Token | her |
Token ID | 607 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
Baby | +0.107 |
dessert | +0.097 |
rice | +0.094 |
homeless | +0.092 |
onda | +0.092 |
Neg logit contributions | |
bs | -0.065 |
counts | -0.065 |
falls | -0.063 |
hops | -0.063 |
gers | -0.063 |
Token | hair |
Token ID | 4190 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
gers | +0.040 |
falls | +0.040 |
counts | +0.039 |
hops | +0.039 |
ways | +0.039 |
Neg logit contributions | |
Baby | -0.055 |
dessert | -0.052 |
rice | -0.048 |
homeless | -0.047 |
Son | -0.047 |
Token | in |
Token ID | 287 |
Feature activation | -8.85e-01 |
Pos logit contributions | |
hops | +0.123 |
counts | +0.122 |
hits | +0.114 |
bs | +0.114 |
falls | +0.113 |
Neg logit contributions | |
Baby | -0.196 |
dessert | -0.182 |
rice | -0.174 |
homeless | -0.172 |
pudding | -0.171 |
Token | a |
Token ID | 257 |
Feature activation | -7.65e-01 |
Pos logit contributions | |
Baby | +0.061 |
onda | +0.056 |
homeless | +0.053 |
essa | +0.052 |
uf | +0.052 |
Neg logit contributions | |
bs | -0.032 |
low | -0.030 |
around | -0.030 |
counts | -0.030 |
gers | -0.030 |
Token | pony |
Token ID | 26902 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
Baby | +0.044 |
dessert | +0.039 |
homeless | +0.038 |
rice | +0.037 |
onda | +0.037 |
Neg logit contributions | |
counts | -0.037 |
bs | -0.037 |
gers | -0.035 |
falls | -0.035 |
hits | -0.035 |
Token | tail |
Token ID | 13199 |
Feature activation | +8.41e+00 |
Pos logit contributions | |
bs | +0.130 |
hits | +0.129 |
falls | +0.128 |
counts | +0.127 |
hops | +0.127 |
Neg logit contributions | |
Baby | -0.247 |
dessert | -0.223 |
homeless | -0.217 |
rice | -0.215 |
onda | -0.215 |
Token | so |
Token ID | 523 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
hops | +0.342 |
spins | +0.330 |
counts | +0.328 |
gers | +0.324 |
bs | +0.323 |
Neg logit contributions | |
Baby | -0.513 |
dessert | -0.469 |
onda | -0.466 |
homeless | -0.460 |
uf | -0.448 |
Token | she |
Token ID | 673 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
Baby | +0.095 |
dessert | +0.087 |
homeless | +0.085 |
onda | +0.084 |
essa | +0.083 |
Neg logit contributions | |
counts | -0.036 |
hits | -0.035 |
gers | -0.034 |
hops | -0.034 |
bs | -0.032 |
Token | could |
Token ID | 714 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
bs | +0.082 |
gers | +0.081 |
counts | +0.080 |
hits | +0.078 |
hops | +0.077 |
Neg logit contributions | |
Baby | -0.166 |
onda | -0.148 |
dessert | -0.145 |
homeless | -0.145 |
essa | -0.142 |
Token | look |
Token ID | 804 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
bs | +0.102 |
gers | +0.099 |
counts | +0.097 |
hits | +0.096 |
falls | +0.096 |
Neg logit contributions | |
Baby | -0.121 |
onda | -0.105 |
dessert | -0.104 |
homeless | -0.102 |
essa | -0.100 |
Token | pretty |
Token ID | 2495 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
Baby | +0.016 |
dessert | +0.015 |
homeless | +0.014 |
rice | +0.014 |
Son | +0.014 |
Neg logit contributions | |
counts | -0.010 |
hops | -0.009 |
gers | -0.009 |
hits | -0.009 |
opl | -0.009 |
Token | with |
Token ID | 351 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
bs | +0.028 |
gers | +0.028 |
hits | +0.027 |
falls | +0.026 |
chairs | +0.026 |
Neg logit contributions | |
Baby | -0.033 |
dessert | -0.031 |
homeless | -0.030 |
onda | -0.029 |
cooking | -0.029 |
Token | glasses |
Token ID | 15232 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
gers | +0.090 |
bs | +0.090 |
counts | +0.088 |
falls | +0.088 |
hits | +0.088 |
Neg logit contributions | |
Baby | -0.075 |
dessert | -0.069 |
homeless | -0.068 |
rice | -0.066 |
onda | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
bs | +0.073 |
around | +0.067 |
counts | +0.067 |
gers | +0.066 |
falls | +0.064 |
Neg logit contributions | |
Baby | -0.238 |
onda | -0.217 |
homeless | -0.210 |
dessert | -0.208 |
Son | -0.207 |
Token | a |
Token ID | 257 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
gers | +0.008 |
ines | +0.008 |
bs | +0.008 |
around | +0.008 |
ang | +0.008 |
Neg logit contributions | |
Baby | -0.006 |
dessert | -0.006 |
rice | -0.006 |
homeless | -0.005 |
essa | -0.005 |
Token | pony |
Token ID | 26902 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
Baby | +0.007 |
dessert | +0.006 |
homeless | +0.006 |
rice | +0.006 |
onda | +0.006 |
Neg logit contributions | |
counts | -0.004 |
bs | -0.004 |
gers | -0.004 |
around | -0.004 |
chairs | -0.004 |
Token | tail |
Token ID | 13199 |
Feature activation | +8.34e+00 |
Pos logit contributions | |
bs | +0.139 |
gers | +0.130 |
counts | +0.130 |
falls | +0.127 |
low | +0.126 |
Neg logit contributions | |
Baby | -0.257 |
dessert | -0.242 |
homeless | -0.238 |
onda | -0.235 |
rice | -0.231 |
Token | . |
Token ID | 13 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
spins | +0.367 |
gers | +0.365 |
hops | +0.363 |
chairs | +0.359 |
bs | +0.358 |
Neg logit contributions | |
homeless | -0.394 |
onda | -0.384 |
dessert | -0.377 |
cooking | -0.375 |
Baby | -0.375 |
Token | She |
Token ID | 1375 |
Feature activation | +8.48e-01 |
Pos logit contributions | |
gers | +0.009 |
counts | +0.009 |
ines | +0.009 |
rounds | +0.009 |
hits | +0.009 |
Neg logit contributions | |
Baby | -0.013 |
rice | -0.011 |
dessert | -0.011 |
Baby | -0.011 |
homeless | -0.011 |
Token | is |
Token ID | 318 |
Feature activation | +6.20e-01 |
Pos logit contributions | |
bs | +0.028 |
gers | +0.027 |
low | +0.025 |
chair | +0.024 |
middle | +0.024 |
Neg logit contributions | |
Baby | -0.058 |
onda | -0.054 |
Sonny | -0.053 |
homeless | -0.053 |
Son | -0.053 |
Token | reading |
Token ID | 3555 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
bs | +0.027 |
around | +0.026 |
counts | +0.026 |
low | +0.026 |
gers | +0.026 |
Neg logit contributions | |
Baby | -0.038 |
onda | -0.034 |
homeless | -0.032 |
dessert | -0.032 |
essa | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
Baby | +0.022 |
dessert | +0.021 |
rice | +0.020 |
homeless | +0.020 |
onda | +0.019 |
Neg logit contributions | |
bs | -0.027 |
hits | -0.026 |
hops | -0.026 |
falls | -0.026 |
counts | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
Baby | +0.177 |
onda | +0.162 |
dessert | +0.161 |
homeless | +0.160 |
Son | +0.157 |
Neg logit contributions | |
bs | -0.050 |
gers | -0.046 |
counts | -0.045 |
hits | -0.042 |
hops | -0.042 |
Token | Emma |
Token ID | 18966 |
Feature activation | -4.71e+00 |
Pos logit contributions | |
Baby | +0.010 |
onda | +0.009 |
homeless | +0.008 |
dessert | +0.008 |
essa | +0.008 |
Neg logit contributions | |
bs | -0.016 |
counts | -0.015 |
gers | -0.015 |
around | -0.015 |
falls | -0.015 |
Token | decided |
Token ID | 3066 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
Baby | +0.351 |
onda | +0.318 |
dessert | +0.313 |
homeless | +0.311 |
essa | +0.303 |
Neg logit contributions | |
bs | -0.148 |
counts | -0.138 |
falls | -0.135 |
around | -0.134 |
gers | -0.134 |
Token | a |
Token ID | 257 |
Feature activation | -9.64e-01 |
Pos logit contributions | |
Baby | +0.021 |
onda | +0.019 |
homeless | +0.018 |
dessert | +0.018 |
Son | +0.017 |
Neg logit contributions | |
bs | -0.013 |
around | -0.012 |
counts | -0.012 |
gers | -0.012 |
falls | -0.012 |
Token | pony |
Token ID | 26902 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
Baby | +0.059 |
onda | +0.053 |
essa | +0.050 |
urg | +0.049 |
homeless | +0.049 |
Neg logit contributions | |
bs | -0.043 |
low | -0.041 |
falls | -0.040 |
gers | -0.040 |
hops | -0.040 |
Token | tail |
Token ID | 13199 |
Feature activation | +8.08e+00 |
Pos logit contributions | |
bs | +0.132 |
gers | +0.123 |
falls | +0.123 |
hits | +0.121 |
low | +0.121 |
Neg logit contributions | |
Baby | -0.216 |
homeless | -0.198 |
onda | -0.196 |
dessert | -0.194 |
essa | -0.190 |
Token | wasn |
Token ID | 2492 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
bs | +0.307 |
gers | +0.298 |
hops | +0.297 |
sideways | +0.293 |
counts | +0.292 |
Neg logit contributions | |
Baby | -0.496 |
onda | -0.464 |
homeless | -0.462 |
dessert | -0.461 |
essa | -0.445 |
Token | 't |
Token ID | 470 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
Baby | +0.046 |
onda | +0.044 |
essa | +0.042 |
rice | +0.042 |
urg | +0.042 |
Neg logit contributions | |
around | -0.025 |
bs | -0.025 |
low | -0.024 |
falls | -0.024 |
hits | -0.024 |
Token | so |
Token ID | 523 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Baby | +0.052 |
onda | +0.048 |
uf | +0.046 |
urg | +0.045 |
Son | +0.045 |
Neg logit contributions | |
around | -0.033 |
low | -0.032 |
bs | -0.031 |
hits | -0.028 |
until | -0.028 |
Token | important |
Token ID | 1593 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
Baby | +0.093 |
homeless | +0.083 |
dessert | +0.082 |
onda | +0.081 |
essa | +0.078 |
Neg logit contributions | |
gers | -0.063 |
counts | -0.063 |
bs | -0.062 |
hits | -0.061 |
hops | -0.061 |
Token | after |
Token ID | 706 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
Baby | +0.146 |
onda | +0.132 |
homeless | +0.129 |
dessert | +0.129 |
essa | +0.126 |
Neg logit contributions | |
bs | -0.059 |
counts | -0.055 |
gers | -0.055 |
falls | -0.053 |
hits | -0.053 |
Token | hands |
Token ID | 2832 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
bs | +0.021 |
gers | +0.020 |
counts | +0.020 |
chair | +0.019 |
hits | +0.019 |
Neg logit contributions | |
Baby | -0.040 |
onda | -0.036 |
dessert | -0.036 |
homeless | -0.035 |
rice | -0.035 |
Token | too |
Token ID | 1165 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
bs | +0.004 |
gers | +0.004 |
counts | +0.004 |
hits | +0.003 |
falls | +0.003 |
Neg logit contributions | |
Baby | -0.010 |
onda | -0.009 |
dessert | -0.009 |
homeless | -0.009 |
essa | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
bs | +0.018 |
counts | +0.017 |
gers | +0.016 |
falls | +0.016 |
around | +0.016 |
Neg logit contributions | |
Baby | -0.057 |
onda | -0.052 |
homeless | -0.051 |
dessert | -0.050 |
essa | -0.050 |
Token | He |
Token ID | 679 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
bs | +0.024 |
gers | +0.023 |
hops | +0.023 |
tail | +0.023 |
lers | +0.023 |
Neg logit contributions | |
Baby | -0.031 |
dessert | -0.027 |
Son | -0.027 |
Baby | -0.026 |
Sonny | -0.026 |
Token | says |
Token ID | 1139 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
bs | +0.015 |
gers | +0.014 |
tail | +0.013 |
chair | +0.013 |
middle | +0.013 |
Neg logit contributions | |
Baby | -0.026 |
pudding | -0.025 |
onda | -0.024 |
dessert | -0.024 |
rice | -0.024 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.55e+00 |
Pos logit contributions | |
bs | +0.057 |
counts | +0.054 |
gers | +0.054 |
hits | +0.051 |
falls | +0.051 |
Neg logit contributions | |
Baby | -0.137 |
dessert | -0.125 |
onda | -0.122 |
homeless | -0.122 |
rice | -0.120 |
Token | o |
Token ID | 78 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
bs | +0.333 |
hops | +0.308 |
gers | +0.304 |
low | +0.304 |
falls | +0.303 |
Neg logit contributions | |
Baby | -0.462 |
onda | -0.420 |
dessert | -0.408 |
essa | -0.399 |
Son | -0.388 |
Token | . |
Token ID | 13 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
Baby | +0.030 |
onda | +0.028 |
dessert | +0.027 |
homeless | +0.027 |
rice | +0.027 |
Neg logit contributions | |
bs | -0.014 |
low | -0.013 |
chair | -0.012 |
gers | -0.012 |
chairs | -0.012 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
bs | +0.025 |
gers | +0.024 |
hops | +0.024 |
counts | +0.024 |
falls | +0.023 |
Neg logit contributions | |
Baby | -0.038 |
dessert | -0.032 |
Son | -0.032 |
homeless | -0.032 |
Sonny | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
bs | +0.034 |
gers | +0.032 |
low | +0.030 |
chair | +0.029 |
middle | +0.029 |
Neg logit contributions | |
Baby | -0.071 |
onda | -0.064 |
dessert | -0.063 |
homeless | -0.062 |
pudding | -0.061 |
Token | Anna |
Token ID | 11735 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
gers | +0.023 |
bs | +0.023 |
opl | +0.022 |
ines | +0.022 |
falls | +0.022 |
Neg logit contributions | |
Baby | -0.022 |
dessert | -0.019 |
homeless | -0.018 |
pudding | -0.018 |
Baby | -0.017 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
bs | +0.038 |
gers | +0.037 |
ines | +0.036 |
chairs | +0.036 |
ars | +0.035 |
Neg logit contributions | |
Baby | -0.052 |
dessert | -0.048 |
homeless | -0.046 |
rice | -0.045 |
Son | -0.044 |
Token | wanted |
Token ID | 2227 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
gers | +0.049 |
bs | +0.049 |
chair | +0.047 |
hits | +0.046 |
counts | +0.046 |
Neg logit contributions | |
Baby | -0.072 |
onda | -0.064 |
homeless | -0.062 |
dessert | -0.061 |
uf | -0.061 |
Token | to |
Token ID | 284 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
bs | +0.043 |
counts | +0.041 |
gers | +0.040 |
falls | +0.040 |
hits | +0.040 |
Neg logit contributions | |
Baby | -0.059 |
onda | -0.052 |
homeless | -0.050 |
dessert | -0.050 |
essa | -0.049 |
Token | see |
Token ID | 766 |
Feature activation | +3.74e+00 |
Pos logit contributions | |
bs | +0.015 |
counts | +0.014 |
gers | +0.014 |
falls | +0.014 |
hits | +0.014 |
Neg logit contributions | |
Baby | -0.024 |
onda | -0.022 |
homeless | -0.021 |
dessert | -0.021 |
essa | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
gers | +0.209 |
bs | +0.209 |
hops | +0.197 |
counts | +0.197 |
falls | +0.197 |
Neg logit contributions | |
Baby | -0.182 |
dessert | -0.162 |
onda | -0.155 |
homeless | -0.154 |
rice | -0.152 |
Token | ze |
Token ID | 41271 |
Feature activation | +7.50e+00 |
Pos logit contributions | |
bs | +0.066 |
counts | +0.059 |
chair | +0.058 |
falls | +0.057 |
gers | +0.057 |
Neg logit contributions | |
Baby | -0.089 |
onda | -0.082 |
homeless | -0.081 |
dessert | -0.080 |
rice | -0.077 |
Token | br |
Token ID | 1671 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
low | +0.211 |
sideways | +0.210 |
gers | +0.209 |
counts | +0.209 |
middle | +0.207 |
Neg logit contributions | |
Baby | -0.511 |
dessert | -0.478 |
onda | -0.474 |
uf | -0.470 |
pudding | -0.464 |
Token | as |
Token ID | 292 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
chair | +0.173 |
falls | +0.173 |
counts | +0.172 |
bs | +0.172 |
gers | +0.171 |
Neg logit contributions | |
Baby | -0.199 |
onda | -0.180 |
urg | -0.167 |
uggets | -0.165 |
essa | -0.165 |
Token | the |
Token ID | 262 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
bs | +0.075 |
counts | +0.070 |
around | +0.069 |
gers | +0.068 |
falls | +0.068 |
Neg logit contributions | |
Baby | -0.209 |
onda | -0.191 |
homeless | -0.185 |
dessert | -0.184 |
essa | -0.181 |
Token | most |
Token ID | 749 |
Feature activation | +4.49e+00 |
Pos logit contributions | |
bs | +0.058 |
chair | +0.051 |
falls | +0.050 |
gers | +0.049 |
counts | +0.049 |
Neg logit contributions | |
Baby | -0.082 |
onda | -0.073 |
dessert | -0.072 |
homeless | -0.071 |
Son | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
bs | +0.195 |
gers | +0.183 |
falls | +0.181 |
counts | +0.177 |
hits | +0.177 |
Neg logit contributions | |
Baby | -0.277 |
dessert | -0.255 |
onda | -0.254 |
homeless | -0.247 |
rice | -0.240 |
Token | . |
Token ID | 13 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
bs | +0.095 |
chairs | +0.093 |
counts | +0.091 |
chair | +0.086 |
hits | +0.084 |
Neg logit contributions | |
Baby | -0.120 |
ful | -0.117 |
dessert | -0.106 |
onda | -0.106 |
rice | -0.105 |
Token | " |
Token ID | 366 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
bs | +0.023 |
gers | +0.022 |
counts | +0.021 |
around | +0.020 |
falls | +0.020 |
Neg logit contributions | |
Baby | -0.045 |
dessert | -0.039 |
homeless | -0.039 |
onda | -0.039 |
Son | -0.039 |
Token | I |
Token ID | 40 |
Feature activation | -7.23e-01 |
Pos logit contributions | |
Baby | +0.017 |
onda | +0.015 |
homeless | +0.014 |
dessert | +0.014 |
Son | +0.014 |
Neg logit contributions | |
bs | -0.018 |
counts | -0.017 |
gers | -0.017 |
falls | -0.017 |
hits | -0.017 |
Token | 'm |
Token ID | 1101 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
Baby | +0.041 |
onda | +0.036 |
homeless | +0.036 |
dessert | +0.035 |
uf | +0.034 |
Neg logit contributions | |
chairs | -0.032 |
chair | -0.032 |
gers | -0.031 |
ways | -0.031 |
lights | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
Baby | +0.132 |
onda | +0.117 |
homeless | +0.114 |
dessert | +0.113 |
essa | +0.110 |
Neg logit contributions | |
bs | -0.094 |
counts | -0.090 |
gers | -0.089 |
falls | -0.088 |
hits | -0.087 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.40e+00 |
Pos logit contributions | |
bs | +0.046 |
counts | +0.042 |
falls | +0.042 |
chair | +0.042 |
gers | +0.042 |
Neg logit contributions | |
Baby | -0.059 |
onda | -0.053 |
homeless | -0.052 |
dessert | -0.051 |
rice | -0.049 |
Token | est |
Token ID | 395 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
bs | +0.291 |
low | +0.263 |
falls | +0.260 |
chairs | +0.256 |
gers | +0.253 |
Neg logit contributions | |
Baby | -0.504 |
onda | -0.459 |
dessert | -0.440 |
essa | -0.434 |
urg | -0.430 |
Token | bird |
Token ID | 6512 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
bs | +0.150 |
counts | +0.137 |
gers | +0.136 |
falls | +0.134 |
low | +0.134 |
Neg logit contributions | |
Baby | -0.176 |
dessert | -0.159 |
onda | -0.159 |
homeless | -0.156 |
rice | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
gers | +0.107 |
bs | +0.107 |
ways | +0.107 |
chairs | +0.104 |
low | +0.103 |
Neg logit contributions | |
Baby | -0.111 |
dessert | -0.097 |
onda | -0.097 |
ful | -0.094 |
homeless | -0.093 |
Token | I |
Token ID | 314 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
gers | +0.029 |
bs | +0.029 |
ines | +0.028 |
rounds | +0.027 |
ls | +0.027 |
Neg logit contributions | |
Baby | -0.028 |
Son | -0.022 |
uf | -0.022 |
Sonny | -0.022 |
Baby | -0.022 |
Token | will |
Token ID | 481 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
Baby | +0.034 |
onda | +0.030 |
homeless | +0.030 |
dessert | +0.029 |
pudding | +0.027 |
Neg logit contributions | |
chair | -0.028 |
gers | -0.028 |
chairs | -0.028 |
ways | -0.027 |
arms | -0.026 |
Token | dal |
Token ID | 31748 |
Feature activation | +8.05e-01 |
Pos logit contributions | |
Baby | +0.034 |
onda | +0.030 |
essa | +0.029 |
uf | +0.029 |
Son | +0.029 |
Neg logit contributions | |
bs | -0.026 |
gers | -0.025 |
low | -0.024 |
counts | -0.024 |
around | -0.024 |
Token | ene |
Token ID | 1734 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
bs | +0.036 |
hits | +0.035 |
around | +0.035 |
falls | +0.034 |
gers | +0.034 |
Neg logit contributions | |
Baby | -0.047 |
dessert | -0.045 |
homeless | -0.044 |
onda | -0.041 |
rice | -0.041 |
Token | will |
Token ID | 481 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
Baby | +0.136 |
onda | +0.121 |
homeless | +0.117 |
essa | +0.116 |
Son | +0.116 |
Neg logit contributions | |
bs | -0.099 |
gers | -0.096 |
low | -0.094 |
counts | -0.094 |
falls | -0.093 |
Token | be |
Token ID | 307 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
bs | +0.095 |
around | +0.093 |
counts | +0.092 |
low | +0.091 |
hops | +0.091 |
Neg logit contributions | |
Baby | -0.135 |
onda | -0.120 |
dessert | -0.115 |
homeless | -0.115 |
essa | -0.112 |
Token | the |
Token ID | 262 |
Feature activation | +6.76e-01 |
Pos logit contributions | |
Baby | +0.056 |
onda | +0.049 |
homeless | +0.049 |
dessert | +0.048 |
Son | +0.046 |
Neg logit contributions | |
bs | -0.051 |
gers | -0.050 |
counts | -0.048 |
falls | -0.047 |
chair | -0.047 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.39e+00 |
Pos logit contributions | |
bs | +0.027 |
counts | +0.027 |
gers | +0.027 |
low | +0.026 |
around | +0.026 |
Neg logit contributions | |
Baby | -0.044 |
onda | -0.039 |
homeless | -0.037 |
dessert | -0.037 |
essa | -0.036 |
Token | est |
Token ID | 395 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
bs | +0.244 |
low | +0.224 |
falls | +0.223 |
counts | +0.219 |
gers | +0.217 |
Neg logit contributions | |
Baby | -0.544 |
onda | -0.493 |
essa | -0.476 |
dessert | -0.475 |
homeless | -0.472 |
Token | , |
Token ID | 11 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
bs | +0.119 |
counts | +0.114 |
gers | +0.113 |
around | +0.111 |
falls | +0.111 |
Neg logit contributions | |
Baby | -0.210 |
onda | -0.188 |
homeless | -0.182 |
dessert | -0.180 |
essa | -0.176 |
Token | kind |
Token ID | 1611 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
bs | +0.010 |
gers | +0.009 |
counts | +0.009 |
hits | +0.009 |
falls | +0.009 |
Neg logit contributions | |
Baby | -0.013 |
onda | -0.011 |
homeless | -0.011 |
dessert | -0.011 |
Son | -0.011 |
Token | est |
Token ID | 395 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
bs | +0.064 |
gers | +0.062 |
around | +0.061 |
counts | +0.060 |
likes | +0.060 |
Neg logit contributions | |
Baby | -0.109 |
onda | -0.097 |
dessert | -0.095 |
homeless | -0.094 |
uf | -0.091 |
Token | and |
Token ID | 290 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
bs | +0.133 |
counts | +0.127 |
gers | +0.127 |
hops | +0.124 |
falls | +0.124 |
Neg logit contributions | |
Baby | -0.192 |
onda | -0.171 |
homeless | -0.166 |
dessert | -0.164 |
essa | -0.160 |
Token | " |
Token ID | 1 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
bs | +0.017 |
counts | +0.017 |
gers | +0.017 |
hits | +0.017 |
hops | +0.016 |
Neg logit contributions | |
Baby | -0.026 |
onda | -0.022 |
homeless | -0.021 |
dessert | -0.021 |
Son | -0.021 |
Token | B |
Token ID | 33 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
urg | +0.034 |
Baby | +0.033 |
uf | +0.032 |
onda | +0.031 |
homeless | +0.031 |
Neg logit contributions | |
gers | -0.023 |
counts | -0.022 |
bs | -0.022 |
likes | -0.022 |
low | -0.021 |
Token | rav |
Token ID | 4108 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
Baby | +0.129 |
onda | +0.118 |
homeless | +0.117 |
dessert | +0.116 |
rice | +0.113 |
Neg logit contributions | |
bs | -0.093 |
gers | -0.089 |
counts | -0.086 |
falls | -0.085 |
likes | -0.083 |
Token | o |
Token ID | 78 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
bs | +0.044 |
counts | +0.041 |
likes | +0.040 |
around | +0.040 |
falls | +0.040 |
Neg logit contributions | |
Baby | -0.077 |
onda | -0.070 |
homeless | -0.069 |
dessert | -0.068 |
Sonny | -0.067 |
Token | , |
Token ID | 11 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
Baby | +0.045 |
onda | +0.041 |
homeless | +0.041 |
dessert | +0.040 |
essa | +0.040 |
Neg logit contributions | |
bs | -0.009 |
gers | -0.008 |
counts | -0.008 |
falls | -0.008 |
around | -0.008 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.33e+00 |
Pos logit contributions | |
bs | +0.030 |
likes | +0.030 |
hops | +0.030 |
around | +0.030 |
falls | +0.029 |
Neg logit contributions | |
Baby | -0.044 |
onda | -0.041 |
homeless | -0.039 |
dessert | -0.038 |
essa | -0.038 |
Token | o |
Token ID | 78 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
bs | +0.366 |
counts | +0.344 |
gers | +0.344 |
falls | +0.343 |
hops | +0.341 |
Neg logit contributions | |
Baby | -0.418 |
onda | -0.364 |
dessert | -0.350 |
essa | -0.345 |
homeless | -0.342 |
Token | , |
Token ID | 11 |
Feature activation | +7.20e-01 |
Pos logit contributions | |
Baby | +0.040 |
onda | +0.036 |
dessert | +0.036 |
homeless | +0.036 |
essa | +0.035 |
Neg logit contributions | |
bs | -0.013 |
gers | -0.012 |
around | -0.012 |
falls | -0.012 |
counts | -0.012 |
Token | you |
Token ID | 345 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
bs | +0.032 |
around | +0.031 |
falls | +0.031 |
likes | +0.031 |
hops | +0.031 |
Neg logit contributions | |
Baby | -0.043 |
onda | -0.040 |
homeless | -0.037 |
essa | -0.036 |
dessert | -0.036 |
Token | are |
Token ID | 389 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
Baby | +0.048 |
onda | +0.043 |
homeless | +0.042 |
dessert | +0.042 |
essa | +0.042 |
Neg logit contributions | |
bs | -0.022 |
counts | -0.021 |
falls | -0.020 |
likes | -0.020 |
hits | -0.020 |
Token | wonderful |
Token ID | 7932 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
Baby | +0.054 |
onda | +0.049 |
obby | +0.048 |
urg | +0.047 |
essa | +0.047 |
Neg logit contributions | |
around | -0.025 |
low | -0.022 |
counts | -0.021 |
likes | -0.020 |
middle | -0.020 |
Token | animals |
Token ID | 4695 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
bs | +0.039 |
falls | +0.036 |
tail | +0.034 |
hits | +0.034 |
chair | +0.033 |
Neg logit contributions | |
Baby | -0.033 |
homeless | -0.030 |
onda | -0.029 |
dessert | -0.029 |
rice | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
bs | +0.098 |
hits | +0.093 |
falls | +0.092 |
gers | +0.091 |
chair | +0.087 |
Neg logit contributions | |
Baby | -0.177 |
dessert | -0.163 |
homeless | -0.163 |
rice | -0.160 |
onda | -0.159 |
Token | like |
Token ID | 588 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
hits | +0.039 |
falls | +0.039 |
bs | +0.038 |
opl | +0.037 |
spins | +0.036 |
Neg logit contributions | |
Baby | -0.058 |
homeless | -0.054 |
dessert | -0.051 |
rice | -0.050 |
onda | -0.048 |
Token | lions |
Token ID | 31420 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
bs | +0.107 |
counts | +0.103 |
gers | +0.102 |
around | +0.099 |
hops | +0.099 |
Neg logit contributions | |
Baby | -0.199 |
onda | -0.179 |
dessert | -0.172 |
homeless | -0.172 |
essa | -0.169 |
Token | , |
Token ID | 11 |
Feature activation | +9.87e-01 |
Pos logit contributions | |
bs | +0.106 |
gers | +0.092 |
hits | +0.089 |
falls | +0.088 |
sideways | +0.086 |
Neg logit contributions | |
Baby | -0.220 |
dessert | -0.202 |
onda | -0.198 |
uf | -0.197 |
homeless | -0.197 |
Token | ze |
Token ID | 41271 |
Feature activation | +7.29e+00 |
Pos logit contributions | |
bs | +0.047 |
gers | +0.040 |
falls | +0.040 |
chair | +0.039 |
hits | +0.039 |
Neg logit contributions | |
Baby | -0.057 |
homeless | -0.051 |
rice | -0.049 |
dessert | -0.049 |
onda | -0.048 |
Token | br |
Token ID | 1671 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
gers | +0.217 |
middle | +0.214 |
sideways | +0.213 |
low | +0.208 |
hits | +0.203 |
Neg logit contributions | |
Baby | -0.498 |
uf | -0.460 |
onda | -0.443 |
dessert | -0.437 |
rice | -0.428 |
Token | as |
Token ID | 292 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
falls | +0.167 |
counts | +0.166 |
gers | +0.166 |
bs | +0.166 |
chair | +0.165 |
Neg logit contributions | |
Baby | -0.206 |
onda | -0.185 |
urg | -0.173 |
essa | -0.172 |
uf | -0.169 |
Token | , |
Token ID | 11 |
Feature activation | +9.90e-01 |
Pos logit contributions | |
bs | +0.068 |
gers | +0.061 |
counts | +0.061 |
falls | +0.060 |
hits | +0.059 |
Neg logit contributions | |
Baby | -0.212 |
onda | -0.193 |
homeless | -0.190 |
dessert | -0.189 |
essa | -0.185 |
Token | and |
Token ID | 290 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
bs | +0.049 |
chair | +0.043 |
gers | +0.043 |
chairs | +0.042 |
falls | +0.040 |
Neg logit contributions | |
Baby | -0.052 |
rice | -0.047 |
homeless | -0.046 |
dessert | -0.045 |
onda | -0.044 |
Token | gir |
Token ID | 37370 |
Feature activation | +6.28e+00 |
Pos logit contributions | |
bs | +0.017 |
gers | +0.015 |
falls | +0.015 |
ines | +0.015 |
chair | +0.015 |
Neg logit contributions | |
Baby | -0.022 |
homeless | -0.020 |
dessert | -0.020 |
rice | -0.020 |
onda | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +9.74e-01 |
Pos logit contributions | |
bs | +0.164 |
gers | +0.158 |
hits | +0.144 |
falls | +0.143 |
hops | +0.140 |
Neg logit contributions | |
Baby | -0.349 |
dessert | -0.317 |
onda | -0.311 |
homeless | -0.311 |
uf | -0.311 |
Token | gir |
Token ID | 37370 |
Feature activation | +6.25e+00 |
Pos logit contributions | |
bs | +0.046 |
gers | +0.041 |
chair | +0.040 |
chairs | +0.039 |
tail | +0.039 |
Neg logit contributions | |
Baby | -0.054 |
homeless | -0.049 |
rice | -0.048 |
dessert | -0.048 |
wasting | -0.046 |
Token | aff |
Token ID | 2001 |
Feature activation | -9.26e-01 |
Pos logit contributions | |
low | +0.439 |
middle | +0.436 |
gers | +0.434 |
bs | +0.434 |
roads | +0.431 |
Neg logit contributions | |
Baby | -0.184 |
uf | -0.157 |
onda | -0.144 |
urg | -0.143 |
essa | -0.137 |
Token | es |
Token ID | 274 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
Baby | +0.063 |
onda | +0.059 |
essa | +0.055 |
uf | +0.055 |
dessert | +0.054 |
Neg logit contributions | |
gers | -0.032 |
bs | -0.031 |
hits | -0.031 |
low | -0.030 |
falls | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
bs | +0.046 |
gers | +0.041 |
hits | +0.041 |
falls | +0.041 |
chair | +0.039 |
Neg logit contributions | |
Baby | -0.086 |
onda | -0.077 |
homeless | -0.077 |
dessert | -0.076 |
rice | -0.075 |
Token | ze |
Token ID | 41271 |
Feature activation | +7.26e+00 |
Pos logit contributions | |
bs | +0.017 |
gers | +0.016 |
ines | +0.016 |
falls | +0.015 |
around | +0.015 |
Neg logit contributions | |
Baby | -0.021 |
dessert | -0.019 |
rice | -0.019 |
homeless | -0.019 |
onda | -0.018 |
Token | br |
Token ID | 1671 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
middle | +0.240 |
gers | +0.237 |
sideways | +0.232 |
low | +0.230 |
hits | +0.225 |
Neg logit contributions | |
Baby | -0.462 |
uf | -0.433 |
onda | -0.423 |
dessert | -0.415 |
essa | -0.400 |
Token | as |
Token ID | 292 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
falls | +0.178 |
chair | +0.178 |
hits | +0.175 |
counts | +0.175 |
gers | +0.175 |
Neg logit contributions | |
Baby | -0.195 |
onda | -0.177 |
urg | -0.167 |
essa | -0.165 |
edd | -0.163 |
Token | . |
Token ID | 13 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
bs | +0.075 |
counts | +0.069 |
gers | +0.069 |
falls | +0.067 |
hits | +0.066 |
Neg logit contributions | |
Baby | -0.206 |
onda | -0.188 |
homeless | -0.183 |
dessert | -0.183 |
essa | -0.179 |
Token | They |
Token ID | 1119 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
hops | +0.023 |
bs | +0.023 |
gers | +0.022 |
tail | +0.022 |
lers | +0.021 |
Neg logit contributions | |
Baby | -0.025 |
Baby | -0.021 |
Son | -0.021 |
dessert | -0.020 |
homeless | -0.019 |
Token | took |
Token ID | 1718 |
Feature activation | -9.86e-01 |
Pos logit contributions | |
Baby | +0.023 |
onda | +0.020 |
homeless | +0.020 |
Sonny | +0.020 |
Son | +0.020 |
Neg logit contributions | |
gers | -0.014 |
bs | -0.014 |
chair | -0.014 |
hits | -0.013 |
falls | -0.013 |
Token | that |
Token ID | 326 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
bs | +0.080 |
hops | +0.075 |
counts | +0.073 |
gers | +0.072 |
rounds | +0.071 |
Neg logit contributions | |
Baby | -0.091 |
homeless | -0.088 |
dessert | -0.087 |
rice | -0.084 |
onda | -0.083 |
Token | says |
Token ID | 1139 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Baby | +0.070 |
rice | +0.067 |
homeless | +0.066 |
dessert | +0.066 |
onda | +0.066 |
Neg logit contributions | |
bs | -0.040 |
tail | -0.038 |
gers | -0.037 |
until | -0.036 |
low | -0.036 |
Token | " |
Token ID | 366 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
bs | +0.078 |
gers | +0.073 |
counts | +0.073 |
hits | +0.071 |
hops | +0.071 |
Neg logit contributions | |
Baby | -0.130 |
onda | -0.118 |
homeless | -0.116 |
dessert | -0.115 |
essa | -0.112 |
Token | Feed |
Token ID | 18332 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
Baby | +0.016 |
onda | +0.015 |
dessert | +0.015 |
Son | +0.014 |
homeless | +0.014 |
Neg logit contributions | |
bs | -0.015 |
gers | -0.014 |
low | -0.014 |
counts | -0.014 |
hits | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +9.37e-01 |
Pos logit contributions | |
Baby | +0.152 |
onda | +0.147 |
homeless | +0.139 |
dessert | +0.138 |
pudding | +0.133 |
Neg logit contributions | |
around | -0.111 |
gers | -0.107 |
bs | -0.105 |
trees | -0.100 |
likes | -0.099 |
Token | ze |
Token ID | 41271 |
Feature activation | +7.24e+00 |
Pos logit contributions | |
bs | +0.039 |
gers | +0.038 |
counts | +0.037 |
hits | +0.037 |
falls | +0.037 |
Neg logit contributions | |
Baby | -0.060 |
onda | -0.053 |
dessert | -0.051 |
homeless | -0.051 |
essa | -0.050 |
Token | br |
Token ID | 1671 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
bs | +0.224 |
counts | +0.218 |
gers | +0.214 |
hits | +0.211 |
low | +0.211 |
Neg logit contributions | |
Baby | -0.523 |
onda | -0.473 |
dessert | -0.462 |
homeless | -0.462 |
essa | -0.454 |
Token | as |
Token ID | 292 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
counts | +0.161 |
bs | +0.160 |
gers | +0.160 |
falls | +0.160 |
hits | +0.158 |
Neg logit contributions | |
Baby | -0.205 |
onda | -0.183 |
essa | -0.172 |
dessert | -0.169 |
Son | -0.169 |
Token | ". |
Token ID | 1911 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
bs | +0.113 |
around | +0.108 |
gers | +0.106 |
falls | +0.105 |
counts | +0.105 |
Neg logit contributions | |
Baby | -0.157 |
onda | -0.141 |
homeless | -0.135 |
dessert | -0.135 |
essa | -0.131 |
Token | He |
Token ID | 679 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
Baby | +0.171 |
onda | +0.152 |
homeless | +0.151 |
dessert | +0.150 |
Son | +0.148 |
Neg logit contributions | |
bs | -0.076 |
counts | -0.071 |
gers | -0.069 |
hits | -0.068 |
falls | -0.068 |
Token | has |
Token ID | 468 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
bs | +0.008 |
gers | +0.007 |
chair | +0.007 |
tail | +0.007 |
low | +0.007 |
Neg logit contributions | |
Baby | -0.017 |
onda | -0.015 |
Son | -0.015 |
Sonny | -0.015 |
uf | -0.015 |
Token | That |
Token ID | 1320 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
Baby | +0.017 |
onda | +0.015 |
homeless | +0.015 |
dessert | +0.015 |
essa | +0.014 |
Neg logit contributions | |
bs | -0.009 |
counts | -0.008 |
gers | -0.008 |
falls | -0.008 |
hits | -0.008 |
Token | is |
Token ID | 318 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
Baby | +0.232 |
onda | +0.213 |
homeless | +0.208 |
dessert | +0.202 |
rice | +0.200 |
Neg logit contributions | |
bs | -0.145 |
gers | -0.129 |
low | -0.127 |
ways | -0.126 |
hits | -0.125 |
Token | only |
Token ID | 691 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
bs | +0.012 |
counts | +0.011 |
gers | +0.011 |
falls | +0.011 |
around | +0.011 |
Neg logit contributions | |
Baby | -0.014 |
onda | -0.013 |
homeless | -0.012 |
dessert | -0.012 |
essa | -0.012 |
Token | for |
Token ID | 329 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
bs | +0.026 |
counts | +0.025 |
around | +0.025 |
likes | +0.024 |
gers | +0.024 |
Neg logit contributions | |
Baby | -0.050 |
onda | -0.045 |
homeless | -0.043 |
dessert | -0.043 |
essa | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
bs | +0.073 |
counts | +0.070 |
hops | +0.070 |
around | +0.070 |
falls | +0.069 |
Neg logit contributions | |
Baby | -0.087 |
onda | -0.078 |
homeless | -0.074 |
essa | -0.072 |
dessert | -0.072 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.24e+00 |
Pos logit contributions | |
bs | +0.020 |
counts | +0.019 |
gers | +0.019 |
falls | +0.018 |
hits | +0.018 |
Neg logit contributions | |
Baby | -0.030 |
onda | -0.026 |
homeless | -0.026 |
dessert | -0.026 |
essa | -0.025 |
Token | est |
Token ID | 395 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
bs | +0.256 |
falls | +0.227 |
gers | +0.227 |
low | +0.227 |
counts | +0.220 |
Neg logit contributions | |
Baby | -0.514 |
onda | -0.466 |
dessert | -0.456 |
homeless | -0.443 |
essa | -0.442 |
Token | of |
Token ID | 286 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
bs | +0.109 |
gers | +0.100 |
counts | +0.099 |
falls | +0.098 |
low | +0.098 |
Neg logit contributions | |
Baby | -0.191 |
onda | -0.171 |
dessert | -0.171 |
homeless | -0.168 |
essa | -0.163 |
Token | creatures |
Token ID | 8109 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
Baby | +0.089 |
onda | +0.080 |
homeless | +0.077 |
dessert | +0.075 |
essa | +0.075 |
Neg logit contributions | |
bs | -0.063 |
counts | -0.060 |
gers | -0.060 |
around | -0.060 |
hops | -0.060 |
Token | ." |
Token ID | 526 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
bs | +0.029 |
gers | +0.028 |
counts | +0.028 |
hits | +0.027 |
falls | +0.027 |
Neg logit contributions | |
Baby | -0.043 |
onda | -0.038 |
dessert | -0.037 |
homeless | -0.037 |
essa | -0.036 |
Token | Fl |
Token ID | 1610 |
Feature activation | -9.49e-01 |
Pos logit contributions | |
Baby | +0.100 |
onda | +0.086 |
homeless | +0.084 |
dessert | +0.084 |
Sonny | +0.083 |
Neg logit contributions | |
bs | -0.060 |
gers | -0.058 |
counts | -0.058 |
hits | -0.056 |
falls | -0.056 |
Token | known |
Token ID | 1900 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
bs | +0.014 |
gers | +0.014 |
counts | +0.013 |
chair | +0.013 |
falls | +0.013 |
Neg logit contributions | |
Baby | -0.015 |
homeless | -0.013 |
dessert | -0.013 |
onda | -0.012 |
ful | -0.012 |
Token | as |
Token ID | 355 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
bs | +0.067 |
counts | +0.066 |
gers | +0.066 |
falls | +0.064 |
hits | +0.064 |
Neg logit contributions | |
Baby | -0.115 |
dessert | -0.103 |
homeless | -0.103 |
onda | -0.099 |
Son | -0.098 |
Token | the |
Token ID | 262 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
Baby | +0.039 |
onda | +0.035 |
homeless | +0.034 |
dessert | +0.034 |
essa | +0.033 |
Neg logit contributions | |
bs | -0.024 |
gers | -0.023 |
counts | -0.023 |
falls | -0.022 |
hits | -0.022 |
Token | fastest |
Token ID | 14162 |
Feature activation | +4.82e+00 |
Pos logit contributions | |
bs | +0.018 |
gers | +0.016 |
counts | +0.016 |
falls | +0.016 |
chair | +0.016 |
Neg logit contributions | |
Baby | -0.026 |
onda | -0.024 |
homeless | -0.023 |
dessert | -0.023 |
Son | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
bs | +0.229 |
gers | +0.210 |
counts | +0.204 |
falls | +0.204 |
low | +0.203 |
Neg logit contributions | |
Baby | -0.258 |
dessert | -0.245 |
onda | -0.239 |
homeless | -0.238 |
rice | -0.235 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.20e+00 |
Pos logit contributions | |
Baby | +0.026 |
homeless | +0.026 |
dessert | +0.024 |
quitting | +0.024 |
wasting | +0.023 |
Neg logit contributions | |
bs | -0.022 |
gers | -0.021 |
ines | -0.021 |
chair | -0.019 |
tail | -0.019 |
Token | est |
Token ID | 395 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
bs | +0.256 |
gers | +0.232 |
low | +0.231 |
falls | +0.231 |
spins | +0.226 |
Neg logit contributions | |
Baby | -0.501 |
onda | -0.464 |
dessert | -0.444 |
essa | -0.438 |
homeless | -0.436 |
Token | knight |
Token ID | 22062 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
bs | +0.157 |
gers | +0.145 |
falls | +0.144 |
sideways | +0.142 |
counts | +0.142 |
Neg logit contributions | |
Baby | -0.141 |
dessert | -0.135 |
onda | -0.132 |
homeless | -0.129 |
rice | -0.126 |
Token | in |
Token ID | 287 |
Feature activation | -9.61e-01 |
Pos logit contributions | |
bs | +0.104 |
hits | +0.099 |
low | +0.099 |
falls | +0.097 |
gers | +0.097 |
Neg logit contributions | |
Baby | -0.120 |
onda | -0.114 |
homeless | -0.112 |
ful | -0.108 |
dessert | -0.107 |
Token | all |
Token ID | 477 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Baby | +0.055 |
dessert | +0.049 |
homeless | +0.048 |
onda | +0.048 |
Son | +0.046 |
Neg logit contributions | |
bs | -0.046 |
gers | -0.045 |
counts | -0.043 |
falls | -0.042 |
chair | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
Baby | +0.075 |
dessert | +0.069 |
onda | +0.068 |
ful | +0.066 |
homeless | +0.064 |
Neg logit contributions | |
bs | -0.078 |
gers | -0.073 |
tail | -0.072 |
chairs | -0.067 |
belts | -0.067 |
Token | monkey |
Token ID | 21657 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
bs | +0.018 |
falls | +0.016 |
gers | +0.016 |
chair | +0.015 |
ways | +0.015 |
Neg logit contributions | |
Baby | -0.022 |
homeless | -0.020 |
dessert | -0.020 |
onda | -0.019 |
uf | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
gers | +0.183 |
bs | +0.174 |
low | +0.163 |
sideways | +0.163 |
hits | +0.159 |
Neg logit contributions | |
Baby | -0.206 |
onda | -0.187 |
homeless | -0.181 |
illa | -0.174 |
ful | -0.172 |
Token | known |
Token ID | 1900 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
bs | +0.011 |
gers | +0.011 |
falls | +0.010 |
counts | +0.010 |
chair | +0.010 |
Neg logit contributions | |
Baby | -0.011 |
homeless | -0.010 |
dessert | -0.010 |
onda | -0.010 |
Son | -0.009 |
Token | as |
Token ID | 355 |
Feature activation | -7.18e-01 |
Pos logit contributions | |
bs | +0.063 |
counts | +0.062 |
gers | +0.060 |
falls | +0.060 |
hits | +0.060 |
Neg logit contributions | |
Baby | -0.112 |
homeless | -0.100 |
dessert | -0.099 |
onda | -0.098 |
Son | -0.096 |
Token | the |
Token ID | 262 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
Baby | +0.047 |
onda | +0.042 |
homeless | +0.041 |
dessert | +0.041 |
essa | +0.040 |
Neg logit contributions | |
bs | -0.029 |
counts | -0.027 |
gers | -0.027 |
falls | -0.026 |
hits | -0.026 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.20e+00 |
Pos logit contributions | |
bs | +0.017 |
falls | +0.015 |
chair | +0.015 |
gers | +0.015 |
counts | +0.015 |
Neg logit contributions | |
Baby | -0.026 |
homeless | -0.023 |
onda | -0.023 |
dessert | -0.022 |
Son | -0.022 |
Token | est |
Token ID | 395 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
bs | +0.270 |
falls | +0.253 |
low | +0.253 |
spins | +0.242 |
chair | +0.241 |
Neg logit contributions | |
Baby | -0.486 |
onda | -0.436 |
uf | -0.425 |
dessert | -0.419 |
essa | -0.418 |
Token | and |
Token ID | 290 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
bs | +0.138 |
falls | +0.131 |
low | +0.125 |
counts | +0.123 |
hits | +0.122 |
Neg logit contributions | |
Baby | -0.160 |
homeless | -0.148 |
dessert | -0.145 |
onda | -0.144 |
rice | -0.136 |
Token | most |
Token ID | 749 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
Baby | +0.035 |
homeless | +0.034 |
dessert | +0.032 |
rice | +0.031 |
quitting | +0.030 |
Neg logit contributions | |
bs | -0.029 |
ines | -0.027 |
gers | -0.026 |
falls | -0.026 |
chair | -0.026 |
Token | impressive |
Token ID | 8036 |
Feature activation | +9.95e-01 |
Pos logit contributions | |
bs | +0.185 |
falls | +0.182 |
hits | +0.178 |
counts | +0.176 |
tail | +0.173 |
Neg logit contributions | |
Baby | -0.221 |
homeless | -0.201 |
onda | -0.191 |
dessert | -0.189 |
rice | -0.181 |
Token | animal |
Token ID | 5044 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
bs | +0.049 |
falls | +0.044 |
low | +0.042 |
counts | +0.042 |
chair | +0.041 |
Neg logit contributions | |
Baby | -0.055 |
dessert | -0.052 |
homeless | -0.051 |
onda | -0.050 |
rice | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
Baby | +0.079 |
Son | +0.073 |
homeless | +0.072 |
dessert | +0.072 |
Sonny | +0.071 |
Neg logit contributions | |
bs | -0.038 |
chairs | -0.036 |
ines | -0.035 |
chair | -0.035 |
counts | -0.033 |
Token | he |
Token ID | 339 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
ines | +0.008 |
bs | +0.008 |
counts | +0.008 |
likes | +0.008 |
gers | +0.007 |
Neg logit contributions | |
Baby | -0.014 |
homeless | -0.013 |
dessert | -0.013 |
Son | -0.012 |
rice | -0.012 |
Token | put |
Token ID | 1234 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
bs | +0.057 |
counts | +0.054 |
gers | +0.054 |
falls | +0.053 |
hits | +0.052 |
Neg logit contributions | |
Baby | -0.107 |
onda | -0.097 |
homeless | -0.094 |
dessert | -0.093 |
essa | -0.091 |
Token | on |
Token ID | 319 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
Baby | +0.099 |
dessert | +0.091 |
rice | +0.088 |
Son | +0.087 |
homeless | +0.086 |
Neg logit contributions | |
bs | -0.067 |
counts | -0.067 |
chair | -0.065 |
spins | -0.063 |
falls | -0.063 |
Token | his |
Token ID | 465 |
Feature activation | -6.25e-02 |
Pos logit contributions | |
bs | +0.106 |
counts | +0.100 |
gers | +0.100 |
falls | +0.098 |
hits | +0.097 |
Neg logit contributions | |
Baby | -0.195 |
onda | -0.175 |
homeless | -0.171 |
dessert | -0.171 |
essa | -0.166 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.19e+00 |
Pos logit contributions | |
Baby | +0.004 |
dessert | +0.004 |
rice | +0.003 |
homeless | +0.003 |
Son | +0.003 |
Neg logit contributions | |
bs | -0.003 |
counts | -0.003 |
chairs | -0.003 |
ways | -0.003 |
chair | -0.003 |
Token | est |
Token ID | 395 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
bs | +0.294 |
falls | +0.255 |
chairs | +0.255 |
ways | +0.255 |
gers | +0.252 |
Neg logit contributions | |
Baby | -0.476 |
dessert | -0.433 |
onda | -0.426 |
essa | -0.421 |
uf | -0.415 |
Token | clothes |
Token ID | 8242 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
bs | +0.115 |
counts | +0.113 |
falls | +0.109 |
hits | +0.107 |
chairs | +0.107 |
Neg logit contributions | |
Baby | -0.173 |
dessert | -0.161 |
rice | -0.156 |
homeless | -0.153 |
onda | -0.151 |
Token | and |
Token ID | 290 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
bs | +0.029 |
counts | +0.028 |
gers | +0.028 |
hops | +0.027 |
hits | +0.027 |
Neg logit contributions | |
Baby | -0.066 |
homeless | -0.060 |
onda | -0.060 |
dessert | -0.059 |
essa | -0.058 |
Token | started |
Token ID | 2067 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Baby | +0.026 |
homeless | +0.023 |
dessert | +0.023 |
rice | +0.023 |
Son | +0.022 |
Neg logit contributions | |
bs | -0.016 |
gers | -0.016 |
low | -0.015 |
hits | -0.015 |
ines | -0.015 |
Token | his |
Token ID | 465 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
Baby | +0.077 |
onda | +0.070 |
essa | +0.065 |
homeless | +0.064 |
dessert | +0.064 |
Neg logit contributions | |
bs | -0.048 |
around | -0.047 |
hops | -0.046 |
hits | -0.046 |
counts | -0.046 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Baby | +0.021 |
dessert | +0.018 |
homeless | +0.018 |
Baby | +0.017 |
onda | +0.017 |
Neg logit contributions | |
bs | -0.021 |
gers | -0.021 |
ars | -0.020 |
falls | -0.020 |
ines | -0.020 |
Token | were |
Token ID | 547 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
Baby | +0.008 |
onda | +0.007 |
homeless | +0.007 |
Sonny | +0.007 |
Son | +0.007 |
Neg logit contributions | |
bs | -0.005 |
gers | -0.005 |
hits | -0.005 |
chair | -0.005 |
counts | -0.005 |
Token | known |
Token ID | 1900 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
Baby | +0.077 |
onda | +0.067 |
homeless | +0.066 |
dessert | +0.065 |
Son | +0.063 |
Neg logit contributions | |
bs | -0.059 |
gers | -0.056 |
counts | -0.055 |
falls | -0.055 |
hits | -0.055 |
Token | as |
Token ID | 355 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
bs | +0.061 |
counts | +0.058 |
gers | +0.058 |
hits | +0.057 |
falls | +0.057 |
Neg logit contributions | |
Baby | -0.118 |
onda | -0.105 |
homeless | -0.104 |
dessert | -0.104 |
essa | -0.101 |
Token | the |
Token ID | 262 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
Baby | +0.039 |
onda | +0.036 |
uf | +0.034 |
dessert | +0.033 |
essa | +0.033 |
Neg logit contributions | |
bs | -0.017 |
low | -0.016 |
around | -0.016 |
counts | -0.016 |
hops | -0.015 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.19e+00 |
Pos logit contributions | |
gers | +0.014 |
low | +0.014 |
counts | +0.013 |
hits | +0.013 |
likes | +0.013 |
Neg logit contributions | |
Baby | -0.028 |
onda | -0.025 |
essa | -0.024 |
urg | -0.024 |
uf | -0.024 |
Token | est |
Token ID | 395 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
bs | +0.257 |
low | +0.250 |
falls | +0.245 |
counts | +0.242 |
hops | +0.240 |
Neg logit contributions | |
Baby | -0.494 |
onda | -0.447 |
dessert | -0.434 |
homeless | -0.429 |
essa | -0.426 |
Token | animals |
Token ID | 4695 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
bs | +0.125 |
counts | +0.119 |
gers | +0.118 |
falls | +0.117 |
hits | +0.116 |
Neg logit contributions | |
Baby | -0.172 |
onda | -0.153 |
homeless | -0.149 |
dessert | -0.149 |
essa | -0.143 |
Token | in |
Token ID | 287 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
bs | +0.095 |
gers | +0.093 |
counts | +0.092 |
hits | +0.092 |
falls | +0.091 |
Neg logit contributions | |
Baby | -0.130 |
onda | -0.117 |
homeless | -0.115 |
dessert | -0.113 |
rice | -0.109 |
Token | the |
Token ID | 262 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
Baby | +0.057 |
onda | +0.050 |
dessert | +0.049 |
homeless | +0.049 |
Son | +0.047 |
Neg logit contributions | |
bs | -0.038 |
gers | -0.037 |
counts | -0.036 |
falls | -0.036 |
hits | -0.035 |
Token | park |
Token ID | 3952 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
bs | +0.017 |
counts | +0.015 |
gers | +0.015 |
falls | +0.015 |
hops | +0.015 |
Neg logit contributions | |
Baby | -0.023 |
onda | -0.021 |
homeless | -0.020 |
dessert | -0.020 |
Son | -0.019 |
Token | So |
Token ID | 2396 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
Baby | +0.039 |
onda | +0.035 |
homeless | +0.035 |
essa | +0.034 |
Son | +0.034 |
Neg logit contributions | |
bs | -0.019 |
gers | -0.019 |
counts | -0.018 |
hits | -0.018 |
ines | -0.018 |
Token | she |
Token ID | 673 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
Baby | +0.217 |
dessert | +0.190 |
Son | +0.188 |
Sonny | +0.186 |
onda | +0.185 |
Neg logit contributions | |
bs | -0.110 |
hits | -0.105 |
falls | -0.104 |
gers | -0.102 |
counts | -0.101 |
Token | put |
Token ID | 1234 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
bs | +0.068 |
gers | +0.068 |
low | +0.065 |
chairs | +0.064 |
hits | +0.064 |
Neg logit contributions | |
Baby | -0.117 |
onda | -0.106 |
homeless | -0.101 |
essa | -0.101 |
dessert | -0.101 |
Token | on |
Token ID | 319 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
Baby | +0.101 |
dessert | +0.092 |
rice | +0.089 |
homeless | +0.085 |
Baby | +0.085 |
Neg logit contributions | |
counts | -0.080 |
gers | -0.078 |
hops | -0.077 |
bs | -0.077 |
likes | -0.077 |
Token | her |
Token ID | 607 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
bs | +0.122 |
counts | +0.113 |
gers | +0.112 |
around | +0.112 |
low | +0.111 |
Neg logit contributions | |
Baby | -0.204 |
onda | -0.186 |
homeless | -0.178 |
dessert | -0.175 |
essa | -0.173 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.19e+00 |
Pos logit contributions | |
counts | +0.034 |
bs | +0.034 |
ways | +0.032 |
gers | +0.032 |
falls | +0.032 |
Neg logit contributions | |
Baby | -0.048 |
dessert | -0.046 |
rice | -0.044 |
homeless | -0.042 |
pudding | -0.041 |
Token | est |
Token ID | 395 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
bs | +0.306 |
ls | +0.284 |
gers | +0.279 |
ways | +0.273 |
falls | +0.271 |
Neg logit contributions | |
Baby | -0.458 |
dessert | -0.407 |
essa | -0.405 |
onda | -0.401 |
lette | -0.399 |
Token | face |
Token ID | 1986 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
bs | +0.114 |
counts | +0.111 |
gers | +0.108 |
falls | +0.106 |
ways | +0.105 |
Neg logit contributions | |
Baby | -0.167 |
dessert | -0.156 |
rice | -0.148 |
homeless | -0.145 |
onda | -0.143 |
Token | and |
Token ID | 290 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
Baby | +0.111 |
dessert | +0.100 |
ful | +0.098 |
homeless | +0.098 |
rice | +0.098 |
Neg logit contributions | |
gers | -0.061 |
bs | -0.059 |
counts | -0.059 |
hits | -0.058 |
hops | -0.057 |
Token | set |
Token ID | 900 |
Feature activation | +7.21e-01 |
Pos logit contributions | |
Baby | +0.025 |
dessert | +0.023 |
rice | +0.023 |
homeless | +0.022 |
Son | +0.022 |
Neg logit contributions | |
gers | -0.017 |
bs | -0.016 |
ines | -0.016 |
low | -0.016 |
sideways | -0.016 |
Token | off |
Token ID | 572 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
bs | +0.033 |
counts | +0.032 |
likes | +0.032 |
falls | +0.032 |
gers | +0.032 |
Neg logit contributions | |
Baby | -0.042 |
dessert | -0.036 |
homeless | -0.036 |
essa | -0.036 |
onda | -0.036 |
Token | he |
Token ID | 339 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Baby | +0.189 |
onda | +0.173 |
homeless | +0.169 |
dessert | +0.169 |
essa | +0.166 |
Neg logit contributions | |
bs | -0.053 |
counts | -0.048 |
gers | -0.048 |
falls | -0.046 |
hits | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
gers | +0.057 |
bs | +0.055 |
low | +0.053 |
chairs | +0.051 |
ways | +0.051 |
Neg logit contributions | |
Baby | -0.090 |
onda | -0.085 |
Son | -0.080 |
Sonny | -0.079 |
homeless | -0.079 |
Token | known |
Token ID | 1900 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
bs | +0.013 |
gers | +0.013 |
counts | +0.012 |
falls | +0.012 |
chair | +0.012 |
Neg logit contributions | |
Baby | -0.014 |
homeless | -0.012 |
dessert | -0.012 |
onda | -0.012 |
Son | -0.011 |
Token | as |
Token ID | 355 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
bs | +0.066 |
counts | +0.064 |
gers | +0.064 |
falls | +0.063 |
hits | +0.063 |
Neg logit contributions | |
Baby | -0.116 |
dessert | -0.103 |
homeless | -0.102 |
onda | -0.102 |
Son | -0.100 |
Token | the |
Token ID | 262 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
Baby | +0.055 |
onda | +0.050 |
homeless | +0.048 |
dessert | +0.048 |
essa | +0.047 |
Neg logit contributions | |
bs | -0.030 |
counts | -0.028 |
gers | -0.028 |
falls | -0.027 |
hits | -0.027 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.17e+00 |
Pos logit contributions | |
bs | +0.010 |
counts | +0.009 |
falls | +0.009 |
gers | +0.009 |
hits | +0.009 |
Neg logit contributions | |
Baby | -0.015 |
onda | -0.013 |
homeless | -0.013 |
dessert | -0.013 |
Son | -0.013 |
Token | est |
Token ID | 395 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
bs | +0.251 |
falls | +0.233 |
low | +0.232 |
counts | +0.222 |
gers | +0.221 |
Neg logit contributions | |
Baby | -0.507 |
onda | -0.458 |
dessert | -0.446 |
essa | -0.445 |
Son | -0.439 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
bs | +0.124 |
counts | +0.115 |
falls | +0.115 |
low | +0.114 |
hits | +0.113 |
Neg logit contributions | |
Baby | -0.165 |
onda | -0.148 |
homeless | -0.146 |
dessert | -0.145 |
essa | -0.142 |
Token | in |
Token ID | 287 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
Baby | +0.084 |
onda | +0.075 |
homeless | +0.074 |
dessert | +0.073 |
essa | +0.071 |
Neg logit contributions | |
bs | -0.051 |
hits | -0.048 |
gers | -0.048 |
low | -0.047 |
falls | -0.047 |
Token | all |
Token ID | 477 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
Baby | +0.073 |
dessert | +0.064 |
onda | +0.064 |
homeless | +0.064 |
Son | +0.062 |
Neg logit contributions | |
bs | -0.053 |
gers | -0.050 |
counts | -0.049 |
falls | -0.049 |
chair | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
Baby | +0.114 |
onda | +0.103 |
dessert | +0.102 |
homeless | +0.100 |
Son | +0.099 |
Neg logit contributions | |
bs | -0.078 |
gers | -0.072 |
tail | -0.069 |
falls | -0.069 |
hits | -0.068 |
Token | met |
Token ID | 1138 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
gers | +0.063 |
lights | +0.060 |
bs | +0.059 |
chair | +0.057 |
chairs | +0.057 |
Neg logit contributions | |
Baby | -0.095 |
onda | -0.091 |
homeless | -0.088 |
essa | -0.086 |
uf | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
bs | +0.005 |
hits | +0.005 |
falls | +0.005 |
counts | +0.005 |
gers | +0.005 |
Neg logit contributions | |
Baby | -0.011 |
dessert | -0.010 |
onda | -0.010 |
homeless | -0.010 |
essa | -0.009 |
Token | He |
Token ID | 679 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
Baby | +0.035 |
onda | +0.032 |
homeless | +0.031 |
dessert | +0.031 |
essa | +0.030 |
Neg logit contributions | |
bs | -0.013 |
counts | -0.012 |
gers | -0.012 |
falls | -0.012 |
hits | -0.012 |
Token | was |
Token ID | 373 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
Baby | +0.015 |
onda | +0.014 |
homeless | +0.013 |
dessert | +0.013 |
Son | +0.013 |
Neg logit contributions | |
bs | -0.010 |
gers | -0.010 |
low | -0.009 |
hits | -0.009 |
sideways | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
bs | +0.012 |
counts | +0.012 |
gers | +0.012 |
falls | +0.011 |
hits | +0.011 |
Neg logit contributions | |
Baby | -0.019 |
onda | -0.017 |
homeless | -0.017 |
dessert | -0.017 |
essa | -0.016 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.16e+00 |
Pos logit contributions | |
bs | +0.014 |
gers | +0.013 |
counts | +0.013 |
around | +0.013 |
low | +0.013 |
Neg logit contributions | |
Baby | -0.026 |
onda | -0.024 |
homeless | -0.023 |
dessert | -0.023 |
essa | -0.023 |
Token | est |
Token ID | 395 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
bs | +0.248 |
falls | +0.220 |
low | +0.218 |
gers | +0.214 |
hops | +0.213 |
Neg logit contributions | |
Baby | -0.525 |
onda | -0.480 |
dessert | -0.466 |
essa | -0.457 |
homeless | -0.455 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
bs | +0.124 |
gers | +0.120 |
counts | +0.118 |
around | +0.117 |
falls | +0.116 |
Neg logit contributions | |
Baby | -0.174 |
onda | -0.154 |
homeless | -0.149 |
dessert | -0.147 |
essa | -0.145 |
Token | around |
Token ID | 1088 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
Baby | +0.074 |
onda | +0.066 |
dessert | +0.065 |
homeless | +0.065 |
Son | +0.062 |
Neg logit contributions | |
bs | -0.047 |
hits | -0.044 |
gers | -0.044 |
counts | -0.043 |
falls | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
bs | +0.107 |
counts | +0.099 |
gers | +0.099 |
falls | +0.097 |
hits | +0.096 |
Neg logit contributions | |
Baby | -0.141 |
onda | -0.126 |
dessert | -0.125 |
homeless | -0.123 |
Son | -0.119 |
Token |
|
Token ID | 198 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Baby | +0.032 |
onda | +0.029 |
dessert | +0.028 |
homeless | +0.028 |
Son | +0.028 |
Neg logit contributions | |
bs | -0.015 |
gers | -0.014 |
counts | -0.014 |
falls | -0.014 |
hits | -0.014 |
Token | known |
Token ID | 1900 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
bs | +0.009 |
gers | +0.009 |
counts | +0.009 |
falls | +0.009 |
hits | +0.009 |
Neg logit contributions | |
Baby | -0.011 |
homeless | -0.009 |
onda | -0.009 |
dessert | -0.009 |
Son | -0.009 |
Token | as |
Token ID | 355 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
bs | +0.063 |
counts | +0.061 |
gers | +0.060 |
falls | +0.059 |
hits | +0.059 |
Neg logit contributions | |
Baby | -0.118 |
homeless | -0.104 |
dessert | -0.104 |
onda | -0.103 |
essa | -0.101 |
Token | the |
Token ID | 262 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
Baby | +0.037 |
onda | +0.033 |
homeless | +0.033 |
dessert | +0.032 |
essa | +0.032 |
Neg logit contributions | |
bs | -0.023 |
counts | -0.022 |
gers | -0.022 |
falls | -0.021 |
hits | -0.021 |
Token | biggest |
Token ID | 4094 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
bs | +0.020 |
falls | +0.019 |
gers | +0.018 |
chair | +0.018 |
counts | +0.018 |
Neg logit contributions | |
Baby | -0.027 |
homeless | -0.024 |
onda | -0.024 |
dessert | -0.023 |
Son | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
bs | +0.181 |
gers | +0.167 |
falls | +0.165 |
ways | +0.157 |
tail | +0.154 |
Neg logit contributions | |
Baby | -0.181 |
dessert | -0.180 |
homeless | -0.179 |
rice | -0.169 |
Son | -0.162 |
Token | brav |
Token ID | 49025 |
Feature activation | +7.15e+00 |
Pos logit contributions | |
Baby | +0.035 |
homeless | +0.032 |
dessert | +0.032 |
rice | +0.030 |
onda | +0.030 |
Neg logit contributions | |
bs | -0.026 |
ines | -0.025 |
falls | -0.024 |
gers | -0.024 |
hits | -0.023 |
Token | est |
Token ID | 395 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
bs | +0.274 |
falls | +0.258 |
hops | +0.252 |
gers | +0.248 |
low | +0.247 |
Neg logit contributions | |
Baby | -0.486 |
onda | -0.446 |
essa | -0.420 |
dessert | -0.419 |
homeless | -0.418 |
Token | helper |
Token ID | 31904 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
bs | +0.154 |
falls | +0.146 |
counts | +0.138 |
hops | +0.137 |
gers | +0.136 |
Neg logit contributions | |
Baby | -0.155 |
homeless | -0.144 |
dessert | -0.144 |
onda | -0.142 |
essa | -0.134 |
Token | in |
Token ID | 287 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Baby | +0.212 |
onda | +0.194 |
homeless | +0.187 |
dessert | +0.185 |
essa | +0.184 |
Neg logit contributions | |
bs | -0.140 |
hits | -0.132 |
falls | -0.132 |
counts | -0.131 |
low | -0.131 |
Token | the |
Token ID | 262 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
Baby | +0.061 |
dessert | +0.053 |
onda | +0.053 |
homeless | +0.052 |
Son | +0.051 |
Neg logit contributions | |
bs | -0.046 |
gers | -0.045 |
counts | -0.044 |
falls | -0.044 |
likes | -0.043 |
Token | jungle |
Token ID | 20712 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
bs | +0.017 |
falls | +0.015 |
ls | +0.015 |
gers | +0.014 |
hops | +0.014 |
Neg logit contributions | |
Baby | -0.019 |
homeless | -0.018 |
dessert | -0.018 |
onda | -0.017 |
Son | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
Baby | +0.026 |
dessert | +0.024 |
homeless | +0.024 |
onda | +0.023 |
essa | +0.023 |
Neg logit contributions | |
bs | -0.012 |
counts | -0.011 |
hits | -0.011 |
gers | -0.011 |
falls | -0.011 |
Token | She |
Token ID | 1375 |
Feature activation | +8.07e-02 |
Pos logit contributions | |
Baby | +0.062 |
onda | +0.056 |
homeless | +0.054 |
dessert | +0.054 |
Son | +0.054 |
Neg logit contributions | |
bs | -0.022 |
gers | -0.021 |
counts | -0.020 |
hits | -0.020 |
falls | -0.020 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
bs | +0.003 |
gers | +0.003 |
opl | +0.003 |
hits | +0.003 |
counts | +0.003 |
Neg logit contributions | |
Baby | -0.005 |
onda | -0.005 |
homeless | -0.004 |
Son | -0.004 |
Baby | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
Baby | +0.109 |
onda | +0.095 |
dessert | +0.095 |
homeless | +0.093 |
Son | +0.092 |
Neg logit contributions | |
bs | -0.073 |
gers | -0.071 |
counts | -0.070 |
falls | -0.068 |
hops | -0.068 |
Token | fire |
Token ID | 2046 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
bs | +0.011 |
counts | +0.011 |
gers | +0.011 |
around | +0.011 |
hits | +0.011 |
Neg logit contributions | |
Baby | -0.012 |
onda | -0.010 |
homeless | -0.010 |
essa | -0.010 |
dessert | -0.010 |
Token | extingu |
Token ID | 24995 |
Feature activation | -9.77e+00 |
Pos logit contributions | |
Baby | +0.159 |
onda | +0.143 |
homeless | +0.139 |
dessert | +0.139 |
essa | +0.136 |
Neg logit contributions | |
bs | -0.078 |
counts | -0.073 |
gers | -0.073 |
falls | -0.071 |
hits | -0.071 |
Token | isher |
Token ID | 4828 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
Baby | +0.404 |
onda | +0.344 |
dessert | +0.336 |
Son | +0.314 |
homeless | +0.312 |
Neg logit contributions | |
hops | -0.624 |
bs | -0.618 |
low | -0.613 |
counts | -0.610 |
falls | -0.601 |
Token | and |
Token ID | 290 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
Baby | +0.273 |
onda | +0.247 |
homeless | +0.243 |
dessert | +0.242 |
essa | +0.237 |
Neg logit contributions | |
bs | -0.109 |
counts | -0.102 |
gers | -0.102 |
hits | -0.098 |
hops | -0.098 |
Token | put |
Token ID | 1234 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
Baby | +0.050 |
onda | +0.043 |
homeless | +0.043 |
dessert | +0.043 |
Son | +0.043 |
Neg logit contributions | |
bs | -0.026 |
gers | -0.026 |
counts | -0.025 |
hops | -0.024 |
hits | -0.024 |
Token | out |
Token ID | 503 |
Feature activation | -7.30e-01 |
Pos logit contributions | |
Baby | +0.088 |
dessert | +0.076 |
Baby | +0.073 |
Son | +0.072 |
rice | +0.071 |
Neg logit contributions | |
counts | -0.084 |
hops | -0.084 |
gers | -0.083 |
likes | -0.083 |
bs | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
Baby | +0.037 |
dessert | +0.033 |
ful | +0.031 |
Baby | +0.031 |
homeless | +0.031 |
Neg logit contributions | |
gers | -0.037 |
counts | -0.037 |
likes | -0.036 |
hops | -0.036 |
bs | -0.036 |
Token | quickly |
Token ID | 2952 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
bs | +0.057 |
gers | +0.055 |
counts | +0.053 |
hits | +0.051 |
falls | +0.051 |
Neg logit contributions | |
Baby | -0.119 |
onda | -0.108 |
homeless | -0.105 |
dessert | -0.103 |
Son | -0.102 |
Token | checked |
Token ID | 10667 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
bs | +0.009 |
counts | +0.009 |
gers | +0.009 |
falls | +0.008 |
hits | +0.008 |
Neg logit contributions | |
Baby | -0.022 |
onda | -0.020 |
homeless | -0.019 |
dessert | -0.019 |
essa | -0.019 |
Token | for |
Token ID | 329 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Baby | +0.070 |
ful | +0.062 |
Son | +0.060 |
dessert | +0.060 |
Baby | +0.059 |
Neg logit contributions | |
gers | -0.078 |
hops | -0.075 |
tail | -0.075 |
bs | -0.073 |
spins | -0.071 |
Token | a |
Token ID | 257 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
bs | +0.061 |
counts | +0.058 |
low | +0.057 |
gers | +0.056 |
hits | +0.056 |
Neg logit contributions | |
Baby | -0.102 |
onda | -0.093 |
homeless | -0.088 |
essa | -0.087 |
dessert | -0.086 |
Token | fire |
Token ID | 2046 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
Baby | +0.042 |
onda | +0.039 |
essa | +0.038 |
urg | +0.037 |
uf | +0.036 |
Neg logit contributions | |
low | -0.027 |
bs | -0.027 |
hits | -0.024 |
falls | -0.024 |
around | -0.024 |
Token | extingu |
Token ID | 24995 |
Feature activation | -9.55e+00 |
Pos logit contributions | |
Baby | +0.124 |
onda | +0.112 |
homeless | +0.108 |
dessert | +0.108 |
essa | +0.106 |
Neg logit contributions | |
bs | -0.068 |
counts | -0.064 |
gers | -0.064 |
falls | -0.063 |
hits | -0.063 |
Token | isher |
Token ID | 4828 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
Baby | +0.395 |
onda | +0.344 |
dessert | +0.340 |
homeless | +0.317 |
Son | +0.310 |
Neg logit contributions | |
bs | -0.599 |
hops | -0.592 |
counts | -0.586 |
low | -0.578 |
falls | -0.574 |
Token | , |
Token ID | 11 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
Baby | +0.266 |
onda | +0.243 |
homeless | +0.237 |
dessert | +0.237 |
essa | +0.232 |
Neg logit contributions | |
bs | -0.104 |
counts | -0.096 |
gers | -0.096 |
falls | -0.093 |
hops | -0.093 |
Token | but |
Token ID | 475 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
bs | +0.007 |
gers | +0.007 |
counts | +0.007 |
hops | +0.006 |
falls | +0.006 |
Neg logit contributions | |
Baby | -0.017 |
homeless | -0.015 |
onda | -0.015 |
dessert | -0.015 |
Son | -0.015 |
Token | there |
Token ID | 612 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
bs | +0.085 |
gers | +0.081 |
counts | +0.078 |
hops | +0.078 |
hits | +0.076 |
Neg logit contributions | |
Baby | -0.140 |
homeless | -0.126 |
dessert | -0.126 |
Son | -0.122 |
onda | -0.119 |
Token | wasn |
Token ID | 2492 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
bs | +0.001 |
gers | +0.001 |
counts | +0.001 |
hops | +0.001 |
hits | +0.001 |
Neg logit contributions | |
Baby | -0.003 |
onda | -0.003 |
homeless | -0.002 |
dessert | -0.002 |
Son | -0.002 |
Token | two |
Token ID | 734 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
bs | +0.018 |
counts | +0.017 |
falls | +0.016 |
gers | +0.016 |
hops | +0.016 |
Neg logit contributions | |
Baby | -0.026 |
onda | -0.023 |
homeless | -0.023 |
dessert | -0.023 |
Son | -0.022 |
Token | squirrel |
Token ID | 33039 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
Baby | +0.171 |
onda | +0.146 |
homeless | +0.144 |
dessert | +0.143 |
rice | +0.137 |
Neg logit contributions | |
bs | -0.146 |
counts | -0.144 |
low | -0.141 |
falls | -0.140 |
sideways | -0.138 |
Token | s |
Token ID | 82 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
low | +0.019 |
middle | +0.018 |
counts | +0.017 |
tail | +0.017 |
sideways | +0.016 |
Neg logit contributions | |
Baby | -0.040 |
onda | -0.035 |
homeless | -0.034 |
Baby | -0.033 |
Sonny | -0.033 |
Token | kept |
Token ID | 4030 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
Baby | +0.101 |
onda | +0.090 |
homeless | +0.089 |
Son | +0.087 |
essa | +0.086 |
Neg logit contributions | |
bs | -0.059 |
counts | -0.058 |
falls | -0.056 |
hits | -0.056 |
gers | -0.056 |
Token | qu |
Token ID | 627 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
Baby | +0.029 |
onda | +0.026 |
homeless | +0.025 |
dessert | +0.025 |
essa | +0.024 |
Neg logit contributions | |
bs | -0.019 |
around | -0.019 |
hits | -0.018 |
gers | -0.018 |
low | -0.018 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.45e+00 |
Pos logit contributions | |
Baby | +0.130 |
essa | +0.120 |
dessert | +0.119 |
homeless | +0.119 |
onda | +0.117 |
Neg logit contributions | |
bs | -0.114 |
around | -0.114 |
gers | -0.111 |
ines | -0.108 |
hops | -0.107 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
Baby | +0.671 |
onda | +0.602 |
dessert | +0.590 |
homeless | +0.590 |
essa | +0.580 |
Neg logit contributions | |
bs | -0.331 |
around | -0.316 |
gers | -0.314 |
counts | -0.308 |
falls | -0.305 |
Token | until |
Token ID | 1566 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
Baby | +0.113 |
dessert | +0.100 |
rice | +0.098 |
homeless | +0.097 |
onda | +0.096 |
Neg logit contributions | |
bs | -0.064 |
counts | -0.061 |
hits | -0.059 |
falls | -0.058 |
chair | -0.057 |
Token | finally |
Token ID | 3443 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
bs | +0.083 |
gers | +0.078 |
around | +0.077 |
low | +0.077 |
falls | +0.077 |
Neg logit contributions | |
Baby | -0.131 |
onda | -0.123 |
homeless | -0.114 |
essa | -0.112 |
uf | -0.111 |
Token | Sammy |
Token ID | 44067 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
bs | +0.016 |
counts | +0.014 |
gers | +0.014 |
around | +0.014 |
falls | +0.014 |
Neg logit contributions | |
Baby | -0.040 |
onda | -0.037 |
homeless | -0.035 |
dessert | -0.035 |
pudding | -0.034 |
Token | said |
Token ID | 531 |
Feature activation | -7.21e-01 |
Pos logit contributions | |
Baby | +0.134 |
onda | +0.121 |
homeless | +0.118 |
dessert | +0.118 |
essa | +0.116 |
Neg logit contributions | |
bs | -0.059 |
counts | -0.055 |
gers | -0.055 |
falls | -0.054 |
hits | -0.053 |
Token | â |
Token ID | 22940 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
Baby | +0.042 |
onda | +0.039 |
rice | +0.039 |
uggets | +0.039 |
essa | +0.038 |
Neg logit contributions | |
around | -0.024 |
low | -0.023 |
falls | -0.022 |
chair | -0.022 |
hits | -0.021 |
Token | € |
Token ID | 26391 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
Baby | +0.173 |
onda | +0.154 |
dessert | +0.151 |
homeless | +0.150 |
rice | +0.145 |
Neg logit contributions | |
bs | -0.135 |
gers | -0.131 |
counts | -0.128 |
around | -0.126 |
falls | -0.126 |
Token | ™ |
Token ID | 8151 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
Baby | +0.070 |
onda | +0.062 |
homeless | +0.060 |
dessert | +0.059 |
essa | +0.059 |
Neg logit contributions | |
bs | -0.050 |
counts | -0.049 |
hits | -0.048 |
gers | -0.048 |
hops | -0.048 |
Token | t |
Token ID | 83 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
bs | +0.075 |
around | +0.067 |
gers | +0.067 |
falls | +0.066 |
counts | +0.066 |
Neg logit contributions | |
Baby | -0.156 |
onda | -0.143 |
dessert | -0.139 |
homeless | -0.138 |
essa | -0.136 |
Token | qu |
Token ID | 627 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
Baby | +0.016 |
onda | +0.014 |
uf | +0.014 |
rice | +0.014 |
pudding | +0.014 |
Neg logit contributions | |
around | -0.010 |
bs | -0.010 |
until | -0.009 |
low | -0.009 |
, | -0.008 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.37e+00 |
Pos logit contributions | |
Baby | +0.102 |
dessert | +0.086 |
essa | +0.085 |
homeless | +0.085 |
Son | +0.083 |
Neg logit contributions | |
bs | -0.152 |
around | -0.149 |
ines | -0.143 |
falls | -0.143 |
gers | -0.143 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
Baby | +0.718 |
uf | +0.651 |
dessert | +0.629 |
onda | +0.629 |
Son | +0.626 |
Neg logit contributions | |
around | -0.317 |
bs | -0.286 |
gers | -0.271 |
falls | -0.267 |
likes | -0.263 |
Token | , |
Token ID | 11 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
Baby | +0.131 |
onda | +0.118 |
homeless | +0.117 |
dessert | +0.116 |
essa | +0.113 |
Neg logit contributions | |
bs | -0.050 |
counts | -0.047 |
gers | -0.046 |
hits | -0.045 |
falls | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
Baby | +0.010 |
onda | +0.009 |
dessert | +0.008 |
homeless | +0.008 |
essa | +0.008 |
Neg logit contributions | |
bs | -0.005 |
gers | -0.005 |
counts | -0.005 |
falls | -0.005 |
around | -0.005 |
Token | furniture |
Token ID | 13091 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
bs | +0.003 |
sideways | +0.003 |
hops | +0.003 |
counts | +0.003 |
hits | +0.003 |
Neg logit contributions | |
Baby | -0.003 |
dessert | -0.002 |
homeless | -0.002 |
rice | -0.002 |
onda | -0.002 |
Token | stayed |
Token ID | 9658 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
Baby | +0.003 |
onda | +0.002 |
homeless | +0.002 |
dessert | +0.002 |
Son | +0.002 |
Neg logit contributions | |
bs | -0.001 |
counts | -0.001 |
gers | -0.001 |
hops | -0.001 |
hits | -0.001 |
Token | both |
Token ID | 1111 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
Baby | +0.022 |
onda | +0.020 |
homeless | +0.019 |
dessert | +0.019 |
Son | +0.019 |
Neg logit contributions | |
bs | -0.013 |
counts | -0.012 |
gers | -0.012 |
falls | -0.012 |
hits | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
Baby | +0.137 |
onda | +0.121 |
homeless | +0.120 |
dessert | +0.119 |
Son | +0.115 |
Neg logit contributions | |
bs | -0.085 |
gers | -0.085 |
counts | -0.084 |
falls | -0.080 |
hits | -0.080 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
bs | +0.001 |
counts | +0.001 |
gers | +0.001 |
falls | +0.001 |
hits | +0.001 |
Neg logit contributions | |
Baby | -0.002 |
onda | -0.002 |
homeless | -0.002 |
dessert | -0.002 |
essa | -0.001 |
Token | stopped |
Token ID | 5025 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
bs | +0.038 |
counts | +0.036 |
gers | +0.036 |
falls | +0.035 |
hits | +0.035 |
Neg logit contributions | |
Baby | -0.088 |
onda | -0.079 |
homeless | -0.077 |
dessert | -0.077 |
essa | -0.075 |
Token | qu |
Token ID | 627 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
Baby | +0.172 |
onda | +0.153 |
homeless | +0.148 |
dessert | +0.148 |
essa | +0.144 |
Neg logit contributions | |
bs | -0.120 |
counts | -0.113 |
around | -0.112 |
gers | -0.112 |
falls | -0.111 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.34e+00 |
Pos logit contributions | |
Baby | +0.133 |
dessert | +0.121 |
homeless | +0.120 |
essa | +0.120 |
Sonny | +0.120 |
Neg logit contributions | |
bs | -0.109 |
around | -0.108 |
c | -0.103 |
ines | -0.102 |
falls | -0.101 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
Baby | +0.725 |
uf | +0.669 |
onda | +0.643 |
essa | +0.642 |
Son | +0.641 |
Neg logit contributions | |
around | -0.296 |
bs | -0.261 |
gers | -0.248 |
falls | -0.247 |
low | -0.240 |
Token | . |
Token ID | 13 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
Baby | +0.126 |
onda | +0.113 |
homeless | +0.111 |
dessert | +0.111 |
essa | +0.108 |
Neg logit contributions | |
bs | -0.057 |
counts | -0.054 |
gers | -0.053 |
falls | -0.052 |
hits | -0.052 |
Token | They |
Token ID | 1119 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
Baby | +0.076 |
onda | +0.070 |
homeless | +0.068 |
dessert | +0.067 |
essa | +0.066 |
Neg logit contributions | |
bs | -0.030 |
counts | -0.028 |
around | -0.028 |
gers | -0.028 |
falls | -0.027 |
Token | decided |
Token ID | 3066 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
Baby | +0.099 |
onda | +0.089 |
homeless | +0.087 |
dessert | +0.085 |
Son | +0.084 |
Neg logit contributions | |
gers | -0.056 |
bs | -0.055 |
counts | -0.054 |
hits | -0.053 |
falls | -0.053 |
Token | to |
Token ID | 284 |
Feature activation | -9.99e-01 |
Pos logit contributions | |
Baby | +0.028 |
onda | +0.025 |
dessert | +0.024 |
uf | +0.023 |
homeless | +0.023 |
Neg logit contributions | |
bs | -0.016 |
around | -0.016 |
counts | -0.015 |
gers | -0.015 |
low | -0.015 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
Baby | +0.169 |
onda | +0.149 |
homeless | +0.146 |
dessert | +0.145 |
rice | +0.139 |
Neg logit contributions | |
bs | -0.120 |
counts | -0.118 |
gers | -0.117 |
falls | -0.115 |
hits | -0.115 |
Token | didn |
Token ID | 1422 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
bs | +0.048 |
gers | +0.048 |
counts | +0.047 |
hops | +0.046 |
hits | +0.046 |
Neg logit contributions | |
Baby | -0.091 |
onda | -0.082 |
homeless | -0.079 |
dessert | -0.078 |
Son | -0.077 |
Token | 't |
Token ID | 470 |
Feature activation | -8.03e-01 |
Pos logit contributions | |
Baby | +0.027 |
onda | +0.025 |
essa | +0.024 |
Son | +0.023 |
homeless | +0.023 |
Neg logit contributions | |
around | -0.015 |
falls | -0.015 |
hits | -0.014 |
low | -0.014 |
bs | -0.014 |
Token | stop |
Token ID | 2245 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
Baby | +0.062 |
onda | +0.058 |
Son | +0.055 |
urg | +0.055 |
rice | +0.055 |
Neg logit contributions | |
bs | -0.024 |
around | -0.024 |
low | -0.023 |
counts | -0.022 |
until | -0.021 |
Token | qu |
Token ID | 627 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
Baby | +0.087 |
onda | +0.081 |
homeless | +0.074 |
essa | +0.074 |
pudding | +0.074 |
Neg logit contributions | |
around | -0.056 |
until | -0.053 |
low | -0.052 |
bs | -0.052 |
hits | -0.049 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.29e+00 |
Pos logit contributions | |
Baby | +0.095 |
essa | +0.082 |
Son | +0.079 |
homeless | +0.079 |
dessert | +0.079 |
Neg logit contributions | |
bs | -0.137 |
around | -0.134 |
hits | -0.132 |
falls | -0.131 |
hops | -0.129 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
Baby | +0.605 |
uf | +0.528 |
onda | +0.522 |
Son | +0.517 |
essa | +0.515 |
Neg logit contributions | |
around | -0.414 |
bs | -0.392 |
falls | -0.377 |
gers | -0.375 |
counts | -0.370 |
Token | and |
Token ID | 290 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
Baby | +0.115 |
onda | +0.104 |
dessert | +0.103 |
homeless | +0.103 |
Son | +0.099 |
Neg logit contributions | |
bs | -0.050 |
counts | -0.047 |
gers | -0.046 |
hits | -0.045 |
falls | -0.045 |
Token | were |
Token ID | 547 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
Baby | +0.051 |
homeless | +0.045 |
onda | +0.045 |
dessert | +0.044 |
Son | +0.043 |
Neg logit contributions | |
bs | -0.031 |
gers | -0.030 |
counts | -0.030 |
hits | -0.028 |
falls | -0.028 |
Token | getting |
Token ID | 1972 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
Baby | +0.108 |
onda | +0.097 |
homeless | +0.094 |
dessert | +0.094 |
essa | +0.092 |
Neg logit contributions | |
bs | -0.057 |
counts | -0.053 |
gers | -0.053 |
around | -0.053 |
falls | -0.052 |
Token | louder |
Token ID | 27089 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
Baby | +0.226 |
onda | +0.209 |
essa | +0.201 |
uf | +0.197 |
Son | +0.195 |
Neg logit contributions | |
bs | -0.104 |
around | -0.101 |
low | -0.097 |
hits | -0.094 |
gers | -0.093 |
Token | the |
Token ID | 262 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
Baby | +0.134 |
onda | +0.120 |
homeless | +0.118 |
dessert | +0.118 |
essa | +0.114 |
Neg logit contributions | |
bs | -0.078 |
gers | -0.073 |
counts | -0.072 |
falls | -0.071 |
hits | -0.071 |
Token | birds |
Token ID | 10087 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
bs | +0.008 |
counts | +0.008 |
gers | +0.008 |
falls | +0.008 |
hits | +0.008 |
Neg logit contributions | |
Baby | -0.011 |
onda | -0.010 |
homeless | -0.010 |
dessert | -0.010 |
essa | -0.009 |
Token | were |
Token ID | 547 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
bs | +0.094 |
gers | +0.094 |
counts | +0.091 |
hits | +0.087 |
chair | +0.086 |
Neg logit contributions | |
Baby | -0.173 |
onda | -0.159 |
homeless | -0.156 |
dessert | -0.151 |
rice | -0.147 |
Token | only |
Token ID | 691 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
Baby | +0.105 |
onda | +0.092 |
uf | +0.089 |
Son | +0.089 |
essa | +0.088 |
Neg logit contributions | |
around | -0.067 |
bs | -0.063 |
low | -0.063 |
counts | -0.060 |
hops | -0.060 |
Token | qu |
Token ID | 627 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
bs | +0.024 |
around | +0.024 |
counts | +0.023 |
hops | +0.022 |
low | +0.022 |
Neg logit contributions | |
Baby | -0.036 |
onda | -0.032 |
homeless | -0.030 |
essa | -0.030 |
dessert | -0.030 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.22e+00 |
Pos logit contributions | |
Baby | +0.093 |
dessert | +0.082 |
essa | +0.080 |
homeless | +0.080 |
Son | +0.079 |
Neg logit contributions | |
bs | -0.143 |
around | -0.136 |
ines | -0.132 |
hops | -0.131 |
falls | -0.130 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
Baby | +0.711 |
onda | +0.635 |
uf | +0.631 |
dessert | +0.625 |
essa | +0.623 |
Neg logit contributions | |
around | -0.281 |
bs | -0.280 |
falls | -0.255 |
gers | -0.253 |
counts | -0.250 |
Token | like |
Token ID | 588 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
Baby | +0.102 |
homeless | +0.091 |
ful | +0.091 |
dessert | +0.091 |
onda | +0.090 |
Neg logit contributions | |
bs | -0.062 |
hits | -0.059 |
counts | -0.059 |
gers | -0.058 |
chair | -0.057 |
Token | little |
Token ID | 1310 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
bs | +0.078 |
counts | +0.072 |
around | +0.072 |
gers | +0.071 |
low | +0.071 |
Neg logit contributions | |
Baby | -0.141 |
onda | -0.127 |
homeless | -0.121 |
dessert | -0.121 |
Son | -0.119 |
Token | children |
Token ID | 1751 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Baby | +0.253 |
homeless | +0.228 |
rice | +0.227 |
dessert | +0.226 |
onda | +0.221 |
Neg logit contributions | |
bs | -0.214 |
chair | -0.202 |
falls | -0.201 |
ars | -0.199 |
hits | -0.198 |
Token | do |
Token ID | 466 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
Baby | +0.095 |
onda | +0.087 |
homeless | +0.085 |
dessert | +0.085 |
rice | +0.082 |
Neg logit contributions | |
bs | -0.040 |
gers | -0.040 |
counts | -0.038 |
hits | -0.038 |
falls | -0.037 |
Token | other |
Token ID | 584 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
Baby | +0.088 |
onda | +0.079 |
homeless | +0.077 |
dessert | +0.077 |
essa | +0.076 |
Neg logit contributions | |
bs | -0.041 |
counts | -0.038 |
gers | -0.038 |
falls | -0.037 |
around | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
bs | +0.039 |
around | +0.037 |
counts | +0.035 |
gers | +0.035 |
falls | +0.034 |
Neg logit contributions | |
Baby | -0.145 |
onda | -0.133 |
homeless | -0.129 |
dessert | -0.128 |
essa | -0.127 |
Token | They |
Token ID | 1119 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Baby | +0.044 |
onda | +0.041 |
homeless | +0.040 |
dessert | +0.040 |
essa | +0.039 |
Neg logit contributions | |
bs | -0.015 |
counts | -0.013 |
gers | -0.013 |
around | -0.013 |
falls | -0.013 |
Token | starting |
Token ID | 3599 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
Baby | +0.083 |
onda | +0.075 |
homeless | +0.073 |
dessert | +0.073 |
essa | +0.071 |
Neg logit contributions | |
bs | -0.041 |
gers | -0.039 |
counts | -0.039 |
falls | -0.038 |
hits | -0.038 |
Token | qu |
Token ID | 627 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
Baby | +0.172 |
onda | +0.154 |
essa | +0.146 |
homeless | +0.145 |
dessert | +0.144 |
Neg logit contributions | |
bs | -0.108 |
around | -0.106 |
hits | -0.104 |
counts | -0.103 |
falls | -0.102 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.22e+00 |
Pos logit contributions | |
Baby | +0.083 |
homeless | +0.071 |
dessert | +0.069 |
essa | +0.069 |
Son | +0.067 |
Neg logit contributions | |
bs | -0.147 |
around | -0.143 |
falls | -0.140 |
hits | -0.139 |
counts | -0.138 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
Baby | +0.672 |
onda | +0.597 |
homeless | +0.592 |
dessert | +0.590 |
uf | +0.584 |
Neg logit contributions | |
bs | -0.312 |
around | -0.305 |
counts | -0.291 |
gers | -0.290 |
falls | -0.289 |
Token | again |
Token ID | 757 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
Baby | +0.098 |
dessert | +0.090 |
homeless | +0.088 |
rice | +0.087 |
onda | +0.087 |
Neg logit contributions | |
bs | -0.055 |
gers | -0.051 |
counts | -0.051 |
hops | -0.049 |
chair | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
bs | +0.001 |
counts | +0.001 |
gers | +0.001 |
falls | +0.001 |
around | +0.001 |
Neg logit contributions | |
Baby | -0.002 |
onda | -0.002 |
homeless | -0.002 |
dessert | -0.002 |
essa | -0.002 |
Token | their |
Token ID | 511 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
Baby | +0.039 |
onda | +0.035 |
homeless | +0.034 |
dessert | +0.034 |
essa | +0.034 |
Neg logit contributions | |
bs | -0.024 |
counts | -0.023 |
around | -0.022 |
falls | -0.022 |
hits | -0.022 |
Token | voices |
Token ID | 10839 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
Baby | +0.132 |
onda | +0.117 |
homeless | +0.113 |
essa | +0.112 |
dessert | +0.111 |
Neg logit contributions | |
bs | -0.093 |
gers | -0.089 |
counts | -0.088 |
hits | -0.087 |
hops | -0.087 |
Token | but |
Token ID | 475 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
gers | +0.008 |
bs | +0.008 |
hits | +0.007 |
chair | +0.007 |
middle | +0.007 |
Neg logit contributions | |
Baby | -0.011 |
homeless | -0.010 |
dessert | -0.010 |
wasting | -0.010 |
rice | -0.010 |
Token | today |
Token ID | 1909 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
gers | +0.101 |
bs | +0.096 |
ways | +0.095 |
falls | +0.093 |
ars | +0.092 |
Neg logit contributions | |
Baby | -0.124 |
dessert | -0.108 |
homeless | -0.108 |
ful | -0.106 |
onda | -0.105 |
Token | they |
Token ID | 484 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
bs | +0.011 |
gers | +0.010 |
falls | +0.010 |
around | +0.010 |
chair | +0.010 |
Neg logit contributions | |
Baby | -0.032 |
onda | -0.029 |
essa | -0.028 |
homeless | -0.028 |
dessert | -0.028 |
Token | were |
Token ID | 547 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Baby | +0.003 |
onda | +0.003 |
dessert | +0.003 |
homeless | +0.003 |
essa | +0.002 |
Neg logit contributions | |
bs | -0.001 |
counts | -0.001 |
gers | -0.001 |
falls | -0.001 |
around | -0.001 |
Token | qu |
Token ID | 627 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
Baby | +0.087 |
onda | +0.079 |
essa | +0.076 |
Son | +0.075 |
uf | +0.074 |
Neg logit contributions | |
around | -0.047 |
low | -0.043 |
bs | -0.042 |
counts | -0.039 |
likes | -0.039 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.18e+00 |
Pos logit contributions | |
Baby | +0.094 |
Sonny | +0.086 |
dessert | +0.086 |
Son | +0.085 |
essa | +0.084 |
Neg logit contributions | |
bs | -0.120 |
around | -0.118 |
ines | -0.114 |
gers | -0.114 |
c | -0.113 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Baby | +0.758 |
uf | +0.691 |
Son | +0.680 |
onda | +0.677 |
essa | +0.675 |
Neg logit contributions | |
around | -0.243 |
bs | -0.220 |
gers | -0.204 |
likes | -0.194 |
falls | -0.193 |
Token | . |
Token ID | 13 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
Baby | +0.100 |
dessert | +0.089 |
onda | +0.089 |
homeless | +0.088 |
rice | +0.087 |
Neg logit contributions | |
bs | -0.057 |
counts | -0.054 |
hits | -0.053 |
hops | -0.052 |
falls | -0.052 |
Token | They |
Token ID | 1119 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Baby | +0.033 |
uf | +0.028 |
Baby | +0.028 |
Son | +0.028 |
Sonny | +0.028 |
Neg logit contributions | |
gers | -0.026 |
bs | -0.024 |
falls | -0.024 |
hits | -0.023 |
low | -0.023 |
Token | were |
Token ID | 547 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Baby | +0.076 |
onda | +0.069 |
dessert | +0.067 |
homeless | +0.067 |
essa | +0.066 |
Neg logit contributions | |
bs | -0.037 |
counts | -0.034 |
gers | -0.034 |
falls | -0.034 |
around | -0.034 |
Token | arguing |
Token ID | 11810 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
Baby | +0.092 |
onda | +0.086 |
essa | +0.083 |
Son | +0.080 |
dessert | +0.079 |
Neg logit contributions | |
around | -0.045 |
low | -0.040 |
bs | -0.039 |
hits | -0.037 |
likes | -0.037 |
Token | good |
Token ID | 922 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
Baby | +0.103 |
onda | +0.091 |
dessert | +0.089 |
homeless | +0.089 |
Son | +0.085 |
Neg logit contributions | |
bs | -0.068 |
gers | -0.065 |
counts | -0.064 |
falls | -0.063 |
hits | -0.062 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
Baby | +0.097 |
onda | +0.088 |
homeless | +0.085 |
dessert | +0.083 |
Son | +0.083 |
Neg logit contributions | |
bs | -0.047 |
gers | -0.043 |
around | -0.043 |
counts | -0.043 |
hits | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
bs | +0.069 |
hits | +0.068 |
hops | +0.065 |
spins | +0.064 |
falls | +0.064 |
Neg logit contributions | |
Baby | -0.105 |
onda | -0.095 |
dessert | -0.095 |
homeless | -0.093 |
ful | -0.092 |
Token | never |
Token ID | 1239 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
Baby | +0.044 |
dessert | +0.039 |
onda | +0.039 |
homeless | +0.039 |
rice | +0.037 |
Neg logit contributions | |
bs | -0.024 |
gers | -0.024 |
counts | -0.023 |
hits | -0.023 |
falls | -0.023 |
Token | qu |
Token ID | 627 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
bs | +0.094 |
counts | +0.089 |
hits | +0.087 |
around | +0.086 |
falls | +0.085 |
Neg logit contributions | |
Baby | -0.232 |
onda | -0.213 |
essa | -0.203 |
homeless | -0.201 |
dessert | -0.201 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.18e+00 |
Pos logit contributions | |
Baby | +0.091 |
essa | +0.080 |
Son | +0.077 |
homeless | +0.077 |
Sonny | +0.076 |
Neg logit contributions | |
bs | -0.145 |
around | -0.142 |
hits | -0.138 |
likes | -0.135 |
falls | -0.135 |
Token | led |
Token ID | 992 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
Baby | +0.833 |
onda | +0.754 |
uf | +0.752 |
Son | +0.750 |
homeless | +0.744 |
Neg logit contributions | |
bs | -0.150 |
around | -0.150 |
gers | -0.127 |
counts | -0.123 |
likes | -0.123 |
Token | again |
Token ID | 757 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Baby | +0.101 |
dessert | +0.092 |
homeless | +0.089 |
onda | +0.087 |
ful | +0.083 |
Neg logit contributions | |
bs | -0.096 |
gers | -0.092 |
counts | -0.092 |
chair | -0.091 |
falls | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
bs | +0.004 |
gers | +0.003 |
counts | +0.003 |
hits | +0.003 |
falls | +0.003 |
Neg logit contributions | |
Baby | -0.008 |
dessert | -0.007 |
onda | -0.007 |
homeless | -0.007 |
rice | -0.007 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
Baby | +0.049 |
onda | +0.044 |
dessert | +0.043 |
homeless | +0.043 |
Son | +0.042 |
Neg logit contributions | |
bs | -0.026 |
gers | -0.025 |
counts | -0.024 |
falls | -0.024 |
hops | -0.024 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
Baby | +0.289 |
onda | +0.264 |
homeless | +0.260 |
dessert | +0.256 |
essa | +0.254 |
Neg logit contributions | |
bs | -0.114 |
counts | -0.112 |
around | -0.112 |
gers | -0.111 |
falls | -0.107 |
Token | little |
Token ID | 1310 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
Baby | +0.148 |
onda | +0.131 |
dessert | +0.127 |
homeless | +0.125 |
essa | +0.124 |
Neg logit contributions | |
bs | -0.109 |
gers | -0.104 |
counts | -0.104 |
hops | -0.102 |
falls | -0.102 |
Token | birds |
Token ID | 10087 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
Baby | +0.252 |
onda | +0.221 |
homeless | +0.215 |
dessert | +0.214 |
essa | +0.207 |
Neg logit contributions | |
bs | -0.211 |
counts | -0.201 |
gers | -0.200 |
falls | -0.198 |
hits | -0.197 |
Token | that |
Token ID | 326 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
bs | +0.120 |
gers | +0.114 |
counts | +0.114 |
falls | +0.111 |
hits | +0.111 |
Neg logit contributions | |
Baby | -0.186 |
onda | -0.166 |
homeless | -0.162 |
dessert | -0.161 |
essa | -0.157 |
Token | were |
Token ID | 547 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
Baby | +0.104 |
homeless | +0.093 |
onda | +0.091 |
Baby | +0.089 |
rice | +0.088 |
Neg logit contributions | |
hits | -0.074 |
gers | -0.074 |
ines | -0.072 |
counts | -0.072 |
rounds | -0.072 |
Token | qu |
Token ID | 627 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
Baby | +0.090 |
onda | +0.079 |
dessert | +0.077 |
Son | +0.077 |
essa | +0.076 |
Neg logit contributions | |
around | -0.052 |
low | -0.051 |
bs | -0.051 |
hops | -0.048 |
hits | -0.048 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.17e+00 |
Pos logit contributions | |
Baby | +0.107 |
dessert | +0.098 |
homeless | +0.097 |
Son | +0.097 |
Sonny | +0.095 |
Neg logit contributions | |
bs | -0.107 |
around | -0.103 |
ines | -0.102 |
c | -0.099 |
gers | -0.099 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
Baby | +0.711 |
uf | +0.652 |
Son | +0.637 |
essa | +0.622 |
dessert | +0.621 |
Neg logit contributions | |
around | -0.307 |
bs | -0.266 |
falls | -0.254 |
gers | -0.254 |
low | -0.248 |
Token | . |
Token ID | 13 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
Baby | +0.099 |
dessert | +0.089 |
onda | +0.089 |
homeless | +0.088 |
ful | +0.086 |
Neg logit contributions | |
bs | -0.053 |
counts | -0.050 |
hits | -0.049 |
gers | -0.049 |
falls | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Baby | +0.025 |
onda | +0.022 |
homeless | +0.021 |
dessert | +0.021 |
Son | +0.021 |
Neg logit contributions | |
bs | -0.011 |
gers | -0.011 |
counts | -0.011 |
falls | -0.011 |
hits | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
bs | +0.004 |
counts | +0.003 |
gers | +0.003 |
falls | +0.003 |
around | +0.003 |
Neg logit contributions | |
Baby | -0.007 |
onda | -0.007 |
homeless | -0.007 |
dessert | -0.007 |
essa | -0.006 |
Token | The |
Token ID | 464 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
bs | +0.001 |
counts | +0.001 |
gers | +0.001 |
falls | +0.001 |
around | +0.001 |
Neg logit contributions | |
Baby | -0.002 |
onda | -0.002 |
dessert | -0.002 |
homeless | -0.002 |
essa | -0.001 |
Token | much |
Token ID | 881 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
around | +0.000 |
bs | +0.000 |
low | +0.000 |
counts | +0.000 |
gers | +0.000 |
Neg logit contributions | |
Baby | -0.000 |
onda | -0.000 |
essa | -0.000 |
uf | -0.000 |
dessert | -0.000 |
Token | more |
Token ID | 517 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
bs | +0.029 |
gers | +0.027 |
counts | +0.027 |
falls | +0.027 |
hits | +0.027 |
Neg logit contributions | |
Baby | -0.038 |
onda | -0.034 |
homeless | -0.033 |
dessert | -0.033 |
rice | -0.032 |
Token | important |
Token ID | 1593 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
bs | +0.113 |
counts | +0.108 |
gers | +0.108 |
falls | +0.107 |
hits | +0.106 |
Neg logit contributions | |
Baby | -0.158 |
onda | -0.140 |
homeless | -0.137 |
dessert | -0.136 |
essa | -0.132 |
Token | than |
Token ID | 621 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
Baby | +0.131 |
dessert | +0.123 |
onda | +0.121 |
essa | +0.118 |
homeless | +0.118 |
Neg logit contributions | |
bs | -0.059 |
low | -0.056 |
falls | -0.054 |
hits | -0.054 |
hops | -0.052 |
Token | qu |
Token ID | 627 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
bs | +0.154 |
around | +0.150 |
counts | +0.144 |
low | +0.144 |
until | +0.144 |
Neg logit contributions | |
Baby | -0.179 |
onda | -0.163 |
essa | -0.147 |
uf | -0.143 |
urg | -0.142 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.16e+00 |
Pos logit contributions | |
Baby | +0.098 |
dessert | +0.087 |
essa | +0.085 |
Son | +0.083 |
homeless | +0.083 |
Neg logit contributions | |
bs | -0.133 |
around | -0.127 |
ines | -0.124 |
gers | -0.121 |
hops | -0.121 |
Token | lling |
Token ID | 2680 |
Feature activation | -6.17e+00 |
Pos logit contributions | |
Baby | +0.678 |
uf | +0.601 |
onda | +0.598 |
essa | +0.590 |
homeless | +0.588 |
Neg logit contributions | |
around | -0.308 |
bs | -0.307 |
gers | -0.284 |
likes | -0.283 |
falls | -0.275 |
Token | . |
Token ID | 13 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
Baby | +0.447 |
onda | +0.405 |
dessert | +0.397 |
homeless | +0.397 |
rice | +0.387 |
Neg logit contributions | |
bs | -0.197 |
counts | -0.184 |
gers | -0.183 |
hits | -0.178 |
falls | -0.177 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
Baby | +0.059 |
onda | +0.054 |
homeless | +0.053 |
dessert | +0.053 |
essa | +0.052 |
Neg logit contributions | |
bs | -0.026 |
counts | -0.024 |
gers | -0.024 |
falls | -0.023 |
around | -0.023 |
Token | One |
Token ID | 3198 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
Baby | +0.267 |
onda | +0.240 |
homeless | +0.234 |
dessert | +0.233 |
essa | +0.228 |
Neg logit contributions | |
bs | -0.133 |
counts | -0.125 |
gers | -0.125 |
falls | -0.122 |
hits | -0.122 |
Token | day |
Token ID | 1110 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
bs | +0.071 |
counts | +0.067 |
gers | +0.067 |
falls | +0.065 |
hits | +0.064 |
Neg logit contributions | |
Baby | -0.145 |
dessert | -0.130 |
onda | -0.127 |
homeless | -0.126 |
Son | -0.123 |
Token | the |
Token ID | 262 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
Baby | +0.112 |
onda | +0.104 |
homeless | +0.103 |
essa | +0.100 |
dessert | +0.100 |
Neg logit contributions | |
around | -0.046 |
counts | -0.043 |
bs | -0.042 |
hits | -0.041 |
likes | -0.040 |
Token | story |
Token ID | 1621 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
counts | +0.012 |
hops | +0.012 |
hits | +0.012 |
likes | +0.012 |
around | +0.012 |
Neg logit contributions | |
Baby | -0.017 |
onda | -0.014 |
uggets | -0.014 |
homeless | -0.014 |
dessert | -0.014 |
Token | is |
Token ID | 318 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
Baby | +0.153 |
onda | +0.146 |
dessert | +0.141 |
homeless | +0.137 |
essa | +0.134 |
Neg logit contributions | |
bs | -0.075 |
gers | -0.070 |
low | -0.069 |
hops | -0.067 |
tail | -0.066 |
Token | : |
Token ID | 25 |
Feature activation | -3.47e+00 |
Pos logit contributions | |
bs | +0.005 |
around | +0.005 |
But | +0.004 |
hits | +0.004 |
counts | +0.004 |
Neg logit contributions | |
Baby | -0.014 |
onda | -0.012 |
dessert | -0.012 |
homeless | -0.012 |
pudding | -0.012 |
Token | Qu |
Token ID | 2264 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
Baby | +0.208 |
onda | +0.185 |
homeless | +0.179 |
dessert | +0.178 |
essa | +0.175 |
Neg logit contributions | |
bs | -0.160 |
counts | -0.153 |
around | -0.150 |
falls | -0.149 |
hits | -0.149 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.15e+00 |
Pos logit contributions | |
Baby | +0.169 |
homeless | +0.151 |
dessert | +0.150 |
Son | +0.145 |
essa | +0.145 |
Neg logit contributions | |
bs | -0.117 |
likes | -0.112 |
hops | -0.109 |
falls | -0.109 |
gers | -0.108 |
Token | lling |
Token ID | 2680 |
Feature activation | -6.10e+00 |
Pos logit contributions | |
Baby | +0.646 |
onda | +0.596 |
dessert | +0.570 |
homeless | +0.568 |
rice | +0.552 |
Neg logit contributions | |
bs | -0.309 |
counts | -0.294 |
gers | -0.293 |
low | -0.284 |
falls | -0.281 |
Token | over |
Token ID | 625 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
Baby | +0.427 |
onda | +0.396 |
homeless | +0.380 |
dessert | +0.379 |
rice | +0.376 |
Neg logit contributions | |
bs | -0.196 |
gers | -0.189 |
low | -0.183 |
counts | -0.181 |
tail | -0.177 |
Token | small |
Token ID | 1402 |
Feature activation | -8.60e-01 |
Pos logit contributions | |
bs | +0.024 |
gers | +0.023 |
chair | +0.023 |
low | +0.022 |
counts | +0.022 |
Neg logit contributions | |
Baby | -0.035 |
dessert | -0.031 |
onda | -0.031 |
homeless | -0.031 |
rice | -0.030 |
Token | things |
Token ID | 1243 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
Baby | +0.053 |
dessert | +0.048 |
onda | +0.048 |
homeless | +0.047 |
rice | +0.046 |
Neg logit contributions | |
bs | -0.036 |
gers | -0.034 |
low | -0.033 |
falls | -0.033 |
chair | -0.033 |
Token | is |
Token ID | 318 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
bs | +0.022 |
gers | +0.021 |
counts | +0.021 |
hits | +0.020 |
hops | +0.020 |
Neg logit contributions | |
Baby | -0.055 |
onda | -0.050 |
dessert | -0.049 |
homeless | -0.049 |
essa | -0.048 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
bs | +0.001 |
counts | +0.001 |
gers | +0.001 |
falls | +0.001 |
hops | +0.001 |
Neg logit contributions | |
Baby | -0.001 |
dessert | -0.001 |
homeless | -0.001 |
onda | -0.001 |
rice | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
Baby | +0.018 |
onda | +0.016 |
homeless | +0.016 |
essa | +0.015 |
Son | +0.015 |
Neg logit contributions | |
bs | -0.011 |
gers | -0.011 |
low | -0.010 |
middle | -0.010 |
roads | -0.010 |
Token | Jack |
Token ID | 3619 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
Baby | +0.034 |
onda | +0.032 |
homeless | +0.030 |
urg | +0.030 |
essa | +0.029 |
Neg logit contributions | |
around | -0.027 |
counts | -0.026 |
bs | -0.026 |
hits | -0.025 |
But | -0.025 |
Token | never |
Token ID | 1239 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
Baby | +0.228 |
onda | +0.206 |
homeless | +0.198 |
dessert | +0.197 |
essa | +0.194 |
Neg logit contributions | |
bs | -0.123 |
gers | -0.121 |
hits | -0.116 |
counts | -0.116 |
falls | -0.115 |
Token | qu |
Token ID | 627 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
bs | +0.097 |
counts | +0.091 |
hits | +0.088 |
around | +0.087 |
falls | +0.087 |
Neg logit contributions | |
Baby | -0.229 |
onda | -0.209 |
homeless | -0.201 |
dessert | -0.200 |
essa | -0.199 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.15e+00 |
Pos logit contributions | |
Baby | +0.085 |
Son | +0.071 |
essa | +0.070 |
homeless | +0.070 |
Sonny | +0.069 |
Neg logit contributions | |
bs | -0.143 |
around | -0.137 |
hits | -0.134 |
falls | -0.132 |
gers | -0.132 |
Token | led |
Token ID | 992 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
Baby | +0.837 |
onda | +0.758 |
uf | +0.757 |
Son | +0.756 |
homeless | +0.752 |
Neg logit contributions | |
bs | -0.142 |
around | -0.141 |
gers | -0.118 |
counts | -0.115 |
hits | -0.114 |
Token | again |
Token ID | 757 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
Baby | +0.107 |
dessert | +0.096 |
homeless | +0.096 |
onda | +0.096 |
rice | +0.091 |
Neg logit contributions | |
bs | -0.083 |
counts | -0.079 |
chair | -0.079 |
gers | -0.078 |
falls | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
bs | +0.005 |
gers | +0.005 |
counts | +0.004 |
falls | +0.004 |
hits | +0.004 |
Neg logit contributions | |
Baby | -0.011 |
onda | -0.010 |
dessert | -0.010 |
homeless | -0.010 |
essa | -0.010 |
Token | They |
Token ID | 1119 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Baby | +0.054 |
onda | +0.049 |
homeless | +0.048 |
essa | +0.047 |
dessert | +0.047 |
Neg logit contributions | |
bs | -0.020 |
counts | -0.019 |
around | -0.018 |
likes | -0.018 |
falls | -0.018 |
Token | learned |
Token ID | 4499 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
Baby | +0.085 |
onda | +0.076 |
homeless | +0.073 |
dessert | +0.071 |
Sonny | +0.070 |
Neg logit contributions | |
gers | -0.059 |
bs | -0.057 |
hits | -0.056 |
chair | -0.056 |
counts | -0.055 |
Token | on |
Token ID | 319 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
bs | +0.037 |
counts | +0.037 |
gers | +0.036 |
hops | +0.035 |
likes | +0.035 |
Neg logit contributions | |
Baby | -0.089 |
dessert | -0.080 |
onda | -0.079 |
homeless | -0.077 |
ful | -0.076 |
Token | , |
Token ID | 11 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
bs | +0.064 |
counts | +0.056 |
gers | +0.056 |
low | +0.053 |
hops | +0.053 |
Neg logit contributions | |
Baby | -0.242 |
dessert | -0.219 |
onda | -0.219 |
homeless | -0.215 |
Son | -0.212 |
Token | they |
Token ID | 484 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
bs | +0.008 |
around | +0.008 |
gers | +0.008 |
counts | +0.008 |
low | +0.007 |
Neg logit contributions | |
Baby | -0.013 |
onda | -0.012 |
essa | -0.011 |
urg | -0.011 |
uf | -0.011 |
Token | never |
Token ID | 1239 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
Baby | +0.019 |
onda | +0.017 |
homeless | +0.017 |
dessert | +0.017 |
essa | +0.016 |
Neg logit contributions | |
bs | -0.008 |
around | -0.007 |
counts | -0.007 |
low | -0.007 |
likes | -0.007 |
Token | qu |
Token ID | 627 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
bs | +0.087 |
counts | +0.083 |
hits | +0.079 |
falls | +0.078 |
around | +0.077 |
Neg logit contributions | |
Baby | -0.258 |
onda | -0.238 |
essa | -0.229 |
Son | -0.227 |
homeless | -0.227 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.15e+00 |
Pos logit contributions | |
Baby | +0.083 |
essa | +0.071 |
homeless | +0.070 |
Son | +0.069 |
Sonny | +0.068 |
Neg logit contributions | |
bs | -0.136 |
around | -0.131 |
hits | -0.128 |
likes | -0.127 |
falls | -0.127 |
Token | led |
Token ID | 992 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
Baby | +0.844 |
onda | +0.766 |
homeless | +0.761 |
Son | +0.758 |
uf | +0.758 |
Neg logit contributions | |
bs | -0.136 |
around | -0.128 |
counts | -0.112 |
gers | -0.112 |
falls | -0.109 |
Token | again |
Token ID | 757 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
Baby | +0.103 |
dessert | +0.092 |
onda | +0.091 |
homeless | +0.091 |
rice | +0.087 |
Neg logit contributions | |
bs | -0.081 |
counts | -0.077 |
gers | -0.076 |
chair | -0.075 |
falls | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
bs | +0.008 |
gers | +0.008 |
counts | +0.008 |
hits | +0.007 |
falls | +0.007 |
Neg logit contributions | |
Baby | -0.019 |
onda | -0.017 |
dessert | -0.017 |
homeless | -0.017 |
essa | -0.016 |
Token | played |
Token ID | 2826 |
Feature activation | +9.91e-01 |
Pos logit contributions | |
Baby | +0.047 |
onda | +0.042 |
homeless | +0.040 |
dessert | +0.040 |
essa | +0.040 |
Neg logit contributions | |
bs | -0.023 |
counts | -0.021 |
around | -0.021 |
falls | -0.021 |
hits | -0.020 |
Token | together |
Token ID | 1978 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
bs | +0.042 |
counts | +0.039 |
falls | +0.038 |
hits | +0.038 |
gers | +0.038 |
Neg logit contributions | |
Baby | -0.062 |
dessert | -0.057 |
onda | -0.055 |
homeless | -0.055 |
rice | -0.054 |
Token | on |
Token ID | 319 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
counts | +0.036 |
bs | +0.035 |
gers | +0.035 |
likes | +0.034 |
hops | +0.034 |
Neg logit contributions | |
Baby | -0.080 |
dessert | -0.070 |
onda | -0.069 |
Sonny | -0.069 |
homeless | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
bs | +0.074 |
counts | +0.067 |
gers | +0.066 |
low | +0.065 |
sideways | +0.064 |
Neg logit contributions | |
Baby | -0.225 |
dessert | -0.205 |
onda | -0.202 |
rice | -0.200 |
homeless | -0.198 |
Token | they |
Token ID | 484 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
bs | +0.008 |
gers | +0.008 |
counts | +0.008 |
falls | +0.008 |
hits | +0.008 |
Neg logit contributions | |
Baby | -0.013 |
dessert | -0.011 |
homeless | -0.011 |
onda | -0.011 |
rice | -0.011 |
Token | never |
Token ID | 1239 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
Baby | +0.008 |
onda | +0.007 |
homeless | +0.007 |
dessert | +0.007 |
essa | +0.007 |
Neg logit contributions | |
bs | -0.004 |
counts | -0.004 |
gers | -0.004 |
around | -0.004 |
falls | -0.004 |
Token | qu |
Token ID | 627 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
bs | +0.103 |
counts | +0.096 |
hits | +0.093 |
falls | +0.092 |
hops | +0.092 |
Neg logit contributions | |
Baby | -0.259 |
onda | -0.237 |
essa | -0.226 |
Son | -0.225 |
dessert | -0.225 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.14e+00 |
Pos logit contributions | |
Baby | +0.078 |
essa | +0.067 |
Son | +0.067 |
Sonny | +0.065 |
homeless | +0.065 |
Neg logit contributions | |
bs | -0.127 |
around | -0.122 |
hits | -0.118 |
likes | -0.117 |
gers | -0.116 |
Token | led |
Token ID | 992 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
Baby | +0.791 |
onda | +0.712 |
uf | +0.706 |
Son | +0.705 |
homeless | +0.705 |
Neg logit contributions | |
bs | -0.188 |
around | -0.182 |
gers | -0.165 |
counts | -0.163 |
falls | -0.159 |
Token | again |
Token ID | 757 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
Baby | +0.099 |
dessert | +0.088 |
homeless | +0.088 |
onda | +0.088 |
rice | +0.083 |
Neg logit contributions | |
bs | -0.077 |
counts | -0.073 |
gers | -0.072 |
chair | -0.072 |
falls | -0.072 |
Token | over |
Token ID | 625 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
bs | +0.008 |
gers | +0.008 |
counts | +0.008 |
hits | +0.008 |
falls | +0.007 |
Neg logit contributions | |
Baby | -0.018 |
onda | -0.017 |
homeless | -0.016 |
dessert | -0.016 |
essa | -0.016 |
Token | clothes |
Token ID | 8242 |
Feature activation | +7.62e-01 |
Pos logit contributions | |
bs | +0.028 |
counts | +0.027 |
gers | +0.026 |
around | +0.026 |
falls | +0.026 |
Neg logit contributions | |
Baby | -0.049 |
onda | -0.044 |
homeless | -0.042 |
dessert | -0.042 |
essa | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
bs | +0.028 |
counts | +0.026 |
gers | +0.026 |
falls | +0.025 |
hits | +0.025 |
Neg logit contributions | |
Baby | -0.053 |
onda | -0.047 |
homeless | -0.046 |
dessert | -0.046 |
essa | -0.045 |
Token | she |
Token ID | 673 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
Baby | +0.048 |
onda | +0.043 |
homeless | +0.042 |
dessert | +0.042 |
essa | +0.041 |
Neg logit contributions | |
bs | -0.029 |
counts | -0.028 |
gers | -0.028 |
falls | -0.027 |
hits | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
gers | +0.086 |
bs | +0.086 |
hops | +0.083 |
low | +0.083 |
counts | +0.083 |
Neg logit contributions | |
Baby | -0.158 |
onda | -0.148 |
homeless | -0.145 |
dessert | -0.141 |
essa | -0.137 |
Token | sorry |
Token ID | 7926 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
around | +0.007 |
bs | +0.007 |
low | +0.006 |
counts | +0.006 |
hits | +0.006 |
Neg logit contributions | |
Baby | -0.012 |
onda | -0.011 |
essa | -0.010 |
urg | -0.010 |
Son | -0.010 |
Token | for |
Token ID | 329 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
Baby | +0.024 |
dessert | +0.022 |
onda | +0.022 |
homeless | +0.022 |
ful | +0.021 |
Neg logit contributions | |
bs | -0.015 |
gers | -0.014 |
counts | -0.014 |
hits | -0.014 |
hops | -0.014 |
Token | qu |
Token ID | 627 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
low | +0.070 |
until | +0.070 |
around | +0.069 |
bs | +0.067 |
counts | +0.063 |
Neg logit contributions | |
Baby | -0.086 |
onda | -0.078 |
uf | -0.071 |
eddy | -0.070 |
obby | -0.069 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.13e+00 |
Pos logit contributions | |
Baby | +0.088 |
Son | +0.073 |
essa | +0.073 |
dessert | +0.073 |
Sonny | +0.073 |
Neg logit contributions | |
bs | -0.142 |
around | -0.136 |
gers | -0.135 |
likes | -0.132 |
hits | -0.131 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
Baby | +0.675 |
onda | +0.603 |
homeless | +0.590 |
dessert | +0.589 |
uf | +0.584 |
Neg logit contributions | |
bs | -0.300 |
around | -0.285 |
gers | -0.279 |
counts | -0.273 |
falls | -0.271 |
Token | . |
Token ID | 13 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
Baby | +0.094 |
dessert | +0.086 |
homeless | +0.085 |
onda | +0.084 |
ful | +0.082 |
Neg logit contributions | |
bs | -0.051 |
counts | -0.048 |
hits | -0.048 |
hops | -0.047 |
gers | -0.047 |
Token | From |
Token ID | 3574 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Baby | +0.048 |
onda | +0.043 |
homeless | +0.042 |
dessert | +0.042 |
essa | +0.041 |
Neg logit contributions | |
bs | -0.024 |
counts | -0.023 |
gers | -0.022 |
falls | -0.022 |
hits | -0.022 |
Token | that |
Token ID | 326 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
Baby | +0.068 |
onda | +0.061 |
homeless | +0.060 |
dessert | +0.059 |
essa | +0.059 |
Neg logit contributions | |
bs | -0.043 |
counts | -0.041 |
hops | -0.040 |
falls | -0.039 |
gers | -0.039 |
Token | day |
Token ID | 1110 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
Baby | +0.149 |
onda | +0.134 |
dessert | +0.133 |
homeless | +0.132 |
Son | +0.129 |
Neg logit contributions | |
bs | -0.069 |
gers | -0.066 |
counts | -0.064 |
falls | -0.062 |
hits | -0.062 |
Token | peaceful |
Token ID | 12309 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
Baby | +0.078 |
onda | +0.069 |
homeless | +0.068 |
dessert | +0.067 |
Son | +0.065 |
Neg logit contributions | |
bs | -0.050 |
gers | -0.050 |
counts | -0.050 |
hits | -0.048 |
hops | -0.047 |
Token | everyone |
Token ID | 2506 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
bs | +0.006 |
counts | +0.006 |
gers | +0.006 |
falls | +0.006 |
hits | +0.006 |
Neg logit contributions | |
Baby | -0.013 |
onda | -0.012 |
dessert | -0.012 |
homeless | -0.012 |
essa | -0.011 |
Token | just |
Token ID | 655 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
Baby | +0.045 |
onda | +0.041 |
homeless | +0.040 |
dessert | +0.040 |
essa | +0.039 |
Neg logit contributions | |
bs | -0.020 |
gers | -0.019 |
counts | -0.019 |
hits | -0.019 |
falls | -0.018 |
Token | stopped |
Token ID | 5025 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
gers | +0.022 |
bs | +0.021 |
chair | +0.020 |
counts | +0.019 |
hits | +0.019 |
Neg logit contributions | |
Baby | -0.039 |
dessert | -0.035 |
homeless | -0.035 |
onda | -0.034 |
Son | -0.034 |
Token | qu |
Token ID | 627 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Baby | +0.147 |
onda | +0.130 |
homeless | +0.126 |
dessert | +0.126 |
essa | +0.122 |
Neg logit contributions | |
bs | -0.114 |
counts | -0.109 |
gers | -0.109 |
falls | -0.107 |
hits | -0.106 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.10e+00 |
Pos logit contributions | |
Baby | +0.113 |
dessert | +0.100 |
Sonny | +0.100 |
homeless | +0.100 |
Son | +0.099 |
Neg logit contributions | |
bs | -0.106 |
around | -0.104 |
c | -0.103 |
hits | -0.098 |
ines | -0.098 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
Baby | +0.710 |
uf | +0.642 |
dessert | +0.620 |
Son | +0.620 |
onda | +0.614 |
Neg logit contributions | |
around | -0.305 |
bs | -0.263 |
gers | -0.261 |
falls | -0.252 |
low | -0.251 |
Token | . |
Token ID | 13 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
Baby | +0.107 |
onda | +0.097 |
dessert | +0.096 |
homeless | +0.095 |
essa | +0.093 |
Neg logit contributions | |
bs | -0.047 |
counts | -0.044 |
gers | -0.043 |
hits | -0.043 |
falls | -0.043 |
Token | |
Token ID | 220 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
Baby | +0.039 |
onda | +0.035 |
homeless | +0.034 |
dessert | +0.034 |
essa | +0.033 |
Neg logit contributions | |
bs | -0.021 |
counts | -0.020 |
gers | -0.020 |
falls | -0.020 |
hits | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -8.48e-03 |
Pos logit contributions | |
bs | +0.056 |
around | +0.052 |
falls | +0.050 |
counts | +0.050 |
But | +0.049 |
Neg logit contributions | |
Baby | -0.195 |
onda | -0.183 |
dessert | -0.183 |
homeless | -0.181 |
rice | -0.180 |
Token |
|
Token ID | 198 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
Baby | +0.001 |
onda | +0.001 |
dessert | +0.001 |
rice | +0.001 |
homeless | +0.001 |
Neg logit contributions | |
bs | -0.000 |
low | -0.000 |
around | -0.000 |
counts | -0.000 |
gers | -0.000 |
Token | Why |
Token ID | 5195 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
Baby | +0.046 |
dessert | +0.046 |
urg | +0.044 |
uf | +0.043 |
rice | +0.043 |
Neg logit contributions | |
bs | -0.038 |
gers | -0.037 |
around | -0.036 |
counts | -0.035 |
low | -0.035 |
Token | are |
Token ID | 389 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
Baby | +0.039 |
onda | +0.035 |
homeless | +0.034 |
dessert | +0.034 |
essa | +0.033 |
Neg logit contributions | |
bs | -0.019 |
counts | -0.018 |
falls | -0.018 |
gers | -0.018 |
likes | -0.018 |
Token | you |
Token ID | 345 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
Baby | +0.050 |
onda | +0.047 |
urg | +0.044 |
essa | +0.043 |
uf | +0.043 |
Neg logit contributions | |
bs | -0.033 |
around | -0.031 |
falls | -0.029 |
gers | -0.029 |
hits | -0.028 |
Token | two |
Token ID | 734 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
Baby | +0.053 |
onda | +0.048 |
essa | +0.045 |
uf | +0.045 |
dessert | +0.044 |
Neg logit contributions | |
around | -0.035 |
bs | -0.031 |
hits | -0.031 |
likes | -0.031 |
low | -0.030 |
Token | qu |
Token ID | 627 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
Baby | +0.132 |
onda | +0.117 |
dessert | +0.115 |
homeless | +0.115 |
essa | +0.111 |
Neg logit contributions | |
bs | -0.101 |
gers | -0.096 |
counts | -0.096 |
hits | -0.094 |
falls | -0.094 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.09e+00 |
Pos logit contributions | |
Baby | +0.086 |
homeless | +0.071 |
dessert | +0.069 |
Son | +0.069 |
essa | +0.069 |
Neg logit contributions | |
bs | -0.141 |
around | -0.133 |
hits | -0.131 |
gers | -0.131 |
ines | -0.131 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
Baby | +0.643 |
onda | +0.568 |
uf | +0.561 |
homeless | +0.558 |
dessert | +0.554 |
Neg logit contributions | |
bs | -0.333 |
around | -0.326 |
gers | -0.309 |
falls | -0.305 |
counts | -0.304 |
Token | ? |
Token ID | 30 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
Baby | +0.089 |
onda | +0.080 |
homeless | +0.077 |
dessert | +0.077 |
essa | +0.075 |
Neg logit contributions | |
bs | -0.058 |
gers | -0.055 |
around | -0.055 |
counts | -0.055 |
falls | -0.054 |
Token | There |
Token ID | 1318 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
Baby | +0.026 |
onda | +0.023 |
homeless | +0.022 |
dessert | +0.022 |
Son | +0.021 |
Neg logit contributions | |
bs | -0.018 |
gers | -0.018 |
counts | -0.017 |
falls | -0.017 |
hops | -0.017 |
Token | are |
Token ID | 389 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
Baby | +0.168 |
onda | +0.152 |
homeless | +0.148 |
dessert | +0.147 |
essa | +0.146 |
Neg logit contributions | |
bs | -0.064 |
counts | -0.060 |
falls | -0.059 |
gers | -0.058 |
around | -0.058 |
Token | plenty |
Token ID | 6088 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
Baby | +0.054 |
onda | +0.049 |
essa | +0.046 |
uf | +0.046 |
Son | +0.045 |
Neg logit contributions | |
bs | -0.030 |
counts | -0.030 |
around | -0.030 |
gers | -0.029 |
falls | -0.029 |
Token | walked |
Token ID | 6807 |
Feature activation | -7.55e-02 |
Pos logit contributions | |
Baby | +0.066 |
onda | +0.060 |
dessert | +0.059 |
homeless | +0.059 |
essa | +0.058 |
Neg logit contributions | |
bs | -0.029 |
around | -0.026 |
likes | -0.026 |
falls | -0.025 |
counts | -0.025 |
Token | in |
Token ID | 287 |
Feature activation | -9.67e-01 |
Pos logit contributions | |
Baby | +0.005 |
onda | +0.004 |
homeless | +0.004 |
dessert | +0.004 |
Son | +0.004 |
Neg logit contributions | |
bs | -0.003 |
counts | -0.003 |
falls | -0.003 |
hops | -0.003 |
chair | -0.003 |
Token | on |
Token ID | 319 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
Baby | +0.060 |
onda | +0.052 |
dessert | +0.052 |
homeless | +0.051 |
Son | +0.049 |
Neg logit contributions | |
bs | -0.043 |
gers | -0.041 |
counts | -0.040 |
falls | -0.039 |
hops | -0.039 |
Token | us |
Token ID | 514 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
bs | +0.124 |
counts | +0.119 |
gers | +0.118 |
falls | +0.116 |
hits | +0.116 |
Neg logit contributions | |
Baby | -0.177 |
onda | -0.158 |
homeless | -0.152 |
dessert | -0.151 |
essa | -0.149 |
Token | qu |
Token ID | 627 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
Baby | +0.077 |
onda | +0.069 |
dessert | +0.068 |
homeless | +0.067 |
Son | +0.065 |
Neg logit contributions | |
bs | -0.044 |
gers | -0.043 |
hits | -0.041 |
counts | -0.041 |
hops | -0.041 |
Token | arre |
Token ID | 9624 |
Feature activation | -9.09e+00 |
Pos logit contributions | |
Baby | +0.091 |
dessert | +0.076 |
homeless | +0.076 |
essa | +0.075 |
Son | +0.074 |
Neg logit contributions | |
bs | -0.134 |
around | -0.131 |
gers | -0.129 |
hits | -0.128 |
hops | -0.128 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Baby | +0.657 |
onda | +0.585 |
homeless | +0.575 |
dessert | +0.575 |
essa | +0.565 |
Neg logit contributions | |
bs | -0.309 |
around | -0.296 |
gers | -0.292 |
counts | -0.288 |
falls | -0.285 |
Token | and |
Token ID | 290 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
Baby | +0.097 |
onda | +0.086 |
dessert | +0.086 |
homeless | +0.085 |
Son | +0.083 |
Neg logit contributions | |
bs | -0.056 |
counts | -0.053 |
gers | -0.052 |
hits | -0.052 |
falls | -0.051 |
Token | heard |
Token ID | 2982 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
Baby | +0.036 |
dessert | +0.031 |
onda | +0.031 |
homeless | +0.031 |
Son | +0.030 |
Neg logit contributions | |
bs | -0.031 |
gers | -0.031 |
counts | -0.029 |
chair | -0.029 |
falls | -0.029 |
Token | us |
Token ID | 514 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
bs | +0.081 |
around | +0.081 |
low | +0.074 |
falls | +0.074 |
counts | +0.073 |
Neg logit contributions | |
Baby | -0.147 |
onda | -0.133 |
essa | -0.125 |
uf | -0.123 |
homeless | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
Baby | +0.087 |
onda | +0.078 |
homeless | +0.076 |
dessert | +0.076 |
essa | +0.075 |
Neg logit contributions | |
bs | -0.035 |
around | -0.035 |
falls | -0.033 |
counts | -0.033 |
low | -0.032 |
Token | there |
Token ID | 612 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
Baby | +0.000 |
onda | +0.000 |
dessert | +0.000 |
homeless | +0.000 |
Son | +0.000 |
Neg logit contributions | |
bs | -0.000 |
counts | -0.000 |
gers | -0.000 |
falls | -0.000 |
hits | -0.000 |
Token | was |
Token ID | 373 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
Baby | +0.025 |
onda | +0.023 |
dessert | +0.022 |
homeless | +0.022 |
Son | +0.022 |
Neg logit contributions | |
bs | -0.013 |
gers | -0.012 |
hits | -0.012 |
counts | -0.012 |
falls | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
bs | +0.041 |
counts | +0.039 |
gers | +0.039 |
falls | +0.039 |
around | +0.039 |
Neg logit contributions | |
Baby | -0.047 |
onda | -0.041 |
homeless | -0.040 |
dessert | -0.039 |
essa | -0.038 |
Token | little |
Token ID | 1310 |
Feature activation | -4.76e+00 |
Pos logit contributions | |
Baby | +0.047 |
onda | +0.042 |
essa | +0.040 |
dessert | +0.039 |
homeless | +0.039 |
Neg logit contributions | |
bs | -0.040 |
low | -0.039 |
gers | -0.038 |
likes | -0.038 |
hops | -0.038 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
Baby | +0.274 |
onda | +0.237 |
dessert | +0.234 |
homeless | +0.233 |
Son | +0.225 |
Neg logit contributions | |
bs | -0.232 |
counts | -0.221 |
gers | -0.217 |
falls | -0.217 |
hits | -0.215 |
Token | named |
Token ID | 3706 |
Feature activation | +4.41e+00 |
Pos logit contributions | |
Baby | +0.002 |
onda | +0.001 |
homeless | +0.001 |
dessert | +0.001 |
essa | +0.001 |
Neg logit contributions | |
hits | -0.001 |
gers | -0.001 |
falls | -0.001 |
bs | -0.001 |
counts | -0.001 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
bs | +0.180 |
counts | +0.170 |
around | +0.168 |
low | +0.165 |
gers | +0.165 |
Neg logit contributions | |
Baby | -0.287 |
onda | -0.261 |
homeless | -0.253 |
dessert | -0.251 |
rice | -0.244 |
Token | . |
Token ID | 13 |
Feature activation | -8.77e-02 |
Pos logit contributions | |
bs | +0.008 |
gers | +0.008 |
hits | +0.007 |
counts | +0.007 |
falls | +0.007 |
Neg logit contributions | |
Baby | -0.018 |
onda | -0.017 |
homeless | -0.016 |
dessert | -0.016 |
essa | -0.016 |
Token | She |
Token ID | 1375 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
Baby | +0.006 |
dessert | +0.006 |
onda | +0.005 |
homeless | +0.005 |
Son | +0.005 |
Neg logit contributions | |
bs | -0.003 |
gers | -0.003 |
hits | -0.003 |
falls | -0.003 |
counts | -0.003 |
Token | loved |
Token ID | 6151 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
bs | +0.019 |
gers | +0.018 |
hits | +0.018 |
counts | +0.017 |
falls | +0.017 |
Neg logit contributions | |
Baby | -0.036 |
onda | -0.032 |
homeless | -0.031 |
dessert | -0.031 |
essa | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
around | +0.014 |
bs | +0.013 |
counts | +0.013 |
spins | +0.013 |
hits | +0.013 |
Neg logit contributions | |
Baby | -0.031 |
onda | -0.030 |
uf | -0.027 |
homeless | -0.027 |
Son | -0.027 |
Token | sweet |
Token ID | 6029 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
bs | +0.051 |
counts | +0.050 |
gers | +0.049 |
hits | +0.048 |
falls | +0.048 |
Neg logit contributions | |
Baby | -0.089 |
onda | -0.078 |
homeless | -0.077 |
dessert | -0.077 |
Son | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
Baby | +0.102 |
onda | +0.096 |
homeless | +0.092 |
dessert | +0.089 |
essa | +0.089 |
Neg logit contributions | |
bs | -0.026 |
around | -0.025 |
gers | -0.024 |
counts | -0.024 |
likes | -0.024 |
Token | juicy |
Token ID | 35075 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
Baby | +0.003 |
dessert | +0.003 |
onda | +0.003 |
homeless | +0.003 |
Son | +0.003 |
Neg logit contributions | |
bs | -0.003 |
gers | -0.003 |
counts | -0.003 |
hits | -0.002 |
falls | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
bs | +0.019 |
counts | +0.018 |
gers | +0.017 |
falls | +0.017 |
hits | +0.017 |
Neg logit contributions | |
Baby | -0.051 |
onda | -0.047 |
dessert | -0.046 |
homeless | -0.045 |
essa | -0.044 |
Token | But |
Token ID | 887 |
Feature activation | +4.21e+00 |
Pos logit contributions | |
Baby | +0.004 |
onda | +0.004 |
homeless | +0.003 |
dessert | +0.003 |
essa | +0.003 |
Neg logit contributions | |
bs | -0.001 |
counts | -0.001 |
gers | -0.001 |
falls | -0.001 |
hits | -0.001 |
Token | one |
Token ID | 530 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
bs | +0.156 |
gers | +0.147 |
counts | +0.143 |
falls | +0.140 |
hits | +0.139 |
Neg logit contributions | |
Baby | -0.292 |
dessert | -0.261 |
onda | -0.257 |
homeless | -0.257 |
Son | -0.251 |
Token | corn |
Token ID | 11676 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
bs | +0.121 |
gers | +0.118 |
ways | +0.118 |
counts | +0.114 |
hits | +0.114 |
Neg logit contributions | |
Baby | -0.226 |
dessert | -0.202 |
onda | -0.195 |
homeless | -0.192 |
rice | -0.190 |
Token | wasn |
Token ID | 2492 |
Feature activation | +6.17e-02 |
Pos logit contributions | |
Baby | +0.110 |
onda | +0.100 |
dessert | +0.098 |
homeless | +0.098 |
rice | +0.095 |
Neg logit contributions | |
bs | -0.051 |
gers | -0.047 |
counts | -0.047 |
low | -0.046 |
hits | -0.045 |
Token | 't |
Token ID | 470 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
low | +0.002 |
around | +0.002 |
bs | +0.002 |
falls | +0.002 |
hits | +0.002 |
Neg logit contributions | |
Baby | -0.004 |
onda | -0.004 |
essa | -0.004 |
keeping | -0.004 |
dessert | -0.004 |
Token | quite |
Token ID | 2407 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
Baby | +0.001 |
onda | +0.001 |
obby | +0.001 |
edd | +0.001 |
uf | +0.001 |
Neg logit contributions | |
around | -0.000 |
low | -0.000 |
bs | -0.000 |
until | -0.000 |
high | -0.000 |
Token | right |
Token ID | 826 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
bs | +0.024 |
low | +0.023 |
around | +0.022 |
likes | +0.021 |
hops | +0.021 |
Neg logit contributions | |
Baby | -0.044 |
onda | -0.042 |
edd | -0.037 |
essa | -0.037 |
obby | -0.037 |
Token | she |
Token ID | 673 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
Baby | +0.126 |
homeless | +0.115 |
Son | +0.110 |
Sonny | +0.109 |
onda | +0.108 |
Neg logit contributions | |
ines | -0.058 |
gers | -0.058 |
counts | -0.057 |
bs | -0.056 |
falls | -0.054 |
Token | said |
Token ID | 531 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
opl | +0.103 |
bs | +0.101 |
gers | +0.101 |
counts | +0.100 |
chair | +0.099 |
Neg logit contributions | |
Baby | -0.163 |
onda | -0.150 |
homeless | -0.146 |
essa | -0.140 |
Son | -0.138 |
Token | . |
Token ID | 13 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
bs | +0.003 |
gers | +0.003 |
counts | +0.003 |
hits | +0.003 |
falls | +0.003 |
Neg logit contributions | |
Baby | -0.009 |
onda | -0.008 |
homeless | -0.008 |
dessert | -0.008 |
essa | -0.008 |
Token | The |
Token ID | 383 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
Baby | +0.027 |
onda | +0.025 |
homeless | +0.024 |
dessert | +0.024 |
essa | +0.024 |
Neg logit contributions | |
bs | -0.012 |
counts | -0.011 |
gers | -0.011 |
falls | -0.011 |
hits | -0.011 |
Token | gentle |
Token ID | 10296 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
bs | +0.037 |
counts | +0.035 |
gers | +0.035 |
falls | +0.034 |
hits | +0.034 |
Neg logit contributions | |
Baby | -0.052 |
onda | -0.046 |
homeless | -0.045 |
dessert | -0.045 |
essa | -0.043 |
Token | pony |
Token ID | 26902 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
bs | +0.073 |
gers | +0.069 |
counts | +0.069 |
around | +0.068 |
hits | +0.068 |
Neg logit contributions | |
Baby | -0.115 |
onda | -0.103 |
dessert | -0.099 |
homeless | -0.099 |
essa | -0.096 |
Token | looked |
Token ID | 3114 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
bs | +0.142 |
gers | +0.137 |
hits | +0.132 |
chairs | +0.132 |
opl | +0.132 |
Neg logit contributions | |
Baby | -0.234 |
onda | -0.211 |
homeless | -0.206 |
Son | -0.197 |
uf | -0.195 |
Token | around |
Token ID | 1088 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
Baby | +0.042 |
homeless | +0.042 |
rice | +0.040 |
dessert | +0.038 |
Son | +0.038 |
Neg logit contributions | |
tail | -0.034 |
ebra | -0.034 |
pter | -0.033 |
chair | -0.033 |
opl | -0.033 |
Token | nervously |
Token ID | 43324 |
Feature activation | -6.63e-01 |
Pos logit contributions | |
bs | +0.107 |
gers | +0.103 |
chairs | +0.103 |
chair | +0.101 |
falls | +0.100 |
Neg logit contributions | |
Baby | -0.139 |
dessert | -0.126 |
homeless | -0.121 |
Son | -0.120 |
Baby | -0.118 |
Token | , |
Token ID | 11 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
Baby | +0.043 |
homeless | +0.039 |
dessert | +0.039 |
Son | +0.038 |
onda | +0.038 |
Neg logit contributions | |
bs | -0.024 |
gers | -0.023 |
counts | -0.022 |
chairs | -0.022 |
hits | -0.022 |
Token | but |
Token ID | 475 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
bs | +0.013 |
gers | +0.013 |
counts | +0.012 |
hits | +0.012 |
falls | +0.012 |
Neg logit contributions | |
Baby | -0.019 |
homeless | -0.017 |
Son | -0.017 |
dessert | -0.016 |
Sonny | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
bs | +0.036 |
spins | +0.036 |
gers | +0.035 |
hops | +0.035 |
low | +0.035 |
Neg logit contributions | |
dessert | -0.058 |
Baby | -0.058 |
homeless | -0.054 |
uf | -0.054 |
rice | -0.054 |
Token | The |
Token ID | 383 |
Feature activation | +9.73e-01 |
Pos logit contributions | |
Baby | +0.016 |
dessert | +0.014 |
homeless | +0.014 |
Son | +0.014 |
onda | +0.014 |
Neg logit contributions | |
bs | -0.008 |
gers | -0.007 |
falls | -0.007 |
hops | -0.007 |
counts | -0.006 |
Token | serious |
Token ID | 2726 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
bs | +0.051 |
gers | +0.045 |
falls | +0.045 |
chairs | +0.044 |
spins | +0.044 |
Neg logit contributions | |
Baby | -0.055 |
dessert | -0.051 |
homeless | -0.050 |
onda | -0.047 |
pudding | -0.045 |
Token | gorilla |
Token ID | 45314 |
Feature activation | +8.23e-01 |
Pos logit contributions | |
bs | +0.059 |
gers | +0.054 |
counts | +0.054 |
falls | +0.054 |
hops | +0.053 |
Neg logit contributions | |
Baby | -0.089 |
onda | -0.079 |
dessert | -0.078 |
homeless | -0.078 |
essa | -0.075 |
Token | said |
Token ID | 531 |
Feature activation | +5.69e-02 |
Pos logit contributions | |
bs | +0.029 |
gers | +0.028 |
counts | +0.026 |
hits | +0.026 |
low | +0.026 |
Neg logit contributions | |
Baby | -0.057 |
onda | -0.053 |
homeless | -0.051 |
dessert | -0.051 |
Son | -0.049 |
Token | no |
Token ID | 645 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
bs | +0.002 |
gers | +0.002 |
chairs | +0.001 |
counts | +0.001 |
hops | +0.001 |
Neg logit contributions | |
Baby | -0.004 |
dessert | -0.004 |
homeless | -0.004 |
rice | -0.004 |
pudding | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
bs | +0.088 |
counts | +0.082 |
gers | +0.080 |
falls | +0.078 |
around | +0.078 |
Neg logit contributions | |
Baby | -0.286 |
onda | -0.262 |
homeless | -0.254 |
dessert | -0.252 |
essa | -0.250 |
Token | The |
Token ID | 383 |
Feature activation | +9.38e-01 |
Pos logit contributions | |
Baby | +0.014 |
dessert | +0.012 |
homeless | +0.012 |
onda | +0.012 |
Son | +0.012 |
Neg logit contributions | |
bs | -0.006 |
gers | -0.005 |
hops | -0.005 |
falls | -0.005 |
counts | -0.005 |
Token | monkey |
Token ID | 21657 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
bs | +0.052 |
chairs | +0.047 |
gers | +0.046 |
falls | +0.045 |
spins | +0.045 |
Neg logit contributions | |
Baby | -0.049 |
dessert | -0.047 |
homeless | -0.046 |
onda | -0.042 |
rice | -0.040 |
Token | was |
Token ID | 373 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
bs | +0.148 |
gers | +0.143 |
counts | +0.137 |
chairs | +0.135 |
spins | +0.135 |
Neg logit contributions | |
Baby | -0.262 |
onda | -0.242 |
homeless | -0.235 |
dessert | -0.231 |
uf | -0.222 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
bs | +0.010 |
counts | +0.009 |
around | +0.009 |
low | +0.009 |
hits | +0.009 |
Neg logit contributions | |
Baby | -0.027 |
onda | -0.024 |
dessert | -0.023 |
homeless | -0.023 |
essa | -0.023 |
Token | ," |
Token ID | 553 |
Feature activation | -7.12e-01 |
Pos logit contributions | |
bs | +0.029 |
counts | +0.027 |
gers | +0.027 |
hops | +0.026 |
hits | +0.026 |
Neg logit contributions | |
Baby | -0.053 |
onda | -0.047 |
dessert | -0.047 |
homeless | -0.046 |
rice | -0.045 |
Token | Lily |
Token ID | 20037 |
Feature activation | +7.47e-01 |
Pos logit contributions | |
Baby | +0.041 |
dessert | +0.036 |
homeless | +0.035 |
rice | +0.035 |
onda | +0.034 |
Neg logit contributions | |
gers | -0.034 |
bs | -0.034 |
ways | -0.033 |
counts | -0.033 |
ines | -0.032 |
Token | said |
Token ID | 531 |
Feature activation | +9.59e-01 |
Pos logit contributions | |
bs | +0.021 |
gers | +0.020 |
low | +0.018 |
hits | +0.018 |
counts | +0.017 |
Neg logit contributions | |
Baby | -0.058 |
onda | -0.053 |
dessert | -0.052 |
homeless | -0.052 |
essa | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +6.95e-01 |
Pos logit contributions | |
bs | +0.024 |
around | +0.021 |
counts | +0.021 |
falls | +0.021 |
low | +0.021 |
Neg logit contributions | |
Baby | -0.079 |
onda | -0.072 |
homeless | -0.070 |
dessert | -0.069 |
essa | -0.068 |
Token | " |
Token ID | 366 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
bs | +0.025 |
gers | +0.023 |
counts | +0.023 |
falls | +0.023 |
hits | +0.023 |
Neg logit contributions | |
Baby | -0.049 |
onda | -0.044 |
homeless | -0.043 |
dessert | -0.043 |
essa | -0.042 |
Token | But |
Token ID | 1537 |
Feature activation | +4.81e+00 |
Pos logit contributions | |
Baby | +0.016 |
onda | +0.014 |
homeless | +0.013 |
dessert | +0.013 |
essa | +0.013 |
Neg logit contributions | |
bs | -0.014 |
counts | -0.014 |
gers | -0.014 |
falls | -0.014 |
hits | -0.014 |
Token | I |
Token ID | 314 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
bs | +0.222 |
counts | +0.212 |
gers | +0.210 |
falls | +0.209 |
hits | +0.207 |
Neg logit contributions | |
Baby | -0.287 |
onda | -0.255 |
homeless | -0.246 |
dessert | -0.244 |
essa | -0.239 |
Token | like |
Token ID | 588 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
Baby | +0.026 |
homeless | +0.023 |
onda | +0.023 |
dessert | +0.023 |
uf | +0.022 |
Neg logit contributions | |
gers | -0.015 |
chair | -0.014 |
bs | -0.014 |
chairs | -0.014 |
counts | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
bs | +0.117 |
counts | +0.111 |
gers | +0.110 |
falls | +0.108 |
hits | +0.108 |
Neg logit contributions | |
Baby | -0.183 |
onda | -0.164 |
homeless | -0.159 |
dessert | -0.158 |
essa | -0.155 |
Token | sun |
Token ID | 4252 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
bs | +0.052 |
counts | +0.049 |
gers | +0.048 |
hops | +0.047 |
falls | +0.047 |
Neg logit contributions | |
Baby | -0.061 |
dessert | -0.055 |
homeless | -0.054 |
onda | -0.053 |
rice | -0.052 |
Token | more |
Token ID | 517 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
bs | +0.025 |
gers | +0.024 |
counts | +0.023 |
hops | +0.023 |
hits | +0.023 |
Neg logit contributions | |
Baby | -0.053 |
dessert | -0.048 |
onda | -0.048 |
homeless | -0.047 |
rice | -0.046 |
Token | not |
Token ID | 407 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
bs | +0.021 |
gers | +0.020 |
counts | +0.020 |
falls | +0.020 |
hits | +0.019 |
Neg logit contributions | |
Baby | -0.031 |
onda | -0.028 |
homeless | -0.027 |
dessert | -0.027 |
essa | -0.026 |
Token | there |
Token ID | 612 |
Feature activation | +7.65e-02 |
Pos logit contributions | |
Baby | +0.014 |
onda | +0.013 |
Son | +0.012 |
uf | +0.012 |
homeless | +0.012 |
Neg logit contributions | |
low | -0.007 |
around | -0.007 |
bs | -0.007 |
counts | -0.006 |
likes | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
bs | +0.003 |
gers | +0.002 |
hops | +0.002 |
counts | +0.002 |
hits | +0.002 |
Neg logit contributions | |
Baby | -0.005 |
dessert | -0.005 |
onda | -0.005 |
homeless | -0.005 |
essa | -0.005 |
Token | He |
Token ID | 679 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
bs | +0.006 |
gers | +0.006 |
counts | +0.006 |
hops | +0.006 |
hits | +0.006 |
Neg logit contributions | |
Baby | -0.018 |
dessert | -0.016 |
homeless | -0.016 |
onda | -0.015 |
Son | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
bs | +0.005 |
opl | +0.005 |
gers | +0.005 |
ways | +0.004 |
counts | +0.004 |
Neg logit contributions | |
Baby | -0.007 |
onda | -0.006 |
homeless | -0.006 |
uf | -0.006 |
dessert | -0.005 |
Token | still |
Token ID | 991 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
bs | +0.022 |
counts | +0.021 |
gers | +0.021 |
falls | +0.021 |
hits | +0.021 |
Neg logit contributions | |
Baby | -0.034 |
onda | -0.030 |
homeless | -0.029 |
dessert | -0.029 |
essa | -0.028 |
Token | in |
Token ID | 287 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
bs | +0.030 |
around | +0.028 |
low | +0.028 |
gers | +0.028 |
counts | +0.028 |
Neg logit contributions | |
Baby | -0.049 |
onda | -0.044 |
homeless | -0.042 |
dessert | -0.042 |
essa | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
Baby | +0.045 |
dessert | +0.040 |
homeless | +0.039 |
onda | +0.039 |
Son | +0.037 |
Neg logit contributions | |
bs | -0.034 |
gers | -0.032 |
counts | -0.032 |
falls | -0.031 |
likes | -0.031 |
Token | park |
Token ID | 3952 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
bs | +0.035 |
counts | +0.033 |
gers | +0.033 |
falls | +0.032 |
around | +0.032 |
Neg logit contributions | |
Baby | -0.046 |
onda | -0.041 |
dessert | -0.040 |
homeless | -0.040 |
rice | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
bs | +0.016 |
hops | +0.014 |
spins | +0.014 |
counts | +0.014 |
hits | +0.014 |
Neg logit contributions | |
Baby | -0.024 |
homeless | -0.023 |
dessert | -0.023 |
ful | -0.021 |
rice | -0.021 |
Token | feeling |
Token ID | 4203 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
bs | +0.032 |
gers | +0.029 |
counts | +0.028 |
ars | +0.027 |
ways | +0.027 |
Neg logit contributions | |
Baby | -0.042 |
dessert | -0.040 |
homeless | -0.039 |
rice | -0.037 |
wasting | -0.036 |
Token | shed |
Token ID | 14999 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
Baby | +0.167 |
dessert | +0.150 |
homeless | +0.150 |
uf | +0.149 |
onda | +0.148 |
Neg logit contributions | |
bs | -0.198 |
counts | -0.191 |
spins | -0.187 |
gers | -0.185 |
falls | -0.185 |
Token | . |
Token ID | 13 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
Baby | +0.097 |
onda | +0.089 |
homeless | +0.088 |
dessert | +0.087 |
essa | +0.085 |
Neg logit contributions | |
bs | -0.041 |
counts | -0.037 |
gers | -0.037 |
falls | -0.036 |
hops | -0.036 |
Token | He |
Token ID | 679 |
Feature activation | -7.26e-02 |
Pos logit contributions | |
Baby | +0.025 |
homeless | +0.022 |
dessert | +0.022 |
onda | +0.022 |
Son | +0.022 |
Neg logit contributions | |
bs | -0.010 |
gers | -0.010 |
falls | -0.009 |
counts | -0.009 |
hops | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | -6.02e-01 |
Pos logit contributions | |
Baby | +0.005 |
onda | +0.005 |
dessert | +0.005 |
homeless | +0.005 |
essa | +0.005 |
Neg logit contributions | |
bs | -0.002 |
counts | -0.002 |
gers | -0.002 |
falls | -0.002 |
hits | -0.002 |
Token | never |
Token ID | 1239 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
Baby | +0.034 |
onda | +0.030 |
homeless | +0.029 |
dessert | +0.029 |
essa | +0.028 |
Neg logit contributions | |
bs | -0.030 |
gers | -0.029 |
counts | -0.028 |
falls | -0.028 |
hops | -0.028 |
Token | seen |
Token ID | 1775 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
bs | +0.120 |
counts | +0.115 |
around | +0.111 |
low | +0.110 |
hits | +0.110 |
Neg logit contributions | |
Baby | -0.228 |
onda | -0.203 |
homeless | -0.194 |
dessert | -0.194 |
Son | -0.193 |
Token | a |
Token ID | 257 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
bs | +0.073 |
gers | +0.070 |
falls | +0.069 |
counts | +0.068 |
hops | +0.068 |
Neg logit contributions | |
Baby | -0.078 |
dessert | -0.069 |
onda | -0.068 |
homeless | -0.068 |
essa | -0.065 |
Token | gun |
Token ID | 2485 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
Baby | +0.039 |
onda | +0.035 |
homeless | +0.034 |
dessert | +0.034 |
essa | +0.033 |
Neg logit contributions | |
bs | -0.023 |
gers | -0.022 |
counts | -0.022 |
falls | -0.021 |
hits | -0.021 |
Token | before |
Token ID | 878 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
bs | +0.021 |
counts | +0.020 |
gers | +0.020 |
around | +0.020 |
falls | +0.020 |
Neg logit contributions | |
Baby | -0.040 |
onda | -0.036 |
homeless | -0.035 |
dessert | -0.035 |
essa | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
bs | +0.006 |
gers | +0.006 |
counts | +0.006 |
hits | +0.006 |
falls | +0.005 |
Neg logit contributions | |
Baby | -0.015 |
dessert | -0.014 |
onda | -0.014 |
homeless | -0.014 |
essa | -0.013 |
Token | so |
Token ID | 523 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
bs | +0.013 |
gers | +0.012 |
counts | +0.012 |
hits | +0.011 |
hops | +0.011 |
Neg logit contributions | |
Baby | -0.022 |
homeless | -0.020 |
dessert | -0.020 |
Son | -0.019 |
rice | -0.019 |
Token | Tom |
Token ID | 13787 |
Feature activation | -8.03e-01 |
Pos logit contributions | |
Baby | +0.016 |
onda | +0.014 |
homeless | +0.014 |
dessert | +0.014 |
essa | +0.014 |
Neg logit contributions | |
bs | -0.012 |
gers | -0.012 |
counts | -0.012 |
falls | -0.012 |
hits | -0.011 |
Token | smiled |
Token ID | 13541 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
Baby | +0.060 |
onda | +0.054 |
dessert | +0.052 |
homeless | +0.052 |
Son | +0.051 |
Neg logit contributions | |
bs | -0.025 |
gers | -0.024 |
counts | -0.023 |
falls | -0.023 |
hits | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
Baby | +0.004 |
onda | +0.003 |
dessert | +0.003 |
homeless | +0.003 |
Son | +0.003 |
Neg logit contributions | |
bs | -0.002 |
counts | -0.002 |
gers | -0.002 |
hits | -0.002 |
falls | -0.002 |
Token | He |
Token ID | 679 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
Baby | +0.046 |
Son | +0.041 |
dessert | +0.040 |
Sonny | +0.039 |
Baby | +0.039 |
Neg logit contributions | |
bs | -0.022 |
gers | -0.020 |
hops | -0.019 |
counts | -0.019 |
tail | -0.018 |
Token | liked |
Token ID | 8288 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
Baby | +0.022 |
onda | +0.019 |
Son | +0.019 |
homeless | +0.019 |
essa | +0.018 |
Neg logit contributions | |
bs | -0.013 |
gers | -0.012 |
hits | -0.012 |
chairs | -0.012 |
falls | -0.012 |
Token | cookies |
Token ID | 14746 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
around | +0.082 |
counts | +0.079 |
hits | +0.078 |
low | +0.078 |
until | +0.075 |
Neg logit contributions | |
Baby | -0.150 |
onda | -0.143 |
essa | -0.133 |
uf | -0.127 |
urg | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
Baby | +0.120 |
onda | +0.109 |
homeless | +0.107 |
dessert | +0.106 |
essa | +0.104 |
Neg logit contributions | |
bs | -0.038 |
counts | -0.035 |
gers | -0.034 |
falls | -0.033 |
around | -0.033 |
Token | So |
Token ID | 1406 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Baby | +0.039 |
Son | +0.035 |
dessert | +0.033 |
Baby | +0.033 |
Sonny | +0.032 |
Neg logit contributions | |
bs | -0.022 |
gers | -0.021 |
hops | -0.020 |
tail | -0.020 |
hits | -0.019 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
Baby | +0.070 |
dessert | +0.065 |
Son | +0.063 |
Baby | +0.060 |
Sonny | +0.059 |
Neg logit contributions | |
bs | -0.049 |
hops | -0.047 |
counts | -0.046 |
webs | -0.046 |
gers | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
bs | +0.005 |
gers | +0.005 |
counts | +0.005 |
falls | +0.005 |
hits | +0.005 |
Neg logit contributions | |
Baby | -0.012 |
onda | -0.010 |
homeless | -0.010 |
dessert | -0.010 |
Son | -0.010 |
Token | his |
Token ID | 465 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
Baby | +0.032 |
dessert | +0.029 |
homeless | +0.028 |
onda | +0.028 |
Son | +0.027 |
Neg logit contributions | |
bs | -0.025 |
gers | -0.024 |
counts | -0.024 |
hops | -0.023 |
falls | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
Baby | +0.030 |
dessert | +0.028 |
homeless | +0.028 |
onda | +0.027 |
rice | +0.026 |
Neg logit contributions | |
bs | -0.019 |
gers | -0.018 |
hops | -0.017 |
counts | -0.017 |
hits | -0.017 |
Token | Mom |
Token ID | 11254 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
gers | +0.010 |
bs | +0.009 |
counts | +0.009 |
hits | +0.009 |
around | +0.009 |
Neg logit contributions | |
Baby | -0.017 |
Son | -0.015 |
Sonny | -0.014 |
dessert | -0.014 |
onda | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
Baby | +0.060 |
onda | +0.055 |
dessert | +0.052 |
homeless | +0.052 |
essa | +0.051 |
Neg logit contributions | |
gers | -0.033 |
bs | -0.033 |
hops | -0.032 |
low | -0.031 |
falls | -0.030 |
Token | Dad |
Token ID | 17415 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
bs | +0.011 |
counts | +0.010 |
hits | +0.010 |
falls | +0.010 |
around | +0.010 |
Neg logit contributions | |
Baby | -0.009 |
onda | -0.008 |
homeless | -0.007 |
dessert | -0.007 |
essa | -0.007 |
Token | gave |
Token ID | 2921 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
Baby | +0.246 |
onda | +0.224 |
homeless | +0.219 |
dessert | +0.218 |
essa | +0.214 |
Neg logit contributions | |
bs | -0.095 |
gers | -0.090 |
counts | -0.089 |
falls | -0.086 |
hits | -0.086 |
Token | him |
Token ID | 683 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
Baby | +0.026 |
onda | +0.022 |
dessert | +0.022 |
homeless | +0.021 |
essa | +0.021 |
Neg logit contributions | |
bs | -0.026 |
counts | -0.025 |
gers | -0.025 |
falls | -0.025 |
hits | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
Baby | +0.026 |
onda | +0.023 |
homeless | +0.022 |
dessert | +0.022 |
essa | +0.022 |
Neg logit contributions | |
bs | -0.022 |
counts | -0.021 |
gers | -0.021 |
falls | -0.021 |
hits | -0.021 |
Token | hug |
Token ID | 16225 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
Baby | +0.027 |
onda | +0.024 |
homeless | +0.023 |
dessert | +0.023 |
essa | +0.023 |
Neg logit contributions | |
bs | -0.015 |
counts | -0.015 |
gers | -0.015 |
falls | -0.014 |
hits | -0.014 |
Token | too |
Token ID | 1165 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
bs | +0.070 |
hops | +0.068 |
hits | +0.066 |
counts | +0.065 |
spins | +0.064 |
Neg logit contributions | |
Baby | -0.119 |
dessert | -0.109 |
homeless | -0.107 |
ful | -0.106 |
onda | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
bs | +0.012 |
gers | +0.011 |
hits | +0.011 |
hops | +0.011 |
counts | +0.011 |
Neg logit contributions | |
Baby | -0.026 |
onda | -0.023 |
dessert | -0.023 |
homeless | -0.023 |
essa | -0.022 |
Token | They |
Token ID | 1119 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
gers | +0.010 |
bs | +0.010 |
counts | +0.010 |
around | +0.010 |
chair | +0.010 |
Neg logit contributions | |
Baby | -0.014 |
Son | -0.012 |
dessert | -0.012 |
rice | -0.012 |
homeless | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
dessert | +0.095 |
Baby | +0.093 |
homeless | +0.089 |
recipe | +0.086 |
rice | +0.085 |
Neg logit contributions | |
hops | -0.072 |
bs | -0.071 |
gers | -0.070 |
hits | -0.068 |
counts | -0.067 |
Token |
|
Token ID | 198 |
Feature activation | +7.56e-01 |
Pos logit contributions | |
bs | +0.024 |
gers | +0.024 |
counts | +0.023 |
hits | +0.023 |
hops | +0.023 |
Neg logit contributions | |
Baby | -0.038 |
dessert | -0.032 |
homeless | -0.032 |
Son | -0.031 |
onda | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
bs | +0.029 |
gers | +0.028 |
counts | +0.028 |
hits | +0.027 |
hops | +0.027 |
Neg logit contributions | |
Baby | -0.051 |
onda | -0.045 |
homeless | -0.044 |
dessert | -0.043 |
essa | -0.042 |
Token | She |
Token ID | 3347 |
Feature activation | +7.81e-01 |
Pos logit contributions | |
bs | +0.031 |
counts | +0.031 |
hits | +0.030 |
hops | +0.030 |
gers | +0.030 |
Neg logit contributions | |
Baby | -0.041 |
onda | -0.034 |
homeless | -0.033 |
dessert | -0.032 |
essa | -0.031 |
Token | sees |
Token ID | 7224 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
bs | +0.029 |
gers | +0.028 |
low | +0.027 |
tail | +0.026 |
sideways | +0.025 |
Neg logit contributions | |
Baby | -0.047 |
wasting | -0.044 |
onda | -0.044 |
dessert | -0.044 |
rice | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
bs | +0.130 |
counts | +0.123 |
gers | +0.123 |
falls | +0.121 |
hits | +0.120 |
Neg logit contributions | |
Baby | -0.180 |
onda | -0.159 |
homeless | -0.155 |
dessert | -0.154 |
essa | -0.150 |
Token | tower |
Token ID | 10580 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
bs | +0.042 |
counts | +0.041 |
falls | +0.041 |
hits | +0.041 |
low | +0.040 |
Neg logit contributions | |
Baby | -0.072 |
onda | -0.065 |
homeless | -0.061 |
essa | -0.061 |
uf | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
bs | +0.047 |
gers | +0.044 |
chairs | +0.039 |
tail | +0.039 |
hops | +0.039 |
Neg logit contributions | |
dessert | -0.059 |
Baby | -0.059 |
pudding | -0.057 |
wasting | -0.056 |
cooking | -0.056 |
Token | It |
Token ID | 632 |
Feature activation | +8.01e-02 |
Pos logit contributions | |
bs | +0.013 |
gers | +0.012 |
counts | +0.012 |
falls | +0.012 |
hits | +0.012 |
Neg logit contributions | |
Baby | -0.030 |
onda | -0.027 |
dessert | -0.026 |
homeless | -0.026 |
Son | -0.026 |
Token | is |
Token ID | 318 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
bs | +0.004 |
gers | +0.004 |
belt | +0.004 |
tail | +0.004 |
ways | +0.004 |
Neg logit contributions | |
pudding | -0.004 |
wasting | -0.003 |
truth | -0.003 |
dessert | -0.003 |
caul | -0.003 |
Token | very |
Token ID | 845 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
bs | +0.024 |
gers | +0.022 |
counts | +0.022 |
hops | +0.021 |
falls | +0.021 |
Neg logit contributions | |
Baby | -0.049 |
onda | -0.044 |
dessert | -0.044 |
homeless | -0.043 |
Son | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e-01 |
Pos logit contributions | |
bs | +0.044 |
low | +0.040 |
hits | +0.039 |
gers | +0.039 |
hops | +0.039 |
Neg logit contributions | |
Baby | -0.084 |
dessert | -0.080 |
onda | -0.077 |
homeless | -0.075 |
pudding | -0.075 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Baby | +0.037 |
Baby | +0.032 |
dessert | +0.031 |
homeless | +0.031 |
Son | +0.031 |
Neg logit contributions | |
gers | -0.026 |
counts | -0.025 |
hops | -0.025 |
hits | -0.024 |
bs | -0.024 |
Token | realized |
Token ID | 6939 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
Baby | +0.009 |
onda | +0.008 |
homeless | +0.008 |
dessert | +0.008 |
essa | +0.008 |
Neg logit contributions | |
bs | -0.005 |
gers | -0.005 |
counts | -0.004 |
low | -0.004 |
hits | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
Baby | +0.002 |
onda | +0.002 |
homeless | +0.002 |
dessert | +0.002 |
essa | +0.002 |
Neg logit contributions | |
bs | -0.001 |
counts | -0.001 |
gers | -0.001 |
falls | -0.001 |
hits | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
bs | +0.011 |
gers | +0.011 |
counts | +0.010 |
low | +0.010 |
hops | +0.010 |
Neg logit contributions | |
Baby | -0.025 |
onda | -0.023 |
homeless | -0.022 |
dessert | -0.022 |
essa | -0.021 |
Token | okay |
Token ID | 8788 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
around | +0.009 |
low | +0.009 |
counts | +0.009 |
bs | +0.009 |
hits | +0.009 |
Neg logit contributions | |
Baby | -0.015 |
onda | -0.014 |
essa | -0.013 |
uf | -0.013 |
urg | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
Baby | +0.172 |
onda | +0.156 |
homeless | +0.152 |
dessert | +0.150 |
Son | +0.148 |
Neg logit contributions | |
bs | -0.073 |
counts | -0.068 |
gers | -0.067 |
falls | -0.066 |
around | -0.066 |
Token | be |
Token ID | 307 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
Baby | +0.040 |
onda | +0.036 |
homeless | +0.034 |
dessert | +0.034 |
essa | +0.034 |
Neg logit contributions | |
bs | -0.028 |
low | -0.027 |
counts | -0.026 |
around | -0.026 |
hits | -0.026 |
Token | silly |
Token ID | 14397 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
Baby | +0.080 |
homeless | +0.071 |
dessert | +0.071 |
onda | +0.071 |
rice | +0.066 |
Neg logit contributions | |
bs | -0.052 |
gers | -0.052 |
counts | -0.051 |
falls | -0.050 |
hops | -0.050 |
Token | sometimes |
Token ID | 3360 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
bs | +0.014 |
gers | +0.013 |
low | +0.013 |
counts | +0.013 |
falls | +0.013 |
Neg logit contributions | |
Baby | -0.031 |
onda | -0.029 |
dessert | -0.028 |
homeless | -0.028 |
essa | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
bs | +0.005 |
gers | +0.005 |
counts | +0.005 |
hits | +0.004 |
falls | +0.004 |
Neg logit contributions | |
Baby | -0.015 |
onda | -0.013 |
dessert | -0.013 |
homeless | -0.013 |
rice | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
bs | +0.035 |
counts | +0.032 |
gers | +0.032 |
falls | +0.032 |
around | +0.032 |
Neg logit contributions | |
Baby | -0.101 |
onda | -0.092 |
homeless | -0.089 |
dessert | -0.089 |
essa | -0.087 |
Token | like |
Token ID | 588 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
bs | +0.003 |
gers | +0.003 |
counts | +0.003 |
hits | +0.003 |
falls | +0.003 |
Neg logit contributions | |
Baby | -0.005 |
homeless | -0.005 |
dessert | -0.005 |
Son | -0.005 |
onda | -0.005 |
Token | he |
Token ID | 339 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
bs | +0.100 |
counts | +0.095 |
gers | +0.094 |
around | +0.093 |
falls | +0.093 |
Neg logit contributions | |
Baby | -0.149 |
onda | -0.133 |
homeless | -0.128 |
dessert | -0.127 |
essa | -0.125 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
bs | +0.066 |
gers | +0.065 |
counts | +0.059 |
chair | +0.059 |
falls | +0.058 |
Neg logit contributions | |
Baby | -0.107 |
onda | -0.098 |
homeless | -0.097 |
dessert | -0.092 |
uf | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
bs | +0.009 |
gers | +0.009 |
counts | +0.009 |
falls | +0.009 |
hits | +0.009 |
Neg logit contributions | |
Baby | -0.011 |
dessert | -0.010 |
onda | -0.010 |
homeless | -0.010 |
essa | -0.009 |
Token | apologize |
Token ID | 16521 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
Baby | +0.029 |
onda | +0.026 |
homeless | +0.025 |
dessert | +0.025 |
essa | +0.025 |
Neg logit contributions | |
bs | -0.020 |
counts | -0.019 |
falls | -0.018 |
hits | -0.018 |
gers | -0.018 |
Token | but |
Token ID | 475 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
Baby | +0.153 |
dessert | +0.146 |
homeless | +0.140 |
essa | +0.140 |
onda | +0.139 |
Neg logit contributions | |
bs | -0.100 |
gers | -0.094 |
counts | -0.093 |
hits | -0.092 |
hops | -0.092 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
bs | +0.070 |
gers | +0.067 |
counts | +0.066 |
hits | +0.064 |
falls | +0.063 |
Neg logit contributions | |
Baby | -0.152 |
homeless | -0.136 |
dessert | -0.136 |
onda | -0.135 |
Son | -0.130 |
Token | 't |
Token ID | 470 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
Baby | +0.001 |
onda | +0.001 |
essa | +0.001 |
homeless | +0.001 |
uggets | +0.001 |
Neg logit contributions | |
falls | -0.001 |
bs | -0.001 |
hits | -0.001 |
around | -0.001 |
chair | -0.001 |
Token | know |
Token ID | 760 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
Baby | +0.031 |
onda | +0.028 |
homeless | +0.027 |
essa | +0.027 |
Son | +0.027 |
Neg logit contributions | |
bs | -0.018 |
around | -0.017 |
counts | -0.017 |
low | -0.016 |
falls | -0.016 |
Token | how |
Token ID | 703 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
Baby | +0.066 |
dessert | +0.058 |
onda | +0.057 |
homeless | +0.055 |
rice | +0.053 |
Neg logit contributions | |
bs | -0.080 |
gers | -0.079 |
hits | -0.076 |
counts | -0.076 |
hops | -0.076 |
Token | still |
Token ID | 991 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
Baby | +0.053 |
homeless | +0.048 |
dessert | +0.047 |
onda | +0.047 |
rice | +0.046 |
Neg logit contributions | |
bs | -0.038 |
chair | -0.038 |
counts | -0.037 |
falls | -0.037 |
hits | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
bs | +0.014 |
counts | +0.013 |
gers | +0.013 |
hits | +0.013 |
falls | +0.012 |
Neg logit contributions | |
Baby | -0.034 |
onda | -0.031 |
homeless | -0.030 |
dessert | -0.030 |
essa | -0.029 |
Token | see |
Token ID | 766 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
Baby | +0.023 |
dessert | +0.020 |
homeless | +0.020 |
onda | +0.019 |
rice | +0.019 |
Neg logit contributions | |
gers | -0.019 |
bs | -0.019 |
counts | -0.018 |
falls | -0.017 |
chairs | -0.017 |
Token | what |
Token ID | 644 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
bs | +0.151 |
gers | +0.147 |
falls | +0.144 |
counts | +0.144 |
hits | +0.143 |
Neg logit contributions | |
Baby | -0.160 |
dessert | -0.139 |
onda | -0.138 |
homeless | -0.137 |
Son | -0.133 |
Token | would |
Token ID | 561 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
Baby | +0.096 |
dessert | +0.088 |
onda | +0.088 |
homeless | +0.087 |
rice | +0.085 |
Neg logit contributions | |
bs | -0.041 |
gers | -0.036 |
falls | -0.036 |
counts | -0.036 |
hits | -0.036 |
Token | happen |
Token ID | 1645 |
Feature activation | -4.75e+00 |
Pos logit contributions | |
around | +0.145 |
counts | +0.144 |
low | +0.141 |
hits | +0.139 |
hops | +0.138 |
Neg logit contributions | |
Baby | -0.134 |
onda | -0.114 |
uf | -0.109 |
Son | -0.105 |
essa | -0.105 |
Token | . |
Token ID | 13 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
Baby | +0.294 |
onda | +0.259 |
dessert | +0.257 |
homeless | +0.254 |
essa | +0.250 |
Neg logit contributions | |
bs | -0.207 |
hits | -0.194 |
falls | -0.194 |
counts | -0.192 |
hops | -0.190 |
Token | He |
Token ID | 679 |
Feature activation | -7.30e-02 |
Pos logit contributions | |
Baby | +0.029 |
Son | +0.026 |
obby | +0.025 |
Sonny | +0.025 |
onda | +0.024 |
Neg logit contributions | |
bs | -0.019 |
gers | -0.019 |
hops | -0.018 |
hits | -0.018 |
falls | -0.018 |
Token | stood |
Token ID | 6204 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
Baby | +0.005 |
onda | +0.004 |
homeless | +0.004 |
Son | +0.004 |
Sonny | +0.004 |
Neg logit contributions | |
bs | -0.003 |
gers | -0.003 |
hits | -0.003 |
falls | -0.003 |
counts | -0.003 |
Token | in |
Token ID | 287 |
Feature activation | -9.86e-01 |
Pos logit contributions | |
Baby | +0.071 |
homeless | +0.063 |
dessert | +0.063 |
Son | +0.063 |
Sonny | +0.062 |
Neg logit contributions | |
bs | -0.063 |
chair | -0.059 |
counts | -0.059 |
falls | -0.057 |
chairs | -0.056 |
Token | his |
Token ID | 465 |
Feature activation | +4.42e-03 |
Pos logit contributions | |
Baby | +0.060 |
dessert | +0.054 |
homeless | +0.052 |
Son | +0.051 |
onda | +0.050 |
Neg logit contributions | |
bs | -0.047 |
gers | -0.043 |
falls | -0.043 |
counts | -0.042 |
chair | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
bs | +0.119 |
counts | +0.113 |
gers | +0.113 |
falls | +0.111 |
around | +0.110 |
Neg logit contributions | |
Baby | -0.197 |
onda | -0.177 |
homeless | -0.171 |
dessert | -0.170 |
essa | -0.167 |
Token | chairs |
Token ID | 18791 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
bs | +0.046 |
counts | +0.043 |
gers | +0.043 |
hops | +0.042 |
hits | +0.042 |
Neg logit contributions | |
Baby | -0.036 |
dessert | -0.031 |
homeless | -0.030 |
onda | -0.029 |
rice | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
hops | +0.050 |
bs | +0.050 |
gers | +0.048 |
spins | +0.047 |
hits | +0.047 |
Neg logit contributions | |
Baby | -0.076 |
dessert | -0.070 |
rice | -0.068 |
homeless | -0.068 |
ful | -0.068 |
Token | pretended |
Token ID | 36766 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
bs | +0.008 |
gers | +0.008 |
hops | +0.007 |
counts | +0.007 |
hits | +0.007 |
Neg logit contributions | |
Baby | -0.011 |
dessert | -0.010 |
rice | -0.010 |
homeless | -0.010 |
onda | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
bs | +0.067 |
counts | +0.063 |
gers | +0.063 |
falls | +0.062 |
hits | +0.061 |
Neg logit contributions | |
Baby | -0.119 |
onda | -0.107 |
dessert | -0.105 |
homeless | -0.104 |
essa | -0.101 |
Token | pour |
Token ID | 12797 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
Baby | +0.016 |
onda | +0.014 |
homeless | +0.013 |
essa | +0.013 |
Son | +0.013 |
Neg logit contributions | |
bs | -0.012 |
counts | -0.011 |
low | -0.011 |
falls | -0.011 |
gers | -0.011 |
Token | them |
Token ID | 606 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
Baby | +0.158 |
onda | +0.138 |
dessert | +0.135 |
homeless | +0.134 |
essa | +0.130 |
Neg logit contributions | |
bs | -0.126 |
counts | -0.120 |
gers | -0.120 |
falls | -0.118 |
hits | -0.117 |
Token | tea |
Token ID | 8887 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
bs | +0.017 |
gers | +0.016 |
counts | +0.016 |
hops | +0.016 |
hits | +0.016 |
Neg logit contributions | |
Baby | -0.031 |
dessert | -0.029 |
onda | -0.028 |
homeless | -0.028 |
rice | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
bs | +0.009 |
counts | +0.008 |
gers | +0.008 |
hops | +0.008 |
hits | +0.008 |
Neg logit contributions | |
Baby | -0.021 |
onda | -0.019 |
dessert | -0.018 |
homeless | -0.018 |
essa | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
bs | +0.021 |
gers | +0.020 |
counts | +0.020 |
hits | +0.020 |
falls | +0.020 |
Neg logit contributions | |
Baby | -0.031 |
dessert | -0.027 |
Son | -0.027 |
homeless | -0.027 |
onda | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +6.11e-01 |
Pos logit contributions | |
bs | +0.023 |
counts | +0.021 |
gers | +0.021 |
falls | +0.021 |
around | +0.021 |
Neg logit contributions | |
Baby | -0.048 |
onda | -0.044 |
homeless | -0.043 |
dessert | -0.043 |
essa | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
Baby | +0.075 |
dessert | +0.070 |
onda | +0.068 |
rice | +0.068 |
homeless | +0.068 |
Neg logit contributions | |
bs | -0.066 |
chair | -0.063 |
hits | -0.062 |
counts | -0.061 |
falls | -0.061 |
Token | bridge |
Token ID | 7696 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
Baby | +0.051 |
onda | +0.045 |
homeless | +0.045 |
dessert | +0.045 |
Son | +0.043 |
Neg logit contributions | |
bs | -0.032 |
counts | -0.030 |
gers | -0.029 |
falls | -0.029 |
hits | -0.029 |
Token | � |
Token ID | 6184 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
bs | +0.102 |
hops | +0.093 |
gers | +0.092 |
counts | +0.091 |
falls | +0.090 |
Neg logit contributions | |
Baby | -0.160 |
dessert | -0.148 |
homeless | -0.147 |
onda | -0.146 |
rice | -0.141 |
Token | � |
Token ID | 95 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
Baby | +0.011 |
onda | +0.010 |
homeless | +0.009 |
dessert | +0.009 |
essa | +0.009 |
Neg logit contributions | |
bs | -0.008 |
gers | -0.008 |
counts | -0.008 |
falls | -0.008 |
hits | -0.008 |
Token | € |
Token ID | 26391 |
Feature activation | -9.67e-01 |
Pos logit contributions | |
Baby | +0.126 |
onda | +0.111 |
homeless | +0.108 |
dessert | +0.108 |
essa | +0.104 |
Neg logit contributions | |
bs | -0.104 |
counts | -0.099 |
gers | -0.099 |
falls | -0.097 |
hits | -0.097 |
Token | � |
Token ID | 447 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
Baby | +0.059 |
onda | +0.054 |
homeless | +0.050 |
essa | +0.049 |
uf | +0.049 |
Neg logit contributions | |
bs | -0.040 |
hops | -0.040 |
gers | -0.040 |
counts | -0.039 |
hits | -0.039 |
Token | � |
Token ID | 250 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
Baby | +0.135 |
onda | +0.118 |
homeless | +0.115 |
dessert | +0.114 |
Son | +0.112 |
Neg logit contributions | |
bs | -0.094 |
counts | -0.090 |
gers | -0.089 |
around | -0.089 |
hops | -0.087 |
Token | stones |
Token ID | 14966 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
Baby | +0.101 |
onda | +0.088 |
homeless | +0.085 |
dessert | +0.085 |
essa | +0.082 |
Neg logit contributions | |
bs | -0.101 |
counts | -0.097 |
gers | -0.097 |
falls | -0.095 |
around | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
bs | +0.059 |
counts | +0.056 |
gers | +0.056 |
hits | +0.054 |
falls | +0.054 |
Neg logit contributions | |
Baby | -0.099 |
onda | -0.089 |
homeless | -0.086 |
dessert | -0.086 |
essa | -0.084 |
Token | tw |
Token ID | 665 |
Feature activation | +7.07e-01 |
Pos logit contributions | |
Baby | +0.002 |
onda | +0.002 |
homeless | +0.002 |
dessert | +0.002 |
Son | +0.002 |
Neg logit contributions | |
bs | -0.001 |
gers | -0.001 |
counts | -0.001 |
hits | -0.001 |
falls | -0.001 |
Token | igs |
Token ID | 9235 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
chair | +0.031 |
counts | +0.030 |
bs | +0.028 |
low | +0.028 |
chairs | +0.027 |
Neg logit contributions | |
Baby | -0.044 |
urg | -0.040 |
onda | -0.039 |
essa | -0.036 |
uf | -0.036 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Baby | +0.294 |
onda | +0.273 |
homeless | +0.255 |
essa | +0.254 |
uf | +0.253 |
Neg logit contributions | |
around | -0.100 |
bs | -0.099 |
likes | -0.092 |
counts | -0.091 |
gers | -0.090 |
Token | |
Token ID | 220 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
Baby | +0.053 |
onda | +0.049 |
homeless | +0.047 |
essa | +0.046 |
dessert | +0.044 |
Neg logit contributions | |
bs | -0.069 |
around | -0.068 |
counts | -0.065 |
But | -0.065 |
chair | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
bs | +0.068 |
around | +0.066 |
falls | +0.059 |
But | +0.059 |
counts | +0.059 |
Neg logit contributions | |
Baby | -0.216 |
onda | -0.203 |
homeless | -0.202 |
dessert | -0.200 |
essa | -0.197 |
Token |
|
Token ID | 198 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
around | +0.007 |
bs | +0.006 |
low | +0.006 |
But | +0.006 |
likes | +0.006 |
Neg logit contributions | |
Baby | -0.020 |
essa | -0.020 |
homeless | -0.020 |
onda | -0.020 |
dessert | -0.019 |
Token | B |
Token ID | 33 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
around | +0.006 |
until | +0.006 |
low | +0.006 |
liked | +0.006 |
likes | +0.005 |
Neg logit contributions | |
essa | -0.013 |
urg | -0.012 |
uf | -0.012 |
onda | -0.012 |
dessert | -0.012 |
Token | obby |
Token ID | 11369 |
Feature activation | -8.27e+00 |
Pos logit contributions | |
rice | +0.074 |
dessert | +0.071 |
homeless | +0.070 |
Baby | +0.064 |
wasting | +0.061 |
Neg logit contributions | |
around | -0.159 |
gers | -0.156 |
bs | -0.156 |
likes | -0.154 |
chair | -0.153 |
Token | was |
Token ID | 373 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
Baby | +0.649 |
onda | +0.593 |
dessert | +0.582 |
homeless | +0.581 |
essa | +0.568 |
Neg logit contributions | |
bs | -0.220 |
counts | -0.211 |
around | -0.207 |
likes | -0.207 |
falls | -0.199 |
Token | so |
Token ID | 523 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
bs | +0.013 |
around | +0.013 |
counts | +0.013 |
low | +0.013 |
gers | +0.012 |
Neg logit contributions | |
Baby | -0.028 |
onda | -0.025 |
homeless | -0.024 |
dessert | -0.024 |
essa | -0.024 |
Token | excited |
Token ID | 6568 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
Baby | +0.087 |
onda | +0.078 |
homeless | +0.078 |
dessert | +0.077 |
Son | +0.075 |
Neg logit contributions | |
bs | -0.038 |
gers | -0.036 |
counts | -0.036 |
hits | -0.035 |
falls | -0.034 |
Token | that |
Token ID | 326 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
Baby | +0.035 |
onda | +0.032 |
homeless | +0.031 |
Sonny | +0.030 |
uf | +0.030 |
Neg logit contributions | |
around | -0.011 |
bs | -0.010 |
gers | -0.010 |
counts | -0.010 |
chair | -0.010 |
Token | he |
Token ID | 339 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Baby | +0.152 |
dessert | +0.138 |
homeless | +0.138 |
onda | +0.137 |
Son | +0.135 |
Neg logit contributions | |
bs | -0.043 |
gers | -0.038 |
counts | -0.038 |
falls | -0.037 |
hits | -0.036 |
Token | who |
Token ID | 508 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
Baby | +0.046 |
onda | +0.041 |
homeless | +0.040 |
dessert | +0.040 |
Son | +0.039 |
Neg logit contributions | |
gers | -0.029 |
bs | -0.028 |
hits | -0.028 |
hops | -0.027 |
counts | -0.027 |
Token | took |
Token ID | 1718 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
Baby | +0.043 |
onda | +0.039 |
homeless | +0.039 |
dessert | +0.038 |
essa | +0.037 |
Neg logit contributions | |
bs | -0.020 |
gers | -0.020 |
counts | -0.019 |
hops | -0.019 |
hits | -0.018 |
Token | an |
Token ID | 281 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
Baby | +0.112 |
onda | +0.106 |
homeless | +0.100 |
essa | +0.098 |
uf | +0.096 |
Neg logit contributions | |
bs | -0.067 |
around | -0.062 |
falls | -0.060 |
counts | -0.059 |
hits | -0.058 |
Token | x |
Token ID | 2124 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
bs | +0.003 |
around | +0.003 |
chair | +0.003 |
hits | +0.003 |
counts | +0.003 |
Neg logit contributions | |
Baby | -0.003 |
onda | -0.003 |
Sonny | -0.002 |
Son | -0.002 |
essa | -0.002 |
Token | - |
Token ID | 12 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
Baby | +0.106 |
dessert | +0.101 |
onda | +0.101 |
ful | +0.100 |
essa | +0.096 |
Neg logit contributions | |
low | -0.080 |
counts | -0.080 |
likes | -0.080 |
sideways | -0.079 |
bs | -0.079 |
Token | ray |
Token ID | 2433 |
Feature activation | -7.40e+00 |
Pos logit contributions | |
bs | +0.207 |
counts | +0.206 |
gers | +0.205 |
hops | +0.205 |
falls | +0.202 |
Neg logit contributions | |
Baby | -0.146 |
onda | -0.120 |
homeless | -0.115 |
dessert | -0.114 |
essa | -0.111 |
Token | of |
Token ID | 286 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
Baby | +0.474 |
dessert | +0.428 |
onda | +0.423 |
homeless | +0.422 |
rice | +0.418 |
Neg logit contributions | |
gers | -0.277 |
bs | -0.273 |
hops | -0.270 |
counts | -0.269 |
likes | -0.257 |
Token | the |
Token ID | 262 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
Baby | +0.105 |
onda | +0.099 |
essa | +0.092 |
homeless | +0.092 |
uf | +0.090 |
Neg logit contributions | |
bs | -0.048 |
low | -0.045 |
hits | -0.045 |
falls | -0.045 |
around | -0.045 |
Token | wing |
Token ID | 8539 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
hits | +0.020 |
chair | +0.020 |
low | +0.020 |
falls | +0.020 |
bs | +0.020 |
Neg logit contributions | |
Baby | -0.030 |
onda | -0.027 |
urg | -0.026 |
essa | -0.025 |
Sonny | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
Baby | +0.079 |
ful | +0.072 |
homeless | +0.072 |
onda | +0.071 |
dessert | +0.071 |
Neg logit contributions | |
gers | -0.046 |
counts | -0.045 |
hops | -0.044 |
bs | -0.044 |
hits | -0.042 |
Token | The |
Token ID | 383 |
Feature activation | +7.67e-01 |
Pos logit contributions | |
Baby | +0.035 |
onda | +0.034 |
dessert | +0.032 |
essa | +0.032 |
urg | +0.032 |
Neg logit contributions | |
bs | -0.007 |
But | -0.007 |
low | -0.006 |
chair | -0.006 |
around | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
Baby | +0.087 |
onda | +0.078 |
homeless | +0.076 |
dessert | +0.076 |
essa | +0.074 |
Neg logit contributions | |
bs | -0.050 |
counts | -0.048 |
gers | -0.047 |
falls | -0.047 |
around | -0.046 |
Token | world |
Token ID | 995 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
Baby | +0.003 |
onda | +0.002 |
homeless | +0.002 |
dessert | +0.002 |
Son | +0.002 |
Neg logit contributions | |
bs | -0.002 |
falls | -0.002 |
gers | -0.002 |
counts | -0.002 |
hops | -0.002 |
Token | ! |
Token ID | 0 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
bs | +0.020 |
hits | +0.019 |
hops | +0.018 |
counts | +0.018 |
gers | +0.018 |
Neg logit contributions | |
Baby | -0.035 |
dessert | -0.032 |
onda | -0.031 |
homeless | -0.031 |
essa | -0.030 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
Baby | +0.088 |
onda | +0.078 |
dessert | +0.076 |
homeless | +0.075 |
Son | +0.074 |
Neg logit contributions | |
bs | -0.050 |
gers | -0.048 |
counts | -0.047 |
falls | -0.047 |
hits | -0.047 |
Token | B |
Token ID | 33 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
Baby | +0.282 |
onda | +0.257 |
homeless | +0.253 |
dessert | +0.249 |
essa | +0.247 |
Neg logit contributions | |
bs | -0.117 |
counts | -0.114 |
around | -0.113 |
gers | -0.113 |
low | -0.109 |
Token | obby |
Token ID | 11369 |
Feature activation | -8.10e+00 |
Pos logit contributions | |
Baby | +0.149 |
dessert | +0.138 |
homeless | +0.137 |
onda | +0.137 |
rice | +0.136 |
Neg logit contributions | |
bs | -0.092 |
around | -0.083 |
gers | -0.082 |
low | -0.082 |
counts | -0.081 |
Token | was |
Token ID | 373 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
Baby | +0.606 |
onda | +0.545 |
dessert | +0.533 |
homeless | +0.532 |
Son | +0.525 |
Neg logit contributions | |
bs | -0.245 |
gers | -0.229 |
counts | -0.221 |
low | -0.218 |
falls | -0.216 |
Token | very |
Token ID | 845 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
bs | +0.018 |
counts | +0.017 |
gers | +0.017 |
falls | +0.016 |
around | +0.016 |
Neg logit contributions | |
Baby | -0.025 |
onda | -0.022 |
homeless | -0.022 |
dessert | -0.021 |
essa | -0.021 |
Token | excited |
Token ID | 6568 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
counts | +0.041 |
bs | +0.041 |
gers | +0.041 |
hits | +0.039 |
falls | +0.039 |
Neg logit contributions | |
Baby | -0.077 |
dessert | -0.068 |
homeless | -0.068 |
Son | -0.066 |
onda | -0.065 |
Token | to |
Token ID | 284 |
Feature activation | -7.27e-01 |
Pos logit contributions | |
Baby | +0.040 |
onda | +0.037 |
homeless | +0.035 |
uf | +0.035 |
Sonny | +0.034 |
Neg logit contributions | |
around | -0.011 |
chair | -0.008 |
until | -0.008 |
gers | -0.008 |
counts | -0.008 |
Token | go |
Token ID | 467 |
Feature activation | +9.43e-02 |
Pos logit contributions | |
Baby | +0.049 |
onda | +0.046 |
homeless | +0.044 |
uf | +0.042 |
essa | +0.042 |
Neg logit contributions | |
bs | -0.027 |
counts | -0.027 |
around | -0.026 |
hops | -0.026 |
low | -0.026 |
Token | The |
Token ID | 383 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
Baby | +0.069 |
onda | +0.059 |
Son | +0.059 |
dessert | +0.059 |
homeless | +0.058 |
Neg logit contributions | |
bs | -0.035 |
gers | -0.034 |
hits | -0.034 |
falls | -0.033 |
counts | -0.032 |
Token | end |
Token ID | 886 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
bs | +0.030 |
gers | +0.029 |
counts | +0.029 |
falls | +0.028 |
around | +0.028 |
Neg logit contributions | |
Baby | -0.035 |
onda | -0.030 |
homeless | -0.029 |
dessert | -0.029 |
essa | -0.028 |
Token | ! |
Token ID | 0 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
Baby | +0.075 |
dessert | +0.068 |
ful | +0.066 |
truth | +0.065 |
onda | +0.065 |
Neg logit contributions | |
bs | -0.068 |
gers | -0.066 |
hops | -0.064 |
hits | -0.063 |
sideways | -0.061 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
Baby | +0.101 |
onda | +0.094 |
dessert | +0.092 |
homeless | +0.092 |
urg | +0.091 |
Neg logit contributions | |
bs | -0.039 |
counts | -0.035 |
around | -0.034 |
chair | -0.034 |
low | -0.034 |
Token | B |
Token ID | 33 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Baby | +0.280 |
onda | +0.256 |
homeless | +0.252 |
dessert | +0.248 |
essa | +0.246 |
Neg logit contributions | |
bs | -0.116 |
counts | -0.113 |
around | -0.112 |
gers | -0.111 |
falls | -0.109 |
Token | obby |
Token ID | 11369 |
Feature activation | -7.95e+00 |
Pos logit contributions | |
Baby | +0.115 |
dessert | +0.109 |
rice | +0.108 |
homeless | +0.107 |
onda | +0.105 |
Neg logit contributions | |
bs | -0.100 |
around | -0.091 |
gers | -0.090 |
chair | -0.089 |
low | -0.089 |
Token | was |
Token ID | 373 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
Baby | +0.600 |
onda | +0.542 |
homeless | +0.529 |
dessert | +0.529 |
Son | +0.518 |
Neg logit contributions | |
bs | -0.238 |
gers | -0.221 |
counts | -0.218 |
falls | -0.212 |
hits | -0.211 |
Token | a |
Token ID | 257 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
bs | +0.028 |
around | +0.026 |
counts | +0.026 |
gers | +0.026 |
low | +0.026 |
Neg logit contributions | |
Baby | -0.043 |
onda | -0.039 |
homeless | -0.037 |
dessert | -0.037 |
essa | -0.036 |
Token | very |
Token ID | 845 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Baby | +0.025 |
onda | +0.022 |
homeless | +0.022 |
dessert | +0.021 |
essa | +0.021 |
Neg logit contributions | |
bs | -0.022 |
counts | -0.021 |
gers | -0.021 |
falls | -0.021 |
hits | -0.020 |
Token | lucky |
Token ID | 9670 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
counts | +0.050 |
chairs | +0.049 |
hits | +0.047 |
bs | +0.047 |
gers | +0.047 |
Neg logit contributions | |
Baby | -0.065 |
dessert | -0.057 |
homeless | -0.057 |
Son | -0.054 |
rice | -0.053 |
Token | boy |
Token ID | 2933 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
Baby | +0.080 |
onda | +0.076 |
urg | +0.071 |
homeless | +0.069 |
essa | +0.068 |
Neg logit contributions | |
around | -0.053 |
likes | -0.052 |
gers | -0.052 |
low | -0.051 |
bs | -0.051 |
Token | He |
Token ID | 679 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
Baby | +0.035 |
onda | +0.032 |
dessert | +0.031 |
homeless | +0.031 |
essa | +0.030 |
Neg logit contributions | |
bs | -0.011 |
counts | -0.010 |
gers | -0.010 |
around | -0.009 |
falls | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
Baby | +0.011 |
onda | +0.010 |
homeless | +0.009 |
dessert | +0.009 |
Son | +0.009 |
Neg logit contributions | |
bs | -0.006 |
gers | -0.005 |
counts | -0.005 |
falls | -0.005 |
hits | -0.005 |
Token | an |
Token ID | 281 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
Baby | +0.035 |
onda | +0.031 |
homeless | +0.029 |
dessert | +0.029 |
essa | +0.029 |
Neg logit contributions | |
bs | -0.026 |
counts | -0.026 |
hits | -0.025 |
around | -0.025 |
falls | -0.025 |
Token | x |
Token ID | 2124 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
bs | +0.006 |
gers | +0.006 |
falls | +0.005 |
counts | +0.005 |
hops | +0.005 |
Neg logit contributions | |
Baby | -0.010 |
dessert | -0.009 |
onda | -0.009 |
homeless | -0.009 |
rice | -0.009 |
Token | - |
Token ID | 12 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
dessert | +0.096 |
Baby | +0.094 |
onda | +0.093 |
urg | +0.091 |
essa | +0.089 |
Neg logit contributions | |
bs | -0.090 |
low | -0.085 |
counts | -0.084 |
sideways | -0.084 |
hops | -0.082 |
Token | ray |
Token ID | 2433 |
Feature activation | -7.41e+00 |
Pos logit contributions | |
hops | +0.227 |
gers | +0.224 |
bs | +0.224 |
counts | +0.223 |
falls | +0.223 |
Neg logit contributions | |
Baby | -0.132 |
onda | -0.104 |
essa | -0.098 |
homeless | -0.097 |
dessert | -0.097 |
Token | that |
Token ID | 326 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
Baby | +0.440 |
dessert | +0.403 |
homeless | +0.403 |
onda | +0.398 |
rice | +0.391 |
Neg logit contributions | |
bs | -0.304 |
gers | -0.300 |
hops | -0.293 |
counts | -0.291 |
low | -0.285 |
Token | he |
Token ID | 339 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
Baby | +0.116 |
homeless | +0.106 |
dessert | +0.106 |
onda | +0.104 |
Son | +0.104 |
Neg logit contributions | |
bs | -0.070 |
gers | -0.067 |
tail | -0.065 |
hops | -0.061 |
ines | -0.060 |
Token | shone |
Token ID | 44193 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
bs | +0.064 |
gers | +0.064 |
counts | +0.059 |
low | +0.059 |
hops | +0.058 |
Neg logit contributions | |
Baby | -0.109 |
onda | -0.099 |
homeless | -0.097 |
dessert | -0.092 |
uf | -0.092 |
Token | on |
Token ID | 319 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
bs | +0.019 |
hops | +0.019 |
counts | +0.019 |
hits | +0.018 |
rounds | +0.018 |
Neg logit contributions | |
Baby | -0.026 |
homeless | -0.024 |
dessert | -0.024 |
rice | -0.023 |
Son | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
bs | +0.129 |
gers | +0.120 |
counts | +0.119 |
hops | +0.117 |
falls | +0.116 |
Neg logit contributions | |
Baby | -0.177 |
dessert | -0.158 |
homeless | -0.154 |
onda | -0.153 |
Son | -0.151 |