Token | aded |
Token ID | 5286 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
rou | +0.028 |
ouncing | +0.023 |
rol | +0.023 |
ucking | +0.022 |
aults | +0.021 |
Neg logit contributions | |
Tim | -0.030 |
pe | -0.030 |
shr | -0.029 |
pepper | -0.028 |
timer | -0.028 |
Token | in |
Token ID | 287 |
Feature activation | +7.79e-02 |
Pos logit contributions | |
pe | +0.022 |
Tim | +0.021 |
pepper | +0.021 |
shr | +0.020 |
fought | +0.020 |
Neg logit contributions | |
rou | -0.029 |
ouncing | -0.025 |
rol | -0.025 |
liness | -0.024 |
Carly | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
help | +0.055 |
pepper | +0.054 |
Tim | +0.050 |
pe | +0.049 |
crack | +0.046 |
Neg logit contributions | |
rou | -0.198 |
ouncing | -0.192 |
rol | -0.179 |
la | -0.176 |
idget | -0.172 |
Token | water |
Token ID | 1660 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
pe | +0.015 |
Tim | +0.014 |
shr | +0.013 |
fought | +0.013 |
pepper | +0.013 |
Neg logit contributions | |
rou | -0.032 |
ouncing | -0.028 |
liness | -0.028 |
rol | -0.027 |
Carly | -0.027 |
Token | delic |
Token ID | 8675 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
pe | +0.033 |
pepper | +0.031 |
help | +0.029 |
Tim | +0.028 |
fought | +0.027 |
Neg logit contributions | |
rou | -0.078 |
ouncing | -0.070 |
Carly | -0.070 |
rol | -0.067 |
aults | -0.067 |
Token | ately |
Token ID | 1286 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
rou | +0.025 |
ouncing | +0.021 |
rol | +0.020 |
ucking | +0.020 |
aults | +0.020 |
Neg logit contributions | |
Tim | -0.025 |
pe | -0.024 |
shr | -0.023 |
pepper | -0.023 |
help | -0.022 |
Token | with |
Token ID | 351 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
pe | +0.211 |
pepper | +0.195 |
Tim | +0.192 |
shr | +0.185 |
help | +0.184 |
Neg logit contributions | |
rou | -0.329 |
ouncing | -0.286 |
rol | -0.278 |
liness | -0.274 |
Carly | -0.272 |
Token | his |
Token ID | 465 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
pe | +0.027 |
Tim | +0.025 |
pepper | +0.025 |
shr | +0.023 |
help | +0.023 |
Neg logit contributions | |
rou | -0.082 |
ouncing | -0.074 |
rol | -0.072 |
liness | -0.071 |
Carly | -0.070 |
Token | fishing |
Token ID | 12478 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
pe | +0.020 |
Tim | +0.020 |
shr | +0.019 |
fought | +0.018 |
timer | +0.017 |
Neg logit contributions | |
rou | -0.027 |
aults | -0.023 |
amaz | -0.022 |
ouncing | -0.022 |
downward | -0.022 |
Token | rod |
Token ID | 15299 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
Tim | +0.045 |
pe | +0.043 |
shr | +0.041 |
fought | +0.039 |
pepper | +0.036 |
Neg logit contributions | |
rou | -0.092 |
ouncing | -0.082 |
liness | -0.081 |
rol | -0.078 |
downward | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
pe | +0.022 |
pepper | +0.020 |
Tim | +0.019 |
help | +0.017 |
shr | +0.017 |
Neg logit contributions | |
rou | -0.078 |
ouncing | -0.070 |
Carly | -0.069 |
rol | -0.069 |
liness | -0.068 |
Token | if |
Token ID | 611 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
rou | +0.011 |
rol | +0.010 |
ouncing | +0.010 |
ucking | +0.010 |
aults | +0.010 |
Neg logit contributions | |
help | -0.004 |
pepper | -0.004 |
tried | -0.003 |
pe | -0.003 |
Tim | -0.003 |
Token | you |
Token ID | 345 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
pe | +0.014 |
pepper | +0.013 |
Tim | +0.013 |
help | +0.013 |
tried | +0.012 |
Neg logit contributions | |
rou | -0.033 |
ouncing | -0.030 |
rol | -0.030 |
liness | -0.029 |
aults | -0.028 |
Token | put |
Token ID | 1234 |
Feature activation | +4.82e-03 |
Pos logit contributions | |
pe | +0.044 |
Tim | +0.042 |
pepper | +0.041 |
shr | +0.039 |
fought | +0.038 |
Neg logit contributions | |
rou | -0.082 |
ouncing | -0.072 |
rol | -0.070 |
liness | -0.069 |
Carly | -0.068 |
Token | un |
Token ID | 555 |
Feature activation | +3.58e-03 |
Pos logit contributions | |
pe | +0.005 |
Tim | +0.005 |
pepper | +0.005 |
shr | +0.005 |
help | +0.005 |
Neg logit contributions | |
rou | -0.010 |
ouncing | -0.009 |
rol | -0.008 |
liness | -0.008 |
Carly | -0.008 |
Token | ri |
Token ID | 380 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
pe | +0.005 |
Tim | +0.005 |
pepper | +0.005 |
shr | +0.004 |
fought | +0.004 |
Neg logit contributions | |
rou | -0.006 |
ouncing | -0.006 |
rol | -0.005 |
liness | -0.005 |
Carly | -0.005 |
Token | pe |
Token ID | 431 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
pe | +0.022 |
pepper | +0.020 |
Tim | +0.020 |
help | +0.019 |
shr | +0.019 |
Neg logit contributions | |
rou | -0.030 |
ouncing | -0.026 |
rol | -0.025 |
liness | -0.025 |
Carly | -0.025 |
Token | fruit |
Token ID | 8234 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
pe | +0.204 |
Tim | +0.193 |
pepper | +0.189 |
shr | +0.183 |
fought | +0.179 |
Neg logit contributions | |
rou | -0.317 |
ouncing | -0.278 |
rol | -0.269 |
liness | -0.267 |
Carly | -0.263 |
Token | in |
Token ID | 287 |
Feature activation | +7.67e-02 |
Pos logit contributions | |
pe | +0.046 |
pepper | +0.043 |
Tim | +0.043 |
help | +0.041 |
shr | +0.040 |
Neg logit contributions | |
rou | -0.081 |
ouncing | -0.072 |
Carly | -0.069 |
rol | -0.068 |
liness | -0.068 |
Token | a |
Token ID | 257 |
Feature activation | +5.52e-03 |
Pos logit contributions | |
pe | +0.083 |
Tim | +0.078 |
pepper | +0.078 |
help | +0.075 |
shr | +0.072 |
Neg logit contributions | |
rou | -0.160 |
ouncing | -0.146 |
rol | -0.139 |
liness | -0.135 |
Carly | -0.134 |
Token | warm |
Token ID | 5814 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
pe | +0.007 |
Tim | +0.007 |
shr | +0.006 |
pepper | +0.006 |
fought | +0.006 |
Neg logit contributions | |
rou | -0.010 |
ouncing | -0.009 |
liness | -0.009 |
rol | -0.009 |
downward | -0.009 |
Token | place |
Token ID | 1295 |
Feature activation | +8.90e-04 |
Pos logit contributions | |
rou | +0.026 |
ouncing | +0.023 |
rol | +0.022 |
aults | +0.022 |
downward | +0.022 |
Neg logit contributions | |
pe | -0.014 |
pepper | -0.013 |
Tim | -0.013 |
help | -0.013 |
tried | -0.012 |
Token | his |
Token ID | 465 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
rou | +0.011 |
ouncing | +0.010 |
ucking | +0.009 |
rol | +0.009 |
liness | +0.009 |
Neg logit contributions | |
pe | -0.004 |
Tim | -0.004 |
pepper | -0.004 |
help | -0.004 |
fought | -0.003 |
Token | mind |
Token ID | 2000 |
Feature activation | +4.33e-02 |
Pos logit contributions | |
pe | +0.019 |
Tim | +0.018 |
pepper | +0.018 |
shr | +0.017 |
help | +0.017 |
Neg logit contributions | |
rou | -0.029 |
ouncing | -0.025 |
rol | -0.025 |
liness | -0.024 |
Carly | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
pe | +0.031 |
help | +0.025 |
pepper | +0.025 |
cube | +0.023 |
Tim | +0.023 |
Neg logit contributions | |
rou | -0.103 |
ouncing | -0.095 |
ucking | -0.093 |
rol | -0.092 |
Carly | -0.092 |
Token | He |
Token ID | 679 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
Tim | +0.008 |
But | +0.007 |
pe | +0.006 |
� | +0.006 |
Tim | +0.005 |
Neg logit contributions | |
rou | -0.035 |
aults | -0.032 |
iet | -0.031 |
ouncing | -0.031 |
liness | -0.031 |
Token | w |
Token ID | 266 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
pe | +0.009 |
tried | +0.009 |
fought | +0.009 |
Tim | +0.008 |
pepper | +0.008 |
Neg logit contributions | |
rou | -0.034 |
ouncing | -0.032 |
rol | -0.031 |
liness | -0.031 |
ucking | -0.031 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
rou | +0.036 |
ouncing | +0.032 |
rol | +0.031 |
liness | +0.030 |
Carly | +0.030 |
Neg logit contributions | |
pe | -0.021 |
Tim | -0.021 |
pepper | -0.020 |
shr | -0.019 |
help | -0.019 |
Token | his |
Token ID | 465 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
pe | +0.068 |
tried | +0.049 |
Tim | +0.041 |
fought | +0.041 |
pepper | +0.035 |
Neg logit contributions | |
rou | -0.474 |
ouncing | -0.433 |
ucking | -0.424 |
rol | -0.423 |
itating | -0.409 |
Token | tail |
Token ID | 7894 |
Feature activation | +8.17e-03 |
Pos logit contributions | |
pe | +0.020 |
Tim | +0.019 |
pepper | +0.019 |
shr | +0.018 |
fought | +0.018 |
Neg logit contributions | |
rou | -0.024 |
ouncing | -0.021 |
liness | -0.020 |
Carly | -0.020 |
rol | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -3.42e-03 |
Pos logit contributions | |
pe | +0.009 |
tried | +0.008 |
help | +0.008 |
pepper | +0.008 |
Tim | +0.007 |
Neg logit contributions | |
rou | -0.018 |
ouncing | -0.016 |
rol | -0.015 |
liness | -0.015 |
idget | -0.015 |
Token | bark |
Token ID | 21405 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
rou | +0.008 |
ouncing | +0.007 |
rol | +0.007 |
liness | +0.007 |
ucking | +0.007 |
Neg logit contributions | |
pe | -0.003 |
Tim | -0.003 |
pepper | -0.003 |
fought | -0.003 |
tried | -0.002 |
Token | ed |
Token ID | 276 |
Feature activation | +6.00e-02 |
Pos logit contributions | |
pe | +0.028 |
Tim | +0.026 |
fought | +0.025 |
help | +0.025 |
tried | +0.025 |
Neg logit contributions | |
rou | -0.055 |
ouncing | -0.049 |
liness | -0.047 |
aj | -0.046 |
Carly | -0.046 |
Token | very |
Token ID | 845 |
Feature activation | +5.24e-02 |
Pos logit contributions | |
pe | +0.009 |
Tim | +0.008 |
shr | +0.008 |
occasion | +0.008 |
pepper | +0.008 |
Neg logit contributions | |
rou | -0.011 |
ouncing | -0.008 |
capt | -0.008 |
rol | -0.008 |
downward | -0.008 |
Token | modest |
Token ID | 12949 |
Feature activation | +6.97e-03 |
Pos logit contributions | |
pe | +0.087 |
Tim | +0.082 |
shr | +0.080 |
occasion | +0.078 |
pepper | +0.077 |
Neg logit contributions | |
rou | -0.080 |
ouncing | -0.064 |
capt | -0.058 |
ucking | -0.058 |
forward | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | -5.67e-04 |
Pos logit contributions | |
pe | +0.007 |
pepper | +0.007 |
Tim | +0.007 |
help | +0.007 |
fought | +0.007 |
Neg logit contributions | |
rou | -0.014 |
rol | -0.013 |
ouncing | -0.013 |
Carly | -0.012 |
aults | -0.012 |
Token | never |
Token ID | 1239 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
rou | +0.001 |
ouncing | +0.001 |
liness | +0.001 |
Carly | +0.001 |
downward | +0.001 |
Neg logit contributions | |
pe | -0.001 |
shr | -0.001 |
� | -0.001 |
Tim | -0.001 |
timer | -0.001 |
Token | br |
Token ID | 865 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
pe | +0.011 |
Tim | +0.010 |
pepper | +0.009 |
� | +0.009 |
shr | +0.009 |
Neg logit contributions | |
rou | -0.024 |
ouncing | -0.022 |
liness | -0.021 |
rol | -0.020 |
aults | -0.020 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
pe | +0.050 |
Tim | +0.047 |
pepper | +0.046 |
shr | +0.045 |
fought | +0.044 |
Neg logit contributions | |
rou | -0.071 |
ouncing | -0.062 |
liness | -0.060 |
rol | -0.060 |
Carly | -0.059 |
Token | about |
Token ID | 546 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
pe | +0.091 |
pepper | +0.084 |
fought | +0.080 |
Tim | +0.078 |
help | +0.074 |
Neg logit contributions | |
rou | -0.428 |
ouncing | -0.394 |
rol | -0.391 |
ucking | -0.384 |
aults | -0.376 |
Token | his |
Token ID | 465 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
rou | +0.144 |
ouncing | +0.120 |
liness | +0.119 |
Carly | +0.115 |
rol | +0.114 |
Neg logit contributions | |
pe | -0.107 |
Tim | -0.102 |
shr | -0.101 |
pepper | -0.097 |
fought | -0.094 |
Token | work |
Token ID | 670 |
Feature activation | +6.57e-02 |
Pos logit contributions | |
Tim | +0.021 |
pe | +0.020 |
shr | +0.020 |
pepper | +0.019 |
fought | +0.018 |
Neg logit contributions | |
rou | -0.023 |
ouncing | -0.019 |
liness | -0.018 |
aj | -0.018 |
amaz | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
pe | +0.054 |
Tim | +0.049 |
pepper | +0.048 |
shr | +0.045 |
help | +0.044 |
Neg logit contributions | |
rou | -0.153 |
ouncing | -0.137 |
rol | -0.134 |
liness | -0.133 |
Carly | -0.132 |
Token | Every |
Token ID | 3887 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
pe | +0.013 |
shr | +0.013 |
pepper | +0.012 |
fought | +0.012 |
timer | +0.012 |
Neg logit contributions | |
rou | -0.026 |
ouncing | -0.023 |
liness | -0.023 |
Carly | -0.023 |
rol | -0.023 |
Token | day |
Token ID | 1110 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
rou | +0.038 |
ouncing | +0.034 |
rol | +0.033 |
liness | +0.033 |
Carly | +0.033 |
Neg logit contributions | |
pe | -0.012 |
Tim | -0.011 |
pepper | -0.011 |
help | -0.010 |
shr | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
pe | +0.039 |
Tim | +0.036 |
pepper | +0.035 |
help | +0.033 |
shr | +0.033 |
Neg logit contributions | |
rou | -0.094 |
ouncing | -0.084 |
rol | -0.082 |
liness | -0.082 |
Carly | -0.080 |
Token | They |
Token ID | 1119 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
Tim | +0.008 |
� | +0.007 |
pe | +0.007 |
But | +0.007 |
pepper | +0.007 |
Neg logit contributions | |
rou | -0.024 |
ouncing | -0.021 |
aults | -0.021 |
iet | -0.021 |
liness | -0.021 |
Token | worked |
Token ID | 3111 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
pe | +0.015 |
Tim | +0.014 |
pepper | +0.013 |
shr | +0.013 |
help | +0.012 |
Neg logit contributions | |
rou | -0.052 |
ouncing | -0.047 |
rol | -0.046 |
liness | -0.046 |
Carly | -0.045 |
Token | delic |
Token ID | 8675 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
pe | +0.042 |
Tim | +0.039 |
pepper | +0.039 |
help | +0.038 |
shr | +0.036 |
Neg logit contributions | |
rou | -0.093 |
ouncing | -0.084 |
rol | -0.082 |
liness | -0.081 |
Carly | -0.079 |
Token | ately |
Token ID | 1286 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
rou | +0.038 |
liness | +0.033 |
ouncing | +0.033 |
rol | +0.032 |
Carly | +0.032 |
Neg logit contributions | |
pe | -0.020 |
Tim | -0.019 |
pepper | -0.018 |
fought | -0.018 |
shr | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -9.24e-03 |
Pos logit contributions | |
pe | +0.196 |
Tim | +0.188 |
pepper | +0.182 |
shr | +0.177 |
fought | +0.171 |
Neg logit contributions | |
rou | -0.321 |
ouncing | -0.282 |
rol | -0.273 |
liness | -0.271 |
Carly | -0.266 |
Token | carefully |
Token ID | 7773 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
rou | +0.020 |
ouncing | +0.018 |
rol | +0.017 |
Carly | +0.017 |
liness | +0.017 |
Neg logit contributions | |
pe | -0.009 |
Tim | -0.008 |
shr | -0.008 |
pepper | -0.008 |
Blood | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -9.83e-03 |
Pos logit contributions | |
pe | +0.083 |
Tim | +0.081 |
shr | +0.077 |
pepper | +0.076 |
fought | +0.070 |
Neg logit contributions | |
rou | -0.151 |
ouncing | -0.133 |
liness | -0.130 |
rol | -0.129 |
Carly | -0.127 |
Token | nothing |
Token ID | 2147 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
rou | +0.021 |
ouncing | +0.019 |
rol | +0.018 |
liness | +0.018 |
Carly | +0.018 |
Neg logit contributions | |
pe | -0.010 |
Tim | -0.009 |
pepper | -0.009 |
shr | -0.009 |
fought | -0.009 |
Token | broke |
Token ID | 6265 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
rou | +0.037 |
ouncing | +0.033 |
rol | +0.032 |
liness | +0.032 |
Carly | +0.032 |
Neg logit contributions | |
pe | -0.014 |
Tim | -0.013 |
pepper | -0.013 |
shr | -0.012 |
help | -0.012 |
Token | Norman |
Token ID | 19636 |
Feature activation | +5.37e-03 |
Pos logit contributions | |
help | +0.012 |
pepper | +0.011 |
pe | +0.011 |
Tim | +0.010 |
fought | +0.009 |
Neg logit contributions | |
rou | -0.039 |
ouncing | -0.036 |
rol | -0.034 |
ucking | -0.034 |
aults | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +7.29e-03 |
Pos logit contributions | |
pe | +0.004 |
help | +0.004 |
pepper | +0.003 |
Tim | +0.003 |
needed | +0.003 |
Neg logit contributions | |
rou | -0.013 |
rol | -0.012 |
amaz | -0.012 |
ouncing | -0.012 |
ayed | -0.012 |
Token | His |
Token ID | 2399 |
Feature activation | +8.22e-02 |
Pos logit contributions | |
pe | +0.004 |
Tim | +0.004 |
pepper | +0.004 |
help | +0.003 |
shr | +0.003 |
Neg logit contributions | |
rou | -0.019 |
ouncing | -0.017 |
aults | -0.017 |
rol | -0.017 |
liness | -0.017 |
Token | tail |
Token ID | 7894 |
Feature activation | +9.80e-03 |
Pos logit contributions | |
pe | +0.095 |
pepper | +0.095 |
help | +0.094 |
Tim | +0.088 |
fought | +0.084 |
Neg logit contributions | |
rou | -0.163 |
rol | -0.148 |
ouncing | -0.148 |
ucking | -0.140 |
liness | -0.139 |
Token | w |
Token ID | 266 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
pe | +0.006 |
tried | +0.005 |
pepper | +0.005 |
fought | +0.004 |
Tim | +0.004 |
Neg logit contributions | |
rou | -0.026 |
ouncing | -0.024 |
Carly | -0.023 |
rol | -0.023 |
aults | -0.023 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
rou | +0.040 |
ouncing | +0.035 |
rol | +0.034 |
liness | +0.034 |
Carly | +0.034 |
Neg logit contributions | |
pe | -0.021 |
Tim | -0.019 |
pepper | -0.019 |
shr | -0.018 |
fought | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
pe | +0.091 |
pepper | +0.062 |
tried | +0.061 |
fought | +0.059 |
thought | +0.052 |
Neg logit contributions | |
rou | -0.450 |
ucking | -0.411 |
ouncing | -0.410 |
rol | -0.410 |
idget | -0.398 |
Token | joy |
Token ID | 8716 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
help | +0.044 |
pe | +0.044 |
Tim | +0.043 |
pepper | +0.043 |
fought | +0.040 |
Neg logit contributions | |
rou | -0.057 |
ouncing | -0.052 |
rol | -0.047 |
liness | -0.046 |
aj | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +8.32e-03 |
Pos logit contributions | |
rou | +0.056 |
rol | +0.051 |
ouncing | +0.050 |
Carly | +0.049 |
amaz | +0.048 |
Neg logit contributions | |
pe | -0.029 |
pepper | -0.025 |
help | -0.024 |
fought | -0.023 |
Tim | -0.022 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
pe | +0.005 |
Tim | +0.005 |
pepper | +0.005 |
help | +0.004 |
shr | +0.004 |
Neg logit contributions | |
rou | -0.021 |
ouncing | -0.019 |
aults | -0.019 |
rol | -0.018 |
liness | -0.018 |
Token | Once |
Token ID | 7454 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
rou | +0.108 |
ouncing | +0.095 |
rol | +0.095 |
liness | +0.095 |
Carly | +0.094 |
Neg logit contributions | |
pe | -0.053 |
pepper | -0.050 |
fought | -0.048 |
Tim | -0.048 |
shr | -0.047 |
Token | stairs |
Token ID | 16046 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
pe | +0.009 |
Tim | +0.009 |
pepper | +0.008 |
shr | +0.008 |
help | +0.008 |
Neg logit contributions | |
rou | -0.023 |
ouncing | -0.021 |
rol | -0.020 |
liness | -0.020 |
Carly | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -4.93e-03 |
Pos logit contributions | |
rou | +0.053 |
ouncing | +0.049 |
Carly | +0.046 |
liness | +0.046 |
aults | +0.046 |
Neg logit contributions | |
pe | -0.025 |
help | -0.023 |
Tim | -0.022 |
tried | -0.022 |
fought | -0.022 |
Token | z |
Token ID | 1976 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
rou | +0.011 |
ucking | +0.010 |
rol | +0.010 |
ouncing | +0.010 |
ayed | +0.009 |
Neg logit contributions | |
tried | -0.005 |
help | -0.005 |
Tim | -0.004 |
fought | -0.004 |
be | -0.004 |
Token | ig |
Token ID | 328 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
rou | +0.064 |
liness | +0.056 |
ouncing | +0.054 |
downward | +0.054 |
rol | +0.054 |
Neg logit contributions | |
pe | -0.051 |
help | -0.048 |
� | -0.048 |
fought | -0.047 |
shr | -0.047 |
Token | z |
Token ID | 89 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
pe | +0.026 |
tried | +0.023 |
help | +0.022 |
Tim | +0.022 |
fought | +0.021 |
Neg logit contributions | |
rou | -0.034 |
amaz | -0.033 |
downward | -0.030 |
reating | -0.029 |
Carly | -0.029 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
rou | +0.060 |
ouncing | +0.052 |
liness | +0.049 |
rol | +0.049 |
Carly | +0.048 |
Neg logit contributions | |
pe | -0.043 |
Tim | -0.041 |
pepper | -0.040 |
help | -0.040 |
shr | -0.039 |
Token | around |
Token ID | 1088 |
Feature activation | -5.64e-03 |
Pos logit contributions | |
be | +0.131 |
Tim | +0.127 |
pe | +0.126 |
Tim | +0.116 |
split | +0.115 |
Neg logit contributions | |
rou | -0.357 |
ucking | -0.350 |
ouncing | -0.342 |
ogging | -0.324 |
ares | -0.322 |
Token | the |
Token ID | 262 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
rou | +0.013 |
ouncing | +0.012 |
rol | +0.011 |
ucking | +0.011 |
idget | +0.011 |
Neg logit contributions | |
pe | -0.005 |
Tim | -0.005 |
help | -0.005 |
pepper | -0.004 |
shr | -0.004 |
Token | hallway |
Token ID | 23959 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
Tim | +0.012 |
pe | +0.012 |
pepper | +0.011 |
help | +0.011 |
shr | +0.010 |
Neg logit contributions | |
rou | -0.022 |
ouncing | -0.021 |
rol | -0.020 |
Carly | -0.019 |
ucking | -0.019 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.55e-03 |
Pos logit contributions | |
rou | +0.062 |
ouncing | +0.057 |
Carly | +0.055 |
idget | +0.055 |
rol | +0.055 |
Neg logit contributions | |
pe | -0.020 |
help | -0.017 |
Tim | -0.017 |
fought | -0.016 |
tried | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +3.25e-03 |
Pos logit contributions | |
rou | +0.003 |
ouncing | +0.003 |
rol | +0.003 |
liness | +0.003 |
Carly | +0.003 |
Neg logit contributions | |
pe | -0.001 |
Tim | -0.001 |
pepper | -0.001 |
shr | -0.001 |
fought | -0.001 |
Token | ggy |
Token ID | 19970 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
pe | +0.004 |
Tim | +0.004 |
pepper | +0.003 |
help | +0.003 |
shr | +0.003 |
Neg logit contributions | |
rou | -0.008 |
ouncing | -0.007 |
liness | -0.007 |
rol | -0.007 |
aults | -0.007 |
Token | win |
Token ID | 1592 |
Feature activation | -4.53e-03 |
Pos logit contributions | |
rou | +0.089 |
ouncing | +0.080 |
rol | +0.078 |
liness | +0.077 |
ucking | +0.076 |
Neg logit contributions | |
pe | -0.027 |
fought | -0.024 |
Tim | -0.024 |
tried | -0.023 |
pepper | -0.023 |
Token | ned |
Token ID | 2817 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
rou | +0.010 |
ouncing | +0.009 |
rol | +0.009 |
liness | +0.009 |
ucking | +0.009 |
Neg logit contributions | |
pe | -0.004 |
Tim | -0.004 |
fought | -0.004 |
tried | -0.004 |
help | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -3.12e-03 |
Pos logit contributions | |
pe | +0.002 |
help | +0.002 |
pepper | +0.002 |
Tim | +0.002 |
fought | +0.002 |
Neg logit contributions | |
rou | -0.004 |
ouncing | -0.003 |
ucking | -0.003 |
rol | -0.003 |
aults | -0.003 |
Token | w |
Token ID | 266 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
rou | +0.007 |
ouncing | +0.006 |
ucking | +0.006 |
rol | +0.006 |
liness | +0.006 |
Neg logit contributions | |
pe | -0.003 |
fought | -0.003 |
tried | -0.003 |
help | -0.002 |
Tim | -0.002 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
rou | +0.037 |
ouncing | +0.033 |
rol | +0.032 |
ayed | +0.030 |
ucking | +0.030 |
Neg logit contributions | |
Tim | -0.024 |
pe | -0.024 |
pepper | -0.023 |
shr | -0.022 |
help | -0.021 |
Token | his |
Token ID | 465 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
pe | +0.137 |
Tim | +0.108 |
fought | +0.107 |
tried | +0.105 |
help | +0.102 |
Neg logit contributions | |
rou | -0.401 |
ouncing | -0.363 |
rol | -0.351 |
ucking | -0.348 |
aj | -0.340 |
Token | tail |
Token ID | 7894 |
Feature activation | +7.49e-03 |
Pos logit contributions | |
Tim | +0.023 |
pepper | +0.022 |
shr | +0.022 |
pe | +0.021 |
fought | +0.021 |
Neg logit contributions | |
rou | -0.019 |
ouncing | -0.016 |
aults | -0.015 |
Carly | -0.015 |
downward | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
pe | +0.005 |
help | +0.004 |
Tim | +0.004 |
fought | +0.004 |
tried | +0.004 |
Neg logit contributions | |
rou | -0.019 |
ouncing | -0.018 |
Carly | -0.017 |
rol | -0.017 |
liness | -0.017 |
Token | Debbie |
Token ID | 31311 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
But | +0.010 |
pe | +0.009 |
His | +0.008 |
Blood | +0.007 |
Tim | +0.007 |
Neg logit contributions | |
rou | -0.026 |
aults | -0.022 |
ouncing | -0.022 |
liness | -0.021 |
rol | -0.021 |
Token | smiled |
Token ID | 13541 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
rou | +0.279 |
ouncing | +0.255 |
rol | +0.249 |
liness | +0.246 |
downward | +0.243 |
Neg logit contributions | |
pe | -0.080 |
fought | -0.068 |
tried | -0.067 |
Tim | -0.065 |
pepper | -0.064 |
Token | his |
Token ID | 465 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
pe | +0.020 |
Tim | +0.019 |
shr | +0.018 |
pepper | +0.018 |
help | +0.017 |
Neg logit contributions | |
rou | -0.035 |
liness | -0.030 |
ouncing | -0.030 |
Carly | -0.030 |
rol | -0.030 |
Token | strength |
Token ID | 4202 |
Feature activation | -4.57e-03 |
Pos logit contributions | |
pe | +0.014 |
Tim | +0.013 |
pepper | +0.013 |
shr | +0.012 |
fought | +0.012 |
Neg logit contributions | |
rou | -0.021 |
ouncing | -0.018 |
liness | -0.017 |
rol | -0.017 |
Carly | -0.017 |
Token | back |
Token ID | 736 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
rou | +0.010 |
ouncing | +0.010 |
iet | +0.009 |
idget | +0.009 |
Carly | +0.009 |
Neg logit contributions | |
help | -0.004 |
pe | -0.004 |
pepper | -0.004 |
tried | -0.003 |
hurt | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -4.22e-03 |
Pos logit contributions | |
rou | +0.052 |
rol | +0.048 |
ouncing | +0.047 |
downward | +0.046 |
Carly | +0.046 |
Neg logit contributions | |
pe | -0.019 |
help | -0.019 |
pepper | -0.017 |
tried | -0.017 |
fought | -0.016 |
Token | w |
Token ID | 266 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
rou | +0.010 |
ouncing | +0.009 |
rol | +0.009 |
aults | +0.008 |
Carly | +0.008 |
Neg logit contributions | |
pe | -0.004 |
pepper | -0.004 |
Tim | -0.003 |
help | -0.003 |
fought | -0.003 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
rou | +0.037 |
ouncing | +0.032 |
rol | +0.031 |
liness | +0.030 |
ucking | +0.030 |
Neg logit contributions | |
pe | -0.027 |
Tim | -0.026 |
pepper | -0.026 |
shr | -0.025 |
fought | -0.024 |
Token | his |
Token ID | 465 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
pe | +0.071 |
help | +0.052 |
pepper | +0.049 |
tried | +0.044 |
fought | +0.044 |
Neg logit contributions | |
rou | -0.454 |
ouncing | -0.426 |
rol | -0.416 |
ucking | -0.416 |
idget | -0.405 |
Token | tail |
Token ID | 7894 |
Feature activation | +5.41e-03 |
Pos logit contributions | |
Tim | +0.017 |
shr | +0.016 |
pepper | +0.016 |
pe | +0.016 |
fought | +0.016 |
Neg logit contributions | |
rou | -0.017 |
ouncing | -0.014 |
liness | -0.014 |
aults | -0.013 |
Carly | -0.013 |
Token | in |
Token ID | 287 |
Feature activation | +7.58e-02 |
Pos logit contributions | |
pe | +0.004 |
help | +0.003 |
pepper | +0.002 |
Tim | +0.002 |
tried | +0.002 |
Neg logit contributions | |
rou | -0.014 |
ouncing | -0.013 |
Carly | -0.013 |
idget | -0.012 |
iet | -0.012 |
Token | excitement |
Token ID | 14067 |
Feature activation | -9.01e-03 |
Pos logit contributions | |
pe | +0.090 |
help | +0.089 |
Tim | +0.089 |
pepper | +0.087 |
fought | +0.081 |
Neg logit contributions | |
rou | -0.150 |
ouncing | -0.140 |
rol | -0.129 |
liness | -0.123 |
downward | -0.122 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
rou | +0.025 |
ouncing | +0.023 |
rol | +0.022 |
idget | +0.022 |
Carly | +0.022 |
Neg logit contributions | |
pe | -0.004 |
help | -0.003 |
pepper | -0.002 |
Tim | -0.002 |
fought | -0.002 |
Token | out |
Token ID | 503 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
rou | +0.008 |
ouncing | +0.007 |
rol | +0.007 |
liness | +0.007 |
Carly | +0.006 |
Neg logit contributions | |
pe | -0.003 |
Tim | -0.003 |
pepper | -0.003 |
shr | -0.003 |
fought | -0.003 |
Token | popped |
Token ID | 22928 |
Feature activation | +4.46e-03 |
Pos logit contributions | |
rou | +0.032 |
ouncing | +0.028 |
liness | +0.027 |
rol | +0.027 |
Carly | +0.026 |
Neg logit contributions | |
pe | -0.019 |
Tim | -0.017 |
pepper | -0.017 |
help | -0.017 |
shr | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | +5.28e-03 |
Pos logit contributions | |
pe | +0.004 |
pepper | +0.004 |
help | +0.004 |
Tim | +0.004 |
Tim | +0.004 |
Neg logit contributions | |
rou | -0.010 |
ouncing | -0.010 |
ogging | -0.009 |
ucking | -0.008 |
aj | -0.008 |
Token | little |
Token ID | 1310 |
Feature activation | +2.16e-04 |
Pos logit contributions | |
pe | +0.006 |
pepper | +0.006 |
Tim | +0.006 |
shr | +0.005 |
help | +0.005 |
Neg logit contributions | |
rou | -0.011 |
ouncing | -0.009 |
rol | -0.009 |
liness | -0.009 |
Carly | -0.009 |
Token | wood |
Token ID | 4898 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
pepper | +0.000 |
pe | +0.000 |
help | +0.000 |
Tim | +0.000 |
shr | +0.000 |
Neg logit contributions | |
rou | -0.000 |
ouncing | -0.000 |
rol | -0.000 |
Carly | -0.000 |
liness | -0.000 |
Token | pe |
Token ID | 431 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
pe | +0.079 |
pepper | +0.068 |
Tim | +0.065 |
help | +0.063 |
fought | +0.061 |
Neg logit contributions | |
rou | -0.178 |
ouncing | -0.163 |
Carly | -0.157 |
rol | -0.156 |
liness | -0.155 |
Token | cker |
Token ID | 15280 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Tim | +0.166 |
pe | +0.150 |
pepper | +0.149 |
shr | +0.145 |
help | +0.142 |
Neg logit contributions | |
rou | -0.350 |
ouncing | -0.305 |
rol | -0.298 |
aults | -0.294 |
liness | -0.291 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
pe | +0.115 |
pepper | +0.105 |
help | +0.101 |
Tim | +0.098 |
fought | +0.098 |
Neg logit contributions | |
rou | -0.209 |
ouncing | -0.190 |
rol | -0.181 |
Carly | -0.179 |
idget | -0.178 |
Token |
|
Token ID | 198 |
Feature activation | +7.02e-03 |
Pos logit contributions | |
pe | +0.012 |
Tim | +0.011 |
pepper | +0.010 |
shr | +0.010 |
help | +0.009 |
Neg logit contributions | |
rou | -0.033 |
ouncing | -0.030 |
rol | -0.029 |
liness | -0.029 |
Carly | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +6.86e-03 |
Pos logit contributions | |
pe | +0.006 |
Tim | +0.005 |
pepper | +0.005 |
shr | +0.005 |
fought | +0.005 |
Neg logit contributions | |
rou | -0.016 |
ouncing | -0.015 |
rol | -0.014 |
liness | -0.014 |
Carly | -0.014 |
Token | " |
Token ID | 1 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
Tim | +0.008 |
pe | +0.007 |
shr | +0.006 |
pepper | +0.006 |
help | +0.006 |
Neg logit contributions | |
rou | -0.015 |
ouncing | -0.013 |
liness | -0.013 |
rol | -0.013 |
la | -0.013 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
rou | +0.012 |
ouncing | +0.011 |
rol | +0.010 |
liness | +0.010 |
aults | +0.010 |
Neg logit contributions | |
pe | -0.004 |
Tim | -0.003 |
fought | -0.003 |
pepper | -0.003 |
tried | -0.003 |
Token | in |
Token ID | 287 |
Feature activation | +7.41e-02 |
Pos logit contributions | |
pe | +0.028 |
help | +0.025 |
Tim | +0.025 |
pepper | +0.024 |
tried | +0.024 |
Neg logit contributions | |
rou | -0.053 |
ouncing | -0.049 |
rol | -0.047 |
liness | -0.045 |
aults | -0.044 |
Token | agreement |
Token ID | 4381 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
pe | +0.123 |
Tim | +0.120 |
help | +0.118 |
pepper | +0.115 |
fought | +0.111 |
Neg logit contributions | |
rou | -0.117 |
ouncing | -0.102 |
rol | -0.095 |
la | -0.090 |
liness | -0.090 |
Token | , |
Token ID | 11 |
Feature activation | -7.07e-03 |
Pos logit contributions | |
rou | +0.032 |
ouncing | +0.028 |
rol | +0.028 |
aults | +0.027 |
liness | +0.027 |
Neg logit contributions | |
pe | -0.017 |
Tim | -0.015 |
pepper | -0.015 |
help | -0.015 |
fought | -0.014 |
Token | w |
Token ID | 266 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
rou | +0.015 |
rol | +0.013 |
liness | +0.013 |
ouncing | +0.013 |
iet | +0.013 |
Neg logit contributions | |
pe | -0.008 |
tried | -0.007 |
Tim | -0.007 |
help | -0.007 |
pepper | -0.007 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
rou | +0.036 |
ouncing | +0.030 |
rol | +0.028 |
ayed | +0.027 |
ucking | +0.027 |
Neg logit contributions | |
Tim | -0.036 |
pepper | -0.036 |
shr | -0.035 |
pe | -0.035 |
fought | -0.033 |
Token | its |
Token ID | 663 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
pe | +0.090 |
tried | +0.068 |
Tim | +0.063 |
fought | +0.061 |
help | +0.057 |
Neg logit contributions | |
rou | -0.455 |
rol | -0.408 |
ouncing | -0.408 |
ucking | -0.392 |
idget | -0.387 |
Token | tail |
Token ID | 7894 |
Feature activation | +6.33e-03 |
Pos logit contributions | |
pe | +0.062 |
Tim | +0.058 |
pepper | +0.057 |
shr | +0.055 |
help | +0.055 |
Neg logit contributions | |
rou | -0.094 |
ouncing | -0.082 |
rol | -0.080 |
liness | -0.079 |
Carly | -0.078 |
Token | and |
Token ID | 290 |
Feature activation | -5.69e-03 |
Pos logit contributions | |
pe | +0.007 |
help | +0.006 |
pepper | +0.006 |
tried | +0.006 |
Tim | +0.006 |
Neg logit contributions | |
rou | -0.014 |
ouncing | -0.012 |
Carly | -0.012 |
rol | -0.012 |
liness | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +8.41e-03 |
Pos logit contributions | |
rou | +0.013 |
ouncing | +0.012 |
rol | +0.011 |
liness | +0.011 |
aults | +0.011 |
Neg logit contributions | |
pe | -0.005 |
Tim | -0.005 |
pepper | -0.004 |
fought | -0.004 |
help | -0.004 |
Token | two |
Token ID | 734 |
Feature activation | -6.39e-02 |
Pos logit contributions | |
pe | +0.009 |
Tim | +0.009 |
shr | +0.009 |
pepper | +0.009 |
fought | +0.008 |
Neg logit contributions | |
rou | -0.017 |
ouncing | -0.015 |
liness | -0.014 |
rol | -0.014 |
Carly | -0.014 |
Token | I |
Token ID | 40 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
rou | +0.052 |
ouncing | +0.045 |
rol | +0.045 |
aults | +0.044 |
downward | +0.044 |
Neg logit contributions | |
pepper | -0.029 |
help | -0.029 |
pe | -0.029 |
fought | -0.028 |
tried | -0.025 |
Token | 'm |
Token ID | 1101 |
Feature activation | +3.27e-04 |
Pos logit contributions | |
pe | +0.043 |
help | +0.040 |
pepper | +0.039 |
Tim | +0.039 |
fought | +0.039 |
Neg logit contributions | |
rou | -0.088 |
ouncing | -0.081 |
liness | -0.079 |
rol | -0.079 |
Carly | -0.078 |
Token | a |
Token ID | 257 |
Feature activation | +9.29e-03 |
Pos logit contributions | |
pe | +0.000 |
pepper | +0.000 |
help | +0.000 |
tried | +0.000 |
Tim | +0.000 |
Neg logit contributions | |
rou | -0.001 |
liness | -0.001 |
ouncing | -0.001 |
rol | -0.001 |
aults | -0.001 |
Token | tight |
Token ID | 5381 |
Feature activation | -7.25e-03 |
Pos logit contributions | |
pe | +0.012 |
Tim | +0.011 |
pepper | +0.011 |
shr | +0.010 |
help | +0.010 |
Neg logit contributions | |
rou | -0.018 |
ouncing | -0.015 |
rol | -0.015 |
liness | -0.015 |
Carly | -0.015 |
Token | ro |
Token ID | 305 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
rou | +0.015 |
ouncing | +0.014 |
rol | +0.014 |
liness | +0.013 |
Carly | +0.013 |
Neg logit contributions | |
pe | -0.008 |
pepper | -0.008 |
help | -0.007 |
Tim | -0.007 |
shr | -0.006 |
Token | pe |
Token ID | 431 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
pe | +0.010 |
Tim | +0.007 |
pepper | +0.005 |
shr | +0.004 |
fought | +0.002 |
Neg logit contributions | |
rou | -0.153 |
ouncing | -0.141 |
rol | -0.138 |
liness | -0.137 |
Carly | -0.136 |
Token | walk |
Token ID | 2513 |
Feature activation | +3.30e-03 |
Pos logit contributions | |
Tim | +0.181 |
pe | +0.173 |
pepper | +0.170 |
shr | +0.165 |
help | +0.160 |
Neg logit contributions | |
rou | -0.333 |
ouncing | -0.293 |
rol | -0.280 |
liness | -0.276 |
aults | -0.272 |
Token | er |
Token ID | 263 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
pe | +0.003 |
Tim | +0.003 |
help | +0.003 |
pepper | +0.003 |
fought | +0.003 |
Neg logit contributions | |
rou | -0.007 |
ouncing | -0.007 |
amaz | -0.006 |
liness | -0.006 |
Carly | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
pe | +0.014 |
Tim | +0.012 |
pepper | +0.012 |
help | +0.011 |
fought | +0.011 |
Neg logit contributions | |
rou | -0.022 |
ouncing | -0.019 |
amaz | -0.019 |
liness | -0.019 |
rol | -0.019 |
Token | Do |
Token ID | 2141 |
Feature activation | +2.69e-03 |
Pos logit contributions | |
pe | +0.018 |
Tim | +0.018 |
pepper | +0.017 |
help | +0.016 |
shr | +0.016 |
Neg logit contributions | |
rou | -0.039 |
ouncing | -0.034 |
rol | -0.033 |
aults | -0.033 |
liness | -0.033 |
Token | you |
Token ID | 345 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
pe | +0.002 |
pepper | +0.002 |
Tim | +0.002 |
shr | +0.002 |
fought | +0.002 |
Neg logit contributions | |
rou | -0.006 |
ouncing | -0.005 |
Carly | -0.005 |
downward | -0.005 |
aults | -0.005 |
Token | work |
Token ID | 670 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
pe | +0.010 |
shr | +0.009 |
Tim | +0.008 |
olph | +0.008 |
pepper | +0.008 |
Neg logit contributions | |
rou | -0.021 |
amaz | -0.019 |
liness | -0.019 |
Carly | -0.018 |
downward | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -3.12e-03 |
Pos logit contributions | |
pe | +0.072 |
Tim | +0.069 |
pepper | +0.068 |
shr | +0.066 |
fought | +0.064 |
Neg logit contributions | |
rou | -0.103 |
ouncing | -0.090 |
rol | -0.087 |
liness | -0.086 |
Carly | -0.085 |
Token | carefully |
Token ID | 7773 |
Feature activation | +7.70e-02 |
Pos logit contributions | |
rou | +0.006 |
ouncing | +0.005 |
rol | +0.005 |
liness | +0.005 |
aults | +0.005 |
Neg logit contributions | |
pe | -0.004 |
pepper | -0.004 |
Tim | -0.004 |
fought | -0.003 |
tried | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -2.70e-03 |
Pos logit contributions | |
Tim | +0.093 |
pe | +0.091 |
shr | +0.088 |
pepper | +0.081 |
occasion | +0.076 |
Neg logit contributions | |
rou | -0.143 |
ouncing | -0.123 |
liness | -0.122 |
amaz | -0.117 |
aults | -0.115 |
Token | delic |
Token ID | 8675 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
rou | +0.006 |
ouncing | +0.005 |
rol | +0.005 |
liness | +0.005 |
Carly | +0.005 |
Neg logit contributions | |
pe | -0.003 |
Tim | -0.003 |
pepper | -0.002 |
shr | -0.002 |
help | -0.002 |
Token | ately |
Token ID | 1286 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
rou | +0.036 |
ouncing | +0.032 |
rol | +0.030 |
la | +0.030 |
ucking | +0.029 |
Neg logit contributions | |
Tim | -0.027 |
pe | -0.027 |
pepper | -0.026 |
shr | -0.026 |
help | -0.024 |
Token | repairing |
Token ID | 39005 |
Feature activation | +8.22e-02 |
Pos logit contributions | |
Tim | +0.171 |
shr | +0.169 |
pe | +0.165 |
pepper | +0.159 |
timer | +0.152 |
Neg logit contributions | |
rou | -0.320 |
ouncing | -0.282 |
liness | -0.280 |
aults | -0.270 |
rol | -0.269 |
Token | the |
Token ID | 262 |
Feature activation | +9.91e-03 |
Pos logit contributions | |
pe | +0.094 |
Tim | +0.088 |
shr | +0.086 |
pepper | +0.085 |
fought | +0.082 |
Neg logit contributions | |
rou | -0.163 |
ouncing | -0.142 |
liness | -0.142 |
rol | -0.139 |
Carly | -0.138 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -7.19e-03 |
Pos logit contributions | |
pe | +0.011 |
Tim | +0.011 |
shr | +0.010 |
pepper | +0.010 |
fought | +0.010 |
Neg logit contributions | |
rou | -0.020 |
ouncing | -0.017 |
liness | -0.017 |
rol | -0.016 |
Carly | -0.016 |
Token | 's |
Token ID | 338 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
rou | +0.017 |
ouncing | +0.016 |
Carly | +0.015 |
rol | +0.015 |
Debbie | +0.015 |
Neg logit contributions | |
pe | -0.006 |
pepper | -0.005 |
help | -0.005 |
Tim | -0.005 |
fought | -0.004 |
Token | wings |
Token ID | 12098 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
rou | +0.081 |
ouncing | +0.070 |
liness | +0.069 |
rol | +0.067 |
downward | +0.066 |
Neg logit contributions | |
pe | -0.048 |
Tim | -0.048 |
shr | -0.045 |
pepper | -0.044 |
fought | -0.043 |
Token | her |
Token ID | 607 |
Feature activation | -7.64e-03 |
Pos logit contributions | |
help | +0.039 |
Tim | +0.038 |
pe | +0.038 |
pepper | +0.037 |
tried | +0.034 |
Neg logit contributions | |
rou | -0.069 |
ouncing | -0.064 |
ucking | -0.059 |
rol | -0.059 |
liness | -0.057 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
rou | +0.015 |
ouncing | +0.013 |
rol | +0.013 |
liness | +0.013 |
ucking | +0.013 |
Neg logit contributions | |
pe | -0.009 |
help | -0.009 |
pepper | -0.009 |
Tim | -0.009 |
fought | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +8.66e-03 |
Pos logit contributions | |
rou | +0.037 |
ouncing | +0.032 |
rol | +0.032 |
ucking | +0.031 |
ayed | +0.031 |
Neg logit contributions | |
Tim | -0.013 |
help | -0.013 |
pe | -0.013 |
fought | -0.012 |
pepper | -0.012 |
Token | She |
Token ID | 1375 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
Tim | +0.007 |
pe | +0.006 |
pepper | +0.005 |
help | +0.005 |
shr | +0.005 |
Neg logit contributions | |
rou | -0.021 |
ouncing | -0.018 |
aults | -0.018 |
liness | -0.018 |
rol | -0.018 |
Token | w |
Token ID | 266 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
rou | +0.063 |
ouncing | +0.058 |
liness | +0.057 |
rol | +0.057 |
aults | +0.056 |
Neg logit contributions | |
tried | -0.015 |
pe | -0.015 |
fought | -0.015 |
Tim | -0.014 |
pepper | -0.013 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
rou | +0.039 |
ouncing | +0.034 |
rol | +0.033 |
liness | +0.032 |
Carly | +0.032 |
Neg logit contributions | |
pe | -0.024 |
Tim | -0.023 |
pepper | -0.023 |
shr | -0.022 |
help | -0.021 |
Token | her |
Token ID | 607 |
Feature activation | -6.76e-03 |
Pos logit contributions | |
pe | +0.129 |
Tim | +0.108 |
tried | +0.107 |
fought | +0.104 |
pepper | +0.092 |
Neg logit contributions | |
rou | -0.410 |
ouncing | -0.363 |
rol | -0.357 |
ucking | -0.352 |
aults | -0.346 |
Token | tail |
Token ID | 7894 |
Feature activation | +9.34e-03 |
Pos logit contributions | |
rou | +0.011 |
ouncing | +0.010 |
liness | +0.009 |
Carly | +0.009 |
rol | +0.009 |
Neg logit contributions | |
pe | -0.009 |
Tim | -0.009 |
pepper | -0.009 |
shr | -0.009 |
fought | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -1.61e-03 |
Pos logit contributions | |
pe | +0.011 |
tried | +0.009 |
help | +0.009 |
Tim | +0.009 |
pepper | +0.009 |
Neg logit contributions | |
rou | -0.020 |
ouncing | -0.018 |
rol | -0.017 |
Carly | -0.017 |
liness | -0.017 |
Token | bark |
Token ID | 21405 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
rou | +0.004 |
ouncing | +0.003 |
rol | +0.003 |
liness | +0.003 |
ucking | +0.003 |
Neg logit contributions | |
pe | -0.001 |
Tim | -0.001 |
pepper | -0.001 |
fought | -0.001 |
help | -0.001 |
Token | ed |
Token ID | 276 |
Feature activation | +6.00e-02 |
Pos logit contributions | |
pe | +0.032 |
Tim | +0.030 |
fought | +0.029 |
help | +0.028 |
tried | +0.028 |
Neg logit contributions | |
rou | -0.058 |
ouncing | -0.051 |
Carly | -0.049 |
liness | -0.049 |
rol | -0.048 |
Token | went |
Token ID | 1816 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
pe | +0.008 |
Tim | +0.008 |
pepper | +0.007 |
shr | +0.007 |
help | +0.006 |
Neg logit contributions | |
rou | -0.022 |
ouncing | -0.020 |
liness | -0.019 |
rol | -0.019 |
Carly | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +1.65e-03 |
Pos logit contributions | |
pe | +0.011 |
Tim | +0.010 |
pepper | +0.010 |
help | +0.010 |
shr | +0.010 |
Neg logit contributions | |
rou | -0.019 |
ouncing | -0.017 |
rol | -0.016 |
liness | -0.016 |
aults | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
pepper | +0.001 |
Tim | +0.001 |
help | +0.001 |
pe | +0.001 |
he | +0.001 |
Neg logit contributions | |
rou | -0.005 |
ouncing | -0.004 |
ucking | -0.004 |
rol | -0.004 |
idget | -0.004 |
Token | tail |
Token ID | 7894 |
Feature activation | +5.19e-03 |
Pos logit contributions | |
pe | +0.013 |
pepper | +0.013 |
Tim | +0.012 |
help | +0.012 |
shr | +0.012 |
Neg logit contributions | |
rou | -0.021 |
ouncing | -0.018 |
rol | -0.018 |
liness | -0.018 |
Carly | -0.017 |
Token | w |
Token ID | 266 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
pe | +0.004 |
Tim | +0.003 |
pepper | +0.003 |
fought | +0.003 |
shr | +0.003 |
Neg logit contributions | |
rou | -0.013 |
ouncing | -0.012 |
rol | -0.011 |
Carly | -0.011 |
liness | -0.011 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
rou | +0.034 |
ouncing | +0.029 |
rol | +0.028 |
liness | +0.027 |
ucking | +0.026 |
Neg logit contributions | |
pe | -0.029 |
Tim | -0.028 |
pepper | -0.027 |
shr | -0.027 |
fought | -0.026 |
Token | happily |
Token ID | 18177 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
pe | +0.149 |
pepper | +0.120 |
Tim | +0.119 |
tried | +0.119 |
cube | +0.114 |
Neg logit contributions | |
rou | -0.383 |
ouncing | -0.348 |
rol | -0.346 |
ucking | -0.336 |
idget | -0.324 |
Token | . |
Token ID | 13 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
pe | +0.103 |
Tim | +0.091 |
pepper | +0.090 |
help | +0.087 |
shr | +0.086 |
Neg logit contributions | |
rou | -0.223 |
ouncing | -0.198 |
rol | -0.196 |
liness | -0.194 |
Carly | -0.189 |
Token |
|
Token ID | 198 |
Feature activation | +6.41e-03 |
Pos logit contributions | |
pe | +0.011 |
Tim | +0.010 |
pepper | +0.010 |
shr | +0.010 |
fought | +0.009 |
Neg logit contributions | |
rou | -0.029 |
ouncing | -0.026 |
rol | -0.025 |
liness | -0.025 |
Carly | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +5.99e-03 |
Pos logit contributions | |
pe | +0.005 |
Tim | +0.005 |
pepper | +0.004 |
shr | +0.004 |
help | +0.004 |
Neg logit contributions | |
rou | -0.015 |
ouncing | -0.014 |
rol | -0.013 |
liness | -0.013 |
Carly | -0.013 |
Token | One |
Token ID | 3198 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
Tim | +0.006 |
pe | +0.006 |
shr | +0.005 |
pepper | +0.005 |
fought | +0.005 |
Neg logit contributions | |
rou | -0.013 |
ouncing | -0.012 |
liness | -0.012 |
rol | -0.011 |
amaz | -0.011 |
Token | as |
Token ID | 355 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
rou | +0.061 |
ouncing | +0.056 |
rol | +0.055 |
Carly | +0.054 |
liness | +0.053 |
Neg logit contributions | |
pe | -0.023 |
help | -0.021 |
pepper | -0.020 |
Tim | -0.019 |
fought | -0.019 |
Token | she |
Token ID | 673 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
help | +0.031 |
Tim | +0.030 |
pepper | +0.027 |
pe | +0.025 |
days | +0.024 |
Neg logit contributions | |
rou | -0.098 |
rol | -0.093 |
ouncing | -0.093 |
liness | -0.090 |
iet | -0.087 |
Token | z |
Token ID | 1976 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
rou | +0.043 |
ouncing | +0.038 |
rol | +0.037 |
liness | +0.037 |
Carly | +0.036 |
Neg logit contributions | |
pe | -0.019 |
Tim | -0.018 |
pepper | -0.018 |
shr | -0.017 |
fought | -0.016 |
Token | ig |
Token ID | 328 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
rou | +0.057 |
liness | +0.049 |
ouncing | +0.049 |
rol | +0.047 |
Carly | +0.046 |
Neg logit contributions | |
pe | -0.056 |
Tim | -0.053 |
help | -0.052 |
shr | -0.052 |
pepper | -0.052 |
Token | z |
Token ID | 1976 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
pe | +0.028 |
help | +0.024 |
Tim | +0.024 |
tried | +0.024 |
fought | +0.023 |
Neg logit contributions | |
rou | -0.033 |
amaz | -0.032 |
liness | -0.030 |
Carly | -0.029 |
Spanish | -0.028 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
rou | +0.054 |
liness | +0.046 |
ouncing | +0.046 |
rol | +0.045 |
Carly | +0.043 |
Neg logit contributions | |
pe | -0.059 |
Tim | -0.055 |
shr | -0.054 |
pepper | -0.054 |
help | -0.054 |
Token | around |
Token ID | 1088 |
Feature activation | -5.72e-03 |
Pos logit contributions | |
pe | +0.146 |
help | +0.117 |
Tim | +0.116 |
pepper | +0.109 |
tried | +0.109 |
Neg logit contributions | |
rou | -0.386 |
ouncing | -0.357 |
rol | -0.343 |
aj | -0.339 |
ucking | -0.334 |
Token | the |
Token ID | 262 |
Feature activation | +9.38e-03 |
Pos logit contributions | |
rou | +0.010 |
ouncing | +0.009 |
liness | +0.009 |
rol | +0.009 |
Carly | +0.009 |
Neg logit contributions | |
pe | -0.007 |
Tim | -0.007 |
pepper | -0.007 |
shr | -0.007 |
fought | -0.007 |
Token | room |
Token ID | 2119 |
Feature activation | +1.89e-03 |
Pos logit contributions | |
pe | +0.011 |
Tim | +0.011 |
pepper | +0.010 |
shr | +0.010 |
help | +0.010 |
Neg logit contributions | |
rou | -0.018 |
ouncing | -0.016 |
rol | -0.016 |
liness | -0.015 |
Carly | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +7.00e-03 |
Pos logit contributions | |
pe | +0.001 |
help | +0.001 |
pepper | +0.001 |
Tim | +0.001 |
fought | +0.001 |
Neg logit contributions | |
rou | -0.005 |
ouncing | -0.005 |
rol | -0.004 |
Carly | -0.004 |
liness | -0.004 |
Token | She |
Token ID | 1375 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
pe | +0.008 |
pepper | +0.007 |
Tim | +0.007 |
shr | +0.007 |
fought | +0.007 |
Neg logit contributions | |
rou | -0.014 |
ouncing | -0.012 |
rol | -0.012 |
liness | -0.012 |
Carly | -0.012 |
Token | dog |
Token ID | 3290 |
Feature activation | +3.98e-03 |
Pos logit contributions | |
rou | +0.044 |
ouncing | +0.039 |
rol | +0.038 |
liness | +0.037 |
aults | +0.037 |
Neg logit contributions | |
pe | -0.020 |
pepper | -0.018 |
Tim | -0.018 |
help | -0.017 |
fought | -0.017 |
Token | !" |
Token ID | 2474 |
Feature activation | -7.10e-04 |
Pos logit contributions | |
pe | +0.005 |
fought | +0.004 |
pepper | +0.004 |
help | +0.004 |
Tim | +0.004 |
Neg logit contributions | |
rou | -0.008 |
Carly | -0.007 |
amaz | -0.007 |
ouncing | -0.007 |
rol | -0.007 |
Token | Max |
Token ID | 5436 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
rou | +0.001 |
Carly | +0.001 |
ouncing | +0.001 |
Lauren | +0.001 |
liness | +0.001 |
Neg logit contributions | |
pe | -0.001 |
shr | -0.001 |
pepper | -0.001 |
Tim | -0.001 |
timer | -0.001 |
Token | just |
Token ID | 655 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
rou | +0.054 |
ouncing | +0.049 |
liness | +0.048 |
rol | +0.048 |
Carly | +0.048 |
Neg logit contributions | |
pe | -0.015 |
Tim | -0.013 |
pepper | -0.013 |
shr | -0.013 |
help | -0.011 |
Token | w |
Token ID | 266 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
rou | +0.021 |
ouncing | +0.019 |
rol | +0.018 |
la | +0.018 |
liness | +0.017 |
Neg logit contributions | |
pe | -0.011 |
shr | -0.011 |
Tim | -0.011 |
� | -0.011 |
pepper | -0.011 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
rou | +0.033 |
ayed | +0.029 |
ouncing | +0.029 |
rol | +0.027 |
ucking | +0.027 |
Neg logit contributions | |
pepper | -0.026 |
Tim | -0.026 |
shr | -0.026 |
help | -0.024 |
pe | -0.024 |
Token | his |
Token ID | 465 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
pe | +0.122 |
Tim | +0.112 |
pepper | +0.105 |
shr | +0.099 |
fought | +0.098 |
Neg logit contributions | |
rou | -0.399 |
ouncing | -0.360 |
rol | -0.353 |
liness | -0.347 |
aults | -0.342 |
Token | tail |
Token ID | 7894 |
Feature activation | +7.90e-03 |
Pos logit contributions | |
shr | +0.032 |
occasion | +0.030 |
pepper | +0.030 |
Tim | +0.030 |
� | +0.029 |
Neg logit contributions | |
rou | -0.020 |
ouncing | -0.018 |
aults | -0.016 |
downward | -0.016 |
iet | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -1.59e-03 |
Pos logit contributions | |
pe | +0.009 |
Tim | +0.008 |
pepper | +0.008 |
help | +0.007 |
fought | +0.007 |
Neg logit contributions | |
rou | -0.016 |
ouncing | -0.015 |
rol | -0.014 |
liness | -0.014 |
Carly | -0.014 |
Token | l |
Token ID | 300 |
Feature activation | -1.51e-03 |
Pos logit contributions | |
rou | +0.003 |
ouncing | +0.003 |
Carly | +0.003 |
liness | +0.003 |
downward | +0.003 |
Neg logit contributions | |
pe | -0.002 |
shr | -0.001 |
Tim | -0.001 |
pepper | -0.001 |
help | -0.001 |
Token | icked |
Token ID | 9484 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
rou | +0.002 |
rol | +0.001 |
ouncing | +0.001 |
ucking | +0.001 |
ayed | +0.001 |
Neg logit contributions | |
pepper | -0.003 |
Tim | -0.003 |
pe | -0.003 |
shr | -0.003 |
occasion | -0.003 |
Token | you |
Token ID | 345 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
rou | +0.000 |
Carly | +0.000 |
Riley | +0.000 |
aults | +0.000 |
liness | +0.000 |
Neg logit contributions | |
pepper | -0.000 |
pe | -0.000 |
timer | -0.000 |
fought | -0.000 |
Tim | -0.000 |
Token | ." |
Token ID | 526 |
Feature activation | -6.90e-04 |
Pos logit contributions | |
help | +0.042 |
pe | +0.038 |
pepper | +0.038 |
Tim | +0.037 |
days | +0.033 |
Neg logit contributions | |
rou | -0.088 |
ouncing | -0.083 |
ucking | -0.080 |
amaz | -0.080 |
idget | -0.079 |
Token | And |
Token ID | 843 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
rou | +0.001 |
ouncing | +0.001 |
rol | +0.001 |
liness | +0.001 |
Carly | +0.001 |
Neg logit contributions | |
pe | -0.001 |
Tim | -0.001 |
pepper | -0.001 |
shr | -0.001 |
fought | -0.001 |
Token | Rover |
Token ID | 36718 |
Feature activation | -9.86e-02 |
Pos logit contributions | |
pe | +0.062 |
help | +0.059 |
Tim | +0.058 |
tried | +0.054 |
pepper | +0.054 |
Neg logit contributions | |
rou | -0.173 |
ouncing | -0.152 |
idget | -0.148 |
rol | -0.148 |
aj | -0.148 |
Token | w |
Token ID | 266 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
rou | +0.227 |
ouncing | +0.204 |
rol | +0.202 |
ucking | +0.197 |
aults | +0.196 |
Neg logit contributions | |
pe | -0.088 |
pepper | -0.076 |
Tim | -0.076 |
fought | -0.074 |
tried | -0.073 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
rou | +0.041 |
ouncing | +0.036 |
rol | +0.035 |
ayed | +0.033 |
liness | +0.033 |
Neg logit contributions | |
Tim | -0.021 |
pe | -0.021 |
pepper | -0.021 |
shr | -0.020 |
help | -0.019 |
Token | his |
Token ID | 465 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
pe | +0.099 |
Tim | +0.077 |
fought | +0.077 |
pepper | +0.075 |
tried | +0.073 |
Neg logit contributions | |
rou | -0.432 |
ouncing | -0.392 |
rol | -0.385 |
ucking | -0.379 |
aults | -0.372 |
Token | tail |
Token ID | 7894 |
Feature activation | +7.14e-03 |
Pos logit contributions | |
shr | +0.024 |
Tim | +0.024 |
pepper | +0.023 |
pe | +0.023 |
fought | +0.023 |
Neg logit contributions | |
rou | -0.023 |
ouncing | -0.019 |
liness | -0.018 |
downward | -0.018 |
Carly | -0.018 |
Token | happily |
Token ID | 18177 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
pe | +0.008 |
help | +0.007 |
pepper | +0.007 |
Tim | +0.006 |
tried | +0.006 |
Neg logit contributions | |
rou | -0.015 |
ouncing | -0.014 |
Carly | -0.013 |
liness | -0.013 |
rol | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
Tim | +0.128 |
pe | +0.124 |
pepper | +0.124 |
shr | +0.121 |
Blood | +0.112 |
Neg logit contributions | |
rou | -0.197 |
ouncing | -0.174 |
rol | -0.165 |
aults | -0.164 |
liness | -0.163 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
pe | +0.012 |
Tim | +0.011 |
pepper | +0.011 |
shr | +0.010 |
help | +0.010 |
Neg logit contributions | |
rou | -0.033 |
ouncing | -0.030 |
rol | -0.029 |
liness | -0.029 |
aults | -0.029 |
Token | around |
Token ID | 1088 |
Feature activation | -6.96e-03 |
Pos logit contributions | |
pe | +0.004 |
Tim | +0.003 |
help | +0.003 |
pepper | +0.003 |
fought | +0.003 |
Neg logit contributions | |
rou | -0.006 |
ouncing | -0.005 |
rol | -0.005 |
liness | -0.005 |
ucking | -0.005 |
Token | on |
Token ID | 319 |
Feature activation | +1.47e-03 |
Pos logit contributions | |
rou | +0.014 |
ouncing | +0.013 |
rol | +0.012 |
liness | +0.012 |
Carly | +0.012 |
Neg logit contributions | |
pe | -0.008 |
Tim | -0.007 |
pepper | -0.007 |
shr | -0.007 |
help | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +5.29e-03 |
Pos logit contributions | |
pe | +0.001 |
Tim | +0.001 |
pepper | +0.001 |
help | +0.001 |
fought | +0.001 |
Neg logit contributions | |
rou | -0.004 |
ouncing | -0.003 |
rol | -0.003 |
ucking | -0.003 |
liness | -0.003 |
Token | tight |
Token ID | 5381 |
Feature activation | -7.13e-03 |
Pos logit contributions | |
pe | +0.006 |
Tim | +0.006 |
pepper | +0.006 |
shr | +0.006 |
help | +0.005 |
Neg logit contributions | |
rou | -0.010 |
ouncing | -0.009 |
rol | -0.009 |
liness | -0.009 |
Carly | -0.009 |
Token | ro |
Token ID | 305 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
rou | +0.015 |
ouncing | +0.013 |
rol | +0.013 |
liness | +0.013 |
Carly | +0.012 |
Neg logit contributions | |
pe | -0.008 |
Tim | -0.007 |
pepper | -0.007 |
shr | -0.007 |
help | -0.007 |
Token | pe |
Token ID | 431 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Tim | +0.024 |
pepper | +0.024 |
shr | +0.022 |
pe | +0.020 |
help | +0.020 |
Neg logit contributions | |
rou | -0.130 |
ouncing | -0.119 |
liness | -0.116 |
aults | -0.115 |
Carly | -0.115 |
Token | , |
Token ID | 11 |
Feature activation | +2.51e-03 |
Pos logit contributions | |
Tim | +0.195 |
pepper | +0.178 |
shr | +0.176 |
pe | +0.172 |
� | +0.171 |
Neg logit contributions | |
rou | -0.307 |
ouncing | -0.274 |
aults | -0.266 |
rol | -0.258 |
liness | -0.256 |
Token | he |
Token ID | 339 |
Feature activation | +9.99e-03 |
Pos logit contributions | |
pe | +0.003 |
pepper | +0.003 |
help | +0.003 |
Tim | +0.003 |
tried | +0.003 |
Neg logit contributions | |
rou | -0.005 |
ouncing | -0.005 |
rol | -0.004 |
liness | -0.004 |
ucking | -0.004 |
Token | could |
Token ID | 714 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
pe | +0.009 |
Tim | +0.009 |
pepper | +0.008 |
shr | +0.008 |
help | +0.008 |
Neg logit contributions | |
rou | -0.022 |
ouncing | -0.020 |
rol | -0.019 |
liness | -0.019 |
Carly | -0.019 |
Token | even |
Token ID | 772 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
pe | +0.049 |
help | +0.046 |
Tim | +0.042 |
fought | +0.040 |
pepper | +0.040 |
Neg logit contributions | |
rou | -0.117 |
ouncing | -0.107 |
liness | -0.103 |
rol | -0.103 |
ucking | -0.099 |
Token | turn |
Token ID | 1210 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
pe | +0.043 |
Tim | +0.042 |
� | +0.040 |
pepper | +0.040 |
shr | +0.038 |
Neg logit contributions | |
rou | -0.061 |
aults | -0.056 |
downward | -0.054 |
liness | -0.053 |
Carly | -0.053 |
Token | was |
Token ID | 373 |
Feature activation | +3.89e-03 |
Pos logit contributions | |
rou | +0.043 |
ouncing | +0.039 |
rol | +0.039 |
liness | +0.038 |
aults | +0.037 |
Neg logit contributions | |
pe | -0.014 |
Tim | -0.013 |
pepper | -0.013 |
fought | -0.013 |
tried | -0.012 |
Token | very |
Token ID | 845 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
hurt | +0.002 |
tried | +0.002 |
finally | +0.002 |
help | +0.002 |
be | +0.002 |
Neg logit contributions | |
ouncing | -0.010 |
rou | -0.010 |
iet | -0.009 |
ancing | -0.009 |
liness | -0.009 |
Token | happy |
Token ID | 3772 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
help | +0.048 |
tried | +0.046 |
fought | +0.045 |
pe | +0.045 |
pepper | +0.044 |
Neg logit contributions | |
rou | -0.103 |
liness | -0.098 |
rol | -0.096 |
ouncing | -0.095 |
la | -0.095 |
Token | and |
Token ID | 290 |
Feature activation | -1.62e-03 |
Pos logit contributions | |
pepper | +0.034 |
pe | +0.034 |
help | +0.033 |
Tim | +0.031 |
tried | +0.031 |
Neg logit contributions | |
rou | -0.090 |
rol | -0.085 |
idget | -0.080 |
ouncing | -0.080 |
iet | -0.079 |
Token | w |
Token ID | 266 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
rou | +0.004 |
rol | +0.004 |
ouncing | +0.003 |
liness | +0.003 |
ucking | +0.003 |
Neg logit contributions | |
pe | -0.001 |
pepper | -0.001 |
Tim | -0.001 |
help | -0.001 |
tried | -0.001 |
Token | agged |
Token ID | 14655 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
rou | +0.040 |
ouncing | +0.035 |
rol | +0.034 |
ayed | +0.033 |
ucking | +0.033 |
Neg logit contributions | |
pe | -0.022 |
Tim | -0.022 |
pepper | -0.022 |
shr | -0.021 |
help | -0.020 |
Token | his |
Token ID | 465 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
pe | +0.056 |
tried | +0.043 |
pepper | +0.039 |
fought | +0.038 |
Tim | +0.037 |
Neg logit contributions | |
rou | -0.471 |
ouncing | -0.430 |
rol | -0.428 |
ucking | -0.419 |
aults | -0.410 |
Token | tail |
Token ID | 7894 |
Feature activation | +6.85e-03 |
Pos logit contributions | |
shr | +0.023 |
Tim | +0.022 |
pe | +0.022 |
fought | +0.021 |
pepper | +0.021 |
Neg logit contributions | |
rou | -0.017 |
ouncing | -0.015 |
aj | -0.014 |
amaz | -0.014 |
capt | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
pe | +0.005 |
pepper | +0.005 |
Tim | +0.005 |
help | +0.005 |
fought | +0.004 |
Neg logit contributions | |
rou | -0.016 |
ouncing | -0.015 |
Carly | -0.014 |
rol | -0.014 |
liness | -0.014 |
Token | He |
Token ID | 679 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
pe | +0.011 |
pepper | +0.010 |
shr | +0.010 |
Tim | +0.010 |
fought | +0.010 |
Neg logit contributions | |
rou | -0.024 |
ouncing | -0.022 |
rol | -0.021 |
liness | -0.021 |
Carly | -0.021 |
Token | shared |
Token ID | 4888 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
pe | +0.010 |
tried | +0.010 |
Tim | +0.010 |
fought | +0.010 |
pepper | +0.010 |
Neg logit contributions | |
rou | -0.036 |
ouncing | -0.033 |
rol | -0.033 |
liness | -0.033 |
ucking | -0.033 |
Token | it |
Token ID | 340 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
rou | +0.026 |
ouncing | +0.023 |
rol | +0.023 |
amaz | +0.023 |
liness | +0.022 |
Neg logit contributions | |
pe | -0.009 |
pepper | -0.009 |
Tim | -0.009 |
help | -0.008 |
tried | -0.008 |
Token | on |
Token ID | 319 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
pe | +0.033 |
Tim | +0.032 |
pepper | +0.031 |
shr | +0.030 |
fought | +0.030 |
Neg logit contributions | |
rou | -0.042 |
ouncing | -0.036 |
rol | -0.035 |
liness | -0.035 |
Carly | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -5.66e-03 |
Pos logit contributions | |
rou | +0.035 |
ouncing | +0.031 |
rol | +0.030 |
liness | +0.029 |
Carly | +0.029 |
Neg logit contributions | |
pe | -0.023 |
Tim | -0.022 |
pepper | -0.022 |
shr | -0.021 |
help | -0.021 |
Token | chickens |
Token ID | 25972 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
rou | +0.009 |
liness | +0.008 |
downward | +0.008 |
aults | +0.008 |
Carly | +0.008 |
Neg logit contributions | |
Tim | -0.008 |
fought | -0.008 |
shr | -0.008 |
pe | -0.008 |
Blood | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -9.08e-03 |
Pos logit contributions | |
rou | +0.109 |
ouncing | +0.098 |
rol | +0.096 |
Carly | +0.095 |
liness | +0.094 |
Neg logit contributions | |
pe | -0.032 |
pepper | -0.028 |
Tim | -0.028 |
help | -0.027 |
shr | -0.025 |
Token | Fel |
Token ID | 13937 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
rou | +0.019 |
aults | +0.016 |
rol | +0.016 |
liness | +0.016 |
ouncing | +0.015 |
Neg logit contributions | |
help | -0.010 |
pe | -0.009 |
But | -0.009 |
shr | -0.009 |
Tim | -0.009 |
Token | icity |
Token ID | 8467 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
rou | +0.240 |
ouncing | +0.198 |
rol | +0.191 |
liness | +0.188 |
Carly | +0.181 |
Neg logit contributions | |
pe | -0.307 |
Tim | -0.294 |
pepper | -0.290 |
shr | -0.285 |
help | -0.280 |
Token | blinked |
Token ID | 40360 |
Feature activation | +2.78e-03 |
Pos logit contributions | |
rou | +0.092 |
ouncing | +0.082 |
liness | +0.079 |
rol | +0.079 |
Carly | +0.077 |
Neg logit contributions | |
pe | -0.034 |
Tim | -0.031 |
pepper | -0.030 |
shr | -0.029 |
help | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
help | +0.002 |
tried | +0.002 |
pe | +0.002 |
pepper | +0.002 |
Tim | +0.002 |
Neg logit contributions | |
rou | -0.006 |
ouncing | -0.006 |
rol | -0.006 |
aults | -0.005 |
packs | -0.005 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.04e-03 |
Pos logit contributions | |
rou | +0.053 |
Carly | +0.045 |
ouncing | +0.045 |
liness | +0.044 |
amaz | +0.044 |
Neg logit contributions | |
pe | -0.032 |
� | -0.028 |
shr | -0.028 |
Tim | -0.028 |
pepper | -0.027 |
Token | that |
Token ID | 326 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
pepper | +0.002 |
help | +0.002 |
pe | +0.002 |
Tim | +0.002 |
tried | +0.001 |
Neg logit contributions | |
rou | -0.005 |
ouncing | -0.004 |
aults | -0.004 |
rol | -0.004 |
aj | -0.004 |
Token | so |
Token ID | 523 |
Feature activation | +5.03e-03 |
Pos logit contributions | |
rou | +0.063 |
ouncing | +0.060 |
rol | +0.058 |
idget | +0.057 |
ucking | +0.057 |
Neg logit contributions | |
help | -0.023 |
pepper | -0.022 |
pe | -0.020 |
Tim | -0.019 |
he | -0.019 |
Token | big |
Token ID | 1263 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
pe | +0.006 |
Tim | +0.005 |
pepper | +0.005 |
help | +0.005 |
shr | +0.005 |
Neg logit contributions | |
rou | -0.010 |
ouncing | -0.009 |
rol | -0.009 |
liness | -0.009 |
aults | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
rou | +0.103 |
ouncing | +0.096 |
rol | +0.095 |
ucking | +0.092 |
idget | +0.091 |
Neg logit contributions | |
pe | -0.030 |
help | -0.029 |
pepper | -0.029 |
tried | -0.027 |
Tim | -0.026 |
Token | special |
Token ID | 2041 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
rou | +0.041 |
ouncing | +0.037 |
rol | +0.036 |
liness | +0.035 |
ucking | +0.034 |
Neg logit contributions | |
pe | -0.026 |
Tim | -0.025 |
pepper | -0.025 |
help | -0.024 |
shr | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -6.85e-03 |
Pos logit contributions | |
rou | +0.082 |
ouncing | +0.076 |
rol | +0.074 |
idget | +0.072 |
ucking | +0.072 |
Neg logit contributions | |
pe | -0.019 |
help | -0.018 |
pepper | -0.017 |
Tim | -0.014 |
tried | -0.013 |
Token | Fel |
Token ID | 13937 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
rou | +0.015 |
aults | +0.013 |
rol | +0.013 |
ouncing | +0.013 |
ayed | +0.012 |
Neg logit contributions | |
Tim | -0.006 |
But | -0.006 |
help | -0.006 |
pe | -0.006 |
shr | -0.006 |
Token | icity |
Token ID | 8467 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
rou | +0.237 |
ouncing | +0.196 |
rol | +0.189 |
liness | +0.186 |
Carly | +0.179 |
Neg logit contributions | |
pe | -0.301 |
Tim | -0.287 |
pepper | -0.283 |
shr | -0.279 |
help | -0.274 |
Token | was |
Token ID | 373 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
rou | +0.082 |
ouncing | +0.075 |
rol | +0.075 |
downward | +0.074 |
liness | +0.073 |
Neg logit contributions | |
pe | -0.026 |
tried | -0.025 |
fought | -0.024 |
pepper | -0.022 |
Tim | -0.022 |
Token | so |
Token ID | 523 |
Feature activation | +3.22e-03 |
Pos logit contributions | |
ouncing | +0.032 |
rou | +0.031 |
aj | +0.030 |
iet | +0.030 |
liness | +0.030 |
Neg logit contributions | |
finally | -0.009 |
tried | -0.009 |
help | -0.009 |
measured | -0.008 |
had | -0.007 |
Token | proud |
Token ID | 6613 |
Feature activation | -4.85e-04 |
Pos logit contributions | |
help | +0.003 |
tried | +0.003 |
pe | +0.003 |
measured | +0.003 |
Tim | +0.003 |
Neg logit contributions | |
rou | -0.007 |
liness | -0.007 |
ouncing | -0.006 |
rol | -0.006 |
aters | -0.006 |
Token | of |
Token ID | 286 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
rou | +0.001 |
aults | +0.001 |
rol | +0.001 |
idget | +0.001 |
liness | +0.001 |
Neg logit contributions | |
he | -0.000 |
help | -0.000 |
pepper | -0.000 |
Tim | -0.000 |
tried | -0.000 |
Token | was |
Token ID | 373 |
Feature activation | -7.65e-03 |
Pos logit contributions | |
rou | +0.087 |
ouncing | +0.078 |
rol | +0.076 |
liness | +0.076 |
Carly | +0.075 |
Neg logit contributions | |
pe | -0.027 |
Tim | -0.025 |
pepper | -0.024 |
shr | -0.022 |
help | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | -4.62e-03 |
Pos logit contributions | |
rou | +0.016 |
ouncing | +0.014 |
rol | +0.014 |
downward | +0.014 |
Carly | +0.013 |
Neg logit contributions | |
pe | -0.008 |
Tim | -0.007 |
pepper | -0.007 |
shr | -0.007 |
fought | -0.007 |
Token | little |
Token ID | 1310 |
Feature activation | -7.33e-03 |
Pos logit contributions | |
rou | +0.008 |
ouncing | +0.007 |
downward | +0.007 |
Carly | +0.007 |
aults | +0.007 |
Neg logit contributions | |
pe | -0.006 |
shr | -0.006 |
fought | -0.006 |
Tim | -0.006 |
pepper | -0.005 |
Token | girl |
Token ID | 2576 |
Feature activation | +4.30e-04 |
Pos logit contributions | |
rou | +0.014 |
ouncing | +0.012 |
downward | +0.011 |
Carly | +0.011 |
liness | +0.011 |
Neg logit contributions | |
pe | -0.010 |
shr | -0.009 |
fought | -0.008 |
Tim | -0.008 |
help | -0.008 |
Token | called |
Token ID | 1444 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
pe | +0.000 |
shr | +0.000 |
pepper | +0.000 |
Tim | +0.000 |
� | +0.000 |
Neg logit contributions | |
rou | -0.001 |
ouncing | -0.001 |
liness | -0.001 |
ucking | -0.001 |
iet | -0.001 |
Token | Fel |
Token ID | 13937 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
rou | +0.066 |
ouncing | +0.058 |
rol | +0.057 |
liness | +0.056 |
Carly | +0.056 |
Neg logit contributions | |
pe | -0.033 |
Tim | -0.031 |
pepper | -0.031 |
shr | -0.030 |
fought | -0.029 |
Token | icia |
Token ID | 33577 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
rou | +0.193 |
la | +0.152 |
ouncing | +0.149 |
Carly | +0.147 |
iet | +0.143 |
Neg logit contributions | |
Tim | -0.329 |
pepper | -0.328 |
pe | -0.326 |
shr | -0.314 |
fought | -0.313 |
Token | . |
Token ID | 13 |
Feature activation | -3.64e-03 |
Pos logit contributions | |
rou | +0.120 |
ouncing | +0.108 |
rol | +0.105 |
liness | +0.105 |
Carly | +0.104 |
Neg logit contributions | |
pe | -0.038 |
Tim | -0.035 |
pepper | -0.034 |
shr | -0.032 |
fought | -0.031 |
Token | She |
Token ID | 1375 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
rou | +0.007 |
aults | +0.006 |
ouncing | +0.006 |
rol | +0.006 |
ucking | +0.006 |
Neg logit contributions | |
Tim | -0.005 |
help | -0.004 |
shr | -0.004 |
pe | -0.004 |
But | -0.004 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
rou | +0.053 |
ouncing | +0.045 |
capt | +0.044 |
iet | +0.042 |
downward | +0.042 |
Neg logit contributions | |
pe | -0.054 |
timer | -0.054 |
pepper | -0.052 |
vinegar | -0.050 |
� | -0.050 |
Token | to |
Token ID | 284 |
Feature activation | +2.57e-03 |
Pos logit contributions | |
pe | +0.013 |
Tim | +0.012 |
pepper | +0.012 |
shr | +0.011 |
help | +0.011 |
Neg logit contributions | |
rou | -0.019 |
ouncing | -0.017 |
rol | -0.016 |
liness | -0.016 |
Carly | -0.016 |
Token | said |
Token ID | 531 |
Feature activation | +2.74e-04 |
Pos logit contributions | |
rou | +0.045 |
ouncing | +0.041 |
liness | +0.040 |
Carly | +0.039 |
rol | +0.039 |
Neg logit contributions | |
pe | -0.017 |
Tim | -0.015 |
shr | -0.015 |
pepper | -0.014 |
fought | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
pe | +0.000 |
Tim | +0.000 |
pepper | +0.000 |
shr | +0.000 |
fought | +0.000 |
Neg logit contributions | |
rou | -0.001 |
ouncing | -0.000 |
rol | -0.000 |
liness | -0.000 |
Carly | -0.000 |
Token | was |
Token ID | 373 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
pe | +0.023 |
Tim | +0.022 |
pepper | +0.021 |
shr | +0.020 |
fought | +0.019 |
Neg logit contributions | |
rou | -0.061 |
ouncing | -0.055 |
liness | -0.054 |
rol | -0.053 |
downward | -0.052 |
Token | not |
Token ID | 407 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
rou | +0.021 |
ouncing | +0.018 |
rol | +0.017 |
liness | +0.017 |
Carly | +0.017 |
Neg logit contributions | |
pe | -0.021 |
Tim | -0.021 |
pepper | -0.020 |
shr | -0.020 |
fought | -0.019 |
Token | his |
Token ID | 465 |
Feature activation | -3.10e-03 |
Pos logit contributions | |
rou | +0.060 |
ouncing | +0.052 |
liness | +0.050 |
rol | +0.050 |
downward | +0.049 |
Neg logit contributions | |
pe | -0.039 |
Tim | -0.037 |
pepper | -0.035 |
shr | -0.035 |
fought | -0.033 |
Token | fault |
Token ID | 8046 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
rou | +0.005 |
rol | +0.004 |
ouncing | +0.004 |
liness | +0.004 |
Carly | +0.004 |
Neg logit contributions | |
pe | -0.005 |
help | -0.005 |
pepper | -0.005 |
fought | -0.005 |
Tim | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +2.78e-03 |
Pos logit contributions | |
rou | +0.385 |
ouncing | +0.346 |
rol | +0.333 |
liness | +0.333 |
downward | +0.328 |
Neg logit contributions | |
pe | -0.142 |
Tim | -0.132 |
shr | -0.130 |
pepper | -0.128 |
fought | -0.119 |
Token | She |
Token ID | 1375 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
Tim | +0.003 |
pepper | +0.003 |
pe | +0.003 |
Blood | +0.003 |
help | +0.003 |
Neg logit contributions | |
rou | -0.006 |
rol | -0.005 |
aults | -0.005 |
ouncing | -0.005 |
liness | -0.005 |
Token | said |
Token ID | 531 |
Feature activation | -8.28e-04 |
Pos logit contributions | |
rou | +0.079 |
ouncing | +0.067 |
la | +0.066 |
liness | +0.064 |
Carly | +0.064 |
Neg logit contributions | |
pe | -0.037 |
shr | -0.035 |
Tim | -0.032 |
� | -0.031 |
pepper | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | -5.14e-03 |
Pos logit contributions | |
rou | +0.002 |
rol | +0.002 |
ouncing | +0.002 |
liness | +0.002 |
aults | +0.002 |
Neg logit contributions | |
pe | -0.001 |
pepper | -0.001 |
Tim | -0.001 |
help | -0.001 |
fought | -0.000 |
Token | bird |
Token ID | 6512 |
Feature activation | +1.36e-03 |
Pos logit contributions | |
rou | +0.011 |
ouncing | +0.009 |
downward | +0.009 |
liness | +0.009 |
Carly | +0.009 |
Neg logit contributions | |
pe | -0.005 |
Tim | -0.005 |
shr | -0.005 |
pepper | -0.005 |
fought | -0.005 |
Token | wrong |
Token ID | 2642 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
rou | +0.068 |
rol | +0.061 |
ouncing | +0.061 |
Carly | +0.060 |
liness | +0.059 |
Neg logit contributions | |
pe | -0.025 |
pepper | -0.024 |
Tim | -0.024 |
help | -0.023 |
fought | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
rou | +0.123 |
rol | +0.113 |
ouncing | +0.112 |
Carly | +0.109 |
aults | +0.109 |
Neg logit contributions | |
pe | -0.026 |
Tim | -0.019 |
help | -0.019 |
pepper | -0.018 |
fought | -0.016 |
Token | it |
Token ID | 340 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
rou | +0.054 |
ouncing | +0.047 |
liness | +0.046 |
rol | +0.044 |
amaz | +0.043 |
Neg logit contributions | |
pe | -0.023 |
shr | -0.020 |
Tim | -0.020 |
� | -0.019 |
pepper | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
pe | +0.021 |
Tim | +0.019 |
pepper | +0.019 |
shr | +0.018 |
help | +0.017 |
Neg logit contributions | |
rou | -0.055 |
ouncing | -0.048 |
liness | -0.047 |
rol | -0.047 |
downward | -0.045 |
Token | his |
Token ID | 465 |
Feature activation | -6.35e-03 |
Pos logit contributions | |
rou | +0.031 |
liness | +0.025 |
ouncing | +0.025 |
rol | +0.025 |
downward | +0.024 |
Neg logit contributions | |
pe | -0.022 |
Tim | -0.021 |
shr | -0.020 |
pepper | -0.019 |
� | -0.019 |
Token | fault |
Token ID | 8046 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
rou | +0.012 |
ouncing | +0.011 |
rol | +0.011 |
liness | +0.010 |
Carly | +0.010 |
Neg logit contributions | |
pe | -0.008 |
pepper | -0.008 |
help | -0.008 |
Tim | -0.007 |
shr | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -8.57e-03 |
Pos logit contributions | |
rou | +0.385 |
ouncing | +0.346 |
rol | +0.338 |
liness | +0.337 |
Carly | +0.332 |
Neg logit contributions | |
pe | -0.143 |
Tim | -0.131 |
pepper | -0.127 |
help | -0.119 |
shr | -0.116 |
Token | He |
Token ID | 679 |
Feature activation | -7.01e-03 |
Pos logit contributions | |
rou | +0.015 |
amaz | +0.013 |
Carly | +0.013 |
ouncing | +0.013 |
liness | +0.013 |
Neg logit contributions | |
pe | -0.011 |
ts | -0.011 |
pepper | -0.011 |
shr | -0.010 |
fought | -0.010 |
Token | learned |
Token ID | 4499 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
rou | +0.015 |
ouncing | +0.013 |
liness | +0.013 |
rol | +0.013 |
la | +0.013 |
Neg logit contributions | |
pe | -0.007 |
Tim | -0.006 |
shr | -0.006 |
pepper | -0.006 |
� | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
pe | +0.015 |
Tim | +0.014 |
shr | +0.014 |
pepper | +0.014 |
fought | +0.013 |
Neg logit contributions | |
rou | -0.026 |
ouncing | -0.022 |
Carly | -0.022 |
liness | -0.022 |
rol | -0.021 |
Token | lesson |
Token ID | 11483 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
rou | +0.020 |
ouncing | +0.017 |
downward | +0.016 |
liness | +0.016 |
Carly | +0.015 |
Neg logit contributions | |
pe | -0.017 |
Tim | -0.016 |
fought | -0.016 |
shr | -0.015 |
pepper | -0.015 |
Token | He |
Token ID | 679 |
Feature activation | -4.25e-03 |
Pos logit contributions | |
rou | +0.010 |
liness | +0.010 |
Carly | +0.010 |
ouncing | +0.009 |
about | +0.009 |
Neg logit contributions | |
ts | -0.007 |
timer | -0.007 |
Tim | -0.007 |
pe | -0.007 |
pepper | -0.007 |
Token | realized |
Token ID | 6939 |
Feature activation | -8.76e-03 |
Pos logit contributions | |
rou | +0.008 |
la | +0.007 |
Carly | +0.007 |
abel | +0.007 |
liness | +0.007 |
Neg logit contributions | |
� | -0.005 |
pe | -0.005 |
Tim | -0.004 |
pepper | -0.004 |
shr | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
rou | +0.021 |
ouncing | +0.019 |
rol | +0.019 |
liness | +0.019 |
aults | +0.018 |
Neg logit contributions | |
pe | -0.007 |
Tim | -0.006 |
pepper | -0.006 |
help | -0.006 |
fought | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
pe | +0.019 |
Tim | +0.017 |
pepper | +0.017 |
shr | +0.015 |
fought | +0.015 |
Neg logit contributions | |
rou | -0.059 |
ouncing | -0.054 |
rol | -0.053 |
liness | -0.052 |
Carly | -0.052 |
Token | his |
Token ID | 465 |
Feature activation | -2.94e-03 |
Pos logit contributions | |
rou | +0.029 |
ouncing | +0.025 |
liness | +0.025 |
rol | +0.024 |
Carly | +0.024 |
Neg logit contributions | |
pe | -0.016 |
Tim | -0.015 |
pepper | -0.014 |
shr | -0.014 |
fought | -0.013 |
Token | fault |
Token ID | 8046 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
rou | +0.005 |
rol | +0.004 |
ouncing | +0.004 |
liness | +0.004 |
ucking | +0.004 |
Neg logit contributions | |
pe | -0.004 |
pepper | -0.004 |
help | -0.004 |
Tim | -0.004 |
cube | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
rou | +0.340 |
ouncing | +0.300 |
liness | +0.295 |
rol | +0.295 |
Carly | +0.287 |
Neg logit contributions | |
pe | -0.188 |
Tim | -0.175 |
pepper | -0.172 |
help | -0.164 |
fought | -0.158 |
Token | he |
Token ID | 339 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
rou | +0.057 |
ouncing | +0.052 |
rol | +0.052 |
liness | +0.051 |
Carly | +0.050 |
Neg logit contributions | |
pe | -0.012 |
Tim | -0.011 |
pepper | -0.011 |
fought | -0.010 |
help | -0.009 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
rou | +0.023 |
ouncing | +0.021 |
rol | +0.020 |
liness | +0.020 |
Carly | +0.020 |
Neg logit contributions | |
pe | -0.009 |
Tim | -0.008 |
pepper | -0.008 |
shr | -0.008 |
help | -0.007 |
Token | regret |
Token ID | 13721 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
pe | +0.016 |
pepper | +0.016 |
help | +0.016 |
Tim | +0.015 |
fought | +0.014 |
Neg logit contributions | |
rou | -0.032 |
ouncing | -0.030 |
rol | -0.028 |
Carly | -0.028 |
liness | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -7.62e-03 |
Pos logit contributions | |
rou | +0.102 |
ouncing | +0.094 |
rol | +0.094 |
Carly | +0.093 |
liness | +0.092 |
Neg logit contributions | |
pe | -0.043 |
help | -0.036 |
pepper | -0.036 |
Tim | -0.035 |
fought | -0.033 |
Token | " |
Token ID | 366 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
rou | +0.030 |
ouncing | +0.026 |
rol | +0.025 |
liness | +0.025 |
Carly | +0.025 |
Neg logit contributions | |
pe | -0.021 |
Tim | -0.020 |
pepper | -0.019 |
shr | -0.019 |
help | -0.018 |
Token | It |
Token ID | 1026 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
rou | +0.092 |
ouncing | +0.080 |
rol | +0.080 |
ucking | +0.077 |
liness | +0.076 |
Neg logit contributions | |
pe | -0.048 |
pepper | -0.048 |
help | -0.046 |
fought | -0.044 |
Tim | -0.043 |
Token | 's |
Token ID | 338 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
pe | +0.057 |
pepper | +0.056 |
tried | +0.054 |
help | +0.054 |
fought | +0.051 |
Neg logit contributions | |
rou | -0.086 |
aults | -0.077 |
Carly | -0.077 |
rol | -0.077 |
ouncing | -0.074 |
Token | not |
Token ID | 407 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
rou | +0.120 |
ouncing | +0.106 |
rol | +0.102 |
liness | +0.102 |
Carly | +0.100 |
Neg logit contributions | |
pe | -0.073 |
Tim | -0.069 |
pepper | -0.068 |
shr | -0.066 |
fought | -0.064 |
Token | my |
Token ID | 616 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
rou | +0.060 |
ouncing | +0.053 |
rol | +0.052 |
liness | +0.051 |
Carly | +0.051 |
Neg logit contributions | |
pe | -0.035 |
pepper | -0.034 |
Tim | -0.034 |
help | -0.033 |
shr | -0.032 |
Token | fault |
Token ID | 8046 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
rou | +0.189 |
ouncing | +0.164 |
rol | +0.161 |
liness | +0.157 |
Carly | +0.155 |
Neg logit contributions | |
pe | -0.148 |
pepper | -0.141 |
Tim | -0.140 |
help | -0.134 |
shr | -0.133 |
Token | . |
Token ID | 13 |
Feature activation | -5.39e-03 |
Pos logit contributions | |
rou | +0.395 |
rol | +0.349 |
amaz | +0.346 |
ouncing | +0.345 |
Carly | +0.343 |
Neg logit contributions | |
pe | -0.136 |
help | -0.122 |
pepper | -0.120 |
Tim | -0.118 |
fought | -0.107 |
Token | I |
Token ID | 314 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
rou | +0.011 |
ouncing | +0.010 |
rol | +0.009 |
liness | +0.009 |
Carly | +0.009 |
Neg logit contributions | |
pe | -0.006 |
Tim | -0.006 |
pepper | -0.006 |
shr | -0.005 |
fought | -0.005 |
Token | can |
Token ID | 460 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
rou | +0.057 |
ouncing | +0.052 |
rol | +0.050 |
liness | +0.050 |
Carly | +0.050 |
Neg logit contributions | |
pe | -0.027 |
Tim | -0.025 |
pepper | -0.025 |
help | -0.024 |
fought | -0.023 |
Token | 't |
Token ID | 470 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
rou | +0.068 |
ouncing | +0.059 |
aults | +0.059 |
rol | +0.058 |
downward | +0.058 |
Neg logit contributions | |
pe | -0.026 |
pepper | -0.025 |
Tim | -0.024 |
shr | -0.023 |
fought | -0.023 |
Token | help |
Token ID | 1037 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
pe | +0.012 |
pepper | +0.012 |
shr | +0.011 |
Blood | +0.011 |
Tim | +0.011 |
Neg logit contributions | |
rou | -0.019 |
aults | -0.017 |
ouncing | -0.016 |
ala | -0.016 |
rol | -0.016 |
Token | sorry |
Token ID | 7926 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
pe | +0.053 |
pepper | +0.053 |
Tim | +0.053 |
fought | +0.051 |
help | +0.049 |
Neg logit contributions | |
rou | -0.073 |
ouncing | -0.065 |
liness | -0.065 |
rol | -0.065 |
Carly | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +1.79e-03 |
Pos logit contributions | |
rou | +0.078 |
rol | +0.076 |
amaz | +0.072 |
Carly | +0.071 |
capt | +0.070 |
Neg logit contributions | |
help | -0.013 |
Tim | -0.012 |
pepper | -0.009 |
be | -0.008 |
pe | -0.008 |
Token | It |
Token ID | 632 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
pe | +0.002 |
Tim | +0.002 |
fought | +0.002 |
shr | +0.002 |
timer | +0.002 |
Neg logit contributions | |
rou | -0.003 |
ouncing | -0.003 |
liness | -0.003 |
Carly | -0.003 |
downward | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
pe | +0.078 |
Tim | +0.075 |
pepper | +0.074 |
shr | +0.072 |
help | +0.070 |
Neg logit contributions | |
rou | -0.088 |
ouncing | -0.076 |
rol | -0.073 |
liness | -0.072 |
Carly | -0.071 |
Token | my |
Token ID | 616 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
rou | +0.025 |
ouncing | +0.021 |
rol | +0.021 |
liness | +0.021 |
downward | +0.020 |
Neg logit contributions | |
pe | -0.016 |
Tim | -0.015 |
shr | -0.014 |
pepper | -0.014 |
fought | -0.014 |
Token | fault |
Token ID | 8046 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
rou | +0.166 |
rol | +0.145 |
ouncing | +0.144 |
liness | +0.140 |
Carly | +0.136 |
Neg logit contributions | |
pe | -0.166 |
Tim | -0.162 |
pepper | -0.161 |
help | -0.160 |
fought | -0.152 |
Token | . |
Token ID | 13 |
Feature activation | +1.72e-03 |
Pos logit contributions | |
rou | +0.381 |
ouncing | +0.342 |
rol | +0.331 |
liness | +0.330 |
Carly | +0.324 |
Neg logit contributions | |
pe | -0.143 |
Tim | -0.133 |
pepper | -0.130 |
shr | -0.127 |
fought | -0.120 |
Token | I |
Token ID | 314 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
pe | +0.002 |
Tim | +0.002 |
pepper | +0.002 |
fought | +0.002 |
shr | +0.002 |
Neg logit contributions | |
rou | -0.003 |
ouncing | -0.003 |
liness | -0.003 |
Carly | -0.003 |
rol | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
rou | +0.057 |
ouncing | +0.048 |
rol | +0.047 |
liness | +0.046 |
ucking | +0.044 |
Neg logit contributions | |
pe | -0.035 |
Tim | -0.033 |
pepper | -0.032 |
shr | -0.031 |
timer | -0.029 |
Token | too |
Token ID | 1165 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
rou | +0.028 |
ouncing | +0.023 |
rol | +0.023 |
liness | +0.022 |
downward | +0.022 |
Neg logit contributions | |
pe | -0.017 |
Tim | -0.016 |
shr | -0.016 |
pepper | -0.015 |
fought | -0.015 |
Token | brave |
Token ID | 14802 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
pe | +0.029 |
Tim | +0.028 |
pepper | +0.028 |
shr | +0.027 |
fought | +0.026 |
Neg logit contributions | |
rou | -0.046 |
rol | -0.041 |
ouncing | -0.041 |
liness | -0.040 |
Carly | -0.040 |
Token | he |
Token ID | 339 |
Feature activation | -8.60e-03 |
Pos logit contributions | |
rou | +0.060 |
ouncing | +0.054 |
rol | +0.054 |
liness | +0.053 |
Carly | +0.051 |
Neg logit contributions | |
pe | -0.020 |
Tim | -0.018 |
pepper | -0.018 |
help | -0.016 |
shr | -0.016 |
Token | knew |
Token ID | 2993 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
rou | +0.016 |
ouncing | +0.013 |
la | +0.013 |
liness | +0.012 |
Carly | +0.012 |
Neg logit contributions | |
pe | -0.011 |
shr | -0.011 |
Tim | -0.010 |
pepper | -0.010 |
� | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
pe | +0.013 |
Tim | +0.012 |
pepper | +0.011 |
help | +0.010 |
fought | +0.009 |
Neg logit contributions | |
rou | -0.056 |
ouncing | -0.051 |
rol | -0.050 |
liness | -0.050 |
aults | -0.049 |
Token | was |
Token ID | 373 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
pe | +0.021 |
Tim | +0.019 |
pepper | +0.019 |
shr | +0.018 |
help | +0.017 |
Neg logit contributions | |
rou | -0.053 |
ouncing | -0.048 |
liness | -0.047 |
rol | -0.046 |
Carly | -0.046 |
Token | his |
Token ID | 465 |
Feature activation | -6.13e-03 |
Pos logit contributions | |
rou | +0.023 |
liness | +0.020 |
ouncing | +0.019 |
downward | +0.019 |
Carly | +0.018 |
Neg logit contributions | |
pe | -0.021 |
Tim | -0.020 |
shr | -0.019 |
pepper | -0.019 |
� | -0.018 |
Token | fault |
Token ID | 8046 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
rou | +0.011 |
ouncing | +0.010 |
liness | +0.009 |
rol | +0.009 |
Carly | +0.009 |
Neg logit contributions | |
pe | -0.008 |
Tim | -0.008 |
pepper | -0.008 |
shr | -0.008 |
fought | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -6.11e-03 |
Pos logit contributions | |
rou | +0.354 |
ouncing | +0.315 |
rol | +0.304 |
liness | +0.303 |
Carly | +0.298 |
Neg logit contributions | |
pe | -0.169 |
Tim | -0.160 |
pepper | -0.155 |
shr | -0.154 |
fought | -0.146 |
Token | In |
Token ID | 554 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
rou | +0.015 |
rol | +0.014 |
aults | +0.014 |
ouncing | +0.014 |
liness | +0.014 |
Neg logit contributions | |
pe | -0.003 |
pepper | -0.003 |
Tim | -0.003 |
help | -0.003 |
shr | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -9.17e-03 |
Pos logit contributions | |
pe | +0.013 |
Tim | +0.012 |
pepper | +0.012 |
shr | +0.011 |
help | +0.011 |
Neg logit contributions | |
rou | -0.027 |
ouncing | -0.024 |
rol | -0.024 |
liness | -0.023 |
Carly | -0.023 |
Token | end |
Token ID | 886 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
rou | +0.017 |
liness | +0.016 |
downward | +0.014 |
ouncing | +0.014 |
amaz | +0.014 |
Neg logit contributions | |
pe | -0.012 |
Tim | -0.011 |
fought | -0.011 |
shr | -0.011 |
ts | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
rou | +0.072 |
rol | +0.067 |
la | +0.064 |
aults | +0.064 |
ouncing | +0.063 |
Neg logit contributions | |
Tim | -0.028 |
help | -0.028 |
pepper | -0.027 |
pe | -0.026 |
hammer | -0.023 |
Token | He |
Token ID | 679 |
Feature activation | +2.00e-03 |
Pos logit contributions | |
pe | +0.004 |
shr | +0.004 |
pepper | +0.004 |
Tim | +0.003 |
fought | +0.003 |
Neg logit contributions | |
rou | -0.008 |
ouncing | -0.007 |
Carly | -0.007 |
liness | -0.007 |
downward | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | +4.61e-04 |
Pos logit contributions | |
pe | +0.002 |
Tim | +0.002 |
shr | +0.002 |
pepper | +0.002 |
help | +0.002 |
Neg logit contributions | |
rou | -0.004 |
ouncing | -0.004 |
rol | -0.004 |
liness | -0.004 |
Carly | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
pe | +0.000 |
pepper | +0.000 |
Tim | +0.000 |
help | +0.000 |
fought | +0.000 |
Neg logit contributions | |
rou | -0.001 |
rol | -0.001 |
ouncing | -0.001 |
liness | -0.001 |
aults | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
pe | +0.027 |
pepper | +0.025 |
Tim | +0.025 |
shr | +0.024 |
help | +0.023 |
Neg logit contributions | |
rou | -0.062 |
ouncing | -0.055 |
liness | -0.054 |
rol | -0.053 |
downward | -0.052 |
Token | her |
Token ID | 607 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
rou | +0.023 |
ouncing | +0.019 |
liness | +0.018 |
downward | +0.018 |
Carly | +0.018 |
Neg logit contributions | |
pe | -0.015 |
Tim | -0.014 |
shr | -0.013 |
pepper | -0.013 |
� | -0.013 |
Token | fault |
Token ID | 8046 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
rou | +0.040 |
rol | +0.038 |
ouncing | +0.037 |
liness | +0.037 |
Carly | +0.036 |
Neg logit contributions | |
help | -0.020 |
pe | -0.019 |
pepper | -0.018 |
hammer | -0.017 |
cube | -0.017 |
Token | for |
Token ID | 329 |
Feature activation | -3.59e-03 |
Pos logit contributions | |
rou | +0.340 |
ouncing | +0.301 |
rol | +0.290 |
liness | +0.289 |
Carly | +0.284 |
Neg logit contributions | |
pe | -0.183 |
Tim | -0.173 |
pepper | -0.169 |
shr | -0.167 |
fought | -0.160 |
Token | not |
Token ID | 407 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
rou | +0.006 |
ouncing | +0.004 |
ayed | +0.004 |
downward | +0.004 |
Carly | +0.004 |
Neg logit contributions | |
pe | -0.005 |
shr | -0.005 |
pepper | -0.005 |
� | -0.005 |
Tim | -0.005 |
Token | moving |
Token ID | 3867 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
rou | +0.056 |
ouncing | +0.048 |
liness | +0.046 |
rol | +0.045 |
aults | +0.045 |
Neg logit contributions | |
pe | -0.042 |
Tim | -0.039 |
pepper | -0.037 |
shr | -0.037 |
� | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +1.93e-03 |
Pos logit contributions | |
rou | +0.088 |
ouncing | +0.081 |
rol | +0.079 |
Carly | +0.077 |
amaz | +0.076 |
Neg logit contributions | |
pe | -0.023 |
Tim | -0.022 |
help | -0.020 |
pepper | -0.020 |
fought | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -7.46e-03 |
Pos logit contributions | |
pe | +0.002 |
shr | +0.002 |
pepper | +0.002 |
fought | +0.002 |
Tim | +0.002 |
Neg logit contributions | |
rou | -0.004 |
ouncing | -0.004 |
Carly | -0.003 |
downward | -0.003 |
liness | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -5.63e-03 |
Pos logit contributions | |
rou | +0.052 |
ouncing | +0.049 |
Carly | +0.047 |
rol | +0.046 |
downward | +0.046 |
Neg logit contributions | |
pe | -0.011 |
help | -0.009 |
pepper | -0.009 |
Tim | -0.008 |
fought | -0.008 |
Token | |
Token ID | 220 |
Feature activation | +8.75e-03 |
Pos logit contributions | |
rou | +0.013 |
aults | +0.011 |
liness | +0.011 |
rol | +0.011 |
ouncing | +0.011 |
Neg logit contributions | |
Tim | -0.005 |
pe | -0.005 |
pepper | -0.004 |
help | -0.004 |
� | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
pe | +0.005 |
Tim | +0.005 |
pepper | +0.005 |
shr | +0.004 |
help | +0.004 |
Neg logit contributions | |
rou | -0.022 |
ouncing | -0.020 |
liness | -0.019 |
rol | -0.019 |
aults | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
rou | +0.025 |
ouncing | +0.023 |
rol | +0.022 |
liness | +0.022 |
Carly | +0.022 |
Neg logit contributions | |
pe | -0.008 |
Tim | -0.008 |
pepper | -0.008 |
shr | -0.007 |
fought | -0.007 |
Token | Then |
Token ID | 6423 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
rou | +0.027 |
liness | +0.024 |
asleep | +0.024 |
rol | +0.023 |
packs | +0.023 |
Neg logit contributions | |
Tim | -0.011 |
pe | -0.010 |
B | -0.008 |
shr | -0.008 |
tried | -0.007 |
Token | Izzy |
Token ID | 46359 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
tried | +0.092 |
help | +0.089 |
be | +0.075 |
Tim | +0.069 |
days | +0.068 |
Neg logit contributions | |
rou | -0.277 |
rol | -0.246 |
ouncing | -0.231 |
reating | -0.230 |
aj | -0.225 |
Token | used |
Token ID | 973 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
rou | +0.402 |
ouncing | +0.370 |
rol | +0.369 |
liness | +0.351 |
downward | +0.351 |
Neg logit contributions | |
tried | -0.125 |
fought | -0.117 |
pepper | -0.116 |
pe | -0.116 |
Tim | -0.112 |
Token | the |
Token ID | 262 |
Feature activation | -7.26e-03 |
Pos logit contributions | |
pe | +0.017 |
Tim | +0.016 |
pepper | +0.016 |
shr | +0.015 |
help | +0.015 |
Neg logit contributions | |
rou | -0.033 |
ouncing | -0.029 |
rol | -0.028 |
liness | -0.028 |
Carly | -0.027 |
Token | tape |
Token ID | 9154 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
rou | +0.014 |
downward | +0.012 |
liness | +0.012 |
ouncing | +0.012 |
amaz | +0.012 |
Neg logit contributions | |
pe | -0.008 |
olph | -0.008 |
fought | -0.008 |
Tim | -0.007 |
shr | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -3.73e-03 |
Pos logit contributions | |
pe | +0.047 |
pepper | +0.044 |
Tim | +0.043 |
help | +0.042 |
shr | +0.041 |
Neg logit contributions | |
rou | -0.075 |
ouncing | -0.066 |
rol | -0.064 |
Carly | -0.063 |
liness | -0.062 |
Token | stick |
Token ID | 4859 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
rou | +0.007 |
aults | +0.006 |
amaz | +0.006 |
liness | +0.006 |
downward | +0.006 |
Neg logit contributions | |
pe | -0.004 |
olph | -0.004 |
shr | -0.004 |
Tim | -0.003 |
fought | -0.003 |
Token | their |
Token ID | 511 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
help | +0.007 |
pe | +0.006 |
Tim | +0.006 |
pepper | +0.006 |
tried | +0.006 |
Neg logit contributions | |
rou | -0.009 |
ouncing | -0.008 |
rol | -0.008 |
liness | -0.008 |
ucking | -0.007 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.85e-04 |
Pos logit contributions | |
help | +0.021 |
pepper | +0.020 |
Tim | +0.019 |
cube | +0.018 |
fight | +0.018 |
Neg logit contributions | |
rou | -0.038 |
ouncing | -0.037 |
rol | -0.037 |
ucking | -0.034 |
liness | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.40e-03 |
Pos logit contributions | |
rou | +0.001 |
ouncing | +0.000 |
rol | +0.000 |
ayed | +0.000 |
ushes | +0.000 |
Neg logit contributions | |
help | -0.000 |
pe | -0.000 |
. | -0.000 |
tried | -0.000 |
he | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -4.55e-03 |
Pos logit contributions | |
rou | +0.003 |
ouncing | +0.003 |
aults | +0.003 |
ayed | +0.003 |
rol | +0.003 |
Neg logit contributions | |
pe | -0.001 |
Tim | -0.001 |
help | -0.001 |
But | -0.001 |
tried | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -4.45e-03 |
Pos logit contributions | |
rou | +0.011 |
ouncing | +0.010 |
rol | +0.009 |
liness | +0.009 |
Carly | +0.009 |
Neg logit contributions | |
pe | -0.004 |
Tim | -0.003 |
pepper | -0.003 |
shr | -0.003 |
help | -0.003 |
Token | Fel |
Token ID | 42493 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
rou | +0.009 |
ouncing | +0.008 |
liness | +0.008 |
rol | +0.008 |
fault | +0.008 |
Neg logit contributions | |
Tim | -0.005 |
pe | -0.005 |
B | -0.005 |
shr | -0.005 |
help | -0.004 |
Token | icity |
Token ID | 8467 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
rou | +0.243 |
ouncing | +0.203 |
rol | +0.197 |
liness | +0.194 |
Carly | +0.185 |
Neg logit contributions | |
pe | -0.282 |
Tim | -0.266 |
pepper | -0.261 |
shr | -0.260 |
fought | -0.255 |
Token | decided |
Token ID | 3066 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
rou | +0.065 |
rol | +0.059 |
ouncing | +0.059 |
liness | +0.057 |
downward | +0.057 |
Neg logit contributions | |
pe | -0.027 |
tried | -0.025 |
fought | -0.025 |
pepper | -0.024 |
Tim | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | +3.37e-03 |
Pos logit contributions | |
rou | +0.057 |
rol | +0.053 |
ouncing | +0.051 |
ucking | +0.050 |
liness | +0.050 |
Neg logit contributions | |
help | -0.019 |
pe | -0.019 |
tried | -0.018 |
Tim | -0.018 |
pepper | -0.016 |
Token | keep |
Token ID | 1394 |
Feature activation | +9.24e-03 |
Pos logit contributions | |
pe | +0.003 |
Tim | +0.003 |
pepper | +0.003 |
help | +0.003 |
shr | +0.003 |
Neg logit contributions | |
rou | -0.008 |
ouncing | -0.007 |
rol | -0.007 |
liness | -0.006 |
Carly | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +4.29e-04 |
Pos logit contributions | |
pe | +0.011 |
Tim | +0.011 |
shr | +0.010 |
pepper | +0.010 |
fought | +0.010 |
Neg logit contributions | |
rou | -0.018 |
ouncing | -0.015 |
liness | -0.015 |
Carly | -0.015 |
rol | -0.015 |
Token | " |
Token ID | 366 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
rou | +0.029 |
ouncing | +0.025 |
Carly | +0.024 |
liness | +0.024 |
rol | +0.024 |
Neg logit contributions | |
pe | -0.023 |
Tim | -0.022 |
shr | -0.022 |
pepper | -0.022 |
fought | -0.021 |
Token | This |
Token ID | 1212 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
rou | +0.095 |
ouncing | +0.083 |
rol | +0.082 |
liness | +0.081 |
Carly | +0.080 |
Neg logit contributions | |
pe | -0.050 |
pepper | -0.046 |
help | -0.044 |
Tim | -0.044 |
shr | -0.043 |
Token | is |
Token ID | 318 |
Feature activation | -6.96e-02 |
Pos logit contributions | |
pe | +0.028 |
pepper | +0.027 |
help | +0.026 |
tried | +0.024 |
Tim | +0.024 |
Neg logit contributions | |
rou | -0.043 |
ouncing | -0.039 |
aults | -0.036 |
rol | -0.036 |
downward | -0.036 |
Token | all |
Token ID | 477 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
rou | +0.145 |
ouncing | +0.132 |
rol | +0.128 |
liness | +0.126 |
ucking | +0.125 |
Neg logit contributions | |
pepper | -0.073 |
pe | -0.072 |
Tim | -0.071 |
help | -0.069 |
shr | -0.067 |
Token | your |
Token ID | 534 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
rou | +0.052 |
ouncing | +0.048 |
Carly | +0.047 |
aults | +0.046 |
liness | +0.046 |
Neg logit contributions | |
pepper | -0.028 |
pe | -0.028 |
Tim | -0.027 |
help | -0.027 |
shr | -0.025 |
Token | fault |
Token ID | 8046 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
rou | +0.101 |
ouncing | +0.088 |
rol | +0.085 |
liness | +0.085 |
Carly | +0.083 |
Neg logit contributions | |
pe | -0.062 |
Tim | -0.059 |
pepper | -0.058 |
shr | -0.056 |
fought | -0.055 |
Token | , |
Token ID | 11 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
rou | +0.343 |
rol | +0.301 |
ouncing | +0.297 |
Carly | +0.295 |
amaz | +0.295 |
Neg logit contributions | |
pe | -0.180 |
pepper | -0.166 |
help | -0.164 |
Tim | -0.163 |
fought | -0.148 |
Token | Mom |
Token ID | 11254 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
rou | +0.039 |
rol | +0.037 |
ouncing | +0.037 |
packs | +0.035 |
la | +0.035 |
Neg logit contributions | |
pepper | -0.018 |
Tim | -0.017 |
help | -0.016 |
cube | -0.015 |
Tim | -0.015 |
Token | my |
Token ID | 1820 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
rou | +0.045 |
Carly | +0.042 |
rol | +0.041 |
aults | +0.041 |
downward | +0.041 |
Neg logit contributions | |
Tim | -0.031 |
pe | -0.030 |
pepper | -0.030 |
cube | -0.028 |
fought | -0.027 |
Token | ! |
Token ID | 0 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
rou | +0.097 |
rol | +0.092 |
Carly | +0.088 |
amaz | +0.088 |
ouncing | +0.087 |
Neg logit contributions | |
pe | -0.041 |
pepper | -0.040 |
Tim | -0.038 |
help | -0.038 |
fought | -0.038 |
Token | You |
Token ID | 921 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
rou | +0.086 |
ouncing | +0.076 |
rol | +0.074 |
liness | +0.073 |
Carly | +0.072 |
Neg logit contributions | |
pe | -0.049 |
Tim | -0.047 |
pepper | -0.046 |
shr | -0.044 |
help | -0.043 |
Token | said |
Token ID | 531 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
rou | +0.038 |
ouncing | +0.033 |
liness | +0.033 |
Carly | +0.033 |
downward | +0.032 |
Neg logit contributions | |
pe | -0.018 |
shr | -0.016 |
Tim | -0.015 |
pepper | -0.014 |
� | -0.014 |
Token | it |
Token ID | 340 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
pe | +0.003 |
pepper | +0.003 |
Tim | +0.003 |
help | +0.003 |
fought | +0.002 |
Neg logit contributions | |
rou | -0.007 |
rol | -0.007 |
ouncing | -0.007 |
liness | -0.006 |
aults | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -9.88e-03 |
Pos logit contributions | |
pe | +0.025 |
Tim | +0.023 |
pepper | +0.022 |
shr | +0.022 |
help | +0.020 |
Neg logit contributions | |
rou | -0.071 |
ouncing | -0.064 |
liness | -0.062 |
rol | -0.062 |
Carly | -0.061 |
Token | not |
Token ID | 407 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
rou | +0.017 |
ouncing | +0.014 |
rol | +0.014 |
liness | +0.014 |
Carly | +0.013 |
Neg logit contributions | |
pe | -0.015 |
Tim | -0.014 |
pepper | -0.014 |
shr | -0.013 |
help | -0.013 |
Token | their |
Token ID | 511 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
rou | +0.052 |
ouncing | +0.046 |
liness | +0.044 |
rol | +0.044 |
downward | +0.043 |
Neg logit contributions | |
pe | -0.036 |
Tim | -0.034 |
shr | -0.033 |
pepper | -0.033 |
� | -0.031 |
Token | fault |
Token ID | 8046 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
help | +0.030 |
pe | +0.027 |
pepper | +0.027 |
fight | +0.026 |
crane | +0.025 |
Neg logit contributions | |
rou | -0.041 |
rol | -0.039 |
ouncing | -0.037 |
ucking | -0.037 |
liness | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +3.56e-03 |
Pos logit contributions | |
rou | +0.386 |
ouncing | +0.348 |
rol | +0.338 |
liness | +0.336 |
Carly | +0.331 |
Neg logit contributions | |
pe | -0.132 |
Tim | -0.122 |
pepper | -0.119 |
shr | -0.116 |
fought | -0.109 |
Token | She |
Token ID | 1375 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
pe | +0.004 |
Tim | +0.004 |
pepper | +0.004 |
shr | +0.004 |
help | +0.004 |
Neg logit contributions | |
rou | -0.007 |
ouncing | -0.006 |
rol | -0.006 |
liness | -0.006 |
aults | -0.006 |
Token | said |
Token ID | 531 |
Feature activation | +2.28e-03 |
Pos logit contributions | |
rou | +0.073 |
ouncing | +0.061 |
la | +0.061 |
Carly | +0.060 |
liness | +0.059 |
Neg logit contributions | |
pe | -0.034 |
shr | -0.033 |
Tim | -0.030 |
� | -0.029 |
pepper | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +1.36e-03 |
Pos logit contributions | |
pe | +0.002 |
pepper | +0.002 |
help | +0.002 |
Tim | +0.002 |
fought | +0.001 |
Neg logit contributions | |
rou | -0.006 |
rol | -0.005 |
ouncing | -0.005 |
liness | -0.005 |
aults | -0.005 |
Token | truck |
Token ID | 7779 |
Feature activation | +6.86e-03 |
Pos logit contributions | |
pe | +0.001 |
Tim | +0.001 |
shr | +0.001 |
pepper | +0.001 |
fought | +0.001 |
Neg logit contributions | |
rou | -0.003 |
ouncing | -0.002 |
liness | -0.002 |
rol | -0.002 |
downward | -0.002 |
Token | From |
Token ID | 3574 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
rou | +0.008 |
rol | +0.007 |
ouncing | +0.007 |
aults | +0.007 |
liness | +0.007 |
Neg logit contributions | |
pe | -0.003 |
Tim | -0.003 |
pepper | -0.003 |
shr | -0.003 |
help | -0.003 |
Token | that |
Token ID | 326 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
rou | +0.040 |
ouncing | +0.036 |
rol | +0.035 |
liness | +0.035 |
Carly | +0.034 |
Neg logit contributions | |
pe | -0.018 |
Tim | -0.017 |
pepper | -0.017 |
shr | -0.016 |
help | -0.016 |
Token | day |
Token ID | 1110 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
rou | +0.056 |
ouncing | +0.051 |
rol | +0.051 |
liness | +0.050 |
Carly | +0.049 |
Neg logit contributions | |
pe | -0.008 |
Tim | -0.006 |
pepper | -0.006 |
help | -0.005 |
shr | -0.005 |
Token | on |
Token ID | 319 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
pe | +0.037 |
Tim | +0.035 |
pepper | +0.034 |
help | +0.034 |
shr | +0.034 |
Neg logit contributions | |
rou | -0.055 |
ouncing | -0.047 |
rol | -0.046 |
liness | -0.045 |
Carly | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
rou | +0.050 |
rol | +0.045 |
capt | +0.042 |
aults | +0.042 |
ouncing | +0.042 |
Neg logit contributions | |
help | -0.014 |
pe | -0.013 |
Tim | -0.012 |
pepper | -0.012 |
tried | -0.011 |
Token | Izzy |
Token ID | 46359 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
rou | +0.043 |
rol | +0.041 |
aj | +0.037 |
la | +0.037 |
aults | +0.036 |
Neg logit contributions | |
help | -0.022 |
Tim | -0.019 |
tried | -0.019 |
But | -0.018 |
days | -0.017 |
Token | worked |
Token ID | 3111 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
rou | +0.405 |
rol | +0.386 |
ouncing | +0.366 |
Carly | +0.358 |
aults | +0.357 |
Neg logit contributions | |
tried | -0.121 |
pepper | -0.119 |
pe | -0.118 |
fought | -0.117 |
help | -0.114 |
Token | hard |
Token ID | 1327 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
pe | +0.032 |
Tim | +0.031 |
pepper | +0.030 |
help | +0.029 |
shr | +0.028 |
Neg logit contributions | |
rou | -0.065 |
ouncing | -0.058 |
rol | -0.057 |
liness | -0.056 |
Carly | -0.056 |
Token | with |
Token ID | 351 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
pe | +0.026 |
Tim | +0.025 |
pepper | +0.024 |
shr | +0.024 |
help | +0.023 |
Neg logit contributions | |
rou | -0.035 |
ouncing | -0.030 |
rol | -0.030 |
liness | -0.029 |
Carly | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | -7.89e-03 |
Pos logit contributions | |
pe | +0.012 |
Tim | +0.011 |
pepper | +0.011 |
help | +0.011 |
shr | +0.010 |
Neg logit contributions | |
rou | -0.021 |
ouncing | -0.019 |
rol | -0.019 |
liness | -0.018 |
aults | -0.018 |
Token | other |
Token ID | 584 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
rou | +0.015 |
ouncing | +0.013 |
liness | +0.013 |
downward | +0.013 |
aults | +0.012 |
Neg logit contributions | |
pe | -0.010 |
Tim | -0.009 |
shr | -0.009 |
fought | -0.009 |
pepper | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -9.59e-03 |
Pos logit contributions | |
rou | +0.022 |
ouncing | +0.019 |
rol | +0.019 |
liness | +0.019 |
Carly | +0.019 |
Neg logit contributions | |
pe | -0.007 |
Tim | -0.007 |
pepper | -0.006 |
shr | -0.006 |
fought | -0.006 |
Token | At |
Token ID | 2953 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
rou | +0.022 |
liness | +0.019 |
ouncing | +0.019 |
rol | +0.019 |
aj | +0.018 |
Neg logit contributions | |
pe | -0.010 |
Tim | -0.010 |
help | -0.009 |
shr | -0.008 |
tried | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -5.72e-03 |
Pos logit contributions | |
pe | +0.011 |
pepper | +0.010 |
Tim | +0.010 |
help | +0.010 |
shr | +0.009 |
Neg logit contributions | |
rou | -0.029 |
ouncing | -0.027 |
rol | -0.026 |
liness | -0.026 |
aults | -0.025 |
Token | farm |
Token ID | 5318 |
Feature activation | +8.07e-03 |
Pos logit contributions | |
rou | +0.011 |
ouncing | +0.010 |
rol | +0.010 |
liness | +0.010 |
Carly | +0.009 |
Neg logit contributions | |
pe | -0.007 |
Tim | -0.006 |
pepper | -0.006 |
shr | -0.006 |
help | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
pe | +0.008 |
help | +0.008 |
pepper | +0.008 |
Tim | +0.007 |
tried | +0.007 |
Neg logit contributions | |
rou | -0.018 |
ouncing | -0.016 |
aults | -0.015 |
rol | -0.015 |
idget | -0.015 |
Token | Izzy |
Token ID | 46359 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
rou | +0.049 |
rol | +0.042 |
aj | +0.042 |
udd | +0.041 |
aters | +0.041 |
Neg logit contributions | |
tried | -0.017 |
he | -0.017 |
help | -0.017 |
Tim | -0.016 |
Tim | -0.015 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.53e-03 |
Pos logit contributions | |
rou | +0.407 |
ouncing | +0.372 |
rol | +0.363 |
ayed | +0.352 |
reating | +0.351 |
Neg logit contributions | |
tried | -0.121 |
pe | -0.118 |
fought | -0.110 |
pepper | -0.104 |
help | -0.100 |
Token | lots |
Token ID | 6041 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
pepper | +0.003 |
help | +0.003 |
pe | +0.003 |
Tim | +0.003 |
tried | +0.002 |
Neg logit contributions | |
rou | -0.012 |
ouncing | -0.011 |
aj | -0.010 |
iding | -0.010 |
aults | -0.010 |
Token | of |
Token ID | 286 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
rou | +0.108 |
aj | +0.101 |
idget | +0.100 |
ouncing | +0.099 |
ancing | +0.098 |
Neg logit contributions | |
pepper | -0.052 |
help | -0.044 |
crack | -0.043 |
Tim | -0.043 |
days | -0.042 |
Token | different |
Token ID | 1180 |
Feature activation | +2.35e-03 |
Pos logit contributions | |
rou | +0.091 |
ouncing | +0.085 |
rol | +0.079 |
ancing | +0.078 |
liness | +0.078 |
Neg logit contributions | |
pepper | -0.042 |
pe | -0.041 |
Tim | -0.040 |
help | -0.039 |
crack | -0.036 |
Token | animals |
Token ID | 4695 |
Feature activation | +3.72e-03 |
Pos logit contributions | |
pepper | +0.002 |
pe | +0.002 |
help | +0.002 |
Tim | +0.002 |
days | +0.001 |
Neg logit contributions | |
rou | -0.006 |
ouncing | -0.005 |
idget | -0.005 |
Carly | -0.005 |
aults | -0.005 |
Token | were |
Token ID | 547 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
pe | +0.017 |
Tim | +0.016 |
pepper | +0.015 |
shr | +0.014 |
fought | +0.014 |
Neg logit contributions | |
rou | -0.033 |
ouncing | -0.029 |
rol | -0.028 |
liness | -0.028 |
Carly | -0.027 |
Token | healed |
Token ID | 28557 |
Feature activation | +6.42e-03 |
Pos logit contributions | |
pe | +0.029 |
pepper | +0.027 |
fought | +0.027 |
help | +0.026 |
tried | +0.026 |
Neg logit contributions | |
rou | -0.053 |
ouncing | -0.051 |
liness | -0.048 |
Carly | -0.048 |
rol | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +3.57e-03 |
Pos logit contributions | |
pe | +0.004 |
Tim | +0.003 |
pepper | +0.003 |
help | +0.003 |
fought | +0.003 |
Neg logit contributions | |
rou | -0.016 |
ouncing | -0.015 |
rol | -0.015 |
Carly | -0.015 |
Debbie | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +2.00e-03 |
Pos logit contributions | |
pe | +0.003 |
Tim | +0.003 |
pepper | +0.003 |
shr | +0.003 |
fought | +0.003 |
Neg logit contributions | |
rou | -0.008 |
ouncing | -0.007 |
rol | -0.007 |
liness | -0.007 |
Carly | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +2.55e-03 |
Pos logit contributions | |
pe | +0.002 |
pepper | +0.002 |
shr | +0.002 |
fought | +0.002 |
Tim | +0.002 |
Neg logit contributions | |
rou | -0.004 |
ouncing | -0.004 |
Carly | -0.004 |
rol | -0.004 |
liness | -0.004 |
Token | Fel |
Token ID | 42493 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
Tim | +0.003 |
pe | +0.002 |
shr | +0.002 |
pepper | +0.002 |
fought | +0.002 |
Neg logit contributions | |
rou | -0.006 |
ouncing | -0.005 |
liness | -0.005 |
rol | -0.005 |
downward | -0.005 |
Token | icity |
Token ID | 8467 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
rou | +0.301 |
ouncing | +0.262 |
rol | +0.250 |
liness | +0.247 |
Carly | +0.246 |
Neg logit contributions | |
pe | -0.213 |
pepper | -0.208 |
Tim | -0.208 |
shr | -0.198 |
help | -0.196 |
Token | was |
Token ID | 373 |
Feature activation | -1.43e-03 |
Pos logit contributions | |
rou | +0.058 |
ouncing | +0.052 |
rol | +0.050 |
liness | +0.050 |
Carly | +0.049 |
Neg logit contributions | |
pe | -0.022 |
Tim | -0.021 |
pepper | -0.020 |
shr | -0.019 |
fought | -0.019 |
Token | so |
Token ID | 523 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
ouncing | +0.003 |
rou | +0.003 |
liness | +0.003 |
downward | +0.003 |
ancing | +0.003 |
Neg logit contributions | |
hurt | -0.001 |
tried | -0.001 |
pe | -0.001 |
help | -0.001 |
finally | -0.001 |
Token | happy |
Token ID | 3772 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
pe | +0.015 |
help | +0.015 |
tried | +0.014 |
pepper | +0.014 |
Tim | +0.014 |
Neg logit contributions | |
rou | -0.026 |
ouncing | -0.024 |
liness | -0.024 |
rol | -0.023 |
downward | -0.023 |
Token | that |
Token ID | 326 |
Feature activation | -9.52e-03 |
Pos logit contributions | |
pe | +0.024 |
help | +0.022 |
pepper | +0.022 |
Tim | +0.022 |
he | +0.021 |
Neg logit contributions | |
rou | -0.085 |
rol | -0.079 |
idget | -0.077 |
aults | -0.076 |
ouncing | -0.075 |
Token |
|
Token ID | 198 |
Feature activation | -1.85e-03 |
Pos logit contributions | |
rou | +0.003 |
ouncing | +0.003 |
Carly | +0.003 |
ucking | +0.002 |
downward | +0.002 |
Neg logit contributions | |
pepper | -0.002 |
fought | -0.002 |
shr | -0.002 |
pe | -0.002 |
timer | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
rou | +0.003 |
ouncing | +0.003 |
liness | +0.002 |
Carly | +0.002 |
rol | +0.002 |
Neg logit contributions | |
pe | -0.003 |
pepper | -0.003 |
shr | -0.003 |
fought | -0.003 |
help | -0.003 |
Token | That |
Token ID | 2504 |
Feature activation | +1.05e-05 |
Pos logit contributions | |
rou | +0.048 |
downward | +0.043 |
aults | +0.042 |
ouncing | +0.042 |
ucking | +0.041 |
Neg logit contributions | |
help | -0.028 |
pe | -0.028 |
pepper | -0.027 |
shr | -0.025 |
occasion | -0.025 |
Token | 's |
Token ID | 338 |
Feature activation | -5.66e-02 |
Pos logit contributions | |
pe | +0.000 |
Tim | +0.000 |
pepper | +0.000 |
shr | +0.000 |
help | +0.000 |
Neg logit contributions | |
rou | -0.000 |
ouncing | -0.000 |
rol | -0.000 |
Carly | -0.000 |
aults | -0.000 |
Token | your |
Token ID | 534 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
rou | +0.096 |
ouncing | +0.080 |
liness | +0.077 |
downward | +0.076 |
aults | +0.076 |
Neg logit contributions | |
pe | -0.085 |
pepper | -0.075 |
shr | -0.074 |
Tim | -0.074 |
occasion | -0.074 |
Token | fault |
Token ID | 8046 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
rou | +0.062 |
rol | +0.058 |
ouncing | +0.054 |
liness | +0.054 |
Carly | +0.052 |
Neg logit contributions | |
pe | -0.089 |
help | -0.089 |
Tim | -0.087 |
pepper | -0.086 |
hammer | -0.084 |
Token | , |
Token ID | 11 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
rou | +0.329 |
rol | +0.287 |
ouncing | +0.286 |
liness | +0.283 |
Carly | +0.280 |
Neg logit contributions | |
pe | -0.192 |
Tim | -0.178 |
pepper | -0.173 |
help | -0.170 |
fought | -0.162 |
Token | Mia |
Token ID | 32189 |
Feature activation | -7.07e-02 |
Pos logit contributions | |
rou | +0.026 |
rol | +0.025 |
packs | +0.024 |
ouncing | +0.024 |
aults | +0.024 |
Neg logit contributions | |
Tim | -0.010 |
help | -0.010 |
pe | -0.010 |
Tim | -0.010 |
be | -0.009 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
rou | +0.153 |
rol | +0.138 |
ouncing | +0.137 |
Carly | +0.132 |
downward | +0.131 |
Neg logit contributions | |
pe | -0.073 |
Tim | -0.068 |
pepper | -0.065 |
help | -0.062 |
fought | -0.062 |
Token | he |
Token ID | 339 |
Feature activation | -3.39e-03 |
Pos logit contributions | |
rou | +0.023 |
ouncing | +0.021 |
Carly | +0.020 |
downward | +0.020 |
rol | +0.019 |
Neg logit contributions | |
pe | -0.018 |
shr | -0.018 |
pepper | -0.017 |
Tim | -0.016 |
occasion | -0.016 |
Token | shouted |
Token ID | 17293 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
rou | +0.008 |
ouncing | +0.007 |
Carly | +0.007 |
rol | +0.007 |
liness | +0.007 |
Neg logit contributions | |
pe | -0.003 |
shr | -0.003 |
pepper | -0.003 |
� | -0.003 |
Tim | -0.003 |
Token | were |
Token ID | 547 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
pe | +0.025 |
Tim | +0.023 |
pepper | +0.023 |
shr | +0.022 |
fought | +0.022 |
Neg logit contributions | |
rou | -0.043 |
ouncing | -0.038 |
rol | -0.036 |
liness | -0.036 |
Carly | -0.036 |
Token | healed |
Token ID | 28557 |
Feature activation | +5.30e-03 |
Pos logit contributions | |
pe | +0.031 |
pepper | +0.028 |
fought | +0.028 |
Tim | +0.028 |
help | +0.027 |
Neg logit contributions | |
rou | -0.055 |
ouncing | -0.052 |
liness | -0.050 |
Carly | -0.050 |
rol | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e-03 |
Pos logit contributions | |
pe | +0.003 |
Tim | +0.003 |
help | +0.002 |
pepper | +0.002 |
tried | +0.002 |
Neg logit contributions | |
rou | -0.014 |
ouncing | -0.013 |
rol | -0.012 |
Carly | -0.012 |
Debbie | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -8.78e-04 |
Pos logit contributions | |
pe | +0.001 |
Tim | +0.001 |
pepper | +0.001 |
help | +0.001 |
Blood | +0.001 |
Neg logit contributions | |
rou | -0.004 |
ouncing | -0.003 |
aults | -0.003 |
rol | -0.003 |
liness | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -7.57e-04 |
Pos logit contributions | |
rou | +0.002 |
ouncing | +0.002 |
rol | +0.002 |
liness | +0.002 |
Carly | +0.002 |
Neg logit contributions | |
pe | -0.001 |
Tim | -0.001 |
pepper | -0.001 |
shr | -0.001 |
fought | -0.001 |
Token | Fel |
Token ID | 42493 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
rou | +0.001 |
ouncing | +0.001 |
liness | +0.001 |
rol | +0.001 |
packs | +0.001 |
Neg logit contributions | |
Tim | -0.001 |
pe | -0.001 |
B | -0.001 |
shr | -0.001 |
help | -0.001 |
Token | icity |
Token ID | 8467 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
rou | +0.232 |
ouncing | +0.193 |
rol | +0.184 |
liness | +0.182 |
Carly | +0.177 |
Neg logit contributions | |
pe | -0.283 |
Tim | -0.272 |
pepper | -0.268 |
shr | -0.263 |
fought | -0.258 |
Token | was |
Token ID | 373 |
Feature activation | -4.22e-03 |
Pos logit contributions | |
rou | +0.058 |
ouncing | +0.053 |
rol | +0.051 |
liness | +0.051 |
Carly | +0.050 |
Neg logit contributions | |
pe | -0.022 |
Tim | -0.020 |
pepper | -0.019 |
fought | -0.018 |
shr | -0.018 |
Token | so |
Token ID | 523 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
ouncing | +0.009 |
rou | +0.009 |
liness | +0.009 |
downward | +0.009 |
Carly | +0.008 |
Neg logit contributions | |
help | -0.004 |
pe | -0.004 |
hurt | -0.004 |
tried | -0.004 |
fought | -0.003 |
Token | happy |
Token ID | 3772 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
pe | +0.014 |
help | +0.014 |
Tim | +0.013 |
pepper | +0.013 |
tried | +0.013 |
Neg logit contributions | |
rou | -0.024 |
ouncing | -0.022 |
rol | -0.021 |
liness | -0.021 |
Carly | -0.020 |
Token | that |
Token ID | 326 |
Feature activation | -9.37e-03 |
Pos logit contributions | |
pe | +0.023 |
help | +0.022 |
pepper | +0.022 |
Tim | +0.022 |
he | +0.021 |
Neg logit contributions | |
rou | -0.091 |
rol | -0.086 |
idget | -0.083 |
aults | -0.083 |
ouncing | -0.081 |
Token | told |
Token ID | 1297 |
Feature activation | -4.65e-02 |
Pos logit contributions | |
rou | +0.046 |
ouncing | +0.042 |
rol | +0.041 |
liness | +0.041 |
downward | +0.040 |
Neg logit contributions | |
pe | -0.011 |
Tim | -0.010 |
pepper | -0.010 |
fought | -0.010 |
tried | -0.009 |
Token | her |
Token ID | 607 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
rou | +0.086 |
ouncing | +0.076 |
rol | +0.075 |
ucking | +0.069 |
capt | +0.068 |
Neg logit contributions | |
help | -0.061 |
Tim | -0.060 |
pe | -0.060 |
pepper | -0.057 |
tried | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
rol | +0.052 |
rou | +0.051 |
idget | +0.048 |
ucking | +0.047 |
ays | +0.047 |
Neg logit contributions | |
help | -0.017 |
tried | -0.014 |
had | -0.013 |
pepper | -0.013 |
needed | -0.012 |
Token | " |
Token ID | 366 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
rou | +0.035 |
rol | +0.031 |
ouncing | +0.031 |
aults | +0.031 |
liness | +0.030 |
Neg logit contributions | |
pe | -0.016 |
Tim | -0.014 |
pepper | -0.014 |
help | -0.014 |
tried | -0.013 |
Token | B |
Token ID | 33 |
Feature activation | +8.40e-02 |
Pos logit contributions | |
rou | +0.100 |
ouncing | +0.089 |
rol | +0.086 |
liness | +0.086 |
Carly | +0.085 |
Neg logit contributions | |
pe | -0.042 |
Tim | -0.040 |
pepper | -0.039 |
shr | -0.037 |
fought | -0.036 |
Token | ouncing |
Token ID | 18155 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
Tim | +0.136 |
pe | +0.130 |
fought | +0.123 |
shr | +0.121 |
help | +0.121 |
Neg logit contributions | |
rou | -0.118 |
rol | -0.107 |
liness | -0.104 |
amaz | -0.102 |
capt | -0.102 |
Token | is |
Token ID | 318 |
Feature activation | -6.81e-02 |
Pos logit contributions | |
rou | +0.359 |
ouncing | +0.311 |
aj | +0.310 |
ucking | +0.309 |
idget | +0.307 |
Neg logit contributions | |
pe | -0.171 |
help | -0.155 |
Tim | -0.148 |
pepper | -0.142 |
tried | -0.142 |
Token | lots |
Token ID | 6041 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
ucking | +0.158 |
rou | +0.154 |
ouncing | +0.150 |
rol | +0.147 |
Debbie | +0.147 |
Neg logit contributions | |
help | -0.056 |
Tim | -0.052 |
tried | -0.052 |
pepper | -0.049 |
be | -0.049 |
Token | of |
Token ID | 286 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
rou | +0.085 |
Carly | +0.079 |
la | +0.078 |
Debbie | +0.078 |
idget | +0.077 |
Neg logit contributions | |
Tim | -0.059 |
help | -0.059 |
pepper | -0.056 |
pe | -0.054 |
days | -0.054 |
Token | fun |
Token ID | 1257 |
Feature activation | -6.43e-03 |
Pos logit contributions | |
rou | +0.089 |
ouncing | +0.080 |
rol | +0.078 |
liness | +0.077 |
Carly | +0.077 |
Neg logit contributions | |
pe | -0.032 |
Tim | -0.031 |
pepper | -0.029 |
help | -0.028 |
shr | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
rou | +0.013 |
rol | +0.012 |
idget | +0.012 |
la | +0.012 |
Carly | +0.012 |
Neg logit contributions | |
pe | -0.007 |
Tim | -0.007 |
help | -0.006 |
pepper | -0.006 |
fought | -0.006 |
Token | ." |
Token ID | 526 |
Feature activation | -6.57e-03 |
Pos logit contributions | |
rou | +0.028 |
ouncing | +0.025 |
rol | +0.025 |
idget | +0.024 |
ayed | +0.024 |
Neg logit contributions | |
pepper | -0.009 |
pe | -0.009 |
help | -0.008 |
cube | -0.008 |
Tim | -0.008 |
Token | They |
Token ID | 1119 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
rou | +0.013 |
ouncing | +0.012 |
liness | +0.011 |
rol | +0.011 |
Carly | +0.011 |
Neg logit contributions | |
pe | -0.008 |
shr | -0.007 |
Tim | -0.007 |
pepper | -0.007 |
fought | -0.006 |
Token | watched |
Token ID | 7342 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
pe | +0.018 |
Tim | +0.016 |
pepper | +0.016 |
shr | +0.015 |
help | +0.014 |
Neg logit contributions | |
rou | -0.044 |
ouncing | -0.039 |
rol | -0.038 |
aults | -0.038 |
Carly | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +3.13e-03 |
Pos logit contributions | |
rou | +0.050 |
ouncing | +0.044 |
rol | +0.042 |
liness | +0.042 |
Carly | +0.042 |
Neg logit contributions | |
pe | -0.027 |
Tim | -0.025 |
pepper | -0.024 |
shr | -0.024 |
fought | -0.023 |
Token | sunset |
Token ID | 26428 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
pe | +0.004 |
Tim | +0.004 |
shr | +0.004 |
fought | +0.004 |
Blood | +0.004 |
Neg logit contributions | |
rou | -0.005 |
amaz | -0.004 |
aults | -0.004 |
ouncing | -0.004 |
capt | -0.004 |
Token | together |
Token ID | 1978 |
Feature activation | +9.49e-02 |
Pos logit contributions | |
pe | +0.062 |
Tim | +0.059 |
pepper | +0.058 |
shr | +0.056 |
help | +0.056 |
Neg logit contributions | |
rou | -0.089 |
ouncing | -0.078 |
rol | -0.075 |
liness | -0.075 |
Carly | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
help | +0.086 |
pe | +0.086 |
pepper | +0.084 |
tried | +0.076 |
Tim | +0.075 |
Neg logit contributions | |
rou | -0.215 |
rol | -0.193 |
ouncing | -0.192 |
liness | -0.186 |
idget | -0.181 |
Token | had |
Token ID | 550 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
rou | +0.031 |
Carly | +0.027 |
la | +0.027 |
ouncing | +0.027 |
liness | +0.026 |
Neg logit contributions | |
pe | -0.018 |
Tim | -0.015 |
shr | -0.015 |
� | -0.015 |
pepper | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -1.50e-03 |
Pos logit contributions | |
Tim | +0.020 |
pepper | +0.020 |
pe | +0.018 |
tried | +0.017 |
help | +0.017 |
Neg logit contributions | |
rou | -0.051 |
ouncing | -0.048 |
liness | -0.045 |
rol | -0.045 |
ucking | -0.045 |
Token | happy |
Token ID | 3772 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
rou | +0.003 |
ouncing | +0.003 |
rol | +0.003 |
liness | +0.002 |
ucking | +0.002 |
Neg logit contributions | |
Tim | -0.002 |
pe | -0.002 |
pepper | -0.002 |
help | -0.002 |
shr | -0.002 |
Token | time |
Token ID | 640 |
Feature activation | -1.31e-03 |
Pos logit contributions | |
pe | +0.027 |
shr | +0.025 |
Tim | +0.024 |
pepper | +0.023 |
fought | +0.023 |
Neg logit contributions | |
rou | -0.072 |
ouncing | -0.064 |
Carly | -0.063 |
amaz | -0.062 |
liness | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | +9.99e-04 |
Pos logit contributions | |
rou | +0.009 |
ouncing | +0.008 |
rol | +0.008 |
Carly | +0.007 |
liness | +0.007 |
Neg logit contributions | |
pe | -0.005 |
Tim | -0.005 |
pepper | -0.005 |
help | -0.005 |
fought | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -4.50e-03 |
Pos logit contributions | |
Tim | +0.001 |
pe | +0.001 |
pepper | +0.001 |
shr | +0.001 |
help | +0.001 |
Neg logit contributions | |
rou | -0.002 |
rol | -0.002 |
aults | -0.002 |
ouncing | -0.002 |
liness | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -4.77e-03 |
Pos logit contributions | |
rou | +0.010 |
ouncing | +0.009 |
Carly | +0.008 |
rol | +0.008 |
downward | +0.008 |
Neg logit contributions | |
pe | -0.004 |
pepper | -0.004 |
fought | -0.004 |
shr | -0.004 |
Tim | -0.004 |
Token | A |
Token ID | 32 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
rou | +0.010 |
rol | +0.009 |
ouncing | +0.008 |
amaz | +0.008 |
liness | +0.008 |
Neg logit contributions | |
Tim | -0.006 |
pe | -0.005 |
shr | -0.005 |
tried | -0.005 |
B | -0.005 |
Token | few |
Token ID | 1178 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
pe | +0.128 |
shr | +0.115 |
pepper | +0.113 |
fought | +0.111 |
Tim | +0.110 |
Neg logit contributions | |
rou | -0.134 |
ouncing | -0.114 |
downward | -0.110 |
Carly | -0.109 |
liness | -0.106 |
Token | days |
Token ID | 1528 |
Feature activation | +9.82e-02 |
Pos logit contributions | |
pe | +0.020 |
shr | +0.017 |
Tim | +0.016 |
� | +0.016 |
pepper | +0.016 |
Neg logit contributions | |
rou | -0.050 |
ouncing | -0.044 |
iet | -0.044 |
liness | -0.044 |
Carly | -0.044 |
Token | later |
Token ID | 1568 |
Feature activation | +8.91e-02 |
Pos logit contributions | |
pe | +0.101 |
Tim | +0.094 |
pepper | +0.092 |
shr | +0.088 |
help | +0.085 |
Neg logit contributions | |
rou | -0.209 |
ouncing | -0.186 |
rol | -0.180 |
liness | -0.179 |
Carly | -0.177 |
Token | , |
Token ID | 11 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
pe | +0.125 |
Tim | +0.115 |
pepper | +0.114 |
shr | +0.114 |
fought | +0.107 |
Neg logit contributions | |
rou | -0.156 |
ouncing | -0.137 |
liness | -0.131 |
Carly | -0.131 |
rol | -0.130 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.55e-03 |
Pos logit contributions | |
rou | +0.029 |
ouncing | +0.026 |
rol | +0.025 |
liness | +0.025 |
Carly | +0.025 |
Neg logit contributions | |
pe | -0.010 |
Tim | -0.009 |
pepper | -0.009 |
shr | -0.008 |
fought | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
rou | +0.006 |
rol | +0.006 |
downward | +0.006 |
ouncing | +0.006 |
ancing | +0.006 |
Neg logit contributions | |
tried | -0.002 |
fought | -0.002 |
needed | -0.002 |
had | -0.002 |
pepper | -0.002 |
Token | Max |
Token ID | 5436 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
rou | +0.044 |
rol | +0.040 |
ucking | +0.038 |
ouncing | +0.038 |
capt | +0.038 |
Neg logit contributions | |
Tim | -0.016 |
help | -0.014 |
fought | -0.014 |
tried | -0.013 |
pe | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +2.07e-03 |
Pos logit contributions | |
tried | +0.009 |
fought | +0.008 |
pepper | +0.007 |
needed | +0.007 |
had | +0.007 |
Neg logit contributions | |
rou | -0.029 |
rol | -0.027 |
Carly | -0.027 |
downward | -0.027 |
aj | -0.027 |
Token | happy |
Token ID | 3772 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
pe | +0.002 |
hurt | +0.002 |
Tim | +0.002 |
tried | +0.002 |
help | +0.002 |
Neg logit contributions | |
rou | -0.004 |
liness | -0.004 |
ouncing | -0.004 |
rol | -0.004 |
downward | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -2.47e-03 |
Pos logit contributions | |
pepper | +0.041 |
pe | +0.041 |
Tim | +0.041 |
help | +0.040 |
fought | +0.039 |
Neg logit contributions | |
rou | -0.089 |
rol | -0.083 |
ouncing | -0.079 |
aults | -0.079 |
Carly | -0.078 |
Token | played |
Token ID | 2826 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
rou | +0.006 |
rol | +0.005 |
ouncing | +0.005 |
liness | +0.005 |
downward | +0.005 |
Neg logit contributions | |
Tim | -0.002 |
pe | -0.002 |
pepper | -0.002 |
help | -0.002 |
fought | -0.002 |
Token | music |
Token ID | 2647 |
Feature activation | +2.57e-03 |
Pos logit contributions | |
pe | +0.029 |
Tim | +0.027 |
pepper | +0.027 |
shr | +0.026 |
fought | +0.025 |
Neg logit contributions | |
rou | -0.041 |
ouncing | -0.036 |
liness | -0.035 |
rol | -0.034 |
downward | -0.034 |
Token | together |
Token ID | 1978 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
pe | +0.002 |
help | +0.002 |
Tim | +0.002 |
fought | +0.002 |
pepper | +0.002 |
Neg logit contributions | |
rou | -0.006 |
ouncing | -0.005 |
Carly | -0.005 |
liness | -0.005 |
rol | -0.005 |
Token | under |
Token ID | 739 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
pe | +0.089 |
Tim | +0.080 |
pepper | +0.078 |
shr | +0.074 |
help | +0.073 |
Neg logit contributions | |
rou | -0.239 |
ouncing | -0.215 |
rol | -0.209 |
liness | -0.207 |
Carly | -0.205 |
Token | the |
Token ID | 262 |
Feature activation | +9.84e-03 |
Pos logit contributions | |
pe | +0.030 |
shr | +0.027 |
Tim | +0.027 |
pepper | +0.026 |
fought | +0.025 |
Neg logit contributions | |
rou | -0.046 |
Carly | -0.040 |
liness | -0.040 |
downward | -0.040 |
ouncing | -0.039 |
Token | big |
Token ID | 1263 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
pe | +0.011 |
Tim | +0.010 |
pepper | +0.010 |
shr | +0.010 |
fought | +0.010 |
Neg logit contributions | |
rou | -0.020 |
ouncing | -0.018 |
rol | -0.017 |
liness | -0.017 |
Carly | -0.017 |
Token | tree |
Token ID | 5509 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
rou | +0.067 |
ouncing | +0.059 |
liness | +0.059 |
Carly | +0.058 |
downward | +0.058 |
Neg logit contributions | |
pe | -0.037 |
shr | -0.034 |
Tim | -0.034 |
fought | -0.032 |
pepper | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +6.79e-03 |
Pos logit contributions | |
pe | +0.037 |
Tim | +0.033 |
pepper | +0.032 |
shr | +0.030 |
fought | +0.030 |
Neg logit contributions | |
rou | -0.115 |
ouncing | -0.103 |
rol | -0.101 |
liness | -0.100 |
Carly | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
rou | +0.049 |
rol | +0.044 |
ouncing | +0.044 |
aults | +0.042 |
liness | +0.041 |
Neg logit contributions | |
tried | -0.020 |
pe | -0.020 |
help | -0.020 |
Tim | -0.019 |
pepper | -0.019 |
Token | said |
Token ID | 531 |
Feature activation | +9.50e-03 |
Pos logit contributions | |
rou | +0.026 |
ouncing | +0.024 |
Carly | +0.023 |
liness | +0.023 |
rol | +0.023 |
Neg logit contributions | |
pe | -0.008 |
Tim | -0.007 |
shr | -0.007 |
pepper | -0.007 |
help | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -5.32e-03 |
Pos logit contributions | |
pe | +0.011 |
Tim | +0.011 |
pepper | +0.011 |
help | +0.010 |
shr | +0.010 |
Neg logit contributions | |
rou | -0.019 |
ouncing | -0.016 |
rol | -0.016 |
liness | -0.016 |
Carly | -0.015 |
Token | " |
Token ID | 366 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
rou | +0.008 |
ouncing | +0.007 |
downward | +0.007 |
Carly | +0.007 |
rol | +0.006 |
Neg logit contributions | |
shr | -0.008 |
pe | -0.008 |
Tim | -0.008 |
fought | -0.008 |
pepper | -0.008 |
Token | Maybe |
Token ID | 13300 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
rou | +0.070 |
ouncing | +0.061 |
downward | +0.060 |
rol | +0.059 |
Carly | +0.059 |
Neg logit contributions | |
pe | -0.044 |
pepper | -0.041 |
fought | -0.039 |
shr | -0.038 |
Tim | -0.037 |
Token | later |
Token ID | 1568 |
Feature activation | +9.50e-02 |
Pos logit contributions | |
rou | +0.049 |
Carly | +0.044 |
ouncing | +0.044 |
liness | +0.043 |
downward | +0.043 |
Neg logit contributions | |
pe | -0.028 |
shr | -0.026 |
Tim | -0.024 |
fought | -0.024 |
pepper | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -6.34e-03 |
Pos logit contributions | |
pe | +0.087 |
Tim | +0.084 |
help | +0.084 |
pepper | +0.079 |
fought | +0.078 |
Neg logit contributions | |
rou | -0.214 |
ouncing | -0.189 |
rol | -0.189 |
liness | -0.181 |
ucking | -0.179 |
Token | but |
Token ID | 475 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
rou | +0.015 |
rol | +0.015 |
ouncing | +0.014 |
idget | +0.014 |
la | +0.014 |
Neg logit contributions | |
help | -0.006 |
pepper | -0.005 |
Tim | -0.005 |
tried | -0.004 |
be | -0.004 |
Token | for |
Token ID | 329 |
Feature activation | +8.13e-03 |
Pos logit contributions | |
rou | +0.028 |
ouncing | +0.024 |
liness | +0.024 |
downward | +0.024 |
Carly | +0.024 |
Neg logit contributions | |
pe | -0.017 |
Tim | -0.016 |
shr | -0.015 |
pepper | -0.014 |
fought | -0.014 |
Token | now |
Token ID | 783 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
pe | +0.010 |
Tim | +0.009 |
shr | +0.009 |
pepper | +0.009 |
cube | +0.008 |
Neg logit contributions | |
rou | -0.015 |
Carly | -0.013 |
liness | -0.013 |
ouncing | -0.013 |
rol | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -7.25e-03 |
Pos logit contributions | |
help | +0.030 |
pe | +0.030 |
pepper | +0.028 |
Tim | +0.028 |
fought | +0.027 |
Neg logit contributions | |
rou | -0.080 |
rol | -0.071 |
ouncing | -0.070 |
idget | -0.066 |
liness | -0.066 |
Token | said |
Token ID | 531 |
Feature activation | +9.18e-03 |
Pos logit contributions | |
rou | +0.010 |
Carly | +0.010 |
rol | +0.009 |
ouncing | +0.009 |
Debbie | +0.009 |
Neg logit contributions | |
pe | -0.005 |
Tim | -0.004 |
shr | -0.004 |
pepper | -0.004 |
fought | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +8.15e-03 |
Pos logit contributions | |
pe | +0.013 |
Tim | +0.012 |
shr | +0.012 |
pepper | +0.011 |
timer | +0.011 |
Neg logit contributions | |
rou | -0.014 |
Carly | -0.014 |
Riley | -0.014 |
Debbie | -0.013 |
Lauren | -0.013 |
Token | bird |
Token ID | 6512 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
pe | +0.011 |
Tim | +0.011 |
fought | +0.010 |
shr | +0.010 |
pepper | +0.010 |
Neg logit contributions | |
rou | -0.014 |
liness | -0.012 |
Carly | -0.012 |
ouncing | -0.011 |
amaz | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +7.51e-03 |
Pos logit contributions | |
pe | +0.009 |
Tim | +0.008 |
pepper | +0.008 |
shr | +0.007 |
help | +0.007 |
Neg logit contributions | |
rou | -0.027 |
ouncing | -0.025 |
rol | -0.024 |
liness | -0.024 |
Carly | -0.024 |
Token | " |
Token ID | 366 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
pe | +0.009 |
Tim | +0.008 |
shr | +0.008 |
pepper | +0.008 |
fought | +0.008 |
Neg logit contributions | |
rou | -0.014 |
ouncing | -0.013 |
Carly | -0.012 |
rol | -0.012 |
liness | -0.012 |
Token | Please |
Token ID | 5492 |
Feature activation | +9.01e-02 |
Pos logit contributions | |
rou | +0.065 |
ouncing | +0.057 |
rol | +0.056 |
liness | +0.055 |
Carly | +0.054 |
Neg logit contributions | |
pe | -0.034 |
pepper | -0.033 |
Tim | -0.031 |
fought | -0.031 |
help | -0.030 |
Token | can |
Token ID | 460 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
pe | +0.090 |
Tim | +0.085 |
pepper | +0.083 |
help | +0.082 |
shr | +0.079 |
Neg logit contributions | |
rou | -0.195 |
ouncing | -0.174 |
rol | -0.169 |
liness | -0.167 |
Carly | -0.165 |
Token | I |
Token ID | 314 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
rou | +0.043 |
ouncing | +0.038 |
Carly | +0.038 |
liness | +0.037 |
rol | +0.037 |
Neg logit contributions | |
pe | -0.019 |
pepper | -0.018 |
shr | -0.017 |
Tim | -0.017 |
fought | -0.016 |
Token | have |
Token ID | 423 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
rou | +0.034 |
rol | +0.029 |
ouncing | +0.029 |
liness | +0.029 |
aults | +0.028 |
Neg logit contributions | |
pe | -0.017 |
Tim | -0.015 |
pepper | -0.015 |
shr | -0.015 |
fought | -0.014 |
Token | some |
Token ID | 617 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
pe | +0.023 |
Tim | +0.020 |
shr | +0.019 |
fought | +0.018 |
� | +0.018 |
Neg logit contributions | |
rou | -0.030 |
rol | -0.025 |
Carly | -0.025 |
liness | -0.024 |
la | -0.024 |
Token | ink |
Token ID | 16882 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
rou | +0.046 |
ouncing | +0.040 |
liness | +0.039 |
rol | +0.039 |
downward | +0.038 |
Neg logit contributions | |
pe | -0.024 |
Tim | -0.021 |
shr | -0.020 |
fought | -0.020 |
pepper | -0.019 |
Token | right |
Token ID | 826 |
Feature activation | -6.58e-02 |
Pos logit contributions | |
pe | +0.006 |
help | +0.006 |
pepper | +0.006 |
tried | +0.006 |
fought | +0.006 |
Neg logit contributions | |
rou | -0.010 |
ouncing | -0.009 |
liness | -0.008 |
rol | -0.008 |
Carly | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +8.57e-03 |
Pos logit contributions | |
rou | +0.165 |
ouncing | +0.156 |
rol | +0.153 |
Carly | +0.147 |
liness | +0.146 |
Neg logit contributions | |
pe | -0.044 |
pepper | -0.035 |
Tim | -0.034 |
help | -0.033 |
fought | -0.029 |
Token | He |
Token ID | 679 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
pe | +0.007 |
Tim | +0.007 |
pepper | +0.006 |
shr | +0.006 |
help | +0.006 |
Neg logit contributions | |
rou | -0.020 |
aults | -0.018 |
ouncing | -0.018 |
liness | -0.017 |
rol | -0.017 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
pe | +0.010 |
tried | +0.010 |
pepper | +0.010 |
fought | +0.009 |
Tim | +0.009 |
Neg logit contributions | |
rou | -0.035 |
ouncing | -0.032 |
liness | -0.031 |
rol | -0.031 |
downward | -0.031 |
Token | 't |
Token ID | 470 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
pe | +0.014 |
Tim | +0.013 |
pepper | +0.013 |
shr | +0.012 |
fought | +0.012 |
Neg logit contributions | |
rou | -0.046 |
ouncing | -0.041 |
rol | -0.040 |
liness | -0.040 |
Carly | -0.040 |
Token | resist |
Token ID | 4180 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
Blood | +0.031 |
pepper | +0.030 |
shr | +0.030 |
Tim | +0.028 |
cubes | +0.027 |
Neg logit contributions | |
aults | -0.039 |
ault | -0.036 |
rou | -0.034 |
abel | -0.033 |
ala | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | -1.49e-03 |
Pos logit contributions | |
pe | +0.048 |
pepper | +0.046 |
help | +0.045 |
tried | +0.041 |
fought | +0.040 |
Neg logit contributions | |
rou | -0.119 |
ouncing | -0.113 |
rol | -0.111 |
idget | -0.109 |
iet | -0.105 |
Token | however |
Token ID | 2158 |
Feature activation | +6.42e-02 |
Pos logit contributions | |
rou | +0.003 |
ouncing | +0.003 |
rol | +0.003 |
idget | +0.003 |
iet | +0.003 |
Neg logit contributions | |
pe | -0.001 |
help | -0.001 |
pepper | -0.001 |
tried | -0.001 |
Tim | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.59e-03 |
Pos logit contributions | |
pe | +0.070 |
Tim | +0.064 |
pepper | +0.064 |
help | +0.060 |
fought | +0.059 |
Neg logit contributions | |
rou | -0.136 |
ouncing | -0.121 |
rol | -0.119 |
liness | -0.116 |
Carly | -0.115 |
Token | adding |
Token ID | 4375 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
rou | +0.003 |
ouncing | +0.003 |
rol | +0.003 |
idget | +0.003 |
ucking | +0.003 |
Neg logit contributions | |
pe | -0.002 |
tried | -0.002 |
pepper | -0.002 |
fought | -0.001 |
help | -0.001 |
Token | that |
Token ID | 326 |
Feature activation | -2.62e-03 |
Pos logit contributions | |
rou | +0.032 |
ouncing | +0.029 |
rol | +0.028 |
Carly | +0.027 |
idget | +0.027 |
Neg logit contributions | |
pepper | -0.013 |
pe | -0.013 |
help | -0.013 |
Tim | -0.012 |
tried | -0.011 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
tried | +0.005 |
pe | +0.004 |
fought | +0.004 |
pepper | +0.004 |
help | +0.004 |
Neg logit contributions | |
rou | -0.016 |
ouncing | -0.015 |
rol | -0.015 |
liness | -0.014 |
ucking | -0.014 |
Token | around |
Token ID | 1088 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
tried | +0.014 |
pepper | +0.013 |
help | +0.012 |
looked | +0.012 |
needed | +0.012 |
Neg logit contributions | |
rou | -0.021 |
ouncing | -0.021 |
aters | -0.020 |
aya | -0.019 |
reating | -0.019 |
Token | for |
Token ID | 329 |
Feature activation | +6.59e-04 |
Pos logit contributions | |
rou | +0.037 |
ouncing | +0.033 |
rol | +0.031 |
liness | +0.031 |
Carly | +0.030 |
Neg logit contributions | |
pe | -0.028 |
Tim | -0.027 |
pepper | -0.026 |
shr | -0.025 |
help | -0.025 |
Token | something |
Token ID | 1223 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
pe | +0.001 |
shr | +0.001 |
Tim | +0.001 |
pepper | +0.001 |
fought | +0.001 |
Neg logit contributions | |
rou | -0.001 |
Carly | -0.001 |
liness | -0.001 |
amaz | -0.001 |
downward | -0.001 |
Token | he |
Token ID | 339 |
Feature activation | +1.15e-04 |
Pos logit contributions | |
rou | +0.056 |
ouncing | +0.050 |
rol | +0.049 |
liness | +0.048 |
Carly | +0.048 |
Neg logit contributions | |
pe | -0.029 |
Tim | -0.027 |
pepper | -0.027 |
help | -0.026 |
fought | -0.025 |
Token | could |
Token ID | 714 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
pe | +0.000 |
shr | +0.000 |
Tim | +0.000 |
pepper | +0.000 |
� | +0.000 |
Neg logit contributions | |
rou | -0.000 |
la | -0.000 |
ouncing | -0.000 |
Carly | -0.000 |
liness | -0.000 |
Token | use |
Token ID | 779 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
pe | +0.038 |
Tim | +0.035 |
help | +0.035 |
pepper | +0.034 |
fought | +0.032 |
Neg logit contributions | |
rou | -0.094 |
ouncing | -0.085 |
rol | -0.083 |
liness | -0.081 |
Carly | -0.080 |
Token | to |
Token ID | 284 |
Feature activation | +7.99e-03 |
Pos logit contributions | |
pe | +0.015 |
pepper | +0.014 |
Tim | +0.014 |
help | +0.013 |
shr | +0.013 |
Neg logit contributions | |
rou | -0.038 |
ouncing | -0.034 |
rol | -0.033 |
liness | -0.033 |
Carly | -0.032 |
Token | make |
Token ID | 787 |
Feature activation | +7.03e-03 |
Pos logit contributions | |
pe | +0.009 |
shr | +0.008 |
Tim | +0.008 |
olph | +0.008 |
� | +0.008 |
Neg logit contributions | |
rou | -0.014 |
amaz | -0.013 |
aults | -0.013 |
liness | -0.013 |
Carly | -0.012 |
Token | steam |
Token ID | 13324 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
pe | +0.006 |
pepper | +0.006 |
Tim | +0.006 |
help | +0.006 |
shr | +0.005 |
Neg logit contributions | |
rou | -0.016 |
ouncing | -0.015 |
rol | -0.014 |
liness | -0.014 |
Carly | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -5.31e-03 |
Pos logit contributions | |
rou | +0.138 |
ouncing | +0.122 |
Carly | +0.120 |
Debbie | +0.117 |
rol | +0.116 |
Neg logit contributions | |
pe | -0.061 |
help | -0.057 |
pepper | -0.055 |
Tim | -0.049 |
fought | -0.046 |
Token | her |
Token ID | 607 |
Feature activation | -9.93e-03 |
Pos logit contributions | |
pe | +0.018 |
shr | +0.015 |
cube | +0.015 |
Tim | +0.015 |
pepper | +0.015 |
Neg logit contributions | |
rou | -0.015 |
Carly | -0.014 |
downward | -0.013 |
Lauren | -0.013 |
Debbie | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
rou | +0.021 |
ouncing | +0.019 |
rol | +0.018 |
liness | +0.018 |
Carly | +0.018 |
Neg logit contributions | |
pe | -0.010 |
Tim | -0.009 |
pepper | -0.009 |
shr | -0.009 |
fought | -0.008 |
Token | climb |
Token ID | 12080 |
Feature activation | -6.46e-03 |
Pos logit contributions | |
pe | +0.015 |
pepper | +0.015 |
shr | +0.014 |
Tim | +0.014 |
fought | +0.013 |
Neg logit contributions | |
rou | -0.020 |
ouncing | -0.017 |
aults | -0.017 |
downward | -0.017 |
liness | -0.016 |
Token | by |
Token ID | 416 |
Feature activation | +6.44e-02 |
Pos logit contributions | |
rou | +0.014 |
ouncing | +0.012 |
rol | +0.012 |
liness | +0.012 |
Carly | +0.012 |
Neg logit contributions | |
pe | -0.006 |
Tim | -0.006 |
pepper | -0.006 |
shr | -0.006 |
help | -0.005 |
Token | herself |
Token ID | 5223 |
Feature activation | -4.34e-03 |
Pos logit contributions | |
pe | +0.056 |
pepper | +0.053 |
Tim | +0.052 |
shr | +0.050 |
fought | +0.048 |
Neg logit contributions | |
rou | -0.145 |
ouncing | -0.129 |
liness | -0.126 |
rol | -0.125 |
Carly | -0.125 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
rou | +0.013 |
rol | +0.012 |
Carly | +0.012 |
Debbie | +0.011 |
iet | +0.011 |
Neg logit contributions | |
pe | -0.001 |
help | -0.000 |
Tim | -0.000 |
be | -0.000 |
pepper | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +3.05e-03 |
Pos logit contributions | |
pe | +0.010 |
Tim | +0.009 |
pepper | +0.009 |
shr | +0.009 |
fought | +0.008 |
Neg logit contributions | |
rou | -0.022 |
ouncing | -0.020 |
rol | -0.019 |
liness | -0.019 |
Carly | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +2.95e-03 |
Pos logit contributions | |
pe | +0.003 |
pepper | +0.002 |
Tim | +0.002 |
shr | +0.002 |
fought | +0.002 |
Neg logit contributions | |
rou | -0.007 |
ouncing | -0.006 |
rol | -0.006 |
liness | -0.006 |
Carly | -0.006 |
Token | She |
Token ID | 3347 |
Feature activation | -8.77e-03 |
Pos logit contributions | |
pe | +0.003 |
Tim | +0.003 |
pepper | +0.003 |
shr | +0.003 |
help | +0.003 |
Neg logit contributions | |
rou | -0.006 |
ouncing | -0.005 |
rol | -0.005 |
liness | -0.005 |
Carly | -0.005 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
rou | +0.016 |
la | +0.015 |
abel | +0.014 |
ouncing | +0.014 |
Carly | +0.014 |
Neg logit contributions | |
pe | -0.010 |
� | -0.010 |
pepper | -0.009 |
occasion | -0.009 |
help | -0.009 |
Token | around |
Token ID | 1088 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
tried | +0.034 |
help | +0.032 |
needed | +0.030 |
pepper | +0.028 |
looked | +0.028 |
Neg logit contributions | |
rou | -0.060 |
lor | -0.059 |
rol | -0.055 |
liness | -0.054 |
ucking | -0.054 |
Token | Jen |
Token ID | 13364 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
rou | +0.024 |
ouncing | +0.021 |
Carly | +0.021 |
rol | +0.021 |
downward | +0.020 |
Neg logit contributions | |
pe | -0.020 |
shr | -0.019 |
Tim | -0.018 |
pepper | -0.018 |
fought | -0.017 |
Token | says |
Token ID | 1139 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
rou | +0.119 |
rol | +0.110 |
aults | +0.104 |
liness | +0.102 |
ayed | +0.101 |
Neg logit contributions | |
tried | -0.047 |
pe | -0.045 |
Tim | -0.042 |
crack | -0.041 |
help | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +7.11e-03 |
Pos logit contributions | |
pe | +0.009 |
Tim | +0.008 |
pepper | +0.008 |
shr | +0.008 |
fought | +0.007 |
Neg logit contributions | |
rou | -0.028 |
ouncing | -0.025 |
rol | -0.024 |
liness | -0.024 |
Carly | -0.024 |
Token | " |
Token ID | 366 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
pe | +0.010 |
Tim | +0.009 |
pepper | +0.009 |
shr | +0.009 |
fought | +0.009 |
Neg logit contributions | |
rou | -0.012 |
ouncing | -0.011 |
rol | -0.010 |
Carly | -0.010 |
liness | -0.010 |
Token | Look |
Token ID | 8567 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
rou | +0.073 |
aults | +0.067 |
ouncing | +0.065 |
downward | +0.065 |
ucking | +0.063 |
Neg logit contributions | |
pepper | -0.048 |
pe | -0.047 |
help | -0.046 |
fought | -0.045 |
hammer | -0.042 |
Token | at |
Token ID | 379 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
Tim | +0.055 |
help | +0.055 |
pepper | +0.054 |
pe | +0.051 |
tried | +0.051 |
Neg logit contributions | |
rou | -0.104 |
rol | -0.093 |
liness | -0.092 |
ouncing | -0.089 |
capt | -0.088 |
Token | our |
Token ID | 674 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
pe | +0.012 |
Tim | +0.012 |
pepper | +0.012 |
help | +0.011 |
shr | +0.011 |
Neg logit contributions | |
rou | -0.023 |
ouncing | -0.021 |
rol | -0.020 |
liness | -0.020 |
aults | -0.019 |
Token | ball |
Token ID | 2613 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
rou | +0.112 |
rol | +0.108 |
la | +0.102 |
ouncing | +0.101 |
liness | +0.100 |
Neg logit contributions | |
fight | -0.038 |
help | -0.037 |
cube | -0.034 |
pepper | -0.033 |
Tim | -0.033 |
Token | go |
Token ID | 467 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
rou | +0.026 |
Carly | +0.023 |
ouncing | +0.023 |
liness | +0.023 |
rol | +0.022 |
Neg logit contributions | |
pe | -0.014 |
Tim | -0.013 |
pepper | -0.013 |
fought | -0.013 |
help | -0.013 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
rou | +0.129 |
ouncing | +0.114 |
liness | +0.111 |
amaz | +0.109 |
rol | +0.108 |
Neg logit contributions | |
pe | -0.086 |
Tim | -0.084 |
shr | -0.078 |
pepper | -0.078 |
fought | -0.076 |
Token |
|
Token ID | 198 |
Feature activation | -3.13e-03 |
Pos logit contributions | |
rou | +0.035 |
ouncing | +0.032 |
Carly | +0.031 |
downward | +0.031 |
rol | +0.030 |
Neg logit contributions | |
pe | -0.014 |
Tim | -0.013 |
shr | -0.013 |
pepper | -0.013 |
fought | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +4.56e-03 |
Pos logit contributions | |
rou | +0.091 |
ouncing | +0.086 |
aj | +0.080 |
amaz | +0.079 |
rol | +0.079 |
Neg logit contributions | |
pe | -0.022 |
Tim | -0.020 |
help | -0.019 |
fought | -0.017 |
tried | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -9.33e-04 |
Pos logit contributions | |
timer | +0.007 |
ts | +0.007 |
pe | +0.006 |
occasion | +0.006 |
shr | +0.006 |
Neg logit contributions | |
rou | -0.007 |
ouncing | -0.007 |
ounce | -0.006 |
Carly | -0.006 |
rol | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -7.27e-04 |
Pos logit contributions | |
rou | +0.002 |
ouncing | +0.001 |
Carly | +0.001 |
ounce | +0.001 |
rol | +0.001 |
Neg logit contributions | |
timer | -0.001 |
pepper | -0.001 |
pe | -0.001 |
fought | -0.001 |
shr | -0.001 |
Token | All |
Token ID | 3237 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
rou | +0.001 |
ouncing | +0.001 |
aults | +0.001 |
ounce | +0.001 |
ucking | +0.001 |
Neg logit contributions | |
timer | -0.001 |
pepper | -0.001 |
pe | -0.001 |
cube | -0.001 |
occasion | -0.001 |
Token | of |
Token ID | 286 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
pe | +0.066 |
Tim | +0.064 |
pepper | +0.062 |
shr | +0.060 |
help | +0.060 |
Neg logit contributions | |
rou | -0.082 |
ouncing | -0.071 |
rol | -0.068 |
liness | -0.068 |
Carly | -0.066 |
Token | Nut |
Token ID | 11959 |
Feature activation | +2.41e-03 |
Pos logit contributions | |
rou | +0.039 |
Carly | +0.034 |
downward | +0.032 |
amaz | +0.031 |
Rita | +0.031 |
Neg logit contributions | |
pe | -0.036 |
timer | -0.035 |
shr | -0.034 |
cube | -0.033 |
occasion | -0.032 |
Token | ty |
Token ID | 774 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
pe | +0.003 |
pepper | +0.003 |
Tim | +0.003 |
shr | +0.003 |
help | +0.003 |
Neg logit contributions | |
rou | -0.004 |
ouncing | -0.004 |
liness | -0.003 |
rol | -0.003 |
Carly | -0.003 |
Token | 's |
Token ID | 338 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
rou | +0.114 |
ouncing | +0.102 |
rol | +0.099 |
liness | +0.099 |
Carly | +0.097 |
Neg logit contributions | |
pe | -0.045 |
Tim | -0.042 |
pepper | -0.041 |
shr | -0.039 |
help | -0.038 |
Token | friends |
Token ID | 2460 |
Feature activation | +7.32e-03 |
Pos logit contributions | |
rou | +0.074 |
ouncing | +0.063 |
liness | +0.060 |
rol | +0.059 |
Carly | +0.059 |
Neg logit contributions | |
pe | -0.063 |
Tim | -0.061 |
pepper | -0.059 |
shr | -0.058 |
fought | -0.056 |
Token | were |
Token ID | 547 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
pe | +0.006 |
Tim | +0.005 |
fought | +0.005 |
pepper | +0.005 |
help | +0.005 |
Neg logit contributions | |
rou | -0.017 |
ouncing | -0.016 |
rol | -0.015 |
liness | -0.015 |
downward | -0.015 |
Token | sad |
Token ID | 6507 |
Feature activation | +5.34e-03 |
Pos logit contributions | |
pe | +0.031 |
hurt | +0.030 |
fought | +0.029 |
tried | +0.029 |
help | +0.028 |
Neg logit contributions | |
rou | -0.065 |
ouncing | -0.062 |
liness | -0.060 |
downward | -0.058 |
rol | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | -2.59e-03 |
Pos logit contributions | |
rou | +0.039 |
ouncing | +0.032 |
rol | +0.032 |
liness | +0.032 |
ucking | +0.031 |
Neg logit contributions | |
pe | -0.024 |
Tim | -0.023 |
help | -0.022 |
tried | -0.021 |
fought | -0.021 |
Token | slide |
Token ID | 10649 |
Feature activation | -9.18e-03 |
Pos logit contributions | |
rou | +0.006 |
ucking | +0.005 |
liness | +0.005 |
ayed | +0.005 |
rol | +0.005 |
Neg logit contributions | |
Tim | -0.002 |
tried | -0.002 |
pe | -0.002 |
help | -0.002 |
fought | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -2.53e-03 |
Pos logit contributions | |
rou | +0.018 |
ouncing | +0.015 |
rol | +0.015 |
liness | +0.015 |
ucking | +0.015 |
Neg logit contributions | |
pe | -0.011 |
Tim | -0.011 |
pepper | -0.010 |
fought | -0.010 |
help | -0.010 |
Token | swing |
Token ID | 9628 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
rou | +0.006 |
liness | +0.005 |
rol | +0.005 |
ouncing | +0.005 |
ayed | +0.005 |
Neg logit contributions | |
pe | -0.003 |
Tim | -0.003 |
tried | -0.003 |
help | -0.003 |
pepper | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -7.31e-03 |
Pos logit contributions | |
rou | +0.076 |
ouncing | +0.064 |
liness | +0.063 |
rol | +0.062 |
ucking | +0.061 |
Neg logit contributions | |
pe | -0.046 |
Tim | -0.045 |
pepper | -0.040 |
help | -0.040 |
fought | -0.039 |
Token | climb |
Token ID | 12080 |
Feature activation | -5.36e-03 |
Pos logit contributions | |
rou | +0.013 |
aults | +0.012 |
ouncing | +0.012 |
downward | +0.012 |
Carly | +0.012 |
Neg logit contributions | |
pe | -0.009 |
pepper | -0.008 |
shr | -0.008 |
Tim | -0.008 |
Blood | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +8.62e-03 |
Pos logit contributions | |
rou | +0.012 |
ouncing | +0.011 |
rol | +0.010 |
liness | +0.010 |
Carly | +0.010 |
Neg logit contributions | |
pe | -0.005 |
Tim | -0.005 |
pepper | -0.005 |
shr | -0.004 |
help | -0.004 |
Token | They |
Token ID | 1119 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
pe | +0.007 |
Tim | +0.007 |
pepper | +0.006 |
shr | +0.006 |
help | +0.006 |
Neg logit contributions | |
rou | -0.020 |
ouncing | -0.018 |
rol | -0.018 |
liness | -0.017 |
Carly | -0.017 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
pe | +0.015 |
Tim | +0.015 |
fought | +0.013 |
tried | +0.013 |
help | +0.012 |
Neg logit contributions | |
rou | -0.066 |
ouncing | -0.060 |
liness | -0.059 |
rol | -0.058 |
ucking | -0.058 |
Token | a |
Token ID | 257 |
Feature activation | +5.42e-03 |
Pos logit contributions | |
Tim | +0.014 |
pepper | +0.014 |
pe | +0.012 |
help | +0.012 |
tried | +0.011 |
Neg logit contributions | |
rou | -0.049 |
ouncing | -0.046 |
liness | -0.044 |
rol | -0.043 |
aults | -0.043 |
Token | big |
Token ID | 1263 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
pepper | +0.006 |
Tim | +0.006 |
pe | +0.006 |
shr | +0.005 |
help | +0.005 |
Neg logit contributions | |
rou | -0.011 |
rol | -0.010 |
ouncing | -0.010 |
liness | -0.010 |
Carly | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -6.57e-03 |
Pos logit contributions | |
pe | +0.014 |
pepper | +0.011 |
Tim | +0.011 |
tried | +0.010 |
help | +0.010 |
Neg logit contributions | |
rou | -0.059 |
Carly | -0.054 |
ouncing | -0.053 |
rol | -0.052 |
liness | -0.052 |
Token | Bill |
Token ID | 3941 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
rou | +0.010 |
ouncing | +0.009 |
Carly | +0.009 |
rol | +0.008 |
downward | +0.008 |
Neg logit contributions | |
shr | -0.009 |
pe | -0.009 |
occasion | -0.009 |
ts | -0.009 |
help | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
rou | +0.024 |
ouncing | +0.022 |
rol | +0.021 |
liness | +0.021 |
Carly | +0.021 |
Neg logit contributions | |
pe | -0.009 |
Tim | -0.008 |
pepper | -0.008 |
shr | -0.008 |
help | -0.008 |
Token | scared |
Token ID | 12008 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
rou | +0.030 |
ouncing | +0.028 |
liness | +0.028 |
rol | +0.027 |
reating | +0.026 |
Neg logit contributions | |
pe | -0.016 |
tried | -0.016 |
fought | -0.015 |
hurt | -0.015 |
help | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -6.90e-03 |
Pos logit contributions | |
rou | +0.065 |
rol | +0.058 |
ouncing | +0.057 |
liness | +0.056 |
aults | +0.056 |
Neg logit contributions | |
Tim | -0.032 |
pe | -0.031 |
help | -0.030 |
pepper | -0.029 |
tried | -0.027 |
Token | He |
Token ID | 679 |
Feature activation | -4.69e-03 |
Pos logit contributions | |
rou | +0.009 |
ouncing | +0.009 |
Carly | +0.009 |
rol | +0.008 |
Debbie | +0.008 |
Neg logit contributions | |
shr | -0.010 |
ts | -0.010 |
pepper | -0.010 |
occasion | -0.010 |
help | -0.009 |
Token | shouted |
Token ID | 17293 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
rou | +0.010 |
ouncing | +0.009 |
Carly | +0.009 |
downward | +0.009 |
la | +0.009 |
Neg logit contributions | |
pe | -0.004 |
Tim | -0.004 |
pepper | -0.004 |
shr | -0.004 |
� | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
help | +0.035 |
pe | +0.030 |
Tim | +0.028 |
pepper | +0.028 |
tried | +0.028 |
Neg logit contributions | |
rou | -0.083 |
rol | -0.076 |
packs | -0.074 |
liness | -0.073 |
la | -0.073 |
Token | " |
Token ID | 366 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
rou | +0.033 |
ouncing | +0.029 |
downward | +0.027 |
Carly | +0.027 |
rol | +0.027 |
Neg logit contributions | |
pe | -0.026 |
Tim | -0.026 |
shr | -0.025 |
pepper | -0.025 |
fought | -0.024 |
Token | The |
Token ID | 464 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
rou | +0.083 |
ouncing | +0.073 |
ucking | +0.072 |
capt | +0.071 |
aults | +0.071 |
Neg logit contributions | |
pepper | -0.066 |
pe | -0.063 |
help | -0.063 |
ts | -0.060 |
cube | -0.060 |
Token | cat |
Token ID | 3797 |
Feature activation | +1.30e-03 |
Pos logit contributions | |
pe | +0.052 |
ts | +0.046 |
fought | +0.046 |
shr | +0.046 |
olph | +0.045 |
Neg logit contributions | |
rou | -0.080 |
ouncing | -0.068 |
liness | -0.065 |
downward | -0.064 |
ucking | -0.062 |
Token | ina |
Token ID | 1437 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
pe | +0.126 |
Tim | +0.115 |
shr | +0.112 |
pepper | +0.110 |
occasion | +0.105 |
Neg logit contributions | |
rou | -0.100 |
downward | -0.098 |
amaz | -0.095 |
aults | -0.090 |
Carly | -0.088 |
Token | , |
Token ID | 11 |
Feature activation | -2.89e-03 |
Pos logit contributions | |
rou | +0.082 |
rol | +0.075 |
downward | +0.072 |
aults | +0.072 |
reating | +0.072 |
Neg logit contributions | |
pe | -0.036 |
help | -0.035 |
pepper | -0.034 |
tried | -0.032 |
fought | -0.032 |
Token | manage |
Token ID | 6687 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
rou | +0.006 |
rol | +0.006 |
ucking | +0.006 |
ouncing | +0.006 |
packs | +0.006 |
Neg logit contributions | |
help | -0.004 |
be | -0.003 |
pepper | -0.003 |
hammer | -0.003 |
wait | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
help | +0.011 |
pepper | +0.010 |
tried | +0.009 |
pe | +0.009 |
be | +0.009 |
Neg logit contributions | |
rol | -0.023 |
rou | -0.023 |
ucking | -0.021 |
ouncing | -0.021 |
idget | -0.021 |
Token | pick |
Token ID | 2298 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
pe | +0.011 |
Tim | +0.010 |
pepper | +0.010 |
shr | +0.010 |
fought | +0.009 |
Neg logit contributions | |
rou | -0.026 |
ouncing | -0.023 |
rol | -0.023 |
liness | -0.023 |
Carly | -0.022 |
Token | one |
Token ID | 530 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
rou | +0.043 |
ouncing | +0.041 |
aults | +0.037 |
ucking | +0.037 |
capt | +0.037 |
Neg logit contributions | |
pepper | -0.018 |
help | -0.015 |
pe | -0.015 |
Tim | -0.015 |
shr | -0.013 |
Token | ". |
Token ID | 1911 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
rou | +0.055 |
ouncing | +0.050 |
rol | +0.049 |
Carly | +0.048 |
liness | +0.047 |
Neg logit contributions | |
pepper | -0.027 |
help | -0.026 |
pe | -0.026 |
Tim | -0.026 |
shr | -0.023 |
Token | Gina |
Token ID | 43002 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
rou | +0.051 |
ouncing | +0.043 |
rol | +0.042 |
aults | +0.041 |
liness | +0.041 |
Neg logit contributions | |
pe | -0.039 |
Tim | -0.038 |
pepper | -0.037 |
help | -0.037 |
shr | -0.036 |
Token | thought |
Token ID | 1807 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
rou | +0.093 |
ouncing | +0.084 |
rol | +0.083 |
liness | +0.081 |
downward | +0.080 |
Neg logit contributions | |
pe | -0.036 |
Tim | -0.032 |
pepper | -0.031 |
fought | -0.030 |
shr | -0.029 |
Token | really |
Token ID | 1107 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
help | +0.030 |
pe | +0.028 |
pepper | +0.027 |
Tim | +0.026 |
tried | +0.025 |
Neg logit contributions | |
rou | -0.054 |
rol | -0.050 |
ouncing | -0.049 |
liness | -0.048 |
aults | -0.048 |
Token | hard |
Token ID | 1327 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
pe | +0.072 |
pepper | +0.070 |
shr | +0.070 |
Tim | +0.068 |
fought | +0.067 |
Neg logit contributions | |
rou | -0.096 |
ouncing | -0.084 |
downward | -0.081 |
Carly | -0.081 |
liness | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | -7.02e-04 |
Pos logit contributions | |
rou | +0.073 |
ouncing | +0.065 |
Carly | +0.062 |
rol | +0.062 |
liness | +0.061 |
Neg logit contributions | |
pe | -0.039 |
Tim | -0.036 |
pepper | -0.035 |
fought | -0.034 |
help | -0.034 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
rou | +0.002 |
ouncing | +0.002 |
rol | +0.001 |
liness | +0.001 |
aults | +0.001 |
Neg logit contributions | |
pe | -0.001 |
Tim | -0.000 |
pepper | -0.000 |
shr | -0.000 |
help | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
pe | +0.026 |
help | +0.025 |
pepper | +0.023 |
tried | +0.023 |
Tim | +0.023 |
Neg logit contributions | |
rou | -0.068 |
ouncing | -0.060 |
rol | -0.058 |
liness | -0.058 |
idget | -0.057 |
Token | lightning |
Token ID | 14357 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
rou | +0.033 |
ouncing | +0.030 |
liness | +0.029 |
rol | +0.029 |
aults | +0.028 |
Neg logit contributions | |
pe | -0.009 |
help | -0.007 |
Tim | -0.007 |
pepper | -0.007 |
shr | -0.007 |
Token | struck |
Token ID | 7425 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
rou | +0.033 |
ouncing | +0.030 |
rol | +0.029 |
liness | +0.029 |
Carly | +0.028 |
Neg logit contributions | |
pe | -0.013 |
Tim | -0.012 |
pepper | -0.011 |
shr | -0.011 |
help | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
pe | +0.027 |
Tim | +0.025 |
pepper | +0.024 |
shr | +0.023 |
help | +0.023 |
Neg logit contributions | |
rou | -0.056 |
ouncing | -0.050 |
liness | -0.048 |
rol | -0.048 |
aults | -0.047 |
Token | El |
Token ID | 2574 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
rou | +0.034 |
ouncing | +0.029 |
rol | +0.029 |
liness | +0.028 |
Carly | +0.028 |
Neg logit contributions | |
pe | -0.024 |
Tim | -0.023 |
pepper | -0.023 |
shr | -0.022 |
fought | -0.022 |
Token | mer |
Token ID | 647 |
Feature activation | +7.07e-02 |
Pos logit contributions | |
rou | +0.148 |
ouncing | +0.121 |
la | +0.118 |
rol | +0.116 |
liness | +0.116 |
Neg logit contributions | |
Tim | -0.161 |
pe | -0.160 |
pepper | -0.154 |
shr | -0.149 |
help | -0.148 |
Token | was |
Token ID | 373 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
pe | +0.054 |
fought | +0.050 |
tried | +0.050 |
Tim | +0.048 |
pepper | +0.047 |
Neg logit contributions | |
rou | -0.172 |
ouncing | -0.155 |
liness | -0.150 |
rol | -0.149 |
downward | -0.148 |
Token | j |
Token ID | 474 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
rou | +0.022 |
ouncing | +0.022 |
liness | +0.022 |
Carly | +0.021 |
packs | +0.020 |
Neg logit contributions | |
hurt | -0.010 |
tried | -0.009 |
hit | -0.009 |
burned | -0.008 |
pe | -0.008 |
Token | ol |
Token ID | 349 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
rou | +0.019 |
ouncing | +0.016 |
rol | +0.015 |
liness | +0.015 |
Carly | +0.014 |
Neg logit contributions | |
pe | -0.030 |
Tim | -0.029 |
pepper | -0.029 |
shr | -0.028 |
help | -0.028 |
Token | course |
Token ID | 1781 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
Blood | +0.000 |
sidewalk | +0.000 |
hammer | +0.000 |
stacked | +0.000 |
pe | +0.000 |
Neg logit contributions | |
rou | -0.000 |
Carly | -0.000 |
la | -0.000 |
liness | -0.000 |
thanks | -0.000 |
Token | ! |
Token ID | 0 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
pe | +0.067 |
Tim | +0.064 |
pepper | +0.062 |
help | +0.060 |
shr | +0.058 |
Neg logit contributions | |
rou | -0.143 |
ouncing | -0.127 |
rol | -0.124 |
liness | -0.123 |
Carly | -0.120 |
Token | Here |
Token ID | 3423 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
rou | +0.074 |
ouncing | +0.066 |
rol | +0.063 |
liness | +0.063 |
Carly | +0.062 |
Neg logit contributions | |
pe | -0.049 |
Tim | -0.046 |
pepper | -0.046 |
shr | -0.044 |
fought | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
pe | +0.019 |
help | +0.017 |
pepper | +0.017 |
Tim | +0.017 |
tried | +0.016 |
Neg logit contributions | |
rou | -0.059 |
ouncing | -0.053 |
rol | -0.052 |
ucking | -0.050 |
idget | -0.050 |
Token | pick |
Token ID | 2298 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
rou | +0.036 |
ouncing | +0.034 |
rol | +0.032 |
ucking | +0.032 |
idget | +0.032 |
Neg logit contributions | |
help | -0.013 |
Tim | -0.011 |
pe | -0.011 |
pepper | -0.011 |
tried | -0.010 |
Token | some |
Token ID | 617 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
rou | +0.071 |
ouncing | +0.065 |
aults | +0.061 |
amaz | +0.061 |
idget | +0.060 |
Neg logit contributions | |
pe | -0.027 |
pepper | -0.027 |
Tim | -0.025 |
help | -0.024 |
tried | -0.022 |
Token | of |
Token ID | 286 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
rou | +0.068 |
ouncing | +0.062 |
liness | +0.058 |
Carly | +0.058 |
rol | +0.058 |
Neg logit contributions | |
pepper | -0.035 |
help | -0.034 |
pe | -0.034 |
Tim | -0.032 |
shr | -0.030 |
Token | these |
Token ID | 777 |
Feature activation | -7.10e-02 |
Pos logit contributions | |
rou | +0.072 |
ouncing | +0.063 |
rol | +0.060 |
liness | +0.060 |
Carly | +0.058 |
Neg logit contributions | |
pe | -0.049 |
pepper | -0.047 |
Tim | -0.047 |
help | -0.045 |
shr | -0.044 |
Token | flowers |
Token ID | 12734 |
Feature activation | +3.57e-03 |
Pos logit contributions | |
rou | +0.148 |
ouncing | +0.131 |
rol | +0.126 |
liness | +0.124 |
Carly | +0.123 |
Neg logit contributions | |
pe | -0.076 |
pepper | -0.073 |
help | -0.073 |
Tim | -0.069 |
shr | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
pe | +0.004 |
pepper | +0.003 |
help | +0.003 |
Tim | +0.003 |
shr | +0.003 |
Neg logit contributions | |
rou | -0.008 |
ouncing | -0.007 |
Carly | -0.007 |
downward | -0.007 |
liness | -0.007 |
Token | put |
Token ID | 1234 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
rou | +0.047 |
ouncing | +0.042 |
rol | +0.041 |
liness | +0.041 |
Carly | +0.040 |
Neg logit contributions | |
pe | -0.022 |
Tim | -0.021 |
pepper | -0.020 |
shr | -0.019 |
help | -0.019 |
Token | of |
Token ID | 286 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
rou | +0.025 |
ouncing | +0.022 |
liness | +0.021 |
Carly | +0.021 |
rol | +0.020 |
Neg logit contributions | |
pe | -0.015 |
Tim | -0.014 |
help | -0.014 |
pepper | -0.014 |
fought | -0.013 |
Token | snow |
Token ID | 6729 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
rou | +0.055 |
liness | +0.047 |
rol | +0.046 |
ouncing | +0.045 |
downward | +0.045 |
Neg logit contributions | |
pe | -0.045 |
shr | -0.042 |
� | -0.041 |
Tim | -0.040 |
fought | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
rou | +0.018 |
ouncing | +0.016 |
liness | +0.015 |
rol | +0.015 |
downward | +0.015 |
Neg logit contributions | |
Tim | -0.013 |
shr | -0.013 |
pe | -0.013 |
pepper | -0.012 |
help | -0.012 |
Token | stack |
Token ID | 8931 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
rou | +0.026 |
ouncing | +0.022 |
aults | +0.022 |
liness | +0.022 |
downward | +0.022 |
Neg logit contributions | |
pe | -0.014 |
Tim | -0.013 |
shr | -0.012 |
pepper | -0.012 |
� | -0.012 |
Token | them |
Token ID | 606 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
rou | +0.035 |
ouncing | +0.031 |
rol | +0.030 |
liness | +0.030 |
Carly | +0.029 |
Neg logit contributions | |
pe | -0.019 |
Tim | -0.018 |
pepper | -0.017 |
shr | -0.017 |
fought | -0.016 |
Token | up |
Token ID | 510 |
Feature activation | -9.27e-02 |
Pos logit contributions | |
pe | +0.045 |
Tim | +0.044 |
pepper | +0.043 |
shr | +0.042 |
fought | +0.041 |
Neg logit contributions | |
rou | -0.060 |
ouncing | -0.052 |
rol | -0.050 |
liness | -0.050 |
downward | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +4.65e-03 |
Pos logit contributions | |
rou | +0.204 |
ouncing | +0.182 |
rol | +0.177 |
liness | +0.176 |
Carly | +0.173 |
Neg logit contributions | |
pe | -0.088 |
Tim | -0.082 |
pepper | -0.080 |
shr | -0.077 |
fought | -0.074 |
Token | They |
Token ID | 1119 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
pe | +0.005 |
shr | +0.005 |
pepper | +0.004 |
fought | +0.004 |
help | +0.004 |
Neg logit contributions | |
rou | -0.009 |
ouncing | -0.009 |
Carly | -0.008 |
rol | -0.008 |
downward | -0.008 |
Token | find |
Token ID | 1064 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
pe | +0.016 |
Tim | +0.015 |
pepper | +0.015 |
help | +0.015 |
fought | +0.014 |
Neg logit contributions | |
rou | -0.047 |
ouncing | -0.042 |
rol | -0.041 |
liness | -0.041 |
Carly | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | -2.14e-03 |
Pos logit contributions | |
pe | +0.013 |
Tim | +0.012 |
pepper | +0.012 |
shr | +0.011 |
help | +0.011 |
Neg logit contributions | |
rou | -0.031 |
ouncing | -0.028 |
rol | -0.027 |
liness | -0.027 |
Carly | -0.026 |
Token | hat |
Token ID | 6877 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
rou | +0.004 |
ouncing | +0.004 |
rol | +0.004 |
liness | +0.004 |
Carly | +0.004 |
Neg logit contributions | |
pe | -0.002 |
pepper | -0.002 |
Tim | -0.002 |
help | -0.002 |
shr | -0.002 |
Token | any |
Token ID | 597 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
rou | +0.044 |
ouncing | +0.042 |
rol | +0.040 |
ucking | +0.039 |
aj | +0.037 |
Neg logit contributions | |
help | -0.017 |
pe | -0.016 |
pepper | -0.016 |
tried | -0.015 |
ts | -0.015 |
Token | struggle |
Token ID | 6531 |
Feature activation | -8.74e-04 |
Pos logit contributions | |
rou | +0.040 |
ouncing | +0.036 |
rol | +0.035 |
Carly | +0.034 |
liness | +0.033 |
Neg logit contributions | |
pe | -0.028 |
pepper | -0.028 |
help | -0.028 |
Tim | -0.027 |
fought | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -2.05e-03 |
Pos logit contributions | |
rou | +0.002 |
ouncing | +0.002 |
Carly | +0.002 |
rol | +0.002 |
Debbie | +0.002 |
Neg logit contributions | |
pe | -0.001 |
help | -0.001 |
pepper | -0.001 |
Tim | -0.001 |
tried | -0.001 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -6.58e-02 |
Pos logit contributions | |
rou | +0.004 |
ouncing | +0.004 |
rol | +0.004 |
liness | +0.004 |
Carly | +0.004 |
Neg logit contributions | |
pe | -0.002 |
pepper | -0.002 |
fought | -0.002 |
shr | -0.002 |
Tim | -0.002 |
Token | Once |
Token ID | 7454 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
rou | +0.136 |
ouncing | +0.120 |
liness | +0.119 |
rol | +0.119 |
Carly | +0.118 |
Neg logit contributions | |
pe | -0.067 |
pepper | -0.062 |
fought | -0.060 |
shr | -0.060 |
Tim | -0.058 |
Token | upon |
Token ID | 2402 |
Feature activation | -9.92e-02 |
Pos logit contributions | |
rou | +0.052 |
ouncing | +0.039 |
rol | +0.039 |
aults | +0.038 |
liness | +0.037 |
Neg logit contributions | |
Tim | -0.116 |
pe | -0.115 |
pepper | -0.112 |
help | -0.112 |
tried | -0.108 |
Token | a |
Token ID | 257 |
Feature activation | -7.27e-03 |
Pos logit contributions | |
rou | +0.173 |
downward | +0.153 |
Carly | +0.146 |
aults | +0.145 |
ouncing | +0.141 |
Neg logit contributions | |
pe | -0.132 |
pepper | -0.121 |
shr | -0.119 |
fought | -0.118 |
timer | -0.111 |
Token | time |
Token ID | 640 |
Feature activation | -3.22e-03 |
Pos logit contributions | |
rou | +0.016 |
ouncing | +0.014 |
rol | +0.014 |
liness | +0.014 |
Carly | +0.013 |
Neg logit contributions | |
pe | -0.007 |
Tim | -0.006 |
pepper | -0.006 |
shr | -0.006 |
fought | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
rou | +0.007 |
rol | +0.006 |
ouncing | +0.006 |
liness | +0.006 |
aults | +0.006 |
Neg logit contributions | |
pe | -0.003 |
Tim | -0.003 |
help | -0.003 |
pepper | -0.003 |
tried | -0.003 |
Token | there |
Token ID | 612 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
rou | +0.018 |
ouncing | +0.015 |
downward | +0.014 |
liness | +0.014 |
rol | +0.013 |
Neg logit contributions | |
pe | -0.025 |
pepper | -0.024 |
shr | -0.023 |
fought | -0.023 |
Tim | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
rou | +0.086 |
ouncing | +0.077 |
liness | +0.076 |
rol | +0.075 |
Carly | +0.074 |
Neg logit contributions | |
pe | -0.034 |
pepper | -0.030 |
Tim | -0.030 |
shr | -0.030 |
fought | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
rou | +0.020 |
ouncing | +0.018 |
liness | +0.017 |
rol | +0.017 |
Carly | +0.017 |
Neg logit contributions | |
pe | -0.010 |
Tim | -0.009 |
pepper | -0.009 |
shr | -0.009 |
fought | -0.008 |
Token | easier |
Token ID | 4577 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
rou | +0.020 |
ouncing | +0.017 |
liness | +0.016 |
downward | +0.015 |
capt | +0.015 |
Neg logit contributions | |
pe | -0.015 |
Tim | -0.014 |
pepper | -0.013 |
shr | -0.013 |
olph | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +2.09e-03 |
Pos logit contributions | |
rou | +0.037 |
rol | +0.033 |
ouncing | +0.033 |
Carly | +0.031 |
ucking | +0.031 |
Neg logit contributions | |
pe | -0.019 |
help | -0.018 |
pepper | -0.018 |
Tim | -0.017 |
fought | -0.016 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.73e-02 |
Pos logit contributions | |
pe | +0.002 |
fought | +0.002 |
pepper | +0.002 |
Tim | +0.002 |
shr | +0.002 |
Neg logit contributions | |
rou | -0.004 |
ouncing | -0.004 |
liness | -0.004 |
rol | -0.004 |
Carly | -0.004 |
Token | Once |
Token ID | 7454 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
rou | +0.120 |
ouncing | +0.106 |
rol | +0.105 |
liness | +0.105 |
Carly | +0.104 |
Neg logit contributions | |
pe | -0.058 |
pepper | -0.053 |
fought | -0.052 |
Tim | -0.051 |
shr | -0.051 |
Token | upon |
Token ID | 2402 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
rou | +0.050 |
aults | +0.040 |
rol | +0.039 |
ouncing | +0.038 |
liness | +0.038 |
Neg logit contributions | |
Tim | -0.089 |
pe | -0.087 |
help | -0.087 |
pepper | -0.085 |
tried | -0.083 |
Token | a |
Token ID | 257 |
Feature activation | +1.61e-03 |
Pos logit contributions | |
rou | +0.165 |
downward | +0.144 |
Carly | +0.139 |
aults | +0.137 |
ouncing | +0.137 |
Neg logit contributions | |
pe | -0.113 |
pepper | -0.104 |
shr | -0.102 |
fought | -0.101 |
Tim | -0.095 |
Token | time |
Token ID | 640 |
Feature activation | +5.04e-03 |
Pos logit contributions | |
pe | +0.001 |
Tim | +0.001 |
pepper | +0.001 |
shr | +0.001 |
help | +0.001 |
Neg logit contributions | |
rou | -0.004 |
ouncing | -0.003 |
rol | -0.003 |
liness | -0.003 |
Carly | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -4.42e-03 |
Pos logit contributions | |
Tim | +0.005 |
pe | +0.004 |
help | +0.004 |
pepper | +0.004 |
tried | +0.004 |
Neg logit contributions | |
rou | -0.012 |
rol | -0.010 |
liness | -0.010 |
aults | -0.010 |
ouncing | -0.010 |
Token | there |
Token ID | 612 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
rou | +0.007 |
ouncing | +0.006 |
rol | +0.006 |
liness | +0.006 |
Carly | +0.006 |
Neg logit contributions | |
pe | -0.007 |
Tim | -0.006 |
pepper | -0.006 |
shr | -0.006 |
fought | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -2.32e-03 |
Pos logit contributions | |
rou | +0.076 |
ouncing | +0.068 |
rol | +0.066 |
liness | +0.066 |
aults | +0.065 |
Neg logit contributions | |
pe | -0.021 |
Tim | -0.020 |
pepper | -0.018 |
shr | -0.017 |
help | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -1.50e-04 |
Pos logit contributions | |
help | +0.010 |
pe | +0.009 |
pepper | +0.009 |
tried | +0.008 |
Tim | +0.007 |
Neg logit contributions | |
rou | -0.031 |
rol | -0.028 |
rig | -0.028 |
idget | -0.028 |
liness | -0.028 |
Token | " |
Token ID | 366 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
rou | +0.000 |
ouncing | +0.000 |
rol | +0.000 |
liness | +0.000 |
Carly | +0.000 |
Neg logit contributions | |
pe | -0.000 |
Tim | -0.000 |
pepper | -0.000 |
help | -0.000 |
shr | -0.000 |
Token | Now |
Token ID | 3844 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
rou | +0.051 |
aults | +0.048 |
ouncing | +0.047 |
downward | +0.045 |
ucking | +0.045 |
Neg logit contributions | |
pepper | -0.038 |
help | -0.037 |
pe | -0.037 |
hammer | -0.037 |
cube | -0.036 |
Token | it |
Token ID | 340 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
help | +0.038 |
be | +0.029 |
Tim | +0.029 |
pepper | +0.028 |
tried | +0.028 |
Neg logit contributions | |
rou | -0.153 |
rol | -0.140 |
ucking | -0.136 |
ouncing | -0.134 |
idget | -0.131 |
Token | 's |
Token ID | 338 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
pe | +0.047 |
pepper | +0.046 |
tried | +0.045 |
help | +0.044 |
fought | +0.043 |
Neg logit contributions | |
rou | -0.083 |
Carly | -0.073 |
ouncing | -0.072 |
aults | -0.070 |
rol | -0.070 |
Token | my |
Token ID | 616 |
Feature activation | -9.32e-02 |
Pos logit contributions | |
rou | +0.093 |
ouncing | +0.081 |
rol | +0.079 |
liness | +0.079 |
downward | +0.077 |
Neg logit contributions | |
pe | -0.052 |
Tim | -0.048 |
pepper | -0.046 |
shr | -0.045 |
fought | -0.044 |
Token | turn |
Token ID | 1210 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
rou | +0.173 |
ouncing | +0.153 |
rol | +0.150 |
liness | +0.147 |
Carly | +0.146 |
Neg logit contributions | |
pe | -0.120 |
pepper | -0.116 |
Tim | -0.113 |
help | -0.111 |
shr | -0.106 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.66e-03 |
Pos logit contributions | |
rou | +0.066 |
Carly | +0.062 |
ouncing | +0.061 |
amaz | +0.060 |
rol | +0.060 |
Neg logit contributions | |
pe | -0.021 |
pepper | -0.021 |
help | -0.020 |
cube | -0.019 |
Tim | -0.019 |
Token | She |
Token ID | 1375 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
rou | +0.003 |
ouncing | +0.003 |
Carly | +0.003 |
downward | +0.003 |
liness | +0.003 |
Neg logit contributions | |
pe | -0.002 |
shr | -0.002 |
fought | -0.002 |
pepper | -0.002 |
Tim | -0.002 |
Token | squ |
Token ID | 2809 |
Feature activation | +4.35e-02 |
Pos logit contributions | |
rou | +0.060 |
ouncing | +0.055 |
rol | +0.054 |
liness | +0.054 |
Carly | +0.053 |
Neg logit contributions | |
pe | -0.020 |
Tim | -0.019 |
pepper | -0.019 |
fought | -0.018 |
tried | -0.017 |
Token | irted |
Token ID | 48357 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
pe | +0.077 |
Tim | +0.075 |
pepper | +0.074 |
shr | +0.072 |
fought | +0.071 |
Neg logit contributions | |
rou | -0.060 |
ouncing | -0.049 |
rol | -0.047 |
liness | -0.047 |
Carly | -0.046 |
Token |
|
Token ID | 198 |
Feature activation | +4.55e-03 |
Pos logit contributions | |
pe | +0.000 |
Tim | +0.000 |
pepper | +0.000 |
shr | +0.000 |
fought | +0.000 |
Neg logit contributions | |
rou | -0.001 |
ouncing | -0.001 |
rol | -0.001 |
liness | -0.001 |
Carly | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +4.41e-03 |
Pos logit contributions | |
pe | +0.004 |
Tim | +0.004 |
pepper | +0.004 |
shr | +0.003 |
help | +0.003 |
Neg logit contributions | |
rou | -0.010 |
ouncing | -0.009 |
rol | -0.009 |
liness | -0.009 |
Carly | -0.009 |
Token | They |
Token ID | 2990 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
pe | +0.005 |
Tim | +0.005 |
pepper | +0.005 |
shr | +0.004 |
help | +0.004 |
Neg logit contributions | |
rou | -0.009 |
ouncing | -0.008 |
rol | -0.008 |
liness | -0.008 |
Carly | -0.008 |
Token | both |
Token ID | 1111 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
pe | +0.017 |
Tim | +0.017 |
help | +0.015 |
pepper | +0.015 |
fought | +0.015 |
Neg logit contributions | |
rou | -0.053 |
ouncing | -0.049 |
liness | -0.047 |
rol | -0.047 |
ucking | -0.046 |
Token | fell |
Token ID | 3214 |
Feature activation | +5.00e-03 |
Pos logit contributions | |
pe | +0.033 |
Tim | +0.031 |
pepper | +0.030 |
shr | +0.028 |
fought | +0.028 |
Neg logit contributions | |
rou | -0.089 |
ouncing | -0.080 |
rol | -0.078 |
liness | -0.077 |
Carly | -0.076 |
Token | asleep |
Token ID | 16039 |
Feature activation | -9.31e-02 |
Pos logit contributions | |
help | +0.010 |
Tim | +0.010 |
tried | +0.010 |
pe | +0.010 |
shr | +0.009 |
Neg logit contributions | |
rou | -0.006 |
aults | -0.004 |
ouncing | -0.004 |
amaz | -0.004 |
Carly | -0.004 |
Token | on |
Token ID | 319 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
rou | +0.135 |
asleep | +0.118 |
ouncing | +0.113 |
aults | +0.109 |
liness | +0.108 |
Neg logit contributions | |
Tim | -0.152 |
shr | -0.145 |
pepper | -0.143 |
pe | -0.141 |
cube | -0.132 |
Token | the |
Token ID | 262 |
Feature activation | +2.43e-03 |
Pos logit contributions | |
rou | +0.021 |
liness | +0.018 |
Carly | +0.018 |
ouncing | +0.018 |
downward | +0.018 |
Neg logit contributions | |
pe | -0.013 |
Tim | -0.012 |
shr | -0.011 |
pepper | -0.011 |
fought | -0.011 |
Token | couch |
Token ID | 18507 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
pe | +0.003 |
Tim | +0.003 |
pepper | +0.003 |
shr | +0.003 |
help | +0.003 |
Neg logit contributions | |
rou | -0.004 |
ouncing | -0.004 |
rol | -0.004 |
liness | -0.004 |
Carly | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -8.76e-03 |
Pos logit contributions | |
rou | +0.062 |
ouncing | +0.055 |
rol | +0.054 |
Carly | +0.053 |
liness | +0.052 |
Neg logit contributions | |
pe | -0.033 |
Tim | -0.030 |
pepper | -0.029 |
help | -0.028 |
fought | -0.028 |
Token | dreaming |
Token ID | 28368 |
Feature activation | -8.93e-02 |
Pos logit contributions | |
rou | +0.018 |
rol | +0.016 |
ouncing | +0.016 |
ayed | +0.015 |
liness | +0.015 |
Neg logit contributions | |
pe | -0.010 |
tried | -0.009 |
Tim | -0.009 |
help | -0.009 |
pepper | -0.009 |