Token | not |
Token ID | 407 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ost | +0.092 |
ric | +0.087 |
highlight | +0.084 |
work | +0.083 |
peac | +0.083 |
Neg logit contributions | |
iddled | -0.128 |
aback | -0.117 |
ocated | -0.112 |
uster | -0.112 |
bench | -0.109 |
Token | play |
Token ID | 711 |
Feature activation | -7.10e-03 |
Pos logit contributions | |
ost | +0.065 |
ric | +0.062 |
highlight | +0.061 |
ostr | +0.059 |
End | +0.058 |
Neg logit contributions | |
iddled | -0.057 |
aback | -0.052 |
ocated | -0.047 |
zip | -0.047 |
uster | -0.046 |
Token | with |
Token ID | 351 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
iddled | +0.000 |
uster | +0.000 |
aback | +0.000 |
bench | +0.000 |
wn | +0.000 |
Neg logit contributions | |
ost | -0.000 |
ric | -0.000 |
ostr | -0.000 |
zoo | -0.000 |
rob | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | +7.73e-01 |
Pos logit contributions | |
iddled | +0.027 |
aback | +0.023 |
uster | +0.022 |
tant | +0.021 |
ocated | +0.021 |
Neg logit contributions | |
ost | -0.029 |
ric | -0.028 |
peac | -0.026 |
ostr | -0.026 |
work | -0.026 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.48e+00 |
Pos logit contributions | |
orse | +0.048 |
highlight | +0.048 |
work | +0.047 |
ric | +0.047 |
riage | +0.045 |
Neg logit contributions | |
tant | -0.034 |
iddled | -0.031 |
aback | -0.029 |
cloth | -0.029 |
scissors | -0.029 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.62e+00 |
Pos logit contributions | |
ost | +0.269 |
ric | +0.254 |
peac | +0.238 |
highlight | +0.238 |
ostr | +0.226 |
Neg logit contributions | |
iddled | -0.497 |
aback | -0.454 |
tant | -0.435 |
uster | -0.433 |
ocated | -0.425 |
Token | ay |
Token ID | 323 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
ric | +0.637 |
work | +0.617 |
ost | +0.608 |
ostr | +0.571 |
peac | +0.570 |
Neg logit contributions | |
iddled | -0.396 |
aback | -0.335 |
uster | -0.313 |
ocated | -0.282 |
bench | -0.281 |
Token | . |
Token ID | 13 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
iddled | +0.101 |
aback | +0.087 |
uster | +0.084 |
bench | +0.076 |
tant | +0.076 |
Neg logit contributions | |
ost | -0.119 |
ric | -0.116 |
ostr | -0.111 |
work | -0.110 |
peac | -0.109 |
Token | It |
Token ID | 632 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
ost | +0.010 |
ric | +0.009 |
highlight | +0.009 |
ostr | +0.009 |
peac | +0.009 |
Neg logit contributions | |
iddled | -0.008 |
aback | -0.007 |
uster | -0.006 |
ocated | -0.006 |
tant | -0.006 |
Token | is |
Token ID | 318 |
Feature activation | -8.00e-03 |
Pos logit contributions | |
iddled | +0.046 |
aback | +0.042 |
ocated | +0.038 |
tant | +0.038 |
sc | +0.036 |
Neg logit contributions | |
ost | -0.049 |
ric | -0.049 |
highlight | -0.048 |
peac | -0.047 |
End | -0.045 |
Token | dangerous |
Token ID | 4923 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
iddled | +0.000 |
aback | +0.000 |
ocated | +0.000 |
wn | +0.000 |
bench | +0.000 |
Neg logit contributions | |
ost | -0.000 |
ric | -0.000 |
peac | -0.000 |
work | -0.000 |
highlight | -0.000 |
Token | not |
Token ID | 407 |
Feature activation | +8.20e-01 |
Pos logit contributions | |
ric | +0.170 |
ost | +0.169 |
peac | +0.157 |
ostr | +0.155 |
work | +0.150 |
Neg logit contributions | |
iddled | -0.230 |
aback | -0.206 |
ocated | -0.191 |
bench | -0.190 |
uster | -0.189 |
Token | played |
Token ID | 2826 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
ric | +0.051 |
ost | +0.050 |
peac | +0.046 |
ostr | +0.046 |
flaming | +0.046 |
Neg logit contributions | |
iddled | -0.045 |
aback | -0.041 |
wn | -0.037 |
bench | -0.036 |
tant | -0.036 |
Token | with |
Token ID | 351 |
Feature activation | -5.73e-01 |
Pos logit contributions | |
iddled | +0.033 |
aback | +0.028 |
uster | +0.028 |
bench | +0.025 |
tant | +0.025 |
Neg logit contributions | |
ost | -0.029 |
work | -0.027 |
ric | -0.027 |
zoo | -0.026 |
moral | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
iddled | +0.025 |
aback | +0.021 |
tant | +0.020 |
uster | +0.019 |
ocated | +0.017 |
Neg logit contributions | |
ric | -0.041 |
peac | -0.040 |
ost | -0.040 |
moral | -0.039 |
highlight | -0.039 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.38e+00 |
Pos logit contributions | |
ric | +0.034 |
ost | +0.032 |
peac | +0.031 |
orse | +0.030 |
work | +0.029 |
Neg logit contributions | |
iddled | -0.027 |
tant | -0.025 |
aback | -0.024 |
bench | -0.023 |
uster | -0.023 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.58e+00 |
Pos logit contributions | |
ost | +0.421 |
ric | +0.417 |
peac | +0.394 |
highlight | +0.393 |
rob | +0.384 |
Neg logit contributions | |
iddled | -0.320 |
aback | -0.284 |
tant | -0.274 |
bench | -0.264 |
uster | -0.260 |
Token | ay |
Token ID | 323 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
ric | +0.632 |
work | +0.626 |
ost | +0.604 |
peac | +0.568 |
ostr | +0.565 |
Neg logit contributions | |
iddled | -0.382 |
aback | -0.321 |
uster | -0.296 |
ocated | -0.277 |
bench | -0.270 |
Token | . |
Token ID | 13 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
iddled | +0.094 |
aback | +0.077 |
tant | +0.072 |
uster | +0.071 |
bench | +0.071 |
Neg logit contributions | |
ost | -0.142 |
ric | -0.138 |
peac | -0.138 |
highlight | -0.134 |
ostr | -0.133 |
Token | They |
Token ID | 1119 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
ost | +0.003 |
ric | +0.003 |
ostr | +0.003 |
peac | +0.003 |
rob | +0.003 |
Neg logit contributions | |
iddled | -0.002 |
aback | -0.002 |
uster | -0.001 |
bench | -0.001 |
tant | -0.001 |
Token | wish |
Token ID | 4601 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
iddled | +0.012 |
aback | +0.011 |
tant | +0.010 |
bench | +0.010 |
ocated | +0.010 |
Neg logit contributions | |
ost | -0.010 |
ric | -0.010 |
work | -0.010 |
peac | -0.009 |
flaming | -0.009 |
Token | they |
Token ID | 484 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
iddled | +0.001 |
bench | +0.000 |
ows | +0.000 |
fully | +0.000 |
uster | +0.000 |
Neg logit contributions | |
ric | -0.001 |
highlight | -0.001 |
rob | -0.001 |
ost | -0.001 |
ostr | -0.001 |
Token | mean |
Token ID | 1612 |
Feature activation | +3.16e+00 |
Pos logit contributions | |
ost | +0.046 |
ric | +0.042 |
ostr | +0.042 |
peac | +0.042 |
moral | +0.040 |
Neg logit contributions | |
iddled | -0.069 |
uster | -0.059 |
aback | -0.059 |
ocated | -0.056 |
tant | -0.055 |
Token | to |
Token ID | 284 |
Feature activation | +8.49e-02 |
Pos logit contributions | |
ost | +0.170 |
ric | +0.155 |
ostr | +0.151 |
peac | +0.145 |
moral | +0.143 |
Neg logit contributions | |
iddled | -0.225 |
uster | -0.192 |
aback | -0.187 |
bench | -0.181 |
ocated | -0.174 |
Token | break |
Token ID | 2270 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
ost | +0.004 |
ric | +0.004 |
ostr | +0.004 |
peac | +0.004 |
rob | +0.004 |
Neg logit contributions | |
iddled | -0.006 |
aback | -0.006 |
uster | -0.005 |
bench | -0.005 |
wn | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
iddled | +0.103 |
aback | +0.089 |
uster | +0.086 |
bench | +0.082 |
ocated | +0.080 |
Neg logit contributions | |
ost | -0.098 |
ric | -0.093 |
ostr | -0.089 |
peac | -0.087 |
work | -0.086 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.44e+00 |
Pos logit contributions | |
highlight | +0.031 |
ric | +0.031 |
ost | +0.030 |
orse | +0.029 |
work | +0.028 |
Neg logit contributions | |
iddled | -0.037 |
tant | -0.034 |
aback | -0.034 |
bench | -0.032 |
ocated | -0.032 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.55e+00 |
Pos logit contributions | |
ost | +0.187 |
ric | +0.173 |
ostr | +0.155 |
peac | +0.155 |
work | +0.145 |
Neg logit contributions | |
iddled | -0.592 |
aback | -0.542 |
uster | -0.527 |
tant | -0.508 |
bench | -0.507 |
Token | ay |
Token ID | 323 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
ric | +0.603 |
work | +0.585 |
ost | +0.565 |
je | +0.537 |
ostr | +0.531 |
Neg logit contributions | |
iddled | -0.410 |
aback | -0.359 |
uster | -0.330 |
bench | -0.301 |
ocated | -0.298 |
Token | . |
Token ID | 13 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
iddled | +0.116 |
aback | +0.101 |
uster | +0.097 |
tant | +0.090 |
ocated | +0.089 |
Neg logit contributions | |
ost | -0.113 |
ric | -0.109 |
ostr | -0.104 |
work | -0.104 |
peac | -0.103 |
Token | They |
Token ID | 1119 |
Feature activation | -5.72e-02 |
Pos logit contributions | |
ost | +0.005 |
ric | +0.005 |
ostr | +0.005 |
peac | +0.005 |
work | +0.004 |
Neg logit contributions | |
iddled | -0.005 |
aback | -0.004 |
uster | -0.004 |
tant | -0.004 |
bench | -0.004 |
Token | did |
Token ID | 750 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.004 |
uster | +0.004 |
bench | +0.003 |
ocated | +0.003 |
Neg logit contributions | |
ost | -0.003 |
ostr | -0.003 |
ric | -0.003 |
peac | -0.003 |
moral | -0.003 |
Token | not |
Token ID | 407 |
Feature activation | +9.08e-01 |
Pos logit contributions | |
ost | +0.020 |
peac | +0.018 |
ostr | +0.018 |
ric | +0.018 |
rob | +0.017 |
Neg logit contributions | |
iddled | -0.023 |
aback | -0.021 |
ocated | -0.020 |
tant | -0.020 |
bench | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.004 |
tant | +0.003 |
ocated | +0.003 |
uster | +0.003 |
Neg logit contributions | |
ric | -0.006 |
ost | -0.006 |
highlight | -0.006 |
work | -0.006 |
oral | -0.006 |
Token | top |
Token ID | 1353 |
Feature activation | +4.43e+00 |
Pos logit contributions | |
ric | +0.032 |
highlight | +0.030 |
orse | +0.030 |
riage | +0.029 |
ost | +0.029 |
Neg logit contributions | |
tant | -0.021 |
cloth | -0.020 |
iddled | -0.019 |
aback | -0.019 |
scissors | -0.019 |
Token | of |
Token ID | 286 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
ost | +0.285 |
ric | +0.283 |
highlight | +0.261 |
peac | +0.261 |
ostr | +0.258 |
Neg logit contributions | |
iddled | -0.253 |
aback | -0.211 |
uster | -0.202 |
bench | -0.199 |
tant | -0.199 |
Token | the |
Token ID | 262 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
ost | +0.061 |
ric | +0.059 |
highlight | +0.059 |
work | +0.058 |
peac | +0.058 |
Neg logit contributions | |
iddled | -0.039 |
aback | -0.033 |
uster | -0.029 |
tant | -0.026 |
bench | -0.026 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.39e+00 |
Pos logit contributions | |
ric | +0.032 |
orse | +0.029 |
highlight | +0.029 |
ost | +0.029 |
riage | +0.029 |
Neg logit contributions | |
tant | -0.023 |
cloth | -0.020 |
away | -0.019 |
aback | -0.019 |
iddled | -0.019 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.54e+00 |
Pos logit contributions | |
ost | +0.233 |
ric | +0.214 |
peac | +0.203 |
highlight | +0.200 |
moral | +0.197 |
Neg logit contributions | |
iddled | -0.533 |
aback | -0.481 |
uster | -0.463 |
bench | -0.460 |
tant | -0.455 |
Token | ay |
Token ID | 323 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
ric | +0.615 |
work | +0.595 |
ost | +0.587 |
ostr | +0.555 |
peac | +0.552 |
Neg logit contributions | |
iddled | -0.380 |
aback | -0.335 |
uster | -0.314 |
bench | -0.287 |
ocated | -0.284 |
Token | fell |
Token ID | 3214 |
Feature activation | +9.45e-02 |
Pos logit contributions | |
iddled | +0.106 |
aback | +0.089 |
uster | +0.085 |
bench | +0.080 |
ocated | +0.079 |
Neg logit contributions | |
ost | -0.130 |
ric | -0.126 |
peac | -0.120 |
highlight | -0.119 |
moral | -0.118 |
Token | off |
Token ID | 572 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
ost | +0.006 |
ric | +0.006 |
work | +0.006 |
ostr | +0.005 |
peac | +0.005 |
Neg logit contributions | |
iddled | -0.005 |
aback | -0.004 |
uster | -0.004 |
tant | -0.004 |
bench | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
iddled | +0.062 |
aback | +0.054 |
uster | +0.052 |
tant | +0.048 |
bench | +0.048 |
Neg logit contributions | |
ost | -0.065 |
ric | -0.062 |
ostr | -0.060 |
peac | -0.059 |
work | -0.058 |
Token | Bill |
Token ID | 3941 |
Feature activation | +7.47e-01 |
Pos logit contributions | |
highlight | +0.017 |
work | +0.017 |
ost | +0.017 |
wedding | +0.016 |
flaming | +0.016 |
Neg logit contributions | |
iddled | -0.013 |
aback | -0.013 |
uster | -0.010 |
bench | -0.010 |
ocated | -0.009 |
Token | did |
Token ID | 750 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
ost | +0.172 |
ric | +0.163 |
ostr | +0.159 |
peac | +0.159 |
work | +0.153 |
Neg logit contributions | |
iddled | -0.151 |
aback | -0.128 |
uster | -0.120 |
tant | -0.118 |
ocated | -0.115 |
Token | you |
Token ID | 345 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
ric | +0.024 |
ost | +0.024 |
highlight | +0.023 |
ostr | +0.023 |
work | +0.023 |
Neg logit contributions | |
iddled | -0.018 |
aback | -0.015 |
ocated | -0.015 |
tant | -0.014 |
uster | -0.014 |
Token | touch |
Token ID | 3638 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
ost | +0.152 |
ric | +0.145 |
ostr | +0.139 |
peac | +0.138 |
work | +0.135 |
Neg logit contributions | |
iddled | -0.151 |
aback | -0.132 |
uster | -0.126 |
tant | -0.119 |
ocated | -0.118 |
Token | my |
Token ID | 616 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
ost | +0.103 |
ric | +0.099 |
ostr | +0.095 |
peac | +0.094 |
work | +0.092 |
Neg logit contributions | |
iddled | -0.093 |
aback | -0.081 |
uster | -0.077 |
tant | -0.072 |
bench | -0.072 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.41e+00 |
Pos logit contributions | |
highlight | +0.032 |
ost | +0.031 |
ric | +0.030 |
Whale | +0.028 |
peac | +0.028 |
Neg logit contributions | |
iddled | -0.031 |
tant | -0.031 |
aback | -0.029 |
away | -0.027 |
uster | -0.026 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.53e+00 |
Pos logit contributions | |
ost | +0.186 |
ric | +0.170 |
ostr | +0.153 |
peac | +0.153 |
work | +0.141 |
Neg logit contributions | |
iddled | -0.590 |
aback | -0.540 |
uster | -0.525 |
tant | -0.508 |
bench | -0.505 |
Token | ay |
Token ID | 323 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ric | +0.635 |
work | +0.611 |
ost | +0.603 |
peac | +0.572 |
ostr | +0.568 |
Neg logit contributions | |
iddled | -0.387 |
aback | -0.327 |
uster | -0.307 |
ocated | -0.279 |
bench | -0.278 |
Token | ? |
Token ID | 30 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
iddled | +0.128 |
aback | +0.113 |
uster | +0.111 |
tant | +0.103 |
ocated | +0.103 |
Neg logit contributions | |
ric | -0.094 |
ost | -0.093 |
work | -0.092 |
ostr | -0.086 |
peac | -0.085 |
Token | Do |
Token ID | 2141 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
ost | +0.067 |
highlight | +0.065 |
ostr | +0.063 |
ric | +0.062 |
peac | +0.062 |
Neg logit contributions | |
iddled | -0.046 |
aback | -0.038 |
uster | -0.036 |
ocated | -0.033 |
zip | -0.032 |
Token | you |
Token ID | 345 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ost | +0.180 |
ric | +0.176 |
work | +0.175 |
highlight | +0.175 |
North | +0.174 |
Neg logit contributions | |
iddled | -0.143 |
ocated | -0.120 |
aback | -0.114 |
tant | -0.112 |
olls | -0.109 |
Token | know |
Token ID | 760 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
ost | +0.153 |
ric | +0.145 |
ostr | +0.140 |
peac | +0.138 |
work | +0.137 |
Neg logit contributions | |
iddled | -0.152 |
aback | -0.134 |
uster | -0.127 |
tant | -0.122 |
bench | -0.119 |
Token |
|
Token ID | 198 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
iddled | +0.007 |
aback | +0.006 |
uster | +0.006 |
ocated | +0.006 |
bench | +0.006 |
Neg logit contributions | |
ost | -0.005 |
ric | -0.004 |
ostr | -0.004 |
peac | -0.004 |
moral | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
iddled | +0.020 |
aback | +0.018 |
uster | +0.017 |
tant | +0.016 |
bench | +0.016 |
Neg logit contributions | |
ost | -0.013 |
ric | -0.013 |
ostr | -0.012 |
peac | -0.012 |
moral | -0.012 |
Token | " |
Token ID | 1 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
iddled | +0.011 |
aback | +0.009 |
uster | +0.009 |
tant | +0.008 |
bench | +0.008 |
Neg logit contributions | |
ost | -0.015 |
ric | -0.014 |
ostr | -0.014 |
peac | -0.014 |
work | -0.013 |
Token | And |
Token ID | 1870 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
iddled | +0.066 |
aback | +0.057 |
uster | +0.054 |
tant | +0.052 |
ocated | +0.052 |
Neg logit contributions | |
ric | -0.067 |
ost | -0.067 |
ostr | -0.062 |
peac | -0.061 |
work | -0.061 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.42e+00 |
Pos logit contributions | |
ric | +0.096 |
ost | +0.091 |
peac | +0.087 |
flaming | +0.085 |
highlight | +0.085 |
Neg logit contributions | |
iddled | -0.072 |
aback | -0.070 |
uster | -0.060 |
ocated | -0.059 |
away | -0.056 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
ost | +0.281 |
ric | +0.266 |
peac | +0.251 |
ostr | +0.248 |
highlight | +0.239 |
Neg logit contributions | |
iddled | -0.495 |
aback | -0.446 |
uster | -0.430 |
tant | -0.412 |
bench | -0.411 |
Token | ays |
Token ID | 592 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
ric | +0.594 |
work | +0.578 |
ost | +0.562 |
peac | +0.516 |
ostr | +0.516 |
Neg logit contributions | |
iddled | -0.399 |
aback | -0.354 |
bench | -0.335 |
uster | -0.333 |
ocated | -0.311 |
Token | are |
Token ID | 389 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
iddled | +0.047 |
aback | +0.040 |
uster | +0.040 |
ocated | +0.037 |
bench | +0.037 |
Neg logit contributions | |
ost | -0.045 |
ric | -0.043 |
ostr | -0.042 |
peac | -0.041 |
moral | -0.040 |
Token | not |
Token ID | 407 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
ric | +0.013 |
ost | +0.013 |
peac | +0.012 |
ostr | +0.012 |
work | +0.012 |
Neg logit contributions | |
iddled | -0.012 |
aback | -0.011 |
uster | -0.010 |
tant | -0.010 |
ocated | -0.010 |
Token | hats |
Token ID | 23910 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
ost | +0.056 |
ric | +0.054 |
ostr | +0.052 |
peac | +0.051 |
work | +0.050 |
Neg logit contributions | |
iddled | -0.054 |
aback | -0.047 |
uster | -0.045 |
tant | -0.042 |
bench | -0.041 |
Token | ," |
Token ID | 553 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
iddled | +0.133 |
aback | +0.115 |
uster | +0.110 |
tant | +0.103 |
bench | +0.102 |
Neg logit contributions | |
ost | -0.148 |
ric | -0.143 |
ostr | -0.137 |
peac | -0.136 |
work | -0.133 |
Token |
|
Token ID | 198 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
ost | +0.019 |
highlight | +0.019 |
work | +0.019 |
peac | +0.019 |
ostr | +0.018 |
Neg logit contributions | |
iddled | -0.010 |
aback | -0.010 |
bench | -0.008 |
uster | -0.007 |
ocated | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
iddled | +0.008 |
aback | +0.007 |
uster | +0.007 |
bench | +0.007 |
ocated | +0.006 |
Neg logit contributions | |
ost | -0.009 |
work | -0.009 |
highlight | -0.009 |
ric | -0.009 |
ostr | -0.009 |
Token | The |
Token ID | 464 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
iddled | +0.003 |
aback | +0.003 |
uster | +0.003 |
bench | +0.002 |
zip | +0.002 |
Neg logit contributions | |
work | -0.007 |
ric | -0.007 |
wedding | -0.006 |
highlight | -0.006 |
ost | -0.006 |
Token | poor |
Token ID | 3595 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ric | +0.203 |
work | +0.176 |
highlight | +0.169 |
Clown | +0.166 |
field | +0.165 |
Neg logit contributions | |
cloth | -0.107 |
tant | -0.106 |
mess | -0.099 |
breath | -0.097 |
scissors | -0.095 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.38e+00 |
Pos logit contributions | |
ric | +0.080 |
highlight | +0.079 |
work | +0.070 |
riage | +0.065 |
happiest | +0.065 |
Neg logit contributions | |
iddled | -0.049 |
cloth | -0.047 |
tant | -0.047 |
ly | -0.045 |
aback | -0.044 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.49e+00 |
Pos logit contributions | |
ost | +0.285 |
ric | +0.275 |
highlight | +0.258 |
peac | +0.243 |
North | +0.237 |
Neg logit contributions | |
iddled | -0.475 |
aback | -0.429 |
bench | -0.409 |
ocated | -0.403 |
tant | -0.401 |
Token | ay |
Token ID | 323 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
ric | +0.622 |
work | +0.608 |
ost | +0.603 |
peac | +0.574 |
ostr | +0.570 |
Neg logit contributions | |
iddled | -0.360 |
aback | -0.313 |
uster | -0.296 |
bench | -0.276 |
ocated | -0.263 |
Token | was |
Token ID | 373 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
iddled | +0.127 |
aback | +0.106 |
uster | +0.098 |
ocated | +0.097 |
bench | +0.094 |
Neg logit contributions | |
ost | -0.112 |
ric | -0.110 |
highlight | -0.107 |
peac | -0.102 |
North | -0.099 |
Token | broken |
Token ID | 5445 |
Feature activation | +9.59e-01 |
Pos logit contributions | |
ost | +0.021 |
ric | +0.020 |
ostr | +0.019 |
peac | +0.019 |
work | +0.018 |
Neg logit contributions | |
iddled | -0.025 |
aback | -0.022 |
uster | -0.021 |
bench | -0.020 |
tant | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
ost | +0.050 |
ric | +0.048 |
work | +0.048 |
ostr | +0.046 |
peac | +0.045 |
Neg logit contributions | |
iddled | -0.064 |
aback | -0.055 |
uster | -0.054 |
tant | -0.052 |
wn | -0.050 |
Token | it |
Token ID | 340 |
Feature activation | -9.07e-01 |
Pos logit contributions | |
ric | +0.001 |
ost | +0.001 |
flaming | +0.001 |
highlight | +0.001 |
work | +0.001 |
Neg logit contributions | |
iddled | -0.001 |
aback | -0.001 |
tant | -0.001 |
bench | -0.001 |
ocated | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
ost | +0.005 |
ric | +0.005 |
ostr | +0.005 |
peac | +0.005 |
moral | +0.005 |
Neg logit contributions | |
iddled | -0.005 |
uster | -0.004 |
aback | -0.004 |
tant | -0.004 |
wn | -0.004 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
ost | +0.009 |
ostr | +0.008 |
ric | +0.008 |
peac | +0.008 |
moral | +0.008 |
Neg logit contributions | |
iddled | -0.015 |
uster | -0.014 |
aback | -0.013 |
tant | -0.013 |
bench | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +6.27e-01 |
Pos logit contributions | |
ost | +0.027 |
ric | +0.026 |
ostr | +0.025 |
peac | +0.025 |
work | +0.024 |
Neg logit contributions | |
iddled | -0.028 |
aback | -0.024 |
uster | -0.024 |
bench | -0.022 |
tant | -0.022 |
Token | broken |
Token ID | 5445 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ric | +0.046 |
highlight | +0.046 |
ost | +0.046 |
riage | +0.045 |
orse | +0.045 |
Neg logit contributions | |
tant | -0.023 |
iddled | -0.021 |
aback | -0.019 |
bench | -0.016 |
ocated | -0.016 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.40e+00 |
Pos logit contributions | |
ost | +0.054 |
highlight | +0.054 |
ric | +0.053 |
happiest | +0.047 |
North | +0.047 |
Neg logit contributions | |
iddled | -0.064 |
aback | -0.057 |
tant | -0.056 |
bench | -0.053 |
uster | -0.052 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.49e+00 |
Pos logit contributions | |
ost | +0.243 |
ric | +0.227 |
peac | +0.213 |
ostr | +0.209 |
highlight | +0.202 |
Neg logit contributions | |
iddled | -0.530 |
aback | -0.480 |
uster | -0.464 |
bench | -0.450 |
tant | -0.450 |
Token | ay |
Token ID | 323 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
ric | +0.590 |
work | +0.575 |
ost | +0.546 |
je | +0.521 |
peac | +0.512 |
Neg logit contributions | |
iddled | -0.399 |
aback | -0.362 |
uster | -0.329 |
bench | -0.307 |
wn | -0.302 |
Token | and |
Token ID | 290 |
Feature activation | +9.56e-04 |
Pos logit contributions | |
iddled | +0.106 |
aback | +0.091 |
uster | +0.086 |
tant | +0.080 |
bench | +0.080 |
Neg logit contributions | |
ost | -0.129 |
ric | -0.124 |
ostr | -0.119 |
peac | -0.119 |
work | -0.116 |
Token | the |
Token ID | 262 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
ost | +0.000 |
ric | +0.000 |
ostr | +0.000 |
peac | +0.000 |
work | +0.000 |
Neg logit contributions | |
iddled | -0.000 |
aback | -0.000 |
uster | -0.000 |
tant | -0.000 |
bench | -0.000 |
Token | mess |
Token ID | 2085 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
highlight | +0.036 |
ost | +0.035 |
ric | +0.035 |
orse | +0.035 |
riage | +0.034 |
Neg logit contributions | |
tant | -0.027 |
iddled | -0.024 |
aback | -0.023 |
bench | -0.022 |
away | -0.020 |
Token | on |
Token ID | 319 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
iddled | +0.226 |
aback | +0.211 |
uster | +0.200 |
tant | +0.190 |
wn | +0.180 |
Neg logit contributions | |
ost | -0.169 |
ric | -0.163 |
work | -0.163 |
ostr | -0.151 |
moral | -0.149 |
Token | they |
Token ID | 484 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
ric | +0.003 |
oral | +0.003 |
North | +0.003 |
rob | +0.003 |
str | +0.003 |
Neg logit contributions | |
iddled | -0.002 |
socks | -0.002 |
away | -0.001 |
her | -0.001 |
inn | -0.001 |
Token | also |
Token ID | 635 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
ost | +0.041 |
ric | +0.039 |
peac | +0.038 |
work | +0.037 |
ostr | +0.036 |
Neg logit contributions | |
iddled | -0.056 |
aback | -0.050 |
uster | -0.047 |
tant | -0.047 |
bench | -0.047 |
Token | see |
Token ID | 766 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
ost | +0.211 |
ric | +0.204 |
peac | +0.198 |
ostr | +0.194 |
work | +0.193 |
Neg logit contributions | |
iddled | -0.160 |
aback | -0.142 |
bench | -0.129 |
tant | -0.127 |
uster | -0.125 |
Token | an |
Token ID | 281 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
iddled | +0.006 |
aback | +0.005 |
tant | +0.004 |
uster | +0.004 |
bench | +0.004 |
Neg logit contributions | |
ric | -0.010 |
ost | -0.010 |
highlight | -0.009 |
str | -0.009 |
peac | -0.009 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.33e+00 |
Pos logit contributions | |
iddled | +0.020 |
aback | +0.018 |
uster | +0.017 |
tant | +0.017 |
bench | +0.017 |
Neg logit contributions | |
ost | -0.016 |
ric | -0.016 |
peac | -0.015 |
ostr | -0.014 |
highlight | -0.014 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.47e+00 |
Pos logit contributions | |
ost | +0.223 |
ric | +0.209 |
highlight | +0.198 |
peac | +0.195 |
moral | +0.188 |
Neg logit contributions | |
iddled | -0.526 |
aback | -0.466 |
uster | -0.457 |
bench | -0.457 |
ocated | -0.448 |
Token | ay |
Token ID | 323 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
work | +0.574 |
ric | +0.556 |
ost | +0.509 |
je | +0.486 |
chim | +0.481 |
Neg logit contributions | |
iddled | -0.418 |
aback | -0.382 |
uster | -0.360 |
ocated | -0.330 |
wn | -0.327 |
Token | on |
Token ID | 319 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
iddled | +0.109 |
aback | +0.091 |
uster | +0.088 |
bench | +0.083 |
ocated | +0.082 |
Neg logit contributions | |
ost | -0.133 |
ric | -0.130 |
peac | -0.124 |
ostr | -0.123 |
highlight | -0.121 |
Token | the |
Token ID | 262 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
ost | +0.057 |
ric | +0.055 |
ostr | +0.052 |
work | +0.051 |
peac | +0.051 |
Neg logit contributions | |
iddled | -0.069 |
aback | -0.061 |
uster | -0.059 |
ocated | -0.055 |
tant | -0.054 |
Token | table |
Token ID | 3084 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
ost | +0.025 |
ric | +0.024 |
peac | +0.023 |
rob | +0.022 |
highlight | +0.022 |
Neg logit contributions | |
iddled | -0.019 |
aback | -0.016 |
uster | -0.016 |
bench | -0.015 |
tant | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
iddled | +0.125 |
aback | +0.107 |
uster | +0.102 |
bench | +0.097 |
tant | +0.096 |
Neg logit contributions | |
ost | -0.135 |
ric | -0.130 |
peac | -0.125 |
ostr | -0.125 |
work | -0.121 |
Token | should |
Token ID | 815 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
ost | +0.163 |
ric | +0.156 |
highlight | +0.153 |
peac | +0.151 |
ostr | +0.151 |
Neg logit contributions | |
iddled | -0.148 |
aback | -0.138 |
uster | -0.125 |
tant | -0.123 |
ocated | -0.121 |
Token | never |
Token ID | 1239 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
ost | +0.096 |
highlight | +0.092 |
work | +0.092 |
ric | +0.089 |
peac | +0.089 |
Neg logit contributions | |
iddled | -0.098 |
aback | -0.090 |
uster | -0.087 |
ocated | -0.086 |
bench | -0.086 |
Token | touch |
Token ID | 3638 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ost | +0.128 |
ric | +0.119 |
End | +0.118 |
highlight | +0.118 |
work | +0.117 |
Neg logit contributions | |
iddled | -0.074 |
aback | -0.067 |
tant | -0.063 |
wn | -0.063 |
zip | -0.062 |
Token | an |
Token ID | 281 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
ost | +0.104 |
ric | +0.102 |
highlight | +0.099 |
ostr | +0.099 |
peac | +0.099 |
Neg logit contributions | |
iddled | -0.069 |
aback | -0.062 |
uster | -0.055 |
tant | -0.055 |
ocated | -0.049 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.38e+00 |
Pos logit contributions | |
iddled | +0.008 |
aback | +0.007 |
uster | +0.007 |
tant | +0.007 |
bench | +0.007 |
Neg logit contributions | |
ost | -0.005 |
ric | -0.005 |
highlight | -0.005 |
peac | -0.005 |
work | -0.005 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.45e+00 |
Pos logit contributions | |
ost | +0.224 |
ric | +0.208 |
peac | +0.190 |
ostr | +0.186 |
highlight | +0.186 |
Neg logit contributions | |
iddled | -0.540 |
aback | -0.495 |
uster | -0.481 |
tant | -0.468 |
ocated | -0.463 |
Token | ay |
Token ID | 323 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
ric | +0.618 |
work | +0.605 |
ost | +0.585 |
peac | +0.558 |
ostr | +0.551 |
Neg logit contributions | |
iddled | -0.391 |
aback | -0.324 |
uster | -0.306 |
bench | -0.282 |
ocated | -0.280 |
Token | ." |
Token ID | 526 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
iddled | +0.112 |
aback | +0.098 |
uster | +0.093 |
tant | +0.088 |
bench | +0.087 |
Neg logit contributions | |
ost | -0.120 |
ric | -0.116 |
ostr | -0.110 |
peac | -0.110 |
highlight | -0.107 |
Token |
|
Token ID | 198 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
iddled | +0.010 |
aback | +0.009 |
uster | +0.008 |
tant | +0.008 |
bench | +0.008 |
Neg logit contributions | |
ost | -0.007 |
ric | -0.006 |
ostr | -0.006 |
peac | -0.006 |
moral | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
iddled | +0.021 |
aback | +0.018 |
uster | +0.018 |
tant | +0.017 |
ocated | +0.017 |
Neg logit contributions | |
ost | -0.013 |
ric | -0.012 |
ostr | -0.011 |
peac | -0.011 |
moral | -0.011 |
Token | S |
Token ID | 50 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
iddled | +0.012 |
aback | +0.010 |
uster | +0.010 |
tant | +0.009 |
bench | +0.009 |
Neg logit contributions | |
ost | -0.015 |
ric | -0.014 |
ostr | -0.014 |
peac | -0.014 |
moral | -0.014 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ost | +0.042 |
ric | +0.039 |
ostr | +0.038 |
peac | +0.038 |
work | +0.037 |
Neg logit contributions | |
iddled | -0.052 |
aback | -0.047 |
uster | -0.046 |
tant | -0.044 |
bench | -0.043 |
Token | at |
Token ID | 379 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
ost | +0.080 |
ostr | +0.074 |
moral | +0.074 |
ric | +0.069 |
hipp | +0.068 |
Neg logit contributions | |
iddled | -0.093 |
ocated | -0.087 |
uster | -0.085 |
bench | -0.083 |
aback | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
ost | +0.090 |
ric | +0.087 |
ostr | +0.084 |
peac | +0.082 |
work | +0.081 |
Neg logit contributions | |
iddled | -0.076 |
aback | -0.064 |
uster | -0.061 |
bench | -0.058 |
ocated | -0.058 |
Token | empty |
Token ID | 6565 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
ost | +0.037 |
ric | +0.035 |
peac | +0.034 |
ostr | +0.033 |
highlight | +0.032 |
Neg logit contributions | |
iddled | -0.039 |
aback | -0.035 |
tant | -0.033 |
uster | -0.033 |
bench | -0.032 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.45e+00 |
Pos logit contributions | |
iddled | +0.008 |
aback | +0.007 |
uster | +0.007 |
tant | +0.006 |
bench | +0.006 |
Neg logit contributions | |
ost | -0.005 |
ric | -0.005 |
ostr | -0.005 |
peac | -0.005 |
work | -0.005 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.45e+00 |
Pos logit contributions | |
ost | +0.251 |
ric | +0.233 |
peac | +0.220 |
highlight | +0.219 |
ostr | +0.211 |
Neg logit contributions | |
iddled | -0.522 |
aback | -0.470 |
bench | -0.454 |
uster | -0.454 |
tant | -0.446 |
Token | ay |
Token ID | 323 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
ric | +0.596 |
work | +0.589 |
ost | +0.553 |
je | +0.527 |
peac | +0.523 |
Neg logit contributions | |
iddled | -0.386 |
aback | -0.346 |
uster | -0.324 |
ocated | -0.300 |
wn | -0.293 |
Token | on |
Token ID | 319 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
iddled | +0.093 |
aback | +0.078 |
uster | +0.073 |
bench | +0.072 |
tant | +0.068 |
Neg logit contributions | |
ost | -0.135 |
ric | -0.131 |
peac | -0.128 |
highlight | -0.125 |
rob | -0.125 |
Token | Sara |
Token ID | 24799 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
ost | +0.054 |
ostr | +0.050 |
zoo | +0.049 |
ric | +0.049 |
moral | +0.047 |
Neg logit contributions | |
iddled | -0.080 |
aback | -0.068 |
uster | -0.068 |
wn | -0.064 |
iddling | -0.064 |
Token | 's |
Token ID | 338 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
iddled | +0.024 |
aback | +0.021 |
uster | +0.021 |
bench | +0.020 |
ocated | +0.019 |
Neg logit contributions | |
ost | -0.019 |
ric | -0.018 |
peac | -0.017 |
ostr | -0.017 |
highlight | -0.017 |
Token | head |
Token ID | 1182 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
ost | +0.155 |
ric | +0.147 |
peac | +0.141 |
highlight | +0.141 |
work | +0.138 |
Neg logit contributions | |
iddled | -0.146 |
aback | -0.129 |
uster | -0.125 |
tant | -0.122 |
bench | -0.119 |
Token | not |
Token ID | 407 |
Feature activation | +8.69e-01 |
Pos logit contributions | |
ost | +0.157 |
ric | +0.152 |
ostr | +0.142 |
peac | +0.141 |
work | +0.137 |
Neg logit contributions | |
iddled | -0.250 |
aback | -0.225 |
uster | -0.212 |
bench | -0.207 |
ocated | -0.207 |
Token | played |
Token ID | 2826 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
ric | +0.055 |
ost | +0.055 |
work | +0.051 |
North | +0.051 |
ostr | +0.051 |
Neg logit contributions | |
iddled | -0.048 |
aback | -0.041 |
wn | -0.037 |
ocated | -0.037 |
bench | -0.036 |
Token | with |
Token ID | 351 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
iddled | +0.032 |
uster | +0.028 |
aback | +0.028 |
bench | +0.024 |
tant | +0.024 |
Neg logit contributions | |
ost | -0.028 |
work | -0.025 |
ric | -0.025 |
zoo | -0.025 |
peac | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
iddled | +0.025 |
aback | +0.020 |
tant | +0.020 |
uster | +0.019 |
ocated | +0.018 |
Neg logit contributions | |
ost | -0.035 |
ric | -0.035 |
peac | -0.034 |
highlight | -0.033 |
ostr | -0.033 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.39e+00 |
Pos logit contributions | |
ric | +0.029 |
highlight | +0.028 |
orse | +0.028 |
work | +0.027 |
ost | +0.027 |
Neg logit contributions | |
iddled | -0.026 |
tant | -0.025 |
aback | -0.024 |
bench | -0.022 |
ocated | -0.021 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.45e+00 |
Pos logit contributions | |
ost | +0.238 |
ric | +0.225 |
highlight | +0.208 |
peac | +0.207 |
rob | +0.197 |
Neg logit contributions | |
iddled | -0.517 |
aback | -0.477 |
tant | -0.455 |
uster | -0.454 |
bench | -0.452 |
Token | ay |
Token ID | 323 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
ric | +0.625 |
work | +0.608 |
ost | +0.600 |
peac | +0.568 |
ostr | +0.563 |
Neg logit contributions | |
iddled | -0.380 |
aback | -0.321 |
uster | -0.298 |
ocated | -0.271 |
bench | -0.267 |
Token | . |
Token ID | 13 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
iddled | +0.107 |
aback | +0.091 |
uster | +0.086 |
tant | +0.083 |
bench | +0.082 |
Neg logit contributions | |
ost | -0.133 |
ric | -0.128 |
peac | -0.126 |
ostr | -0.124 |
highlight | -0.122 |
Token | They |
Token ID | 1119 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
iddled | +0.001 |
aback | +0.001 |
uster | +0.001 |
tant | +0.001 |
bench | +0.001 |
Neg logit contributions | |
ost | -0.001 |
ric | -0.001 |
ostr | -0.001 |
peac | -0.001 |
work | -0.001 |
Token | wished |
Token ID | 16555 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
iddled | +0.008 |
aback | +0.007 |
tant | +0.007 |
uster | +0.007 |
bench | +0.007 |
Neg logit contributions | |
ost | -0.007 |
ric | -0.007 |
work | -0.006 |
peac | -0.006 |
ostr | -0.006 |
Token | they |
Token ID | 484 |
Feature activation | +8.63e-01 |
Pos logit contributions | |
iddled | +0.039 |
her | +0.035 |
fully | +0.035 |
them | +0.032 |
him | +0.030 |
Neg logit contributions | |
ric | -0.081 |
ostr | -0.080 |
ost | -0.079 |
str | -0.078 |
rob | -0.078 |
Token | They |
Token ID | 1119 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
ost | +0.020 |
ric | +0.019 |
peac | +0.019 |
ostr | +0.018 |
highlight | +0.018 |
Neg logit contributions | |
iddled | -0.014 |
aback | -0.012 |
uster | -0.012 |
tant | -0.010 |
bench | -0.010 |
Token | took |
Token ID | 1718 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
iddled | +0.004 |
uster | +0.004 |
aback | +0.004 |
bench | +0.004 |
tant | +0.004 |
Neg logit contributions | |
ost | -0.003 |
ostr | -0.003 |
peac | -0.003 |
ric | -0.003 |
moral | -0.003 |
Token | off |
Token ID | 572 |
Feature activation | -9.57e-01 |
Pos logit contributions | |
ost | +0.020 |
ostr | +0.018 |
work | +0.018 |
ric | +0.018 |
highlight | +0.017 |
Neg logit contributions | |
iddled | -0.025 |
uster | -0.021 |
bench | -0.021 |
aback | -0.021 |
ocated | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
iddled | +0.043 |
aback | +0.037 |
tant | +0.033 |
uster | +0.033 |
bench | +0.030 |
Neg logit contributions | |
ric | -0.072 |
ost | -0.069 |
peac | -0.066 |
highlight | -0.066 |
ostr | -0.065 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.39e+00 |
Pos logit contributions | |
ric | +0.035 |
ost | +0.034 |
highlight | +0.033 |
work | +0.031 |
peac | +0.031 |
Neg logit contributions | |
iddled | -0.047 |
aback | -0.043 |
tant | -0.041 |
uster | -0.040 |
bench | -0.039 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.44e+00 |
Pos logit contributions | |
ost | +0.158 |
ric | +0.145 |
ostr | +0.132 |
peac | +0.127 |
work | +0.124 |
Neg logit contributions | |
iddled | -0.620 |
aback | -0.566 |
uster | -0.553 |
bench | -0.530 |
tant | -0.529 |
Token | ay |
Token ID | 323 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
ric | +0.556 |
work | +0.549 |
je | +0.506 |
ost | +0.505 |
peac | +0.478 |
Neg logit contributions | |
iddled | -0.409 |
aback | -0.360 |
ocated | -0.335 |
uster | -0.333 |
bench | -0.319 |
Token | and |
Token ID | 290 |
Feature activation | -8.82e-02 |
Pos logit contributions | |
iddled | +0.098 |
aback | +0.086 |
uster | +0.081 |
bench | +0.076 |
tant | +0.075 |
Neg logit contributions | |
ost | -0.132 |
ric | -0.127 |
peac | -0.124 |
highlight | -0.123 |
ostr | -0.122 |
Token | the |
Token ID | 262 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
iddled | +0.005 |
aback | +0.004 |
uster | +0.004 |
tant | +0.004 |
bench | +0.004 |
Neg logit contributions | |
ost | -0.006 |
ric | -0.006 |
peac | -0.006 |
ostr | -0.005 |
highlight | -0.005 |
Token | book |
Token ID | 1492 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
ric | +0.042 |
highlight | +0.042 |
orse | +0.041 |
work | +0.041 |
ost | +0.040 |
Neg logit contributions | |
tant | -0.030 |
iddled | -0.029 |
cloth | -0.026 |
aback | -0.026 |
bench | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
ost | +0.122 |
ric | +0.114 |
ostr | +0.112 |
peac | +0.111 |
work | +0.111 |
Neg logit contributions | |
iddled | -0.135 |
aback | -0.118 |
uster | -0.115 |
tant | -0.105 |
ocated | -0.104 |
Token | , |
Token ID | 11 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
ost | +0.080 |
ostr | +0.073 |
peac | +0.072 |
ric | +0.071 |
moral | +0.069 |
Neg logit contributions | |
iddled | -0.112 |
aback | -0.098 |
uster | -0.096 |
bench | -0.094 |
ocated | -0.089 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
ost | +0.004 |
ostr | +0.004 |
moral | +0.004 |
peac | +0.004 |
flaming | +0.004 |
Neg logit contributions | |
iddled | -0.005 |
aback | -0.005 |
uster | -0.005 |
bench | -0.004 |
wn | -0.004 |
Token | found |
Token ID | 1043 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
iddled | +0.101 |
aback | +0.089 |
uster | +0.085 |
bench | +0.081 |
ocated | +0.080 |
Neg logit contributions | |
ost | -0.083 |
ric | -0.081 |
peac | -0.075 |
ostr | -0.074 |
work | -0.073 |
Token | an |
Token ID | 281 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
iddled | +0.180 |
uster | +0.159 |
aback | +0.155 |
bench | +0.153 |
tant | +0.151 |
Neg logit contributions | |
ost | -0.119 |
ostr | -0.110 |
peac | -0.107 |
ric | -0.105 |
rob | -0.104 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.19e+00 |
Pos logit contributions | |
iddled | +0.035 |
aback | +0.031 |
uster | +0.030 |
tant | +0.029 |
bench | +0.029 |
Neg logit contributions | |
ost | -0.026 |
ric | -0.024 |
ostr | -0.024 |
peac | -0.023 |
work | -0.022 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.44e+00 |
Pos logit contributions | |
ost | +0.210 |
ric | +0.200 |
peac | +0.177 |
highlight | +0.176 |
ostr | +0.169 |
Neg logit contributions | |
iddled | -0.528 |
aback | -0.477 |
uster | -0.467 |
bench | -0.458 |
tant | -0.455 |
Token | ay |
Token ID | 323 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
work | +0.564 |
ric | +0.549 |
ost | +0.513 |
je | +0.497 |
peac | +0.494 |
Neg logit contributions | |
iddled | -0.413 |
aback | -0.368 |
uster | -0.337 |
ocated | -0.317 |
bench | -0.315 |
Token | in |
Token ID | 287 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
iddled | +0.104 |
aback | +0.083 |
uster | +0.080 |
bench | +0.076 |
ocated | +0.076 |
Neg logit contributions | |
ost | -0.156 |
ric | -0.153 |
peac | -0.146 |
highlight | -0.144 |
rob | -0.143 |
Token | the |
Token ID | 262 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
ost | +0.022 |
zoo | +0.020 |
ostr | +0.020 |
ric | +0.020 |
rob | +0.020 |
Neg logit contributions | |
iddled | -0.029 |
uster | -0.026 |
aback | -0.025 |
wn | -0.024 |
tant | -0.023 |
Token | world |
Token ID | 995 |
Feature activation | +9.64e-01 |
Pos logit contributions | |
ric | +0.018 |
ost | +0.018 |
work | +0.017 |
peac | +0.016 |
orse | +0.016 |
Neg logit contributions | |
iddled | -0.022 |
aback | -0.021 |
uster | -0.019 |
tant | -0.019 |
bench | -0.019 |
Token | behind |
Token ID | 2157 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ost | +0.064 |
ric | +0.062 |
peac | +0.059 |
ostr | +0.059 |
highlight | +0.058 |
Neg logit contributions | |
iddled | -0.053 |
aback | -0.046 |
uster | -0.042 |
bench | -0.041 |
ocated | -0.040 |
Token | room |
Token ID | 2119 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
ost | +0.017 |
ric | +0.016 |
peac | +0.015 |
work | +0.015 |
flaming | +0.015 |
Neg logit contributions | |
iddled | -0.022 |
aback | -0.020 |
uster | -0.020 |
tant | -0.019 |
ocated | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
iddled | +0.117 |
aback | +0.101 |
uster | +0.098 |
tant | +0.092 |
bench | +0.090 |
Neg logit contributions | |
ost | -0.120 |
ric | -0.115 |
ostr | -0.111 |
peac | -0.111 |
work | -0.110 |
Token | saw |
Token ID | 2497 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
iddled | +0.016 |
aback | +0.014 |
uster | +0.012 |
ocated | +0.012 |
tant | +0.012 |
Neg logit contributions | |
ost | -0.019 |
ric | -0.018 |
ostr | -0.017 |
peac | -0.017 |
flaming | -0.017 |
Token | an |
Token ID | 281 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
ost | +0.006 |
ric | +0.006 |
peac | +0.006 |
ostr | +0.005 |
work | +0.005 |
Neg logit contributions | |
iddled | -0.005 |
aback | -0.005 |
uster | -0.004 |
tant | -0.004 |
ocated | -0.004 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.14e+00 |
Pos logit contributions | |
iddled | +0.031 |
aback | +0.028 |
uster | +0.027 |
tant | +0.025 |
bench | +0.025 |
Neg logit contributions | |
ric | -0.027 |
ost | -0.027 |
peac | -0.025 |
work | -0.025 |
highlight | -0.024 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.44e+00 |
Pos logit contributions | |
ost | +0.233 |
ric | +0.220 |
peac | +0.210 |
highlight | +0.207 |
moral | +0.205 |
Neg logit contributions | |
iddled | -0.493 |
aback | -0.434 |
bench | -0.428 |
uster | -0.426 |
ocated | -0.420 |
Token | ay |
Token ID | 323 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
work | +0.557 |
ric | +0.555 |
ost | +0.504 |
je | +0.488 |
chim | +0.484 |
Neg logit contributions | |
iddled | -0.409 |
aback | -0.360 |
uster | -0.340 |
ocated | -0.318 |
wn | -0.315 |
Token | . |
Token ID | 13 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
iddled | +0.110 |
aback | +0.088 |
uster | +0.086 |
ocated | +0.082 |
bench | +0.080 |
Neg logit contributions | |
ost | -0.149 |
ric | -0.144 |
peac | -0.142 |
ostr | -0.139 |
rob | -0.138 |
Token | She |
Token ID | 1375 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
iddled | +0.005 |
aback | +0.004 |
tant | +0.004 |
bench | +0.004 |
uster | +0.004 |
Neg logit contributions | |
ost | -0.004 |
ric | -0.004 |
peac | -0.004 |
ostr | -0.004 |
moral | -0.004 |
Token | remembered |
Token ID | 12086 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
iddled | +0.010 |
aback | +0.008 |
uster | +0.008 |
tant | +0.008 |
ocated | +0.008 |
Neg logit contributions | |
ost | -0.009 |
ric | -0.009 |
work | -0.009 |
peac | -0.008 |
highlight | -0.008 |
Token | her |
Token ID | 607 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ric | +0.056 |
ost | +0.055 |
work | +0.054 |
highlight | +0.053 |
flaming | +0.052 |
Neg logit contributions | |
iddled | -0.036 |
aback | -0.030 |
uster | -0.024 |
ocated | -0.024 |
tant | -0.024 |
Token | backyard |
Token ID | 24296 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
ric | +0.019 |
ost | +0.019 |
work | +0.017 |
peac | +0.017 |
flaming | +0.017 |
Neg logit contributions | |
iddled | -0.023 |
aback | -0.020 |
uster | -0.020 |
tant | -0.019 |
bench | -0.019 |
Token | there |
Token ID | 612 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
iddled | +0.013 |
aback | +0.009 |
bench | +0.008 |
ocated | +0.007 |
anked | +0.007 |
Neg logit contributions | |
highlight | -0.026 |
ric | -0.026 |
ost | -0.026 |
peac | -0.025 |
flaming | -0.025 |
Token | was |
Token ID | 373 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
iddled | +0.084 |
aback | +0.073 |
ocated | +0.067 |
bench | +0.066 |
uster | +0.066 |
Neg logit contributions | |
ost | -0.053 |
ric | -0.051 |
peac | -0.048 |
ostr | -0.046 |
work | -0.045 |
Token | an |
Token ID | 281 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.003 |
ocated | +0.003 |
bench | +0.003 |
tant | +0.003 |
Neg logit contributions | |
ric | -0.007 |
ost | -0.007 |
work | -0.007 |
peac | -0.007 |
orse | -0.007 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.10e+00 |
Pos logit contributions | |
iddled | +0.034 |
uster | +0.029 |
aback | +0.028 |
tant | +0.027 |
bench | +0.027 |
Neg logit contributions | |
ric | -0.041 |
work | -0.034 |
highlight | -0.034 |
fly | -0.033 |
ost | -0.033 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.42e+00 |
Pos logit contributions | |
ost | +0.257 |
ric | +0.252 |
highlight | +0.241 |
moral | +0.230 |
peac | +0.229 |
Neg logit contributions | |
iddled | -0.443 |
bench | -0.391 |
aback | -0.388 |
uster | -0.381 |
tant | -0.377 |
Token | ay |
Token ID | 323 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
work | +0.589 |
ric | +0.586 |
ost | +0.563 |
peac | +0.537 |
ostr | +0.532 |
Neg logit contributions | |
iddled | -0.383 |
aback | -0.341 |
uster | -0.309 |
ocated | -0.296 |
tant | -0.291 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
iddled | +0.106 |
uster | +0.081 |
aback | +0.079 |
bench | +0.078 |
ocated | +0.075 |
Neg logit contributions | |
ost | -0.157 |
ric | -0.156 |
highlight | -0.150 |
rob | -0.146 |
peac | -0.146 |
Token | It |
Token ID | 632 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ost | +0.006 |
ric | +0.006 |
peac | +0.006 |
ostr | +0.006 |
work | +0.006 |
Neg logit contributions | |
iddled | -0.007 |
aback | -0.006 |
uster | -0.005 |
bench | -0.005 |
ocated | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | -8.07e-02 |
Pos logit contributions | |
iddled | +0.077 |
aback | +0.065 |
bench | +0.061 |
tant | +0.059 |
ocated | +0.058 |
Neg logit contributions | |
ost | -0.067 |
ric | -0.067 |
highlight | -0.064 |
peac | -0.061 |
work | -0.060 |
Token | the |
Token ID | 262 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
iddled | +0.005 |
aback | +0.005 |
uster | +0.004 |
bench | +0.004 |
tant | +0.004 |
Neg logit contributions | |
ost | -0.005 |
ric | -0.004 |
peac | -0.004 |
ostr | -0.004 |
work | -0.004 |
Token | He |
Token ID | 679 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
ost | +0.004 |
ric | +0.004 |
ostr | +0.003 |
peac | +0.003 |
flaming | +0.003 |
Neg logit contributions | |
iddled | -0.005 |
aback | -0.005 |
uster | -0.004 |
tant | -0.004 |
ocated | -0.004 |
Token | picks |
Token ID | 11103 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
iddled | +0.012 |
aback | +0.011 |
uster | +0.011 |
tant | +0.010 |
bench | +0.010 |
Neg logit contributions | |
ost | -0.010 |
ric | -0.010 |
ostr | -0.009 |
peac | -0.009 |
moral | -0.009 |
Token | up |
Token ID | 510 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
ost | +0.066 |
ric | +0.063 |
ostr | +0.062 |
work | +0.061 |
rob | +0.060 |
Neg logit contributions | |
iddled | -0.068 |
uster | -0.062 |
aback | -0.058 |
bench | -0.057 |
tant | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
ost | +0.082 |
ostr | +0.074 |
ric | +0.074 |
peac | +0.074 |
flaming | +0.072 |
Neg logit contributions | |
iddled | -0.115 |
uster | -0.104 |
aback | -0.100 |
ocated | -0.099 |
bench | -0.095 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.27e+00 |
Pos logit contributions | |
ost | +0.023 |
ric | +0.023 |
highlight | +0.022 |
peac | +0.021 |
work | +0.021 |
Neg logit contributions | |
iddled | -0.034 |
aback | -0.031 |
tant | -0.029 |
uster | -0.028 |
bench | -0.027 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.41e+00 |
Pos logit contributions | |
ost | +0.195 |
ric | +0.179 |
peac | +0.163 |
ostr | +0.159 |
highlight | +0.157 |
Neg logit contributions | |
iddled | -0.563 |
aback | -0.516 |
uster | -0.498 |
tant | -0.483 |
bench | -0.483 |
Token | ay |
Token ID | 323 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
work | +0.536 |
ric | +0.514 |
je | +0.454 |
ost | +0.448 |
chim | +0.435 |
Neg logit contributions | |
iddled | -0.430 |
aback | -0.386 |
uster | -0.369 |
ocated | -0.358 |
wn | -0.349 |
Token | and |
Token ID | 290 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
iddled | +0.107 |
aback | +0.091 |
uster | +0.085 |
bench | +0.080 |
tant | +0.079 |
Neg logit contributions | |
ost | -0.135 |
ric | -0.130 |
peac | -0.126 |
ostr | -0.125 |
highlight | -0.124 |
Token | shakes |
Token ID | 29959 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
iddled | +0.026 |
aback | +0.022 |
uster | +0.021 |
tant | +0.020 |
bench | +0.020 |
Neg logit contributions | |
ost | -0.029 |
ric | -0.028 |
ostr | -0.027 |
peac | -0.027 |
work | -0.026 |
Token | it |
Token ID | 340 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
iddled | +0.126 |
uster | +0.117 |
aback | +0.111 |
ocated | +0.106 |
bench | +0.105 |
Neg logit contributions | |
ric | -0.109 |
ost | -0.109 |
ostr | -0.099 |
moral | -0.098 |
flaming | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
iddled | +0.059 |
uster | +0.053 |
aback | +0.051 |
tant | +0.047 |
ocated | +0.047 |
Neg logit contributions | |
ost | -0.038 |
work | -0.037 |
moral | -0.037 |
ric | -0.035 |
ostr | -0.035 |
Token | around |
Token ID | 1088 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
ost | +0.062 |
ostr | +0.058 |
peac | +0.057 |
work | +0.057 |
ric | +0.057 |
Neg logit contributions | |
iddled | -0.043 |
uster | -0.036 |
ocated | -0.034 |
bench | -0.034 |
aback | -0.034 |
Token | town |
Token ID | 3240 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
iddled | +0.116 |
aback | +0.103 |
uster | +0.089 |
tant | +0.088 |
bench | +0.086 |
Neg logit contributions | |
ost | -0.159 |
ric | -0.158 |
highlight | -0.148 |
ostr | -0.147 |
peac | -0.146 |
Token | on |
Token ID | 319 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
iddled | +0.005 |
aback | +0.004 |
bench | +0.004 |
away | +0.003 |
uster | +0.003 |
Neg logit contributions | |
ost | -0.009 |
ric | -0.009 |
highlight | -0.009 |
moral | -0.009 |
peac | -0.008 |
Token | his |
Token ID | 465 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
ost | +0.063 |
ric | +0.062 |
ostr | +0.059 |
peac | +0.059 |
highlight | +0.058 |
Neg logit contributions | |
iddled | -0.041 |
aback | -0.037 |
tant | -0.032 |
uster | -0.031 |
bench | -0.030 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.27e+00 |
Pos logit contributions | |
breath | +0.054 |
away | +0.052 |
iddled | +0.049 |
tant | +0.048 |
apr | +0.048 |
Neg logit contributions | |
ric | -0.087 |
riage | -0.080 |
ost | -0.080 |
rob | -0.079 |
oral | -0.079 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.41e+00 |
Pos logit contributions | |
ost | +0.250 |
highlight | +0.224 |
ric | +0.222 |
moral | +0.221 |
rob | +0.214 |
Neg logit contributions | |
iddled | -0.486 |
aback | -0.438 |
bench | -0.427 |
uster | -0.426 |
tant | -0.418 |
Token | ay |
Token ID | 323 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
ric | +0.604 |
work | +0.575 |
ost | +0.560 |
peac | +0.536 |
je | +0.531 |
Neg logit contributions | |
iddled | -0.380 |
aback | -0.336 |
uster | -0.310 |
ocated | -0.300 |
bench | -0.289 |
Token | , |
Token ID | 11 |
Feature activation | +8.99e-02 |
Pos logit contributions | |
iddled | +0.096 |
aback | +0.079 |
uster | +0.075 |
bench | +0.072 |
tant | +0.065 |
Neg logit contributions | |
ost | -0.157 |
ric | -0.145 |
peac | -0.145 |
highlight | -0.145 |
ostr | -0.144 |
Token | z |
Token ID | 1976 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
ost | +0.005 |
ric | +0.005 |
moral | +0.005 |
ostr | +0.005 |
peac | +0.005 |
Neg logit contributions | |
iddled | -0.006 |
aback | -0.005 |
uster | -0.005 |
bench | -0.004 |
ocated | -0.004 |
Token | ipping |
Token ID | 4501 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ost | +0.092 |
ric | +0.090 |
peac | +0.086 |
ostr | +0.085 |
work | +0.084 |
Neg logit contributions | |
iddled | -0.071 |
aback | -0.062 |
uster | -0.057 |
bench | -0.055 |
tant | -0.055 |
Token | everywhere |
Token ID | 8347 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
iddled | +0.060 |
uster | +0.050 |
aback | +0.049 |
ocated | +0.046 |
bench | +0.046 |
Neg logit contributions | |
ost | -0.082 |
ric | -0.077 |
ostr | -0.077 |
peac | -0.076 |
work | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | +9.50e-02 |
Pos logit contributions | |
ost | +0.081 |
ostr | +0.074 |
ric | +0.072 |
moral | +0.072 |
peac | +0.071 |
Neg logit contributions | |
iddled | -0.108 |
uster | -0.097 |
aback | -0.095 |
tant | -0.089 |
ocated | -0.089 |
Token | he |
Token ID | 339 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
ost | +0.004 |
moral | +0.004 |
ostr | +0.004 |
highlight | +0.004 |
work | +0.004 |
Neg logit contributions | |
iddled | -0.007 |
uster | -0.007 |
aback | -0.006 |
ocated | -0.006 |
bench | -0.006 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
iddled | +0.011 |
aback | +0.009 |
tant | +0.009 |
bench | +0.008 |
ocated | +0.008 |
Neg logit contributions | |
ric | -0.015 |
ost | -0.015 |
work | -0.014 |
highlight | -0.013 |
peac | -0.013 |
Token | an |
Token ID | 281 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
ost | +0.001 |
ric | +0.001 |
peac | +0.001 |
highlight | +0.001 |
work | +0.001 |
Neg logit contributions | |
iddled | -0.001 |
aback | -0.001 |
uster | -0.001 |
tant | -0.001 |
bench | -0.001 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.16e+00 |
Pos logit contributions | |
iddled | +0.037 |
aback | +0.032 |
uster | +0.032 |
tant | +0.030 |
bench | +0.029 |
Neg logit contributions | |
ric | -0.032 |
ost | -0.032 |
peac | -0.029 |
highlight | -0.029 |
work | -0.029 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.41e+00 |
Pos logit contributions | |
ost | +0.237 |
ric | +0.225 |
highlight | +0.214 |
peac | +0.208 |
moral | +0.204 |
Neg logit contributions | |
iddled | -0.485 |
aback | -0.424 |
bench | -0.422 |
uster | -0.416 |
tant | -0.411 |
Token | ay |
Token ID | 323 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
work | +0.560 |
ric | +0.554 |
ost | +0.511 |
je | +0.496 |
peac | +0.485 |
Neg logit contributions | |
iddled | -0.409 |
aback | -0.371 |
uster | -0.341 |
ocated | -0.318 |
wn | -0.314 |
Token | on |
Token ID | 319 |
Feature activation | +7.74e-01 |
Pos logit contributions | |
iddled | +0.105 |
aback | +0.082 |
uster | +0.080 |
bench | +0.077 |
ocated | +0.077 |
Neg logit contributions | |
ost | -0.154 |
ric | -0.152 |
peac | -0.144 |
highlight | -0.144 |
rob | -0.142 |
Token | the |
Token ID | 262 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
ost | +0.042 |
ric | +0.041 |
ostr | +0.040 |
peac | +0.038 |
zoo | +0.038 |
Neg logit contributions | |
iddled | -0.053 |
aback | -0.046 |
uster | -0.045 |
ocated | -0.042 |
bench | -0.042 |
Token | table |
Token ID | 3084 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
ric | +0.021 |
ost | +0.021 |
peac | +0.020 |
orse | +0.020 |
work | +0.019 |
Neg logit contributions | |
iddled | -0.016 |
uster | -0.015 |
aback | -0.015 |
bench | -0.015 |
ocated | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
iddled | +0.115 |
aback | +0.097 |
uster | +0.092 |
bench | +0.091 |
tant | +0.087 |
Neg logit contributions | |
ost | -0.141 |
ric | -0.136 |
peac | -0.131 |
ostr | -0.130 |
rob | -0.126 |
Token | tried |
Token ID | 3088 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
iddled | +0.001 |
aback | +0.001 |
uster | +0.001 |
tant | +0.001 |
ocated | +0.001 |
Neg logit contributions | |
ost | -0.001 |
ric | -0.001 |
work | -0.001 |
highlight | -0.001 |
peac | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
iddled | +0.081 |
aback | +0.074 |
uster | +0.062 |
tant | +0.061 |
bench | +0.059 |
Neg logit contributions | |
ost | -0.124 |
ric | -0.118 |
ostr | -0.118 |
peac | -0.115 |
work | -0.112 |
Token | fix |
Token ID | 4259 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
iddled | +0.005 |
aback | +0.005 |
uster | +0.005 |
bench | +0.005 |
tant | +0.004 |
Neg logit contributions | |
ost | -0.004 |
ric | -0.004 |
ostr | -0.004 |
rob | -0.004 |
moral | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
iddled | +0.026 |
uster | +0.022 |
aback | +0.022 |
bench | +0.020 |
tant | +0.020 |
Neg logit contributions | |
ost | -0.028 |
ric | -0.027 |
ostr | -0.026 |
peac | -0.025 |
flaming | -0.024 |
Token | ash |
Token ID | 12530 |
Feature activation | +6.29e+00 |
Pos logit contributions | |
ric | +0.027 |
highlight | +0.027 |
orse | +0.027 |
Whale | +0.026 |
ost | +0.026 |
Neg logit contributions | |
iddled | -0.023 |
aback | -0.023 |
tant | -0.022 |
ocated | -0.021 |
away | -0.021 |
Token | tr |
Token ID | 2213 |
Feature activation | +8.41e+00 |
Pos logit contributions | |
ost | +0.190 |
ric | +0.174 |
peac | +0.159 |
ostr | +0.157 |
highlight | +0.148 |
Neg logit contributions | |
iddled | -0.570 |
aback | -0.524 |
uster | -0.507 |
tant | -0.490 |
bench | -0.489 |
Token | ay |
Token ID | 323 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
ric | +0.608 |
work | +0.580 |
ost | +0.573 |
peac | +0.539 |
ostr | +0.539 |
Neg logit contributions | |
iddled | -0.385 |
aback | -0.326 |
uster | -0.307 |
bench | -0.282 |
ocated | -0.277 |
Token | , |
Token ID | 11 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
iddled | +0.117 |
aback | +0.102 |
uster | +0.096 |
tant | +0.090 |
bench | +0.090 |
Neg logit contributions | |
ost | -0.131 |
ric | -0.125 |
peac | -0.121 |
ostr | -0.120 |
highlight | -0.117 |
Token | but |
Token ID | 475 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
ost | +0.014 |
ric | +0.014 |
ostr | +0.013 |
peac | +0.013 |
work | +0.013 |
Neg logit contributions | |
iddled | -0.008 |
aback | -0.007 |
uster | -0.006 |
tant | -0.006 |
bench | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
iddled | +0.047 |
aback | +0.040 |
ocated | +0.031 |
away | +0.028 |
uster | +0.027 |
Neg logit contributions | |
ric | -0.118 |
ost | -0.117 |
flaming | -0.113 |
peac | -0.112 |
zoo | -0.111 |
Token | was |
Token ID | 373 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
iddled | +0.061 |
aback | +0.055 |
ocated | +0.050 |
tant | +0.049 |
uster | +0.049 |
Neg logit contributions | |
ost | -0.053 |
ric | -0.051 |
peac | -0.048 |
zoo | -0.047 |
highlight | -0.047 |
Token | too |
Token ID | 1165 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
ost | +0.128 |
ric | +0.123 |
riage | +0.121 |
highlight | +0.120 |
North | +0.119 |
Neg logit contributions | |
iddled | -0.048 |
aback | -0.043 |
away | -0.040 |
ly | -0.031 |
acks | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -9.36e-02 |
Pos logit contributions | |
iddled | +0.003 |
aback | +0.003 |
uster | +0.003 |
bench | +0.002 |
tant | +0.002 |
Neg logit contributions | |
ost | -0.006 |
ric | -0.006 |
highlight | -0.005 |
work | -0.005 |
peac | -0.005 |
Token | he |
Token ID | 339 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
iddled | +0.005 |
aback | +0.004 |
tant | +0.004 |
uster | +0.004 |
ocated | +0.003 |
Neg logit contributions | |
ric | -0.006 |
ost | -0.006 |
highlight | -0.006 |
work | -0.006 |
North | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
iddled | +0.014 |
aback | +0.012 |
tant | +0.010 |
uster | +0.010 |
zip | +0.010 |
Neg logit contributions | |
ost | -0.016 |
ric | -0.016 |
highlight | -0.015 |
work | -0.015 |
zoo | -0.015 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
iddled | +0.006 |
aback | +0.005 |
uster | +0.004 |
bench | +0.004 |
tant | +0.004 |
Neg logit contributions | |
ost | -0.007 |
ric | -0.007 |
work | -0.007 |
oral | -0.007 |
peac | -0.007 |
Token | aback |
Token ID | 50147 |
Feature activation | -8.41e+00 |
Pos logit contributions | |
iddled | +0.104 |
uster | +0.088 |
bench | +0.085 |
ocated | +0.081 |
tant | +0.077 |
Neg logit contributions | |
ost | -0.103 |
rob | -0.097 |
ostr | -0.096 |
peac | -0.096 |
moral | -0.094 |
Token | by |
Token ID | 416 |
Feature activation | +9.84e-01 |
Pos logit contributions | |
iddled | +0.492 |
aback | +0.424 |
uster | +0.406 |
tant | +0.379 |
bench | +0.379 |
Neg logit contributions | |
ost | -0.532 |
ric | -0.512 |
ostr | -0.491 |
peac | -0.488 |
work | -0.476 |
Token | how |
Token ID | 703 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
ost | +0.075 |
highlight | +0.074 |
oral | +0.072 |
str | +0.070 |
work | +0.070 |
Neg logit contributions | |
iddled | -0.037 |
aback | -0.032 |
tant | -0.025 |
uster | -0.025 |
breath | -0.024 |
Token | attractive |
Token ID | 10966 |
Feature activation | +9.47e-01 |
Pos logit contributions | |
iddled | +0.101 |
aback | +0.090 |
uster | +0.078 |
tant | +0.078 |
bench | +0.076 |
Neg logit contributions | |
ric | -0.112 |
ost | -0.112 |
work | -0.105 |
ostr | -0.104 |
peac | -0.104 |
Token | her |
Token ID | 607 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
ric | +0.072 |
ost | +0.071 |
work | +0.067 |
peac | +0.067 |
highlight | +0.066 |
Neg logit contributions | |
iddled | -0.042 |
aback | -0.035 |
uster | -0.031 |
tant | -0.029 |
bench | -0.028 |
Token | wings |
Token ID | 12098 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
iddled | +0.064 |
aback | +0.059 |
knot | +0.056 |
breath | +0.055 |
away | +0.053 |
Neg logit contributions | |
ric | -0.137 |
ost | -0.136 |
highlight | -0.125 |
zoo | -0.124 |
North | -0.123 |
Token | The |
Token ID | 383 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
iddled | +0.000 |
aback | +0.000 |
uster | +0.000 |
tant | +0.000 |
ocated | +0.000 |
Neg logit contributions | |
ost | -0.000 |
peac | -0.000 |
work | -0.000 |
ric | -0.000 |
ostr | -0.000 |
Token | little |
Token ID | 1310 |
Feature activation | -7.49e-02 |
Pos logit contributions | |
ric | +0.186 |
ost | +0.178 |
work | +0.159 |
peac | +0.159 |
fly | +0.155 |
Neg logit contributions | |
tant | -0.092 |
iddled | -0.084 |
aback | -0.078 |
uster | -0.076 |
ly | -0.072 |
Token | boy |
Token ID | 2933 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.004 |
uster | +0.004 |
tant | +0.004 |
bench | +0.003 |
Neg logit contributions | |
ric | -0.005 |
ost | -0.005 |
peac | -0.004 |
highlight | -0.004 |
work | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
iddled | +0.041 |
aback | +0.039 |
away | +0.034 |
tant | +0.033 |
shoved | +0.032 |
Neg logit contributions | |
ric | -0.064 |
ost | -0.063 |
highlight | -0.061 |
peac | -0.057 |
North | -0.057 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
ost | +0.013 |
ric | +0.013 |
peac | +0.013 |
ostr | +0.013 |
work | +0.013 |
Neg logit contributions | |
iddled | -0.011 |
aback | -0.010 |
away | -0.008 |
tant | -0.008 |
bench | -0.008 |
Token | aback |
Token ID | 50147 |
Feature activation | -8.25e+00 |
Pos logit contributions | |
iddled | +0.084 |
uster | +0.068 |
bench | +0.066 |
ocated | +0.065 |
iddling | +0.058 |
Neg logit contributions | |
ost | -0.116 |
rob | -0.110 |
moral | -0.110 |
ostr | -0.109 |
happiest | -0.108 |
Token | and |
Token ID | 290 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
iddled | +0.472 |
aback | +0.402 |
uster | +0.386 |
bench | +0.361 |
tant | +0.359 |
Neg logit contributions | |
ost | -0.534 |
ric | -0.514 |
ostr | -0.495 |
peac | -0.489 |
work | -0.478 |
Token | felt |
Token ID | 2936 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
iddled | +0.009 |
aback | +0.009 |
tant | +0.007 |
uster | +0.007 |
ocated | +0.007 |
Neg logit contributions | |
ric | -0.012 |
ost | -0.011 |
peac | -0.010 |
highlight | -0.010 |
work | -0.010 |
Token | guilty |
Token ID | 6717 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
iddled | +0.023 |
aback | +0.022 |
uster | +0.020 |
tant | +0.019 |
bench | +0.019 |
Neg logit contributions | |
ost | -0.019 |
ric | -0.019 |
ostr | -0.018 |
peac | -0.018 |
work | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
iddled | +0.001 |
aback | +0.001 |
uster | +0.001 |
tant | +0.001 |
bench | +0.001 |
Neg logit contributions | |
ost | -0.001 |
ric | -0.001 |
ostr | -0.001 |
peac | -0.001 |
work | -0.001 |
Token | He |
Token ID | 679 |
Feature activation | -6.85e-02 |
Pos logit contributions | |
ost | +0.004 |
peac | +0.004 |
ric | +0.004 |
work | +0.004 |
ostr | +0.004 |
Neg logit contributions | |
iddled | -0.003 |
aback | -0.003 |
uster | -0.002 |
tant | -0.002 |
bench | -0.002 |
Token | The |
Token ID | 464 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
iddled | +0.021 |
aback | +0.018 |
uster | +0.016 |
bench | +0.015 |
ocated | +0.015 |
Neg logit contributions | |
ost | -0.024 |
ric | -0.023 |
ostr | -0.023 |
work | -0.022 |
peac | -0.021 |
Token | man |
Token ID | 582 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
ric | +0.209 |
ost | +0.179 |
fly | +0.177 |
work | +0.175 |
highlight | +0.167 |
Neg logit contributions | |
iddled | -0.107 |
bench | -0.100 |
uster | -0.092 |
aback | -0.088 |
eal | -0.077 |
Token | was |
Token ID | 373 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
iddled | +0.034 |
aback | +0.030 |
away | +0.025 |
led | +0.024 |
shoved | +0.024 |
Neg logit contributions | |
ric | -0.053 |
ost | -0.048 |
highlight | -0.048 |
North | -0.046 |
ich | -0.044 |
Token | so |
Token ID | 523 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
ost | +0.008 |
ric | +0.008 |
peac | +0.008 |
ostr | +0.008 |
work | +0.008 |
Neg logit contributions | |
iddled | -0.007 |
aback | -0.007 |
bench | -0.006 |
uster | -0.006 |
wn | -0.005 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
ric | +0.001 |
ost | +0.001 |
peac | +0.001 |
ostr | +0.001 |
zoo | +0.001 |
Neg logit contributions | |
iddled | -0.001 |
aback | -0.001 |
uster | -0.001 |
bench | -0.001 |
tant | -0.001 |
Token | aback |
Token ID | 50147 |
Feature activation | -8.24e+00 |
Pos logit contributions | |
iddled | +0.102 |
uster | +0.088 |
ocated | +0.084 |
bench | +0.081 |
irms | +0.080 |
Neg logit contributions | |
ost | -0.103 |
work | -0.103 |
rob | -0.096 |
ostr | -0.096 |
moral | -0.096 |
Token | by |
Token ID | 416 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
iddled | +0.467 |
aback | +0.395 |
uster | +0.384 |
bench | +0.357 |
ocated | +0.356 |
Neg logit contributions | |
ost | -0.537 |
ric | -0.515 |
ostr | -0.498 |
peac | -0.493 |
work | -0.485 |
Token | Maria |
Token ID | 14200 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
ost | +0.074 |
ric | +0.071 |
highlight | +0.070 |
ostr | +0.070 |
North | +0.068 |
Neg logit contributions | |
iddled | -0.051 |
aback | -0.045 |
uster | -0.036 |
fully | -0.035 |
tant | -0.034 |
Token | 's |
Token ID | 338 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
aback | +0.003 |
iddled | +0.002 |
uster | +0.002 |
away | +0.002 |
bench | +0.002 |
Neg logit contributions | |
ost | -0.003 |
ric | -0.003 |
peac | -0.002 |
flaming | -0.002 |
zoo | -0.002 |
Token | charming |
Token ID | 23332 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
ric | +0.131 |
ost | +0.126 |
str | +0.117 |
peac | +0.114 |
highlight | +0.113 |
Neg logit contributions | |
iddled | -0.098 |
aback | -0.095 |
uster | -0.087 |
bench | -0.082 |
tant | -0.082 |
Token | request |
Token ID | 2581 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
iddled | +0.026 |
aback | +0.023 |
uster | +0.023 |
tant | +0.022 |
bench | +0.021 |
Neg logit contributions | |
ric | -0.025 |
ost | -0.024 |
highlight | -0.022 |
work | -0.022 |
peac | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
iddled | +0.003 |
aback | +0.002 |
uster | +0.002 |
tant | +0.002 |
ocated | +0.002 |
Neg logit contributions | |
ost | -0.003 |
ric | -0.003 |
ostr | -0.003 |
peac | -0.003 |
work | -0.003 |
Token | entrance |
Token ID | 10384 |
Feature activation | -3.85e-01 |
Pos logit contributions | |
ric | +0.017 |
work | +0.016 |
ost | +0.016 |
highlight | +0.016 |
orse | +0.016 |
Neg logit contributions | |
aback | -0.015 |
iddled | -0.014 |
tant | -0.012 |
away | -0.012 |
ly | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
iddled | +0.019 |
aback | +0.017 |
uster | +0.015 |
ocated | +0.014 |
bench | +0.014 |
Neg logit contributions | |
ost | -0.027 |
ric | -0.027 |
peac | -0.025 |
ostr | -0.025 |
flaming | -0.025 |
Token | was |
Token ID | 373 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
iddled | +0.020 |
aback | +0.018 |
uster | +0.016 |
ocated | +0.015 |
tant | +0.015 |
Neg logit contributions | |
ost | -0.022 |
ric | -0.022 |
ostr | -0.020 |
peac | -0.020 |
moral | -0.020 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ost | +0.001 |
ric | +0.001 |
peac | +0.001 |
ostr | +0.001 |
work | +0.001 |
Neg logit contributions | |
iddled | -0.001 |
aback | -0.001 |
uster | -0.001 |
ocated | -0.001 |
bench | -0.001 |
Token | aback |
Token ID | 50147 |
Feature activation | -8.24e+00 |
Pos logit contributions | |
iddled | +0.115 |
bench | +0.106 |
uster | +0.104 |
tant | +0.101 |
arl | +0.098 |
Neg logit contributions | |
ost | -0.097 |
rob | -0.089 |
guardian | -0.087 |
ostr | -0.085 |
peac | -0.085 |
Token | by |
Token ID | 416 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
iddled | +0.443 |
aback | +0.394 |
uster | +0.364 |
tant | +0.345 |
ocated | +0.334 |
Neg logit contributions | |
ost | -0.538 |
ric | -0.532 |
peac | -0.509 |
ostr | -0.498 |
work | -0.495 |
Token | how |
Token ID | 703 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ric | +0.077 |
ost | +0.076 |
ostr | +0.073 |
highlight | +0.073 |
work | +0.073 |
Neg logit contributions | |
iddled | -0.044 |
aback | -0.039 |
ocated | -0.030 |
uster | -0.030 |
tant | -0.028 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +8.59e-01 |
Pos logit contributions | |
iddled | +0.087 |
aback | +0.076 |
uster | +0.066 |
ocated | +0.064 |
tant | +0.063 |
Neg logit contributions | |
ric | -0.113 |
ost | -0.110 |
ostr | -0.104 |
work | -0.104 |
peac | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
ric | +0.063 |
ost | +0.063 |
peac | +0.059 |
ostr | +0.059 |
work | +0.059 |
Neg logit contributions | |
iddled | -0.040 |
aback | -0.033 |
uster | -0.029 |
tant | -0.027 |
ocated | -0.027 |
Token | enchanted |
Token ID | 42282 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ric | +0.018 |
work | +0.015 |
orse | +0.014 |
highlight | +0.014 |
fly | +0.014 |
Neg logit contributions | |
aback | -0.013 |
breath | -0.012 |
iddled | -0.012 |
piles | -0.012 |
ly | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
iddled | +0.030 |
aback | +0.027 |
uster | +0.026 |
ocated | +0.025 |
tant | +0.024 |
Neg logit contributions | |
ost | -0.018 |
ric | -0.018 |
ostr | -0.016 |
peac | -0.016 |
moral | -0.016 |
Token | His |
Token ID | 6653 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
iddled | +0.017 |
aback | +0.014 |
uster | +0.013 |
tant | +0.013 |
bench | +0.012 |
Neg logit contributions | |
ost | -0.026 |
ric | -0.025 |
peac | -0.024 |
ostr | -0.024 |
work | -0.024 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
ric | +0.050 |
ost | +0.048 |
peac | +0.044 |
work | +0.043 |
rob | +0.042 |
Neg logit contributions | |
tant | -0.036 |
iddled | -0.034 |
bench | -0.030 |
aback | -0.029 |
uster | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
ost | +0.142 |
ric | +0.140 |
highlight | +0.132 |
peac | +0.130 |
North | +0.129 |
Neg logit contributions | |
iddled | -0.115 |
aback | -0.098 |
uster | -0.087 |
ocated | -0.085 |
tant | -0.084 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
ost | +0.006 |
ric | +0.006 |
ostr | +0.006 |
peac | +0.006 |
work | +0.006 |
Neg logit contributions | |
iddled | -0.009 |
aback | -0.008 |
uster | -0.008 |
tant | -0.008 |
bench | -0.008 |
Token | aback |
Token ID | 50147 |
Feature activation | -8.20e+00 |
Pos logit contributions | |
iddled | +0.142 |
ows | +0.125 |
uster | +0.122 |
ocated | +0.122 |
iddling | +0.120 |
Neg logit contributions | |
ost | -0.060 |
je | -0.059 |
moral | -0.056 |
ostr | -0.055 |
hipp | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +9.50e-02 |
Pos logit contributions | |
iddled | +0.543 |
aback | +0.442 |
uster | +0.432 |
ocated | +0.429 |
wn | +0.422 |
Neg logit contributions | |
ost | -0.482 |
ric | -0.461 |
ostr | -0.449 |
moral | -0.414 |
peac | -0.411 |
Token | She |
Token ID | 1375 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
ost | +0.005 |
ric | +0.005 |
ostr | +0.004 |
zoo | +0.004 |
moral | +0.004 |
Neg logit contributions | |
iddled | -0.006 |
aback | -0.006 |
uster | -0.006 |
tant | -0.005 |
wn | -0.005 |
Token | said |
Token ID | 531 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
iddled | +0.002 |
aback | +0.002 |
uster | +0.002 |
tant | +0.002 |
ocated | +0.002 |
Neg logit contributions | |
ost | -0.003 |
ric | -0.003 |
highlight | -0.003 |
peac | -0.003 |
work | -0.002 |
Token | calmly |
Token ID | 30180 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
iddled | +0.002 |
aback | +0.002 |
uster | +0.002 |
tant | +0.002 |
ocated | +0.002 |
Neg logit contributions | |
ost | -0.003 |
ric | -0.003 |
peac | -0.003 |
highlight | -0.003 |
ostr | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -1.18e-04 |
Pos logit contributions | |
iddled | +0.031 |
aback | +0.026 |
uster | +0.024 |
bench | +0.022 |
ocated | +0.021 |
Neg logit contributions | |
ost | -0.055 |
ric | -0.054 |
highlight | -0.053 |
peac | -0.051 |
ostr | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
iddled | +0.007 |
aback | +0.006 |
ocated | +0.006 |
uster | +0.006 |
tant | +0.005 |
Neg logit contributions | |
ric | -0.009 |
ost | -0.009 |
highlight | -0.008 |
work | -0.008 |
ostr | -0.008 |
Token | little |
Token ID | 1310 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
ric | +0.034 |
ost | +0.031 |
orse | +0.030 |
work | +0.030 |
peac | +0.029 |
Neg logit contributions | |
iddled | -0.026 |
tant | -0.024 |
aback | -0.023 |
uster | -0.022 |
ocated | -0.021 |
Token | girl |
Token ID | 2576 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
ric | +0.013 |
ost | +0.013 |
peac | +0.012 |
highlight | +0.012 |
work | +0.012 |
Neg logit contributions | |
iddled | -0.013 |
aback | -0.012 |
uster | -0.011 |
tant | -0.011 |
ocated | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
iddled | +0.021 |
aback | +0.018 |
ocated | +0.015 |
uster | +0.015 |
tant | +0.015 |
Neg logit contributions | |
ric | -0.026 |
ost | -0.026 |
highlight | -0.025 |
work | -0.025 |
moral | -0.024 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
ost | +0.014 |
ric | +0.014 |
peac | +0.013 |
ostr | +0.013 |
work | +0.013 |
Neg logit contributions | |
iddled | -0.016 |
aback | -0.014 |
uster | -0.012 |
ocated | -0.012 |
bench | -0.012 |
Token | aback |
Token ID | 50147 |
Feature activation | -8.19e+00 |
Pos logit contributions | |
iddled | +0.091 |
uster | +0.079 |
tant | +0.078 |
bench | +0.076 |
irms | +0.073 |
Neg logit contributions | |
ost | -0.086 |
� | -0.080 |
guardian | -0.078 |
ostr | -0.077 |
peac | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
iddled | +0.517 |
aback | +0.436 |
uster | +0.427 |
bench | +0.406 |
ocated | +0.401 |
Neg logit contributions | |
ost | -0.491 |
ric | -0.463 |
ostr | -0.452 |
peac | -0.437 |
work | -0.431 |
Token | Inside |
Token ID | 14384 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
ost | +0.007 |
ric | +0.006 |
ostr | +0.006 |
peac | +0.006 |
work | +0.006 |
Neg logit contributions | |
iddled | -0.006 |
aback | -0.005 |
uster | -0.005 |
ocated | -0.004 |
bench | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
iddled | +0.176 |
aback | +0.154 |
uster | +0.147 |
tant | +0.139 |
bench | +0.139 |
Neg logit contributions | |
ost | -0.161 |
ric | -0.155 |
ostr | -0.147 |
peac | -0.147 |
work | -0.143 |
Token | room |
Token ID | 2119 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
ric | +0.042 |
ost | +0.038 |
orse | +0.038 |
peac | +0.037 |
flaming | +0.036 |
Neg logit contributions | |
iddled | -0.023 |
aback | -0.022 |
tant | -0.021 |
ocated | -0.019 |
uster | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
iddled | +0.110 |
aback | +0.095 |
ocated | +0.084 |
bench | +0.084 |
uster | +0.084 |
Neg logit contributions | |
ric | -0.122 |
ost | -0.121 |
highlight | -0.112 |
ostr | -0.111 |
peac | -0.111 |
Token | The |
Token ID | 464 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
iddled | +0.022 |
aback | +0.019 |
uster | +0.018 |
bench | +0.017 |
ocated | +0.017 |
Neg logit contributions | |
ost | -0.021 |
ric | -0.021 |
ostr | -0.019 |
work | -0.019 |
peac | -0.019 |
Token | small |
Token ID | 1402 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
ric | +0.234 |
ost | +0.216 |
North | +0.198 |
highlight | +0.196 |
peac | +0.195 |
Neg logit contributions | |
iddled | -0.074 |
tant | -0.069 |
aback | -0.066 |
tug | -0.059 |
knot | -0.058 |
Token | boy |
Token ID | 2933 |
Feature activation | -8.09e-01 |
Pos logit contributions | |
iddled | +0.076 |
aback | +0.071 |
tant | +0.067 |
bench | +0.060 |
uster | +0.059 |
Neg logit contributions | |
ric | -0.101 |
ost | -0.097 |
highlight | -0.092 |
peac | -0.088 |
North | -0.087 |
Token | was |
Token ID | 373 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
iddled | +0.040 |
aback | +0.038 |
tant | +0.030 |
away | +0.030 |
eal | +0.030 |
Neg logit contributions | |
ost | -0.058 |
ric | -0.057 |
highlight | -0.054 |
peac | -0.051 |
North | -0.051 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
ost | +0.018 |
peac | +0.017 |
ric | +0.017 |
ostr | +0.017 |
work | +0.017 |
Neg logit contributions | |
iddled | -0.015 |
aback | -0.014 |
away | -0.011 |
bench | -0.011 |
uster | -0.010 |
Token | aback |
Token ID | 50147 |
Feature activation | -8.13e+00 |
Pos logit contributions | |
iddled | +0.066 |
uster | +0.048 |
bench | +0.045 |
ocated | +0.045 |
iddling | +0.040 |
Neg logit contributions | |
ost | -0.132 |
rob | -0.126 |
ostr | -0.125 |
moral | -0.123 |
work | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
iddled | +0.436 |
aback | +0.376 |
uster | +0.356 |
tant | +0.332 |
bench | +0.328 |
Neg logit contributions | |
ost | -0.547 |
ric | -0.530 |
peac | -0.509 |
ostr | -0.507 |
work | -0.496 |
Token | He |
Token ID | 679 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
ost | +0.009 |
peac | +0.008 |
ric | +0.008 |
ostr | +0.008 |
flaming | +0.008 |
Neg logit contributions | |
iddled | -0.005 |
aback | -0.004 |
uster | -0.004 |
bench | -0.003 |
ocated | -0.003 |
Token | thanked |
Token ID | 26280 |
Feature activation | -7.15e-01 |
Pos logit contributions | |
iddled | +0.002 |
aback | +0.002 |
tant | +0.002 |
uster | +0.002 |
ocated | +0.002 |
Neg logit contributions | |
ost | -0.003 |
ric | -0.003 |
highlight | -0.002 |
work | -0.002 |
peac | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
iddled | +0.027 |
aback | +0.025 |
uster | +0.020 |
fully | +0.018 |
tant | +0.018 |
Neg logit contributions | |
ric | -0.056 |
ost | -0.056 |
highlight | -0.055 |
ostr | -0.052 |
str | -0.052 |
Token | saw |
Token ID | 2497 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
ric | +0.032 |
ost | +0.028 |
orse | +0.026 |
highlight | +0.026 |
work | +0.025 |
Neg logit contributions | |
aback | -0.016 |
tant | -0.014 |
iddled | -0.013 |
bench | -0.013 |
away | -0.011 |
Token | Jane |
Token ID | 41083 |
Feature activation | -8.75e-04 |
Pos logit contributions | |
iddled | +0.005 |
aback | +0.004 |
uster | +0.004 |
bench | +0.004 |
tant | +0.004 |
Neg logit contributions | |
ost | -0.007 |
ric | -0.006 |
ostr | -0.006 |
peac | -0.006 |
work | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +4.87e-02 |
Pos logit contributions | |
iddled | +0.000 |
aback | +0.000 |
tant | +0.000 |
led | +0.000 |
ly | +0.000 |
Neg logit contributions | |
ost | -0.000 |
highlight | -0.000 |
ric | -0.000 |
zoo | -0.000 |
North | -0.000 |
Token | a |
Token ID | 257 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
ost | +0.003 |
ric | +0.003 |
ostr | +0.003 |
highlight | +0.003 |
peac | +0.003 |
Neg logit contributions | |
iddled | -0.003 |
aback | -0.003 |
uster | -0.002 |
bench | -0.002 |
tant | -0.002 |
Token | little |
Token ID | 1310 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
iddled | +0.017 |
aback | +0.016 |
tant | +0.016 |
uster | +0.015 |
bench | +0.014 |
Neg logit contributions | |
ric | -0.022 |
ost | -0.021 |
highlight | -0.019 |
work | -0.019 |
riage | -0.019 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
ric | +0.003 |
ost | +0.003 |
highlight | +0.003 |
moral | +0.003 |
North | +0.003 |
Neg logit contributions | |
iddled | -0.003 |
aback | -0.003 |
uster | -0.003 |
tant | -0.003 |
bench | -0.003 |
Token | aback |
Token ID | 50147 |
Feature activation | -8.12e+00 |
Pos logit contributions | |
iddled | +0.088 |
uster | +0.075 |
ocated | +0.072 |
bench | +0.069 |
irms | +0.068 |
Neg logit contributions | |
ost | -0.107 |
work | -0.102 |
rob | -0.100 |
peac | -0.099 |
ostr | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
iddled | +0.452 |
aback | +0.392 |
uster | +0.371 |
tant | +0.347 |
bench | +0.345 |
Neg logit contributions | |
ost | -0.532 |
ric | -0.516 |
peac | -0.493 |
ostr | -0.491 |
work | -0.480 |
Token | thought |
Token ID | 1807 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
ric | +0.020 |
ost | +0.020 |
work | +0.019 |
flaming | +0.018 |
ostr | +0.018 |
Neg logit contributions | |
iddled | -0.013 |
aback | -0.012 |
uster | -0.010 |
bench | -0.010 |
tant | -0.009 |
Token | for |
Token ID | 329 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
ost | +0.029 |
ric | +0.028 |
ostr | +0.027 |
peac | +0.027 |
work | +0.026 |
Neg logit contributions | |
iddled | -0.026 |
aback | -0.022 |
uster | -0.021 |
tant | -0.020 |
bench | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
iddled | +0.059 |
aback | +0.051 |
uster | +0.048 |
bench | +0.045 |
ocated | +0.045 |
Neg logit contributions | |
ost | -0.063 |
ric | -0.061 |
ostr | -0.058 |
peac | -0.058 |
work | -0.056 |
Token | moment |
Token ID | 2589 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
aback | +0.018 |
iddled | +0.017 |
tant | +0.017 |
uster | +0.016 |
bench | +0.015 |
Neg logit contributions | |
ric | -0.025 |
ost | -0.024 |
work | -0.021 |
North | -0.021 |
highlight | -0.021 |
Token | The |
Token ID | 383 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ost | +0.012 |
ric | +0.011 |
peac | +0.011 |
ostr | +0.011 |
highlight | +0.011 |
Neg logit contributions | |
iddled | -0.007 |
aback | -0.007 |
uster | -0.006 |
bench | -0.005 |
acks | -0.005 |
Token | bear |
Token ID | 6842 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
ric | +0.180 |
ost | +0.172 |
work | +0.166 |
highlight | +0.162 |
str | +0.156 |
Neg logit contributions | |
iddled | -0.101 |
aback | -0.092 |
uster | -0.089 |
tant | -0.083 |
knot | -0.080 |
Token | was |
Token ID | 373 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
ric | +0.032 |
highlight | +0.030 |
ost | +0.030 |
riage | +0.028 |
North | +0.028 |
Neg logit contributions | |
iddled | -0.020 |
aback | -0.017 |
led | -0.015 |
away | -0.014 |
tant | -0.014 |
Token | so |
Token ID | 523 |
Feature activation | +9.55e-03 |
Pos logit contributions | |
ost | +0.009 |
ric | +0.009 |
work | +0.009 |
peac | +0.009 |
ostr | +0.009 |
Neg logit contributions | |
iddled | -0.009 |
aback | -0.008 |
uster | -0.007 |
bench | -0.007 |
wn | -0.007 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
ric | +0.001 |
ost | +0.001 |
work | +0.001 |
peac | +0.001 |
highlight | +0.001 |
Neg logit contributions | |
iddled | -0.001 |
aback | -0.001 |
tant | -0.000 |
uster | -0.000 |
bench | -0.000 |
Token | aback |
Token ID | 50147 |
Feature activation | -8.07e+00 |
Pos logit contributions | |
iddled | +0.083 |
uster | +0.064 |
bench | +0.062 |
ocated | +0.060 |
iddling | +0.059 |
Neg logit contributions | |
ost | -0.109 |
ostr | -0.105 |
moral | -0.104 |
rob | -0.104 |
work | -0.102 |
Token | by |
Token ID | 416 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
iddled | +0.456 |
aback | +0.380 |
uster | +0.370 |
bench | +0.345 |
ocated | +0.345 |
Neg logit contributions | |
ost | -0.533 |
ric | -0.510 |
ostr | -0.496 |
peac | -0.486 |
work | -0.479 |
Token | such |
Token ID | 884 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
ost | +0.077 |
highlight | +0.072 |
ric | +0.072 |
peac | +0.071 |
ostr | +0.071 |
Neg logit contributions | |
iddled | -0.052 |
aback | -0.047 |
uster | -0.041 |
tant | -0.038 |
ocated | -0.036 |
Token | kindness |
Token ID | 23887 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
iddled | +0.113 |
aback | +0.100 |
uster | +0.094 |
tant | +0.089 |
bench | +0.088 |
Neg logit contributions | |
ost | -0.115 |
ric | -0.112 |
peac | -0.105 |
ostr | -0.105 |
work | -0.104 |
Token | it |
Token ID | 340 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ost | +0.069 |
ric | +0.067 |
peac | +0.064 |
ostr | +0.064 |
flaming | +0.062 |
Neg logit contributions | |
iddled | -0.054 |
aback | -0.048 |
uster | -0.045 |
tant | -0.041 |
bench | -0.041 |
Token | gave |
Token ID | 2921 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
iddled | +0.066 |
aback | +0.062 |
uster | +0.053 |
bench | +0.052 |
ocated | +0.051 |
Neg logit contributions | |
ost | -0.079 |
ric | -0.079 |
peac | -0.074 |
flaming | -0.072 |
ostr | -0.072 |
Token | touched |
Token ID | 12615 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
ost | +0.056 |
ric | +0.054 |
ostr | +0.051 |
peac | +0.051 |
work | +0.050 |
Neg logit contributions | |
iddled | -0.058 |
aback | -0.051 |
uster | -0.049 |
tant | -0.046 |
bench | -0.046 |
Token | !" |
Token ID | 2474 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
ric | +0.025 |
ost | +0.024 |
highlight | +0.024 |
peac | +0.024 |
ostr | +0.023 |
Neg logit contributions | |
iddled | -0.015 |
aback | -0.014 |
uster | -0.012 |
tant | -0.012 |
ocated | -0.011 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
iddled | +0.044 |
aback | +0.039 |
uster | +0.036 |
tant | +0.035 |
bench | +0.034 |
Neg logit contributions | |
ost | -0.042 |
ric | -0.041 |
ostr | -0.038 |
peac | -0.038 |
highlight | -0.037 |
Token | was |
Token ID | 373 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
iddled | +0.082 |
aback | +0.075 |
uster | +0.063 |
tant | +0.062 |
ocated | +0.060 |
Neg logit contributions | |
ost | -0.083 |
ric | -0.079 |
highlight | -0.077 |
zoo | -0.075 |
peac | -0.075 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
ost | +0.020 |
ric | +0.020 |
peac | +0.019 |
ostr | +0.019 |
work | +0.019 |
Neg logit contributions | |
iddled | -0.020 |
aback | -0.018 |
uster | -0.016 |
ocated | -0.015 |
bench | -0.015 |
Token | aback |
Token ID | 50147 |
Feature activation | -8.06e+00 |
Pos logit contributions | |
iddled | +0.076 |
bench | +0.065 |
oons | +0.062 |
uster | +0.061 |
resses | +0.055 |
Neg logit contributions | |
ost | -0.097 |
guardian | -0.093 |
rob | -0.093 |
� | -0.092 |
moral | -0.092 |
Token | and |
Token ID | 290 |
Feature activation | -9.99e-02 |
Pos logit contributions | |
iddled | +0.498 |
aback | +0.411 |
uster | +0.402 |
bench | +0.386 |
ocated | +0.383 |
Neg logit contributions | |
ost | -0.499 |
ric | -0.470 |
ostr | -0.463 |
peac | -0.445 |
moral | -0.437 |
Token | stopped |
Token ID | 5025 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
iddled | +0.006 |
aback | +0.005 |
uster | +0.005 |
ocated | +0.005 |
tant | +0.005 |
Neg logit contributions | |
ost | -0.006 |
ric | -0.006 |
ostr | -0.006 |
peac | -0.006 |
highlight | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
highlight | +0.106 |
ost | +0.102 |
ric | +0.099 |
moral | +0.099 |
work | +0.099 |
Neg logit contributions | |
iddled | -0.093 |
aback | -0.086 |
uster | -0.070 |
away | -0.069 |
iddling | -0.067 |
Token | her |
Token ID | 607 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
ost | +0.030 |
peac | +0.028 |
ric | +0.027 |
moral | +0.027 |
ostr | +0.027 |
Neg logit contributions | |
iddled | -0.040 |
uster | -0.034 |
aback | -0.033 |
ocated | -0.032 |
wn | -0.032 |
Token | tracks |
Token ID | 8339 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
aback | +0.058 |
tant | +0.057 |
iddled | +0.057 |
breath | +0.055 |
away | +0.051 |
Neg logit contributions | |
ost | -0.099 |
ric | -0.096 |
peac | -0.090 |
flaming | -0.089 |
highlight | -0.088 |
Token | ." |
Token ID | 526 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
iddled | +0.091 |
aback | +0.078 |
tant | +0.068 |
bench | +0.067 |
uster | +0.065 |
Neg logit contributions | |
ost | -0.142 |
ric | -0.134 |
ostr | -0.131 |
peac | -0.129 |
North | -0.125 |
Token | |
Token ID | 220 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
ost | +0.008 |
ric | +0.008 |
highlight | +0.008 |
work | +0.007 |
peac | +0.007 |
Neg logit contributions | |
iddled | -0.005 |
aback | -0.005 |
tant | -0.004 |
uster | -0.004 |
bench | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
ost | +0.012 |
ric | +0.012 |
ostr | +0.011 |
peac | +0.011 |
work | +0.011 |
Neg logit contributions | |
iddled | -0.017 |
aback | -0.015 |
uster | -0.014 |
tant | -0.013 |
bench | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
iddled | +0.044 |
aback | +0.039 |
uster | +0.037 |
bench | +0.035 |
tant | +0.035 |
Neg logit contributions | |
ost | -0.039 |
ric | -0.037 |
ostr | -0.036 |
peac | -0.036 |
work | -0.036 |
Token | B |
Token ID | 33 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
iddled | +0.026 |
aback | +0.021 |
uster | +0.020 |
bench | +0.018 |
tant | +0.018 |
Neg logit contributions | |
ost | -0.040 |
ric | -0.039 |
ostr | -0.037 |
work | -0.037 |
peac | -0.036 |
Token | uster |
Token ID | 5819 |
Feature activation | -8.03e+00 |
Pos logit contributions | |
iddled | +0.153 |
aback | +0.131 |
uster | +0.127 |
ocated | +0.109 |
tant | +0.109 |
Neg logit contributions | |
ost | -0.236 |
ric | -0.223 |
ostr | -0.215 |
peac | -0.214 |
flaming | -0.212 |
Token | felt |
Token ID | 2936 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
iddled | +0.456 |
aback | +0.386 |
uster | +0.335 |
tant | +0.328 |
bench | +0.316 |
Neg logit contributions | |
ost | -0.528 |
ric | -0.513 |
highlight | -0.505 |
peac | -0.489 |
North | -0.482 |
Token | so |
Token ID | 523 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
iddled | +0.038 |
aback | +0.034 |
uster | +0.031 |
tant | +0.031 |
bench | +0.030 |
Neg logit contributions | |
ost | -0.033 |
ric | -0.032 |
peac | -0.030 |
ostr | -0.030 |
work | -0.029 |
Token | proud |
Token ID | 6613 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
iddled | +0.017 |
aback | +0.016 |
uster | +0.014 |
tant | +0.014 |
bench | +0.014 |
Neg logit contributions | |
ost | -0.015 |
ric | -0.015 |
peac | -0.014 |
work | -0.014 |
ostr | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
ost | +0.011 |
ric | +0.011 |
ostr | +0.010 |
peac | +0.010 |
highlight | +0.010 |
Neg logit contributions | |
iddled | -0.010 |
aback | -0.009 |
uster | -0.008 |
tant | -0.008 |
bench | -0.008 |
Token | hum |
Token ID | 1311 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
iddled | +0.017 |
aback | +0.015 |
bench | +0.013 |
tant | +0.012 |
ocated | +0.012 |
Neg logit contributions | |
ric | -0.021 |
ost | -0.019 |
work | -0.018 |
flaming | -0.017 |
zoo | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
iddled | +0.024 |
aback | +0.021 |
uster | +0.020 |
tant | +0.019 |
bench | +0.019 |
Neg logit contributions | |
ost | -0.023 |
ric | -0.022 |
ostr | -0.021 |
peac | -0.021 |
work | -0.021 |
Token | |
Token ID | 220 |
Feature activation | +9.90e-01 |
Pos logit contributions | |
ost | +0.001 |
ric | +0.001 |
peac | +0.001 |
ostr | +0.001 |
work | +0.001 |
Neg logit contributions | |
iddled | -0.001 |
aback | -0.001 |
uster | -0.001 |
tant | -0.001 |
bench | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
ost | +0.043 |
ric | +0.041 |
ostr | +0.037 |
moral | +0.037 |
peac | +0.037 |
Neg logit contributions | |
iddled | -0.076 |
aback | -0.067 |
uster | -0.065 |
tant | -0.064 |
wn | -0.063 |
Token |
|
Token ID | 198 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
iddled | +0.030 |
aback | +0.027 |
uster | +0.026 |
tant | +0.025 |
bench | +0.025 |
Neg logit contributions | |
ost | -0.018 |
ric | -0.017 |
peac | -0.016 |
ostr | -0.016 |
moral | -0.016 |
Token | B |
Token ID | 33 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
iddled | +0.018 |
aback | +0.015 |
uster | +0.014 |
tant | +0.014 |
bench | +0.014 |
Neg logit contributions | |
ost | -0.025 |
ric | -0.024 |
peac | -0.024 |
ostr | -0.023 |
moral | -0.023 |
Token | uster |
Token ID | 5819 |
Feature activation | -7.81e+00 |
Pos logit contributions | |
iddled | +0.154 |
bench | +0.136 |
tant | +0.133 |
aback | +0.119 |
apr | +0.116 |
Neg logit contributions | |
ric | -0.210 |
ich | -0.186 |
work | -0.186 |
ost | -0.186 |
je | -0.184 |
Token | thought |
Token ID | 1807 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
iddled | +0.391 |
aback | +0.332 |
uster | +0.290 |
eal | +0.262 |
tant | +0.257 |
Neg logit contributions | |
ost | -0.554 |
highlight | -0.542 |
ric | -0.541 |
North | -0.511 |
peac | -0.502 |
Token | for |
Token ID | 329 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ost | +0.021 |
ric | +0.020 |
ostr | +0.019 |
peac | +0.019 |
moral | +0.019 |
Neg logit contributions | |
iddled | -0.021 |
aback | -0.019 |
uster | -0.017 |
tant | -0.017 |
bench | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
iddled | +0.069 |
aback | +0.060 |
uster | +0.059 |
tant | +0.057 |
bench | +0.055 |
Neg logit contributions | |
ost | -0.058 |
ric | -0.055 |
peac | -0.052 |
ostr | -0.052 |
work | -0.052 |
Token | moment |
Token ID | 2589 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
iddled | +0.023 |
aback | +0.023 |
uster | +0.021 |
tant | +0.021 |
bench | +0.020 |
Neg logit contributions | |
ric | -0.024 |
ost | -0.023 |
work | -0.021 |
ostr | -0.019 |
North | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
ost | +0.137 |
ric | +0.135 |
peac | +0.127 |
ostr | +0.127 |
rob | +0.125 |
Neg logit contributions | |
iddled | -0.095 |
aback | -0.083 |
uster | -0.077 |
bench | -0.073 |
tant | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
iddled | +0.027 |
aback | +0.023 |
tant | +0.023 |
uster | +0.022 |
bench | +0.022 |
Neg logit contributions | |
ost | -0.023 |
ric | -0.022 |
ostr | -0.021 |
peac | -0.021 |
moral | -0.020 |
Token | |
Token ID | 220 |
Feature activation | +9.67e-01 |
Pos logit contributions | |
ost | +0.006 |
ric | +0.006 |
peac | +0.006 |
ostr | +0.006 |
work | +0.006 |
Neg logit contributions | |
iddled | -0.010 |
aback | -0.009 |
tant | -0.009 |
uster | -0.009 |
bench | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
ost | +0.043 |
ric | +0.041 |
ostr | +0.037 |
peac | +0.037 |
moral | +0.036 |
Neg logit contributions | |
iddled | -0.073 |
aback | -0.065 |
uster | -0.063 |
tant | -0.062 |
bench | -0.060 |
Token |
|
Token ID | 198 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
iddled | +0.029 |
tant | +0.025 |
aback | +0.025 |
wn | +0.024 |
uster | +0.024 |
Neg logit contributions | |
ost | -0.013 |
ric | -0.012 |
moral | -0.011 |
peac | -0.011 |
ostr | -0.011 |
Token | B |
Token ID | 33 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
iddled | +0.020 |
tant | +0.017 |
aback | +0.016 |
bench | +0.016 |
wn | +0.016 |
Neg logit contributions | |
ost | -0.018 |
peac | -0.017 |
moral | -0.016 |
str | -0.016 |
ostr | -0.016 |
Token | uster |
Token ID | 5819 |
Feature activation | -7.79e+00 |
Pos logit contributions | |
iddled | +0.164 |
tant | +0.137 |
bench | +0.137 |
aback | +0.126 |
ags | +0.119 |
Neg logit contributions | |
ric | -0.204 |
ost | -0.193 |
work | -0.186 |
peac | -0.186 |
ostr | -0.183 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
iddled | +0.384 |
aback | +0.319 |
uster | +0.281 |
away | +0.252 |
ly | +0.251 |
Neg logit contributions | |
highlight | -0.564 |
ost | -0.540 |
ric | -0.528 |
zoo | -0.506 |
North | -0.501 |
Token | and |
Token ID | 290 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
ost | +0.075 |
ric | +0.072 |
ostr | +0.070 |
peac | +0.070 |
work | +0.068 |
Neg logit contributions | |
iddled | -0.051 |
aback | -0.042 |
uster | -0.040 |
bench | -0.037 |
tant | -0.037 |
Token | saw |
Token ID | 2497 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
iddled | +0.011 |
aback | +0.010 |
uster | +0.009 |
tant | +0.009 |
bench | +0.009 |
Neg logit contributions | |
ost | -0.010 |
ric | -0.010 |
ostr | -0.009 |
peac | -0.009 |
work | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
iddled | +0.002 |
aback | +0.002 |
uster | +0.001 |
tant | +0.001 |
ocated | +0.001 |
Neg logit contributions | |
ric | -0.002 |
ost | -0.002 |
highlight | -0.002 |
ostr | -0.002 |
work | -0.002 |
Token | small |
Token ID | 1402 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
iddled | +0.017 |
aback | +0.016 |
uster | +0.015 |
tant | +0.015 |
bench | +0.013 |
Neg logit contributions | |
ric | -0.025 |
ost | -0.024 |
highlight | -0.023 |
work | -0.022 |
riage | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
ost | +0.035 |
ric | +0.033 |
peac | +0.032 |
ostr | +0.032 |
work | +0.031 |
Neg logit contributions | |
iddled | -0.035 |
uster | -0.031 |
aback | -0.030 |
tant | -0.028 |
ocated | -0.028 |
Token | forest |
Token ID | 8222 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
tant | +0.010 |
away | +0.009 |
iddled | +0.009 |
aback | +0.009 |
sc | +0.009 |
Neg logit contributions | |
ric | -0.011 |
orse | -0.011 |
ost | -0.010 |
flaming | -0.010 |
riage | -0.009 |
Token | looking |
Token ID | 2045 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
ric | +0.056 |
ost | +0.055 |
ostr | +0.052 |
flaming | +0.052 |
peac | +0.052 |
Neg logit contributions | |
iddled | -0.028 |
aback | -0.024 |
bench | -0.020 |
ows | -0.019 |
away | -0.018 |
Token | for |
Token ID | 329 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ost | +0.012 |
ric | +0.011 |
ostr | +0.011 |
peac | +0.011 |
work | +0.011 |
Neg logit contributions | |
iddled | -0.009 |
aback | -0.008 |
uster | -0.008 |
tant | -0.007 |
bench | -0.007 |
Token | all |
Token ID | 477 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
iddled | +0.088 |
aback | +0.077 |
uster | +0.072 |
tant | +0.068 |
bench | +0.067 |
Neg logit contributions | |
ost | -0.096 |
ric | -0.095 |
ostr | -0.089 |
peac | -0.089 |
work | -0.088 |
Token | sorts |
Token ID | 10524 |
Feature activation | -7.49e+00 |
Pos logit contributions | |
iddled | +0.173 |
aback | +0.155 |
bench | +0.130 |
uster | +0.129 |
wn | +0.125 |
Neg logit contributions | |
ost | -0.244 |
ric | -0.242 |
peac | -0.237 |
work | -0.231 |
flaming | -0.229 |
Token | of |
Token ID | 286 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
iddled | +0.480 |
aback | +0.419 |
uster | +0.405 |
tant | +0.383 |
bench | +0.382 |
Neg logit contributions | |
ost | -0.429 |
ric | -0.406 |
ostr | -0.392 |
peac | -0.390 |
work | -0.381 |
Token | exciting |
Token ID | 7895 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ric | +0.009 |
ost | +0.009 |
work | +0.008 |
peac | +0.008 |
ostr | +0.008 |
Neg logit contributions | |
iddled | -0.009 |
aback | -0.008 |
uster | -0.007 |
tant | -0.007 |
bench | -0.006 |
Token | things |
Token ID | 1243 |
Feature activation | -9.61e-01 |
Pos logit contributions | |
iddled | +0.079 |
aback | +0.070 |
uster | +0.065 |
tant | +0.062 |
bench | +0.061 |
Neg logit contributions | |
ost | -0.079 |
ric | -0.079 |
peac | -0.074 |
ostr | -0.073 |
work | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
iddled | +0.052 |
aback | +0.046 |
uster | +0.042 |
bench | +0.040 |
ocated | +0.039 |
Neg logit contributions | |
ost | -0.063 |
ric | -0.062 |
ostr | -0.059 |
peac | -0.059 |
moral | -0.058 |
Token | She |
Token ID | 1375 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
iddled | +0.027 |
aback | +0.025 |
uster | +0.023 |
bench | +0.020 |
ocated | +0.019 |
Neg logit contributions | |
ost | -0.041 |
work | -0.038 |
peac | -0.038 |
ostr | -0.037 |
flaming | -0.037 |
Token | so |
Token ID | 523 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
ost | +0.204 |
ric | +0.198 |
ostr | +0.189 |
peac | +0.188 |
work | +0.183 |
Neg logit contributions | |
iddled | -0.182 |
aback | -0.158 |
uster | -0.147 |
bench | -0.139 |
tant | -0.138 |
Token | much |
Token ID | 881 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
iddled | +0.002 |
tant | +0.002 |
aback | +0.002 |
ushes | +0.002 |
wn | +0.002 |
Neg logit contributions | |
ric | -0.004 |
ost | -0.004 |
peac | -0.004 |
ostr | -0.004 |
ich | -0.004 |
Token | fun |
Token ID | 1257 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
iddled | +0.013 |
aback | +0.011 |
uster | +0.010 |
bench | +0.010 |
tant | +0.009 |
Neg logit contributions | |
ost | -0.020 |
ric | -0.020 |
ostr | -0.019 |
peac | -0.019 |
work | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
iddled | +0.173 |
aback | +0.139 |
bench | +0.123 |
tant | +0.121 |
uster | +0.121 |
Neg logit contributions | |
ost | -0.287 |
ric | -0.281 |
ostr | -0.278 |
peac | -0.268 |
rob | -0.264 |
Token | K |
Token ID | 509 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
iddled | +0.012 |
aback | +0.011 |
zip | +0.009 |
uster | +0.009 |
ocated | +0.008 |
Neg logit contributions | |
work | -0.013 |
highlight | -0.013 |
peac | -0.013 |
whale | -0.013 |
orse | -0.013 |
Token | ait |
Token ID | 4548 |
Feature activation | -7.45e+00 |
Pos logit contributions | |
ost | +0.228 |
ric | +0.223 |
peac | +0.218 |
ostr | +0.214 |
work | +0.213 |
Neg logit contributions | |
iddled | -0.115 |
aback | -0.093 |
uster | -0.090 |
tant | -0.078 |
ocated | -0.075 |
Token | lyn |
Token ID | 6213 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
iddled | +0.482 |
aback | +0.426 |
uster | +0.417 |
fully | +0.390 |
away | +0.390 |
Neg logit contributions | |
peac | -0.385 |
ost | -0.359 |
highlight | -0.351 |
orse | -0.344 |
ric | -0.343 |
Token | was |
Token ID | 373 |
Feature activation | +9.08e-02 |
Pos logit contributions | |
ost | +0.156 |
ric | +0.150 |
highlight | +0.141 |
flaming | +0.135 |
hipp | +0.134 |
Neg logit contributions | |
iddled | -0.138 |
aback | -0.121 |
tant | -0.101 |
bench | -0.100 |
eal | -0.100 |
Token | so |
Token ID | 523 |
Feature activation | -7.24e-02 |
Pos logit contributions | |
ost | +0.007 |
ric | +0.006 |
oral | +0.006 |
ostr | +0.006 |
peac | +0.006 |
Neg logit contributions | |
aback | -0.004 |
iddled | -0.004 |
away | -0.003 |
ocated | -0.003 |
wn | -0.003 |
Token | happy |
Token ID | 3772 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.004 |
tant | +0.003 |
uster | +0.003 |
eal | +0.003 |
Neg logit contributions | |
ost | -0.005 |
ric | -0.005 |
ostr | -0.004 |
zoo | -0.004 |
flaming | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
ric | +0.034 |
ost | +0.034 |
highlight | +0.032 |
ostr | +0.032 |
peac | +0.031 |
Neg logit contributions | |
iddled | -0.026 |
aback | -0.024 |
uster | -0.020 |
ocated | -0.020 |
bench | -0.019 |
Token | so |
Token ID | 523 |
Feature activation | -5.75e-02 |
Pos logit contributions | |
ost | +0.004 |
ric | +0.004 |
ostr | +0.004 |
peac | +0.004 |
moral | +0.004 |
Neg logit contributions | |
iddled | -0.003 |
aback | -0.003 |
bench | -0.002 |
uster | -0.002 |
ocated | -0.002 |
Token | happy |
Token ID | 3772 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.004 |
uster | +0.003 |
tant | +0.003 |
bench | +0.003 |
Neg logit contributions | |
ost | -0.003 |
ric | -0.003 |
ostr | -0.003 |
peac | -0.003 |
work | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
ost | +0.037 |
ric | +0.037 |
highlight | +0.035 |
ostr | +0.035 |
peac | +0.035 |
Neg logit contributions | |
iddled | -0.026 |
aback | -0.023 |
uster | -0.020 |
ocated | -0.020 |
bench | -0.018 |
Token | see |
Token ID | 766 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
iddled | +0.016 |
aback | +0.015 |
uster | +0.013 |
ocated | +0.012 |
zip | +0.012 |
Neg logit contributions | |
ost | -0.020 |
ric | -0.019 |
highlight | -0.018 |
ostr | -0.018 |
ich | -0.017 |
Token | K |
Token ID | 509 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
iddled | +0.021 |
aback | +0.019 |
uster | +0.018 |
ocated | +0.017 |
tant | +0.017 |
Neg logit contributions | |
ric | -0.016 |
ost | -0.015 |
peac | -0.015 |
work | -0.015 |
highlight | -0.015 |
Token | ait |
Token ID | 4548 |
Feature activation | -7.42e+00 |
Pos logit contributions | |
ost | +0.237 |
ric | +0.234 |
peac | +0.231 |
ostr | +0.224 |
work | +0.223 |
Neg logit contributions | |
iddled | -0.105 |
uster | -0.083 |
aback | -0.083 |
tant | -0.067 |
ocated | -0.067 |
Token | lyn |
Token ID | 6213 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
iddled | +0.437 |
aback | +0.404 |
away | +0.387 |
fully | +0.386 |
uster | +0.375 |
Neg logit contributions | |
peac | -0.422 |
ost | -0.389 |
highlight | -0.386 |
orse | -0.378 |
ric | -0.372 |
Token | with |
Token ID | 351 |
Feature activation | -8.52e-01 |
Pos logit contributions | |
ost | +0.175 |
ostr | +0.162 |
peac | +0.158 |
hipp | +0.155 |
flaming | +0.155 |
Neg logit contributions | |
iddled | -0.110 |
aback | -0.094 |
ows | -0.086 |
away | -0.084 |
ets | -0.084 |
Token | a |
Token ID | 257 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
iddled | +0.050 |
aback | +0.043 |
uster | +0.041 |
tant | +0.039 |
bench | +0.038 |
Neg logit contributions | |
ost | -0.054 |
ric | -0.052 |
ostr | -0.050 |
peac | -0.049 |
work | -0.048 |
Token | smile |
Token ID | 8212 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
iddled | +0.022 |
aback | +0.020 |
tant | +0.019 |
uster | +0.019 |
bench | +0.018 |
Neg logit contributions | |
ost | -0.025 |
ric | -0.024 |
ostr | -0.023 |
work | -0.022 |
peac | -0.021 |
Token | on |
Token ID | 319 |
Feature activation | +7.07e-01 |
Pos logit contributions | |
iddled | +0.014 |
aback | +0.011 |
uster | +0.011 |
bench | +0.011 |
tant | +0.010 |
Neg logit contributions | |
ost | -0.026 |
ric | -0.025 |
peac | -0.023 |
ostr | -0.023 |
highlight | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
iddled | +0.140 |
aback | +0.119 |
bench | +0.106 |
uster | +0.105 |
ocated | +0.098 |
Neg logit contributions | |
ost | -0.209 |
ric | -0.207 |
peac | -0.197 |
ostr | -0.197 |
flaming | -0.193 |
Token | thrilled |
Token ID | 20536 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
iddled | +0.030 |
aback | +0.027 |
uster | +0.023 |
tant | +0.023 |
bench | +0.023 |
Neg logit contributions | |
ost | -0.028 |
ric | -0.027 |
ostr | -0.026 |
peac | -0.026 |
oral | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
iddled | +0.017 |
aback | +0.015 |
uster | +0.014 |
tant | +0.013 |
ocated | +0.012 |
Neg logit contributions | |
ost | -0.022 |
ric | -0.022 |
ostr | -0.021 |
peac | -0.021 |
highlight | -0.020 |
Token | discover |
Token ID | 7073 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
iddled | +0.040 |
aback | +0.035 |
bench | +0.030 |
uster | +0.030 |
ocated | +0.030 |
Neg logit contributions | |
ost | -0.047 |
ric | -0.045 |
work | -0.043 |
ostr | -0.043 |
peac | -0.042 |
Token | all |
Token ID | 477 |
Feature activation | -3.37e+00 |
Pos logit contributions | |
ost | +0.043 |
ric | +0.043 |
peac | +0.040 |
ostr | +0.040 |
work | +0.040 |
Neg logit contributions | |
iddled | -0.042 |
aback | -0.037 |
uster | -0.034 |
tant | -0.032 |
bench | -0.031 |
Token | sorts |
Token ID | 10524 |
Feature activation | -7.41e+00 |
Pos logit contributions | |
iddled | +0.187 |
aback | +0.164 |
uster | +0.150 |
bench | +0.143 |
tant | +0.142 |
Neg logit contributions | |
ost | -0.219 |
ric | -0.214 |
peac | -0.207 |
ostr | -0.204 |
work | -0.201 |
Token | of |
Token ID | 286 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
iddled | +0.465 |
aback | +0.408 |
uster | +0.391 |
tant | +0.369 |
bench | +0.367 |
Neg logit contributions | |
ost | -0.434 |
ric | -0.415 |
ostr | -0.397 |
peac | -0.396 |
work | -0.385 |
Token | new |
Token ID | 649 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
ric | +0.013 |
ost | +0.013 |
work | +0.013 |
peac | +0.012 |
ostr | +0.012 |
Neg logit contributions | |
iddled | -0.013 |
aback | -0.012 |
uster | -0.011 |
tant | -0.010 |
bench | -0.010 |
Token | things |
Token ID | 1243 |
Feature activation | -7.71e-01 |
Pos logit contributions | |
iddled | +0.034 |
aback | +0.030 |
uster | +0.027 |
tant | +0.026 |
bench | +0.025 |
Neg logit contributions | |
ost | -0.037 |
ric | -0.037 |
peac | -0.034 |
ostr | -0.034 |
work | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
iddled | +0.041 |
aback | +0.036 |
uster | +0.032 |
bench | +0.031 |
ocated | +0.030 |
Neg logit contributions | |
ost | -0.052 |
ric | -0.052 |
ostr | -0.049 |
peac | -0.049 |
moral | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -8.19e-01 |
Pos logit contributions | |
iddled | +0.027 |
aback | +0.025 |
uster | +0.023 |
bench | +0.020 |
ocated | +0.019 |
Neg logit contributions | |
ost | -0.038 |
peac | -0.035 |
ostr | -0.035 |
ric | -0.034 |
flaming | -0.034 |
Token | they |
Token ID | 484 |
Feature activation | +6.59e-01 |
Pos logit contributions | |
iddled | +0.018 |
away | +0.016 |
aback | +0.015 |
breath | +0.014 |
her | +0.013 |
Neg logit contributions | |
ric | -0.047 |
ost | -0.044 |
rob | -0.043 |
oral | -0.043 |
ostr | -0.042 |
Token | both |
Token ID | 1111 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
ost | +0.039 |
ric | +0.038 |
ostr | +0.035 |
peac | +0.034 |
work | +0.033 |
Neg logit contributions | |
iddled | -0.041 |
aback | -0.036 |
uster | -0.033 |
tant | -0.033 |
ocated | -0.032 |
Token | were |
Token ID | 547 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
iddled | +0.017 |
aback | +0.015 |
tant | +0.014 |
away | +0.013 |
ets | +0.013 |
Neg logit contributions | |
ost | -0.021 |
ric | -0.021 |
ich | -0.019 |
moral | -0.018 |
ostr | -0.018 |
Token | out |
Token ID | 503 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
ost | +0.020 |
ric | +0.019 |
ostr | +0.019 |
peac | +0.018 |
work | +0.018 |
Neg logit contributions | |
iddled | -0.021 |
aback | -0.018 |
uster | -0.017 |
bench | -0.016 |
ocated | -0.016 |
Token | of |
Token ID | 286 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
iddled | +0.065 |
aback | +0.057 |
tant | +0.050 |
away | +0.049 |
breath | +0.045 |
Neg logit contributions | |
ost | -0.132 |
ostr | -0.131 |
ric | -0.130 |
str | -0.128 |
peac | -0.126 |
Token | breath |
Token ID | 8033 |
Feature activation | -7.40e+00 |
Pos logit contributions | |
ost | +0.039 |
ric | +0.037 |
peac | +0.036 |
ostr | +0.036 |
work | +0.035 |
Neg logit contributions | |
iddled | -0.042 |
aback | -0.037 |
uster | -0.036 |
tant | -0.033 |
ocated | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
iddled | +0.393 |
aback | +0.344 |
uster | +0.321 |
tant | +0.301 |
bench | +0.299 |
Neg logit contributions | |
ost | -0.503 |
ric | -0.490 |
peac | -0.466 |
ostr | -0.465 |
rob | -0.451 |
Token | |
Token ID | 220 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
iddled | +0.008 |
aback | +0.007 |
uster | +0.007 |
acks | +0.006 |
ocated | +0.006 |
Neg logit contributions | |
ost | -0.012 |
ostr | -0.011 |
highlight | -0.011 |
peac | -0.011 |
ric | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -6.80e-01 |
Pos logit contributions | |
ost | +0.032 |
ric | +0.030 |
ostr | +0.029 |
peac | +0.029 |
work | +0.028 |
Neg logit contributions | |
iddled | -0.041 |
aback | -0.036 |
uster | -0.035 |
bench | -0.033 |
tant | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
iddled | +0.044 |
aback | +0.038 |
uster | +0.036 |
bench | +0.034 |
ocated | +0.034 |
Neg logit contributions | |
ost | -0.040 |
ric | -0.037 |
ostr | -0.037 |
work | -0.036 |
peac | -0.036 |
Token | B |
Token ID | 33 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
iddled | +0.028 |
aback | +0.022 |
uster | +0.022 |
ocated | +0.020 |
bench | +0.019 |
Neg logit contributions | |
ost | -0.042 |
ric | -0.041 |
ostr | -0.040 |
work | -0.040 |
peac | -0.038 |
Token | match |
Token ID | 2872 |
Feature activation | -6.63e-01 |
Pos logit contributions | |
iddled | +0.013 |
aback | +0.011 |
tant | +0.010 |
wn | +0.010 |
bench | +0.010 |
Neg logit contributions | |
ost | -0.015 |
ric | -0.015 |
peac | -0.014 |
work | -0.014 |
ostr | -0.014 |
Token | her |
Token ID | 607 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
iddled | +0.035 |
aback | +0.030 |
uster | +0.029 |
bench | +0.027 |
tant | +0.027 |
Neg logit contributions | |
ost | -0.045 |
ric | -0.044 |
ostr | -0.042 |
peac | -0.041 |
work | -0.041 |
Token | moves |
Token ID | 6100 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
iddled | +0.082 |
tant | +0.071 |
aback | +0.070 |
uster | +0.065 |
wn | +0.062 |
Neg logit contributions | |
ost | -0.109 |
ric | -0.105 |
peac | -0.103 |
ostr | -0.097 |
str | -0.097 |
Token | with |
Token ID | 351 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
iddled | +0.135 |
aback | +0.116 |
uster | +0.110 |
tant | +0.103 |
bench | +0.102 |
Neg logit contributions | |
ost | -0.163 |
ric | -0.158 |
ostr | -0.151 |
peac | -0.151 |
highlight | -0.145 |
Token | her |
Token ID | 607 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
iddled | +0.050 |
aback | +0.042 |
uster | +0.040 |
bench | +0.037 |
tant | +0.036 |
Neg logit contributions | |
ost | -0.076 |
ric | -0.074 |
ostr | -0.071 |
peac | -0.071 |
work | -0.069 |
Token | breath |
Token ID | 8033 |
Feature activation | -7.40e+00 |
Pos logit contributions | |
iddled | +0.085 |
tant | +0.074 |
aback | +0.073 |
uster | +0.071 |
bench | +0.064 |
Neg logit contributions | |
ost | -0.108 |
ric | -0.105 |
peac | -0.102 |
ostr | -0.098 |
str | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
iddled | +0.433 |
aback | +0.375 |
uster | +0.358 |
tant | +0.336 |
bench | +0.334 |
Neg logit contributions | |
ost | -0.461 |
ric | -0.444 |
ostr | -0.425 |
peac | -0.424 |
work | -0.415 |
Token | One |
Token ID | 1881 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
iddled | +0.002 |
aback | +0.002 |
uster | +0.001 |
tant | +0.001 |
bench | +0.001 |
Neg logit contributions | |
ost | -0.002 |
peac | -0.002 |
ostr | -0.002 |
ric | -0.002 |
work | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
ric | +0.115 |
ost | +0.102 |
peac | +0.100 |
work | +0.093 |
flaming | +0.090 |
Neg logit contributions | |
iddled | -0.141 |
tant | -0.126 |
aback | -0.126 |
uster | -0.121 |
bench | -0.118 |
Token | , |
Token ID | 11 |
Feature activation | -8.12e-02 |
Pos logit contributions | |
ost | +0.094 |
ric | +0.092 |
peac | +0.087 |
ostr | +0.087 |
work | +0.085 |
Neg logit contributions | |
iddled | -0.083 |
aback | -0.072 |
uster | -0.068 |
tant | -0.064 |
ocated | -0.063 |
Token | while |
Token ID | 981 |
Feature activation | -4.08e+00 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.004 |
uster | +0.003 |
tant | +0.003 |
ocated | +0.003 |
Neg logit contributions | |
ric | -0.006 |
ost | -0.006 |
peac | -0.005 |
ostr | -0.005 |
moral | -0.005 |
Token | things |
Token ID | 1243 |
Feature activation | -9.67e-01 |
Pos logit contributions | |
iddled | +0.051 |
tant | +0.048 |
aback | +0.047 |
bench | +0.042 |
uster | +0.042 |
Neg logit contributions | |
ric | -0.056 |
ost | -0.053 |
highlight | -0.052 |
str | -0.048 |
riage | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
iddled | +0.051 |
aback | +0.044 |
uster | +0.039 |
bench | +0.039 |
ocated | +0.037 |
Neg logit contributions | |
ost | -0.065 |
ric | -0.064 |
peac | -0.061 |
ostr | -0.061 |
flaming | -0.059 |
Token | They |
Token ID | 1119 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
iddled | +0.024 |
aback | +0.021 |
uster | +0.019 |
bench | +0.017 |
zip | +0.016 |
Neg logit contributions | |
ost | -0.043 |
ric | -0.040 |
highlight | -0.039 |
peac | -0.039 |
ostr | -0.039 |
Token | met |
Token ID | 1138 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
iddled | +0.054 |
aback | +0.046 |
tant | +0.042 |
bench | +0.041 |
uster | +0.041 |
Neg logit contributions | |
ost | -0.054 |
ric | -0.052 |
highlight | -0.048 |
flaming | -0.048 |
ich | -0.048 |
Token | all |
Token ID | 477 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
ric | +0.106 |
ost | +0.101 |
peac | +0.096 |
ostr | +0.094 |
highlight | +0.094 |
Neg logit contributions | |
iddled | -0.073 |
aback | -0.063 |
tant | -0.053 |
uster | -0.052 |
ocated | -0.048 |
Token | kinds |
Token ID | 6982 |
Feature activation | -7.38e+00 |
Pos logit contributions | |
iddled | +0.143 |
aback | +0.128 |
tant | +0.107 |
bench | +0.105 |
away | +0.096 |
Neg logit contributions | |
ric | -0.238 |
ich | -0.233 |
ost | -0.233 |
peac | -0.230 |
flaming | -0.225 |
Token | of |
Token ID | 286 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
iddled | +0.351 |
aback | +0.301 |
uster | +0.269 |
tant | +0.251 |
bench | +0.250 |
Neg logit contributions | |
ric | -0.559 |
ost | -0.525 |
North | -0.498 |
highlight | -0.496 |
ostr | -0.494 |
Token | animals |
Token ID | 4695 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
ric | +0.026 |
work | +0.024 |
peac | +0.023 |
ost | +0.023 |
highlight | +0.023 |
Neg logit contributions | |
iddled | -0.020 |
aback | -0.017 |
tant | -0.016 |
wn | -0.014 |
uster | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
ric | +0.171 |
moral | +0.163 |
ost | +0.161 |
highlight | +0.159 |
North | +0.158 |
Neg logit contributions | |
iddled | -0.080 |
aback | -0.069 |
ows | -0.055 |
ets | -0.053 |
uster | -0.051 |
Token | from |
Token ID | 422 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
iddled | +0.016 |
aback | +0.014 |
tant | +0.013 |
uster | +0.012 |
wn | +0.012 |
Neg logit contributions | |
ric | -0.018 |
work | -0.016 |
oral | -0.016 |
riage | -0.016 |
ost | -0.016 |
Token | tiny |
Token ID | 7009 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
ric | +0.087 |
str | +0.084 |
ost | +0.083 |
highlight | +0.083 |
peac | +0.081 |
Neg logit contributions | |
iddled | -0.047 |
aback | -0.041 |
tant | -0.039 |
away | -0.036 |
bench | -0.031 |
Token | moral |
Token ID | 6573 |
Feature activation | +6.83e+00 |
Pos logit contributions | |
ric | +0.022 |
orse | +0.021 |
work | +0.020 |
riage | +0.018 |
grand | +0.018 |
Neg logit contributions | |
aback | -0.016 |
knot | -0.015 |
away | -0.015 |
iddled | -0.014 |
breath | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
ric | +0.476 |
ost | +0.453 |
peac | +0.430 |
highlight | +0.427 |
flaming | +0.426 |
Neg logit contributions | |
iddled | -0.355 |
aback | -0.325 |
uster | -0.290 |
bench | -0.281 |
wn | -0.263 |
Token | their |
Token ID | 511 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
ost | +0.047 |
ostr | +0.044 |
peac | +0.044 |
highlight | +0.043 |
ric | +0.043 |
Neg logit contributions | |
iddled | -0.029 |
aback | -0.025 |
uster | -0.022 |
bench | -0.021 |
tant | -0.019 |
Token | story |
Token ID | 1621 |
Feature activation | +5.47e+00 |
Pos logit contributions | |
breath | +0.017 |
aback | +0.016 |
knot | +0.016 |
iddled | +0.015 |
off | +0.015 |
Neg logit contributions | |
ric | -0.021 |
flower | -0.019 |
ost | -0.017 |
ich | -0.016 |
highlight | -0.016 |
Token | : |
Token ID | 25 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
ost | +0.332 |
ric | +0.325 |
peac | +0.309 |
ostr | +0.307 |
flaming | +0.301 |
Neg logit contributions | |
iddled | -0.330 |
aback | -0.294 |
bench | -0.272 |
uster | -0.269 |
ocated | -0.253 |
Token | Just |
Token ID | 2329 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
ost | +0.267 |
flaming | +0.251 |
str | +0.251 |
ostr | +0.250 |
highlight | +0.243 |
Neg logit contributions | |
iddled | -0.195 |
aback | -0.182 |
uster | -0.137 |
ocated | -0.135 |
If | -0.130 |
Token | because |
Token ID | 780 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
ost | +0.273 |
flaming | +0.253 |
ich | +0.244 |
ric | +0.237 |
peac | +0.233 |
Neg logit contributions | |
aback | -0.201 |
iddled | -0.192 |
away | -0.181 |
wn | -0.169 |
uster | -0.164 |
Token | something |
Token ID | 1223 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
ric | +0.041 |
flaming | +0.039 |
ost | +0.038 |
highlight | +0.037 |
ostr | +0.037 |
Neg logit contributions | |
iddled | -0.018 |
aback | -0.015 |
ocated | -0.012 |
uster | -0.012 |
bench | -0.011 |
Token | is |
Token ID | 318 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
iddled | +0.003 |
aback | +0.003 |
uster | +0.002 |
tant | +0.002 |
ocated | +0.002 |
Neg logit contributions | |
ric | -0.003 |
ost | -0.003 |
peac | -0.003 |
highlight | -0.003 |
flaming | -0.003 |
Token | difficult |
Token ID | 2408 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
iddled | +0.021 |
aback | +0.021 |
bench | +0.017 |
away | +0.017 |
wn | +0.017 |
Neg logit contributions | |
ric | -0.026 |
ost | -0.025 |
flaming | -0.024 |
peac | -0.024 |
orse | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
iddled | +0.076 |
aback | +0.066 |
uster | +0.063 |
tant | +0.059 |
bench | +0.059 |
Neg logit contributions | |
ost | -0.084 |
ric | -0.081 |
ostr | -0.078 |
peac | -0.077 |
work | -0.075 |
Token | in |
Token ID | 287 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
ric | +0.397 |
highlight | +0.392 |
ost | +0.391 |
moral | +0.379 |
oral | +0.374 |
Neg logit contributions | |
iddled | -0.153 |
away | -0.114 |
aback | -0.105 |
bench | -0.105 |
wn | -0.093 |
Token | his |
Token ID | 465 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
ost | +0.020 |
ric | +0.020 |
ostr | +0.019 |
peac | +0.019 |
rob | +0.018 |
Neg logit contributions | |
iddled | -0.017 |
aback | -0.015 |
uster | -0.015 |
tant | -0.013 |
ocated | -0.013 |
Token | garden |
Token ID | 11376 |
Feature activation | +5.68e-01 |
Pos logit contributions | |
iddled | +0.090 |
aback | +0.082 |
tant | +0.075 |
uster | +0.073 |
bench | +0.072 |
Neg logit contributions | |
ost | -0.093 |
ric | -0.090 |
work | -0.085 |
peac | -0.085 |
highlight | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
ost | +0.038 |
ric | +0.037 |
ostr | +0.035 |
peac | +0.035 |
work | +0.035 |
Neg logit contributions | |
iddled | -0.030 |
aback | -0.026 |
uster | -0.024 |
bench | -0.023 |
tant | -0.023 |
Token | The |
Token ID | 383 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
ost | +0.007 |
ric | +0.006 |
ostr | +0.006 |
peac | +0.006 |
rob | +0.006 |
Neg logit contributions | |
iddled | -0.013 |
uster | -0.012 |
aback | -0.012 |
tant | -0.011 |
ocated | -0.011 |
Token | angel |
Token ID | 18304 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
ric | +0.110 |
ost | +0.099 |
orse | +0.087 |
highlight | +0.087 |
peac | +0.086 |
Neg logit contributions | |
iddled | -0.161 |
aback | -0.149 |
tant | -0.143 |
bench | -0.136 |
uster | -0.132 |
Token | was |
Token ID | 373 |
Feature activation | -6.62e-02 |
Pos logit contributions | |
ric | +0.332 |
highlight | +0.317 |
ost | +0.301 |
North | +0.290 |
Whale | +0.282 |
Neg logit contributions | |
iddled | -0.223 |
led | -0.186 |
aback | -0.175 |
zip | -0.172 |
away | -0.164 |
Token | very |
Token ID | 845 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.003 |
bench | +0.003 |
uster | +0.003 |
tant | +0.003 |
Neg logit contributions | |
ric | -0.004 |
ost | -0.004 |
work | -0.004 |
highlight | -0.004 |
ostr | -0.004 |
Token | pretty |
Token ID | 2495 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
ric | +0.013 |
ost | +0.013 |
work | +0.012 |
highlight | +0.012 |
peac | +0.012 |
Neg logit contributions | |
iddled | -0.010 |
aback | -0.008 |
tant | -0.008 |
uster | -0.007 |
bench | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
iddled | +0.073 |
aback | +0.063 |
uster | +0.061 |
tant | +0.056 |
bench | +0.056 |
Neg logit contributions | |
ost | -0.071 |
ric | -0.067 |
ostr | -0.065 |
peac | -0.064 |
work | -0.064 |
Token | had |
Token ID | 550 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
iddled | +0.025 |
aback | +0.022 |
bench | +0.020 |
uster | +0.020 |
ocated | +0.020 |
Neg logit contributions | |
ric | -0.023 |
ost | -0.022 |
highlight | -0.020 |
work | -0.020 |
peac | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
ost | +0.043 |
ric | +0.040 |
ostr | +0.039 |
work | +0.039 |
peac | +0.039 |
Neg logit contributions | |
iddled | -0.042 |
aback | -0.036 |
uster | -0.035 |
tant | -0.033 |
bench | -0.032 |
Token | buy |
Token ID | 2822 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
iddled | +0.023 |
aback | +0.020 |
uster | +0.019 |
ocated | +0.018 |
bench | +0.018 |
Neg logit contributions | |
ost | -0.026 |
ric | -0.025 |
peac | -0.024 |
ostr | -0.024 |
moral | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
iddled | +0.002 |
aback | +0.002 |
uster | +0.002 |
ocated | +0.002 |
tant | +0.002 |
Neg logit contributions | |
ost | -0.002 |
ric | -0.002 |
ostr | -0.002 |
peac | -0.002 |
moral | -0.002 |
Token | bright |
Token ID | 6016 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
iddled | +0.026 |
aback | +0.024 |
uster | +0.023 |
tant | +0.022 |
bench | +0.021 |
Neg logit contributions | |
ric | -0.033 |
ost | -0.033 |
work | -0.031 |
peac | -0.030 |
highlight | -0.029 |
Token | purple |
Token ID | 14032 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
ost | +0.008 |
ric | +0.008 |
highlight | +0.007 |
peac | +0.007 |
work | +0.007 |
Neg logit contributions | |
iddled | -0.008 |
aback | -0.007 |
uster | -0.007 |
tant | -0.006 |
bench | -0.006 |
Token | je |
Token ID | 11223 |
Feature activation | +4.91e+00 |
Pos logit contributions | |
ric | +0.102 |
ost | +0.102 |
highlight | +0.097 |
work | +0.089 |
moral | +0.088 |
Neg logit contributions | |
iddled | -0.113 |
aback | -0.104 |
uster | -0.102 |
tant | -0.102 |
bench | -0.093 |
Token | ep |
Token ID | 538 |
Feature activation | -8.03e-01 |
Pos logit contributions | |
ost | +0.348 |
ric | +0.337 |
ostr | +0.324 |
peac | +0.323 |
work | +0.315 |
Neg logit contributions | |
iddled | -0.247 |
aback | -0.210 |
uster | -0.198 |
tant | -0.184 |
bench | -0.182 |
Token | . |
Token ID | 13 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
iddled | +0.042 |
aback | +0.035 |
uster | +0.034 |
bench | +0.031 |
ocated | +0.029 |
Neg logit contributions | |
ost | -0.057 |
ric | -0.054 |
peac | -0.053 |
moral | -0.052 |
ostr | -0.052 |
Token | It |
Token ID | 632 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
iddled | +0.016 |
aback | +0.014 |
uster | +0.013 |
tant | +0.012 |
bench | +0.012 |
Neg logit contributions | |
ost | -0.018 |
ric | -0.017 |
ostr | -0.016 |
peac | -0.016 |
work | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -6.65e-02 |
Pos logit contributions | |
iddled | +0.091 |
aback | +0.081 |
uster | +0.075 |
bench | +0.074 |
ocated | +0.073 |
Neg logit contributions | |
ost | -0.060 |
ric | -0.058 |
peac | -0.053 |
ostr | -0.053 |
highlight | -0.051 |
Token | so |
Token ID | 523 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.004 |
uster | +0.003 |
tant | +0.003 |
bench | +0.003 |
Neg logit contributions | |
ost | -0.004 |
ric | -0.004 |
ostr | -0.004 |
peac | -0.004 |
work | -0.003 |
Token | 's |
Token ID | 338 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
iddled | +0.159 |
aback | +0.138 |
ocated | +0.121 |
uster | +0.120 |
bench | +0.115 |
Neg logit contributions | |
ost | -0.227 |
ric | -0.223 |
peac | -0.216 |
highlight | -0.214 |
ostr | -0.212 |
Token | face |
Token ID | 1986 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
ric | +0.159 |
ost | +0.157 |
work | +0.149 |
highlight | +0.144 |
str | +0.144 |
Neg logit contributions | |
iddled | -0.115 |
aback | -0.106 |
tant | -0.104 |
uster | -0.096 |
ocated | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | +8.51e-02 |
Pos logit contributions | |
iddled | +0.040 |
aback | +0.032 |
uster | +0.029 |
ocated | +0.028 |
ows | +0.027 |
Neg logit contributions | |
ost | -0.063 |
ric | -0.062 |
peac | -0.060 |
ostr | -0.059 |
rob | -0.058 |
Token | He |
Token ID | 679 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
ost | +0.004 |
ric | +0.004 |
moral | +0.004 |
zoo | +0.004 |
flaming | +0.004 |
Neg logit contributions | |
iddled | -0.005 |
uster | -0.005 |
aback | -0.005 |
wn | -0.005 |
ocated | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.003 |
uster | +0.003 |
tant | +0.003 |
ocated | +0.003 |
Neg logit contributions | |
ost | -0.004 |
ric | -0.004 |
peac | -0.003 |
highlight | -0.003 |
work | -0.003 |
Token | smiling |
Token ID | 16755 |
Feature activation | +4.63e+00 |
Pos logit contributions | |
ost | +0.026 |
ric | +0.026 |
peac | +0.024 |
ostr | +0.024 |
work | +0.024 |
Neg logit contributions | |
iddled | -0.032 |
aback | -0.029 |
uster | -0.027 |
bench | -0.026 |
ocated | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
ost | +0.262 |
work | +0.247 |
ostr | +0.247 |
peac | +0.244 |
ric | +0.244 |
Neg logit contributions | |
iddled | -0.288 |
aback | -0.248 |
uster | -0.246 |
bench | -0.239 |
tant | -0.229 |
Token | holding |
Token ID | 4769 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
iddled | +0.008 |
aback | +0.007 |
bench | +0.007 |
uster | +0.007 |
ocated | +0.007 |
Neg logit contributions | |
ost | -0.006 |
ric | -0.006 |
highlight | -0.006 |
ostr | -0.006 |
work | -0.006 |
Token | something |
Token ID | 1223 |
Feature activation | +7.09e-03 |
Pos logit contributions | |
ost | +0.031 |
flaming | +0.029 |
ric | +0.029 |
work | +0.028 |
ostr | +0.028 |
Neg logit contributions | |
iddled | -0.035 |
bench | -0.031 |
uster | -0.031 |
aback | -0.029 |
ocated | -0.029 |
Token | in |
Token ID | 287 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
ost | +0.000 |
ric | +0.000 |
ostr | +0.000 |
peac | +0.000 |
work | +0.000 |
Neg logit contributions | |
iddled | -0.000 |
aback | -0.000 |
uster | -0.000 |
tant | -0.000 |
bench | -0.000 |
Token | his |
Token ID | 465 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ost | +0.029 |
zoo | +0.027 |
rob | +0.027 |
North | +0.027 |
ostr | +0.027 |
Neg logit contributions | |
iddled | -0.039 |
uster | -0.035 |
ocated | -0.032 |
ows | -0.031 |
iddling | -0.031 |
Token | of |
Token ID | 286 |
Feature activation | +8.94e-01 |
Pos logit contributions | |
iddled | +0.080 |
aback | +0.069 |
uster | +0.065 |
tant | +0.061 |
bench | +0.060 |
Neg logit contributions | |
ric | -0.091 |
ost | -0.091 |
ostr | -0.085 |
peac | -0.085 |
highlight | -0.084 |
Token | the |
Token ID | 262 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
ost | +0.043 |
zoo | +0.039 |
ostr | +0.039 |
peac | +0.037 |
flaming | +0.036 |
Neg logit contributions | |
iddled | -0.070 |
uster | -0.064 |
ocated | -0.059 |
aback | -0.058 |
iddling | -0.058 |
Token | trees |
Token ID | 7150 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
ric | +0.031 |
highlight | +0.030 |
ost | +0.030 |
work | +0.029 |
known | +0.029 |
Neg logit contributions | |
iddled | -0.021 |
aback | -0.021 |
tant | -0.021 |
away | -0.021 |
bench | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ost | +0.138 |
peac | +0.137 |
ric | +0.136 |
highlight | +0.135 |
moral | +0.134 |
Neg logit contributions | |
iddled | -0.063 |
aback | -0.053 |
bench | -0.049 |
away | -0.043 |
ows | -0.043 |
Token | The |
Token ID | 383 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
ost | +0.007 |
ric | +0.007 |
ostr | +0.007 |
peac | +0.007 |
highlight | +0.006 |
Neg logit contributions | |
iddled | -0.005 |
aback | -0.005 |
uster | -0.004 |
bench | -0.004 |
tant | -0.004 |
Token | giant |
Token ID | 6175 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
ric | +0.140 |
highlight | +0.125 |
ost | +0.119 |
orse | +0.118 |
str | +0.114 |
Neg logit contributions | |
tant | -0.107 |
iddled | -0.093 |
ly | -0.093 |
knot | -0.092 |
breath | -0.091 |
Token | had |
Token ID | 550 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
ric | +0.333 |
highlight | +0.324 |
riage | +0.290 |
work | +0.282 |
ost | +0.282 |
Neg logit contributions | |
iddled | -0.187 |
ly | -0.159 |
les | -0.156 |
led | -0.155 |
shoved | -0.145 |
Token | long |
Token ID | 890 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ric | +0.198 |
ost | +0.197 |
work | +0.186 |
highlight | +0.182 |
peac | +0.181 |
Neg logit contributions | |
iddled | -0.193 |
aback | -0.169 |
uster | -0.154 |
ocated | -0.149 |
bench | -0.149 |
Token | hair |
Token ID | 4190 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
iddled | +0.081 |
aback | +0.070 |
tant | +0.064 |
bench | +0.061 |
uster | +0.060 |
Neg logit contributions | |
ost | -0.121 |
ric | -0.118 |
work | -0.118 |
highlight | -0.117 |
peac | -0.114 |
Token | and |
Token ID | 290 |
Feature activation | -7.49e-02 |
Pos logit contributions | |
ost | +0.031 |
highlight | +0.030 |
moral | +0.030 |
peac | +0.030 |
ric | +0.030 |
Neg logit contributions | |
iddled | -0.018 |
aback | -0.013 |
ows | -0.012 |
uster | -0.012 |
bench | -0.012 |
Token | long |
Token ID | 890 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.003 |
tant | +0.003 |
bench | +0.003 |
uster | +0.002 |
Neg logit contributions | |
ric | -0.005 |
ost | -0.005 |
highlight | -0.005 |
work | -0.005 |
str | -0.005 |
Token | snow |
Token ID | 6729 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
work | +0.013 |
ric | +0.013 |
ost | +0.012 |
peac | +0.012 |
rob | +0.012 |
Neg logit contributions | |
iddled | -0.014 |
aback | -0.013 |
uster | -0.012 |
tant | -0.012 |
bench | -0.012 |
Token | house |
Token ID | 2156 |
Feature activation | -8.87e-01 |
Pos logit contributions | |
ost | +0.061 |
ric | +0.059 |
ostr | +0.057 |
peac | +0.056 |
work | +0.055 |
Neg logit contributions | |
iddled | -0.056 |
aback | -0.048 |
uster | -0.046 |
tant | -0.043 |
bench | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
iddled | +0.051 |
aback | +0.044 |
uster | +0.042 |
tant | +0.039 |
bench | +0.039 |
Neg logit contributions | |
ost | -0.057 |
ric | -0.055 |
ostr | -0.053 |
peac | -0.052 |
work | -0.051 |
Token | Even |
Token ID | 3412 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
ost | +0.009 |
peac | +0.008 |
ric | +0.008 |
ostr | +0.008 |
flaming | +0.008 |
Neg logit contributions | |
iddled | -0.006 |
aback | -0.005 |
uster | -0.005 |
acks | -0.004 |
tant | -0.004 |
Token | when |
Token ID | 618 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
iddled | +0.001 |
aback | +0.001 |
uster | +0.001 |
tant | +0.001 |
bench | +0.001 |
Neg logit contributions | |
ost | -0.002 |
ric | -0.002 |
peac | -0.002 |
ostr | -0.002 |
work | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
ric | +0.044 |
work | +0.042 |
peac | +0.042 |
ostr | +0.040 |
highlight | +0.040 |
Neg logit contributions | |
breath | -0.013 |
iddled | -0.013 |
ts | -0.013 |
it | -0.012 |
his | -0.011 |
Token | weather |
Token ID | 6193 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
ric | +0.017 |
str | +0.015 |
work | +0.015 |
rob | +0.014 |
peac | +0.014 |
Neg logit contributions | |
tant | -0.011 |
iddled | -0.010 |
uster | -0.010 |
breath | -0.010 |
socks | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | +8.79e-02 |
Pos logit contributions | |
iddled | +0.034 |
aback | +0.030 |
bench | +0.027 |
uster | +0.027 |
ocated | +0.026 |
Neg logit contributions | |
ric | -0.031 |
ost | -0.029 |
peac | -0.028 |
ostr | -0.027 |
highlight | -0.026 |
Token | cold |
Token ID | 4692 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
ost | +0.005 |
ric | +0.005 |
peac | +0.005 |
ostr | +0.005 |
flaming | +0.005 |
Neg logit contributions | |
iddled | -0.005 |
aback | -0.005 |
uster | -0.004 |
bench | -0.004 |
tant | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
iddled | +0.061 |
aback | +0.051 |
uster | +0.050 |
tant | +0.048 |
bench | +0.048 |
Neg logit contributions | |
ost | -0.052 |
ric | -0.048 |
work | -0.047 |
ostr | -0.047 |
peac | -0.046 |
Token | they |
Token ID | 484 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
iddled | +0.010 |
aback | +0.009 |
tant | +0.008 |
uster | +0.008 |
away | +0.007 |
Neg logit contributions | |
ric | -0.019 |
peac | -0.018 |
str | -0.018 |
oral | -0.018 |
work | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
ost | +0.046 |
ric | +0.044 |
ostr | +0.043 |
peac | +0.042 |
moral | +0.041 |
Neg logit contributions | |
iddled | -0.042 |
aback | -0.035 |
uster | -0.035 |
bench | -0.032 |
ocated | -0.032 |
Token | bed |
Token ID | 3996 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
ric | +0.007 |
ost | +0.007 |
work | +0.007 |
peac | +0.007 |
rob | +0.007 |
Neg logit contributions | |
iddled | -0.006 |
uster | -0.005 |
aback | -0.005 |
tant | -0.005 |
ocated | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
iddled | +0.203 |
aback | +0.177 |
uster | +0.169 |
tant | +0.159 |
ocated | +0.157 |
Neg logit contributions | |
ost | -0.162 |
work | -0.155 |
ostr | -0.153 |
ric | -0.152 |
zoo | -0.151 |
Token | she |
Token ID | 673 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
iddled | +0.021 |
aback | +0.019 |
uster | +0.018 |
tant | +0.017 |
bench | +0.017 |
Neg logit contributions | |
ost | -0.018 |
ric | -0.017 |
ostr | -0.017 |
peac | -0.017 |
highlight | -0.016 |
Token | hopped |
Token ID | 45230 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
iddled | +0.010 |
aback | +0.009 |
uster | +0.008 |
tant | +0.008 |
ocated | +0.008 |
Neg logit contributions | |
ost | -0.009 |
ric | -0.009 |
peac | -0.008 |
work | -0.008 |
ostr | -0.008 |
Token | on |
Token ID | 319 |
Feature activation | +7.02e-01 |
Pos logit contributions | |
iddled | +0.057 |
tant | +0.048 |
aback | +0.047 |
uster | +0.047 |
bench | +0.046 |
Neg logit contributions | |
ost | -0.039 |
moral | -0.038 |
ostr | -0.037 |
zoo | -0.036 |
North | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
ost | +0.045 |
ric | +0.042 |
ostr | +0.041 |
peac | +0.040 |
moral | +0.040 |
Neg logit contributions | |
iddled | -0.042 |
aback | -0.035 |
uster | -0.035 |
bench | -0.033 |
ocated | -0.032 |
Token | sofa |
Token ID | 34902 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
ost | +0.012 |
ric | +0.011 |
work | +0.011 |
peac | +0.011 |
rob | +0.011 |
Neg logit contributions | |
iddled | -0.009 |
aback | -0.007 |
uster | -0.007 |
tant | -0.007 |
ocated | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
iddled | +0.162 |
aback | +0.140 |
uster | +0.137 |
tant | +0.130 |
ocated | +0.125 |
Neg logit contributions | |
ost | -0.141 |
ric | -0.134 |
ostr | -0.131 |
work | -0.129 |
zoo | -0.129 |
Token | she |
Token ID | 673 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
iddled | +0.022 |
aback | +0.019 |
uster | +0.018 |
tant | +0.017 |
bench | +0.017 |
Neg logit contributions | |
ost | -0.023 |
ric | -0.022 |
ostr | -0.021 |
peac | -0.021 |
highlight | -0.020 |
Token | hopped |
Token ID | 45230 |
Feature activation | -8.54e-01 |
Pos logit contributions | |
iddled | +0.011 |
aback | +0.010 |
uster | +0.009 |
tant | +0.009 |
ocated | +0.009 |
Neg logit contributions | |
ost | -0.010 |
ric | -0.010 |
peac | -0.009 |
ostr | -0.009 |
work | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
iddled | +0.163 |
aback | +0.146 |
uster | +0.135 |
tant | +0.124 |
ocated | +0.123 |
Neg logit contributions | |
ost | -0.204 |
ric | -0.195 |
peac | -0.192 |
ostr | -0.190 |
flaming | -0.188 |
Token | important |
Token ID | 1593 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
ric | +0.015 |
ost | +0.014 |
ostr | +0.014 |
peac | +0.014 |
flaming | +0.014 |
Neg logit contributions | |
iddled | -0.013 |
aback | -0.013 |
wn | -0.010 |
uster | -0.010 |
away | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
ost | +0.053 |
ric | +0.051 |
ostr | +0.049 |
peac | +0.049 |
work | +0.048 |
Neg logit contributions | |
iddled | -0.049 |
aback | -0.042 |
uster | -0.040 |
tant | -0.038 |
bench | -0.037 |
Token | He |
Token ID | 679 |
Feature activation | -6.20e-02 |
Pos logit contributions | |
aback | +0.003 |
iddled | +0.003 |
uster | +0.002 |
bench | +0.002 |
acks | +0.002 |
Neg logit contributions | |
ost | -0.005 |
peac | -0.005 |
work | -0.005 |
highlight | -0.005 |
ric | -0.005 |
Token | learned |
Token ID | 4499 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
iddled | +0.003 |
aback | +0.003 |
uster | +0.002 |
away | +0.002 |
eal | +0.002 |
Neg logit contributions | |
ric | -0.004 |
ost | -0.004 |
highlight | -0.004 |
flaming | -0.004 |
work | -0.004 |
Token | that |
Token ID | 326 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
ost | +0.094 |
ric | +0.094 |
flaming | +0.090 |
peac | +0.088 |
ostr | +0.088 |
Neg logit contributions | |
iddled | -0.082 |
aback | -0.077 |
uster | -0.067 |
tant | -0.063 |
bench | -0.062 |
Token | day |
Token ID | 1110 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
aback | +0.012 |
iddled | +0.011 |
bench | +0.009 |
ocated | +0.009 |
tant | +0.009 |
Neg logit contributions | |
ric | -0.022 |
ost | -0.020 |
flaming | -0.020 |
ostr | -0.019 |
work | -0.019 |
Token | that |
Token ID | 326 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
ost | +0.094 |
ric | +0.092 |
ostr | +0.087 |
peac | +0.087 |
flaming | +0.086 |
Neg logit contributions | |
iddled | -0.095 |
aback | -0.085 |
uster | -0.080 |
bench | -0.075 |
ocated | -0.075 |
Token | if |
Token ID | 611 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
aback | +0.015 |
iddled | +0.015 |
ocated | +0.013 |
bench | +0.013 |
tant | +0.012 |
Neg logit contributions | |
ric | -0.020 |
ost | -0.017 |
flaming | -0.017 |
ostr | -0.017 |
work | -0.017 |
Token | you |
Token ID | 345 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
aback | +0.061 |
iddled | +0.050 |
ocated | +0.044 |
away | +0.041 |
ly | +0.038 |
Neg logit contributions | |
ric | -0.132 |
work | -0.114 |
flaming | -0.114 |
flower | -0.112 |
peac | -0.112 |
Token | make |
Token ID | 787 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ost | +0.147 |
ric | +0.136 |
flaming | +0.133 |
ostr | +0.133 |
peac | +0.132 |
Neg logit contributions | |
iddled | -0.131 |
aback | -0.122 |
uster | -0.112 |
tant | -0.104 |
bench | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
ost | +0.027 |
ric | +0.027 |
peac | +0.025 |
ostr | +0.025 |
highlight | +0.025 |
Neg logit contributions | |
iddled | -0.021 |
aback | -0.018 |
uster | -0.017 |
tant | -0.016 |
bench | -0.016 |
Token | whispers |
Token ID | 37806 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
iddled | +0.013 |
aback | +0.011 |
uster | +0.010 |
tant | +0.010 |
ocated | +0.009 |
Neg logit contributions | |
ost | -0.017 |
ric | -0.017 |
peac | -0.017 |
highlight | -0.016 |
ostr | -0.016 |
Token | in |
Token ID | 287 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
iddled | +0.092 |
aback | +0.080 |
uster | +0.076 |
tant | +0.071 |
bench | +0.070 |
Neg logit contributions | |
ost | -0.104 |
ric | -0.102 |
ostr | -0.097 |
peac | -0.097 |
work | -0.094 |
Token | their |
Token ID | 511 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
ost | +0.033 |
ric | +0.033 |
ostr | +0.031 |
peac | +0.031 |
work | +0.031 |
Neg logit contributions | |
iddled | -0.031 |
aback | -0.028 |
uster | -0.026 |
tant | -0.025 |
bench | -0.024 |
Token | ears |
Token ID | 11368 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
iddled | +0.014 |
tant | +0.013 |
aback | +0.012 |
uster | +0.012 |
ags | +0.012 |
Neg logit contributions | |
ric | -0.020 |
ost | -0.020 |
peac | -0.020 |
work | -0.020 |
highlight | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
iddled | +0.034 |
aback | +0.029 |
uster | +0.028 |
bench | +0.026 |
tant | +0.026 |
Neg logit contributions | |
ost | -0.047 |
ric | -0.046 |
peac | -0.044 |
ostr | -0.044 |
highlight | -0.043 |
Token | He |
Token ID | 679 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
iddled | +0.018 |
aback | +0.015 |
uster | +0.014 |
bench | +0.014 |
ocated | +0.014 |
Neg logit contributions | |
ost | -0.019 |
ric | -0.018 |
ostr | -0.017 |
peac | -0.017 |
moral | -0.017 |
Token | says |
Token ID | 1139 |
Feature activation | -6.32e-02 |
Pos logit contributions | |
iddled | +0.010 |
aback | +0.008 |
tant | +0.007 |
uster | +0.007 |
irms | +0.007 |
Neg logit contributions | |
ost | -0.014 |
ric | -0.014 |
highlight | -0.013 |
peac | -0.013 |
work | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
iddled | +0.002 |
aback | +0.002 |
tant | +0.002 |
uster | +0.002 |
ocated | +0.002 |
Neg logit contributions | |
ric | -0.005 |
highlight | -0.005 |
work | -0.005 |
colony | -0.004 |
ich | -0.004 |
Token | secret |
Token ID | 3200 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
ric | +0.022 |
ost | +0.020 |
work | +0.020 |
highlight | +0.020 |
orse | +0.020 |
Neg logit contributions | |
iddled | -0.022 |
uster | -0.020 |
bench | -0.020 |
tant | -0.020 |
aback | -0.020 |
Token | is |
Token ID | 318 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
highlight | +0.014 |
ric | +0.014 |
ost | +0.013 |
rob | +0.013 |
str | +0.013 |
Neg logit contributions | |
iddled | -0.012 |
aback | -0.010 |
ows | -0.009 |
bench | -0.009 |
uster | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
iddled | +0.023 |
aback | +0.020 |
bench | +0.016 |
uster | +0.015 |
ocated | +0.015 |
Neg logit contributions | |
ost | -0.044 |
ric | -0.043 |
ostr | -0.042 |
peac | -0.041 |
highlight | -0.041 |
Token | using |
Token ID | 1262 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
iddled | +0.007 |
aback | +0.007 |
uster | +0.006 |
bench | +0.006 |
tant | +0.006 |
Neg logit contributions | |
ric | -0.008 |
ost | -0.008 |
ostr | -0.008 |
work | -0.008 |
oral | -0.007 |
Token | his |
Token ID | 465 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
ost | +0.015 |
ric | +0.014 |
peac | +0.014 |
ostr | +0.013 |
work | +0.013 |
Neg logit contributions | |
iddled | -0.013 |
uster | -0.011 |
aback | -0.011 |
ocated | -0.010 |
bench | -0.010 |
Token | reversing |
Token ID | 37556 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
iddled | +0.073 |
aback | +0.069 |
tant | +0.063 |
uster | +0.061 |
wn | +0.060 |
Neg logit contributions | |
ric | -0.097 |
ost | -0.096 |
ostr | -0.089 |
work | -0.088 |
North | -0.086 |
Token | powers |
Token ID | 5635 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
ost | +0.140 |
ric | +0.135 |
ostr | +0.130 |
peac | +0.129 |
work | +0.126 |
Neg logit contributions | |
iddled | -0.099 |
aback | -0.084 |
uster | -0.079 |
tant | -0.074 |
bench | -0.074 |
Token | to |
Token ID | 284 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
iddled | +0.222 |
aback | +0.194 |
uster | +0.182 |
bench | +0.170 |
tant | +0.170 |
Neg logit contributions | |
ost | -0.260 |
ric | -0.250 |
ostr | -0.241 |
peac | -0.240 |
flaming | -0.232 |
Token | help |
Token ID | 1037 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
iddled | +0.019 |
aback | +0.017 |
uster | +0.016 |
bench | +0.015 |
ocated | +0.015 |
Neg logit contributions | |
ost | -0.019 |
ric | -0.018 |
ostr | -0.018 |
peac | -0.017 |
work | -0.017 |
Token | out |
Token ID | 503 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
iddled | +0.096 |
aback | +0.080 |
uster | +0.077 |
tant | +0.071 |
bench | +0.071 |
Neg logit contributions | |
ost | -0.124 |
ric | -0.119 |
peac | -0.115 |
ostr | -0.115 |
work | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
iddled | +0.074 |
aback | +0.068 |
uster | +0.057 |
tant | +0.057 |
away | +0.053 |
Neg logit contributions | |
ost | -0.116 |
ric | -0.114 |
ostr | -0.112 |
str | -0.110 |
flaming | -0.108 |
Token | He |
Token ID | 679 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
ost | +0.003 |
peac | +0.003 |
ostr | +0.003 |
flaming | +0.003 |
ric | +0.003 |
Neg logit contributions | |
iddled | -0.002 |
aback | -0.002 |
uster | -0.002 |
bench | -0.002 |
acks | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
iddled | +0.011 |
aback | +0.010 |
uster | +0.008 |
bench | +0.008 |
ocated | +0.008 |
Neg logit contributions | |
ost | -0.010 |
ric | -0.009 |
highlight | -0.008 |
flaming | -0.008 |
peac | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
iddled | +0.057 |
aback | +0.049 |
uster | +0.047 |
tant | +0.044 |
bench | +0.044 |
Neg logit contributions | |
ost | -0.059 |
ric | -0.056 |
peac | -0.054 |
ostr | -0.054 |
work | -0.052 |
Token | time |
Token ID | 640 |
Feature activation | -9.00e-01 |
Pos logit contributions | |
iddled | +0.043 |
aback | +0.037 |
uster | +0.036 |
ocated | +0.035 |
bench | +0.033 |
Neg logit contributions | |
ost | -0.025 |
moral | -0.024 |
ric | -0.024 |
ostr | -0.023 |
highlight | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
iddled | +0.053 |
aback | +0.046 |
uster | +0.044 |
tant | +0.041 |
bench | +0.041 |
Neg logit contributions | |
ost | -0.057 |
ric | -0.054 |
ostr | -0.052 |
peac | -0.052 |
work | -0.051 |
Token | there |
Token ID | 612 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ost | +0.004 |
ric | +0.004 |
peac | +0.004 |
ostr | +0.004 |
work | +0.004 |
Neg logit contributions | |
iddled | -0.003 |
aback | -0.003 |
uster | -0.002 |
tant | -0.002 |
ocated | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
iddled | +0.090 |
aback | +0.081 |
uster | +0.079 |
tant | +0.075 |
bench | +0.075 |
Neg logit contributions | |
ost | -0.046 |
ric | -0.043 |
ostr | -0.040 |
peac | -0.040 |
work | -0.038 |
Token | a |
Token ID | 257 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
iddled | +0.003 |
aback | +0.002 |
uster | +0.002 |
tant | +0.002 |
bench | +0.002 |
Neg logit contributions | |
ost | -0.003 |
ric | -0.003 |
peac | -0.002 |
ostr | -0.002 |
work | -0.002 |
Token | little |
Token ID | 1310 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
iddled | +0.025 |
uster | +0.022 |
tant | +0.022 |
aback | +0.020 |
ocated | +0.020 |
Neg logit contributions | |
ric | -0.038 |
ost | -0.036 |
work | -0.034 |
highlight | -0.033 |
peac | -0.031 |
Token | girl |
Token ID | 2576 |
Feature activation | -8.17e-01 |
Pos logit contributions | |
iddled | +0.017 |
aback | +0.015 |
uster | +0.015 |
tant | +0.014 |
bench | +0.014 |
Neg logit contributions | |
ric | -0.014 |
ost | -0.014 |
highlight | -0.013 |
peac | -0.013 |
work | -0.012 |
Token | named |
Token ID | 3706 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
iddled | +0.057 |
aback | +0.051 |
uster | +0.049 |
bench | +0.049 |
tant | +0.047 |
Neg logit contributions | |
ric | -0.038 |
highlight | -0.036 |
ost | -0.036 |
work | -0.032 |
moral | -0.032 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
ost | +0.180 |
ric | +0.173 |
ostr | +0.169 |
peac | +0.168 |
work | +0.164 |
Neg logit contributions | |
iddled | -0.127 |
aback | -0.106 |
uster | -0.101 |
tant | -0.093 |
bench | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
iddled | +0.093 |
aback | +0.080 |
uster | +0.076 |
bench | +0.072 |
tant | +0.071 |
Neg logit contributions | |
ost | -0.109 |
ric | -0.105 |
peac | -0.100 |
ostr | -0.099 |
highlight | -0.098 |
Token | and |
Token ID | 290 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
ost | +0.062 |
ric | +0.061 |
moral | +0.058 |
work | +0.058 |
ostr | +0.057 |
Neg logit contributions | |
iddled | -0.076 |
aback | -0.066 |
uster | -0.064 |
bench | -0.061 |
ocated | -0.061 |
Token | started |
Token ID | 2067 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
iddled | +0.018 |
aback | +0.016 |
uster | +0.015 |
tant | +0.014 |
ocated | +0.014 |
Neg logit contributions | |
ost | -0.016 |
ric | -0.015 |
ostr | -0.015 |
peac | -0.015 |
moral | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
iddled | +0.064 |
aback | +0.055 |
tant | +0.050 |
bench | +0.048 |
uster | +0.047 |
Neg logit contributions | |
ost | -0.092 |
ric | -0.087 |
ostr | -0.086 |
peac | -0.085 |
highlight | -0.084 |
Token | cry |
Token ID | 3960 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
iddled | +0.019 |
aback | +0.017 |
uster | +0.016 |
tant | +0.015 |
bench | +0.015 |
Neg logit contributions | |
ost | -0.016 |
ric | -0.015 |
ostr | -0.015 |
peac | -0.015 |
work | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +9.01e-02 |
Pos logit contributions | |
iddled | +0.041 |
aback | +0.036 |
uster | +0.034 |
bench | +0.032 |
tant | +0.032 |
Neg logit contributions | |
ost | -0.040 |
ric | -0.039 |
ostr | -0.037 |
peac | -0.037 |
work | -0.036 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ost | +0.005 |
ric | +0.005 |
moral | +0.005 |
ostr | +0.005 |
peac | +0.005 |
Neg logit contributions | |
iddled | -0.005 |
aback | -0.005 |
uster | -0.004 |
ocated | -0.004 |
tant | -0.004 |
Token | felt |
Token ID | 2936 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
iddled | +0.087 |
aback | +0.076 |
tant | +0.066 |
ocated | +0.066 |
uster | +0.064 |
Neg logit contributions | |
ost | -0.097 |
ric | -0.096 |
highlight | -0.089 |
peac | -0.087 |
work | -0.086 |
Token | bad |
Token ID | 2089 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
iddled | +0.031 |
uster | +0.027 |
aback | +0.027 |
bench | +0.026 |
ocated | +0.026 |
Neg logit contributions | |
ost | -0.019 |
ric | -0.018 |
ostr | -0.017 |
peac | -0.017 |
moral | -0.017 |
Token | but |
Token ID | 475 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
ost | +0.007 |
ric | +0.007 |
work | +0.007 |
moral | +0.006 |
ostr | +0.006 |
Neg logit contributions | |
iddled | -0.008 |
aback | -0.007 |
uster | -0.007 |
bench | -0.006 |
tant | -0.006 |
Token | she |
Token ID | 673 |
Feature activation | -1.17e-03 |
Pos logit contributions | |
iddled | +0.094 |
aback | +0.081 |
ocated | +0.072 |
uster | +0.070 |
bench | +0.069 |
Neg logit contributions | |
ric | -0.109 |
ost | -0.108 |
ostr | -0.098 |
peac | -0.097 |
flaming | -0.096 |
Token | still |
Token ID | 991 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
iddled | +0.000 |
aback | +0.000 |
uster | +0.000 |
tant | +0.000 |
ocated | +0.000 |
Neg logit contributions | |
ost | -0.000 |
ric | -0.000 |
peac | -0.000 |
ostr | -0.000 |
work | -0.000 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ric | +0.075 |
ost | +0.071 |
highlight | +0.068 |
work | +0.067 |
North | +0.066 |
Neg logit contributions | |
iddled | -0.031 |
aback | -0.027 |
tant | -0.024 |
bench | -0.022 |
shoved | -0.019 |
Token | said |
Token ID | 531 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
iddled | +0.075 |
uster | +0.067 |
aback | +0.064 |
ocated | +0.062 |
wn | +0.060 |
Neg logit contributions | |
ost | -0.058 |
ostr | -0.055 |
peac | -0.054 |
moral | -0.054 |
ric | -0.053 |
Token | , |
Token ID | 11 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
ost | +0.021 |
ric | +0.020 |
work | +0.020 |
ostr | +0.019 |
peac | +0.019 |
Neg logit contributions | |
iddled | -0.028 |
aback | -0.024 |
uster | -0.023 |
tant | -0.022 |
ocated | -0.022 |
Token | rubbing |
Token ID | 27714 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
ost | +0.021 |
ric | +0.020 |
peac | +0.019 |
ostr | +0.019 |
work | +0.018 |
Neg logit contributions | |
iddled | -0.030 |
aback | -0.027 |
uster | -0.026 |
bench | -0.025 |
ocated | -0.025 |
Token | her |
Token ID | 607 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
ost | +0.048 |
ric | +0.048 |
ostr | +0.044 |
flaming | +0.044 |
moral | +0.044 |
Neg logit contributions | |
iddled | -0.039 |
aback | -0.034 |
uster | -0.033 |
ocated | -0.031 |
bench | -0.030 |
Token | belly |
Token ID | 19921 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
iddled | +0.065 |
aback | +0.058 |
uster | +0.054 |
tant | +0.053 |
bench | +0.051 |
Neg logit contributions | |
ost | -0.067 |
ric | -0.064 |
peac | -0.062 |
ostr | -0.061 |
work | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
iddled | +0.048 |
aback | +0.041 |
uster | +0.040 |
ocated | +0.038 |
tant | +0.037 |
Neg logit contributions | |
ost | -0.043 |
ric | -0.041 |
ostr | -0.040 |
peac | -0.039 |
work | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
ost | +0.011 |
ric | +0.011 |
peac | +0.010 |
ostr | +0.010 |
work | +0.010 |
Neg logit contributions | |
iddled | -0.016 |
aback | -0.014 |
uster | -0.014 |
tant | -0.013 |
bench | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
iddled | +0.016 |
aback | +0.015 |
uster | +0.014 |
tant | +0.014 |
ocated | +0.013 |
Neg logit contributions | |
ost | -0.009 |
ric | -0.009 |
ostr | -0.008 |
peac | -0.008 |
moral | -0.008 |
Token | Sam |
Token ID | 16305 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
iddled | +0.010 |
aback | +0.008 |
uster | +0.008 |
tant | +0.008 |
bench | +0.007 |
Neg logit contributions | |
ost | -0.011 |
ric | -0.011 |
peac | -0.010 |
ostr | -0.010 |
moral | -0.010 |
Token | felt |
Token ID | 2936 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
iddled | +0.040 |
aback | +0.035 |
uster | +0.034 |
tant | +0.032 |
bench | +0.032 |
Neg logit contributions | |
ost | -0.033 |
ric | -0.032 |
ostr | -0.031 |
peac | -0.031 |
work | -0.030 |
Token | yellow |
Token ID | 7872 |
Feature activation | +6.21e-01 |
Pos logit contributions | |
ost | +0.046 |
ric | +0.046 |
work | +0.044 |
peac | +0.043 |
highlight | +0.042 |
Neg logit contributions | |
iddled | -0.047 |
uster | -0.041 |
aback | -0.040 |
tant | -0.040 |
bench | -0.038 |
Token | scarf |
Token ID | 36953 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
ost | +0.045 |
ric | +0.043 |
peac | +0.042 |
work | +0.041 |
ostr | +0.041 |
Neg logit contributions | |
iddled | -0.029 |
aback | -0.024 |
uster | -0.024 |
tant | -0.022 |
bench | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
iddled | +0.130 |
aback | +0.118 |
uster | +0.117 |
tant | +0.109 |
ocated | +0.107 |
Neg logit contributions | |
ost | -0.061 |
ric | -0.057 |
ostr | -0.056 |
peac | -0.054 |
work | -0.053 |
Token | soft |
Token ID | 2705 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
ost | +0.012 |
ric | +0.011 |
ostr | +0.011 |
peac | +0.011 |
moral | +0.011 |
Neg logit contributions | |
iddled | -0.016 |
aback | -0.014 |
uster | -0.013 |
ocated | -0.013 |
tant | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
iddled | +0.192 |
uster | +0.163 |
aback | +0.160 |
tant | +0.155 |
wn | +0.152 |
Neg logit contributions | |
work | -0.105 |
ost | -0.103 |
ostr | -0.100 |
North | -0.098 |
zoo | -0.096 |
Token | warm |
Token ID | 5814 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
iddled | +0.014 |
aback | +0.013 |
uster | +0.012 |
tant | +0.012 |
ocated | +0.011 |
Neg logit contributions | |
ost | -0.011 |
ric | -0.011 |
ostr | -0.011 |
peac | -0.010 |
moral | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
iddled | +0.150 |
uster | +0.128 |
aback | +0.126 |
tant | +0.122 |
ocated | +0.118 |
Neg logit contributions | |
ost | -0.094 |
work | -0.093 |
ostr | -0.088 |
ric | -0.087 |
hipp | -0.082 |
Token | His |
Token ID | 2399 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
ost | +0.002 |
ric | +0.002 |
ostr | +0.002 |
moral | +0.002 |
peac | +0.002 |
Neg logit contributions | |
iddled | -0.002 |
aback | -0.002 |
uster | -0.002 |
ocated | -0.002 |
tant | -0.002 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
ost | +0.054 |
ric | +0.053 |
work | +0.051 |
peac | +0.051 |
highlight | +0.050 |
Neg logit contributions | |
iddled | -0.037 |
aback | -0.030 |
uster | -0.030 |
tant | -0.029 |
bench | -0.028 |
Token | gave |
Token ID | 2921 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
ost | +0.132 |
ric | +0.126 |
ostr | +0.122 |
peac | +0.121 |
work | +0.118 |
Neg logit contributions | |
iddled | -0.151 |
aback | -0.133 |
uster | -0.129 |
tant | -0.122 |
bench | -0.120 |
Token | it |
Token ID | 340 |
Feature activation | -9.83e-01 |
Pos logit contributions | |
ost | +0.021 |
ostr | +0.020 |
ric | +0.019 |
hipp | +0.018 |
flaming | +0.018 |
Neg logit contributions | |
iddled | -0.019 |
uster | -0.016 |
aback | -0.015 |
ocated | -0.015 |
bench | -0.015 |
Token | is |
Token ID | 318 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
iddled | +0.020 |
uster | +0.019 |
bench | +0.017 |
aback | +0.017 |
wn | +0.017 |
Neg logit contributions | |
ost | -0.013 |
ostr | -0.012 |
moral | -0.012 |
ric | -0.012 |
zoo | -0.011 |
Token | not |
Token ID | 407 |
Feature activation | +7.85e-01 |
Pos logit contributions | |
iddled | +0.013 |
aback | +0.011 |
uster | +0.011 |
tant | +0.011 |
bench | +0.011 |
Neg logit contributions | |
ost | -0.008 |
ric | -0.008 |
ostr | -0.007 |
peac | -0.007 |
moral | -0.007 |
Token | scared |
Token ID | 12008 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
ost | +0.032 |
ostr | +0.029 |
flaming | +0.027 |
rob | +0.027 |
peac | +0.027 |
Neg logit contributions | |
iddled | -0.066 |
uster | -0.058 |
tant | -0.057 |
aback | -0.054 |
ocated | -0.053 |
Token | of |
Token ID | 286 |
Feature activation | +9.73e-01 |
Pos logit contributions | |
ost | +0.016 |
ric | +0.014 |
ostr | +0.014 |
moral | +0.014 |
peac | +0.014 |
Neg logit contributions | |
iddled | -0.023 |
aback | -0.019 |
uster | -0.018 |
bench | -0.018 |
ocated | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
ost | +0.059 |
ric | +0.058 |
peac | +0.055 |
ostr | +0.054 |
work | +0.054 |
Neg logit contributions | |
iddled | -0.058 |
aback | -0.051 |
uster | -0.048 |
tant | -0.045 |
ocated | -0.045 |
Token | dog |
Token ID | 3290 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
ost | +0.035 |
ric | +0.032 |
ostr | +0.031 |
peac | +0.031 |
moral | +0.030 |
Neg logit contributions | |
iddled | -0.057 |
aback | -0.051 |
uster | -0.049 |
bench | -0.047 |
ocated | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +8.38e-02 |
Pos logit contributions | |
iddled | +0.023 |
aback | +0.021 |
uster | +0.019 |
ocated | +0.019 |
tant | +0.019 |
Neg logit contributions | |
ric | -0.036 |
ost | -0.036 |
peac | -0.034 |
highlight | -0.033 |
ostr | -0.033 |
Token | He |
Token ID | 679 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
ost | +0.005 |
ric | +0.005 |
ostr | +0.005 |
peac | +0.004 |
rob | +0.004 |
Neg logit contributions | |
iddled | -0.005 |
uster | -0.004 |
aback | -0.004 |
bench | -0.004 |
ocated | -0.004 |
Token | walks |
Token ID | 11114 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
iddled | +0.010 |
aback | +0.009 |
uster | +0.008 |
tant | +0.008 |
ocated | +0.007 |
Neg logit contributions | |
ost | -0.011 |
ric | -0.011 |
peac | -0.010 |
ostr | -0.010 |
work | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
iddled | +0.030 |
uster | +0.026 |
aback | +0.023 |
bench | +0.022 |
ocated | +0.022 |
Neg logit contributions | |
ost | -0.031 |
ostr | -0.028 |
moral | -0.028 |
zoo | -0.028 |
work | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +6.91e-01 |
Pos logit contributions | |
ost | +0.001 |
ric | +0.001 |
ostr | +0.001 |
peac | +0.001 |
work | +0.001 |
Neg logit contributions | |
iddled | -0.001 |
aback | -0.001 |
uster | -0.001 |
tant | -0.001 |
bench | -0.001 |
Token | with |
Token ID | 351 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
moral | +0.045 |
ost | +0.043 |
zoo | +0.042 |
hipp | +0.041 |
work | +0.041 |
Neg logit contributions | |
iddled | -0.062 |
uster | -0.056 |
aback | -0.053 |
wn | -0.048 |
bench | -0.048 |
Token | him |
Token ID | 683 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
iddled | +0.049 |
aback | +0.042 |
uster | +0.040 |
tant | +0.037 |
bench | +0.037 |
Neg logit contributions | |
ost | -0.068 |
ric | -0.065 |
ostr | -0.063 |
peac | -0.063 |
work | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -7.48e-02 |
Pos logit contributions | |
iddled | +0.092 |
aback | +0.079 |
uster | +0.072 |
tant | +0.072 |
ocated | +0.069 |
Neg logit contributions | |
ost | -0.133 |
ric | -0.127 |
peac | -0.126 |
ostr | -0.123 |
rob | -0.121 |
Token | Then |
Token ID | 3244 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.004 |
uster | +0.004 |
tant | +0.004 |
bench | +0.003 |
Neg logit contributions | |
ost | -0.005 |
ric | -0.004 |
peac | -0.004 |
ostr | -0.004 |
highlight | -0.004 |
Token | an |
Token ID | 281 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
ric | +0.102 |
ost | +0.095 |
highlight | +0.092 |
work | +0.091 |
ostr | +0.089 |
Neg logit contributions | |
iddled | -0.040 |
aback | -0.038 |
away | -0.029 |
tant | -0.026 |
ocated | -0.026 |
Token | idea |
Token ID | 2126 |
Feature activation | -5.34e+00 |
Pos logit contributions | |
iddled | +0.041 |
aback | +0.035 |
uster | +0.034 |
tant | +0.032 |
bench | +0.031 |
Neg logit contributions | |
ost | -0.043 |
ric | -0.042 |
peac | -0.039 |
ostr | -0.039 |
work | -0.039 |
Token | came |
Token ID | 1625 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
iddled | +0.356 |
uster | +0.324 |
aback | +0.319 |
tant | +0.309 |
bench | +0.291 |
Neg logit contributions | |
ost | -0.280 |
ostr | -0.260 |
peac | -0.257 |
work | -0.255 |
hipp | -0.241 |
Token | to |
Token ID | 284 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
ost | +0.029 |
ric | +0.028 |
ostr | +0.027 |
peac | +0.027 |
highlight | +0.027 |
Neg logit contributions | |
iddled | -0.018 |
aback | -0.016 |
uster | -0.014 |
ocated | -0.012 |
bench | -0.012 |
Token | him |
Token ID | 683 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
iddled | +0.023 |
aback | +0.020 |
uster | +0.018 |
tant | +0.017 |
bench | +0.017 |
Neg logit contributions | |
ost | -0.028 |
ric | -0.027 |
ostr | -0.026 |
peac | -0.026 |
highlight | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
iddled | +0.087 |
aback | +0.076 |
uster | +0.069 |
bench | +0.064 |
ocated | +0.063 |
Neg logit contributions | |
ost | -0.139 |
ric | -0.131 |
peac | -0.129 |
ostr | -0.128 |
flaming | -0.126 |
Token | He |
Token ID | 679 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
iddled | +0.011 |
uster | +0.009 |
aback | +0.009 |
tant | +0.009 |
wn | +0.009 |
Neg logit contributions | |
ost | -0.009 |
ric | -0.008 |
ostr | -0.008 |
work | -0.008 |
peac | -0.008 |
Token | desk |
Token ID | 6915 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
ric | +0.030 |
ost | +0.030 |
peac | +0.027 |
str | +0.026 |
orse | +0.026 |
Neg logit contributions | |
iddled | -0.019 |
aback | -0.017 |
tant | -0.017 |
uster | -0.017 |
ocated | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
ost | +0.026 |
ric | +0.025 |
peac | +0.024 |
ostr | +0.024 |
highlight | +0.023 |
Neg logit contributions | |
iddled | -0.021 |
aback | -0.018 |
uster | -0.017 |
bench | -0.016 |
ocated | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
iddled | +0.002 |
aback | +0.002 |
uster | +0.001 |
ocated | +0.001 |
tant | +0.001 |
Neg logit contributions | |
ost | -0.004 |
ric | -0.004 |
highlight | -0.003 |
str | -0.003 |
peac | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
iddled | +0.010 |
aback | +0.009 |
ocated | +0.008 |
tant | +0.008 |
uster | +0.008 |
Neg logit contributions | |
ric | -0.014 |
ost | -0.014 |
str | -0.013 |
ostr | -0.013 |
peac | -0.013 |
Token | wet |
Token ID | 9583 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
tant | +0.018 |
iddled | +0.017 |
aback | +0.016 |
knot | +0.015 |
uster | +0.015 |
Neg logit contributions | |
ric | -0.028 |
ost | -0.028 |
work | -0.026 |
highlight | -0.025 |
riage | -0.024 |
Token | spot |
Token ID | 4136 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
iddled | +0.054 |
aback | +0.048 |
uster | +0.046 |
tant | +0.044 |
bench | +0.043 |
Neg logit contributions | |
ost | -0.062 |
ric | -0.059 |
highlight | -0.055 |
peac | -0.055 |
ostr | -0.055 |
Token | on |
Token ID | 319 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
iddled | +0.280 |
aback | +0.241 |
uster | +0.226 |
bench | +0.215 |
tant | +0.211 |
Neg logit contributions | |
ost | -0.352 |
ric | -0.342 |
peac | -0.327 |
ostr | -0.326 |
highlight | -0.317 |
Token | the |
Token ID | 262 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
ost | +0.054 |
ric | +0.052 |
ostr | +0.050 |
peac | +0.049 |
work | +0.048 |
Neg logit contributions | |
iddled | -0.052 |
aback | -0.046 |
uster | -0.044 |
tant | -0.041 |
bench | -0.041 |
Token | rug |
Token ID | 14477 |
Feature activation | -6.71e-02 |
Pos logit contributions | |
ric | +0.031 |
ost | +0.031 |
peac | +0.028 |
rob | +0.027 |
str | +0.027 |
Neg logit contributions | |
iddled | -0.021 |
aback | -0.019 |
tant | -0.019 |
uster | -0.019 |
ocated | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
iddled | +0.004 |
aback | +0.003 |
uster | +0.003 |
bench | +0.003 |
ocated | +0.003 |
Neg logit contributions | |
ost | -0.005 |
ric | -0.004 |
peac | -0.004 |
ostr | -0.004 |
rob | -0.004 |
Token | Who |
Token ID | 5338 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
ost | +0.002 |
ric | +0.002 |
peac | +0.002 |
ostr | +0.002 |
highlight | +0.002 |
Neg logit contributions | |
iddled | -0.002 |
aback | -0.002 |
uster | -0.002 |
ocated | -0.002 |
zip | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
ost | +0.129 |
ric | +0.124 |
ostr | +0.119 |
peac | +0.118 |
work | +0.117 |
Neg logit contributions | |
iddled | -0.113 |
aback | -0.097 |
uster | -0.093 |
bench | -0.086 |
tant | -0.086 |
Token | but |
Token ID | 475 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
iddled | +0.008 |
aback | +0.007 |
tant | +0.006 |
uster | +0.006 |
bench | +0.006 |
Neg logit contributions | |
ric | -0.015 |
ost | -0.015 |
peac | -0.014 |
work | -0.013 |
ostr | -0.013 |
Token | Jamie |
Token ID | 17826 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
iddled | +0.088 |
aback | +0.074 |
ocated | +0.060 |
bench | +0.059 |
uster | +0.058 |
Neg logit contributions | |
ric | -0.115 |
ost | -0.109 |
peac | -0.106 |
flaming | -0.103 |
ostr | -0.101 |
Token | had |
Token ID | 550 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
iddled | +0.043 |
aback | +0.038 |
bench | +0.034 |
uster | +0.033 |
tant | +0.033 |
Neg logit contributions | |
ric | -0.054 |
ost | -0.054 |
peac | -0.049 |
highlight | -0.049 |
North | -0.048 |
Token | an |
Token ID | 281 |
Feature activation | -5.53e-01 |
Pos logit contributions | |
ost | +0.175 |
ric | +0.170 |
ostr | +0.160 |
peac | +0.156 |
work | +0.156 |
Neg logit contributions | |
iddled | -0.200 |
aback | -0.181 |
uster | -0.164 |
tant | -0.163 |
bench | -0.161 |
Token | idea |
Token ID | 2126 |
Feature activation | -5.38e+00 |
Pos logit contributions | |
iddled | +0.030 |
aback | +0.025 |
uster | +0.024 |
ocated | +0.023 |
tant | +0.022 |
Neg logit contributions | |
ost | -0.036 |
ostr | -0.035 |
moral | -0.035 |
ric | -0.034 |
peac | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
iddled | +0.333 |
aback | +0.290 |
uster | +0.282 |
tant | +0.266 |
bench | +0.260 |
Neg logit contributions | |
ost | -0.320 |
ric | -0.303 |
ostr | -0.293 |
peac | -0.291 |
work | -0.285 |
Token |
|
Token ID | 198 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
ost | +0.003 |
ric | +0.002 |
ostr | +0.002 |
peac | +0.002 |
work | +0.002 |
Neg logit contributions | |
iddled | -0.003 |
aback | -0.002 |
uster | -0.002 |
tant | -0.002 |
ocated | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
iddled | +0.028 |
uster | +0.025 |
aback | +0.025 |
tant | +0.025 |
wn | +0.024 |
Neg logit contributions | |
ost | -0.012 |
ric | -0.011 |
moral | -0.011 |
ostr | -0.010 |
peac | -0.010 |
Token | He |
Token ID | 1544 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
iddled | +0.018 |
aback | +0.016 |
uster | +0.016 |
tant | +0.016 |
ocated | +0.015 |
Neg logit contributions | |
ost | -0.017 |
moral | -0.016 |
ric | -0.016 |
ostr | -0.016 |
peac | -0.016 |
Token | said |
Token ID | 531 |
Feature activation | -9.95e-02 |
Pos logit contributions | |
ost | +0.024 |
ric | +0.023 |
ostr | +0.023 |
peac | +0.023 |
moral | +0.022 |
Neg logit contributions | |
iddled | -0.026 |
aback | -0.023 |
uster | -0.022 |
tant | -0.021 |
bench | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
iddled | +0.043 |
aback | +0.037 |
uster | +0.035 |
bench | +0.033 |
tant | +0.032 |
Neg logit contributions | |
ost | -0.058 |
ric | -0.057 |
peac | -0.055 |
ostr | -0.055 |
rob | -0.054 |
Token | put |
Token ID | 1234 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
iddled | +0.025 |
aback | +0.022 |
uster | +0.021 |
tant | +0.020 |
ocated | +0.020 |
Neg logit contributions | |
ost | -0.018 |
ric | -0.017 |
ostr | -0.017 |
peac | -0.016 |
moral | -0.016 |
Token | it |
Token ID | 340 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
iddled | +0.165 |
aback | +0.145 |
uster | +0.143 |
ocated | +0.139 |
tant | +0.139 |
Neg logit contributions | |
ost | -0.137 |
ric | -0.127 |
peac | -0.123 |
work | -0.121 |
ostr | -0.121 |
Token | on |
Token ID | 319 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
iddled | +0.069 |
aback | +0.061 |
uster | +0.058 |
tant | +0.055 |
bench | +0.054 |
Neg logit contributions | |
ost | -0.067 |
ric | -0.064 |
peac | -0.061 |
ostr | -0.061 |
work | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
ost | +0.046 |
ric | +0.045 |
ostr | +0.043 |
peac | +0.043 |
work | +0.042 |
Neg logit contributions | |
iddled | -0.038 |
aback | -0.033 |
uster | -0.031 |
tant | -0.029 |
bench | -0.029 |
Token | First |
Token ID | 3274 |
Feature activation | -5.94e+00 |
Pos logit contributions | |
ost | +0.005 |
peac | +0.005 |
ric | +0.005 |
work | +0.005 |
highlight | +0.005 |
Neg logit contributions | |
iddled | -0.004 |
aback | -0.004 |
uster | -0.003 |
ocated | -0.003 |
bench | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
iddled | +0.314 |
aback | +0.270 |
uster | +0.254 |
bench | +0.237 |
ocated | +0.235 |
Neg logit contributions | |
ost | -0.400 |
ric | -0.399 |
peac | -0.375 |
work | -0.370 |
highlight | -0.369 |
Token | her |
Token ID | 607 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
iddled | +0.006 |
aback | +0.005 |
uster | +0.005 |
ocated | +0.005 |
tant | +0.005 |
Neg logit contributions | |
ric | -0.007 |
ost | -0.006 |
peac | -0.006 |
work | -0.006 |
moral | -0.006 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
iddled | +0.103 |
uster | +0.088 |
aback | +0.087 |
tant | +0.085 |
ocated | +0.078 |
Neg logit contributions | |
ost | -0.092 |
ric | -0.090 |
peac | -0.080 |
work | -0.079 |
ostr | -0.078 |
Token | helped |
Token ID | 4193 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
ost | +0.145 |
ric | +0.140 |
ostr | +0.133 |
peac | +0.133 |
work | +0.129 |
Neg logit contributions | |
iddled | -0.144 |
aback | -0.126 |
uster | -0.120 |
tant | -0.113 |
bench | -0.113 |
Token | her |
Token ID | 607 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
iddled | +0.133 |
aback | +0.109 |
uster | +0.109 |
bench | +0.103 |
tant | +0.102 |
Neg logit contributions | |
ost | -0.193 |
ostr | -0.179 |
ric | -0.174 |
rob | -0.169 |
work | -0.163 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
ost | +0.002 |
ric | +0.002 |
ostr | +0.002 |
peac | +0.002 |
work | +0.002 |
Neg logit contributions | |
iddled | -0.002 |
aback | -0.002 |
uster | -0.002 |
tant | -0.002 |
bench | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
ost | +0.090 |
peac | +0.083 |
ric | +0.083 |
ostr | +0.083 |
moral | +0.080 |
Neg logit contributions | |
iddled | -0.103 |
aback | -0.088 |
uster | -0.088 |
bench | -0.083 |
tant | -0.082 |
Token | Molly |
Token ID | 30236 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
iddled | +0.001 |
aback | +0.001 |
uster | +0.001 |
tant | +0.001 |
bench | +0.001 |
Neg logit contributions | |
ost | -0.002 |
ric | -0.001 |
ostr | -0.001 |
peac | -0.001 |
work | -0.001 |
Token | had |
Token ID | 550 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
iddled | +0.117 |
aback | +0.100 |
ocated | +0.091 |
uster | +0.090 |
tant | +0.087 |
Neg logit contributions | |
ric | -0.146 |
ost | -0.143 |
highlight | -0.137 |
work | -0.135 |
riage | -0.133 |
Token | an |
Token ID | 281 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
ost | +0.182 |
ric | +0.175 |
ostr | +0.167 |
peac | +0.166 |
work | +0.162 |
Neg logit contributions | |
iddled | -0.205 |
aback | -0.181 |
uster | -0.173 |
tant | -0.163 |
bench | -0.162 |
Token | idea |
Token ID | 2126 |
Feature activation | -5.30e+00 |
Pos logit contributions | |
iddled | +0.025 |
aback | +0.021 |
uster | +0.020 |
tant | +0.019 |
ocated | +0.019 |
Neg logit contributions | |
ost | -0.030 |
ostr | -0.029 |
moral | -0.029 |
peac | -0.027 |
ric | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
iddled | +0.331 |
aback | +0.287 |
uster | +0.282 |
tant | +0.270 |
bench | +0.259 |
Neg logit contributions | |
ost | -0.314 |
ric | -0.293 |
ostr | -0.287 |
peac | -0.284 |
work | -0.277 |
Token | She |
Token ID | 1375 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
ost | +0.001 |
ric | +0.001 |
moral | +0.001 |
peac | +0.001 |
ostr | +0.001 |
Neg logit contributions | |
iddled | -0.001 |
uster | -0.001 |
tant | -0.001 |
aback | -0.001 |
wn | -0.001 |
Token | went |
Token ID | 1816 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
ost | +0.004 |
ric | +0.004 |
ostr | +0.003 |
peac | +0.003 |
work | +0.003 |
Neg logit contributions | |
iddled | -0.003 |
aback | -0.003 |
uster | -0.003 |
tant | -0.003 |
bench | -0.003 |
Token | inside |
Token ID | 2641 |
Feature activation | +7.78e-01 |
Pos logit contributions | |
ost | +0.023 |
ric | +0.021 |
ostr | +0.021 |
peac | +0.021 |
work | +0.020 |
Neg logit contributions | |
iddled | -0.024 |
aback | -0.020 |
uster | -0.020 |
tant | -0.019 |
bench | -0.019 |
Token | her |
Token ID | 607 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
ost | +0.043 |
ric | +0.042 |
moral | +0.041 |
ostr | +0.040 |
work | +0.039 |
Neg logit contributions | |
iddled | -0.049 |
uster | -0.043 |
aback | -0.041 |
ocated | -0.040 |
wn | -0.039 |