Token | happily |
Token ID | 18177 |
Feature activation | -9.47e-03 |
Pos logit contributions | |
envelop | +0.092 |
Offic | +0.087 |
ons | +0.082 |
Amen | +0.078 |
read | +0.077 |
Neg logit contributions | |
tub | -0.123 |
gan | -0.113 |
ider | -0.113 |
ses | -0.104 |
Moe | -0.103 |
Token | started |
Token ID | 2067 |
Feature activation | +5.92e-02 |
Pos logit contributions | |
envelop | +0.017 |
Offic | +0.015 |
Amen | +0.015 |
ons | +0.014 |
oc | +0.013 |
Neg logit contributions | |
tub | -0.019 |
gan | -0.019 |
Moe | -0.017 |
ider | -0.017 |
ahl | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
tub | +0.106 |
gan | +0.100 |
Moe | +0.090 |
ider | +0.087 |
Mil | +0.085 |
Neg logit contributions | |
envelop | -0.123 |
Offic | -0.114 |
Amen | -0.113 |
ons | -0.106 |
Mic | -0.104 |
Token | go |
Token ID | 467 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
envelop | +0.042 |
Offic | +0.039 |
Amen | +0.038 |
ons | +0.036 |
Mic | +0.035 |
Neg logit contributions | |
tub | -0.041 |
gan | -0.039 |
Moe | -0.035 |
ider | -0.034 |
Mil | -0.034 |
Token | down |
Token ID | 866 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
tub | +0.024 |
gan | +0.023 |
ider | +0.020 |
Moe | +0.020 |
ses | +0.020 |
Neg logit contributions | |
envelop | -0.027 |
Offic | -0.025 |
Amen | -0.025 |
ons | -0.024 |
Mic | -0.022 |
Token | it |
Token ID | 340 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
tub | +0.159 |
gan | +0.151 |
ider | +0.136 |
ses | +0.133 |
Moe | +0.132 |
Neg logit contributions | |
envelop | -0.156 |
Offic | -0.144 |
Amen | -0.140 |
ons | -0.137 |
oc | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
tub | +0.253 |
gan | +0.248 |
ses | +0.230 |
sed | +0.223 |
Moe | +0.222 |
Neg logit contributions | |
envelop | -0.217 |
ons | -0.192 |
Offic | -0.190 |
read | -0.187 |
Amen | -0.182 |
Token |
|
Token ID | 198 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
tub | +0.023 |
gan | +0.021 |
ider | +0.020 |
Moe | +0.019 |
ahl | +0.018 |
Neg logit contributions | |
envelop | -0.028 |
Offic | -0.026 |
Amen | -0.026 |
ons | -0.025 |
Mic | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -9.88e-03 |
Pos logit contributions | |
envelop | +0.035 |
Offic | +0.032 |
Amen | +0.032 |
ons | +0.030 |
itivity | +0.030 |
Neg logit contributions | |
tub | -0.017 |
gan | -0.015 |
Moe | -0.013 |
ahl | -0.012 |
Mil | -0.012 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
envelop | +0.019 |
Amen | +0.017 |
ons | +0.016 |
Offic | +0.016 |
parcels | +0.015 |
Neg logit contributions | |
tub | -0.019 |
gan | -0.018 |
ider | -0.017 |
Moe | -0.016 |
ahl | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
tub | +0.131 |
gan | +0.115 |
ider | +0.115 |
ipples | +0.113 |
ses | +0.108 |
Neg logit contributions | |
envelop | -0.078 |
ons | -0.076 |
read | -0.064 |
Offic | -0.061 |
Amen | -0.059 |
Token | when |
Token ID | 618 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
tub | +0.124 |
gan | +0.113 |
ider | +0.104 |
ses | +0.103 |
ahl | +0.101 |
Neg logit contributions | |
envelop | -0.139 |
ons | -0.128 |
Offic | -0.124 |
Amen | -0.118 |
read | -0.114 |
Token | she |
Token ID | 673 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
tub | +0.093 |
gan | +0.088 |
ider | +0.078 |
Moe | +0.076 |
ahl | +0.074 |
Neg logit contributions | |
envelop | -0.122 |
Offic | -0.114 |
Amen | -0.111 |
ons | -0.108 |
Mic | -0.102 |
Token | was |
Token ID | 373 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
tub | +0.085 |
gan | +0.083 |
ses | +0.073 |
ider | +0.073 |
Moe | +0.071 |
Neg logit contributions | |
envelop | -0.110 |
Offic | -0.104 |
Amen | -0.103 |
ons | -0.102 |
oc | -0.098 |
Token | walking |
Token ID | 6155 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
tub | +0.076 |
gan | +0.073 |
Moe | +0.067 |
ider | +0.066 |
ahl | +0.065 |
Neg logit contributions | |
envelop | -0.058 |
Offic | -0.053 |
Amen | -0.052 |
ons | -0.050 |
Mic | -0.045 |
Token | outside |
Token ID | 2354 |
Feature activation | +8.09e-02 |
Pos logit contributions | |
gan | +0.046 |
tub | +0.045 |
ses | +0.044 |
fault | +0.043 |
sed | +0.042 |
Neg logit contributions | |
envelop | -0.029 |
Amen | -0.027 |
ons | -0.026 |
read | -0.026 |
saddle | -0.025 |
Token | it |
Token ID | 340 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
gan | +0.168 |
tub | +0.161 |
ider | +0.154 |
ses | +0.153 |
ahl | +0.147 |
Neg logit contributions | |
envelop | -0.134 |
ons | -0.132 |
read | -0.120 |
Amen | -0.118 |
Offic | -0.114 |
Token | started |
Token ID | 2067 |
Feature activation | +8.42e-02 |
Pos logit contributions | |
gan | +0.310 |
fault | +0.294 |
ses | +0.292 |
sed | +0.274 |
ider | +0.273 |
Neg logit contributions | |
read | -0.173 |
ons | -0.172 |
envelop | -0.143 |
Offic | -0.131 |
chat | -0.131 |
Token | to |
Token ID | 284 |
Feature activation | -8.38e-03 |
Pos logit contributions | |
gan | +0.184 |
tub | +0.180 |
ahl | +0.167 |
ses | +0.167 |
Moe | +0.165 |
Neg logit contributions | |
envelop | -0.127 |
ons | -0.115 |
Offic | -0.108 |
read | -0.104 |
Amen | -0.102 |
Token | snow |
Token ID | 6729 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
envelop | +0.016 |
Offic | +0.014 |
Amen | +0.014 |
read | +0.013 |
ons | +0.013 |
Neg logit contributions | |
gan | -0.016 |
tub | -0.016 |
ider | -0.015 |
Moe | -0.014 |
ahl | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
gan | +0.076 |
ses | +0.071 |
ahl | +0.070 |
tub | +0.070 |
Moe | +0.070 |
Neg logit contributions | |
envelop | -0.062 |
ons | -0.060 |
Offic | -0.056 |
Amen | -0.056 |
oc | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
tub | +0.065 |
gan | +0.061 |
ider | +0.060 |
ahl | +0.057 |
ses | +0.057 |
Neg logit contributions | |
envelop | -0.057 |
ons | -0.051 |
Offic | -0.050 |
read | -0.047 |
Amen | -0.047 |
Token | almost |
Token ID | 2048 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
envelop | +0.015 |
Offic | +0.015 |
Amen | +0.014 |
Mic | +0.013 |
oc | +0.012 |
Neg logit contributions | |
tub | -0.012 |
gan | -0.011 |
Moe | -0.010 |
Mil | -0.009 |
ider | -0.009 |
Token | every |
Token ID | 790 |
Feature activation | -7.87e-04 |
Pos logit contributions | |
tub | +0.073 |
gan | +0.067 |
ahl | +0.061 |
ses | +0.061 |
Moe | +0.061 |
Neg logit contributions | |
envelop | -0.055 |
ons | -0.053 |
Offic | -0.052 |
read | -0.049 |
rode | -0.048 |
Token | day |
Token ID | 1110 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
envelop | +0.002 |
Offic | +0.002 |
Amen | +0.002 |
ons | +0.001 |
Mic | +0.001 |
Neg logit contributions | |
tub | -0.001 |
gan | -0.001 |
Moe | -0.001 |
ider | -0.001 |
ahl | -0.001 |
Token | sliding |
Token ID | 22292 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
tub | +0.079 |
gan | +0.074 |
ider | +0.067 |
Moe | +0.066 |
Mil | +0.062 |
Neg logit contributions | |
envelop | -0.102 |
Offic | -0.096 |
Amen | -0.094 |
ons | -0.089 |
Mic | -0.087 |
Token | down |
Token ID | 866 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
tub | +0.060 |
gan | +0.059 |
ses | +0.057 |
ider | +0.055 |
sed | +0.054 |
Neg logit contributions | |
envelop | -0.043 |
ons | -0.039 |
Offic | -0.037 |
saddle | -0.035 |
Amen | -0.035 |
Token | it |
Token ID | 340 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
tub | +0.176 |
gan | +0.174 |
ses | +0.161 |
ider | +0.159 |
sed | +0.158 |
Neg logit contributions | |
envelop | -0.146 |
ons | -0.130 |
Offic | -0.129 |
Amen | -0.120 |
oc | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
gan | +0.231 |
tub | +0.230 |
ses | +0.204 |
Moe | +0.203 |
ider | +0.200 |
Neg logit contributions | |
envelop | -0.238 |
ons | -0.214 |
Offic | -0.209 |
Amen | -0.201 |
oc | -0.192 |
Token | One |
Token ID | 1881 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
tub | +0.031 |
ider | +0.030 |
gan | +0.030 |
ses | +0.029 |
sed | +0.029 |
Neg logit contributions | |
envelop | -0.025 |
Amen | -0.023 |
oc | -0.023 |
Offic | -0.023 |
ons | -0.020 |
Token | day |
Token ID | 1110 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
tub | +0.025 |
ider | +0.024 |
ses | +0.024 |
gan | +0.024 |
sed | +0.024 |
Neg logit contributions | |
envelop | -0.015 |
ons | -0.014 |
Offic | -0.014 |
oc | -0.014 |
read | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
tub | +0.141 |
ses | +0.138 |
gan | +0.130 |
ider | +0.127 |
sed | +0.125 |
Neg logit contributions | |
envelop | -0.104 |
ons | -0.098 |
oc | -0.092 |
Offic | -0.087 |
rode | -0.087 |
Token | was |
Token ID | 373 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
ses | +0.241 |
ider | +0.227 |
sed | +0.218 |
tub | +0.213 |
gan | +0.210 |
Neg logit contributions | |
rode | -0.138 |
read | -0.131 |
Offic | -0.130 |
envelop | -0.128 |
ons | -0.125 |
Token | When |
Token ID | 1649 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
tub | +0.027 |
gan | +0.025 |
ider | +0.024 |
ahl | +0.023 |
ses | +0.023 |
Neg logit contributions | |
envelop | -0.031 |
Offic | -0.030 |
Amen | -0.028 |
ons | -0.027 |
Mic | -0.027 |
Token | it |
Token ID | 340 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
tub | +0.105 |
gan | +0.100 |
ider | +0.096 |
ses | +0.091 |
ahl | +0.090 |
Neg logit contributions | |
envelop | -0.068 |
Offic | -0.061 |
ons | -0.059 |
Amen | -0.058 |
read | -0.050 |
Token | reached |
Token ID | 4251 |
Feature activation | +4.04e-02 |
Pos logit contributions | |
ses | +0.226 |
tub | +0.221 |
fault | +0.216 |
gan | +0.214 |
ider | +0.206 |
Neg logit contributions | |
rode | -0.144 |
ons | -0.140 |
Offic | -0.140 |
read | -0.139 |
oc | -0.130 |
Token | the |
Token ID | 262 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
ider | +0.118 |
tub | +0.114 |
inter | +0.110 |
gan | +0.108 |
auc | +0.105 |
Neg logit contributions | |
envelop | -0.035 |
ons | -0.024 |
Offic | -0.018 |
exchange | -0.018 |
helicopters | -0.017 |
Token | bottom |
Token ID | 4220 |
Feature activation | +9.96e-02 |
Pos logit contributions | |
gan | +0.105 |
tub | +0.105 |
ider | +0.093 |
ahl | +0.092 |
Mil | +0.090 |
Neg logit contributions | |
envelop | -0.119 |
Offic | -0.110 |
Amen | -0.102 |
saddle | -0.098 |
oc | -0.097 |
Token | it |
Token ID | 340 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
ses | +0.213 |
gan | +0.205 |
tub | +0.204 |
ahl | +0.201 |
sed | +0.199 |
Neg logit contributions | |
ons | -0.149 |
envelop | -0.137 |
Offic | -0.109 |
oc | -0.108 |
read | -0.106 |
Token | went |
Token ID | 1816 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
gan | +0.266 |
ses | +0.262 |
tub | +0.252 |
ider | +0.244 |
fault | +0.244 |
Neg logit contributions | |
ons | -0.163 |
rode | -0.162 |
read | -0.152 |
envelop | -0.134 |
chat | -0.132 |
Token | rolling |
Token ID | 10708 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
gan | +0.068 |
tub | +0.067 |
ses | +0.065 |
sed | +0.060 |
Mil | +0.059 |
Neg logit contributions | |
envelop | -0.055 |
Amen | -0.050 |
Offic | -0.049 |
ons | -0.047 |
saddle | -0.045 |
Token | off |
Token ID | 572 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
tub | +0.053 |
gan | +0.052 |
ider | +0.051 |
ses | +0.049 |
sed | +0.047 |
Neg logit contributions | |
envelop | -0.034 |
ons | -0.029 |
saddle | -0.027 |
Amen | -0.026 |
Offic | -0.026 |
Token | into |
Token ID | 656 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
tub | +0.071 |
gan | +0.070 |
ider | +0.065 |
ses | +0.063 |
ahl | +0.062 |
Neg logit contributions | |
envelop | -0.052 |
ons | -0.045 |
Offic | -0.043 |
Amen | -0.041 |
arrest | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
tub | +0.049 |
gan | +0.047 |
ider | +0.044 |
Moe | +0.041 |
ahl | +0.041 |
Neg logit contributions | |
envelop | -0.042 |
Offic | -0.038 |
Amen | -0.036 |
ons | -0.035 |
Mic | -0.032 |
Token | he |
Token ID | 339 |
Feature activation | +5.44e-02 |
Pos logit contributions | |
tub | +0.078 |
gan | +0.072 |
Moe | +0.063 |
ider | +0.063 |
ahl | +0.060 |
Neg logit contributions | |
envelop | -0.118 |
Offic | -0.111 |
Amen | -0.108 |
ons | -0.105 |
Mic | -0.101 |
Token | climbed |
Token ID | 19952 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
tub | +0.101 |
gan | +0.096 |
Moe | +0.085 |
ider | +0.085 |
ahl | +0.082 |
Neg logit contributions | |
envelop | -0.108 |
Offic | -0.101 |
Amen | -0.098 |
ons | -0.095 |
Mic | -0.090 |
Token | on |
Token ID | 319 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
tub | +0.037 |
gan | +0.036 |
ider | +0.035 |
ses | +0.034 |
sed | +0.034 |
Neg logit contributions | |
ons | -0.016 |
envelop | -0.016 |
saddle | -0.015 |
Offic | -0.015 |
oc | -0.014 |
Token | top |
Token ID | 1353 |
Feature activation | +7.48e-02 |
Pos logit contributions | |
tub | +0.022 |
gan | +0.021 |
ider | +0.019 |
Moe | +0.018 |
ahl | +0.018 |
Neg logit contributions | |
envelop | -0.019 |
Offic | -0.017 |
Amen | -0.017 |
ons | -0.017 |
oc | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | +8.67e-04 |
Pos logit contributions | |
ses | +0.210 |
gan | +0.208 |
sed | +0.206 |
tub | +0.201 |
ider | +0.196 |
Neg logit contributions | |
ons | -0.071 |
envelop | -0.057 |
oc | -0.052 |
Offic | -0.041 |
arrest | -0.039 |
Token | it |
Token ID | 340 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
tub | +0.002 |
gan | +0.001 |
Moe | +0.001 |
ider | +0.001 |
ahl | +0.001 |
Neg logit contributions | |
envelop | -0.002 |
Offic | -0.002 |
Amen | -0.002 |
ons | -0.002 |
Mic | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
ses | +0.336 |
gan | +0.334 |
sed | +0.317 |
tub | +0.302 |
fault | +0.300 |
Neg logit contributions | |
ons | -0.123 |
read | -0.102 |
oc | -0.088 |
course | -0.085 |
rode | -0.075 |
Token | he |
Token ID | 339 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
tub | +0.056 |
gan | +0.054 |
ider | +0.047 |
Moe | +0.046 |
ahl | +0.045 |
Neg logit contributions | |
envelop | -0.073 |
Offic | -0.067 |
Amen | -0.066 |
ons | -0.064 |
Mic | -0.061 |
Token | realized |
Token ID | 6939 |
Feature activation | +5.23e-03 |
Pos logit contributions | |
tub | +0.045 |
gan | +0.043 |
Moe | +0.039 |
ider | +0.039 |
ses | +0.038 |
Neg logit contributions | |
envelop | -0.040 |
Offic | -0.038 |
ons | -0.037 |
oc | -0.034 |
read | -0.034 |
Token | that |
Token ID | 326 |
Feature activation | +8.89e-03 |
Pos logit contributions | |
tub | +0.011 |
gan | +0.010 |
ider | +0.009 |
ses | +0.009 |
ahl | +0.009 |
Neg logit contributions | |
envelop | -0.010 |
Offic | -0.009 |
ons | -0.009 |
Amen | -0.008 |
oc | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
tub | +0.016 |
gan | +0.015 |
Moe | +0.013 |
ider | +0.013 |
ahl | +0.013 |
Neg logit contributions | |
envelop | -0.018 |
Offic | -0.017 |
Amen | -0.017 |
ons | -0.016 |
Mic | -0.015 |
Token | hill |
Token ID | 12788 |
Feature activation | +5.58e-02 |
Pos logit contributions | |
envelop | +0.075 |
Offic | +0.069 |
Amen | +0.068 |
ons | +0.065 |
Mic | +0.062 |
Neg logit contributions | |
tub | -0.067 |
gan | -0.064 |
ider | -0.057 |
Moe | -0.057 |
ahl | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
gan | +0.107 |
tub | +0.103 |
ses | +0.098 |
Moe | +0.097 |
ahl | +0.097 |
Neg logit contributions | |
envelop | -0.092 |
ons | -0.092 |
Offic | -0.082 |
Amen | -0.080 |
oc | -0.077 |
Token | wanted |
Token ID | 2227 |
Feature activation | -8.54e-03 |
Pos logit contributions | |
envelop | +0.083 |
Offic | +0.076 |
Amen | +0.073 |
ons | +0.071 |
oc | +0.066 |
Neg logit contributions | |
tub | -0.093 |
gan | -0.090 |
ider | -0.084 |
Moe | -0.080 |
ahl | -0.080 |
Token | to |
Token ID | 284 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
envelop | +0.015 |
ons | +0.014 |
Offic | +0.014 |
Amen | +0.013 |
parcels | +0.012 |
Neg logit contributions | |
tub | -0.018 |
gan | -0.016 |
ider | -0.016 |
ahl | -0.015 |
Moe | -0.015 |
Token | climb |
Token ID | 12080 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
envelop | +0.064 |
Offic | +0.060 |
Amen | +0.059 |
Mic | +0.055 |
ons | +0.055 |
Neg logit contributions | |
tub | -0.059 |
gan | -0.057 |
Moe | -0.052 |
Mil | -0.050 |
ahl | -0.048 |
Token | it |
Token ID | 340 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
tub | +0.035 |
gan | +0.033 |
ider | +0.032 |
ses | +0.031 |
sed | +0.030 |
Neg logit contributions | |
envelop | -0.020 |
ons | -0.017 |
Offic | -0.016 |
Amen | -0.016 |
saddle | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
tub | +0.242 |
gan | +0.240 |
ses | +0.231 |
sed | +0.227 |
ider | +0.217 |
Neg logit contributions | |
envelop | -0.209 |
ons | -0.190 |
Offic | -0.176 |
Amen | -0.176 |
oc | -0.169 |
Token | but |
Token ID | 475 |
Feature activation | +3.32e-03 |
Pos logit contributions | |
tub | +0.034 |
gan | +0.032 |
ider | +0.028 |
ahl | +0.027 |
Moe | +0.027 |
Neg logit contributions | |
envelop | -0.052 |
Offic | -0.049 |
Amen | -0.048 |
ons | -0.046 |
Mic | -0.044 |
Token | it |
Token ID | 340 |
Feature activation | +7.32e-02 |
Pos logit contributions | |
tub | +0.007 |
ider | +0.006 |
gan | +0.006 |
ahl | +0.006 |
ses | +0.006 |
Neg logit contributions | |
envelop | -0.005 |
Offic | -0.005 |
ons | -0.005 |
Amen | -0.005 |
read | -0.004 |
Token | looked |
Token ID | 3114 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
tub | +0.133 |
ider | +0.117 |
gan | +0.117 |
Moe | +0.111 |
ses | +0.109 |
Neg logit contributions | |
ons | -0.131 |
envelop | -0.130 |
Offic | -0.127 |
oc | -0.124 |
read | -0.118 |
Token | too |
Token ID | 1165 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
tub | +0.071 |
gan | +0.067 |
ider | +0.063 |
ahl | +0.061 |
ses | +0.060 |
Neg logit contributions | |
envelop | -0.063 |
Amen | -0.052 |
ons | -0.051 |
Offic | -0.050 |
parcels | -0.049 |
Token | that |
Token ID | 326 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
envelop | +0.084 |
Offic | +0.079 |
Amen | +0.077 |
ons | +0.074 |
Mic | +0.069 |
Neg logit contributions | |
tub | -0.089 |
gan | -0.082 |
ider | -0.076 |
ahl | -0.072 |
ses | -0.072 |
Token | day |
Token ID | 1110 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
tub | +0.040 |
gan | +0.038 |
ider | +0.033 |
Moe | +0.033 |
ahl | +0.032 |
Neg logit contributions | |
envelop | -0.048 |
Offic | -0.045 |
Amen | -0.044 |
ons | -0.043 |
oc | -0.041 |
Token | on |
Token ID | 319 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
tub | +0.086 |
gan | +0.083 |
Moe | +0.075 |
ider | +0.075 |
ahl | +0.072 |
Neg logit contributions | |
envelop | -0.069 |
Offic | -0.063 |
Amen | -0.062 |
ons | -0.059 |
Mic | -0.055 |
Token | , |
Token ID | 11 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
tub | +0.023 |
gan | +0.023 |
ider | +0.020 |
ses | +0.020 |
ahl | +0.020 |
Neg logit contributions | |
envelop | -0.024 |
ons | -0.023 |
Amen | -0.021 |
Offic | -0.020 |
oc | -0.020 |
Token | whenever |
Token ID | 8797 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
tub | +0.051 |
gan | +0.047 |
ider | +0.044 |
ses | +0.042 |
ahl | +0.042 |
Neg logit contributions | |
envelop | -0.056 |
Offic | -0.052 |
Amen | -0.050 |
ons | -0.049 |
oc | -0.047 |
Token | it |
Token ID | 340 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
tub | +0.106 |
gan | +0.098 |
ider | +0.093 |
ses | +0.091 |
sed | +0.089 |
Neg logit contributions | |
envelop | -0.096 |
Offic | -0.089 |
Amen | -0.087 |
ons | -0.082 |
read | -0.079 |
Token | r |
Token ID | 374 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
gan | +0.236 |
tub | +0.223 |
ses | +0.214 |
Moe | +0.210 |
ider | +0.205 |
Neg logit contributions | |
ons | -0.215 |
Offic | -0.203 |
oc | -0.203 |
envelop | -0.203 |
Amen | -0.199 |
Token | ained |
Token ID | 1328 |
Feature activation | +6.47e-02 |
Pos logit contributions | |
tub | +0.090 |
gan | +0.080 |
Moe | +0.079 |
Mil | +0.075 |
ahl | +0.075 |
Neg logit contributions | |
envelop | -0.125 |
ons | -0.123 |
Offic | -0.119 |
Amen | -0.111 |
Mic | -0.110 |
Token | , |
Token ID | 11 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
gan | +0.134 |
ses | +0.119 |
tub | +0.119 |
ahl | +0.119 |
ider | +0.116 |
Neg logit contributions | |
ons | -0.102 |
envelop | -0.096 |
Amen | -0.089 |
lending | -0.088 |
course | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
tub | +0.065 |
gan | +0.061 |
ider | +0.059 |
ses | +0.057 |
ahl | +0.056 |
Neg logit contributions | |
envelop | -0.062 |
Offic | -0.056 |
ons | -0.053 |
Amen | -0.052 |
Mic | -0.049 |
Token | gir |
Token ID | 37370 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
tub | +0.096 |
gan | +0.093 |
ider | +0.084 |
Moe | +0.084 |
ahl | +0.082 |
Neg logit contributions | |
envelop | -0.078 |
Amen | -0.072 |
Offic | -0.070 |
ons | -0.067 |
Mic | -0.062 |
Token | could |
Token ID | 714 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
gan | +0.029 |
tub | +0.029 |
Moe | +0.026 |
ider | +0.025 |
ahl | +0.025 |
Neg logit contributions | |
envelop | -0.036 |
ons | -0.035 |
Amen | -0.032 |
Offic | -0.031 |
oc | -0.031 |
Token | appear |
Token ID | 1656 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
tub | +0.029 |
gan | +0.027 |
Moe | +0.025 |
ider | +0.024 |
ahl | +0.024 |
Neg logit contributions | |
envelop | -0.029 |
Offic | -0.027 |
Amen | -0.026 |
ons | -0.025 |
Mic | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
tub | +0.091 |
gan | +0.091 |
ses | +0.081 |
Moe | +0.081 |
ider | +0.081 |
Neg logit contributions | |
envelop | -0.088 |
ons | -0.076 |
Offic | -0.076 |
Amen | -0.074 |
oc | -0.070 |
Token | disappear |
Token ID | 10921 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
envelop | +0.082 |
Offic | +0.076 |
Amen | +0.074 |
ons | +0.071 |
Mic | +0.068 |
Neg logit contributions | |
tub | -0.084 |
gan | -0.079 |
Moe | -0.071 |
ider | -0.071 |
ahl | -0.069 |
Token | whenever |
Token ID | 8797 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
gan | +0.078 |
tub | +0.075 |
Moe | +0.072 |
ses | +0.072 |
ider | +0.072 |
Neg logit contributions | |
envelop | -0.067 |
ons | -0.054 |
armor | -0.050 |
parcels | -0.050 |
airplanes | -0.049 |
Token | it |
Token ID | 340 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
tub | +0.121 |
gan | +0.116 |
ider | +0.109 |
ses | +0.106 |
inter | +0.105 |
Neg logit contributions | |
envelop | -0.074 |
Amen | -0.067 |
Offic | -0.066 |
ons | -0.064 |
read | -0.056 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
gan | +0.231 |
tub | +0.227 |
Moe | +0.209 |
ider | +0.201 |
Mil | +0.191 |
Neg logit contributions | |
envelop | -0.231 |
ons | -0.219 |
Offic | -0.209 |
Amen | -0.200 |
oc | -0.198 |
Token | . |
Token ID | 13 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
tub | +0.025 |
gan | +0.024 |
ider | +0.021 |
Moe | +0.021 |
ahl | +0.020 |
Neg logit contributions | |
envelop | -0.027 |
Offic | -0.024 |
Amen | -0.023 |
ons | -0.023 |
oc | -0.021 |
Token | The |
Token ID | 383 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
tub | +0.029 |
gan | +0.028 |
ider | +0.026 |
Moe | +0.024 |
ahl | +0.024 |
Neg logit contributions | |
envelop | -0.034 |
Amen | -0.032 |
Offic | -0.032 |
ons | -0.030 |
Mic | -0.029 |
Token | people |
Token ID | 661 |
Feature activation | -8.24e-03 |
Pos logit contributions | |
tub | +0.038 |
gan | +0.037 |
Moe | +0.033 |
ider | +0.033 |
ahl | +0.032 |
Neg logit contributions | |
envelop | -0.039 |
Offic | -0.035 |
Amen | -0.033 |
ons | -0.033 |
Mic | -0.032 |
Token | in |
Token ID | 287 |
Feature activation | +9.63e-03 |
Pos logit contributions | |
envelop | +0.018 |
Offic | +0.017 |
Amen | +0.017 |
ons | +0.016 |
Mic | +0.015 |
Neg logit contributions | |
tub | -0.013 |
gan | -0.013 |
Moe | -0.011 |
ider | -0.011 |
ahl | -0.011 |
Token | tree |
Token ID | 5509 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
envelop | +0.114 |
Offic | +0.106 |
Amen | +0.104 |
ons | +0.100 |
Mic | +0.096 |
Neg logit contributions | |
tub | -0.095 |
gan | -0.089 |
Moe | -0.079 |
ider | -0.078 |
ahl | -0.076 |
Token | and |
Token ID | 290 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
tub | +0.030 |
gan | +0.029 |
Moe | +0.026 |
ider | +0.025 |
ahl | +0.025 |
Neg logit contributions | |
envelop | -0.032 |
Offic | -0.030 |
Amen | -0.029 |
ons | -0.028 |
oc | -0.026 |
Token | decided |
Token ID | 3066 |
Feature activation | +2.29e-03 |
Pos logit contributions | |
envelop | +0.104 |
Offic | +0.096 |
Amen | +0.094 |
ons | +0.091 |
Mic | +0.085 |
Neg logit contributions | |
tub | -0.096 |
gan | -0.091 |
ider | -0.082 |
Moe | -0.081 |
ahl | -0.079 |
Token | to |
Token ID | 284 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
tub | +0.005 |
gan | +0.005 |
ses | +0.005 |
ider | +0.005 |
ahl | +0.004 |
Neg logit contributions | |
envelop | -0.004 |
Offic | -0.003 |
Amen | -0.003 |
ons | -0.003 |
read | -0.003 |
Token | climb |
Token ID | 12080 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
envelop | +0.046 |
Offic | +0.044 |
Amen | +0.042 |
Mic | +0.039 |
ons | +0.039 |
Neg logit contributions | |
tub | -0.044 |
gan | -0.042 |
Moe | -0.038 |
ahl | -0.036 |
Mil | -0.035 |
Token | it |
Token ID | 340 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
tub | +0.044 |
gan | +0.043 |
ider | +0.041 |
ses | +0.038 |
sed | +0.038 |
Neg logit contributions | |
envelop | -0.024 |
ons | -0.021 |
Offic | -0.020 |
Amen | -0.020 |
oc | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +7.97e-03 |
Pos logit contributions | |
gan | +0.247 |
tub | +0.243 |
ses | +0.233 |
sed | +0.228 |
ider | +0.219 |
Neg logit contributions | |
envelop | -0.208 |
ons | -0.187 |
Offic | -0.177 |
oc | -0.169 |
read | -0.169 |
Token | Tim |
Token ID | 5045 |
Feature activation | -7.79e-03 |
Pos logit contributions | |
tub | +0.013 |
ider | +0.013 |
gan | +0.012 |
ses | +0.012 |
ahl | +0.012 |
Neg logit contributions | |
envelop | -0.016 |
Offic | -0.015 |
Amen | -0.014 |
parcels | -0.013 |
ons | -0.013 |
Token | went |
Token ID | 1816 |
Feature activation | +8.36e-03 |
Pos logit contributions | |
envelop | +0.014 |
Offic | +0.013 |
Amen | +0.013 |
ons | +0.012 |
Mic | +0.011 |
Neg logit contributions | |
tub | -0.016 |
gan | -0.015 |
Moe | -0.014 |
ider | -0.014 |
ahl | -0.013 |
Token | first |
Token ID | 717 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
tub | +0.014 |
gan | +0.014 |
ses | +0.012 |
ider | +0.012 |
Moe | +0.012 |
Neg logit contributions | |
envelop | -0.017 |
Offic | -0.016 |
Amen | -0.016 |
ons | -0.015 |
saddle | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | -5.35e-02 |
Pos logit contributions | |
tub | +0.047 |
gan | +0.044 |
ider | +0.041 |
ses | +0.041 |
ahl | +0.039 |
Neg logit contributions | |
envelop | -0.041 |
ons | -0.037 |
Offic | -0.037 |
read | -0.034 |
oc | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
envelop | +0.020 |
Amen | +0.019 |
Offic | +0.019 |
ons | +0.017 |
Mic | +0.017 |
Neg logit contributions | |
tub | -0.018 |
gan | -0.017 |
Moe | -0.016 |
Mil | -0.015 |
ahl | -0.014 |
Token | afternoon |
Token ID | 6672 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
tub | +0.081 |
gan | +0.077 |
Moe | +0.068 |
ider | +0.068 |
ahl | +0.066 |
Neg logit contributions | |
envelop | -0.082 |
Offic | -0.076 |
Amen | -0.074 |
ons | -0.073 |
Mic | -0.068 |
Token | bouncing |
Token ID | 30713 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
tub | +0.089 |
gan | +0.085 |
ider | +0.076 |
Moe | +0.076 |
ahl | +0.075 |
Neg logit contributions | |
envelop | -0.075 |
Offic | -0.068 |
Amen | -0.066 |
ons | -0.065 |
Mic | -0.060 |
Token | until |
Token ID | 1566 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
gan | +0.035 |
tub | +0.034 |
ider | +0.033 |
ses | +0.030 |
ahl | +0.029 |
Neg logit contributions | |
envelop | -0.029 |
ons | -0.027 |
Offic | -0.024 |
oc | -0.023 |
Amen | -0.023 |
Token | soon |
Token ID | 2582 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
tub | +0.104 |
gan | +0.098 |
ider | +0.097 |
ses | +0.091 |
ahl | +0.091 |
Neg logit contributions | |
envelop | -0.077 |
ons | -0.068 |
Offic | -0.067 |
Amen | -0.066 |
lending | -0.058 |
Token | it |
Token ID | 340 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
ider | +0.120 |
tub | +0.120 |
ses | +0.114 |
gan | +0.113 |
sed | +0.106 |
Neg logit contributions | |
envelop | -0.085 |
ons | -0.076 |
Offic | -0.068 |
arrest | -0.064 |
itivity | -0.062 |
Token | was |
Token ID | 373 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
ses | +0.268 |
ider | +0.258 |
gan | +0.252 |
spine | +0.250 |
sing | +0.246 |
Neg logit contributions | |
ons | -0.161 |
envelop | -0.137 |
rode | -0.136 |
read | -0.130 |
oc | -0.129 |
Token | time |
Token ID | 640 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
tub | +0.097 |
gan | +0.093 |
ider | +0.089 |
ahl | +0.084 |
ses | +0.084 |
Neg logit contributions | |
envelop | -0.076 |
Amen | -0.065 |
Offic | -0.064 |
ons | -0.063 |
Mic | -0.057 |
Token | to |
Token ID | 284 |
Feature activation | -6.13e-03 |
Pos logit contributions | |
tub | +0.111 |
gan | +0.105 |
Moe | +0.097 |
ahl | +0.096 |
ider | +0.095 |
Neg logit contributions | |
envelop | -0.090 |
ons | -0.076 |
Amen | -0.076 |
Offic | -0.073 |
permitting | -0.067 |
Token | go |
Token ID | 467 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
envelop | +0.010 |
Offic | +0.009 |
Amen | +0.009 |
ons | +0.008 |
Mic | +0.008 |
Neg logit contributions | |
tub | -0.013 |
gan | -0.013 |
ider | -0.012 |
Moe | -0.012 |
ahl | -0.011 |
Token | home |
Token ID | 1363 |
Feature activation | +7.38e-02 |
Pos logit contributions | |
gan | +0.054 |
tub | +0.054 |
ider | +0.051 |
ses | +0.049 |
sed | +0.048 |
Neg logit contributions | |
envelop | -0.044 |
ons | -0.038 |
Amen | -0.038 |
Offic | -0.037 |
saddle | -0.034 |
Token | tree |
Token ID | 5509 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
envelop | +0.110 |
Offic | +0.102 |
Amen | +0.100 |
ons | +0.096 |
Mic | +0.092 |
Neg logit contributions | |
tub | -0.090 |
gan | -0.085 |
Moe | -0.075 |
ider | -0.075 |
ahl | -0.072 |
Token | with |
Token ID | 351 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
tub | +0.024 |
gan | +0.023 |
Moe | +0.021 |
ider | +0.020 |
ahl | +0.020 |
Neg logit contributions | |
envelop | -0.022 |
Offic | -0.020 |
Amen | -0.020 |
ons | -0.019 |
oc | -0.018 |
Token | many |
Token ID | 867 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
envelop | +0.045 |
Offic | +0.043 |
Amen | +0.043 |
ons | +0.040 |
Mic | +0.038 |
Neg logit contributions | |
tub | -0.042 |
gan | -0.040 |
Moe | -0.036 |
ider | -0.034 |
ahl | -0.034 |
Token | mushrooms |
Token ID | 23452 |
Feature activation | -8.50e-03 |
Pos logit contributions | |
envelop | +0.086 |
Amen | +0.085 |
Offic | +0.082 |
Mic | +0.075 |
ons | +0.073 |
Neg logit contributions | |
tub | -0.064 |
gan | -0.063 |
Moe | -0.055 |
ider | -0.054 |
ahl | -0.053 |
Token | under |
Token ID | 739 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
envelop | +0.017 |
Offic | +0.016 |
ons | +0.016 |
Amen | +0.015 |
Mic | +0.014 |
Neg logit contributions | |
tub | -0.015 |
gan | -0.015 |
Moe | -0.013 |
ahl | -0.013 |
ider | -0.012 |
Token | it |
Token ID | 340 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
tub | +0.036 |
gan | +0.034 |
Moe | +0.031 |
Mil | +0.030 |
ider | +0.030 |
Neg logit contributions | |
envelop | -0.034 |
Offic | -0.032 |
Amen | -0.032 |
ons | -0.030 |
Mic | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
gan | +0.236 |
tub | +0.236 |
Moe | +0.211 |
ider | +0.206 |
ahl | +0.204 |
Neg logit contributions | |
envelop | -0.229 |
ons | -0.211 |
Offic | -0.197 |
Amen | -0.183 |
oc | -0.179 |
Token | Tim |
Token ID | 5045 |
Feature activation | -5.27e-03 |
Pos logit contributions | |
tub | +0.021 |
ider | +0.020 |
gan | +0.019 |
ahl | +0.018 |
ses | +0.018 |
Neg logit contributions | |
envelop | -0.019 |
Offic | -0.019 |
Amen | -0.018 |
ons | -0.017 |
oc | -0.017 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
envelop | +0.010 |
Offic | +0.009 |
Amen | +0.009 |
ons | +0.009 |
Mic | +0.008 |
Neg logit contributions | |
tub | -0.010 |
gan | -0.010 |
ider | -0.009 |
Moe | -0.009 |
ahl | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
envelop | +0.031 |
Offic | +0.029 |
Amen | +0.029 |
ons | +0.027 |
Mic | +0.026 |
Neg logit contributions | |
tub | -0.025 |
gan | -0.023 |
Moe | -0.021 |
ider | -0.020 |
ahl | -0.020 |
Token | look |
Token ID | 804 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
envelop | +0.073 |
Amen | +0.070 |
Offic | +0.069 |
Mic | +0.064 |
ons | +0.063 |
Neg logit contributions | |
gan | -0.058 |
tub | -0.055 |
Moe | -0.052 |
Mil | -0.050 |
ahl | -0.048 |
Token | tree |
Token ID | 5509 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
envelop | +0.102 |
Offic | +0.097 |
Amen | +0.095 |
ons | +0.091 |
Mic | +0.089 |
Neg logit contributions | |
tub | -0.080 |
gan | -0.072 |
Moe | -0.066 |
ses | -0.064 |
ider | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
tub | +0.037 |
gan | +0.035 |
Moe | +0.032 |
ider | +0.031 |
ahl | +0.031 |
Neg logit contributions | |
envelop | -0.040 |
Offic | -0.037 |
Amen | -0.037 |
ons | -0.035 |
Mic | -0.033 |
Token | decided |
Token ID | 3066 |
Feature activation | +6.37e-03 |
Pos logit contributions | |
envelop | +0.095 |
Offic | +0.089 |
Amen | +0.087 |
ons | +0.083 |
Mic | +0.078 |
Neg logit contributions | |
tub | -0.087 |
gan | -0.082 |
ider | -0.073 |
Moe | -0.073 |
ahl | -0.071 |
Token | to |
Token ID | 284 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
tub | +0.014 |
gan | +0.012 |
ider | +0.012 |
ses | +0.012 |
Moe | +0.011 |
Neg logit contributions | |
envelop | -0.011 |
Offic | -0.010 |
Amen | -0.010 |
ons | -0.009 |
read | -0.009 |
Token | climb |
Token ID | 12080 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
envelop | +0.038 |
Offic | +0.035 |
Amen | +0.034 |
ons | +0.032 |
Mic | +0.032 |
Neg logit contributions | |
tub | -0.034 |
gan | -0.033 |
Moe | -0.030 |
Mil | -0.028 |
ahl | -0.028 |
Token | it |
Token ID | 340 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
tub | +0.043 |
gan | +0.041 |
ider | +0.039 |
ses | +0.037 |
sed | +0.036 |
Neg logit contributions | |
envelop | -0.030 |
ons | -0.026 |
Offic | -0.026 |
Amen | -0.026 |
oc | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
tub | +0.245 |
gan | +0.245 |
ses | +0.234 |
sed | +0.228 |
Moe | +0.219 |
Neg logit contributions | |
envelop | -0.201 |
ons | -0.183 |
Offic | -0.174 |
Amen | -0.169 |
oc | -0.168 |
Token |
|
Token ID | 198 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
tub | +0.019 |
gan | +0.017 |
ider | +0.017 |
ses | +0.016 |
ahl | +0.015 |
Neg logit contributions | |
envelop | -0.025 |
Offic | -0.024 |
Amen | -0.023 |
ons | -0.022 |
Mic | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
envelop | +0.033 |
Amen | +0.030 |
Offic | +0.030 |
ons | +0.028 |
itivity | +0.028 |
Neg logit contributions | |
tub | -0.017 |
gan | -0.016 |
Moe | -0.014 |
Mil | -0.013 |
ahl | -0.013 |
Token | As |
Token ID | 1722 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
envelop | +0.021 |
Amen | +0.019 |
ons | +0.018 |
Offic | +0.017 |
parcels | +0.016 |
Neg logit contributions | |
tub | -0.021 |
gan | -0.020 |
Moe | -0.018 |
ider | -0.018 |
Mil | -0.017 |
Token | Tom |
Token ID | 4186 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
tub | +0.041 |
gan | +0.037 |
ider | +0.035 |
ses | +0.034 |
ahl | +0.033 |
Neg logit contributions | |
envelop | -0.048 |
Offic | -0.045 |
ons | -0.043 |
Amen | -0.043 |
Mic | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -5.79e-02 |
Pos logit contributions | |
envelop | +0.021 |
Offic | +0.019 |
Amen | +0.018 |
ons | +0.018 |
oc | +0.017 |
Neg logit contributions | |
tub | -0.025 |
gan | -0.024 |
ider | -0.021 |
Moe | -0.021 |
ahl | -0.021 |
Token | seek |
Token ID | 5380 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
envelop | +0.108 |
Offic | +0.099 |
Amen | +0.096 |
ons | +0.093 |
Mic | +0.088 |
Neg logit contributions | |
tub | -0.114 |
gan | -0.109 |
ider | -0.098 |
Moe | -0.096 |
ahl | -0.094 |
Token | all |
Token ID | 477 |
Feature activation | -9.18e-03 |
Pos logit contributions | |
envelop | +0.046 |
ons | +0.040 |
Offic | +0.040 |
Amen | +0.039 |
oc | +0.037 |
Neg logit contributions | |
tub | -0.044 |
gan | -0.043 |
ses | -0.038 |
ahl | -0.038 |
ider | -0.037 |
Token | afternoon |
Token ID | 6672 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
envelop | +0.016 |
ons | +0.014 |
Amen | +0.014 |
Offic | +0.014 |
oc | +0.013 |
Neg logit contributions | |
tub | -0.019 |
gan | -0.018 |
ider | -0.018 |
ses | -0.017 |
sed | -0.016 |
Token | until |
Token ID | 1566 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
tub | +0.087 |
gan | +0.084 |
ahl | +0.073 |
Moe | +0.072 |
ider | +0.071 |
Neg logit contributions | |
envelop | -0.111 |
Offic | -0.100 |
Amen | -0.098 |
ons | -0.098 |
Mic | -0.091 |
Token | it |
Token ID | 340 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
tub | +0.115 |
gan | +0.109 |
ider | +0.102 |
ses | +0.096 |
ahl | +0.096 |
Neg logit contributions | |
envelop | -0.103 |
Offic | -0.093 |
Amen | -0.090 |
ons | -0.090 |
Mic | -0.079 |
Token | was |
Token ID | 373 |
Feature activation | +4.46e-02 |
Pos logit contributions | |
ses | +0.241 |
ider | +0.229 |
gan | +0.227 |
tub | +0.223 |
fault | +0.218 |
Neg logit contributions | |
envelop | -0.181 |
Offic | -0.178 |
ons | -0.175 |
read | -0.169 |
oc | -0.167 |
Token | time |
Token ID | 640 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
tub | +0.086 |
gan | +0.082 |
Moe | +0.073 |
ider | +0.073 |
ahl | +0.071 |
Neg logit contributions | |
envelop | -0.086 |
Offic | -0.079 |
Amen | -0.077 |
ons | -0.074 |
Mic | -0.070 |
Token | for |
Token ID | 329 |
Feature activation | -6.75e-03 |
Pos logit contributions | |
tub | +0.106 |
gan | +0.100 |
Moe | +0.094 |
ahl | +0.092 |
ider | +0.090 |
Neg logit contributions | |
envelop | -0.083 |
Amen | -0.071 |
Offic | -0.068 |
ons | -0.068 |
lending | -0.063 |
Token | Sam |
Token ID | 3409 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
envelop | +0.015 |
Offic | +0.014 |
Amen | +0.014 |
oc | +0.013 |
ons | +0.013 |
Neg logit contributions | |
tub | -0.011 |
gan | -0.010 |
Moe | -0.010 |
Mil | -0.009 |
ahl | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
tub | +0.021 |
gan | +0.020 |
ider | +0.018 |
Moe | +0.018 |
ahl | +0.017 |
Neg logit contributions | |
envelop | -0.023 |
Offic | -0.021 |
Amen | -0.020 |
ons | -0.020 |
Mic | -0.019 |
Token | Finally |
Token ID | 11158 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
envelop | +0.016 |
Offic | +0.014 |
Amen | +0.014 |
ons | +0.013 |
oc | +0.012 |
Neg logit contributions | |
tub | -0.016 |
gan | -0.016 |
ider | -0.014 |
Moe | -0.014 |
ahl | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
tub | +0.097 |
gan | +0.086 |
ses | +0.082 |
inter | +0.079 |
ider | +0.079 |
Neg logit contributions | |
envelop | -0.102 |
ons | -0.091 |
Offic | -0.088 |
Amen | -0.085 |
read | -0.085 |
Token | day |
Token ID | 1110 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
tub | +0.093 |
gan | +0.090 |
Moe | +0.079 |
ider | +0.079 |
ahl | +0.078 |
Neg logit contributions | |
envelop | -0.086 |
Offic | -0.078 |
Amen | -0.077 |
ons | -0.074 |
Mic | -0.069 |
Token | came |
Token ID | 1625 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
tub | +0.118 |
gan | +0.111 |
ses | +0.108 |
ider | +0.105 |
sed | +0.103 |
Neg logit contributions | |
envelop | -0.103 |
ons | -0.093 |
Offic | -0.091 |
read | -0.082 |
oc | -0.081 |
Token | when |
Token ID | 618 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
tub | +0.114 |
ses | +0.105 |
ider | +0.101 |
gan | +0.100 |
auc | +0.100 |
Neg logit contributions | |
envelop | -0.080 |
ons | -0.063 |
Offic | -0.063 |
read | -0.062 |
lending | -0.062 |
Token | it |
Token ID | 340 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
tub | +0.096 |
gan | +0.087 |
ider | +0.083 |
inter | +0.081 |
ses | +0.080 |
Neg logit contributions | |
envelop | -0.103 |
Offic | -0.092 |
Amen | -0.092 |
ons | -0.091 |
read | -0.086 |
Token | was |
Token ID | 373 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
ses | +0.254 |
gan | +0.247 |
tub | +0.238 |
fault | +0.233 |
ider | +0.227 |
Neg logit contributions | |
envelop | -0.184 |
Offic | -0.180 |
ons | -0.179 |
read | -0.174 |
oc | -0.165 |
Token | time |
Token ID | 640 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
tub | +0.072 |
gan | +0.069 |
Moe | +0.060 |
ider | +0.059 |
ahl | +0.057 |
Neg logit contributions | |
envelop | -0.086 |
Offic | -0.082 |
Amen | -0.079 |
ons | -0.076 |
oc | -0.076 |
Token | for |
Token ID | 329 |
Feature activation | -5.85e-03 |
Pos logit contributions | |
tub | +0.121 |
gan | +0.115 |
Moe | +0.107 |
ider | +0.104 |
ahl | +0.103 |
Neg logit contributions | |
envelop | -0.090 |
ons | -0.074 |
Offic | -0.071 |
Amen | -0.071 |
read | -0.065 |
Token | Anna |
Token ID | 11735 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
envelop | +0.012 |
Offic | +0.011 |
Amen | +0.011 |
ons | +0.011 |
Mic | +0.010 |
Neg logit contributions | |
tub | -0.010 |
gan | -0.010 |
Moe | -0.009 |
ider | -0.009 |
ahl | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
tub | +0.045 |
gan | +0.043 |
Moe | +0.038 |
ider | +0.038 |
ahl | +0.037 |
Neg logit contributions | |
envelop | -0.052 |
Offic | -0.047 |
Amen | -0.047 |
ons | -0.046 |
Mic | -0.043 |
Token | that |
Token ID | 326 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
envelop | +0.067 |
Offic | +0.064 |
Amen | +0.062 |
ons | +0.060 |
read | +0.056 |
Neg logit contributions | |
tub | -0.079 |
gan | -0.074 |
ider | -0.068 |
Moe | -0.065 |
ses | -0.064 |
Token | day |
Token ID | 1110 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
tub | +0.029 |
gan | +0.023 |
Moe | +0.021 |
ahl | +0.020 |
rains | +0.019 |
Neg logit contributions | |
envelop | -0.065 |
Offic | -0.061 |
Amen | -0.058 |
ons | -0.057 |
parcels | -0.057 |
Token | on |
Token ID | 319 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
tub | +0.076 |
gan | +0.074 |
ider | +0.067 |
Moe | +0.063 |
ahl | +0.063 |
Neg logit contributions | |
envelop | -0.096 |
Offic | -0.089 |
Amen | -0.086 |
parcels | -0.080 |
Mic | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
tub | +0.028 |
gan | +0.026 |
ider | +0.024 |
ses | +0.024 |
ipples | +0.023 |
Neg logit contributions | |
envelop | -0.026 |
ons | -0.025 |
Amen | -0.022 |
Offic | -0.022 |
read | -0.021 |
Token | whenever |
Token ID | 8797 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
tub | +0.041 |
gan | +0.038 |
ider | +0.032 |
Moe | +0.032 |
ahl | +0.030 |
Neg logit contributions | |
envelop | -0.089 |
Offic | -0.084 |
Amen | -0.083 |
ons | -0.081 |
Mic | -0.077 |
Token | it |
Token ID | 340 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
tub | +0.097 |
gan | +0.093 |
ider | +0.083 |
ses | +0.080 |
ahl | +0.077 |
Neg logit contributions | |
envelop | -0.112 |
Offic | -0.103 |
Amen | -0.102 |
ons | -0.099 |
read | -0.095 |
Token | r |
Token ID | 374 |
Feature activation | +5.83e-02 |
Pos logit contributions | |
gan | +0.212 |
tub | +0.192 |
ses | +0.191 |
Moe | +0.189 |
ider | +0.187 |
Neg logit contributions | |
ons | -0.223 |
read | -0.217 |
Offic | -0.211 |
envelop | -0.210 |
Amen | -0.207 |
Token | ained |
Token ID | 1328 |
Feature activation | +6.29e-02 |
Pos logit contributions | |
tub | +0.086 |
Moe | +0.081 |
gan | +0.080 |
Mil | +0.076 |
ahl | +0.075 |
Neg logit contributions | |
envelop | -0.122 |
ons | -0.120 |
Offic | -0.116 |
Amen | -0.110 |
Mic | -0.105 |
Token | , |
Token ID | 11 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
gan | +0.143 |
tub | +0.127 |
ses | +0.126 |
ider | +0.125 |
ahl | +0.123 |
Neg logit contributions | |
ons | -0.086 |
envelop | -0.075 |
course | -0.073 |
lending | -0.071 |
read | -0.071 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
tub | +0.074 |
gan | +0.073 |
ider | +0.072 |
ses | +0.066 |
sed | +0.064 |
Neg logit contributions | |
envelop | -0.078 |
ons | -0.070 |
Offic | -0.067 |
read | -0.064 |
Amen | -0.063 |
Token | would |
Token ID | 561 |
Feature activation | -8.84e-03 |
Pos logit contributions | |
tub | +0.059 |
gan | +0.058 |
ider | +0.054 |
Moe | +0.054 |
ahl | +0.051 |
Neg logit contributions | |
envelop | -0.053 |
Offic | -0.049 |
ons | -0.047 |
Amen | -0.047 |
read | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -9.16e-03 |
Pos logit contributions | |
envelop | +0.030 |
Offic | +0.028 |
Amen | +0.027 |
ons | +0.026 |
Mic | +0.026 |
Neg logit contributions | |
tub | -0.016 |
gan | -0.014 |
Moe | -0.013 |
ahl | -0.012 |
Mil | -0.012 |
Token | One |
Token ID | 3198 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
envelop | +0.017 |
Amen | +0.015 |
Offic | +0.014 |
ons | +0.014 |
parcels | +0.013 |
Neg logit contributions | |
tub | -0.019 |
gan | -0.016 |
Moe | -0.015 |
ider | -0.015 |
ahl | -0.014 |
Token | day |
Token ID | 1110 |
Feature activation | +6.67e-02 |
Pos logit contributions | |
tub | +0.021 |
gan | +0.021 |
ider | +0.020 |
ses | +0.019 |
sed | +0.019 |
Neg logit contributions | |
Amen | -0.017 |
envelop | -0.017 |
Offic | -0.016 |
oc | -0.015 |
ons | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +4.91e-02 |
Pos logit contributions | |
tub | +0.099 |
gan | +0.095 |
ider | +0.081 |
Moe | +0.080 |
ses | +0.079 |
Neg logit contributions | |
envelop | -0.158 |
Offic | -0.147 |
Amen | -0.142 |
ons | -0.142 |
oc | -0.133 |
Token | when |
Token ID | 618 |
Feature activation | +6.08e-02 |
Pos logit contributions | |
tub | +0.081 |
gan | +0.075 |
Moe | +0.067 |
ider | +0.065 |
Mil | +0.062 |
Neg logit contributions | |
envelop | -0.108 |
Offic | -0.101 |
Amen | -0.101 |
ons | -0.095 |
Mic | -0.092 |
Token | it |
Token ID | 340 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
tub | +0.105 |
gan | +0.099 |
Moe | +0.087 |
ider | +0.086 |
ahl | +0.083 |
Neg logit contributions | |
envelop | -0.130 |
Offic | -0.121 |
Amen | -0.118 |
ons | -0.114 |
Mic | -0.109 |
Token | was |
Token ID | 373 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
gan | +0.233 |
ses | +0.225 |
fault | +0.213 |
tub | +0.206 |
ider | +0.205 |
Neg logit contributions | |
Amen | -0.205 |
oc | -0.200 |
envelop | -0.194 |
ons | -0.194 |
Offic | -0.191 |
Token | very |
Token ID | 845 |
Feature activation | +4.17e-03 |
Pos logit contributions | |
tub | +0.083 |
gan | +0.079 |
Moe | +0.070 |
ider | +0.070 |
ahl | +0.067 |
Neg logit contributions | |
envelop | -0.092 |
Offic | -0.085 |
Amen | -0.084 |
ons | -0.080 |
Mic | -0.076 |
Token | warm |
Token ID | 5814 |
Feature activation | +1.35e-03 |
Pos logit contributions | |
tub | +0.008 |
gan | +0.008 |
ses | +0.007 |
Moe | +0.007 |
ahl | +0.007 |
Neg logit contributions | |
envelop | -0.007 |
Amen | -0.007 |
Offic | -0.007 |
ons | -0.007 |
Mic | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
gan | +0.003 |
ahl | +0.003 |
tub | +0.003 |
ses | +0.003 |
ider | +0.003 |
Neg logit contributions | |
envelop | -0.002 |
ons | -0.002 |
Amen | -0.002 |
Offic | -0.002 |
pa | -0.002 |
Token | sunny |
Token ID | 27737 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
envelop | +0.083 |
Offic | +0.077 |
Amen | +0.076 |
ons | +0.073 |
Mic | +0.070 |
Neg logit contributions | |
tub | -0.064 |
gan | -0.059 |
Moe | -0.052 |
ider | -0.051 |
ahl | -0.049 |
Token | that |
Token ID | 326 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
envelop | +0.067 |
Offic | +0.063 |
ons | +0.061 |
Amen | +0.058 |
read | +0.055 |
Neg logit contributions | |
tub | -0.092 |
gan | -0.081 |
ider | -0.078 |
inter | -0.075 |
ses | -0.074 |
Token | day |
Token ID | 1110 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
tub | +0.035 |
gan | +0.030 |
Moe | +0.027 |
ider | +0.025 |
ahl | +0.025 |
Neg logit contributions | |
envelop | -0.064 |
Offic | -0.060 |
Amen | -0.057 |
ons | -0.056 |
parcels | -0.054 |
Token | on |
Token ID | 319 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
tub | +0.073 |
gan | +0.071 |
ider | +0.065 |
Moe | +0.062 |
ahl | +0.061 |
Neg logit contributions | |
envelop | -0.095 |
Offic | -0.089 |
Amen | -0.085 |
parcels | -0.080 |
Mic | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
tub | +0.028 |
gan | +0.027 |
ider | +0.024 |
ses | +0.024 |
ipples | +0.023 |
Neg logit contributions | |
envelop | -0.027 |
ons | -0.026 |
Amen | -0.023 |
Offic | -0.023 |
lending | -0.022 |
Token | whenever |
Token ID | 8797 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
tub | +0.046 |
gan | +0.043 |
ider | +0.038 |
ses | +0.037 |
Moe | +0.035 |
Neg logit contributions | |
envelop | -0.079 |
Offic | -0.072 |
ons | -0.071 |
Amen | -0.070 |
Mic | -0.067 |
Token | it |
Token ID | 340 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
tub | +0.099 |
gan | +0.093 |
ider | +0.087 |
ses | +0.085 |
inter | +0.083 |
Neg logit contributions | |
envelop | -0.114 |
ons | -0.101 |
Offic | -0.100 |
Amen | -0.099 |
read | -0.098 |
Token | r |
Token ID | 374 |
Feature activation | +5.96e-02 |
Pos logit contributions | |
gan | +0.234 |
tub | +0.221 |
ses | +0.221 |
Moe | +0.215 |
ider | +0.208 |
Neg logit contributions | |
ons | -0.206 |
read | -0.195 |
rode | -0.184 |
envelop | -0.183 |
Offic | -0.182 |
Token | ained |
Token ID | 1328 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
tub | +0.093 |
Moe | +0.087 |
gan | +0.084 |
ahl | +0.082 |
Mil | +0.081 |
Neg logit contributions | |
ons | -0.118 |
envelop | -0.118 |
Offic | -0.110 |
Amen | -0.102 |
Mic | -0.100 |
Token | , |
Token ID | 11 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
gan | +0.138 |
ses | +0.125 |
tub | +0.125 |
ahl | +0.123 |
ider | +0.121 |
Neg logit contributions | |
ons | -0.093 |
envelop | -0.084 |
course | -0.078 |
lending | -0.074 |
read | -0.073 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
tub | +0.066 |
gan | +0.065 |
ider | +0.063 |
ses | +0.060 |
sed | +0.058 |
Neg logit contributions | |
envelop | -0.076 |
ons | -0.068 |
Offic | -0.065 |
read | -0.061 |
Amen | -0.060 |
Token | would |
Token ID | 561 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
tub | +0.052 |
gan | +0.051 |
ider | +0.047 |
Moe | +0.047 |
ahl | +0.045 |
Neg logit contributions | |
envelop | -0.049 |
Offic | -0.045 |
ons | -0.043 |
Amen | -0.043 |
oc | -0.040 |
Token | rolling |
Token ID | 18886 |
Feature activation | +6.89e-03 |
Pos logit contributions | |
envelop | +0.005 |
Offic | +0.005 |
ons | +0.004 |
Amen | +0.004 |
Mic | +0.004 |
Neg logit contributions | |
tub | -0.005 |
gan | -0.005 |
Moe | -0.004 |
ses | -0.004 |
Mil | -0.004 |
Token | through |
Token ID | 832 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
gan | +0.013 |
tub | +0.013 |
ses | +0.013 |
sed | +0.012 |
ider | +0.012 |
Neg logit contributions | |
envelop | -0.012 |
Offic | -0.011 |
ons | -0.010 |
oc | -0.010 |
Amen | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
tub | +0.057 |
gan | +0.056 |
ider | +0.051 |
ahl | +0.049 |
Moe | +0.049 |
Neg logit contributions | |
envelop | -0.049 |
Offic | -0.044 |
ons | -0.041 |
Amen | -0.041 |
oc | -0.038 |
Token | corridors |
Token ID | 32807 |
Feature activation | +6.84e-02 |
Pos logit contributions | |
tub | +0.090 |
gan | +0.088 |
ider | +0.078 |
Moe | +0.077 |
ses | +0.075 |
Neg logit contributions | |
envelop | -0.100 |
Offic | -0.092 |
Amen | -0.089 |
ons | -0.087 |
Mic | -0.081 |
Token | when |
Token ID | 618 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
tub | +0.128 |
gan | +0.127 |
ahl | +0.115 |
Moe | +0.115 |
ses | +0.114 |
Neg logit contributions | |
envelop | -0.127 |
ons | -0.119 |
Offic | -0.115 |
oc | -0.104 |
Amen | -0.103 |
Token | it |
Token ID | 340 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
tub | +0.110 |
gan | +0.107 |
ahl | +0.101 |
ider | +0.099 |
inter | +0.097 |
Neg logit contributions | |
envelop | -0.077 |
ons | -0.071 |
Amen | -0.071 |
Offic | -0.069 |
read | -0.065 |
Token | ran |
Token ID | 4966 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
gan | +0.290 |
fault | +0.287 |
ses | +0.286 |
tub | +0.276 |
Moe | +0.273 |
Neg logit contributions | |
envelop | -0.144 |
read | -0.141 |
Offic | -0.140 |
ons | -0.133 |
rode | -0.113 |
Token | into |
Token ID | 656 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
gan | +0.035 |
tub | +0.034 |
ses | +0.031 |
ider | +0.031 |
Moe | +0.030 |
Neg logit contributions | |
envelop | -0.033 |
Offic | -0.028 |
Amen | -0.026 |
ons | -0.025 |
Mic | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
tub | +0.024 |
gan | +0.022 |
Moe | +0.020 |
Mil | +0.019 |
ahl | +0.018 |
Neg logit contributions | |
envelop | -0.032 |
Amen | -0.031 |
Offic | -0.031 |
ons | -0.029 |
Mic | -0.029 |
Token | tiny |
Token ID | 7009 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
envelop | +0.045 |
Offic | +0.042 |
Amen | +0.041 |
ons | +0.040 |
Mic | +0.038 |
Neg logit contributions | |
tub | -0.035 |
gan | -0.032 |
Moe | -0.029 |
ider | -0.029 |
ahl | -0.028 |
Token | mouse |
Token ID | 10211 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
envelop | +0.046 |
Offic | +0.043 |
Amen | +0.042 |
ons | +0.041 |
Mic | +0.039 |
Neg logit contributions | |
tub | -0.037 |
gan | -0.035 |
ider | -0.031 |
Moe | -0.031 |
ahl | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
tub | +0.091 |
gan | +0.083 |
Moe | +0.073 |
ider | +0.072 |
Mil | +0.067 |
Neg logit contributions | |
envelop | -0.153 |
Offic | -0.148 |
Amen | -0.138 |
ons | -0.135 |
Mic | -0.133 |
Token | One |
Token ID | 1881 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
tub | +0.036 |
gan | +0.036 |
ider | +0.034 |
Moe | +0.031 |
ses | +0.030 |
Neg logit contributions | |
envelop | -0.031 |
Amen | -0.030 |
Offic | -0.029 |
ons | -0.028 |
Mic | -0.026 |
Token | day |
Token ID | 1110 |
Feature activation | +7.23e-02 |
Pos logit contributions | |
tub | +0.036 |
gan | +0.035 |
ider | +0.034 |
Moe | +0.032 |
ses | +0.031 |
Neg logit contributions | |
envelop | -0.031 |
Offic | -0.029 |
Amen | -0.029 |
ons | -0.028 |
oc | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
tub | +0.139 |
gan | +0.131 |
ider | +0.116 |
Moe | +0.112 |
fault | +0.111 |
Neg logit contributions | |
envelop | -0.134 |
ons | -0.133 |
Amen | -0.120 |
Offic | -0.119 |
oc | -0.114 |
Token | when |
Token ID | 618 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
tub | +0.086 |
gan | +0.082 |
ider | +0.073 |
Moe | +0.070 |
ses | +0.068 |
Neg logit contributions | |
envelop | -0.109 |
Offic | -0.102 |
Amen | -0.102 |
ons | -0.101 |
Mic | -0.095 |
Token | it |
Token ID | 340 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
tub | +0.110 |
gan | +0.107 |
ider | +0.095 |
Moe | +0.090 |
ses | +0.088 |
Neg logit contributions | |
envelop | -0.127 |
Amen | -0.122 |
Offic | -0.120 |
ons | -0.118 |
Mic | -0.108 |
Token | was |
Token ID | 373 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
ses | +0.243 |
gan | +0.232 |
fault | +0.230 |
sed | +0.223 |
tub | +0.220 |
Neg logit contributions | |
ons | -0.185 |
read | -0.180 |
Amen | -0.176 |
oc | -0.170 |
Offic | -0.168 |
Token | very |
Token ID | 845 |
Feature activation | +5.30e-03 |
Pos logit contributions | |
tub | +0.076 |
gan | +0.073 |
Moe | +0.065 |
ider | +0.065 |
ahl | +0.063 |
Neg logit contributions | |
envelop | -0.085 |
Offic | -0.078 |
Amen | -0.077 |
ons | -0.075 |
Mic | -0.071 |
Token | cold |
Token ID | 4692 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
tub | +0.010 |
gan | +0.010 |
ses | +0.009 |
Moe | +0.009 |
ahl | +0.009 |
Neg logit contributions | |
envelop | -0.008 |
Amen | -0.008 |
ons | -0.008 |
Offic | -0.007 |
read | -0.007 |
Token | outside |
Token ID | 2354 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
gan | +0.094 |
tub | +0.089 |
ider | +0.086 |
ahl | +0.084 |
ses | +0.084 |
Neg logit contributions | |
envelop | -0.071 |
ons | -0.070 |
Amen | -0.067 |
Offic | -0.062 |
Mic | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | +4.87e-02 |
Pos logit contributions | |
gan | +0.170 |
ider | +0.162 |
tub | +0.160 |
ses | +0.158 |
ahl | +0.153 |
Neg logit contributions | |
envelop | -0.147 |
ons | -0.144 |
Amen | -0.129 |
read | -0.120 |
pa | -0.118 |
Token | that |
Token ID | 326 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
envelop | +0.083 |
Offic | +0.078 |
Amen | +0.077 |
ons | +0.074 |
Mic | +0.069 |
Neg logit contributions | |
tub | -0.090 |
gan | -0.081 |
ider | -0.076 |
Moe | -0.072 |
ses | -0.071 |
Token | day |
Token ID | 1110 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
tub | +0.029 |
gan | +0.023 |
Moe | +0.019 |
ahl | +0.019 |
ider | +0.018 |
Neg logit contributions | |
envelop | -0.065 |
Offic | -0.061 |
ons | -0.057 |
itivity | -0.057 |
Amen | -0.057 |
Token | on |
Token ID | 319 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
tub | +0.071 |
gan | +0.069 |
ider | +0.063 |
Moe | +0.059 |
ahl | +0.059 |
Neg logit contributions | |
envelop | -0.093 |
Offic | -0.086 |
Amen | -0.082 |
parcels | -0.077 |
Mic | -0.077 |
Token | , |
Token ID | 11 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
tub | +0.026 |
gan | +0.025 |
ider | +0.023 |
ses | +0.022 |
ahl | +0.021 |
Neg logit contributions | |
envelop | -0.028 |
ons | -0.027 |
Amen | -0.024 |
Offic | -0.024 |
Mic | -0.023 |
Token | whenever |
Token ID | 8797 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
tub | +0.037 |
gan | +0.034 |
ider | +0.029 |
Moe | +0.028 |
ahl | +0.027 |
Neg logit contributions | |
envelop | -0.076 |
Offic | -0.071 |
Amen | -0.070 |
ons | -0.068 |
Mic | -0.065 |
Token | it |
Token ID | 340 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
tub | +0.096 |
gan | +0.091 |
ider | +0.083 |
ses | +0.080 |
inter | +0.078 |
Neg logit contributions | |
envelop | -0.114 |
Offic | -0.102 |
ons | -0.101 |
Amen | -0.101 |
read | -0.097 |
Token | r |
Token ID | 374 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
gan | +0.222 |
tub | +0.216 |
Moe | +0.208 |
ses | +0.192 |
Mil | +0.192 |
Neg logit contributions | |
envelop | -0.222 |
ons | -0.218 |
Amen | -0.215 |
Offic | -0.215 |
oc | -0.199 |
Token | ained |
Token ID | 1328 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
tub | +0.083 |
gan | +0.077 |
Moe | +0.075 |
Mil | +0.071 |
ahl | +0.070 |
Neg logit contributions | |
envelop | -0.136 |
ons | -0.129 |
Offic | -0.128 |
Amen | -0.124 |
Mic | -0.118 |
Token | , |
Token ID | 11 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
gan | +0.136 |
tub | +0.123 |
ses | +0.120 |
ahl | +0.120 |
ider | +0.116 |
Neg logit contributions | |
ons | -0.095 |
envelop | -0.090 |
Amen | -0.085 |
lending | -0.082 |
course | -0.080 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
tub | +0.061 |
gan | +0.060 |
ider | +0.057 |
ses | +0.054 |
sed | +0.053 |
Neg logit contributions | |
envelop | -0.075 |
ons | -0.067 |
Offic | -0.065 |
read | -0.060 |
Amen | -0.060 |
Token | would |
Token ID | 561 |
Feature activation | -9.25e-03 |
Pos logit contributions | |
tub | +0.049 |
gan | +0.047 |
Moe | +0.043 |
ider | +0.042 |
ahl | +0.041 |
Neg logit contributions | |
envelop | -0.048 |
Offic | -0.045 |
Amen | -0.043 |
ons | -0.042 |
Mic | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
envelop | +0.104 |
Offic | +0.097 |
Amen | +0.095 |
ons | +0.092 |
Mic | +0.087 |
Neg logit contributions | |
tub | -0.075 |
gan | -0.071 |
Moe | -0.062 |
ider | -0.062 |
ahl | -0.060 |
Token | He |
Token ID | 679 |
Feature activation | -3.85e-03 |
Pos logit contributions | |
tub | +0.021 |
gan | +0.020 |
ider | +0.018 |
ahl | +0.018 |
ses | +0.018 |
Neg logit contributions | |
envelop | -0.030 |
Offic | -0.028 |
Amen | -0.028 |
Mic | -0.025 |
ons | -0.025 |
Token | also |
Token ID | 635 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
envelop | +0.008 |
Offic | +0.007 |
Amen | +0.007 |
ons | +0.007 |
oc | +0.007 |
Neg logit contributions | |
tub | -0.007 |
gan | -0.007 |
ider | -0.006 |
Moe | -0.006 |
ahl | -0.006 |
Token | has |
Token ID | 468 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
envelop | +0.027 |
Offic | +0.025 |
Amen | +0.024 |
ons | +0.023 |
Mic | +0.022 |
Neg logit contributions | |
tub | -0.026 |
gan | -0.024 |
Moe | -0.022 |
ider | -0.022 |
ahl | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
envelop | +0.034 |
Offic | +0.033 |
Amen | +0.033 |
ons | +0.031 |
oc | +0.030 |
Neg logit contributions | |
tub | -0.026 |
gan | -0.025 |
Moe | -0.022 |
ider | -0.021 |
Mil | -0.020 |
Token | sle |
Token ID | 3133 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
envelop | +0.097 |
Offic | +0.094 |
ons | +0.090 |
Amen | +0.090 |
Mic | +0.083 |
Neg logit contributions | |
tub | -0.089 |
gan | -0.078 |
Moe | -0.071 |
ider | -0.070 |
ses | -0.069 |
Token | igh |
Token ID | 394 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
Offic | +0.160 |
envelop | +0.157 |
Mic | +0.151 |
pic | +0.129 |
inviting | +0.127 |
Neg logit contributions | |
tub | -0.314 |
ider | -0.314 |
gan | -0.307 |
ses | -0.299 |
sed | -0.294 |
Token | and |
Token ID | 290 |
Feature activation | -7.63e-02 |
Pos logit contributions | |
envelop | +0.103 |
Offic | +0.097 |
Amen | +0.095 |
ons | +0.089 |
Mic | +0.087 |
Neg logit contributions | |
tub | -0.092 |
gan | -0.086 |
ider | -0.077 |
Moe | -0.076 |
ahl | -0.072 |
Token | re |
Token ID | 302 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
envelop | +0.163 |
Offic | +0.155 |
Amen | +0.151 |
ons | +0.148 |
Mic | +0.144 |
Neg logit contributions | |
tub | -0.119 |
gan | -0.108 |
Moe | -0.098 |
ses | -0.097 |
Mil | -0.095 |
Token | ind |
Token ID | 521 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
envelop | +0.147 |
Offic | +0.141 |
Mic | +0.139 |
lending | +0.125 |
permitting | +0.122 |
Neg logit contributions | |
gan | -0.212 |
tub | -0.210 |
flies | -0.193 |
ider | -0.189 |
uffy | -0.187 |
Token | eer |
Token ID | 28153 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
envelop | +0.110 |
Offic | +0.101 |
Amen | +0.099 |
ons | +0.095 |
Mic | +0.089 |
Neg logit contributions | |
tub | -0.119 |
gan | -0.114 |
Moe | -0.103 |
ider | -0.102 |
ahl | -0.099 |
Token | burn |
Token ID | 4245 |
Feature activation | -5.24e-03 |
Pos logit contributions | |
tub | +0.020 |
gan | +0.019 |
ider | +0.017 |
Moe | +0.017 |
ahl | +0.017 |
Neg logit contributions | |
envelop | -0.019 |
Amen | -0.017 |
Offic | -0.017 |
ons | -0.016 |
Mic | -0.015 |
Token | you |
Token ID | 345 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
envelop | +0.009 |
ons | +0.008 |
Amen | +0.008 |
Offic | +0.007 |
oc | +0.007 |
Neg logit contributions | |
tub | -0.011 |
gan | -0.011 |
ider | -0.010 |
Moe | -0.010 |
ahl | -0.010 |
Token | !" |
Token ID | 2474 |
Feature activation | +4.38e-03 |
Pos logit contributions | |
tub | +0.045 |
gan | +0.043 |
Moe | +0.040 |
ider | +0.039 |
ahl | +0.039 |
Neg logit contributions | |
envelop | -0.046 |
Amen | -0.041 |
ons | -0.041 |
Offic | -0.040 |
oc | -0.038 |
Token | They |
Token ID | 1119 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
tub | +0.007 |
gan | +0.006 |
ider | +0.006 |
Moe | +0.006 |
ahl | +0.005 |
Neg logit contributions | |
envelop | -0.010 |
Offic | -0.009 |
Amen | -0.009 |
ons | -0.009 |
Mic | -0.008 |
Token | wear |
Token ID | 5806 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
envelop | +0.030 |
Offic | +0.027 |
Amen | +0.026 |
ons | +0.025 |
Mic | +0.024 |
Neg logit contributions | |
tub | -0.036 |
gan | -0.034 |
ider | -0.031 |
Moe | -0.031 |
ahl | -0.030 |
Token | apr |
Token ID | 46593 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
envelop | +0.079 |
Offic | +0.076 |
Amen | +0.074 |
ons | +0.070 |
Mic | +0.068 |
Neg logit contributions | |
tub | -0.073 |
gan | -0.069 |
Moe | -0.061 |
ider | -0.059 |
ahl | -0.058 |
Token | ons |
Token ID | 684 |
Feature activation | -6.58e-02 |
Pos logit contributions | |
Offic | +0.129 |
Amen | +0.122 |
envelop | +0.112 |
Mic | +0.089 |
saddle | +0.088 |
Neg logit contributions | |
tub | -0.351 |
gan | -0.347 |
Moe | -0.330 |
ia | -0.320 |
icky | -0.311 |
Token | and |
Token ID | 290 |
Feature activation | -8.29e-02 |
Pos logit contributions | |
envelop | +0.123 |
Offic | +0.108 |
Amen | +0.106 |
ons | +0.106 |
Mic | +0.096 |
Neg logit contributions | |
tub | -0.130 |
gan | -0.125 |
ider | -0.114 |
Moe | -0.113 |
ahl | -0.111 |
Token | hats |
Token ID | 23910 |
Feature activation | -7.19e-02 |
Pos logit contributions | |
Offic | +0.154 |
Amen | +0.154 |
envelop | +0.154 |
Mic | +0.141 |
ons | +0.137 |
Neg logit contributions | |
tub | -0.147 |
gan | -0.142 |
ses | -0.124 |
Mil | -0.122 |
Moe | -0.122 |
Token | and |
Token ID | 290 |
Feature activation | -7.69e-02 |
Pos logit contributions | |
envelop | +0.139 |
Offic | +0.122 |
Amen | +0.120 |
ons | +0.119 |
oc | +0.107 |
Neg logit contributions | |
tub | -0.136 |
gan | -0.130 |
Moe | -0.119 |
ahl | -0.117 |
ider | -0.117 |
Token | act |
Token ID | 719 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
Offic | +0.140 |
Amen | +0.140 |
envelop | +0.139 |
ons | +0.126 |
Mic | +0.126 |
Neg logit contributions | |
tub | -0.139 |
gan | -0.136 |
ses | -0.120 |
sed | -0.117 |
Mil | -0.116 |
Token | I |
Token ID | 314 |
Feature activation | -1.82e-03 |
Pos logit contributions | |
tub | +0.006 |
gan | +0.006 |
Moe | +0.005 |
ider | +0.005 |
ahl | +0.005 |
Neg logit contributions | |
envelop | -0.010 |
Amen | -0.009 |
Offic | -0.009 |
ons | -0.009 |
oc | -0.009 |
Token | also |
Token ID | 635 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
envelop | +0.003 |
Offic | +0.003 |
Amen | +0.003 |
ons | +0.003 |
oc | +0.003 |
Neg logit contributions | |
tub | -0.004 |
gan | -0.003 |
Moe | -0.003 |
ider | -0.003 |
ahl | -0.003 |
Token | have |
Token ID | 423 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
envelop | +0.024 |
Offic | +0.022 |
Amen | +0.021 |
Mic | +0.021 |
ons | +0.020 |
Neg logit contributions | |
tub | -0.025 |
gan | -0.024 |
ider | -0.022 |
ses | -0.021 |
Moe | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
envelop | +0.054 |
Offic | +0.053 |
Amen | +0.053 |
Mic | +0.049 |
ons | +0.049 |
Neg logit contributions | |
tub | -0.037 |
gan | -0.036 |
Moe | -0.030 |
ider | -0.030 |
Mil | -0.029 |
Token | magic |
Token ID | 5536 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
Offic | +0.091 |
ons | +0.089 |
envelop | +0.088 |
Mic | +0.083 |
Amen | +0.081 |
Neg logit contributions | |
tub | -0.092 |
ses | -0.078 |
gan | -0.078 |
sed | -0.077 |
fault | -0.075 |
Token | sle |
Token ID | 3133 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
envelop | +0.023 |
Offic | +0.022 |
Amen | +0.020 |
ons | +0.020 |
Mic | +0.019 |
Neg logit contributions | |
tub | -0.028 |
gan | -0.026 |
ider | -0.024 |
Moe | -0.023 |
ses | -0.023 |
Token | igh |
Token ID | 394 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
Offic | +0.157 |
envelop | +0.155 |
Mic | +0.147 |
pic | +0.123 |
inviting | +0.121 |
Neg logit contributions | |
ider | -0.311 |
tub | -0.303 |
gan | -0.300 |
ses | -0.290 |
sed | -0.282 |
Token | and |
Token ID | 290 |
Feature activation | -7.26e-02 |
Pos logit contributions | |
envelop | +0.107 |
Offic | +0.099 |
Amen | +0.097 |
ons | +0.093 |
Mic | +0.088 |
Neg logit contributions | |
tub | -0.097 |
gan | -0.092 |
Moe | -0.082 |
ider | -0.081 |
ahl | -0.079 |
Token | re |
Token ID | 302 |
Feature activation | -9.99e-02 |
Pos logit contributions | |
envelop | +0.140 |
Offic | +0.135 |
Amen | +0.130 |
ons | +0.125 |
Mic | +0.125 |
Neg logit contributions | |
tub | -0.126 |
gan | -0.118 |
ses | -0.107 |
ider | -0.107 |
Moe | -0.105 |
Token | ind |
Token ID | 521 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
Mic | +0.152 |
envelop | +0.152 |
Offic | +0.148 |
lending | +0.134 |
permitting | +0.134 |
Neg logit contributions | |
gan | -0.187 |
tub | -0.176 |
flies | -0.171 |
ider | -0.171 |
uffy | -0.167 |
Token | eer |
Token ID | 28153 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
envelop | +0.116 |
Offic | +0.108 |
Amen | +0.104 |
ons | +0.100 |
Mic | +0.097 |
Neg logit contributions | |
tub | -0.114 |
gan | -0.109 |
ider | -0.097 |
Moe | -0.096 |
ahl | -0.093 |
Token | Tom |
Token ID | 4186 |
Feature activation | -6.15e-03 |
Pos logit contributions | |
tub | +0.006 |
gan | +0.005 |
ider | +0.005 |
Moe | +0.005 |
ahl | +0.005 |
Neg logit contributions | |
envelop | -0.009 |
Offic | -0.008 |
Amen | -0.008 |
ons | -0.007 |
Mic | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
envelop | +0.011 |
Offic | +0.010 |
Amen | +0.009 |
ons | +0.009 |
oc | +0.009 |
Neg logit contributions | |
tub | -0.013 |
gan | -0.012 |
ider | -0.011 |
Moe | -0.011 |
ahl | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -8.86e-03 |
Pos logit contributions | |
envelop | +0.047 |
Amen | +0.042 |
Offic | +0.039 |
ons | +0.038 |
oc | +0.038 |
Neg logit contributions | |
tub | -0.050 |
ses | -0.044 |
gan | -0.044 |
ahl | -0.043 |
ider | -0.042 |
Token | holding |
Token ID | 4769 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
envelop | +0.017 |
Offic | +0.015 |
ons | +0.015 |
Amen | +0.015 |
Mic | +0.014 |
Neg logit contributions | |
tub | -0.017 |
gan | -0.017 |
ider | -0.015 |
Moe | -0.015 |
Mil | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
envelop | +0.047 |
Offic | +0.044 |
Amen | +0.043 |
ons | +0.041 |
Mic | +0.039 |
Neg logit contributions | |
tub | -0.045 |
gan | -0.043 |
Moe | -0.039 |
ider | -0.038 |
ahl | -0.037 |
Token | stuffed |
Token ID | 22259 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
Offic | +0.103 |
envelop | +0.101 |
Amen | +0.100 |
ons | +0.097 |
itivity | +0.093 |
Neg logit contributions | |
tub | -0.088 |
gan | -0.075 |
ses | -0.070 |
sed | -0.069 |
ahl | -0.067 |
Token | dog |
Token ID | 3290 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
envelop | +0.289 |
Offic | +0.281 |
Amen | +0.275 |
ons | +0.273 |
itivity | +0.264 |
Neg logit contributions | |
tub | -0.194 |
gan | -0.173 |
Moe | -0.164 |
ahl | -0.162 |
ter | -0.161 |
Token | as |
Token ID | 355 |
Feature activation | -3.93e-03 |
Pos logit contributions | |
envelop | +0.127 |
Offic | +0.124 |
Amen | +0.121 |
ons | +0.116 |
Mic | +0.113 |
Neg logit contributions | |
tub | -0.086 |
gan | -0.081 |
ider | -0.075 |
Moe | -0.073 |
ses | -0.066 |
Token | his |
Token ID | 465 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
envelop | +0.008 |
Offic | +0.007 |
ons | +0.007 |
Amen | +0.007 |
parcels | +0.006 |
Neg logit contributions | |
tub | -0.007 |
gan | -0.007 |
ahl | -0.006 |
ider | -0.006 |
Moe | -0.006 |
Token | bird |
Token ID | 6512 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
tub | +0.024 |
gan | +0.023 |
Moe | +0.020 |
ider | +0.020 |
ahl | +0.020 |
Neg logit contributions | |
envelop | -0.020 |
Offic | -0.019 |
Amen | -0.018 |
ons | -0.017 |
Mic | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.37e-03 |
Pos logit contributions | |
envelop | +0.026 |
Offic | +0.024 |
Amen | +0.024 |
ons | +0.023 |
Mic | +0.022 |
Neg logit contributions | |
tub | -0.024 |
gan | -0.023 |
ider | -0.020 |
Moe | -0.020 |
ahl | -0.020 |
Token | outside |
Token ID | 2354 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
tub | +0.044 |
gan | +0.042 |
Moe | +0.037 |
ider | +0.035 |
ahl | +0.034 |
Neg logit contributions | |
envelop | -0.064 |
Amen | -0.063 |
Offic | -0.063 |
ons | -0.059 |
Mic | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | -7.23e-02 |
Pos logit contributions | |
tub | +0.073 |
gan | +0.068 |
Moe | +0.061 |
ider | +0.060 |
Mil | +0.058 |
Neg logit contributions | |
envelop | -0.079 |
Offic | -0.074 |
Amen | -0.072 |
ons | -0.069 |
Mic | -0.068 |
Token | showed |
Token ID | 3751 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
envelop | +0.129 |
Offic | +0.119 |
Amen | +0.117 |
ons | +0.111 |
Mic | +0.105 |
Neg logit contributions | |
tub | -0.148 |
gan | -0.142 |
Moe | -0.128 |
ider | -0.127 |
ahl | -0.123 |
Token | them |
Token ID | 606 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
envelop | +0.068 |
Offic | +0.061 |
Amen | +0.059 |
ons | +0.055 |
Mic | +0.051 |
Neg logit contributions | |
tub | -0.122 |
gan | -0.118 |
ider | -0.108 |
Moe | -0.108 |
ahl | -0.105 |
Token | his |
Token ID | 465 |
Feature activation | +6.50e-03 |
Pos logit contributions | |
tub | +0.045 |
gan | +0.042 |
ider | +0.041 |
ses | +0.039 |
sed | +0.038 |
Neg logit contributions | |
envelop | -0.032 |
Offic | -0.027 |
ons | -0.026 |
Amen | -0.025 |
parcels | -0.023 |
Token | sle |
Token ID | 3133 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
tub | +0.013 |
gan | +0.012 |
ses | +0.011 |
sed | +0.011 |
ider | +0.011 |
Neg logit contributions | |
Offic | -0.011 |
envelop | -0.011 |
Mic | -0.011 |
ons | -0.010 |
Amen | -0.010 |
Token | igh |
Token ID | 394 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
Offic | +0.150 |
envelop | +0.143 |
Mic | +0.138 |
permitting | +0.114 |
inviting | +0.112 |
Neg logit contributions | |
tub | -0.310 |
gan | -0.308 |
ider | -0.299 |
uffy | -0.293 |
ses | -0.292 |
Token | and |
Token ID | 290 |
Feature activation | -8.15e-02 |
Pos logit contributions | |
envelop | +0.107 |
Offic | +0.100 |
Amen | +0.098 |
ons | +0.094 |
Mic | +0.091 |
Neg logit contributions | |
tub | -0.080 |
gan | -0.075 |
ider | -0.066 |
Moe | -0.066 |
ahl | -0.063 |
Token | re |
Token ID | 302 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
Amen | +0.136 |
Offic | +0.134 |
envelop | +0.129 |
Mic | +0.125 |
ons | +0.124 |
Neg logit contributions | |
tub | -0.156 |
gan | -0.143 |
ses | -0.136 |
sed | -0.135 |
Moe | -0.134 |
Token | ind |
Token ID | 521 |
Feature activation | -6.64e-02 |
Pos logit contributions | |
envelop | +0.157 |
Offic | +0.151 |
Mic | +0.149 |
permitting | +0.137 |
lending | +0.135 |
Neg logit contributions | |
gan | -0.197 |
tub | -0.187 |
uffy | -0.180 |
flies | -0.170 |
anda | -0.170 |
Token | eer |
Token ID | 28153 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
envelop | +0.126 |
Offic | +0.117 |
Amen | +0.113 |
ons | +0.109 |
Mic | +0.103 |
Neg logit contributions | |
tub | -0.129 |
gan | -0.123 |
Moe | -0.110 |
ider | -0.110 |
ahl | -0.106 |
Token | put |
Token ID | 1234 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
envelop | +0.006 |
Offic | +0.006 |
Amen | +0.005 |
ons | +0.005 |
Mic | +0.005 |
Neg logit contributions | |
tub | -0.005 |
gan | -0.005 |
Moe | -0.004 |
ider | -0.004 |
ahl | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
envelop | +0.072 |
Offic | +0.066 |
Amen | +0.065 |
ons | +0.061 |
Mic | +0.058 |
Neg logit contributions | |
tub | -0.090 |
gan | -0.086 |
Moe | -0.078 |
ider | -0.078 |
ahl | -0.076 |
Token | shiny |
Token ID | 22441 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
envelop | +0.083 |
Offic | +0.083 |
Amen | +0.081 |
ons | +0.076 |
Mic | +0.072 |
Neg logit contributions | |
tub | -0.085 |
gan | -0.077 |
ider | -0.069 |
Moe | -0.068 |
ahl | -0.068 |
Token | saddle |
Token ID | 33083 |
Feature activation | -9.04e-02 |
Pos logit contributions | |
envelop | +0.108 |
Offic | +0.097 |
Amen | +0.095 |
ons | +0.091 |
Mic | +0.085 |
Neg logit contributions | |
tub | -0.133 |
gan | -0.128 |
Moe | -0.115 |
ider | -0.115 |
ahl | -0.112 |
Token | and |
Token ID | 290 |
Feature activation | -7.42e-02 |
Pos logit contributions | |
envelop | +0.165 |
Offic | +0.150 |
Amen | +0.148 |
ons | +0.142 |
Mic | +0.132 |
Neg logit contributions | |
tub | -0.183 |
gan | -0.174 |
Moe | -0.157 |
ider | -0.156 |
ahl | -0.152 |
Token | brid |
Token ID | 38265 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
envelop | +0.146 |
Amen | +0.141 |
Offic | +0.137 |
ons | +0.127 |
Mic | +0.125 |
Neg logit contributions | |
tub | -0.132 |
gan | -0.128 |
Moe | -0.113 |
ider | -0.110 |
Mil | -0.110 |
Token | le |
Token ID | 293 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
envelop | +0.252 |
Offic | +0.235 |
Amen | +0.228 |
Mic | +0.211 |
ons | +0.210 |
Neg logit contributions | |
tub | -0.249 |
gan | -0.247 |
Moe | -0.221 |
ider | -0.220 |
ahl | -0.214 |
Token | on |
Token ID | 319 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
envelop | +0.023 |
ons | +0.020 |
Offic | +0.020 |
Amen | +0.019 |
oc | +0.018 |
Neg logit contributions | |
tub | -0.023 |
gan | -0.022 |
ider | -0.020 |
Moe | -0.020 |
ahl | -0.020 |
Token | him |
Token ID | 683 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
envelop | +0.046 |
Amen | +0.046 |
Offic | +0.045 |
ons | +0.041 |
itivity | +0.040 |
Neg logit contributions | |
tub | -0.041 |
gan | -0.040 |
Moe | -0.037 |
Mil | -0.036 |
ahl | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
tub | +0.046 |
gan | +0.043 |
ider | +0.038 |
Moe | +0.038 |
ahl | +0.036 |
Neg logit contributions | |
envelop | -0.056 |
Offic | -0.053 |
Amen | -0.051 |
ons | -0.049 |
Mic | -0.047 |
Token | Finally |
Token ID | 9461 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
tub | +0.024 |
gan | +0.022 |
ider | +0.021 |
ahl | +0.019 |
Moe | +0.019 |
Neg logit contributions | |
envelop | -0.032 |
Offic | -0.030 |
Amen | -0.028 |
Mic | -0.027 |
ons | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
tub | +0.046 |
gan | +0.043 |
ider | +0.038 |
Moe | +0.037 |
ahl | +0.036 |
Neg logit contributions | |
envelop | -0.061 |
Offic | -0.056 |
Amen | -0.055 |
ons | -0.053 |
Mic | -0.051 |
Token | She |
Token ID | 1375 |
Feature activation | -1.53e-03 |
Pos logit contributions | |
tub | +0.020 |
ider | +0.017 |
gan | +0.016 |
ulu | +0.016 |
auc | +0.016 |
Neg logit contributions | |
envelop | -0.020 |
Amen | -0.020 |
Offic | -0.018 |
parcels | -0.017 |
read | -0.016 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
envelop | +0.003 |
Offic | +0.003 |
Amen | +0.003 |
ons | +0.003 |
Mic | +0.003 |
Neg logit contributions | |
tub | -0.003 |
gan | -0.003 |
Moe | -0.002 |
ider | -0.002 |
ahl | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
envelop | +0.053 |
Offic | +0.052 |
Amen | +0.051 |
ons | +0.047 |
oc | +0.047 |
Neg logit contributions | |
tub | -0.045 |
gan | -0.044 |
Moe | -0.040 |
Mil | -0.038 |
ahl | -0.037 |
Token | fire |
Token ID | 2046 |
Feature activation | -3.49e-03 |
Pos logit contributions | |
tub | +0.047 |
gan | +0.046 |
Moe | +0.041 |
ider | +0.041 |
ahl | +0.041 |
Neg logit contributions | |
envelop | -0.042 |
Amen | -0.038 |
Offic | -0.038 |
ons | -0.035 |
oc | -0.033 |
Token | extingu |
Token ID | 24995 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
envelop | +0.006 |
Amen | +0.005 |
Offic | +0.005 |
oc | +0.005 |
ons | +0.005 |
Neg logit contributions | |
tub | -0.007 |
gan | -0.007 |
ahl | -0.006 |
Moe | -0.006 |
ider | -0.006 |
Token | isher |
Token ID | 4828 |
Feature activation | +3.21e-03 |
Pos logit contributions | |
envelop | +0.307 |
Offic | +0.298 |
Amen | +0.282 |
Mic | +0.279 |
ons | +0.274 |
Neg logit contributions | |
gan | -0.190 |
tub | -0.186 |
ider | -0.162 |
Moe | -0.161 |
ahl | -0.150 |
Token | and |
Token ID | 290 |
Feature activation | -7.00e-02 |
Pos logit contributions | |
tub | +0.006 |
gan | +0.006 |
Moe | +0.006 |
ider | +0.006 |
ahl | +0.005 |
Neg logit contributions | |
envelop | -0.006 |
Offic | -0.005 |
Amen | -0.005 |
ons | -0.005 |
oc | -0.005 |
Token | put |
Token ID | 1234 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
envelop | +0.132 |
Amen | +0.123 |
Offic | +0.123 |
ons | +0.114 |
Mic | +0.112 |
Neg logit contributions | |
tub | -0.135 |
gan | -0.130 |
Moe | -0.116 |
ider | -0.112 |
Mil | -0.111 |
Token | out |
Token ID | 503 |
Feature activation | +3.70e-03 |
Pos logit contributions | |
envelop | +0.067 |
Offic | +0.064 |
Amen | +0.061 |
ons | +0.057 |
Mic | +0.056 |
Neg logit contributions | |
tub | -0.078 |
gan | -0.075 |
Moe | -0.068 |
ider | -0.065 |
ahl | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
tub | +0.008 |
gan | +0.007 |
Moe | +0.007 |
ider | +0.007 |
ahl | +0.007 |
Neg logit contributions | |
envelop | -0.006 |
Offic | -0.006 |
Amen | -0.006 |
ons | -0.005 |
Mic | -0.005 |
Token | put |
Token ID | 1234 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
envelop | +0.006 |
Amen | +0.005 |
Offic | +0.005 |
ons | +0.005 |
Mic | +0.005 |
Neg logit contributions | |
tub | -0.005 |
gan | -0.005 |
Moe | -0.004 |
ider | -0.004 |
Mil | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
envelop | +0.074 |
Offic | +0.071 |
Amen | +0.068 |
ons | +0.064 |
Mic | +0.063 |
Neg logit contributions | |
tub | -0.072 |
gan | -0.070 |
Moe | -0.065 |
ses | -0.061 |
ahl | -0.060 |
Token | saddle |
Token ID | 33083 |
Feature activation | -9.27e-02 |
Pos logit contributions | |
ons | +0.089 |
Offic | +0.089 |
envelop | +0.089 |
Amen | +0.088 |
itivity | +0.085 |
Neg logit contributions | |
tub | -0.079 |
gan | -0.067 |
ider | -0.062 |
sed | -0.062 |
ses | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
envelop | +0.179 |
Offic | +0.167 |
Amen | +0.162 |
ons | +0.156 |
Mic | +0.151 |
Neg logit contributions | |
tub | -0.177 |
gan | -0.169 |
ider | -0.152 |
Moe | -0.149 |
ahl | -0.144 |
Token | a |
Token ID | 257 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
envelop | +0.153 |
Amen | +0.149 |
Offic | +0.142 |
ons | +0.136 |
Mic | +0.132 |
Neg logit contributions | |
tub | -0.131 |
gan | -0.128 |
Moe | -0.113 |
ses | -0.111 |
Mil | -0.110 |
Token | brid |
Token ID | 38265 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
envelop | +0.086 |
Offic | +0.084 |
ons | +0.084 |
Amen | +0.082 |
itivity | +0.076 |
Neg logit contributions | |
tub | -0.101 |
gan | -0.090 |
ider | -0.086 |
ses | -0.082 |
sed | -0.081 |
Token | le |
Token ID | 293 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
envelop | +0.270 |
Offic | +0.248 |
Amen | +0.240 |
Mic | +0.236 |
parcels | +0.226 |
Neg logit contributions | |
tub | -0.222 |
gan | -0.220 |
ider | -0.204 |
Moe | -0.198 |
ahl | -0.192 |
Token | on |
Token ID | 319 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
envelop | +0.031 |
Offic | +0.028 |
Amen | +0.027 |
ons | +0.026 |
oc | +0.024 |
Neg logit contributions | |
tub | -0.033 |
gan | -0.032 |
Moe | -0.029 |
ider | -0.028 |
ahl | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
Amen | +0.040 |
envelop | +0.039 |
Offic | +0.038 |
ons | +0.036 |
itivity | +0.036 |
Neg logit contributions | |
tub | -0.029 |
gan | -0.029 |
Moe | -0.027 |
Mil | -0.026 |
ses | -0.025 |
Token | horse |
Token ID | 8223 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
tub | +0.076 |
gan | +0.073 |
ses | +0.067 |
sed | +0.066 |
ider | +0.066 |
Neg logit contributions | |
envelop | -0.056 |
ons | -0.055 |
Offic | -0.054 |
Amen | -0.052 |
Mic | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
envelop | +0.071 |
Offic | +0.069 |
Amen | +0.068 |
Mic | +0.064 |
ons | +0.063 |
Neg logit contributions | |
tub | -0.039 |
gan | -0.035 |
ider | -0.031 |
Moe | -0.029 |
ses | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
tub | +0.006 |
gan | +0.004 |
Moe | +0.004 |
Mil | +0.004 |
ider | +0.004 |
Neg logit contributions | |
Offic | -0.013 |
Mic | -0.013 |
envelop | -0.013 |
itivity | -0.012 |
ons | -0.012 |
Token | He |
Token ID | 679 |
Feature activation | -6.93e-03 |
Pos logit contributions | |
tub | +0.017 |
gan | +0.016 |
ider | +0.014 |
ter | +0.014 |
ses | +0.014 |
Neg logit contributions | |
envelop | -0.022 |
Amen | -0.021 |
Offic | -0.020 |
read | -0.018 |
oc | -0.018 |
Token | gave |
Token ID | 2921 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
envelop | +0.013 |
Offic | +0.012 |
Amen | +0.012 |
ons | +0.012 |
Mic | +0.011 |
Neg logit contributions | |
tub | -0.013 |
gan | -0.013 |
Moe | -0.011 |
ider | -0.011 |
ahl | -0.011 |
Token | her |
Token ID | 607 |
Feature activation | -2.86e-03 |
Pos logit contributions | |
envelop | +0.115 |
Amen | +0.102 |
Offic | +0.098 |
ons | +0.097 |
Mic | +0.085 |
Neg logit contributions | |
tub | -0.154 |
gan | -0.143 |
ider | -0.137 |
ses | -0.131 |
Moe | -0.129 |
Token | a |
Token ID | 257 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
envelop | +0.006 |
Offic | +0.006 |
Amen | +0.006 |
ons | +0.006 |
Mic | +0.006 |
Neg logit contributions | |
tub | -0.004 |
gan | -0.004 |
Moe | -0.004 |
ider | -0.004 |
ahl | -0.003 |
Token | stuffed |
Token ID | 22259 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
envelop | +0.100 |
Offic | +0.094 |
Amen | +0.090 |
ons | +0.086 |
Mic | +0.081 |
Neg logit contributions | |
tub | -0.123 |
gan | -0.115 |
Moe | -0.105 |
ider | -0.104 |
ahl | -0.103 |
Token | bear |
Token ID | 6842 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
envelop | +0.281 |
Offic | +0.267 |
Amen | +0.259 |
ons | +0.249 |
Mic | +0.242 |
Neg logit contributions | |
tub | -0.217 |
gan | -0.201 |
Moe | -0.179 |
ahl | -0.177 |
Mil | -0.171 |
Token | and |
Token ID | 290 |
Feature activation | -8.09e-02 |
Pos logit contributions | |
envelop | +0.108 |
Offic | +0.104 |
Amen | +0.101 |
ons | +0.097 |
Mic | +0.093 |
Neg logit contributions | |
tub | -0.090 |
gan | -0.086 |
ider | -0.075 |
Moe | -0.075 |
ahl | -0.071 |
Token | a |
Token ID | 257 |
Feature activation | -5.77e-02 |
Pos logit contributions | |
envelop | +0.177 |
Offic | +0.166 |
Amen | +0.163 |
ons | +0.157 |
Mic | +0.150 |
Neg logit contributions | |
tub | -0.133 |
gan | -0.125 |
Moe | -0.110 |
ider | -0.109 |
ahl | -0.105 |
Token | night |
Token ID | 1755 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
envelop | +0.099 |
Offic | +0.092 |
Amen | +0.089 |
ons | +0.085 |
Mic | +0.080 |
Neg logit contributions | |
tub | -0.122 |
gan | -0.115 |
ider | -0.105 |
Moe | -0.104 |
ahl | -0.102 |
Token | light |
Token ID | 1657 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
gan | +0.074 |
tub | +0.070 |
Moe | +0.065 |
ahl | +0.065 |
ses | +0.063 |
Neg logit contributions | |
envelop | -0.074 |
Amen | -0.064 |
ons | -0.064 |
Offic | -0.063 |
oc | -0.061 |
Token | trial |
Token ID | 4473 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
tub | +0.096 |
ider | +0.083 |
gan | +0.081 |
ses | +0.079 |
sed | +0.075 |
Neg logit contributions | |
Amen | -0.122 |
envelop | -0.115 |
Offic | -0.112 |
itivity | -0.106 |
ons | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
tub | +0.019 |
gan | +0.018 |
ider | +0.016 |
Moe | +0.016 |
ses | +0.015 |
Neg logit contributions | |
envelop | -0.020 |
Offic | -0.019 |
Amen | -0.019 |
ons | -0.017 |
Mic | -0.017 |
Token | They |
Token ID | 1119 |
Feature activation | -9.96e-03 |
Pos logit contributions | |
tub | +0.020 |
gan | +0.019 |
ider | +0.018 |
ahl | +0.017 |
ble | +0.017 |
Neg logit contributions | |
envelop | -0.027 |
Offic | -0.023 |
Amen | -0.022 |
parcels | -0.021 |
read | -0.020 |
Token | used |
Token ID | 973 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
Amen | +0.019 |
envelop | +0.019 |
Offic | +0.018 |
ons | +0.017 |
Mic | +0.016 |
Neg logit contributions | |
gan | -0.018 |
tub | -0.018 |
Moe | -0.016 |
ahl | -0.015 |
Mil | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
Amen | +0.103 |
Offic | +0.103 |
envelop | +0.100 |
Mic | +0.092 |
oc | +0.092 |
Neg logit contributions | |
tub | -0.083 |
gan | -0.078 |
Moe | -0.072 |
ses | -0.069 |
Mil | -0.067 |
Token | stuffed |
Token ID | 22259 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
envelop | +0.093 |
Amen | +0.092 |
Offic | +0.091 |
ons | +0.085 |
Mic | +0.080 |
Neg logit contributions | |
tub | -0.095 |
gan | -0.084 |
ses | -0.078 |
ider | -0.077 |
Moe | -0.077 |
Token | dog |
Token ID | 3290 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
Amen | +0.269 |
envelop | +0.262 |
Offic | +0.258 |
ons | +0.244 |
Mic | +0.243 |
Neg logit contributions | |
tub | -0.201 |
gan | -0.179 |
ses | -0.178 |
sed | -0.168 |
Moe | -0.168 |
Token | and |
Token ID | 290 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
envelop | +0.121 |
Offic | +0.113 |
Amen | +0.110 |
ons | +0.106 |
Mic | +0.101 |
Neg logit contributions | |
tub | -0.105 |
gan | -0.099 |
Moe | -0.088 |
ider | -0.088 |
ahl | -0.085 |
Token | a |
Token ID | 257 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
envelop | +0.162 |
Amen | +0.155 |
Offic | +0.154 |
ons | +0.145 |
Mic | +0.139 |
Neg logit contributions | |
tub | -0.133 |
gan | -0.127 |
Moe | -0.112 |
ider | -0.109 |
ses | -0.108 |
Token | stuffed |
Token ID | 22259 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
envelop | +0.091 |
Offic | +0.086 |
Amen | +0.085 |
ons | +0.081 |
Mic | +0.076 |
Neg logit contributions | |
tub | -0.095 |
gan | -0.088 |
ider | -0.080 |
Moe | -0.079 |
ses | -0.078 |
Token | cat |
Token ID | 3797 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
envelop | +0.260 |
Amen | +0.251 |
Offic | +0.246 |
ons | +0.242 |
Mic | +0.229 |
Neg logit contributions | |
tub | -0.209 |
gan | -0.192 |
ses | -0.178 |
Moe | -0.177 |
sed | -0.170 |
Token | come |
Token ID | 1282 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
envelop | +0.030 |
Offic | +0.028 |
Amen | +0.027 |
ons | +0.027 |
Mic | +0.025 |
Neg logit contributions | |
tub | -0.022 |
gan | -0.020 |
ider | -0.018 |
Moe | -0.018 |
ahl | -0.017 |
Token | in |
Token ID | 287 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
tub | +0.056 |
gan | +0.053 |
Moe | +0.046 |
ider | +0.046 |
ahl | +0.044 |
Neg logit contributions | |
envelop | -0.073 |
Offic | -0.068 |
Amen | -0.066 |
ons | -0.064 |
Mic | -0.061 |
Token | ," |
Token ID | 553 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
tub | +0.035 |
gan | +0.033 |
Moe | +0.029 |
ider | +0.029 |
ahl | +0.027 |
Neg logit contributions | |
envelop | -0.045 |
Offic | -0.043 |
Amen | -0.041 |
ons | -0.040 |
Mic | -0.039 |
Token | said |
Token ID | 531 |
Feature activation | -6.80e-03 |
Pos logit contributions | |
envelop | +0.054 |
Offic | +0.049 |
Amen | +0.049 |
ons | +0.047 |
Mic | +0.045 |
Neg logit contributions | |
tub | -0.043 |
gan | -0.039 |
ider | -0.036 |
ahl | -0.035 |
Moe | -0.034 |
Token | his |
Token ID | 465 |
Feature activation | -6.59e-03 |
Pos logit contributions | |
envelop | +0.013 |
Amen | +0.012 |
Offic | +0.012 |
ons | +0.011 |
Mic | +0.010 |
Neg logit contributions | |
tub | -0.013 |
gan | -0.012 |
ider | -0.011 |
ahl | -0.011 |
ses | -0.011 |
Token | grand |
Token ID | 4490 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
envelop | +0.012 |
Amen | +0.011 |
Offic | +0.010 |
ons | +0.009 |
Mic | +0.009 |
Neg logit contributions | |
tub | -0.014 |
gan | -0.012 |
ahl | -0.011 |
Moe | -0.011 |
ider | -0.011 |
Token | pa |
Token ID | 8957 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
envelop | +0.105 |
Offic | +0.102 |
ons | +0.081 |
Mic | +0.080 |
Amen | +0.079 |
Neg logit contributions | |
gan | -0.360 |
tub | -0.359 |
Moe | -0.331 |
ider | -0.327 |
ses | -0.324 |
Token | . |
Token ID | 13 |
Feature activation | +4.93e-04 |
Pos logit contributions | |
envelop | +0.164 |
Amen | +0.151 |
Offic | +0.150 |
ons | +0.145 |
oc | +0.136 |
Neg logit contributions | |
tub | -0.138 |
gan | -0.127 |
ider | -0.114 |
Moe | -0.114 |
ahl | -0.112 |
Token | They |
Token ID | 1119 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
tub | +0.001 |
gan | +0.001 |
ider | +0.001 |
Moe | +0.001 |
ahl | +0.001 |
Neg logit contributions | |
envelop | -0.001 |
Offic | -0.001 |
Amen | -0.001 |
ons | -0.001 |
Mic | -0.001 |
Token | all |
Token ID | 477 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
envelop | +0.050 |
Amen | +0.047 |
Offic | +0.046 |
ons | +0.044 |
Mic | +0.043 |
Neg logit contributions | |
tub | -0.045 |
gan | -0.044 |
Moe | -0.039 |
Mil | -0.036 |
ahl | -0.036 |
Token | turned |
Token ID | 2900 |
Feature activation | +8.76e-03 |
Pos logit contributions | |
envelop | +0.029 |
Offic | +0.027 |
Amen | +0.026 |
ons | +0.025 |
Mic | +0.024 |
Neg logit contributions | |
tub | -0.030 |
gan | -0.029 |
Moe | -0.026 |
ider | -0.025 |
ahl | -0.024 |
Token | so |
Token ID | 523 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
tub | +0.021 |
gan | +0.020 |
Moe | +0.017 |
ider | +0.016 |
Mil | +0.016 |
Neg logit contributions | |
envelop | -0.033 |
Offic | -0.031 |
Amen | -0.031 |
ons | -0.029 |
Mic | -0.028 |
Token | he |
Token ID | 339 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
envelop | +0.054 |
Offic | +0.051 |
Amen | +0.050 |
ons | +0.049 |
Mic | +0.046 |
Neg logit contributions | |
tub | -0.035 |
gan | -0.032 |
ider | -0.028 |
Moe | -0.028 |
ahl | -0.027 |
Token | gave |
Token ID | 2921 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
tub | +0.036 |
gan | +0.035 |
Moe | +0.031 |
ider | +0.031 |
ahl | +0.030 |
Neg logit contributions | |
envelop | -0.032 |
Offic | -0.029 |
Amen | -0.029 |
ons | -0.027 |
Mic | -0.026 |
Token | her |
Token ID | 607 |
Feature activation | -4.80e-04 |
Pos logit contributions | |
envelop | +0.110 |
Offic | +0.108 |
Amen | +0.104 |
Mic | +0.094 |
oc | +0.093 |
Neg logit contributions | |
gan | -0.128 |
tub | -0.126 |
Moe | -0.120 |
Mil | -0.112 |
ahl | -0.110 |
Token | a |
Token ID | 257 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
envelop | +0.001 |
Offic | +0.001 |
Amen | +0.001 |
ons | +0.001 |
Mic | +0.001 |
Neg logit contributions | |
tub | -0.001 |
gan | -0.001 |
ider | -0.001 |
Moe | -0.001 |
ahl | -0.001 |
Token | stuffed |
Token ID | 22259 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
envelop | +0.103 |
Offic | +0.097 |
Amen | +0.094 |
ons | +0.090 |
Mic | +0.085 |
Neg logit contributions | |
tub | -0.116 |
gan | -0.108 |
Moe | -0.098 |
ider | -0.096 |
ahl | -0.095 |
Token | bear |
Token ID | 6842 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
envelop | +0.274 |
Offic | +0.260 |
Amen | +0.257 |
ons | +0.247 |
Mic | +0.239 |
Neg logit contributions | |
tub | -0.214 |
gan | -0.199 |
Moe | -0.181 |
ahl | -0.172 |
Mil | -0.167 |
Token | . |
Token ID | 13 |
Feature activation | +8.26e-03 |
Pos logit contributions | |
envelop | +0.108 |
Offic | +0.102 |
Amen | +0.100 |
ons | +0.096 |
Mic | +0.092 |
Neg logit contributions | |
tub | -0.084 |
gan | -0.079 |
Moe | -0.069 |
ider | -0.069 |
ahl | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
tub | +0.014 |
gan | +0.013 |
ider | +0.012 |
ses | +0.011 |
ahl | +0.011 |
Neg logit contributions | |
envelop | -0.018 |
Offic | -0.017 |
Amen | -0.016 |
ons | -0.016 |
Mic | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
envelop | +0.055 |
Amen | +0.050 |
Offic | +0.050 |
parcels | +0.047 |
itivity | +0.047 |
Neg logit contributions | |
tub | -0.025 |
gan | -0.022 |
Moe | -0.020 |
Mil | -0.018 |
ahl | -0.018 |
Token | The |
Token ID | 464 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
envelop | +0.039 |
Amen | +0.035 |
ons | +0.033 |
parcels | +0.032 |
pa | +0.032 |
Neg logit contributions | |
tub | -0.024 |
gan | -0.023 |
Moe | -0.022 |
ulu | -0.020 |
ider | -0.020 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.10e-03 |
Pos logit contributions | |
envelop | +0.045 |
Amen | +0.042 |
Offic | +0.042 |
ons | +0.039 |
oc | +0.039 |
Neg logit contributions | |
tub | -0.029 |
gan | -0.029 |
ider | -0.026 |
Moe | -0.025 |
ahl | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
tub | +0.002 |
gan | +0.002 |
Mil | +0.002 |
ider | +0.002 |
Moe | +0.002 |
Neg logit contributions | |
envelop | -0.002 |
Offic | -0.002 |
parcels | -0.002 |
Amen | -0.002 |
itivity | -0.002 |
Token | she |
Token ID | 673 |
Feature activation | +6.95e-03 |
Pos logit contributions | |
envelop | +0.112 |
Amen | +0.107 |
Offic | +0.104 |
ons | +0.099 |
Mic | +0.093 |
Neg logit contributions | |
tub | -0.115 |
gan | -0.107 |
ider | -0.099 |
Moe | -0.096 |
ahl | -0.093 |
Token | had |
Token ID | 550 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
tub | +0.014 |
gan | +0.014 |
Moe | +0.013 |
ider | +0.013 |
ahl | +0.012 |
Neg logit contributions | |
envelop | -0.012 |
Amen | -0.011 |
Offic | -0.011 |
ons | -0.011 |
oc | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
envelop | +0.025 |
Offic | +0.023 |
Amen | +0.023 |
ons | +0.022 |
Mic | +0.021 |
Neg logit contributions | |
tub | -0.019 |
gan | -0.018 |
ider | -0.016 |
Moe | -0.015 |
ahl | -0.015 |
Token | stuffed |
Token ID | 22259 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
envelop | +0.096 |
Offic | +0.089 |
Amen | +0.088 |
ons | +0.083 |
Mic | +0.080 |
Neg logit contributions | |
tub | -0.090 |
gan | -0.086 |
Moe | -0.077 |
ider | -0.077 |
ahl | -0.074 |
Token | bear |
Token ID | 6842 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
envelop | +0.264 |
Offic | +0.262 |
ons | +0.253 |
Amen | +0.253 |
itivity | +0.239 |
Neg logit contributions | |
tub | -0.207 |
gan | -0.185 |
ahl | -0.166 |
Moe | -0.166 |
ter | -0.165 |
Token | called |
Token ID | 1444 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
envelop | +0.090 |
Offic | +0.087 |
Amen | +0.083 |
ons | +0.079 |
Mic | +0.077 |
Neg logit contributions | |
tub | -0.102 |
gan | -0.094 |
ider | -0.088 |
Moe | -0.084 |
ses | -0.081 |
Token | Bear |
Token ID | 14732 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
envelop | +0.087 |
Offic | +0.079 |
Amen | +0.078 |
ons | +0.076 |
Mic | +0.072 |
Neg logit contributions | |
tub | -0.070 |
gan | -0.066 |
ider | -0.061 |
Moe | -0.057 |
ahl | -0.057 |
Token | y |
Token ID | 88 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
envelop | +0.093 |
Amen | +0.084 |
Offic | +0.084 |
ons | +0.080 |
oc | +0.077 |
Neg logit contributions | |
tub | -0.073 |
gan | -0.070 |
ider | -0.063 |
Moe | -0.062 |
ahl | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
envelop | +0.051 |
Amen | +0.047 |
Offic | +0.046 |
ons | +0.044 |
oc | +0.042 |
Neg logit contributions | |
tub | -0.042 |
gan | -0.040 |
ider | -0.036 |
Moe | -0.035 |
ahl | -0.035 |
Token | quickly |
Token ID | 2952 |
Feature activation | -6.55e-04 |
Pos logit contributions | |
tub | +0.019 |
gan | +0.019 |
Moe | +0.016 |
ider | +0.016 |
ahl | +0.016 |
Neg logit contributions | |
envelop | -0.022 |
Offic | -0.020 |
Amen | -0.020 |
ons | -0.019 |
Mic | -0.018 |
Token | checked |
Token ID | 10667 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
envelop | +0.001 |
Amen | +0.001 |
Offic | +0.001 |
ons | +0.001 |
Mic | +0.001 |
Neg logit contributions | |
tub | -0.001 |
gan | -0.001 |
Moe | -0.001 |
ider | -0.001 |
ahl | -0.001 |
Token | for |
Token ID | 329 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
envelop | +0.035 |
Offic | +0.032 |
Amen | +0.032 |
ons | +0.030 |
Mic | +0.029 |
Neg logit contributions | |
tub | -0.037 |
gan | -0.035 |
Moe | -0.032 |
ider | -0.031 |
ahl | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
envelop | +0.063 |
Amen | +0.062 |
Offic | +0.060 |
Mic | +0.057 |
oc | +0.057 |
Neg logit contributions | |
tub | -0.043 |
gan | -0.041 |
Moe | -0.038 |
Mil | -0.035 |
ider | -0.034 |
Token | fire |
Token ID | 2046 |
Feature activation | -5.62e-03 |
Pos logit contributions | |
envelop | +0.076 |
Offic | +0.071 |
Amen | +0.070 |
ons | +0.068 |
Mic | +0.064 |
Neg logit contributions | |
tub | -0.064 |
gan | -0.059 |
ider | -0.053 |
Moe | -0.053 |
ahl | -0.050 |
Token | extingu |
Token ID | 24995 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
envelop | +0.011 |
Offic | +0.009 |
Amen | +0.009 |
oc | +0.009 |
ons | +0.009 |
Neg logit contributions | |
gan | -0.011 |
tub | -0.010 |
Moe | -0.010 |
ahl | -0.010 |
Mil | -0.009 |
Token | isher |
Token ID | 4828 |
Feature activation | -2.74e-03 |
Pos logit contributions | |
envelop | +0.296 |
Offic | +0.285 |
Amen | +0.273 |
Mic | +0.264 |
ons | +0.264 |
Neg logit contributions | |
tub | -0.186 |
gan | -0.183 |
ider | -0.158 |
Moe | -0.156 |
ses | -0.147 |
Token | , |
Token ID | 11 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
envelop | +0.006 |
Offic | +0.005 |
Amen | +0.005 |
ons | +0.005 |
oc | +0.005 |
Neg logit contributions | |
tub | -0.005 |
gan | -0.005 |
Moe | -0.004 |
ahl | -0.004 |
ider | -0.004 |
Token | but |
Token ID | 475 |
Feature activation | +6.25e-03 |
Pos logit contributions | |
tub | +0.026 |
gan | +0.024 |
Moe | +0.021 |
ider | +0.020 |
ahl | +0.019 |
Neg logit contributions | |
envelop | -0.046 |
Offic | -0.043 |
Amen | -0.043 |
ons | -0.041 |
Mic | -0.040 |
Token | there |
Token ID | 612 |
Feature activation | +5.10e-02 |
Pos logit contributions | |
tub | +0.013 |
gan | +0.012 |
ider | +0.011 |
ahl | +0.011 |
Moe | +0.011 |
Neg logit contributions | |
envelop | -0.011 |
Offic | -0.010 |
Amen | -0.010 |
ons | -0.010 |
read | -0.009 |
Token | wasn |
Token ID | 2492 |
Feature activation | +9.35e-03 |
Pos logit contributions | |
tub | +0.081 |
gan | +0.079 |
ider | +0.077 |
ahl | +0.068 |
ses | +0.068 |
Neg logit contributions | |
envelop | -0.112 |
ons | -0.104 |
Offic | -0.102 |
Amen | -0.097 |
oc | -0.093 |
Token | Tom |
Token ID | 4186 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
tub | +0.015 |
ider | +0.015 |
ses | +0.014 |
gan | +0.014 |
ter | +0.014 |
Neg logit contributions | |
envelop | -0.020 |
Offic | -0.018 |
Amen | -0.017 |
ons | -0.017 |
Mic | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -7.05e-02 |
Pos logit contributions | |
envelop | +0.020 |
Offic | +0.019 |
Amen | +0.019 |
ons | +0.017 |
Mic | +0.017 |
Neg logit contributions | |
tub | -0.020 |
gan | -0.019 |
Moe | -0.017 |
ider | -0.017 |
ahl | -0.016 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.34e-03 |
Pos logit contributions | |
envelop | +0.195 |
Offic | +0.186 |
Amen | +0.183 |
ons | +0.179 |
Mic | +0.172 |
Neg logit contributions | |
tub | -0.076 |
gan | -0.069 |
Moe | -0.056 |
ider | -0.053 |
ahl | -0.050 |
Token | thanked |
Token ID | 26280 |
Feature activation | -5.98e-02 |
Pos logit contributions | |
envelop | +0.007 |
Amen | +0.006 |
Offic | +0.006 |
ons | +0.006 |
Mic | +0.006 |
Neg logit contributions | |
tub | -0.006 |
gan | -0.006 |
Moe | -0.005 |
Mil | -0.005 |
ahl | -0.005 |
Token | their |
Token ID | 511 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
envelop | +0.118 |
Offic | +0.117 |
oc | +0.108 |
Amen | +0.108 |
Mic | +0.107 |
Neg logit contributions | |
gan | -0.098 |
tub | -0.096 |
Moe | -0.095 |
Mil | -0.090 |
Mia | -0.080 |
Token | grand |
Token ID | 4490 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
envelop | +0.045 |
ons | +0.043 |
Offic | +0.042 |
Amen | +0.041 |
Mic | +0.041 |
Neg logit contributions | |
gan | -0.044 |
tub | -0.044 |
Moe | -0.039 |
sed | -0.039 |
ses | -0.039 |
Token | pa |
Token ID | 8957 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
envelop | +0.123 |
Offic | +0.115 |
Mic | +0.098 |
ons | +0.094 |
Amen | +0.093 |
Neg logit contributions | |
gan | -0.341 |
tub | -0.335 |
Moe | -0.315 |
ses | -0.314 |
ia | -0.310 |
Token | and |
Token ID | 290 |
Feature activation | -9.78e-02 |
Pos logit contributions | |
envelop | +0.175 |
Offic | +0.162 |
Amen | +0.159 |
ons | +0.153 |
Mic | +0.145 |
Neg logit contributions | |
tub | -0.146 |
gan | -0.138 |
Moe | -0.122 |
ider | -0.122 |
ahl | -0.117 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
envelop | +0.188 |
Offic | +0.181 |
Amen | +0.179 |
ons | +0.175 |
Mic | +0.171 |
Neg logit contributions | |
gan | -0.166 |
tub | -0.164 |
Moe | -0.144 |
Mil | -0.139 |
ses | -0.136 |
Token | to |
Token ID | 284 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
envelop | +0.034 |
Offic | +0.033 |
Amen | +0.032 |
ons | +0.031 |
Mic | +0.030 |
Neg logit contributions | |
tub | -0.023 |
gan | -0.023 |
Moe | -0.021 |
ider | -0.019 |
ahl | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
envelop | +0.092 |
Offic | +0.090 |
ons | +0.089 |
oc | +0.088 |
Mic | +0.088 |
Neg logit contributions | |
Moe | -0.051 |
tub | -0.048 |
gan | -0.048 |
Mil | -0.046 |
Mia | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -8.67e-02 |
Pos logit contributions | |
gan | +0.021 |
tub | +0.021 |
Moe | +0.018 |
ider | +0.017 |
ahl | +0.016 |
Neg logit contributions | |
envelop | -0.027 |
Offic | -0.027 |
Amen | -0.025 |
ons | -0.024 |
Mic | -0.023 |
Token | thanked |
Token ID | 26280 |
Feature activation | -5.63e-02 |
Pos logit contributions | |
envelop | +0.171 |
Offic | +0.162 |
Amen | +0.161 |
ons | +0.153 |
Mic | +0.151 |
Neg logit contributions | |
tub | -0.153 |
gan | -0.148 |
Moe | -0.132 |
ses | -0.130 |
Mil | -0.128 |
Token | her |
Token ID | 607 |
Feature activation | -7.45e-03 |
Pos logit contributions | |
envelop | +0.098 |
Offic | +0.096 |
Amen | +0.094 |
Mic | +0.091 |
oc | +0.091 |
Neg logit contributions | |
gan | -0.105 |
tub | -0.104 |
Mil | -0.097 |
Moe | -0.095 |
ahl | -0.087 |
Token | grandma |
Token ID | 49890 |
Feature activation | -9.10e-02 |
Pos logit contributions | |
envelop | +0.014 |
Offic | +0.013 |
ons | +0.013 |
Amen | +0.013 |
Mic | +0.012 |
Neg logit contributions | |
tub | -0.014 |
gan | -0.013 |
ider | -0.012 |
ses | -0.011 |
Moe | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
envelop | +0.213 |
Offic | +0.204 |
Amen | +0.198 |
ons | +0.192 |
Mic | +0.186 |
Neg logit contributions | |
tub | -0.133 |
gan | -0.127 |
Moe | -0.109 |
ider | -0.107 |
ahl | -0.102 |
Token | grand |
Token ID | 4490 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
envelop | +0.194 |
Amen | +0.188 |
ons | +0.188 |
Offic | +0.186 |
oc | +0.184 |
Neg logit contributions | |
gan | -0.172 |
tub | -0.167 |
ses | -0.162 |
sed | -0.157 |
Mil | -0.146 |
Token | pa |
Token ID | 8957 |
Feature activation | -8.26e-02 |
Pos logit contributions | |
Offic | +0.114 |
envelop | +0.112 |
Mic | +0.103 |
Amen | +0.089 |
ons | +0.089 |
Neg logit contributions | |
gan | -0.335 |
tub | -0.326 |
ses | -0.315 |
fault | -0.307 |
Moe | -0.306 |
Token | . |
Token ID | 13 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
envelop | +0.185 |
Offic | +0.171 |
Amen | +0.169 |
ons | +0.163 |
Mic | +0.154 |
Neg logit contributions | |
tub | -0.133 |
gan | -0.125 |
ider | -0.109 |
Moe | -0.109 |
ahl | -0.105 |
Token | She |
Token ID | 1375 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
tub | +0.015 |
gan | +0.014 |
ider | +0.013 |
Moe | +0.012 |
ahl | +0.012 |
Neg logit contributions | |
envelop | -0.023 |
Offic | -0.022 |
Amen | -0.022 |
ons | -0.020 |
Mic | -0.020 |
Token | said |
Token ID | 531 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
envelop | +0.033 |
Offic | +0.031 |
Amen | +0.030 |
ons | +0.029 |
Mic | +0.028 |
Neg logit contributions | |
tub | -0.030 |
gan | -0.029 |
Moe | -0.026 |
ider | -0.025 |
ahl | -0.024 |
Token | they |
Token ID | 484 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
envelop | +0.075 |
Amen | +0.070 |
Offic | +0.070 |
ons | +0.067 |
Mic | +0.063 |
Neg logit contributions | |
tub | -0.053 |
gan | -0.049 |
ider | -0.044 |
Moe | -0.042 |
ahl | -0.042 |
Token | They |
Token ID | 1119 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
tub | +0.025 |
gan | +0.025 |
ider | +0.024 |
ahl | +0.022 |
sed | +0.021 |
Neg logit contributions | |
envelop | -0.030 |
Offic | -0.028 |
Amen | -0.027 |
parcels | -0.024 |
oc | -0.024 |
Token | thanked |
Token ID | 26280 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
envelop | +0.032 |
Amen | +0.031 |
Offic | +0.030 |
ons | +0.029 |
Mic | +0.028 |
Neg logit contributions | |
tub | -0.028 |
gan | -0.027 |
Moe | -0.024 |
ider | -0.023 |
Mil | -0.023 |
Token | their |
Token ID | 511 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
Offic | +0.122 |
envelop | +0.120 |
Mic | +0.117 |
Amen | +0.116 |
ons | +0.116 |
Neg logit contributions | |
gan | -0.093 |
tub | -0.091 |
Mil | -0.084 |
Moe | -0.084 |
ahl | -0.070 |
Token | grandma |
Token ID | 49890 |
Feature activation | -8.65e-02 |
Pos logit contributions | |
envelop | +0.043 |
Offic | +0.041 |
Amen | +0.041 |
ons | +0.040 |
Mic | +0.039 |
Neg logit contributions | |
tub | -0.040 |
gan | -0.039 |
Moe | -0.035 |
ses | -0.034 |
ider | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
envelop | +0.184 |
Offic | +0.180 |
Amen | +0.174 |
ons | +0.167 |
Mic | +0.167 |
Neg logit contributions | |
tub | -0.139 |
gan | -0.135 |
Moe | -0.116 |
ider | -0.111 |
Mil | -0.109 |
Token | grand |
Token ID | 4490 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
envelop | +0.194 |
Amen | +0.189 |
ons | +0.187 |
Offic | +0.184 |
Mic | +0.179 |
Neg logit contributions | |
gan | -0.188 |
tub | -0.188 |
ses | -0.167 |
Moe | -0.164 |
sed | -0.162 |
Token | pa |
Token ID | 8957 |
Feature activation | -8.14e-02 |
Pos logit contributions | |
Offic | +0.117 |
envelop | +0.111 |
Mic | +0.107 |
Amen | +0.092 |
ons | +0.088 |
Neg logit contributions | |
gan | -0.335 |
tub | -0.326 |
ses | -0.313 |
Moe | -0.306 |
fault | -0.305 |
Token | in |
Token ID | 287 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
envelop | +0.174 |
Offic | +0.165 |
Amen | +0.160 |
ons | +0.154 |
Mic | +0.149 |
Neg logit contributions | |
tub | -0.137 |
gan | -0.130 |
Moe | -0.114 |
ider | -0.113 |
ahl | -0.109 |
Token | their |
Token ID | 511 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
ons | +0.037 |
Offic | +0.037 |
Mic | +0.037 |
itivity | +0.034 |
envelop | +0.034 |
Neg logit contributions | |
tub | -0.019 |
Moe | -0.019 |
Mil | -0.018 |
gan | -0.017 |
ses | -0.016 |
Token | hearts |
Token ID | 11954 |
Feature activation | +1.41e-03 |
Pos logit contributions | |
Offic | +0.034 |
Amen | +0.033 |
ons | +0.033 |
Mic | +0.032 |
envelop | +0.031 |
Neg logit contributions | |
tub | -0.027 |
gan | -0.025 |
ses | -0.024 |
sed | -0.023 |
Moe | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -8.78e-02 |
Pos logit contributions | |
tub | +0.002 |
gan | +0.002 |
Moe | +0.002 |
ider | +0.002 |
ahl | +0.002 |
Neg logit contributions | |
envelop | -0.003 |
Offic | -0.003 |
Amen | -0.003 |
ons | -0.003 |
Mic | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
envelop | +0.051 |
Offic | +0.050 |
Amen | +0.046 |
ons | +0.046 |
parcels | +0.046 |
Neg logit contributions | |
gan | -0.021 |
tub | -0.021 |
Mil | -0.016 |
Moe | -0.015 |
ahl | -0.015 |
Token | She |
Token ID | 1375 |
Feature activation | -3.02e-03 |
Pos logit contributions | |
tub | +0.026 |
gan | +0.026 |
Moe | +0.020 |
ses | +0.020 |
ahl | +0.020 |
Neg logit contributions | |
envelop | -0.053 |
Offic | -0.050 |
Amen | -0.048 |
ons | -0.047 |
Mic | -0.046 |
Token | had |
Token ID | 550 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
envelop | +0.005 |
Amen | +0.005 |
Offic | +0.005 |
ons | +0.005 |
Mic | +0.004 |
Neg logit contributions | |
tub | -0.006 |
gan | -0.006 |
Moe | -0.005 |
ider | -0.005 |
ahl | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
envelop | +0.037 |
Offic | +0.035 |
Amen | +0.035 |
ons | +0.033 |
Mic | +0.032 |
Neg logit contributions | |
tub | -0.026 |
gan | -0.025 |
Moe | -0.021 |
ider | -0.021 |
ahl | -0.020 |
Token | blue |
Token ID | 4171 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
envelop | +0.097 |
Offic | +0.090 |
Amen | +0.088 |
ons | +0.085 |
Mic | +0.081 |
Neg logit contributions | |
tub | -0.085 |
gan | -0.081 |
Moe | -0.072 |
ider | -0.071 |
ahl | -0.069 |
Token | stuffed |
Token ID | 22259 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
envelop | +0.080 |
Amen | +0.072 |
Offic | +0.071 |
ons | +0.067 |
Mic | +0.063 |
Neg logit contributions | |
tub | -0.100 |
gan | -0.097 |
Moe | -0.088 |
ider | -0.087 |
ahl | -0.086 |
Token | animal |
Token ID | 5044 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
envelop | +0.269 |
Offic | +0.267 |
Amen | +0.256 |
ons | +0.252 |
itivity | +0.244 |
Neg logit contributions | |
tub | -0.204 |
gan | -0.182 |
Moe | -0.163 |
ahl | -0.161 |
ter | -0.159 |
Token | named |
Token ID | 3706 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
envelop | +0.048 |
Offic | +0.042 |
Amen | +0.041 |
ons | +0.039 |
Mic | +0.036 |
Neg logit contributions | |
tub | -0.080 |
gan | -0.077 |
Moe | -0.071 |
ider | -0.070 |
ahl | -0.069 |
Token | Paul |
Token ID | 3362 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
envelop | +0.095 |
Offic | +0.088 |
Amen | +0.085 |
ons | +0.082 |
Mic | +0.080 |
Neg logit contributions | |
tub | -0.081 |
gan | -0.075 |
ider | -0.069 |
Moe | -0.066 |
ahl | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
envelop | +0.077 |
Amen | +0.069 |
Offic | +0.069 |
ons | +0.067 |
oc | +0.064 |
Neg logit contributions | |
tub | -0.067 |
gan | -0.063 |
ider | -0.057 |
Moe | -0.056 |
ahl | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
tub | +0.015 |
gan | +0.013 |
Moe | +0.012 |
ahl | +0.010 |
Mil | +0.010 |
Neg logit contributions | |
envelop | -0.032 |
Offic | -0.029 |
Amen | -0.028 |
ons | -0.028 |
Mic | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
Amen | +0.103 |
Offic | +0.103 |
envelop | +0.100 |
Mic | +0.092 |
oc | +0.092 |
Neg logit contributions | |
tub | -0.083 |
gan | -0.078 |
Moe | -0.072 |
ses | -0.069 |
Mil | -0.067 |
Token | stuffed |
Token ID | 22259 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
envelop | +0.093 |
Amen | +0.092 |
Offic | +0.091 |
ons | +0.085 |
Mic | +0.080 |
Neg logit contributions | |
tub | -0.095 |
gan | -0.084 |
ses | -0.078 |
ider | -0.077 |
Moe | -0.077 |
Token | dog |
Token ID | 3290 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
Amen | +0.269 |
envelop | +0.262 |
Offic | +0.258 |
ons | +0.244 |
Mic | +0.243 |
Neg logit contributions | |
tub | -0.201 |
gan | -0.179 |
ses | -0.178 |
sed | -0.168 |
Moe | -0.168 |
Token | and |
Token ID | 290 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
envelop | +0.121 |
Offic | +0.113 |
Amen | +0.110 |
ons | +0.106 |
Mic | +0.101 |
Neg logit contributions | |
tub | -0.105 |
gan | -0.099 |
Moe | -0.088 |
ider | -0.088 |
ahl | -0.085 |
Token | a |
Token ID | 257 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
envelop | +0.162 |
Amen | +0.155 |
Offic | +0.154 |
ons | +0.145 |
Mic | +0.139 |
Neg logit contributions | |
tub | -0.133 |
gan | -0.127 |
Moe | -0.112 |
ider | -0.109 |
ses | -0.108 |
Token | stuffed |
Token ID | 22259 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
envelop | +0.091 |
Offic | +0.086 |
Amen | +0.085 |
ons | +0.081 |
Mic | +0.076 |
Neg logit contributions | |
tub | -0.095 |
gan | -0.088 |
ider | -0.080 |
Moe | -0.079 |
ses | -0.078 |
Token | cat |
Token ID | 3797 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
envelop | +0.260 |
Amen | +0.251 |
Offic | +0.246 |
ons | +0.242 |
Mic | +0.229 |
Neg logit contributions | |
tub | -0.209 |
gan | -0.192 |
ses | -0.178 |
Moe | -0.177 |
sed | -0.170 |
Token | as |
Token ID | 355 |
Feature activation | +4.03e-03 |
Pos logit contributions | |
envelop | +0.111 |
Offic | +0.105 |
Amen | +0.103 |
ons | +0.099 |
Mic | +0.095 |
Neg logit contributions | |
tub | -0.082 |
gan | -0.077 |
Moe | -0.067 |
ider | -0.067 |
ahl | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
tub | +0.007 |
gan | +0.007 |
ider | +0.006 |
Moe | +0.006 |
ahl | +0.006 |
Neg logit contributions | |
envelop | -0.008 |
Offic | -0.008 |
Amen | -0.008 |
ons | -0.007 |
Mic | -0.007 |
Token | witnesses |
Token ID | 11432 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
tub | +0.082 |
gan | +0.077 |
ider | +0.070 |
Moe | +0.069 |
ses | +0.067 |
Neg logit contributions | |
envelop | -0.094 |
Offic | -0.088 |
Amen | -0.087 |
ons | -0.084 |
Mic | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | +8.84e-03 |
Pos logit contributions | |
envelop | +0.091 |
Offic | +0.085 |
Amen | +0.083 |
ons | +0.080 |
Mic | +0.076 |
Neg logit contributions | |
tub | -0.073 |
gan | -0.069 |
Moe | -0.060 |
ider | -0.060 |
ahl | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
envelop | +0.045 |
Offic | +0.042 |
Amen | +0.042 |
ons | +0.040 |
Mic | +0.038 |
Neg logit contributions | |
tub | -0.035 |
gan | -0.033 |
Moe | -0.029 |
ider | -0.029 |
ahl | -0.028 |
Token | The |
Token ID | 464 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
envelop | +0.033 |
Amen | +0.029 |
Offic | +0.029 |
ons | +0.028 |
Mic | +0.025 |
Neg logit contributions | |
tub | -0.036 |
gan | -0.035 |
Moe | -0.032 |
ider | -0.032 |
Mil | -0.030 |
Token | kid |
Token ID | 5141 |
Feature activation | -4.98e-03 |
Pos logit contributions | |
tub | +0.030 |
gan | +0.029 |
Moe | +0.026 |
ider | +0.026 |
ahl | +0.025 |
Neg logit contributions | |
envelop | -0.029 |
Offic | -0.027 |
Amen | -0.026 |
ons | -0.024 |
Mic | -0.024 |
Token | had |
Token ID | 550 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
envelop | +0.010 |
Offic | +0.009 |
Amen | +0.009 |
ons | +0.009 |
oc | +0.008 |
Neg logit contributions | |
tub | -0.009 |
gan | -0.009 |
Moe | -0.008 |
ider | -0.008 |
ahl | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
envelop | +0.038 |
Amen | +0.038 |
Offic | +0.037 |
ons | +0.035 |
Mic | +0.035 |
Neg logit contributions | |
tub | -0.027 |
gan | -0.026 |
Moe | -0.023 |
Mil | -0.022 |
ahl | -0.022 |
Token | grand |
Token ID | 4490 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
envelop | +0.099 |
Offic | +0.093 |
ons | +0.090 |
Amen | +0.090 |
Mic | +0.082 |
Neg logit contributions | |
tub | -0.098 |
gan | -0.089 |
Moe | -0.082 |
ses | -0.081 |
ider | -0.080 |
Token | old |
Token ID | 1468 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
envelop | +0.189 |
Offic | +0.184 |
Mic | +0.172 |
Amen | +0.169 |
ons | +0.157 |
Neg logit contributions | |
tub | -0.270 |
gan | -0.260 |
ses | -0.251 |
sed | -0.247 |
Moe | -0.244 |
Token | time |
Token ID | 640 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
envelop | +0.067 |
Offic | +0.060 |
Amen | +0.059 |
ons | +0.056 |
oc | +0.052 |
Neg logit contributions | |
tub | -0.082 |
gan | -0.080 |
Moe | -0.072 |
ider | -0.071 |
ahl | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +8.08e-03 |
Pos logit contributions | |
tub | +0.055 |
gan | +0.052 |
ider | +0.047 |
Moe | +0.047 |
ahl | +0.044 |
Neg logit contributions | |
envelop | -0.057 |
Offic | -0.053 |
Amen | -0.052 |
ons | -0.049 |
Mic | -0.048 |
Token | He |
Token ID | 679 |
Feature activation | -3.48e-03 |
Pos logit contributions | |
tub | +0.013 |
gan | +0.013 |
ider | +0.013 |
ahl | +0.011 |
arling | +0.011 |
Neg logit contributions | |
envelop | -0.017 |
Offic | -0.016 |
Amen | -0.015 |
oc | -0.015 |
parcels | -0.014 |
Token | laughed |
Token ID | 13818 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
envelop | +0.007 |
Offic | +0.007 |
Amen | +0.007 |
ons | +0.006 |
Mic | +0.006 |
Neg logit contributions | |
tub | -0.006 |
gan | -0.006 |
Moe | -0.005 |
ider | -0.005 |
ahl | -0.005 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
tub | +0.031 |
gan | +0.029 |
Moe | +0.025 |
ider | +0.025 |
ahl | +0.024 |
Neg logit contributions | |
envelop | -0.047 |
Offic | -0.044 |
Amen | -0.043 |
ons | -0.042 |
Mic | -0.040 |
Token | Once |
Token ID | 7454 |
Feature activation | +6.83e-02 |
Pos logit contributions | |
gan | +0.050 |
tub | +0.045 |
ider | +0.045 |
ses | +0.045 |
sed | +0.044 |
Neg logit contributions | |
envelop | -0.055 |
ons | -0.053 |
Amen | -0.052 |
Offic | -0.052 |
oc | -0.048 |
Token | upon |
Token ID | 2402 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
ider | +0.105 |
tub | +0.102 |
gan | +0.097 |
ses | +0.094 |
sed | +0.088 |
Neg logit contributions | |
ons | -0.145 |
envelop | -0.143 |
Amen | -0.130 |
pa | -0.127 |
Offic | -0.124 |
Token | a |
Token ID | 257 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
tub | +0.082 |
gan | +0.077 |
Moe | +0.067 |
ider | +0.066 |
ahl | +0.063 |
Neg logit contributions | |
envelop | -0.124 |
Offic | -0.116 |
Amen | -0.113 |
ons | -0.110 |
Mic | -0.105 |
Token | time |
Token ID | 640 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
envelop | +0.029 |
Offic | +0.027 |
Amen | +0.026 |
ons | +0.025 |
Mic | +0.023 |
Neg logit contributions | |
tub | -0.025 |
gan | -0.024 |
Moe | -0.021 |
ider | -0.021 |
ahl | -0.020 |
Token | there |
Token ID | 612 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
tub | +0.185 |
gan | +0.173 |
ipples | +0.171 |
ider | +0.168 |
ses | +0.166 |
Neg logit contributions | |
ons | -0.107 |
envelop | -0.095 |
Amen | -0.093 |
oc | -0.087 |
lending | -0.084 |
Token | was |
Token ID | 373 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
ider | +0.213 |
tub | +0.187 |
gan | +0.184 |
ses | +0.181 |
arling | +0.175 |
Neg logit contributions | |
ons | -0.130 |
oc | -0.124 |
envelop | -0.121 |
read | -0.113 |
Amen | -0.110 |
Token | a |
Token ID | 257 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
tub | +0.051 |
gan | +0.048 |
ider | +0.042 |
Moe | +0.042 |
ahl | +0.040 |
Neg logit contributions | |
envelop | -0.066 |
Offic | -0.061 |
Amen | -0.060 |
ons | -0.058 |
Mic | -0.056 |
Token | yellow |
Token ID | 7872 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
envelop | +0.079 |
Offic | +0.074 |
Amen | +0.072 |
ons | +0.070 |
Mic | +0.067 |
Neg logit contributions | |
tub | -0.053 |
gan | -0.049 |
Moe | -0.043 |
ider | -0.043 |
ahl | -0.041 |
Token | bear |
Token ID | 6842 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
envelop | +0.065 |
Amen | +0.059 |
Offic | +0.058 |
ons | +0.055 |
oc | +0.053 |
Neg logit contributions | |
tub | -0.065 |
gan | -0.063 |
ider | -0.057 |
Moe | -0.057 |
ahl | -0.056 |
Token | who |
Token ID | 508 |
Feature activation | -8.26e-03 |
Pos logit contributions | |
envelop | +0.038 |
Offic | +0.034 |
Amen | +0.032 |
ons | +0.030 |
Mic | +0.029 |
Neg logit contributions | |
tub | -0.062 |
gan | -0.059 |
ider | -0.054 |
Moe | -0.054 |
Mil | -0.052 |
Token | was |
Token ID | 373 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
envelop | +0.114 |
Offic | +0.105 |
Amen | +0.103 |
ons | +0.098 |
Mic | +0.094 |
Neg logit contributions | |
tub | -0.116 |
gan | -0.111 |
Moe | -0.099 |
ider | -0.098 |
ahl | -0.095 |
Token | very |
Token ID | 845 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
tub | +0.027 |
gan | +0.026 |
Moe | +0.023 |
ider | +0.023 |
Mil | +0.022 |
Neg logit contributions | |
envelop | -0.031 |
Offic | -0.029 |
Amen | -0.028 |
ons | -0.027 |
Mic | -0.026 |
Token | careful |
Token ID | 8161 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
envelop | +0.041 |
Offic | +0.037 |
Amen | +0.037 |
ons | +0.035 |
Mic | +0.033 |
Neg logit contributions | |
tub | -0.045 |
gan | -0.043 |
Moe | -0.039 |
ider | -0.039 |
ahl | -0.038 |
Token | to |
Token ID | 284 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
tub | +0.030 |
gan | +0.028 |
ahl | +0.025 |
Moe | +0.025 |
ider | +0.025 |
Neg logit contributions | |
envelop | -0.033 |
Amen | -0.030 |
ons | -0.029 |
Offic | -0.029 |
lending | -0.026 |
Token | do |
Token ID | 466 |
Feature activation | +5.07e-03 |
Pos logit contributions | |
envelop | +0.065 |
Offic | +0.059 |
Amen | +0.058 |
ons | +0.055 |
Mic | +0.052 |
Neg logit contributions | |
tub | -0.077 |
gan | -0.073 |
ider | -0.067 |
Moe | -0.066 |
ahl | -0.065 |
Token | it |
Token ID | 340 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
tub | +0.011 |
gan | +0.010 |
ider | +0.009 |
Moe | +0.009 |
ses | +0.009 |
Neg logit contributions | |
envelop | -0.009 |
ons | -0.008 |
Offic | -0.008 |
Amen | -0.008 |
oc | -0.007 |
Token | just |
Token ID | 655 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
tub | +0.166 |
gan | +0.164 |
ses | +0.155 |
sed | +0.150 |
ahl | +0.144 |
Neg logit contributions | |
envelop | -0.142 |
ons | -0.129 |
Offic | -0.118 |
read | -0.118 |
lending | -0.115 |
Token | like |
Token ID | 588 |
Feature activation | +3.85e-03 |
Pos logit contributions | |
tub | +0.072 |
gan | +0.064 |
ider | +0.058 |
ahl | +0.057 |
Moe | +0.057 |
Neg logit contributions | |
envelop | -0.073 |
Offic | -0.063 |
ons | -0.063 |
read | -0.062 |
Amen | -0.060 |
Token | her |
Token ID | 607 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
tub | +0.007 |
gan | +0.007 |
Moe | +0.006 |
ider | +0.006 |
ahl | +0.006 |
Neg logit contributions | |
envelop | -0.007 |
Offic | -0.007 |
Amen | -0.007 |
ons | -0.006 |
Mic | -0.006 |
Token | mummy |
Token ID | 38567 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
tub | +0.060 |
gan | +0.058 |
ahl | +0.052 |
ider | +0.051 |
Moe | +0.051 |
Neg logit contributions | |
envelop | -0.064 |
Amen | -0.059 |
ons | -0.055 |
Offic | -0.054 |
oc | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
envelop | +0.085 |
Offic | +0.078 |
Amen | +0.077 |
ons | +0.074 |
Mic | +0.070 |
Neg logit contributions | |
tub | -0.077 |
gan | -0.073 |
Moe | -0.065 |
ider | -0.065 |
ahl | -0.063 |
Token | day |
Token ID | 1110 |
Feature activation | +6.09e-02 |
Pos logit contributions | |
envelop | +0.004 |
Amen | +0.003 |
ons | +0.003 |
Offic | +0.003 |
oc | +0.003 |
Neg logit contributions | |
tub | -0.005 |
gan | -0.005 |
ses | -0.005 |
ider | -0.005 |
sed | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
tub | +0.116 |
gan | +0.107 |
ses | +0.106 |
ider | +0.101 |
fault | +0.099 |
Neg logit contributions | |
envelop | -0.120 |
ons | -0.106 |
Offic | -0.103 |
Amen | -0.095 |
oc | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
tub | +0.075 |
gan | +0.068 |
ider | +0.066 |
ses | +0.065 |
ahl | +0.063 |
Neg logit contributions | |
envelop | -0.079 |
Offic | -0.072 |
ons | -0.067 |
Amen | -0.066 |
Mic | -0.064 |
Token | boy |
Token ID | 2933 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
tub | +0.079 |
gan | +0.079 |
ahl | +0.070 |
Moe | +0.069 |
ider | +0.068 |
Neg logit contributions | |
envelop | -0.083 |
Offic | -0.075 |
Amen | -0.074 |
ons | -0.070 |
Mic | -0.069 |
Token | walked |
Token ID | 6807 |
Feature activation | +2.67e-03 |
Pos logit contributions | |
tub | +0.029 |
gan | +0.028 |
ider | +0.026 |
Moe | +0.025 |
ahl | +0.025 |
Neg logit contributions | |
envelop | -0.026 |
Offic | -0.024 |
ons | -0.022 |
Amen | -0.022 |
oc | -0.022 |
Token | down |
Token ID | 866 |
Feature activation | +6.91e-02 |
Pos logit contributions | |
tub | +0.005 |
gan | +0.005 |
ses | +0.005 |
ider | +0.004 |
sed | +0.004 |
Neg logit contributions | |
envelop | -0.005 |
Offic | -0.004 |
Amen | -0.004 |
ons | -0.004 |
saddle | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +7.09e-02 |
Pos logit contributions | |
gan | +0.148 |
tub | +0.147 |
ider | +0.135 |
ses | +0.128 |
ahl | +0.126 |
Neg logit contributions | |
envelop | -0.118 |
Offic | -0.107 |
Amen | -0.098 |
ons | -0.096 |
read | -0.089 |
Token | street |
Token ID | 4675 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
gan | +0.154 |
tub | +0.147 |
ider | +0.138 |
Moe | +0.135 |
ahl | +0.135 |
Neg logit contributions | |
envelop | -0.115 |
Offic | -0.107 |
Mic | -0.097 |
Amen | -0.097 |
read | -0.093 |
Token | and |
Token ID | 290 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
tub | +0.085 |
gan | +0.085 |
ahl | +0.079 |
ses | +0.077 |
ider | +0.076 |
Neg logit contributions | |
envelop | -0.078 |
Offic | -0.070 |
ons | -0.065 |
Amen | -0.063 |
read | -0.062 |
Token | stopped |
Token ID | 5025 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
envelop | +0.082 |
Offic | +0.077 |
read | +0.069 |
ons | +0.066 |
Amen | +0.065 |
Neg logit contributions | |
tub | -0.113 |
gan | -0.107 |
ider | -0.106 |
fault | -0.102 |
ses | -0.101 |
Token | in |
Token ID | 287 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
tub | +0.095 |
gan | +0.093 |
ider | +0.084 |
ahl | +0.082 |
Moe | +0.081 |
Neg logit contributions | |
envelop | -0.081 |
Offic | -0.071 |
Amen | -0.071 |
ons | -0.068 |
lending | -0.064 |
Token | her |
Token ID | 607 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
Amen | +0.040 |
Offic | +0.039 |
oc | +0.038 |
envelop | +0.038 |
Mic | +0.037 |
Neg logit contributions | |
tub | -0.028 |
Moe | -0.025 |
gan | -0.025 |
Mil | -0.023 |
ahl | -0.021 |
Token | pocket |
Token ID | 10000 |
Feature activation | +2.29e-03 |
Pos logit contributions | |
tub | +0.061 |
gan | +0.057 |
Moe | +0.051 |
ider | +0.051 |
ahl | +0.049 |
Neg logit contributions | |
envelop | -0.069 |
Offic | -0.065 |
Amen | -0.064 |
ons | -0.060 |
Mic | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +7.92e-03 |
Pos logit contributions | |
tub | +0.004 |
gan | +0.004 |
ider | +0.003 |
Moe | +0.003 |
ahl | +0.003 |
Neg logit contributions | |
envelop | -0.005 |
Offic | -0.005 |
Amen | -0.005 |
ons | -0.004 |
Mic | -0.004 |
Token | She |
Token ID | 1375 |
Feature activation | -4.74e-03 |
Pos logit contributions | |
tub | +0.012 |
gan | +0.012 |
ider | +0.011 |
ses | +0.011 |
ahl | +0.011 |
Neg logit contributions | |
envelop | -0.017 |
Amen | -0.016 |
Offic | -0.015 |
ons | -0.015 |
parcels | -0.014 |
Token | put |
Token ID | 1234 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
envelop | +0.010 |
Amen | +0.010 |
Offic | +0.010 |
Mic | +0.009 |
ons | +0.009 |
Neg logit contributions | |
tub | -0.008 |
gan | -0.007 |
Moe | -0.006 |
Mil | -0.006 |
ider | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | +6.46e-02 |
Pos logit contributions | |
Offic | +0.081 |
envelop | +0.081 |
Amen | +0.079 |
Mic | +0.072 |
oc | +0.070 |
Neg logit contributions | |
tub | -0.081 |
gan | -0.078 |
Moe | -0.072 |
Mil | -0.068 |
ahl | -0.068 |
Token | in |
Token ID | 287 |
Feature activation | -1.04e-03 |
Pos logit contributions | |
tub | +0.118 |
gan | +0.112 |
Moe | +0.102 |
ider | +0.100 |
ses | +0.100 |
Neg logit contributions | |
envelop | -0.129 |
Offic | -0.114 |
ons | -0.113 |
Amen | -0.111 |
Mic | -0.104 |
Token | a |
Token ID | 257 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
Offic | +0.002 |
Amen | +0.002 |
oc | +0.002 |
ons | +0.002 |
Mic | +0.002 |
Neg logit contributions | |
tub | -0.002 |
Moe | -0.001 |
gan | -0.001 |
Mil | -0.001 |
ahl | -0.001 |
Token | special |
Token ID | 2041 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
Offic | +0.080 |
Amen | +0.080 |
envelop | +0.078 |
ons | +0.076 |
Mic | +0.072 |
Neg logit contributions | |
tub | -0.056 |
gan | -0.045 |
Moe | -0.044 |
ahl | -0.041 |
ider | -0.040 |
Token | box |
Token ID | 3091 |
Feature activation | -3.44e-03 |
Pos logit contributions | |
Offic | +0.077 |
envelop | +0.077 |
Amen | +0.074 |
ons | +0.072 |
Mic | +0.072 |
Neg logit contributions | |
tub | -0.060 |
gan | -0.051 |
Moe | -0.048 |
ider | -0.045 |
ahl | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
envelop | +0.007 |
Offic | +0.006 |
Amen | +0.006 |
ons | +0.006 |
Mic | +0.006 |
Neg logit contributions | |
tub | -0.006 |
gan | -0.006 |
Moe | -0.005 |
ider | -0.005 |
ahl | -0.005 |
Token | y |
Token ID | 331 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
envelop | +0.028 |
Offic | +0.025 |
Amen | +0.024 |
ons | +0.023 |
Mic | +0.022 |
Neg logit contributions | |
tub | -0.036 |
gan | -0.035 |
Moe | -0.032 |
ider | -0.031 |
ahl | -0.031 |
Token | ummy |
Token ID | 13513 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
envelop | +0.102 |
ons | +0.097 |
Offic | +0.093 |
Amen | +0.092 |
Mic | +0.091 |
Neg logit contributions | |
tub | -0.074 |
gan | -0.067 |
ses | -0.061 |
ider | -0.058 |
ahl | -0.058 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
envelop | +0.044 |
Offic | +0.040 |
Amen | +0.038 |
ons | +0.038 |
oc | +0.035 |
Neg logit contributions | |
tub | -0.055 |
gan | -0.052 |
ider | -0.047 |
ahl | -0.046 |
Moe | -0.046 |
Token | she |
Token ID | 673 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
tub | +0.021 |
gan | +0.020 |
Moe | +0.017 |
ider | +0.016 |
Mil | +0.016 |
Neg logit contributions | |
envelop | -0.027 |
Offic | -0.026 |
Amen | -0.025 |
ons | -0.023 |
Mic | -0.023 |
Token | ate |
Token ID | 15063 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
tub | +0.024 |
gan | +0.023 |
Moe | +0.020 |
ider | +0.020 |
Mil | +0.019 |
Neg logit contributions | |
envelop | -0.028 |
Offic | -0.026 |
Amen | -0.025 |
ons | -0.024 |
Mic | -0.023 |
Token | it |
Token ID | 340 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
envelop | +0.040 |
Offic | +0.038 |
Amen | +0.037 |
ons | +0.034 |
Mic | +0.033 |
Neg logit contributions | |
tub | -0.045 |
gan | -0.044 |
Moe | -0.040 |
ider | -0.038 |
ahl | -0.038 |
Token | all |
Token ID | 477 |
Feature activation | -9.73e-03 |
Pos logit contributions | |
tub | +0.156 |
gan | +0.148 |
ses | +0.136 |
ider | +0.136 |
Moe | +0.133 |
Neg logit contributions | |
envelop | -0.132 |
ons | -0.116 |
Offic | -0.115 |
Amen | -0.113 |
oc | -0.104 |
Token | up |
Token ID | 510 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
envelop | +0.016 |
ons | +0.016 |
Amen | +0.015 |
read | +0.014 |
Offic | +0.013 |
Neg logit contributions | |
tub | -0.020 |
ses | -0.019 |
ider | -0.019 |
sed | -0.018 |
gan | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
tub | +0.092 |
gan | +0.086 |
ider | +0.078 |
Moe | +0.077 |
ahl | +0.075 |
Neg logit contributions | |
envelop | -0.100 |
Offic | -0.091 |
Amen | -0.089 |
ons | -0.087 |
Mic | -0.081 |
Token |
|
Token ID | 198 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
tub | +0.033 |
gan | +0.030 |
ider | +0.029 |
Moe | +0.027 |
ses | +0.026 |
Neg logit contributions | |
envelop | -0.036 |
Offic | -0.034 |
Amen | -0.034 |
ons | -0.032 |
Mic | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
envelop | +0.042 |
Offic | +0.038 |
Amen | +0.037 |
ons | +0.035 |
parcels | +0.035 |
Neg logit contributions | |
tub | -0.023 |
gan | -0.021 |
Moe | -0.019 |
ahl | -0.018 |
Mil | -0.017 |
Token | She |
Token ID | 1375 |
Feature activation | -9.33e-03 |
Pos logit contributions | |
tub | +0.025 |
ider | +0.023 |
gan | +0.022 |
ahl | +0.020 |
flies | +0.020 |
Neg logit contributions | |
envelop | -0.022 |
Offic | -0.021 |
Amen | -0.020 |
Mic | -0.018 |
ons | -0.018 |
Token | does |
Token ID | 857 |
Feature activation | +4.62e-03 |
Pos logit contributions | |
envelop | +0.019 |
Offic | +0.018 |
Amen | +0.017 |
ons | +0.017 |
Mic | +0.016 |
Neg logit contributions | |
tub | -0.017 |
gan | -0.016 |
Moe | -0.014 |
ider | -0.014 |
ahl | -0.014 |
Token | not |
Token ID | 407 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
tub | +0.005 |
gan | +0.005 |
Moe | +0.004 |
ider | +0.004 |
Mil | +0.004 |
Neg logit contributions | |
envelop | -0.012 |
Offic | -0.011 |
Amen | -0.011 |
ons | -0.011 |
Mic | -0.011 |
Token | know |
Token ID | 760 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
envelop | +0.020 |
Offic | +0.019 |
Amen | +0.018 |
ons | +0.018 |
Mic | +0.016 |
Neg logit contributions | |
tub | -0.026 |
gan | -0.025 |
Moe | -0.022 |
ider | -0.022 |
ahl | -0.022 |
Token | what |
Token ID | 644 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
envelop | +0.042 |
ons | +0.037 |
Offic | +0.036 |
Amen | +0.033 |
read | +0.030 |
Neg logit contributions | |
tub | -0.070 |
gan | -0.061 |
ider | -0.060 |
ahl | -0.059 |
ses | -0.058 |
Token | she |
Token ID | 673 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
tub | +0.036 |
gan | +0.034 |
ider | +0.031 |
Moe | +0.030 |
ahl | +0.030 |
Neg logit contributions | |
envelop | -0.036 |
Offic | -0.032 |
ons | -0.031 |
Amen | -0.031 |
parcels | -0.028 |
Token | did |
Token ID | 750 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
tub | +0.062 |
gan | +0.058 |
Moe | +0.050 |
ider | +0.050 |
ahl | +0.047 |
Neg logit contributions | |
envelop | -0.099 |
Offic | -0.094 |
Amen | -0.092 |
ons | -0.089 |
oc | -0.085 |
Token | wrong |
Token ID | 2642 |
Feature activation | +8.14e-02 |
Pos logit contributions | |
tub | +0.055 |
gan | +0.052 |
Moe | +0.045 |
ider | +0.045 |
ahl | +0.043 |
Neg logit contributions | |
envelop | -0.078 |
Offic | -0.073 |
Amen | -0.072 |
ons | -0.069 |
Mic | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
tub | +0.155 |
gan | +0.148 |
ahl | +0.133 |
Moe | +0.132 |
ider | +0.130 |
Neg logit contributions | |
envelop | -0.159 |
Amen | -0.137 |
Offic | -0.137 |
ons | -0.135 |
oc | -0.127 |
Token | She |
Token ID | 1375 |
Feature activation | -8.23e-03 |
Pos logit contributions | |
tub | +0.034 |
ider | +0.029 |
gan | +0.028 |
flies | +0.027 |
ahl | +0.026 |
Neg logit contributions | |
envelop | -0.028 |
Offic | -0.028 |
Amen | -0.027 |
Mic | -0.023 |
ons | -0.023 |
Token | pulls |
Token ID | 16194 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
envelop | +0.017 |
Offic | +0.016 |
Amen | +0.015 |
ons | +0.015 |
oc | +0.014 |
Neg logit contributions | |
tub | -0.015 |
gan | -0.014 |
ider | -0.013 |
Moe | -0.013 |
ahl | -0.012 |
Token | every |
Token ID | 790 |
Feature activation | -3.92e-03 |
Pos logit contributions | |
envelop | +0.014 |
Offic | +0.013 |
Amen | +0.013 |
ons | +0.012 |
Mic | +0.012 |
Neg logit contributions | |
tub | -0.013 |
gan | -0.013 |
Moe | -0.011 |
ider | -0.011 |
ahl | -0.011 |
Token | hot |
Token ID | 3024 |
Feature activation | -4.95e-03 |
Pos logit contributions | |
envelop | +0.009 |
Offic | +0.009 |
Amen | +0.009 |
Mic | +0.008 |
ons | +0.008 |
Neg logit contributions | |
tub | -0.006 |
gan | -0.005 |
ider | -0.004 |
Moe | -0.004 |
ses | -0.004 |
Token | day |
Token ID | 1110 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
envelop | +0.010 |
Offic | +0.009 |
Amen | +0.009 |
ons | +0.008 |
Mic | +0.008 |
Neg logit contributions | |
tub | -0.009 |
gan | -0.009 |
Moe | -0.008 |
ider | -0.008 |
ahl | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
tub | +0.076 |
gan | +0.072 |
Moe | +0.063 |
ider | +0.063 |
ahl | +0.060 |
Neg logit contributions | |
envelop | -0.104 |
Offic | -0.098 |
Amen | -0.096 |
ons | -0.092 |
Mic | -0.088 |
Token | had |
Token ID | 550 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
envelop | +0.122 |
Offic | +0.115 |
Amen | +0.115 |
ons | +0.107 |
Mic | +0.105 |
Neg logit contributions | |
tub | -0.107 |
gan | -0.101 |
Moe | -0.091 |
Mil | -0.086 |
ahl | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
envelop | +0.025 |
Amen | +0.025 |
Offic | +0.025 |
oc | +0.022 |
ons | +0.022 |
Neg logit contributions | |
tub | -0.026 |
gan | -0.025 |
Moe | -0.022 |
Mil | -0.021 |
ahl | -0.021 |
Token | best |
Token ID | 1266 |
Feature activation | -7.24e-03 |
Pos logit contributions | |
tub | +0.030 |
gan | +0.028 |
ider | +0.024 |
Moe | +0.024 |
ahl | +0.022 |
Neg logit contributions | |
envelop | -0.066 |
Offic | -0.060 |
Amen | -0.060 |
ons | -0.059 |
Mic | -0.056 |
Token | time |
Token ID | 640 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
envelop | +0.015 |
Offic | +0.015 |
Amen | +0.014 |
Mic | +0.013 |
ons | +0.012 |
Neg logit contributions | |
tub | -0.013 |
gan | -0.012 |
Moe | -0.011 |
ider | -0.011 |
ses | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +9.50e-03 |
Pos logit contributions | |
tub | +0.064 |
gan | +0.061 |
Moe | +0.053 |
ider | +0.053 |
Mil | +0.050 |
Neg logit contributions | |
envelop | -0.089 |
Amen | -0.085 |
Offic | -0.085 |
Mic | -0.078 |
ons | -0.077 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
tub | +0.015 |
gan | +0.014 |
ider | +0.013 |
Moe | +0.012 |
ahl | +0.012 |
Neg logit contributions | |
envelop | -0.022 |
Offic | -0.020 |
Amen | -0.020 |
ons | -0.019 |
Mic | -0.018 |
Token | Once |
Token ID | 7454 |
Feature activation | +6.92e-02 |
Pos logit contributions | |
tub | +0.044 |
gan | +0.044 |
ider | +0.040 |
ses | +0.038 |
sed | +0.037 |
Neg logit contributions | |
envelop | -0.049 |
Offic | -0.045 |
Amen | -0.045 |
ons | -0.044 |
oc | -0.041 |
Token | happy |
Token ID | 3772 |
Feature activation | +4.22e-04 |
Pos logit contributions | |
envelop | +0.117 |
Offic | +0.105 |
Amen | +0.104 |
ons | +0.104 |
Mic | +0.096 |
Neg logit contributions | |
tub | -0.112 |
gan | -0.110 |
Moe | -0.098 |
Mil | -0.093 |
ider | -0.093 |
Token | to |
Token ID | 284 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
tub | +0.001 |
gan | +0.001 |
ahl | +0.001 |
ider | +0.001 |
ses | +0.001 |
Neg logit contributions | |
envelop | -0.001 |
ons | -0.001 |
Offic | -0.001 |
Amen | -0.001 |
lending | -0.001 |
Token | hear |
Token ID | 3285 |
Feature activation | -1.11e-03 |
Pos logit contributions | |
envelop | +0.086 |
Offic | +0.078 |
ons | +0.077 |
Amen | +0.075 |
Mic | +0.070 |
Neg logit contributions | |
tub | -0.084 |
gan | -0.083 |
Moe | -0.075 |
ahl | -0.072 |
Mil | -0.072 |
Token | from |
Token ID | 422 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
envelop | +0.002 |
Offic | +0.002 |
Amen | +0.002 |
Mic | +0.002 |
ons | +0.002 |
Neg logit contributions | |
tub | -0.002 |
gan | -0.002 |
Moe | -0.001 |
Mil | -0.001 |
der | -0.001 |
Token | him |
Token ID | 683 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
envelop | +0.052 |
ons | +0.050 |
Offic | +0.049 |
Amen | +0.048 |
oc | +0.046 |
Neg logit contributions | |
gan | -0.046 |
tub | -0.046 |
Moe | -0.044 |
ses | -0.041 |
Mil | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
tub | +0.041 |
gan | +0.038 |
ider | +0.033 |
Moe | +0.033 |
ahl | +0.031 |
Neg logit contributions | |
envelop | -0.067 |
Offic | -0.064 |
Amen | -0.063 |
ons | -0.061 |
Mic | -0.059 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
tub | +0.015 |
gan | +0.014 |
ider | +0.013 |
Moe | +0.012 |
ses | +0.012 |
Neg logit contributions | |
envelop | -0.024 |
Offic | -0.022 |
Amen | -0.022 |
ons | -0.020 |
Mic | -0.020 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
tub | +0.027 |
gan | +0.026 |
Moe | +0.023 |
ider | +0.023 |
ahl | +0.022 |
Neg logit contributions | |
envelop | -0.028 |
Offic | -0.025 |
Amen | -0.025 |
ons | -0.024 |
Mic | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -5.86e-03 |
Pos logit contributions | |
envelop | +0.035 |
Offic | +0.032 |
Amen | +0.032 |
ons | +0.029 |
Mic | +0.028 |
Neg logit contributions | |
tub | -0.041 |
gan | -0.039 |
ider | -0.035 |
Moe | -0.035 |
ahl | -0.035 |
Token | knowing |
Token ID | 6970 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
envelop | +0.013 |
ons | +0.012 |
Amen | +0.012 |
Offic | +0.011 |
igh | +0.011 |
Neg logit contributions | |
gan | -0.009 |
tub | -0.009 |
der | -0.007 |
sed | -0.007 |
Moe | -0.007 |
Token | that |
Token ID | 326 |
Feature activation | +3.25e-03 |
Pos logit contributions | |
envelop | +0.049 |
Amen | +0.046 |
Offic | +0.046 |
ons | +0.044 |
oc | +0.043 |
Neg logit contributions | |
tub | -0.023 |
gan | -0.023 |
Moe | -0.021 |
Mil | -0.018 |
der | -0.018 |
Token | like |
Token ID | 588 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
envelop | +0.036 |
Offic | +0.033 |
Amen | +0.033 |
ons | +0.031 |
Mic | +0.029 |
Neg logit contributions | |
tub | -0.033 |
gan | -0.031 |
ider | -0.028 |
Moe | -0.028 |
ahl | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
envelop | +0.058 |
Offic | +0.056 |
Amen | +0.056 |
oc | +0.053 |
ons | +0.052 |
Neg logit contributions | |
tub | -0.034 |
gan | -0.030 |
Moe | -0.026 |
Mil | -0.025 |
ses | -0.025 |
Token | play |
Token ID | 711 |
Feature activation | +6.46e-03 |
Pos logit contributions | |
envelop | +0.052 |
Offic | +0.049 |
Amen | +0.046 |
ons | +0.043 |
Mic | +0.042 |
Neg logit contributions | |
tub | -0.068 |
gan | -0.064 |
Moe | -0.059 |
ider | -0.057 |
Mil | -0.056 |
Token | in |
Token ID | 287 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
tub | +0.012 |
gan | +0.011 |
Moe | +0.010 |
ider | +0.010 |
ahl | +0.010 |
Neg logit contributions | |
envelop | -0.013 |
Offic | -0.012 |
Amen | -0.012 |
ons | -0.011 |
Mic | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
tub | +0.037 |
gan | +0.032 |
Moe | +0.031 |
Mil | +0.029 |
ahl | +0.028 |
Neg logit contributions | |
envelop | -0.047 |
Offic | -0.046 |
Amen | -0.045 |
Mic | -0.042 |
ons | -0.042 |
Token | park |
Token ID | 3952 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
gan | +0.077 |
tub | +0.076 |
ider | +0.070 |
Moe | +0.066 |
ses | +0.065 |
Neg logit contributions | |
envelop | -0.082 |
Offic | -0.074 |
Amen | -0.071 |
ons | -0.070 |
Mic | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
tub | +0.085 |
gan | +0.081 |
Moe | +0.073 |
ahl | +0.073 |
ider | +0.071 |
Neg logit contributions | |
envelop | -0.097 |
Offic | -0.088 |
Amen | -0.086 |
ons | -0.085 |
oc | -0.081 |
Token | One |
Token ID | 1881 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
tub | +0.036 |
ider | +0.033 |
gan | +0.033 |
ahl | +0.031 |
ter | +0.030 |
Neg logit contributions | |
envelop | -0.029 |
Amen | -0.026 |
Offic | -0.025 |
oc | -0.023 |
ons | -0.022 |
Token | day |
Token ID | 1110 |
Feature activation | +6.81e-02 |
Pos logit contributions | |
ider | +0.036 |
tub | +0.035 |
gan | +0.034 |
der | +0.033 |
sed | +0.033 |
Neg logit contributions | |
envelop | -0.019 |
read | -0.018 |
ons | -0.018 |
Offic | -0.017 |
oc | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
tub | +0.144 |
ses | +0.131 |
ahl | +0.123 |
ider | +0.121 |
sing | +0.120 |
Neg logit contributions | |
envelop | -0.109 |
ons | -0.102 |
read | -0.087 |
Amen | -0.087 |
exchange | -0.086 |
Token | they |
Token ID | 484 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
tub | +0.093 |
ider | +0.085 |
ses | +0.083 |
ahl | +0.081 |
sed | +0.076 |
Neg logit contributions | |
envelop | -0.086 |
ons | -0.072 |
Offic | -0.070 |
Amen | -0.070 |
read | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
tub | +0.020 |
gan | +0.019 |
ider | +0.017 |
ahl | +0.016 |
Moe | +0.016 |
Neg logit contributions | |
envelop | -0.024 |
Offic | -0.023 |
ons | -0.022 |
Amen | -0.021 |
oc | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
envelop | +0.046 |
Offic | +0.044 |
Amen | +0.042 |
ons | +0.041 |
Mic | +0.039 |
Neg logit contributions | |
tub | -0.038 |
gan | -0.036 |
ider | -0.032 |
Moe | -0.031 |
ses | -0.030 |
Token | Tom |
Token ID | 13787 |
Feature activation | -3.56e-03 |
Pos logit contributions | |
envelop | +0.034 |
Amen | +0.031 |
Offic | +0.029 |
ons | +0.028 |
parcels | +0.027 |
Neg logit contributions | |
tub | -0.034 |
gan | -0.034 |
Moe | -0.031 |
ider | -0.030 |
ahl | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -7.02e-02 |
Pos logit contributions | |
envelop | +0.007 |
Offic | +0.007 |
ons | +0.007 |
Amen | +0.007 |
oc | +0.006 |
Neg logit contributions | |
tub | -0.006 |
gan | -0.006 |
ider | -0.005 |
Moe | -0.005 |
ahl | -0.005 |
Token | Mia |
Token ID | 32189 |
Feature activation | -4.62e-03 |
Pos logit contributions | |
envelop | +0.192 |
Offic | +0.184 |
Amen | +0.176 |
ons | +0.175 |
oc | +0.166 |
Neg logit contributions | |
tub | -0.084 |
gan | -0.069 |
ider | -0.063 |
Moe | -0.055 |
ahl | -0.051 |
Token | get |
Token ID | 651 |
Feature activation | +3.72e-03 |
Pos logit contributions | |
envelop | +0.009 |
Offic | +0.008 |
Amen | +0.008 |
ons | +0.008 |
oc | +0.008 |
Neg logit contributions | |
tub | -0.009 |
gan | -0.008 |
ider | -0.007 |
Moe | -0.007 |
ahl | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
tub | +0.006 |
gan | +0.006 |
Moe | +0.005 |
Mil | +0.005 |
ahl | +0.004 |
Neg logit contributions | |
Offic | -0.008 |
envelop | -0.008 |
Amen | -0.008 |
Mic | -0.008 |
ons | -0.007 |
Token | je |
Token ID | 11223 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
tub | +0.069 |
gan | +0.063 |
ses | +0.056 |
ider | +0.055 |
ahl | +0.055 |
Neg logit contributions | |
envelop | -0.089 |
Amen | -0.087 |
Offic | -0.087 |
ons | -0.081 |
Mic | -0.079 |
Token | ep |
Token ID | 538 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
Amen | +0.048 |
envelop | +0.046 |
Offic | +0.041 |
Mic | +0.038 |
lending | +0.038 |
Neg logit contributions | |
tub | -0.095 |
gan | -0.090 |
ider | -0.089 |
anger | -0.086 |
ahl | -0.085 |
Token | back |
Token ID | 736 |
Feature activation | +5.08e-02 |
Pos logit contributions | |
envelop | +0.032 |
Offic | +0.030 |
Amen | +0.029 |
ons | +0.028 |
oc | +0.027 |
Neg logit contributions | |
tub | -0.026 |
gan | -0.025 |
Moe | -0.022 |
ider | -0.022 |
ahl | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +6.94e-03 |
Pos logit contributions | |
tub | +0.085 |
gan | +0.081 |
Moe | +0.072 |
ider | +0.071 |
ahl | +0.069 |
Neg logit contributions | |
envelop | -0.111 |
Offic | -0.101 |
Amen | -0.098 |
ons | -0.098 |
oc | -0.091 |
Token | took |
Token ID | 1718 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
envelop | +0.094 |
Offic | +0.086 |
Amen | +0.086 |
ons | +0.081 |
Mic | +0.078 |
Neg logit contributions | |
tub | -0.088 |
gan | -0.085 |
Moe | -0.075 |
ider | -0.074 |
ahl | -0.072 |
Token | out |
Token ID | 503 |
Feature activation | +7.79e-03 |
Pos logit contributions | |
Offic | +0.122 |
envelop | +0.122 |
Amen | +0.122 |
ons | +0.112 |
oc | +0.110 |
Neg logit contributions | |
tub | -0.083 |
gan | -0.081 |
Moe | -0.074 |
Mil | -0.068 |
ahl | -0.068 |
Token | two |
Token ID | 734 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
tub | +0.012 |
gan | +0.012 |
Moe | +0.011 |
Mil | +0.010 |
ahl | +0.010 |
Neg logit contributions | |
Amen | -0.017 |
envelop | -0.017 |
Offic | -0.017 |
oc | -0.016 |
Mic | -0.015 |
Token | past |
Token ID | 1613 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
Amen | +0.110 |
Offic | +0.106 |
Mic | +0.096 |
envelop | +0.092 |
oc | +0.091 |
Neg logit contributions | |
tub | -0.081 |
gan | -0.073 |
Moe | -0.071 |
ider | -0.066 |
ses | -0.066 |
Token | els |
Token ID | 1424 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
envelop | +0.049 |
Offic | +0.046 |
Amen | +0.045 |
ons | +0.040 |
Mic | +0.039 |
Neg logit contributions | |
tub | -0.082 |
gan | -0.079 |
Moe | -0.073 |
ider | -0.071 |
ahl | -0.070 |
Token | from |
Token ID | 422 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
envelop | +0.080 |
Offic | +0.073 |
Amen | +0.071 |
ons | +0.070 |
Mic | +0.064 |
Neg logit contributions | |
tub | -0.079 |
gan | -0.075 |
ider | -0.067 |
Moe | -0.067 |
ahl | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
envelop | +0.073 |
Offic | +0.068 |
Amen | +0.066 |
ons | +0.063 |
Mic | +0.060 |
Neg logit contributions | |
tub | -0.070 |
gan | -0.066 |
ider | -0.059 |
Moe | -0.059 |
ahl | -0.057 |
Token | box |
Token ID | 3091 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
tub | +0.066 |
gan | +0.063 |
Moe | +0.057 |
ider | +0.057 |
ahl | +0.054 |
Neg logit contributions | |
envelop | -0.066 |
Offic | -0.059 |
Amen | -0.056 |
ons | -0.055 |
Mic | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +4.89e-03 |
Pos logit contributions | |
tub | +0.034 |
gan | +0.032 |
Moe | +0.028 |
ider | +0.028 |
ahl | +0.027 |
Neg logit contributions | |
envelop | -0.041 |
Offic | -0.038 |
Amen | -0.037 |
ons | -0.036 |
Mic | -0.034 |
Token | One |
Token ID | 1881 |
Feature activation | +3.74e-03 |
Pos logit contributions | |
tub | +0.008 |
gan | +0.007 |
ider | +0.007 |
ahl | +0.006 |
Moe | +0.006 |
Neg logit contributions | |
envelop | -0.011 |
Offic | -0.010 |
Amen | -0.010 |
ons | -0.009 |
Mic | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
tub | +0.007 |
gan | +0.007 |
Moe | +0.007 |
ider | +0.007 |
ahl | +0.006 |
Neg logit contributions | |
envelop | -0.007 |
ons | -0.006 |
Offic | -0.006 |
Amen | -0.006 |
oc | -0.006 |
Token | big |
Token ID | 1263 |
Feature activation | -6.47e-02 |
Pos logit contributions | |
tub | +0.060 |
gan | +0.059 |
ider | +0.053 |
Moe | +0.052 |
ahl | +0.052 |
Neg logit contributions | |
envelop | -0.061 |
Offic | -0.053 |
Amen | -0.052 |
ons | -0.052 |
oc | -0.049 |
Token | claws |
Token ID | 28421 |
Feature activation | -3.93e-03 |
Pos logit contributions | |
envelop | +0.132 |
Offic | +0.121 |
Amen | +0.119 |
ons | +0.114 |
Mic | +0.108 |
Neg logit contributions | |
tub | -0.117 |
gan | -0.112 |
ider | -0.099 |
Moe | -0.099 |
ahl | -0.096 |
Token | and |
Token ID | 290 |
Feature activation | -6.19e-02 |
Pos logit contributions | |
envelop | +0.007 |
ons | +0.006 |
Offic | +0.005 |
parcels | +0.005 |
saddle | +0.005 |
Neg logit contributions | |
tub | -0.008 |
gan | -0.008 |
Moe | -0.007 |
ahl | -0.007 |
Mil | -0.007 |
Token | picked |
Token ID | 6497 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
envelop | +0.121 |
Offic | +0.112 |
Amen | +0.109 |
ons | +0.105 |
Mic | +0.099 |
Neg logit contributions | |
tub | -0.117 |
gan | -0.111 |
ider | -0.100 |
Moe | -0.099 |
ahl | -0.096 |
Token | up |
Token ID | 510 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
envelop | +0.026 |
ons | +0.022 |
Offic | +0.021 |
Amen | +0.021 |
parcels | +0.019 |
Neg logit contributions | |
tub | -0.031 |
gan | -0.030 |
ider | -0.029 |
ses | -0.026 |
ahl | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
tub | +0.047 |
gan | +0.045 |
ider | +0.041 |
Moe | +0.040 |
ahl | +0.039 |
Neg logit contributions | |
envelop | -0.045 |
Offic | -0.040 |
Amen | -0.039 |
ons | -0.038 |
oc | -0.035 |
Token | big |
Token ID | 1263 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
envelop | +0.063 |
Offic | +0.059 |
Amen | +0.057 |
ons | +0.055 |
Mic | +0.052 |
Neg logit contributions | |
tub | -0.055 |
gan | -0.052 |
Moe | -0.046 |
ider | -0.046 |
ahl | -0.044 |
Token | fish |
Token ID | 5916 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
envelop | +0.126 |
Offic | +0.118 |
Amen | +0.115 |
ons | +0.111 |
Mic | +0.107 |
Neg logit contributions | |
tub | -0.098 |
gan | -0.092 |
Moe | -0.082 |
ider | -0.080 |
ahl | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +4.01e-03 |
Pos logit contributions | |
envelop | +0.053 |
Offic | +0.049 |
Amen | +0.048 |
ons | +0.047 |
oc | +0.044 |
Neg logit contributions | |
tub | -0.043 |
gan | -0.041 |
Moe | -0.036 |
ider | -0.036 |
ahl | -0.035 |
Token | |
Token ID | 220 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
tub | +0.007 |
gan | +0.006 |
ider | +0.006 |
ahl | +0.006 |
ses | +0.006 |
Neg logit contributions | |
envelop | -0.008 |
Offic | -0.008 |
Amen | -0.008 |
ons | -0.007 |
oc | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
tub | +0.073 |
ider | +0.073 |
gan | +0.071 |
ses | +0.068 |
sed | +0.067 |
Neg logit contributions | |
envelop | -0.058 |
ons | -0.055 |
Offic | -0.051 |
oc | -0.048 |
Amen | -0.048 |
Token | it |
Token ID | 340 |
Feature activation | +7.71e-02 |
Pos logit contributions | |
envelop | +0.006 |
Offic | +0.005 |
Amen | +0.005 |
ons | +0.005 |
Mic | +0.005 |
Neg logit contributions | |
tub | -0.005 |
gan | -0.005 |
Moe | -0.004 |
ider | -0.004 |
ahl | -0.004 |
Token | must |
Token ID | 1276 |
Feature activation | +5.31e-03 |
Pos logit contributions | |
gan | +0.148 |
tub | +0.146 |
ider | +0.135 |
Moe | +0.131 |
ahl | +0.129 |
Neg logit contributions | |
envelop | -0.143 |
ons | -0.135 |
Offic | -0.125 |
Amen | -0.125 |
oc | -0.122 |
Token | have |
Token ID | 423 |
Feature activation | -7.06e-03 |
Pos logit contributions | |
tub | +0.012 |
gan | +0.011 |
ider | +0.010 |
Moe | +0.010 |
ses | +0.010 |
Neg logit contributions | |
envelop | -0.009 |
Amen | -0.008 |
Offic | -0.008 |
ons | -0.007 |
oc | -0.007 |
Token | blown |
Token ID | 16318 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
envelop | +0.013 |
Offic | +0.011 |
ons | +0.011 |
Amen | +0.011 |
parcels | +0.010 |
Neg logit contributions | |
tub | -0.014 |
gan | -0.014 |
ider | -0.013 |
Moe | -0.012 |
ahl | -0.012 |
Token | in |
Token ID | 287 |
Feature activation | +2.05e-03 |
Pos logit contributions | |
envelop | +0.071 |
Offic | +0.064 |
Amen | +0.063 |
ons | +0.061 |
Mic | +0.057 |
Neg logit contributions | |
tub | -0.083 |
gan | -0.079 |
ider | -0.072 |
Moe | -0.071 |
ahl | -0.069 |
Token | from |
Token ID | 422 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
tub | +0.003 |
gan | +0.003 |
Moe | +0.003 |
Mil | +0.003 |
ses | +0.002 |
Neg logit contributions | |
Offic | -0.004 |
envelop | -0.004 |
Amen | -0.004 |
ons | -0.004 |
Mic | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
envelop | +0.068 |
Offic | +0.064 |
Amen | +0.062 |
ons | +0.059 |
Mic | +0.057 |
Neg logit contributions | |
tub | -0.062 |
gan | -0.059 |
Moe | -0.053 |
ider | -0.051 |
ses | -0.050 |
Token | picnic |
Token ID | 35715 |
Feature activation | -4.42e-03 |
Pos logit contributions | |
gan | +0.040 |
tub | +0.040 |
ahl | +0.035 |
ider | +0.035 |
Moe | +0.034 |
Neg logit contributions | |
envelop | -0.047 |
Offic | -0.041 |
Amen | -0.041 |
ons | -0.038 |
parcels | -0.037 |
Token | they |
Token ID | 484 |
Feature activation | +5.82e-03 |
Pos logit contributions | |
envelop | +0.009 |
Offic | +0.008 |
Amen | +0.008 |
ons | +0.008 |
oc | +0.007 |
Neg logit contributions | |
tub | -0.008 |
gan | -0.008 |
ider | -0.007 |
Moe | -0.007 |
ahl | -0.007 |
Token | were |
Token ID | 547 |
Feature activation | +5.56e-03 |
Pos logit contributions | |
tub | +0.009 |
gan | +0.008 |
Moe | +0.007 |
ider | +0.007 |
ses | +0.007 |
Neg logit contributions | |
envelop | -0.013 |
Offic | -0.013 |
Amen | -0.013 |
ons | -0.012 |
Mic | -0.012 |
Token | planning |
Token ID | 5410 |
Feature activation | +5.42e-03 |
Pos logit contributions | |
tub | +0.011 |
gan | +0.010 |
Moe | +0.009 |
ider | +0.009 |
ses | +0.008 |
Neg logit contributions | |
envelop | -0.011 |
Offic | -0.010 |
Amen | -0.010 |
ons | -0.009 |
Mic | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
tub | +0.083 |
gan | +0.080 |
Moe | +0.071 |
ider | +0.069 |
ahl | +0.069 |
Neg logit contributions | |
envelop | -0.104 |
Offic | -0.095 |
Amen | -0.093 |
ons | -0.091 |
oc | -0.085 |
Token | and |
Token ID | 290 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
tub | +0.016 |
gan | +0.015 |
ider | +0.013 |
Moe | +0.013 |
ahl | +0.013 |
Neg logit contributions | |
envelop | -0.024 |
Offic | -0.023 |
Amen | -0.022 |
ons | -0.021 |
Mic | -0.020 |
Token | then |
Token ID | 788 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
envelop | +0.093 |
Offic | +0.085 |
Amen | +0.082 |
ons | +0.079 |
read | +0.075 |
Neg logit contributions | |
tub | -0.095 |
gan | -0.088 |
ider | -0.084 |
ahl | -0.081 |
Moe | -0.080 |
Token | he |
Token ID | 339 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
tub | +0.041 |
gan | +0.038 |
ider | +0.036 |
ahl | +0.035 |
ses | +0.034 |
Neg logit contributions | |
envelop | -0.038 |
Offic | -0.033 |
ons | -0.033 |
read | -0.032 |
Amen | -0.031 |
Token | started |
Token ID | 2067 |
Feature activation | +6.15e-02 |
Pos logit contributions | |
tub | +0.037 |
gan | +0.035 |
Moe | +0.032 |
ider | +0.031 |
ahl | +0.030 |
Neg logit contributions | |
envelop | -0.038 |
Offic | -0.035 |
Amen | -0.033 |
ons | -0.033 |
Mic | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
tub | +0.110 |
gan | +0.104 |
ider | +0.093 |
Moe | +0.092 |
ahl | +0.089 |
Neg logit contributions | |
envelop | -0.127 |
Offic | -0.118 |
Amen | -0.115 |
ons | -0.111 |
Mic | -0.105 |
Token | play |
Token ID | 711 |
Feature activation | +7.78e-03 |
Pos logit contributions | |
envelop | +0.036 |
Offic | +0.033 |
Amen | +0.032 |
ons | +0.030 |
Mic | +0.029 |
Neg logit contributions | |
tub | -0.041 |
gan | -0.039 |
ider | -0.035 |
Moe | -0.035 |
ahl | -0.034 |
Token | on |
Token ID | 319 |
Feature activation | +6.51e-03 |
Pos logit contributions | |
gan | +0.015 |
tub | +0.015 |
ses | +0.013 |
ider | +0.013 |
Moe | +0.013 |
Neg logit contributions | |
envelop | -0.015 |
Offic | -0.013 |
Amen | -0.013 |
ons | -0.013 |
oc | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
tub | +0.013 |
gan | +0.013 |
ider | +0.012 |
Moe | +0.011 |
ahl | +0.011 |
Neg logit contributions | |
envelop | -0.012 |
Offic | -0.011 |
Amen | -0.010 |
ons | -0.010 |
Mic | -0.009 |
Token | stair |
Token ID | 18761 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
tub | +0.091 |
gan | +0.087 |
ider | +0.077 |
Moe | +0.076 |
ses | +0.073 |
Neg logit contributions | |
envelop | -0.119 |
Offic | -0.113 |
ons | -0.108 |
Amen | -0.105 |
Mic | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
tub | +0.118 |
gan | +0.114 |
Moe | +0.102 |
ses | +0.101 |
ahl | +0.100 |
Neg logit contributions | |
envelop | -0.138 |
Offic | -0.126 |
ons | -0.122 |
Amen | -0.122 |
oc | -0.114 |
Token | carry |
Token ID | 3283 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
envelop | +0.005 |
Offic | +0.004 |
Amen | +0.004 |
ons | +0.004 |
Mic | +0.004 |
Neg logit contributions | |
tub | -0.005 |
gan | -0.005 |
Moe | -0.005 |
ider | -0.005 |
ahl | -0.004 |
Token | Anna |
Token ID | 11735 |
Feature activation | +4.63e-03 |
Pos logit contributions | |
envelop | +0.056 |
Offic | +0.049 |
ons | +0.048 |
Amen | +0.048 |
Mic | +0.044 |
Neg logit contributions | |
tub | -0.066 |
gan | -0.062 |
ider | -0.057 |
Moe | -0.055 |
ahl | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | -8.45e-02 |
Pos logit contributions | |
tub | +0.009 |
gan | +0.008 |
ider | +0.007 |
Moe | +0.007 |
ahl | +0.007 |
Neg logit contributions | |
envelop | -0.009 |
Offic | -0.008 |
Amen | -0.008 |
ons | -0.008 |
oc | -0.007 |
Token | he |
Token ID | 339 |
Feature activation | -6.94e-04 |
Pos logit contributions | |
envelop | +0.191 |
Offic | +0.177 |
Amen | +0.174 |
ons | +0.167 |
Mic | +0.159 |
Neg logit contributions | |
tub | -0.139 |
gan | -0.127 |
ider | -0.112 |
Moe | -0.110 |
ahl | -0.107 |
Token | can |
Token ID | 460 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
envelop | +0.002 |
Offic | +0.001 |
Amen | +0.001 |
ons | +0.001 |
oc | +0.001 |
Neg logit contributions | |
tub | -0.001 |
gan | -0.001 |
Moe | -0.001 |
ider | -0.001 |
ahl | -0.001 |
Token | 't |
Token ID | 470 |
Feature activation | -5.05e-03 |
Pos logit contributions | |
tub | +0.020 |
gan | +0.019 |
Moe | +0.017 |
ider | +0.017 |
ahl | +0.017 |
Neg logit contributions | |
envelop | -0.020 |
Offic | -0.018 |
Amen | -0.018 |
ons | -0.017 |
Mic | -0.016 |
Token | leave |
Token ID | 2666 |
Feature activation | +9.96e-03 |
Pos logit contributions | |
envelop | +0.009 |
Offic | +0.008 |
Amen | +0.008 |
ons | +0.008 |
Mic | +0.007 |
Neg logit contributions | |
tub | -0.010 |
gan | -0.010 |
Moe | -0.009 |
ider | -0.009 |
ahl | -0.008 |
Token | her |
Token ID | 607 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
tub | +0.019 |
gan | +0.018 |
Moe | +0.016 |
ider | +0.016 |
ahl | +0.016 |
Neg logit contributions | |
envelop | -0.019 |
Offic | -0.018 |
Amen | -0.017 |
ons | -0.017 |
Mic | -0.016 |
Token | alone |
Token ID | 3436 |
Feature activation | +6.91e-02 |
Pos logit contributions | |
tub | +0.065 |
gan | +0.062 |
Moe | +0.056 |
ider | +0.055 |
ahl | +0.054 |
Neg logit contributions | |
envelop | -0.077 |
Amen | -0.067 |
Offic | -0.067 |
ons | -0.065 |
oc | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
tub | +0.131 |
gan | +0.127 |
ahl | +0.117 |
Moe | +0.116 |
ses | +0.113 |
Neg logit contributions | |
envelop | -0.130 |
Amen | -0.116 |
Offic | -0.111 |
ons | -0.110 |
oc | -0.105 |
Token | He |
Token ID | 679 |
Feature activation | -5.71e-03 |
Pos logit contributions | |
tub | +0.024 |
gan | +0.021 |
anger | +0.020 |
ider | +0.020 |
sed | +0.020 |
Neg logit contributions | |
Amen | -0.023 |
envelop | -0.023 |
Offic | -0.022 |
oc | -0.021 |
Mic | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
tub | +0.002 |
gan | +0.002 |
ses | +0.002 |
Moe | +0.002 |
ahl | +0.002 |
Neg logit contributions | |
envelop | -0.002 |
Offic | -0.002 |
Amen | -0.002 |
ons | -0.002 |
oc | -0.002 |
Token | his |
Token ID | 465 |
Feature activation | +9.08e-03 |
Pos logit contributions | |
envelop | +0.086 |
Offic | +0.081 |
Amen | +0.077 |
ons | +0.076 |
Mic | +0.074 |
Neg logit contributions | |
tub | -0.069 |
gan | -0.063 |
Moe | -0.059 |
Mil | -0.057 |
ahl | -0.052 |
Token | mom |
Token ID | 1995 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
tub | +0.015 |
gan | +0.015 |
Moe | +0.013 |
ider | +0.013 |
ahl | +0.012 |
Neg logit contributions | |
envelop | -0.020 |
Offic | -0.019 |
Amen | -0.018 |
ons | -0.017 |
Mic | -0.017 |
Token | 's |
Token ID | 338 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
envelop | +0.139 |
Offic | +0.132 |
Amen | +0.128 |
ons | +0.122 |
Mic | +0.119 |
Neg logit contributions | |
tub | -0.123 |
gan | -0.116 |
Moe | -0.103 |
ider | -0.102 |
ahl | -0.097 |
Token | warning |
Token ID | 6509 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
envelop | +0.027 |
Offic | +0.025 |
ons | +0.024 |
Amen | +0.024 |
Mic | +0.023 |
Neg logit contributions | |
tub | -0.026 |
gan | -0.024 |
ider | -0.022 |
Moe | -0.022 |
ses | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
envelop | +0.041 |
Offic | +0.036 |
ons | +0.036 |
Amen | +0.035 |
lending | +0.032 |
Neg logit contributions | |
tub | -0.050 |
gan | -0.049 |
ider | -0.045 |
ahl | -0.044 |
Moe | -0.043 |
Token | kept |
Token ID | 4030 |
Feature activation | +9.73e-03 |
Pos logit contributions | |
envelop | +0.145 |
Offic | +0.134 |
Amen | +0.131 |
ons | +0.125 |
Mic | +0.119 |
Neg logit contributions | |
tub | -0.150 |
gan | -0.143 |
Moe | -0.128 |
ider | -0.127 |
ahl | -0.123 |
Token | his |
Token ID | 465 |
Feature activation | +3.67e-03 |
Pos logit contributions | |
tub | +0.016 |
gan | +0.015 |
Moe | +0.014 |
Mil | +0.013 |
ter | +0.013 |
Neg logit contributions | |
envelop | -0.020 |
Offic | -0.019 |
oc | -0.019 |
Amen | -0.018 |
Mic | -0.018 |
Token | lungs |
Token ID | 21726 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
tub | +0.007 |
gan | +0.006 |
ider | +0.006 |
Moe | +0.006 |
ses | +0.005 |
Neg logit contributions | |
envelop | -0.007 |
Offic | -0.007 |
Amen | -0.007 |
ons | -0.007 |
Mic | -0.006 |
Token | safe |
Token ID | 3338 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
envelop | +0.070 |
Offic | +0.063 |
Amen | +0.063 |
ons | +0.060 |
Mic | +0.056 |
Neg logit contributions | |
tub | -0.073 |
gan | -0.070 |
Moe | -0.063 |
ider | -0.062 |
ahl | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
gan | +0.029 |
tub | +0.029 |
ahl | +0.027 |
ider | +0.026 |
Moe | +0.026 |
Neg logit contributions | |
envelop | -0.030 |
Amen | -0.027 |
Offic | -0.025 |
ons | -0.025 |
oc | -0.025 |
Token | waved |
Token ID | 26834 |
Feature activation | -6.39e-02 |
Pos logit contributions | |
envelop | +0.115 |
Offic | +0.106 |
Amen | +0.104 |
ons | +0.099 |
Mic | +0.094 |
Neg logit contributions | |
tub | -0.118 |
gan | -0.112 |
Moe | -0.100 |
ider | -0.100 |
ahl | -0.097 |
Token | their |
Token ID | 511 |
Feature activation | -3.95e-03 |
Pos logit contributions | |
envelop | +0.136 |
Offic | +0.128 |
Amen | +0.125 |
ons | +0.122 |
Mic | +0.117 |
Neg logit contributions | |
tub | -0.108 |
gan | -0.103 |
Moe | -0.091 |
ider | -0.089 |
ahl | -0.087 |
Token | fins |
Token ID | 42080 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
envelop | +0.007 |
Offic | +0.007 |
Amen | +0.007 |
ons | +0.007 |
Mic | +0.006 |
Neg logit contributions | |
tub | -0.008 |
gan | -0.007 |
Moe | -0.007 |
ider | -0.007 |
ses | -0.006 |
Token | at |
Token ID | 379 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
envelop | +0.071 |
Offic | +0.064 |
Amen | +0.063 |
ons | +0.061 |
oc | +0.057 |
Neg logit contributions | |
tub | -0.064 |
gan | -0.061 |
Moe | -0.054 |
ider | -0.054 |
ahl | -0.053 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
envelop | +0.063 |
Offic | +0.059 |
Amen | +0.058 |
ons | +0.055 |
Mic | +0.052 |
Neg logit contributions | |
tub | -0.071 |
gan | -0.067 |
Moe | -0.061 |
ider | -0.060 |
ahl | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -8.33e-02 |
Pos logit contributions | |
tub | +0.026 |
gan | +0.025 |
Moe | +0.022 |
ider | +0.022 |
ahl | +0.021 |
Neg logit contributions | |
envelop | -0.036 |
Offic | -0.033 |
Amen | -0.033 |
ons | -0.032 |
Mic | -0.030 |
Token | Mia |
Token ID | 32189 |
Feature activation | +2.88e-03 |
Pos logit contributions | |
envelop | +0.177 |
Offic | +0.165 |
Amen | +0.158 |
ons | +0.146 |
oc | +0.145 |
Neg logit contributions | |
tub | -0.145 |
gan | -0.130 |
ider | -0.127 |
anger | -0.113 |
ahl | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | +5.42e-03 |
Pos logit contributions | |
tub | +0.005 |
gan | +0.004 |
Moe | +0.004 |
ider | +0.004 |
ahl | +0.004 |
Neg logit contributions | |
envelop | -0.006 |
Offic | -0.006 |
Amen | -0.006 |
ons | -0.006 |
oc | -0.005 |
Token | They |
Token ID | 1119 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
gan | +0.010 |
tub | +0.009 |
ider | +0.009 |
ahl | +0.008 |
ter | +0.008 |
Neg logit contributions | |
envelop | -0.010 |
Offic | -0.010 |
Amen | -0.009 |
oc | -0.008 |
exchange | -0.008 |
Token | thanked |
Token ID | 26280 |
Feature activation | -6.02e-02 |
Pos logit contributions | |
envelop | +0.039 |
Amen | +0.036 |
Offic | +0.035 |
ons | +0.034 |
Mic | +0.032 |
Neg logit contributions | |
tub | -0.038 |
gan | -0.036 |
Moe | -0.033 |
ider | -0.031 |
ahl | -0.031 |
Token | them |
Token ID | 606 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
envelop | +0.103 |
Offic | +0.095 |
Amen | +0.092 |
ons | +0.088 |
Mic | +0.084 |
Neg logit contributions | |
tub | -0.127 |
gan | -0.123 |
Moe | -0.115 |
Mil | -0.110 |
ahl | -0.107 |
Token |
|
Token ID | 198 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
envelop | +0.063 |
Offic | +0.060 |
Amen | +0.058 |
ons | +0.056 |
oc | +0.053 |
Neg logit contributions | |
tub | -0.059 |
gan | -0.056 |
ider | -0.052 |
Moe | -0.049 |
ses | -0.048 |
Token | M |
Token ID | 44 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
envelop | +0.045 |
Offic | +0.042 |
Amen | +0.040 |
ons | +0.038 |
Mic | +0.036 |
Neg logit contributions | |
tub | -0.062 |
gan | -0.058 |
ider | -0.053 |
Moe | -0.053 |
ses | -0.052 |
Token | ia |
Token ID | 544 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
envelop | +0.084 |
ons | +0.083 |
Offic | +0.078 |
pa | +0.076 |
Amen | +0.073 |
Neg logit contributions | |
tub | -0.050 |
Moe | -0.035 |
robe | -0.034 |
gan | -0.034 |
ter | -0.033 |
Token | hugged |
Token ID | 41130 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
envelop | +0.022 |
Offic | +0.020 |
ons | +0.019 |
Amen | +0.019 |
read | +0.018 |
Neg logit contributions | |
tub | -0.026 |
gan | -0.023 |
ider | -0.022 |
ahl | -0.022 |
Moe | -0.022 |
Token | her |
Token ID | 607 |
Feature activation | -4.71e-03 |
Pos logit contributions | |
envelop | +0.117 |
Offic | +0.107 |
Amen | +0.105 |
ons | +0.099 |
Mic | +0.093 |
Neg logit contributions | |
tub | -0.144 |
gan | -0.139 |
Moe | -0.127 |
ider | -0.124 |
ahl | -0.122 |
Token | mom |
Token ID | 1995 |
Feature activation | -7.73e-02 |
Pos logit contributions | |
envelop | +0.011 |
Offic | +0.010 |
Amen | +0.010 |
ons | +0.010 |
Mic | +0.009 |
Neg logit contributions | |
tub | -0.007 |
gan | -0.007 |
Moe | -0.006 |
ider | -0.006 |
ses | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -9.99e-02 |
Pos logit contributions | |
envelop | +0.172 |
Offic | +0.171 |
Amen | +0.162 |
ons | +0.156 |
Mic | +0.153 |
Neg logit contributions | |
gan | -0.119 |
tub | -0.113 |
Moe | -0.099 |
ider | -0.091 |
ahl | -0.090 |
Token | dad |
Token ID | 9955 |
Feature activation | -8.76e-02 |
Pos logit contributions | |
envelop | +0.198 |
Amen | +0.187 |
Offic | +0.186 |
ons | +0.177 |
oc | +0.173 |
Neg logit contributions | |
gan | -0.176 |
tub | -0.175 |
Moe | -0.154 |
Mil | -0.149 |
ses | -0.146 |
Token | . |
Token ID | 13 |
Feature activation | +4.23e-03 |
Pos logit contributions | |
envelop | +0.190 |
Offic | +0.174 |
Amen | +0.172 |
ons | +0.167 |
oc | +0.157 |
Neg logit contributions | |
tub | -0.150 |
gan | -0.139 |
ider | -0.124 |
Moe | -0.122 |
ahl | -0.119 |
Token | She |
Token ID | 1375 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
tub | +0.007 |
gan | +0.007 |
ider | +0.007 |
ahl | +0.006 |
anger | +0.006 |
Neg logit contributions | |
envelop | -0.009 |
Offic | -0.008 |
Amen | -0.008 |
Mic | -0.007 |
ons | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
envelop | +0.035 |
Offic | +0.033 |
Amen | +0.032 |
ons | +0.031 |
Mic | +0.030 |
Neg logit contributions | |
tub | -0.024 |
gan | -0.023 |
Moe | -0.020 |
ider | -0.020 |
ahl | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
gan | +0.048 |
tub | +0.046 |
ider | +0.043 |
ses | +0.042 |
ahl | +0.041 |
Neg logit contributions | |
envelop | -0.056 |
Offic | -0.050 |
ons | -0.050 |
Amen | -0.046 |
oc | -0.045 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
tub | +0.028 |
gan | +0.026 |
ider | +0.026 |
ses | +0.025 |
ahl | +0.023 |
Neg logit contributions | |
envelop | -0.039 |
Offic | -0.035 |
ons | -0.034 |
Amen | -0.034 |
read | -0.032 |
Token | reminded |
Token ID | 14516 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
tub | +0.037 |
gan | +0.035 |
Moe | +0.031 |
ider | +0.031 |
ahl | +0.030 |
Neg logit contributions | |
envelop | -0.035 |
Offic | -0.033 |
Amen | -0.033 |
ons | -0.031 |
Mic | -0.029 |
Token | her |
Token ID | 607 |
Feature activation | +3.31e-03 |
Pos logit contributions | |
envelop | +0.054 |
Offic | +0.051 |
Amen | +0.049 |
oc | +0.048 |
ons | +0.044 |
Neg logit contributions | |
gan | -0.063 |
tub | -0.062 |
Moe | -0.058 |
Mil | -0.054 |
ahl | -0.054 |
Token | dad |
Token ID | 9955 |
Feature activation | -8.79e-02 |
Pos logit contributions | |
tub | +0.005 |
gan | +0.005 |
Moe | +0.004 |
Mil | +0.003 |
ider | +0.003 |
Neg logit contributions | |
envelop | -0.008 |
Offic | -0.008 |
Amen | -0.008 |
Mic | -0.007 |
oc | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -7.63e-02 |
Pos logit contributions | |
envelop | +0.178 |
Offic | +0.163 |
Amen | +0.160 |
ons | +0.157 |
Mic | +0.144 |
Neg logit contributions | |
tub | -0.161 |
gan | -0.152 |
ider | -0.139 |
Moe | -0.134 |
ahl | -0.131 |
Token | he |
Token ID | 339 |
Feature activation | -9.49e-03 |
Pos logit contributions | |
envelop | +0.151 |
Offic | +0.146 |
Amen | +0.140 |
Mic | +0.135 |
oc | +0.134 |
Neg logit contributions | |
tub | -0.137 |
gan | -0.132 |
Moe | -0.117 |
Mil | -0.111 |
ahl | -0.105 |
Token | thanked |
Token ID | 26280 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
envelop | +0.019 |
Offic | +0.017 |
Amen | +0.017 |
ons | +0.016 |
Mic | +0.015 |
Neg logit contributions | |
tub | -0.018 |
gan | -0.017 |
ider | -0.015 |
Moe | -0.015 |
ahl | -0.014 |
Token | her |
Token ID | 607 |
Feature activation | -4.60e-03 |
Pos logit contributions | |
Offic | +0.126 |
envelop | +0.122 |
oc | +0.118 |
saddle | +0.117 |
ons | +0.114 |
Neg logit contributions | |
Moe | -0.091 |
gan | -0.088 |
tub | -0.086 |
Mil | -0.086 |
ter | -0.076 |
Token | for |
Token ID | 329 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
envelop | +0.011 |
Offic | +0.010 |
Amen | +0.010 |
ons | +0.010 |
Mic | +0.009 |
Neg logit contributions | |
tub | -0.007 |
gan | -0.007 |
Moe | -0.006 |
ider | -0.006 |
ahl | -0.005 |
Token | being |
Token ID | 852 |
Feature activation | +2.46e-03 |
Pos logit contributions | |
oc | +0.084 |
envelop | +0.075 |
Offic | +0.072 |
Amen | +0.071 |
Mic | +0.071 |
Neg logit contributions | |
tub | -0.061 |
gan | -0.059 |
Moe | -0.057 |
ter | -0.057 |
Mil | -0.052 |
Token | to |
Token ID | 284 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
envelop | +0.109 |
Offic | +0.104 |
Amen | +0.102 |
ons | +0.097 |
Mic | +0.094 |
Neg logit contributions | |
tub | -0.081 |
gan | -0.076 |
Moe | -0.066 |
ider | -0.066 |
ahl | -0.064 |
Token | their |
Token ID | 511 |
Feature activation | -6.17e-03 |
Pos logit contributions | |
envelop | +0.082 |
ons | +0.080 |
Amen | +0.079 |
Offic | +0.079 |
Mic | +0.074 |
Neg logit contributions | |
tub | -0.062 |
gan | -0.059 |
Moe | -0.054 |
Mil | -0.051 |
ses | -0.049 |
Token | room |
Token ID | 2119 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
envelop | +0.014 |
Offic | +0.013 |
Amen | +0.013 |
ons | +0.012 |
Mic | +0.012 |
Neg logit contributions | |
tub | -0.010 |
gan | -0.009 |
Moe | -0.008 |
ider | -0.008 |
ses | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
tub | +0.028 |
gan | +0.026 |
ider | +0.022 |
Moe | +0.022 |
ahl | +0.021 |
Neg logit contributions | |
envelop | -0.040 |
Offic | -0.038 |
Amen | -0.037 |
ons | -0.036 |
Mic | -0.035 |
Token | They |
Token ID | 1119 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
tub | +0.027 |
gan | +0.027 |
ider | +0.024 |
ahl | +0.023 |
Moe | +0.023 |
Neg logit contributions | |
envelop | -0.035 |
Amen | -0.032 |
Offic | -0.032 |
ons | -0.029 |
oc | -0.028 |
Token | gave |
Token ID | 2921 |
Feature activation | -7.26e-02 |
Pos logit contributions | |
envelop | +0.043 |
Amen | +0.041 |
Offic | +0.041 |
ons | +0.039 |
Mic | +0.038 |
Neg logit contributions | |
tub | -0.041 |
gan | -0.040 |
Moe | -0.035 |
Mil | -0.033 |
ahl | -0.033 |
Token | him |
Token ID | 683 |
Feature activation | -1.14e-03 |
Pos logit contributions | |
Offic | +0.145 |
envelop | +0.139 |
Amen | +0.134 |
Mic | +0.129 |
ons | +0.125 |
Neg logit contributions | |
tub | -0.125 |
gan | -0.124 |
Moe | -0.113 |
ses | -0.104 |
Mil | -0.103 |
Token | food |
Token ID | 2057 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
envelop | +0.002 |
Offic | +0.002 |
Amen | +0.002 |
ons | +0.002 |
Mic | +0.002 |
Neg logit contributions | |
tub | -0.002 |
gan | -0.002 |
Moe | -0.002 |
ider | -0.002 |
ahl | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
envelop | +0.081 |
Offic | +0.071 |
Amen | +0.069 |
ons | +0.069 |
oc | +0.064 |
Neg logit contributions | |
tub | -0.079 |
gan | -0.077 |
ider | -0.070 |
Moe | -0.069 |
ahl | -0.068 |
Token | water |
Token ID | 1660 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
Offic | +0.159 |
envelop | +0.159 |
Amen | +0.158 |
Mic | +0.145 |
ons | +0.145 |
Neg logit contributions | |
tub | -0.167 |
gan | -0.156 |
Moe | -0.141 |
ses | -0.141 |
sed | -0.135 |
Token | . |
Token ID | 13 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
envelop | +0.028 |
Offic | +0.024 |
ons | +0.024 |
Amen | +0.024 |
oc | +0.022 |
Neg logit contributions | |
tub | -0.024 |
gan | -0.024 |
ider | -0.022 |
Moe | -0.020 |
ahl | -0.020 |