Token | support |
Token ID | 1104 |
Feature activation | +8.86e-02 |
Pos logit contributions | |
| +0.387 |
Catalan | +0.376 |
Zi | +0.375 |
Niet | +0.367 |
Moss | +0.357 |
Neg logit contributions | |
WORK | -0.319 |
liv | -0.245 |
WARE | -0.234 |
awatts | -0.233 |
aturdays | -0.225 |
Token | the |
Token ID | 262 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
WORK | +0.079 |
liv | +0.064 |
awatts | +0.063 |
WARE | +0.062 |
aturdays | +0.061 |
Neg logit contributions | |
| -0.059 |
Zi | -0.056 |
Catalan | -0.056 |
Niet | -0.055 |
Moss | -0.053 |
Token | mission |
Token ID | 4365 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
| +0.138 |
Zi | +0.134 |
Catalan | +0.134 |
Niet | +0.133 |
wikipedia | +0.128 |
Neg logit contributions | |
WORK | -0.091 |
liv | -0.067 |
awatts | -0.063 |
WARE | -0.063 |
aturdays | -0.059 |
Token | of |
Token ID | 286 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
| +0.049 |
Zi | +0.047 |
Catalan | +0.047 |
Niet | +0.046 |
wikipedia | +0.045 |
Neg logit contributions | |
WORK | -0.066 |
liv | -0.053 |
awatts | -0.051 |
WARE | -0.051 |
aturdays | -0.050 |
Token | New |
Token ID | 968 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
| +0.036 |
Catalan | +0.034 |
Zi | +0.034 |
Niet | +0.033 |
Moss | +0.032 |
Neg logit contributions | |
WORK | -0.055 |
awatts | -0.045 |
liv | -0.045 |
WARE | -0.044 |
aturdays | -0.043 |
Token | Advent |
Token ID | 33732 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
| +0.167 |
Catalan | +0.162 |
Zi | +0.162 |
Niet | +0.160 |
wikipedia | +0.153 |
Neg logit contributions | |
WORK | -0.176 |
liv | -0.141 |
awatts | -0.135 |
WARE | -0.135 |
aturdays | -0.130 |
Token | and |
Token ID | 290 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
WORK | +1.397 |
liv | +1.104 |
awatts | +1.047 |
WARE | +1.041 |
aturdays | +1.003 |
Neg logit contributions | |
| -1.511 |
Catalan | -1.460 |
Zi | -1.459 |
Niet | -1.428 |
Moss | -1.389 |
Token | get |
Token ID | 651 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
| +0.196 |
Catalan | +0.190 |
Zi | +0.190 |
Niet | +0.185 |
Moss | +0.181 |
Neg logit contributions | |
WORK | -0.167 |
liv | -0.124 |
awatts | -0.123 |
WARE | -0.121 |
aturdays | -0.118 |
Token | the |
Token ID | 262 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
| +0.008 |
Catalan | +0.008 |
Zi | +0.008 |
Niet | +0.008 |
Moss | +0.007 |
Neg logit contributions | |
WORK | -0.012 |
liv | -0.010 |
awatts | -0.010 |
WARE | -0.010 |
Appl | -0.009 |
Token | full |
Token ID | 1336 |
Feature activation | +7.42e-03 |
Pos logit contributions | |
| +0.111 |
Zi | +0.107 |
Catalan | +0.106 |
Niet | +0.106 |
wikipedia | +0.100 |
Neg logit contributions | |
WORK | -0.156 |
liv | -0.128 |
awatts | -0.124 |
WARE | -0.123 |
aturdays | -0.119 |
Token | contents |
Token ID | 10154 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
WORK | +0.005 |
liv | +0.004 |
WARE | +0.004 |
awatts | +0.004 |
Appl | +0.004 |
Neg logit contributions | |
| -0.006 |
Zi | -0.006 |
Niet | -0.006 |
Catalan | -0.006 |
wikipedia | -0.006 |
Token | other |
Token ID | 584 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
| +0.070 |
Zi | +0.067 |
Catalan | +0.067 |
Niet | +0.067 |
wikipedia | +0.064 |
Neg logit contributions | |
WORK | -0.079 |
liv | -0.064 |
awatts | -0.061 |
WARE | -0.061 |
Appl | -0.059 |
Token | characters |
Token ID | 3435 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
| +0.193 |
Zi | +0.187 |
Catalan | +0.186 |
Niet | +0.186 |
wikipedia | +0.177 |
Neg logit contributions | |
WORK | -0.212 |
liv | -0.170 |
awatts | -0.163 |
WARE | -0.163 |
aturdays | -0.157 |
Token | from |
Token ID | 422 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
| +0.210 |
Zi | +0.204 |
Catalan | +0.202 |
Niet | +0.201 |
wikipedia | +0.192 |
Neg logit contributions | |
WORK | -0.229 |
liv | -0.184 |
awatts | -0.176 |
WARE | -0.176 |
aturdays | -0.170 |
Token | non |
Token ID | 1729 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
| +0.103 |
Zi | +0.099 |
Catalan | +0.099 |
Niet | +0.098 |
wikipedia | +0.093 |
Neg logit contributions | |
WORK | -0.126 |
liv | -0.103 |
awatts | -0.099 |
WARE | -0.098 |
aturdays | -0.095 |
Token | - |
Token ID | 12 |
Feature activation | +8.20e-03 |
Pos logit contributions | |
WORK | +0.195 |
liv | +0.156 |
awatts | +0.149 |
WARE | +0.148 |
aturdays | +0.143 |
Neg logit contributions | |
| -0.179 |
Zi | -0.175 |
Catalan | -0.174 |
Niet | -0.174 |
wikipedia | -0.165 |
Token | Friend |
Token ID | 23331 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
WORK | +0.007 |
liv | +0.005 |
awatts | +0.005 |
WARE | +0.005 |
aturdays | +0.005 |
Neg logit contributions | |
| -0.006 |
Zi | -0.006 |
Catalan | -0.006 |
Niet | -0.006 |
wikipedia | -0.006 |
Token | ship |
Token ID | 6720 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
WORK | +1.383 |
liv | +1.099 |
awatts | +1.045 |
WARE | +1.043 |
aturdays | +1.011 |
Neg logit contributions | |
| -1.352 |
Zi | -1.329 |
Catalan | -1.312 |
Niet | -1.306 |
wikipedia | -1.252 |
Token | is |
Token ID | 318 |
Feature activation | -8.63e-02 |
Pos logit contributions | |
| +0.226 |
Zi | +0.219 |
Catalan | +0.218 |
Niet | +0.217 |
wikipedia | +0.207 |
Neg logit contributions | |
WORK | -0.243 |
liv | -0.195 |
awatts | -0.187 |
WARE | -0.185 |
aturdays | -0.178 |
Token | Magic |
Token ID | 6139 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
| +0.062 |
Zi | +0.060 |
Catalan | +0.059 |
Niet | +0.059 |
wikipedia | +0.056 |
Neg logit contributions | |
WORK | -0.072 |
liv | -0.058 |
WARE | -0.056 |
awatts | -0.055 |
aturdays | -0.053 |
Token | iterations |
Token ID | 34820 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
WORK | +0.199 |
liv | +0.159 |
awatts | +0.152 |
WARE | +0.151 |
aturdays | +0.145 |
Neg logit contributions | |
| -0.193 |
Zi | -0.187 |
Catalan | -0.186 |
Niet | -0.185 |
wikipedia | -0.176 |
Token | of |
Token ID | 286 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
WORK | +0.176 |
liv | +0.141 |
awatts | +0.135 |
WARE | +0.134 |
aturdays | +0.130 |
Neg logit contributions | |
| -0.166 |
Zi | -0.161 |
Catalan | -0.160 |
Niet | -0.159 |
wikipedia | -0.152 |
Token | released |
Token ID | 2716 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
WORK | +0.180 |
liv | +0.152 |
awatts | +0.147 |
WARE | +0.146 |
aturdays | +0.142 |
Neg logit contributions | |
| -0.095 |
Zi | -0.091 |
Catalan | -0.090 |
Niet | -0.090 |
wikipedia | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
| +0.278 |
Zi | +0.270 |
Catalan | +0.268 |
Niet | +0.267 |
wikipedia | +0.255 |
Neg logit contributions | |
WORK | -0.295 |
liv | -0.237 |
awatts | -0.226 |
WARE | -0.225 |
aturdays | -0.217 |
Token |
|
Token ID | 198 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
WORK | +0.215 |
liv | +0.165 |
awatts | +0.156 |
WARE | +0.155 |
Appl | +0.151 |
Neg logit contributions | |
| -0.259 |
Zi | -0.252 |
Catalan | -0.252 |
Niet | -0.251 |
wikipedia | -0.240 |
Token |
|
Token ID | 198 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
WORK | +0.259 |
liv | +0.207 |
aturdays | +0.202 |
awatts | +0.201 |
WARE | +0.200 |
Neg logit contributions | |
| -0.266 |
Zi | -0.254 |
Catalan | -0.253 |
Niet | -0.245 |
wikipedia | -0.240 |
Token | R |
Token ID | 49 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
WORK | +0.240 |
liv | +0.189 |
awatts | +0.180 |
WARE | +0.180 |
Appl | +0.172 |
Neg logit contributions | |
| -0.250 |
Zi | -0.243 |
Catalan | -0.242 |
Niet | -0.241 |
wikipedia | -0.230 |
Token | OB |
Token ID | 9864 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
WORK | +0.142 |
liv | +0.112 |
awatts | +0.106 |
WARE | +0.106 |
aturdays | +0.102 |
Neg logit contributions | |
| -0.154 |
Zi | -0.149 |
Catalan | -0.148 |
Niet | -0.148 |
wikipedia | -0.141 |
Token | OT |
Token ID | 2394 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
WORK | +1.335 |
liv | +1.049 |
WARE | +1.006 |
awatts | +0.993 |
Appl | +0.963 |
Neg logit contributions | |
| -1.321 |
Zi | -1.281 |
Niet | -1.275 |
Catalan | -1.275 |
wikipedia | -1.218 |
Token |
|
Token ID | 198 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
| +0.037 |
Zi | +0.037 |
Catalan | +0.036 |
Niet | +0.036 |
Moss | +0.034 |
Neg logit contributions | |
WORK | -0.037 |
liv | -0.029 |
awatts | -0.028 |
WARE | -0.028 |
aturdays | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
WORK | +0.136 |
liv | +0.109 |
aturdays | +0.107 |
WARE | +0.106 |
awatts | +0.106 |
Neg logit contributions | |
| -0.138 |
Zi | -0.132 |
Catalan | -0.131 |
Niet | -0.127 |
wikipedia | -0.125 |
Token | + |
Token ID | 10 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
WORK | +0.191 |
liv | +0.156 |
WARE | +0.150 |
awatts | +0.150 |
Appl | +0.145 |
Neg logit contributions | |
| -0.130 |
Zi | -0.126 |
Niet | -0.125 |
Catalan | -0.125 |
wikipedia | -0.118 |
Token | Law |
Token ID | 3854 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
WORK | +0.528 |
liv | +0.432 |
awatts | +0.414 |
WARE | +0.412 |
Appl | +0.403 |
Neg logit contributions | |
| -0.407 |
Zi | -0.395 |
Catalan | -0.391 |
Niet | -0.390 |
wikipedia | -0.372 |
Token | . |
Token ID | 13 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
| +0.287 |
Zi | +0.279 |
Niet | +0.278 |
Catalan | +0.277 |
wikipedia | +0.264 |
Neg logit contributions | |
WORK | -0.297 |
liv | -0.238 |
awatts | -0.227 |
WARE | -0.225 |
Appl | -0.218 |
Token | Available |
Token ID | 14898 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
| +0.028 |
Zi | +0.027 |
Catalan | +0.027 |
Niet | +0.027 |
wikipedia | +0.025 |
Neg logit contributions | |
WORK | -0.037 |
liv | -0.030 |
awatts | -0.029 |
WARE | -0.029 |
Appl | -0.028 |
Token | at |
Token ID | 379 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
| +0.130 |
Zi | +0.126 |
Catalan | +0.125 |
Niet | +0.123 |
wikipedia | +0.118 |
Neg logit contributions | |
WORK | -0.131 |
liv | -0.103 |
WARE | -0.100 |
awatts | -0.100 |
aturdays | -0.098 |
Token | http |
Token ID | 2638 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
| +0.177 |
Zi | +0.172 |
Catalan | +0.172 |
Niet | +0.172 |
wikipedia | +0.163 |
Neg logit contributions | |
WORK | -0.170 |
liv | -0.136 |
awatts | -0.128 |
WARE | -0.127 |
Appl | -0.123 |
Token | :// |
Token ID | 1378 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
WORK | +0.193 |
liv | +0.165 |
WARE | +0.160 |
awatts | +0.160 |
aturdays | +0.159 |
Neg logit contributions | |
| -0.092 |
Zi | -0.086 |
Catalan | -0.085 |
Niet | -0.083 |
wikipedia | -0.080 |
Token | df |
Token ID | 7568 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
| +0.230 |
Zi | +0.223 |
Catalan | +0.222 |
Niet | +0.221 |
wikipedia | +0.212 |
Neg logit contributions | |
WORK | -0.207 |
liv | -0.163 |
awatts | -0.155 |
WARE | -0.154 |
aturdays | -0.148 |
Token | cons |
Token ID | 5936 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
WORK | +1.313 |
liv | +1.058 |
awatts | +1.014 |
WARE | +1.000 |
aturdays | +0.971 |
Neg logit contributions | |
| -1.199 |
Zi | -1.163 |
Catalan | -1.159 |
Niet | -1.157 |
wikipedia | -1.091 |
Token | ult |
Token ID | 586 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
WORK | +0.622 |
liv | +0.520 |
awatts | +0.502 |
WARE | +0.494 |
Appl | +0.484 |
Neg logit contributions | |
| -0.370 |
Zi | -0.360 |
Niet | -0.357 |
Catalan | -0.357 |
Moss | -0.330 |
Token | ing |
Token ID | 278 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
| +0.411 |
Zi | +0.399 |
Catalan | +0.397 |
Niet | +0.396 |
wikipedia | +0.378 |
Neg logit contributions | |
WORK | -0.393 |
liv | -0.311 |
awatts | -0.296 |
WARE | -0.294 |
Appl | -0.283 |
Token | . |
Token ID | 13 |
Feature activation | -6.20e-02 |
Pos logit contributions | |
WORK | +0.138 |
liv | +0.111 |
awatts | +0.106 |
WARE | +0.105 |
Appl | +0.101 |
Neg logit contributions | |
| -0.132 |
Zi | -0.129 |
Catalan | -0.128 |
Niet | -0.128 |
wikipedia | -0.121 |
Token | info |
Token ID | 10951 |
Feature activation | +9.83e-02 |
Pos logit contributions | |
| +0.045 |
Zi | +0.044 |
Catalan | +0.043 |
Niet | +0.043 |
wikipedia | +0.041 |
Neg logit contributions | |
WORK | -0.051 |
liv | -0.041 |
awatts | -0.039 |
WARE | -0.039 |
Appl | -0.038 |
Token | We |
Token ID | 1135 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
WORK | +0.160 |
liv | +0.129 |
awatts | +0.123 |
WARE | +0.123 |
Appl | +0.119 |
Neg logit contributions | |
| -0.143 |
Zi | -0.139 |
Catalan | -0.138 |
Niet | -0.138 |
wikipedia | -0.130 |
Token | ber |
Token ID | 527 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
WORK | +0.242 |
liv | +0.196 |
awatts | +0.187 |
WARE | +0.185 |
aturdays | +0.180 |
Neg logit contributions | |
| -0.214 |
Zi | -0.207 |
Catalan | -0.206 |
Niet | -0.206 |
wikipedia | -0.195 |
Token | ( |
Token ID | 357 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
| +0.016 |
Zi | +0.016 |
Catalan | +0.016 |
Niet | +0.016 |
wikipedia | +0.015 |
Neg logit contributions | |
WORK | -0.018 |
liv | -0.014 |
awatts | -0.014 |
WARE | -0.013 |
Appl | -0.013 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
WORK | +0.175 |
liv | +0.140 |
awatts | +0.133 |
WARE | +0.132 |
aturdays | +0.128 |
Neg logit contributions | |
| -0.179 |
Zi | -0.173 |
Catalan | -0.172 |
Niet | -0.171 |
wikipedia | -0.164 |
Token | ivers |
Token ID | 1191 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
WORK | +0.082 |
liv | +0.060 |
awatts | +0.057 |
WARE | +0.056 |
aturdays | +0.054 |
Neg logit contributions | |
| -0.127 |
Zi | -0.124 |
Catalan | -0.124 |
Niet | -0.123 |
wikipedia | -0.119 |
Token | ité |
Token ID | 43816 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
WORK | +0.792 |
liv | +0.625 |
awatts | +0.594 |
WARE | +0.591 |
aturdays | +0.568 |
Neg logit contributions | |
| -0.849 |
Zi | -0.825 |
Catalan | -0.820 |
Niet | -0.818 |
wikipedia | -0.782 |
Token | de |
Token ID | 390 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
WORK | +1.638 |
liv | +1.376 |
awatts | +1.338 |
WARE | +1.334 |
aturdays | +1.298 |
Neg logit contributions | |
| -0.876 |
Zi | -0.839 |
Catalan | -0.836 |
Niet | -0.829 |
wikipedia | -0.771 |
Token | T |
Token ID | 309 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
WORK | +0.227 |
liv | +0.188 |
awatts | +0.185 |
WARE | +0.185 |
aturdays | +0.180 |
Neg logit contributions | |
| -0.128 |
Catalan | -0.123 |
Zi | -0.122 |
Niet | -0.120 |
Moss | -0.114 |
Token | oul |
Token ID | 2852 |
Feature activation | +5.22e-01 |
Pos logit contributions | |
WORK | +0.025 |
liv | +0.019 |
WARE | +0.018 |
awatts | +0.018 |
terday | +0.017 |
Neg logit contributions | |
| -0.033 |
Catalan | -0.032 |
Zi | -0.032 |
Niet | -0.031 |
wikipedia | -0.030 |
Token | ouse |
Token ID | 1076 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
WORK | +0.302 |
liv | +0.221 |
awatts | +0.205 |
WARE | +0.202 |
Appl | +0.194 |
Neg logit contributions | |
| -0.502 |
Zi | -0.492 |
Niet | -0.489 |
Catalan | -0.489 |
wikipedia | -0.470 |
Token | ) |
Token ID | 8 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
WORK | +0.031 |
liv | +0.025 |
awatts | +0.024 |
WARE | +0.024 |
aturdays | +0.023 |
Neg logit contributions | |
| -0.030 |
Zi | -0.029 |
Catalan | -0.029 |
Niet | -0.029 |
wikipedia | -0.027 |
Token | 1 |
Token ID | 16 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
WORK | +0.183 |
liv | +0.151 |
awatts | +0.145 |
WARE | +0.144 |
aturdays | +0.140 |
Neg logit contributions | |
| -0.138 |
Zi | -0.132 |
Catalan | -0.131 |
Niet | -0.131 |
wikipedia | -0.124 |
Token | ) |
Token ID | 8 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
WORK | +0.083 |
liv | +0.067 |
awatts | +0.063 |
WARE | +0.063 |
Appl | +0.061 |
Neg logit contributions | |
| -0.078 |
Zi | -0.076 |
Niet | -0.076 |
Catalan | -0.075 |
wikipedia | -0.072 |
Token |
|
Token ID | 198 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
WORK | +0.300 |
liv | +0.233 |
awatts | +0.219 |
WARE | +0.218 |
Appl | +0.211 |
Neg logit contributions | |
| -0.358 |
Zi | -0.351 |
Niet | -0.349 |
Catalan | -0.348 |
wikipedia | -0.334 |
Token |
|
Token ID | 198 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
WORK | +0.182 |
liv | +0.145 |
aturdays | +0.142 |
WARE | +0.141 |
awatts | +0.140 |
Neg logit contributions | |
| -0.187 |
Zi | -0.179 |
Catalan | -0.178 |
Niet | -0.172 |
wikipedia | -0.169 |
Token | Ex |
Token ID | 3109 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
WORK | +0.142 |
liv | +0.112 |
awatts | +0.106 |
WARE | +0.105 |
aturdays | +0.102 |
Neg logit contributions | |
| -0.170 |
Zi | -0.164 |
Catalan | -0.163 |
Niet | -0.162 |
wikipedia | -0.157 |
Token | amine |
Token ID | 9862 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
WORK | +0.043 |
liv | +0.034 |
awatts | +0.032 |
WARE | +0.032 |
aturdays | +0.031 |
Neg logit contributions | |
| -0.047 |
Zi | -0.045 |
Catalan | -0.045 |
Niet | -0.045 |
wikipedia | -0.043 |
Token | all |
Token ID | 477 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
WORK | +1.339 |
liv | +1.090 |
awatts | +1.035 |
WARE | +1.031 |
Appl | +1.006 |
Neg logit contributions | |
| -1.146 |
Zi | -1.108 |
Niet | -1.105 |
Catalan | -1.103 |
wikipedia | -1.053 |
Token | the |
Token ID | 262 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
WORK | +0.037 |
liv | +0.030 |
awatts | +0.029 |
WARE | +0.029 |
aturdays | +0.028 |
Neg logit contributions | |
| -0.031 |
Zi | -0.030 |
Catalan | -0.029 |
Niet | -0.029 |
wikipedia | -0.028 |
Token | vit |
Token ID | 9090 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
WORK | +0.085 |
liv | +0.068 |
awatts | +0.066 |
WARE | +0.066 |
aturdays | +0.063 |
Neg logit contributions | |
| -0.076 |
Zi | -0.074 |
Catalan | -0.073 |
Niet | -0.073 |
Moss | -0.069 |
Token | rified |
Token ID | 41301 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
WORK | +0.227 |
liv | +0.183 |
awatts | +0.175 |
WARE | +0.175 |
aturdays | +0.169 |
Neg logit contributions | |
| -0.197 |
Zi | -0.191 |
Catalan | -0.190 |
Niet | -0.189 |
wikipedia | -0.180 |
Token | test |
Token ID | 1332 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
WORK | +0.334 |
liv | +0.267 |
awatts | +0.256 |
WARE | +0.255 |
aturdays | +0.246 |
Neg logit contributions | |
| -0.317 |
Zi | -0.307 |
Catalan | -0.306 |
Niet | -0.304 |
wikipedia | -0.290 |
Token | wrap |
Token ID | 37150 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
| +0.068 |
Zi | +0.066 |
Catalan | +0.065 |
Niet | +0.065 |
wikipedia | +0.062 |
Neg logit contributions | |
WORK | -0.073 |
liv | -0.058 |
awatts | -0.056 |
WARE | -0.056 |
aturdays | -0.054 |
Token | ()) |
Token ID | 32865 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
WORK | +0.638 |
liv | +0.530 |
awatts | +0.511 |
WARE | +0.508 |
aturdays | +0.495 |
Neg logit contributions | |
| -0.399 |
Zi | -0.385 |
Niet | -0.384 |
Catalan | -0.382 |
wikipedia | -0.358 |
Token | } |
Token ID | 1782 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
| +0.093 |
Zi | +0.090 |
Catalan | +0.089 |
Niet | +0.089 |
wikipedia | +0.085 |
Neg logit contributions | |
WORK | -0.099 |
liv | -0.079 |
awatts | -0.076 |
WARE | -0.076 |
aturdays | -0.073 |
Token | } |
Token ID | 1782 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
| +0.440 |
Zi | +0.426 |
Catalan | +0.423 |
Niet | +0.420 |
Moss | +0.400 |
Neg logit contributions | |
WORK | -0.443 |
liv | -0.353 |
awatts | -0.337 |
WARE | -0.335 |
aturdays | -0.324 |
Token | pub |
Token ID | 2240 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
| +0.349 |
Zi | +0.339 |
Catalan | +0.337 |
Niet | +0.335 |
wikipedia | +0.320 |
Neg logit contributions | |
WORK | -0.335 |
liv | -0.265 |
awatts | -0.253 |
WARE | -0.251 |
aturdays | -0.242 |
Token | fn |
Token ID | 24714 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
WORK | +0.303 |
liv | +0.243 |
awatts | +0.230 |
WARE | +0.228 |
Appl | +0.221 |
Neg logit contributions | |
| -0.300 |
Zi | -0.291 |
Niet | -0.291 |
Catalan | -0.290 |
wikipedia | -0.276 |
Token | apply |
Token ID | 4174 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
WORK | +1.244 |
liv | +0.998 |
awatts | +0.945 |
WARE | +0.937 |
Appl | +0.909 |
Neg logit contributions | |
| -1.201 |
Zi | -1.169 |
Niet | -1.166 |
Catalan | -1.164 |
wikipedia | -1.109 |
Token | _ |
Token ID | 62 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
| +0.156 |
Zi | +0.149 |
Niet | +0.148 |
Catalan | +0.148 |
wikipedia | +0.135 |
Neg logit contributions | |
WORK | -0.369 |
liv | -0.315 |
awatts | -0.305 |
WARE | -0.303 |
Appl | -0.297 |
Token | move |
Token ID | 21084 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
WORK | +0.015 |
liv | +0.012 |
awatts | +0.011 |
WARE | +0.011 |
Appl | +0.011 |
Neg logit contributions | |
| -0.014 |
Zi | -0.014 |
Niet | -0.014 |
Catalan | -0.014 |
Moss | -0.013 |
Token | ( |
Token ID | 357 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
| +0.354 |
Niet | +0.344 |
Zi | +0.342 |
Catalan | +0.342 |
Moss | +0.320 |
Neg logit contributions | |
WORK | -0.552 |
liv | -0.458 |
awatts | -0.441 |
WARE | -0.435 |
aturdays | -0.428 |
Token | & |
Token ID | 1222 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
| +0.013 |
Zi | +0.013 |
Catalan | +0.013 |
Niet | +0.013 |
wikipedia | +0.013 |
Neg logit contributions | |
WORK | -0.009 |
liv | -0.006 |
awatts | -0.006 |
WARE | -0.006 |
Appl | -0.006 |
Token | Guard |
Token ID | 4932 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
WORK | +0.160 |
liv | +0.131 |
awatts | +0.125 |
WARE | +0.125 |
Appl | +0.121 |
Neg logit contributions | |
| -0.126 |
Zi | -0.122 |
Catalan | -0.121 |
Niet | -0.121 |
wikipedia | -0.115 |
Token | { |
Token ID | 1391 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
| +0.232 |
Zi | +0.223 |
Catalan | +0.222 |
Niet | +0.221 |
wikipedia | +0.208 |
Neg logit contributions | |
WORK | -0.374 |
liv | -0.312 |
awatts | -0.301 |
WARE | -0.300 |
Appl | -0.291 |
Token | ... |
Token ID | 2644 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
| +0.244 |
Zi | +0.236 |
Catalan | +0.234 |
Niet | +0.233 |
wikipedia | +0.222 |
Neg logit contributions | |
WORK | -0.281 |
liv | -0.228 |
awatts | -0.218 |
WARE | -0.217 |
aturdays | -0.210 |
Token | } |
Token ID | 1782 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
| +0.140 |
Zi | +0.135 |
Catalan | +0.134 |
Niet | +0.133 |
wikipedia | +0.127 |
Neg logit contributions | |
WORK | -0.154 |
liv | -0.124 |
awatts | -0.119 |
WARE | -0.118 |
aturdays | -0.114 |
Token | pub |
Token ID | 2240 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
| +0.390 |
Zi | +0.378 |
Catalan | +0.376 |
Niet | +0.374 |
wikipedia | +0.357 |
Neg logit contributions | |
WORK | -0.387 |
liv | -0.308 |
awatts | -0.294 |
WARE | -0.292 |
aturdays | -0.282 |
Token | fn |
Token ID | 24714 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
WORK | +0.332 |
liv | +0.268 |
awatts | +0.254 |
WARE | +0.252 |
Appl | +0.246 |
Neg logit contributions | |
| -0.307 |
Niet | -0.298 |
Zi | -0.298 |
Catalan | -0.296 |
wikipedia | -0.283 |
Token | pin |
Token ID | 6757 |
Feature activation | +7.10e-02 |
Pos logit contributions | |
WORK | +1.209 |
liv | +0.970 |
awatts | +0.913 |
WARE | +0.905 |
Appl | +0.881 |
Neg logit contributions | |
| -1.212 |
Zi | -1.175 |
Niet | -1.175 |
Catalan | -1.172 |
wikipedia | -1.122 |
Token | () |
Token ID | 7499 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
WORK | +0.066 |
liv | +0.054 |
awatts | +0.052 |
WARE | +0.052 |
Appl | +0.050 |
Neg logit contributions | |
| -0.044 |
Niet | -0.043 |
Catalan | -0.043 |
Zi | -0.043 |
Moss | -0.040 |
Token | -> |
Token ID | 4613 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
| +0.130 |
Zi | +0.126 |
Catalan | +0.125 |
Niet | +0.124 |
wikipedia | +0.117 |
Neg logit contributions | |
WORK | -0.180 |
liv | -0.147 |
awatts | -0.142 |
WARE | -0.142 |
aturdays | -0.137 |
Token | Guard |
Token ID | 4932 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
| +0.152 |
Zi | +0.147 |
Catalan | +0.146 |
Niet | +0.145 |
Moss | +0.138 |
Neg logit contributions | |
WORK | -0.178 |
liv | -0.144 |
awatts | -0.138 |
WARE | -0.138 |
aturdays | -0.133 |
Token | ; |
Token ID | 2162 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
| +0.301 |
Zi | +0.290 |
Catalan | +0.287 |
Niet | +0.286 |
wikipedia | +0.269 |
Neg logit contributions | |
WORK | -0.484 |
liv | -0.404 |
awatts | -0.390 |
WARE | -0.388 |
aturdays | -0.377 |
Token | omsday |
Token ID | 33415 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
| +0.039 |
Zi | +0.038 |
Catalan | +0.038 |
Niet | +0.037 |
Moss | +0.035 |
Neg logit contributions | |
WORK | -0.043 |
liv | -0.034 |
awatts | -0.033 |
WARE | -0.033 |
aturdays | -0.032 |
Token | cult |
Token ID | 2285 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
| +0.441 |
Zi | +0.430 |
Catalan | +0.428 |
Niet | +0.426 |
wikipedia | +0.409 |
Neg logit contributions | |
WORK | -0.337 |
liv | -0.258 |
awatts | -0.244 |
WARE | -0.242 |
aturdays | -0.232 |
Token | , |
Token ID | 11 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
| +0.101 |
Zi | +0.097 |
Catalan | +0.097 |
Niet | +0.097 |
wikipedia | +0.092 |
Neg logit contributions | |
WORK | -0.131 |
liv | -0.108 |
awatts | -0.103 |
WARE | -0.103 |
Appl | -0.100 |
Token | releases |
Token ID | 10050 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
WORK | +0.020 |
liv | +0.016 |
awatts | +0.016 |
WARE | +0.015 |
aturdays | +0.015 |
Neg logit contributions | |
| -0.017 |
Zi | -0.016 |
Catalan | -0.016 |
Niet | -0.016 |
wikipedia | -0.015 |
Token | s |
Token ID | 264 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
| +0.115 |
Zi | +0.112 |
Catalan | +0.111 |
Niet | +0.110 |
Moss | +0.104 |
Neg logit contributions | |
WORK | -0.142 |
liv | -0.115 |
awatts | -0.111 |
WARE | -0.110 |
aturdays | -0.107 |
Token | arin |
Token ID | 17714 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
WORK | +0.435 |
liv | +0.355 |
awatts | +0.340 |
WARE | +0.339 |
Appl | +0.328 |
Neg logit contributions | |
| -0.346 |
Zi | -0.335 |
Catalan | -0.332 |
Niet | -0.331 |
wikipedia | -0.314 |
Token | gas |
Token ID | 3623 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
WORK | +1.095 |
liv | +0.854 |
WARE | +0.804 |
awatts | +0.803 |
Appl | +0.771 |
Neg logit contributions | |
| -1.296 |
Niet | -1.261 |
Zi | -1.256 |
Catalan | -1.254 |
wikipedia | -1.205 |
Token | in |
Token ID | 287 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
| +0.167 |
Zi | +0.162 |
Catalan | +0.161 |
Niet | +0.160 |
wikipedia | +0.152 |
Neg logit contributions | |
WORK | -0.178 |
liv | -0.142 |
awatts | -0.136 |
WARE | -0.136 |
aturdays | -0.131 |
Token | a |
Token ID | 257 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
WORK | +0.100 |
liv | +0.083 |
awatts | +0.080 |
WARE | +0.079 |
aturdays | +0.077 |
Neg logit contributions | |
| -0.068 |
Zi | -0.066 |
Catalan | -0.065 |
Niet | -0.065 |
wikipedia | -0.061 |
Token | Tokyo |
Token ID | 11790 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
WORK | +0.087 |
liv | +0.069 |
awatts | +0.067 |
WARE | +0.067 |
aturdays | +0.065 |
Neg logit contributions | |
| -0.079 |
Zi | -0.077 |
Catalan | -0.076 |
Niet | -0.076 |
Moss | -0.072 |
Token | subway |
Token ID | 19612 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
WORK | +0.417 |
liv | +0.321 |
awatts | +0.308 |
WARE | +0.304 |
aturdays | +0.293 |
Neg logit contributions | |
| -0.505 |
Zi | -0.494 |
Catalan | -0.491 |
Niet | -0.488 |
Moss | -0.468 |
Token | state |
Token ID | 5219 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
| +0.052 |
Zi | +0.049 |
Catalan | +0.049 |
Niet | +0.048 |
wikipedia | +0.048 |
Neg logit contributions | |
WORK | -0.047 |
liv | -0.037 |
WARE | -0.036 |
aturdays | -0.036 |
awatts | -0.035 |
Token | ); |
Token ID | 5619 |
Feature activation | +7.93e-02 |
Pos logit contributions | |
WORK | +0.130 |
liv | +0.109 |
awatts | +0.105 |
WARE | +0.105 |
Appl | +0.102 |
Neg logit contributions | |
| -0.076 |
Zi | -0.073 |
Niet | -0.072 |
Catalan | -0.072 |
wikipedia | -0.068 |
Token | ... |
Token ID | 2644 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
WORK | +0.066 |
liv | +0.053 |
awatts | +0.051 |
WARE | +0.051 |
aturdays | +0.050 |
Neg logit contributions | |
| -0.057 |
Zi | -0.055 |
Catalan | -0.055 |
Niet | -0.054 |
wikipedia | -0.052 |
Token | } |
Token ID | 1782 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
| +0.160 |
Zi | +0.155 |
Catalan | +0.154 |
Niet | +0.153 |
wikipedia | +0.146 |
Neg logit contributions | |
WORK | -0.164 |
liv | -0.130 |
awatts | -0.125 |
WARE | -0.124 |
aturdays | -0.120 |
Token | pub |
Token ID | 2240 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
| +0.214 |
Zi | +0.208 |
Catalan | +0.206 |
Niet | +0.205 |
wikipedia | +0.196 |
Neg logit contributions | |
WORK | -0.205 |
liv | -0.162 |
awatts | -0.155 |
WARE | -0.154 |
aturdays | -0.148 |
Token | fn |
Token ID | 24714 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
WORK | +0.285 |
liv | +0.228 |
awatts | +0.215 |
WARE | +0.213 |
Appl | +0.207 |
Neg logit contributions | |
| -0.288 |
Niet | -0.280 |
Zi | -0.280 |
Catalan | -0.279 |
wikipedia | -0.267 |
Token | undo |
Token ID | 23981 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
WORK | +1.235 |
liv | +0.996 |
awatts | +0.943 |
WARE | +0.934 |
Appl | +0.908 |
Neg logit contributions | |
| -1.143 |
Zi | -1.113 |
Niet | -1.110 |
Catalan | -1.107 |
wikipedia | -1.054 |
Token | _ |
Token ID | 62 |
Feature activation | -2.88e-03 |
Pos logit contributions | |
| +0.149 |
Zi | +0.140 |
Catalan | +0.138 |
Niet | +0.135 |
wikipedia | +0.127 |
Neg logit contributions | |
WORK | -0.384 |
liv | -0.331 |
awatts | -0.321 |
WARE | -0.320 |
aturdays | -0.316 |
Token | move |
Token ID | 21084 |
Feature activation | -6.33e-01 |
Pos logit contributions | |
| +0.002 |
Zi | +0.002 |
Catalan | +0.002 |
wikipedia | +0.002 |
Niet | +0.002 |
Neg logit contributions | |
WORK | -0.002 |
liv | -0.002 |
WARE | -0.002 |
awatts | -0.002 |
aturdays | -0.002 |
Token | ( |
Token ID | 357 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
| +0.396 |
Niet | +0.383 |
Zi | +0.383 |
Catalan | +0.381 |
Moss | +0.357 |
Neg logit contributions | |
WORK | -0.583 |
liv | -0.480 |
awatts | -0.462 |
WARE | -0.458 |
aturdays | -0.448 |
Token | & |
Token ID | 1222 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
WORK | +0.010 |
liv | +0.007 |
awatts | +0.007 |
WARE | +0.007 |
Appl | +0.006 |
Neg logit contributions | |
| -0.016 |
Zi | -0.016 |
Niet | -0.016 |
Catalan | -0.016 |
wikipedia | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
WORK | +0.269 |
liv | +0.214 |
awatts | +0.203 |
WARE | +0.202 |
aturdays | +0.195 |
Neg logit contributions | |
| -0.277 |
Zi | -0.269 |
Catalan | -0.267 |
Niet | -0.267 |
wikipedia | -0.255 |
Token | is |
Token ID | 318 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
WORK | +0.039 |
liv | +0.031 |
awatts | +0.030 |
WARE | +0.030 |
Appl | +0.029 |
Neg logit contributions | |
| -0.034 |
Zi | -0.033 |
Catalan | -0.033 |
Niet | -0.033 |
wikipedia | -0.031 |
Token | all |
Token ID | 477 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
WORK | +0.099 |
liv | +0.081 |
awatts | +0.077 |
WARE | +0.076 |
Appl | +0.074 |
Neg logit contributions | |
| -0.086 |
Zi | -0.084 |
Catalan | -0.083 |
Niet | -0.083 |
wikipedia | -0.079 |
Token | that |
Token ID | 326 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
WORK | +0.017 |
liv | +0.014 |
awatts | +0.013 |
WARE | +0.013 |
Appl | +0.013 |
Neg logit contributions | |
| -0.015 |
Zi | -0.015 |
Catalan | -0.015 |
Niet | -0.015 |
wikipedia | -0.014 |
Token | is |
Token ID | 318 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
WORK | +0.100 |
liv | +0.081 |
awatts | +0.077 |
WARE | +0.077 |
Appl | +0.074 |
Neg logit contributions | |
| -0.090 |
Zi | -0.087 |
Catalan | -0.086 |
Niet | -0.086 |
wikipedia | -0.082 |
Token | wrong |
Token ID | 2642 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
WORK | +0.161 |
liv | +0.131 |
awatts | +0.124 |
WARE | +0.123 |
Appl | +0.119 |
Neg logit contributions | |
| -0.148 |
Zi | -0.144 |
Catalan | -0.143 |
Niet | -0.142 |
wikipedia | -0.136 |
Token | with |
Token ID | 351 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
WORK | +1.163 |
liv | +0.932 |
awatts | +0.878 |
WARE | +0.868 |
Appl | +0.838 |
Neg logit contributions | |
| -1.189 |
Zi | -1.164 |
Niet | -1.155 |
Catalan | -1.153 |
wikipedia | -1.104 |
Token | modern |
Token ID | 3660 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
WORK | +0.062 |
liv | +0.052 |
awatts | +0.049 |
WARE | +0.049 |
Appl | +0.047 |
Neg logit contributions | |
| -0.045 |
Zi | -0.044 |
Catalan | -0.043 |
Niet | -0.043 |
wikipedia | -0.041 |
Token | culture |
Token ID | 3968 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
| +0.400 |
Zi | +0.387 |
Catalan | +0.385 |
Niet | +0.384 |
wikipedia | +0.366 |
Neg logit contributions | |
WORK | -0.412 |
liv | -0.328 |
awatts | -0.314 |
WARE | -0.313 |
Appl | -0.301 |
Token | . |
Token ID | 13 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
| +0.076 |
Zi | +0.074 |
Catalan | +0.073 |
Niet | +0.073 |
wikipedia | +0.070 |
Neg logit contributions | |
WORK | -0.082 |
liv | -0.066 |
awatts | -0.063 |
WARE | -0.063 |
aturdays | -0.061 |
Token |
|
Token ID | 198 |
Feature activation | -7.79e-02 |
Pos logit contributions | |
| +0.028 |
Zi | +0.027 |
Catalan | +0.027 |
Niet | +0.027 |
wikipedia | +0.026 |
Neg logit contributions | |
WORK | -0.027 |
liv | -0.021 |
awatts | -0.020 |
WARE | -0.020 |
Appl | -0.019 |
Token | we |
Token ID | 356 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
WORK | +0.024 |
liv | +0.019 |
awatts | +0.018 |
WARE | +0.018 |
Appl | +0.017 |
Neg logit contributions | |
| -0.027 |
Zi | -0.026 |
Catalan | -0.026 |
Niet | -0.026 |
wikipedia | -0.025 |
Token | will |
Token ID | 481 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
WORK | +0.085 |
liv | +0.069 |
awatts | +0.066 |
WARE | +0.066 |
aturdays | +0.063 |
Neg logit contributions | |
| -0.073 |
Zi | -0.070 |
Catalan | -0.070 |
Niet | -0.070 |
wikipedia | -0.066 |
Token | use |
Token ID | 779 |
Feature activation | +2.64e-03 |
Pos logit contributions | |
WORK | +0.082 |
liv | +0.064 |
awatts | +0.060 |
WARE | +0.060 |
aturdays | +0.058 |
Neg logit contributions | |
| -0.096 |
Zi | -0.094 |
Catalan | -0.093 |
Niet | -0.093 |
wikipedia | -0.089 |
Token | the |
Token ID | 262 |
Feature activation | +7.97e-02 |
Pos logit contributions | |
WORK | +0.002 |
liv | +0.002 |
WARE | +0.002 |
awatts | +0.002 |
Appl | +0.002 |
Neg logit contributions | |
| -0.002 |
Zi | -0.002 |
Catalan | -0.002 |
Niet | -0.002 |
wikipedia | -0.002 |
Token | � |
Token ID | 18074 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
WORK | +0.070 |
liv | +0.057 |
awatts | +0.055 |
WARE | +0.055 |
Appl | +0.053 |
Neg logit contributions | |
| -0.053 |
Zi | -0.051 |
Catalan | -0.051 |
Niet | -0.051 |
wikipedia | -0.048 |
Token | � |
Token ID | 229 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
| +0.266 |
Zi | +0.256 |
Catalan | +0.255 |
Niet | +0.254 |
wikipedia | +0.240 |
Neg logit contributions | |
WORK | -0.385 |
liv | -0.319 |
awatts | -0.306 |
WARE | -0.305 |
Appl | -0.295 |
Token | 2 |
Token ID | 17 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
WORK | +1.179 |
liv | +0.944 |
WARE | +0.885 |
awatts | +0.879 |
Appl | +0.865 |
Neg logit contributions | |
| -1.185 |
Niet | -1.149 |
Catalan | -1.137 |
Zi | -1.129 |
wikipedia | -1.090 |
Token | test |
Token ID | 1332 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
| +0.224 |
Zi | +0.216 |
Catalan | +0.215 |
Niet | +0.214 |
wikipedia | +0.203 |
Neg logit contributions | |
WORK | -0.271 |
liv | -0.220 |
awatts | -0.211 |
WARE | -0.210 |
Appl | -0.203 |
Token | for |
Token ID | 329 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
WORK | +0.186 |
liv | +0.150 |
awatts | +0.142 |
WARE | +0.142 |
Appl | +0.138 |
Neg logit contributions | |
| -0.170 |
Zi | -0.164 |
Niet | -0.164 |
Catalan | -0.163 |
wikipedia | -0.156 |
Token | each |
Token ID | 1123 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
WORK | +0.088 |
liv | +0.072 |
awatts | +0.068 |
WARE | +0.068 |
Appl | +0.066 |
Neg logit contributions | |
| -0.070 |
Zi | -0.068 |
Niet | -0.067 |
Catalan | -0.067 |
wikipedia | -0.064 |
Token | sub |
Token ID | 850 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
WORK | +0.166 |
liv | +0.135 |
awatts | +0.128 |
WARE | +0.128 |
Appl | +0.125 |
Neg logit contributions | |
| -0.136 |
Zi | -0.131 |
Niet | -0.131 |
Catalan | -0.130 |
wikipedia | -0.124 |
Token | I |
Token ID | 314 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
| +0.040 |
Zi | +0.039 |
Catalan | +0.039 |
Niet | +0.039 |
wikipedia | +0.037 |
Neg logit contributions | |
WORK | -0.038 |
liv | -0.030 |
awatts | -0.029 |
WARE | -0.029 |
aturdays | -0.028 |
Token | put |
Token ID | 1234 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
WORK | +0.119 |
liv | +0.092 |
awatts | +0.089 |
WARE | +0.089 |
aturdays | +0.086 |
Neg logit contributions | |
| -0.131 |
Zi | -0.127 |
Catalan | -0.126 |
Niet | -0.125 |
Moss | -0.120 |
Token | out |
Token ID | 503 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
WORK | +0.127 |
liv | +0.102 |
awatts | +0.097 |
WARE | +0.097 |
Appl | +0.094 |
Neg logit contributions | |
| -0.117 |
Zi | -0.113 |
Catalan | -0.113 |
Niet | -0.113 |
wikipedia | -0.107 |
Token | a |
Token ID | 257 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
| +0.036 |
Zi | +0.035 |
Catalan | +0.035 |
Niet | +0.035 |
wikipedia | +0.033 |
Neg logit contributions | |
WORK | -0.041 |
liv | -0.033 |
awatts | -0.032 |
WARE | -0.032 |
Appl | -0.031 |
Token | little |
Token ID | 1310 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
WORK | +0.072 |
liv | +0.057 |
awatts | +0.055 |
WARE | +0.054 |
aturdays | +0.052 |
Neg logit contributions | |
| -0.072 |
Zi | -0.070 |
Catalan | -0.070 |
Niet | -0.069 |
wikipedia | -0.066 |
Token | shed |
Token ID | 14999 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
| +0.063 |
Zi | +0.061 |
Catalan | +0.061 |
Niet | +0.061 |
Moss | +0.058 |
Neg logit contributions | |
WORK | -0.058 |
liv | -0.045 |
awatts | -0.044 |
WARE | -0.043 |
aturdays | -0.042 |
Token | in |
Token ID | 287 |
Feature activation | +8.98e-02 |
Pos logit contributions | |
WORK | +1.214 |
liv | +0.975 |
awatts | +0.928 |
WARE | +0.923 |
Appl | +0.892 |
Neg logit contributions | |
| -1.147 |
Zi | -1.115 |
Catalan | -1.109 |
Niet | -1.109 |
wikipedia | -1.056 |
Token | a |
Token ID | 257 |
Feature activation | +7.71e-02 |
Pos logit contributions | |
WORK | +0.075 |
liv | +0.061 |
awatts | +0.058 |
WARE | +0.058 |
Appl | +0.056 |
Neg logit contributions | |
| -0.063 |
Zi | -0.061 |
Niet | -0.061 |
Catalan | -0.061 |
wikipedia | -0.058 |
Token | parking |
Token ID | 7647 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
WORK | +0.056 |
liv | +0.044 |
awatts | +0.042 |
WARE | +0.041 |
aturdays | +0.040 |
Neg logit contributions | |
| -0.063 |
Zi | -0.062 |
Catalan | -0.061 |
Niet | -0.061 |
wikipedia | -0.058 |
Token | lot |
Token ID | 1256 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
WORK | +0.068 |
liv | +0.050 |
awatts | +0.047 |
WARE | +0.047 |
aturdays | +0.045 |
Neg logit contributions | |
| -0.100 |
Zi | -0.098 |
Catalan | -0.097 |
Niet | -0.097 |
wikipedia | -0.093 |
Token | and |
Token ID | 290 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
WORK | +0.041 |
liv | +0.032 |
awatts | +0.031 |
WARE | +0.031 |
aturdays | +0.030 |
Neg logit contributions | |
| -0.039 |
Zi | -0.038 |
Catalan | -0.037 |
Niet | -0.037 |
wikipedia | -0.035 |
Token | The |
Token ID | 383 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
WORK | +0.075 |
liv | +0.060 |
awatts | +0.058 |
WARE | +0.057 |
aturdays | +0.055 |
Neg logit contributions | |
| -0.072 |
Zi | -0.070 |
Catalan | -0.069 |
Niet | -0.069 |
Moss | -0.066 |
Token | K |
Token ID | 509 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
| +0.007 |
Zi | +0.007 |
Catalan | +0.007 |
Niet | +0.007 |
wikipedia | +0.006 |
Neg logit contributions | |
WORK | -0.010 |
liv | -0.008 |
awatts | -0.008 |
WARE | -0.008 |
aturdays | -0.008 |
Token | EF |
Token ID | 25425 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
WORK | +0.132 |
liv | +0.109 |
awatts | +0.105 |
WARE | +0.104 |
aturdays | +0.101 |
Neg logit contributions | |
| -0.098 |
Zi | -0.095 |
Catalan | -0.094 |
Niet | -0.094 |
wikipedia | -0.089 |
Token | s |
Token ID | 82 |
Feature activation | -9.31e-02 |
Pos logit contributions | |
WORK | +0.672 |
liv | +0.540 |
awatts | +0.516 |
WARE | +0.514 |
aturdays | +0.496 |
Neg logit contributions | |
| -0.620 |
Zi | -0.601 |
Catalan | -0.597 |
Niet | -0.595 |
wikipedia | -0.567 |
Token | , |
Token ID | 11 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
| +0.072 |
Zi | +0.070 |
Catalan | +0.069 |
Niet | +0.069 |
Moss | +0.066 |
Neg logit contributions | |
WORK | -0.072 |
liv | -0.057 |
awatts | -0.055 |
WARE | -0.055 |
aturdays | -0.053 |
Token | Def |
Token ID | 2896 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
| +0.079 |
Zi | +0.077 |
Catalan | +0.076 |
Niet | +0.076 |
Moss | +0.072 |
Neg logit contributions | |
WORK | -0.082 |
liv | -0.065 |
awatts | -0.063 |
WARE | -0.062 |
aturdays | -0.060 |
Token | Tech |
Token ID | 17760 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
WORK | +0.974 |
liv | +0.734 |
awatts | +0.690 |
WARE | +0.679 |
Appl | +0.652 |
Neg logit contributions | |
| -1.384 |
Zi | -1.353 |
Niet | -1.346 |
Catalan | -1.344 |
wikipedia | -1.292 |
Token | s |
Token ID | 82 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
WORK | +0.010 |
liv | +0.008 |
awatts | +0.008 |
WARE | +0.008 |
aturdays | +0.008 |
Neg logit contributions | |
| -0.011 |
Zi | -0.011 |
Catalan | -0.011 |
Niet | -0.010 |
Moss | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
| +0.017 |
Zi | +0.017 |
Catalan | +0.017 |
Niet | +0.016 |
wikipedia | +0.016 |
Neg logit contributions | |
WORK | -0.016 |
liv | -0.012 |
awatts | -0.012 |
WARE | -0.012 |
aturdays | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
| +0.086 |
Zi | +0.084 |
Catalan | +0.083 |
Niet | +0.083 |
wikipedia | +0.080 |
Neg logit contributions | |
WORK | -0.078 |
liv | -0.061 |
awatts | -0.058 |
WARE | -0.058 |
aturdays | -0.055 |
Token | top |
Token ID | 1353 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
| +0.086 |
Zi | +0.083 |
Catalan | +0.083 |
Niet | +0.082 |
wikipedia | +0.077 |
Neg logit contributions | |
WORK | -0.137 |
liv | -0.114 |
awatts | -0.110 |
WARE | -0.110 |
aturdays | -0.107 |
Token | risks |
Token ID | 7476 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
| +0.119 |
Zi | +0.115 |
Catalan | +0.114 |
Niet | +0.114 |
wikipedia | +0.108 |
Neg logit contributions | |
WORK | -0.153 |
liv | -0.125 |
awatts | -0.120 |
WARE | -0.120 |
aturdays | -0.116 |
Token | of |
Token ID | 286 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
WORK | +0.138 |
liv | +0.112 |
awatts | +0.107 |
WARE | +0.106 |
Appl | +0.103 |
Neg logit contributions | |
| -0.122 |
Zi | -0.119 |
Niet | -0.118 |
Catalan | -0.118 |
wikipedia | -0.112 |
Token | a |
Token ID | 257 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
| +0.029 |
Zi | +0.028 |
Catalan | +0.028 |
Niet | +0.028 |
wikipedia | +0.026 |
Neg logit contributions | |
WORK | -0.028 |
liv | -0.022 |
awatts | -0.021 |
WARE | -0.021 |
aturdays | -0.020 |
Token | Deep |
Token ID | 10766 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
| +0.087 |
Zi | +0.085 |
Catalan | +0.084 |
Niet | +0.084 |
wikipedia | +0.080 |
Neg logit contributions | |
WORK | -0.078 |
liv | -0.061 |
awatts | -0.058 |
WARE | -0.058 |
aturdays | -0.056 |
Token | water |
Token ID | 7050 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
WORK | +0.123 |
liv | +0.106 |
awatts | +0.102 |
WARE | +0.102 |
Appl | +0.100 |
Neg logit contributions | |
| -0.049 |
Zi | -0.046 |
Catalan | -0.046 |
Niet | -0.046 |
wikipedia | -0.042 |
Token | Horizon |
Token ID | 22776 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
| +0.113 |
Zi | +0.107 |
Catalan | +0.106 |
Niet | +0.105 |
Moss | +0.098 |
Neg logit contributions | |
WORK | -0.231 |
liv | -0.196 |
awatts | -0.190 |
WARE | -0.190 |
aturdays | -0.186 |
Token | - |
Token ID | 12 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
WORK | +1.049 |
liv | +0.816 |
awatts | +0.765 |
WARE | +0.759 |
Appl | +0.732 |
Neg logit contributions | |
| -1.283 |
Zi | -1.255 |
Niet | -1.249 |
Catalan | -1.247 |
wikipedia | -1.197 |
Token | style |
Token ID | 7635 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
| +0.071 |
Zi | +0.069 |
Catalan | +0.068 |
Niet | +0.068 |
wikipedia | +0.065 |
Neg logit contributions | |
WORK | -0.061 |
liv | -0.048 |
awatts | -0.045 |
WARE | -0.045 |
aturdays | -0.043 |
Token | blow |
Token ID | 6611 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
| +0.193 |
Zi | +0.188 |
Catalan | +0.187 |
Niet | +0.186 |
Moss | +0.179 |
Neg logit contributions | |
WORK | -0.151 |
liv | -0.114 |
awatts | -0.110 |
WARE | -0.109 |
aturdays | -0.105 |
Token | out |
Token ID | 448 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
| +0.042 |
Zi | +0.041 |
Catalan | +0.041 |
Niet | +0.041 |
wikipedia | +0.040 |
Neg logit contributions | |
WORK | -0.030 |
liv | -0.022 |
awatts | -0.021 |
WARE | -0.021 |
Appl | -0.020 |
Token | late |
Token ID | 2739 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
| +0.176 |
Zi | +0.170 |
Catalan | +0.169 |
Niet | +0.168 |
wikipedia | +0.160 |
Neg logit contributions | |
WORK | -0.191 |
liv | -0.153 |
awatts | -0.147 |
WARE | -0.146 |
aturdays | -0.142 |
Token | ends |
Token ID | 5645 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
| +0.006 |
Zi | +0.006 |
Catalan | +0.006 |
Niet | +0.006 |
wikipedia | +0.005 |
Neg logit contributions | |
WORK | -0.006 |
liv | -0.005 |
awatts | -0.005 |
WARE | -0.005 |
Appl | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
| +0.199 |
Zi | +0.192 |
Catalan | +0.191 |
Niet | +0.190 |
wikipedia | +0.182 |
Neg logit contributions | |
WORK | -0.215 |
liv | -0.172 |
WARE | -0.165 |
awatts | -0.164 |
aturdays | -0.158 |
Token |
|
Token ID | 198 |
Feature activation | -7.84e-02 |
Pos logit contributions | |
| +0.128 |
Zi | +0.124 |
Catalan | +0.123 |
Niet | +0.123 |
wikipedia | +0.117 |
Neg logit contributions | |
WORK | -0.121 |
liv | -0.095 |
awatts | -0.091 |
WARE | -0.091 |
aturdays | -0.088 |
Token |
|
Token ID | 198 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
| +0.062 |
Zi | +0.059 |
Catalan | +0.058 |
Niet | +0.057 |
wikipedia | +0.056 |
Neg logit contributions | |
WORK | -0.061 |
liv | -0.049 |
aturdays | -0.048 |
awatts | -0.047 |
WARE | -0.047 |
Token | R |
Token ID | 49 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
| +0.010 |
Zi | +0.010 |
Catalan | +0.010 |
Niet | +0.010 |
wikipedia | +0.009 |
Neg logit contributions | |
WORK | -0.008 |
liv | -0.006 |
awatts | -0.006 |
WARE | -0.006 |
aturdays | -0.006 |
Token | OB |
Token ID | 9864 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
| +0.096 |
Zi | +0.092 |
Catalan | +0.091 |
Niet | +0.091 |
wikipedia | +0.087 |
Neg logit contributions | |
WORK | -0.111 |
liv | -0.090 |
awatts | -0.086 |
WARE | -0.086 |
Appl | -0.082 |
Token | ERT |
Token ID | 17395 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
WORK | +1.232 |
liv | +0.989 |
WARE | +0.941 |
awatts | +0.936 |
Appl | +0.912 |
Neg logit contributions | |
| -1.109 |
Zi | -1.075 |
Catalan | -1.069 |
Niet | -1.068 |
wikipedia | -1.018 |
Token | P |
Token ID | 350 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
| +0.011 |
Zi | +0.010 |
Catalan | +0.010 |
Niet | +0.010 |
Moss | +0.010 |
Neg logit contributions | |
WORK | -0.020 |
liv | -0.017 |
awatts | -0.017 |
WARE | -0.016 |
aturdays | -0.016 |
Token | OR |
Token ID | 1581 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
| +0.075 |
Zi | +0.073 |
Catalan | +0.072 |
Niet | +0.072 |
wikipedia | +0.069 |
Neg logit contributions | |
WORK | -0.078 |
liv | -0.062 |
awatts | -0.059 |
WARE | -0.059 |
Appl | -0.057 |
Token | ATH |
Token ID | 12599 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
WORK | +0.022 |
liv | +0.016 |
awatts | +0.015 |
WARE | +0.015 |
aturdays | +0.014 |
Neg logit contributions | |
| -0.037 |
Zi | -0.036 |
Catalan | -0.036 |
Niet | -0.036 |
Moss | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
WORK | +0.478 |
liv | +0.383 |
awatts | +0.367 |
WARE | +0.361 |
aturdays | +0.352 |
Neg logit contributions | |
| -0.473 |
Zi | -0.460 |
Catalan | -0.456 |
Niet | -0.454 |
wikipedia | -0.432 |
Token | Jr |
Token ID | 7504 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
WORK | +0.318 |
liv | +0.263 |
awatts | +0.252 |
WARE | +0.251 |
Appl | +0.244 |
Neg logit contributions | |
| -0.229 |
Zi | -0.222 |
Catalan | -0.220 |
Niet | -0.219 |
wikipedia | -0.208 |
Token | .' |
Token ID | 2637 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
| +0.080 |
Zi | +0.078 |
Catalan | +0.078 |
Niet | +0.077 |
wikipedia | +0.074 |
Neg logit contributions | |
WORK | -0.083 |
liv | -0.066 |
awatts | -0.063 |
WARE | -0.063 |
aturdays | -0.061 |
Token | s |
Token ID | 82 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
| +0.110 |
Zi | +0.106 |
Catalan | +0.106 |
Niet | +0.105 |
wikipedia | +0.101 |
Neg logit contributions | |
WORK | -0.113 |
liv | -0.090 |
WARE | -0.086 |
awatts | -0.086 |
aturdays | -0.084 |
Token | first |
Token ID | 717 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
| +0.035 |
Zi | +0.034 |
Catalan | +0.034 |
Niet | +0.034 |
Moss | +0.032 |
Neg logit contributions | |
WORK | -0.037 |
liv | -0.030 |
awatts | -0.028 |
WARE | -0.028 |
aturdays | -0.027 |
Token | four |
Token ID | 1440 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
| +0.034 |
Zi | +0.032 |
Catalan | +0.032 |
Niet | +0.032 |
wikipedia | +0.031 |
Neg logit contributions | |
WORK | -0.038 |
liv | -0.030 |
awatts | -0.029 |
WARE | -0.029 |
aturdays | -0.028 |
Token | singles |
Token ID | 25287 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
| +0.262 |
Zi | +0.254 |
Catalan | +0.252 |
Niet | +0.252 |
Moss | +0.239 |
Neg logit contributions | |
WORK | -0.274 |
liv | -0.217 |
awatts | -0.209 |
WARE | -0.209 |
aturdays | -0.201 |
Token | on |
Token ID | 319 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
WORK | +1.158 |
liv | +0.929 |
awatts | +0.879 |
WARE | +0.874 |
Appl | +0.848 |
Neg logit contributions | |
| -1.141 |
Zi | -1.109 |
Catalan | -1.105 |
Niet | -1.100 |
wikipedia | -1.057 |
Token | S |
Token ID | 311 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
WORK | +0.047 |
liv | +0.038 |
awatts | +0.037 |
WARE | +0.037 |
aturdays | +0.035 |
Neg logit contributions | |
| -0.040 |
Zi | -0.039 |
Catalan | -0.038 |
Niet | -0.038 |
wikipedia | -0.036 |
Token | ST |
Token ID | 2257 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
WORK | +0.155 |
liv | +0.125 |
awatts | +0.120 |
WARE | +0.119 |
aturdays | +0.115 |
Neg logit contributions | |
| -0.136 |
Zi | -0.132 |
Catalan | -0.131 |
Niet | -0.131 |
wikipedia | -0.124 |
Token | and |
Token ID | 290 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
WORK | +0.164 |
liv | +0.132 |
awatts | +0.129 |
WARE | +0.127 |
aturdays | +0.123 |
Neg logit contributions | |
| -0.153 |
Zi | -0.147 |
Catalan | -0.147 |
Niet | -0.146 |
Moss | -0.139 |
Token | Hom |
Token ID | 8074 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
WORK | +0.031 |
liv | +0.025 |
awatts | +0.024 |
WARE | +0.024 |
aturdays | +0.023 |
Neg logit contributions | |
| -0.025 |
Zi | -0.024 |
Catalan | -0.024 |
Niet | -0.024 |
wikipedia | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
WORK | +0.047 |
liv | +0.038 |
WARE | +0.037 |
awatts | +0.037 |
aturdays | +0.036 |
Neg logit contributions | |
| -0.040 |
Zi | -0.038 |
Catalan | -0.038 |
Niet | -0.038 |
wikipedia | -0.036 |
Token | amazed |
Token ID | 24562 |
Feature activation | +7.49e-03 |
Pos logit contributions | |
WORK | +0.130 |
liv | +0.103 |
awatts | +0.099 |
WARE | +0.098 |
aturdays | +0.095 |
Neg logit contributions | |
| -0.130 |
Zi | -0.126 |
Catalan | -0.125 |
Niet | -0.125 |
wikipedia | -0.119 |
Token | to |
Token ID | 284 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
WORK | +0.006 |
liv | +0.005 |
awatts | +0.004 |
WARE | +0.004 |
Appl | +0.004 |
Neg logit contributions | |
| -0.006 |
Zi | -0.006 |
Catalan | -0.006 |
Niet | -0.006 |
wikipedia | -0.005 |
Token | see |
Token ID | 766 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
| +0.008 |
Zi | +0.008 |
Catalan | +0.008 |
Niet | +0.008 |
Moss | +0.008 |
Neg logit contributions | |
WORK | -0.009 |
liv | -0.007 |
awatts | -0.007 |
WARE | -0.007 |
aturdays | -0.007 |
Token | how |
Token ID | 703 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
| +0.079 |
Zi | +0.076 |
Catalan | +0.076 |
Niet | +0.076 |
wikipedia | +0.072 |
Neg logit contributions | |
WORK | -0.093 |
liv | -0.075 |
awatts | -0.072 |
WARE | -0.072 |
aturdays | -0.070 |
Token | wrong |
Token ID | 2642 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
| +0.012 |
Zi | +0.011 |
Catalan | +0.011 |
Niet | +0.011 |
wikipedia | +0.011 |
Neg logit contributions | |
WORK | -0.012 |
liv | -0.010 |
awatts | -0.009 |
WARE | -0.009 |
aturdays | -0.009 |
Token | I |
Token ID | 314 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
WORK | +1.156 |
liv | +0.926 |
WARE | +0.879 |
awatts | +0.877 |
Appl | +0.841 |
Neg logit contributions | |
| -1.107 |
Zi | -1.077 |
Niet | -1.077 |
Catalan | -1.076 |
wikipedia | -1.025 |
Token | was |
Token ID | 373 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
WORK | +0.227 |
liv | +0.186 |
awatts | +0.179 |
WARE | +0.178 |
aturdays | +0.173 |
Neg logit contributions | |
| -0.176 |
Zi | -0.170 |
Catalan | -0.169 |
Niet | -0.168 |
wikipedia | -0.159 |
Token | . |
Token ID | 13 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
WORK | +0.167 |
liv | +0.134 |
awatts | +0.129 |
WARE | +0.128 |
aturdays | +0.124 |
Neg logit contributions | |
| -0.151 |
Zi | -0.146 |
Catalan | -0.145 |
Niet | -0.145 |
wikipedia | -0.138 |
Token | The |
Token ID | 383 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
WORK | +0.033 |
liv | +0.027 |
awatts | +0.025 |
WARE | +0.025 |
aturdays | +0.024 |
Neg logit contributions | |
| -0.032 |
Zi | -0.031 |
Catalan | -0.031 |
Niet | -0.031 |
wikipedia | -0.029 |
Token | generosity |
Token ID | 31569 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
| +0.071 |
Zi | +0.068 |
Catalan | +0.068 |
Niet | +0.068 |
wikipedia | +0.064 |
Neg logit contributions | |
WORK | -0.080 |
liv | -0.065 |
awatts | -0.062 |
WARE | -0.062 |
aturdays | -0.060 |
Token | The |
Token ID | 383 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
| +0.064 |
Zi | +0.062 |
Catalan | +0.062 |
Niet | +0.062 |
wikipedia | +0.058 |
Neg logit contributions | |
WORK | -0.082 |
liv | -0.068 |
awatts | -0.064 |
WARE | -0.064 |
Appl | -0.063 |
Token | Red |
Token ID | 2297 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
WORK | +0.010 |
liv | +0.008 |
awatts | +0.008 |
WARE | +0.008 |
Appl | +0.008 |
Neg logit contributions | |
| -0.007 |
Zi | -0.007 |
Catalan | -0.007 |
Niet | -0.007 |
wikipedia | -0.006 |
Token | Tent |
Token ID | 30765 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
WORK | +0.435 |
liv | +0.356 |
awatts | +0.340 |
WARE | +0.338 |
Appl | +0.329 |
Neg logit contributions | |
| -0.352 |
Zi | -0.340 |
Niet | -0.338 |
Catalan | -0.338 |
wikipedia | -0.320 |
Token | : |
Token ID | 25 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
WORK | +0.028 |
liv | +0.022 |
awatts | +0.022 |
WARE | +0.021 |
Appl | +0.021 |
Neg logit contributions | |
| -0.025 |
Zi | -0.024 |
Catalan | -0.024 |
Niet | -0.024 |
wikipedia | -0.023 |
Token | A |
Token ID | 317 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
WORK | +0.020 |
liv | +0.016 |
awatts | +0.016 |
WARE | +0.016 |
Appl | +0.015 |
Neg logit contributions | |
| -0.016 |
Zi | -0.016 |
Catalan | -0.016 |
Niet | -0.016 |
wikipedia | -0.015 |
Token | Novel |
Token ID | 24467 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
| +0.045 |
Zi | +0.044 |
Catalan | +0.044 |
Niet | +0.043 |
wikipedia | +0.041 |
Neg logit contributions | |
WORK | -0.061 |
liv | -0.050 |
awatts | -0.048 |
WARE | -0.048 |
Appl | -0.046 |
Token | ( |
Token ID | 357 |
Feature activation | +7.35e-02 |
Pos logit contributions | |
WORK | +1.191 |
liv | +0.959 |
awatts | +0.915 |
WARE | +0.911 |
Appl | +0.881 |
Neg logit contributions | |
| -1.084 |
Zi | -1.051 |
Catalan | -1.045 |
Niet | -1.042 |
wikipedia | -0.993 |
Token | New |
Token ID | 3791 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
WORK | +0.056 |
liv | +0.044 |
awatts | +0.042 |
WARE | +0.042 |
aturdays | +0.040 |
Neg logit contributions | |
| -0.058 |
Zi | -0.056 |
Catalan | -0.056 |
Niet | -0.056 |
wikipedia | -0.053 |
Token | York |
Token ID | 1971 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
WORK | +0.167 |
liv | +0.136 |
awatts | +0.130 |
WARE | +0.129 |
Appl | +0.126 |
Neg logit contributions | |
| -0.141 |
Zi | -0.137 |
Catalan | -0.136 |
Niet | -0.135 |
wikipedia | -0.129 |
Token | : |
Token ID | 25 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
WORK | +0.011 |
liv | +0.009 |
awatts | +0.009 |
WARE | +0.009 |
aturdays | +0.008 |
Neg logit contributions | |
| -0.010 |
Zi | -0.010 |
Catalan | -0.010 |
Niet | -0.009 |
Moss | -0.009 |
Token | St |
Token ID | 520 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
WORK | +0.024 |
liv | +0.020 |
awatts | +0.020 |
WARE | +0.020 |
Appl | +0.019 |
Neg logit contributions | |
| -0.015 |
Zi | -0.015 |
Catalan | -0.015 |
Niet | -0.014 |
wikipedia | -0.014 |
Token | respond |
Token ID | 3031 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
WORK | +0.064 |
liv | +0.048 |
awatts | +0.046 |
WARE | +0.046 |
aturdays | +0.044 |
Neg logit contributions | |
| -0.084 |
Zi | -0.082 |
Catalan | -0.081 |
Niet | -0.081 |
wikipedia | -0.077 |
Token | to |
Token ID | 284 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
WORK | +0.355 |
liv | +0.288 |
awatts | +0.275 |
WARE | +0.271 |
Appl | +0.264 |
Neg logit contributions | |
| -0.320 |
Zi | -0.313 |
Niet | -0.311 |
Catalan | -0.310 |
wikipedia | -0.296 |
Token | the |
Token ID | 262 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
WORK | +0.099 |
liv | +0.081 |
awatts | +0.078 |
WARE | +0.077 |
Appl | +0.075 |
Neg logit contributions | |
| -0.078 |
Zi | -0.076 |
Catalan | -0.075 |
Niet | -0.075 |
wikipedia | -0.071 |
Token | Deep |
Token ID | 10766 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
| +0.032 |
Zi | +0.031 |
Catalan | +0.031 |
Niet | +0.031 |
wikipedia | +0.029 |
Neg logit contributions | |
WORK | -0.032 |
liv | -0.025 |
awatts | -0.024 |
WARE | -0.024 |
aturdays | -0.023 |
Token | water |
Token ID | 7050 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
WORK | +0.162 |
liv | +0.139 |
awatts | +0.135 |
WARE | +0.135 |
aturdays | +0.132 |
Neg logit contributions | |
| -0.060 |
Zi | -0.056 |
Catalan | -0.056 |
Niet | -0.055 |
wikipedia | -0.050 |
Token | Horizon |
Token ID | 22776 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
| +0.089 |
Zi | +0.086 |
Catalan | +0.085 |
Niet | +0.084 |
wikipedia | +0.079 |
Neg logit contributions | |
WORK | -0.175 |
liv | -0.148 |
awatts | -0.143 |
WARE | -0.142 |
aturdays | -0.139 |
Token | oil |
Token ID | 3056 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
WORK | +1.023 |
liv | +0.795 |
awatts | +0.750 |
WARE | +0.743 |
Appl | +0.716 |
Neg logit contributions | |
| -1.234 |
Zi | -1.205 |
Catalan | -1.198 |
Niet | -1.197 |
wikipedia | -1.149 |
Token | spill |
Token ID | 19431 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
WORK | +0.236 |
liv | +0.176 |
awatts | +0.167 |
WARE | +0.166 |
aturdays | +0.158 |
Neg logit contributions | |
| -0.341 |
Zi | -0.332 |
Catalan | -0.330 |
Niet | -0.329 |
Moss | -0.316 |
Token | . |
Token ID | 13 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
WORK | +0.580 |
liv | +0.465 |
awatts | +0.442 |
WARE | +0.438 |
Appl | +0.424 |
Neg logit contributions | |
| -0.564 |
Zi | -0.550 |
Niet | -0.546 |
Catalan | -0.545 |
wikipedia | -0.522 |
Token | Shell |
Token ID | 17537 |
Feature activation | -2.09e-03 |
Pos logit contributions | |
WORK | +0.022 |
liv | +0.017 |
awatts | +0.016 |
WARE | +0.016 |
aturdays | +0.016 |
Neg logit contributions | |
| -0.022 |
Zi | -0.022 |
Catalan | -0.022 |
Niet | -0.022 |
wikipedia | -0.021 |
Token | is |
Token ID | 318 |
Feature activation | +8.68e-02 |
Pos logit contributions | |
| +0.001 |
Zi | +0.001 |
Catalan | +0.001 |
Niet | +0.001 |
wikipedia | +0.001 |
Neg logit contributions | |
WORK | -0.002 |
liv | -0.001 |
awatts | -0.001 |
WARE | -0.001 |
aturdays | -0.001 |
Token | Hide |
Token ID | 10415 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
| +0.042 |
Zi | +0.041 |
Catalan | +0.041 |
Niet | +0.041 |
wikipedia | +0.039 |
Neg logit contributions | |
WORK | -0.044 |
liv | -0.035 |
awatts | -0.033 |
WARE | -0.033 |
aturdays | -0.032 |
Token | Caption |
Token ID | 11260 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
| +0.091 |
Catalan | +0.088 |
Zi | +0.088 |
Niet | +0.086 |
Moss | +0.084 |
Neg logit contributions | |
WORK | -0.072 |
liv | -0.056 |
awatts | -0.054 |
WARE | -0.054 |
aturdays | -0.054 |
Token | 10 |
Token ID | 838 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
| +0.026 |
Zi | +0.025 |
Catalan | +0.025 |
Niet | +0.024 |
wikipedia | +0.023 |
Neg logit contributions | |
WORK | -0.030 |
liv | -0.024 |
aturdays | -0.024 |
WARE | -0.023 |
awatts | -0.023 |
Token | of |
Token ID | 286 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
| +0.111 |
Zi | +0.106 |
Catalan | +0.105 |
Niet | +0.100 |
wikipedia | +0.099 |
Neg logit contributions | |
WORK | -0.157 |
liv | -0.129 |
aturdays | -0.127 |
WARE | -0.126 |
awatts | -0.125 |
Token | 15 |
Token ID | 1315 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
| +0.153 |
Catalan | +0.144 |
Zi | +0.143 |
Niet | +0.139 |
wikipedia | +0.134 |
Neg logit contributions | |
WORK | -0.207 |
WARE | -0.167 |
aturdays | -0.167 |
liv | -0.166 |
awatts | -0.166 |
Token | Photos |
Token ID | 9434 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
| +0.204 |
Catalan | +0.197 |
Zi | +0.196 |
Niet | +0.191 |
Moss | +0.187 |
Neg logit contributions | |
WORK | -0.183 |
liv | -0.143 |
WARE | -0.141 |
aturdays | -0.140 |
awatts | -0.137 |
Token | : |
Token ID | 25 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
| +1.227 |
Zi | +1.167 |
Catalan | +1.151 |
Niet | +1.099 |
wikipedia | +1.088 |
Neg logit contributions | |
WORK | -1.588 |
aturdays | -1.315 |
liv | -1.302 |
WARE | -1.281 |
awatts | -1.270 |
Token | Carly |
Token ID | 35982 |
Feature activation | -7.56e-01 |
Pos logit contributions | |
WORK | +0.095 |
liv | +0.078 |
aturdays | +0.077 |
WARE | +0.077 |
awatts | +0.076 |
Neg logit contributions | |
| -0.068 |
Zi | -0.065 |
Catalan | -0.065 |
Niet | -0.063 |
Moss | -0.060 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
| +0.347 |
Zi | +0.329 |
Catalan | +0.327 |
Niet | +0.319 |
Moss | +0.298 |
Neg logit contributions | |
WORK | -0.832 |
liv | -0.708 |
awatts | -0.691 |
aturdays | -0.691 |
WARE | -0.689 |
Token | 's |
Token ID | 338 |
Feature activation | -4.85e-04 |
Pos logit contributions | |
| +0.136 |
Zi | +0.130 |
Catalan | +0.130 |
Niet | +0.125 |
Moss | +0.122 |
Neg logit contributions | |
WORK | -0.133 |
aturdays | -0.104 |
liv | -0.103 |
WARE | -0.103 |
awatts | -0.102 |
Token | political |
Token ID | 1964 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
| +0.000 |
Catalan | +0.000 |
Zi | +0.000 |
Niet | +0.000 |
wikipedia | +0.000 |
Neg logit contributions | |
WORK | -0.000 |
liv | -0.000 |
awatts | -0.000 |
WARE | -0.000 |
aturdays | -0.000 |
Token | — |
Token ID | 851 |
Feature activation | -4.82e-03 |
Pos logit contributions | |
WORK | +0.004 |
liv | +0.003 |
awatts | +0.003 |
WARE | +0.003 |
Appl | +0.003 |
Neg logit contributions | |
| -0.004 |
Niet | -0.004 |
Zi | -0.004 |
Catalan | -0.004 |
wikipedia | -0.004 |
Token | New |
Token ID | 968 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
| +0.003 |
Zi | +0.003 |
Niet | +0.003 |
Catalan | +0.003 |
wikipedia | +0.003 |
Neg logit contributions | |
WORK | -0.004 |
liv | -0.004 |
awatts | -0.003 |
WARE | -0.003 |
Appl | -0.003 |
Token | South |
Token ID | 2520 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
| +0.213 |
Zi | +0.207 |
Niet | +0.207 |
Catalan | +0.206 |
wikipedia | +0.194 |
Neg logit contributions | |
WORK | -0.298 |
liv | -0.245 |
awatts | -0.236 |
WARE | -0.231 |
Appl | -0.229 |
Token | Wales |
Token ID | 11769 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
| +0.006 |
Zi | +0.005 |
Catalan | +0.005 |
Niet | +0.005 |
wikipedia | +0.005 |
Neg logit contributions | |
WORK | -0.020 |
liv | -0.017 |
awatts | -0.017 |
WARE | -0.017 |
Appl | -0.016 |
Token | ( |
Token ID | 357 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
WORK | +0.382 |
liv | +0.301 |
awatts | +0.285 |
WARE | +0.283 |
Appl | +0.272 |
Neg logit contributions | |
| -0.417 |
Zi | -0.407 |
Niet | -0.405 |
Catalan | -0.405 |
wikipedia | -0.387 |
Token | NS |
Token ID | 8035 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
WORK | +0.017 |
liv | +0.013 |
awatts | +0.013 |
WARE | +0.013 |
aturdays | +0.012 |
Neg logit contributions | |
| -0.015 |
Zi | -0.015 |
Catalan | -0.015 |
Niet | -0.015 |
wikipedia | -0.014 |
Token | W |
Token ID | 54 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
| +1.353 |
Zi | +1.313 |
Catalan | +1.302 |
Niet | +1.287 |
Moss | +1.233 |
Neg logit contributions | |
WORK | -1.401 |
liv | -1.129 |
awatts | -1.076 |
WARE | -1.073 |
aturdays | -1.037 |
Token | ), |
Token ID | 828 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
WORK | +0.322 |
liv | +0.256 |
awatts | +0.244 |
WARE | +0.242 |
aturdays | +0.233 |
Neg logit contributions | |
| -0.327 |
Zi | -0.318 |
Catalan | -0.316 |
Niet | -0.315 |
wikipedia | -0.301 |
Token | Queensland |
Token ID | 21105 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
WORK | +0.025 |
liv | +0.021 |
awatts | +0.020 |
WARE | +0.020 |
Appl | +0.019 |
Neg logit contributions | |
| -0.018 |
Zi | -0.018 |
Catalan | -0.018 |
Niet | -0.018 |
wikipedia | -0.017 |
Token | ( |
Token ID | 357 |
Feature activation | +9.55e-04 |
Pos logit contributions | |
| +0.379 |
Zi | +0.369 |
Niet | +0.366 |
Catalan | +0.366 |
wikipedia | +0.350 |
Neg logit contributions | |
WORK | -0.397 |
liv | -0.318 |
awatts | -0.302 |
WARE | -0.299 |
Appl | -0.290 |
Token | QL |
Token ID | 9711 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
WORK | +0.001 |
liv | +0.001 |
awatts | +0.001 |
WARE | +0.001 |
aturdays | +0.001 |
Neg logit contributions | |
| -0.001 |
Zi | -0.001 |
Catalan | -0.001 |
Niet | -0.001 |
wikipedia | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
| +0.263 |
Zi | +0.253 |
Catalan | +0.252 |
Niet | +0.249 |
wikipedia | +0.237 |
Neg logit contributions | |
WORK | -0.311 |
liv | -0.249 |
WARE | -0.242 |
awatts | -0.241 |
aturdays | -0.235 |
Token | Donetsk |
Token ID | 31819 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
| +0.056 |
Zi | +0.054 |
Catalan | +0.054 |
Niet | +0.054 |
wikipedia | +0.051 |
Neg logit contributions | |
WORK | -0.058 |
liv | -0.047 |
awatts | -0.045 |
WARE | -0.044 |
aturdays | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
WORK | +0.023 |
liv | +0.018 |
awatts | +0.018 |
WARE | +0.017 |
aturdays | +0.017 |
Neg logit contributions | |
| -0.023 |
Zi | -0.023 |
Catalan | -0.023 |
Niet | -0.023 |
wikipedia | -0.022 |
Token | Lu |
Token ID | 6026 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
| +0.126 |
Zi | +0.121 |
Catalan | +0.120 |
Niet | +0.119 |
wikipedia | +0.112 |
Neg logit contributions | |
WORK | -0.221 |
liv | -0.185 |
awatts | -0.179 |
WARE | -0.178 |
aturdays | -0.173 |
Token | h |
Token ID | 71 |
Feature activation | +8.76e-02 |
Pos logit contributions | |
WORK | +0.091 |
liv | +0.074 |
awatts | +0.071 |
WARE | +0.071 |
aturdays | +0.069 |
Neg logit contributions | |
| -0.074 |
Zi | -0.071 |
Catalan | -0.071 |
Niet | -0.070 |
wikipedia | -0.067 |
Token | ansk |
Token ID | 34738 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
WORK | +0.098 |
liv | +0.084 |
awatts | +0.081 |
WARE | +0.081 |
aturdays | +0.079 |
Neg logit contributions | |
| -0.038 |
Zi | -0.036 |
Catalan | -0.035 |
Niet | -0.035 |
wikipedia | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | -8.16e-02 |
Pos logit contributions | |
| +1.330 |
Catalan | +1.283 |
Zi | +1.280 |
Niet | +1.267 |
wikipedia | +1.206 |
Neg logit contributions | |
WORK | -1.433 |
liv | -1.130 |
WARE | -1.099 |
awatts | -1.092 |
aturdays | -1.054 |
Token | however |
Token ID | 2158 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
| +0.063 |
Zi | +0.061 |
Catalan | +0.060 |
Niet | +0.060 |
wikipedia | +0.057 |
Neg logit contributions | |
WORK | -0.064 |
liv | -0.051 |
awatts | -0.048 |
WARE | -0.048 |
aturdays | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
| +0.100 |
Zi | +0.097 |
Niet | +0.097 |
Catalan | +0.096 |
wikipedia | +0.093 |
Neg logit contributions | |
WORK | -0.087 |
liv | -0.070 |
awatts | -0.065 |
WARE | -0.064 |
Appl | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
| +0.009 |
Zi | +0.009 |
Catalan | +0.009 |
Niet | +0.009 |
wikipedia | +0.008 |
Neg logit contributions | |
WORK | -0.011 |
liv | -0.009 |
awatts | -0.009 |
WARE | -0.009 |
aturdays | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
| +0.165 |
Zi | +0.160 |
Catalan | +0.159 |
Niet | +0.158 |
wikipedia | +0.150 |
Neg logit contributions | |
WORK | -0.197 |
liv | -0.159 |
awatts | -0.153 |
WARE | -0.152 |
aturdays | -0.147 |
Token |
|
Token ID | 198 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
| +0.329 |
Zi | +0.313 |
Catalan | +0.312 |
Niet | +0.300 |
wikipedia | +0.295 |
Neg logit contributions | |
WORK | -0.330 |
liv | -0.265 |
aturdays | -0.262 |
WARE | -0.258 |
awatts | -0.257 |
Token | Google |
Token ID | 11708 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
| +0.292 |
Zi | +0.281 |
Catalan | +0.280 |
Niet | +0.277 |
wikipedia | +0.269 |
Neg logit contributions | |
WORK | -0.313 |
liv | -0.257 |
awatts | -0.245 |
WARE | -0.241 |
aturdays | -0.241 |
Token | Scholar |
Token ID | 11713 |
Feature activation | -6.11e-01 |
Pos logit contributions | |
| +0.211 |
Zi | +0.196 |
Catalan | +0.196 |
Niet | +0.188 |
wikipedia | +0.180 |
Neg logit contributions | |
WORK | -0.416 |
liv | -0.353 |
awatts | -0.348 |
aturdays | -0.347 |
WARE | -0.346 |
Token | S |
Token ID | 311 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
| +0.386 |
Zi | +0.372 |
Catalan | +0.370 |
Niet | +0.369 |
wikipedia | +0.348 |
Neg logit contributions | |
WORK | -0.560 |
liv | -0.463 |
awatts | -0.445 |
WARE | -0.442 |
Appl | -0.431 |
Token | AGE |
Token ID | 11879 |
Feature activation | +6.97e-02 |
Pos logit contributions | |
WORK | +0.019 |
liv | +0.016 |
WARE | +0.015 |
awatts | +0.015 |
Appl | +0.014 |
Neg logit contributions | |
| -0.017 |
Zi | -0.016 |
Niet | -0.016 |
Catalan | -0.016 |
wikipedia | -0.015 |
Token | Journals |
Token ID | 48608 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
WORK | +0.072 |
liv | +0.061 |
awatts | +0.059 |
WARE | +0.058 |
Appl | +0.057 |
Neg logit contributions | |
| -0.036 |
Zi | -0.035 |
Catalan | -0.034 |
Niet | -0.034 |
wikipedia | -0.032 |
Token | | |
Token ID | 930 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
| +1.364 |
Zi | +1.325 |
Catalan | +1.318 |
Niet | +1.314 |
wikipedia | +1.256 |
Neg logit contributions | |
WORK | -1.277 |
liv | -1.008 |
awatts | -0.959 |
WARE | -0.954 |
aturdays | -0.917 |
Token | ISI |
Token ID | 22048 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
| +0.025 |
Zi | +0.023 |
Catalan | +0.022 |
Niet | +0.022 |
wikipedia | +0.018 |
Neg logit contributions | |
WORK | -0.142 |
liv | -0.125 |
awatts | -0.123 |
WARE | -0.122 |
aturdays | -0.120 |
Token |
|
Token ID | 198 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
| +1.095 |
Zi | +1.067 |
Catalan | +1.061 |
Niet | +1.060 |
wikipedia | +1.015 |
Neg logit contributions | |
WORK | -0.958 |
liv | -0.747 |
awatts | -0.710 |
WARE | -0.704 |
Appl | -0.676 |
Token |
|
Token ID | 198 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
| +0.296 |
Zi | +0.283 |
Catalan | +0.281 |
Niet | +0.272 |
Moss | +0.266 |
Neg logit contributions | |
WORK | -0.292 |
liv | -0.234 |
aturdays | -0.228 |
WARE | -0.227 |
awatts | -0.226 |
Token | R |
Token ID | 49 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
| +0.253 |
Zi | +0.245 |
Catalan | +0.244 |
Niet | +0.244 |
wikipedia | +0.231 |
Neg logit contributions | |
WORK | -0.295 |
liv | -0.237 |
awatts | -0.227 |
WARE | -0.227 |
Appl | -0.218 |
Token | Hide |
Token ID | 10415 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
WORK | +0.190 |
liv | +0.150 |
awatts | +0.143 |
WARE | +0.142 |
aturdays | +0.137 |
Neg logit contributions | |
| -0.200 |
Zi | -0.194 |
Catalan | -0.193 |
Niet | -0.193 |
wikipedia | -0.184 |
Token | Caption |
Token ID | 11260 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
| +0.098 |
Catalan | +0.094 |
Zi | +0.094 |
Niet | +0.092 |
Moss | +0.090 |
Neg logit contributions | |
WORK | -0.079 |
liv | -0.061 |
awatts | -0.059 |
aturdays | -0.058 |
WARE | -0.058 |
Token | 7 |
Token ID | 767 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
WORK | +0.024 |
liv | +0.020 |
WARE | +0.020 |
awatts | +0.020 |
aturdays | +0.019 |
Neg logit contributions | |
| -0.016 |
Zi | -0.015 |
Catalan | -0.015 |
Niet | -0.015 |
wikipedia | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
WORK | +0.031 |
liv | +0.025 |
aturdays | +0.025 |
WARE | +0.025 |
awatts | +0.025 |
Neg logit contributions | |
| -0.022 |
Zi | -0.021 |
Catalan | -0.021 |
Niet | -0.020 |
wikipedia | -0.020 |
Token | 15 |
Token ID | 1315 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
| +0.143 |
Catalan | +0.136 |
Zi | +0.135 |
Niet | +0.131 |
wikipedia | +0.126 |
Neg logit contributions | |
WORK | -0.193 |
WARE | -0.155 |
liv | -0.155 |
awatts | -0.155 |
aturdays | -0.155 |
Token | Photos |
Token ID | 9434 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
| +0.147 |
Catalan | +0.143 |
Zi | +0.142 |
Niet | +0.138 |
Moss | +0.135 |
Neg logit contributions | |
WORK | -0.132 |
liv | -0.103 |
WARE | -0.102 |
aturdays | -0.101 |
awatts | -0.099 |
Token | : |
Token ID | 25 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
| +1.145 |
Zi | +1.090 |
Catalan | +1.075 |
Niet | +1.030 |
wikipedia | +1.017 |
Neg logit contributions | |
WORK | -1.472 |
aturdays | -1.214 |
liv | -1.208 |
WARE | -1.187 |
awatts | -1.176 |
Token | Carly |
Token ID | 35982 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
WORK | +0.120 |
liv | +0.099 |
aturdays | +0.098 |
WARE | +0.097 |
awatts | +0.097 |
Neg logit contributions | |
| -0.086 |
Zi | -0.082 |
Catalan | -0.082 |
Niet | -0.080 |
Moss | -0.076 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | -8.70e-02 |
Pos logit contributions | |
| +0.314 |
Zi | +0.297 |
Catalan | +0.295 |
Niet | +0.288 |
Moss | +0.269 |
Neg logit contributions | |
WORK | -0.758 |
liv | -0.645 |
awatts | -0.630 |
aturdays | -0.630 |
WARE | -0.628 |
Token | 's |
Token ID | 338 |
Feature activation | +8.28e-02 |
Pos logit contributions | |
| +0.069 |
Zi | +0.066 |
Catalan | +0.066 |
Niet | +0.063 |
wikipedia | +0.062 |
Neg logit contributions | |
WORK | -0.067 |
aturdays | -0.053 |
liv | -0.052 |
WARE | -0.052 |
awatts | -0.052 |
Token | political |
Token ID | 1964 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
WORK | +0.074 |
liv | +0.060 |
awatts | +0.060 |
WARE | +0.060 |
aturdays | +0.059 |
Neg logit contributions | |
| -0.055 |
Catalan | -0.052 |
Zi | -0.052 |
Niet | -0.051 |
wikipedia | -0.049 |
Token | Hide |
Token ID | 10415 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
| +0.005 |
Zi | +0.005 |
Catalan | +0.005 |
Niet | +0.005 |
Moss | +0.005 |
Neg logit contributions | |
WORK | -0.006 |
liv | -0.005 |
awatts | -0.004 |
WARE | -0.004 |
aturdays | -0.004 |
Token | Caption |
Token ID | 11260 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
| +0.043 |
Catalan | +0.041 |
Zi | +0.041 |
Niet | +0.040 |
Moss | +0.039 |
Neg logit contributions | |
WORK | -0.034 |
liv | -0.026 |
awatts | -0.025 |
aturdays | -0.025 |
WARE | -0.025 |
Token | 9 |
Token ID | 860 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
WORK | +0.049 |
liv | +0.040 |
aturdays | +0.040 |
WARE | +0.040 |
awatts | +0.039 |
Neg logit contributions | |
| -0.035 |
Zi | -0.034 |
Catalan | -0.033 |
Niet | -0.032 |
Moss | -0.031 |
Token | of |
Token ID | 286 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
| +0.097 |
Zi | +0.093 |
Catalan | +0.092 |
Niet | +0.088 |
wikipedia | +0.087 |
Neg logit contributions | |
WORK | -0.138 |
liv | -0.113 |
aturdays | -0.112 |
WARE | -0.111 |
awatts | -0.110 |
Token | 15 |
Token ID | 1315 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
| +0.182 |
Catalan | +0.173 |
Zi | +0.171 |
Niet | +0.166 |
wikipedia | +0.161 |
Neg logit contributions | |
WORK | -0.248 |
WARE | -0.200 |
aturdays | -0.200 |
awatts | -0.199 |
liv | -0.199 |
Token | Photos |
Token ID | 9434 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
| +0.204 |
Catalan | +0.197 |
Zi | +0.196 |
Niet | +0.191 |
Moss | +0.187 |
Neg logit contributions | |
WORK | -0.183 |
liv | -0.143 |
WARE | -0.141 |
aturdays | -0.141 |
awatts | -0.138 |
Token | : |
Token ID | 25 |
Feature activation | +7.33e-02 |
Pos logit contributions | |
| +1.141 |
Zi | +1.085 |
Catalan | +1.070 |
Niet | +1.023 |
wikipedia | +1.013 |
Neg logit contributions | |
WORK | -1.468 |
aturdays | -1.212 |
liv | -1.204 |
WARE | -1.183 |
awatts | -1.173 |
Token | Carly |
Token ID | 35982 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
WORK | +0.067 |
liv | +0.055 |
aturdays | +0.054 |
WARE | +0.054 |
awatts | +0.054 |
Neg logit contributions | |
| -0.048 |
Zi | -0.046 |
Catalan | -0.046 |
Niet | -0.045 |
Moss | -0.042 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
| +0.336 |
Zi | +0.319 |
Catalan | +0.316 |
Niet | +0.309 |
Moss | +0.288 |
Neg logit contributions | |
WORK | -0.806 |
liv | -0.686 |
awatts | -0.670 |
aturdays | -0.669 |
WARE | -0.668 |
Token | 's |
Token ID | 338 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
| +0.114 |
Zi | +0.109 |
Catalan | +0.108 |
Niet | +0.105 |
wikipedia | +0.102 |
Neg logit contributions | |
WORK | -0.111 |
aturdays | -0.087 |
liv | -0.086 |
WARE | -0.086 |
awatts | -0.086 |
Token | political |
Token ID | 1964 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
WORK | +0.044 |
liv | +0.036 |
awatts | +0.036 |
WARE | +0.036 |
aturdays | +0.035 |
Neg logit contributions | |
| -0.033 |
Catalan | -0.031 |
Zi | -0.031 |
Niet | -0.031 |
wikipedia | -0.029 |
Token | Hide |
Token ID | 10415 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
WORK | +0.086 |
liv | +0.068 |
awatts | +0.065 |
WARE | +0.065 |
aturdays | +0.062 |
Neg logit contributions | |
| -0.088 |
Zi | -0.085 |
Catalan | -0.085 |
Niet | -0.084 |
wikipedia | -0.081 |
Token | Caption |
Token ID | 11260 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
| +0.031 |
Catalan | +0.030 |
Zi | +0.030 |
Niet | +0.030 |
Moss | +0.029 |
Neg logit contributions | |
WORK | -0.025 |
liv | -0.019 |
awatts | -0.019 |
WARE | -0.018 |
aturdays | -0.018 |
Token | 8 |
Token ID | 807 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
WORK | +0.016 |
liv | +0.013 |
aturdays | +0.013 |
WARE | +0.013 |
awatts | +0.013 |
Neg logit contributions | |
| -0.015 |
Zi | -0.015 |
Catalan | -0.015 |
Niet | -0.014 |
Moss | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
WORK | +0.115 |
liv | +0.094 |
aturdays | +0.092 |
WARE | +0.092 |
awatts | +0.091 |
Neg logit contributions | |
| -0.083 |
Zi | -0.079 |
Catalan | -0.078 |
Niet | -0.076 |
wikipedia | -0.074 |
Token | 15 |
Token ID | 1315 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
| +0.097 |
Catalan | +0.092 |
Zi | +0.091 |
Niet | +0.089 |
wikipedia | +0.086 |
Neg logit contributions | |
WORK | -0.131 |
WARE | -0.106 |
aturdays | -0.106 |
liv | -0.105 |
awatts | -0.105 |
Token | Photos |
Token ID | 9434 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
| +0.146 |
Catalan | +0.141 |
Zi | +0.140 |
Niet | +0.136 |
Moss | +0.134 |
Neg logit contributions | |
WORK | -0.131 |
liv | -0.102 |
WARE | -0.101 |
aturdays | -0.100 |
awatts | -0.098 |
Token | : |
Token ID | 25 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
| +1.136 |
Zi | +1.080 |
Catalan | +1.066 |
Niet | +1.020 |
wikipedia | +1.009 |
Neg logit contributions | |
WORK | -1.457 |
aturdays | -1.201 |
liv | -1.194 |
WARE | -1.174 |
awatts | -1.162 |
Token | Carly |
Token ID | 35982 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
WORK | +0.100 |
liv | +0.083 |
aturdays | +0.082 |
WARE | +0.081 |
awatts | +0.081 |
Neg logit contributions | |
| -0.072 |
Zi | -0.069 |
Catalan | -0.069 |
Niet | -0.067 |
Moss | -0.064 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | -2.12e-03 |
Pos logit contributions | |
| +0.318 |
Zi | +0.301 |
Catalan | +0.299 |
Niet | +0.292 |
Moss | +0.272 |
Neg logit contributions | |
WORK | -0.763 |
liv | -0.649 |
awatts | -0.634 |
aturdays | -0.633 |
WARE | -0.632 |
Token | 's |
Token ID | 338 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
| +0.002 |
Zi | +0.002 |
Catalan | +0.002 |
Niet | +0.002 |
wikipedia | +0.002 |
Neg logit contributions | |
WORK | -0.002 |
aturdays | -0.001 |
liv | -0.001 |
WARE | -0.001 |
awatts | -0.001 |
Token | political |
Token ID | 1964 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
WORK | +0.114 |
liv | +0.093 |
awatts | +0.092 |
WARE | +0.092 |
aturdays | +0.090 |
Neg logit contributions | |
| -0.084 |
Catalan | -0.080 |
Zi | -0.080 |
Niet | -0.078 |
wikipedia | -0.075 |
Token | ch |
Token ID | 442 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
| +0.117 |
Zi | +0.114 |
Catalan | +0.113 |
Niet | +0.112 |
Moss | +0.108 |
Neg logit contributions | |
WORK | -0.110 |
liv | -0.086 |
awatts | -0.083 |
WARE | -0.083 |
aturdays | -0.081 |
Token | mod |
Token ID | 4666 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
| +0.305 |
Zi | +0.294 |
Catalan | +0.293 |
Niet | +0.290 |
wikipedia | +0.278 |
Neg logit contributions | |
WORK | -0.325 |
liv | -0.258 |
WARE | -0.251 |
awatts | -0.247 |
aturdays | -0.241 |
Token | + |
Token ID | 1343 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
| +0.145 |
Zi | +0.140 |
Catalan | +0.139 |
Niet | +0.139 |
wikipedia | +0.130 |
Neg logit contributions | |
WORK | -0.230 |
liv | -0.192 |
awatts | -0.185 |
WARE | -0.184 |
Appl | -0.179 |
Token | x |
Token ID | 87 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
| +0.353 |
Zi | +0.343 |
Catalan | +0.341 |
Niet | +0.340 |
wikipedia | +0.325 |
Neg logit contributions | |
WORK | -0.341 |
liv | -0.270 |
awatts | -0.257 |
WARE | -0.256 |
aturdays | -0.247 |
Token | / |
Token ID | 1220 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
| +0.259 |
Zi | +0.252 |
Catalan | +0.250 |
Niet | +0.250 |
wikipedia | +0.238 |
Neg logit contributions | |
WORK | -0.220 |
liv | -0.170 |
awatts | -0.163 |
WARE | -0.162 |
aturdays | -0.157 |
Token | etc |
Token ID | 14784 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
| +0.172 |
Zi | +0.168 |
Catalan | +0.167 |
Niet | +0.167 |
wikipedia | +0.160 |
Neg logit contributions | |
WORK | -0.121 |
liv | -0.091 |
awatts | -0.086 |
WARE | -0.085 |
Appl | -0.081 |
Token | / |
Token ID | 14 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
| +1.108 |
Zi | +1.051 |
Catalan | +1.039 |
Niet | +1.008 |
wikipedia | +0.983 |
Neg logit contributions | |
WORK | -1.468 |
liv | -1.199 |
aturdays | -1.187 |
WARE | -1.177 |
awatts | -1.166 |
Token | pm |
Token ID | 4426 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
| +0.399 |
Zi | +0.387 |
Catalan | +0.384 |
Niet | +0.383 |
wikipedia | +0.368 |
Neg logit contributions | |
WORK | -0.367 |
liv | -0.290 |
WARE | -0.276 |
awatts | -0.274 |
aturdays | -0.266 |
Token | / |
Token ID | 14 |
Feature activation | -4.41e-01 |
Pos logit contributions | |
WORK | +0.141 |
liv | +0.114 |
awatts | +0.110 |
WARE | +0.110 |
aturdays | +0.107 |
Neg logit contributions | |
| -0.121 |
Zi | -0.117 |
Catalan | -0.116 |
Niet | -0.115 |
wikipedia | -0.110 |
Token | power |
Token ID | 6477 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
| +0.438 |
Zi | +0.427 |
Catalan | +0.424 |
Niet | +0.424 |
wikipedia | +0.410 |
Neg logit contributions | |
WORK | -0.244 |
liv | -0.175 |
awatts | -0.162 |
WARE | -0.161 |
aturdays | -0.153 |
Token | . |
Token ID | 13 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
WORK | +0.120 |
liv | +0.095 |
awatts | +0.091 |
WARE | +0.090 |
Appl | +0.087 |
Neg logit contributions | |
| -0.116 |
Zi | -0.113 |
Niet | -0.112 |
Catalan | -0.112 |
wikipedia | -0.107 |
Token | Hide |
Token ID | 10415 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
| +0.056 |
Zi | +0.054 |
Catalan | +0.054 |
Niet | +0.054 |
wikipedia | +0.052 |
Neg logit contributions | |
WORK | -0.054 |
liv | -0.043 |
awatts | -0.041 |
WARE | -0.041 |
Appl | -0.039 |
Token | Caption |
Token ID | 11260 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
| +0.294 |
Zi | +0.287 |
Catalan | +0.286 |
Niet | +0.285 |
wikipedia | +0.274 |
Neg logit contributions | |
WORK | -0.215 |
liv | -0.162 |
awatts | -0.153 |
WARE | -0.153 |
aturdays | -0.146 |
Token | 4 |
Token ID | 604 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
WORK | +0.177 |
liv | +0.145 |
awatts | +0.139 |
WARE | +0.138 |
Appl | +0.134 |
Neg logit contributions | |
| -0.134 |
Zi | -0.130 |
Catalan | -0.129 |
Niet | -0.129 |
wikipedia | -0.122 |
Token | of |
Token ID | 286 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
| +0.041 |
Zi | +0.040 |
Catalan | +0.040 |
Niet | +0.039 |
wikipedia | +0.037 |
Neg logit contributions | |
WORK | -0.054 |
liv | -0.044 |
awatts | -0.043 |
WARE | -0.042 |
aturdays | -0.042 |
Token | 15 |
Token ID | 1315 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
| +0.153 |
Zi | +0.148 |
Catalan | +0.147 |
Niet | +0.146 |
wikipedia | +0.138 |
Neg logit contributions | |
WORK | -0.209 |
liv | -0.172 |
awatts | -0.165 |
WARE | -0.165 |
aturdays | -0.160 |
Token | Photos |
Token ID | 9434 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
| +0.212 |
Zi | +0.206 |
Catalan | +0.205 |
Niet | +0.205 |
wikipedia | +0.197 |
Neg logit contributions | |
WORK | -0.167 |
liv | -0.128 |
awatts | -0.121 |
WARE | -0.120 |
Appl | -0.116 |
Token | : |
Token ID | 25 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
| +1.142 |
Zi | +1.099 |
Catalan | +1.087 |
Niet | +1.052 |
Moss | +1.026 |
Neg logit contributions | |
WORK | -1.438 |
liv | -1.170 |
aturdays | -1.144 |
WARE | -1.142 |
awatts | -1.133 |
Token | Carly |
Token ID | 35982 |
Feature activation | -6.63e-01 |
Pos logit contributions | |
WORK | +0.019 |
liv | +0.015 |
awatts | +0.015 |
WARE | +0.014 |
aturdays | +0.014 |
Neg logit contributions | |
| -0.017 |
Zi | -0.017 |
Catalan | -0.017 |
Niet | -0.016 |
wikipedia | -0.016 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
| +0.332 |
Zi | +0.317 |
Catalan | +0.314 |
Niet | +0.313 |
wikipedia | +0.289 |
Neg logit contributions | |
WORK | -0.693 |
liv | -0.588 |
awatts | -0.570 |
WARE | -0.569 |
aturdays | -0.554 |
Token | 's |
Token ID | 338 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
WORK | +0.137 |
liv | +0.107 |
awatts | +0.104 |
WARE | +0.103 |
aturdays | +0.100 |
Neg logit contributions | |
| -0.138 |
Zi | -0.135 |
Catalan | -0.133 |
Niet | -0.133 |
Moss | -0.127 |
Token | political |
Token ID | 1964 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
WORK | +0.052 |
liv | +0.040 |
awatts | +0.038 |
WARE | +0.038 |
aturdays | +0.036 |
Neg logit contributions | |
| -0.060 |
Zi | -0.059 |
Catalan | -0.058 |
Niet | -0.058 |
wikipedia | -0.056 |
Token | be |
Token ID | 307 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
WORK | +0.017 |
liv | +0.014 |
awatts | +0.013 |
WARE | +0.013 |
Appl | +0.013 |
Neg logit contributions | |
| -0.016 |
Zi | -0.015 |
Catalan | -0.015 |
Niet | -0.015 |
wikipedia | -0.014 |
Token | thr |
Token ID | 5636 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
| +0.030 |
Zi | +0.029 |
Catalan | +0.029 |
Niet | +0.029 |
wikipedia | +0.027 |
Neg logit contributions | |
WORK | -0.036 |
liv | -0.029 |
awatts | -0.028 |
WARE | -0.028 |
aturdays | -0.027 |
Token | ongs |
Token ID | 28079 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
| +0.412 |
Zi | +0.400 |
Catalan | +0.397 |
Niet | +0.395 |
wikipedia | +0.380 |
Neg logit contributions | |
WORK | -0.358 |
liv | -0.279 |
WARE | -0.267 |
awatts | -0.262 |
aturdays | -0.252 |
Token | of |
Token ID | 286 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
| +0.393 |
Zi | +0.380 |
Catalan | +0.377 |
Niet | +0.375 |
wikipedia | +0.357 |
Neg logit contributions | |
WORK | -0.480 |
liv | -0.390 |
awatts | -0.375 |
WARE | -0.374 |
aturdays | -0.362 |
Token | ad |
Token ID | 512 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
| +0.029 |
Zi | +0.028 |
Catalan | +0.028 |
Niet | +0.028 |
wikipedia | +0.027 |
Neg logit contributions | |
WORK | -0.033 |
liv | -0.026 |
awatts | -0.025 |
WARE | -0.025 |
aturdays | -0.024 |
Token | oring |
Token ID | 3255 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
| +0.177 |
Zi | +0.169 |
Catalan | +0.168 |
Niet | +0.167 |
wikipedia | +0.155 |
Neg logit contributions | |
WORK | -0.369 |
liv | -0.313 |
awatts | -0.303 |
WARE | -0.302 |
aturdays | -0.294 |
Token | fans |
Token ID | 3296 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
| +1.242 |
Zi | +1.203 |
Catalan | +1.200 |
Niet | +1.198 |
Moss | +1.138 |
Neg logit contributions | |
WORK | -1.316 |
liv | -1.038 |
WARE | -1.012 |
awatts | -1.007 |
aturdays | -0.979 |
Token | wanting |
Token ID | 10291 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
| +0.124 |
Zi | +0.120 |
Catalan | +0.119 |
Niet | +0.118 |
wikipedia | +0.113 |
Neg logit contributions | |
WORK | -0.142 |
liv | -0.114 |
awatts | -0.110 |
WARE | -0.110 |
aturdays | -0.106 |
Token | his |
Token ID | 465 |
Feature activation | +6.42e-03 |
Pos logit contributions | |
WORK | +0.256 |
liv | +0.208 |
awatts | +0.199 |
WARE | +0.198 |
Appl | +0.191 |
Neg logit contributions | |
| -0.217 |
Zi | -0.210 |
Catalan | -0.209 |
Niet | -0.208 |
wikipedia | -0.198 |
Token | aut |
Token ID | 1960 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
WORK | +0.005 |
liv | +0.004 |
awatts | +0.004 |
WARE | +0.004 |
aturdays | +0.004 |
Neg logit contributions | |
| -0.005 |
Zi | -0.005 |
Catalan | -0.005 |
Niet | -0.005 |
Moss | -0.005 |
Token | ograph |
Token ID | 2384 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
| +0.306 |
Zi | +0.299 |
Catalan | +0.297 |
Niet | +0.297 |
wikipedia | +0.286 |
Neg logit contributions | |
WORK | -0.183 |
liv | -0.133 |
awatts | -0.124 |
WARE | -0.123 |
aturdays | -0.116 |
Token | Hide |
Token ID | 10415 |
Feature activation | +7.15e-03 |
Pos logit contributions | |
WORK | +0.226 |
liv | +0.179 |
awatts | +0.170 |
WARE | +0.169 |
aturdays | +0.163 |
Neg logit contributions | |
| -0.234 |
Zi | -0.228 |
Catalan | -0.227 |
Niet | -0.226 |
wikipedia | -0.216 |
Token | Caption |
Token ID | 11260 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
WORK | +0.005 |
liv | +0.004 |
awatts | +0.004 |
WARE | +0.004 |
aturdays | +0.004 |
Neg logit contributions | |
| -0.006 |
Catalan | -0.006 |
Zi | -0.006 |
Niet | -0.006 |
Moss | -0.006 |
Token | 6 |
Token ID | 718 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
WORK | +0.178 |
liv | +0.148 |
awatts | +0.144 |
WARE | +0.144 |
aturdays | +0.141 |
Neg logit contributions | |
| -0.113 |
Zi | -0.108 |
Catalan | -0.108 |
Niet | -0.105 |
wikipedia | -0.100 |
Token | of |
Token ID | 286 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
WORK | +0.136 |
liv | +0.112 |
WARE | +0.109 |
aturdays | +0.109 |
awatts | +0.108 |
Neg logit contributions | |
| -0.099 |
Zi | -0.095 |
Catalan | -0.094 |
Niet | -0.091 |
wikipedia | -0.089 |
Token | 15 |
Token ID | 1315 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
| +0.139 |
Catalan | +0.133 |
Zi | +0.131 |
Niet | +0.127 |
wikipedia | +0.123 |
Neg logit contributions | |
WORK | -0.191 |
WARE | -0.155 |
aturdays | -0.154 |
awatts | -0.154 |
liv | -0.154 |
Token | Photos |
Token ID | 9434 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
| +0.153 |
Catalan | +0.149 |
Zi | +0.148 |
Niet | +0.144 |
Moss | +0.141 |
Neg logit contributions | |
WORK | -0.138 |
liv | -0.108 |
WARE | -0.107 |
aturdays | -0.106 |
awatts | -0.104 |
Token | : |
Token ID | 25 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
| +1.120 |
Zi | +1.065 |
Catalan | +1.052 |
Niet | +1.007 |
wikipedia | +0.994 |
Neg logit contributions | |
WORK | -1.441 |
aturdays | -1.188 |
liv | -1.182 |
WARE | -1.163 |
awatts | -1.151 |
Token | Carly |
Token ID | 35982 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
WORK | +0.204 |
liv | +0.167 |
aturdays | +0.165 |
WARE | +0.165 |
awatts | +0.164 |
Neg logit contributions | |
| -0.146 |
Zi | -0.140 |
Catalan | -0.139 |
Niet | -0.136 |
Moss | -0.129 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
| +0.261 |
Zi | +0.247 |
Catalan | +0.245 |
Niet | +0.239 |
Moss | +0.223 |
Neg logit contributions | |
WORK | -0.629 |
liv | -0.535 |
awatts | -0.523 |
aturdays | -0.522 |
WARE | -0.521 |
Token | 's |
Token ID | 338 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
WORK | +0.040 |
liv | +0.031 |
aturdays | +0.031 |
WARE | +0.031 |
awatts | +0.031 |
Neg logit contributions | |
| -0.042 |
Zi | -0.040 |
Catalan | -0.040 |
Niet | -0.039 |
wikipedia | -0.038 |
Token | political |
Token ID | 1964 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
WORK | +0.120 |
liv | +0.098 |
awatts | +0.096 |
WARE | +0.096 |
aturdays | +0.094 |
Neg logit contributions | |
| -0.089 |
Catalan | -0.085 |
Zi | -0.085 |
Niet | -0.083 |
wikipedia | -0.079 |
Token | said |
Token ID | 531 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
| +0.001 |
Zi | +0.001 |
Niet | +0.001 |
Catalan | +0.001 |
wikipedia | +0.001 |
Neg logit contributions | |
WORK | -0.001 |
liv | -0.001 |
awatts | -0.001 |
WARE | -0.001 |
Appl | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -3.00e-04 |
Pos logit contributions | |
| +0.142 |
Zi | +0.138 |
Catalan | +0.137 |
Niet | +0.137 |
wikipedia | +0.131 |
Neg logit contributions | |
WORK | -0.147 |
liv | -0.118 |
awatts | -0.112 |
WARE | -0.112 |
Appl | -0.108 |
Token | Although |
Token ID | 4900 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
| +0.000 |
Zi | +0.000 |
Catalan | +0.000 |
Niet | +0.000 |
wikipedia | +0.000 |
Neg logit contributions | |
WORK | -0.000 |
liv | -0.000 |
awatts | -0.000 |
WARE | -0.000 |
Appl | -0.000 |
Token | most |
Token ID | 749 |
Feature activation | +4.43e-01 |
Pos logit contributions | |
WORK | +0.187 |
liv | +0.152 |
awatts | +0.145 |
WARE | +0.145 |
aturdays | +0.140 |
Neg logit contributions | |
| -0.162 |
Zi | -0.157 |
Catalan | -0.156 |
Niet | -0.156 |
wikipedia | -0.148 |
Token | suff |
Token ID | 3027 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
WORK | +0.352 |
liv | +0.283 |
awatts | +0.270 |
WARE | +0.268 |
Appl | +0.259 |
Neg logit contributions | |
| -0.332 |
Zi | -0.322 |
Catalan | -0.320 |
Niet | -0.319 |
wikipedia | -0.305 |
Token | erers |
Token ID | 19288 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
| +0.129 |
Zi | +0.120 |
Catalan | +0.118 |
Niet | +0.117 |
wikipedia | +0.103 |
Neg logit contributions | |
WORK | -0.506 |
liv | -0.441 |
WARE | -0.429 |
awatts | -0.429 |
aturdays | -0.420 |
Token | experience |
Token ID | 1998 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
| +1.205 |
Zi | +1.160 |
Catalan | +1.137 |
Niet | +1.127 |
Moss | +1.076 |
Neg logit contributions | |
WORK | -1.352 |
liv | -1.069 |
WARE | -1.055 |
awatts | -1.041 |
aturdays | -1.023 |
Token | normal |
Token ID | 3487 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
| +0.270 |
Zi | +0.261 |
Catalan | +0.259 |
Niet | +0.258 |
wikipedia | +0.244 |
Neg logit contributions | |
WORK | -0.311 |
liv | -0.250 |
awatts | -0.241 |
WARE | -0.240 |
aturdays | -0.231 |
Token | flu |
Token ID | 6562 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
WORK | +0.035 |
liv | +0.028 |
awatts | +0.027 |
WARE | +0.026 |
aturdays | +0.025 |
Neg logit contributions | |
| -0.038 |
Zi | -0.037 |
Catalan | -0.037 |
Niet | -0.037 |
wikipedia | -0.035 |
Token | symptoms |
Token ID | 7460 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
| +0.528 |
Zi | +0.513 |
Catalan | +0.510 |
Niet | +0.508 |
wikipedia | +0.486 |
Neg logit contributions | |
WORK | -0.511 |
liv | -0.405 |
awatts | -0.386 |
WARE | -0.384 |
aturdays | -0.369 |
Token | and |
Token ID | 290 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
| +0.337 |
Zi | +0.326 |
Catalan | +0.325 |
Niet | +0.324 |
wikipedia | +0.309 |
Neg logit contributions | |
WORK | -0.322 |
liv | -0.254 |
awatts | -0.243 |
WARE | -0.242 |
aturdays | -0.232 |
Token | � |
Token ID | 447 |
Feature activation | -5.92e-02 |
Pos logit contributions | |
| +0.034 |
Zi | +0.033 |
Catalan | +0.033 |
Niet | +0.033 |
wikipedia | +0.031 |
Neg logit contributions | |
WORK | -0.032 |
liv | -0.025 |
awatts | -0.024 |
WARE | -0.024 |
aturdays | -0.023 |
Token | � |
Token ID | 251 |
Feature activation | -7.65e-03 |
Pos logit contributions | |
| +0.045 |
Zi | +0.044 |
Catalan | +0.044 |
Niet | +0.043 |
wikipedia | +0.041 |
Neg logit contributions | |
WORK | -0.047 |
liv | -0.037 |
awatts | -0.035 |
WARE | -0.035 |
aturdays | -0.034 |
Token | � |
Token ID | 564 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
| +0.006 |
Zi | +0.006 |
Catalan | +0.006 |
Niet | +0.006 |
wikipedia | +0.006 |
Neg logit contributions | |
WORK | -0.006 |
liv | -0.004 |
awatts | -0.004 |
WARE | -0.004 |
aturdays | -0.004 |
Token | � |
Token ID | 250 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
WORK | +0.231 |
liv | +0.190 |
awatts | +0.183 |
WARE | +0.180 |
aturdays | +0.179 |
Neg logit contributions | |
| -0.177 |
Zi | -0.169 |
Catalan | -0.168 |
Niet | -0.165 |
wikipedia | -0.159 |
Token | Comm |
Token ID | 6935 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
| +0.166 |
Zi | +0.161 |
Catalan | +0.160 |
Niet | +0.160 |
wikipedia | +0.155 |
Neg logit contributions | |
WORK | -0.121 |
liv | -0.093 |
awatts | -0.088 |
WARE | -0.087 |
aturdays | -0.085 |
Token | ander |
Token ID | 4066 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
| +0.379 |
Zi | +0.366 |
Catalan | +0.363 |
Niet | +0.362 |
wikipedia | +0.342 |
Neg logit contributions | |
WORK | -0.521 |
liv | -0.429 |
awatts | -0.412 |
WARE | -0.411 |
aturdays | -0.398 |
Token | of |
Token ID | 286 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
| +1.174 |
Zi | +1.136 |
Catalan | +1.125 |
Niet | +1.122 |
wikipedia | +1.066 |
Neg logit contributions | |
WORK | -1.333 |
liv | -1.073 |
awatts | -1.033 |
WARE | -1.030 |
aturdays | -0.995 |
Token | the |
Token ID | 262 |
Feature activation | -9.66e-02 |
Pos logit contributions | |
| +0.198 |
Zi | +0.192 |
Catalan | +0.191 |
Niet | +0.190 |
wikipedia | +0.180 |
Neg logit contributions | |
WORK | -0.250 |
liv | -0.204 |
awatts | -0.196 |
WARE | -0.195 |
Appl | -0.189 |
Token | On |
Token ID | 1550 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
| +0.067 |
Zi | +0.065 |
Catalan | +0.064 |
Niet | +0.064 |
wikipedia | +0.061 |
Neg logit contributions | |
WORK | -0.082 |
liv | -0.067 |
awatts | -0.064 |
WARE | -0.064 |
Appl | -0.062 |
Token | mits |
Token ID | 24883 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
| +0.342 |
Zi | +0.331 |
Catalan | +0.330 |
Niet | +0.329 |
wikipedia | +0.314 |
Neg logit contributions | |
WORK | -0.352 |
liv | -0.281 |
awatts | -0.268 |
WARE | -0.266 |
Appl | -0.257 |
Token | uk |
Token ID | 2724 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
| +0.339 |
Zi | +0.328 |
Catalan | +0.325 |
Niet | +0.324 |
wikipedia | +0.309 |
Neg logit contributions | |
WORK | -0.364 |
liv | -0.291 |
WARE | -0.278 |
awatts | -0.278 |
aturdays | -0.268 |
Token | Hide |
Token ID | 10415 |
Feature activation | -7.90e-03 |
Pos logit contributions | |
WORK | +0.238 |
liv | +0.189 |
awatts | +0.180 |
WARE | +0.179 |
aturdays | +0.172 |
Neg logit contributions | |
| -0.246 |
Zi | -0.239 |
Catalan | -0.237 |
Niet | -0.237 |
wikipedia | -0.226 |
Token | Caption |
Token ID | 11260 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
| +0.007 |
Catalan | +0.007 |
Zi | +0.007 |
Niet | +0.006 |
Moss | +0.006 |
Neg logit contributions | |
WORK | -0.005 |
liv | -0.004 |
aturdays | -0.004 |
awatts | -0.004 |
WARE | -0.004 |
Token | 5 |
Token ID | 642 |
Feature activation | +9.34e-02 |
Pos logit contributions | |
WORK | +0.192 |
liv | +0.159 |
awatts | +0.153 |
WARE | +0.153 |
aturdays | +0.149 |
Neg logit contributions | |
| -0.128 |
Zi | -0.123 |
Catalan | -0.122 |
Niet | -0.122 |
wikipedia | -0.115 |
Token | of |
Token ID | 286 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
WORK | +0.084 |
liv | +0.069 |
aturdays | +0.067 |
WARE | +0.067 |
awatts | +0.067 |
Neg logit contributions | |
| -0.061 |
Zi | -0.059 |
Catalan | -0.058 |
Niet | -0.056 |
wikipedia | -0.055 |
Token | 15 |
Token ID | 1315 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
| +0.083 |
Catalan | +0.079 |
Zi | +0.078 |
Niet | +0.076 |
wikipedia | +0.073 |
Neg logit contributions | |
WORK | -0.114 |
WARE | -0.092 |
aturdays | -0.092 |
awatts | -0.091 |
liv | -0.091 |
Token | Photos |
Token ID | 9434 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
| +0.112 |
Catalan | +0.109 |
Zi | +0.108 |
Niet | +0.106 |
Moss | +0.103 |
Neg logit contributions | |
WORK | -0.101 |
liv | -0.079 |
WARE | -0.078 |
aturdays | -0.077 |
awatts | -0.076 |
Token | : |
Token ID | 25 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
| +1.104 |
Zi | +1.047 |
Catalan | +1.036 |
Niet | +0.989 |
wikipedia | +0.978 |
Neg logit contributions | |
WORK | -1.418 |
aturdays | -1.170 |
liv | -1.163 |
WARE | -1.144 |
awatts | -1.132 |
Token | Carly |
Token ID | 35982 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
WORK | +0.254 |
liv | +0.208 |
awatts | +0.204 |
WARE | +0.204 |
aturdays | +0.203 |
Neg logit contributions | |
| -0.185 |
Zi | -0.178 |
Catalan | -0.176 |
Niet | -0.174 |
Moss | -0.165 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
| +0.298 |
Zi | +0.282 |
Catalan | +0.279 |
Niet | +0.274 |
Moss | +0.256 |
Neg logit contributions | |
WORK | -0.735 |
liv | -0.621 |
awatts | -0.612 |
WARE | -0.609 |
aturdays | -0.609 |
Token | 's |
Token ID | 338 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
WORK | +0.018 |
liv | +0.014 |
WARE | +0.014 |
awatts | +0.014 |
aturdays | +0.014 |
Neg logit contributions | |
| -0.018 |
Zi | -0.018 |
Catalan | -0.018 |
Niet | -0.017 |
wikipedia | -0.017 |
Token | political |
Token ID | 1964 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
WORK | +0.192 |
liv | +0.156 |
awatts | +0.154 |
WARE | +0.154 |
aturdays | +0.151 |
Neg logit contributions | |
| -0.142 |
Catalan | -0.136 |
Zi | -0.135 |
Niet | -0.133 |
wikipedia | -0.127 |
Token | Pelosi |
Token ID | 29611 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
WORK | +0.137 |
liv | +0.113 |
awatts | +0.109 |
WARE | +0.109 |
aturdays | +0.106 |
Neg logit contributions | |
| -0.099 |
Zi | -0.096 |
Catalan | -0.095 |
Niet | -0.095 |
Moss | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
WORK | +0.201 |
liv | +0.160 |
awatts | +0.153 |
WARE | +0.152 |
Appl | +0.146 |
Neg logit contributions | |
| -0.195 |
Zi | -0.190 |
Catalan | -0.189 |
Niet | -0.189 |
wikipedia | -0.180 |
Token | producing |
Token ID | 9194 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
| +0.114 |
Zi | +0.110 |
Catalan | +0.110 |
Niet | +0.109 |
wikipedia | +0.104 |
Neg logit contributions | |
WORK | -0.131 |
liv | -0.105 |
awatts | -0.101 |
WARE | -0.101 |
aturdays | -0.097 |
Token | a |
Token ID | 257 |
Feature activation | -9.14e-02 |
Pos logit contributions | |
| +0.245 |
Zi | +0.237 |
Catalan | +0.235 |
Niet | +0.234 |
wikipedia | +0.222 |
Neg logit contributions | |
WORK | -0.315 |
liv | -0.258 |
awatts | -0.248 |
WARE | -0.247 |
aturdays | -0.239 |
Token | miss |
Token ID | 2051 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
| +0.068 |
Zi | +0.066 |
Catalan | +0.065 |
Niet | +0.065 |
wikipedia | +0.062 |
Neg logit contributions | |
WORK | -0.074 |
liv | -0.059 |
awatts | -0.056 |
WARE | -0.056 |
Appl | -0.054 |
Token | hap |
Token ID | 45897 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
| +0.555 |
Zi | +0.538 |
Catalan | +0.535 |
Niet | +0.533 |
wikipedia | +0.510 |
Neg logit contributions | |
WORK | -0.540 |
liv | -0.427 |
awatts | -0.407 |
WARE | -0.405 |
aturdays | -0.390 |
Token | en |
Token ID | 268 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
| +1.333 |
Zi | +1.295 |
Catalan | +1.288 |
Niet | +1.273 |
Moss | +1.237 |
Neg logit contributions | |
WORK | -1.117 |
liv | -0.879 |
awatts | -0.833 |
WARE | -0.824 |
aturdays | -0.807 |
Token | mass |
Token ID | 2347 |
Feature activation | -9.87e-03 |
Pos logit contributions | |
| +0.101 |
Zi | +0.098 |
Catalan | +0.098 |
Niet | +0.097 |
wikipedia | +0.093 |
Neg logit contributions | |
WORK | -0.102 |
liv | -0.081 |
awatts | -0.077 |
WARE | -0.077 |
aturdays | -0.074 |
Token | of |
Token ID | 286 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
| +0.007 |
Zi | +0.007 |
Catalan | +0.007 |
Niet | +0.007 |
wikipedia | +0.006 |
Neg logit contributions | |
WORK | -0.008 |
liv | -0.007 |
awatts | -0.006 |
WARE | -0.006 |
Appl | -0.006 |
Token | special |
Token ID | 2041 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
| +0.141 |
Zi | +0.137 |
Catalan | +0.136 |
Niet | +0.135 |
wikipedia | +0.129 |
Neg logit contributions | |
WORK | -0.153 |
liv | -0.123 |
awatts | -0.118 |
WARE | -0.117 |
aturdays | -0.113 |
Token | - |
Token ID | 12 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
| +0.402 |
Zi | +0.389 |
Catalan | +0.388 |
Niet | +0.384 |
wikipedia | +0.364 |
Neg logit contributions | |
WORK | -0.466 |
liv | -0.373 |
awatts | -0.361 |
WARE | -0.359 |
aturdays | -0.347 |
Token | Hide |
Token ID | 10415 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
WORK | +0.065 |
liv | +0.052 |
awatts | +0.049 |
WARE | +0.049 |
aturdays | +0.047 |
Neg logit contributions | |
| -0.067 |
Zi | -0.065 |
Catalan | -0.065 |
Niet | -0.065 |
wikipedia | -0.062 |
Token | Caption |
Token ID | 11260 |
Feature activation | +8.66e-02 |
Pos logit contributions | |
WORK | +0.037 |
liv | +0.029 |
awatts | +0.028 |
WARE | +0.028 |
aturdays | +0.027 |
Neg logit contributions | |
| -0.047 |
Catalan | -0.045 |
Zi | -0.045 |
Niet | -0.044 |
Moss | -0.043 |
Token | 12 |
Token ID | 1105 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
WORK | +0.084 |
liv | +0.070 |
WARE | +0.068 |
awatts | +0.068 |
aturdays | +0.067 |
Neg logit contributions | |
| -0.051 |
Zi | -0.049 |
Catalan | -0.049 |
Niet | -0.047 |
wikipedia | -0.045 |
Token | of |
Token ID | 286 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
WORK | +0.201 |
liv | +0.165 |
WARE | +0.161 |
aturdays | +0.161 |
awatts | +0.160 |
Neg logit contributions | |
| -0.146 |
Zi | -0.139 |
Catalan | -0.138 |
Niet | -0.133 |
wikipedia | -0.131 |
Token | 15 |
Token ID | 1315 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
| +0.190 |
Catalan | +0.179 |
Zi | +0.177 |
Niet | +0.172 |
wikipedia | +0.167 |
Neg logit contributions | |
WORK | -0.259 |
WARE | -0.209 |
aturdays | -0.209 |
liv | -0.208 |
awatts | -0.208 |
Token | Photos |
Token ID | 9434 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
| +0.161 |
Catalan | +0.156 |
Zi | +0.156 |
Niet | +0.152 |
Moss | +0.148 |
Neg logit contributions | |
WORK | -0.144 |
liv | -0.113 |
WARE | -0.111 |
aturdays | -0.110 |
awatts | -0.108 |
Token | : |
Token ID | 25 |
Feature activation | +7.70e-02 |
Pos logit contributions | |
| +1.087 |
Zi | +1.036 |
Catalan | +1.022 |
Niet | +0.981 |
wikipedia | +0.966 |
Neg logit contributions | |
WORK | -1.387 |
aturdays | -1.139 |
liv | -1.136 |
WARE | -1.115 |
awatts | -1.105 |
Token | Carly |
Token ID | 35982 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
WORK | +0.070 |
liv | +0.058 |
aturdays | +0.057 |
WARE | +0.057 |
awatts | +0.056 |
Neg logit contributions | |
| -0.050 |
Zi | -0.048 |
Catalan | -0.048 |
Niet | -0.047 |
Moss | -0.045 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
| +0.269 |
Zi | +0.255 |
Catalan | +0.253 |
Niet | +0.247 |
Moss | +0.231 |
Neg logit contributions | |
WORK | -0.636 |
liv | -0.541 |
awatts | -0.528 |
WARE | -0.527 |
aturdays | -0.527 |
Token | 's |
Token ID | 338 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
WORK | +0.043 |
aturdays | +0.034 |
liv | +0.034 |
WARE | +0.034 |
awatts | +0.033 |
Neg logit contributions | |
| -0.044 |
Zi | -0.043 |
Catalan | -0.042 |
Niet | -0.041 |
Moss | -0.040 |
Token | political |
Token ID | 1964 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
WORK | +0.034 |
liv | +0.027 |
awatts | +0.027 |
WARE | +0.027 |
aturdays | +0.026 |
Neg logit contributions | |
| -0.025 |
Catalan | -0.024 |
Zi | -0.024 |
Niet | -0.023 |
wikipedia | -0.022 |
Token | Hide |
Token ID | 10415 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
WORK | +0.117 |
liv | +0.092 |
awatts | +0.088 |
WARE | +0.088 |
aturdays | +0.084 |
Neg logit contributions | |
| -0.122 |
Zi | -0.118 |
Catalan | -0.118 |
Niet | -0.117 |
wikipedia | -0.112 |
Token | Caption |
Token ID | 11260 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
WORK | +0.064 |
liv | +0.050 |
awatts | +0.047 |
WARE | +0.047 |
aturdays | +0.047 |
Neg logit contributions | |
| -0.082 |
Catalan | -0.079 |
Zi | -0.079 |
Niet | -0.077 |
Moss | -0.076 |
Token | 13 |
Token ID | 1511 |
Feature activation | -7.84e-02 |
Pos logit contributions | |
WORK | +0.124 |
liv | +0.102 |
awatts | +0.099 |
WARE | +0.099 |
aturdays | +0.096 |
Neg logit contributions | |
| -0.089 |
Zi | -0.086 |
Catalan | -0.085 |
Niet | -0.084 |
wikipedia | -0.080 |
Token | of |
Token ID | 286 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
| +0.051 |
Zi | +0.049 |
Catalan | +0.048 |
Niet | +0.046 |
wikipedia | +0.046 |
Neg logit contributions | |
WORK | -0.071 |
liv | -0.059 |
aturdays | -0.057 |
WARE | -0.057 |
awatts | -0.057 |
Token | 15 |
Token ID | 1315 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
| +0.173 |
Catalan | +0.163 |
Zi | +0.162 |
Niet | +0.157 |
wikipedia | +0.153 |
Neg logit contributions | |
WORK | -0.232 |
liv | -0.187 |
aturdays | -0.187 |
WARE | -0.187 |
awatts | -0.186 |
Token | Photos |
Token ID | 9434 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
| +0.195 |
Catalan | +0.189 |
Zi | +0.188 |
Niet | +0.184 |
Moss | +0.179 |
Neg logit contributions | |
WORK | -0.174 |
liv | -0.136 |
WARE | -0.133 |
aturdays | -0.132 |
awatts | -0.131 |
Token | : |
Token ID | 25 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
| +1.083 |
Zi | +1.033 |
Catalan | +1.019 |
Niet | +0.978 |
wikipedia | +0.963 |
Neg logit contributions | |
WORK | -1.383 |
aturdays | -1.137 |
liv | -1.133 |
WARE | -1.112 |
awatts | -1.102 |
Token | Carly |
Token ID | 35982 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
WORK | +0.060 |
liv | +0.049 |
aturdays | +0.048 |
WARE | +0.048 |
awatts | +0.048 |
Neg logit contributions | |
| -0.043 |
Zi | -0.041 |
Catalan | -0.041 |
Niet | -0.040 |
Moss | -0.038 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
| +0.228 |
Zi | +0.217 |
Catalan | +0.215 |
Niet | +0.211 |
Moss | +0.196 |
Neg logit contributions | |
WORK | -0.538 |
liv | -0.458 |
awatts | -0.447 |
WARE | -0.445 |
aturdays | -0.445 |
Token | 's |
Token ID | 338 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
WORK | +0.104 |
aturdays | +0.081 |
liv | +0.081 |
WARE | +0.080 |
awatts | +0.080 |
Neg logit contributions | |
| -0.107 |
Zi | -0.103 |
Catalan | -0.102 |
Niet | -0.099 |
Moss | -0.097 |
Token | political |
Token ID | 1964 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
WORK | +0.030 |
liv | +0.025 |
awatts | +0.024 |
WARE | +0.024 |
aturdays | +0.024 |
Neg logit contributions | |
| -0.022 |
Catalan | -0.021 |
Zi | -0.021 |
Niet | -0.021 |
wikipedia | -0.020 |
Token | be |
Token ID | 307 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
| +0.193 |
Zi | +0.186 |
Catalan | +0.185 |
Niet | +0.183 |
wikipedia | +0.175 |
Neg logit contributions | |
WORK | -0.200 |
liv | -0.159 |
WARE | -0.153 |
awatts | -0.152 |
aturdays | -0.148 |
Token | looking |
Token ID | 2045 |
Feature activation | +6.94e-03 |
Pos logit contributions | |
| +0.091 |
Zi | +0.088 |
Catalan | +0.088 |
Niet | +0.087 |
wikipedia | +0.083 |
Neg logit contributions | |
WORK | -0.096 |
liv | -0.077 |
awatts | -0.074 |
WARE | -0.073 |
aturdays | -0.071 |
Token | to |
Token ID | 284 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
WORK | +0.005 |
liv | +0.004 |
awatts | +0.004 |
WARE | +0.004 |
aturdays | +0.004 |
Neg logit contributions | |
| -0.006 |
Zi | -0.006 |
Catalan | -0.006 |
Niet | -0.005 |
wikipedia | -0.005 |
Token | g |
Token ID | 308 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
| +0.082 |
Zi | +0.079 |
Catalan | +0.079 |
Niet | +0.078 |
Moss | +0.075 |
Neg logit contributions | |
WORK | -0.076 |
liv | -0.059 |
awatts | -0.057 |
WARE | -0.056 |
aturdays | -0.054 |
Token | ank |
Token ID | 962 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
| +0.366 |
Zi | +0.355 |
Catalan | +0.352 |
Niet | +0.351 |
wikipedia | +0.336 |
Neg logit contributions | |
WORK | -0.366 |
liv | -0.290 |
WARE | -0.277 |
awatts | -0.276 |
Appl | -0.266 |
Token | immedi |
Token ID | 2729 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
| +0.145 |
Zi | +0.140 |
Catalan | +0.139 |
Niet | +0.139 |
wikipedia | +0.132 |
Neg logit contributions | |
WORK | -0.162 |
liv | -0.131 |
awatts | -0.125 |
WARE | -0.125 |
aturdays | -0.120 |
Token | at |
Token ID | 265 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
| +1.448 |
Zi | +1.408 |
Catalan | +1.399 |
Niet | +1.396 |
wikipedia | +1.343 |
Neg logit contributions | |
WORK | -0.995 |
liv | -0.742 |
WARE | -0.708 |
awatts | -0.687 |
Appl | -0.658 |
Token | ly |
Token ID | 306 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
WORK | +0.034 |
liv | +0.026 |
awatts | +0.025 |
WARE | +0.025 |
aturdays | +0.024 |
Neg logit contributions | |
| -0.041 |
Zi | -0.039 |
Catalan | -0.039 |
Niet | -0.039 |
wikipedia | -0.037 |
Token | after |
Token ID | 706 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
WORK | +0.154 |
liv | +0.122 |
awatts | +0.117 |
WARE | +0.116 |
aturdays | +0.112 |
Neg logit contributions | |
| -0.151 |
Zi | -0.147 |
Catalan | -0.146 |
Niet | -0.145 |
wikipedia | -0.139 |
Token | 6 |
Token ID | 718 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
| +0.083 |
Zi | +0.080 |
Catalan | +0.080 |
Niet | +0.079 |
wikipedia | +0.075 |
Neg logit contributions | |
WORK | -0.091 |
liv | -0.073 |
awatts | -0.070 |
WARE | -0.070 |
aturdays | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -8.76e-02 |
Pos logit contributions | |
| +0.017 |
Zi | +0.017 |
Catalan | +0.017 |
Niet | +0.017 |
wikipedia | +0.016 |
Neg logit contributions | |
WORK | -0.016 |
liv | -0.012 |
awatts | -0.012 |
WARE | -0.012 |
aturdays | -0.011 |
Token | Hide |
Token ID | 10415 |
Feature activation | +7.72e-02 |
Pos logit contributions | |
WORK | +0.017 |
liv | +0.013 |
awatts | +0.013 |
WARE | +0.013 |
aturdays | +0.012 |
Neg logit contributions | |
| -0.016 |
Zi | -0.016 |
Catalan | -0.016 |
Niet | -0.016 |
wikipedia | -0.015 |
Token | Caption |
Token ID | 11260 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
WORK | +0.053 |
liv | +0.041 |
awatts | +0.039 |
WARE | +0.039 |
aturdays | +0.039 |
Neg logit contributions | |
| -0.067 |
Catalan | -0.065 |
Zi | -0.065 |
Niet | -0.063 |
Moss | -0.062 |
Token | 11 |
Token ID | 1367 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
WORK | +0.106 |
liv | +0.086 |
awatts | +0.084 |
WARE | +0.084 |
aturdays | +0.083 |
Neg logit contributions | |
| -0.088 |
Zi | -0.085 |
Catalan | -0.084 |
Niet | -0.083 |
Moss | -0.080 |
Token | of |
Token ID | 286 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
WORK | +0.214 |
liv | +0.175 |
aturdays | +0.172 |
WARE | +0.172 |
awatts | +0.171 |
Neg logit contributions | |
| -0.154 |
Zi | -0.147 |
Catalan | -0.146 |
Niet | -0.141 |
wikipedia | -0.138 |
Token | 15 |
Token ID | 1315 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
| +0.137 |
Catalan | +0.130 |
Zi | +0.129 |
Niet | +0.125 |
wikipedia | +0.121 |
Neg logit contributions | |
WORK | -0.185 |
aturdays | -0.149 |
liv | -0.149 |
WARE | -0.149 |
awatts | -0.149 |
Token | Photos |
Token ID | 9434 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
| +0.153 |
Catalan | +0.148 |
Zi | +0.147 |
Niet | +0.143 |
Moss | +0.140 |
Neg logit contributions | |
WORK | -0.136 |
liv | -0.107 |
WARE | -0.105 |
aturdays | -0.104 |
awatts | -0.102 |
Token | : |
Token ID | 25 |
Feature activation | +9.70e-02 |
Pos logit contributions | |
| +1.076 |
Zi | +1.025 |
Catalan | +1.011 |
Niet | +0.970 |
wikipedia | +0.956 |
Neg logit contributions | |
WORK | -1.374 |
aturdays | -1.128 |
liv | -1.125 |
WARE | -1.105 |
awatts | -1.095 |
Token | Carly |
Token ID | 35982 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
WORK | +0.088 |
liv | +0.073 |
aturdays | +0.072 |
WARE | +0.071 |
awatts | +0.071 |
Neg logit contributions | |
| -0.063 |
Zi | -0.061 |
Catalan | -0.061 |
Niet | -0.059 |
Moss | -0.056 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | +7.54e-03 |
Pos logit contributions | |
| +0.285 |
Zi | +0.270 |
Catalan | +0.268 |
Niet | +0.262 |
Moss | +0.244 |
Neg logit contributions | |
WORK | -0.675 |
liv | -0.574 |
awatts | -0.561 |
aturdays | -0.560 |
WARE | -0.559 |
Token | 's |
Token ID | 338 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
WORK | +0.006 |
aturdays | +0.005 |
liv | +0.005 |
WARE | +0.005 |
awatts | +0.004 |
Neg logit contributions | |
| -0.006 |
Zi | -0.006 |
Catalan | -0.006 |
Niet | -0.006 |
Moss | -0.005 |
Token | political |
Token ID | 1964 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
WORK | +0.034 |
liv | +0.028 |
awatts | +0.028 |
WARE | +0.028 |
aturdays | +0.027 |
Neg logit contributions | |
| -0.025 |
Catalan | -0.024 |
Zi | -0.024 |
Niet | -0.024 |
wikipedia | -0.022 |
Token | Hide |
Token ID | 10415 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
WORK | +0.106 |
liv | +0.085 |
awatts | +0.081 |
WARE | +0.080 |
aturdays | +0.078 |
Neg logit contributions | |
| -0.103 |
Zi | -0.100 |
Catalan | -0.100 |
Niet | -0.099 |
wikipedia | -0.095 |
Token | Caption |
Token ID | 11260 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
WORK | +0.088 |
liv | +0.068 |
awatts | +0.065 |
WARE | +0.065 |
aturdays | +0.065 |
Neg logit contributions | |
| -0.112 |
Catalan | -0.108 |
Zi | -0.108 |
Niet | -0.106 |
Moss | -0.104 |
Token | 14 |
Token ID | 1478 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
WORK | +0.247 |
liv | +0.204 |
WARE | +0.198 |
awatts | +0.198 |
aturdays | +0.197 |
Neg logit contributions | |
| -0.161 |
Zi | -0.155 |
Catalan | -0.154 |
Niet | -0.150 |
Moss | -0.143 |
Token | of |
Token ID | 286 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
WORK | +0.032 |
liv | +0.027 |
aturdays | +0.026 |
WARE | +0.026 |
awatts | +0.026 |
Neg logit contributions | |
| -0.023 |
Zi | -0.022 |
Catalan | -0.022 |
Niet | -0.021 |
wikipedia | -0.021 |
Token | 15 |
Token ID | 1315 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
| +0.125 |
Catalan | +0.118 |
Zi | +0.117 |
Niet | +0.114 |
wikipedia | +0.110 |
Neg logit contributions | |
WORK | -0.167 |
liv | -0.135 |
aturdays | -0.135 |
WARE | -0.134 |
awatts | -0.134 |
Token | Photos |
Token ID | 9434 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
| +0.132 |
Catalan | +0.128 |
Zi | +0.128 |
Niet | +0.124 |
Moss | +0.121 |
Neg logit contributions | |
WORK | -0.118 |
liv | -0.092 |
WARE | -0.090 |
aturdays | -0.090 |
awatts | -0.088 |
Token | : |
Token ID | 25 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
| +1.056 |
Zi | +1.006 |
Catalan | +0.993 |
Niet | +0.953 |
wikipedia | +0.938 |
Neg logit contributions | |
WORK | -1.349 |
aturdays | -1.110 |
liv | -1.106 |
WARE | -1.085 |
awatts | -1.076 |
Token | Carly |
Token ID | 35982 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
WORK | +0.138 |
liv | +0.114 |
aturdays | +0.112 |
WARE | +0.112 |
awatts | +0.111 |
Neg logit contributions | |
| -0.099 |
Zi | -0.095 |
Catalan | -0.095 |
Niet | -0.092 |
Moss | -0.088 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
| +0.204 |
Zi | +0.194 |
Catalan | +0.193 |
Niet | +0.188 |
Moss | +0.176 |
Neg logit contributions | |
WORK | -0.480 |
liv | -0.408 |
awatts | -0.398 |
WARE | -0.397 |
aturdays | -0.396 |
Token | 's |
Token ID | 338 |
Feature activation | +7.80e-02 |
Pos logit contributions | |
WORK | +0.110 |
liv | +0.085 |
aturdays | +0.085 |
WARE | +0.085 |
awatts | +0.084 |
Neg logit contributions | |
| -0.114 |
Zi | -0.110 |
Catalan | -0.109 |
Niet | -0.105 |
Moss | -0.103 |
Token | political |
Token ID | 1964 |
Feature activation | -9.80e-03 |
Pos logit contributions | |
WORK | +0.070 |
liv | +0.057 |
awatts | +0.056 |
WARE | +0.056 |
aturdays | +0.055 |
Neg logit contributions | |
| -0.052 |
Catalan | -0.050 |
Zi | -0.049 |
Niet | -0.048 |
wikipedia | -0.046 |
Token | � |
Token ID | 134 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
WORK | +0.466 |
liv | +0.378 |
awatts | +0.359 |
WARE | +0.351 |
aturdays | +0.339 |
Neg logit contributions | |
| -0.460 |
Zi | -0.447 |
Niet | -0.447 |
Catalan | -0.446 |
wikipedia | -0.424 |
Token | � |
Token ID | 119 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
| +0.417 |
Zi | +0.405 |
Catalan | +0.403 |
Niet | +0.397 |
wikipedia | +0.384 |
Neg logit contributions | |
WORK | -0.325 |
liv | -0.246 |
awatts | -0.235 |
WARE | -0.234 |
aturdays | -0.227 |
Token | o |
Token ID | 78 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
WORK | +0.168 |
liv | +0.133 |
WARE | +0.127 |
awatts | +0.127 |
Appl | +0.122 |
Neg logit contributions | |
| -0.167 |
Zi | -0.162 |
Catalan | -0.161 |
Niet | -0.161 |
wikipedia | -0.154 |
Token | Hal |
Token ID | 11023 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
WORK | +0.106 |
liv | +0.087 |
awatts | +0.084 |
WARE | +0.083 |
aturdays | +0.081 |
Neg logit contributions | |
| -0.080 |
Zi | -0.077 |
Catalan | -0.076 |
Niet | -0.076 |
wikipedia | -0.072 |
Token | ep |
Token ID | 538 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
WORK | +0.452 |
liv | +0.375 |
awatts | +0.361 |
WARE | +0.357 |
Appl | +0.348 |
Neg logit contributions | |
| -0.312 |
Zi | -0.301 |
Catalan | -0.300 |
Niet | -0.298 |
wikipedia | -0.281 |
Token | ā |
Token ID | 10235 |
Feature activation | +9.75e-01 |
Pos logit contributions | |
| +0.034 |
Zi | +0.033 |
Catalan | +0.033 |
Niet | +0.033 |
wikipedia | +0.031 |
Neg logit contributions | |
WORK | -0.037 |
liv | -0.029 |
awatts | -0.028 |
WARE | -0.028 |
aturdays | -0.027 |
Token | nih |
Token ID | 37373 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
WORK | +0.871 |
liv | +0.719 |
awatts | +0.691 |
WARE | +0.685 |
Appl | +0.665 |
Neg logit contributions | |
| -0.635 |
Zi | -0.613 |
Catalan | -0.611 |
Niet | -0.607 |
wikipedia | -0.573 |
Token | o |
Token ID | 78 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
WORK | +0.084 |
liv | +0.067 |
awatts | +0.064 |
WARE | +0.064 |
aturdays | +0.062 |
Neg logit contributions | |
| -0.076 |
Zi | -0.074 |
Catalan | -0.073 |
Niet | -0.073 |
wikipedia | -0.070 |
Token | m |
Token ID | 285 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
WORK | +0.011 |
liv | +0.009 |
awatts | +0.009 |
WARE | +0.009 |
aturdays | +0.008 |
Neg logit contributions | |
| -0.010 |
Zi | -0.009 |
Catalan | -0.009 |
Niet | -0.009 |
wikipedia | -0.009 |
Token | ai |
Token ID | 1872 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
WORK | +0.197 |
liv | +0.160 |
awatts | +0.153 |
WARE | +0.153 |
aturdays | +0.148 |
Neg logit contributions | |
| -0.165 |
Zi | -0.160 |
Catalan | -0.159 |
Niet | -0.158 |
wikipedia | -0.150 |
Token | , |
Token ID | 11 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
WORK | +0.709 |
liv | +0.581 |
awatts | +0.556 |
WARE | +0.552 |
Appl | +0.535 |
Neg logit contributions | |
| -0.575 |
Zi | -0.556 |
Catalan | -0.553 |
Niet | -0.552 |
wikipedia | -0.523 |
Token | about |
Token ID | 546 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
WORK | +0.096 |
liv | +0.077 |
awatts | +0.074 |
WARE | +0.073 |
aturdays | +0.071 |
Neg logit contributions | |
| -0.086 |
Zi | -0.083 |
Catalan | -0.083 |
Niet | -0.082 |
wikipedia | -0.078 |
Token | LGBT |
Token ID | 9916 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
WORK | +0.133 |
liv | +0.105 |
awatts | +0.103 |
WARE | +0.102 |
aturdays | +0.099 |
Neg logit contributions | |
| -0.122 |
Zi | -0.118 |
Catalan | -0.118 |
Niet | -0.116 |
Moss | -0.111 |
Token | youth |
Token ID | 6205 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
| +0.284 |
Zi | +0.276 |
Catalan | +0.274 |
Niet | +0.273 |
Moss | +0.262 |
Neg logit contributions | |
WORK | -0.192 |
liv | -0.143 |
awatts | -0.136 |
WARE | -0.135 |
aturdays | -0.129 |
Token | that |
Token ID | 326 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
| +0.178 |
Zi | +0.172 |
Catalan | +0.171 |
Niet | +0.169 |
wikipedia | +0.162 |
Neg logit contributions | |
WORK | -0.164 |
liv | -0.128 |
awatts | -0.123 |
WARE | -0.123 |
aturdays | -0.118 |
Token | takes |
Token ID | 2753 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
WORK | +0.178 |
liv | +0.142 |
awatts | +0.137 |
WARE | +0.137 |
aturdays | +0.133 |
Neg logit contributions | |
| -0.159 |
Zi | -0.153 |
Catalan | -0.153 |
Niet | -0.151 |
Moss | -0.144 |
Token | account |
Token ID | 1848 |
Feature activation | +9.69e-01 |
Pos logit contributions | |
WORK | +0.242 |
liv | +0.192 |
awatts | +0.185 |
WARE | +0.184 |
aturdays | +0.177 |
Neg logit contributions | |
| -0.241 |
Zi | -0.233 |
Catalan | -0.232 |
Niet | -0.231 |
Moss | -0.220 |
Token | of |
Token ID | 286 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
WORK | +0.800 |
liv | +0.657 |
awatts | +0.620 |
WARE | +0.614 |
aturdays | +0.592 |
Neg logit contributions | |
| -0.688 |
Zi | -0.674 |
Niet | -0.669 |
Catalan | -0.669 |
wikipedia | -0.637 |
Token | their |
Token ID | 511 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
WORK | +0.129 |
liv | +0.103 |
awatts | +0.099 |
WARE | +0.099 |
aturdays | +0.095 |
Neg logit contributions | |
| -0.120 |
Zi | -0.116 |
Catalan | -0.116 |
Niet | -0.115 |
Moss | -0.109 |
Token | experiences |
Token ID | 6461 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
WORK | +0.093 |
liv | +0.073 |
awatts | +0.071 |
WARE | +0.071 |
aturdays | +0.068 |
Neg logit contributions | |
| -0.096 |
Zi | -0.093 |
Catalan | -0.093 |
Niet | -0.092 |
Moss | -0.088 |
Token | and |
Token ID | 290 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
WORK | +0.033 |
liv | +0.026 |
awatts | +0.025 |
WARE | +0.024 |
aturdays | +0.023 |
Neg logit contributions | |
| -0.035 |
Zi | -0.034 |
Catalan | -0.033 |
Niet | -0.033 |
wikipedia | -0.032 |
Token | needs |
Token ID | 2476 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
WORK | +0.151 |
liv | +0.119 |
awatts | +0.115 |
WARE | +0.114 |
aturdays | +0.110 |
Neg logit contributions | |
| -0.154 |
Zi | -0.149 |
Catalan | -0.149 |
Niet | -0.148 |
wikipedia | -0.141 |
Token | as |
Token ID | 355 |
Feature activation | +5.61e-02 |
Pos logit contributions | |
| +0.149 |
Zi | +0.143 |
Catalan | +0.143 |
Niet | +0.142 |
wikipedia | +0.136 |
Neg logit contributions | |
WORK | -0.144 |
liv | -0.113 |
awatts | -0.109 |
WARE | -0.108 |
aturdays | -0.104 |
Token | she |
Token ID | 673 |
Feature activation | -6.15e-03 |
Pos logit contributions | |
WORK | +0.041 |
liv | +0.032 |
awatts | +0.030 |
WARE | +0.030 |
aturdays | +0.029 |
Neg logit contributions | |
| -0.046 |
Zi | -0.045 |
Catalan | -0.044 |
Niet | -0.044 |
wikipedia | -0.042 |
Token | bid |
Token ID | 8406 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
| +0.004 |
Zi | +0.004 |
Catalan | +0.004 |
Niet | +0.004 |
wikipedia | +0.004 |
Neg logit contributions | |
WORK | -0.005 |
liv | -0.004 |
WARE | -0.004 |
awatts | -0.004 |
aturdays | -0.004 |
Token | an |
Token ID | 281 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
WORK | +0.404 |
liv | +0.309 |
awatts | +0.292 |
WARE | +0.291 |
aturdays | +0.277 |
Neg logit contributions | |
| -0.524 |
Zi | -0.511 |
Catalan | -0.508 |
Niet | -0.507 |
wikipedia | -0.486 |
Token | early |
Token ID | 1903 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
| +0.182 |
Catalan | +0.173 |
Zi | +0.173 |
Niet | +0.173 |
Moss | +0.164 |
Neg logit contributions | |
WORK | -0.194 |
liv | -0.154 |
WARE | -0.150 |
awatts | -0.148 |
terday | -0.142 |
Token | farewell |
Token ID | 33458 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
| +0.377 |
Zi | +0.367 |
Catalan | +0.367 |
Niet | +0.364 |
Moss | +0.347 |
Neg logit contributions | |
WORK | -0.326 |
liv | -0.251 |
awatts | -0.245 |
WARE | -0.242 |
aturdays | -0.229 |
Token | to |
Token ID | 284 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
WORK | +0.903 |
liv | +0.729 |
awatts | +0.697 |
WARE | +0.694 |
Appl | +0.670 |
Neg logit contributions | |
| -0.803 |
Zi | -0.778 |
Catalan | -0.773 |
Niet | -0.771 |
wikipedia | -0.734 |
Token | the |
Token ID | 262 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
| +0.071 |
Zi | +0.068 |
Catalan | +0.068 |
Niet | +0.068 |
wikipedia | +0.064 |
Neg logit contributions | |
WORK | -0.074 |
liv | -0.059 |
awatts | -0.056 |
WARE | -0.056 |
aturdays | -0.054 |
Token | White |
Token ID | 2635 |
Feature activation | +6.01e-02 |
Pos logit contributions | |
| +0.020 |
Zi | +0.019 |
Catalan | +0.019 |
Niet | +0.019 |
Moss | +0.018 |
Neg logit contributions | |
WORK | -0.023 |
liv | -0.018 |
WARE | -0.017 |
awatts | -0.017 |
aturdays | -0.017 |
Token | House |
Token ID | 2097 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
WORK | +0.058 |
liv | +0.049 |
WARE | +0.048 |
awatts | +0.048 |
aturdays | +0.047 |
Neg logit contributions | |
| -0.035 |
Zi | -0.033 |
Catalan | -0.033 |
Niet | -0.032 |
Moss | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
| +0.248 |
Zi | +0.238 |
Catalan | +0.237 |
Niet | +0.235 |
Moss | +0.224 |
Neg logit contributions | |
WORK | -0.265 |
liv | -0.213 |
awatts | -0.205 |
WARE | -0.204 |
aturdays | -0.197 |
Token | So |
Token ID | 1406 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
| +0.119 |
Zi | +0.116 |
Catalan | +0.115 |
Niet | +0.115 |
wikipedia | +0.110 |
Neg logit contributions | |
WORK | -0.110 |
liv | -0.087 |
awatts | -0.083 |
WARE | -0.082 |
aturdays | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | -9.20e-02 |
Pos logit contributions | |
| +0.197 |
Zi | +0.191 |
Catalan | +0.190 |
Niet | +0.189 |
wikipedia | +0.181 |
Neg logit contributions | |
WORK | -0.199 |
liv | -0.158 |
awatts | -0.151 |
WARE | -0.150 |
aturdays | -0.145 |
Token | if |
Token ID | 611 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
| +0.070 |
Zi | +0.068 |
Catalan | +0.067 |
Niet | +0.067 |
wikipedia | +0.064 |
Neg logit contributions | |
WORK | -0.073 |
liv | -0.058 |
awatts | -0.056 |
WARE | -0.055 |
aturdays | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
WORK | +0.022 |
liv | +0.018 |
awatts | +0.017 |
WARE | +0.017 |
aturdays | +0.017 |
Neg logit contributions | |
| -0.020 |
Zi | -0.019 |
Catalan | -0.019 |
Niet | -0.019 |
wikipedia | -0.018 |
Token | data |
Token ID | 1366 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
WORK | +0.035 |
liv | +0.028 |
awatts | +0.027 |
WARE | +0.027 |
Appl | +0.026 |
Neg logit contributions | |
| -0.028 |
Zi | -0.027 |
Catalan | -0.027 |
Niet | -0.026 |
wikipedia | -0.025 |
Token | file |
Token ID | 2393 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
WORK | +0.500 |
liv | +0.404 |
awatts | +0.386 |
WARE | +0.385 |
Appl | +0.372 |
Neg logit contributions | |
| -0.439 |
Zi | -0.425 |
Catalan | -0.422 |
Niet | -0.421 |
wikipedia | -0.400 |
Token | is |
Token ID | 318 |
Feature activation | +3.65e-02 |
Pos logit contributions | |
WORK | +0.977 |
liv | +0.791 |
awatts | +0.750 |
WARE | +0.745 |
Appl | +0.727 |
Neg logit contributions | |
| -0.882 |
Zi | -0.851 |
Catalan | -0.849 |
Niet | -0.847 |
wikipedia | -0.807 |
Token | data |
Token ID | 1366 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
WORK | +0.030 |
liv | +0.024 |
awatts | +0.023 |
WARE | +0.023 |
Appl | +0.022 |
Neg logit contributions | |
| -0.026 |
Zi | -0.026 |
Catalan | -0.025 |
Niet | -0.025 |
wikipedia | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
WORK | +0.408 |
liv | +0.329 |
awatts | +0.315 |
WARE | +0.313 |
Appl | +0.303 |
Neg logit contributions | |
| -0.355 |
Zi | -0.345 |
Catalan | -0.342 |
Niet | -0.342 |
wikipedia | -0.324 |
Token | json |
Token ID | 17752 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
| +0.076 |
Zi | +0.073 |
Catalan | +0.073 |
Niet | +0.072 |
wikipedia | +0.068 |
Neg logit contributions | |
WORK | -0.127 |
liv | -0.106 |
awatts | -0.103 |
WARE | -0.102 |
aturdays | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | -9.96e-02 |
Pos logit contributions | |
| +0.277 |
Zi | +0.269 |
Catalan | +0.267 |
Niet | +0.266 |
wikipedia | +0.252 |
Neg logit contributions | |
WORK | -0.338 |
liv | -0.275 |
awatts | -0.264 |
WARE | -0.263 |
Appl | -0.254 |
Token | recreational |
Token ID | 18136 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
| +0.014 |
Zi | +0.013 |
Catalan | +0.013 |
Niet | +0.013 |
wikipedia | +0.013 |
Neg logit contributions | |
WORK | -0.014 |
liv | -0.011 |
awatts | -0.010 |
WARE | -0.010 |
Appl | -0.010 |
Token | marijuana |
Token ID | 5727 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
| +0.227 |
Zi | +0.221 |
Catalan | +0.220 |
Niet | +0.219 |
wikipedia | +0.209 |
Neg logit contributions | |
WORK | -0.222 |
liv | -0.177 |
awatts | -0.167 |
WARE | -0.166 |
aturdays | -0.160 |
Token | . |
Token ID | 13 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
WORK | +0.142 |
liv | +0.113 |
awatts | +0.108 |
WARE | +0.107 |
aturdays | +0.103 |
Neg logit contributions | |
| -0.137 |
Zi | -0.133 |
Catalan | -0.132 |
Niet | -0.132 |
wikipedia | -0.126 |
Token |
|
Token ID | 198 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
WORK | +0.151 |
liv | +0.117 |
awatts | +0.111 |
WARE | +0.109 |
Appl | +0.107 |
Neg logit contributions | |
| -0.182 |
Zi | -0.179 |
Niet | -0.177 |
Catalan | -0.177 |
wikipedia | -0.170 |
Token |
|
Token ID | 198 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
WORK | +0.117 |
liv | +0.093 |
aturdays | +0.090 |
WARE | +0.090 |
awatts | +0.090 |
Neg logit contributions | |
| -0.121 |
Zi | -0.116 |
Catalan | -0.116 |
Niet | -0.113 |
wikipedia | -0.110 |
Token | • |
Token ID | 3581 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
WORK | +0.097 |
liv | +0.074 |
awatts | +0.071 |
WARE | +0.070 |
Appl | +0.067 |
Neg logit contributions | |
| -0.116 |
Zi | -0.114 |
Niet | -0.113 |
Catalan | -0.113 |
Moss | -0.107 |
Token | MP |
Token ID | 4904 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
WORK | +0.953 |
liv | +0.798 |
awatts | +0.764 |
WARE | +0.761 |
Appl | +0.747 |
Neg logit contributions | |
| -0.582 |
Zi | -0.564 |
Niet | -0.560 |
Catalan | -0.558 |
wikipedia | -0.526 |
Token | P |
Token ID | 47 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
| +0.260 |
Zi | +0.248 |
Catalan | +0.247 |
Niet | +0.241 |
wikipedia | +0.233 |
Neg logit contributions | |
WORK | -0.316 |
liv | -0.258 |
awatts | -0.251 |
aturdays | -0.248 |
WARE | -0.245 |
Token | grade |
Token ID | 9559 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
WORK | +0.104 |
liv | +0.073 |
awatts | +0.070 |
WARE | +0.069 |
aturdays | +0.066 |
Neg logit contributions | |
| -0.187 |
Zi | -0.182 |
Catalan | -0.182 |
Niet | -0.180 |
wikipedia | -0.174 |
Token | : |
Token ID | 25 |
Feature activation | +8.96e-02 |
Pos logit contributions | |
WORK | +0.482 |
liv | +0.388 |
awatts | +0.376 |
WARE | +0.375 |
aturdays | +0.367 |
Neg logit contributions | |
| -0.420 |
Zi | -0.406 |
Catalan | -0.403 |
Niet | -0.398 |
wikipedia | -0.381 |
Token | C |
Token ID | 327 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
WORK | +0.077 |
liv | +0.063 |
awatts | +0.060 |
WARE | +0.059 |
Appl | +0.058 |
Neg logit contributions | |
| -0.062 |
Zi | -0.060 |
Catalan | -0.059 |
Niet | -0.059 |
wikipedia | -0.056 |
Token | into |
Token ID | 656 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
WORK | +0.192 |
liv | +0.154 |
awatts | +0.146 |
WARE | +0.146 |
aturdays | +0.140 |
Neg logit contributions | |
| -0.188 |
Zi | -0.183 |
Niet | -0.182 |
Catalan | -0.182 |
wikipedia | -0.173 |
Token | another |
Token ID | 1194 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
| +0.072 |
Zi | +0.070 |
Catalan | +0.069 |
Niet | +0.069 |
wikipedia | +0.066 |
Neg logit contributions | |
WORK | -0.086 |
liv | -0.070 |
awatts | -0.067 |
WARE | -0.067 |
Appl | -0.065 |
Token | brewery |
Token ID | 22120 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
| +0.283 |
Zi | +0.275 |
Catalan | +0.274 |
Niet | +0.274 |
wikipedia | +0.262 |
Neg logit contributions | |
WORK | -0.265 |
liv | -0.210 |
awatts | -0.199 |
WARE | -0.198 |
Appl | -0.191 |
Token | . |
Token ID | 13 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
WORK | +0.118 |
liv | +0.095 |
awatts | +0.091 |
WARE | +0.091 |
aturdays | +0.087 |
Neg logit contributions | |
| -0.108 |
Zi | -0.105 |
Catalan | -0.104 |
Niet | -0.103 |
wikipedia | -0.098 |
Token | If |
Token ID | 1002 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
| +0.133 |
Zi | +0.129 |
Catalan | +0.128 |
Niet | +0.127 |
wikipedia | +0.122 |
Neg logit contributions | |
WORK | -0.124 |
liv | -0.098 |
awatts | -0.094 |
WARE | -0.093 |
aturdays | -0.090 |
Token | you |
Token ID | 345 |
Feature activation | +9.50e-02 |
Pos logit contributions | |
| +0.229 |
Zi | +0.222 |
Catalan | +0.220 |
Niet | +0.219 |
wikipedia | +0.209 |
Neg logit contributions | |
WORK | -0.256 |
liv | -0.207 |
awatts | -0.198 |
WARE | -0.197 |
aturdays | -0.190 |
Token | aren |
Token ID | 3588 |
Feature activation | -1.77e-03 |
Pos logit contributions | |
WORK | +0.076 |
liv | +0.060 |
WARE | +0.058 |
awatts | +0.058 |
aturdays | +0.056 |
Neg logit contributions | |
| -0.072 |
Zi | -0.069 |
Catalan | -0.069 |
Niet | -0.068 |
wikipedia | -0.065 |
Token | � |
Token ID | 447 |
Feature activation | +5.74e-02 |
Pos logit contributions | |
| +0.001 |
Zi | +0.001 |
Catalan | +0.001 |
Niet | +0.001 |
wikipedia | +0.001 |
Neg logit contributions | |
WORK | -0.002 |
liv | -0.001 |
aturdays | -0.001 |
WARE | -0.001 |
awatts | -0.001 |
Token | � |
Token ID | 247 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
WORK | +0.060 |
liv | +0.051 |
awatts | +0.049 |
WARE | +0.049 |
aturdays | +0.048 |
Neg logit contributions | |
| -0.029 |
Zi | -0.028 |
Catalan | -0.027 |
Niet | -0.027 |
wikipedia | -0.025 |
Token | t |
Token ID | 83 |
Feature activation | +5.28e-03 |
Pos logit contributions | |
WORK | +0.012 |
liv | +0.009 |
WARE | +0.009 |
awatts | +0.009 |
aturdays | +0.008 |
Neg logit contributions | |
| -0.013 |
Zi | -0.013 |
Catalan | -0.012 |
Niet | -0.012 |
wikipedia | -0.012 |
Token | trying |
Token ID | 2111 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
WORK | +0.004 |
liv | +0.004 |
awatts | +0.003 |
WARE | +0.003 |
Appl | +0.003 |
Neg logit contributions | |
| -0.004 |
Zi | -0.004 |
Catalan | -0.004 |
Niet | -0.004 |
wikipedia | -0.003 |
Token | Disclosure |
Token ID | 41806 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
| +0.119 |
Zi | +0.115 |
Catalan | +0.114 |
Niet | +0.114 |
wikipedia | +0.109 |
Neg logit contributions | |
WORK | -0.119 |
liv | -0.094 |
awatts | -0.090 |
WARE | -0.090 |
aturdays | -0.087 |
Token | Form |
Token ID | 5178 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
WORK | +0.331 |
liv | +0.266 |
awatts | +0.254 |
WARE | +0.253 |
aturdays | +0.244 |
Neg logit contributions | |
| -0.310 |
Zi | -0.300 |
Catalan | -0.298 |
Niet | -0.297 |
wikipedia | -0.283 |
Token | ] |
Token ID | 60 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
WORK | +0.343 |
liv | +0.275 |
awatts | +0.262 |
WARE | +0.261 |
aturdays | +0.252 |
Neg logit contributions | |
| -0.323 |
Zi | -0.313 |
Catalan | -0.311 |
Niet | -0.310 |
wikipedia | -0.296 |
Token | to |
Token ID | 284 |
Feature activation | -1.52e-03 |
Pos logit contributions | |
| +0.219 |
Zi | +0.212 |
Catalan | +0.210 |
Niet | +0.210 |
wikipedia | +0.200 |
Neg logit contributions | |
WORK | -0.245 |
liv | -0.197 |
awatts | -0.190 |
WARE | -0.189 |
aturdays | -0.182 |
Token | become |
Token ID | 1716 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
| +0.001 |
Zi | +0.001 |
Catalan | +0.001 |
Niet | +0.001 |
Moss | +0.001 |
Neg logit contributions | |
WORK | -0.001 |
liv | -0.001 |
awatts | -0.001 |
WARE | -0.001 |
aturdays | -0.001 |
Token | Senator |
Token ID | 8962 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
WORK | +0.135 |
liv | +0.111 |
awatts | +0.106 |
WARE | +0.106 |
Appl | +0.103 |
Neg logit contributions | |
| -0.103 |
Zi | -0.100 |
Catalan | -0.099 |
Niet | -0.099 |
wikipedia | -0.093 |
Token | John |
Token ID | 1757 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
WORK | +0.340 |
liv | +0.282 |
awatts | +0.269 |
WARE | +0.268 |
Appl | +0.261 |
Neg logit contributions | |
| -0.249 |
Zi | -0.240 |
Catalan | -0.239 |
Niet | -0.239 |
wikipedia | -0.225 |
Token | McCain |
Token ID | 14264 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
| +0.068 |
Zi | +0.064 |
Catalan | +0.063 |
Niet | +0.063 |
wikipedia | +0.057 |
Neg logit contributions | |
WORK | -0.205 |
liv | -0.177 |
awatts | -0.172 |
WARE | -0.171 |
Appl | -0.168 |
Token | 's |
Token ID | 338 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
| +0.114 |
Zi | +0.111 |
Catalan | +0.110 |
Niet | +0.110 |
wikipedia | +0.105 |
Neg logit contributions | |
WORK | -0.104 |
liv | -0.082 |
awatts | -0.078 |
WARE | -0.077 |
aturdays | -0.074 |
Token | chief |
Token ID | 4039 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
WORK | +0.044 |
liv | +0.036 |
awatts | +0.035 |
WARE | +0.035 |
aturdays | +0.034 |
Neg logit contributions | |
| -0.034 |
Zi | -0.033 |
Catalan | -0.032 |
Niet | -0.032 |
wikipedia | -0.031 |
Token | campaign |
Token ID | 1923 |
Feature activation | +9.12e-02 |
Pos logit contributions | |
| +0.020 |
Zi | +0.020 |
Catalan | +0.020 |
Niet | +0.019 |
wikipedia | +0.018 |
Neg logit contributions | |
WORK | -0.026 |
liv | -0.021 |
awatts | -0.020 |
WARE | -0.020 |
Appl | -0.020 |
Token | optimized |
Token ID | 23392 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
WORK | +0.065 |
liv | +0.050 |
awatts | +0.048 |
WARE | +0.048 |
aturdays | +0.046 |
Neg logit contributions | |
| -0.072 |
Zi | -0.070 |
Catalan | -0.069 |
Niet | -0.069 |
Moss | -0.066 |
Token | Store |
Token ID | 9363 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
WORK | +0.279 |
liv | +0.221 |
awatts | +0.214 |
WARE | +0.211 |
aturdays | +0.205 |
Neg logit contributions | |
| -0.270 |
Zi | -0.260 |
Catalan | -0.258 |
Niet | -0.257 |
Moss | -0.244 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
| +0.199 |
Zi | +0.192 |
Catalan | +0.191 |
Niet | +0.190 |
Moss | +0.181 |
Neg logit contributions | |
WORK | -0.202 |
liv | -0.161 |
awatts | -0.155 |
WARE | -0.153 |
aturdays | -0.149 |
Token | As |
Token ID | 1081 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
WORK | +0.076 |
liv | +0.060 |
awatts | +0.057 |
WARE | +0.056 |
aturdays | +0.055 |
Neg logit contributions | |
| -0.084 |
Zi | -0.081 |
Catalan | -0.080 |
Niet | -0.080 |
wikipedia | -0.077 |
Token | you |
Token ID | 345 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
WORK | +0.025 |
liv | +0.020 |
awatts | +0.019 |
WARE | +0.019 |
aturdays | +0.018 |
Neg logit contributions | |
| -0.022 |
Zi | -0.021 |
Catalan | -0.021 |
Niet | -0.021 |
wikipedia | -0.020 |
Token | land |
Token ID | 1956 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
WORK | +0.134 |
liv | +0.106 |
WARE | +0.103 |
awatts | +0.103 |
aturdays | +0.099 |
Neg logit contributions | |
| -0.134 |
Zi | -0.129 |
Catalan | -0.128 |
Niet | -0.126 |
wikipedia | -0.121 |
Token | on |
Token ID | 319 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
WORK | +0.091 |
liv | +0.074 |
awatts | +0.071 |
WARE | +0.070 |
aturdays | +0.068 |
Neg logit contributions | |
| -0.080 |
Zi | -0.077 |
Catalan | -0.077 |
Niet | -0.076 |
wikipedia | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | +7.50e-02 |
Pos logit contributions | |
WORK | +0.072 |
liv | +0.059 |
WARE | +0.057 |
awatts | +0.057 |
Appl | +0.055 |
Neg logit contributions | |
| -0.058 |
Zi | -0.056 |
Catalan | -0.056 |
Niet | -0.056 |
wikipedia | -0.053 |
Token | Store |
Token ID | 9363 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
WORK | +0.060 |
liv | +0.048 |
awatts | +0.046 |
WARE | +0.045 |
aturdays | +0.044 |
Neg logit contributions | |
| -0.056 |
Zi | -0.055 |
Catalan | -0.054 |
Niet | -0.054 |
wikipedia | -0.052 |
Token | from |
Token ID | 422 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
| +0.088 |
Zi | +0.085 |
Catalan | +0.084 |
Niet | +0.084 |
wikipedia | +0.080 |
Neg logit contributions | |
WORK | -0.085 |
liv | -0.067 |
awatts | -0.064 |
WARE | -0.063 |
aturdays | -0.062 |
Token | Home |
Token ID | 5995 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
WORK | +0.028 |
liv | +0.023 |
awatts | +0.022 |
WARE | +0.022 |
aturdays | +0.021 |
Neg logit contributions | |
| -0.025 |
Zi | -0.024 |
Catalan | -0.024 |
Niet | -0.024 |
Moss | -0.023 |
Token | status |
Token ID | 3722 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
WORK | +0.000 |
liv | +0.000 |
awatts | +0.000 |
WARE | +0.000 |
aturdays | +0.000 |
Neg logit contributions | |
| -0.000 |
Zi | -0.000 |
Catalan | -0.000 |
Niet | -0.000 |
Moss | -0.000 |
Token | Indians |
Token ID | 12746 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
WORK | +0.222 |
liv | +0.176 |
awatts | +0.167 |
WARE | +0.166 |
aturdays | +0.160 |
Neg logit contributions | |
| -0.229 |
Zi | -0.223 |
Catalan | -0.221 |
Niet | -0.221 |
wikipedia | -0.211 |
Token | are |
Token ID | 389 |
Feature activation | +9.14e-02 |
Pos logit contributions | |
WORK | +0.147 |
liv | +0.116 |
awatts | +0.111 |
WARE | +0.111 |
aturdays | +0.107 |
Neg logit contributions | |
| -0.150 |
Zi | -0.145 |
Catalan | -0.144 |
Niet | -0.143 |
wikipedia | -0.137 |
Token | largely |
Token ID | 5688 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
WORK | +0.074 |
liv | +0.060 |
awatts | +0.057 |
WARE | +0.057 |
aturdays | +0.055 |
Neg logit contributions | |
| -0.067 |
Zi | -0.065 |
Catalan | -0.065 |
Niet | -0.064 |
wikipedia | -0.061 |
Token | a |
Token ID | 257 |
Feature activation | +6.71e-02 |
Pos logit contributions | |
WORK | +0.252 |
liv | +0.200 |
awatts | +0.191 |
WARE | +0.190 |
aturdays | +0.183 |
Neg logit contributions | |
| -0.252 |
Zi | -0.245 |
Catalan | -0.244 |
Niet | -0.243 |
wikipedia | -0.232 |
Token | federal |
Token ID | 2717 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
WORK | +0.053 |
liv | +0.042 |
awatts | +0.041 |
WARE | +0.040 |
aturdays | +0.039 |
Neg logit contributions | |
| -0.051 |
Zi | -0.049 |
Catalan | -0.049 |
Niet | -0.049 |
wikipedia | -0.046 |
Token | responsibility |
Token ID | 5798 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
WORK | +0.188 |
liv | +0.148 |
awatts | +0.140 |
WARE | +0.140 |
Appl | +0.134 |
Neg logit contributions | |
| -0.204 |
Zi | -0.199 |
Catalan | -0.197 |
Niet | -0.197 |
wikipedia | -0.188 |
Token | . |
Token ID | 13 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
WORK | +0.272 |
liv | +0.218 |
awatts | +0.209 |
WARE | +0.208 |
aturdays | +0.201 |
Neg logit contributions | |
| -0.257 |
Zi | -0.248 |
Catalan | -0.247 |
Niet | -0.246 |
wikipedia | -0.234 |
Token | Access |
Token ID | 8798 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
WORK | +0.089 |
liv | +0.071 |
awatts | +0.068 |
WARE | +0.068 |
aturdays | +0.066 |
Neg logit contributions | |
| -0.090 |
Zi | -0.087 |
Catalan | -0.087 |
Niet | -0.086 |
wikipedia | -0.082 |
Token | ing |
Token ID | 278 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
| +0.632 |
Zi | +0.612 |
Catalan | +0.608 |
Niet | +0.606 |
wikipedia | +0.575 |
Neg logit contributions | |
WORK | -0.747 |
liv | -0.606 |
awatts | -0.581 |
WARE | -0.578 |
aturdays | -0.559 |
Token | resources |
Token ID | 4133 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
WORK | +0.200 |
liv | +0.157 |
awatts | +0.151 |
WARE | +0.150 |
aturdays | +0.145 |
Neg logit contributions | |
| -0.207 |
Zi | -0.201 |
Catalan | -0.200 |
Niet | -0.199 |
wikipedia | -0.190 |
Token | match |
Token ID | 2872 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
| +0.037 |
Zi | +0.036 |
Catalan | +0.035 |
Niet | +0.035 |
wikipedia | +0.033 |
Neg logit contributions | |
WORK | -0.053 |
liv | -0.044 |
awatts | -0.042 |
WARE | -0.042 |
Appl | -0.041 |
Token | situations |
Token ID | 7445 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
WORK | +0.029 |
liv | +0.023 |
awatts | +0.022 |
WARE | +0.022 |
aturdays | +0.022 |
Neg logit contributions | |
| -0.023 |
Zi | -0.022 |
Catalan | -0.022 |
Niet | -0.022 |
wikipedia | -0.021 |
Token | that |
Token ID | 326 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
WORK | +0.147 |
liv | +0.119 |
awatts | +0.113 |
WARE | +0.113 |
Appl | +0.109 |
Neg logit contributions | |
| -0.134 |
Zi | -0.130 |
Niet | -0.130 |
Catalan | -0.130 |
wikipedia | -0.124 |
Token | Pakistan |
Token ID | 7648 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
| +0.032 |
Zi | +0.031 |
Catalan | +0.031 |
Niet | +0.031 |
wikipedia | +0.029 |
Neg logit contributions | |
WORK | -0.039 |
liv | -0.031 |
awatts | -0.030 |
WARE | -0.030 |
Appl | -0.029 |
Token | � |
Token ID | 447 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
| +0.217 |
Zi | +0.210 |
Catalan | +0.208 |
Niet | +0.208 |
wikipedia | +0.197 |
Neg logit contributions | |
WORK | -0.286 |
liv | -0.235 |
awatts | -0.226 |
WARE | -0.225 |
aturdays | -0.218 |
Token | � |
Token ID | 247 |
Feature activation | +9.48e-02 |
Pos logit contributions | |
WORK | +0.138 |
liv | +0.117 |
awatts | +0.113 |
WARE | +0.113 |
aturdays | +0.111 |
Neg logit contributions | |
| -0.066 |
Zi | -0.063 |
Catalan | -0.062 |
Niet | -0.061 |
wikipedia | -0.057 |
Token | s |
Token ID | 82 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
WORK | +0.076 |
liv | +0.061 |
WARE | +0.059 |
aturdays | +0.058 |
awatts | +0.058 |
Neg logit contributions | |
| -0.072 |
Zi | -0.069 |
Catalan | -0.068 |
Niet | -0.067 |
wikipedia | -0.065 |
Token | chase |
Token ID | 15505 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
| +0.090 |
Zi | +0.087 |
Catalan | +0.086 |
Niet | +0.086 |
wikipedia | +0.082 |
Neg logit contributions | |
WORK | -0.104 |
liv | -0.084 |
awatts | -0.081 |
WARE | -0.080 |
aturdays | -0.078 |
Token | went |
Token ID | 1816 |
Feature activation | +9.90e-02 |
Pos logit contributions | |
WORK | +0.125 |
liv | +0.102 |
awatts | +0.098 |
WARE | +0.097 |
aturdays | +0.094 |
Neg logit contributions | |
| -0.101 |
Zi | -0.098 |
Catalan | -0.097 |
Niet | -0.097 |
wikipedia | -0.092 |
Token | through |
Token ID | 832 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
WORK | +0.071 |
liv | +0.056 |
awatts | +0.053 |
WARE | +0.053 |
Appl | +0.050 |
Neg logit contributions | |
| -0.081 |
Zi | -0.079 |
Catalan | -0.079 |
Niet | -0.079 |
wikipedia | -0.075 |
Token | in |
Token ID | 287 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
WORK | +0.265 |
liv | +0.213 |
awatts | +0.202 |
WARE | +0.201 |
Appl | +0.194 |
Neg logit contributions | |
| -0.255 |
Zi | -0.248 |
Niet | -0.247 |
Catalan | -0.247 |
wikipedia | -0.234 |
Token | , |
Token ID | 11 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
| +0.669 |
Zi | +0.648 |
Catalan | +0.644 |
Niet | +0.642 |
Moss | +0.613 |
Neg logit contributions | |
WORK | -0.575 |
liv | -0.447 |
WARE | -0.428 |
awatts | -0.427 |
aturdays | -0.410 |
Token | and |
Token ID | 290 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
WORK | +0.056 |
liv | +0.044 |
awatts | +0.042 |
WARE | +0.042 |
aturdays | +0.040 |
Neg logit contributions | |
| -0.060 |
Zi | -0.058 |
Catalan | -0.058 |
Niet | -0.057 |
wikipedia | -0.055 |
Token | second |
Token ID | 1218 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
WORK | +0.031 |
liv | +0.025 |
awatts | +0.024 |
WARE | +0.024 |
aturdays | +0.023 |
Neg logit contributions | |
| -0.027 |
Zi | -0.026 |
Catalan | -0.026 |
Niet | -0.026 |
wikipedia | -0.024 |
Token | Way |
Token ID | 6378 |
Feature activation | -7.84e-03 |
Pos logit contributions | |
| +0.238 |
Zi | +0.230 |
Catalan | +0.229 |
Niet | +0.228 |
wikipedia | +0.216 |
Neg logit contributions | |
WORK | -0.280 |
liv | -0.226 |
awatts | -0.218 |
WARE | -0.218 |
aturdays | -0.211 |
Token | man |
Token ID | 805 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
| +0.006 |
Zi | +0.006 |
Catalan | +0.006 |
Niet | +0.006 |
wikipedia | +0.005 |
Neg logit contributions | |
WORK | -0.006 |
liv | -0.005 |
WARE | -0.005 |
awatts | -0.005 |
aturdays | -0.005 |
Token | T |
Token ID | 309 |
Feature activation | -4.92e-03 |
Pos logit contributions | |
| +0.487 |
Zi | +0.472 |
Catalan | +0.468 |
Niet | +0.467 |
wikipedia | +0.444 |
Neg logit contributions | |
WORK | -0.563 |
liv | -0.455 |
awatts | -0.437 |
WARE | -0.436 |
aturdays | -0.422 |
Token | isd |
Token ID | 9409 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
| +0.002 |
Zi | +0.002 |
Catalan | +0.002 |
Niet | +0.002 |
wikipedia | +0.002 |
Neg logit contributions | |
WORK | -0.006 |
liv | -0.005 |
WARE | -0.005 |
awatts | -0.005 |
aturdays | -0.005 |
Token | ale |
Token ID | 1000 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
| +0.468 |
Catalan | +0.454 |
Zi | +0.452 |
Niet | +0.441 |
wikipedia | +0.434 |
Neg logit contributions | |
WORK | -0.434 |
liv | -0.342 |
WARE | -0.332 |
aturdays | -0.321 |
awatts | -0.318 |
Token | National |
Token ID | 2351 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
WORK | +0.480 |
liv | +0.398 |
awatts | +0.385 |
WARE | +0.384 |
aturdays | +0.373 |
Neg logit contributions | |
| -0.314 |
Zi | -0.302 |
Catalan | -0.300 |
Niet | -0.299 |
wikipedia | -0.281 |
Token | Fresh |
Token ID | 20138 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
| +0.121 |
Zi | +0.118 |
Catalan | +0.117 |
Niet | +0.117 |
Moss | +0.111 |
Neg logit contributions | |
WORK | -0.105 |
liv | -0.082 |
awatts | -0.079 |
WARE | -0.079 |
aturdays | -0.076 |
Token | man |
Token ID | 805 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
| +0.167 |
Zi | +0.159 |
Catalan | +0.158 |
Niet | +0.155 |
wikipedia | +0.150 |
Neg logit contributions | |
WORK | -0.186 |
liv | -0.150 |
WARE | -0.147 |
aturdays | -0.145 |
awatts | -0.145 |
Token | concerning |
Token ID | 9305 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
| +0.328 |
Zi | +0.318 |
Catalan | +0.316 |
Niet | +0.314 |
wikipedia | +0.299 |
Neg logit contributions | |
WORK | -0.342 |
liv | -0.273 |
awatts | -0.262 |
WARE | -0.261 |
aturdays | -0.252 |
Token | how |
Token ID | 703 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
| +0.146 |
Zi | +0.142 |
Catalan | +0.141 |
Niet | +0.141 |
wikipedia | +0.134 |
Neg logit contributions | |
WORK | -0.149 |
liv | -0.119 |
awatts | -0.114 |
WARE | -0.113 |
aturdays | -0.109 |
Token | the |
Token ID | 262 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
| +0.012 |
Zi | +0.011 |
Catalan | +0.011 |
Niet | +0.011 |
wikipedia | +0.011 |
Neg logit contributions | |
WORK | -0.012 |
liv | -0.009 |
awatts | -0.009 |
WARE | -0.009 |
Appl | -0.008 |
Token | chapter |
Token ID | 6843 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
WORK | +0.101 |
liv | +0.079 |
awatts | +0.076 |
WARE | +0.076 |
aturdays | +0.073 |
Neg logit contributions | |
| -0.106 |
Zi | -0.103 |
Catalan | -0.102 |
Niet | -0.102 |
wikipedia | -0.097 |
Token | operates |
Token ID | 14051 |
Feature activation | +2.44e-03 |
Pos logit contributions | |
WORK | +0.504 |
liv | +0.401 |
awatts | +0.378 |
WARE | +0.376 |
Appl | +0.363 |
Neg logit contributions | |
| -0.525 |
Zi | -0.510 |
Niet | -0.508 |
Catalan | -0.508 |
wikipedia | -0.484 |
Token | , |
Token ID | 11 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
WORK | +0.002 |
liv | +0.002 |
awatts | +0.001 |
WARE | +0.001 |
aturdays | +0.001 |
Neg logit contributions | |
| -0.002 |
Zi | -0.002 |
Catalan | -0.002 |
Niet | -0.002 |
wikipedia | -0.002 |
Token | which |
Token ID | 543 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
| +0.034 |
Zi | +0.032 |
Catalan | +0.032 |
Niet | +0.032 |
wikipedia | +0.030 |
Neg logit contributions | |
WORK | -0.048 |
liv | -0.040 |
awatts | -0.038 |
WARE | -0.038 |
Appl | -0.037 |
Token | is |
Token ID | 318 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
| +0.126 |
Zi | +0.122 |
Catalan | +0.121 |
Niet | +0.121 |
wikipedia | +0.115 |
Neg logit contributions | |
WORK | -0.154 |
liv | -0.126 |
awatts | -0.121 |
WARE | -0.120 |
aturdays | -0.116 |
Token | another |
Token ID | 1194 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
WORK | +0.024 |
liv | +0.019 |
awatts | +0.018 |
WARE | +0.018 |
aturdays | +0.018 |
Neg logit contributions | |
| -0.019 |
Zi | -0.019 |
Catalan | -0.019 |
Niet | -0.019 |
wikipedia | -0.018 |
Token | impetus |
Token ID | 43007 |
Feature activation | -9.05e-02 |
Pos logit contributions | |
| +0.063 |
Zi | +0.061 |
Catalan | +0.061 |
Niet | +0.060 |
wikipedia | +0.057 |
Neg logit contributions | |
WORK | -0.069 |
liv | -0.056 |
awatts | -0.054 |
WARE | -0.053 |
aturdays | -0.051 |
Token | for |
Token ID | 329 |
Feature activation | +7.70e-02 |
Pos logit contributions | |
| +0.073 |
Zi | +0.071 |
Catalan | +0.070 |
Niet | +0.070 |
wikipedia | +0.067 |
Neg logit contributions | |
WORK | -0.067 |
liv | -0.053 |
awatts | -0.051 |
WARE | -0.050 |
aturdays | -0.048 |
Token | It |
Token ID | 1026 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
WORK | +0.044 |
liv | +0.034 |
awatts | +0.031 |
WARE | +0.031 |
aturdays | +0.030 |
Neg logit contributions | |
| -0.063 |
Zi | -0.061 |
Catalan | -0.061 |
Niet | -0.060 |
wikipedia | -0.059 |
Token | was |
Token ID | 373 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
| +0.019 |
Zi | +0.018 |
Catalan | +0.018 |
Niet | +0.018 |
wikipedia | +0.017 |
Neg logit contributions | |
WORK | -0.024 |
liv | -0.019 |
WARE | -0.019 |
awatts | -0.019 |
aturdays | -0.018 |
Token | not |
Token ID | 407 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
WORK | +0.099 |
liv | +0.081 |
awatts | +0.077 |
WARE | +0.077 |
Appl | +0.075 |
Neg logit contributions | |
| -0.075 |
Zi | -0.073 |
Catalan | -0.072 |
Niet | -0.072 |
wikipedia | -0.069 |
Token | an |
Token ID | 281 |
Feature activation | -9.01e-02 |
Pos logit contributions | |
WORK | +0.067 |
liv | +0.054 |
awatts | +0.052 |
WARE | +0.052 |
aturdays | +0.050 |
Neg logit contributions | |
| -0.055 |
Zi | -0.053 |
Catalan | -0.052 |
Niet | -0.052 |
wikipedia | -0.050 |
Token | adjustment |
Token ID | 15068 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
| +0.062 |
Zi | +0.060 |
Catalan | +0.060 |
Niet | +0.059 |
Moss | +0.056 |
Neg logit contributions | |
WORK | -0.078 |
liv | -0.063 |
WARE | -0.061 |
awatts | -0.060 |
aturdays | -0.058 |
Token | whatsoever |
Token ID | 16014 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
| +0.505 |
Zi | +0.488 |
Catalan | +0.487 |
Niet | +0.481 |
Moss | +0.459 |
Neg logit contributions | |
WORK | -0.502 |
liv | -0.395 |
awatts | -0.383 |
WARE | -0.381 |
aturdays | -0.368 |
Token | on |
Token ID | 319 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
| +0.398 |
Zi | +0.385 |
Catalan | +0.383 |
Niet | +0.380 |
wikipedia | +0.361 |
Neg logit contributions | |
WORK | -0.420 |
liv | -0.335 |
awatts | -0.323 |
WARE | -0.322 |
aturdays | -0.310 |
Token | Sundays |
Token ID | 32714 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
WORK | +0.020 |
liv | +0.017 |
awatts | +0.016 |
WARE | +0.016 |
Appl | +0.015 |
Neg logit contributions | |
| -0.018 |
Zi | -0.017 |
Catalan | -0.017 |
Niet | -0.017 |
wikipedia | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
WORK | +0.115 |
liv | +0.092 |
awatts | +0.088 |
WARE | +0.088 |
Appl | +0.084 |
Neg logit contributions | |
| -0.109 |
Zi | -0.105 |
Catalan | -0.105 |
Niet | -0.104 |
wikipedia | -0.099 |
Token | Erin |
Token ID | 28894 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
| +0.092 |
Zi | +0.089 |
Catalan | +0.088 |
Niet | +0.088 |
Moss | +0.084 |
Neg logit contributions | |
WORK | -0.082 |
liv | -0.064 |
awatts | -0.061 |
WARE | -0.061 |
aturdays | -0.059 |
Token | has |
Token ID | 468 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
| +0.030 |
Zi | +0.029 |
Catalan | +0.028 |
Niet | +0.028 |
wikipedia | +0.027 |
Neg logit contributions | |
WORK | -0.035 |
liv | -0.029 |
awatts | -0.027 |
WARE | -0.027 |
Appl | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
| +0.072 |
Zi | +0.070 |
Catalan | +0.069 |
Niet | +0.069 |
wikipedia | +0.066 |
Neg logit contributions | |
WORK | -0.071 |
liv | -0.056 |
awatts | -0.054 |
WARE | -0.054 |
aturdays | -0.052 |
Token | they |
Token ID | 484 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
| +0.009 |
Zi | +0.009 |
Catalan | +0.009 |
Niet | +0.009 |
wikipedia | +0.008 |
Neg logit contributions | |
WORK | -0.010 |
liv | -0.008 |
awatts | -0.008 |
WARE | -0.008 |
aturdays | -0.007 |
Token | have |
Token ID | 423 |
Feature activation | -8.23e-02 |
Pos logit contributions | |
WORK | +0.033 |
liv | +0.027 |
awatts | +0.026 |
WARE | +0.026 |
aturdays | +0.025 |
Neg logit contributions | |
| -0.028 |
Zi | -0.027 |
Catalan | -0.027 |
Niet | -0.026 |
wikipedia | -0.025 |
Token | now |
Token ID | 783 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
| +0.056 |
Zi | +0.054 |
Catalan | +0.054 |
Niet | +0.054 |
wikipedia | +0.051 |
Neg logit contributions | |
WORK | -0.071 |
liv | -0.058 |
awatts | -0.055 |
WARE | -0.055 |
Appl | -0.053 |
Token | returned |
Token ID | 4504 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
WORK | +0.259 |
liv | +0.212 |
awatts | +0.202 |
WARE | +0.201 |
Appl | +0.195 |
Neg logit contributions | |
| -0.206 |
Zi | -0.200 |
Niet | -0.199 |
Catalan | -0.198 |
wikipedia | -0.188 |
Token | to |
Token ID | 284 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
WORK | +0.388 |
liv | +0.314 |
awatts | +0.299 |
WARE | +0.298 |
Appl | +0.289 |
Neg logit contributions | |
| -0.339 |
Zi | -0.329 |
Catalan | -0.327 |
Niet | -0.327 |
wikipedia | -0.310 |
Token | Ireland |
Token ID | 7517 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
WORK | +0.030 |
liv | +0.026 |
awatts | +0.025 |
WARE | +0.025 |
aturdays | +0.024 |
Neg logit contributions | |
| -0.016 |
Zi | -0.015 |
Catalan | -0.015 |
Niet | -0.015 |
wikipedia | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
| +0.164 |
Zi | +0.158 |
Catalan | +0.157 |
Niet | +0.156 |
wikipedia | +0.149 |
Neg logit contributions | |
WORK | -0.168 |
liv | -0.134 |
awatts | -0.129 |
WARE | -0.128 |
aturdays | -0.124 |
Token | set |
Token ID | 900 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
| +0.017 |
Zi | +0.017 |
Catalan | +0.017 |
Niet | +0.017 |
Moss | +0.016 |
Neg logit contributions | |
WORK | -0.018 |
liv | -0.015 |
awatts | -0.014 |
WARE | -0.014 |
aturdays | -0.014 |
Token | up |
Token ID | 510 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
| +0.325 |
Zi | +0.316 |
Niet | +0.314 |
Catalan | +0.314 |
wikipedia | +0.300 |
Neg logit contributions | |
WORK | -0.322 |
liv | -0.256 |
awatts | -0.243 |
WARE | -0.242 |
Appl | -0.233 |
Token | Black |
Token ID | 2619 |
Feature activation | -9.88e-02 |
Pos logit contributions | |
WORK | +0.367 |
liv | +0.296 |
awatts | +0.278 |
WARE | +0.277 |
Appl | +0.270 |
Neg logit contributions | |
| -0.348 |
Niet | -0.340 |
Zi | -0.339 |
Catalan | -0.337 |
wikipedia | -0.324 |
Token | FT |
Token ID | 9792 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
WORK | +0.038 |
liv | +0.031 |
awatts | +0.029 |
WARE | +0.029 |
aturdays | +0.028 |
Neg logit contributions | |
| -0.032 |
Zi | -0.031 |
Catalan | -0.031 |
Niet | -0.031 |
wikipedia | -0.029 |
Token | should |
Token ID | 815 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
WORK | +0.287 |
liv | +0.233 |
awatts | +0.224 |
WARE | +0.222 |
aturdays | +0.215 |
Neg logit contributions | |
| -0.244 |
Zi | -0.236 |
Catalan | -0.234 |
Niet | -0.234 |
wikipedia | -0.222 |
Token | take |
Token ID | 1011 |
Feature activation | -6.37e-03 |
Pos logit contributions | |
| +0.140 |
Zi | +0.136 |
Catalan | +0.134 |
Niet | +0.133 |
Moss | +0.127 |
Neg logit contributions | |
WORK | -0.145 |
liv | -0.114 |
WARE | -0.111 |
awatts | -0.111 |
aturdays | -0.107 |
Token | their |
Token ID | 511 |
Feature activation | +1.79e-03 |
Pos logit contributions | |
| +0.004 |
Zi | +0.004 |
Catalan | +0.004 |
Niet | +0.004 |
wikipedia | +0.004 |
Neg logit contributions | |
WORK | -0.005 |
liv | -0.004 |
awatts | -0.004 |
WARE | -0.004 |
Appl | -0.004 |
Token | business |
Token ID | 1597 |
Feature activation | +5.52e-01 |
Pos logit contributions | |
WORK | +0.001 |
liv | +0.001 |
awatts | +0.001 |
WARE | +0.001 |
aturdays | +0.001 |
Neg logit contributions | |
| -0.002 |
Zi | -0.002 |
Catalan | -0.001 |
Niet | -0.001 |
Moss | -0.001 |
Token | here |
Token ID | 994 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
WORK | +0.467 |
liv | +0.380 |
awatts | +0.364 |
WARE | +0.362 |
aturdays | +0.350 |
Neg logit contributions | |
| -0.386 |
Zi | -0.374 |
Catalan | -0.371 |
Niet | -0.370 |
wikipedia | -0.351 |
Token | instead |
Token ID | 2427 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
| +0.237 |
Zi | +0.229 |
Catalan | +0.227 |
Niet | +0.226 |
wikipedia | +0.216 |
Neg logit contributions | |
WORK | -0.248 |
liv | -0.198 |
awatts | -0.189 |
WARE | -0.189 |
aturdays | -0.182 |
Token | of |
Token ID | 286 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
| +0.032 |
Zi | +0.032 |
Niet | +0.031 |
Catalan | +0.031 |
wikipedia | +0.030 |
Neg logit contributions | |
WORK | -0.035 |
liv | -0.028 |
awatts | -0.027 |
WARE | -0.027 |
aturdays | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
| +0.015 |
Zi | +0.015 |
Catalan | +0.014 |
Niet | +0.014 |
wikipedia | +0.014 |
Neg logit contributions | |
WORK | -0.017 |
liv | -0.014 |
awatts | -0.013 |
WARE | -0.013 |
aturdays | -0.013 |
Token | Wat |
Token ID | 12242 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
WORK | +0.009 |
liv | +0.007 |
awatts | +0.007 |
WARE | +0.007 |
aturdays | +0.006 |
Neg logit contributions | |
| -0.008 |
Zi | -0.008 |
Catalan | -0.008 |
Niet | -0.008 |
wikipedia | -0.008 |
Token | erville |
Token ID | 33487 |
Feature activation | +6.57e-02 |
Pos logit contributions | |
WORK | +0.310 |
liv | +0.261 |
WARE | +0.255 |
awatts | +0.252 |
aturdays | +0.247 |
Neg logit contributions | |
| -0.157 |
Zi | -0.148 |
Catalan | -0.147 |
Niet | -0.143 |
wikipedia | -0.137 |
Token | Did |
Token ID | 11633 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
WORK | +0.043 |
liv | +0.033 |
awatts | +0.032 |
WARE | +0.032 |
Appl | +0.030 |
Neg logit contributions | |
| -0.049 |
Zi | -0.048 |
Catalan | -0.048 |
Niet | -0.048 |
wikipedia | -0.045 |
Token | Marc |
Token ID | 13067 |
Feature activation | -5.74e-01 |
Pos logit contributions | |
| +0.090 |
Zi | +0.087 |
Catalan | +0.086 |
Niet | +0.086 |
wikipedia | +0.082 |
Neg logit contributions | |
WORK | -0.101 |
liv | -0.081 |
awatts | -0.078 |
WARE | -0.077 |
Appl | -0.075 |
Token | - |
Token ID | 12 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
| +0.380 |
Zi | +0.367 |
Catalan | +0.365 |
Niet | +0.364 |
wikipedia | +0.344 |
Neg logit contributions | |
WORK | -0.507 |
liv | -0.416 |
awatts | -0.400 |
WARE | -0.399 |
aturdays | -0.386 |
Token | Andre |
Token ID | 31258 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
| +0.248 |
Zi | +0.238 |
Catalan | +0.236 |
Niet | +0.235 |
wikipedia | +0.224 |
Neg logit contributions | |
WORK | -0.316 |
liv | -0.259 |
awatts | -0.248 |
WARE | -0.247 |
aturdays | -0.241 |
Token | Fle |
Token ID | 12005 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
| +0.053 |
Zi | +0.051 |
Catalan | +0.050 |
Niet | +0.050 |
Moss | +0.046 |
Neg logit contributions | |
WORK | -0.128 |
liv | -0.108 |
awatts | -0.106 |
WARE | -0.106 |
aturdays | -0.103 |
Token | ury |
Token ID | 1601 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
| +0.172 |
Zi | +0.167 |
Catalan | +0.166 |
Niet | +0.166 |
wikipedia | +0.160 |
Neg logit contributions | |
WORK | -0.114 |
liv | -0.084 |
WARE | -0.080 |
awatts | -0.079 |
aturdays | -0.075 |
Token | solve |
Token ID | 8494 |
Feature activation | -4.93e-01 |
Pos logit contributions | |
| +0.843 |
Zi | +0.816 |
Catalan | +0.806 |
Niet | +0.798 |
Moss | +0.769 |
Neg logit contributions | |
WORK | -0.791 |
liv | -0.611 |
awatts | -0.595 |
WARE | -0.594 |
aturdays | -0.572 |
Token | the |
Token ID | 262 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
| +0.337 |
Zi | +0.326 |
Catalan | +0.324 |
Niet | +0.323 |
wikipedia | +0.306 |
Neg logit contributions | |
WORK | -0.426 |
liv | -0.347 |
awatts | -0.334 |
WARE | -0.332 |
aturdays | -0.321 |
Token | Pittsburgh |
Token ID | 12767 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
| +0.103 |
Zi | +0.100 |
Catalan | +0.099 |
Niet | +0.099 |
wikipedia | +0.093 |
Neg logit contributions | |
WORK | -0.143 |
liv | -0.118 |
awatts | -0.114 |
WARE | -0.113 |
aturdays | -0.110 |
Token | Penguins |
Token ID | 29755 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
WORK | +0.148 |
liv | +0.127 |
awatts | +0.124 |
WARE | +0.124 |
aturdays | +0.121 |
Neg logit contributions | |
| -0.050 |
Zi | -0.047 |
Catalan | -0.047 |
Niet | -0.046 |
wikipedia | -0.042 |
Token | expansion |
Token ID | 7118 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
WORK | +0.039 |
liv | +0.030 |
awatts | +0.029 |
WARE | +0.029 |
aturdays | +0.027 |
Neg logit contributions | |
| -0.047 |
Zi | -0.046 |
Catalan | -0.046 |
Niet | -0.045 |
wikipedia | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -6.94e-02 |
Pos logit contributions | |
| +0.113 |
Zi | +0.110 |
Catalan | +0.110 |
Niet | +0.109 |
wikipedia | +0.105 |
Neg logit contributions | |
WORK | -0.100 |
liv | -0.078 |
awatts | -0.074 |
WARE | -0.074 |
Appl | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
| +0.055 |
Zi | +0.052 |
Catalan | +0.052 |
Niet | +0.050 |
wikipedia | +0.049 |
Neg logit contributions | |
WORK | -0.054 |
liv | -0.043 |
aturdays | -0.042 |
WARE | -0.041 |
awatts | -0.041 |
Token | Twitter |
Token ID | 14254 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
WORK | +0.012 |
liv | +0.009 |
awatts | +0.009 |
WARE | +0.009 |
aturdays | +0.008 |
Neg logit contributions | |
| -0.014 |
Zi | -0.014 |
Catalan | -0.014 |
Niet | -0.014 |
wikipedia | -0.013 |
Token | : |
Token ID | 25 |
Feature activation | +9.60e-02 |
Pos logit contributions | |
WORK | +0.062 |
liv | +0.050 |
awatts | +0.048 |
WARE | +0.048 |
aturdays | +0.046 |
Neg logit contributions | |
| -0.055 |
Zi | -0.053 |
Catalan | -0.053 |
Niet | -0.052 |
wikipedia | -0.050 |
Token | @ |
Token ID | 2488 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
WORK | +0.097 |
liv | +0.082 |
awatts | +0.079 |
WARE | +0.078 |
Appl | +0.076 |
Neg logit contributions | |
| -0.052 |
Zi | -0.050 |
Catalan | -0.049 |
Niet | -0.049 |
wikipedia | -0.046 |
Token | brand |
Token ID | 17938 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
WORK | +0.285 |
liv | +0.229 |
awatts | +0.219 |
WARE | +0.218 |
aturdays | +0.211 |
Neg logit contributions | |
| -0.255 |
Zi | -0.247 |
Catalan | -0.246 |
Niet | -0.244 |
wikipedia | -0.232 |
Token | t |
Token ID | 83 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
| +0.844 |
Zi | +0.818 |
Catalan | +0.813 |
Niet | +0.811 |
wikipedia | +0.775 |
Neg logit contributions | |
WORK | -0.859 |
liv | -0.685 |
WARE | -0.653 |
awatts | -0.652 |
Appl | -0.627 |
Token | st |
Token ID | 301 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
| +0.104 |
Zi | +0.101 |
Catalan | +0.101 |
Niet | +0.100 |
wikipedia | +0.095 |
Neg logit contributions | |
WORK | -0.121 |
liv | -0.098 |
awatts | -0.094 |
WARE | -0.093 |
aturdays | -0.091 |
Token | rib |
Token ID | 822 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
WORK | +0.178 |
liv | +0.144 |
awatts | +0.137 |
WARE | +0.136 |
aturdays | +0.132 |
Neg logit contributions | |
| -0.160 |
Zi | -0.156 |
Catalan | -0.155 |
Niet | -0.154 |
wikipedia | -0.146 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
WORK | +0.281 |
liv | +0.221 |
awatts | +0.209 |
WARE | +0.208 |
Appl | +0.200 |
Neg logit contributions | |
| -0.312 |
Zi | -0.304 |
Catalan | -0.302 |
Niet | -0.301 |
wikipedia | -0.288 |
Token | + |
Token ID | 10 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
WORK | +0.036 |
liv | +0.028 |
awatts | +0.027 |
WARE | +0.027 |
aturdays | +0.026 |
Neg logit contributions | |
| -0.042 |
Zi | -0.041 |
Catalan | -0.040 |
Niet | -0.040 |
wikipedia | -0.039 |
Token | Thirty |
Token ID | 29948 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
WORK | +0.089 |
liv | +0.071 |
awatts | +0.068 |
WARE | +0.068 |
aturdays | +0.065 |
Neg logit contributions | |
| -0.092 |
Zi | -0.089 |
Catalan | -0.089 |
Niet | -0.088 |
wikipedia | -0.084 |
Token | chief |
Token ID | 4039 |
Feature activation | +7.20e-02 |
Pos logit contributions | |
| +0.090 |
Zi | +0.087 |
Niet | +0.087 |
Catalan | +0.086 |
wikipedia | +0.083 |
Neg logit contributions | |
WORK | -0.095 |
liv | -0.077 |
awatts | -0.074 |
WARE | -0.073 |
Appl | -0.070 |
Token | executives |
Token ID | 12353 |
Feature activation | +5.64e-02 |
Pos logit contributions | |
WORK | +0.057 |
liv | +0.046 |
awatts | +0.044 |
WARE | +0.044 |
Appl | +0.042 |
Neg logit contributions | |
| -0.054 |
Zi | -0.052 |
Catalan | -0.052 |
Niet | -0.052 |
wikipedia | -0.050 |
Token | signed |
Token ID | 4488 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
WORK | +0.046 |
liv | +0.037 |
awatts | +0.035 |
WARE | +0.035 |
Appl | +0.034 |
Neg logit contributions | |
| -0.041 |
Zi | -0.040 |
Catalan | -0.040 |
Niet | -0.040 |
wikipedia | -0.038 |
Token | a |
Token ID | 257 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
| +0.256 |
Zi | +0.248 |
Niet | +0.246 |
Catalan | +0.246 |
wikipedia | +0.233 |
Neg logit contributions | |
WORK | -0.343 |
liv | -0.282 |
awatts | -0.270 |
WARE | -0.268 |
Appl | -0.261 |
Token | letter |
Token ID | 3850 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
WORK | +0.083 |
liv | +0.069 |
awatts | +0.065 |
WARE | +0.065 |
Appl | +0.063 |
Neg logit contributions | |
| -0.060 |
Zi | -0.058 |
Niet | -0.058 |
Catalan | -0.058 |
wikipedia | -0.055 |
Token | saying |
Token ID | 2282 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
| +0.678 |
Zi | +0.658 |
Niet | +0.654 |
Catalan | +0.654 |
wikipedia | +0.624 |
Neg logit contributions | |
WORK | -0.724 |
liv | -0.582 |
awatts | -0.552 |
WARE | -0.551 |
Appl | -0.535 |
Token | that |
Token ID | 326 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
WORK | +0.207 |
liv | +0.167 |
awatts | +0.159 |
WARE | +0.158 |
Appl | +0.153 |
Neg logit contributions | |
| -0.187 |
Zi | -0.181 |
Catalan | -0.180 |
Niet | -0.180 |
wikipedia | -0.172 |
Token | breaking |
Token ID | 7163 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
WORK | +0.031 |
liv | +0.025 |
awatts | +0.024 |
WARE | +0.024 |
aturdays | +0.023 |
Neg logit contributions | |
| -0.026 |
Zi | -0.025 |
Catalan | -0.025 |
Niet | -0.025 |
wikipedia | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +6.54e-03 |
Pos logit contributions | |
WORK | +0.568 |
liv | +0.458 |
awatts | +0.429 |
WARE | +0.428 |
Appl | +0.416 |
Neg logit contributions | |
| -0.547 |
Niet | -0.535 |
Zi | -0.535 |
Catalan | -0.531 |
wikipedia | -0.508 |
Token | arms |
Token ID | 5101 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
WORK | +0.006 |
liv | +0.005 |
awatts | +0.005 |
WARE | +0.005 |
Appl | +0.004 |
Neg logit contributions | |
| -0.004 |
Zi | -0.004 |
Niet | -0.004 |
Catalan | -0.004 |
wikipedia | -0.004 |
Token | ( |
Token ID | 357 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
| +0.391 |
Zi | +0.379 |
Catalan | +0.376 |
Niet | +0.375 |
wikipedia | +0.356 |
Neg logit contributions | |
WORK | -0.449 |
liv | -0.363 |
awatts | -0.348 |
WARE | -0.347 |
aturdays | -0.334 |
Token | 1989 |
Token ID | 25475 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
| +0.161 |
Zi | +0.156 |
Catalan | +0.155 |
Niet | +0.154 |
wikipedia | +0.147 |
Neg logit contributions | |
WORK | -0.171 |
liv | -0.138 |
awatts | -0.131 |
WARE | -0.131 |
aturdays | -0.126 |
Token | – |
Token ID | 1906 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
| +0.257 |
Zi | +0.249 |
Catalan | +0.247 |
Niet | +0.246 |
wikipedia | +0.235 |
Neg logit contributions | |
WORK | -0.282 |
liv | -0.227 |
awatts | -0.217 |
WARE | -0.216 |
aturdays | -0.209 |
Token | 99 |
Token ID | 2079 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
| +0.084 |
Zi | +0.081 |
Catalan | +0.081 |
Niet | +0.080 |
wikipedia | +0.077 |
Neg logit contributions | |
WORK | -0.094 |
liv | -0.076 |
awatts | -0.073 |
WARE | -0.073 |
aturdays | -0.071 |
Token | ) |
Token ID | 8 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
| +0.145 |
Zi | +0.140 |
Catalan | +0.140 |
Niet | +0.139 |
wikipedia | +0.133 |
Neg logit contributions | |
WORK | -0.151 |
liv | -0.121 |
awatts | -0.116 |
WARE | -0.115 |
Appl | -0.111 |
Token | turned |
Token ID | 2900 |
Feature activation | -9.42e-01 |
Pos logit contributions | |
| +0.120 |
Zi | +0.116 |
Catalan | +0.116 |
Niet | +0.115 |
wikipedia | +0.109 |
Neg logit contributions | |
WORK | -0.150 |
liv | -0.122 |
awatts | -0.117 |
WARE | -0.117 |
aturdays | -0.113 |
Token | to |
Token ID | 284 |
Feature activation | +8.86e-02 |
Pos logit contributions | |
| +0.714 |
Zi | +0.692 |
Catalan | +0.688 |
Niet | +0.684 |
Moss | +0.651 |
Neg logit contributions | |
WORK | -0.748 |
liv | -0.594 |
awatts | -0.573 |
WARE | -0.570 |
aturdays | -0.550 |
Token | a |
Token ID | 257 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
WORK | +0.072 |
liv | +0.058 |
awatts | +0.056 |
WARE | +0.055 |
Appl | +0.054 |
Neg logit contributions | |
| -0.065 |
Zi | -0.063 |
Catalan | -0.062 |
Niet | -0.062 |
wikipedia | -0.059 |
Token | neoliberal |
Token ID | 31972 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
WORK | +0.040 |
liv | +0.032 |
awatts | +0.031 |
WARE | +0.031 |
Appl | +0.029 |
Neg logit contributions | |
| -0.038 |
Zi | -0.037 |
Catalan | -0.036 |
Niet | -0.036 |
wikipedia | -0.035 |
Token | model |
Token ID | 2746 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
| +0.136 |
Zi | +0.132 |
Catalan | +0.132 |
Niet | +0.131 |
wikipedia | +0.126 |
Neg logit contributions | |
WORK | -0.125 |
liv | -0.098 |
awatts | -0.093 |
WARE | -0.093 |
Appl | -0.089 |
Token | only |
Token ID | 691 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
| +0.362 |
Zi | +0.351 |
Catalan | +0.349 |
Niet | +0.348 |
wikipedia | +0.331 |
Neg logit contributions | |
WORK | -0.406 |
liv | -0.328 |
awatts | -0.314 |
WARE | -0.312 |
aturdays | -0.301 |
Token | and |
Token ID | 290 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
WORK | +0.096 |
liv | +0.076 |
awatts | +0.073 |
WARE | +0.072 |
aturdays | +0.070 |
Neg logit contributions | |
| -0.097 |
Zi | -0.094 |
Catalan | -0.093 |
Niet | -0.093 |
wikipedia | -0.089 |
Token | is |
Token ID | 318 |
Feature activation | +7.33e-02 |
Pos logit contributions | |
WORK | +0.084 |
liv | +0.067 |
awatts | +0.064 |
WARE | +0.064 |
aturdays | +0.062 |
Neg logit contributions | |
| -0.077 |
Zi | -0.074 |
Catalan | -0.074 |
Niet | -0.074 |
wikipedia | -0.070 |
Token | flanked |
Token ID | 49059 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
WORK | +0.062 |
liv | +0.050 |
awatts | +0.049 |
WARE | +0.048 |
aturdays | +0.047 |
Neg logit contributions | |
| -0.051 |
Zi | -0.050 |
Catalan | -0.049 |
Niet | -0.049 |
wikipedia | -0.047 |
Token | by |
Token ID | 416 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
WORK | +0.701 |
liv | +0.548 |
awatts | +0.514 |
WARE | +0.505 |
Appl | +0.490 |
Neg logit contributions | |
| -0.851 |
Zi | -0.834 |
Niet | -0.833 |
Catalan | -0.831 |
wikipedia | -0.797 |
Token | the |
Token ID | 262 |
Feature activation | -8.59e-03 |
Pos logit contributions | |
WORK | +0.027 |
liv | +0.022 |
awatts | +0.021 |
WARE | +0.021 |
Appl | +0.020 |
Neg logit contributions | |
| -0.021 |
Zi | -0.020 |
Niet | -0.020 |
Catalan | -0.020 |
wikipedia | -0.019 |
Token | Caller |
Token ID | 10244 |
Feature activation | -8.87e-01 |
Pos logit contributions | |
| +0.006 |
Zi | +0.006 |
Catalan | +0.006 |
Niet | +0.006 |
wikipedia | +0.006 |
Neg logit contributions | |
WORK | -0.007 |
liv | -0.006 |
awatts | -0.006 |
WARE | -0.006 |
Appl | -0.005 |
Token | � |
Token ID | 447 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
| +0.712 |
Zi | +0.691 |
Catalan | +0.687 |
Niet | +0.684 |
wikipedia | +0.653 |
Neg logit contributions | |
WORK | -0.659 |
liv | -0.519 |
awatts | -0.496 |
WARE | -0.493 |
aturdays | -0.473 |
Token | � |
Token ID | 247 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
| +0.082 |
Niet | +0.080 |
Zi | +0.080 |
Catalan | +0.079 |
wikipedia | +0.075 |
Neg logit contributions | |
WORK | -0.119 |
liv | -0.099 |
awatts | -0.095 |
WARE | -0.095 |
Appl | -0.092 |
Token | s |
Token ID | 82 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
WORK | +0.042 |
liv | +0.033 |
awatts | +0.032 |
WARE | +0.032 |
aturdays | +0.030 |
Neg logit contributions | |
| -0.043 |
Zi | -0.041 |
Catalan | -0.041 |
Niet | -0.041 |
wikipedia | -0.039 |
Token | video |
Token ID | 2008 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
| +0.171 |
Zi | +0.166 |
Catalan | +0.165 |
Niet | +0.164 |
wikipedia | +0.156 |
Neg logit contributions | |
WORK | -0.188 |
liv | -0.151 |
awatts | -0.145 |
WARE | -0.144 |
aturdays | -0.139 |
Token | producer |
Token ID | 9920 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
WORK | +0.192 |
liv | +0.156 |
awatts | +0.149 |
WARE | +0.149 |
Appl | +0.144 |
Neg logit contributions | |
| -0.156 |
Zi | -0.151 |
Niet | -0.150 |
Catalan | -0.150 |
wikipedia | -0.142 |