yard | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -2.99e-03 |
Pos logit contributions | |
street | +0.054 |
garage | +0.052 |
yard | +0.050 |
distance | +0.050 |
driveway | +0.048 |
Neg logit contributions | |
od | -0.044 |
abella | -0.043 |
mal | -0.043 |
ip | -0.041 |
oll | -0.040 |
Token | She |
Token ID | 1375 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
walk | +0.005 |
garage | +0.004 |
street | +0.004 |
distance | +0.004 |
yard | +0.004 |
Neg logit contributions | |
abella | -0.005 |
od | -0.005 |
ip | -0.004 |
mal | -0.004 |
ls | -0.004 |
Token | fierce |
Token ID | 14800 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
abella | +0.030 |
mal | +0.029 |
od | +0.028 |
ip | +0.028 |
oll | +0.028 |
Neg logit contributions | |
street | -0.018 |
walk | -0.018 |
shout | -0.018 |
scream | -0.017 |
push | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +1.42e-03 |
Pos logit contributions | |
street | +0.048 |
garage | +0.047 |
walk | +0.046 |
distance | +0.046 |
shout | +0.043 |
Neg logit contributions | |
mal | -0.069 |
oll | -0.067 |
od | -0.067 |
abella | -0.067 |
ils | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +7.07e-03 |
Pos logit contributions | |
ils | +0.003 |
mal | +0.003 |
od | +0.003 |
abella | +0.003 |
oll | +0.003 |
Neg logit contributions | |
yard | -0.001 |
street | -0.001 |
garage | -0.001 |
shout | -0.001 |
grill | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -6.35e-04 |
Pos logit contributions | |
mal | +0.017 |
ils | +0.016 |
oll | +0.016 |
abella | +0.016 |
od | +0.016 |
Neg logit contributions | |
garage | -0.006 |
street | -0.006 |
yard | -0.005 |
walk | -0.005 |
shout | -0.005 |
Token | " |
Token ID | 1 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
street | +0.001 |
garage | +0.001 |
walk | +0.001 |
push | +0.001 |
yard | +0.001 |
Neg logit contributions | |
oll | -0.001 |
abella | -0.001 |
od | -0.001 |
mal | -0.001 |
unic | -0.001 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
abella | +0.033 |
mal | +0.032 |
oll | +0.031 |
od | +0.031 |
ip | +0.031 |
Neg logit contributions | |
street | -0.022 |
walk | -0.022 |
garage | -0.021 |
push | -0.021 |
shout | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
garage | +0.182 |
street | +0.170 |
yard | +0.166 |
shed | +0.159 |
walk | +0.159 |
Neg logit contributions | |
mal | -0.152 |
abella | -0.151 |
ip | -0.148 |
ls | -0.145 |
od | -0.139 |
Token | Lily |
Token ID | 20037 |
Feature activation | +5.86e-03 |
Pos logit contributions | |
walk | +0.041 |
push | +0.041 |
scream | +0.040 |
shout | +0.040 |
call | +0.040 |
Neg logit contributions | |
abella | -0.046 |
od | -0.044 |
mal | -0.044 |
ip | -0.043 |
ret | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
od | +0.007 |
ip | +0.007 |
mal | +0.007 |
abella | +0.007 |
ls | +0.007 |
Neg logit contributions | |
street | -0.011 |
distance | -0.011 |
garage | -0.011 |
walk | -0.010 |
driveway | -0.010 |
Token | run |
Token ID | 1057 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
street | +0.022 |
garage | +0.020 |
distance | +0.019 |
yard | +0.018 |
spine | +0.016 |
Neg logit contributions | |
mal | -0.066 |
abella | -0.065 |
ip | -0.062 |
oll | -0.062 |
od | -0.061 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
garage | +0.071 |
street | +0.068 |
yard | +0.067 |
nod | +0.066 |
walk | +0.066 |
Neg logit contributions | |
mal | -0.050 |
abella | -0.049 |
ip | -0.048 |
ls | -0.047 |
ret | -0.044 |
Token | scared |
Token ID | 12008 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
garage | +0.019 |
street | +0.019 |
walk | +0.018 |
distance | +0.017 |
grill | +0.017 |
Neg logit contributions | |
mal | -0.026 |
abella | -0.026 |
od | -0.025 |
oll | -0.025 |
ls | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +2.20e-05 |
Pos logit contributions | |
street | +0.048 |
garage | +0.048 |
walk | +0.047 |
yard | +0.044 |
push | +0.044 |
Neg logit contributions | |
abella | -0.060 |
mal | -0.058 |
od | -0.057 |
oll | -0.056 |
ip | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | +6.62e-03 |
Pos logit contributions | |
mal | +0.000 |
abella | +0.000 |
oll | +0.000 |
od | +0.000 |
ip | +0.000 |
Neg logit contributions | |
street | -0.000 |
walk | -0.000 |
garage | -0.000 |
shout | -0.000 |
yard | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -1.85e-03 |
Pos logit contributions | |
mal | +0.013 |
abella | +0.013 |
oll | +0.013 |
od | +0.013 |
ip | +0.013 |
Neg logit contributions | |
street | -0.008 |
walk | -0.007 |
garage | -0.007 |
yard | -0.007 |
shout | -0.007 |
Token | " |
Token ID | 1 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
street | +0.003 |
garage | +0.003 |
walk | +0.003 |
scream | +0.003 |
yard | +0.003 |
Neg logit contributions | |
abella | -0.003 |
mal | -0.003 |
ip | -0.003 |
od | -0.003 |
oll | -0.003 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
abella | +0.030 |
mal | +0.030 |
od | +0.028 |
ip | +0.027 |
oll | +0.026 |
Neg logit contributions | |
garage | -0.021 |
street | -0.021 |
walk | -0.021 |
yard | -0.020 |
push | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
garage | +0.173 |
street | +0.161 |
yard | +0.156 |
shed | +0.151 |
workshop | +0.151 |
Neg logit contributions | |
abella | -0.161 |
mal | -0.159 |
ip | -0.155 |
ls | -0.154 |
od | -0.146 |
Token | Ben |
Token ID | 3932 |
Feature activation | +6.94e-03 |
Pos logit contributions | |
walk | +0.034 |
street | +0.034 |
push | +0.033 |
shout | +0.033 |
garage | +0.033 |
Neg logit contributions | |
abella | -0.040 |
mal | -0.039 |
od | -0.038 |
ip | -0.037 |
oll | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
ip | +0.009 |
mal | +0.009 |
ls | +0.008 |
od | +0.008 |
abella | +0.008 |
Neg logit contributions | |
garage | -0.013 |
street | -0.013 |
driveway | -0.012 |
distance | -0.012 |
spine | -0.012 |
Token | run |
Token ID | 1057 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
street | +0.022 |
garage | +0.021 |
yard | +0.020 |
distance | +0.019 |
spine | +0.018 |
Neg logit contributions | |
abella | -0.056 |
mal | -0.056 |
od | -0.052 |
ip | -0.052 |
oll | -0.052 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
garage | +0.070 |
street | +0.068 |
yard | +0.066 |
walk | +0.065 |
nod | +0.065 |
Neg logit contributions | |
abella | -0.054 |
mal | -0.054 |
ip | -0.052 |
ls | -0.051 |
ils | -0.048 |
Token | them |
Token ID | 606 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
street | +0.008 |
garage | +0.008 |
walk | +0.008 |
push | +0.008 |
shout | +0.008 |
Neg logit contributions | |
abella | -0.014 |
mal | -0.014 |
od | -0.014 |
ip | -0.013 |
oll | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
street | +0.044 |
garage | +0.044 |
walk | +0.042 |
yard | +0.041 |
push | +0.041 |
Neg logit contributions | |
abella | -0.052 |
mal | -0.051 |
od | -0.049 |
ip | -0.048 |
oll | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e-03 |
Pos logit contributions | |
street | +0.018 |
garage | +0.018 |
walk | +0.018 |
push | +0.017 |
yard | +0.017 |
Neg logit contributions | |
abella | -0.028 |
mal | -0.028 |
od | -0.027 |
ip | -0.026 |
oll | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -4.25e-03 |
Pos logit contributions | |
street | +0.002 |
garage | +0.002 |
yard | +0.002 |
walk | +0.002 |
shout | +0.002 |
Neg logit contributions | |
mal | -0.004 |
abella | -0.004 |
oll | -0.004 |
od | -0.004 |
ils | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
street | +0.006 |
garage | +0.006 |
walk | +0.006 |
push | +0.006 |
yard | +0.006 |
Neg logit contributions | |
abella | -0.007 |
mal | -0.007 |
od | -0.006 |
ip | -0.006 |
oll | -0.006 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
abella | +0.019 |
mal | +0.019 |
od | +0.018 |
ip | +0.017 |
oll | +0.017 |
Neg logit contributions | |
street | -0.017 |
garage | -0.017 |
walk | -0.017 |
yard | -0.016 |
scream | -0.016 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
garage | +0.182 |
workshop | +0.167 |
street | +0.164 |
yard | +0.163 |
shed | +0.159 |
Neg logit contributions | |
ip | -0.147 |
abella | -0.146 |
ls | -0.145 |
mal | -0.139 |
l | -0.137 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
abella | +0.021 |
mal | +0.020 |
od | +0.019 |
ip | +0.019 |
oll | +0.019 |
Neg logit contributions | |
street | -0.018 |
garage | -0.017 |
walk | -0.017 |
yard | -0.016 |
push | -0.016 |
Token | said |
Token ID | 531 |
Feature activation | +1.91e-03 |
Pos logit contributions | |
abella | +0.015 |
ls | +0.015 |
ip | +0.015 |
l | +0.014 |
mal | +0.014 |
Neg logit contributions | |
garage | -0.020 |
street | -0.018 |
yard | -0.017 |
driveway | -0.017 |
distance | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -9.95e-03 |
Pos logit contributions | |
abella | +0.003 |
ip | +0.003 |
od | +0.003 |
mal | +0.003 |
oll | +0.003 |
Neg logit contributions | |
garage | -0.003 |
street | -0.003 |
yard | -0.003 |
distance | -0.003 |
walk | -0.003 |
Token | He |
Token ID | 679 |
Feature activation | +4.96e-03 |
Pos logit contributions | |
walk | +0.015 |
garage | +0.014 |
street | +0.014 |
yard | +0.014 |
urge | +0.014 |
Neg logit contributions | |
abella | -0.017 |
od | -0.015 |
mal | -0.015 |
ip | -0.015 |
l | -0.014 |
Token | them |
Token ID | 606 |
Feature activation | -9.65e-03 |
Pos logit contributions | |
walk | +0.035 |
street | +0.034 |
push | +0.033 |
call | +0.033 |
shout | +0.032 |
Neg logit contributions | |
oll | -0.037 |
mal | -0.035 |
abella | -0.034 |
ip | -0.034 |
od | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +3.32e-03 |
Pos logit contributions | |
street | +0.014 |
garage | +0.014 |
walk | +0.013 |
yard | +0.013 |
grill | +0.012 |
Neg logit contributions | |
mal | -0.016 |
abella | -0.016 |
od | -0.016 |
ip | -0.015 |
oll | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +8.07e-03 |
Pos logit contributions | |
abella | +0.006 |
mal | +0.006 |
od | +0.006 |
oll | +0.006 |
ip | +0.006 |
Neg logit contributions | |
street | -0.004 |
garage | -0.004 |
walk | -0.004 |
yard | -0.004 |
push | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +8.84e-04 |
Pos logit contributions | |
abella | +0.016 |
mal | +0.015 |
od | +0.015 |
oll | +0.015 |
ip | +0.015 |
Neg logit contributions | |
street | -0.010 |
garage | -0.010 |
walk | -0.010 |
yard | -0.009 |
push | -0.009 |
Token | " |
Token ID | 1 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
abella | +0.001 |
mal | +0.001 |
ip | +0.001 |
od | +0.001 |
oll | +0.001 |
Neg logit contributions | |
street | -0.001 |
walk | -0.001 |
garage | -0.001 |
scream | -0.001 |
shout | -0.001 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
abella | +0.038 |
mal | +0.037 |
oll | +0.037 |
od | +0.036 |
ip | +0.036 |
Neg logit contributions | |
street | -0.019 |
walk | -0.019 |
garage | -0.018 |
push | -0.018 |
distance | -0.018 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
garage | +0.185 |
street | +0.170 |
yard | +0.165 |
workshop | +0.164 |
driveway | +0.162 |
Neg logit contributions | |
mal | -0.145 |
ls | -0.144 |
abella | -0.143 |
ip | -0.142 |
ret | -0.134 |
Token | Tom |
Token ID | 4186 |
Feature activation | +9.66e-03 |
Pos logit contributions | |
abella | +0.016 |
mal | +0.016 |
od | +0.015 |
ip | +0.015 |
oll | +0.015 |
Neg logit contributions | |
street | -0.017 |
walk | -0.016 |
garage | -0.016 |
push | -0.016 |
yard | -0.016 |
Token | said |
Token ID | 531 |
Feature activation | -6.40e-03 |
Pos logit contributions | |
abella | +0.015 |
ip | +0.015 |
mal | +0.015 |
od | +0.014 |
ls | +0.014 |
Neg logit contributions | |
garage | -0.016 |
street | -0.014 |
driveway | -0.014 |
yard | -0.014 |
basement | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -5.20e-03 |
Pos logit contributions | |
garage | +0.011 |
street | +0.011 |
yard | +0.011 |
distance | +0.010 |
driveway | +0.010 |
Neg logit contributions | |
abella | -0.009 |
mal | -0.009 |
od | -0.009 |
ip | -0.009 |
ls | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
walk | +0.007 |
garage | +0.007 |
street | +0.006 |
yard | +0.006 |
distance | +0.006 |
Neg logit contributions | |
abella | -0.010 |
mal | -0.009 |
ip | -0.009 |
od | -0.009 |
oll | -0.009 |
Token | them |
Token ID | 606 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
abella | +0.045 |
mal | +0.043 |
oll | +0.041 |
ip | +0.041 |
od | +0.041 |
Neg logit contributions | |
walk | -0.037 |
street | -0.036 |
garage | -0.036 |
push | -0.036 |
grill | -0.035 |
Token | ! |
Token ID | 0 |
Feature activation | -2.80e-03 |
Pos logit contributions | |
street | +0.031 |
garage | +0.030 |
walk | +0.030 |
push | +0.029 |
yard | +0.029 |
Neg logit contributions | |
abella | -0.047 |
mal | -0.046 |
od | -0.045 |
ip | -0.044 |
oll | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | +8.03e-03 |
Pos logit contributions | |
street | +0.003 |
garage | +0.003 |
walk | +0.003 |
shout | +0.003 |
push | +0.003 |
Neg logit contributions | |
abella | -0.005 |
mal | -0.005 |
oll | -0.005 |
od | -0.005 |
ip | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +2.00e-03 |
Pos logit contributions | |
abella | +0.016 |
mal | +0.016 |
oll | +0.016 |
od | +0.016 |
ip | +0.015 |
Neg logit contributions | |
street | -0.009 |
garage | -0.009 |
walk | -0.009 |
yard | -0.008 |
shout | -0.008 |
Token | " |
Token ID | 1 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
abella | +0.004 |
mal | +0.004 |
od | +0.003 |
oll | +0.003 |
ip | +0.003 |
Neg logit contributions | |
street | -0.003 |
garage | -0.003 |
walk | -0.003 |
push | -0.003 |
yard | -0.002 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
abella | +0.049 |
mal | +0.048 |
od | +0.046 |
ip | +0.046 |
oll | +0.046 |
Neg logit contributions | |
street | -0.019 |
walk | -0.018 |
garage | -0.018 |
push | -0.017 |
shout | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
garage | +0.172 |
street | +0.163 |
yard | +0.153 |
workshop | +0.151 |
driveway | +0.150 |
Neg logit contributions | |
mal | -0.154 |
abella | -0.153 |
ls | -0.151 |
ip | -0.150 |
ret | -0.142 |
Token | Mia |
Token ID | 32189 |
Feature activation | -2.87e-03 |
Pos logit contributions | |
walk | +0.040 |
street | +0.040 |
garage | +0.040 |
push | +0.039 |
shout | +0.039 |
Neg logit contributions | |
abella | -0.045 |
mal | -0.043 |
od | -0.042 |
ip | -0.041 |
oll | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
street | +0.004 |
garage | +0.004 |
distance | +0.004 |
walk | +0.004 |
driveway | +0.003 |
Neg logit contributions | |
ip | -0.005 |
od | -0.005 |
mal | -0.005 |
abella | -0.005 |
ls | -0.005 |
Token | run |
Token ID | 1057 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
street | +0.023 |
garage | +0.021 |
yard | +0.019 |
distance | +0.019 |
driveway | +0.017 |
Neg logit contributions | |
mal | -0.063 |
abella | -0.062 |
ip | -0.058 |
oll | -0.058 |
od | -0.058 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
garage | +0.066 |
street | +0.065 |
yard | +0.062 |
walk | +0.062 |
nod | +0.061 |
Neg logit contributions | |
mal | -0.053 |
abella | -0.052 |
ip | -0.050 |
ls | -0.049 |
od | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +2.27e-03 |
Pos logit contributions | |
garage | +0.027 |
street | +0.027 |
distance | +0.025 |
yard | +0.024 |
grill | +0.024 |
Neg logit contributions | |
abella | -0.032 |
mal | -0.030 |
ip | -0.029 |
od | -0.029 |
oll | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
abella | +0.004 |
mal | +0.004 |
od | +0.004 |
ip | +0.004 |
oll | +0.004 |
Neg logit contributions | |
street | -0.003 |
garage | -0.003 |
walk | -0.003 |
yard | -0.003 |
push | -0.003 |
Token | lion |
Token ID | 18744 |
Feature activation | -1.17e-03 |
Pos logit contributions | |
street | +0.022 |
walk | +0.022 |
garage | +0.022 |
push | +0.021 |
shout | +0.021 |
Neg logit contributions | |
abella | -0.036 |
mal | -0.035 |
ip | -0.033 |
od | -0.033 |
oll | -0.033 |
Token | were |
Token ID | 547 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
street | +0.002 |
garage | +0.001 |
walk | +0.001 |
yard | +0.001 |
push | +0.001 |
Neg logit contributions | |
abella | -0.002 |
mal | -0.002 |
od | -0.002 |
ip | -0.002 |
oll | -0.002 |
Token | best |
Token ID | 1266 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
street | +0.063 |
walk | +0.061 |
garage | +0.061 |
shout | +0.059 |
push | +0.059 |
Neg logit contributions | |
abella | -0.096 |
mal | -0.096 |
ip | -0.091 |
oll | -0.090 |
od | -0.090 |
Token | buddies |
Token ID | 35548 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
street | +0.092 |
garage | +0.091 |
walk | +0.090 |
push | +0.087 |
shout | +0.087 |
Neg logit contributions | |
abella | -0.065 |
mal | -0.062 |
od | -0.060 |
ip | -0.059 |
oll | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -5.36e-03 |
Pos logit contributions | |
street | +0.133 |
walk | +0.129 |
garage | +0.129 |
push | +0.123 |
yard | +0.122 |
Neg logit contributions | |
abella | -0.200 |
mal | -0.196 |
od | -0.188 |
ip | -0.185 |
oll | -0.185 |
Token | They |
Token ID | 1119 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
street | +0.006 |
garage | +0.006 |
walk | +0.006 |
yard | +0.006 |
shout | +0.006 |
Neg logit contributions | |
mal | -0.011 |
abella | -0.011 |
od | -0.010 |
oll | -0.010 |
ils | -0.010 |
Token | played |
Token ID | 2826 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
walk | +0.043 |
street | +0.042 |
garage | +0.041 |
push | +0.040 |
shout | +0.040 |
Neg logit contributions | |
abella | -0.058 |
mal | -0.057 |
oll | -0.054 |
od | -0.054 |
ip | -0.053 |
Token | together |
Token ID | 1978 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
street | +0.038 |
garage | +0.037 |
walk | +0.037 |
push | +0.035 |
yard | +0.035 |
Neg logit contributions | |
abella | -0.042 |
mal | -0.041 |
od | -0.040 |
ip | -0.039 |
oll | -0.038 |
Token | every |
Token ID | 790 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
garage | +0.023 |
street | +0.023 |
walk | +0.022 |
yard | +0.021 |
shout | +0.021 |
Neg logit contributions | |
abella | -0.023 |
mal | -0.022 |
od | -0.022 |
ip | -0.022 |
oll | -0.021 |
Token | volcano |
Token ID | 32410 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
garage | +0.003 |
walk | +0.003 |
street | +0.003 |
distance | +0.003 |
Step | +0.003 |
Neg logit contributions | |
od | -0.003 |
ls | -0.003 |
abella | -0.003 |
mal | -0.003 |
unic | -0.003 |
Token | ! |
Token ID | 0 |
Feature activation | -2.93e-03 |
Pos logit contributions | |
garage | +0.029 |
street | +0.028 |
nod | +0.028 |
walk | +0.028 |
yard | +0.027 |
Neg logit contributions | |
abella | -0.023 |
mal | -0.023 |
od | -0.022 |
ls | -0.022 |
ip | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +5.85e-03 |
Pos logit contributions | |
street | +0.003 |
garage | +0.003 |
walk | +0.003 |
shout | +0.003 |
yard | +0.003 |
Neg logit contributions | |
mal | -0.006 |
abella | -0.006 |
od | -0.005 |
oll | -0.005 |
ils | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -1.55e-04 |
Pos logit contributions | |
mal | +0.012 |
abella | +0.012 |
oll | +0.012 |
od | +0.012 |
ip | +0.011 |
Neg logit contributions | |
street | -0.006 |
garage | -0.006 |
walk | -0.006 |
shout | -0.006 |
yard | -0.006 |
Token | " |
Token ID | 1 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
street | +0.000 |
garage | +0.000 |
walk | +0.000 |
push | +0.000 |
yard | +0.000 |
Neg logit contributions | |
mal | -0.000 |
abella | -0.000 |
oll | -0.000 |
od | -0.000 |
ip | -0.000 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
abella | +0.044 |
mal | +0.043 |
ip | +0.043 |
oll | +0.043 |
od | +0.042 |
Neg logit contributions | |
street | -0.022 |
walk | -0.022 |
garage | -0.021 |
distance | -0.020 |
push | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
garage | +0.173 |
street | +0.164 |
yard | +0.159 |
workshop | +0.153 |
distance | +0.151 |
Neg logit contributions | |
abella | -0.152 |
mal | -0.145 |
ip | -0.141 |
ls | -0.141 |
od | -0.136 |
Token | run |
Token ID | 1057 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
street | +0.035 |
garage | +0.034 |
walk | +0.034 |
push | +0.032 |
yard | +0.032 |
Neg logit contributions | |
abella | -0.059 |
mal | -0.058 |
od | -0.055 |
ip | -0.055 |
oll | -0.055 |
Token | , |
Token ID | 11 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
garage | +0.065 |
street | +0.064 |
yard | +0.062 |
walk | +0.062 |
nod | +0.061 |
Neg logit contributions | |
abella | -0.059 |
mal | -0.057 |
ip | -0.055 |
ls | -0.054 |
od | -0.052 |
Token | run |
Token ID | 1057 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
street | +0.025 |
garage | +0.024 |
yard | +0.023 |
walk | +0.022 |
distance | +0.022 |
Neg logit contributions | |
mal | -0.056 |
abella | -0.055 |
od | -0.052 |
oll | -0.051 |
ip | -0.051 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
garage | +0.064 |
street | +0.062 |
yard | +0.061 |
nod | +0.060 |
walk | +0.060 |
Neg logit contributions | |
abella | -0.051 |
mal | -0.050 |
ip | -0.048 |
ls | -0.047 |
od | -0.045 |
Token | them |
Token ID | 606 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
walk | +0.029 |
street | +0.029 |
push | +0.028 |
garage | +0.028 |
call | +0.028 |
Neg logit contributions | |
oll | -0.036 |
mal | -0.035 |
ip | -0.035 |
abella | -0.034 |
od | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -3.78e-03 |
Pos logit contributions | |
street | +0.016 |
garage | +0.015 |
walk | +0.015 |
distance | +0.015 |
yard | +0.014 |
Neg logit contributions | |
od | -0.016 |
mal | -0.016 |
abella | -0.016 |
ip | -0.015 |
oll | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +9.33e-03 |
Pos logit contributions | |
street | +0.004 |
garage | +0.004 |
shout | +0.003 |
scream | +0.003 |
call | +0.003 |
Neg logit contributions | |
mal | -0.008 |
ils | -0.008 |
abella | -0.008 |
od | -0.008 |
ret | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +2.46e-03 |
Pos logit contributions | |
mal | +0.021 |
abella | +0.020 |
od | +0.020 |
oll | +0.020 |
ils | +0.019 |
Neg logit contributions | |
street | -0.009 |
garage | -0.009 |
yard | -0.009 |
walk | -0.009 |
shout | -0.008 |
Token | " |
Token ID | 1 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
abella | +0.005 |
od | +0.005 |
mal | +0.005 |
oll | +0.005 |
ret | +0.005 |
Neg logit contributions | |
street | -0.003 |
garage | -0.003 |
walk | -0.003 |
push | -0.003 |
yard | -0.003 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
abella | +0.041 |
mal | +0.040 |
oll | +0.040 |
od | +0.040 |
ip | +0.040 |
Neg logit contributions | |
street | -0.026 |
walk | -0.026 |
push | -0.025 |
distance | -0.025 |
garage | -0.025 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
garage | +0.162 |
street | +0.153 |
yard | +0.145 |
workshop | +0.142 |
shed | +0.141 |
Neg logit contributions | |
abella | -0.163 |
mal | -0.162 |
ip | -0.159 |
ls | -0.158 |
ret | -0.150 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
abella | +0.023 |
mal | +0.022 |
od | +0.021 |
ip | +0.021 |
oll | +0.021 |
Neg logit contributions | |
street | -0.024 |
walk | -0.024 |
garage | -0.024 |
push | -0.023 |
shout | -0.023 |
Token | shouts |
Token ID | 34757 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
ls | +0.025 |
abella | +0.024 |
od | +0.024 |
mal | +0.023 |
ip | +0.022 |
Neg logit contributions | |
garage | -0.031 |
distance | -0.030 |
street | -0.029 |
walk | -0.029 |
spine | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -5.75e-03 |
Pos logit contributions | |
street | +0.050 |
garage | +0.050 |
yard | +0.046 |
distance | +0.045 |
walk | +0.045 |
Neg logit contributions | |
mal | -0.066 |
abella | -0.066 |
od | -0.065 |
ip | -0.064 |
oll | -0.063 |
Token | She |
Token ID | 1375 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
garage | +0.009 |
street | +0.009 |
walk | +0.009 |
distance | +0.008 |
yard | +0.008 |
Neg logit contributions | |
abella | -0.009 |
ip | -0.009 |
od | -0.008 |
mal | -0.008 |
ls | -0.008 |
Token | them |
Token ID | 606 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
abella | +0.044 |
mal | +0.042 |
oll | +0.041 |
od | +0.041 |
ret | +0.040 |
Neg logit contributions | |
walk | -0.042 |
push | -0.041 |
grill | -0.040 |
shout | -0.039 |
pull | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +3.13e-03 |
Pos logit contributions | |
walk | +0.038 |
street | +0.036 |
push | +0.035 |
garage | +0.035 |
yard | +0.034 |
Neg logit contributions | |
abella | -0.070 |
mal | -0.068 |
od | -0.064 |
oll | -0.064 |
ret | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | +7.81e-03 |
Pos logit contributions | |
abella | +0.006 |
mal | +0.006 |
od | +0.006 |
oll | +0.006 |
ip | +0.006 |
Neg logit contributions | |
street | -0.004 |
garage | -0.004 |
walk | -0.004 |
shout | -0.003 |
push | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +3.91e-04 |
Pos logit contributions | |
mal | +0.016 |
abella | +0.016 |
oll | +0.015 |
od | +0.015 |
ip | +0.015 |
Neg logit contributions | |
street | -0.009 |
walk | -0.009 |
garage | -0.009 |
shout | -0.008 |
push | -0.008 |
Token | " |
Token ID | 1 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
abella | +0.001 |
mal | +0.001 |
od | +0.001 |
ip | +0.001 |
oll | +0.001 |
Neg logit contributions | |
street | -0.001 |
garage | -0.001 |
walk | -0.001 |
yard | -0.001 |
push | -0.001 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
abella | +0.035 |
mal | +0.034 |
od | +0.032 |
ip | +0.032 |
oll | +0.031 |
Neg logit contributions | |
street | -0.031 |
garage | -0.031 |
walk | -0.030 |
yard | -0.029 |
scream | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
garage | +0.169 |
street | +0.157 |
yard | +0.152 |
workshop | +0.147 |
driveway | +0.145 |
Neg logit contributions | |
abella | -0.161 |
ip | -0.159 |
mal | -0.158 |
ls | -0.153 |
ils | -0.144 |
Token | run |
Token ID | 1057 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
street | +0.025 |
walk | +0.025 |
garage | +0.025 |
push | +0.024 |
shout | +0.024 |
Neg logit contributions | |
abella | -0.053 |
mal | -0.052 |
od | -0.051 |
ip | -0.050 |
oll | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
garage | +0.066 |
street | +0.064 |
yard | +0.062 |
walk | +0.062 |
nod | +0.060 |
Neg logit contributions | |
abella | -0.061 |
mal | -0.061 |
ip | -0.059 |
ls | -0.056 |
od | -0.055 |
Token | run |
Token ID | 1057 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
street | +0.017 |
garage | +0.016 |
yard | +0.014 |
walk | +0.014 |
distance | +0.014 |
Neg logit contributions | |
mal | -0.054 |
abella | -0.053 |
ip | -0.051 |
oll | -0.051 |
od | -0.051 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
garage | +0.068 |
street | +0.065 |
yard | +0.064 |
walk | +0.063 |
nod | +0.063 |
Neg logit contributions | |
mal | -0.051 |
abella | -0.051 |
ip | -0.050 |
ls | -0.048 |
od | -0.045 |
Token | them |
Token ID | 606 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
walk | +0.021 |
push | +0.021 |
shout | +0.021 |
pull | +0.020 |
run | +0.020 |
Neg logit contributions | |
oll | -0.034 |
abella | -0.033 |
mm | -0.031 |
mal | -0.030 |
unic | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -1.16e-04 |
Pos logit contributions | |
garage | +0.032 |
street | +0.032 |
grill | +0.030 |
yard | +0.029 |
walk | +0.029 |
Neg logit contributions | |
mal | -0.031 |
od | -0.031 |
ls | -0.030 |
ip | -0.029 |
abella | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
garage | +0.000 |
street | +0.000 |
shout | +0.000 |
yard | +0.000 |
call | +0.000 |
Neg logit contributions | |
mal | -0.000 |
ils | -0.000 |
abella | -0.000 |
oll | -0.000 |
ip | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +4.82e-03 |
Pos logit contributions | |
mal | +0.027 |
ils | +0.026 |
abella | +0.026 |
oll | +0.026 |
od | +0.025 |
Neg logit contributions | |
garage | -0.010 |
street | -0.010 |
shout | -0.010 |
yard | -0.010 |
push | -0.010 |
Token | " |
Token ID | 1 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
abella | +0.010 |
oll | +0.010 |
mal | +0.010 |
od | +0.010 |
ils | +0.009 |
Neg logit contributions | |
garage | -0.005 |
street | -0.005 |
push | -0.005 |
yard | -0.005 |
walk | -0.005 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
abella | +0.047 |
oll | +0.046 |
ip | +0.046 |
mal | +0.045 |
ret | +0.045 |
Neg logit contributions | |
walk | -0.019 |
street | -0.019 |
push | -0.019 |
garage | -0.018 |
shout | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
garage | +0.151 |
street | +0.140 |
workshop | +0.133 |
yard | +0.132 |
driveway | +0.129 |
Neg logit contributions | |
ls | -0.172 |
mal | -0.169 |
abella | -0.169 |
ip | -0.165 |
od | -0.161 |
Token | L |
Token ID | 406 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
push | +0.036 |
call | +0.035 |
shout | +0.035 |
scream | +0.035 |
walk | +0.034 |
Neg logit contributions | |
abella | -0.052 |
mal | -0.049 |
oll | -0.049 |
ip | -0.049 |
od | -0.049 |
Token | ila |
Token ID | 10102 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
abella | +0.038 |
mal | +0.036 |
ils | +0.032 |
unic | +0.031 |
od | +0.031 |
Neg logit contributions | |
walk | -0.091 |
push | -0.090 |
garage | -0.090 |
street | -0.090 |
shout | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
ip | +0.019 |
od | +0.018 |
ls | +0.018 |
mal | +0.018 |
abella | +0.018 |
Neg logit contributions | |
garage | -0.015 |
street | -0.015 |
distance | -0.014 |
walk | -0.014 |
driveway | -0.013 |
Token | run |
Token ID | 1057 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
street | +0.021 |
garage | +0.019 |
yard | +0.017 |
distance | +0.017 |
walk | +0.015 |
Neg logit contributions | |
abella | -0.070 |
mal | -0.070 |
od | -0.066 |
ip | -0.066 |
oll | -0.066 |
Token | shell |
Token ID | 7582 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
street | +0.004 |
garage | +0.004 |
walk | +0.004 |
push | +0.004 |
shout | +0.004 |
Neg logit contributions | |
abella | -0.005 |
mal | -0.005 |
od | -0.005 |
oll | -0.004 |
ip | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.85e-03 |
Pos logit contributions | |
abella | +0.102 |
mal | +0.101 |
od | +0.096 |
ip | +0.096 |
oll | +0.095 |
Neg logit contributions | |
street | -0.070 |
garage | -0.068 |
walk | -0.068 |
shout | -0.064 |
push | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
street | +0.003 |
garage | +0.003 |
shout | +0.003 |
push | +0.003 |
yard | +0.003 |
Neg logit contributions | |
abella | -0.006 |
mal | -0.006 |
ils | -0.006 |
oll | -0.005 |
od | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +3.11e-03 |
Pos logit contributions | |
mal | +0.023 |
abella | +0.022 |
oll | +0.022 |
od | +0.021 |
ils | +0.021 |
Neg logit contributions | |
street | -0.011 |
garage | -0.011 |
walk | -0.011 |
yard | -0.011 |
shout | -0.011 |
Token | " |
Token ID | 1 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
abella | +0.006 |
mal | +0.006 |
oll | +0.006 |
od | +0.006 |
ip | +0.006 |
Neg logit contributions | |
street | -0.004 |
garage | -0.003 |
walk | -0.003 |
yard | -0.003 |
push | -0.003 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
abella | +0.043 |
mal | +0.042 |
od | +0.040 |
ip | +0.040 |
oll | +0.040 |
Neg logit contributions | |
street | -0.026 |
garage | -0.025 |
walk | -0.025 |
push | -0.024 |
yard | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
garage | +0.173 |
street | +0.161 |
yard | +0.153 |
workshop | +0.151 |
driveway | +0.150 |
Neg logit contributions | |
mal | -0.153 |
abella | -0.152 |
ip | -0.152 |
ls | -0.152 |
ret | -0.142 |
Token | Mia |
Token ID | 32189 |
Feature activation | -2.58e-03 |
Pos logit contributions | |
walk | +0.035 |
push | +0.034 |
scream | +0.034 |
shout | +0.034 |
call | +0.034 |
Neg logit contributions | |
abella | -0.048 |
mal | -0.046 |
od | -0.045 |
ip | -0.045 |
oll | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
street | +0.004 |
garage | +0.003 |
distance | +0.003 |
walk | +0.003 |
driveway | +0.003 |
Neg logit contributions | |
ip | -0.005 |
abella | -0.005 |
od | -0.005 |
mal | -0.005 |
ls | -0.004 |
Token | run |
Token ID | 1057 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
street | +0.023 |
garage | +0.021 |
distance | +0.020 |
yard | +0.020 |
driveway | +0.018 |
Neg logit contributions | |
mal | -0.059 |
abella | -0.058 |
ip | -0.055 |
oll | -0.055 |
od | -0.054 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
garage | +0.064 |
street | +0.063 |
yard | +0.060 |
walk | +0.060 |
nod | +0.059 |
Neg logit contributions | |
mal | -0.053 |
abella | -0.053 |
ip | -0.051 |
ls | -0.049 |
od | -0.048 |
Token | shouted |
Token ID | 17293 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
od | +0.028 |
abella | +0.028 |
mal | +0.026 |
oll | +0.025 |
ip | +0.025 |
Neg logit contributions | |
garage | -0.023 |
street | -0.022 |
distance | -0.022 |
walk | -0.022 |
signal | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +1.88e-03 |
Pos logit contributions | |
street | +0.059 |
garage | +0.058 |
yard | +0.054 |
distance | +0.054 |
driveway | +0.053 |
Neg logit contributions | |
od | -0.054 |
mal | -0.052 |
abella | -0.052 |
ip | -0.051 |
oll | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | +9.77e-03 |
Pos logit contributions | |
abella | +0.003 |
od | +0.003 |
mal | +0.003 |
ip | +0.003 |
ls | +0.002 |
Neg logit contributions | |
walk | -0.003 |
street | -0.003 |
garage | -0.003 |
distance | -0.003 |
spine | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +2.56e-03 |
Pos logit contributions | |
abella | +0.016 |
mal | +0.015 |
od | +0.014 |
ip | +0.014 |
ret | +0.014 |
Neg logit contributions | |
street | -0.015 |
garage | -0.015 |
walk | -0.015 |
distance | -0.014 |
spine | -0.014 |
Token | " |
Token ID | 1 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
abella | +0.003 |
mal | +0.003 |
ip | +0.003 |
od | +0.003 |
ls | +0.003 |
Neg logit contributions | |
spine | -0.005 |
street | -0.005 |
distance | -0.005 |
garage | -0.005 |
calling | -0.005 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
abella | +0.029 |
mal | +0.028 |
ip | +0.028 |
oll | +0.028 |
ret | +0.028 |
Neg logit contributions | |
street | -0.021 |
walk | -0.021 |
garage | -0.020 |
push | -0.020 |
shout | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
garage | +0.141 |
street | +0.134 |
yard | +0.125 |
workshop | +0.122 |
driveway | +0.119 |
Neg logit contributions | |
abella | -0.180 |
mal | -0.178 |
ls | -0.175 |
ip | -0.172 |
od | -0.171 |
Token | run |
Token ID | 1057 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
street | +0.044 |
garage | +0.043 |
walk | +0.043 |
yard | +0.041 |
distance | +0.041 |
Neg logit contributions | |
abella | -0.043 |
mal | -0.043 |
od | -0.040 |
ip | -0.040 |
oll | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
garage | +0.064 |
street | +0.062 |
yard | +0.060 |
walk | +0.059 |
nod | +0.058 |
Neg logit contributions | |
mal | -0.064 |
abella | -0.063 |
ls | -0.061 |
ip | -0.061 |
od | -0.058 |
Token | run |
Token ID | 1057 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
street | +0.022 |
garage | +0.022 |
yard | +0.020 |
distance | +0.019 |
walk | +0.019 |
Neg logit contributions | |
mal | -0.058 |
abella | -0.057 |
ip | -0.054 |
od | -0.054 |
oll | -0.054 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
garage | +0.067 |
street | +0.064 |
yard | +0.062 |
nod | +0.062 |
walk | +0.061 |
Neg logit contributions | |
mal | -0.053 |
ls | -0.051 |
abella | -0.051 |
ip | -0.051 |
ret | -0.047 |
Token | teeth |
Token ID | 9941 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
abella | +0.046 |
mal | +0.044 |
od | +0.042 |
ip | +0.042 |
oll | +0.041 |
Neg logit contributions | |
street | -0.042 |
walk | -0.042 |
garage | -0.041 |
push | -0.040 |
scream | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +8.15e-04 |
Pos logit contributions | |
od | +0.033 |
oll | +0.033 |
mal | +0.032 |
abella | +0.032 |
ip | +0.032 |
Neg logit contributions | |
street | -0.043 |
garage | -0.043 |
walk | -0.039 |
driveway | -0.039 |
yard | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
abella | +0.002 |
mal | +0.002 |
ils | +0.002 |
oll | +0.002 |
od | +0.002 |
Neg logit contributions | |
street | -0.001 |
garage | -0.001 |
walk | -0.001 |
shout | -0.001 |
yard | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +3.14e-03 |
Pos logit contributions | |
mal | +0.021 |
abella | +0.021 |
oll | +0.020 |
od | +0.020 |
ip | +0.019 |
Neg logit contributions | |
street | -0.011 |
garage | -0.011 |
walk | -0.011 |
shout | -0.011 |
yard | -0.011 |
Token | " |
Token ID | 1 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
abella | +0.006 |
mal | +0.005 |
od | +0.005 |
oll | +0.005 |
ip | +0.005 |
Neg logit contributions | |
street | -0.004 |
garage | -0.004 |
walk | -0.004 |
push | -0.004 |
shout | -0.004 |
Token | Run |
Token ID | 10987 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
abella | +0.052 |
mal | +0.050 |
od | +0.049 |
oll | +0.049 |
ip | +0.049 |
Neg logit contributions | |
street | -0.023 |
walk | -0.022 |
garage | -0.022 |
push | -0.021 |
shout | -0.020 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
garage | +0.162 |
street | +0.152 |
yard | +0.145 |
workshop | +0.140 |
driveway | +0.139 |
Neg logit contributions | |
mal | -0.162 |
abella | -0.162 |
ls | -0.159 |
ip | -0.158 |
od | -0.149 |
Token | L |
Token ID | 406 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
abella | +0.024 |
mal | +0.024 |
od | +0.023 |
ip | +0.022 |
oll | +0.022 |
Neg logit contributions | |
street | -0.020 |
walk | -0.019 |
garage | -0.019 |
shout | -0.018 |
push | -0.018 |
Token | ila |
Token ID | 10102 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
mal | +0.062 |
ils | +0.061 |
unic | +0.060 |
abella | +0.057 |
ip | +0.055 |
Neg logit contributions | |
shout | -0.110 |
scream | -0.108 |
push | -0.108 |
call | -0.106 |
calling | -0.103 |
Token | shouted |
Token ID | 17293 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
abella | +0.038 |
od | +0.037 |
ip | +0.037 |
ls | +0.037 |
mal | +0.035 |
Neg logit contributions | |
garage | -0.037 |
street | -0.034 |
driveway | -0.033 |
walk | -0.032 |
distance | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -4.14e-03 |
Pos logit contributions | |
street | +0.054 |
garage | +0.053 |
yard | +0.050 |
distance | +0.049 |
driveway | +0.048 |
Neg logit contributions | |
od | -0.047 |
abella | -0.046 |
mal | -0.046 |
ip | -0.045 |
oll | -0.044 |
Token | dreams |
Token ID | 10625 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
abella | +0.043 |
mal | +0.041 |
od | +0.039 |
ip | +0.039 |
oll | +0.039 |
Neg logit contributions | |
walk | -0.028 |
street | -0.028 |
garage | -0.027 |
yard | -0.027 |
push | -0.026 |
Token | come |
Token ID | 1282 |
Feature activation | +3.44e-03 |
Pos logit contributions | |
abella | +0.044 |
mal | +0.043 |
od | +0.043 |
oll | +0.041 |
ip | +0.041 |
Neg logit contributions | |
street | -0.025 |
garage | -0.025 |
walk | -0.024 |
yard | -0.023 |
grill | -0.023 |
Token | true |
Token ID | 2081 |
Feature activation | +8.68e-02 |
Pos logit contributions | |
abella | +0.006 |
mal | +0.005 |
oll | +0.005 |
ip | +0.005 |
od | +0.005 |
Neg logit contributions | |
walk | -0.006 |
street | -0.005 |
garage | -0.005 |
push | -0.005 |
call | -0.005 |
Token | ! |
Token ID | 0 |
Feature activation | +8.63e-04 |
Pos logit contributions | |
abella | +0.174 |
mal | +0.170 |
od | +0.161 |
ip | +0.161 |
oll | +0.160 |
Neg logit contributions | |
street | -0.101 |
walk | -0.100 |
garage | -0.099 |
yard | -0.094 |
push | -0.094 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -7.35e-03 |
Pos logit contributions | |
abella | +0.002 |
mal | +0.002 |
od | +0.001 |
ip | +0.001 |
oll | +0.001 |
Neg logit contributions | |
street | -0.001 |
garage | -0.001 |
walk | -0.001 |
push | -0.001 |
scream | -0.001 |
Token | Once |
Token ID | 7454 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
street | +0.009 |
garage | +0.009 |
walk | +0.009 |
yard | +0.009 |
shout | +0.009 |
Neg logit contributions | |
abella | -0.014 |
mal | -0.013 |
od | -0.013 |
oll | -0.013 |
ip | -0.013 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
mal | +0.184 |
ils | +0.178 |
abella | +0.178 |
oll | +0.171 |
od | +0.168 |
Neg logit contributions | |
walk | -0.080 |
garage | -0.079 |
yard | -0.076 |
street | -0.075 |
push | -0.072 |
Token | a |
Token ID | 257 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
abella | +0.025 |
mal | +0.025 |
od | +0.024 |
oll | +0.023 |
ip | +0.023 |
Neg logit contributions | |
street | -0.018 |
garage | -0.018 |
walk | -0.017 |
push | -0.017 |
yard | -0.017 |
Token | time |
Token ID | 640 |
Feature activation | -9.08e-03 |
Pos logit contributions | |
abella | +0.050 |
oll | +0.048 |
ip | +0.048 |
mal | +0.047 |
od | +0.046 |
Neg logit contributions | |
street | -0.042 |
garage | -0.041 |
shed | -0.039 |
calling | -0.039 |
shout | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +3.35e-03 |
Pos logit contributions | |
street | +0.009 |
garage | +0.008 |
walk | +0.008 |
distance | +0.007 |
yard | +0.007 |
Neg logit contributions | |
abella | -0.020 |
mal | -0.020 |
od | -0.020 |
ip | -0.019 |
oll | -0.019 |
Token | there |
Token ID | 612 |
Feature activation | -5.38e-04 |
Pos logit contributions | |
abella | +0.006 |
mal | +0.005 |
od | +0.005 |
ip | +0.005 |
oll | +0.005 |
Neg logit contributions | |
street | -0.005 |
garage | -0.005 |
walk | -0.005 |
push | -0.005 |
yard | -0.005 |
Token | w |
Token ID | 266 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
street | +0.017 |
walk | +0.017 |
garage | +0.016 |
push | +0.016 |
shout | +0.016 |
Neg logit contributions | |
abella | -0.026 |
mal | -0.026 |
od | -0.025 |
ip | -0.024 |
oll | -0.024 |
Token | agged |
Token ID | 14655 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
abella | +0.046 |
mal | +0.044 |
od | +0.042 |
ip | +0.041 |
oll | +0.041 |
Neg logit contributions | |
street | -0.051 |
walk | -0.050 |
garage | -0.050 |
push | -0.049 |
yard | -0.048 |
Token | its |
Token ID | 663 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
street | +0.042 |
garage | +0.041 |
walk | +0.041 |
yard | +0.039 |
push | +0.039 |
Neg logit contributions | |
abella | -0.036 |
mal | -0.034 |
od | -0.033 |
ip | -0.032 |
oll | -0.032 |
Token | tail |
Token ID | 7894 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
mal | +0.054 |
ret | +0.051 |
abella | +0.051 |
ip | +0.051 |
od | +0.051 |
Neg logit contributions | |
garage | -0.034 |
street | -0.033 |
walk | -0.032 |
calling | -0.031 |
shout | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
garage | +0.019 |
street | +0.019 |
yard | +0.016 |
driveway | +0.016 |
grill | +0.016 |
Neg logit contributions | |
ip | -0.024 |
od | -0.024 |
oll | -0.024 |
abella | -0.023 |
mal | -0.023 |
Token | l |
Token ID | 300 |
Feature activation | +6.99e-02 |
Pos logit contributions | |
abella | +0.024 |
mal | +0.024 |
od | +0.024 |
oll | +0.024 |
ret | +0.024 |
Neg logit contributions | |
walk | -0.014 |
street | -0.014 |
shout | -0.013 |
scream | -0.013 |
push | -0.013 |
Token | icked |
Token ID | 9484 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
abella | +0.057 |
mal | +0.053 |
od | +0.047 |
ip | +0.045 |
ret | +0.041 |
Neg logit contributions | |
walk | -0.159 |
push | -0.158 |
garage | -0.157 |
shout | -0.157 |
street | -0.157 |
Token | her |
Token ID | 607 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
abella | +0.039 |
mal | +0.039 |
oll | +0.036 |
od | +0.035 |
cast | +0.035 |
Neg logit contributions | |
push | -0.024 |
shout | -0.024 |
walk | -0.023 |
scream | -0.023 |
pull | -0.023 |
Token | hand |
Token ID | 1021 |
Feature activation | +5.22e-03 |
Pos logit contributions | |
abella | +0.062 |
mal | +0.060 |
od | +0.058 |
ip | +0.058 |
oll | +0.058 |
Neg logit contributions | |
street | -0.029 |
garage | -0.028 |
walk | -0.028 |
yard | -0.026 |
push | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e-03 |
Pos logit contributions | |
abella | +0.010 |
mal | +0.009 |
od | +0.009 |
ip | +0.009 |
oll | +0.009 |
Neg logit contributions | |
street | -0.007 |
garage | -0.007 |
walk | -0.007 |
yard | -0.006 |
push | -0.006 |
Token | The |
Token ID | 383 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
mal | +0.002 |
abella | +0.002 |
od | +0.002 |
ils | +0.002 |
oll | +0.002 |
Neg logit contributions | |
street | -0.001 |
walk | -0.001 |
garage | -0.001 |
shout | -0.001 |
yard | -0.001 |
Token | be |
Token ID | 307 |
Feature activation | -6.80e-03 |
Pos logit contributions | |
walk | +0.004 |
street | +0.004 |
call | +0.004 |
shout | +0.004 |
push | +0.004 |
Neg logit contributions | |
mal | -0.006 |
abella | -0.006 |
ip | -0.006 |
oll | -0.006 |
od | -0.006 |
Token | so |
Token ID | 523 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
street | +0.009 |
garage | +0.008 |
walk | +0.008 |
push | +0.008 |
yard | +0.008 |
Neg logit contributions | |
abella | -0.013 |
mal | -0.013 |
od | -0.012 |
ip | -0.012 |
oll | -0.012 |
Token | intelligent |
Token ID | 12661 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
abella | +0.038 |
mal | +0.038 |
od | +0.036 |
ip | +0.036 |
oll | +0.035 |
Neg logit contributions | |
walk | -0.022 |
street | -0.021 |
urge | -0.020 |
shout | -0.020 |
scream | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +2.63e-03 |
Pos logit contributions | |
street | +0.024 |
garage | +0.024 |
walk | +0.023 |
yard | +0.022 |
push | +0.022 |
Neg logit contributions | |
abella | -0.032 |
mal | -0.031 |
od | -0.030 |
oll | -0.030 |
ip | -0.030 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.37e-04 |
Pos logit contributions | |
abella | +0.005 |
mal | +0.005 |
od | +0.005 |
oll | +0.005 |
ip | +0.005 |
Neg logit contributions | |
street | -0.003 |
garage | -0.003 |
walk | -0.003 |
yard | -0.003 |
shout | -0.003 |
Token | Once |
Token ID | 7454 |
Feature activation | +8.56e-02 |
Pos logit contributions | |
street | +0.001 |
garage | +0.001 |
walk | +0.001 |
yard | +0.001 |
shout | +0.000 |
Neg logit contributions | |
abella | -0.001 |
mal | -0.001 |
od | -0.001 |
oll | -0.001 |
ip | -0.001 |
Token | there |
Token ID | 612 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
mal | +0.196 |
ils | +0.194 |
abella | +0.191 |
oll | +0.184 |
od | +0.180 |
Neg logit contributions | |
garage | -0.075 |
walk | -0.074 |
yard | -0.072 |
street | -0.070 |
push | -0.065 |
Token | was |
Token ID | 373 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
abella | +0.030 |
mal | +0.030 |
od | +0.028 |
ip | +0.028 |
oll | +0.028 |
Neg logit contributions | |
street | -0.023 |
garage | -0.023 |
walk | -0.023 |
push | -0.021 |
yard | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | +9.55e-03 |
Pos logit contributions | |
abella | +0.020 |
oll | +0.019 |
mal | +0.019 |
od | +0.019 |
ip | +0.019 |
Neg logit contributions | |
street | -0.017 |
garage | -0.016 |
distance | -0.016 |
walk | -0.016 |
push | -0.015 |
Token | fit |
Token ID | 4197 |
Feature activation | +3.16e-03 |
Pos logit contributions | |
abella | +0.019 |
mal | +0.019 |
ip | +0.019 |
od | +0.018 |
oll | +0.018 |
Neg logit contributions | |
street | -0.010 |
garage | -0.010 |
walk | -0.010 |
shout | -0.010 |
push | -0.009 |
Token | man |
Token ID | 582 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
mal | +0.006 |
abella | +0.006 |
ip | +0.005 |
oll | +0.005 |
od | +0.005 |
Neg logit contributions | |
street | -0.004 |
walk | -0.004 |
garage | -0.004 |
shout | -0.004 |
scream | -0.004 |
Token | about |
Token ID | 546 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
abella | +0.042 |
mal | +0.041 |
od | +0.039 |
ip | +0.039 |
oll | +0.039 |
Neg logit contributions | |
street | -0.023 |
garage | -0.022 |
walk | -0.022 |
push | -0.021 |
yard | -0.021 |
Token | her |
Token ID | 607 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
abella | +0.058 |
mal | +0.055 |
oll | +0.054 |
ip | +0.054 |
ret | +0.049 |
Neg logit contributions | |
walk | -0.039 |
calling | -0.039 |
street | -0.038 |
garage | -0.038 |
yard | -0.037 |
Token | adventures |
Token ID | 17545 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
abella | +0.051 |
mal | +0.047 |
ip | +0.044 |
oll | +0.044 |
od | +0.043 |
Neg logit contributions | |
street | -0.041 |
walk | -0.041 |
garage | -0.040 |
yard | -0.040 |
call | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +2.24e-03 |
Pos logit contributions | |
street | +0.016 |
walk | +0.016 |
garage | +0.016 |
yard | +0.015 |
push | +0.015 |
Neg logit contributions | |
abella | -0.034 |
mal | -0.033 |
ip | -0.031 |
oll | -0.031 |
od | -0.031 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.97e-03 |
Pos logit contributions | |
ils | +0.006 |
abella | +0.006 |
mal | +0.006 |
ls | +0.005 |
oll | +0.005 |
Neg logit contributions | |
yard | -0.002 |
garage | -0.001 |
street | -0.001 |
walk | -0.001 |
scream | -0.001 |
Token | Once |
Token ID | 7454 |
Feature activation | +9.11e-02 |
Pos logit contributions | |
abella | +0.012 |
mal | +0.012 |
od | +0.011 |
ip | +0.011 |
oll | +0.011 |
Neg logit contributions | |
street | -0.007 |
walk | -0.007 |
garage | -0.007 |
yard | -0.007 |
shout | -0.007 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
mal | +0.194 |
abella | +0.191 |
ils | +0.186 |
od | +0.180 |
oll | +0.179 |
Neg logit contributions | |
walk | -0.096 |
garage | -0.094 |
street | -0.091 |
yard | -0.091 |
push | -0.086 |
Token | a |
Token ID | 257 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
abella | +0.019 |
mal | +0.019 |
od | +0.018 |
oll | +0.018 |
ip | +0.017 |
Neg logit contributions | |
street | -0.016 |
garage | -0.015 |
walk | -0.015 |
push | -0.015 |
shout | -0.015 |
Token | time |
Token ID | 640 |
Feature activation | -6.20e-03 |
Pos logit contributions | |
abella | +0.052 |
oll | +0.052 |
ip | +0.051 |
mal | +0.050 |
od | +0.049 |
Neg logit contributions | |
street | -0.045 |
garage | -0.044 |
calling | -0.043 |
shed | -0.042 |
shout | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | +9.95e-03 |
Pos logit contributions | |
street | +0.006 |
garage | +0.006 |
walk | +0.006 |
distance | +0.006 |
grill | +0.005 |
Neg logit contributions | |
abella | -0.013 |
mal | -0.013 |
od | -0.013 |
ip | -0.013 |
oll | -0.013 |
Token | there |
Token ID | 612 |
Feature activation | +1.76e-03 |
Pos logit contributions | |
abella | +0.017 |
mal | +0.016 |
od | +0.016 |
ip | +0.016 |
oll | +0.016 |
Neg logit contributions | |
street | -0.014 |
walk | -0.014 |
garage | -0.014 |
push | -0.014 |
yard | -0.014 |
Token | could |
Token ID | 714 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
abella | +0.058 |
mal | +0.058 |
od | +0.057 |
ip | +0.055 |
oll | +0.054 |
Neg logit contributions | |
walk | -0.044 |
street | -0.042 |
call | -0.041 |
garage | -0.041 |
scream | -0.041 |
Token | do |
Token ID | 466 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
abella | +0.022 |
mal | +0.021 |
od | +0.020 |
oll | +0.020 |
ip | +0.019 |
Neg logit contributions | |
street | -0.029 |
garage | -0.028 |
grill | -0.027 |
yard | -0.027 |
walk | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | -9.96e-03 |
Pos logit contributions | |
abella | +0.068 |
mal | +0.068 |
oll | +0.066 |
ip | +0.066 |
od | +0.064 |
Neg logit contributions | |
walk | -0.066 |
street | -0.063 |
garage | -0.062 |
push | -0.061 |
scream | -0.060 |
Token | cry |
Token ID | 3960 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
street | +0.015 |
garage | +0.015 |
yard | +0.014 |
grill | +0.014 |
walk | +0.014 |
Neg logit contributions | |
abella | -0.016 |
mal | -0.015 |
oll | -0.014 |
ip | -0.014 |
od | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -6.27e-05 |
Pos logit contributions | |
street | +0.027 |
walk | +0.027 |
garage | +0.027 |
push | +0.026 |
scream | +0.025 |
Neg logit contributions | |
abella | -0.048 |
mal | -0.047 |
od | -0.044 |
ip | -0.044 |
oll | -0.044 |
Token | |
Token ID | 220 |
Feature activation | +7.00e-02 |
Pos logit contributions | |
street | +0.000 |
garage | +0.000 |
walk | +0.000 |
push | +0.000 |
scream | +0.000 |
Neg logit contributions | |
abella | -0.000 |
mal | -0.000 |
od | -0.000 |
ip | -0.000 |
oll | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +5.42e-03 |
Pos logit contributions | |
abella | +0.121 |
mal | +0.115 |
od | +0.113 |
ip | +0.112 |
oll | +0.111 |
Neg logit contributions | |
street | -0.101 |
walk | -0.100 |
garage | -0.099 |
push | -0.095 |
distance | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | -1.82e-03 |
Pos logit contributions | |
abella | +0.010 |
mal | +0.010 |
od | +0.009 |
ip | +0.009 |
oll | +0.009 |
Neg logit contributions | |
street | -0.007 |
garage | -0.007 |
walk | -0.007 |
push | -0.007 |
yard | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
street | +0.003 |
garage | +0.003 |
walk | +0.003 |
push | +0.003 |
yard | +0.003 |
Neg logit contributions | |
abella | -0.003 |
mal | -0.003 |
od | -0.003 |
ip | -0.003 |
oll | -0.003 |
Token | next |
Token ID | 1306 |
Feature activation | -6.81e-03 |
Pos logit contributions | |
street | +0.015 |
push | +0.015 |
shout | +0.015 |
garage | +0.015 |
walk | +0.015 |
Neg logit contributions | |
abella | -0.021 |
mal | -0.021 |
oll | -0.021 |
unic | -0.020 |
ip | -0.020 |
Token | day |
Token ID | 1110 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
street | +0.008 |
garage | +0.008 |
walk | +0.008 |
yard | +0.007 |
shout | +0.007 |
Neg logit contributions | |
abella | -0.014 |
mal | -0.013 |
ip | -0.013 |
od | -0.013 |
oll | -0.013 |
Token | They |
Token ID | 1119 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
abella | +0.003 |
mal | +0.003 |
ip | +0.003 |
od | +0.003 |
oll | +0.003 |
Neg logit contributions | |
walk | -0.002 |
street | -0.002 |
garage | -0.002 |
push | -0.002 |
scream | -0.002 |
Token | were |
Token ID | 547 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
walk | +0.034 |
street | +0.033 |
garage | +0.032 |
shout | +0.032 |
push | +0.032 |
Neg logit contributions | |
abella | -0.053 |
mal | -0.053 |
oll | -0.051 |
od | -0.051 |
ip | -0.051 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
street | +0.040 |
walk | +0.039 |
garage | +0.038 |
yard | +0.038 |
scream | +0.038 |
Neg logit contributions | |
abella | -0.062 |
mal | -0.061 |
ip | -0.058 |
od | -0.058 |
oll | -0.057 |
Token | that |
Token ID | 326 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
street | +0.018 |
walk | +0.018 |
garage | +0.018 |
push | +0.017 |
yard | +0.017 |
Neg logit contributions | |
abella | -0.028 |
mal | -0.028 |
od | -0.026 |
ip | -0.026 |
oll | -0.026 |
Token | they |
Token ID | 484 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
abella | +0.059 |
mal | +0.058 |
oll | +0.055 |
ip | +0.055 |
od | +0.055 |
Neg logit contributions | |
street | -0.054 |
walk | -0.054 |
garage | -0.053 |
push | -0.051 |
scream | -0.050 |
Token | couldn |
Token ID | 3521 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
street | +0.018 |
walk | +0.018 |
garage | +0.017 |
scream | +0.017 |
shout | +0.017 |
Neg logit contributions | |
abella | -0.020 |
mal | -0.020 |
od | -0.019 |
ip | -0.019 |
oll | -0.019 |
Token | 't |
Token ID | 470 |
Feature activation | +7.93e-03 |
Pos logit contributions | |
abella | +0.079 |
mal | +0.077 |
oll | +0.074 |
ils | +0.073 |
ip | +0.073 |
Neg logit contributions | |
street | -0.050 |
walk | -0.050 |
garage | -0.049 |
push | -0.048 |
shout | -0.048 |
Token | see |
Token ID | 766 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
abella | +0.019 |
mal | +0.019 |
od | +0.018 |
ip | +0.018 |
oll | +0.018 |
Neg logit contributions | |
street | -0.006 |
walk | -0.006 |
garage | -0.006 |
push | -0.005 |
yard | -0.005 |
Token | it |
Token ID | 340 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
abella | +0.039 |
mal | +0.038 |
oll | +0.037 |
ip | +0.037 |
od | +0.035 |
Neg logit contributions | |
street | -0.028 |
walk | -0.028 |
garage | -0.027 |
push | -0.027 |
calling | -0.027 |
Token | anymore |
Token ID | 7471 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
abella | +0.041 |
mal | +0.040 |
od | +0.038 |
ip | +0.037 |
oll | +0.037 |
Neg logit contributions | |
street | -0.059 |
garage | -0.058 |
walk | -0.057 |
yard | -0.056 |
push | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +3.81e-03 |
Pos logit contributions | |
abella | +0.038 |
mal | +0.038 |
ip | +0.036 |
oll | +0.036 |
od | +0.035 |
Neg logit contributions | |
street | -0.019 |
walk | -0.019 |
garage | -0.019 |
yard | -0.018 |
scream | -0.018 |
Token | " |
Token ID | 1 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
abella | +0.010 |
mal | +0.009 |
od | +0.009 |
ls | +0.008 |
ip | +0.008 |
Neg logit contributions | |
urge | -0.016 |
street | -0.015 |
spine | -0.015 |
call | -0.015 |
scream | -0.015 |
Token | Maybe |
Token ID | 13300 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
abella | +0.035 |
mal | +0.034 |
od | +0.033 |
ip | +0.033 |
oll | +0.033 |
Neg logit contributions | |
street | -0.027 |
walk | -0.026 |
garage | -0.026 |
push | -0.024 |
yard | -0.024 |
Token | it |
Token ID | 340 |
Feature activation | +5.41e-03 |
Pos logit contributions | |
walk | +0.057 |
street | +0.056 |
garage | +0.056 |
push | +0.054 |
call | +0.053 |
Neg logit contributions | |
abella | -0.052 |
mal | -0.051 |
ip | -0.048 |
oll | -0.048 |
od | -0.048 |
Token | was |
Token ID | 373 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
abella | +0.009 |
mal | +0.009 |
od | +0.009 |
ip | +0.009 |
oll | +0.009 |
Neg logit contributions | |
street | -0.008 |
walk | -0.008 |
garage | -0.008 |
push | -0.007 |
yard | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -9.12e-03 |
Pos logit contributions | |
street | +0.043 |
walk | +0.042 |
garage | +0.042 |
push | +0.040 |
yard | +0.040 |
Neg logit contributions | |
abella | -0.056 |
mal | -0.055 |
od | -0.053 |
ip | -0.053 |
oll | -0.052 |
Token | princess |
Token ID | 21752 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
street | +0.014 |
walk | +0.014 |
garage | +0.014 |
push | +0.014 |
shout | +0.013 |
Neg logit contributions | |
abella | -0.014 |
mal | -0.014 |
od | -0.013 |
ip | -0.013 |
oll | -0.013 |
Token | or |
Token ID | 393 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
abella | +0.043 |
mal | +0.041 |
od | +0.039 |
ip | +0.039 |
oll | +0.038 |
Neg logit contributions | |
street | -0.022 |
walk | -0.022 |
garage | -0.021 |
call | -0.021 |
push | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
abella | +0.062 |
mal | +0.060 |
ip | +0.060 |
od | +0.059 |
oll | +0.058 |
Neg logit contributions | |
walk | -0.032 |
street | -0.032 |
garage | -0.031 |
push | -0.030 |
call | -0.030 |
Token | fairy |
Token ID | 25607 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
abella | +0.030 |
mal | +0.029 |
od | +0.028 |
ip | +0.027 |
oll | +0.027 |
Neg logit contributions | |
street | -0.031 |
garage | -0.030 |
walk | -0.030 |
yard | -0.029 |
push | -0.029 |
Token | ," |
Token ID | 553 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
abella | +0.077 |
mal | +0.075 |
od | +0.072 |
ip | +0.071 |
oll | +0.071 |
Neg logit contributions | |
street | -0.047 |
walk | -0.046 |
garage | -0.045 |
push | -0.043 |
shout | -0.043 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
abella | +0.053 |
mal | +0.052 |
od | +0.049 |
ip | +0.049 |
oll | +0.048 |
Neg logit contributions | |
street | -0.045 |
walk | -0.044 |
garage | -0.044 |
push | -0.042 |
shout | -0.042 |
Token | spark |
Token ID | 9009 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
street | +0.041 |
walk | +0.041 |
garage | +0.040 |
push | +0.038 |
yard | +0.038 |
Neg logit contributions | |
abella | -0.064 |
mal | -0.062 |
ip | -0.059 |
oll | -0.059 |
od | -0.059 |
Token | le |
Token ID | 293 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
abella | +0.066 |
mal | +0.057 |
years | +0.056 |
oll | +0.055 |
unic | +0.054 |
Neg logit contributions | |
walk | -0.046 |
push | -0.044 |
while | -0.043 |
call | -0.042 |
pull | -0.040 |
Token | ." |
Token ID | 526 |
Feature activation | +9.07e-03 |
Pos logit contributions | |
walk | +0.013 |
street | +0.013 |
grill | +0.012 |
push | +0.012 |
call | +0.012 |
Neg logit contributions | |
abella | -0.028 |
mal | -0.027 |
oll | -0.026 |
ip | -0.026 |
ret | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +8.59e-03 |
Pos logit contributions | |
abella | +0.016 |
mal | +0.016 |
ip | +0.015 |
od | +0.015 |
oll | +0.015 |
Neg logit contributions | |
street | -0.013 |
walk | -0.012 |
garage | -0.012 |
push | -0.012 |
scream | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +2.01e-03 |
Pos logit contributions | |
abella | +0.017 |
mal | +0.017 |
od | +0.016 |
oll | +0.016 |
ip | +0.016 |
Neg logit contributions | |
street | -0.010 |
garage | -0.010 |
walk | -0.010 |
yard | -0.010 |
push | -0.010 |
Token | L |
Token ID | 43 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
abella | +0.003 |
mal | +0.003 |
od | +0.002 |
ip | +0.002 |
oll | +0.002 |
Neg logit contributions | |
street | -0.004 |
garage | -0.004 |
walk | -0.004 |
push | -0.003 |
yard | -0.003 |
Token | ily |
Token ID | 813 |
Feature activation | +5.45e-03 |
Pos logit contributions | |
abella | +0.038 |
mal | +0.036 |
oll | +0.034 |
od | +0.034 |
ip | +0.032 |
Neg logit contributions | |
street | -0.108 |
walk | -0.106 |
garage | -0.104 |
distance | -0.104 |
shout | -0.103 |
Token | and |
Token ID | 290 |
Feature activation | +2.18e-03 |
Pos logit contributions | |
abella | +0.011 |
mal | +0.010 |
od | +0.010 |
ip | +0.010 |
oll | +0.010 |
Neg logit contributions | |
street | -0.007 |
distance | -0.006 |
garage | -0.006 |
ditch | -0.006 |
spine | -0.006 |
Token | Max |
Token ID | 5436 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
abella | +0.003 |
mal | +0.003 |
ip | +0.003 |
od | +0.003 |
oll | +0.003 |
Neg logit contributions | |
distance | -0.003 |
street | -0.003 |
spine | -0.003 |
ditch | -0.003 |
corner | -0.003 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
mal | +0.037 |
abella | +0.036 |
od | +0.035 |
ils | +0.034 |
oll | +0.033 |
Neg logit contributions | |
street | -0.031 |
garage | -0.030 |
distance | -0.028 |
yard | -0.028 |
calling | -0.028 |
Token | at |
Token ID | 379 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
abella | +0.023 |
mal | +0.023 |
od | +0.021 |
ip | +0.021 |
oll | +0.021 |
Neg logit contributions | |
street | -0.026 |
walk | -0.025 |
garage | -0.025 |
push | -0.024 |
yard | -0.024 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
mal | +0.004 |
abella | +0.004 |
od | +0.004 |
oll | +0.004 |
ret | +0.004 |
Neg logit contributions | |
street | -0.002 |
garage | -0.002 |
walk | -0.002 |
yard | -0.002 |
shout | -0.001 |
Token | felt |
Token ID | 2936 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
abella | +0.030 |
mal | +0.030 |
od | +0.028 |
ip | +0.028 |
oll | +0.028 |
Neg logit contributions | |
street | -0.024 |
garage | -0.023 |
walk | -0.023 |
push | -0.022 |
yard | -0.022 |
Token | proud |
Token ID | 6613 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
abella | +0.047 |
mal | +0.046 |
od | +0.044 |
ip | +0.043 |
oll | +0.043 |
Neg logit contributions | |
street | -0.039 |
walk | -0.038 |
garage | -0.038 |
push | -0.036 |
shout | -0.036 |
Token | that |
Token ID | 326 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
street | +0.020 |
walk | +0.020 |
garage | +0.020 |
push | +0.019 |
yard | +0.019 |
Neg logit contributions | |
abella | -0.031 |
mal | -0.031 |
od | -0.029 |
ip | -0.029 |
oll | -0.029 |
Token | she |
Token ID | 673 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
abella | +0.026 |
mal | +0.025 |
od | +0.024 |
ip | +0.024 |
oll | +0.024 |
Neg logit contributions | |
street | -0.025 |
garage | -0.024 |
walk | -0.024 |
push | -0.023 |
yard | -0.023 |
Token | could |
Token ID | 714 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
abella | +0.028 |
mal | +0.027 |
ip | +0.026 |
od | +0.026 |
oll | +0.026 |
Neg logit contributions | |
street | -0.023 |
walk | -0.023 |
garage | -0.022 |
call | -0.022 |
push | -0.022 |
Token | suggest |
Token ID | 1950 |
Feature activation | +4.58e-03 |
Pos logit contributions | |
abella | +0.023 |
od | +0.022 |
mal | +0.022 |
oll | +0.021 |
ip | +0.021 |
Neg logit contributions | |
street | -0.012 |
garage | -0.011 |
distance | -0.011 |
spine | -0.010 |
yard | -0.010 |
Token | something |
Token ID | 1223 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
abella | +0.007 |
mal | +0.007 |
od | +0.007 |
ip | +0.007 |
oll | +0.006 |
Neg logit contributions | |
street | -0.007 |
walk | -0.007 |
garage | -0.007 |
push | -0.007 |
yard | -0.007 |
Token | helpful |
Token ID | 7613 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
abella | +0.073 |
mal | +0.072 |
oll | +0.068 |
ip | +0.067 |
ret | +0.066 |
Neg logit contributions | |
walk | -0.055 |
street | -0.053 |
push | -0.052 |
garage | -0.052 |
call | -0.052 |
Token | to |
Token ID | 284 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
abella | +0.047 |
od | +0.046 |
ils | +0.046 |
mal | +0.045 |
ip | +0.044 |
Neg logit contributions | |
street | -0.045 |
garage | -0.044 |
ditch | -0.042 |
walk | -0.041 |
yard | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
abella | +0.035 |
mal | +0.035 |
ip | +0.033 |
oll | +0.033 |
od | +0.033 |
Neg logit contributions | |
street | -0.029 |
walk | -0.029 |
garage | -0.029 |
push | -0.028 |
shout | -0.028 |
Token | He |
Token ID | 679 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
street | +0.005 |
walk | +0.005 |
garage | +0.005 |
push | +0.005 |
yard | +0.005 |
Neg logit contributions | |
abella | -0.008 |
mal | -0.008 |
od | -0.008 |
oll | -0.008 |
ip | -0.008 |
Token | sees |
Token ID | 7224 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
abella | +0.032 |
mal | +0.031 |
od | +0.030 |
ip | +0.030 |
oll | +0.029 |
Neg logit contributions | |
street | -0.018 |
walk | -0.018 |
garage | -0.017 |
push | -0.017 |
scream | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
mal | +0.034 |
abella | +0.033 |
ip | +0.033 |
od | +0.032 |
ret | +0.032 |
Neg logit contributions | |
street | -0.024 |
scream | -0.024 |
garage | -0.023 |
walk | -0.023 |
push | -0.023 |
Token | cut |
Token ID | 2005 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
walk | +0.023 |
street | +0.023 |
shout | +0.022 |
push | +0.021 |
scream | +0.021 |
Neg logit contributions | |
mal | -0.029 |
abella | -0.027 |
od | -0.027 |
unic | -0.027 |
oll | -0.027 |
Token | on |
Token ID | 319 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
abella | +0.048 |
mal | +0.047 |
od | +0.045 |
oll | +0.045 |
ip | +0.045 |
Neg logit contributions | |
street | -0.045 |
walk | -0.044 |
garage | -0.043 |
yard | -0.041 |
shout | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
abella | +0.029 |
mal | +0.028 |
ip | +0.027 |
oll | +0.027 |
od | +0.027 |
Neg logit contributions | |
street | -0.022 |
walk | -0.022 |
garage | -0.022 |
push | -0.021 |
scream | -0.021 |
Token | finger |
Token ID | 7660 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
mal | +0.082 |
oll | +0.080 |
od | +0.077 |
ip | +0.075 |
ret | +0.072 |
Neg logit contributions | |
street | -0.050 |
walk | -0.050 |
calling | -0.048 |
proceed | -0.047 |
shout | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
mal | +0.078 |
abella | +0.077 |
oll | +0.077 |
od | +0.076 |
ip | +0.076 |
Neg logit contributions | |
street | -0.076 |
walk | -0.073 |
garage | -0.072 |
distance | -0.068 |
yard | -0.068 |
Token | He |
Token ID | 679 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
street | +0.010 |
garage | +0.010 |
walk | +0.009 |
scream | +0.009 |
yard | +0.009 |
Neg logit contributions | |
abella | -0.022 |
mal | -0.022 |
od | -0.021 |
ils | -0.021 |
ip | -0.021 |
Token | hugs |
Token ID | 40657 |
Feature activation | +5.73e-03 |
Pos logit contributions | |
abella | +0.022 |
mal | +0.021 |
od | +0.021 |
ip | +0.020 |
oll | +0.020 |
Neg logit contributions | |
walk | -0.013 |
push | -0.013 |
scream | -0.013 |
street | -0.013 |
garage | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
mal | +0.011 |
abella | +0.011 |
ip | +0.010 |
od | +0.010 |
ret | +0.010 |
Neg logit contributions | |
walk | -0.008 |
push | -0.007 |
scream | -0.007 |
street | -0.007 |
shout | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -7.67e-03 |
Pos logit contributions | |
garage | +0.008 |
street | +0.008 |
grill | +0.007 |
walk | +0.007 |
pray | +0.007 |
Neg logit contributions | |
od | -0.010 |
oll | -0.010 |
mal | -0.009 |
ip | -0.009 |
abella | -0.009 |
Token | He |
Token ID | 679 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
street | +0.008 |
garage | +0.008 |
yard | +0.008 |
walk | +0.007 |
shout | +0.007 |
Neg logit contributions | |
mal | -0.016 |
abella | -0.016 |
od | -0.016 |
ils | -0.016 |
ret | -0.015 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
abella | +0.024 |
mal | +0.024 |
od | +0.023 |
ret | +0.023 |
ip | +0.023 |
Neg logit contributions | |
walk | -0.013 |
street | -0.013 |
call | -0.012 |
scream | -0.012 |
push | -0.012 |
Token | very |
Token ID | 845 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
abella | +0.030 |
mal | +0.029 |
od | +0.028 |
oll | +0.027 |
ip | +0.027 |
Neg logit contributions | |
street | -0.029 |
walk | -0.029 |
garage | -0.028 |
shout | -0.028 |
push | -0.028 |
Token | proud |
Token ID | 6613 |
Feature activation | -7.77e-03 |
Pos logit contributions | |
abella | +0.054 |
mal | +0.053 |
od | +0.051 |
ip | +0.050 |
oll | +0.049 |
Neg logit contributions | |
street | -0.034 |
walk | -0.033 |
shout | -0.032 |
scream | -0.032 |
call | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -2.80e-03 |
Pos logit contributions | |
street | +0.010 |
garage | +0.010 |
walk | +0.010 |
push | +0.009 |
shout | +0.009 |
Neg logit contributions | |
abella | -0.015 |
mal | -0.014 |
od | -0.014 |
oll | -0.014 |
ip | -0.014 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
street | +0.003 |
garage | +0.003 |
yard | +0.003 |
shout | +0.002 |
scream | +0.002 |
Neg logit contributions | |
mal | -0.006 |
abella | -0.006 |
ils | -0.006 |
od | -0.006 |
ret | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
abella | +0.032 |
mal | +0.032 |
ip | +0.030 |
od | +0.030 |
oll | +0.030 |
Neg logit contributions | |
garage | -0.018 |
street | -0.018 |
walk | -0.016 |
spine | -0.016 |
grill | -0.016 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
od | +0.021 |
ret | +0.020 |
abella | +0.020 |
mal | +0.019 |
oll | +0.019 |
Neg logit contributions | |
walk | -0.015 |
push | -0.014 |
street | -0.014 |
call | -0.014 |
scream | -0.014 |
Token | realized |
Token ID | 6939 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
mal | +0.040 |
abella | +0.039 |
od | +0.039 |
ip | +0.037 |
oll | +0.036 |
Neg logit contributions | |
street | -0.042 |
garage | -0.042 |
grill | -0.039 |
spine | -0.039 |
ditch | -0.039 |
Token | what |
Token ID | 644 |
Feature activation | +5.26e-02 |
Pos logit contributions | |
mal | +0.081 |
ret | +0.078 |
od | +0.076 |
years | +0.076 |
ils | +0.075 |
Neg logit contributions | |
calling | -0.051 |
walk | -0.050 |
garage | -0.048 |
call | -0.048 |
scream | -0.047 |
Token | happened |
Token ID | 3022 |
Feature activation | -3.56e-03 |
Pos logit contributions | |
abella | +0.007 |
mal | +0.007 |
od | +0.006 |
oll | +0.006 |
ip | +0.006 |
Neg logit contributions | |
street | -0.004 |
garage | -0.004 |
walk | -0.004 |
distance | -0.004 |
yard | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +2.65e-04 |
Pos logit contributions | |
street | +0.005 |
garage | +0.005 |
walk | +0.005 |
push | +0.005 |
shout | +0.005 |
Neg logit contributions | |
abella | -0.006 |
mal | -0.006 |
od | -0.006 |
ip | -0.006 |
oll | -0.005 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
abella | +0.000 |
mal | +0.000 |
od | +0.000 |
ip | +0.000 |
oll | +0.000 |
Neg logit contributions | |
street | -0.000 |
walk | -0.000 |
garage | -0.000 |
push | -0.000 |
yard | -0.000 |
Token | 's |
Token ID | 338 |
Feature activation | +9.44e-03 |
Pos logit contributions | |
abella | +0.022 |
mal | +0.022 |
ip | +0.021 |
od | +0.021 |
ils | +0.021 |
Neg logit contributions | |
street | -0.019 |
garage | -0.018 |
distance | -0.018 |
walk | -0.018 |
push | -0.018 |
Token | tum |
Token ID | 11814 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
abella | +0.013 |
mal | +0.013 |
od | +0.012 |
oll | +0.012 |
ip | +0.012 |
Neg logit contributions | |
street | -0.016 |
garage | -0.016 |
shout | -0.016 |
walk | -0.016 |
distance | -0.015 |
Token | my |
Token ID | 1820 |
Feature activation | -2.94e-03 |
Pos logit contributions | |
abella | +0.027 |
mal | +0.026 |
od | +0.025 |
ils | +0.025 |
ip | +0.025 |
Neg logit contributions | |
street | -0.007 |
garage | -0.006 |
yard | -0.006 |
walk | -0.006 |
distance | -0.005 |
Token | started |
Token ID | 2067 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
street | +0.004 |
garage | +0.004 |
walk | +0.004 |
yard | +0.004 |
shout | +0.004 |
Neg logit contributions | |
abella | -0.005 |
mal | -0.005 |
ip | -0.005 |
od | -0.005 |
oll | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | -7.47e-03 |
Pos logit contributions | |
street | +0.039 |
walk | +0.038 |
garage | +0.038 |
scream | +0.036 |
yard | +0.036 |
Neg logit contributions | |
abella | -0.071 |
mal | -0.069 |
od | -0.066 |
oll | -0.066 |
ip | -0.066 |
Token | hurt |
Token ID | 5938 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
garage | +0.010 |
street | +0.009 |
yard | +0.009 |
spine | +0.009 |
shed | +0.008 |
Neg logit contributions | |
mal | -0.013 |
abella | -0.013 |
od | -0.013 |
unic | -0.012 |
ip | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
street | +0.018 |
garage | +0.018 |
walk | +0.017 |
yard | +0.017 |
push | +0.017 |
Neg logit contributions | |
abella | -0.029 |
mal | -0.028 |
od | -0.027 |
ip | -0.027 |
oll | -0.026 |
Token | lot |
Token ID | 1256 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
abella | +0.034 |
mal | +0.034 |
ip | +0.032 |
od | +0.032 |
oll | +0.032 |
Neg logit contributions | |
street | -0.031 |
garage | -0.031 |
walk | -0.030 |
yard | -0.029 |
shout | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
od | +0.010 |
mal | +0.010 |
oll | +0.009 |
abella | +0.009 |
ils | +0.009 |
Neg logit contributions | |
street | -0.007 |
push | -0.007 |
walk | -0.007 |
shout | -0.006 |
garage | -0.006 |
Token | watch |
Token ID | 2342 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
walk | +0.018 |
garage | +0.018 |
street | +0.017 |
yard | +0.017 |
push | +0.016 |
Neg logit contributions | |
abella | -0.033 |
mal | -0.032 |
l | -0.030 |
oll | -0.030 |
ip | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | -1.73e-03 |
Pos logit contributions | |
abella | +0.027 |
mal | +0.026 |
od | +0.025 |
ip | +0.025 |
oll | +0.025 |
Neg logit contributions | |
street | -0.024 |
garage | -0.024 |
walk | -0.023 |
yard | -0.022 |
shout | -0.022 |
Token | too |
Token ID | 1165 |
Feature activation | -8.55e-03 |
Pos logit contributions | |
street | +0.003 |
garage | +0.003 |
walk | +0.003 |
yard | +0.003 |
grill | +0.003 |
Neg logit contributions | |
abella | -0.002 |
mal | -0.002 |
oll | -0.002 |
ip | -0.002 |
ls | -0.002 |
Token | big |
Token ID | 1263 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
garage | +0.014 |
spine | +0.014 |
street | +0.014 |
driveway | +0.013 |
yard | +0.013 |
Neg logit contributions | |
ls | -0.012 |
abella | -0.011 |
l | -0.011 |
ip | -0.011 |
mal | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +8.79e-03 |
Pos logit contributions | |
street | +0.041 |
garage | +0.040 |
distance | +0.038 |
walk | +0.038 |
yard | +0.037 |
Neg logit contributions | |
ip | -0.035 |
oll | -0.035 |
mal | -0.035 |
abella | -0.035 |
ls | -0.034 |
Token | heavy |
Token ID | 4334 |
Feature activation | -7.56e-03 |
Pos logit contributions | |
od | +0.016 |
abella | +0.016 |
mal | +0.016 |
oll | +0.015 |
ret | +0.015 |
Neg logit contributions | |
street | -0.012 |
walk | -0.012 |
push | -0.012 |
garage | -0.011 |
shout | -0.011 |
Token | for |
Token ID | 329 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
garage | +0.013 |
street | +0.013 |
distance | +0.012 |
yard | +0.012 |
walk | +0.012 |
Neg logit contributions | |
abella | -0.011 |
mal | -0.011 |
ip | -0.011 |
oll | -0.011 |
ls | -0.010 |
Token | them |
Token ID | 606 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
abella | +0.016 |
mal | +0.016 |
od | +0.015 |
ip | +0.015 |
oll | +0.014 |
Neg logit contributions | |
street | -0.022 |
walk | -0.022 |
garage | -0.022 |
push | -0.021 |
shout | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +1.46e-03 |
Pos logit contributions | |
street | +0.025 |
garage | +0.024 |
walk | +0.023 |
yard | +0.022 |
grill | +0.022 |
Neg logit contributions | |
abella | -0.026 |
mal | -0.026 |
ip | -0.025 |
od | -0.025 |
oll | -0.024 |
Token | It |
Token ID | 632 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
abella | +0.003 |
mal | +0.003 |
od | +0.003 |
oll | +0.003 |
ip | +0.003 |
Neg logit contributions | |
street | -0.002 |
garage | -0.002 |
walk | -0.002 |
yard | -0.002 |
push | -0.002 |
Token | knew |
Token ID | 2993 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
street | +0.005 |
garage | +0.005 |
walk | +0.005 |
yard | +0.005 |
distance | +0.005 |
Neg logit contributions | |
abella | -0.007 |
mal | -0.007 |
ip | -0.007 |
od | -0.006 |
oll | -0.006 |
Token | how |
Token ID | 703 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
mal | +0.067 |
abella | +0.064 |
ret | +0.063 |
od | +0.063 |
ip | +0.062 |
Neg logit contributions | |
walk | -0.048 |
calling | -0.048 |
garage | -0.047 |
push | -0.046 |
call | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | -3.16e-03 |
Pos logit contributions | |
abella | +0.107 |
mal | +0.105 |
od | +0.100 |
oll | +0.099 |
ip | +0.098 |
Neg logit contributions | |
street | -0.064 |
walk | -0.064 |
garage | -0.062 |
push | -0.060 |
shout | -0.059 |
Token | measure |
Token ID | 3953 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
spine | +0.003 |
distance | +0.003 |
street | +0.003 |
coat | +0.003 |
garage | +0.003 |
Neg logit contributions | |
od | -0.006 |
mal | -0.006 |
ils | -0.006 |
abella | -0.005 |
ret | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
abella | +0.030 |
mal | +0.029 |
oll | +0.028 |
ip | +0.027 |
od | +0.026 |
Neg logit contributions | |
push | -0.023 |
street | -0.023 |
garage | -0.023 |
walk | -0.023 |
yard | -0.022 |
Token | hill |
Token ID | 12788 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
calling | +0.031 |
shout | +0.030 |
call | +0.029 |
spine | +0.029 |
scream | +0.029 |
Neg logit contributions | |
mal | -0.035 |
od | -0.033 |
unic | -0.032 |
ip | -0.032 |
ls | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -6.64e-03 |
Pos logit contributions | |
street | +0.027 |
walk | +0.026 |
garage | +0.026 |
scream | +0.025 |
shout | +0.025 |
Neg logit contributions | |
abella | -0.037 |
mal | -0.036 |
od | -0.035 |
ip | -0.034 |
oll | -0.034 |
Token | He |
Token ID | 679 |
Feature activation | +4.21e-04 |
Pos logit contributions | |
street | +0.008 |
walk | +0.008 |
garage | +0.008 |
push | +0.008 |
shout | +0.007 |
Neg logit contributions | |
abella | -0.013 |
mal | -0.012 |
od | -0.012 |
ip | -0.012 |
oll | -0.012 |
Token | took |
Token ID | 1718 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
abella | +0.001 |
mal | +0.001 |
od | +0.001 |
ip | +0.001 |
oll | +0.001 |
Neg logit contributions | |
street | -0.000 |
garage | -0.000 |
walk | -0.000 |
yard | -0.000 |
push | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | -9.57e-03 |
Pos logit contributions | |
abella | +0.024 |
mal | +0.023 |
od | +0.022 |
oll | +0.022 |
ip | +0.022 |
Neg logit contributions | |
push | -0.015 |
walk | -0.014 |
shout | -0.014 |
garage | -0.014 |
street | -0.014 |
Token | string |
Token ID | 4731 |
Feature activation | -2.64e-03 |
Pos logit contributions | |
calling | +0.016 |
whine | +0.015 |
spine | +0.015 |
shout | +0.014 |
call | +0.014 |
Neg logit contributions | |
mal | -0.012 |
ip | -0.011 |
penn | -0.011 |
jelly | -0.011 |
ret | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +1.27e-03 |
Pos logit contributions | |
abella | +0.129 |
mal | +0.124 |
od | +0.120 |
ip | +0.119 |
oll | +0.118 |
Neg logit contributions | |
street | -0.098 |
walk | -0.097 |
garage | -0.095 |
push | -0.091 |
yard | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | -6.67e-03 |
Pos logit contributions | |
mal | +0.003 |
abella | +0.002 |
oll | +0.002 |
od | +0.002 |
ip | +0.002 |
Neg logit contributions | |
street | -0.001 |
walk | -0.001 |
garage | -0.001 |
shout | -0.001 |
push | -0.001 |
Token | Todd |
Token ID | 42817 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
street | +0.010 |
walk | +0.009 |
garage | +0.009 |
shout | +0.009 |
push | +0.009 |
Neg logit contributions | |
mal | -0.011 |
abella | -0.011 |
oll | -0.011 |
ip | -0.011 |
od | -0.011 |
Token | ran |
Token ID | 4966 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
abella | +0.050 |
mal | +0.048 |
od | +0.046 |
ip | +0.046 |
oll | +0.045 |
Neg logit contributions | |
street | -0.031 |
garage | -0.030 |
walk | -0.029 |
yard | -0.028 |
push | -0.028 |
Token | as |
Token ID | 355 |
Feature activation | -3.36e-03 |
Pos logit contributions | |
garage | +0.045 |
street | +0.042 |
yard | +0.042 |
grill | +0.041 |
walk | +0.040 |
Neg logit contributions | |
abella | -0.044 |
mal | -0.041 |
ls | -0.038 |
ip | -0.038 |
od | -0.037 |
Token | fast |
Token ID | 3049 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
street | +0.004 |
walk | +0.004 |
garage | +0.004 |
push | +0.004 |
shout | +0.004 |
Neg logit contributions | |
abella | -0.007 |
mal | -0.006 |
od | -0.006 |
oll | -0.006 |
ip | -0.006 |
Token | as |
Token ID | 355 |
Feature activation | -8.80e-03 |
Pos logit contributions | |
garage | +0.026 |
street | +0.025 |
yard | +0.024 |
grill | +0.022 |
shed | +0.022 |
Neg logit contributions | |
abella | -0.038 |
ip | -0.037 |
mal | -0.037 |
oll | -0.035 |
ls | -0.035 |
Token | he |
Token ID | 339 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
street | +0.012 |
walk | +0.012 |
garage | +0.011 |
distance | +0.011 |
push | +0.011 |
Neg logit contributions | |
mal | -0.016 |
abella | -0.016 |
od | -0.015 |
oll | -0.015 |
ils | -0.015 |
Token | could |
Token ID | 714 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
abella | +0.021 |
od | +0.021 |
ils | +0.020 |
mal | +0.020 |
oll | +0.020 |
Neg logit contributions | |
walk | -0.017 |
push | -0.016 |
call | -0.016 |
shout | -0.016 |
scream | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -8.20e-04 |
Pos logit contributions | |
od | +0.024 |
abella | +0.023 |
l | +0.022 |
ls | +0.022 |
ret | +0.022 |
Neg logit contributions | |
garage | -0.023 |
street | -0.023 |
workshop | -0.022 |
grill | -0.021 |
pray | -0.020 |
Token | but |
Token ID | 475 |
Feature activation | +7.36e-03 |
Pos logit contributions | |
garage | +0.001 |
grill | +0.001 |
proceed | +0.001 |
pray | +0.001 |
street | +0.001 |
Neg logit contributions | |
abella | -0.001 |
mal | -0.001 |
cast | -0.001 |
od | -0.001 |
ip | -0.001 |
Token | cow |
Token ID | 9875 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
garage | +0.079 |
spine | +0.079 |
street | +0.078 |
push | +0.078 |
shout | +0.078 |
Neg logit contributions | |
mal | -0.080 |
ils | -0.079 |
abella | -0.079 |
ls | -0.077 |
ip | -0.075 |
Token | boys |
Token ID | 13202 |
Feature activation | -9.26e-03 |
Pos logit contributions | |
street | +0.016 |
garage | +0.016 |
walk | +0.015 |
push | +0.015 |
distance | +0.015 |
Neg logit contributions | |
mal | -0.022 |
abella | -0.021 |
od | -0.021 |
ip | -0.021 |
ils | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -8.82e-03 |
Pos logit contributions | |
street | +0.014 |
garage | +0.013 |
distance | +0.013 |
yard | +0.013 |
urge | +0.012 |
Neg logit contributions | |
mal | -0.015 |
ip | -0.015 |
od | -0.014 |
ils | -0.014 |
abella | -0.014 |
Token | Everyone |
Token ID | 11075 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
walk | +0.014 |
brake | +0.014 |
push | +0.014 |
distance | +0.014 |
coat | +0.013 |
Neg logit contributions | |
abella | -0.013 |
ip | -0.012 |
mal | -0.012 |
lee | -0.012 |
od | -0.012 |
Token | was |
Token ID | 373 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
garage | +0.063 |
street | +0.063 |
walk | +0.060 |
yard | +0.059 |
spine | +0.058 |
Neg logit contributions | |
abella | -0.075 |
mal | -0.073 |
od | -0.071 |
ip | -0.069 |
ils | -0.068 |
Token | busy |
Token ID | 8179 |
Feature activation | -5.25e-02 |
Pos logit contributions | |
garage | +0.040 |
street | +0.040 |
walk | +0.039 |
yard | +0.037 |
spine | +0.036 |
Neg logit contributions | |
abella | -0.053 |
mal | -0.051 |
od | -0.050 |
ip | -0.048 |
oll | -0.048 |
Token | carrying |
Token ID | 6872 |
Feature activation | -2.20e-03 |
Pos logit contributions | |
garage | +0.062 |
street | +0.060 |
walk | +0.057 |
distance | +0.056 |
yard | +0.056 |
Neg logit contributions | |
od | -0.094 |
mal | -0.094 |
abella | -0.093 |
ip | -0.090 |
ils | -0.087 |
Token | cattle |
Token ID | 17025 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
street | +0.003 |
walk | +0.003 |
garage | +0.003 |
push | +0.003 |
shout | +0.003 |
Neg logit contributions | |
abella | -0.004 |
mal | -0.004 |
od | -0.004 |
oll | -0.004 |
ip | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
street | +0.062 |
garage | +0.062 |
distance | +0.059 |
urge | +0.057 |
spine | +0.056 |
Neg logit contributions | |
ip | -0.087 |
mal | -0.087 |
ils | -0.084 |
od | -0.081 |
oll | -0.081 |
Token | hay |
Token ID | 27678 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
garage | +0.024 |
street | +0.023 |
walk | +0.023 |
yard | +0.022 |
spine | +0.021 |
Neg logit contributions | |
mal | -0.037 |
abella | -0.037 |
ip | -0.035 |
od | -0.035 |
unic | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -3.98e-03 |
Pos logit contributions | |
street | +0.045 |
garage | +0.044 |
walk | +0.043 |
distance | +0.042 |
nod | +0.042 |
Neg logit contributions | |
mal | -0.067 |
ip | -0.066 |
ils | -0.064 |
abella | -0.063 |
od | -0.063 |
Token | daddy |
Token ID | 35822 |
Feature activation | -8.04e-03 |
Pos logit contributions | |
street | +0.038 |
shout | +0.038 |
scream | +0.037 |
calling | +0.037 |
walk | +0.037 |
Neg logit contributions | |
mal | -0.043 |
abella | -0.042 |
oll | -0.039 |
ip | -0.038 |
od | -0.037 |
Token | said |
Token ID | 531 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
street | +0.011 |
garage | +0.011 |
walk | +0.010 |
distance | +0.010 |
yard | +0.010 |
Neg logit contributions | |
abella | -0.015 |
mal | -0.015 |
ip | -0.014 |
od | -0.014 |
oll | -0.014 |
Token | " |
Token ID | 366 |
Feature activation | +2.57e-04 |
Pos logit contributions | |
street | +0.038 |
distance | +0.036 |
garage | +0.036 |
walk | +0.036 |
yard | +0.035 |
Neg logit contributions | |
abella | -0.057 |
mal | -0.054 |
ip | -0.053 |
od | -0.052 |
oll | -0.052 |
Token | ok |
Token ID | 482 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
abella | +0.000 |
mal | +0.000 |
oll | +0.000 |
od | +0.000 |
ip | +0.000 |
Neg logit contributions | |
street | -0.000 |
walk | -0.000 |
garage | -0.000 |
push | -0.000 |
distance | -0.000 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
street | +0.107 |
garage | +0.105 |
distance | +0.105 |
yard | +0.102 |
walk | +0.102 |
Neg logit contributions | |
ip | -0.078 |
abella | -0.076 |
mal | -0.075 |
ret | -0.072 |
oll | -0.071 |
Token | Then |
Token ID | 3244 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
walk | +0.014 |
street | +0.014 |
garage | +0.014 |
calling | +0.014 |
distance | +0.014 |
Neg logit contributions | |
ip | -0.016 |
abella | -0.015 |
mal | -0.015 |
od | -0.015 |
oll | -0.015 |
Token | he |
Token ID | 339 |
Feature activation | -3.90e-04 |
Pos logit contributions | |
street | +0.053 |
walk | +0.053 |
garage | +0.052 |
push | +0.050 |
shout | +0.049 |
Neg logit contributions | |
abella | -0.114 |
mal | -0.112 |
od | -0.109 |
oll | -0.108 |
ip | -0.107 |
Token | went |
Token ID | 1816 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
street | +0.000 |
distance | +0.000 |
spine | +0.000 |
grill | +0.000 |
scream | +0.000 |
Neg logit contributions | |
abella | -0.001 |
mal | -0.001 |
ip | -0.001 |
oll | -0.001 |
od | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +3.77e-03 |
Pos logit contributions | |
garage | +0.049 |
street | +0.048 |
spine | +0.048 |
yard | +0.047 |
scream | +0.047 |
Neg logit contributions | |
abella | -0.052 |
mal | -0.050 |
ip | -0.047 |
ls | -0.043 |
oll | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | -6.91e-03 |
Pos logit contributions | |
abella | +0.004 |
mal | +0.004 |
od | +0.003 |
unic | +0.003 |
Qu | +0.003 |
Neg logit contributions | |
spine | -0.007 |
coat | -0.007 |
street | -0.007 |
distance | -0.007 |
yard | -0.006 |
Token | garage |
Token ID | 15591 |
Feature activation | +1.14e-03 |
Pos logit contributions | |
spine | +0.006 |
shout | +0.006 |
scream | +0.006 |
push | +0.006 |
nod | +0.006 |
Neg logit contributions | |
abella | -0.015 |
mal | -0.014 |
ip | -0.013 |
oll | -0.013 |
od | -0.013 |
Token | some |
Token ID | 617 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
abella | +0.029 |
mal | +0.029 |
ip | +0.028 |
ret | +0.028 |
oll | +0.027 |
Neg logit contributions | |
street | -0.027 |
walk | -0.025 |
garage | -0.025 |
yard | -0.025 |
scream | -0.024 |
Token | animals |
Token ID | 4695 |
Feature activation | -6.90e-02 |
Pos logit contributions | |
garage | +0.027 |
street | +0.027 |
walk | +0.026 |
push | +0.025 |
shout | +0.025 |
Neg logit contributions | |
abella | -0.027 |
mal | -0.026 |
oll | -0.025 |
ip | -0.025 |
od | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -8.44e-03 |
Pos logit contributions | |
garage | +0.103 |
street | +0.102 |
distance | +0.096 |
walk | +0.095 |
spine | +0.095 |
Neg logit contributions | |
mal | -0.109 |
od | -0.109 |
ip | -0.105 |
abella | -0.104 |
ls | -0.103 |
Token | She |
Token ID | 1375 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
walk | +0.011 |
street | +0.011 |
garage | +0.011 |
distance | +0.011 |
push | +0.011 |
Neg logit contributions | |
abella | -0.015 |
mal | -0.015 |
ip | -0.014 |
od | -0.014 |
oll | -0.014 |
Token | saw |
Token ID | 2497 |
Feature activation | +7.10e-03 |
Pos logit contributions | |
abella | +0.031 |
mal | +0.030 |
od | +0.029 |
ip | +0.029 |
oll | +0.028 |
Neg logit contributions | |
street | -0.018 |
garage | -0.018 |
walk | -0.017 |
push | -0.017 |
shout | -0.017 |
Token | cows |
Token ID | 22575 |
Feature activation | -6.95e-02 |
Pos logit contributions | |
mal | +0.014 |
abella | +0.014 |
ip | +0.014 |
ret | +0.014 |
oll | +0.014 |
Neg logit contributions | |
street | -0.008 |
walk | -0.008 |
garage | -0.008 |
scream | -0.007 |
yard | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -5.63e-03 |
Pos logit contributions | |
street | +0.081 |
garage | +0.081 |
distance | +0.076 |
spine | +0.076 |
driveway | +0.072 |
Neg logit contributions | |
mal | -0.131 |
od | -0.128 |
ip | -0.127 |
abella | -0.122 |
ls | -0.120 |
Token | sheep |
Token ID | 15900 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
garage | +0.008 |
street | +0.008 |
spine | +0.007 |
walk | +0.007 |
shout | +0.007 |
Neg logit contributions | |
abella | -0.010 |
mal | -0.010 |
ip | -0.009 |
od | -0.009 |
oll | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -8.30e-03 |
Pos logit contributions | |
garage | +0.034 |
street | +0.034 |
distance | +0.032 |
walk | +0.031 |
spine | +0.030 |
Neg logit contributions | |
mal | -0.073 |
abella | -0.072 |
ip | -0.072 |
od | -0.071 |
ils | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | -7.51e-03 |
Pos logit contributions | |
spine | +0.011 |
garage | +0.011 |
street | +0.010 |
shout | +0.010 |
intervene | +0.010 |
Neg logit contributions | |
mal | -0.014 |
ip | -0.014 |
abella | -0.014 |
unic | -0.014 |
od | -0.013 |
Token | chickens |
Token ID | 25972 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
street | +0.010 |
walk | +0.009 |
garage | +0.009 |
push | +0.009 |
yard | +0.009 |
Neg logit contributions | |
abella | -0.014 |
mal | -0.014 |
od | -0.013 |
ip | -0.013 |
oll | -0.013 |
Token | worry |
Token ID | 5490 |
Feature activation | -7.50e-03 |
Pos logit contributions | |
street | +0.029 |
garage | +0.028 |
yard | +0.027 |
distance | +0.027 |
shed | +0.025 |
Neg logit contributions | |
od | -0.061 |
mal | -0.061 |
abella | -0.058 |
ret | -0.057 |
unic | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
street | +0.008 |
garage | +0.008 |
distance | +0.008 |
walk | +0.007 |
yard | +0.007 |
Neg logit contributions | |
abella | -0.015 |
od | -0.015 |
mal | -0.015 |
ip | -0.015 |
ls | -0.015 |
Token | Tim |
Token ID | 5045 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
street | +0.051 |
garage | +0.050 |
walk | +0.049 |
yard | +0.047 |
shout | +0.047 |
Neg logit contributions | |
abella | -0.067 |
mal | -0.066 |
od | -0.063 |
ip | -0.062 |
oll | -0.062 |
Token | my |
Token ID | 1820 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
street | +0.023 |
walk | +0.022 |
garage | +0.021 |
shout | +0.021 |
scream | +0.020 |
Neg logit contributions | |
mal | -0.041 |
abella | -0.040 |
od | -0.039 |
ip | -0.039 |
oll | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
street | +0.023 |
garage | +0.023 |
walk | +0.022 |
distance | +0.022 |
yard | +0.022 |
Neg logit contributions | |
mal | -0.026 |
abella | -0.026 |
od | -0.025 |
ip | -0.025 |
ret | -0.024 |
Token | We |
Token ID | 775 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
garage | +0.018 |
street | +0.018 |
walk | +0.017 |
push | +0.017 |
shout | +0.017 |
Neg logit contributions | |
abella | -0.026 |
mal | -0.025 |
od | -0.025 |
oll | -0.024 |
ils | -0.024 |
Token | can |
Token ID | 460 |
Feature activation | -6.76e-03 |
Pos logit contributions | |
street | +0.097 |
garage | +0.091 |
walk | +0.091 |
distance | +0.089 |
yard | +0.089 |
Neg logit contributions | |
mal | -0.110 |
abella | -0.110 |
od | -0.104 |
ip | -0.104 |
oll | -0.102 |
Token | fix |
Token ID | 4259 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
street | +0.007 |
distance | +0.007 |
garage | +0.006 |
yard | +0.006 |
spine | +0.006 |
Neg logit contributions | |
mal | -0.014 |
abella | -0.014 |
od | -0.013 |
ip | -0.013 |
oll | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
abella | +0.040 |
oll | +0.040 |
mal | +0.036 |
od | +0.035 |
years | +0.035 |
Neg logit contributions | |
push | -0.027 |
grill | -0.026 |
pull | -0.025 |
garage | -0.025 |
walk | -0.024 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.36e-03 |
Pos logit contributions | |
abella | +0.026 |
mal | +0.026 |
od | +0.024 |
ip | +0.024 |
oll | +0.024 |
Neg logit contributions | |
street | -0.024 |
garage | -0.023 |
walk | -0.023 |
shout | -0.022 |
scream | -0.022 |
Token | Tim |
Token ID | 5045 |
Feature activation | -9.30e-03 |
Pos logit contributions | |
ip | +0.004 |
abella | +0.004 |
mal | +0.004 |
od | +0.004 |
mm | +0.004 |
Neg logit contributions | |
walk | -0.006 |
garage | -0.006 |
distance | -0.006 |
street | -0.005 |
proceed | -0.005 |
Token | Let |
Token ID | 5756 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
abella | +0.003 |
mal | +0.003 |
ls | +0.003 |
oll | +0.003 |
ret | +0.002 |
Neg logit contributions | |
street | -0.002 |
walk | -0.002 |
distance | -0.002 |
push | -0.002 |
garage | -0.002 |
Token | 's |
Token ID | 338 |
Feature activation | -7.58e-03 |
Pos logit contributions | |
street | +0.045 |
walk | +0.044 |
garage | +0.044 |
push | +0.041 |
yard | +0.041 |
Neg logit contributions | |
abella | -0.070 |
mal | -0.068 |
od | -0.066 |
ip | -0.065 |
oll | -0.065 |
Token | buy |
Token ID | 2822 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
street | +0.007 |
spine | +0.007 |
garage | +0.007 |
distance | +0.007 |
yard | +0.006 |
Neg logit contributions | |
mal | -0.016 |
abella | -0.016 |
od | -0.015 |
ip | -0.015 |
oll | -0.015 |
Token | it |
Token ID | 340 |
Feature activation | +4.94e-03 |
Pos logit contributions | |
abella | +0.047 |
oll | +0.045 |
mal | +0.045 |
od | +0.043 |
cast | +0.041 |
Neg logit contributions | |
push | -0.026 |
grill | -0.025 |
rescue | -0.025 |
walk | -0.025 |
pull | -0.024 |
Token | so |
Token ID | 523 |
Feature activation | +4.04e-02 |
Pos logit contributions | |
abella | +0.009 |
mal | +0.009 |
od | +0.009 |
oll | +0.009 |
ip | +0.009 |
Neg logit contributions | |
street | -0.006 |
walk | -0.006 |
garage | -0.006 |
yard | -0.006 |
push | -0.006 |
Token | we |
Token ID | 356 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
mal | +0.076 |
od | +0.074 |
abella | +0.073 |
oll | +0.071 |
ils | +0.071 |
Neg logit contributions | |
push | -0.052 |
street | -0.052 |
walk | -0.052 |
while | -0.051 |
pull | -0.050 |
Token | can |
Token ID | 460 |
Feature activation | +3.50e-03 |
Pos logit contributions | |
street | +0.096 |
garage | +0.092 |
walk | +0.092 |
scream | +0.089 |
yard | +0.089 |
Neg logit contributions | |
abella | -0.102 |
mal | -0.102 |
ip | -0.095 |
od | -0.095 |
oll | -0.094 |
Token | go |
Token ID | 467 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
abella | +0.008 |
mal | +0.008 |
od | +0.008 |
ip | +0.008 |
oll | +0.008 |
Neg logit contributions | |
street | -0.003 |
garage | -0.003 |
walk | -0.003 |
yard | -0.002 |
push | -0.002 |
Token | home |
Token ID | 1363 |
Feature activation | +7.49e-03 |
Pos logit contributions | |
walk | +0.040 |
street | +0.040 |
garage | +0.038 |
push | +0.037 |
yard | +0.036 |
Neg logit contributions | |
abella | -0.066 |
oll | -0.063 |
od | -0.063 |
mal | -0.062 |
ret | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | +4.41e-03 |
Pos logit contributions | |
abella | +0.013 |
mal | +0.013 |
od | +0.012 |
oll | +0.012 |
ip | +0.012 |
Neg logit contributions | |
street | -0.010 |
walk | -0.010 |
garage | -0.010 |
yard | -0.010 |
push | -0.010 |
Token | make |
Token ID | 787 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
abella | +0.010 |
mal | +0.009 |
od | +0.009 |
ip | +0.009 |
oll | +0.009 |
Neg logit contributions | |
street | -0.004 |
garage | -0.004 |
walk | -0.004 |
yard | -0.004 |
scream | -0.004 |