Token | . |
Token ID | 13 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
nodded | +0.087 |
youngest | +0.085 |
ver | +0.085 |
led | +0.082 |
So | +0.082 |
Neg logit contributions | |
poisonous | -0.118 |
ocol | -0.114 |
ocket | -0.112 |
arter | -0.110 |
bamboo | -0.103 |
Token |
|
Token ID | 198 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
poisonous | +0.020 |
ocol | +0.018 |
ocket | +0.018 |
aster | +0.018 |
arter | +0.018 |
Neg logit contributions | |
nodded | -0.012 |
ver | -0.012 |
youngest | -0.012 |
acts | -0.012 |
Let | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
Let | +0.013 |
So | +0.013 |
ver | +0.013 |
led | +0.012 |
nodded | +0.011 |
Neg logit contributions | |
poisonous | -0.039 |
ocol | -0.038 |
ocket | -0.037 |
arter | -0.037 |
injuries | -0.034 |
Token | He |
Token ID | 1544 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
nodded | +0.024 |
Let | +0.023 |
ver | +0.023 |
youngest | +0.023 |
bars | +0.022 |
Neg logit contributions | |
poisonous | -0.027 |
ocket | -0.025 |
ocol | -0.025 |
arter | -0.024 |
aster | -0.023 |
Token | t |
Token ID | 256 |
Feature activation | -8.20e-01 |
Pos logit contributions | |
poisonous | +0.112 |
aster | +0.103 |
ocket | +0.102 |
ocol | +0.100 |
bamboo | +0.100 |
Neg logit contributions | |
ver | -0.067 |
folk | -0.067 |
youngest | -0.064 |
Let | -0.061 |
oldest | -0.060 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
poisonous | +0.054 |
ocol | +0.050 |
ocket | +0.050 |
aster | +0.049 |
arter | +0.048 |
Neg logit contributions | |
bars | -0.015 |
ver | -0.015 |
nodded | -0.015 |
youngest | -0.015 |
Let | -0.014 |
Token | oes |
Token ID | 3028 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
nodded | +0.424 |
ver | +0.424 |
Let | +0.421 |
bars | +0.419 |
youngest | +0.404 |
Neg logit contributions | |
poisonous | -0.443 |
ocol | -0.407 |
ocket | -0.403 |
arter | -0.385 |
aster | -0.377 |
Token | to |
Token ID | 284 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
youngest | +0.197 |
acts | +0.193 |
folk | +0.192 |
oldest | +0.191 |
ver | +0.189 |
Neg logit contributions | |
poisonous | -0.169 |
aster | -0.149 |
ocket | -0.149 |
ocol | -0.144 |
bamboo | -0.143 |
Token | her |
Token ID | 607 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
ocket | +0.012 |
poisonous | +0.012 |
ocol | +0.012 |
arter | +0.012 |
endant | +0.011 |
Neg logit contributions | |
led | -0.007 |
Let | -0.007 |
youngest | -0.007 |
oldest | -0.006 |
younger | -0.006 |
Token | bed |
Token ID | 3996 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
poisonous | +0.120 |
ocol | +0.118 |
arter | +0.108 |
aster | +0.108 |
ocket | +0.106 |
Neg logit contributions | |
youngest | -0.078 |
nodded | -0.077 |
acts | -0.076 |
led | -0.075 |
So | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
poisonous | +0.228 |
ocol | +0.211 |
ocket | +0.207 |
aster | +0.202 |
arter | +0.200 |
Neg logit contributions | |
nodded | -0.192 |
ver | -0.192 |
youngest | -0.190 |
acts | -0.189 |
Let | -0.186 |
Token | naughty |
Token ID | 45128 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
ocol | +0.021 |
poisonous | +0.020 |
ocket | +0.020 |
aster | +0.020 |
arter | +0.019 |
Neg logit contributions | |
nodded | -0.016 |
youngest | -0.016 |
led | -0.016 |
Let | -0.016 |
nic | -0.015 |
Token | idea |
Token ID | 2126 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
poisonous | +0.014 |
ocol | +0.013 |
ocket | +0.013 |
aster | +0.012 |
arter | +0.012 |
Neg logit contributions | |
acts | -0.008 |
nodded | -0.008 |
youngest | -0.007 |
led | -0.007 |
nods | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
led | +0.138 |
ver | +0.131 |
nodded | +0.131 |
Let | +0.129 |
's | +0.129 |
Neg logit contributions | |
poisonous | -0.209 |
ocol | -0.206 |
ocket | -0.194 |
arter | -0.194 |
aster | -0.188 |
Token | She |
Token ID | 1375 |
Feature activation | -3.23e+00 |
Pos logit contributions | |
poisonous | +0.014 |
aster | +0.012 |
ocket | +0.012 |
ocol | +0.012 |
bamboo | +0.012 |
Neg logit contributions | |
acts | -0.013 |
youngest | -0.012 |
ver | -0.012 |
folk | -0.012 |
oldest | -0.012 |
Token | t |
Token ID | 256 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
poisonous | +0.182 |
ocket | +0.163 |
aster | +0.158 |
bamboo | +0.156 |
ocol | +0.153 |
Neg logit contributions | |
folk | -0.091 |
ver | -0.089 |
bars | -0.086 |
acts | -0.083 |
youngest | -0.083 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
poisonous | +0.036 |
ocol | +0.032 |
ocket | +0.032 |
aster | +0.031 |
arter | +0.030 |
Neg logit contributions | |
ver | -0.037 |
nodded | -0.037 |
youngest | -0.037 |
bars | -0.036 |
Let | -0.036 |
Token | o |
Token ID | 78 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
nodded | +0.439 |
ver | +0.439 |
Let | +0.436 |
bars | +0.432 |
youngest | +0.423 |
Neg logit contributions | |
poisonous | -0.419 |
ocol | -0.384 |
ocket | -0.380 |
arter | -0.363 |
aster | -0.359 |
Token | ed |
Token ID | 276 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
poisonous | +0.156 |
ocket | +0.138 |
bamboo | +0.129 |
aster | +0.128 |
eds | +0.128 |
Neg logit contributions | |
acts | -0.144 |
folk | -0.143 |
youngest | -0.143 |
mun | -0.139 |
oldest | -0.138 |
Token | to |
Token ID | 284 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
led | +0.084 |
Let | +0.084 |
nodded | +0.084 |
ver | +0.081 |
youngest | +0.079 |
Neg logit contributions | |
poisonous | -0.126 |
ocol | -0.125 |
arter | -0.123 |
ocket | -0.122 |
aster | -0.118 |
Token | his |
Token ID | 465 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
poisonous | +0.012 |
ocket | +0.011 |
ocol | +0.011 |
arter | +0.011 |
endant | +0.011 |
Neg logit contributions | |
led | -0.007 |
Let | -0.007 |
youngest | -0.007 |
asked | -0.006 |
asks | -0.006 |
Token | bed |
Token ID | 3996 |
Feature activation | -5.06e+00 |
Pos logit contributions | |
ocol | +0.020 |
poisonous | +0.020 |
aster | +0.018 |
ocket | +0.018 |
arter | +0.017 |
Neg logit contributions | |
youngest | -0.015 |
oldest | -0.014 |
acts | -0.013 |
led | -0.013 |
eldest | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
nodded | +0.083 |
acts | +0.081 |
led | +0.081 |
youngest | +0.080 |
ver | +0.079 |
Neg logit contributions | |
poisonous | -0.123 |
ocket | -0.118 |
ocol | -0.117 |
aster | -0.114 |
arter | -0.112 |
Token | baby |
Token ID | 5156 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
youngest | +0.056 |
acts | +0.055 |
Let | +0.054 |
oldest | +0.054 |
nodded | +0.054 |
Neg logit contributions | |
poisonous | -0.048 |
aster | -0.047 |
ocol | -0.047 |
arter | -0.045 |
ocket | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
poisonous | +0.046 |
ocol | +0.042 |
ocket | +0.042 |
aster | +0.041 |
arter | +0.040 |
Neg logit contributions | |
acts | -0.042 |
youngest | -0.042 |
nodded | -0.042 |
ver | -0.041 |
Let | -0.041 |
Token | They |
Token ID | 1119 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
poisonous | +0.025 |
bamboo | +0.023 |
arter | +0.023 |
heaven | +0.023 |
ocol | +0.023 |
Neg logit contributions | |
ver | -0.015 |
mun | -0.015 |
acts | -0.015 |
bars | -0.015 |
folk | -0.014 |
Token | t |
Token ID | 256 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
sets | +0.021 |
bars | +0.020 |
folk | +0.019 |
ver | +0.019 |
acts | +0.018 |
Neg logit contributions | |
poisonous | -0.037 |
ocket | -0.034 |
bamboo | -0.032 |
aster | -0.031 |
injuries | -0.031 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
poisonous | +0.052 |
ocol | +0.048 |
ocket | +0.048 |
aster | +0.047 |
arter | +0.047 |
Neg logit contributions | |
ver | -0.016 |
nodded | -0.016 |
youngest | -0.016 |
acts | -0.016 |
Let | -0.015 |
Token | o |
Token ID | 78 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
ver | +0.450 |
nodded | +0.450 |
youngest | +0.446 |
acts | +0.446 |
Let | +0.443 |
Neg logit contributions | |
poisonous | -0.398 |
ocol | -0.358 |
ocket | -0.356 |
aster | -0.344 |
arter | -0.339 |
Token | ed |
Token ID | 276 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
poisonous | +0.155 |
ocket | +0.140 |
eds | +0.131 |
bamboo | +0.130 |
sm | +0.127 |
Neg logit contributions | |
acts | -0.141 |
folk | -0.137 |
basics | -0.136 |
youngest | -0.136 |
mun | -0.136 |
Token | inside |
Token ID | 2641 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
nodded | +0.097 |
Let | +0.096 |
led | +0.096 |
ver | +0.096 |
youngest | +0.095 |
Neg logit contributions | |
poisonous | -0.114 |
ocol | -0.107 |
ocket | -0.106 |
arter | -0.104 |
aster | -0.104 |
Token | and |
Token ID | 290 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
poisonous | +0.081 |
ocol | +0.074 |
ocket | +0.074 |
aster | +0.072 |
arter | +0.071 |
Neg logit contributions | |
ver | -0.073 |
nodded | -0.073 |
acts | -0.073 |
youngest | -0.073 |
Let | -0.072 |
Token | put |
Token ID | 1234 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
poisonous | +0.008 |
ocket | +0.007 |
aster | +0.007 |
ocol | +0.007 |
bamboo | +0.007 |
Neg logit contributions | |
acts | -0.005 |
ver | -0.005 |
folk | -0.005 |
bars | -0.005 |
youngest | -0.005 |
Token | aked |
Token ID | 4335 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
poisonous | +0.001 |
arter | +0.001 |
ocol | +0.001 |
ocket | +0.001 |
brand | +0.001 |
Neg logit contributions | |
nodded | -0.000 |
acts | -0.000 |
Let | -0.000 |
ver | -0.000 |
led | -0.000 |
Token | out |
Token ID | 503 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
youngest | +0.094 |
Let | +0.093 |
nodded | +0.092 |
led | +0.092 |
ver | +0.091 |
Neg logit contributions | |
poisonous | -0.109 |
ocket | -0.102 |
ocol | -0.101 |
arter | -0.099 |
aster | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
youngest | +0.023 |
led | +0.023 |
ver | +0.023 |
acts | +0.023 |
Let | +0.022 |
Neg logit contributions | |
poisonous | -0.030 |
ocket | -0.029 |
ocol | -0.028 |
aster | -0.027 |
arter | -0.027 |
Token | They |
Token ID | 1119 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
poisonous | +0.070 |
ocol | +0.065 |
ocket | +0.065 |
aster | +0.064 |
arter | +0.063 |
Neg logit contributions | |
acts | -0.028 |
ver | -0.028 |
nodded | -0.028 |
youngest | -0.028 |
folk | -0.027 |
Token | t |
Token ID | 256 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
acts | +0.013 |
youngest | +0.013 |
folk | +0.013 |
bars | +0.013 |
ver | +0.012 |
Neg logit contributions | |
poisonous | -0.023 |
ocket | -0.021 |
aster | -0.021 |
ocol | -0.020 |
bamboo | -0.020 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
poisonous | +0.057 |
ocol | +0.053 |
ocket | +0.053 |
aster | +0.053 |
arter | +0.052 |
Neg logit contributions | |
ver | -0.017 |
youngest | -0.017 |
nodded | -0.017 |
acts | -0.017 |
Let | -0.017 |
Token | o |
Token ID | 78 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
ver | +0.439 |
nodded | +0.437 |
Let | +0.434 |
bars | +0.429 |
youngest | +0.427 |
Neg logit contributions | |
poisonous | -0.415 |
ocol | -0.377 |
ocket | -0.375 |
aster | -0.358 |
arter | -0.357 |
Token | ed |
Token ID | 276 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
poisonous | +0.157 |
ocket | +0.142 |
eds | +0.133 |
aster | +0.133 |
injuries | +0.130 |
Neg logit contributions | |
acts | -0.144 |
youngest | -0.137 |
folk | -0.135 |
mun | -0.134 |
oldest | -0.134 |
Token | to |
Token ID | 284 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
led | +0.089 |
Let | +0.089 |
nodded | +0.088 |
ver | +0.087 |
youngest | +0.087 |
Neg logit contributions | |
poisonous | -0.119 |
ocol | -0.114 |
ocket | -0.113 |
arter | -0.111 |
aster | -0.109 |
Token | the |
Token ID | 262 |
Feature activation | +9.74e-01 |
Pos logit contributions | |
poisonous | +0.018 |
ocket | +0.018 |
ocol | +0.017 |
affle | +0.017 |
arter | +0.017 |
Neg logit contributions | |
youngest | -0.009 |
Let | -0.009 |
led | -0.009 |
oldest | -0.008 |
younger | -0.008 |
Token | living |
Token ID | 2877 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
youngest | +0.032 |
oldest | +0.031 |
Let | +0.029 |
eldest | +0.029 |
way | +0.029 |
Neg logit contributions | |
ocol | -0.053 |
ocket | -0.050 |
poisonous | -0.050 |
ez | -0.049 |
arter | -0.048 |
Token | he |
Token ID | 339 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
poisonous | +0.126 |
ocol | +0.116 |
ocket | +0.115 |
aster | +0.111 |
arter | +0.111 |
Neg logit contributions | |
nodded | -0.097 |
acts | -0.097 |
ver | -0.096 |
youngest | -0.095 |
Let | -0.095 |
Token | was |
Token ID | 373 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
youngest | +0.043 |
ver | +0.043 |
acts | +0.042 |
nodded | +0.042 |
Let | +0.041 |
Neg logit contributions | |
poisonous | -0.115 |
ocket | -0.107 |
ocol | -0.107 |
aster | -0.105 |
arter | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | -9.75e-01 |
Pos logit contributions | |
led | +0.002 |
ver | +0.002 |
acts | +0.002 |
nodded | +0.002 |
youngest | +0.002 |
Neg logit contributions | |
poisonous | -0.002 |
ocol | -0.002 |
ocket | -0.002 |
aster | -0.002 |
arter | -0.002 |
Token | She |
Token ID | 1375 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
poisonous | +0.047 |
ocket | +0.044 |
ocol | +0.043 |
aster | +0.043 |
arter | +0.042 |
Neg logit contributions | |
youngest | -0.032 |
acts | -0.032 |
ver | -0.032 |
folk | -0.031 |
nodded | -0.031 |
Token | t |
Token ID | 256 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
poisonous | +0.173 |
ocket | +0.162 |
aster | +0.152 |
bamboo | +0.152 |
ocol | +0.150 |
Neg logit contributions | |
folk | -0.092 |
ver | -0.090 |
bars | -0.089 |
youngest | -0.088 |
acts | -0.088 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.059 |
ocol | +0.054 |
aster | +0.054 |
ocket | +0.054 |
arter | +0.053 |
Neg logit contributions | |
ver | -0.016 |
nodded | -0.016 |
bars | -0.016 |
youngest | -0.016 |
Let | -0.015 |
Token | o |
Token ID | 78 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
ver | +0.440 |
nodded | +0.440 |
Let | +0.434 |
bars | +0.428 |
youngest | +0.420 |
Neg logit contributions | |
poisonous | -0.416 |
ocol | -0.383 |
ocket | -0.375 |
arter | -0.360 |
aster | -0.359 |
Token | ed |
Token ID | 276 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
poisonous | +0.156 |
ocket | +0.144 |
bamboo | +0.135 |
eds | +0.130 |
aster | +0.129 |
Neg logit contributions | |
acts | -0.135 |
youngest | -0.135 |
folk | -0.134 |
mun | -0.132 |
basics | -0.130 |
Token | to |
Token ID | 284 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
nodded | +0.091 |
led | +0.091 |
Let | +0.090 |
ver | +0.086 |
acts | +0.086 |
Neg logit contributions | |
poisonous | -0.121 |
ocol | -0.116 |
arter | -0.115 |
ocket | -0.114 |
aster | -0.111 |
Token | the |
Token ID | 262 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
poisonous | +0.010 |
ocket | +0.010 |
ocol | +0.010 |
arter | +0.009 |
affle | +0.009 |
Neg logit contributions | |
led | -0.005 |
Let | -0.004 |
youngest | -0.004 |
asks | -0.004 |
asked | -0.004 |
Token | big |
Token ID | 1263 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
youngest | +0.042 |
oldest | +0.041 |
Let | +0.041 |
led | +0.040 |
smallest | +0.040 |
Neg logit contributions | |
ocol | -0.071 |
aster | -0.066 |
ocket | -0.066 |
arter | -0.065 |
poisonous | -0.064 |
Token | of |
Token ID | 286 |
Feature activation | -8.86e-01 |
Pos logit contributions | |
led | +0.022 |
acts | +0.021 |
nodded | +0.020 |
leads | +0.020 |
youngest | +0.020 |
Neg logit contributions | |
poisonous | -0.031 |
ocket | -0.030 |
ocol | -0.030 |
arter | -0.030 |
affle | -0.028 |
Token | her |
Token ID | 607 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
poisonous | +0.044 |
ocket | +0.043 |
affle | +0.043 |
arter | +0.043 |
aker | +0.042 |
Neg logit contributions | |
led | -0.030 |
way | -0.028 |
acts | -0.027 |
leads | -0.026 |
youngest | -0.026 |
Token | bed |
Token ID | 3996 |
Feature activation | -5.08e+00 |
Pos logit contributions | |
poisonous | +0.110 |
ocol | +0.102 |
aster | +0.101 |
arter | +0.100 |
ocket | +0.098 |
Neg logit contributions | |
youngest | -0.099 |
led | -0.099 |
acts | -0.097 |
oldest | -0.095 |
nodded | -0.095 |
Token | and |
Token ID | 290 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
poisonous | +0.229 |
ocol | +0.213 |
ocket | +0.207 |
aster | +0.203 |
arter | +0.201 |
Neg logit contributions | |
nodded | -0.190 |
ver | -0.189 |
youngest | -0.188 |
acts | -0.187 |
led | -0.184 |
Token | t |
Token ID | 256 |
Feature activation | -8.80e-01 |
Pos logit contributions | |
poisonous | +0.008 |
aster | +0.008 |
ocket | +0.008 |
bamboo | +0.007 |
heaven | +0.007 |
Neg logit contributions | |
folk | -0.005 |
ver | -0.005 |
acts | -0.005 |
youngest | -0.005 |
basics | -0.005 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.059 |
ocol | +0.054 |
ocket | +0.054 |
aster | +0.053 |
arter | +0.053 |
Neg logit contributions | |
bars | -0.015 |
youngest | -0.015 |
nodded | -0.015 |
ver | -0.015 |
Let | -0.014 |
Token | o |
Token ID | 78 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
Let | +0.406 |
nodded | +0.406 |
bars | +0.405 |
ver | +0.404 |
led | +0.379 |
Neg logit contributions | |
poisonous | -0.460 |
ocol | -0.426 |
ocket | -0.419 |
arter | -0.403 |
aster | -0.387 |
Token | ed |
Token ID | 276 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
poisonous | +0.164 |
ocket | +0.152 |
aster | +0.144 |
bamboo | +0.141 |
injuries | +0.139 |
Neg logit contributions | |
acts | -0.134 |
folk | -0.133 |
youngest | -0.130 |
mun | -0.129 |
ver | -0.129 |
Token | to |
Token ID | 284 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
led | +0.087 |
Let | +0.083 |
nodded | +0.083 |
youngest | +0.078 |
leads | +0.077 |
Neg logit contributions | |
arter | -0.126 |
poisonous | -0.125 |
ocol | -0.124 |
ocket | -0.121 |
aster | -0.116 |
Token | the |
Token ID | 262 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
poisonous | +0.017 |
affle | +0.016 |
arter | +0.016 |
ocol | +0.016 |
ocket | +0.016 |
Neg logit contributions | |
led | -0.009 |
Let | -0.008 |
youngest | -0.008 |
asked | -0.008 |
acts | -0.008 |
Token | wardrobe |
Token ID | 35593 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
youngest | +0.040 |
oldest | +0.040 |
way | +0.038 |
nic | +0.037 |
led | +0.036 |
Neg logit contributions | |
ocol | -0.070 |
arter | -0.067 |
aster | -0.064 |
ez | -0.064 |
poisonous | -0.064 |
Token | She |
Token ID | 1375 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
poisonous | +0.015 |
bamboo | +0.013 |
aster | +0.013 |
ocol | +0.013 |
arter | +0.013 |
Neg logit contributions | |
acts | -0.012 |
folk | -0.012 |
ver | -0.012 |
youngest | -0.012 |
mun | -0.012 |
Token | stood |
Token ID | 6204 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
poisonous | +0.142 |
ocket | +0.126 |
aster | +0.124 |
bamboo | +0.120 |
isk | +0.116 |
Neg logit contributions | |
folk | -0.110 |
sets | -0.104 |
acts | -0.104 |
bars | -0.102 |
youngest | -0.098 |
Token | on |
Token ID | 319 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
nodded | +0.031 |
nic | +0.028 |
youngest | +0.027 |
best | +0.027 |
Let | +0.027 |
Neg logit contributions | |
poisonous | -0.053 |
ocol | -0.052 |
arter | -0.051 |
ocket | -0.051 |
aster | -0.050 |
Token | her |
Token ID | 607 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
poisonous | +0.051 |
ocket | +0.050 |
arter | +0.049 |
ocol | +0.048 |
affle | +0.045 |
Neg logit contributions | |
Let | -0.034 |
led | -0.033 |
youngest | -0.033 |
acts | -0.032 |
nodded | -0.032 |
Token | t |
Token ID | 256 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
poisonous | +0.098 |
ocol | +0.093 |
ocket | +0.089 |
aster | +0.087 |
arter | +0.087 |
Neg logit contributions | |
youngest | -0.089 |
nodded | -0.087 |
acts | -0.086 |
oldest | -0.086 |
led | -0.085 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.046 |
ocol | +0.041 |
ocket | +0.039 |
bamboo | +0.038 |
arter | +0.038 |
Neg logit contributions | |
bars | -0.032 |
youngest | -0.030 |
So | -0.029 |
eldest | -0.028 |
oldest | -0.028 |
Token | oes |
Token ID | 3028 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
ver | +0.282 |
bars | +0.280 |
nodded | +0.270 |
Let | +0.267 |
youngest | +0.253 |
Neg logit contributions | |
poisonous | -0.587 |
ocol | -0.550 |
ocket | -0.543 |
arter | -0.523 |
aster | -0.513 |
Token | and |
Token ID | 290 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
acts | +0.188 |
youngest | +0.187 |
nodded | +0.183 |
folk | +0.183 |
Let | +0.183 |
Neg logit contributions | |
poisonous | -0.188 |
ocket | -0.169 |
ocol | -0.165 |
aster | -0.164 |
arter | -0.158 |
Token | reached |
Token ID | 4251 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
poisonous | +0.006 |
aster | +0.006 |
ocket | +0.006 |
bamboo | +0.006 |
jam | +0.006 |
Neg logit contributions | |
acts | -0.005 |
folk | -0.005 |
youngest | -0.005 |
basics | -0.005 |
smallest | -0.005 |
Token | up |
Token ID | 510 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
bars | +0.139 |
led | +0.137 |
youngest | +0.137 |
ver | +0.136 |
oldest | +0.133 |
Neg logit contributions | |
poisonous | -0.112 |
ocket | -0.104 |
arter | -0.099 |
ocol | -0.098 |
aster | -0.091 |
Token | high |
Token ID | 1029 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
led | +0.083 |
nodded | +0.082 |
Let | +0.081 |
ver | +0.080 |
youngest | +0.079 |
Neg logit contributions | |
poisonous | -0.123 |
ocket | -0.121 |
ocol | -0.118 |
arter | -0.117 |
aster | -0.109 |
Token | her |
Token ID | 607 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
ocol | +0.159 |
arter | +0.153 |
poisonous | +0.151 |
ocket | +0.150 |
eds | +0.145 |
Neg logit contributions | |
led | -0.119 |
youngest | -0.112 |
neither | -0.110 |
Let | -0.109 |
oldest | -0.108 |
Token | chance |
Token ID | 2863 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
poisonous | +0.127 |
ocol | +0.125 |
ocket | +0.119 |
aster | +0.119 |
arter | +0.117 |
Neg logit contributions | |
youngest | -0.082 |
nodded | -0.078 |
oldest | -0.078 |
acts | -0.077 |
led | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
nodded | +0.027 |
acts | +0.027 |
youngest | +0.027 |
ver | +0.027 |
Let | +0.026 |
Neg logit contributions | |
poisonous | -0.031 |
ocol | -0.029 |
ocket | -0.028 |
aster | -0.028 |
arter | -0.027 |
Token | She |
Token ID | 1375 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
poisonous | +0.027 |
ocket | +0.025 |
ocol | +0.025 |
aster | +0.024 |
arter | +0.024 |
Neg logit contributions | |
acts | -0.018 |
ver | -0.018 |
folk | -0.018 |
youngest | -0.018 |
nodded | -0.018 |
Token | t |
Token ID | 256 |
Feature activation | -8.77e-01 |
Pos logit contributions | |
poisonous | +0.164 |
ocket | +0.154 |
aster | +0.146 |
bamboo | +0.141 |
ocol | +0.139 |
Neg logit contributions | |
folk | -0.105 |
ver | -0.102 |
sets | -0.099 |
acts | -0.099 |
youngest | -0.096 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.053 |
ocol | +0.049 |
ocket | +0.048 |
aster | +0.048 |
arter | +0.048 |
Neg logit contributions | |
youngest | -0.020 |
ver | -0.020 |
nodded | -0.020 |
bars | -0.020 |
Let | -0.020 |
Token | o |
Token ID | 78 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
ver | +0.409 |
Let | +0.407 |
nodded | +0.407 |
bars | +0.403 |
youngest | +0.374 |
Neg logit contributions | |
poisonous | -0.461 |
ocol | -0.434 |
ocket | -0.420 |
arter | -0.405 |
aster | -0.392 |
Token | ed |
Token ID | 276 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
poisonous | +0.166 |
ocket | +0.154 |
bamboo | +0.143 |
aster | +0.142 |
injuries | +0.139 |
Neg logit contributions | |
acts | -0.126 |
youngest | -0.124 |
folk | -0.124 |
ver | -0.121 |
mun | -0.121 |
Token | to |
Token ID | 284 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
led | +0.088 |
Let | +0.086 |
nodded | +0.085 |
youngest | +0.081 |
ver | +0.080 |
Neg logit contributions | |
poisonous | -0.125 |
ocol | -0.123 |
arter | -0.122 |
ocket | -0.119 |
aster | -0.118 |
Token | the |
Token ID | 262 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
poisonous | +0.020 |
ocol | +0.019 |
affle | +0.018 |
arter | +0.018 |
ocket | +0.018 |
Neg logit contributions | |
led | -0.011 |
Let | -0.010 |
youngest | -0.010 |
oldest | -0.010 |
way | -0.009 |
Token | oven |
Token ID | 14361 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
youngest | +0.035 |
oldest | +0.035 |
Let | +0.034 |
nic | +0.032 |
way | +0.032 |
Neg logit contributions | |
ocol | -0.061 |
ocket | -0.057 |
poisonous | -0.057 |
arter | -0.057 |
ez | -0.056 |
Token | time |
Token ID | 640 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
led | +0.013 |
youngest | +0.013 |
nodded | +0.012 |
Let | +0.012 |
ver | +0.012 |
Neg logit contributions | |
ocket | -0.021 |
ocol | -0.021 |
poisonous | -0.021 |
arter | -0.020 |
aster | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -8.87e-04 |
Pos logit contributions | |
ver | +0.068 |
nodded | +0.066 |
led | +0.066 |
Let | +0.066 |
youngest | +0.065 |
Neg logit contributions | |
poisonous | -0.079 |
ocol | -0.077 |
ocket | -0.072 |
arter | -0.072 |
bamboo | -0.071 |
Token | She |
Token ID | 1375 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
poisonous | +0.000 |
aster | +0.000 |
heaven | +0.000 |
bamboo | +0.000 |
what | +0.000 |
Neg logit contributions | |
acts | -0.000 |
youngest | -0.000 |
oldest | -0.000 |
bars | -0.000 |
spe | -0.000 |
Token | quietly |
Token ID | 12703 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
poisonous | +0.152 |
ocket | +0.135 |
aster | +0.134 |
bamboo | +0.131 |
ocol | +0.124 |
Neg logit contributions | |
folk | -0.086 |
ver | -0.083 |
sets | -0.083 |
bars | -0.079 |
acts | -0.079 |
Token | t |
Token ID | 256 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
ver | +0.136 |
folk | +0.134 |
youngest | +0.128 |
acts | +0.126 |
oldest | +0.126 |
Neg logit contributions | |
poisonous | -0.227 |
aster | -0.210 |
ocket | -0.205 |
ocol | -0.201 |
bamboo | -0.192 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.048 |
ocol | +0.045 |
ocket | +0.045 |
aster | +0.045 |
arter | +0.044 |
Neg logit contributions | |
ver | -0.013 |
nodded | -0.013 |
youngest | -0.013 |
acts | -0.013 |
Let | -0.013 |
Token | o |
Token ID | 78 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
ver | +0.416 |
nodded | +0.414 |
bars | +0.412 |
Let | +0.408 |
youngest | +0.396 |
Neg logit contributions | |
poisonous | -0.443 |
ocol | -0.404 |
ocket | -0.401 |
arter | -0.387 |
aster | -0.377 |
Token | ed |
Token ID | 276 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
poisonous | +0.153 |
ocket | +0.140 |
aster | +0.132 |
bamboo | +0.127 |
injuries | +0.126 |
Neg logit contributions | |
acts | -0.119 |
youngest | -0.115 |
folk | -0.114 |
ver | -0.114 |
oldest | -0.113 |
Token | down |
Token ID | 866 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
nodded | +0.087 |
led | +0.087 |
Let | +0.086 |
youngest | +0.084 |
ver | +0.083 |
Neg logit contributions | |
poisonous | -0.123 |
arter | -0.118 |
ocol | -0.117 |
ocket | -0.116 |
aster | -0.113 |
Token | the |
Token ID | 262 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
poisonous | +0.013 |
ocol | +0.012 |
arter | +0.012 |
ocket | +0.012 |
aster | +0.011 |
Neg logit contributions | |
led | -0.009 |
nodded | -0.008 |
Let | -0.008 |
ver | -0.008 |
youngest | -0.008 |
Token | street |
Token ID | 4675 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
youngest | +0.049 |
bars | +0.048 |
oldest | +0.047 |
Let | +0.047 |
led | +0.045 |
Neg logit contributions | |
ocol | -0.082 |
poisonous | -0.081 |
ocket | -0.079 |
arter | -0.077 |
aster | -0.073 |
Token |
|
Token ID | 198 |
Feature activation | +8.64e-01 |
Pos logit contributions | |
acts | +0.005 |
folk | +0.005 |
ver | +0.005 |
bars | +0.005 |
nodded | +0.005 |
Neg logit contributions | |
poisonous | -0.005 |
ocol | -0.005 |
ocket | -0.005 |
bamboo | -0.005 |
aster | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +8.64e-01 |
Pos logit contributions | |
acts | +0.036 |
youngest | +0.034 |
folk | +0.034 |
nodded | +0.034 |
bars | +0.034 |
Neg logit contributions | |
poisonous | -0.035 |
aster | -0.031 |
ocket | -0.031 |
ocol | -0.031 |
bamboo | -0.030 |
Token | M |
Token ID | 44 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
acts | +0.037 |
folk | +0.035 |
bars | +0.034 |
ver | +0.034 |
youngest | +0.034 |
Neg logit contributions | |
poisonous | -0.036 |
aster | -0.032 |
ocol | -0.031 |
ocket | -0.031 |
bamboo | -0.030 |
Token | olly |
Token ID | 5098 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
ver | +0.035 |
acts | +0.035 |
nodded | +0.035 |
youngest | +0.035 |
Let | +0.034 |
Neg logit contributions | |
poisonous | -0.043 |
ocol | -0.039 |
ocket | -0.039 |
aster | -0.038 |
arter | -0.037 |
Token | t |
Token ID | 256 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
ver | +0.022 |
bars | +0.021 |
sets | +0.021 |
folk | +0.021 |
mun | +0.020 |
Neg logit contributions | |
poisonous | -0.039 |
ocket | -0.036 |
aster | -0.036 |
bamboo | -0.034 |
ez | -0.033 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.041 |
ocol | +0.039 |
ocket | +0.038 |
aster | +0.038 |
arter | +0.037 |
Neg logit contributions | |
acts | -0.020 |
ver | -0.019 |
nodded | -0.019 |
youngest | -0.019 |
Let | -0.019 |
Token | o |
Token ID | 78 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
acts | +0.452 |
youngest | +0.451 |
ver | +0.450 |
nodded | +0.450 |
Let | +0.443 |
Neg logit contributions | |
poisonous | -0.386 |
ocol | -0.344 |
ocket | -0.343 |
aster | -0.334 |
arter | -0.327 |
Token | ed |
Token ID | 276 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
poisonous | +0.146 |
ocket | +0.132 |
bamboo | +0.124 |
aster | +0.124 |
eds | +0.122 |
Neg logit contributions | |
acts | -0.115 |
ver | -0.111 |
basics | -0.110 |
folk | -0.110 |
youngest | -0.110 |
Token | around |
Token ID | 1088 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
nodded | +0.094 |
youngest | +0.093 |
ver | +0.092 |
Let | +0.092 |
acts | +0.092 |
Neg logit contributions | |
poisonous | -0.114 |
ocol | -0.106 |
ocket | -0.105 |
arter | -0.104 |
aster | -0.103 |
Token | the |
Token ID | 262 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
nodded | +0.067 |
led | +0.067 |
Let | +0.065 |
acts | +0.064 |
youngest | +0.062 |
Neg logit contributions | |
poisonous | -0.096 |
arter | -0.091 |
ocket | -0.090 |
ocol | -0.088 |
aster | -0.084 |
Token | corner |
Token ID | 5228 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
youngest | +0.057 |
oldest | +0.055 |
Let | +0.054 |
nodded | +0.052 |
eldest | +0.052 |
Neg logit contributions | |
poisonous | -0.084 |
ocol | -0.083 |
ocket | -0.081 |
arter | -0.079 |
aster | -0.074 |
Token | with |
Token ID | 351 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
nodded | +0.155 |
's | +0.148 |
acts | +0.144 |
mun | +0.137 |
So | +0.137 |
Neg logit contributions | |
poisonous | -0.221 |
ocol | -0.209 |
injuries | -0.203 |
ocket | -0.203 |
arter | -0.202 |
Token | it |
Token ID | 340 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
poisonous | +0.006 |
arter | +0.005 |
ocket | +0.005 |
ocol | +0.005 |
prise | +0.005 |
Neg logit contributions | |
acts | -0.004 |
led | -0.004 |
nodded | -0.004 |
youngest | -0.004 |
Let | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
poisonous | +0.025 |
ocol | +0.024 |
aster | +0.023 |
ocket | +0.023 |
arter | +0.022 |
Neg logit contributions | |
youngest | -0.024 |
led | -0.024 |
nodded | -0.024 |
ver | -0.024 |
acts | -0.024 |
Token | She |
Token ID | 1375 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
poisonous | +0.022 |
aster | +0.020 |
bamboo | +0.020 |
ocol | +0.019 |
ocket | +0.019 |
Neg logit contributions | |
acts | -0.018 |
folk | -0.018 |
ver | -0.017 |
youngest | -0.017 |
oldest | -0.017 |
Token | t |
Token ID | 256 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
poisonous | +0.167 |
ocket | +0.152 |
bamboo | +0.149 |
aster | +0.147 |
injuries | +0.139 |
Neg logit contributions | |
folk | -0.103 |
bars | -0.098 |
ver | -0.097 |
sets | -0.095 |
acts | -0.095 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.060 |
ocket | +0.056 |
ocol | +0.056 |
aster | +0.055 |
arter | +0.055 |
Neg logit contributions | |
ver | -0.015 |
nodded | -0.015 |
youngest | -0.015 |
acts | -0.014 |
Let | -0.014 |
Token | o |
Token ID | 78 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
Let | +0.418 |
ver | +0.417 |
nodded | +0.415 |
bars | +0.408 |
led | +0.386 |
Neg logit contributions | |
poisonous | -0.447 |
ocol | -0.416 |
ocket | -0.408 |
arter | -0.392 |
aster | -0.381 |
Token | ed |
Token ID | 276 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
poisonous | +0.161 |
ocket | +0.144 |
bamboo | +0.138 |
eds | +0.136 |
aster | +0.135 |
Neg logit contributions | |
acts | -0.135 |
folk | -0.130 |
youngest | -0.127 |
bars | -0.127 |
basics | -0.126 |
Token | closer |
Token ID | 5699 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
led | +0.087 |
Let | +0.086 |
nodded | +0.085 |
youngest | +0.081 |
ver | +0.080 |
Neg logit contributions | |
poisonous | -0.122 |
ocol | -0.120 |
arter | -0.119 |
ocket | -0.118 |
aster | -0.115 |
Token | and |
Token ID | 290 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
nodded | +0.108 |
ver | +0.108 |
acts | +0.107 |
youngest | +0.107 |
Let | +0.106 |
Neg logit contributions | |
poisonous | -0.123 |
ocol | -0.112 |
ocket | -0.111 |
aster | -0.108 |
arter | -0.107 |
Token | reached |
Token ID | 4251 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
poisonous | +0.007 |
ocket | +0.006 |
aster | +0.006 |
ocol | +0.006 |
bamboo | +0.006 |
Neg logit contributions | |
acts | -0.005 |
youngest | -0.005 |
folk | -0.005 |
ver | -0.005 |
bars | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
acts | +0.098 |
youngest | +0.094 |
led | +0.093 |
Let | +0.093 |
oldest | +0.090 |
Neg logit contributions | |
poisonous | -0.136 |
ocket | -0.126 |
ocol | -0.124 |
arter | -0.122 |
aster | -0.118 |
Token | table |
Token ID | 3084 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
oldest | +0.036 |
way | +0.036 |
Let | +0.036 |
's | +0.036 |
youngest | +0.035 |
Neg logit contributions | |
ocol | -0.070 |
ez | -0.066 |
poisonous | -0.065 |
ocket | -0.064 |
arter | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
poisonous | +0.084 |
ocol | +0.077 |
ocket | +0.074 |
aster | +0.072 |
arter | +0.072 |
Neg logit contributions | |
nodded | -0.082 |
acts | -0.082 |
youngest | -0.081 |
led | -0.081 |
ver | -0.081 |
Token | They |
Token ID | 1119 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
poisonous | +0.022 |
aster | +0.020 |
bamboo | +0.020 |
ores | +0.020 |
ocket | +0.020 |
Neg logit contributions | |
bars | -0.020 |
sets | -0.020 |
folk | -0.020 |
mun | -0.019 |
lest | -0.019 |
Token | t |
Token ID | 256 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
sets | +0.027 |
bars | +0.027 |
basics | +0.026 |
acts | +0.026 |
hardest | +0.025 |
Neg logit contributions | |
poisonous | -0.036 |
ocket | -0.034 |
aster | -0.032 |
injuries | -0.031 |
what | -0.031 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.061 |
ocol | +0.057 |
ocket | +0.057 |
aster | +0.056 |
arter | +0.056 |
Neg logit contributions | |
acts | -0.018 |
youngest | -0.018 |
ver | -0.018 |
nodded | -0.018 |
Let | -0.017 |
Token | o |
Token ID | 78 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
acts | +0.462 |
youngest | +0.461 |
ver | +0.460 |
nodded | +0.460 |
Let | +0.453 |
Neg logit contributions | |
poisonous | -0.376 |
ocol | -0.334 |
ocket | -0.333 |
aster | -0.324 |
arter | -0.317 |
Token | ed |
Token ID | 276 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
poisonous | +0.159 |
ocket | +0.147 |
eds | +0.133 |
aster | +0.131 |
bamboo | +0.131 |
Neg logit contributions | |
acts | -0.144 |
youngest | -0.142 |
basics | -0.139 |
folk | -0.138 |
oldest | -0.138 |
Token | to |
Token ID | 284 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
nodded | +0.094 |
ver | +0.094 |
acts | +0.093 |
Let | +0.093 |
youngest | +0.093 |
Neg logit contributions | |
poisonous | -0.111 |
ocol | -0.102 |
ocket | -0.102 |
aster | -0.099 |
arter | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
poisonous | +0.014 |
ocket | +0.013 |
ocol | +0.013 |
arter | +0.012 |
affle | +0.012 |
Neg logit contributions | |
Let | -0.009 |
led | -0.009 |
youngest | -0.009 |
acts | -0.008 |
nodded | -0.008 |
Token | pot |
Token ID | 1787 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
youngest | +0.042 |
oldest | +0.041 |
Let | +0.040 |
led | +0.039 |
eldest | +0.039 |
Neg logit contributions | |
ocol | -0.064 |
poisonous | -0.061 |
ocket | -0.059 |
arter | -0.059 |
aster | -0.057 |
Token | to |
Token ID | 284 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
poisonous | +0.059 |
bamboo | +0.049 |
ocket | +0.045 |
aster | +0.043 |
what | +0.043 |
Neg logit contributions | |
bars | -0.096 |
folk | -0.093 |
acts | -0.093 |
ver | -0.093 |
lest | -0.090 |
Token | stand |
Token ID | 1302 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
poisonous | +0.007 |
ocol | +0.006 |
aster | +0.006 |
ocket | +0.006 |
arter | +0.006 |
Neg logit contributions | |
ver | -0.005 |
nodded | -0.005 |
acts | -0.005 |
youngest | -0.005 |
bars | -0.005 |
Token | on |
Token ID | 319 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
poisonous | +0.047 |
ocol | +0.045 |
aster | +0.044 |
ocket | +0.044 |
arter | +0.042 |
Neg logit contributions | |
nodded | -0.041 |
ver | -0.041 |
led | -0.040 |
Let | -0.039 |
youngest | -0.039 |
Token | their |
Token ID | 511 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
poisonous | +0.044 |
ocol | +0.040 |
ocket | +0.040 |
aster | +0.039 |
arter | +0.039 |
Neg logit contributions | |
acts | -0.043 |
youngest | -0.042 |
nodded | -0.042 |
ver | -0.042 |
Let | -0.042 |
Token | t |
Token ID | 256 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
poisonous | +0.006 |
ocket | +0.005 |
ocol | +0.005 |
aster | +0.005 |
arter | +0.005 |
Neg logit contributions | |
ver | -0.003 |
nodded | -0.003 |
acts | -0.003 |
youngest | -0.003 |
folk | -0.003 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.037 |
ocol | +0.033 |
ocket | +0.032 |
aster | +0.032 |
arter | +0.030 |
Neg logit contributions | |
bars | -0.037 |
youngest | -0.036 |
So | -0.035 |
oldest | -0.035 |
ver | -0.035 |
Token | oes |
Token ID | 3028 |
Feature activation | +4.59e+00 |
Pos logit contributions | |
ver | +0.214 |
nodded | +0.209 |
youngest | +0.209 |
acts | +0.205 |
Let | +0.204 |
Neg logit contributions | |
poisonous | -0.628 |
ocol | -0.588 |
ocket | -0.586 |
aster | -0.572 |
arter | -0.568 |
Token | to |
Token ID | 284 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
acts | +0.195 |
youngest | +0.194 |
nodded | +0.192 |
Let | +0.191 |
folk | +0.190 |
Neg logit contributions | |
poisonous | -0.178 |
ocket | -0.159 |
aster | -0.153 |
ocol | -0.151 |
bamboo | -0.150 |
Token | see |
Token ID | 766 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
poisonous | +0.007 |
ocol | +0.007 |
ocket | +0.007 |
arter | +0.006 |
aster | +0.006 |
Neg logit contributions | |
acts | -0.006 |
youngest | -0.006 |
Let | -0.005 |
led | -0.005 |
nodded | -0.005 |
Token | some |
Token ID | 617 |
Feature activation | +3.27e+00 |
Pos logit contributions | |
poisonous | +0.144 |
ocol | +0.129 |
ocket | +0.126 |
aster | +0.122 |
bamboo | +0.120 |
Neg logit contributions | |
nodded | -0.171 |
ver | -0.170 |
youngest | -0.169 |
acts | -0.168 |
folk | -0.167 |
Token | of |
Token ID | 286 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
acts | +0.099 |
youngest | +0.098 |
nodded | +0.097 |
Let | +0.095 |
oldest | +0.094 |
Neg logit contributions | |
poisonous | -0.169 |
ocket | -0.159 |
ocol | -0.157 |
aster | -0.156 |
arter | -0.154 |
Token | the |
Token ID | 262 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
Let | +0.062 |
youngest | +0.058 |
acts | +0.058 |
nodded | +0.057 |
led | +0.056 |
Neg logit contributions | |
ocket | -0.106 |
poisonous | -0.105 |
arter | -0.103 |
affle | -0.099 |
ocol | -0.096 |
Token | wind |
Token ID | 2344 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
Let | +0.040 |
nodded | +0.038 |
youngest | +0.037 |
oldest | +0.036 |
way | +0.034 |
Neg logit contributions | |
ocol | -0.095 |
arter | -0.094 |
poisonous | -0.094 |
ocket | -0.091 |
affle | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
nodded | +0.137 |
youngest | +0.135 |
ver | +0.135 |
acts | +0.135 |
bars | +0.133 |
Neg logit contributions | |
poisonous | -0.137 |
ocol | -0.125 |
ocket | -0.125 |
aster | -0.121 |
arter | -0.119 |
Token | He |
Token ID | 679 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
poisonous | +0.001 |
aster | +0.001 |
bamboo | +0.001 |
ocol | +0.000 |
ocket | +0.000 |
Neg logit contributions | |
acts | -0.001 |
mun | -0.001 |
youngest | -0.001 |
ver | -0.001 |
folk | -0.001 |
Token | t |
Token ID | 256 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
poisonous | +0.060 |
ocket | +0.054 |
aster | +0.053 |
bamboo | +0.052 |
ocol | +0.050 |
Neg logit contributions | |
folk | -0.033 |
ver | -0.032 |
acts | -0.031 |
sets | -0.031 |
youngest | -0.030 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.057 |
ocol | +0.053 |
ocket | +0.053 |
aster | +0.053 |
arter | +0.052 |
Neg logit contributions | |
nodded | -0.013 |
ver | -0.013 |
youngest | -0.012 |
bars | -0.012 |
Let | -0.012 |
Token | o |
Token ID | 78 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
nodded | +0.429 |
ver | +0.429 |
Let | +0.426 |
bars | +0.420 |
youngest | +0.409 |
Neg logit contributions | |
poisonous | -0.425 |
ocol | -0.391 |
ocket | -0.386 |
arter | -0.367 |
aster | -0.366 |
Token | ed |
Token ID | 276 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
poisonous | +0.150 |
ocket | +0.137 |
bamboo | +0.129 |
sm | +0.125 |
aster | +0.125 |
Neg logit contributions | |
acts | -0.120 |
folk | -0.119 |
youngest | -0.119 |
mun | -0.117 |
ver | -0.117 |
Token | closer |
Token ID | 5699 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
nodded | +0.087 |
Let | +0.086 |
led | +0.085 |
ver | +0.085 |
acts | +0.085 |
Neg logit contributions | |
poisonous | -0.119 |
ocol | -0.111 |
ocket | -0.110 |
arter | -0.109 |
aster | -0.109 |
Token | and |
Token ID | 290 |
Feature activation | +8.42e-02 |
Pos logit contributions | |
youngest | +0.113 |
acts | +0.112 |
ver | +0.112 |
folk | +0.111 |
nodded | +0.110 |
Neg logit contributions | |
poisonous | -0.119 |
ocket | -0.106 |
ocol | -0.106 |
aster | -0.103 |
arter | -0.101 |
Token | took |
Token ID | 1718 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
acts | +0.003 |
youngest | +0.003 |
folk | +0.003 |
ver | +0.003 |
bars | +0.003 |
Neg logit contributions | |
poisonous | -0.004 |
ocket | -0.004 |
aster | -0.004 |
ocol | -0.003 |
bamboo | -0.003 |
Token | room |
Token ID | 2119 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
ocol | +0.133 |
poisonous | +0.128 |
ocket | +0.120 |
arter | +0.118 |
aster | +0.117 |
Neg logit contributions | |
youngest | -0.090 |
led | -0.089 |
's | -0.088 |
So | -0.087 |
oldest | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | -9.18e-01 |
Pos logit contributions | |
poisonous | +0.106 |
ocol | +0.100 |
ocket | +0.096 |
arter | +0.094 |
aster | +0.093 |
Neg logit contributions | |
nodded | -0.101 |
ver | -0.098 |
youngest | -0.098 |
led | -0.098 |
acts | -0.097 |
Token | Then |
Token ID | 3244 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
poisonous | +0.054 |
ocol | +0.052 |
ocket | +0.052 |
aster | +0.049 |
arter | +0.049 |
Neg logit contributions | |
Let | -0.024 |
So | -0.023 |
nodded | -0.023 |
led | -0.022 |
youngest | -0.022 |
Token | they |
Token ID | 484 |
Feature activation | +4.43e+00 |
Pos logit contributions | |
youngest | +0.007 |
acts | +0.007 |
ver | +0.007 |
nodded | +0.007 |
Let | +0.007 |
Neg logit contributions | |
poisonous | -0.014 |
ocol | -0.013 |
ocket | -0.013 |
aster | -0.013 |
arter | -0.012 |
Token | t |
Token ID | 256 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
nodded | +0.115 |
ver | +0.114 |
youngest | +0.114 |
acts | +0.113 |
Let | +0.111 |
Neg logit contributions | |
poisonous | -0.250 |
ocol | -0.234 |
ocket | -0.231 |
aster | -0.227 |
arter | -0.226 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.054 |
ocol | +0.050 |
ocket | +0.050 |
aster | +0.049 |
arter | +0.048 |
Neg logit contributions | |
ver | -0.019 |
nodded | -0.019 |
bars | -0.019 |
youngest | -0.019 |
Let | -0.019 |
Token | o |
Token ID | 78 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
ver | +0.418 |
nodded | +0.417 |
Let | +0.413 |
bars | +0.411 |
youngest | +0.404 |
Neg logit contributions | |
poisonous | -0.432 |
ocol | -0.395 |
ocket | -0.391 |
arter | -0.375 |
aster | -0.371 |
Token | ed |
Token ID | 276 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
poisonous | +0.160 |
ocket | +0.146 |
aster | +0.139 |
bamboo | +0.135 |
sm | +0.134 |
Neg logit contributions | |
acts | -0.139 |
folk | -0.135 |
youngest | -0.132 |
mun | -0.130 |
basics | -0.129 |
Token | out |
Token ID | 503 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
led | +0.090 |
Let | +0.090 |
nodded | +0.090 |
ver | +0.088 |
youngest | +0.088 |
Neg logit contributions | |
poisonous | -0.114 |
ocol | -0.110 |
ocket | -0.107 |
arter | -0.106 |
aster | -0.104 |
Token | of |
Token ID | 286 |
Feature activation | -8.91e-01 |
Pos logit contributions | |
nodded | +0.026 |
led | +0.026 |
ver | +0.026 |
youngest | +0.026 |
acts | +0.026 |
Neg logit contributions | |
poisonous | -0.032 |
ocket | -0.029 |
ocol | -0.029 |
aster | -0.028 |
arter | -0.028 |
Token | their |
Token ID | 511 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
ocket | +0.051 |
poisonous | +0.050 |
ocol | +0.048 |
arter | +0.048 |
affle | +0.047 |
Neg logit contributions | |
led | -0.024 |
way | -0.022 |
youngest | -0.021 |
Let | -0.021 |
leads | -0.020 |
Token | sword |
Token ID | 8429 |
Feature activation | +9.07e-01 |
Pos logit contributions | |
poisonous | +0.022 |
ocol | +0.020 |
arter | +0.020 |
aster | +0.020 |
ores | +0.020 |
Neg logit contributions | |
acts | -0.022 |
youngest | -0.021 |
led | -0.020 |
oldest | -0.020 |
leads | -0.020 |
Token | too |
Token ID | 1165 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
Let | +0.045 |
youngest | +0.044 |
acts | +0.044 |
oldest | +0.043 |
mun | +0.043 |
Neg logit contributions | |
poisonous | -0.030 |
ocket | -0.027 |
bamboo | -0.024 |
aster | -0.024 |
ocol | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -7.97e-01 |
Pos logit contributions | |
Let | +0.013 |
acts | +0.013 |
youngest | +0.013 |
folk | +0.013 |
oldest | +0.013 |
Neg logit contributions | |
poisonous | -0.008 |
aster | -0.006 |
injuries | -0.006 |
ocket | -0.006 |
bamboo | -0.006 |
Token | They |
Token ID | 1119 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
poisonous | +0.037 |
arter | +0.034 |
bamboo | +0.034 |
aster | +0.033 |
ocol | +0.032 |
Neg logit contributions | |
acts | -0.028 |
youngest | -0.028 |
folk | -0.027 |
bars | -0.027 |
oldest | -0.027 |
Token | t |
Token ID | 256 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
acts | +0.022 |
sets | +0.022 |
folk | +0.021 |
basics | +0.021 |
bars | +0.021 |
Neg logit contributions | |
poisonous | -0.032 |
ocket | -0.030 |
bamboo | -0.029 |
aster | -0.028 |
injuries | -0.027 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.053 |
ocket | +0.049 |
ocol | +0.049 |
aster | +0.048 |
arter | +0.048 |
Neg logit contributions | |
ver | -0.016 |
nodded | -0.016 |
youngest | -0.016 |
acts | -0.016 |
Let | -0.016 |
Token | o |
Token ID | 78 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
ver | +0.455 |
nodded | +0.454 |
acts | +0.451 |
youngest | +0.451 |
Let | +0.448 |
Neg logit contributions | |
poisonous | -0.384 |
ocol | -0.344 |
ocket | -0.342 |
aster | -0.331 |
arter | -0.326 |
Token | ed |
Token ID | 276 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
poisonous | +0.124 |
ocket | +0.108 |
bamboo | +0.106 |
eds | +0.106 |
sm | +0.104 |
Neg logit contributions | |
acts | -0.164 |
mun | -0.156 |
youngest | -0.156 |
folk | -0.155 |
oldest | -0.155 |
Token | to |
Token ID | 284 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
ver | +0.096 |
nodded | +0.096 |
acts | +0.096 |
youngest | +0.096 |
Let | +0.095 |
Neg logit contributions | |
poisonous | -0.105 |
ocol | -0.095 |
ocket | -0.095 |
aster | -0.092 |
arter | -0.091 |
Token | the |
Token ID | 262 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
poisonous | +0.009 |
ocket | +0.009 |
ocol | +0.009 |
arter | +0.008 |
affle | +0.008 |
Neg logit contributions | |
youngest | -0.005 |
Let | -0.005 |
led | -0.005 |
oldest | -0.005 |
acts | -0.005 |
Token | shed |
Token ID | 14999 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
youngest | +0.046 |
oldest | +0.045 |
way | +0.044 |
smallest | +0.043 |
leader | +0.042 |
Neg logit contributions | |
ocol | -0.074 |
ez | -0.070 |
oof | -0.068 |
ake | -0.067 |
poisonous | -0.067 |
Token | brave |
Token ID | 14802 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
acts | +0.014 |
ver | +0.014 |
nodded | +0.014 |
youngest | +0.014 |
Let | +0.014 |
Neg logit contributions | |
poisonous | -0.018 |
ocol | -0.017 |
ocket | -0.017 |
aster | -0.016 |
arter | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
ver | +0.170 |
youngest | +0.169 |
acts | +0.167 |
nodded | +0.167 |
Let | +0.166 |
Neg logit contributions | |
poisonous | -0.174 |
ocket | -0.156 |
ocol | -0.155 |
aster | -0.150 |
arter | -0.148 |
Token | so |
Token ID | 523 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
ver | +0.012 |
youngest | +0.012 |
acts | +0.012 |
nodded | +0.012 |
folk | +0.012 |
Neg logit contributions | |
poisonous | -0.013 |
ocket | -0.012 |
ocol | -0.012 |
aster | -0.012 |
bamboo | -0.011 |
Token | she |
Token ID | 673 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
acts | +0.017 |
youngest | +0.017 |
ver | +0.017 |
bars | +0.017 |
folk | +0.017 |
Neg logit contributions | |
poisonous | -0.023 |
ocket | -0.021 |
aster | -0.021 |
ocol | -0.020 |
arter | -0.019 |
Token | t |
Token ID | 256 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
bars | +0.017 |
acts | +0.017 |
folk | +0.017 |
mun | +0.016 |
sets | +0.016 |
Neg logit contributions | |
poisonous | -0.025 |
ocket | -0.023 |
aster | -0.021 |
bamboo | -0.021 |
buried | -0.019 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.046 |
ocket | +0.043 |
ocol | +0.043 |
aster | +0.042 |
arter | +0.042 |
Neg logit contributions | |
acts | -0.018 |
nodded | -0.018 |
youngest | -0.018 |
ver | -0.018 |
Let | -0.018 |
Token | o |
Token ID | 78 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
ver | +0.437 |
nodded | +0.437 |
youngest | +0.432 |
acts | +0.431 |
Let | +0.430 |
Neg logit contributions | |
poisonous | -0.402 |
ocol | -0.363 |
ocket | -0.360 |
aster | -0.348 |
arter | -0.345 |
Token | ed |
Token ID | 276 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
poisonous | +0.133 |
ocket | +0.121 |
eds | +0.116 |
aster | +0.113 |
bamboo | +0.112 |
Neg logit contributions | |
acts | -0.133 |
mun | -0.125 |
folk | -0.124 |
youngest | -0.123 |
basics | -0.121 |
Token | around |
Token ID | 1088 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
led | +0.086 |
Let | +0.085 |
nodded | +0.085 |
youngest | +0.082 |
ver | +0.081 |
Neg logit contributions | |
poisonous | -0.119 |
arter | -0.115 |
ocol | -0.115 |
ocket | -0.113 |
aster | -0.109 |
Token | and |
Token ID | 290 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
led | +0.062 |
nodded | +0.061 |
Let | +0.060 |
acts | +0.059 |
youngest | +0.057 |
Neg logit contributions | |
poisonous | -0.091 |
arter | -0.085 |
ocol | -0.084 |
ocket | -0.083 |
affle | -0.080 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
acts | +0.001 |
youngest | +0.001 |
folk | +0.001 |
ver | +0.001 |
bars | +0.001 |
Neg logit contributions | |
poisonous | -0.002 |
ocket | -0.002 |
aster | -0.002 |
ocol | -0.002 |
bamboo | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
poisonous | +0.030 |
ocket | +0.027 |
aster | +0.026 |
ocol | +0.025 |
arter | +0.025 |
Neg logit contributions | |
mun | -0.037 |
ver | -0.037 |
youngest | -0.037 |
acts | -0.037 |
Let | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
ver | +0.004 |
youngest | +0.004 |
acts | +0.004 |
nodded | +0.004 |
folk | +0.004 |
Neg logit contributions | |
poisonous | -0.006 |
ocket | -0.005 |
ocol | -0.005 |
aster | -0.005 |
arter | -0.005 |
Token | then |
Token ID | 788 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
poisonous | +0.008 |
ocket | +0.007 |
ocol | +0.007 |
aster | +0.007 |
arter | +0.006 |
Neg logit contributions | |
acts | -0.007 |
ver | -0.007 |
youngest | -0.007 |
nodded | -0.006 |
folk | -0.006 |
Token | they |
Token ID | 484 |
Feature activation | +4.46e+00 |
Pos logit contributions | |
poisonous | +0.006 |
aster | +0.006 |
ocket | +0.006 |
ocol | +0.005 |
injuries | +0.005 |
Neg logit contributions | |
acts | -0.004 |
youngest | -0.004 |
folk | -0.004 |
mun | -0.004 |
ver | -0.004 |
Token | t |
Token ID | 256 |
Feature activation | -8.59e-01 |
Pos logit contributions | |
acts | +0.134 |
ver | +0.131 |
youngest | +0.131 |
nodded | +0.128 |
Let | +0.128 |
Neg logit contributions | |
poisonous | -0.234 |
ocket | -0.217 |
ocol | -0.213 |
aster | -0.212 |
arter | -0.205 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.053 |
ocol | +0.050 |
ocket | +0.050 |
aster | +0.049 |
arter | +0.049 |
Neg logit contributions | |
acts | -0.017 |
nodded | -0.017 |
ver | -0.017 |
youngest | -0.017 |
Let | -0.016 |
Token | o |
Token ID | 78 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
ver | +0.424 |
nodded | +0.423 |
youngest | +0.420 |
acts | +0.419 |
Let | +0.417 |
Neg logit contributions | |
poisonous | -0.414 |
ocol | -0.374 |
ocket | -0.373 |
aster | -0.360 |
arter | -0.357 |
Token | ed |
Token ID | 276 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
poisonous | +0.156 |
ocket | +0.144 |
aster | +0.135 |
eds | +0.131 |
injuries | +0.130 |
Neg logit contributions | |
acts | -0.140 |
mun | -0.134 |
folk | -0.132 |
youngest | -0.132 |
oldest | -0.130 |
Token | to |
Token ID | 284 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
nodded | +0.096 |
Let | +0.095 |
ver | +0.095 |
youngest | +0.095 |
led | +0.094 |
Neg logit contributions | |
poisonous | -0.113 |
ocol | -0.105 |
ocket | -0.105 |
arter | -0.103 |
aster | -0.101 |
Token | the |
Token ID | 262 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
poisonous | +0.016 |
ocket | +0.015 |
ocol | +0.015 |
arter | +0.014 |
affle | +0.014 |
Neg logit contributions | |
Let | -0.008 |
youngest | -0.008 |
led | -0.008 |
oldest | -0.008 |
acts | -0.008 |
Token | table |
Token ID | 3084 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
youngest | +0.035 |
Let | +0.034 |
oldest | +0.033 |
way | +0.032 |
eldest | +0.031 |
Neg logit contributions | |
ocol | -0.066 |
arter | -0.063 |
ez | -0.061 |
poisonous | -0.061 |
ocket | -0.061 |
Token | and |
Token ID | 290 |
Feature activation | -4.57e-03 |
Pos logit contributions | |
poisonous | +0.065 |
ocol | +0.062 |
ocket | +0.061 |
arter | +0.060 |
aster | +0.059 |
Neg logit contributions | |
ver | -0.045 |
led | -0.044 |
acts | -0.043 |
nodded | -0.043 |
bars | -0.043 |
Token | stood |
Token ID | 6204 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
poisonous | +0.000 |
ocket | +0.000 |
aster | +0.000 |
ocol | +0.000 |
arter | +0.000 |
Neg logit contributions | |
youngest | -0.000 |
acts | -0.000 |
ver | -0.000 |
folk | -0.000 |
nodded | -0.000 |
Token | on |
Token ID | 319 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
nodded | +0.029 |
nic | +0.029 |
best | +0.028 |
hardest | +0.028 |
Let | +0.027 |
Neg logit contributions | |
arter | -0.057 |
poisonous | -0.056 |
ocket | -0.054 |
aster | -0.054 |
ocol | -0.053 |
Token | her |
Token ID | 607 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
poisonous | +0.052 |
arter | +0.050 |
ocket | +0.049 |
ocol | +0.049 |
aster | +0.046 |
Neg logit contributions | |
led | -0.034 |
acts | -0.034 |
Let | -0.034 |
youngest | -0.033 |
nodded | -0.032 |
Token | t |
Token ID | 256 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
poisonous | +0.096 |
ocol | +0.093 |
aster | +0.087 |
arter | +0.087 |
ocket | +0.086 |
Neg logit contributions | |
youngest | -0.088 |
oldest | -0.085 |
led | -0.084 |
acts | -0.084 |
way | -0.084 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.038 |
ocol | +0.032 |
ocket | +0.031 |
aster | +0.030 |
arter | +0.030 |
Neg logit contributions | |
bars | -0.040 |
youngest | -0.037 |
ver | -0.037 |
So | -0.036 |
eldest | -0.035 |
Token | oes |
Token ID | 3028 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
bars | +0.250 |
ver | +0.249 |
nodded | +0.231 |
Let | +0.228 |
youngest | +0.219 |
Neg logit contributions | |
poisonous | -0.619 |
ocol | -0.581 |
ocket | -0.576 |
arter | -0.554 |
aster | -0.546 |
Token | to |
Token ID | 284 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
ver | +0.165 |
nodded | +0.162 |
youngest | +0.159 |
led | +0.159 |
acts | +0.159 |
Neg logit contributions | |
poisonous | -0.215 |
ocol | -0.199 |
ocket | -0.194 |
arter | -0.192 |
aster | -0.192 |
Token | make |
Token ID | 787 |
Feature activation | +8.84e-01 |
Pos logit contributions | |
poisonous | +0.008 |
ocket | +0.008 |
arter | +0.008 |
ocol | +0.008 |
affle | +0.007 |
Neg logit contributions | |
youngest | -0.005 |
led | -0.005 |
Let | -0.005 |
oldest | -0.005 |
acts | -0.005 |
Token | sure |
Token ID | 1654 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
youngest | +0.032 |
led | +0.032 |
acts | +0.032 |
nodded | +0.031 |
basics | +0.031 |
Neg logit contributions | |
poisonous | -0.040 |
ocket | -0.038 |
arter | -0.038 |
ocol | -0.037 |
aster | -0.035 |
Token | they |
Token ID | 484 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
youngest | +0.098 |
acts | +0.097 |
nodded | +0.096 |
ver | +0.096 |
Let | +0.095 |
Neg logit contributions | |
poisonous | -0.093 |
ocket | -0.083 |
ocol | -0.082 |
aster | -0.081 |
arter | -0.078 |
Token | cream |
Token ID | 36277 |
Feature activation | -4.58e+00 |
Pos logit contributions | |
poisonous | +0.092 |
ocket | +0.083 |
ocol | +0.081 |
arter | +0.080 |
aster | +0.077 |
Neg logit contributions | |
ver | -0.126 |
youngest | -0.126 |
nodded | -0.126 |
acts | -0.126 |
Let | -0.125 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
poisonous | +0.155 |
ocol | +0.135 |
ocket | +0.135 |
aster | +0.133 |
bamboo | +0.127 |
Neg logit contributions | |
acts | -0.220 |
youngest | -0.218 |
nodded | -0.217 |
Let | -0.216 |
ver | -0.213 |
Token | He |
Token ID | 679 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
aster | +0.004 |
ocket | +0.004 |
poisonous | +0.004 |
heaven | +0.004 |
ocol | +0.004 |
Neg logit contributions | |
mun | -0.004 |
sets | -0.004 |
lest | -0.004 |
folk | -0.004 |
ver | -0.004 |
Token | carefully |
Token ID | 7773 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
poisonous | +0.060 |
aster | +0.055 |
ocket | +0.055 |
bamboo | +0.053 |
ocol | +0.053 |
Neg logit contributions | |
ver | -0.032 |
folk | -0.032 |
acts | -0.031 |
sets | -0.031 |
mun | -0.030 |
Token | t |
Token ID | 256 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
ver | +0.061 |
acts | +0.060 |
youngest | +0.060 |
folk | +0.060 |
oldest | +0.058 |
Neg logit contributions | |
poisonous | -0.093 |
aster | -0.088 |
ocol | -0.085 |
ocket | -0.085 |
bamboo | -0.080 |
Token | ipt |
Token ID | 10257 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
poisonous | +0.052 |
ocket | +0.048 |
arter | +0.047 |
ocol | +0.047 |
aster | +0.047 |
Neg logit contributions | |
bars | -0.011 |
nodded | -0.011 |
ver | -0.011 |
youngest | -0.011 |
Let | -0.011 |
Token | o |
Token ID | 78 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
ver | +0.362 |
bars | +0.359 |
nodded | +0.355 |
Let | +0.350 |
youngest | +0.339 |
Neg logit contributions | |
poisonous | -0.502 |
ocol | -0.457 |
ocket | -0.454 |
arter | -0.440 |
aster | -0.429 |
Token | ed |
Token ID | 276 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
poisonous | +0.151 |
ocket | +0.142 |
aster | +0.137 |
ocol | +0.132 |
bamboo | +0.131 |
Neg logit contributions | |
acts | -0.118 |
youngest | -0.114 |
oldest | -0.111 |
mun | -0.110 |
ver | -0.110 |
Token | out |
Token ID | 503 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
led | +0.081 |
nodded | +0.081 |
Let | +0.080 |
youngest | +0.078 |
ver | +0.076 |
Neg logit contributions | |
poisonous | -0.124 |
arter | -0.121 |
ocket | -0.117 |
ocol | -0.117 |
aster | -0.114 |
Token | of |
Token ID | 286 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
nodded | +0.033 |
ver | +0.033 |
acts | +0.033 |
youngest | +0.033 |
Let | +0.032 |
Neg logit contributions | |
poisonous | -0.034 |
ocol | -0.030 |
ocket | -0.030 |
aster | -0.029 |
arter | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
arter | +0.033 |
ocket | +0.033 |
poisonous | +0.032 |
ocol | +0.029 |
alo | +0.029 |
Neg logit contributions | |
led | -0.017 |
way | -0.016 |
Let | -0.016 |
youngest | -0.016 |
bars | -0.015 |
Token | giving |
Token ID | 3501 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
acts | +0.106 |
youngest | +0.105 |
way | +0.101 |
means | +0.101 |
led | +0.100 |
Neg logit contributions | |
ocket | -0.103 |
poisonous | -0.102 |
arter | -0.094 |
ocol | -0.092 |
sm | -0.086 |
Token | him |
Token ID | 683 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
acts | +0.034 |
Let | +0.034 |
nodded | +0.033 |
youngest | +0.033 |
means | +0.032 |
Neg logit contributions | |
poisonous | -0.033 |
arter | -0.032 |
ocket | -0.030 |
ocol | -0.030 |
aster | -0.029 |
Token | too |
Token ID | 1165 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
arter | +0.064 |
ocket | +0.063 |
ocol | +0.063 |
poisonous | +0.062 |
aster | +0.059 |
Neg logit contributions | |
acts | -0.055 |
youngest | -0.055 |
nodded | -0.055 |
nic | -0.054 |
's | -0.053 |
Token | many |
Token ID | 867 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
nodded | +0.023 |
ver | +0.023 |
acts | +0.023 |
youngest | +0.023 |
Let | +0.022 |
Neg logit contributions | |
poisonous | -0.015 |
ocket | -0.014 |
ocol | -0.014 |
arter | -0.013 |
aster | -0.013 |
Token | ch |
Token ID | 442 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
poisonous | +0.028 |
arter | +0.027 |
ocket | +0.027 |
aster | +0.026 |
ocol | +0.026 |
Neg logit contributions | |
acts | -0.026 |
nodded | -0.024 |
youngest | -0.024 |
led | -0.023 |
Let | -0.023 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
acts | +0.093 |
folk | +0.088 |
nodded | +0.087 |
bars | +0.087 |
youngest | +0.086 |
Neg logit contributions | |
poisonous | -0.059 |
ocket | -0.050 |
aster | -0.049 |
brand | -0.047 |
bamboo | -0.047 |
Token | ates |
Token ID | 689 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
poisonous | +0.412 |
ocket | +0.354 |
ocol | +0.353 |
aster | +0.346 |
arter | +0.342 |
Neg logit contributions | |
ver | -0.577 |
acts | -0.572 |
nodded | -0.571 |
youngest | -0.569 |
folk | -0.562 |
Token | ." |
Token ID | 526 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ver | +0.009 |
bars | +0.008 |
acts | +0.008 |
nodded | +0.008 |
youngest | +0.008 |
Neg logit contributions | |
poisonous | -0.009 |
ocol | -0.009 |
arter | -0.008 |
ocket | -0.008 |
aster | -0.008 |
Token | The |
Token ID | 383 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
acts | +0.038 |
ver | +0.038 |
youngest | +0.038 |
bars | +0.037 |
nodded | +0.037 |
Neg logit contributions | |
poisonous | -0.055 |
ocol | -0.050 |
aster | -0.050 |
ocket | -0.050 |
arter | -0.048 |
Token | king |
Token ID | 5822 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
youngest | +0.001 |
oldest | +0.001 |
nodded | +0.001 |
acts | +0.001 |
eldest | +0.001 |
Neg logit contributions | |
poisonous | -0.002 |
ocol | -0.002 |
arter | -0.002 |
ocket | -0.002 |
aster | -0.002 |
Token | learned |
Token ID | 4499 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
poisonous | +0.099 |
ocket | +0.092 |
aster | +0.089 |
ocol | +0.086 |
bamboo | +0.086 |
Neg logit contributions | |
Let | -0.046 |
youngest | -0.046 |
folk | -0.045 |
ver | -0.044 |
oldest | -0.044 |
Token | as |
Token ID | 355 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
ver | +0.002 |
Let | +0.002 |
nodded | +0.002 |
youngest | +0.002 |
acts | +0.002 |
Neg logit contributions | |
poisonous | -0.002 |
ocket | -0.002 |
ocol | -0.002 |
aster | -0.002 |
arter | -0.002 |
Token | he |
Token ID | 339 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
acts | +0.027 |
youngest | +0.027 |
nodded | +0.027 |
ver | +0.027 |
Let | +0.026 |
Neg logit contributions | |
poisonous | -0.043 |
ocol | -0.039 |
ocket | -0.039 |
aster | -0.039 |
arter | -0.038 |
Token | opened |
Token ID | 4721 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
youngest | +0.052 |
bars | +0.051 |
ver | +0.051 |
folk | +0.051 |
oldest | +0.050 |
Neg logit contributions | |
poisonous | -0.115 |
ocket | -0.107 |
aster | -0.104 |
ocol | -0.104 |
bamboo | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
youngest | +0.071 |
nodded | +0.070 |
acts | +0.069 |
led | +0.069 |
bars | +0.068 |
Neg logit contributions | |
poisonous | -0.088 |
ocket | -0.084 |
arter | -0.081 |
ocol | -0.078 |
aster | -0.078 |
Token | ch |
Token ID | 442 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
youngest | +0.061 |
oldest | +0.059 |
acts | +0.058 |
bars | +0.058 |
nic | +0.057 |
Neg logit contributions | |
poisonous | -0.084 |
ocol | -0.080 |
aster | -0.080 |
ocket | -0.079 |
arter | -0.079 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
acts | +0.123 |
nodded | +0.122 |
youngest | +0.121 |
folk | +0.121 |
bars | +0.119 |
Neg logit contributions | |
poisonous | -0.049 |
ocket | -0.040 |
aster | -0.036 |
arter | -0.035 |
ocol | -0.035 |
Token | ates |
Token ID | 689 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
poisonous | +0.410 |
ocket | +0.353 |
ocol | +0.350 |
aster | +0.343 |
arter | +0.340 |
Neg logit contributions | |
ver | -0.578 |
acts | -0.574 |
nodded | -0.573 |
youngest | -0.573 |
folk | -0.565 |
Token | and |
Token ID | 290 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
acts | +0.009 |
youngest | +0.009 |
nodded | +0.009 |
Let | +0.009 |
ver | +0.009 |
Neg logit contributions | |
poisonous | -0.007 |
ocket | -0.007 |
ocol | -0.006 |
aster | -0.006 |
arter | -0.006 |
Token | started |
Token ID | 2067 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
poisonous | +0.011 |
ocket | +0.010 |
ocol | +0.010 |
aster | +0.010 |
arter | +0.010 |
Neg logit contributions | |
ver | -0.008 |
acts | -0.008 |
youngest | -0.008 |
nodded | -0.007 |
folk | -0.007 |
Token | eating |
Token ID | 6600 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
nodded | +0.141 |
youngest | +0.140 |
acts | +0.140 |
Let | +0.138 |
ver | +0.138 |
Neg logit contributions | |
poisonous | -0.154 |
ocket | -0.141 |
ocol | -0.140 |
arter | -0.135 |
aster | -0.135 |
Token | them |
Token ID | 606 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
poisonous | +0.227 |
ocket | +0.217 |
arter | +0.215 |
aster | +0.203 |
ocol | +0.202 |
Neg logit contributions | |
nodded | -0.199 |
acts | -0.197 |
Let | -0.195 |
led | -0.193 |
youngest | -0.190 |
Token | not |
Token ID | 407 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
poisonous | +0.140 |
ocol | +0.126 |
aster | +0.126 |
ocket | +0.126 |
arter | +0.123 |
Neg logit contributions | |
ver | -0.106 |
youngest | -0.106 |
acts | -0.104 |
folk | -0.104 |
nodded | -0.103 |
Token | eat |
Token ID | 4483 |
Feature activation | -5.57e+00 |
Pos logit contributions | |
acts | +0.026 |
nodded | +0.026 |
youngest | +0.026 |
Let | +0.026 |
ver | +0.025 |
Neg logit contributions | |
poisonous | -0.036 |
ocket | -0.034 |
ocol | -0.034 |
arter | -0.032 |
aster | -0.032 |
Token | too |
Token ID | 1165 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
poisonous | +0.232 |
ocket | +0.227 |
aster | +0.225 |
arter | +0.218 |
ocol | +0.212 |
Neg logit contributions | |
acts | -0.213 |
led | -0.204 |
nodded | -0.202 |
hardest | -0.199 |
ver | -0.198 |
Token | many |
Token ID | 867 |
Feature activation | -7.95e-01 |
Pos logit contributions | |
ver | +0.011 |
acts | +0.011 |
bars | +0.011 |
nodded | +0.011 |
So | +0.011 |
Neg logit contributions | |
ocket | -0.008 |
poisonous | -0.008 |
ocol | -0.008 |
arter | -0.007 |
aster | -0.007 |
Token | ch |
Token ID | 442 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
poisonous | +0.038 |
arter | +0.035 |
ocket | +0.035 |
aster | +0.035 |
ocol | +0.034 |
Neg logit contributions | |
acts | -0.028 |
nodded | -0.027 |
led | -0.026 |
youngest | -0.026 |
folk | -0.026 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
acts | +0.109 |
folk | +0.105 |
ver | +0.104 |
bars | +0.104 |
nodded | +0.103 |
Neg logit contributions | |
poisonous | -0.028 |
ocket | -0.020 |
aster | -0.018 |
bamboo | -0.016 |
brand | -0.015 |
Token | ates |
Token ID | 689 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
poisonous | +0.427 |
ocket | +0.362 |
ocol | +0.361 |
aster | +0.359 |
arter | +0.355 |
Neg logit contributions | |
ver | -0.569 |
acts | -0.557 |
nodded | -0.555 |
youngest | -0.553 |
folk | -0.550 |
Token | and |
Token ID | 290 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
ver | +0.008 |
nodded | +0.008 |
acts | +0.008 |
bars | +0.008 |
folk | +0.008 |
Neg logit contributions | |
poisonous | -0.009 |
ocol | -0.008 |
aster | -0.008 |
arter | -0.008 |
ocket | -0.008 |
Token | his |
Token ID | 465 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
poisonous | +0.033 |
ocol | +0.030 |
ocket | +0.030 |
aster | +0.029 |
arter | +0.029 |
Neg logit contributions | |
ver | -0.026 |
youngest | -0.025 |
nodded | -0.025 |
acts | -0.025 |
Let | -0.025 |
Token | friend |
Token ID | 1545 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
poisonous | +0.033 |
aster | +0.031 |
ocket | +0.030 |
ocol | +0.030 |
arter | +0.030 |
Neg logit contributions | |
acts | -0.023 |
youngest | -0.023 |
nodded | -0.022 |
oldest | -0.022 |
eldest | -0.021 |
Token | learned |
Token ID | 4499 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
youngest | +0.063 |
bars | +0.062 |
folk | +0.061 |
ver | +0.061 |
oldest | +0.060 |
Neg logit contributions | |
poisonous | -0.128 |
ocket | -0.116 |
ocol | -0.113 |
aster | -0.113 |
bamboo | -0.110 |
Token | reward |
Token ID | 6721 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
ver | +0.067 |
nodded | +0.066 |
youngest | +0.066 |
acts | +0.065 |
Let | +0.064 |
Neg logit contributions | |
poisonous | -0.094 |
ocol | -0.086 |
ocket | -0.086 |
aster | -0.084 |
arter | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
nodded | +0.010 |
ver | +0.010 |
acts | +0.010 |
youngest | +0.010 |
Let | +0.010 |
Neg logit contributions | |
poisonous | -0.009 |
ocol | -0.008 |
ocket | -0.008 |
aster | -0.008 |
arter | -0.008 |
Token | It |
Token ID | 632 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
acts | +0.039 |
ver | +0.038 |
youngest | +0.037 |
nodded | +0.037 |
oldest | +0.036 |
Neg logit contributions | |
poisonous | -0.056 |
ocol | -0.053 |
ocket | -0.053 |
aster | -0.053 |
bamboo | -0.051 |
Token | was |
Token ID | 373 |
Feature activation | +7.13e-01 |
Pos logit contributions | |
poisonous | +0.089 |
ocket | +0.079 |
bamboo | +0.077 |
aster | +0.074 |
sm | +0.072 |
Neg logit contributions | |
folk | -0.053 |
youngest | -0.052 |
bars | -0.051 |
ver | -0.051 |
eldest | -0.050 |
Token | ch |
Token ID | 442 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
youngest | +0.026 |
led | +0.026 |
oldest | +0.025 |
nodded | +0.025 |
ver | +0.025 |
Neg logit contributions | |
poisonous | -0.031 |
ocket | -0.030 |
ocol | -0.030 |
arter | -0.030 |
aster | -0.030 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.19e+01 |
Pos logit contributions | |
acts | +0.080 |
ver | +0.079 |
folk | +0.077 |
youngest | +0.077 |
nodded | +0.077 |
Neg logit contributions | |
poisonous | -0.103 |
aster | -0.100 |
ocket | -0.095 |
injuries | -0.090 |
bamboo | -0.090 |
Token | ates |
Token ID | 689 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
poisonous | +0.440 |
ocket | +0.371 |
arter | +0.369 |
aster | +0.367 |
ocol | +0.360 |
Neg logit contributions | |
ver | -0.555 |
acts | -0.541 |
nodded | -0.541 |
youngest | -0.539 |
folk | -0.538 |
Token | and |
Token ID | 290 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
ver | +0.018 |
nodded | +0.017 |
youngest | +0.017 |
acts | +0.017 |
bars | +0.017 |
Neg logit contributions | |
poisonous | -0.013 |
ocol | -0.012 |
ocket | -0.012 |
aster | -0.012 |
arter | -0.011 |
Token | sweets |
Token ID | 42402 |
Feature activation | -3.39e+00 |
Pos logit contributions | |
poisonous | +0.027 |
ocol | +0.025 |
ocket | +0.025 |
aster | +0.024 |
arter | +0.024 |
Neg logit contributions | |
nodded | -0.022 |
acts | -0.022 |
youngest | -0.021 |
ver | -0.021 |
Let | -0.021 |
Token | ! |
Token ID | 0 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
poisonous | +0.131 |
ocol | +0.122 |
ocket | +0.117 |
aster | +0.117 |
arter | +0.115 |
Neg logit contributions | |
ver | -0.148 |
nodded | -0.148 |
acts | -0.146 |
youngest | -0.145 |
led | -0.145 |
Token |
|
Token ID | 198 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
poisonous | +0.014 |
ocol | +0.013 |
aster | +0.013 |
ocket | +0.013 |
bamboo | +0.012 |
Neg logit contributions | |
acts | -0.011 |
youngest | -0.010 |
ver | -0.010 |
nodded | -0.010 |
oldest | -0.010 |
Token | because |
Token ID | 780 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ver | +0.215 |
acts | +0.214 |
bars | +0.205 |
nodded | +0.203 |
led | +0.202 |
Neg logit contributions | |
poisonous | -0.248 |
aster | -0.239 |
ocol | -0.224 |
ocket | -0.217 |
what | -0.207 |
Token | of |
Token ID | 286 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
poisonous | +0.059 |
ocol | +0.054 |
aster | +0.053 |
ocket | +0.052 |
arter | +0.051 |
Neg logit contributions | |
ver | -0.054 |
nodded | -0.054 |
acts | -0.053 |
youngest | -0.053 |
folk | -0.052 |
Token | his |
Token ID | 465 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
acts | +0.001 |
nodded | +0.001 |
youngest | +0.001 |
ver | +0.001 |
Let | +0.001 |
Neg logit contributions | |
poisonous | -0.001 |
ocket | -0.001 |
ocol | -0.001 |
aster | -0.001 |
arter | -0.001 |
Token | big |
Token ID | 1263 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
youngest | +0.014 |
acts | +0.014 |
oldest | +0.014 |
nodded | +0.014 |
eldest | +0.013 |
Neg logit contributions | |
poisonous | -0.019 |
ocket | -0.018 |
ocol | -0.017 |
aster | -0.017 |
arter | -0.017 |
Token | bus |
Token ID | 1323 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
poisonous | +0.134 |
ocket | +0.123 |
ocol | +0.122 |
aster | +0.120 |
arter | +0.118 |
Neg logit contributions | |
acts | -0.094 |
youngest | -0.093 |
nodded | -0.092 |
ver | -0.090 |
Let | -0.089 |
Token | hy |
Token ID | 12114 |
Feature activation | -1.19e+01 |
Pos logit contributions | |
poisonous | +0.086 |
ocket | +0.077 |
ocol | +0.077 |
aster | +0.075 |
arter | +0.074 |
Neg logit contributions | |
youngest | -0.091 |
acts | -0.091 |
nodded | -0.091 |
ver | -0.091 |
Let | -0.090 |
Token | mustache |
Token ID | 49303 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
poisonous | +0.508 |
ocol | +0.460 |
ocket | +0.455 |
aster | +0.443 |
arter | +0.443 |
Neg logit contributions | |
nodded | -0.480 |
youngest | -0.470 |
acts | -0.467 |
folk | -0.467 |
ver | -0.466 |
Token | . |
Token ID | 13 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
poisonous | +0.094 |
ocol | +0.085 |
ocket | +0.084 |
aster | +0.082 |
arter | +0.080 |
Neg logit contributions | |
ver | -0.102 |
acts | -0.102 |
nodded | -0.102 |
youngest | -0.102 |
Let | -0.101 |
Token |
|
Token ID | 198 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
poisonous | +0.029 |
ocol | +0.027 |
aster | +0.027 |
ocket | +0.027 |
arter | +0.026 |
Neg logit contributions | |
acts | -0.020 |
nodded | -0.020 |
youngest | -0.019 |
ver | -0.019 |
led | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
Let | +0.010 |
So | +0.010 |
ver | +0.009 |
First | +0.009 |
basics | +0.009 |
Neg logit contributions | |
poisonous | -0.026 |
ocket | -0.025 |
ocol | -0.024 |
arter | -0.024 |
injuries | -0.023 |
Token | One |
Token ID | 3198 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
Let | +0.015 |
ver | +0.014 |
nodded | +0.014 |
youngest | +0.014 |
So | +0.014 |
Neg logit contributions | |
poisonous | -0.021 |
ocket | -0.019 |
ocol | -0.019 |
arter | -0.019 |
injuries | -0.017 |
Token | colorful |
Token ID | 20239 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
poisonous | +0.030 |
ocol | +0.028 |
ocket | +0.027 |
bamboo | +0.026 |
aster | +0.025 |
Neg logit contributions | |
ver | -0.028 |
bars | -0.026 |
nodded | -0.026 |
Let | -0.026 |
youngest | -0.026 |
Token | cand |
Token ID | 2658 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
poisonous | +0.053 |
ocol | +0.049 |
ocket | +0.048 |
aster | +0.047 |
arter | +0.046 |
Neg logit contributions | |
ver | -0.035 |
nodded | -0.034 |
youngest | -0.034 |
acts | -0.034 |
Let | -0.034 |
Token | ies |
Token ID | 444 |
Feature activation | +7.81e-01 |
Pos logit contributions | |
poisonous | +0.188 |
aster | +0.175 |
ocol | +0.174 |
ocket | +0.173 |
arter | +0.171 |
Neg logit contributions | |
nodded | -0.095 |
ver | -0.094 |
youngest | -0.094 |
acts | -0.093 |
led | -0.091 |
Token | and |
Token ID | 290 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
nodded | +0.033 |
acts | +0.032 |
ver | +0.032 |
youngest | +0.032 |
Let | +0.032 |
Neg logit contributions | |
poisonous | -0.032 |
ocol | -0.029 |
ocket | -0.028 |
aster | -0.028 |
arter | -0.028 |
Token | ch |
Token ID | 442 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
poisonous | +0.008 |
ocol | +0.008 |
ocket | +0.008 |
aster | +0.008 |
arter | +0.008 |
Neg logit contributions | |
nodded | -0.005 |
youngest | -0.005 |
acts | -0.005 |
ver | -0.005 |
Let | -0.005 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
acts | +0.100 |
nodded | +0.099 |
folk | +0.099 |
youngest | +0.099 |
ver | +0.097 |
Neg logit contributions | |
poisonous | -0.035 |
aster | -0.027 |
ocket | -0.027 |
arter | -0.024 |
bamboo | -0.023 |
Token | ates |
Token ID | 689 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
poisonous | +0.420 |
aster | +0.360 |
ocket | +0.358 |
ocol | +0.355 |
arter | +0.351 |
Neg logit contributions | |
ver | -0.551 |
nodded | -0.543 |
acts | -0.542 |
youngest | -0.541 |
folk | -0.536 |
Token | . |
Token ID | 13 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
nodded | +0.016 |
ver | +0.015 |
acts | +0.015 |
youngest | +0.015 |
led | +0.015 |
Neg logit contributions | |
poisonous | -0.016 |
ocol | -0.015 |
aster | -0.015 |
ocket | -0.015 |
arter | -0.015 |
Token | She |
Token ID | 1375 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
poisonous | +0.018 |
ocol | +0.017 |
ocket | +0.016 |
aster | +0.016 |
bamboo | +0.016 |
Neg logit contributions | |
ver | -0.017 |
acts | -0.017 |
folk | -0.016 |
nodded | -0.016 |
mun | -0.016 |
Token | said |
Token ID | 531 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
poisonous | +0.180 |
ocket | +0.166 |
aster | +0.163 |
ocol | +0.162 |
bamboo | +0.158 |
Neg logit contributions | |
ver | -0.074 |
acts | -0.072 |
folk | -0.072 |
bars | -0.070 |
youngest | -0.068 |
Token | to |
Token ID | 284 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
youngest | +0.040 |
acts | +0.040 |
bars | +0.040 |
oldest | +0.039 |
folk | +0.039 |
Neg logit contributions | |
poisonous | -0.039 |
ocket | -0.035 |
ocol | -0.035 |
bamboo | -0.034 |
aster | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
poisonous | +0.009 |
ocket | +0.008 |
ocol | +0.008 |
aster | +0.007 |
arter | +0.007 |
Neg logit contributions | |
acts | -0.011 |
youngest | -0.011 |
Let | -0.011 |
nodded | -0.011 |
ver | -0.011 |
Token | cup |
Token ID | 6508 |
Feature activation | -4.83e+00 |
Pos logit contributions | |
nodded | +0.005 |
Let | +0.005 |
youngest | +0.005 |
acts | +0.005 |
ver | +0.005 |
Neg logit contributions | |
poisonous | -0.006 |
ocket | -0.005 |
ocol | -0.005 |
arter | -0.005 |
aster | -0.005 |
Token | cakes |
Token ID | 37263 |
Feature activation | -9.75e-01 |
Pos logit contributions | |
poisonous | +0.221 |
ocol | +0.202 |
ocket | +0.201 |
aster | +0.196 |
arter | +0.194 |
Neg logit contributions | |
nodded | -0.176 |
ver | -0.176 |
acts | -0.175 |
youngest | -0.175 |
Let | -0.172 |
Token | and |
Token ID | 290 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
poisonous | +0.042 |
ocol | +0.038 |
ocket | +0.038 |
aster | +0.038 |
arter | +0.036 |
Neg logit contributions | |
acts | -0.038 |
youngest | -0.038 |
ver | -0.037 |
nodded | -0.037 |
Let | -0.037 |
Token | ch |
Token ID | 442 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
poisonous | +0.015 |
ocol | +0.013 |
aster | +0.013 |
ocket | +0.013 |
bamboo | +0.012 |
Neg logit contributions | |
ver | -0.015 |
acts | -0.015 |
youngest | -0.015 |
nodded | -0.014 |
folk | -0.014 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
nodded | +0.116 |
folk | +0.116 |
acts | +0.116 |
bars | +0.113 |
youngest | +0.113 |
Neg logit contributions | |
poisonous | -0.037 |
ocket | -0.030 |
arter | -0.025 |
aster | -0.023 |
injuries | -0.022 |
Token | ates |
Token ID | 689 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
poisonous | +0.462 |
arter | +0.390 |
aster | +0.388 |
ocket | +0.388 |
ocol | +0.368 |
Neg logit contributions | |
ver | -0.519 |
folk | -0.504 |
nodded | -0.502 |
youngest | -0.499 |
acts | -0.499 |
Token | . |
Token ID | 13 |
Feature activation | -7.39e-01 |
Pos logit contributions | |
nodded | +0.019 |
ver | +0.019 |
youngest | +0.018 |
acts | +0.018 |
bars | +0.018 |
Neg logit contributions | |
poisonous | -0.019 |
ocol | -0.017 |
ocket | -0.017 |
arter | -0.017 |
aster | -0.017 |
Token | |
Token ID | 220 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
poisonous | +0.036 |
ocol | +0.033 |
aster | +0.033 |
ocket | +0.032 |
arter | +0.031 |
Neg logit contributions | |
acts | -0.025 |
ver | -0.024 |
youngest | -0.024 |
nodded | -0.024 |
oldest | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
acts | +0.120 |
youngest | +0.119 |
nodded | +0.119 |
ver | +0.116 |
folk | +0.115 |
Neg logit contributions | |
poisonous | -0.168 |
ocol | -0.155 |
ocket | -0.154 |
aster | -0.151 |
arter | -0.149 |
Token |
|
Token ID | 198 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
ver | +0.021 |
nodded | +0.020 |
Let | +0.020 |
youngest | +0.020 |
acts | +0.020 |
Neg logit contributions | |
poisonous | -0.035 |
ocket | -0.032 |
ocol | -0.032 |
arter | -0.031 |
aster | -0.031 |
Token | oll |
Token ID | 692 |
Feature activation | -8.83e+00 |
Pos logit contributions | |
poisonous | +0.196 |
aster | +0.171 |
arter | +0.162 |
injuries | +0.160 |
eds | +0.159 |
Neg logit contributions | |
ver | -0.269 |
nodded | -0.253 |
bars | -0.246 |
youngest | -0.246 |
folk | -0.245 |
Token | ip |
Token ID | 541 |
Feature activation | -4.58e+00 |
Pos logit contributions | |
poisonous | +0.364 |
arter | +0.335 |
aster | +0.329 |
ocol | +0.322 |
ocket | +0.320 |
Neg logit contributions | |
ver | -0.365 |
nodded | -0.360 |
bars | -0.352 |
acts | -0.351 |
mun | -0.337 |
Token | ops |
Token ID | 2840 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
poisonous | +0.210 |
ocol | +0.195 |
ocket | +0.193 |
arter | +0.191 |
aster | +0.189 |
Neg logit contributions | |
acts | -0.165 |
nodded | -0.165 |
ver | -0.164 |
youngest | -0.161 |
bars | -0.159 |
Token | , |
Token ID | 11 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
poisonous | +0.015 |
ocol | +0.014 |
ocket | +0.013 |
arter | +0.013 |
aster | +0.013 |
Neg logit contributions | |
ver | -0.013 |
nodded | -0.013 |
led | -0.013 |
bars | -0.013 |
folk | -0.012 |
Token | ch |
Token ID | 442 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
nodded | +0.005 |
Let | +0.005 |
led | +0.004 |
acts | +0.004 |
youngest | +0.004 |
Neg logit contributions | |
poisonous | -0.006 |
ocket | -0.005 |
ocol | -0.005 |
arter | -0.005 |
aster | -0.005 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
acts | +0.101 |
nodded | +0.100 |
bars | +0.097 |
folk | +0.095 |
nods | +0.094 |
Neg logit contributions | |
poisonous | -0.052 |
heaven | -0.044 |
ocket | -0.044 |
prise | -0.042 |
brand | -0.042 |
Token | ates |
Token ID | 689 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
poisonous | +0.503 |
arter | +0.429 |
aster | +0.420 |
ocket | +0.410 |
brand | +0.405 |
Neg logit contributions | |
ver | -0.485 |
nodded | -0.461 |
acts | -0.457 |
youngest | -0.457 |
folk | -0.456 |
Token | , |
Token ID | 11 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
nodded | +0.014 |
ver | +0.014 |
acts | +0.014 |
led | +0.013 |
bars | +0.013 |
Neg logit contributions | |
poisonous | -0.019 |
ocol | -0.017 |
aster | -0.017 |
ocket | -0.017 |
arter | -0.017 |
Token | g |
Token ID | 308 |
Feature activation | +8.69e-01 |
Pos logit contributions | |
poisonous | +0.001 |
arter | +0.001 |
ocket | +0.001 |
ocol | +0.001 |
aster | +0.001 |
Neg logit contributions | |
nodded | -0.001 |
Let | -0.001 |
acts | -0.001 |
led | -0.001 |
bars | -0.001 |
Token | ummies |
Token ID | 39578 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
youngest | +0.031 |
nodded | +0.030 |
bars | +0.029 |
oldest | +0.029 |
eldest | +0.029 |
Neg logit contributions | |
poisonous | -0.041 |
aster | -0.036 |
ocket | -0.036 |
injuries | -0.036 |
handcuffs | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
poisonous | +0.091 |
ocol | +0.085 |
aster | +0.084 |
ocket | +0.084 |
arter | +0.081 |
Neg logit contributions | |
nodded | -0.064 |
acts | -0.062 |
led | -0.060 |
ver | -0.060 |
youngest | -0.060 |
Token | he |
Token ID | 339 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
poisonous | +0.148 |
aster | +0.134 |
ocket | +0.134 |
ocol | +0.133 |
arter | +0.128 |
Neg logit contributions | |
ver | -0.124 |
acts | -0.123 |
youngest | -0.122 |
nodded | -0.121 |
folk | -0.121 |
Token | ate |
Token ID | 15063 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
acts | +0.052 |
ver | +0.051 |
folk | +0.050 |
youngest | +0.049 |
bars | +0.049 |
Neg logit contributions | |
poisonous | -0.111 |
ocket | -0.103 |
ocol | -0.099 |
aster | -0.099 |
bamboo | -0.097 |
Token | all |
Token ID | 477 |
Feature activation | +5.34e+00 |
Pos logit contributions | |
arter | +0.154 |
ocket | +0.152 |
poisonous | +0.149 |
ocol | +0.140 |
eds | +0.136 |
Neg logit contributions | |
nodded | -0.099 |
Let | -0.095 |
led | -0.095 |
nic | -0.093 |
however | -0.090 |
Token | the |
Token ID | 262 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
nodded | +0.207 |
acts | +0.206 |
Let | +0.206 |
youngest | +0.204 |
ver | +0.203 |
Neg logit contributions | |
poisonous | -0.231 |
ocket | -0.216 |
ocol | -0.211 |
arter | -0.207 |
injuries | -0.199 |
Token | ch |
Token ID | 442 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
acts | +0.048 |
nodded | +0.047 |
mun | +0.047 |
youngest | +0.047 |
nic | +0.046 |
Neg logit contributions | |
arter | -0.046 |
poisonous | -0.045 |
ocket | -0.043 |
aster | -0.042 |
ocol | -0.041 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
acts | +0.130 |
folk | +0.128 |
nodded | +0.127 |
bars | +0.126 |
youngest | +0.125 |
Neg logit contributions | |
poisonous | -0.025 |
ocket | -0.016 |
arter | -0.012 |
injuries | -0.012 |
brand | -0.011 |
Token | ates |
Token ID | 689 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
poisonous | +0.428 |
ocket | +0.361 |
arter | +0.359 |
aster | +0.354 |
ocol | +0.353 |
Neg logit contributions | |
ver | -0.539 |
nodded | -0.527 |
acts | -0.524 |
folk | -0.523 |
youngest | -0.522 |
Token | and |
Token ID | 290 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
ver | +0.023 |
nodded | +0.023 |
acts | +0.023 |
youngest | +0.023 |
Let | +0.023 |
Neg logit contributions | |
poisonous | -0.023 |
ocol | -0.021 |
ocket | -0.021 |
aster | -0.020 |
arter | -0.020 |
Token | sweets |
Token ID | 42402 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
poisonous | +0.023 |
ocol | +0.022 |
ocket | +0.022 |
arter | +0.022 |
aster | +0.021 |
Neg logit contributions | |
nodded | -0.012 |
youngest | -0.012 |
Let | -0.012 |
acts | -0.012 |
led | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -5.63e-01 |
Pos logit contributions | |
poisonous | +0.145 |
ocol | +0.134 |
ocket | +0.131 |
arter | +0.129 |
aster | +0.128 |
Neg logit contributions | |
nodded | -0.145 |
ver | -0.144 |
youngest | -0.143 |
acts | -0.141 |
led | -0.141 |
Token | Then |
Token ID | 3244 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
poisonous | +0.025 |
ocol | +0.023 |
aster | +0.023 |
ocket | +0.023 |
arter | +0.023 |
Neg logit contributions | |
acts | -0.020 |
ver | -0.020 |
folk | -0.019 |
nodded | -0.019 |
youngest | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
poisonous | +0.018 |
aster | +0.016 |
ocol | +0.016 |
ocket | +0.016 |
bamboo | +0.015 |
Neg logit contributions | |
youngest | -0.022 |
eldest | -0.021 |
oldest | -0.021 |
acts | -0.021 |
basics | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
youngest | +0.025 |
nodded | +0.025 |
ver | +0.024 |
Let | +0.024 |
acts | +0.024 |
Neg logit contributions | |
poisonous | -0.024 |
ocol | -0.022 |
ocket | -0.022 |
arter | -0.021 |
aster | -0.021 |
Token | box |
Token ID | 3091 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
ocol | +0.007 |
poisonous | +0.007 |
ocket | +0.007 |
arter | +0.007 |
aster | +0.006 |
Neg logit contributions | |
nodded | -0.005 |
youngest | -0.005 |
Let | -0.005 |
oldest | -0.005 |
led | -0.005 |
Token | of |
Token ID | 286 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
nodded | +0.052 |
bars | +0.052 |
led | +0.052 |
ver | +0.051 |
youngest | +0.051 |
Neg logit contributions | |
poisonous | -0.060 |
ocol | -0.056 |
ocket | -0.054 |
aster | -0.053 |
arter | -0.053 |
Token | ch |
Token ID | 442 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
poisonous | +0.022 |
arter | +0.021 |
ocket | +0.021 |
aster | +0.020 |
ocol | +0.019 |
Neg logit contributions | |
acts | -0.016 |
youngest | -0.016 |
Let | -0.016 |
oldest | -0.016 |
nodded | -0.015 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
acts | +0.112 |
bars | +0.109 |
folk | +0.109 |
youngest | +0.107 |
nodded | +0.106 |
Neg logit contributions | |
poisonous | -0.039 |
ocket | -0.030 |
bamboo | -0.028 |
arter | -0.026 |
aster | -0.026 |
Token | ates |
Token ID | 689 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
poisonous | +0.430 |
ocket | +0.363 |
ocol | +0.360 |
arter | +0.359 |
aster | +0.359 |
Neg logit contributions | |
ver | -0.533 |
acts | -0.523 |
nodded | -0.522 |
youngest | -0.522 |
folk | -0.519 |
Token | . |
Token ID | 13 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
ver | +0.028 |
nodded | +0.028 |
bars | +0.027 |
youngest | +0.027 |
acts | +0.027 |
Neg logit contributions | |
poisonous | -0.027 |
ocol | -0.024 |
aster | -0.024 |
ocket | -0.023 |
arter | -0.023 |
Token | " |
Token ID | 366 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
poisonous | +0.002 |
ocol | +0.002 |
aster | +0.002 |
ocket | +0.002 |
bamboo | +0.002 |
Neg logit contributions | |
acts | -0.002 |
youngest | -0.002 |
nodded | -0.002 |
ver | -0.002 |
oldest | -0.002 |
Token | For |
Token ID | 1890 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
nodded | +0.085 |
youngest | +0.085 |
ver | +0.085 |
acts | +0.085 |
Let | +0.084 |
Neg logit contributions | |
poisonous | -0.136 |
ocol | -0.125 |
ocket | -0.124 |
aster | -0.121 |
arter | -0.121 |
Token | me |
Token ID | 502 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
poisonous | +0.099 |
ocket | +0.090 |
arter | +0.088 |
ocol | +0.087 |
aster | +0.083 |
Neg logit contributions | |
youngest | -0.094 |
acts | -0.093 |
oldest | -0.091 |
led | -0.090 |
nodded | -0.090 |
Token | a |
Token ID | 257 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
poisonous | +0.129 |
ocol | +0.124 |
arter | +0.121 |
ocket | +0.119 |
aster | +0.115 |
Neg logit contributions | |
nodded | -0.131 |
youngest | -0.131 |
led | -0.130 |
Let | -0.129 |
acts | -0.127 |
Token | big |
Token ID | 1263 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
ocol | +0.017 |
poisonous | +0.016 |
ocket | +0.016 |
arter | +0.016 |
aster | +0.016 |
Neg logit contributions | |
nodded | -0.010 |
youngest | -0.010 |
Let | -0.010 |
oldest | -0.010 |
led | -0.010 |
Token | box |
Token ID | 3091 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
poisonous | +0.157 |
ocket | +0.151 |
ocol | +0.144 |
arter | +0.143 |
aster | +0.141 |
Neg logit contributions | |
acts | -0.094 |
nodded | -0.092 |
led | -0.086 |
folk | -0.085 |
youngest | -0.085 |
Token | of |
Token ID | 286 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
nodded | +0.053 |
bars | +0.052 |
led | +0.052 |
ver | +0.052 |
youngest | +0.052 |
Neg logit contributions | |
poisonous | -0.056 |
ocol | -0.052 |
ocket | -0.050 |
arter | -0.049 |
aster | -0.049 |
Token | ch |
Token ID | 442 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
arter | +0.027 |
poisonous | +0.027 |
ocket | +0.026 |
aster | +0.026 |
isk | +0.025 |
Neg logit contributions | |
Let | -0.018 |
acts | -0.018 |
youngest | -0.018 |
oldest | -0.017 |
led | -0.017 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
acts | +0.097 |
folk | +0.096 |
bars | +0.094 |
youngest | +0.092 |
towns | +0.090 |
Neg logit contributions | |
poisonous | -0.048 |
ocket | -0.036 |
bamboo | -0.036 |
brand | -0.036 |
aster | -0.034 |
Token | ates |
Token ID | 689 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
poisonous | +0.420 |
ocket | +0.355 |
ocol | +0.353 |
aster | +0.351 |
arter | +0.349 |
Neg logit contributions | |
ver | -0.539 |
youngest | -0.530 |
acts | -0.529 |
nodded | -0.529 |
folk | -0.526 |
Token | . |
Token ID | 13 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
ver | +0.023 |
nodded | +0.023 |
youngest | +0.023 |
acts | +0.023 |
bars | +0.023 |
Neg logit contributions | |
poisonous | -0.022 |
ocol | -0.020 |
ocket | -0.019 |
aster | -0.019 |
arter | -0.019 |
Token | It |
Token ID | 632 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
poisonous | +0.024 |
ocol | +0.023 |
ocket | +0.022 |
aster | +0.022 |
arter | +0.022 |
Neg logit contributions | |
acts | -0.018 |
ver | -0.018 |
nodded | -0.018 |
youngest | -0.018 |
folk | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
poisonous | +0.128 |
bamboo | +0.114 |
ocket | +0.114 |
ocol | +0.109 |
aster | +0.108 |
Neg logit contributions | |
ver | -0.076 |
folk | -0.076 |
acts | -0.074 |
youngest | -0.074 |
bars | -0.072 |
Token | so |
Token ID | 523 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
led | +0.006 |
youngest | +0.006 |
oldest | +0.005 |
younger | +0.005 |
best | +0.005 |
Neg logit contributions | |
eds | -0.009 |
ocket | -0.009 |
arter | -0.009 |
poisonous | -0.009 |
aster | -0.009 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
ver | +0.002 |
acts | +0.002 |
bars | +0.002 |
basics | +0.002 |
youngest | +0.002 |
Neg logit contributions | |
poisonous | -0.004 |
aster | -0.004 |
ocket | -0.004 |
ocol | -0.004 |
bamboo | -0.004 |
Token | her |
Token ID | 607 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
poisonous | +0.156 |
ocket | +0.149 |
arter | +0.149 |
ocol | +0.139 |
aster | +0.133 |
Neg logit contributions | |
nodded | -0.112 |
Let | -0.108 |
led | -0.104 |
youngest | -0.102 |
acts | -0.100 |
Token | cool |
Token ID | 3608 |
Feature activation | -6.64e-02 |
Pos logit contributions | |
poisonous | +0.091 |
ocol | +0.082 |
ocket | +0.082 |
arter | +0.081 |
aster | +0.077 |
Neg logit contributions | |
acts | -0.110 |
nodded | -0.110 |
youngest | -0.109 |
Let | -0.107 |
ver | -0.106 |
Token | , |
Token ID | 11 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
poisonous | +0.003 |
ocket | +0.003 |
arter | +0.003 |
ocol | +0.002 |
aster | +0.002 |
Neg logit contributions | |
acts | -0.003 |
nodded | -0.003 |
Let | -0.003 |
youngest | -0.003 |
nods | -0.003 |
Token | ch |
Token ID | 442 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
nodded | +0.013 |
Let | +0.013 |
youngest | +0.013 |
acts | +0.013 |
led | +0.012 |
Neg logit contributions | |
poisonous | -0.014 |
ocol | -0.013 |
ocket | -0.013 |
arter | -0.012 |
aster | -0.012 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
acts | +0.089 |
nodded | +0.089 |
folk | +0.087 |
youngest | +0.085 |
bars | +0.085 |
Neg logit contributions | |
poisonous | -0.071 |
ocket | -0.060 |
bamboo | -0.059 |
aster | -0.057 |
injuries | -0.055 |
Token | at |
Token ID | 265 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
poisonous | +0.431 |
ocket | +0.361 |
aster | +0.361 |
ocol | +0.357 |
arter | +0.357 |
Neg logit contributions | |
ver | -0.538 |
nodded | -0.528 |
acts | -0.524 |
youngest | -0.523 |
folk | -0.519 |
Token | y |
Token ID | 88 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
poisonous | +0.141 |
ocol | +0.127 |
ocket | +0.127 |
aster | +0.124 |
arter | +0.121 |
Neg logit contributions | |
acts | -0.132 |
ver | -0.131 |
youngest | -0.131 |
nodded | -0.131 |
Let | -0.129 |
Token | treat |
Token ID | 2190 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
poisonous | +0.084 |
ocol | +0.077 |
ocket | +0.077 |
arter | +0.075 |
aster | +0.075 |
Neg logit contributions | |
nodded | -0.073 |
youngest | -0.073 |
acts | -0.073 |
Let | -0.071 |
ver | -0.070 |
Token | ! |
Token ID | 0 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
poisonous | +0.093 |
arter | +0.085 |
ocket | +0.085 |
ocol | +0.084 |
aster | +0.082 |
Neg logit contributions | |
nodded | -0.094 |
acts | -0.091 |
Let | -0.091 |
ver | -0.091 |
led | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | +7.93e-01 |
Pos logit contributions | |
poisonous | +0.024 |
ocol | +0.022 |
ocket | +0.022 |
aster | +0.022 |
arter | +0.021 |
Neg logit contributions | |
acts | -0.016 |
youngest | -0.015 |
ver | -0.015 |
nodded | -0.015 |
oldest | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
poisonous | +0.078 |
ocol | +0.071 |
ocket | +0.071 |
aster | +0.069 |
arter | +0.068 |
Neg logit contributions | |
youngest | -0.072 |
nodded | -0.072 |
acts | -0.072 |
ver | -0.072 |
Let | -0.071 |
Token | large |
Token ID | 1588 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ocol | +0.022 |
ocket | +0.022 |
aster | +0.022 |
poisonous | +0.022 |
arter | +0.022 |
Neg logit contributions | |
nic | -0.015 |
way | -0.015 |
youngest | -0.015 |
Let | -0.014 |
oldest | -0.014 |
Token | selection |
Token ID | 6356 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
poisonous | +0.057 |
ocket | +0.054 |
arter | +0.052 |
aster | +0.052 |
ocol | +0.051 |
Neg logit contributions | |
acts | -0.041 |
nodded | -0.040 |
youngest | -0.039 |
led | -0.039 |
Let | -0.038 |
Token | of |
Token ID | 286 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
poisonous | +0.095 |
ocol | +0.085 |
ocket | +0.085 |
aster | +0.083 |
arter | +0.082 |
Neg logit contributions | |
ver | -0.093 |
acts | -0.092 |
nodded | -0.092 |
youngest | -0.092 |
Let | -0.091 |
Token | ch |
Token ID | 442 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
poisonous | +0.017 |
arter | +0.017 |
ocket | +0.016 |
aster | +0.016 |
injuries | +0.015 |
Neg logit contributions | |
acts | -0.014 |
youngest | -0.013 |
Let | -0.013 |
led | -0.013 |
nodded | -0.013 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
acts | +0.133 |
folk | +0.131 |
bars | +0.129 |
youngest | +0.129 |
oldest | +0.126 |
Neg logit contributions | |
poisonous | -0.027 |
ocket | -0.016 |
arter | -0.013 |
aster | -0.012 |
bamboo | -0.012 |
Token | ates |
Token ID | 689 |
Feature activation | +6.53e-01 |
Pos logit contributions | |
poisonous | +0.447 |
arter | +0.375 |
ocket | +0.368 |
aster | +0.364 |
ocol | +0.355 |
Neg logit contributions | |
ver | -0.521 |
acts | -0.510 |
folk | -0.508 |
youngest | -0.506 |
nodded | -0.502 |
Token | . |
Token ID | 13 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ver | +0.027 |
nodded | +0.026 |
acts | +0.026 |
youngest | +0.026 |
bars | +0.026 |
Neg logit contributions | |
poisonous | -0.027 |
ocol | -0.025 |
ocket | -0.024 |
aster | -0.024 |
arter | -0.024 |
Token | Ell |
Token ID | 7122 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
poisonous | +0.009 |
ocol | +0.008 |
ocket | +0.008 |
aster | +0.008 |
arter | +0.008 |
Neg logit contributions | |
acts | -0.006 |
nodded | -0.006 |
ver | -0.006 |
youngest | -0.006 |
folk | -0.005 |
Token | a |
Token ID | 64 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
poisonous | +0.072 |
ocket | +0.067 |
ocol | +0.064 |
aster | +0.064 |
sm | +0.060 |
Neg logit contributions | |
acts | -0.064 |
folk | -0.062 |
led | -0.062 |
Let | -0.062 |
youngest | -0.062 |
Token | picked |
Token ID | 6497 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
poisonous | +0.018 |
ocket | +0.017 |
aster | +0.016 |
ocol | +0.015 |
sm | +0.015 |
Neg logit contributions | |
bars | -0.013 |
sets | -0.013 |
folk | -0.013 |
ver | -0.013 |
acts | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
poisonous | +0.138 |
ocket | +0.126 |
bamboo | +0.124 |
aster | +0.118 |
ocol | +0.115 |
Neg logit contributions | |
folk | -0.082 |
youngest | -0.078 |
ver | -0.077 |
acts | -0.076 |
mun | -0.074 |
Token | a |
Token ID | 257 |
Feature activation | -6.03e-01 |
Pos logit contributions | |
eds | +0.001 |
aster | +0.001 |
ez | +0.001 |
arter | +0.001 |
poisonous | +0.001 |
Neg logit contributions | |
led | -0.001 |
best | -0.001 |
younger | -0.001 |
nic | -0.001 |
way | -0.001 |
Token | box |
Token ID | 3091 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
ocol | +0.031 |
ez | +0.030 |
aster | +0.030 |
isk | +0.029 |
eds | +0.029 |
Neg logit contributions | |
way | -0.020 |
best | -0.019 |
boy | -0.018 |
leader | -0.017 |
led | -0.017 |
Token | of |
Token ID | 286 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
bars | +0.046 |
ver | +0.046 |
led | +0.046 |
's | +0.045 |
youngest | +0.045 |
Neg logit contributions | |
poisonous | -0.060 |
ocol | -0.058 |
ocket | -0.054 |
aster | -0.053 |
arter | -0.052 |
Token | ch |
Token ID | 442 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
poisonous | +0.036 |
arter | +0.036 |
aster | +0.035 |
ocket | +0.035 |
isk | +0.034 |
Neg logit contributions | |
acts | -0.024 |
Let | -0.024 |
youngest | -0.024 |
led | -0.023 |
oldest | -0.023 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
acts | +0.098 |
folk | +0.094 |
bars | +0.093 |
teams | +0.091 |
towns | +0.091 |
Neg logit contributions | |
poisonous | -0.045 |
brand | -0.034 |
prank | -0.033 |
ocket | -0.032 |
liking | -0.031 |
Token | ates |
Token ID | 689 |
Feature activation | +6.97e-01 |
Pos logit contributions | |
poisonous | +0.444 |
arter | +0.370 |
ocket | +0.368 |
aster | +0.364 |
ocol | +0.357 |
Neg logit contributions | |
ver | -0.516 |
folk | -0.506 |
acts | -0.504 |
youngest | -0.502 |
nodded | -0.501 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
ver | +0.029 |
bars | +0.028 |
youngest | +0.028 |
nodded | +0.028 |
acts | +0.028 |
Neg logit contributions | |
poisonous | -0.029 |
ocol | -0.026 |
aster | -0.025 |
arter | -0.025 |
ocket | -0.025 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
poisonous | +0.055 |
ocol | +0.051 |
ocket | +0.051 |
aster | +0.050 |
arter | +0.049 |
Neg logit contributions | |
nodded | -0.035 |
ver | -0.035 |
acts | -0.035 |
youngest | -0.034 |
Let | -0.034 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
poisonous | +0.201 |
ocket | +0.191 |
aster | +0.177 |
bamboo | +0.176 |
isk | +0.168 |
Neg logit contributions | |
folk | -0.153 |
bars | -0.144 |
hardest | -0.143 |
ver | -0.141 |
youngest | -0.140 |
Token | ch |
Token ID | 442 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
poisonous | +0.082 |
ocket | +0.080 |
arter | +0.080 |
ocol | +0.074 |
aster | +0.073 |
Neg logit contributions | |
acts | -0.059 |
oldest | -0.058 |
youngest | -0.058 |
led | -0.058 |
asks | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
poisonous | +0.028 |
ocol | +0.026 |
ocket | +0.025 |
aster | +0.025 |
arter | +0.024 |
Neg logit contributions | |
nodded | -0.029 |
ver | -0.029 |
acts | -0.029 |
led | -0.029 |
youngest | -0.029 |
Token | filled |
Token ID | 5901 |
Feature activation | +8.57e-01 |
Pos logit contributions | |
led | +0.015 |
nodded | +0.015 |
youngest | +0.015 |
ver | +0.014 |
nic | +0.014 |
Neg logit contributions | |
ocol | -0.016 |
poisonous | -0.016 |
arter | -0.015 |
ocket | -0.015 |
aster | -0.015 |
Token | with |
Token ID | 351 |
Feature activation | +8.46e-02 |
Pos logit contributions | |
Let | +0.032 |
ver | +0.032 |
nodded | +0.031 |
youngest | +0.031 |
acts | +0.031 |
Neg logit contributions | |
poisonous | -0.038 |
ocol | -0.036 |
arter | -0.035 |
ocket | -0.035 |
aster | -0.033 |
Token | delicious |
Token ID | 12625 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
acts | +0.003 |
youngest | +0.003 |
nodded | +0.003 |
Let | +0.003 |
led | +0.003 |
Neg logit contributions | |
poisonous | -0.004 |
arter | -0.003 |
ocket | -0.003 |
aster | -0.003 |
ocol | -0.003 |
Token | ch |
Token ID | 442 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
poisonous | +0.141 |
ocket | +0.132 |
aster | +0.130 |
arter | +0.127 |
ocol | +0.122 |
Neg logit contributions | |
acts | -0.135 |
nodded | -0.131 |
youngest | -0.127 |
Let | -0.125 |
led | -0.124 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
acts | +0.101 |
folk | +0.100 |
youngest | +0.099 |
nodded | +0.096 |
oldest | +0.096 |
Neg logit contributions | |
poisonous | -0.035 |
ocket | -0.028 |
aster | -0.027 |
bamboo | -0.024 |
injuries | -0.023 |
Token | ates |
Token ID | 689 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
poisonous | +0.402 |
ocket | +0.344 |
ocol | +0.344 |
aster | +0.342 |
arter | +0.335 |
Neg logit contributions | |
ver | -0.546 |
acts | -0.540 |
nodded | -0.540 |
youngest | -0.538 |
folk | -0.532 |
Token | . |
Token ID | 13 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
ver | +0.027 |
nodded | +0.027 |
youngest | +0.027 |
acts | +0.026 |
bars | +0.026 |
Neg logit contributions | |
poisonous | -0.026 |
ocol | -0.023 |
aster | -0.023 |
arter | -0.023 |
ocket | -0.022 |
Token | He |
Token ID | 679 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
poisonous | +0.001 |
aster | +0.001 |
ocol | +0.001 |
ocket | +0.001 |
bamboo | +0.001 |
Neg logit contributions | |
acts | -0.001 |
youngest | -0.001 |
ver | -0.001 |
oldest | -0.001 |
nodded | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
poisonous | +0.064 |
ocket | +0.059 |
ocol | +0.058 |
aster | +0.058 |
bamboo | +0.056 |
Neg logit contributions | |
ver | -0.033 |
acts | -0.032 |
youngest | -0.032 |
folk | -0.032 |
bars | -0.032 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
nodded | +0.013 |
youngest | +0.013 |
ver | +0.012 |
oldest | +0.012 |
Let | +0.012 |
Neg logit contributions | |
poisonous | -0.018 |
ocol | -0.017 |
ocket | -0.016 |
arter | -0.016 |
aster | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
acts | +0.064 |
ver | +0.063 |
youngest | +0.062 |
folk | +0.062 |
mun | +0.062 |
Neg logit contributions | |
poisonous | -0.056 |
ocket | -0.051 |
aster | -0.049 |
ocol | -0.049 |
arter | -0.047 |
Token | started |
Token ID | 2067 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
nodded | +0.006 |
youngest | +0.006 |
ver | +0.006 |
acts | +0.006 |
Let | +0.006 |
Neg logit contributions | |
poisonous | -0.011 |
ocol | -0.010 |
ocket | -0.010 |
aster | -0.010 |
arter | -0.010 |
Token | eating |
Token ID | 6600 |
Feature activation | -5.13e+00 |
Pos logit contributions | |
led | +0.116 |
Let | +0.115 |
nodded | +0.114 |
youngest | +0.113 |
acts | +0.109 |
Neg logit contributions | |
poisonous | -0.165 |
ocket | -0.156 |
arter | -0.154 |
ocol | -0.152 |
bamboo | -0.141 |
Token | the |
Token ID | 262 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
arter | +0.252 |
ocket | +0.251 |
poisonous | +0.250 |
ocol | +0.230 |
eds | +0.227 |
Neg logit contributions | |
nodded | -0.171 |
Let | -0.164 |
led | -0.163 |
acts | -0.159 |
So | -0.159 |
Token | ch |
Token ID | 442 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
nic | +0.055 |
youngest | +0.055 |
acts | +0.055 |
Let | +0.054 |
led | +0.054 |
Neg logit contributions | |
poisonous | -0.061 |
arter | -0.060 |
aster | -0.059 |
ocol | -0.059 |
ocket | -0.058 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
acts | +0.123 |
folk | +0.117 |
bars | +0.114 |
youngest | +0.114 |
lest | +0.110 |
Neg logit contributions | |
poisonous | -0.037 |
bamboo | -0.024 |
ocket | -0.021 |
brand | -0.021 |
injuries | -0.021 |
Token | ates |
Token ID | 689 |
Feature activation | +7.71e-01 |
Pos logit contributions | |
poisonous | +0.450 |
arter | +0.372 |
ocket | +0.370 |
ocol | +0.361 |
aster | +0.360 |
Neg logit contributions | |
ver | -0.512 |
folk | -0.493 |
acts | -0.492 |
nodded | -0.491 |
youngest | -0.490 |
Token | . |
Token ID | 13 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
ver | +0.032 |
nodded | +0.032 |
youngest | +0.031 |
acts | +0.031 |
bars | +0.031 |
Neg logit contributions | |
poisonous | -0.032 |
ocol | -0.029 |
ocket | -0.028 |
arter | -0.028 |
aster | -0.028 |
Token | The |
Token ID | 383 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
acts | +0.007 |
ver | +0.007 |
nodded | +0.007 |
youngest | +0.006 |
folk | +0.006 |
Neg logit contributions | |
poisonous | -0.010 |
ocol | -0.010 |
ocket | -0.010 |
aster | -0.009 |
arter | -0.009 |
Token | shop |
Token ID | 6128 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
youngest | +0.013 |
oldest | +0.012 |
nic | +0.012 |
nodded | +0.012 |
eldest | +0.012 |
Neg logit contributions | |
arter | -0.017 |
poisonous | -0.016 |
ocket | -0.015 |
ocol | -0.015 |
aster | -0.015 |
Token | keeper |
Token ID | 13884 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
acts | +0.078 |
ver | +0.076 |
youngest | +0.072 |
folk | +0.072 |
basics | +0.071 |
Neg logit contributions | |
poisonous | -0.117 |
ocket | -0.108 |
aster | -0.104 |
ocol | -0.102 |
bamboo | -0.099 |
Token | ates |
Token ID | 689 |
Feature activation | +6.97e-01 |
Pos logit contributions | |
poisonous | +0.444 |
arter | +0.370 |
ocket | +0.368 |
aster | +0.364 |
ocol | +0.357 |
Neg logit contributions | |
ver | -0.516 |
folk | -0.506 |
acts | -0.504 |
youngest | -0.502 |
nodded | -0.501 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
ver | +0.029 |
bars | +0.028 |
youngest | +0.028 |
nodded | +0.028 |
acts | +0.028 |
Neg logit contributions | |
poisonous | -0.029 |
ocol | -0.026 |
aster | -0.025 |
arter | -0.025 |
ocket | -0.025 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
poisonous | +0.055 |
ocol | +0.051 |
ocket | +0.051 |
aster | +0.050 |
arter | +0.049 |
Neg logit contributions | |
nodded | -0.035 |
ver | -0.035 |
acts | -0.035 |
youngest | -0.034 |
Let | -0.034 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
poisonous | +0.201 |
ocket | +0.191 |
aster | +0.177 |
bamboo | +0.176 |
isk | +0.168 |
Neg logit contributions | |
folk | -0.153 |
bars | -0.144 |
hardest | -0.143 |
ver | -0.141 |
youngest | -0.140 |
Token | ch |
Token ID | 442 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
poisonous | +0.082 |
ocket | +0.080 |
arter | +0.080 |
ocol | +0.074 |
aster | +0.073 |
Neg logit contributions | |
acts | -0.059 |
oldest | -0.058 |
youngest | -0.058 |
led | -0.058 |
asks | -0.058 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.14e+01 |
Pos logit contributions | |
acts | +0.106 |
folk | +0.103 |
ver | +0.101 |
bars | +0.099 |
nodded | +0.098 |
Neg logit contributions | |
poisonous | -0.036 |
ocket | -0.026 |
brand | -0.026 |
injuries | -0.024 |
aster | -0.023 |
Token | ates |
Token ID | 689 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
poisonous | +0.459 |
arter | +0.381 |
aster | +0.377 |
ocket | +0.376 |
ocol | +0.366 |
Neg logit contributions | |
ver | -0.501 |
folk | -0.485 |
acts | -0.485 |
nodded | -0.482 |
youngest | -0.480 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
nodded | +0.027 |
ver | +0.027 |
acts | +0.026 |
's | +0.026 |
led | +0.026 |
Neg logit contributions | |
poisonous | -0.036 |
ocol | -0.035 |
aster | -0.033 |
arter | -0.033 |
ocket | -0.032 |
Token | She |
Token ID | 1375 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
poisonous | +0.050 |
ocol | +0.047 |
ocket | +0.046 |
aster | +0.046 |
arter | +0.046 |
Neg logit contributions | |
youngest | -0.039 |
ver | -0.039 |
folk | -0.038 |
acts | -0.038 |
nodded | -0.037 |
Token | thanked |
Token ID | 26280 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
poisonous | +0.175 |
ocket | +0.164 |
aster | +0.157 |
bamboo | +0.156 |
ocol | +0.151 |
Neg logit contributions | |
folk | -0.101 |
ver | -0.096 |
bars | -0.095 |
youngest | -0.093 |
sets | -0.093 |
Token | Ben |
Token ID | 3932 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
youngest | +0.056 |
acts | +0.056 |
nodded | +0.056 |
oldest | +0.054 |
Let | +0.054 |
Neg logit contributions | |
poisonous | -0.067 |
ocket | -0.059 |
ocol | -0.057 |
arter | -0.057 |
aster | -0.053 |
Token | 't |
Token ID | 470 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
poisonous | +0.025 |
ocol | +0.024 |
ocket | +0.023 |
arter | +0.022 |
aster | +0.022 |
Neg logit contributions | |
ver | -0.017 |
nodded | -0.017 |
youngest | -0.017 |
Let | -0.016 |
acts | -0.016 |
Token | have |
Token ID | 423 |
Feature activation | -3.37e+00 |
Pos logit contributions | |
nodded | +0.089 |
Let | +0.088 |
youngest | +0.088 |
acts | +0.087 |
ver | +0.086 |
Neg logit contributions | |
poisonous | -0.131 |
ocol | -0.122 |
ocket | -0.121 |
arter | -0.117 |
bamboo | -0.113 |
Token | all |
Token ID | 477 |
Feature activation | +5.30e+00 |
Pos logit contributions | |
poisonous | +0.137 |
ocol | +0.125 |
ocket | +0.124 |
aster | +0.120 |
arter | +0.119 |
Neg logit contributions | |
nodded | -0.139 |
acts | -0.139 |
youngest | -0.139 |
ver | -0.139 |
Let | -0.137 |
Token | these |
Token ID | 777 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
ver | +0.246 |
youngest | +0.241 |
nodded | +0.240 |
acts | +0.236 |
folk | +0.235 |
Neg logit contributions | |
poisonous | -0.200 |
ocol | -0.175 |
aster | -0.173 |
ocket | -0.167 |
arter | -0.162 |
Token | ch |
Token ID | 442 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
ver | +0.011 |
nodded | +0.011 |
youngest | +0.011 |
acts | +0.011 |
Let | +0.011 |
Neg logit contributions | |
poisonous | -0.012 |
ocol | -0.011 |
ocket | -0.011 |
aster | -0.010 |
arter | -0.010 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
acts | +0.119 |
folk | +0.115 |
youngest | +0.115 |
bars | +0.114 |
nodded | +0.114 |
Neg logit contributions | |
poisonous | -0.023 |
ocket | -0.015 |
bamboo | -0.012 |
aster | -0.012 |
arter | -0.010 |
Token | ates |
Token ID | 689 |
Feature activation | +7.38e-01 |
Pos logit contributions | |
poisonous | +0.416 |
ocket | +0.352 |
aster | +0.349 |
arter | +0.347 |
ocol | +0.345 |
Neg logit contributions | |
ver | -0.527 |
acts | -0.517 |
nodded | -0.513 |
youngest | -0.513 |
folk | -0.512 |
Token | ! |
Token ID | 0 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
ver | +0.031 |
nodded | +0.030 |
acts | +0.030 |
youngest | +0.030 |
Let | +0.030 |
Neg logit contributions | |
poisonous | -0.029 |
ocol | -0.027 |
ocket | -0.026 |
aster | -0.026 |
arter | -0.026 |
Token | Now |
Token ID | 2735 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
poisonous | +0.030 |
ocol | +0.029 |
aster | +0.028 |
ocket | +0.027 |
heaven | +0.027 |
Neg logit contributions | |
nodded | -0.029 |
youngest | -0.029 |
led | -0.029 |
acts | -0.028 |
ver | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
poisonous | +0.062 |
ocket | +0.055 |
aster | +0.055 |
ocol | +0.055 |
bamboo | +0.051 |
Neg logit contributions | |
youngest | -0.061 |
ver | -0.060 |
bars | -0.059 |
folk | -0.059 |
acts | -0.059 |
Token | go |
Token ID | 467 |
Feature activation | -6.47e-01 |
Pos logit contributions | |
youngest | +0.011 |
acts | +0.011 |
folk | +0.011 |
ver | +0.011 |
nodded | +0.011 |
Neg logit contributions | |
poisonous | -0.012 |
aster | -0.010 |
ocket | -0.010 |
ocol | -0.010 |
arter | -0.010 |
Token | had |
Token ID | 550 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
ocol | +0.065 |
poisonous | +0.063 |
arter | +0.062 |
ocket | +0.061 |
aster | +0.061 |
Neg logit contributions | |
led | -0.019 |
Let | -0.019 |
nodded | -0.019 |
acts | -0.017 |
youngest | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
poisonous | +0.076 |
ocol | +0.073 |
ocket | +0.070 |
arter | +0.069 |
aster | +0.068 |
Neg logit contributions | |
Let | -0.076 |
youngest | -0.075 |
nodded | -0.074 |
led | -0.074 |
oldest | -0.073 |
Token | big |
Token ID | 1263 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
ocol | +0.014 |
poisonous | +0.014 |
ocket | +0.013 |
aster | +0.013 |
arter | +0.012 |
Neg logit contributions | |
youngest | -0.011 |
nodded | -0.011 |
oldest | -0.010 |
Let | -0.010 |
led | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
poisonous | +0.142 |
ocol | +0.131 |
ocket | +0.131 |
aster | +0.128 |
arter | +0.124 |
Neg logit contributions | |
nodded | -0.107 |
acts | -0.106 |
youngest | -0.105 |
ver | -0.103 |
Let | -0.102 |
Token | bus |
Token ID | 1323 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
nodded | +0.001 |
Let | +0.001 |
youngest | +0.001 |
led | +0.001 |
nic | +0.001 |
Neg logit contributions | |
poisonous | -0.002 |
ocol | -0.002 |
ocket | -0.002 |
aster | -0.002 |
arter | -0.002 |
Token | hy |
Token ID | 12114 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
poisonous | +0.082 |
ocol | +0.079 |
ocket | +0.072 |
arter | +0.070 |
aster | +0.069 |
Neg logit contributions | |
nodded | -0.092 |
ver | -0.091 |
led | -0.091 |
bars | -0.090 |
Let | -0.089 |
Token | tail |
Token ID | 7894 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
poisonous | +0.499 |
ocol | +0.460 |
ocket | +0.455 |
aster | +0.443 |
arter | +0.426 |
Neg logit contributions | |
nodded | -0.449 |
youngest | -0.429 |
acts | -0.421 |
folk | -0.419 |
ver | -0.417 |
Token | and |
Token ID | 290 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
poisonous | +0.002 |
ocket | +0.002 |
ocol | +0.002 |
aster | +0.002 |
arter | +0.002 |
Neg logit contributions | |
acts | -0.003 |
ver | -0.003 |
youngest | -0.002 |
nodded | -0.002 |
folk | -0.002 |
Token | sharp |
Token ID | 7786 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
poisonous | +0.012 |
ocket | +0.010 |
ocol | +0.010 |
aster | +0.010 |
bamboo | +0.010 |
Neg logit contributions | |
ver | -0.010 |
acts | -0.010 |
youngest | -0.010 |
folk | -0.010 |
nodded | -0.010 |
Token | teeth |
Token ID | 9941 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
poisonous | +0.046 |
ocol | +0.041 |
ocket | +0.041 |
aster | +0.041 |
arter | +0.040 |
Neg logit contributions | |
acts | -0.038 |
nodded | -0.038 |
youngest | -0.037 |
ver | -0.037 |
Let | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
ocol | +0.088 |
poisonous | +0.087 |
arter | +0.082 |
ocket | +0.082 |
aster | +0.081 |
Neg logit contributions | |
nodded | -0.062 |
ver | -0.062 |
led | -0.061 |
bars | -0.058 |
Let | -0.058 |
Token | son |
Token ID | 3367 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
poisonous | +0.119 |
aster | +0.110 |
ocket | +0.110 |
arter | +0.110 |
ocol | +0.109 |
Neg logit contributions | |
youngest | -0.087 |
acts | -0.086 |
nodded | -0.085 |
oldest | -0.083 |
nic | -0.080 |
Token | is |
Token ID | 318 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
youngest | +0.103 |
ver | +0.102 |
acts | +0.101 |
folk | +0.101 |
nodded | +0.101 |
Neg logit contributions | |
poisonous | -0.125 |
ocket | -0.112 |
ocol | -0.111 |
aster | -0.110 |
arter | -0.107 |
Token | eating |
Token ID | 6600 |
Feature activation | -5.22e+00 |
Pos logit contributions | |
youngest | +0.004 |
nodded | +0.004 |
oldest | +0.004 |
led | +0.004 |
ver | +0.004 |
Neg logit contributions | |
poisonous | -0.004 |
ocket | -0.004 |
ocol | -0.004 |
arter | -0.004 |
aster | -0.004 |
Token | our |
Token ID | 674 |
Feature activation | -9.41e-01 |
Pos logit contributions | |
poisonous | +0.225 |
arter | +0.221 |
ocket | +0.221 |
ocol | +0.203 |
aster | +0.196 |
Neg logit contributions | |
Let | -0.189 |
nodded | -0.186 |
acts | -0.185 |
youngest | -0.185 |
nic | -0.183 |
Token | ch |
Token ID | 442 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
poisonous | +0.038 |
ocket | +0.036 |
arter | +0.035 |
aster | +0.035 |
ocol | +0.034 |
Neg logit contributions | |
acts | -0.040 |
youngest | -0.039 |
oldest | -0.038 |
nodded | -0.038 |
nic | -0.038 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
acts | +0.120 |
youngest | +0.114 |
folk | +0.113 |
bars | +0.112 |
oldest | +0.111 |
Neg logit contributions | |
poisonous | -0.032 |
ocket | -0.020 |
bamboo | -0.020 |
aster | -0.017 |
brand | -0.016 |
Token | ates |
Token ID | 689 |
Feature activation | +7.75e-01 |
Pos logit contributions | |
poisonous | +0.443 |
arter | +0.368 |
ocket | +0.364 |
aster | +0.363 |
ocol | +0.359 |
Neg logit contributions | |
ver | -0.507 |
acts | -0.490 |
nodded | -0.487 |
youngest | -0.486 |
folk | -0.484 |
Token | !" |
Token ID | 2474 |
Feature activation | +5.88e-01 |
Pos logit contributions | |
ver | +0.034 |
nodded | +0.033 |
youngest | +0.033 |
acts | +0.033 |
Let | +0.033 |
Neg logit contributions | |
poisonous | -0.030 |
ocol | -0.027 |
ocket | -0.027 |
aster | -0.026 |
arter | -0.026 |
Token | |
Token ID | 220 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
acts | +0.020 |
youngest | +0.020 |
ver | +0.020 |
oldest | +0.019 |
folk | +0.019 |
Neg logit contributions | |
poisonous | -0.028 |
aster | -0.026 |
ocol | -0.026 |
ocket | -0.026 |
arter | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +9.59e-01 |
Pos logit contributions | |
acts | +0.156 |
youngest | +0.151 |
folk | +0.149 |
nodded | +0.149 |
oldest | +0.148 |
Neg logit contributions | |
poisonous | -0.154 |
aster | -0.142 |
ocol | -0.140 |
ocket | -0.137 |
arter | -0.134 |
Token |
|
Token ID | 198 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
nodded | +0.032 |
ver | +0.032 |
Let | +0.031 |
youngest | +0.031 |
acts | +0.030 |
Neg logit contributions | |
poisonous | -0.048 |
ocol | -0.044 |
ocket | -0.044 |
arter | -0.043 |
aster | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
youngest | +0.124 |
acts | +0.124 |
ver | +0.124 |
nodded | +0.123 |
Let | +0.121 |
Neg logit contributions | |
poisonous | -0.127 |
ocket | -0.113 |
ocol | -0.112 |
aster | -0.111 |
arter | -0.109 |
Token | jumped |
Token ID | 11687 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
poisonous | +0.038 |
aster | +0.034 |
bamboo | +0.033 |
ocket | +0.033 |
ocol | +0.032 |
Neg logit contributions | |
ver | -0.024 |
folk | -0.024 |
youngest | -0.023 |
bars | -0.023 |
acts | -0.022 |
Token | in |
Token ID | 287 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
Let | +0.049 |
led | +0.049 |
nodded | +0.047 |
neither | +0.045 |
ver | +0.045 |
Neg logit contributions | |
ocol | -0.070 |
ocket | -0.070 |
poisonous | -0.069 |
aster | -0.068 |
arter | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
Let | +0.063 |
led | +0.062 |
acts | +0.060 |
nodded | +0.060 |
So | +0.058 |
Neg logit contributions | |
ocket | -0.098 |
poisonous | -0.097 |
arter | -0.094 |
ocol | -0.093 |
affle | -0.088 |
Token | Anna |
Token ID | 11735 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
poisonous | +0.072 |
ocket | +0.068 |
ocol | +0.068 |
arter | +0.064 |
aster | +0.064 |
Neg logit contributions | |
nodded | -0.034 |
Let | -0.034 |
youngest | -0.033 |
ver | -0.033 |
acts | -0.032 |
Token | followed |
Token ID | 3940 |
Feature activation | +5.48e+00 |
Pos logit contributions | |
poisonous | +0.084 |
ocket | +0.072 |
bamboo | +0.072 |
injuries | +0.071 |
aster | +0.071 |
Neg logit contributions | |
sets | -0.074 |
youngest | -0.070 |
folk | -0.070 |
bars | -0.070 |
oldest | -0.069 |
Token | him |
Token ID | 683 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
youngest | +0.223 |
Let | +0.219 |
nodded | +0.217 |
acts | +0.217 |
neither | +0.212 |
Neg logit contributions | |
poisonous | -0.239 |
arter | -0.208 |
ocol | -0.206 |
ocket | -0.206 |
brand | -0.190 |
Token | . |
Token ID | 13 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
poisonous | +0.071 |
ocket | +0.063 |
ocol | +0.062 |
aster | +0.061 |
arter | +0.060 |
Neg logit contributions | |
youngest | -0.072 |
nodded | -0.072 |
acts | -0.072 |
ver | -0.072 |
folk | -0.071 |
Token | She |
Token ID | 1375 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
poisonous | +0.069 |
ocol | +0.064 |
ocket | +0.064 |
aster | +0.063 |
arter | +0.062 |
Neg logit contributions | |
ver | -0.037 |
acts | -0.036 |
nodded | -0.036 |
youngest | -0.036 |
Let | -0.036 |
Token | did |
Token ID | 750 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
poisonous | +0.179 |
bamboo | +0.158 |
ocket | +0.157 |
aster | +0.153 |
injuries | +0.152 |
Neg logit contributions | |
folk | -0.121 |
sets | -0.118 |
bars | -0.116 |
ver | -0.115 |
youngest | -0.115 |
Token | not |
Token ID | 407 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
ocol | +0.084 |
ocket | +0.083 |
poisonous | +0.077 |
arter | +0.076 |
affle | +0.073 |
Neg logit contributions | |
neither | -0.039 |
nodded | -0.038 |
ver | -0.038 |
led | -0.037 |
hardest | -0.034 |
Token | in |
Token ID | 287 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
acts | +0.041 |
led | +0.040 |
hardest | +0.039 |
youngest | +0.038 |
means | +0.038 |
Neg logit contributions | |
poisonous | -0.063 |
aster | -0.059 |
arter | -0.058 |
ocket | -0.056 |
isk | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | -6.20e-01 |
Pos logit contributions | |
led | +0.061 |
acts | +0.060 |
Let | +0.059 |
youngest | +0.058 |
way | +0.057 |
Neg logit contributions | |
poisonous | -0.096 |
arter | -0.089 |
ocket | -0.088 |
affle | -0.086 |
aster | -0.084 |
Token | meet |
Token ID | 1826 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
poisonous | +0.027 |
ocol | +0.025 |
ocket | +0.025 |
aster | +0.024 |
arter | +0.024 |
Neg logit contributions | |
ver | -0.024 |
acts | -0.024 |
youngest | -0.024 |
nodded | -0.023 |
folk | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
poisonous | +0.075 |
ocol | +0.068 |
ocket | +0.067 |
aster | +0.066 |
arter | +0.065 |
Neg logit contributions | |
ver | -0.065 |
acts | -0.065 |
nodded | -0.065 |
youngest | -0.065 |
Let | -0.064 |
Token | other |
Token ID | 584 |
Feature activation | +4.14e+00 |
Pos logit contributions | |
ver | +0.051 |
nodded | +0.049 |
acts | +0.048 |
bars | +0.048 |
Let | +0.048 |
Neg logit contributions | |
poisonous | -0.055 |
ocket | -0.049 |
ocol | -0.049 |
bamboo | -0.047 |
aster | -0.046 |
Token | boats |
Token ID | 16659 |
Feature activation | +5.91e+00 |
Pos logit contributions | |
ver | +0.173 |
nodded | +0.163 |
bars | +0.163 |
Let | +0.155 |
folk | +0.151 |
Neg logit contributions | |
poisonous | -0.177 |
ocol | -0.164 |
bamboo | -0.163 |
ocket | -0.160 |
arter | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
youngest | +0.260 |
acts | +0.260 |
ver | +0.259 |
nodded | +0.259 |
Let | +0.258 |
Neg logit contributions | |
poisonous | -0.222 |
ocket | -0.198 |
ocol | -0.195 |
aster | -0.190 |
arter | -0.187 |
Token | The |
Token ID | 383 |
Feature activation | -8.31e-02 |
Pos logit contributions | |
aster | +0.029 |
poisonous | +0.028 |
bamboo | +0.027 |
ocol | +0.026 |
heaven | +0.026 |
Neg logit contributions | |
acts | -0.033 |
oldest | -0.032 |
youngest | -0.032 |
ver | -0.032 |
bars | -0.032 |
Token | yacht |
Token ID | 43207 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
poisonous | +0.004 |
ocket | +0.004 |
bamboo | +0.004 |
ocol | +0.004 |
aster | +0.003 |
Neg logit contributions | |
ver | -0.003 |
bars | -0.003 |
nodded | -0.003 |
So | -0.003 |
acts | -0.003 |
Token | quickly |
Token ID | 2952 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
First | +0.080 |
First | +0.076 |
Und | +0.075 |
acts | +0.074 |
mun | +0.074 |
Neg logit contributions | |
poisonous | -0.095 |
ocket | -0.084 |
buried | -0.082 |
bamboo | -0.080 |
brand | -0.079 |
Token | gathered |
Token ID | 9272 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
mun | +0.008 |
ver | +0.008 |
sets | +0.008 |
oldest | +0.008 |
folk | +0.008 |
Neg logit contributions | |
poisonous | -0.011 |
ocket | -0.010 |
aster | -0.010 |
bamboo | -0.010 |
ocol | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
youngest | +0.004 |
Let | +0.004 |
acts | +0.004 |
led | +0.004 |
oldest | +0.004 |
Neg logit contributions | |
poisonous | -0.005 |
ocket | -0.005 |
ocol | -0.005 |
arter | -0.005 |
injuries | -0.004 |
Token | beach |
Token ID | 10481 |
Feature activation | +9.21e-01 |
Pos logit contributions | |
youngest | +0.054 |
oldest | +0.052 |
Let | +0.051 |
eldest | +0.048 |
nic | +0.047 |
Neg logit contributions | |
ocol | -0.097 |
poisonous | -0.096 |
ocket | -0.092 |
arter | -0.088 |
injuries | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
ver | +0.045 |
acts | +0.045 |
youngest | +0.045 |
folk | +0.044 |
nodded | +0.044 |
Neg logit contributions | |
poisonous | -0.030 |
aster | -0.027 |
ocket | -0.026 |
ocol | -0.025 |
bamboo | -0.025 |
Token | He |
Token ID | 679 |
Feature activation | -9.44e-01 |
Pos logit contributions | |
acts | +0.001 |
ver | +0.001 |
mun | +0.001 |
folk | +0.001 |
youngest | +0.001 |
Neg logit contributions | |
poisonous | -0.001 |
aster | -0.001 |
ocol | -0.001 |
arter | -0.001 |
ocket | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
poisonous | +0.038 |
aster | +0.034 |
bamboo | +0.033 |
ocket | +0.033 |
sm | +0.031 |
Neg logit contributions | |
folk | -0.039 |
sets | -0.038 |
acts | -0.038 |
youngest | -0.037 |
ver | -0.037 |
Token | excited |
Token ID | 6568 |
Feature activation | +5.09e+00 |
Pos logit contributions | |
youngest | +0.021 |
nodded | +0.021 |
ver | +0.021 |
Let | +0.021 |
oldest | +0.021 |
Neg logit contributions | |
poisonous | -0.021 |
ocol | -0.020 |
ocket | -0.020 |
aster | -0.019 |
arter | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
ver | +0.222 |
acts | +0.220 |
nodded | +0.218 |
youngest | +0.218 |
folk | +0.216 |
Neg logit contributions | |
poisonous | -0.199 |
ocket | -0.175 |
aster | -0.174 |
ocol | -0.174 |
arter | -0.169 |
Token | explore |
Token ID | 7301 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
ver | +0.002 |
acts | +0.002 |
folk | +0.002 |
nodded | +0.002 |
bars | +0.002 |
Neg logit contributions | |
poisonous | -0.002 |
aster | -0.002 |
ocol | -0.002 |
ocket | -0.002 |
arter | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
poisonous | +0.023 |
ocket | +0.021 |
ocol | +0.021 |
arter | +0.020 |
aster | +0.020 |
Neg logit contributions | |
nodded | -0.018 |
acts | -0.018 |
led | -0.018 |
Let | -0.017 |
bars | -0.017 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
ver | +0.062 |
nodded | +0.059 |
acts | +0.059 |
folk | +0.058 |
youngest | +0.057 |
Neg logit contributions | |
poisonous | -0.063 |
ocol | -0.056 |
ocket | -0.056 |
aster | -0.054 |
bamboo | -0.054 |
Token | blue |
Token ID | 4171 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
ver | +0.013 |
nodded | +0.013 |
acts | +0.013 |
youngest | +0.013 |
Let | +0.013 |
Neg logit contributions | |
poisonous | -0.017 |
ocol | -0.016 |
ocket | -0.016 |
aster | -0.015 |
arter | -0.015 |
Token | € |
Token ID | 26391 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
youngest | +0.130 |
ver | +0.129 |
acts | +0.128 |
oldest | +0.125 |
nodded | +0.125 |
Neg logit contributions | |
poisonous | -0.179 |
ocket | -0.164 |
aster | -0.164 |
ocol | -0.164 |
arter | -0.156 |
Token | |
Token ID | 220 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
poisonous | +0.035 |
ocket | +0.033 |
ocol | +0.032 |
aster | +0.032 |
arter | +0.031 |
Neg logit contributions | |
youngest | -0.026 |
acts | -0.025 |
nodded | -0.025 |
oldest | -0.025 |
ver | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +8.68e-01 |
Pos logit contributions | |
acts | +0.125 |
nodded | +0.125 |
youngest | +0.124 |
ver | +0.124 |
Let | +0.121 |
Neg logit contributions | |
poisonous | -0.188 |
ocol | -0.173 |
ocket | -0.172 |
aster | -0.169 |
arter | -0.166 |
Token |
|
Token ID | 198 |
Feature activation | +8.85e-01 |
Pos logit contributions | |
Let | +0.025 |
ver | +0.025 |
nodded | +0.024 |
youngest | +0.023 |
led | +0.023 |
Neg logit contributions | |
poisonous | -0.048 |
ocol | -0.044 |
ocket | -0.044 |
arter | -0.044 |
aster | -0.042 |
Token | The |
Token ID | 464 |
Feature activation | +6.87e-01 |
Pos logit contributions | |
Let | +0.032 |
nodded | +0.030 |
ver | +0.030 |
youngest | +0.030 |
folk | +0.029 |
Neg logit contributions | |
poisonous | -0.043 |
ocket | -0.040 |
arter | -0.039 |
ocol | -0.039 |
aster | -0.037 |
Token | older |
Token ID | 4697 |
Feature activation | +5.85e+00 |
Pos logit contributions | |
ver | +0.021 |
nodded | +0.020 |
acts | +0.020 |
folk | +0.020 |
bars | +0.019 |
Neg logit contributions | |
poisonous | -0.037 |
ocol | -0.034 |
ocket | -0.033 |
aster | -0.033 |
arter | -0.032 |
Token | girl |
Token ID | 2576 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
ver | +0.173 |
First | +0.153 |
led | +0.149 |
bars | +0.149 |
Let | +0.146 |
Neg logit contributions | |
poisonous | -0.297 |
aster | -0.290 |
ocol | -0.289 |
ocket | -0.275 |
bamboo | -0.265 |
Token | smiled |
Token ID | 13541 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
ver | +0.017 |
youngest | +0.017 |
bars | +0.017 |
folk | +0.017 |
acts | +0.017 |
Neg logit contributions | |
poisonous | -0.032 |
ocket | -0.030 |
aster | -0.029 |
ocol | -0.028 |
bamboo | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
ver | +0.166 |
acts | +0.166 |
youngest | +0.165 |
nodded | +0.164 |
Let | +0.163 |
Neg logit contributions | |
poisonous | -0.173 |
ocol | -0.156 |
ocket | -0.155 |
aster | -0.151 |
arter | -0.149 |
Token | said |
Token ID | 531 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
folk | +0.001 |
ver | +0.001 |
acts | +0.001 |
bars | +0.001 |
youngest | +0.001 |
Neg logit contributions | |
poisonous | -0.002 |
ocket | -0.001 |
aster | -0.001 |
bamboo | -0.001 |
ocol | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
youngest | +0.053 |
acts | +0.052 |
oldest | +0.051 |
bars | +0.051 |
folk | +0.051 |
Neg logit contributions | |
poisonous | -0.050 |
ocol | -0.045 |
ocket | -0.045 |
aster | -0.043 |
bamboo | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ver | +0.067 |
bars | +0.065 |
led | +0.064 |
acts | +0.062 |
nodded | +0.062 |
Neg logit contributions | |
poisonous | -0.073 |
ocol | -0.071 |
aster | -0.067 |
arter | -0.066 |
ocket | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | -8.33e-02 |
Pos logit contributions | |
poisonous | +0.058 |
ocket | +0.054 |
ocol | +0.053 |
aster | +0.052 |
arter | +0.051 |
Neg logit contributions | |
youngest | -0.043 |
acts | -0.043 |
folk | -0.042 |
nodded | -0.042 |
bars | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
poisonous | +0.005 |
ocol | +0.005 |
ocket | +0.005 |
arter | +0.005 |
affle | +0.005 |
Neg logit contributions | |
Let | -0.002 |
So | -0.002 |
ver | -0.002 |
nodded | -0.002 |
led | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
poisonous | +0.003 |
aster | +0.003 |
ocket | +0.003 |
ocol | +0.003 |
bamboo | +0.003 |
Neg logit contributions | |
acts | -0.003 |
youngest | -0.003 |
ver | -0.003 |
oldest | -0.003 |
bars | -0.003 |
Token | Can |
Token ID | 6090 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
ver | +0.038 |
bars | +0.038 |
youngest | +0.037 |
acts | +0.037 |
folk | +0.036 |
Neg logit contributions | |
poisonous | -0.046 |
ocket | -0.041 |
aster | -0.041 |
ocol | -0.040 |
bamboo | -0.038 |
Token | we |
Token ID | 356 |
Feature activation | +5.67e+00 |
Pos logit contributions | |
poisonous | +0.179 |
aster | +0.164 |
ocket | +0.163 |
ocol | +0.160 |
bamboo | +0.156 |
Neg logit contributions | |
youngest | -0.098 |
acts | -0.095 |
nodded | -0.093 |
ver | -0.092 |
folk | -0.092 |
Token | play |
Token ID | 711 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
ver | +0.194 |
youngest | +0.193 |
nodded | +0.191 |
led | +0.188 |
acts | +0.186 |
Neg logit contributions | |
poisonous | -0.277 |
ocol | -0.257 |
ocket | -0.251 |
arter | -0.245 |
aster | -0.241 |
Token | with |
Token ID | 351 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
poisonous | +0.226 |
ocol | +0.223 |
arter | +0.222 |
ocket | +0.204 |
eds | +0.201 |
Neg logit contributions | |
's | -0.115 |
nodded | -0.111 |
nic | -0.110 |
led | -0.107 |
Let | -0.103 |
Token | you |
Token ID | 345 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
acts | +0.001 |
's | +0.001 |
Let | +0.001 |
led | +0.001 |
folk | +0.001 |
Neg logit contributions | |
poisonous | -0.001 |
arter | -0.001 |
ocket | -0.001 |
affle | -0.001 |
ocol | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
ver | +0.067 |
nodded | +0.066 |
led | +0.061 |
's | +0.061 |
means | +0.060 |
Neg logit contributions | |
poisonous | -0.106 |
ocol | -0.105 |
ocket | -0.099 |
arter | -0.099 |
bones | -0.096 |
Token | bird |
Token ID | 6512 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
poisonous | +0.003 |
ocol | +0.002 |
ocket | +0.002 |
aster | +0.002 |
arter | +0.002 |
Neg logit contributions | |
ver | -0.001 |
acts | -0.001 |
nodded | -0.001 |
youngest | -0.001 |
Let | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +7.51e-02 |
Pos logit contributions | |
nodded | +0.071 |
ver | +0.070 |
youngest | +0.069 |
acts | +0.068 |
led | +0.066 |
Neg logit contributions | |
poisonous | -0.117 |
ocket | -0.109 |
ocol | -0.107 |
aster | -0.105 |
arter | -0.103 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
nodded | +0.002 |
acts | +0.002 |
ver | +0.002 |
Let | +0.002 |
youngest | +0.002 |
Neg logit contributions | |
poisonous | -0.004 |
ocol | -0.004 |
ocket | -0.004 |
aster | -0.003 |
arter | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +8.47e-02 |
Pos logit contributions | |
poisonous | +0.163 |
ocket | +0.142 |
ocol | +0.138 |
aster | +0.135 |
arter | +0.135 |
Neg logit contributions | |
youngest | -0.187 |
bars | -0.184 |
folk | -0.183 |
oldest | -0.182 |
ver | -0.181 |
Token | we |
Token ID | 356 |
Feature activation | +5.65e+00 |
Pos logit contributions | |
youngest | +0.003 |
acts | +0.003 |
nodded | +0.003 |
ver | +0.003 |
folk | +0.003 |
Neg logit contributions | |
poisonous | -0.004 |
ocket | -0.004 |
ocol | -0.004 |
aster | -0.004 |
arter | -0.003 |
Token | did |
Token ID | 750 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
acts | +0.154 |
youngest | +0.153 |
ver | +0.153 |
nodded | +0.152 |
folk | +0.150 |
Neg logit contributions | |
poisonous | -0.311 |
ocket | -0.288 |
ocol | -0.288 |
aster | -0.283 |
arter | -0.278 |
Token | it |
Token ID | 340 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
poisonous | +0.065 |
ocket | +0.063 |
ocol | +0.061 |
arter | +0.060 |
isk | +0.059 |
Neg logit contributions | |
So | -0.052 |
nodded | -0.051 |
neither | -0.051 |
's | -0.050 |
ver | -0.050 |
Token | ! |
Token ID | 0 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
poisonous | +0.022 |
ocol | +0.019 |
ocket | +0.019 |
aster | +0.018 |
arter | +0.017 |
Neg logit contributions | |
ver | -0.037 |
nodded | -0.037 |
acts | -0.037 |
youngest | -0.037 |
Let | -0.037 |
Token | We |
Token ID | 775 |
Feature activation | +3.77e+00 |
Pos logit contributions | |
arter | +0.041 |
ocol | +0.040 |
poisonous | +0.039 |
aster | +0.039 |
oof | +0.038 |
Neg logit contributions | |
mun | -0.044 |
youngest | -0.042 |
folk | -0.042 |
led | -0.041 |
nodded | -0.041 |
Token | are |
Token ID | 389 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
bars | +0.118 |
folk | +0.116 |
youngest | +0.115 |
acts | +0.114 |
ver | +0.112 |
Neg logit contributions | |
poisonous | -0.192 |
ocket | -0.178 |
ocol | -0.175 |
aster | -0.174 |
arter | -0.166 |
Token | the |
Token ID | 262 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
poisonous | +0.025 |
ocol | +0.023 |
ocket | +0.023 |
aster | +0.022 |
arter | +0.022 |
Neg logit contributions | |
youngest | -0.020 |
oldest | -0.020 |
nodded | -0.019 |
ver | -0.019 |
nic | -0.019 |
Token | best |
Token ID | 1266 |
Feature activation | +5.88e+00 |
Pos logit contributions | |
acts | +0.023 |
nodded | +0.022 |
ver | +0.021 |
folk | +0.021 |
led | +0.021 |
Neg logit contributions | |
poisonous | -0.069 |
aster | -0.064 |
ocket | -0.062 |
ocol | -0.062 |
bamboo | -0.062 |
Token | started |
Token ID | 2067 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
poisonous | +0.118 |
ocket | +0.108 |
aster | +0.106 |
ez | +0.099 |
eds | +0.098 |
Neg logit contributions | |
folk | -0.108 |
ver | -0.105 |
Und | -0.104 |
bars | -0.104 |
quickest | -0.103 |
Token | playing |
Token ID | 2712 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
youngest | +0.126 |
acts | +0.126 |
nodded | +0.125 |
Let | +0.125 |
led | +0.123 |
Neg logit contributions | |
poisonous | -0.138 |
ocket | -0.127 |
ocol | -0.126 |
arter | -0.120 |
aster | -0.119 |
Token | with |
Token ID | 351 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
poisonous | +0.137 |
ocket | +0.132 |
arter | +0.131 |
ocol | +0.130 |
injuries | +0.121 |
Neg logit contributions | |
Let | -0.080 |
led | -0.079 |
nodded | -0.078 |
's | -0.074 |
acts | -0.072 |
Token | the |
Token ID | 262 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
youngest | +0.004 |
led | +0.004 |
acts | +0.004 |
Let | +0.004 |
oldest | +0.004 |
Neg logit contributions | |
poisonous | -0.006 |
ocket | -0.006 |
arter | -0.006 |
prise | -0.006 |
alo | -0.006 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
youngest | +0.060 |
acts | +0.059 |
Let | +0.059 |
nodded | +0.058 |
oldest | +0.058 |
Neg logit contributions | |
poisonous | -0.071 |
ocol | -0.066 |
ocket | -0.066 |
aster | -0.065 |
arter | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
poisonous | +0.111 |
ocol | +0.103 |
ocket | +0.101 |
aster | +0.098 |
arter | +0.097 |
Neg logit contributions | |
youngest | -0.087 |
acts | -0.087 |
ver | -0.086 |
Let | -0.086 |
nodded | -0.085 |
Token | He |
Token ID | 679 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
youngest | +0.008 |
Let | +0.007 |
nodded | +0.007 |
acts | +0.007 |
ver | +0.007 |
Neg logit contributions | |
poisonous | -0.015 |
ocket | -0.014 |
ocol | -0.014 |
aster | -0.013 |
arter | -0.013 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
poisonous | +0.048 |
ocket | +0.044 |
aster | +0.043 |
bamboo | +0.040 |
brand | +0.040 |
Neg logit contributions | |
folk | -0.040 |
ver | -0.038 |
bars | -0.038 |
acts | -0.038 |
Und | -0.037 |
Token | watching |
Token ID | 4964 |
Feature activation | +7.33e-01 |
Pos logit contributions | |
poisonous | +0.088 |
ocket | +0.086 |
ocol | +0.083 |
arter | +0.082 |
prise | +0.078 |
Neg logit contributions | |
Let | -0.061 |
led | -0.058 |
acts | -0.056 |
basics | -0.055 |
would | -0.055 |
Token | it |
Token ID | 340 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
's | +0.022 |
acts | +0.021 |
youngest | +0.021 |
led | +0.021 |
means | +0.020 |
Neg logit contributions | |
poisonous | -0.040 |
ocket | -0.037 |
prise | -0.036 |
arter | -0.036 |
injuries | -0.035 |
Token | print |
Token ID | 3601 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
poisonous | +0.019 |
ocket | +0.018 |
ocol | +0.018 |
aster | +0.017 |
injuries | +0.017 |
Neg logit contributions | |
led | -0.013 |
acts | -0.013 |
youngest | -0.013 |
means | -0.012 |
nodded | -0.012 |
Token | so |
Token ID | 523 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
nodded | +0.009 |
acts | +0.009 |
Let | +0.009 |
youngest | +0.009 |
ver | +0.009 |
Neg logit contributions | |
poisonous | -0.010 |
ocol | -0.009 |
ocket | -0.009 |
arter | -0.009 |
aster | -0.009 |
Token | she |
Token ID | 673 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
nodded | +0.011 |
Let | +0.010 |
led | +0.010 |
ver | +0.010 |
youngest | +0.010 |
Neg logit contributions | |
poisonous | -0.019 |
ocol | -0.018 |
ocket | -0.018 |
arter | -0.018 |
aster | -0.017 |
Token | asked |
Token ID | 1965 |
Feature activation | +4.53e+00 |
Pos logit contributions | |
acts | +0.012 |
folk | +0.012 |
ver | +0.012 |
bars | +0.011 |
sets | +0.011 |
Neg logit contributions | |
poisonous | -0.028 |
ocket | -0.025 |
aster | -0.024 |
bamboo | -0.024 |
ocol | -0.023 |
Token | if |
Token ID | 611 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
nodded | +0.197 |
ver | +0.197 |
acts | +0.194 |
led | +0.193 |
Let | +0.190 |
Neg logit contributions | |
poisonous | -0.184 |
ocol | -0.168 |
arter | -0.167 |
ocket | -0.162 |
aster | -0.156 |
Token | she |
Token ID | 673 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
youngest | +0.042 |
bars | +0.040 |
acts | +0.040 |
folk | +0.040 |
eldest | +0.040 |
Neg logit contributions | |
poisonous | -0.043 |
ocket | -0.039 |
aster | -0.039 |
ocol | -0.037 |
bamboo | -0.036 |
Token | could |
Token ID | 714 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
folk | +0.010 |
youngest | +0.010 |
acts | +0.010 |
mun | +0.010 |
bars | +0.010 |
Neg logit contributions | |
poisonous | -0.029 |
ocket | -0.027 |
ocol | -0.026 |
aster | -0.026 |
arter | -0.025 |
Token | borrow |
Token ID | 8804 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
ver | +0.029 |
nodded | +0.027 |
acts | +0.027 |
led | +0.027 |
Let | +0.027 |
Neg logit contributions | |
poisonous | -0.032 |
ocol | -0.030 |
arter | -0.028 |
ocket | -0.028 |
bamboo | -0.027 |
Token | it |
Token ID | 340 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
arter | +0.237 |
poisonous | +0.227 |
ocket | +0.207 |
affle | +0.206 |
ocol | +0.206 |
Neg logit contributions | |
's | -0.110 |
means | -0.108 |
acts | -0.105 |
led | -0.102 |
So | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
poisonous | +0.024 |
ocol | +0.024 |
arter | +0.023 |
aster | +0.023 |
ocket | +0.022 |
Neg logit contributions | |
ver | -0.014 |
So | -0.014 |
's | -0.014 |
led | -0.014 |
means | -0.014 |
Token | |
Token ID | 220 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
aster | +0.006 |
poisonous | +0.006 |
heaven | +0.005 |
bamboo | +0.005 |
ocket | +0.005 |
Neg logit contributions | |
acts | -0.009 |
ver | -0.009 |
youngest | -0.009 |
folk | -0.008 |
hardest | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +6.48e-01 |
Pos logit contributions | |
acts | +0.137 |
youngest | +0.136 |
nodded | +0.133 |
oldest | +0.132 |
folk | +0.131 |
Neg logit contributions | |
poisonous | -0.154 |
ocket | -0.141 |
ocol | -0.139 |
aster | -0.138 |
arter | -0.134 |
Token | big |
Token ID | 1263 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
ocol | +0.019 |
aster | +0.018 |
ocket | +0.018 |
poisonous | +0.018 |
ez | +0.018 |
Neg logit contributions | |
youngest | -0.011 |
oldest | -0.010 |
Let | -0.010 |
nodded | -0.010 |
led | -0.010 |
Token | brown |
Token ID | 7586 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
poisonous | +0.163 |
ocket | +0.159 |
ocol | +0.153 |
aster | +0.148 |
injuries | +0.148 |
Neg logit contributions | |
acts | -0.088 |
nodded | -0.085 |
way | -0.082 |
led | -0.081 |
oldest | -0.081 |
Token | truck |
Token ID | 7779 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
poisonous | +0.068 |
ocket | +0.065 |
ocol | +0.064 |
injuries | +0.062 |
bones | +0.062 |
Neg logit contributions | |
acts | -0.035 |
youngest | -0.034 |
boy | -0.034 |
Let | -0.034 |
nodded | -0.034 |
Token | outside |
Token ID | 2354 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
Let | +0.062 |
led | +0.061 |
bars | +0.060 |
youngest | +0.060 |
nodded | +0.060 |
Neg logit contributions | |
poisonous | -0.068 |
ocol | -0.064 |
ocket | -0.062 |
arter | -0.058 |
aster | -0.057 |
Token | her |
Token ID | 607 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
poisonous | +0.010 |
ocket | +0.010 |
ocol | +0.009 |
affle | +0.009 |
arter | +0.009 |
Neg logit contributions | |
Let | -0.009 |
led | -0.008 |
nodded | -0.008 |
acts | -0.008 |
's | -0.008 |
Token | window |
Token ID | 4324 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
poisonous | +0.127 |
ocol | +0.124 |
ocket | +0.116 |
aster | +0.114 |
bamboo | +0.109 |
Neg logit contributions | |
youngest | -0.081 |
oldest | -0.075 |
So | -0.073 |
eldest | -0.073 |
bars | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
nodded | +0.009 |
youngest | +0.009 |
acts | +0.009 |
Let | +0.009 |
ver | +0.009 |
Neg logit contributions | |
poisonous | -0.009 |
ocol | -0.008 |
ocket | -0.008 |
aster | -0.008 |
arter | -0.008 |
Token | |
Token ID | 220 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
poisonous | +0.010 |
aster | +0.009 |
ocket | +0.009 |
heaven | +0.009 |
ocol | +0.009 |
Neg logit contributions | |
acts | -0.011 |
ver | -0.010 |
youngest | -0.010 |
folk | -0.010 |
mun | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
ver | +0.110 |
nodded | +0.107 |
Let | +0.106 |
youngest | +0.106 |
acts | +0.106 |
Neg logit contributions | |
poisonous | -0.184 |
ocol | -0.170 |
ocket | -0.169 |
aster | -0.165 |
arter | -0.163 |
Token |
|
Token ID | 198 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
acts | +0.019 |
youngest | +0.019 |
nodded | +0.019 |
ver | +0.019 |
folk | +0.019 |
Neg logit contributions | |
poisonous | -0.029 |
ocket | -0.026 |
ocol | -0.026 |
aster | -0.026 |
arter | -0.025 |
Token | Em |
Token ID | 10161 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
nodded | +0.022 |
ver | +0.022 |
youngest | +0.022 |
acts | +0.022 |
Let | +0.022 |
Neg logit contributions | |
poisonous | -0.028 |
ocol | -0.026 |
ocket | -0.026 |
aster | -0.025 |
arter | -0.025 |
Token | runners |
Token ID | 19323 |
Feature activation | +6.43e+00 |
Pos logit contributions | |
nodded | +0.316 |
ver | +0.312 |
youngest | +0.311 |
eldest | +0.307 |
oldest | +0.307 |
Neg logit contributions | |
poisonous | -0.175 |
aster | -0.167 |
ocol | -0.160 |
bamboo | -0.147 |
ocket | -0.145 |
Token | in |
Token ID | 287 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
youngest | +0.305 |
acts | +0.304 |
oldest | +0.300 |
nodded | +0.298 |
Let | +0.295 |
Neg logit contributions | |
poisonous | -0.207 |
ocket | -0.188 |
aster | -0.187 |
ocol | -0.184 |
bamboo | -0.167 |
Token | the |
Token ID | 262 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
acts | +0.076 |
youngest | +0.075 |
ver | +0.075 |
Let | +0.074 |
folk | +0.074 |
Neg logit contributions | |
poisonous | -0.098 |
ocket | -0.090 |
ocol | -0.088 |
arter | -0.086 |
aster | -0.083 |
Token | world |
Token ID | 995 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
ver | +0.055 |
nodded | +0.050 |
acts | +0.049 |
bars | +0.048 |
Let | +0.048 |
Neg logit contributions | |
poisonous | -0.055 |
aster | -0.051 |
bamboo | -0.047 |
ocol | -0.047 |
brand | -0.045 |
Token | ! |
Token ID | 0 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
youngest | +0.013 |
acts | +0.012 |
oldest | +0.012 |
nodded | +0.012 |
eldest | +0.012 |
Neg logit contributions | |
poisonous | -0.007 |
aster | -0.006 |
ocket | -0.006 |
ocol | -0.006 |
bamboo | -0.006 |
Token | Even |
Token ID | 3412 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
poisonous | +0.025 |
ocol | +0.024 |
ocket | +0.023 |
aster | +0.023 |
arter | +0.023 |
Neg logit contributions | |
acts | -0.016 |
nodded | -0.016 |
youngest | -0.016 |
ver | -0.015 |
Let | -0.015 |
Token | though |
Token ID | 996 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
acts | +0.058 |
ver | +0.058 |
youngest | +0.058 |
nodded | +0.058 |
Let | +0.057 |
Neg logit contributions | |
poisonous | -0.053 |
ocol | -0.047 |
aster | -0.045 |
bamboo | -0.044 |
ocket | -0.043 |
Token | he |
Token ID | 339 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
youngest | +0.059 |
acts | +0.059 |
nodded | +0.059 |
ver | +0.059 |
Let | +0.058 |
Neg logit contributions | |
poisonous | -0.068 |
ocol | -0.062 |
aster | -0.060 |
ocket | -0.059 |
bamboo | -0.059 |
Token | was |
Token ID | 373 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
bars | +0.058 |
youngest | +0.058 |
folk | +0.057 |
oldest | +0.056 |
eldest | +0.055 |
Neg logit contributions | |
poisonous | -0.099 |
ocket | -0.090 |
aster | -0.088 |
ocol | -0.087 |
ez | -0.084 |
Token | nervous |
Token ID | 10927 |
Feature activation | +3.69e+00 |
Pos logit contributions | |
acts | +0.016 |
bars | +0.015 |
Let | +0.015 |
nodded | +0.015 |
mun | +0.015 |
Neg logit contributions | |
poisonous | -0.014 |
aster | -0.012 |
ocol | -0.012 |
ocket | -0.011 |
bamboo | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
Let | +0.160 |
youngest | +0.157 |
nic | +0.155 |
nodded | +0.154 |
ver | +0.154 |
Neg logit contributions | |
poisonous | -0.142 |
ocol | -0.127 |
arter | -0.123 |
aster | -0.116 |
ocket | -0.116 |
Token | He |
Token ID | 679 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
acts | +0.065 |
youngest | +0.064 |
ver | +0.064 |
nodded | +0.062 |
oldest | +0.061 |
Neg logit contributions | |
poisonous | -0.095 |
ocol | -0.089 |
aster | -0.089 |
ocket | -0.088 |
arter | -0.086 |
Token | was |
Token ID | 373 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
poisonous | +0.011 |
ocket | +0.010 |
aster | +0.010 |
ocol | +0.010 |
bamboo | +0.010 |
Neg logit contributions | |
ver | -0.005 |
youngest | -0.005 |
folk | -0.005 |
bars | -0.005 |
oldest | -0.005 |
Token | wearing |
Token ID | 5762 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
youngest | +0.046 |
nodded | +0.045 |
oldest | +0.045 |
ver | +0.043 |
Let | +0.043 |
Neg logit contributions | |
poisonous | -0.056 |
ocol | -0.054 |
ocket | -0.053 |
arter | -0.052 |
aster | -0.051 |
Token | a |
Token ID | 257 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
poisonous | +0.119 |
arter | +0.114 |
ocol | +0.112 |
ocket | +0.111 |
aster | +0.107 |
Neg logit contributions | |
nodded | -0.102 |
youngest | -0.100 |
Let | -0.098 |
nic | -0.097 |
led | -0.097 |
Token | warm |
Token ID | 5814 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
nic | +0.006 |
nodded | +0.006 |
best | +0.005 |
way | +0.005 |
youngest | +0.005 |
Neg logit contributions | |
ocol | -0.010 |
isk | -0.009 |
poisonous | -0.009 |
arter | -0.009 |
ocket | -0.009 |
Token | coat |
Token ID | 13209 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
poisonous | +0.040 |
ocol | +0.038 |
ocket | +0.037 |
arter | +0.037 |
bones | +0.035 |
Neg logit contributions | |
nodded | -0.031 |
way | -0.029 |
led | -0.028 |
youngest | -0.027 |
nods | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
poisonous | +0.134 |
ocol | +0.125 |
ocket | +0.120 |
aster | +0.118 |
arter | +0.118 |
Neg logit contributions | |
ver | -0.120 |
nodded | -0.119 |
youngest | -0.118 |
bars | -0.117 |
Let | -0.115 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
poisonous | +0.016 |
ocol | +0.015 |
ocket | +0.014 |
arter | +0.014 |
aster | +0.014 |
Neg logit contributions | |
nodded | -0.013 |
led | -0.013 |
Let | -0.013 |
youngest | -0.013 |
acts | -0.013 |
Token | very |
Token ID | 845 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
nodded | +0.061 |
youngest | +0.060 |
nic | +0.058 |
oldest | +0.057 |
Let | +0.056 |
Neg logit contributions | |
poisonous | -0.086 |
ocket | -0.084 |
aster | -0.084 |
arter | -0.082 |
ocol | -0.081 |
Token | wise |
Token ID | 10787 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
poisonous | +0.151 |
ocket | +0.138 |
ocol | +0.138 |
aster | +0.135 |
arter | +0.131 |
Neg logit contributions | |
nodded | -0.148 |
youngest | -0.147 |
acts | -0.146 |
ver | -0.146 |
Let | -0.143 |
Token | . |
Token ID | 13 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
ver | +0.187 |
Let | +0.184 |
youngest | +0.183 |
nodded | +0.183 |
mun | +0.182 |
Neg logit contributions | |
poisonous | -0.125 |
ocket | -0.110 |
ocol | -0.106 |
bamboo | -0.101 |
arter | -0.098 |
Token | ran |
Token ID | 4966 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
poisonous | +0.003 |
ocket | +0.003 |
aster | +0.003 |
bamboo | +0.003 |
ocol | +0.003 |
Neg logit contributions | |
youngest | -0.002 |
ver | -0.002 |
acts | -0.002 |
folk | -0.002 |
mun | -0.002 |
Token | away |
Token ID | 1497 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
nodded | +0.075 |
Let | +0.073 |
So | +0.073 |
youngest | +0.073 |
led | +0.072 |
Neg logit contributions | |
poisonous | -0.123 |
ocol | -0.123 |
arter | -0.117 |
ocket | -0.113 |
aster | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
poisonous | +0.108 |
ocol | +0.100 |
ocket | +0.099 |
aster | +0.097 |
arter | +0.094 |
Neg logit contributions | |
nodded | -0.110 |
youngest | -0.109 |
ver | -0.109 |
acts | -0.108 |
led | -0.107 |
Token | |
Token ID | 220 |
Feature activation | +3.70e+00 |
Pos logit contributions | |
poisonous | +0.022 |
aster | +0.021 |
ocket | +0.020 |
arter | +0.020 |
ocol | +0.020 |
Neg logit contributions | |
acts | -0.020 |
mun | -0.020 |
youngest | -0.020 |
ver | -0.019 |
folk | -0.019 |
Token | He |
Token ID | 679 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
acts | +0.136 |
youngest | +0.134 |
nodded | +0.133 |
oldest | +0.130 |
folk | +0.130 |
Neg logit contributions | |
poisonous | -0.166 |
ocol | -0.152 |
ocket | -0.151 |
aster | -0.151 |
arter | -0.147 |
Token | had |
Token ID | 550 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
poisonous | +0.055 |
ocket | +0.051 |
bamboo | +0.048 |
aster | +0.047 |
injuries | +0.045 |
Neg logit contributions | |
folk | -0.031 |
ver | -0.030 |
bars | -0.030 |
youngest | -0.030 |
acts | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
poisonous | +0.069 |
ocket | +0.059 |
aster | +0.057 |
bamboo | +0.055 |
ocol | +0.055 |
Neg logit contributions | |
acts | -0.071 |
ver | -0.071 |
folk | -0.070 |
bars | -0.070 |
mun | -0.069 |
Token | big |
Token ID | 1263 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
poisonous | +0.014 |
ocol | +0.014 |
ocket | +0.013 |
arter | +0.013 |
aster | +0.013 |
Neg logit contributions | |
nodded | -0.014 |
youngest | -0.013 |
ver | -0.013 |
acts | -0.013 |
Let | -0.013 |
Token | smile |
Token ID | 8212 |
Feature activation | -6.06e-01 |
Pos logit contributions | |
poisonous | +0.146 |
ocol | +0.136 |
ocket | +0.135 |
aster | +0.130 |
arter | +0.129 |
Neg logit contributions | |
nodded | -0.106 |
acts | -0.106 |
youngest | -0.103 |
ver | -0.101 |
Let | -0.100 |
Token | on |
Token ID | 319 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
poisonous | +0.027 |
ocol | +0.026 |
ocket | +0.024 |
arter | +0.024 |
aster | +0.023 |
Neg logit contributions | |
nodded | -0.024 |
ver | -0.023 |
led | -0.023 |
youngest | -0.023 |
Let | -0.023 |
Token | his |
Token ID | 465 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
poisonous | +0.054 |
prise | +0.051 |
ocket | +0.051 |
ocol | +0.048 |
arter | +0.048 |
Neg logit contributions | |
nodded | -0.033 |
So | -0.031 |
acts | -0.031 |
youngest | -0.030 |
Let | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
poisonous | +0.032 |
ocol | +0.030 |
aster | +0.030 |
what | +0.030 |
bamboo | +0.029 |
Neg logit contributions | |
folk | -0.052 |
youngest | -0.052 |
bars | -0.052 |
nic | -0.052 |
oldest | -0.052 |
Token | they |
Token ID | 484 |
Feature activation | +4.43e+00 |
Pos logit contributions | |
poisonous | +0.003 |
aster | +0.003 |
bamboo | +0.003 |
ocket | +0.003 |
ocol | +0.003 |
Neg logit contributions | |
acts | -0.002 |
folk | -0.002 |
bars | -0.001 |
basics | -0.001 |
youngest | -0.001 |
Token | went |
Token ID | 1816 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
bars | +0.201 |
acts | +0.195 |
hardest | +0.192 |
basics | +0.191 |
sets | +0.185 |
Neg logit contributions | |
poisonous | -0.169 |
ocket | -0.152 |
aster | -0.144 |
sm | -0.135 |
what | -0.131 |
Token | to |
Token ID | 284 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
led | +0.001 |
Let | +0.001 |
nodded | +0.001 |
ver | +0.001 |
youngest | +0.001 |
Neg logit contributions | |
ocol | -0.001 |
poisonous | -0.001 |
ocket | -0.001 |
aster | -0.001 |
arter | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
poisonous | +0.010 |
ocket | +0.009 |
ocol | +0.009 |
arter | +0.009 |
aster | +0.009 |
Neg logit contributions | |
youngest | -0.008 |
Let | -0.008 |
acts | -0.008 |
led | -0.008 |
nodded | -0.008 |
Token | park |
Token ID | 3952 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
youngest | +0.044 |
Let | +0.043 |
oldest | +0.043 |
led | +0.042 |
nodded | +0.041 |
Neg logit contributions | |
poisonous | -0.069 |
ocol | -0.068 |
ocket | -0.066 |
arter | -0.064 |
aster | -0.063 |
Token | with |
Token ID | 351 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
poisonous | +0.037 |
aster | +0.033 |
ocket | +0.031 |
bamboo | +0.031 |
heaven | +0.029 |
Neg logit contributions | |
folk | -0.062 |
acts | -0.062 |
bars | -0.061 |
youngest | -0.061 |
ver | -0.060 |
Token | their |
Token ID | 511 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
poisonous | +0.008 |
ocol | +0.008 |
ocket | +0.008 |
aster | +0.007 |
arter | +0.007 |
Neg logit contributions | |
youngest | -0.006 |
nodded | -0.006 |
acts | -0.006 |
Let | -0.006 |
ver | -0.006 |
Token | k |
Token ID | 479 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
poisonous | +0.026 |
bamboo | +0.024 |
injuries | +0.023 |
ocol | +0.021 |
ocket | +0.021 |
Neg logit contributions | |
ver | -0.024 |
acts | -0.022 |
folk | -0.022 |
mun | -0.022 |
bars | -0.022 |
Token | ites |
Token ID | 2737 |
Feature activation | +5.96e+00 |
Pos logit contributions | |
youngest | +0.011 |
nodded | +0.011 |
acts | +0.011 |
ver | +0.011 |
Let | +0.010 |
Neg logit contributions | |
poisonous | -0.023 |
ocol | -0.022 |
ocket | -0.022 |
aster | -0.021 |
arter | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
acts | +0.324 |
ver | +0.316 |
Let | +0.313 |
folk | +0.313 |
youngest | +0.313 |
Neg logit contributions | |
poisonous | -0.150 |
bamboo | -0.128 |
ocket | -0.127 |
aster | -0.113 |
arter | -0.112 |
Token | cast |
Token ID | 3350 |
Feature activation | +5.92e-02 |
Pos logit contributions | |
bars | +0.020 |
ver | +0.019 |
folk | +0.019 |
mun | +0.019 |
sets | +0.018 |
Neg logit contributions | |
poisonous | -0.032 |
ocket | -0.031 |
aster | -0.028 |
bamboo | -0.028 |
sm | -0.027 |
Token | her |
Token ID | 607 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
ver | +0.002 |
nodded | +0.002 |
youngest | +0.002 |
acts | +0.002 |
Let | +0.002 |
Neg logit contributions | |
poisonous | -0.003 |
ocol | -0.003 |
ocket | -0.003 |
aster | -0.003 |
arter | -0.003 |
Token | eyes |
Token ID | 2951 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
poisonous | +0.118 |
ocol | +0.108 |
ocket | +0.108 |
aster | +0.106 |
arter | +0.104 |
Neg logit contributions | |
acts | -0.073 |
youngest | -0.073 |
nodded | -0.073 |
ver | -0.073 |
Let | -0.071 |
Token | on |
Token ID | 319 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
acts | +0.017 |
youngest | +0.016 |
ver | +0.016 |
nodded | +0.016 |
folk | +0.016 |
Neg logit contributions | |
poisonous | -0.018 |
ocket | -0.016 |
ocol | -0.016 |
aster | -0.015 |
arter | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
poisonous | +0.053 |
ocol | +0.051 |
arter | +0.050 |
ocket | +0.049 |
prise | +0.048 |
Neg logit contributions | |
led | -0.038 |
Let | -0.037 |
youngest | -0.037 |
acts | -0.037 |
nodded | -0.036 |
Token | wreck |
Token ID | 15228 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
youngest | +0.053 |
acts | +0.052 |
nodded | +0.052 |
ver | +0.051 |
Let | +0.051 |
Neg logit contributions | |
poisonous | -0.066 |
ocol | -0.061 |
ocket | -0.060 |
aster | -0.059 |
arter | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
poisonous | +0.090 |
ocket | +0.080 |
aster | +0.079 |
ocol | +0.078 |
arter | +0.076 |
Neg logit contributions | |
youngest | -0.093 |
acts | -0.093 |
Let | -0.092 |
ver | -0.092 |
nodded | -0.091 |
Token | wondered |
Token ID | 14028 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
poisonous | +0.003 |
ocket | +0.003 |
aster | +0.003 |
ocol | +0.003 |
bamboo | +0.003 |
Neg logit contributions | |
acts | -0.002 |
ver | -0.002 |
youngest | -0.002 |
folk | -0.002 |
bars | -0.002 |
Token | what |
Token ID | 644 |
Feature activation | -7.14e+00 |
Pos logit contributions | |
ver | +0.176 |
nodded | +0.173 |
acts | +0.170 |
youngest | +0.170 |
led | +0.170 |
Neg logit contributions | |
poisonous | -0.062 |
ocol | -0.053 |
aster | -0.049 |
ocket | -0.048 |
arter | -0.047 |
Token | it |
Token ID | 340 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
poisonous | +0.240 |
ocket | +0.203 |
aster | +0.197 |
what | +0.191 |
sm | +0.190 |
Neg logit contributions | |
youngest | -0.358 |
eldest | -0.351 |
ver | -0.347 |
folk | -0.346 |
bars | -0.346 |
Token | could |
Token ID | 714 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
poisonous | +0.025 |
ocket | +0.023 |
what | +0.022 |
bamboo | +0.021 |
sm | +0.021 |
Neg logit contributions | |
folk | -0.017 |
youngest | -0.017 |
bars | -0.017 |
eldest | -0.016 |
mun | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
poisonous | +0.107 |
ocol | +0.097 |
ocket | +0.096 |
aster | +0.094 |
arter | +0.092 |
Neg logit contributions | |
acts | -0.114 |
ver | -0.114 |
nodded | -0.114 |
youngest | -0.114 |
Let | -0.112 |
Token | dry |
Token ID | 5894 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
poisonous | +0.009 |
aster | +0.008 |
injuries | +0.007 |
bamboo | +0.007 |
ocket | +0.007 |
Neg logit contributions | |
ver | -0.011 |
folk | -0.011 |
youngest | -0.011 |
acts | -0.011 |
bars | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
poisonous | +0.057 |
ocket | +0.054 |
ocol | +0.052 |
aster | +0.051 |
arter | +0.050 |
Neg logit contributions | |
acts | -0.055 |
nodded | -0.055 |
youngest | -0.055 |
Let | -0.055 |
bars | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
poisonous | +0.015 |
bamboo | +0.013 |
brand | +0.012 |
what | +0.012 |
arter | +0.012 |
Neg logit contributions | |
bars | -0.023 |
folk | -0.022 |
acts | -0.021 |
youngest | -0.021 |
ver | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
ver | +0.009 |
Let | +0.009 |
nodded | +0.009 |
led | +0.009 |
youngest | +0.009 |
Neg logit contributions | |
poisonous | -0.017 |
ocol | -0.016 |
ocket | -0.016 |
arter | -0.015 |
aster | -0.015 |
Token | After |
Token ID | 3260 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
acts | +0.015 |
ver | +0.015 |
youngest | +0.015 |
nodded | +0.015 |
bars | +0.015 |
Neg logit contributions | |
poisonous | -0.013 |
ocol | -0.011 |
ocket | -0.011 |
aster | -0.011 |
arter | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ocket | +0.172 |
poisonous | +0.164 |
ocol | +0.163 |
arter | +0.159 |
eds | +0.152 |
Neg logit contributions | |
led | -0.104 |
Let | -0.098 |
neither | -0.093 |
nodded | -0.092 |
they | -0.090 |
Token | rain |
Token ID | 6290 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ver | +0.068 |
bars | +0.066 |
folk | +0.064 |
acts | +0.062 |
nodded | +0.061 |
Neg logit contributions | |
poisonous | -0.039 |
bamboo | -0.032 |
injuries | -0.030 |
arter | -0.028 |
aster | -0.026 |
Token | stopped |
Token ID | 5025 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
bars | +0.050 |
acts | +0.048 |
nic | +0.047 |
youngest | +0.047 |
oldest | +0.046 |
Neg logit contributions | |
poisonous | -0.042 |
bamboo | -0.034 |
isk | -0.034 |
injuries | -0.033 |
aster | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
poisonous | +0.025 |
ocol | +0.022 |
aster | +0.021 |
arter | +0.021 |
ocket | +0.021 |
Neg logit contributions | |
youngest | -0.023 |
ver | -0.023 |
acts | -0.023 |
nodded | -0.023 |
folk | -0.023 |
Token | Benny |
Token ID | 44275 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
bars | +0.008 |
acts | +0.008 |
folk | +0.008 |
ver | +0.007 |
nodded | +0.007 |
Neg logit contributions | |
poisonous | -0.005 |
aster | -0.004 |
bamboo | -0.004 |
arter | -0.003 |
ocol | -0.003 |
Token | Comfort |
Token ID | 45769 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
acts | +0.005 |
youngest | +0.005 |
oldest | +0.005 |
mun | +0.005 |
ver | +0.005 |
Neg logit contributions | |
poisonous | -0.005 |
aster | -0.005 |
ocket | -0.005 |
ocol | -0.005 |
arter | -0.005 |
Token | ably |
Token ID | 1346 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
acts | +0.013 |
youngest | +0.013 |
ver | +0.013 |
nodded | +0.013 |
Let | +0.013 |
Neg logit contributions | |
poisonous | -0.024 |
ocket | -0.022 |
ocol | -0.022 |
aster | -0.022 |
arter | -0.021 |
Token | eating |
Token ID | 6600 |
Feature activation | -5.26e+00 |
Pos logit contributions | |
acts | +0.018 |
youngest | +0.018 |
ver | +0.018 |
bars | +0.018 |
Let | +0.018 |
Neg logit contributions | |
poisonous | -0.031 |
ocket | -0.028 |
ocol | -0.028 |
aster | -0.028 |
arter | -0.027 |
Token | his |
Token ID | 465 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
poisonous | +0.243 |
arter | +0.238 |
ocket | +0.230 |
ocol | +0.224 |
eds | +0.215 |
Neg logit contributions | |
nodded | -0.191 |
led | -0.184 |
acts | -0.180 |
Let | -0.179 |
ver | -0.178 |
Token | l |
Token ID | 300 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
arter | +0.022 |
poisonous | +0.021 |
ocol | +0.020 |
aster | +0.019 |
ores | +0.019 |
Neg logit contributions | |
nic | -0.021 |
acts | -0.020 |
youngest | -0.020 |
nodded | -0.020 |
oldest | -0.020 |
Token | oll |
Token ID | 692 |
Feature activation | -8.20e+00 |
Pos logit contributions | |
poisonous | +0.219 |
arter | +0.204 |
prank | +0.187 |
aster | +0.186 |
heaven | +0.186 |
Neg logit contributions | |
ver | -0.262 |
bars | -0.241 |
nodded | -0.240 |
acts | -0.212 |
nic | -0.211 |
Token | ipop |
Token ID | 42800 |
Feature activation | -4.85e+00 |
Pos logit contributions | |
poisonous | +0.357 |
arter | +0.342 |
heaven | +0.329 |
aster | +0.324 |
ocol | +0.306 |
Neg logit contributions | |
ver | -0.333 |
nodded | -0.322 |
bars | -0.313 |
acts | -0.298 |
led | -0.284 |
Token | , |
Token ID | 11 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
poisonous | +0.225 |
ocol | +0.216 |
arter | +0.206 |
ocket | +0.204 |
aster | +0.202 |
Neg logit contributions | |
nodded | -0.177 |
ver | -0.176 |
youngest | -0.170 |
led | -0.170 |
acts | -0.167 |
Token | John |
Token ID | 1757 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
nodded | +0.016 |
Let | +0.015 |
led | +0.015 |
ver | +0.015 |
youngest | +0.015 |
Neg logit contributions | |
poisonous | -0.019 |
ocol | -0.018 |
ocket | -0.018 |
arter | -0.018 |
aster | -0.017 |
Token | happily |
Token ID | 18177 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
acts | +0.043 |
youngest | +0.043 |
ver | +0.043 |
nodded | +0.043 |
Let | +0.042 |
Neg logit contributions | |
poisonous | -0.088 |
ocket | -0.082 |
ocol | -0.081 |
aster | -0.080 |
arter | -0.078 |
Token | went |
Token ID | 1816 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
nodded | +0.034 |
ver | +0.034 |
acts | +0.034 |
youngest | +0.033 |
Let | +0.033 |
Neg logit contributions | |
poisonous | -0.063 |
ocol | -0.059 |
ocket | -0.058 |
aster | -0.057 |
arter | -0.057 |
Token | hole |
Token ID | 7604 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
Let | +0.051 |
youngest | +0.051 |
acts | +0.050 |
led | +0.050 |
nodded | +0.049 |
Neg logit contributions | |
poisonous | -0.061 |
ocol | -0.060 |
ocket | -0.059 |
aster | -0.058 |
arter | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
poisonous | +0.029 |
ocket | +0.025 |
ocol | +0.025 |
aster | +0.024 |
arter | +0.024 |
Neg logit contributions | |
acts | -0.037 |
youngest | -0.037 |
nodded | -0.037 |
ver | -0.037 |
Let | -0.036 |
Token | But |
Token ID | 887 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
poisonous | +0.006 |
bamboo | +0.005 |
aster | +0.005 |
ocol | +0.005 |
ocket | +0.005 |
Neg logit contributions | |
acts | -0.005 |
youngest | -0.004 |
ver | -0.004 |
oldest | -0.004 |
nodded | -0.004 |
Token | Spot |
Token ID | 15899 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
acts | +0.114 |
youngest | +0.113 |
ver | +0.112 |
nodded | +0.112 |
Let | +0.110 |
Neg logit contributions | |
poisonous | -0.121 |
ocket | -0.107 |
ocol | -0.107 |
aster | -0.106 |
arter | -0.102 |
Token | remembered |
Token ID | 12086 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
poisonous | +0.122 |
ocket | +0.107 |
aster | +0.103 |
bamboo | +0.102 |
injuries | +0.101 |
Neg logit contributions | |
bars | -0.072 |
folk | -0.069 |
acts | -0.067 |
youngest | -0.067 |
oldest | -0.065 |
Token | what |
Token ID | 644 |
Feature activation | -7.24e+00 |
Pos logit contributions | |
poisonous | +0.019 |
ocol | +0.018 |
ocket | +0.017 |
arter | +0.017 |
aster | +0.017 |
Neg logit contributions | |
nodded | -0.027 |
acts | -0.027 |
ver | -0.027 |
youngest | -0.027 |
led | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
poisonous | +0.237 |
ocket | +0.190 |
aster | +0.189 |
ocol | +0.180 |
bamboo | +0.178 |
Neg logit contributions | |
folk | -0.374 |
youngest | -0.371 |
bars | -0.370 |
acts | -0.361 |
eldest | -0.361 |
Token | family |
Token ID | 1641 |
Feature activation | +9.85e-01 |
Pos logit contributions | |
ver | +0.066 |
folk | +0.063 |
bars | +0.062 |
mun | +0.062 |
nodded | +0.060 |
Neg logit contributions | |
poisonous | -0.057 |
ocket | -0.048 |
bamboo | -0.046 |
injuries | -0.046 |
arter | -0.044 |
Token | said |
Token ID | 531 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
youngest | +0.034 |
eldest | +0.033 |
Und | +0.033 |
oldest | +0.032 |
bars | +0.032 |
Neg logit contributions | |
poisonous | -0.047 |
ocket | -0.044 |
aster | -0.041 |
ocol | -0.040 |
what | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
youngest | +0.046 |
oldest | +0.045 |
acts | +0.045 |
ver | +0.045 |
Let | +0.044 |
Neg logit contributions | |
poisonous | -0.040 |
ocket | -0.036 |
ocol | -0.035 |
aster | -0.034 |
arter | -0.034 |
Token | stayed |
Token ID | 9658 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
poisonous | +0.001 |
ocket | +0.001 |
aster | +0.001 |
what | +0.001 |
bamboo | +0.001 |
Neg logit contributions | |
bars | -0.000 |
ver | -0.000 |
youngest | -0.000 |
folk | -0.000 |
oldest | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
acts | +0.041 |
nodded | +0.041 |
So | +0.041 |
ver | +0.040 |
means | +0.040 |
Neg logit contributions | |
poisonous | -0.062 |
ocket | -0.061 |
ocol | -0.061 |
aster | -0.058 |
arter | -0.057 |
Token | She |
Token ID | 1375 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
poisonous | +0.014 |
ocket | +0.013 |
ocol | +0.013 |
aster | +0.013 |
arter | +0.012 |
Neg logit contributions | |
ver | -0.012 |
mun | -0.011 |
acts | -0.011 |
folk | -0.011 |
youngest | -0.011 |
Token | wanted |
Token ID | 2227 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
poisonous | +0.181 |
ocket | +0.167 |
ocol | +0.165 |
aster | +0.164 |
arter | +0.160 |
Neg logit contributions | |
ver | -0.064 |
folk | -0.063 |
acts | -0.063 |
youngest | -0.063 |
bars | -0.061 |
Token | to |
Token ID | 284 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
Let | +0.018 |
led | +0.018 |
acts | +0.018 |
youngest | +0.017 |
oldest | +0.017 |
Neg logit contributions | |
poisonous | -0.027 |
ocol | -0.027 |
ocket | -0.026 |
arter | -0.026 |
aster | -0.024 |
Token | know |
Token ID | 760 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
poisonous | +0.010 |
ocol | +0.010 |
ocket | +0.009 |
arter | +0.009 |
bamboo | +0.009 |
Neg logit contributions | |
youngest | -0.007 |
nodded | -0.007 |
acts | -0.007 |
oldest | -0.007 |
ver | -0.007 |
Token | what |
Token ID | 644 |
Feature activation | -7.25e+00 |
Pos logit contributions | |
poisonous | +0.090 |
ocol | +0.078 |
ocket | +0.077 |
arter | +0.074 |
aster | +0.073 |
Neg logit contributions | |
ver | -0.191 |
nodded | -0.190 |
acts | -0.189 |
led | -0.187 |
youngest | -0.187 |
Token | was |
Token ID | 373 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
poisonous | +0.335 |
aster | +0.299 |
ocket | +0.293 |
sm | +0.276 |
bamboo | +0.276 |
Neg logit contributions | |
folk | -0.277 |
youngest | -0.276 |
bars | -0.268 |
eldest | -0.265 |
ver | -0.259 |
Token | inside |
Token ID | 2641 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
ver | +0.010 |
youngest | +0.010 |
led | +0.010 |
acts | +0.010 |
nodded | +0.010 |
Neg logit contributions | |
poisonous | -0.009 |
ocol | -0.009 |
aster | -0.009 |
ocket | -0.009 |
arter | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
poisonous | +0.096 |
ocket | +0.089 |
ocol | +0.088 |
arter | +0.086 |
aster | +0.085 |
Neg logit contributions | |
acts | -0.067 |
ver | -0.066 |
led | -0.066 |
bars | -0.065 |
nodded | -0.064 |
Token | brain |
Token ID | 3632 |
Feature activation | +8.53e-01 |
Pos logit contributions | |
nodded | +0.061 |
ver | +0.060 |
folk | +0.057 |
mun | +0.057 |
acts | +0.057 |
Neg logit contributions | |
poisonous | -0.063 |
ocket | -0.057 |
ocol | -0.055 |
aster | -0.055 |
arter | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
youngest | +0.035 |
nodded | +0.034 |
acts | +0.034 |
Let | +0.034 |
ver | +0.034 |
Neg logit contributions | |
poisonous | -0.035 |
ocket | -0.032 |
ocol | -0.031 |
aster | -0.031 |
arter | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
youngest | +0.195 |
ver | +0.190 |
nodded | +0.189 |
nic | +0.187 |
oldest | +0.187 |
Neg logit contributions | |
poisonous | -0.355 |
ocket | -0.355 |
ocol | -0.346 |
arter | -0.339 |
aster | -0.332 |
Token | good |
Token ID | 922 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
ocol | +0.016 |
ocket | +0.016 |
poisonous | +0.015 |
arter | +0.015 |
aster | +0.015 |
Neg logit contributions | |
youngest | -0.013 |
Let | -0.013 |
oldest | -0.013 |
basics | -0.013 |
nodded | -0.013 |
Token | thing |
Token ID | 1517 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
poisonous | +0.091 |
ocol | +0.083 |
ocket | +0.083 |
aster | +0.082 |
arter | +0.081 |
Neg logit contributions | |
ver | -0.049 |
folk | -0.048 |
youngest | -0.047 |
nodded | -0.047 |
bars | -0.047 |
Token | you |
Token ID | 345 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
poisonous | +0.044 |
ocket | +0.039 |
ocol | +0.038 |
aster | +0.037 |
arter | +0.037 |
Neg logit contributions | |
youngest | -0.050 |
acts | -0.049 |
nodded | -0.049 |
ver | -0.049 |
Let | -0.049 |
Token | know |
Token ID | 760 |
Feature activation | -3.39e+00 |
Pos logit contributions | |
ver | +0.065 |
acts | +0.064 |
youngest | +0.064 |
nodded | +0.064 |
Let | +0.063 |
Neg logit contributions | |
poisonous | -0.106 |
ocol | -0.098 |
ocket | -0.097 |
aster | -0.095 |
arter | -0.094 |
Token | what |
Token ID | 644 |
Feature activation | -7.22e+00 |
Pos logit contributions | |
poisonous | +0.130 |
ocket | +0.117 |
ocol | +0.117 |
aster | +0.113 |
arter | +0.113 |
Neg logit contributions | |
ver | -0.148 |
nodded | -0.147 |
acts | -0.147 |
youngest | -0.145 |
Let | -0.144 |
Token | you |
Token ID | 345 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
poisonous | +0.286 |
aster | +0.258 |
ocket | +0.249 |
ocol | +0.241 |
heaven | +0.229 |
Neg logit contributions | |
bars | -0.317 |
youngest | -0.317 |
folk | -0.315 |
ver | -0.308 |
eldest | -0.306 |
Token | want |
Token ID | 765 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
bars | +0.076 |
folk | +0.074 |
youngest | +0.073 |
ver | +0.073 |
acts | +0.072 |
Neg logit contributions | |
poisonous | -0.096 |
ocket | -0.089 |
ocol | -0.087 |
aster | -0.086 |
arter | -0.084 |
Token | ! |
Token ID | 0 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
poisonous | +0.076 |
ocol | +0.069 |
ocket | +0.068 |
aster | +0.067 |
arter | +0.066 |
Neg logit contributions | |
acts | -0.072 |
ver | -0.072 |
nodded | -0.072 |
youngest | -0.072 |
Let | -0.071 |
Token | Let |
Token ID | 3914 |
Feature activation | +4.51e+00 |
Pos logit contributions | |
ocol | +0.027 |
aster | +0.027 |
poisonous | +0.026 |
arter | +0.025 |
ocket | +0.025 |
Neg logit contributions | |
youngest | -0.035 |
led | -0.035 |
mun | -0.035 |
folk | -0.035 |
nodded | -0.034 |
Token | 's |
Token ID | 338 |
Feature activation | +6.88e+00 |
Pos logit contributions | |
bars | +0.142 |
ver | +0.137 |
folk | +0.133 |
spe | +0.119 |
acts | +0.115 |
Neg logit contributions | |
poisonous | -0.248 |
aster | -0.230 |
what | -0.216 |
heaven | -0.211 |
jam | -0.210 |
Token | Jack |
Token ID | 3619 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
bars | +0.009 |
acts | +0.008 |
folk | +0.008 |
mun | +0.008 |
oldest | +0.008 |
Neg logit contributions | |
poisonous | -0.007 |
what | -0.007 |
aster | -0.006 |
ocket | -0.006 |
brand | -0.006 |
Token | said |
Token ID | 531 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
bars | +0.084 |
ver | +0.082 |
folk | +0.081 |
hardest | +0.081 |
youngest | +0.079 |
Neg logit contributions | |
poisonous | -0.066 |
ocket | -0.057 |
aster | -0.055 |
arter | -0.051 |
heaven | -0.051 |
Token | to |
Token ID | 284 |
Feature activation | -5.75e-02 |
Pos logit contributions | |
bars | +0.055 |
youngest | +0.054 |
acts | +0.054 |
oldest | +0.053 |
mun | +0.053 |
Neg logit contributions | |
poisonous | -0.041 |
ocket | -0.036 |
aster | -0.035 |
ocol | -0.034 |
arter | -0.033 |
Token | Jill |
Token ID | 22772 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
poisonous | +0.003 |
ocol | +0.003 |
ocket | +0.003 |
bamboo | +0.002 |
injuries | +0.002 |
Neg logit contributions | |
youngest | -0.002 |
Let | -0.002 |
nodded | -0.002 |
led | -0.002 |
oldest | -0.002 |
Token | " |
Token ID | 366 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
youngest | +0.017 |
Let | +0.017 |
acts | +0.016 |
nodded | +0.016 |
oldest | +0.016 |
Neg logit contributions | |
poisonous | -0.018 |
ocket | -0.016 |
ocol | -0.016 |
aster | -0.016 |
arter | -0.016 |
Token | What |
Token ID | 2061 |
Feature activation | -7.56e+00 |
Pos logit contributions | |
acts | +0.056 |
ver | +0.055 |
nodded | +0.053 |
youngest | +0.053 |
bars | +0.052 |
Neg logit contributions | |
poisonous | -0.155 |
ocket | -0.145 |
aster | -0.145 |
ocol | -0.144 |
arter | -0.139 |
Token | shall |
Token ID | 2236 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
what | +0.271 |
ocket | +0.264 |
poisonous | +0.261 |
aster | +0.248 |
sm | +0.237 |
Neg logit contributions | |
folk | -0.350 |
youngest | -0.332 |
bars | -0.330 |
eldest | -0.330 |
lest | -0.310 |
Token | we |
Token ID | 356 |
Feature activation | +5.84e+00 |
Pos logit contributions | |
ver | +0.064 |
mun | +0.063 |
nodded | +0.062 |
youngest | +0.062 |
acts | +0.060 |
Neg logit contributions | |
poisonous | -0.077 |
aster | -0.073 |
what | -0.072 |
ocol | -0.071 |
arter | -0.070 |
Token | do |
Token ID | 466 |
Feature activation | -4.56e+00 |
Pos logit contributions | |
mun | +0.273 |
bars | +0.267 |
nodded | +0.265 |
folk | +0.262 |
acts | +0.259 |
Neg logit contributions | |
poisonous | -0.195 |
aster | -0.191 |
ocket | -0.187 |
ocol | -0.183 |
arter | -0.179 |
Token | today |
Token ID | 1909 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
poisonous | +0.169 |
ocket | +0.152 |
ocol | +0.151 |
aster | +0.150 |
arter | +0.144 |
Neg logit contributions | |
nodded | -0.202 |
youngest | -0.201 |
acts | -0.200 |
bars | -0.198 |
folk | -0.197 |
Token | ?" |
Token ID | 1701 |
Feature activation | -8.99e-01 |
Pos logit contributions | |
poisonous | +0.021 |
aster | +0.020 |
ocket | +0.019 |
ocol | +0.019 |
arter | +0.019 |
Neg logit contributions | |
acts | -0.028 |
nodded | -0.028 |
ver | -0.028 |
youngest | -0.028 |
folk | -0.028 |