Token | to |
Token ID | 284 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
ows | +0.085 |
pering | +0.084 |
obb | +0.083 |
arl | +0.081 |
iving | +0.079 |
Neg logit contributions | |
penny | -0.055 |
flavor | -0.054 |
ornament | -0.052 |
peanut | -0.052 |
cube | -0.051 |
Token | Africa |
Token ID | 5478 |
Feature activation | +5.11e+00 |
Pos logit contributions | |
ows | +0.033 |
pering | +0.031 |
obb | +0.031 |
arl | +0.030 |
iving | +0.029 |
Neg logit contributions | |
flavor | -0.027 |
penny | -0.027 |
ornament | -0.026 |
peanut | -0.026 |
cube | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
pering | +0.190 |
arl | +0.189 |
obb | +0.189 |
sons | +0.184 |
ows | +0.176 |
Neg logit contributions | |
flavor | -0.125 |
ornament | -0.120 |
penny | -0.115 |
nut | -0.109 |
peanut | -0.109 |
Token | see |
Token ID | 766 |
Feature activation | +6.23e-02 |
Pos logit contributions | |
penny | +0.007 |
flavor | +0.007 |
ornament | +0.007 |
peanut | +0.006 |
cube | +0.006 |
Neg logit contributions | |
ows | -0.006 |
pering | -0.006 |
obb | -0.006 |
arl | -0.006 |
iving | -0.006 |
Token | more |
Token ID | 517 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
ows | +0.002 |
obb | +0.002 |
pering | +0.002 |
arl | +0.002 |
iving | +0.002 |
Neg logit contributions | |
flavor | -0.002 |
ornament | -0.002 |
penny | -0.002 |
peanut | -0.002 |
cube | -0.001 |
Token | ze |
Token ID | 41271 |
Feature activation | +9.39e+00 |
Pos logit contributions | |
flavor | +0.022 |
peanut | +0.021 |
penny | +0.021 |
ornament | +0.020 |
color | +0.020 |
Neg logit contributions | |
obb | -0.026 |
pering | -0.024 |
ows | -0.024 |
arl | -0.024 |
adders | -0.022 |
Token | br |
Token ID | 1671 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
pering | +0.411 |
sons | +0.404 |
realize | +0.397 |
obb | +0.397 |
icia | +0.387 |
Neg logit contributions | |
flavor | -0.131 |
nut | -0.128 |
penny | -0.127 |
peanut | -0.126 |
ornament | -0.122 |
Token | as |
Token ID | 292 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
pering | +0.013 |
obb | +0.012 |
ows | +0.011 |
announcements | +0.009 |
arl | +0.009 |
Neg logit contributions | |
flavor | -0.058 |
penny | -0.057 |
ornament | -0.056 |
cube | -0.056 |
peanut | -0.056 |
Token | ." |
Token ID | 526 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
obb | +0.159 |
ows | +0.140 |
pering | +0.140 |
arl | +0.138 |
rig | +0.138 |
Neg logit contributions | |
ornament | -0.101 |
penny | -0.099 |
cube | -0.096 |
flavor | -0.092 |
c | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | -8.50e-02 |
Pos logit contributions | |
penny | +0.003 |
flavor | +0.003 |
ornament | +0.003 |
cube | +0.003 |
peanut | +0.003 |
Neg logit contributions | |
ows | -0.004 |
pering | -0.004 |
obb | -0.004 |
arl | -0.004 |
iving | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
flavor | +0.002 |
penny | +0.002 |
ornament | +0.002 |
peanut | +0.002 |
cube | +0.001 |
Neg logit contributions | |
ows | -0.004 |
pering | -0.004 |
obb | -0.004 |
iving | -0.003 |
oms | -0.003 |
Token | did |
Token ID | 750 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
pering | +0.060 |
ows | +0.059 |
arl | +0.058 |
obb | +0.056 |
iving | +0.056 |
Neg logit contributions | |
flavor | -0.043 |
penny | -0.043 |
peanut | -0.041 |
cube | -0.040 |
ornament | -0.040 |
Token | not |
Token ID | 407 |
Feature activation | +9.73e-01 |
Pos logit contributions | |
flavor | +0.010 |
penny | +0.010 |
cube | +0.009 |
peanut | +0.009 |
ornament | +0.009 |
Neg logit contributions | |
arl | -0.018 |
pering | -0.017 |
ows | -0.017 |
iving | -0.017 |
obb | -0.017 |
Token | see |
Token ID | 766 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
ows | +0.025 |
pering | +0.025 |
obb | +0.024 |
arl | +0.023 |
oms | +0.022 |
Neg logit contributions | |
flavor | -0.036 |
penny | -0.036 |
peanut | -0.035 |
ornament | -0.034 |
cube | -0.034 |
Token | any |
Token ID | 597 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
ows | +0.034 |
pering | +0.033 |
arl | +0.033 |
iving | +0.033 |
obb | +0.032 |
Neg logit contributions | |
penny | -0.024 |
flavor | -0.024 |
peanut | -0.023 |
ornament | -0.023 |
cube | -0.022 |
Token | baby |
Token ID | 5156 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
ows | +0.048 |
pering | +0.047 |
obb | +0.046 |
arl | +0.045 |
iving | +0.043 |
Neg logit contributions | |
flavor | -0.044 |
penny | -0.044 |
ornament | -0.043 |
peanut | -0.042 |
cube | -0.042 |
Token | ze |
Token ID | 41271 |
Feature activation | +9.03e+00 |
Pos logit contributions | |
arl | +0.010 |
obb | +0.010 |
ows | +0.009 |
pering | +0.009 |
uits | +0.009 |
Neg logit contributions | |
penny | -0.012 |
flavor | -0.011 |
ornament | -0.011 |
peanut | -0.011 |
cube | -0.011 |
Token | br |
Token ID | 1671 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
sons | +0.396 |
pering | +0.393 |
ows | +0.375 |
realize | +0.373 |
arl | +0.373 |
Neg logit contributions | |
nut | -0.126 |
flavor | -0.125 |
penny | -0.123 |
on | -0.121 |
peanut | -0.121 |
Token | as |
Token ID | 292 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
pering | +0.004 |
ows | +0.004 |
obb | +0.004 |
arl | +0.003 |
iving | +0.003 |
Neg logit contributions | |
flavor | -0.025 |
penny | -0.025 |
ornament | -0.024 |
cube | -0.024 |
peanut | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +6.16e-02 |
Pos logit contributions | |
obb | +0.164 |
ows | +0.149 |
announcements | +0.146 |
rig | +0.146 |
arl | +0.146 |
Neg logit contributions | |
penny | -0.092 |
cube | -0.089 |
ornament | -0.088 |
flavor | -0.081 |
peanut | -0.081 |
Token |
|
Token ID | 198 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
pering | +0.002 |
ows | +0.002 |
obb | +0.002 |
iving | +0.002 |
arl | +0.002 |
Neg logit contributions | |
flavor | -0.002 |
peanut | -0.001 |
penny | -0.001 |
cube | -0.001 |
ornament | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
flavor | +0.008 |
penny | +0.007 |
peanut | +0.007 |
cube | +0.007 |
ornament | +0.007 |
Neg logit contributions | |
obb | -0.022 |
pering | -0.022 |
ows | -0.022 |
iving | -0.022 |
arl | -0.021 |
Token | he |
Token ID | 339 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
ows | +0.213 |
arl | +0.202 |
obb | +0.196 |
pering | +0.191 |
adders | +0.186 |
Neg logit contributions | |
penny | -0.099 |
flavor | -0.099 |
cube | -0.091 |
peanut | -0.091 |
ornament | -0.089 |
Token | was |
Token ID | 373 |
Feature activation | -9.80e-02 |
Pos logit contributions | |
ows | +0.124 |
arl | +0.124 |
pering | +0.122 |
obb | +0.117 |
adders | +0.117 |
Neg logit contributions | |
flavor | -0.049 |
penny | -0.049 |
peanut | -0.045 |
cube | -0.042 |
ornament | -0.042 |
Token | playing |
Token ID | 2712 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
penny | +0.002 |
flavor | +0.002 |
ornament | +0.002 |
peanut | +0.002 |
cube | +0.002 |
Neg logit contributions | |
ows | -0.004 |
arl | -0.004 |
pering | -0.003 |
obb | -0.003 |
oms | -0.003 |
Token | make |
Token ID | 787 |
Feature activation | +6.78e-01 |
Pos logit contributions | |
arl | +0.090 |
ows | +0.088 |
pering | +0.088 |
obb | +0.086 |
oms | +0.085 |
Neg logit contributions | |
ornament | -0.037 |
flavor | -0.036 |
penny | -0.036 |
cube | -0.035 |
peanut | -0.031 |
Token | - |
Token ID | 12 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
arl | +0.029 |
pering | +0.028 |
obb | +0.028 |
ows | +0.026 |
oms | +0.026 |
Neg logit contributions | |
flavor | -0.017 |
ornament | -0.015 |
penny | -0.015 |
peanut | -0.015 |
cube | -0.015 |
Token | bel |
Token ID | 6667 |
Feature activation | +8.90e+00 |
Pos logit contributions | |
pering | +0.021 |
ows | +0.021 |
obb | +0.020 |
arl | +0.019 |
iving | +0.018 |
Neg logit contributions | |
flavor | -0.044 |
penny | -0.044 |
ornament | -0.044 |
cube | -0.043 |
peanut | -0.043 |
Token | ieve |
Token ID | 12311 |
Feature activation | +8.99e-01 |
Pos logit contributions | |
pering | +0.174 |
ows | +0.167 |
arl | +0.165 |
iving | +0.145 |
obb | +0.143 |
Neg logit contributions | |
ornament | -0.388 |
flavor | -0.382 |
penny | -0.379 |
cube | -0.365 |
pencil | -0.349 |
Token | at |
Token ID | 379 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
arl | +0.036 |
pering | +0.035 |
obb | +0.035 |
ows | +0.033 |
ighters | +0.032 |
Neg logit contributions | |
flavor | -0.023 |
penny | -0.023 |
ornament | -0.023 |
cube | -0.022 |
pencil | -0.020 |
Token | home |
Token ID | 1363 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
pering | +0.031 |
ows | +0.030 |
obb | +0.029 |
oms | +0.028 |
arl | +0.028 |
Neg logit contributions | |
penny | -0.036 |
flavor | -0.036 |
peanut | -0.035 |
ornament | -0.035 |
cube | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
flavor | +0.027 |
penny | +0.025 |
cube | +0.025 |
ornament | +0.025 |
color | +0.023 |
Neg logit contributions | |
arl | -0.045 |
obb | -0.044 |
pering | -0.042 |
oms | -0.041 |
ighters | -0.040 |
Token | too |
Token ID | 1165 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
ows | +0.021 |
obb | +0.021 |
arl | +0.021 |
pering | +0.020 |
oms | +0.019 |
Neg logit contributions | |
penny | -0.017 |
flavor | -0.017 |
ornament | -0.016 |
peanut | -0.016 |
cube | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
arl | +0.004 |
pering | +0.004 |
obb | +0.004 |
ows | +0.004 |
adders | +0.004 |
Neg logit contributions | |
flavor | -0.002 |
penny | -0.002 |
peanut | -0.002 |
ornament | -0.002 |
cube | -0.002 |
Token | ready |
Token ID | 3492 |
Feature activation | +9.18e-01 |
Pos logit contributions | |
ows | +0.002 |
arl | +0.002 |
pering | +0.002 |
obb | +0.002 |
iving | +0.002 |
Neg logit contributions | |
penny | -0.002 |
flavor | -0.002 |
ornament | -0.002 |
peanut | -0.002 |
cube | -0.002 |
Token | for |
Token ID | 329 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
obb | +0.042 |
pering | +0.041 |
ows | +0.041 |
ighters | +0.041 |
arl | +0.040 |
Neg logit contributions | |
flavor | -0.016 |
penny | -0.015 |
cube | -0.014 |
peanut | -0.013 |
ornament | -0.013 |
Token | make |
Token ID | 787 |
Feature activation | +6.93e-01 |
Pos logit contributions | |
ows | +0.058 |
arl | +0.056 |
obb | +0.054 |
iving | +0.054 |
pering | +0.054 |
Neg logit contributions | |
flavor | -0.035 |
penny | -0.033 |
peanut | -0.032 |
ornament | -0.032 |
cube | -0.031 |
Token | - |
Token ID | 12 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
arl | +0.032 |
pering | +0.029 |
obb | +0.029 |
ows | +0.029 |
oms | +0.028 |
Neg logit contributions | |
flavor | -0.016 |
ornament | -0.014 |
penny | -0.014 |
peanut | -0.014 |
cube | -0.013 |
Token | bel |
Token ID | 6667 |
Feature activation | +8.75e+00 |
Pos logit contributions | |
pering | +0.036 |
ows | +0.036 |
arl | +0.035 |
obb | +0.035 |
oms | +0.032 |
Neg logit contributions | |
flavor | -0.037 |
penny | -0.036 |
ornament | -0.036 |
cube | -0.035 |
peanut | -0.035 |
Token | ieve |
Token ID | 12311 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
arl | +0.216 |
pering | +0.213 |
ows | +0.204 |
iving | +0.181 |
oms | +0.177 |
Neg logit contributions | |
ornament | -0.339 |
flavor | -0.338 |
penny | -0.326 |
cube | -0.310 |
pencil | -0.302 |
Token | fun |
Token ID | 1257 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
arl | +0.056 |
obb | +0.053 |
pering | +0.052 |
ows | +0.050 |
adders | +0.048 |
Neg logit contributions | |
flavor | -0.035 |
cube | -0.033 |
ornament | -0.033 |
penny | -0.033 |
color | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
arl | +0.019 |
pering | +0.018 |
obb | +0.018 |
ows | +0.017 |
ighters | +0.017 |
Neg logit contributions | |
flavor | -0.013 |
cube | -0.012 |
ornament | -0.012 |
color | -0.012 |
penny | -0.012 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.33e+00 |
Pos logit contributions | |
pering | +0.011 |
ows | +0.011 |
obb | +0.011 |
arl | +0.011 |
iving | +0.010 |
Neg logit contributions | |
flavor | -0.007 |
penny | -0.007 |
ornament | -0.007 |
cube | -0.007 |
peanut | -0.007 |
Token | Once |
Token ID | 7454 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
pering | +0.143 |
ows | +0.143 |
obb | +0.140 |
arl | +0.136 |
announcements | +0.131 |
Neg logit contributions | |
flavor | -0.126 |
penny | -0.123 |
ornament | -0.121 |
cube | -0.119 |
peanut | -0.117 |
Token | a |
Token ID | 257 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
flavor | +0.048 |
penny | +0.048 |
ornament | +0.047 |
peanut | +0.046 |
cube | +0.045 |
Neg logit contributions | |
ows | -0.055 |
pering | -0.054 |
obb | -0.052 |
arl | -0.052 |
iving | -0.050 |
Token | world |
Token ID | 995 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
pering | +0.040 |
ows | +0.037 |
obb | +0.036 |
announcements | +0.036 |
aults | +0.035 |
Neg logit contributions | |
flavor | -0.043 |
penny | -0.042 |
cube | -0.040 |
ornament | -0.040 |
pol | -0.039 |
Token | of |
Token ID | 286 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
obb | +0.102 |
pering | +0.102 |
arl | +0.102 |
ows | +0.098 |
ighters | +0.095 |
Neg logit contributions | |
ornament | -0.089 |
flavor | -0.088 |
penny | -0.088 |
cube | -0.084 |
peanut | -0.081 |
Token | make |
Token ID | 787 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
ows | +0.051 |
pering | +0.050 |
obb | +0.048 |
arl | +0.048 |
iving | +0.046 |
Neg logit contributions | |
flavor | -0.042 |
penny | -0.041 |
ornament | -0.041 |
peanut | -0.040 |
cube | -0.039 |
Token | - |
Token ID | 12 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
arl | +0.038 |
pering | +0.034 |
obb | +0.034 |
ows | +0.031 |
oms | +0.031 |
Neg logit contributions | |
flavor | -0.018 |
ornament | -0.018 |
peanut | -0.017 |
color | -0.016 |
c | -0.016 |
Token | bel |
Token ID | 6667 |
Feature activation | +8.74e+00 |
Pos logit contributions | |
pering | +0.025 |
ows | +0.025 |
obb | +0.024 |
arl | +0.024 |
iving | +0.022 |
Neg logit contributions | |
flavor | -0.044 |
ornament | -0.043 |
penny | -0.043 |
peanut | -0.042 |
cube | -0.042 |
Token | ieve |
Token ID | 12311 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
pering | +0.163 |
arl | +0.159 |
ows | +0.158 |
iving | +0.139 |
obb | +0.138 |
Neg logit contributions | |
ornament | -0.381 |
flavor | -0.377 |
penny | -0.370 |
cube | -0.356 |
pencil | -0.345 |
Token | . |
Token ID | 13 |
Feature activation | -9.56e-02 |
Pos logit contributions | |
arl | +0.030 |
obb | +0.029 |
pering | +0.028 |
ows | +0.027 |
oms | +0.026 |
Neg logit contributions | |
ornament | -0.018 |
flavor | -0.018 |
penny | -0.017 |
color | -0.016 |
symbol | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
penny | +0.003 |
flavor | +0.003 |
ornament | +0.003 |
cube | +0.003 |
peanut | +0.003 |
Neg logit contributions | |
ows | -0.003 |
pering | -0.003 |
arl | -0.003 |
oms | -0.003 |
obb | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
penny | +0.009 |
flavor | +0.009 |
ornament | +0.009 |
cube | +0.008 |
peanut | +0.008 |
Neg logit contributions | |
ows | -0.012 |
pering | -0.012 |
arl | -0.011 |
obb | -0.011 |
oms | -0.011 |
Token | She |
Token ID | 3347 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
ornament | +0.013 |
penny | +0.013 |
flavor | +0.013 |
peanut | +0.012 |
cube | +0.012 |
Neg logit contributions | |
ows | -0.014 |
pering | -0.013 |
obb | -0.013 |
arl | -0.013 |
oms | -0.012 |
Token | L |
Token ID | 43 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
penny | +0.015 |
flavor | +0.015 |
ornament | +0.015 |
cube | +0.015 |
peanut | +0.014 |
Neg logit contributions | |
ows | -0.024 |
pering | -0.024 |
obb | -0.023 |
oms | -0.023 |
adders | -0.022 |
Token | ily |
Token ID | 813 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
pering | +0.098 |
announcements | +0.095 |
ows | +0.087 |
aults | +0.087 |
obb | +0.086 |
Neg logit contributions | |
ornament | -0.125 |
flavor | -0.124 |
cube | -0.120 |
penny | -0.120 |
peanut | -0.117 |
Token | watched |
Token ID | 7342 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
penny | +0.046 |
flavor | +0.046 |
ornament | +0.044 |
cube | +0.043 |
peanut | +0.043 |
Neg logit contributions | |
ows | -0.073 |
pering | -0.073 |
arl | -0.070 |
obb | -0.070 |
iving | -0.068 |
Token | as |
Token ID | 355 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
ows | +0.117 |
obb | +0.109 |
arl | +0.108 |
pering | +0.106 |
adders | +0.105 |
Neg logit contributions | |
ornament | -0.070 |
penny | -0.068 |
cube | -0.066 |
flavor | -0.065 |
peanut | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
ows | +0.166 |
iving | +0.149 |
arl | +0.149 |
immers | +0.149 |
ets | +0.146 |
Neg logit contributions | |
penny | -0.051 |
ornament | -0.050 |
mint | -0.045 |
avocado | -0.044 |
pencil | -0.044 |
Token | ze |
Token ID | 41271 |
Feature activation | +8.69e+00 |
Pos logit contributions | |
ows | +0.079 |
pering | +0.078 |
obb | +0.075 |
arl | +0.073 |
iving | +0.072 |
Neg logit contributions | |
flavor | -0.069 |
penny | -0.068 |
peanut | -0.066 |
ornament | -0.065 |
cube | -0.065 |
Token | br |
Token ID | 1671 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
pering | +0.380 |
sons | +0.360 |
ows | +0.357 |
arl | +0.355 |
stations | +0.352 |
Neg logit contributions | |
nut | -0.141 |
flavor | -0.140 |
peanut | -0.136 |
penny | -0.136 |
ornament | -0.130 |
Token | as |
Token ID | 292 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
pering | +0.006 |
ows | +0.005 |
obb | +0.005 |
arl | +0.005 |
announcements | +0.004 |
Neg logit contributions | |
flavor | -0.020 |
penny | -0.019 |
ornament | -0.019 |
cube | -0.019 |
peanut | -0.019 |
Token | ran |
Token ID | 4966 |
Feature activation | +4.40e+00 |
Pos logit contributions | |
pering | +0.144 |
arl | +0.142 |
obb | +0.141 |
ows | +0.138 |
ighters | +0.132 |
Neg logit contributions | |
penny | -0.086 |
cube | -0.083 |
flavor | -0.082 |
ornament | -0.079 |
comb | -0.075 |
Token | as |
Token ID | 355 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
ighters | +0.176 |
obb | +0.174 |
arl | +0.171 |
ows | +0.168 |
ains | +0.166 |
Neg logit contributions | |
penny | -0.098 |
cube | -0.091 |
ornament | -0.088 |
mine | -0.088 |
flavor | -0.083 |
Token | fast |
Token ID | 3049 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
ows | +0.116 |
arl | +0.111 |
iving | +0.110 |
pering | +0.110 |
oms | +0.103 |
Neg logit contributions | |
penny | -0.089 |
ornament | -0.084 |
flavor | -0.083 |
cube | -0.081 |
peanut | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
flavor | +0.018 |
penny | +0.018 |
ornament | +0.017 |
cube | +0.017 |
peanut | +0.017 |
Neg logit contributions | |
pering | -0.021 |
arl | -0.021 |
obb | -0.020 |
ows | -0.020 |
announcements | -0.019 |
Token | they |
Token ID | 484 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
flavor | +0.033 |
penny | +0.032 |
ornament | +0.031 |
peanut | +0.031 |
cube | +0.030 |
Neg logit contributions | |
pering | -0.044 |
ows | -0.043 |
obb | -0.042 |
arl | -0.042 |
oms | -0.040 |
Token | played |
Token ID | 2826 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
ows | +0.069 |
pering | +0.069 |
obb | +0.066 |
arl | +0.065 |
iving | +0.064 |
Neg logit contributions | |
penny | -0.046 |
flavor | -0.045 |
ornament | -0.044 |
peanut | -0.043 |
cube | -0.043 |
Token | make |
Token ID | 787 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
arl | +0.103 |
pering | +0.101 |
obb | +0.099 |
oms | +0.097 |
ows | +0.095 |
Neg logit contributions | |
flavor | -0.047 |
ornament | -0.045 |
cube | -0.043 |
penny | -0.043 |
color | -0.039 |
Token | - |
Token ID | 12 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
arl | +0.022 |
pering | +0.022 |
obb | +0.021 |
oms | +0.019 |
ows | +0.019 |
Neg logit contributions | |
flavor | -0.010 |
ornament | -0.009 |
peanut | -0.009 |
pencil | -0.008 |
cube | -0.008 |
Token | bel |
Token ID | 6667 |
Feature activation | +8.66e+00 |
Pos logit contributions | |
pering | +0.019 |
ows | +0.019 |
obb | +0.018 |
arl | +0.017 |
iving | +0.016 |
Neg logit contributions | |
flavor | -0.046 |
penny | -0.045 |
ornament | -0.045 |
cube | -0.044 |
peanut | -0.044 |
Token | ieve |
Token ID | 12311 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
pering | +0.175 |
arl | +0.167 |
ows | +0.163 |
iving | +0.146 |
obb | +0.143 |
Neg logit contributions | |
ornament | -0.374 |
flavor | -0.368 |
penny | -0.361 |
cube | -0.350 |
pencil | -0.335 |
Token | . |
Token ID | 13 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
arl | +0.019 |
pering | +0.019 |
obb | +0.018 |
ows | +0.018 |
oms | +0.017 |
Neg logit contributions | |
flavor | -0.011 |
ornament | -0.010 |
penny | -0.010 |
cube | -0.010 |
color | -0.009 |
Token | Finally |
Token ID | 9461 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
pering | +0.001 |
ows | +0.001 |
obb | +0.001 |
arl | +0.001 |
iving | +0.001 |
Neg logit contributions | |
flavor | -0.002 |
penny | -0.002 |
peanut | -0.002 |
ornament | -0.002 |
cube | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
ows | +0.101 |
pering | +0.099 |
arl | +0.094 |
adders | +0.093 |
announcements | +0.093 |
Neg logit contributions | |
penny | -0.047 |
flavor | -0.046 |
peanut | -0.045 |
cube | -0.041 |
ornament | -0.038 |
Token | they |
Token ID | 484 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
ows | +0.037 |
pering | +0.036 |
arl | +0.034 |
obb | +0.033 |
adders | +0.033 |
Neg logit contributions | |
penny | -0.054 |
flavor | -0.054 |
peanut | -0.052 |
cube | -0.050 |
ornament | -0.050 |
Token | that |
Token ID | 326 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
ows | +0.013 |
pering | +0.013 |
arl | +0.012 |
obb | +0.012 |
oms | +0.012 |
Neg logit contributions | |
flavor | -0.013 |
penny | -0.013 |
ornament | -0.012 |
peanut | -0.012 |
cube | -0.012 |
Token | says |
Token ID | 1139 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
penny | +0.019 |
flavor | +0.019 |
ornament | +0.019 |
peanut | +0.019 |
cube | +0.019 |
Neg logit contributions | |
pering | -0.017 |
ows | -0.016 |
obb | -0.016 |
arl | -0.015 |
iving | -0.015 |
Token | " |
Token ID | 366 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
pering | +0.013 |
obb | +0.013 |
ows | +0.013 |
arl | +0.013 |
iving | +0.012 |
Neg logit contributions | |
flavor | -0.007 |
penny | -0.007 |
ornament | -0.007 |
cube | -0.007 |
peanut | -0.007 |
Token | Feed |
Token ID | 18332 |
Feature activation | +9.62e-02 |
Pos logit contributions | |
pering | +0.001 |
ows | +0.001 |
arl | +0.001 |
iving | +0.001 |
obb | +0.001 |
Neg logit contributions | |
penny | -0.001 |
ornament | -0.001 |
flavor | -0.001 |
cube | -0.001 |
peanut | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +9.56e-01 |
Pos logit contributions | |
ows | +0.003 |
pering | +0.003 |
obb | +0.003 |
arl | +0.003 |
iving | +0.003 |
Neg logit contributions | |
flavor | -0.003 |
penny | -0.003 |
ornament | -0.003 |
peanut | -0.003 |
cube | -0.003 |
Token | ze |
Token ID | 41271 |
Feature activation | +8.62e+00 |
Pos logit contributions | |
ows | +0.034 |
pering | +0.034 |
obb | +0.032 |
arl | +0.032 |
iving | +0.031 |
Neg logit contributions | |
penny | -0.026 |
flavor | -0.026 |
peanut | -0.025 |
ornament | -0.025 |
cube | -0.024 |
Token | br |
Token ID | 1671 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
pering | +0.304 |
obb | +0.292 |
ows | +0.289 |
announcements | +0.282 |
sons | +0.275 |
Neg logit contributions | |
flavor | -0.216 |
penny | -0.209 |
peanut | -0.205 |
ornament | -0.198 |
nut | -0.197 |
Token | as |
Token ID | 292 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
flavor | +0.008 |
penny | +0.008 |
ornament | +0.008 |
cube | +0.008 |
peanut | +0.008 |
Neg logit contributions | |
pering | -0.002 |
ows | -0.002 |
obb | -0.002 |
arl | -0.002 |
iving | -0.002 |
Token | ". |
Token ID | 1911 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
obb | +0.138 |
pering | +0.135 |
announcements | +0.132 |
ows | +0.130 |
arl | +0.124 |
Neg logit contributions | |
flavor | -0.110 |
penny | -0.109 |
ornament | -0.108 |
cube | -0.106 |
color | -0.099 |
Token | He |
Token ID | 679 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
peanut | +0.013 |
penny | +0.011 |
His | +0.011 |
flavor | +0.011 |
cube | +0.011 |
Neg logit contributions | |
pering | -0.030 |
obb | -0.030 |
aults | -0.029 |
ows | -0.029 |
ault | -0.029 |
Token | has |
Token ID | 468 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
pering | +0.041 |
arl | +0.041 |
iving | +0.039 |
ows | +0.039 |
obb | +0.038 |
Neg logit contributions | |
flavor | -0.033 |
penny | -0.033 |
peanut | -0.033 |
nut | -0.031 |
ornament | -0.031 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
flavor | +0.082 |
penny | +0.082 |
ornament | +0.079 |
cube | +0.077 |
peanut | +0.076 |
Neg logit contributions | |
ows | -0.135 |
pering | -0.135 |
obb | -0.130 |
arl | -0.130 |
iving | -0.126 |
Token | to |
Token ID | 284 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
ows | +0.043 |
pering | +0.043 |
obb | +0.040 |
arl | +0.040 |
oms | +0.040 |
Neg logit contributions | |
penny | -0.032 |
ornament | -0.031 |
flavor | -0.031 |
cube | -0.030 |
peanut | -0.029 |
Token | see |
Token ID | 766 |
Feature activation | +6.22e-02 |
Pos logit contributions | |
pering | +0.001 |
ows | +0.001 |
obb | +0.001 |
arl | +0.001 |
iving | +0.001 |
Neg logit contributions | |
flavor | -0.001 |
penny | -0.001 |
peanut | -0.001 |
ornament | -0.001 |
cube | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
pering | +0.002 |
obb | +0.002 |
arl | +0.002 |
ows | +0.002 |
iving | +0.002 |
Neg logit contributions | |
flavor | -0.002 |
penny | -0.002 |
peanut | -0.002 |
ornament | -0.002 |
cube | -0.002 |
Token | baby |
Token ID | 5156 |
Feature activation | -8.19e-02 |
Pos logit contributions | |
ows | +0.051 |
pering | +0.049 |
obb | +0.048 |
arl | +0.047 |
adders | +0.046 |
Neg logit contributions | |
penny | -0.044 |
flavor | -0.044 |
peanut | -0.043 |
ornament | -0.043 |
cube | -0.042 |
Token | ze |
Token ID | 41271 |
Feature activation | +8.55e+00 |
Pos logit contributions | |
flavor | +0.002 |
ornament | +0.002 |
penny | +0.002 |
cube | +0.002 |
peanut | +0.002 |
Neg logit contributions | |
obb | -0.003 |
arl | -0.003 |
pering | -0.003 |
ows | -0.003 |
iving | -0.002 |
Token | br |
Token ID | 1671 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
pering | +0.358 |
sons | +0.347 |
obb | +0.341 |
arl | +0.337 |
realize | +0.335 |
Neg logit contributions | |
flavor | -0.155 |
penny | -0.144 |
ornament | -0.144 |
peanut | -0.143 |
nut | -0.141 |
Token | as |
Token ID | 292 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
pering | +0.001 |
ows | +0.001 |
obb | +0.001 |
arl | +0.001 |
iving | +0.001 |
Neg logit contributions | |
flavor | -0.004 |
penny | -0.004 |
ornament | -0.004 |
cube | -0.004 |
peanut | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
obb | +0.133 |
arl | +0.125 |
pering | +0.123 |
ows | +0.120 |
adders | +0.117 |
Neg logit contributions | |
penny | -0.089 |
ornament | -0.088 |
flavor | -0.088 |
cube | -0.086 |
peanut | -0.081 |
Token | They |
Token ID | 1119 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
pering | +0.005 |
ows | +0.005 |
obb | +0.005 |
iving | +0.005 |
oms | +0.005 |
Neg logit contributions | |
flavor | -0.004 |
peanut | -0.004 |
penny | -0.004 |
cube | -0.004 |
ornament | -0.003 |
Token | were |
Token ID | 547 |
Feature activation | -4.66e-01 |
Pos logit contributions | |
pering | +0.081 |
ows | +0.080 |
arl | +0.078 |
obb | +0.076 |
iving | +0.075 |
Neg logit contributions | |
flavor | -0.054 |
penny | -0.054 |
ornament | -0.051 |
peanut | -0.051 |
cube | -0.050 |
Token | He |
Token ID | 679 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ows | +0.002 |
obb | +0.002 |
pering | +0.002 |
arl | +0.002 |
iving | +0.002 |
Neg logit contributions | |
flavor | -0.001 |
peanut | -0.001 |
penny | -0.001 |
cube | -0.001 |
ornament | -0.001 |
Token | would |
Token ID | 561 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
arl | +0.048 |
pering | +0.047 |
ows | +0.045 |
obb | +0.044 |
iving | +0.044 |
Neg logit contributions | |
cube | -0.018 |
flavor | -0.018 |
penny | -0.018 |
ornament | -0.017 |
peanut | -0.016 |
Token | race |
Token ID | 3234 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
arl | +0.027 |
ows | +0.027 |
pering | +0.026 |
obb | +0.025 |
iving | +0.025 |
Neg logit contributions | |
ornament | -0.017 |
cube | -0.017 |
penny | -0.017 |
flavor | -0.017 |
color | -0.016 |
Token | make |
Token ID | 787 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
arl | +0.042 |
pering | +0.041 |
ows | +0.040 |
obb | +0.040 |
ighters | +0.038 |
Neg logit contributions | |
flavor | -0.023 |
penny | -0.023 |
cube | -0.023 |
ornament | -0.022 |
color | -0.021 |
Token | - |
Token ID | 12 |
Feature activation | +6.55e-01 |
Pos logit contributions | |
arl | +0.019 |
pering | +0.018 |
obb | +0.017 |
ows | +0.016 |
oms | +0.016 |
Neg logit contributions | |
ornament | -0.011 |
flavor | -0.010 |
toy | -0.010 |
peanut | -0.010 |
cube | -0.010 |
Token | bel |
Token ID | 6667 |
Feature activation | +8.42e+00 |
Pos logit contributions | |
pering | +0.018 |
ows | +0.017 |
arl | +0.017 |
obb | +0.017 |
oms | +0.015 |
Neg logit contributions | |
flavor | -0.023 |
ornament | -0.023 |
penny | -0.023 |
cube | -0.022 |
peanut | -0.022 |
Token | ieve |
Token ID | 12311 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
pering | +0.207 |
arl | +0.204 |
ows | +0.192 |
iving | +0.173 |
oms | +0.165 |
Neg logit contributions | |
ornament | -0.329 |
flavor | -0.322 |
penny | -0.311 |
cube | -0.299 |
pencil | -0.298 |
Token | cars |
Token ID | 5006 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
arl | +0.053 |
pering | +0.049 |
obb | +0.048 |
ows | +0.047 |
announcements | +0.045 |
Neg logit contributions | |
cube | -0.029 |
ornament | -0.029 |
flavor | -0.029 |
penny | -0.029 |
pencil | -0.026 |
Token | around |
Token ID | 1088 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
arl | +0.103 |
pering | +0.103 |
obb | +0.102 |
ighters | +0.096 |
announcements | +0.095 |
Neg logit contributions | |
cube | -0.058 |
ornament | -0.056 |
flavor | -0.056 |
penny | -0.054 |
color | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
ows | +0.020 |
arl | +0.020 |
obb | +0.019 |
pering | +0.019 |
iving | +0.018 |
Neg logit contributions | |
ornament | -0.012 |
cube | -0.012 |
penny | -0.012 |
flavor | -0.012 |
magnet | -0.011 |
Token | track |
Token ID | 2610 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
ows | +0.061 |
arl | +0.059 |
obb | +0.056 |
pering | +0.056 |
oms | +0.055 |
Neg logit contributions | |
ornament | -0.039 |
cube | -0.039 |
penny | -0.038 |
flavor | -0.037 |
peanut | -0.036 |
Token | watch |
Token ID | 2342 |
Feature activation | +3.74e+00 |
Pos logit contributions | |
arl | +0.002 |
ows | +0.002 |
pering | +0.002 |
iving | +0.002 |
obb | +0.002 |
Neg logit contributions | |
penny | -0.001 |
flavor | -0.001 |
peanut | -0.001 |
ornament | -0.001 |
cube | -0.001 |
Token | as |
Token ID | 355 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
arl | +0.142 |
pering | +0.138 |
ows | +0.137 |
obb | +0.136 |
iving | +0.128 |
Neg logit contributions | |
flavor | -0.100 |
ornament | -0.098 |
penny | -0.094 |
cube | -0.093 |
peanut | -0.092 |
Token | the |
Token ID | 262 |
Feature activation | +3.15e+00 |
Pos logit contributions | |
arl | +0.185 |
ows | +0.184 |
pering | +0.179 |
iving | +0.168 |
obb | +0.167 |
Neg logit contributions | |
penny | -0.082 |
flavor | -0.081 |
cube | -0.080 |
ornament | -0.079 |
copper | -0.074 |
Token | wild |
Token ID | 4295 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ows | +0.108 |
pering | +0.107 |
obb | +0.106 |
arl | +0.103 |
oms | +0.100 |
Neg logit contributions | |
penny | -0.084 |
flavor | -0.083 |
ornament | -0.081 |
peanut | -0.079 |
cube | -0.079 |
Token | sugar |
Token ID | 7543 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ows | +0.043 |
pering | +0.043 |
arl | +0.042 |
obb | +0.041 |
iving | +0.040 |
Neg logit contributions | |
flavor | -0.024 |
penny | -0.024 |
ornament | -0.023 |
peanut | -0.022 |
cube | -0.022 |
Token | scam |
Token ID | 19126 |
Feature activation | +8.40e+00 |
Pos logit contributions | |
flavor | +0.032 |
penny | +0.032 |
ornament | +0.031 |
cube | +0.030 |
peanut | +0.029 |
Neg logit contributions | |
pering | -0.053 |
arl | -0.052 |
ows | -0.050 |
obb | -0.049 |
announcements | -0.048 |
Token | pered |
Token ID | 13653 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
announcements | +0.163 |
amused | +0.150 |
sons | +0.135 |
realize | +0.135 |
ows | +0.134 |
Neg logit contributions | |
peanut | -0.333 |
penny | -0.333 |
nut | -0.326 |
ornament | -0.323 |
avocado | -0.323 |
Token | away |
Token ID | 1497 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
obb | +0.113 |
pering | +0.113 |
arl | +0.113 |
ows | +0.112 |
oms | +0.109 |
Neg logit contributions | |
penny | -0.075 |
ornament | -0.072 |
flavor | -0.071 |
cube | -0.068 |
peanut | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
penny | +0.004 |
flavor | +0.004 |
cube | +0.003 |
ornament | +0.003 |
symbol | +0.003 |
Neg logit contributions | |
arl | -0.007 |
obb | -0.007 |
ows | -0.007 |
adders | -0.007 |
ighters | -0.007 |
Token | |
Token ID | 220 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
ows | +0.015 |
pering | +0.015 |
arl | +0.014 |
obb | +0.014 |
iving | +0.014 |
Neg logit contributions | |
penny | -0.012 |
flavor | -0.012 |
ornament | -0.012 |
peanut | -0.011 |
cube | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
flavor | +0.020 |
cube | +0.019 |
penny | +0.017 |
ornament | +0.016 |
His | +0.016 |
Neg logit contributions | |
announcements | -0.093 |
obb | -0.093 |
pering | -0.092 |
aults | -0.088 |
arl | -0.087 |
Token | But |
Token ID | 1537 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
penny | +0.006 |
flavor | +0.006 |
ornament | +0.006 |
cube | +0.006 |
peanut | +0.006 |
Neg logit contributions | |
ows | -0.007 |
pering | -0.007 |
obb | -0.007 |
arl | -0.006 |
oms | -0.006 |
Token | as |
Token ID | 355 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
flavor | +0.015 |
penny | +0.014 |
ornament | +0.014 |
cube | +0.014 |
peanut | +0.014 |
Neg logit contributions | |
pering | -0.011 |
ows | -0.010 |
obb | -0.010 |
arl | -0.010 |
iving | -0.010 |
Token | they |
Token ID | 484 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
ows | +0.119 |
pering | +0.117 |
arl | +0.109 |
obb | +0.109 |
oms | +0.106 |
Neg logit contributions | |
penny | -0.094 |
flavor | -0.094 |
ornament | -0.089 |
peanut | -0.089 |
cube | -0.086 |
Token | approached |
Token ID | 10448 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ows | +0.076 |
pering | +0.074 |
obb | +0.073 |
oms | +0.072 |
iving | +0.072 |
Neg logit contributions | |
ornament | -0.083 |
penny | -0.081 |
flavor | -0.080 |
cube | -0.079 |
peanut | -0.078 |
Token | the |
Token ID | 262 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
ows | +0.032 |
pering | +0.032 |
obb | +0.031 |
arl | +0.030 |
iving | +0.029 |
Neg logit contributions | |
penny | -0.037 |
flavor | -0.037 |
ornament | -0.037 |
cube | -0.036 |
peanut | -0.036 |
Token | ze |
Token ID | 41271 |
Feature activation | +8.38e+00 |
Pos logit contributions | |
ows | +0.055 |
pering | +0.054 |
obb | +0.053 |
arl | +0.052 |
iving | +0.051 |
Neg logit contributions | |
flavor | -0.043 |
penny | -0.043 |
ornament | -0.042 |
cube | -0.041 |
peanut | -0.041 |
Token | br |
Token ID | 1671 |
Feature activation | -4.30e-03 |
Pos logit contributions | |
pering | +0.347 |
sons | +0.335 |
arl | +0.333 |
ows | +0.333 |
obb | +0.328 |
Neg logit contributions | |
penny | -0.154 |
nut | -0.153 |
flavor | -0.152 |
peanut | -0.152 |
ornament | -0.146 |
Token | as |
Token ID | 292 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
flavor | +0.000 |
penny | +0.000 |
ornament | +0.000 |
cube | +0.000 |
peanut | +0.000 |
Neg logit contributions | |
pering | -0.000 |
ows | -0.000 |
obb | -0.000 |
arl | -0.000 |
announcements | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
obb | +0.131 |
arl | +0.129 |
pering | +0.126 |
ows | +0.123 |
announcements | +0.121 |
Neg logit contributions | |
cube | -0.066 |
penny | -0.065 |
ornament | -0.064 |
flavor | -0.061 |
peanut | -0.056 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
ows | +0.031 |
pering | +0.031 |
arl | +0.030 |
obb | +0.029 |
iving | +0.028 |
Neg logit contributions | |
penny | -0.022 |
flavor | -0.021 |
peanut | -0.021 |
ornament | -0.020 |
cube | -0.019 |
Token | got |
Token ID | 1392 |
Feature activation | -9.93e-02 |
Pos logit contributions | |
penny | +0.011 |
peanut | +0.010 |
flavor | +0.010 |
ornament | +0.009 |
cube | +0.009 |
Neg logit contributions | |
arl | -0.022 |
pering | -0.021 |
obb | -0.021 |
ows | -0.021 |
adders | -0.020 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
flavor | +0.069 |
penny | +0.068 |
peanut | +0.067 |
ornament | +0.067 |
cube | +0.065 |
Neg logit contributions | |
ows | -0.070 |
pering | -0.069 |
obb | -0.067 |
arl | -0.065 |
iving | -0.064 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
flavor | +0.041 |
penny | +0.039 |
peanut | +0.038 |
cube | +0.037 |
nut | +0.037 |
Neg logit contributions | |
pering | -0.048 |
arl | -0.047 |
iving | -0.044 |
obb | -0.044 |
oms | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
ornament | +0.001 |
penny | +0.001 |
magnet | +0.001 |
orn | +0.001 |
cube | +0.001 |
Neg logit contributions | |
men | -0.001 |
announcements | -0.001 |
adders | -0.001 |
dolphins | -0.001 |
ows | -0.001 |
Token | see |
Token ID | 766 |
Feature activation | -6.02e-01 |
Pos logit contributions | |
penny | +0.021 |
ornament | +0.020 |
flavor | +0.020 |
cube | +0.019 |
peanut | +0.019 |
Neg logit contributions | |
ows | -0.020 |
pering | -0.020 |
obb | -0.019 |
arl | -0.019 |
announcements | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
flavor | +0.019 |
ornament | +0.018 |
penny | +0.018 |
peanut | +0.018 |
cube | +0.017 |
Neg logit contributions | |
pering | -0.020 |
obb | -0.020 |
ows | -0.020 |
arl | -0.019 |
iving | -0.019 |
Token | ze |
Token ID | 41271 |
Feature activation | +8.37e+00 |
Pos logit contributions | |
ows | +0.040 |
pering | +0.040 |
obb | +0.038 |
adders | +0.037 |
iving | +0.037 |
Neg logit contributions | |
flavor | -0.032 |
peanut | -0.032 |
cube | -0.031 |
penny | -0.030 |
ornament | -0.030 |
Token | br |
Token ID | 1671 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
pering | +0.335 |
obb | +0.314 |
sons | +0.311 |
arl | +0.311 |
ows | +0.307 |
Neg logit contributions | |
flavor | -0.175 |
peanut | -0.166 |
nut | -0.162 |
penny | -0.158 |
cube | -0.157 |
Token | as |
Token ID | 292 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
flavor | +0.001 |
penny | +0.001 |
ornament | +0.001 |
cube | +0.001 |
peanut | +0.001 |
Neg logit contributions | |
pering | -0.000 |
ows | -0.000 |
obb | -0.000 |
arl | -0.000 |
iving | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
obb | +0.153 |
arl | +0.143 |
announcements | +0.140 |
pering | +0.140 |
ows | +0.138 |
Neg logit contributions | |
cube | -0.097 |
ornament | -0.097 |
penny | -0.095 |
flavor | -0.094 |
peanut | -0.086 |
Token | most |
Token ID | 749 |
Feature activation | +6.17e-02 |
Pos logit contributions | |
obb | +0.066 |
ows | +0.065 |
pering | +0.065 |
oms | +0.064 |
arl | +0.063 |
Neg logit contributions | |
flavor | -0.050 |
peanut | -0.049 |
cube | -0.048 |
ornament | -0.047 |
penny | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
pering | +0.002 |
arl | +0.002 |
ows | +0.002 |
obb | +0.002 |
adders | +0.002 |
Neg logit contributions | |
flavor | -0.002 |
penny | -0.002 |
ornament | -0.002 |
cube | -0.002 |
peanut | -0.002 |
Token | we |
Token ID | 356 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
ows | +0.023 |
pering | +0.022 |
obb | +0.021 |
arl | +0.021 |
oms | +0.021 |
Neg logit contributions | |
peanut | -0.016 |
ornament | -0.016 |
penny | -0.016 |
flavor | -0.016 |
cube | -0.015 |
Token | can |
Token ID | 460 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
pering | +0.002 |
ows | +0.002 |
arl | +0.002 |
obb | +0.002 |
iving | +0.002 |
Neg logit contributions | |
flavor | -0.002 |
penny | -0.002 |
ornament | -0.002 |
peanut | -0.002 |
cube | -0.002 |
Token | 't |
Token ID | 470 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
penny | +0.043 |
flavor | +0.043 |
ornament | +0.041 |
peanut | +0.041 |
cube | +0.040 |
Neg logit contributions | |
ows | -0.047 |
pering | -0.046 |
obb | -0.044 |
arl | -0.043 |
oms | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
penny | +0.006 |
flavor | +0.006 |
ornament | +0.006 |
cube | +0.006 |
peanut | +0.006 |
Neg logit contributions | |
ows | -0.005 |
pering | -0.005 |
obb | -0.005 |
arl | -0.005 |
iving | -0.005 |
Token | The |
Token ID | 383 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
pering | +0.017 |
ows | +0.017 |
obb | +0.016 |
arl | +0.016 |
iving | +0.015 |
Neg logit contributions | |
flavor | -0.017 |
penny | -0.017 |
ornament | -0.017 |
peanut | -0.016 |
cube | -0.016 |
Token | ze |
Token ID | 41271 |
Feature activation | +8.35e+00 |
Pos logit contributions | |
ows | +0.041 |
pering | +0.041 |
arl | +0.038 |
obb | +0.038 |
iving | +0.038 |
Neg logit contributions | |
flavor | -0.031 |
penny | -0.031 |
ornament | -0.030 |
peanut | -0.030 |
cube | -0.029 |
Token | br |
Token ID | 1671 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
pering | +0.352 |
sons | +0.327 |
ows | +0.319 |
arl | +0.319 |
stations | +0.318 |
Neg logit contributions | |
nut | -0.155 |
peanut | -0.152 |
flavor | -0.152 |
ornament | -0.144 |
penny | -0.144 |
Token | as |
Token ID | 292 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
flavor | +0.012 |
penny | +0.012 |
ornament | +0.012 |
cube | +0.012 |
peanut | +0.012 |
Neg logit contributions | |
pering | -0.004 |
ows | -0.004 |
obb | -0.004 |
arl | -0.003 |
iving | -0.003 |
Token | have |
Token ID | 423 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
pering | +0.133 |
obb | +0.121 |
ows | +0.118 |
arl | +0.118 |
announcements | +0.115 |
Neg logit contributions | |
penny | -0.110 |
ornament | -0.109 |
flavor | -0.108 |
cube | -0.107 |
peanut | -0.101 |
Token | their |
Token ID | 511 |
Feature activation | -7.34e-01 |
Pos logit contributions | |
flavor | +0.042 |
penny | +0.041 |
peanut | +0.040 |
ornament | +0.040 |
cube | +0.038 |
Neg logit contributions | |
pering | -0.060 |
ows | -0.059 |
arl | -0.057 |
obb | -0.057 |
iving | -0.055 |
Token | own |
Token ID | 898 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
flavor | +0.019 |
penny | +0.019 |
ornament | +0.018 |
peanut | +0.018 |
cube | +0.018 |
Neg logit contributions | |
pering | -0.027 |
ows | -0.027 |
obb | -0.026 |
arl | -0.026 |
iving | -0.025 |
Token | pen |
Token ID | 3112 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
obb | +0.060 |
pering | +0.059 |
arl | +0.058 |
ows | +0.057 |
iving | +0.054 |
Neg logit contributions | |
penny | -0.051 |
flavor | -0.050 |
ornament | -0.050 |
cube | -0.049 |
peanut | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
penny | +0.040 |
flavor | +0.040 |
ornament | +0.040 |
cube | +0.038 |
peanut | +0.038 |
Neg logit contributions | |
pering | -0.042 |
obb | -0.041 |
ows | -0.040 |
arl | -0.040 |
announcements | -0.038 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ornament | +0.035 |
penny | +0.034 |
flavor | +0.034 |
cube | +0.033 |
peanut | +0.031 |
Neg logit contributions | |
ows | -0.033 |
pering | -0.031 |
announcements | -0.030 |
obb | -0.029 |
adders | -0.029 |
Token | for |
Token ID | 329 |
Feature activation | +8.59e-01 |
Pos logit contributions | |
penny | +0.040 |
cube | +0.039 |
peanut | +0.039 |
flavor | +0.038 |
ornament | +0.037 |
Neg logit contributions | |
pering | -0.064 |
obb | -0.062 |
arl | -0.062 |
iving | -0.060 |
ows | -0.059 |
Token | the |
Token ID | 262 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
ows | +0.029 |
pering | +0.029 |
obb | +0.028 |
arl | +0.028 |
iving | +0.027 |
Neg logit contributions | |
penny | -0.024 |
flavor | -0.024 |
ornament | -0.023 |
peanut | -0.023 |
cube | -0.023 |
Token | ze |
Token ID | 41271 |
Feature activation | +8.33e+00 |
Pos logit contributions | |
ows | +0.035 |
pering | +0.035 |
obb | +0.033 |
arl | +0.033 |
iving | +0.032 |
Neg logit contributions | |
flavor | -0.028 |
penny | -0.028 |
ornament | -0.027 |
peanut | -0.027 |
cube | -0.026 |
Token | br |
Token ID | 1671 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
pering | +0.352 |
obb | +0.336 |
arl | +0.331 |
sons | +0.330 |
realize | +0.326 |
Neg logit contributions | |
flavor | -0.149 |
peanut | -0.144 |
penny | -0.144 |
ornament | -0.138 |
nut | -0.134 |
Token | as |
Token ID | 292 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
pering | +0.000 |
ows | +0.000 |
obb | +0.000 |
arl | +0.000 |
iving | +0.000 |
Neg logit contributions | |
flavor | -0.001 |
penny | -0.001 |
ornament | -0.001 |
cube | -0.001 |
peanut | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
obb | +0.140 |
arl | +0.126 |
pering | +0.126 |
announcements | +0.123 |
ows | +0.119 |
Neg logit contributions | |
penny | -0.086 |
cube | -0.083 |
ornament | -0.082 |
flavor | -0.080 |
peanut | -0.078 |
Token | They |
Token ID | 1119 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
flavor | +0.009 |
peanut | +0.008 |
penny | +0.008 |
cube | +0.008 |
ornament | +0.007 |
Neg logit contributions | |
pering | -0.015 |
arl | -0.014 |
obb | -0.014 |
bling | -0.014 |
iving | -0.014 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
pering | +0.064 |
ows | +0.063 |
arl | +0.062 |
obb | +0.061 |
iving | +0.060 |
Neg logit contributions | |
flavor | -0.041 |
penny | -0.041 |
ornament | -0.039 |
peanut | -0.039 |
cube | -0.038 |
Token | to |
Token ID | 284 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
ows | +0.062 |
pering | +0.062 |
obb | +0.059 |
arl | +0.059 |
oms | +0.057 |
Neg logit contributions | |
penny | -0.044 |
flavor | -0.043 |
ornament | -0.042 |
peanut | -0.041 |
cube | -0.041 |
Token | stay |
Token ID | 2652 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ows | +0.011 |
pering | +0.011 |
arl | +0.010 |
obb | +0.010 |
iving | +0.010 |
Neg logit contributions | |
penny | -0.009 |
flavor | -0.009 |
ornament | -0.009 |
cube | -0.009 |
peanut | -0.009 |
Token | away |
Token ID | 1497 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
ows | +0.041 |
obb | +0.040 |
arl | +0.040 |
oms | +0.039 |
pering | +0.039 |
Neg logit contributions | |
penny | -0.032 |
flavor | -0.030 |
ornament | -0.030 |
peanut | -0.029 |
cube | -0.028 |
Token | from |
Token ID | 422 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
penny | +0.010 |
ornament | +0.009 |
color | +0.009 |
cube | +0.009 |
symbol | +0.009 |
Neg logit contributions | |
obb | -0.020 |
announcements | -0.019 |
arl | -0.019 |
oms | -0.019 |
adies | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
ows | +0.088 |
arl | +0.081 |
iving | +0.080 |
obb | +0.078 |
oms | +0.078 |
Neg logit contributions | |
penny | -0.059 |
ornament | -0.059 |
flavor | -0.056 |
peanut | -0.055 |
cube | -0.053 |
Token | ze |
Token ID | 41271 |
Feature activation | +8.33e+00 |
Pos logit contributions | |
ows | +0.044 |
pering | +0.042 |
obb | +0.041 |
arl | +0.041 |
iving | +0.040 |
Neg logit contributions | |
penny | -0.039 |
flavor | -0.039 |
peanut | -0.038 |
ornament | -0.037 |
cube | -0.037 |
Token | br |
Token ID | 1671 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
pering | +0.305 |
ows | +0.299 |
sons | +0.299 |
arl | +0.295 |
stations | +0.290 |
Neg logit contributions | |
on | -0.183 |
penny | -0.179 |
peanut | -0.174 |
ornament | -0.170 |
flavor | -0.169 |
Token | as |
Token ID | 292 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
pering | +0.005 |
obb | +0.005 |
ows | +0.004 |
announcements | +0.004 |
arl | +0.004 |
Neg logit contributions | |
flavor | -0.020 |
penny | -0.019 |
cube | -0.019 |
ornament | -0.019 |
peanut | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
obb | +0.127 |
arl | +0.121 |
announcements | +0.119 |
pering | +0.118 |
ows | +0.116 |
Neg logit contributions | |
ornament | -0.071 |
penny | -0.071 |
cube | -0.068 |
c | -0.064 |
flavor | -0.063 |
Token | The |
Token ID | 383 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
ows | +0.020 |
pering | +0.020 |
obb | +0.019 |
arl | +0.019 |
iving | +0.018 |
Neg logit contributions | |
penny | -0.014 |
flavor | -0.014 |
ornament | -0.014 |
peanut | -0.013 |
cube | -0.013 |
Token | end |
Token ID | 886 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
ows | +0.067 |
pering | +0.065 |
arl | +0.062 |
obb | +0.062 |
iving | +0.061 |
Neg logit contributions | |
flavor | -0.044 |
penny | -0.044 |
peanut | -0.042 |
ornament | -0.042 |
cube | -0.041 |
Token | John |
Token ID | 1757 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
pering | +0.007 |
ows | +0.007 |
obb | +0.007 |
arl | +0.007 |
iving | +0.007 |
Neg logit contributions | |
flavor | -0.004 |
penny | -0.004 |
ornament | -0.004 |
cube | -0.004 |
peanut | -0.004 |
Token | began |
Token ID | 2540 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
penny | +0.023 |
flavor | +0.022 |
ornament | +0.021 |
peanut | +0.021 |
cube | +0.021 |
Neg logit contributions | |
pering | -0.042 |
ows | -0.041 |
arl | -0.041 |
obb | -0.041 |
iving | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
ows | +0.018 |
pering | +0.018 |
obb | +0.018 |
arl | +0.018 |
iving | +0.017 |
Neg logit contributions | |
penny | -0.009 |
flavor | -0.009 |
ornament | -0.008 |
peanut | -0.008 |
cube | -0.008 |
Token | make |
Token ID | 787 |
Feature activation | -7.24e-02 |
Pos logit contributions | |
ows | +0.019 |
pering | +0.018 |
obb | +0.018 |
arl | +0.018 |
adders | +0.018 |
Neg logit contributions | |
flavor | -0.012 |
ornament | -0.012 |
penny | -0.012 |
peanut | -0.011 |
cube | -0.011 |
Token | - |
Token ID | 12 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
flavor | +0.002 |
peanut | +0.002 |
penny | +0.002 |
ornament | +0.002 |
gu | +0.001 |
Neg logit contributions | |
arl | -0.003 |
obb | -0.003 |
pering | -0.003 |
ows | -0.003 |
oms | -0.003 |
Token | bel |
Token ID | 6667 |
Feature activation | +8.33e+00 |
Pos logit contributions | |
ows | +0.015 |
pering | +0.015 |
obb | +0.014 |
arl | +0.014 |
oms | +0.012 |
Neg logit contributions | |
flavor | -0.022 |
penny | -0.022 |
ornament | -0.021 |
cube | -0.021 |
peanut | -0.021 |
Token | ieve |
Token ID | 12311 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
pering | +0.167 |
arl | +0.164 |
ows | +0.162 |
iving | +0.144 |
oms | +0.141 |
Neg logit contributions | |
ornament | -0.353 |
flavor | -0.352 |
penny | -0.346 |
cube | -0.333 |
pencil | -0.321 |
Token | he |
Token ID | 339 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
arl | +0.017 |
obb | +0.016 |
pering | +0.016 |
ows | +0.015 |
ighters | +0.015 |
Neg logit contributions | |
penny | -0.009 |
flavor | -0.009 |
ornament | -0.009 |
cube | -0.009 |
pencil | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
arl | +0.119 |
pering | +0.117 |
ows | +0.113 |
obb | +0.113 |
iving | +0.108 |
Neg logit contributions | |
flavor | -0.054 |
penny | -0.051 |
ornament | -0.050 |
cube | -0.048 |
peanut | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
flavor | +0.001 |
penny | +0.001 |
ornament | +0.001 |
peanut | +0.001 |
cube | +0.001 |
Neg logit contributions | |
arl | -0.001 |
obb | -0.001 |
ows | -0.001 |
pering | -0.001 |
iving | -0.001 |
Token | hero |
Token ID | 4293 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
ows | +0.072 |
pering | +0.071 |
arl | +0.068 |
obb | +0.068 |
oms | +0.066 |
Neg logit contributions | |
penny | -0.055 |
flavor | -0.053 |
ornament | -0.052 |
cube | -0.051 |
peanut | -0.051 |
Token | Finally |
Token ID | 9461 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
ows | +0.009 |
pering | +0.009 |
obb | +0.009 |
arl | +0.008 |
iving | +0.008 |
Neg logit contributions | |
penny | -0.005 |
flavor | -0.005 |
ornament | -0.005 |
peanut | -0.005 |
cube | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
ows | +0.105 |
pering | +0.103 |
arl | +0.102 |
obb | +0.097 |
rig | +0.097 |
Neg logit contributions | |
penny | -0.047 |
peanut | -0.043 |
flavor | -0.043 |
ornament | -0.041 |
mint | -0.037 |
Token | her |
Token ID | 607 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
ows | +0.050 |
pering | +0.050 |
arl | +0.048 |
obb | +0.047 |
oms | +0.046 |
Neg logit contributions | |
penny | -0.040 |
flavor | -0.039 |
ornament | -0.038 |
peanut | -0.038 |
cube | -0.036 |
Token | make |
Token ID | 787 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
pering | +0.014 |
ows | +0.013 |
arl | +0.013 |
obb | +0.013 |
iving | +0.012 |
Neg logit contributions | |
penny | -0.009 |
flavor | -0.009 |
ornament | -0.009 |
cube | -0.009 |
peanut | -0.009 |
Token | - |
Token ID | 12 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
arl | +0.008 |
pering | +0.007 |
obb | +0.007 |
iving | +0.007 |
ows | +0.007 |
Neg logit contributions | |
flavor | -0.004 |
ornament | -0.004 |
peanut | -0.004 |
penny | -0.004 |
color | -0.003 |
Token | bel |
Token ID | 6667 |
Feature activation | +8.32e+00 |
Pos logit contributions | |
ows | +0.007 |
pering | +0.007 |
obb | +0.007 |
arl | +0.007 |
iving | +0.006 |
Neg logit contributions | |
penny | -0.010 |
flavor | -0.010 |
ornament | -0.009 |
cube | -0.009 |
peanut | -0.009 |
Token | ieve |
Token ID | 12311 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
pering | +0.170 |
ows | +0.166 |
arl | +0.164 |
iving | +0.145 |
oms | +0.144 |
Neg logit contributions | |
ornament | -0.353 |
flavor | -0.347 |
penny | -0.343 |
cube | -0.326 |
pencil | -0.319 |
Token | adventures |
Token ID | 17545 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
pering | +0.007 |
arl | +0.007 |
obb | +0.006 |
ows | +0.006 |
adders | +0.006 |
Neg logit contributions | |
ornament | -0.004 |
penny | -0.004 |
flavor | -0.004 |
cube | -0.004 |
peanut | -0.003 |
Token | were |
Token ID | 547 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
arl | +0.115 |
pering | +0.112 |
obb | +0.112 |
ows | +0.103 |
ighters | +0.102 |
Neg logit contributions | |
flavor | -0.081 |
ornament | -0.080 |
penny | -0.079 |
cube | -0.076 |
peanut | -0.071 |
Token | recorded |
Token ID | 6264 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
arl | +0.057 |
pering | +0.056 |
ows | +0.056 |
obb | +0.056 |
iving | +0.052 |
Neg logit contributions | |
flavor | -0.038 |
penny | -0.037 |
ornament | -0.037 |
peanut | -0.035 |
cube | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
arl | +0.095 |
obb | +0.092 |
pering | +0.090 |
ighters | +0.082 |
adders | +0.082 |
Neg logit contributions | |
ornament | -0.048 |
flavor | -0.047 |
color | -0.045 |
penny | -0.042 |
pencil | -0.041 |
Token | aff |
Token ID | 2001 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
pering | +0.054 |
ows | +0.052 |
obb | +0.051 |
arl | +0.050 |
iving | +0.049 |
Neg logit contributions | |
flavor | -0.059 |
penny | -0.058 |
peanut | -0.058 |
ornament | -0.057 |
cube | -0.056 |
Token | es |
Token ID | 274 |
Feature activation | -3.63e-03 |
Pos logit contributions | |
obb | +0.040 |
pering | +0.040 |
ows | +0.038 |
arl | +0.038 |
iving | +0.035 |
Neg logit contributions | |
flavor | -0.059 |
penny | -0.058 |
ornament | -0.057 |
peanut | -0.057 |
cube | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | +6.11e-02 |
Pos logit contributions | |
flavor | +0.000 |
penny | +0.000 |
peanut | +0.000 |
ornament | +0.000 |
nut | +0.000 |
Neg logit contributions | |
obb | -0.000 |
arl | -0.000 |
pering | -0.000 |
iving | -0.000 |
adders | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | -9.96e-01 |
Pos logit contributions | |
obb | +0.002 |
arl | +0.002 |
iving | +0.002 |
pering | +0.002 |
ows | +0.002 |
Neg logit contributions | |
peanut | -0.001 |
flavor | -0.001 |
penny | -0.001 |
ornament | -0.001 |
nut | -0.001 |
Token | fast |
Token ID | 3049 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
peanut | +0.029 |
flavor | +0.029 |
ornament | +0.028 |
penny | +0.028 |
nut | +0.027 |
Neg logit contributions | |
ows | -0.036 |
arl | -0.036 |
obb | -0.035 |
pering | -0.035 |
iving | -0.033 |
Token | ze |
Token ID | 41271 |
Feature activation | +8.21e+00 |
Pos logit contributions | |
peanut | +0.004 |
nut | +0.003 |
flavor | +0.003 |
color | +0.003 |
penny | +0.003 |
Neg logit contributions | |
obb | -0.006 |
bling | -0.006 |
ows | -0.006 |
arl | -0.005 |
pering | -0.005 |
Token | br |
Token ID | 1671 |
Feature activation | +4.87e-02 |
Pos logit contributions | |
pering | +0.335 |
sons | +0.320 |
obb | +0.317 |
arl | +0.313 |
announcements | +0.311 |
Neg logit contributions | |
nut | -0.141 |
peanut | -0.140 |
flavor | -0.140 |
penny | -0.126 |
ornament | -0.125 |
Token | as |
Token ID | 292 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
pering | +0.001 |
obb | +0.000 |
ows | +0.000 |
arl | +0.000 |
announcements | +0.000 |
Neg logit contributions | |
flavor | -0.002 |
penny | -0.002 |
ornament | -0.002 |
cube | -0.002 |
peanut | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
obb | +0.159 |
pering | +0.141 |
arl | +0.140 |
announcements | +0.139 |
rig | +0.132 |
Neg logit contributions | |
cube | -0.085 |
ornament | -0.081 |
penny | -0.081 |
flavor | -0.079 |
c | -0.075 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
penny | +0.028 |
ornament | +0.026 |
flavor | +0.026 |
peanut | +0.025 |
pencil | +0.025 |
Neg logit contributions | |
ows | -0.023 |
pering | -0.021 |
arl | -0.020 |
adders | -0.020 |
oms | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
penny | +0.029 |
flavor | +0.028 |
ornament | +0.027 |
peanut | +0.026 |
cube | +0.026 |
Neg logit contributions | |
ows | -0.046 |
pering | -0.044 |
obb | -0.042 |
iving | -0.041 |
arl | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
penny | +0.024 |
flavor | +0.024 |
ornament | +0.024 |
cube | +0.023 |
peanut | +0.023 |
Neg logit contributions | |
obb | -0.028 |
pering | -0.027 |
arl | -0.027 |
ows | -0.027 |
announcements | -0.026 |
Token | They |
Token ID | 1119 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
ows | +0.011 |
pering | +0.011 |
obb | +0.011 |
arl | +0.011 |
iving | +0.010 |
Neg logit contributions | |
flavor | -0.006 |
penny | -0.006 |
ornament | -0.006 |
peanut | -0.005 |
cube | -0.005 |
Token | saw |
Token ID | 2497 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
pering | +0.097 |
ows | +0.094 |
arl | +0.094 |
obb | +0.092 |
iving | +0.092 |
Neg logit contributions | |
flavor | -0.046 |
penny | -0.045 |
peanut | -0.044 |
cube | -0.043 |
ornament | -0.041 |
Token | four |
Token ID | 1440 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
pering | +0.022 |
ows | +0.022 |
arl | +0.021 |
obb | +0.021 |
iving | +0.021 |
Neg logit contributions | |
flavor | -0.016 |
penny | -0.016 |
peanut | -0.016 |
ornament | -0.016 |
cube | -0.015 |
Token | baby |
Token ID | 5156 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
arl | +0.004 |
obb | +0.004 |
ows | +0.004 |
pering | +0.004 |
iving | +0.004 |
Neg logit contributions | |
flavor | -0.003 |
penny | -0.003 |
peanut | -0.003 |
ornament | -0.003 |
cube | -0.003 |
Token | ze |
Token ID | 41271 |
Feature activation | +8.18e+00 |
Pos logit contributions | |
penny | +0.013 |
peanut | +0.013 |
flavor | +0.013 |
ornament | +0.013 |
cube | +0.013 |
Neg logit contributions | |
arl | -0.016 |
obb | -0.016 |
ows | -0.016 |
pering | -0.015 |
adders | -0.014 |
Token | br |
Token ID | 1671 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
pering | +0.326 |
ows | +0.313 |
arl | +0.303 |
obb | +0.303 |
announcements | +0.298 |
Neg logit contributions | |
flavor | -0.167 |
penny | -0.166 |
peanut | -0.164 |
ornament | -0.156 |
nut | -0.154 |
Token | as |
Token ID | 292 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
ows | +0.001 |
pering | +0.001 |
obb | +0.001 |
arl | +0.001 |
iving | +0.001 |
Neg logit contributions | |
flavor | -0.007 |
penny | -0.007 |
ornament | -0.007 |
peanut | -0.007 |
cube | -0.007 |
Token | running |
Token ID | 2491 |
Feature activation | +3.93e+00 |
Pos logit contributions | |
obb | +0.120 |
arl | +0.107 |
ows | +0.107 |
rig | +0.105 |
pering | +0.103 |
Neg logit contributions | |
penny | -0.072 |
cube | -0.072 |
ornament | -0.069 |
peanut | -0.069 |
flavor | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | +9.67e-01 |
Pos logit contributions | |
obb | +0.150 |
ows | +0.148 |
arl | +0.146 |
pering | +0.144 |
oms | +0.142 |
Neg logit contributions | |
penny | -0.095 |
cube | -0.091 |
peanut | -0.089 |
flavor | -0.088 |
ornament | -0.088 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
ows | +0.033 |
pering | +0.032 |
obb | +0.032 |
arl | +0.032 |
iving | +0.031 |
Neg logit contributions | |
flavor | -0.027 |
penny | -0.027 |
peanut | -0.026 |
ornament | -0.026 |
cube | -0.026 |
Token | orange |
Token ID | 10912 |
Feature activation | -5.11e+00 |
Pos logit contributions | |
flavor | +0.011 |
cube | +0.010 |
penny | +0.010 |
seed | +0.010 |
nut | +0.010 |
Neg logit contributions | |
ows | -0.010 |
pering | -0.010 |
arl | -0.010 |
Super | -0.009 |
adders | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
flavor | +0.185 |
penny | +0.181 |
seed | +0.176 |
peanut | +0.175 |
ornament | +0.174 |
Neg logit contributions | |
ows | -0.121 |
pering | -0.111 |
arl | -0.101 |
iving | -0.101 |
adders | -0.100 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
penny | +0.035 |
flavor | +0.032 |
cube | +0.032 |
ornament | +0.031 |
wallet | +0.031 |
Neg logit contributions | |
ows | -0.023 |
arl | -0.022 |
oms | -0.020 |
obb | -0.019 |
iving | -0.019 |
Token | wore |
Token ID | 12408 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
penny | +0.054 |
peanut | +0.053 |
flavor | +0.052 |
nut | +0.052 |
dime | +0.051 |
Neg logit contributions | |
ows | -0.031 |
as | -0.025 |
ends | -0.022 |
ila | -0.022 |
follows | -0.021 |
Token | her |
Token ID | 607 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
cube | +0.083 |
flavor | +0.079 |
seed | +0.076 |
penny | +0.075 |
ornament | +0.074 |
Neg logit contributions | |
ows | -0.065 |
obb | -0.059 |
Super | -0.058 |
announcements | -0.058 |
pering | -0.056 |
Token | orange |
Token ID | 10912 |
Feature activation | -8.06e+00 |
Pos logit contributions | |
ornament | +0.078 |
seed | +0.077 |
flavor | +0.076 |
mar | +0.074 |
penny | +0.074 |
Neg logit contributions | |
ows | -0.094 |
Super | -0.089 |
obb | -0.086 |
arl | -0.083 |
oms | -0.079 |
Token | belt |
Token ID | 10999 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
seed | +0.242 |
flavour | +0.240 |
flavor | +0.240 |
gu | +0.235 |
mattered | +0.234 |
Neg logit contributions | |
ows | -0.202 |
Super | -0.181 |
arl | -0.174 |
uits | -0.173 |
ropes | -0.166 |
Token | every |
Token ID | 790 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
flavor | +0.090 |
penny | +0.090 |
ornament | +0.087 |
peanut | +0.086 |
cube | +0.085 |
Neg logit contributions | |
ows | -0.091 |
pering | -0.085 |
obb | -0.083 |
arl | -0.081 |
iving | -0.078 |
Token | day |
Token ID | 1110 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
arl | +0.059 |
obb | +0.059 |
pering | +0.057 |
iving | +0.057 |
oms | +0.055 |
Neg logit contributions | |
seed | -0.047 |
flavor | -0.046 |
peanut | -0.045 |
cube | -0.045 |
dime | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ows | +0.042 |
pering | +0.041 |
obb | +0.038 |
arl | +0.038 |
iving | +0.038 |
Neg logit contributions | |
flavor | -0.050 |
penny | -0.050 |
ornament | -0.049 |
cube | -0.048 |
peanut | -0.047 |
Token | It |
Token ID | 632 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
penny | +0.043 |
flavor | +0.039 |
cube | +0.039 |
wallet | +0.038 |
ornament | +0.038 |
Neg logit contributions | |
ows | -0.025 |
arl | -0.024 |
oms | -0.021 |
iving | -0.020 |
Super | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
penny | +0.082 |
nut | +0.077 |
ornament | +0.077 |
ruby | +0.076 |
flavor | +0.075 |
Neg logit contributions | |
ows | -0.067 |
arl | -0.062 |
iving | -0.062 |
pering | -0.059 |
as | -0.056 |
Token | He |
Token ID | 679 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
penny | +0.070 |
ornament | +0.065 |
peanut | +0.064 |
flavor | +0.064 |
ruby | +0.062 |
Neg logit contributions | |
ows | -0.043 |
arl | -0.042 |
pering | -0.038 |
iving | -0.038 |
uits | -0.036 |
Token | wore |
Token ID | 12408 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
ows | +0.006 |
pering | +0.005 |
obb | +0.005 |
adders | +0.005 |
announcements | +0.005 |
Neg logit contributions | |
penny | -0.004 |
ornament | -0.004 |
flavor | -0.004 |
cube | -0.004 |
peanut | -0.004 |
Token | his |
Token ID | 465 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
penny | +0.093 |
ornament | +0.092 |
cube | +0.091 |
flavor | +0.088 |
symbol | +0.085 |
Neg logit contributions | |
ows | -0.063 |
obb | -0.059 |
uits | -0.059 |
announcements | -0.058 |
aults | -0.056 |
Token | favorite |
Token ID | 4004 |
Feature activation | -6.33e+00 |
Pos logit contributions | |
ornament | +0.092 |
flavor | +0.088 |
checked | +0.088 |
penny | +0.087 |
shr | +0.086 |
Neg logit contributions | |
ows | -0.107 |
obb | -0.104 |
arl | -0.102 |
sons | -0.094 |
Super | -0.094 |
Token | red |
Token ID | 2266 |
Feature activation | -8.04e+00 |
Pos logit contributions | |
checked | +0.216 |
shr | +0.205 |
pol | +0.203 |
Lem | +0.202 |
mar | +0.201 |
Neg logit contributions | |
loud | -0.123 |
men | -0.121 |
dolphins | -0.119 |
oms | -0.117 |
ows | -0.117 |
Token | shirt |
Token ID | 10147 |
Feature activation | -6.18e+00 |
Pos logit contributions | |
mattered | +0.256 |
pol | +0.255 |
possession | +0.254 |
ornament | +0.252 |
checked | +0.246 |
Neg logit contributions | |
arl | -0.197 |
ows | -0.175 |
uits | -0.175 |
oms | -0.172 |
iving | -0.160 |
Token | and |
Token ID | 290 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
penny | +0.216 |
flavor | +0.214 |
ornament | +0.211 |
peanut | +0.206 |
nut | +0.200 |
Neg logit contributions | |
ows | -0.177 |
pering | -0.142 |
obb | -0.141 |
arl | -0.140 |
adders | -0.138 |
Token | blue |
Token ID | 4171 |
Feature activation | -6.04e+00 |
Pos logit contributions | |
ornament | +0.051 |
penny | +0.050 |
cube | +0.048 |
flavor | +0.047 |
nut | +0.045 |
Neg logit contributions | |
ows | -0.055 |
obb | -0.050 |
arl | -0.049 |
adders | -0.049 |
pering | -0.048 |
Token | shorts |
Token ID | 22078 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
ornament | +0.221 |
penny | +0.211 |
possession | +0.205 |
symbol | +0.205 |
dime | +0.205 |
Neg logit contributions | |
ows | -0.143 |
oms | -0.135 |
arl | -0.132 |
uits | -0.125 |
iving | -0.116 |
Token | every |
Token ID | 790 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
ornament | +0.120 |
flavor | +0.119 |
penny | +0.119 |
peanut | +0.118 |
nut | +0.114 |
Neg logit contributions | |
ows | -0.081 |
pering | -0.068 |
obb | -0.067 |
arl | -0.065 |
oms | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
ows | +0.008 |
pering | +0.008 |
iving | +0.007 |
oms | +0.007 |
arl | +0.007 |
Neg logit contributions | |
penny | -0.006 |
flavor | -0.006 |
ornament | -0.006 |
peanut | -0.006 |
cube | -0.006 |
Token | She |
Token ID | 1375 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
penny | +0.065 |
cube | +0.062 |
flavor | +0.061 |
peanut | +0.061 |
ornament | +0.060 |
Neg logit contributions | |
ows | -0.044 |
pering | -0.038 |
arl | -0.038 |
adders | -0.035 |
iving | -0.035 |
Token | wore |
Token ID | 12408 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
ows | +0.013 |
pering | +0.013 |
obb | +0.012 |
arl | +0.012 |
iving | +0.012 |
Neg logit contributions | |
penny | -0.009 |
flavor | -0.009 |
ornament | -0.009 |
peanut | -0.008 |
cube | -0.008 |
Token | her |
Token ID | 607 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
cube | +0.105 |
penny | +0.099 |
flavor | +0.096 |
ornament | +0.096 |
magnet | +0.096 |
Neg logit contributions | |
ows | -0.054 |
announcements | -0.049 |
obb | -0.048 |
uits | -0.047 |
Super | -0.046 |
Token | favorite |
Token ID | 4004 |
Feature activation | -6.26e+00 |
Pos logit contributions | |
flavor | +0.077 |
ornament | +0.074 |
penny | +0.073 |
peanut | +0.072 |
cube | +0.071 |
Neg logit contributions | |
ows | -0.107 |
obb | -0.104 |
arl | -0.102 |
pering | -0.099 |
oms | -0.097 |
Token | red |
Token ID | 2266 |
Feature activation | -8.03e+00 |
Pos logit contributions | |
checked | +0.217 |
shr | +0.198 |
Lem | +0.196 |
ake | +0.191 |
pol | +0.190 |
Neg logit contributions | |
loud | -0.125 |
dancers | -0.121 |
ows | -0.121 |
ropes | -0.120 |
oms | -0.115 |
Token | m |
Token ID | 285 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
checked | +0.245 |
pol | +0.244 |
wealth | +0.242 |
mattered | +0.242 |
cavity | +0.236 |
Neg logit contributions | |
arl | -0.191 |
ows | -0.178 |
uits | -0.174 |
oms | -0.174 |
iving | -0.165 |
Token | itten |
Token ID | 2621 |
Feature activation | -4.92e+00 |
Pos logit contributions | |
ornament | +0.142 |
penny | +0.142 |
gem | +0.140 |
ruby | +0.138 |
diamond | +0.138 |
Neg logit contributions | |
ows | -0.082 |
arl | -0.072 |
adders | -0.064 |
ules | -0.064 |
as | -0.062 |
Token | on |
Token ID | 319 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
flavor | +0.199 |
penny | +0.194 |
peanut | +0.191 |
ornament | +0.188 |
dime | +0.185 |
Neg logit contributions | |
ows | -0.111 |
pering | -0.085 |
adders | -0.079 |
obb | -0.074 |
arl | -0.073 |
Token | her |
Token ID | 607 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
penny | +0.081 |
flavor | +0.080 |
ornament | +0.077 |
cube | +0.076 |
peanut | +0.075 |
Neg logit contributions | |
ows | -0.124 |
pering | -0.123 |
obb | -0.119 |
arl | -0.118 |
iving | -0.115 |
Token | hand |
Token ID | 1021 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
flavor | +0.049 |
penny | +0.048 |
ornament | +0.047 |
shr | +0.046 |
mar | +0.045 |
Neg logit contributions | |
ows | -0.050 |
pering | -0.047 |
obb | -0.042 |
aults | -0.042 |
sons | -0.040 |
Token | my |
Token ID | 1820 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
penny | +0.049 |
dime | +0.044 |
flavor | +0.043 |
peanut | +0.042 |
ornament | +0.041 |
Neg logit contributions | |
ows | -0.083 |
obb | -0.077 |
pering | -0.074 |
realize | -0.073 |
arl | -0.073 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
penny | +0.087 |
dime | +0.083 |
flavor | +0.082 |
ornament | +0.082 |
ruby | +0.079 |
Neg logit contributions | |
ows | -0.062 |
obb | -0.053 |
realize | -0.053 |
sons | -0.052 |
pering | -0.051 |
Token | wearing |
Token ID | 5762 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
c | +0.021 |
penny | +0.020 |
ornament | +0.020 |
ruby | +0.020 |
orn | +0.020 |
Neg logit contributions | |
dolphins | -0.022 |
pering | -0.021 |
men | -0.020 |
dancers | -0.020 |
announcements | -0.020 |
Token | his |
Token ID | 465 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
flavor | +0.080 |
penny | +0.079 |
ornament | +0.078 |
cube | +0.076 |
pol | +0.071 |
Neg logit contributions | |
ows | -0.091 |
obb | -0.090 |
pering | -0.088 |
announcements | -0.087 |
arl | -0.087 |
Token | favorite |
Token ID | 4004 |
Feature activation | -6.29e+00 |
Pos logit contributions | |
ornament | +0.090 |
flavor | +0.087 |
checked | +0.086 |
pol | +0.085 |
mar | +0.085 |
Neg logit contributions | |
obb | -0.102 |
ows | -0.101 |
arl | -0.099 |
Super | -0.094 |
pering | -0.094 |
Token | red |
Token ID | 2266 |
Feature activation | -7.92e+00 |
Pos logit contributions | |
checked | +0.195 |
pol | +0.189 |
mar | +0.182 |
wealth | +0.181 |
shr | +0.180 |
Neg logit contributions | |
dolphins | -0.152 |
dancers | -0.144 |
tent | -0.136 |
oms | -0.136 |
loud | -0.136 |
Token | socks |
Token ID | 24359 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
pol | +0.242 |
mattered | +0.237 |
wealth | +0.235 |
checked | +0.233 |
ornament | +0.229 |
Neg logit contributions | |
arl | -0.191 |
oms | -0.178 |
dolphins | -0.172 |
uits | -0.172 |
ows | -0.171 |
Token | every |
Token ID | 790 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
flavor | +0.103 |
penny | +0.103 |
ornament | +0.103 |
peanut | +0.101 |
nut | +0.098 |
Neg logit contributions | |
ows | -0.067 |
pering | -0.057 |
obb | -0.057 |
oms | -0.054 |
iving | -0.052 |
Token | day |
Token ID | 1110 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
obb | +0.030 |
arl | +0.029 |
iving | +0.029 |
pering | +0.029 |
dolphins | +0.028 |
Neg logit contributions | |
ornament | -0.028 |
seed | -0.027 |
dime | -0.027 |
flavor | -0.026 |
peanut | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
ows | +0.028 |
pering | +0.028 |
obb | +0.025 |
iving | +0.025 |
arl | +0.024 |
Neg logit contributions | |
ornament | -0.043 |
penny | -0.043 |
flavor | -0.043 |
cube | -0.041 |
peanut | -0.041 |
Token | He |
Token ID | 679 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
penny | +0.063 |
ornament | +0.057 |
peanut | +0.056 |
seed | +0.055 |
flavor | +0.055 |
Neg logit contributions | |
ows | -0.050 |
arl | -0.048 |
pering | -0.043 |
iving | -0.042 |
Super | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -9.86e-01 |
Pos logit contributions | |
penny | +0.005 |
flavor | +0.005 |
ornament | +0.005 |
cube | +0.005 |
peanut | +0.005 |
Neg logit contributions | |
pering | -0.006 |
ows | -0.006 |
obb | -0.006 |
arl | -0.006 |
iving | -0.005 |
Token | He |
Token ID | 679 |
Feature activation | +7.87e-01 |
Pos logit contributions | |
penny | +0.024 |
flavor | +0.024 |
cube | +0.023 |
ornament | +0.023 |
peanut | +0.023 |
Neg logit contributions | |
ows | -0.037 |
pering | -0.037 |
obb | -0.035 |
arl | -0.035 |
iving | -0.034 |
Token | shows |
Token ID | 2523 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
pering | +0.030 |
arl | +0.028 |
obb | +0.027 |
ows | +0.027 |
adders | +0.026 |
Neg logit contributions | |
penny | -0.021 |
cube | -0.020 |
flavor | -0.020 |
ornament | -0.020 |
pencil | -0.019 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
penny | +0.068 |
flavor | +0.067 |
ornament | +0.065 |
peanut | +0.063 |
cube | +0.063 |
Neg logit contributions | |
pering | -0.112 |
ows | -0.110 |
arl | -0.106 |
obb | -0.105 |
iving | -0.103 |
Token | his |
Token ID | 465 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
flavor | +0.047 |
penny | +0.046 |
ornament | +0.045 |
peanut | +0.044 |
cube | +0.043 |
Neg logit contributions | |
ows | -0.055 |
pering | -0.051 |
oms | -0.048 |
iving | -0.047 |
announcements | -0.046 |
Token | red |
Token ID | 2266 |
Feature activation | -7.90e+00 |
Pos logit contributions | |
flavor | +0.076 |
penny | +0.075 |
ornament | +0.073 |
peanut | +0.071 |
cube | +0.069 |
Neg logit contributions | |
ows | -0.121 |
pering | -0.117 |
arl | -0.113 |
obb | -0.113 |
oms | -0.111 |
Token | train |
Token ID | 4512 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
penny | +0.238 |
flavor | +0.233 |
peanut | +0.233 |
ornament | +0.229 |
pol | +0.225 |
Neg logit contributions | |
ows | -0.219 |
arl | -0.199 |
oms | -0.198 |
uits | -0.196 |
men | -0.184 |
Token | with |
Token ID | 351 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
penny | +0.014 |
flavor | +0.014 |
ornament | +0.014 |
cube | +0.014 |
peanut | +0.013 |
Neg logit contributions | |
pering | -0.016 |
ows | -0.016 |
obb | -0.016 |
arl | -0.015 |
iving | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
flavor | +0.076 |
penny | +0.076 |
ornament | +0.074 |
cube | +0.073 |
peanut | +0.071 |
Neg logit contributions | |
ows | -0.097 |
pering | -0.097 |
obb | -0.092 |
arl | -0.092 |
oms | -0.090 |
Token | yellow |
Token ID | 7872 |
Feature activation | -5.45e+00 |
Pos logit contributions | |
ows | +0.011 |
pering | +0.010 |
arl | +0.010 |
oms | +0.010 |
announcements | +0.010 |
Neg logit contributions | |
flavor | -0.009 |
ornament | -0.009 |
penny | -0.009 |
peanut | -0.009 |
seed | -0.008 |
Token | bow |
Token ID | 9563 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
flavor | +0.145 |
penny | +0.143 |
mar | +0.142 |
ornament | +0.141 |
pol | +0.141 |
Neg logit contributions | |
ows | -0.162 |
oms | -0.158 |
arl | -0.157 |
uits | -0.149 |
iving | -0.142 |
Token | my |
Token ID | 1820 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
penny | +0.045 |
flavor | +0.040 |
dime | +0.040 |
peanut | +0.038 |
ornament | +0.037 |
Neg logit contributions | |
ows | -0.083 |
obb | -0.077 |
pering | -0.075 |
arl | -0.073 |
realize | -0.073 |
Token | loved |
Token ID | 6151 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
penny | +0.080 |
flavor | +0.077 |
dime | +0.075 |
ornament | +0.075 |
ruby | +0.071 |
Neg logit contributions | |
ows | -0.063 |
obb | -0.055 |
sons | -0.053 |
realize | -0.053 |
pering | -0.053 |
Token | wearing |
Token ID | 5762 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
penny | +0.014 |
c | +0.014 |
flavor | +0.013 |
ornament | +0.013 |
ruby | +0.013 |
Neg logit contributions | |
pering | -0.015 |
dolphins | -0.015 |
announcements | -0.014 |
men | -0.014 |
arl | -0.014 |
Token | his |
Token ID | 465 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
flavor | +0.071 |
penny | +0.070 |
ornament | +0.068 |
cube | +0.066 |
peanut | +0.062 |
Neg logit contributions | |
ows | -0.089 |
obb | -0.087 |
pering | -0.087 |
arl | -0.086 |
announcements | -0.085 |
Token | favorite |
Token ID | 4004 |
Feature activation | -6.18e+00 |
Pos logit contributions | |
flavor | +0.086 |
ornament | +0.086 |
penny | +0.083 |
checked | +0.083 |
mar | +0.082 |
Neg logit contributions | |
ows | -0.096 |
obb | -0.094 |
arl | -0.091 |
sons | -0.086 |
Super | -0.085 |
Token | red |
Token ID | 2266 |
Feature activation | -7.87e+00 |
Pos logit contributions | |
checked | +0.196 |
pol | +0.186 |
mar | +0.186 |
Lem | +0.185 |
iw | +0.184 |
Neg logit contributions | |
loud | -0.141 |
dolphins | -0.135 |
dancers | -0.133 |
men | -0.131 |
tent | -0.130 |
Token | cap |
Token ID | 1451 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
pol | +0.238 |
mattered | +0.233 |
checked | +0.232 |
Lem | +0.231 |
wealth | +0.229 |
Neg logit contributions | |
arl | -0.185 |
uits | -0.173 |
ows | -0.169 |
oms | -0.165 |
men | -0.161 |
Token | everywhere |
Token ID | 8347 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
flavor | +0.091 |
penny | +0.091 |
ornament | +0.088 |
peanut | +0.087 |
cube | +0.085 |
Neg logit contributions | |
ows | -0.071 |
pering | -0.058 |
obb | -0.058 |
adders | -0.056 |
iving | -0.055 |
Token | he |
Token ID | 339 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ows | +0.020 |
pering | +0.019 |
obb | +0.019 |
arl | +0.018 |
iving | +0.018 |
Neg logit contributions | |
penny | -0.024 |
flavor | -0.024 |
ornament | -0.023 |
peanut | -0.023 |
cube | -0.022 |
Token | went |
Token ID | 1816 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
ows | +0.043 |
obb | +0.042 |
pering | +0.042 |
arl | +0.041 |
iving | +0.040 |
Neg logit contributions | |
ornament | -0.048 |
flavor | -0.048 |
penny | -0.048 |
cube | -0.047 |
peanut | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
ows | +0.112 |
pering | +0.110 |
obb | +0.106 |
arl | +0.105 |
iving | +0.102 |
Neg logit contributions | |
penny | -0.099 |
flavor | -0.098 |
ornament | -0.096 |
cube | -0.094 |
peanut | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
penny | +0.062 |
flavor | +0.061 |
ornament | +0.058 |
peanut | +0.058 |
flavour | +0.056 |
Neg logit contributions | |
pering | -0.028 |
ows | -0.027 |
arl | -0.026 |
iving | -0.023 |
obb | -0.023 |
Token | Tim |
Token ID | 5045 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
penny | +0.093 |
ornament | +0.086 |
jewel | +0.081 |
peanut | +0.081 |
ruby | +0.081 |
Neg logit contributions | |
ows | -0.055 |
arl | -0.054 |
pering | -0.052 |
obb | -0.051 |
adders | -0.048 |
Token | my |
Token ID | 1820 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
penny | +0.056 |
dime | +0.050 |
flavor | +0.049 |
peanut | +0.048 |
mint | +0.046 |
Neg logit contributions | |
ows | -0.085 |
obb | -0.078 |
pering | -0.074 |
realize | -0.074 |
arl | -0.074 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
penny | +0.090 |
dime | +0.088 |
flavor | +0.085 |
gu | +0.083 |
ruby | +0.082 |
Neg logit contributions | |
ows | -0.050 |
realize | -0.042 |
obb | -0.041 |
sons | -0.040 |
Super | -0.039 |
Token | his |
Token ID | 465 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
c | +0.021 |
ruby | +0.021 |
penny | +0.020 |
flavor | +0.020 |
ornament | +0.020 |
Neg logit contributions | |
dolphins | -0.019 |
pering | -0.019 |
men | -0.018 |
dancers | -0.018 |
Super | -0.018 |
Token | red |
Token ID | 2266 |
Feature activation | -7.86e+00 |
Pos logit contributions | |
seed | +0.072 |
c | +0.070 |
checked | +0.068 |
mar | +0.068 |
pol | +0.066 |
Neg logit contributions | |
arl | -0.092 |
sons | -0.088 |
Super | -0.086 |
obb | -0.085 |
ows | -0.084 |
Token | shirt |
Token ID | 10147 |
Feature activation | -6.08e+00 |
Pos logit contributions | |
mattered | +0.223 |
checked | +0.221 |
pol | +0.220 |
wealth | +0.217 |
mar | +0.214 |
Neg logit contributions | |
arl | -0.182 |
men | -0.168 |
loud | -0.165 |
ows | -0.164 |
ropes | -0.164 |
Token | and |
Token ID | 290 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
penny | +0.193 |
flavor | +0.189 |
shr | +0.188 |
peanut | +0.187 |
seed | +0.186 |
Neg logit contributions | |
ows | -0.175 |
obb | -0.149 |
pering | -0.138 |
adders | -0.134 |
iving | -0.133 |
Token | wore |
Token ID | 12408 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
ornament | +0.046 |
penny | +0.045 |
shr | +0.044 |
ake | +0.043 |
ase | +0.043 |
Neg logit contributions | |
Super | -0.042 |
ows | -0.042 |
realize | -0.041 |
uits | -0.039 |
announcements | -0.039 |
Token | it |
Token ID | 340 |
Feature activation | -4.68e+00 |
Pos logit contributions | |
ornament | +0.077 |
flavor | +0.075 |
cube | +0.075 |
penny | +0.073 |
seed | +0.072 |
Neg logit contributions | |
ows | -0.069 |
Super | -0.066 |
announcements | -0.066 |
aults | -0.066 |
uits | -0.065 |
Token | every |
Token ID | 790 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
flavor | +0.153 |
penny | +0.151 |
ornament | +0.148 |
peanut | +0.146 |
cube | +0.143 |
Neg logit contributions | |
ows | -0.137 |
obb | -0.127 |
pering | -0.124 |
arl | -0.118 |
iving | -0.114 |
Token | . |
Token ID | 13 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
penny | +0.086 |
flavor | +0.085 |
ornament | +0.084 |
peanut | +0.081 |
cube | +0.080 |
Neg logit contributions | |
ows | -0.057 |
pering | -0.056 |
arl | -0.052 |
iving | -0.051 |
oms | -0.049 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
penny | +0.074 |
ornament | +0.071 |
flavor | +0.069 |
peanut | +0.068 |
ruby | +0.066 |
Neg logit contributions | |
ows | -0.036 |
arl | -0.035 |
pering | -0.032 |
obb | -0.030 |
iving | -0.030 |
Token | wore |
Token ID | 12408 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
flavor | +0.037 |
penny | +0.037 |
ornament | +0.035 |
cube | +0.034 |
peanut | +0.034 |
Neg logit contributions | |
pering | -0.080 |
ows | -0.080 |
arl | -0.078 |
obb | -0.077 |
iving | -0.075 |
Token | his |
Token ID | 465 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
ornament | +0.100 |
penny | +0.100 |
cube | +0.099 |
flavor | +0.097 |
occasion | +0.095 |
Neg logit contributions | |
ows | -0.061 |
obb | -0.057 |
Super | -0.055 |
announcements | -0.055 |
uits | -0.054 |
Token | big |
Token ID | 1263 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
ornament | +0.086 |
flavor | +0.085 |
checked | +0.083 |
penny | +0.082 |
shr | +0.080 |
Neg logit contributions | |
ows | -0.112 |
obb | -0.111 |
arl | -0.109 |
oms | -0.102 |
pering | -0.102 |
Token | red |
Token ID | 2266 |
Feature activation | -7.84e+00 |
Pos logit contributions | |
checked | +0.100 |
shr | +0.099 |
pol | +0.098 |
ornament | +0.096 |
mar | +0.096 |
Neg logit contributions | |
oms | -0.087 |
men | -0.087 |
arl | -0.085 |
obb | -0.085 |
ows | -0.083 |
Token | coat |
Token ID | 13209 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
checked | +0.238 |
pol | +0.236 |
ornament | +0.234 |
mattered | +0.233 |
flavour | +0.227 |
Neg logit contributions | |
arl | -0.201 |
uits | -0.185 |
oms | -0.184 |
ows | -0.183 |
obb | -0.166 |
Token | because |
Token ID | 780 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
penny | +0.108 |
flavor | +0.108 |
ornament | +0.106 |
peanut | +0.103 |
cube | +0.103 |
Neg logit contributions | |
ows | -0.091 |
pering | -0.088 |
obb | -0.083 |
arl | -0.082 |
iving | -0.079 |
Token | it |
Token ID | 340 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
flavor | +0.005 |
penny | +0.005 |
ornament | +0.005 |
peanut | +0.005 |
cube | +0.005 |
Neg logit contributions | |
ows | -0.006 |
pering | -0.006 |
adders | -0.006 |
obb | -0.006 |
arl | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
flavor | +0.092 |
dime | +0.091 |
peanut | +0.089 |
penny | +0.089 |
ornament | +0.088 |
Neg logit contributions | |
ows | -0.113 |
obb | -0.102 |
Super | -0.094 |
pering | -0.092 |
announcements | -0.092 |
Token | cold |
Token ID | 4692 |
Feature activation | +7.86e-02 |
Pos logit contributions | |
penny | +0.005 |
flavor | +0.005 |
ornament | +0.005 |
cube | +0.005 |
peanut | +0.005 |
Neg logit contributions | |
ows | -0.006 |
pering | -0.006 |
obb | -0.005 |
arl | -0.005 |
iving | -0.005 |
Token | He |
Token ID | 679 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
penny | +0.096 |
ornament | +0.088 |
peanut | +0.083 |
possession | +0.083 |
jewel | +0.082 |
Neg logit contributions | |
ows | -0.059 |
arl | -0.057 |
pering | -0.054 |
obb | -0.053 |
adders | -0.051 |
Token | loved |
Token ID | 6151 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
ows | +0.001 |
obb | +0.001 |
Super | +0.001 |
pering | +0.001 |
announcements | +0.001 |
Neg logit contributions | |
penny | -0.001 |
flavor | -0.001 |
ornament | -0.001 |
dime | -0.001 |
seed | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
penny | +0.010 |
flavor | +0.010 |
ornament | +0.010 |
c | +0.010 |
orn | +0.010 |
Neg logit contributions | |
pering | -0.011 |
dolphins | -0.011 |
announcements | -0.011 |
men | -0.010 |
dancers | -0.010 |
Token | wear |
Token ID | 5806 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
penny | +0.035 |
flavor | +0.033 |
ornament | +0.033 |
peanut | +0.032 |
cube | +0.032 |
Neg logit contributions | |
ows | -0.045 |
pering | -0.044 |
obb | -0.044 |
arl | -0.043 |
iving | -0.042 |
Token | his |
Token ID | 465 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
flavor | +0.077 |
penny | +0.077 |
ornament | +0.076 |
cube | +0.073 |
peanut | +0.070 |
Neg logit contributions | |
ows | -0.114 |
pering | -0.111 |
obb | -0.110 |
arl | -0.109 |
oms | -0.104 |
Token | red |
Token ID | 2266 |
Feature activation | -7.80e+00 |
Pos logit contributions | |
ornament | +0.085 |
flavor | +0.083 |
checked | +0.080 |
penny | +0.079 |
seed | +0.079 |
Neg logit contributions | |
ows | -0.103 |
obb | -0.103 |
arl | -0.102 |
Super | -0.095 |
sons | -0.094 |
Token | hat |
Token ID | 6877 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
checked | +0.248 |
pol | +0.236 |
mattered | +0.232 |
wealth | +0.231 |
seed | +0.229 |
Neg logit contributions | |
arl | -0.181 |
uits | -0.165 |
ows | -0.164 |
oms | -0.164 |
Super | -0.154 |
Token | everywhere |
Token ID | 8347 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
flavor | +0.120 |
penny | +0.119 |
ornament | +0.114 |
peanut | +0.113 |
cube | +0.112 |
Neg logit contributions | |
ows | -0.109 |
pering | -0.097 |
obb | -0.095 |
arl | -0.094 |
iving | -0.093 |
Token | he |
Token ID | 339 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ows | +0.014 |
pering | +0.014 |
obb | +0.013 |
arl | +0.013 |
iving | +0.013 |
Neg logit contributions | |
penny | -0.013 |
flavor | -0.013 |
ornament | -0.013 |
peanut | -0.013 |
cube | -0.012 |
Token | went |
Token ID | 1816 |
Feature activation | +3.40e+00 |
Pos logit contributions | |
ows | +0.037 |
obb | +0.035 |
pering | +0.035 |
arl | +0.034 |
oms | +0.034 |
Neg logit contributions | |
penny | -0.038 |
flavor | -0.038 |
ornament | -0.038 |
cube | -0.037 |
peanut | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
ows | +0.109 |
pering | +0.109 |
obb | +0.104 |
arl | +0.103 |
iving | +0.100 |
Neg logit contributions | |
penny | -0.102 |
flavor | -0.101 |
ornament | -0.099 |
peanut | -0.096 |
cube | -0.096 |
Token | She |
Token ID | 1375 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
penny | +0.098 |
flavor | +0.093 |
ornament | +0.089 |
cube | +0.088 |
peanut | +0.087 |
Neg logit contributions | |
ows | -0.059 |
arl | -0.056 |
pering | -0.055 |
obb | -0.051 |
iving | -0.051 |
Token | loved |
Token ID | 6151 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
ows | +0.003 |
obb | +0.003 |
pering | +0.003 |
announcements | +0.003 |
Super | +0.002 |
Neg logit contributions | |
penny | -0.003 |
flavor | -0.003 |
ornament | -0.003 |
dime | -0.003 |
peanut | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
penny | +0.012 |
ruby | +0.011 |
flavor | +0.011 |
orn | +0.011 |
c | +0.011 |
Neg logit contributions | |
dancers | -0.011 |
pering | -0.011 |
dolphins | -0.011 |
arl | -0.011 |
announcements | -0.011 |
Token | wear |
Token ID | 5806 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
penny | +0.038 |
flavor | +0.036 |
ornament | +0.035 |
cube | +0.034 |
peanut | +0.034 |
Neg logit contributions | |
ows | -0.046 |
obb | -0.045 |
pering | -0.044 |
arl | -0.044 |
iving | -0.042 |
Token | her |
Token ID | 607 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
cube | +0.091 |
flavor | +0.091 |
penny | +0.090 |
ornament | +0.087 |
nut | +0.084 |
Neg logit contributions | |
ows | -0.096 |
obb | -0.093 |
arl | -0.092 |
pering | -0.089 |
announcements | -0.087 |
Token | orange |
Token ID | 10912 |
Feature activation | -7.79e+00 |
Pos logit contributions | |
flavor | +0.080 |
nut | +0.075 |
ornament | +0.074 |
cube | +0.074 |
seed | +0.073 |
Neg logit contributions | |
ows | -0.094 |
obb | -0.092 |
arl | -0.091 |
oms | -0.085 |
Super | -0.084 |
Token | bath |
Token ID | 7837 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
dime | +0.228 |
checked | +0.227 |
pol | +0.225 |
flavour | +0.223 |
wealth | +0.221 |
Neg logit contributions | |
dancers | -0.172 |
oms | -0.166 |
arl | -0.164 |
ows | -0.163 |
ropes | -0.158 |
Token | robe |
Token ID | 25481 |
Feature activation | -9.49e-01 |
Pos logit contributions | |
penny | +0.081 |
flavor | +0.080 |
ornament | +0.078 |
peanut | +0.076 |
cube | +0.076 |
Neg logit contributions | |
ows | -0.091 |
pering | -0.090 |
obb | -0.086 |
arl | -0.085 |
iving | -0.083 |
Token | after |
Token ID | 706 |
Feature activation | +9.11e-01 |
Pos logit contributions | |
penny | +0.028 |
flavor | +0.028 |
ornament | +0.027 |
peanut | +0.026 |
cube | +0.026 |
Neg logit contributions | |
ows | -0.031 |
pering | -0.031 |
obb | -0.030 |
arl | -0.029 |
iving | -0.028 |
Token | taking |
Token ID | 2263 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
pering | +0.033 |
ows | +0.033 |
obb | +0.032 |
arl | +0.031 |
iving | +0.030 |
Neg logit contributions | |
penny | -0.023 |
flavor | -0.023 |
ornament | -0.022 |
cube | -0.022 |
peanut | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
ows | +0.074 |
pering | +0.072 |
obb | +0.072 |
announcements | +0.071 |
arl | +0.068 |
Neg logit contributions | |
cube | -0.077 |
flavor | -0.076 |
ornament | -0.074 |
penny | -0.074 |
magnet | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
ows | +0.011 |
pering | +0.011 |
iving | +0.010 |
arl | +0.010 |
oms | +0.010 |
Neg logit contributions | |
penny | -0.016 |
flavor | -0.016 |
ornament | -0.016 |
peanut | -0.015 |
cube | -0.015 |
Token | She |
Token ID | 1375 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
penny | +0.052 |
flavor | +0.052 |
ornament | +0.050 |
cube | +0.050 |
peanut | +0.048 |
Neg logit contributions | |
ows | -0.044 |
pering | -0.043 |
arl | -0.041 |
iving | -0.039 |
obb | -0.039 |
Token | wore |
Token ID | 12408 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
ows | +0.013 |
pering | +0.013 |
obb | +0.013 |
oms | +0.012 |
announcements | +0.012 |
Neg logit contributions | |
penny | -0.011 |
flavor | -0.011 |
ornament | -0.011 |
cube | -0.010 |
peanut | -0.010 |
Token | her |
Token ID | 607 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
cube | +0.099 |
flavor | +0.094 |
penny | +0.093 |
ornament | +0.090 |
magnet | +0.088 |
Neg logit contributions | |
ows | -0.054 |
obb | -0.049 |
announcements | -0.049 |
aults | -0.048 |
uits | -0.048 |
Token | favorite |
Token ID | 4004 |
Feature activation | -6.25e+00 |
Pos logit contributions | |
flavor | +0.081 |
ornament | +0.078 |
penny | +0.076 |
cube | +0.074 |
peanut | +0.074 |
Neg logit contributions | |
ows | -0.106 |
obb | -0.102 |
arl | -0.098 |
pering | -0.095 |
adders | -0.092 |
Token | pink |
Token ID | 11398 |
Feature activation | -7.78e+00 |
Pos logit contributions | |
checked | +0.210 |
Lem | +0.195 |
shr | +0.193 |
Smith | +0.193 |
mar | +0.190 |
Neg logit contributions | |
ows | -0.128 |
loud | -0.128 |
ropes | -0.120 |
iet | -0.119 |
dancers | -0.116 |
Token | boots |
Token ID | 14412 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
checked | +0.248 |
pol | +0.243 |
wealth | +0.243 |
Lem | +0.243 |
flavor | +0.240 |
Neg logit contributions | |
ows | -0.156 |
arl | -0.154 |
oms | -0.151 |
uits | -0.148 |
dancers | -0.147 |
Token | , |
Token ID | 11 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
flavor | +0.050 |
penny | +0.049 |
ornament | +0.048 |
peanut | +0.048 |
cube | +0.047 |
Neg logit contributions | |
ows | -0.040 |
pering | -0.037 |
obb | -0.035 |
arl | -0.035 |
iving | -0.034 |
Token | which |
Token ID | 543 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
penny | +0.025 |
flavor | +0.024 |
ornament | +0.024 |
peanut | +0.023 |
cube | +0.023 |
Neg logit contributions | |
ows | -0.025 |
pering | -0.024 |
obb | -0.023 |
arl | -0.023 |
iving | -0.022 |
Token | made |
Token ID | 925 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
penny | +0.050 |
flavor | +0.050 |
ornament | +0.049 |
cube | +0.048 |
peanut | +0.048 |
Neg logit contributions | |
ows | -0.077 |
pering | -0.072 |
obb | -0.071 |
adders | -0.070 |
oms | -0.069 |
Token | her |
Token ID | 607 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
flavor | +0.008 |
ornament | +0.008 |
peanut | +0.007 |
penny | +0.007 |
color | +0.007 |
Neg logit contributions | |
pering | -0.034 |
arl | -0.032 |
ows | -0.031 |
obb | -0.031 |
sons | -0.030 |
Token | who |
Token ID | 508 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
pering | +0.003 |
ows | +0.003 |
oms | +0.003 |
iving | +0.003 |
obb | +0.003 |
Neg logit contributions | |
penny | -0.007 |
flavor | -0.007 |
ornament | -0.007 |
peanut | -0.007 |
cube | -0.007 |
Token | liked |
Token ID | 8288 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
penny | +0.072 |
flavor | +0.070 |
ornament | +0.070 |
peanut | +0.067 |
cube | +0.065 |
Neg logit contributions | |
ows | -0.096 |
pering | -0.091 |
obb | -0.091 |
arl | -0.087 |
iving | -0.086 |
Token | to |
Token ID | 284 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
penny | +0.029 |
flavor | +0.029 |
ornament | +0.028 |
peanut | +0.026 |
cube | +0.026 |
Neg logit contributions | |
pering | -0.053 |
ows | -0.052 |
arl | -0.050 |
obb | -0.050 |
iving | -0.050 |
Token | wear |
Token ID | 5806 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
flavor | +0.039 |
penny | +0.038 |
ornament | +0.037 |
cube | +0.036 |
peanut | +0.036 |
Neg logit contributions | |
pering | -0.054 |
ows | -0.053 |
arl | -0.051 |
obb | -0.050 |
iving | -0.049 |
Token | his |
Token ID | 465 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
penny | +0.064 |
flavor | +0.064 |
ornament | +0.062 |
cube | +0.059 |
peanut | +0.059 |
Neg logit contributions | |
ows | -0.128 |
pering | -0.127 |
obb | -0.123 |
arl | -0.122 |
iving | -0.119 |
Token | red |
Token ID | 2266 |
Feature activation | -7.77e+00 |
Pos logit contributions | |
flavor | +0.074 |
ornament | +0.071 |
penny | +0.070 |
peanut | +0.068 |
seed | +0.066 |
Neg logit contributions | |
ows | -0.105 |
obb | -0.105 |
arl | -0.103 |
pering | -0.100 |
oms | -0.100 |
Token | cap |
Token ID | 1451 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
flavor | +0.212 |
flavour | +0.209 |
dime | +0.209 |
pol | +0.208 |
checked | +0.207 |
Neg logit contributions | |
ows | -0.211 |
arl | -0.211 |
oms | -0.202 |
uits | -0.198 |
iving | -0.184 |
Token | . |
Token ID | 13 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
penny | +0.089 |
flavor | +0.089 |
ornament | +0.087 |
peanut | +0.085 |
cube | +0.084 |
Neg logit contributions | |
ows | -0.077 |
pering | -0.075 |
obb | -0.072 |
arl | -0.071 |
iving | -0.069 |
Token | He |
Token ID | 679 |
Feature activation | +9.48e-02 |
Pos logit contributions | |
penny | +0.060 |
ornament | +0.058 |
peanut | +0.056 |
flavor | +0.056 |
cube | +0.054 |
Neg logit contributions | |
ows | -0.062 |
arl | -0.060 |
pering | -0.059 |
obb | -0.059 |
iving | -0.057 |
Token | always |
Token ID | 1464 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
ows | +0.004 |
pering | +0.003 |
obb | +0.003 |
oms | +0.003 |
arl | +0.003 |
Neg logit contributions | |
penny | -0.002 |
flavor | -0.002 |
ornament | -0.002 |
peanut | -0.002 |
cube | -0.002 |
Token | did |
Token ID | 750 |
Feature activation | -6.11e-01 |
Pos logit contributions | |
penny | +0.044 |
ornament | +0.044 |
flavor | +0.043 |
peanut | +0.041 |
cube | +0.041 |
Neg logit contributions | |
ows | -0.079 |
pering | -0.075 |
obb | -0.074 |
arl | -0.073 |
announcements | -0.072 |
Token | who |
Token ID | 508 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
flavor | +0.017 |
penny | +0.017 |
ornament | +0.017 |
peanut | +0.017 |
cube | +0.017 |
Neg logit contributions | |
ows | -0.011 |
pering | -0.011 |
oms | -0.009 |
iving | -0.009 |
arl | -0.009 |
Token | liked |
Token ID | 8288 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
penny | +0.068 |
flavor | +0.067 |
ornament | +0.065 |
cube | +0.063 |
peanut | +0.063 |
Neg logit contributions | |
ows | -0.117 |
pering | -0.116 |
obb | -0.112 |
arl | -0.111 |
iving | -0.108 |
Token | to |
Token ID | 284 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
penny | +0.027 |
flavor | +0.027 |
ornament | +0.026 |
cube | +0.025 |
peanut | +0.025 |
Neg logit contributions | |
pering | -0.052 |
ows | -0.051 |
arl | -0.049 |
obb | -0.049 |
iving | -0.049 |
Token | wear |
Token ID | 5806 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
flavor | +0.043 |
penny | +0.042 |
ornament | +0.041 |
peanut | +0.040 |
cube | +0.040 |
Neg logit contributions | |
ows | -0.054 |
pering | -0.054 |
obb | -0.051 |
arl | -0.051 |
iving | -0.050 |
Token | her |
Token ID | 607 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
penny | +0.078 |
flavor | +0.077 |
ornament | +0.075 |
cube | +0.074 |
peanut | +0.072 |
Neg logit contributions | |
ows | -0.116 |
pering | -0.114 |
obb | -0.111 |
arl | -0.110 |
oms | -0.107 |
Token | red |
Token ID | 2266 |
Feature activation | -7.75e+00 |
Pos logit contributions | |
flavor | +0.069 |
penny | +0.067 |
ornament | +0.067 |
cube | +0.065 |
peanut | +0.064 |
Neg logit contributions | |
ows | -0.104 |
obb | -0.099 |
pering | -0.098 |
arl | -0.098 |
oms | -0.097 |
Token | dress |
Token ID | 6576 |
Feature activation | -5.63e+00 |
Pos logit contributions | |
dime | +0.216 |
flavor | +0.214 |
penny | +0.211 |
checked | +0.209 |
pol | +0.209 |
Neg logit contributions | |
ows | -0.214 |
arl | -0.213 |
oms | -0.206 |
uits | -0.194 |
iving | -0.188 |
Token | and |
Token ID | 290 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
penny | +0.173 |
flavor | +0.173 |
ornament | +0.169 |
cube | +0.164 |
peanut | +0.164 |
Neg logit contributions | |
pering | -0.177 |
ows | -0.173 |
obb | -0.167 |
arl | -0.165 |
iving | -0.160 |
Token | her |
Token ID | 607 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
penny | +0.044 |
flavor | +0.043 |
ornament | +0.043 |
cube | +0.043 |
peanut | +0.040 |
Neg logit contributions | |
ows | -0.051 |
pering | -0.049 |
arl | -0.048 |
announcements | -0.047 |
obb | -0.047 |
Token | shiny |
Token ID | 22441 |
Feature activation | -5.86e+00 |
Pos logit contributions | |
flavor | +0.056 |
penny | +0.055 |
ornament | +0.053 |
peanut | +0.052 |
cube | +0.052 |
Neg logit contributions | |
ows | -0.098 |
pering | -0.093 |
obb | -0.093 |
arl | -0.093 |
oms | -0.090 |
Token | shoes |
Token ID | 10012 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
flavor | +0.142 |
peanut | +0.136 |
penny | +0.135 |
ornament | +0.133 |
cube | +0.132 |
Neg logit contributions | |
ows | -0.209 |
arl | -0.194 |
obb | -0.192 |
oms | -0.192 |
pering | -0.184 |
Token | outside |
Token ID | 2354 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
ows | +0.013 |
pering | +0.013 |
obb | +0.013 |
iving | +0.013 |
arl | +0.013 |
Neg logit contributions | |
penny | -0.013 |
flavor | -0.013 |
ornament | -0.012 |
peanut | -0.012 |
cube | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
penny | +0.006 |
flavor | +0.006 |
ornament | +0.006 |
peanut | +0.006 |
cube | +0.006 |
Neg logit contributions | |
ows | -0.007 |
pering | -0.007 |
obb | -0.007 |
arl | -0.007 |
iving | -0.007 |
Token | especially |
Token ID | 2592 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
flavor | +0.035 |
cube | +0.034 |
penny | +0.034 |
ornament | +0.033 |
peanut | +0.033 |
Neg logit contributions | |
ows | -0.037 |
arl | -0.036 |
pering | -0.036 |
obb | -0.036 |
oms | -0.034 |
Token | with |
Token ID | 351 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
penny | +0.094 |
flavor | +0.093 |
ornament | +0.091 |
cube | +0.089 |
peanut | +0.089 |
Neg logit contributions | |
pering | -0.084 |
ows | -0.084 |
arl | -0.079 |
obb | -0.079 |
iving | -0.078 |
Token | his |
Token ID | 465 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
penny | +0.064 |
flavor | +0.064 |
ornament | +0.062 |
cube | +0.059 |
peanut | +0.059 |
Neg logit contributions | |
ows | -0.128 |
pering | -0.128 |
obb | -0.123 |
arl | -0.123 |
iving | -0.120 |
Token | red |
Token ID | 2266 |
Feature activation | -7.75e+00 |
Pos logit contributions | |
flavor | +0.090 |
penny | +0.089 |
ornament | +0.089 |
peanut | +0.084 |
cube | +0.081 |
Neg logit contributions | |
ows | -0.105 |
pering | -0.101 |
arl | -0.101 |
obb | -0.100 |
iving | -0.096 |
Token | flag |
Token ID | 6056 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
penny | +0.232 |
pol | +0.230 |
ornament | +0.230 |
flavor | +0.227 |
dime | +0.220 |
Neg logit contributions | |
arl | -0.206 |
ows | -0.196 |
iving | -0.187 |
oms | -0.184 |
obb | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
flavor | +0.128 |
penny | +0.128 |
ornament | +0.124 |
peanut | +0.122 |
cube | +0.120 |
Neg logit contributions | |
ows | -0.096 |
pering | -0.092 |
arl | -0.087 |
obb | -0.086 |
iving | -0.084 |
Token | One |
Token ID | 1881 |
Feature activation | +3.37e-03 |
Pos logit contributions | |
penny | +0.057 |
ornament | +0.052 |
flavor | +0.050 |
peanut | +0.050 |
jewel | +0.048 |
Neg logit contributions | |
ows | -0.047 |
arl | -0.045 |
pering | -0.042 |
iving | -0.041 |
Super | -0.040 |
Token | day |
Token ID | 1110 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
iving | +0.000 |
obb | +0.000 |
ila | +0.000 |
wing | +0.000 |
ows | +0.000 |
Neg logit contributions | |
dime | -0.000 |
penny | -0.000 |
ornament | -0.000 |
ruby | -0.000 |
magnet | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
ornament | +0.006 |
penny | +0.006 |
flavor | +0.006 |
ruby | +0.006 |
c | +0.006 |
Neg logit contributions | |
iving | -0.003 |
arl | -0.003 |
oms | -0.002 |
obb | -0.002 |
ows | -0.002 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
penny | +0.085 |
dime | +0.080 |
flavor | +0.080 |
ornament | +0.079 |
ruby | +0.075 |
Neg logit contributions | |
ows | -0.065 |
obb | -0.056 |
realize | -0.054 |
pering | -0.054 |
sons | -0.053 |
Token | to |
Token ID | 284 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
c | +0.021 |
ruby | +0.021 |
penny | +0.020 |
orn | +0.020 |
ornament | +0.020 |
Neg logit contributions | |
dolphins | -0.020 |
dancers | -0.019 |
men | -0.019 |
pering | -0.019 |
announcements | -0.019 |
Token | play |
Token ID | 711 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
penny | +0.047 |
flavor | +0.045 |
ornament | +0.045 |
peanut | +0.044 |
cube | +0.043 |
Neg logit contributions | |
ows | -0.063 |
pering | -0.061 |
obb | -0.060 |
arl | -0.060 |
oms | -0.058 |
Token | with |
Token ID | 351 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
ows | +0.010 |
pering | +0.009 |
obb | +0.009 |
arl | +0.009 |
iving | +0.009 |
Neg logit contributions | |
penny | -0.008 |
flavor | -0.008 |
ornament | -0.008 |
peanut | -0.008 |
cube | -0.008 |
Token | his |
Token ID | 465 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
penny | +0.068 |
flavor | +0.067 |
ornament | +0.064 |
peanut | +0.061 |
cube | +0.059 |
Neg logit contributions | |
obb | -0.095 |
ows | -0.093 |
pering | -0.093 |
arl | -0.091 |
adders | -0.090 |
Token | red |
Token ID | 2266 |
Feature activation | -7.73e+00 |
Pos logit contributions | |
flavor | +0.072 |
penny | +0.069 |
ornament | +0.068 |
peanut | +0.066 |
pol | +0.065 |
Neg logit contributions | |
arl | -0.105 |
ows | -0.104 |
obb | -0.104 |
pering | -0.103 |
sons | -0.102 |
Token | ball |
Token ID | 2613 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
pol | +0.237 |
mattered | +0.233 |
wealth | +0.232 |
mar | +0.228 |
checked | +0.219 |
Neg logit contributions | |
arl | -0.186 |
men | -0.171 |
ows | -0.167 |
iving | -0.166 |
oms | -0.166 |
Token | outside |
Token ID | 2354 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
flavor | +0.150 |
penny | +0.149 |
ornament | +0.142 |
peanut | +0.142 |
cube | +0.137 |
Neg logit contributions | |
ows | -0.126 |
pering | -0.121 |
obb | -0.118 |
arl | -0.113 |
iving | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
flavor | +0.021 |
cube | +0.021 |
ornament | +0.021 |
penny | +0.021 |
peanut | +0.019 |
Neg logit contributions | |
ows | -0.029 |
pering | -0.028 |
arl | -0.027 |
obb | -0.027 |
oms | -0.026 |
Token | One |
Token ID | 1881 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
penny | +0.067 |
ornament | +0.061 |
peanut | +0.061 |
seed | +0.060 |
flavor | +0.059 |
Neg logit contributions | |
ows | -0.052 |
arl | -0.050 |
pering | -0.046 |
iving | -0.046 |
Super | -0.045 |
Token | day |
Token ID | 1110 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
dime | +0.005 |
penny | +0.004 |
ruby | +0.004 |
ornament | +0.004 |
cand | +0.004 |
Neg logit contributions | |
iving | -0.005 |
obb | -0.004 |
arl | -0.004 |
ila | -0.004 |
ows | -0.004 |
Token | loved |
Token ID | 6151 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
ows | +0.001 |
obb | +0.001 |
pering | +0.000 |
announcements | +0.000 |
arl | +0.000 |
Neg logit contributions | |
penny | -0.001 |
flavor | -0.001 |
ornament | -0.001 |
dime | -0.001 |
peanut | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
penny | +0.014 |
flavor | +0.014 |
ruby | +0.014 |
orn | +0.014 |
ornament | +0.014 |
Neg logit contributions | |
pering | -0.014 |
dancers | -0.014 |
dolphins | -0.014 |
announcements | -0.014 |
arl | -0.013 |
Token | wear |
Token ID | 5806 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
penny | +0.038 |
flavor | +0.037 |
ornament | +0.037 |
cube | +0.035 |
peanut | +0.035 |
Neg logit contributions | |
ows | -0.050 |
pering | -0.049 |
obb | -0.048 |
arl | -0.047 |
iving | -0.046 |
Token | her |
Token ID | 607 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
flavor | +0.090 |
penny | +0.089 |
cube | +0.089 |
ornament | +0.088 |
nut | +0.082 |
Neg logit contributions | |
ows | -0.102 |
obb | -0.097 |
arl | -0.097 |
pering | -0.096 |
announcements | -0.092 |
Token | favorite |
Token ID | 4004 |
Feature activation | -6.28e+00 |
Pos logit contributions | |
flavor | +0.083 |
ornament | +0.081 |
penny | +0.077 |
nut | +0.076 |
cube | +0.075 |
Neg logit contributions | |
ows | -0.099 |
obb | -0.098 |
arl | -0.096 |
oms | -0.092 |
pering | -0.089 |
Token | pink |
Token ID | 11398 |
Feature activation | -7.71e+00 |
Pos logit contributions | |
checked | +0.212 |
shr | +0.198 |
mar | +0.195 |
Smith | +0.194 |
pol | +0.191 |
Neg logit contributions | |
loud | -0.121 |
dancers | -0.120 |
iet | -0.119 |
dolphins | -0.119 |
ropes | -0.119 |
Token | dress |
Token ID | 6576 |
Feature activation | -5.62e+00 |
Pos logit contributions | |
checked | +0.231 |
pol | +0.231 |
wealth | +0.227 |
flavour | +0.223 |
flavor | +0.223 |
Neg logit contributions | |
oms | -0.168 |
arl | -0.168 |
dancers | -0.159 |
uits | -0.157 |
ows | -0.155 |
Token | every |
Token ID | 790 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
penny | +0.187 |
flavor | +0.185 |
ornament | +0.180 |
peanut | +0.179 |
cube | +0.177 |
Neg logit contributions | |
ows | -0.167 |
pering | -0.148 |
obb | -0.146 |
arl | -0.144 |
iving | -0.139 |
Token | day |
Token ID | 1110 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
arl | +0.040 |
obb | +0.039 |
iving | +0.038 |
oms | +0.037 |
pering | +0.037 |
Neg logit contributions | |
flavor | -0.034 |
dime | -0.033 |
seed | -0.033 |
peanut | -0.033 |
cube | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
ows | +0.029 |
pering | +0.028 |
obb | +0.025 |
arl | +0.025 |
oms | +0.025 |
Neg logit contributions | |
penny | -0.045 |
flavor | -0.045 |
ornament | -0.045 |
cube | -0.044 |
nut | -0.043 |
Token | One |
Token ID | 1881 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
penny | +0.061 |
cube | +0.053 |
possession | +0.052 |
symbol | +0.052 |
nut | +0.052 |
Neg logit contributions | |
ows | -0.049 |
arl | -0.046 |
oms | -0.041 |
iving | -0.040 |
ains | -0.039 |
Token | loved |
Token ID | 6151 |
Feature activation | -6.63e-01 |
Pos logit contributions | |
penny | +0.085 |
dime | +0.082 |
flavor | +0.081 |
ornament | +0.079 |
gu | +0.078 |
Neg logit contributions | |
ows | -0.054 |
realize | -0.047 |
obb | -0.047 |
sons | -0.045 |
Super | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
c | +0.018 |
penny | +0.018 |
ruby | +0.018 |
ornament | +0.018 |
seed | +0.017 |
Neg logit contributions | |
dolphins | -0.018 |
pering | -0.018 |
men | -0.017 |
announcements | -0.017 |
dancers | -0.017 |
Token | play |
Token ID | 711 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
penny | +0.042 |
flavor | +0.041 |
ornament | +0.040 |
peanut | +0.039 |
cube | +0.038 |
Neg logit contributions | |
ows | -0.060 |
pering | -0.059 |
obb | -0.058 |
arl | -0.057 |
iving | -0.056 |
Token | with |
Token ID | 351 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
ows | +0.011 |
pering | +0.011 |
obb | +0.010 |
iving | +0.010 |
arl | +0.010 |
Neg logit contributions | |
penny | -0.009 |
flavor | -0.009 |
ornament | -0.009 |
peanut | -0.009 |
cube | -0.009 |
Token | his |
Token ID | 465 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
penny | +0.071 |
flavor | +0.069 |
ornament | +0.067 |
mar | +0.064 |
peanut | +0.063 |
Neg logit contributions | |
obb | -0.088 |
ows | -0.084 |
pering | -0.083 |
adders | -0.083 |
arl | -0.082 |
Token | red |
Token ID | 2266 |
Feature activation | -7.71e+00 |
Pos logit contributions | |
flavor | +0.080 |
mar | +0.078 |
ornament | +0.077 |
c | +0.077 |
pol | +0.077 |
Neg logit contributions | |
sons | -0.099 |
arl | -0.094 |
obb | -0.093 |
ows | -0.092 |
men | -0.092 |
Token | wagon |
Token ID | 31860 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
pol | +0.243 |
mar | +0.243 |
wealth | +0.239 |
mattered | +0.237 |
checked | +0.233 |
Neg logit contributions | |
arl | -0.168 |
men | -0.158 |
ropes | -0.156 |
uits | -0.147 |
iving | -0.146 |
Token | . |
Token ID | 13 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
flavor | +0.142 |
penny | +0.140 |
peanut | +0.136 |
ornament | +0.133 |
cube | +0.126 |
Neg logit contributions | |
ows | -0.102 |
pering | -0.093 |
obb | -0.088 |
adders | -0.085 |
iving | -0.084 |
Token | He |
Token ID | 679 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
penny | +0.067 |
ornament | +0.062 |
peanut | +0.061 |
seed | +0.060 |
jewel | +0.058 |
Neg logit contributions | |
ows | -0.048 |
arl | -0.046 |
iving | -0.042 |
pering | -0.041 |
Super | -0.041 |
Token | would |
Token ID | 561 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
ows | +0.001 |
adders | +0.001 |
obb | +0.001 |
as | +0.001 |
uits | +0.001 |
Neg logit contributions | |
penny | -0.001 |
flavor | -0.001 |
ornament | -0.001 |
peanut | -0.001 |
avocado | -0.001 |
Token | pull |
Token ID | 2834 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ows | +0.008 |
pering | +0.008 |
arl | +0.008 |
obb | +0.008 |
oms | +0.008 |
Neg logit contributions | |
penny | -0.009 |
flavor | -0.009 |
ornament | -0.008 |
cube | -0.008 |
peanut | -0.008 |
Token | loved |
Token ID | 6151 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
penny | +0.060 |
flavor | +0.056 |
dime | +0.055 |
peanut | +0.054 |
ornament | +0.053 |
Neg logit contributions | |
ows | -0.061 |
obb | -0.055 |
pering | -0.052 |
adders | -0.051 |
realize | -0.050 |
Token | to |
Token ID | 284 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
c | +0.016 |
penny | +0.016 |
ornament | +0.016 |
seed | +0.015 |
orn | +0.015 |
Neg logit contributions | |
dolphins | -0.016 |
pering | -0.016 |
men | -0.016 |
dancers | -0.016 |
travelers | -0.015 |
Token | play |
Token ID | 711 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
penny | +0.037 |
flavor | +0.036 |
ornament | +0.035 |
peanut | +0.034 |
cube | +0.034 |
Neg logit contributions | |
ows | -0.054 |
pering | -0.053 |
obb | -0.052 |
arl | -0.051 |
iving | -0.050 |
Token | with |
Token ID | 351 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
ows | +0.014 |
pering | +0.013 |
obb | +0.013 |
arl | +0.013 |
iving | +0.013 |
Neg logit contributions | |
penny | -0.010 |
flavor | -0.010 |
ornament | -0.010 |
peanut | -0.010 |
cube | -0.010 |
Token | his |
Token ID | 465 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
penny | +0.068 |
flavor | +0.065 |
ornament | +0.064 |
mar | +0.060 |
peanut | +0.060 |
Neg logit contributions | |
obb | -0.086 |
ows | -0.084 |
adders | -0.082 |
pering | -0.081 |
arl | -0.080 |
Token | red |
Token ID | 2266 |
Feature activation | -7.70e+00 |
Pos logit contributions | |
flavor | +0.064 |
mar | +0.062 |
c | +0.062 |
seed | +0.061 |
peanut | +0.060 |
Neg logit contributions | |
sons | -0.098 |
arl | -0.096 |
obb | -0.096 |
ows | -0.094 |
men | -0.094 |
Token | blocks |
Token ID | 7021 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
mar | +0.250 |
seed | +0.249 |
pol | +0.245 |
mattered | +0.240 |
checked | +0.239 |
Neg logit contributions | |
arl | -0.168 |
men | -0.156 |
oms | -0.154 |
ropes | -0.151 |
iving | -0.148 |
Token | . |
Token ID | 13 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
penny | +0.091 |
flavor | +0.090 |
ornament | +0.088 |
peanut | +0.088 |
cube | +0.085 |
Neg logit contributions | |
ows | -0.064 |
pering | -0.060 |
obb | -0.058 |
arl | -0.057 |
iving | -0.056 |
Token | Every |
Token ID | 3887 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
penny | +0.065 |
ornament | +0.058 |
peanut | +0.058 |
seed | +0.055 |
jewel | +0.055 |
Neg logit contributions | |
ows | -0.047 |
arl | -0.042 |
adders | -0.040 |
obb | -0.039 |
iving | -0.039 |
Token | day |
Token ID | 1110 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
obb | +0.068 |
arl | +0.064 |
iving | +0.064 |
ows | +0.064 |
pering | +0.062 |
Neg logit contributions | |
penny | -0.045 |
ornament | -0.045 |
peanut | -0.043 |
seed | -0.043 |
dime | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
iving | +0.006 |
men | +0.006 |
oms | +0.006 |
obb | +0.006 |
adders | +0.006 |
Neg logit contributions | |
ornament | -0.019 |
c | -0.017 |
possession | -0.017 |
seed | -0.017 |
penny | -0.017 |
Token | my |
Token ID | 1820 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
penny | +0.044 |
dime | +0.039 |
flavor | +0.038 |
peanut | +0.037 |
ornament | +0.037 |
Neg logit contributions | |
ows | -0.082 |
obb | -0.075 |
pering | -0.074 |
arl | -0.072 |
realize | -0.071 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.26e-01 |
Pos logit contributions | |
penny | +0.083 |
dime | +0.079 |
flavor | +0.079 |
ornament | +0.078 |
peanut | +0.074 |
Neg logit contributions | |
ows | -0.065 |
obb | -0.055 |
pering | -0.054 |
realize | -0.054 |
sons | -0.053 |
Token | to |
Token ID | 284 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
c | +0.019 |
ruby | +0.019 |
penny | +0.019 |
flavor | +0.019 |
ornament | +0.019 |
Neg logit contributions | |
dolphins | -0.020 |
pering | -0.020 |
dancers | -0.019 |
men | -0.019 |
announcements | -0.019 |
Token | wear |
Token ID | 5806 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
penny | +0.047 |
flavor | +0.045 |
ornament | +0.045 |
peanut | +0.043 |
cube | +0.043 |
Neg logit contributions | |
ows | -0.060 |
pering | -0.058 |
obb | -0.058 |
arl | -0.057 |
oms | -0.055 |
Token | his |
Token ID | 465 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
flavor | +0.088 |
penny | +0.088 |
ornament | +0.087 |
cube | +0.084 |
peanut | +0.079 |
Neg logit contributions | |
ows | -0.115 |
arl | -0.111 |
obb | -0.111 |
pering | -0.111 |
iving | -0.105 |
Token | red |
Token ID | 2266 |
Feature activation | -7.69e+00 |
Pos logit contributions | |
ornament | +0.095 |
flavor | +0.093 |
checked | +0.092 |
pol | +0.088 |
dime | +0.088 |
Neg logit contributions | |
obb | -0.107 |
arl | -0.106 |
ows | -0.105 |
Super | -0.098 |
oms | -0.095 |
Token | cap |
Token ID | 1451 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
checked | +0.241 |
mattered | +0.234 |
pol | +0.232 |
wealth | +0.228 |
flavour | +0.226 |
Neg logit contributions | |
arl | -0.192 |
ows | -0.167 |
oms | -0.166 |
uits | -0.164 |
ropes | -0.156 |
Token | everywhere |
Token ID | 8347 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
flavor | +0.076 |
penny | +0.076 |
ornament | +0.074 |
peanut | +0.072 |
cube | +0.071 |
Neg logit contributions | |
ows | -0.074 |
obb | -0.066 |
pering | -0.066 |
arl | -0.065 |
iving | -0.063 |
Token | he |
Token ID | 339 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ows | +0.019 |
pering | +0.018 |
obb | +0.017 |
arl | +0.016 |
iving | +0.016 |
Neg logit contributions | |
penny | -0.025 |
flavor | -0.024 |
ornament | -0.024 |
peanut | -0.024 |
cube | -0.023 |
Token | went |
Token ID | 1816 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
ows | +0.041 |
obb | +0.041 |
pering | +0.040 |
arl | +0.039 |
iving | +0.038 |
Neg logit contributions | |
flavor | -0.045 |
ornament | -0.045 |
penny | -0.045 |
cube | -0.044 |
peanut | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ows | +0.119 |
pering | +0.118 |
obb | +0.114 |
arl | +0.112 |
iving | +0.109 |
Neg logit contributions | |
penny | -0.099 |
flavor | -0.099 |
ornament | -0.096 |
peanut | -0.094 |
cube | -0.094 |
Token | swim |
Token ID | 9422 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
flavor | +0.049 |
penny | +0.047 |
ornament | +0.047 |
cube | +0.045 |
peanut | +0.045 |
Neg logit contributions | |
ows | -0.074 |
pering | -0.073 |
obb | -0.070 |
arl | -0.070 |
iving | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
pering | +0.045 |
obb | +0.043 |
ows | +0.043 |
arl | +0.042 |
announcements | +0.042 |
Neg logit contributions | |
penny | -0.027 |
flavor | -0.026 |
cube | -0.026 |
peanut | -0.024 |
ornament | -0.024 |
Token | She |
Token ID | 1375 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
flavor | +0.062 |
penny | +0.061 |
ornament | +0.060 |
cube | +0.058 |
peanut | +0.057 |
Neg logit contributions | |
ows | -0.090 |
pering | -0.090 |
obb | -0.086 |
arl | -0.085 |
iving | -0.082 |
Token | had |
Token ID | 550 |
Feature activation | -4.11e+00 |
Pos logit contributions | |
flavor | +0.000 |
penny | +0.000 |
cube | +0.000 |
peanut | +0.000 |
color | +0.000 |
Neg logit contributions | |
pering | -0.002 |
arl | -0.002 |
ows | -0.002 |
obb | -0.002 |
adders | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
flavor | +0.102 |
peanut | +0.099 |
penny | +0.098 |
ornament | +0.089 |
cube | +0.089 |
Neg logit contributions | |
arl | -0.161 |
ows | -0.160 |
pering | -0.159 |
obb | -0.157 |
iving | -0.149 |
Token | red |
Token ID | 2266 |
Feature activation | -7.69e+00 |
Pos logit contributions | |
flavor | +0.047 |
ornament | +0.046 |
penny | +0.046 |
peanut | +0.044 |
cube | +0.043 |
Neg logit contributions | |
ows | -0.091 |
pering | -0.089 |
obb | -0.088 |
arl | -0.087 |
oms | -0.085 |
Token | swimming |
Token ID | 14899 |
Feature activation | +7.58e-01 |
Pos logit contributions | |
ornament | +0.206 |
pol | +0.204 |
checked | +0.203 |
penny | +0.200 |
flavor | +0.198 |
Neg logit contributions | |
ows | -0.219 |
arl | -0.215 |
oms | -0.214 |
uits | -0.205 |
pering | -0.205 |
Token | suit |
Token ID | 6050 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
pering | +0.029 |
ows | +0.028 |
announcements | +0.028 |
ighters | +0.027 |
obb | +0.027 |
Neg logit contributions | |
flavor | -0.019 |
penny | -0.019 |
cube | -0.019 |
peanut | -0.017 |
ornament | -0.017 |
Token | ready |
Token ID | 3492 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
flavor | +0.062 |
penny | +0.061 |
ornament | +0.058 |
cube | +0.058 |
color | +0.055 |
Neg logit contributions | |
pering | -0.101 |
arl | -0.097 |
obb | -0.096 |
ighters | -0.093 |
announcements | -0.092 |
Token | , |
Token ID | 11 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
pering | +0.023 |
obb | +0.022 |
arl | +0.022 |
ows | +0.022 |
ighters | +0.021 |
Neg logit contributions | |
flavor | -0.013 |
penny | -0.012 |
ornament | -0.012 |
color | -0.011 |
peanut | -0.011 |
Token | her |
Token ID | 607 |
Feature activation | -6.45e-01 |
Pos logit contributions | |
obb | +0.015 |
arl | +0.015 |
ows | +0.015 |
pering | +0.014 |
ighters | +0.014 |
Neg logit contributions | |
peanut | -0.010 |
penny | -0.010 |
flavor | -0.010 |
ornament | -0.010 |
color | -0.010 |
Token | racing |
Token ID | 11717 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
flavor | +0.057 |
penny | +0.056 |
cube | +0.055 |
ornament | +0.055 |
color | +0.053 |
Neg logit contributions | |
arl | -0.091 |
obb | -0.090 |
ows | -0.089 |
pering | -0.087 |
adders | -0.082 |
Token | each |
Token ID | 1123 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
pering | +0.042 |
obb | +0.041 |
arl | +0.041 |
ows | +0.041 |
oms | +0.038 |
Neg logit contributions | |
flavor | -0.027 |
penny | -0.027 |
ornament | -0.026 |
cube | -0.026 |
color | -0.024 |
Token | other |
Token ID | 584 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
penny | +0.074 |
flavor | +0.074 |
ornament | +0.072 |
cube | +0.070 |
peanut | +0.068 |
Neg logit contributions | |
ows | -0.088 |
pering | -0.085 |
arl | -0.082 |
obb | -0.082 |
oms | -0.079 |
Token | and |
Token ID | 290 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
penny | +0.025 |
ornament | +0.024 |
flavor | +0.024 |
cube | +0.023 |
color | +0.022 |
Neg logit contributions | |
arl | -0.038 |
obb | -0.037 |
pering | -0.036 |
oms | -0.034 |
announcements | -0.034 |
Token | making |
Token ID | 1642 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
penny | +0.006 |
flavor | +0.006 |
ornament | +0.006 |
peanut | +0.006 |
cube | +0.006 |
Neg logit contributions | |
ows | -0.006 |
pering | -0.006 |
obb | -0.006 |
arl | -0.006 |
iving | -0.006 |
Token | noises |
Token ID | 26782 |
Feature activation | +5.62e+00 |
Pos logit contributions | |
arl | +0.011 |
pering | +0.010 |
obb | +0.010 |
ows | +0.010 |
oms | +0.010 |
Neg logit contributions | |
flavor | -0.006 |
ornament | -0.006 |
penny | -0.006 |
toy | -0.006 |
peanut | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
arl | +0.273 |
obb | +0.257 |
pering | +0.244 |
ighters | +0.236 |
adders | +0.235 |
Neg logit contributions | |
color | -0.093 |
flavor | -0.091 |
ornament | -0.090 |
cube | -0.090 |
penny | -0.084 |
Token |
|
Token ID | 198 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
penny | +0.020 |
ornament | +0.020 |
flavor | +0.019 |
peanut | +0.019 |
cube | +0.018 |
Neg logit contributions | |
ows | -0.022 |
pering | -0.021 |
arl | -0.020 |
adders | -0.020 |
obb | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
penny | +0.032 |
flavor | +0.031 |
ornament | +0.031 |
peanut | +0.030 |
cube | +0.029 |
Neg logit contributions | |
ows | -0.041 |
pering | -0.041 |
arl | -0.039 |
obb | -0.039 |
oms | -0.038 |
Token | One |
Token ID | 3198 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
penny | +0.027 |
ornament | +0.027 |
flavor | +0.025 |
cube | +0.025 |
peanut | +0.025 |
Neg logit contributions | |
pering | -0.046 |
ows | -0.045 |
arl | -0.044 |
obb | -0.044 |
iving | -0.044 |
Token | day |
Token ID | 1110 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
pering | +0.014 |
ows | +0.014 |
obb | +0.014 |
arl | +0.013 |
ighters | +0.013 |
Neg logit contributions | |
flavor | -0.002 |
penny | -0.002 |
cube | -0.002 |
ornament | -0.002 |
peanut | -0.002 |
Token | I |
Token ID | 314 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
pering | +0.009 |
ows | +0.009 |
obb | +0.009 |
arl | +0.009 |
oms | +0.009 |
Neg logit contributions | |
flavor | -0.006 |
penny | -0.006 |
ornament | -0.006 |
cube | -0.006 |
peanut | -0.006 |
Token | love |
Token ID | 1842 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
flavor | +0.025 |
penny | +0.025 |
ornament | +0.024 |
peanut | +0.023 |
cube | +0.023 |
Neg logit contributions | |
pering | -0.029 |
ows | -0.028 |
arl | -0.027 |
obb | -0.026 |
iving | -0.026 |
Token | seeing |
Token ID | 4379 |
Feature activation | +9.32e-01 |
Pos logit contributions | |
penny | +0.029 |
flavor | +0.029 |
ornament | +0.028 |
peanut | +0.027 |
cube | +0.027 |
Neg logit contributions | |
ows | -0.039 |
pering | -0.039 |
obb | -0.038 |
arl | -0.037 |
iving | -0.036 |
Token | all |
Token ID | 477 |
Feature activation | -9.81e-01 |
Pos logit contributions | |
pering | +0.036 |
ows | +0.036 |
obb | +0.036 |
arl | +0.036 |
iving | +0.033 |
Neg logit contributions | |
ornament | -0.024 |
flavor | -0.024 |
penny | -0.023 |
cube | -0.022 |
peanut | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
flavor | +0.024 |
ornament | +0.024 |
color | +0.024 |
cube | +0.023 |
penny | +0.022 |
Neg logit contributions | |
obb | -0.039 |
ows | -0.039 |
pering | -0.039 |
arl | -0.038 |
ighters | -0.035 |
Token | animals |
Token ID | 4695 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
pering | +0.058 |
ows | +0.058 |
arl | +0.057 |
obb | +0.056 |
oms | +0.053 |
Neg logit contributions | |
flavor | -0.042 |
ornament | -0.040 |
penny | -0.040 |
cube | -0.039 |
peanut | -0.039 |
Token | !" |
Token ID | 2474 |
Feature activation | -8.18e-02 |
Pos logit contributions | |
obb | +0.204 |
arl | +0.200 |
pering | +0.189 |
ows | +0.188 |
announcements | +0.182 |
Neg logit contributions | |
ornament | -0.144 |
flavor | -0.144 |
penny | -0.141 |
cube | -0.139 |
color | -0.133 |
Token | |
Token ID | 220 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
penny | +0.002 |
flavor | +0.002 |
ornament | +0.002 |
cube | +0.002 |
peanut | +0.002 |
Neg logit contributions | |
ows | -0.003 |
pering | -0.003 |
arl | -0.003 |
obb | -0.003 |
oms | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
flavor | +0.013 |
ornament | +0.012 |
cube | +0.011 |
penny | +0.011 |
peanut | +0.011 |
Neg logit contributions | |
pering | -0.054 |
obb | -0.053 |
announcements | -0.053 |
aults | -0.051 |
ault | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | +5.06e-03 |
Pos logit contributions | |
ows | +0.009 |
pering | +0.009 |
arl | +0.008 |
obb | +0.008 |
iving | +0.008 |
Neg logit contributions | |
penny | -0.008 |
flavor | -0.008 |
ornament | -0.008 |
peanut | -0.007 |
cube | -0.007 |
Token | Her |
Token ID | 9360 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
pering | +0.000 |
ows | +0.000 |
arl | +0.000 |
obb | +0.000 |
iving | +0.000 |
Neg logit contributions | |
penny | -0.000 |
flavor | -0.000 |
peanut | -0.000 |
ornament | -0.000 |
cube | -0.000 |
Token | He |
Token ID | 679 |
Feature activation | +8.90e-01 |
Pos logit contributions | |
penny | +0.021 |
flavor | +0.021 |
ornament | +0.020 |
peanut | +0.020 |
cube | +0.019 |
Neg logit contributions | |
ows | -0.023 |
pering | -0.023 |
arl | -0.022 |
obb | -0.022 |
iving | -0.021 |
Token | was |
Token ID | 373 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ows | +0.035 |
pering | +0.035 |
obb | +0.034 |
arl | +0.033 |
oms | +0.032 |
Neg logit contributions | |
penny | -0.020 |
flavor | -0.020 |
ornament | -0.019 |
peanut | -0.018 |
cube | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
penny | +0.030 |
flavor | +0.030 |
ornament | +0.029 |
cube | +0.028 |
peanut | +0.028 |
Neg logit contributions | |
ows | -0.043 |
pering | -0.042 |
arl | -0.041 |
obb | -0.041 |
iving | -0.039 |
Token | best |
Token ID | 1266 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
ows | +0.021 |
pering | +0.020 |
arl | +0.019 |
obb | +0.019 |
oms | +0.019 |
Neg logit contributions | |
penny | -0.018 |
flavor | -0.017 |
ornament | -0.017 |
cube | -0.017 |
peanut | -0.016 |
Token | n |
Token ID | 299 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
penny | +0.044 |
flavor | +0.042 |
cube | +0.042 |
ornament | +0.042 |
peanut | +0.037 |
Neg logit contributions | |
ows | -0.068 |
pering | -0.067 |
arl | -0.064 |
obb | -0.063 |
announcements | -0.060 |
Token | oder |
Token ID | 12342 |
Feature activation | +5.05e+00 |
Pos logit contributions | |
ows | +0.008 |
pering | +0.008 |
obb | +0.007 |
announcements | +0.007 |
arl | +0.007 |
Neg logit contributions | |
penny | -0.010 |
flavor | -0.010 |
ornament | -0.010 |
cube | -0.010 |
peanut | -0.010 |
Token | in |
Token ID | 287 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
pering | +0.167 |
ows | +0.162 |
obb | +0.161 |
arl | +0.159 |
announcements | +0.154 |
Neg logit contributions | |
flavor | -0.160 |
cube | -0.153 |
penny | -0.152 |
ornament | -0.146 |
color | -0.142 |
Token | the |
Token ID | 262 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
penny | +0.021 |
flavor | +0.021 |
ornament | +0.020 |
cube | +0.020 |
peanut | +0.019 |
Neg logit contributions | |
ows | -0.032 |
pering | -0.028 |
obb | -0.028 |
oms | -0.027 |
arl | -0.027 |
Token | world |
Token ID | 995 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
ows | +0.027 |
pering | +0.027 |
obb | +0.025 |
arl | +0.025 |
iving | +0.024 |
Neg logit contributions | |
penny | -0.022 |
flavor | -0.022 |
ornament | -0.021 |
peanut | -0.021 |
cube | -0.021 |
Token | ! |
Token ID | 0 |
Feature activation | -7.05e-02 |
Pos logit contributions | |
arl | +0.117 |
pering | +0.116 |
obb | +0.115 |
ows | +0.111 |
ighters | +0.107 |
Neg logit contributions | |
flavor | -0.085 |
penny | -0.084 |
cube | -0.083 |
ornament | -0.083 |
peanut | -0.076 |
Token |
|
Token ID | 198 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
penny | +0.002 |
flavor | +0.002 |
ornament | +0.002 |
peanut | +0.002 |
cube | +0.002 |
Neg logit contributions | |
ows | -0.002 |
pering | -0.002 |
arl | -0.002 |
obb | -0.002 |
iving | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
penny | +0.068 |
flavor | +0.067 |
ornament | +0.066 |
peanut | +0.064 |
cube | +0.064 |
Neg logit contributions | |
pering | -0.069 |
ows | -0.069 |
obb | -0.068 |
arl | -0.067 |
iving | -0.065 |
Token | They |
Token ID | 1119 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
penny | +0.007 |
ornament | +0.007 |
flavor | +0.007 |
cube | +0.007 |
peanut | +0.007 |
Neg logit contributions | |
ows | -0.006 |
pering | -0.006 |
arl | -0.005 |
obb | -0.005 |
oms | -0.005 |
Token | pretend |
Token ID | 16614 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
pering | +0.074 |
arl | +0.070 |
iving | +0.067 |
obb | +0.066 |
ows | +0.065 |
Neg logit contributions | |
flavor | -0.053 |
ornament | -0.053 |
penny | -0.052 |
peanut | -0.052 |
nut | -0.051 |
Token | they |
Token ID | 484 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
pering | +0.004 |
arl | +0.004 |
ows | +0.003 |
obb | +0.003 |
iving | +0.003 |
Neg logit contributions | |
flavor | -0.002 |
penny | -0.002 |
ornament | -0.002 |
peanut | -0.002 |
cube | -0.002 |
Token | are |
Token ID | 389 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
pering | +0.104 |
arl | +0.102 |
obb | +0.095 |
iving | +0.093 |
ows | +0.092 |
Neg logit contributions | |
flavor | -0.100 |
ornament | -0.097 |
penny | -0.096 |
color | -0.092 |
peanut | -0.091 |
Token | animals |
Token ID | 4695 |
Feature activation | +5.43e+00 |
Pos logit contributions | |
color | +0.008 |
ornament | +0.008 |
flavor | +0.008 |
peanut | +0.008 |
penny | +0.007 |
Neg logit contributions | |
obb | -0.016 |
arl | -0.016 |
ows | -0.015 |
rig | -0.015 |
oms | -0.015 |
Token | or |
Token ID | 393 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
obb | +0.216 |
arl | +0.208 |
pering | +0.200 |
ows | +0.190 |
announcements | +0.187 |
Neg logit contributions | |
ornament | -0.132 |
penny | -0.132 |
flavor | -0.130 |
peanut | -0.128 |
nut | -0.124 |
Token | pirates |
Token ID | 27516 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
flavor | +0.006 |
peanut | +0.005 |
penny | +0.005 |
ornament | +0.005 |
cube | +0.005 |
Neg logit contributions | |
ows | -0.005 |
pering | -0.005 |
obb | -0.005 |
arl | -0.005 |
iving | -0.005 |
Token | or |
Token ID | 393 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
obb | +0.094 |
pering | +0.089 |
arl | +0.089 |
rig | +0.084 |
oms | +0.083 |
Neg logit contributions | |
flavor | -0.051 |
penny | -0.051 |
ornament | -0.049 |
peanut | -0.049 |
cube | -0.045 |
Token | fair |
Token ID | 3148 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
flavor | +0.016 |
penny | +0.015 |
peanut | +0.015 |
ornament | +0.015 |
cube | +0.015 |
Neg logit contributions | |
ows | -0.015 |
pering | -0.015 |
obb | -0.014 |
arl | -0.014 |
iving | -0.014 |
Token | ies |
Token ID | 444 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
flavor | +0.054 |
color | +0.053 |
penny | +0.051 |
ornament | +0.049 |
cube | +0.048 |
Neg logit contributions | |
obb | -0.123 |
pering | -0.114 |
arl | -0.110 |
uits | -0.109 |
Run | -0.109 |
Token | who |
Token ID | 508 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
flavor | +0.007 |
ornament | +0.007 |
penny | +0.006 |
peanut | +0.006 |
cube | +0.006 |
Neg logit contributions | |
pering | -0.009 |
obb | -0.008 |
arl | -0.008 |
ows | -0.008 |
iving | -0.008 |
Token | can |
Token ID | 460 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
penny | +0.038 |
flavor | +0.038 |
cube | +0.037 |
ornament | +0.037 |
c | +0.034 |
Neg logit contributions | |
pering | -0.060 |
obb | -0.057 |
arl | -0.056 |
ows | -0.054 |
iving | -0.051 |
Token | spot |
Token ID | 4136 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
flavor | +0.028 |
penny | +0.027 |
ornament | +0.027 |
cube | +0.027 |
magnet | +0.025 |
Neg logit contributions | |
pering | -0.036 |
ows | -0.036 |
arl | -0.035 |
obb | -0.035 |
iving | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
flavor | +0.022 |
penny | +0.021 |
ornament | +0.021 |
peanut | +0.021 |
cube | +0.020 |
Neg logit contributions | |
pering | -0.031 |
obb | -0.031 |
arl | -0.030 |
iving | -0.030 |
ows | -0.030 |
Token | most |
Token ID | 749 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
ows | +0.081 |
pering | +0.078 |
arl | +0.075 |
obb | +0.073 |
iving | +0.073 |
Neg logit contributions | |
penny | -0.061 |
flavor | -0.059 |
ornament | -0.059 |
cube | -0.056 |
peanut | -0.055 |
Token | animals |
Token ID | 4695 |
Feature activation | +5.46e+00 |
Pos logit contributions | |
flavor | +0.005 |
ornament | +0.005 |
penny | +0.005 |
cube | +0.004 |
peanut | +0.004 |
Neg logit contributions | |
arl | -0.007 |
pering | -0.007 |
ows | -0.007 |
obb | -0.006 |
attering | -0.006 |
Token | on |
Token ID | 319 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
pering | +0.180 |
obb | +0.178 |
arl | +0.173 |
ows | +0.172 |
iving | +0.162 |
Neg logit contributions | |
flavor | -0.161 |
penny | -0.160 |
ornament | -0.158 |
cube | -0.150 |
peanut | -0.150 |
Token | the |
Token ID | 262 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
ornament | +0.066 |
penny | +0.066 |
flavor | +0.066 |
color | +0.059 |
peanut | +0.057 |
Neg logit contributions | |
ows | -0.111 |
pering | -0.106 |
arl | -0.104 |
obb | -0.103 |
iving | -0.101 |
Token | way |
Token ID | 835 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ows | +0.021 |
arl | +0.021 |
oms | +0.021 |
pering | +0.020 |
iving | +0.020 |
Neg logit contributions | |
ornament | -0.015 |
penny | -0.015 |
flavor | -0.015 |
cube | -0.014 |
peanut | -0.014 |
Token | ," |
Token ID | 553 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
ornament | +0.026 |
penny | +0.025 |
flavor | +0.025 |
cube | +0.025 |
color | +0.023 |
Neg logit contributions | |
obb | -0.046 |
pering | -0.042 |
arl | -0.041 |
announcements | -0.040 |
rig | -0.040 |
Token | dad |
Token ID | 9955 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
ornament | +0.005 |
penny | +0.005 |
cube | +0.005 |
flavor | +0.005 |
orn | +0.005 |
Neg logit contributions | |
announcements | -0.003 |
ows | -0.003 |
ila | -0.003 |
men | -0.002 |
obb | -0.002 |
Token | Her |
Token ID | 2332 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
pering | +0.001 |
obb | +0.001 |
ows | +0.001 |
announcements | +0.001 |
aults | +0.001 |
Neg logit contributions | |
penny | -0.001 |
peanut | -0.000 |
flavor | -0.000 |
mint | -0.000 |
cube | -0.000 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
flavor | +0.052 |
ornament | +0.050 |
penny | +0.050 |
peanut | +0.049 |
cube | +0.048 |
Neg logit contributions | |
ows | -0.054 |
obb | -0.050 |
pering | -0.050 |
arl | -0.049 |
oms | -0.048 |
Token | my |
Token ID | 1820 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
penny | +0.024 |
flavor | +0.023 |
ornament | +0.022 |
peanut | +0.021 |
nut | +0.020 |
Neg logit contributions | |
ows | -0.064 |
pering | -0.060 |
oms | -0.058 |
arl | -0.058 |
iving | -0.057 |
Token | left |
Token ID | 1364 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
flavor | +0.031 |
penny | +0.031 |
ornament | +0.030 |
cube | +0.029 |
peanut | +0.029 |
Neg logit contributions | |
pering | -0.055 |
ows | -0.054 |
obb | -0.053 |
arl | -0.053 |
iving | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
penny | +0.011 |
flavor | +0.010 |
ornament | +0.010 |
peanut | +0.010 |
cube | +0.009 |
Neg logit contributions | |
ows | -0.015 |
pering | -0.014 |
obb | -0.014 |
iving | -0.014 |
arl | -0.014 |
Token | little |
Token ID | 1310 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
pering | +0.072 |
ows | +0.071 |
obb | +0.070 |
arl | +0.068 |
iving | +0.066 |
Neg logit contributions | |
flavor | -0.052 |
penny | -0.049 |
peanut | -0.048 |
ornament | -0.047 |
cube | -0.046 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ows | +0.083 |
pering | +0.081 |
arl | +0.080 |
obb | +0.079 |
adders | +0.075 |
Neg logit contributions | |
penny | -0.064 |
flavor | -0.060 |
ornament | -0.060 |
cube | -0.059 |
peanut | -0.058 |
Token | alone |
Token ID | 3436 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
arl | +0.040 |
pering | +0.040 |
ows | +0.040 |
obb | +0.040 |
iving | +0.038 |
Neg logit contributions | |
penny | -0.027 |
flavor | -0.027 |
ornament | -0.026 |
cube | -0.026 |
peanut | -0.026 |
Token | with |
Token ID | 351 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
arl | +0.094 |
obb | +0.093 |
ighters | +0.091 |
ows | +0.090 |
announcements | +0.090 |
Neg logit contributions | |
penny | -0.048 |
flavor | -0.043 |
cube | -0.043 |
ornament | -0.041 |
color | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
penny | +0.058 |
flavor | +0.058 |
ornament | +0.056 |
cube | +0.055 |
peanut | +0.054 |
Neg logit contributions | |
ows | -0.061 |
pering | -0.060 |
obb | -0.057 |
arl | -0.057 |
oms | -0.055 |
Token | box |
Token ID | 3091 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
pering | +0.014 |
ows | +0.014 |
arl | +0.014 |
obb | +0.014 |
iving | +0.013 |
Neg logit contributions | |
penny | -0.009 |
flavor | -0.008 |
ornament | -0.008 |
cube | -0.008 |
peanut | -0.008 |
Token | tree |
Token ID | 5509 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
pering | +0.056 |
ows | +0.056 |
obb | +0.054 |
arl | +0.053 |
iving | +0.052 |
Neg logit contributions | |
flavor | -0.035 |
penny | -0.035 |
ornament | -0.034 |
cube | -0.033 |
peanut | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
penny | +0.034 |
flavor | +0.033 |
ornament | +0.033 |
cube | +0.033 |
peanut | +0.032 |
Neg logit contributions | |
obb | -0.040 |
arl | -0.039 |
pering | -0.039 |
ows | -0.038 |
announcements | -0.037 |
Token | The |
Token ID | 383 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ows | +0.021 |
pering | +0.021 |
aults | +0.021 |
arl | +0.021 |
announcements | +0.021 |
Neg logit contributions | |
peanut | -0.009 |
penny | -0.008 |
flavor | -0.008 |
cube | -0.008 |
nut | -0.008 |
Token | tree |
Token ID | 5509 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
pering | +0.067 |
ows | +0.067 |
obb | +0.066 |
arl | +0.065 |
iving | +0.064 |
Neg logit contributions | |
penny | -0.023 |
flavor | -0.023 |
ornament | -0.022 |
cube | -0.021 |
peanut | -0.021 |
Token | yelled |
Token ID | 22760 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
flavor | +0.042 |
penny | +0.042 |
ornament | +0.040 |
cube | +0.039 |
peanut | +0.039 |
Neg logit contributions | |
ows | -0.060 |
pering | -0.058 |
obb | -0.056 |
oms | -0.054 |
iving | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
arl | +0.089 |
announcements | +0.089 |
pering | +0.089 |
ows | +0.088 |
obb | +0.088 |
Neg logit contributions | |
penny | -0.053 |
peanut | -0.052 |
cube | -0.050 |
flavor | -0.049 |
ornament | -0.048 |
Token | " |
Token ID | 366 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
ows | +0.064 |
obb | +0.062 |
iving | +0.061 |
pering | +0.061 |
arl | +0.061 |
Neg logit contributions | |
penny | -0.033 |
flavor | -0.030 |
peanut | -0.030 |
cube | -0.030 |
ornament | -0.030 |
Token | No |
Token ID | 2949 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
penny | +0.006 |
flavor | +0.006 |
ornament | +0.006 |
cube | +0.006 |
peanut | +0.006 |
Neg logit contributions | |
pering | -0.006 |
ows | -0.006 |
obb | -0.005 |
arl | -0.005 |
iving | -0.005 |
Token | ! |
Token ID | 0 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
flavor | +0.016 |
penny | +0.015 |
peanut | +0.014 |
ornament | +0.014 |
cube | +0.014 |
Neg logit contributions | |
ows | -0.020 |
pering | -0.020 |
arl | -0.019 |
obb | -0.019 |
ighters | -0.018 |
Token | Stop |
Token ID | 13707 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
ows | +0.036 |
pering | +0.036 |
obb | +0.035 |
announcements | +0.034 |
arl | +0.033 |
Neg logit contributions | |
penny | -0.025 |
flavor | -0.025 |
ornament | -0.024 |
peanut | -0.024 |
cube | -0.023 |
Token | ! |
Token ID | 0 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
ows | +0.068 |
arl | +0.064 |
obb | +0.063 |
oms | +0.062 |
adders | +0.062 |
Neg logit contributions | |
penny | -0.047 |
peanut | -0.045 |
flavor | -0.045 |
ornament | -0.044 |
cube | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
pering | +0.027 |
obb | +0.027 |
ows | +0.027 |
arl | +0.027 |
announcements | +0.025 |
Neg logit contributions | |
flavor | -0.012 |
penny | -0.011 |
ornament | -0.011 |
cube | -0.010 |
peanut | -0.010 |
Token | draw |
Token ID | 3197 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
ows | +0.042 |
pering | +0.042 |
obb | +0.040 |
arl | +0.040 |
iving | +0.038 |
Neg logit contributions | |
penny | -0.033 |
flavor | -0.033 |
ornament | -0.032 |
peanut | -0.031 |
cube | -0.031 |
Token | on |
Token ID | 319 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
flavor | +0.002 |
ornament | +0.002 |
penny | +0.002 |
peanut | +0.002 |
cube | +0.002 |
Neg logit contributions | |
pering | -0.002 |
arl | -0.002 |
ows | -0.002 |
obb | -0.002 |
iving | -0.002 |
Token | it |
Token ID | 340 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
flavor | +0.065 |
penny | +0.064 |
ornament | +0.063 |
peanut | +0.061 |
cube | +0.060 |
Neg logit contributions | |
ows | -0.085 |
pering | -0.084 |
arl | -0.081 |
obb | -0.080 |
iving | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
flavor | +0.085 |
ornament | +0.084 |
penny | +0.083 |
cube | +0.079 |
peanut | +0.079 |
Neg logit contributions | |
pering | -0.091 |
arl | -0.088 |
ows | -0.088 |
obb | -0.087 |
announcements | -0.082 |
Token | She |
Token ID | 1375 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
flavor | +0.003 |
penny | +0.003 |
ornament | +0.003 |
peanut | +0.003 |
cube | +0.003 |
Neg logit contributions | |
pering | -0.005 |
ows | -0.005 |
obb | -0.005 |
arl | -0.005 |
iving | -0.005 |
Token | took |
Token ID | 1718 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
ows | +0.040 |
pering | +0.038 |
obb | +0.038 |
oms | +0.037 |
arl | +0.036 |
Neg logit contributions | |
penny | -0.028 |
flavor | -0.028 |
ornament | -0.027 |
cube | -0.026 |
peanut | -0.026 |
Token | her |
Token ID | 607 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
ows | +0.015 |
pering | +0.015 |
obb | +0.014 |
arl | +0.014 |
oms | +0.013 |
Neg logit contributions | |
penny | -0.014 |
flavor | -0.014 |
ornament | -0.014 |
cube | -0.013 |
peanut | -0.013 |
Token | cr |
Token ID | 1067 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
flavor | +0.020 |
penny | +0.019 |
ornament | +0.019 |
peanut | +0.018 |
cube | +0.018 |
Neg logit contributions | |
ows | -0.045 |
pering | -0.044 |
obb | -0.043 |
oms | -0.043 |
arl | -0.042 |
Token | ay |
Token ID | 323 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
penny | +0.121 |
flavor | +0.117 |
ornament | +0.113 |
peanut | +0.112 |
cube | +0.111 |
Neg logit contributions | |
ows | -0.110 |
arl | -0.102 |
pering | -0.098 |
iving | -0.097 |
obb | -0.096 |
Token | ons |
Token ID | 684 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
penny | +0.075 |
flavor | +0.071 |
ornament | +0.070 |
nut | +0.066 |
peanut | +0.064 |
Neg logit contributions | |
ows | -0.120 |
oms | -0.115 |
obb | -0.111 |
arl | -0.110 |
adders | -0.109 |
Token |
|
Token ID | 198 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
ows | +0.003 |
obb | +0.003 |
pering | +0.003 |
arl | +0.003 |
iving | +0.003 |
Neg logit contributions | |
flavor | -0.002 |
penny | -0.002 |
peanut | -0.002 |
cube | -0.001 |
ornament | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
flavor | +0.009 |
penny | +0.008 |
ornament | +0.008 |
cube | +0.008 |
peanut | +0.008 |
Neg logit contributions | |
ows | -0.020 |
pering | -0.020 |
obb | -0.020 |
iving | -0.019 |
arl | -0.019 |
Token | The |
Token ID | 464 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
penny | +0.013 |
flavor | +0.013 |
ornament | +0.012 |
cube | +0.012 |
peanut | +0.012 |
Neg logit contributions | |
ows | -0.019 |
pering | -0.017 |
obb | -0.017 |
oms | -0.017 |
arl | -0.017 |
Token | goat |
Token ID | 26917 |
Feature activation | +9.65e-01 |
Pos logit contributions | |
ows | +0.045 |
pering | +0.040 |
obb | +0.040 |
arl | +0.040 |
iving | +0.039 |
Neg logit contributions | |
penny | -0.030 |
peanut | -0.027 |
ornament | -0.027 |
flavor | -0.026 |
cube | -0.026 |
Token | felt |
Token ID | 2936 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
arl | +0.043 |
pering | +0.042 |
ows | +0.041 |
obb | +0.040 |
iving | +0.040 |
Neg logit contributions | |
penny | -0.018 |
flavor | -0.018 |
peanut | -0.017 |
ornament | -0.017 |
cube | -0.017 |
Token | sad |
Token ID | 6507 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
ows | +0.079 |
arl | +0.076 |
iving | +0.074 |
pering | +0.074 |
oms | +0.072 |
Neg logit contributions | |
penny | -0.059 |
peanut | -0.057 |
flavor | -0.055 |
nut | -0.054 |
mint | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
pering | +0.037 |
obb | +0.037 |
announcements | +0.037 |
adders | +0.036 |
adies | +0.036 |
Neg logit contributions | |
penny | -0.022 |
flavor | -0.021 |
cube | -0.021 |
ornament | -0.020 |
peanut | -0.020 |
Token | didn |
Token ID | 1422 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
ows | +0.050 |
pering | +0.047 |
arl | +0.047 |
iving | +0.046 |
obb | +0.045 |
Neg logit contributions | |
penny | -0.045 |
peanut | -0.044 |
cube | -0.042 |
flavor | -0.042 |
ornament | -0.042 |
Token | 't |
Token ID | 470 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ows | +0.190 |
pering | +0.179 |
announcements | +0.175 |
iving | +0.175 |
arl | +0.174 |
Neg logit contributions | |
penny | -0.087 |
flavor | -0.080 |
ornament | -0.080 |
cube | -0.075 |
peanut | -0.073 |
Token | know |
Token ID | 760 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
ows | +0.026 |
pering | +0.026 |
obb | +0.024 |
arl | +0.024 |
oms | +0.023 |
Neg logit contributions | |
flavor | -0.038 |
penny | -0.038 |
ornament | -0.037 |
peanut | -0.037 |
cube | -0.036 |
Token | what |
Token ID | 644 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
flavor | +0.005 |
ornament | +0.005 |
penny | +0.005 |
cube | +0.005 |
peanut | +0.005 |
Neg logit contributions | |
arl | -0.006 |
pering | -0.006 |
obb | -0.006 |
ows | -0.006 |
iving | -0.005 |
Token | hair |
Token ID | 4190 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
penny | +0.044 |
flavor | +0.044 |
ornament | +0.043 |
peanut | +0.042 |
cube | +0.041 |
Neg logit contributions | |
ows | -0.054 |
pering | -0.054 |
obb | -0.051 |
arl | -0.051 |
iving | -0.050 |
Token | again |
Token ID | 757 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
penny | +0.075 |
flavor | +0.075 |
ornament | +0.073 |
peanut | +0.070 |
cube | +0.070 |
Neg logit contributions | |
pering | -0.076 |
ows | -0.075 |
arl | -0.072 |
obb | -0.071 |
iving | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
pering | +0.018 |
ows | +0.018 |
obb | +0.017 |
arl | +0.016 |
iving | +0.016 |
Neg logit contributions | |
penny | -0.022 |
flavor | -0.021 |
ornament | -0.021 |
cube | -0.021 |
peanut | -0.020 |
Token | always |
Token ID | 1464 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
pering | +0.036 |
ows | +0.036 |
obb | +0.036 |
iving | +0.034 |
oms | +0.033 |
Neg logit contributions | |
flavor | -0.029 |
penny | -0.029 |
ornament | -0.029 |
cube | -0.027 |
peanut | -0.027 |
Token | felt |
Token ID | 2936 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
penny | +0.016 |
ornament | +0.015 |
peanut | +0.015 |
flavor | +0.015 |
possession | +0.014 |
Neg logit contributions | |
ows | -0.016 |
realize | -0.016 |
pering | -0.015 |
obb | -0.015 |
oms | -0.014 |
Token | guilty |
Token ID | 6717 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
ows | +0.082 |
pering | +0.081 |
arl | +0.077 |
obb | +0.076 |
iving | +0.075 |
Neg logit contributions | |
penny | -0.077 |
flavor | -0.076 |
ornament | -0.074 |
peanut | -0.073 |
cube | -0.072 |
Token | for |
Token ID | 329 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ows | +0.090 |
pering | +0.089 |
arl | +0.088 |
obb | +0.087 |
announcements | +0.086 |
Neg logit contributions | |
penny | -0.058 |
ornament | -0.055 |
flavor | -0.055 |
cube | -0.054 |
peanut | -0.050 |
Token | what |
Token ID | 644 |
Feature activation | +6.14e-01 |
Pos logit contributions | |
ows | +0.054 |
pering | +0.054 |
obb | +0.052 |
arl | +0.051 |
iving | +0.050 |
Neg logit contributions | |
penny | -0.044 |
flavor | -0.044 |
ornament | -0.043 |
peanut | -0.041 |
cube | -0.041 |
Token | he |
Token ID | 339 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
ows | +0.014 |
pering | +0.014 |
obb | +0.014 |
men | +0.014 |
announcements | +0.014 |
Neg logit contributions | |
ornament | -0.020 |
penny | -0.019 |
flavor | -0.019 |
peanut | -0.019 |
c | -0.018 |
Token | had |
Token ID | 550 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
ows | +0.065 |
obb | +0.064 |
pering | +0.063 |
arl | +0.059 |
realize | +0.058 |
Neg logit contributions | |
ornament | -0.069 |
penny | -0.069 |
flavor | -0.065 |
cube | -0.065 |
peanut | -0.064 |
Token | done |
Token ID | 1760 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
ornament | +0.087 |
penny | +0.086 |
flavor | +0.084 |
peanut | +0.083 |
cube | +0.082 |
Neg logit contributions | |
ows | -0.085 |
pering | -0.082 |
obb | -0.081 |
iving | -0.081 |
arl | -0.079 |
Token | made |
Token ID | 925 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
flavor | +0.004 |
penny | +0.004 |
ornament | +0.004 |
peanut | +0.004 |
cube | +0.004 |
Neg logit contributions | |
pering | -0.007 |
ows | -0.007 |
arl | -0.007 |
obb | -0.007 |
announcements | -0.007 |
Token | her |
Token ID | 607 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
flavor | +0.025 |
ornament | +0.024 |
penny | +0.023 |
peanut | +0.022 |
color | +0.022 |
Neg logit contributions | |
pering | -0.035 |
obb | -0.034 |
arl | -0.034 |
ows | -0.033 |
announcements | -0.032 |
Token | so |
Token ID | 523 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
penny | +0.014 |
flavor | +0.014 |
ornament | +0.013 |
peanut | +0.013 |
cube | +0.013 |
Neg logit contributions | |
ows | -0.018 |
pering | -0.018 |
obb | -0.017 |
arl | -0.017 |
iving | -0.017 |
Token | excited |
Token ID | 6568 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
penny | +0.044 |
flavor | +0.043 |
ornament | +0.042 |
cube | +0.042 |
peanut | +0.042 |
Neg logit contributions | |
pering | -0.066 |
ows | -0.065 |
arl | -0.061 |
chers | -0.060 |
iving | -0.059 |
Token | ! |
Token ID | 0 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
pering | +0.001 |
ows | +0.001 |
obb | +0.001 |
arl | +0.001 |
adders | +0.001 |
Neg logit contributions | |
flavor | -0.001 |
penny | -0.001 |
ornament | -0.001 |
peanut | -0.001 |
cube | -0.001 |
Token | There |
Token ID | 1318 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
penny | +0.014 |
flavor | +0.014 |
peanut | +0.013 |
cube | +0.013 |
ornament | +0.013 |
Neg logit contributions | |
ows | -0.011 |
pering | -0.011 |
arl | -0.010 |
obb | -0.010 |
adders | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
ows | +0.001 |
pering | +0.001 |
oms | +0.001 |
obb | +0.001 |
adders | +0.001 |
Neg logit contributions | |
penny | -0.001 |
flavor | -0.001 |
peanut | -0.001 |
ornament | -0.001 |
cube | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
flavor | +0.005 |
ornament | +0.005 |
penny | +0.005 |
peanut | +0.005 |
cube | +0.004 |
Neg logit contributions | |
pering | -0.007 |
arl | -0.007 |
ows | -0.007 |
obb | -0.007 |
iving | -0.007 |
Token | little |
Token ID | 1310 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
pering | +0.051 |
ows | +0.049 |
arl | +0.048 |
obb | +0.047 |
announcements | +0.046 |
Neg logit contributions | |
flavor | -0.071 |
penny | -0.069 |
ornament | -0.069 |
cube | -0.066 |
peanut | -0.066 |
Token | puppy |
Token ID | 26188 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
pering | +0.049 |
ows | +0.048 |
oms | +0.046 |
iving | +0.045 |
arl | +0.045 |
Neg logit contributions | |
flavor | -0.041 |
penny | -0.040 |
ornament | -0.038 |
peanut | -0.038 |
cube | -0.037 |
Token | outside |
Token ID | 2354 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
obb | +0.002 |
ows | +0.002 |
arl | +0.002 |
pering | +0.002 |
adders | +0.002 |
Neg logit contributions | |
penny | -0.001 |
ornament | -0.001 |
peanut | -0.001 |
flavor | -0.001 |
cube | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
flavor | +0.021 |
penny | +0.021 |
ornament | +0.020 |
peanut | +0.020 |
cube | +0.020 |
Neg logit contributions | |
pering | -0.040 |
ows | -0.040 |
arl | -0.039 |
obb | -0.038 |
oms | -0.037 |
Token | her |
Token ID | 607 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
cube | +0.075 |
penny | +0.074 |
flavor | +0.073 |
c | +0.070 |
dime | +0.070 |
Neg logit contributions | |
realize | -0.047 |
announcements | -0.046 |
Super | -0.046 |
arl | -0.046 |
obb | -0.046 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
flavor | +0.062 |
penny | +0.062 |
ornament | +0.060 |
peanut | +0.059 |
cube | +0.058 |
Neg logit contributions | |
ows | -0.076 |
pering | -0.074 |
arl | -0.072 |
obb | -0.072 |
iving | -0.069 |
Token | my |
Token ID | 1820 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
penny | +0.048 |
ornament | +0.045 |
possession | +0.044 |
ruby | +0.044 |
flavor | +0.043 |
Neg logit contributions | |
ows | -0.062 |
iving | -0.061 |
arl | -0.059 |
pering | -0.059 |
realize | -0.058 |
Token | went |
Token ID | 1816 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
penny | +0.051 |
flavor | +0.050 |
ornament | +0.049 |
cube | +0.048 |
peanut | +0.046 |
Neg logit contributions | |
ows | -0.063 |
pering | -0.061 |
obb | -0.059 |
arl | -0.058 |
iving | -0.058 |
Token | outside |
Token ID | 2354 |
Feature activation | -6.06e-02 |
Pos logit contributions | |
pering | +0.093 |
ows | +0.093 |
arl | +0.089 |
oms | +0.087 |
iving | +0.086 |
Neg logit contributions | |
flavor | -0.059 |
penny | -0.057 |
ornament | -0.056 |
peanut | -0.053 |
cube | -0.053 |
Token | to |
Token ID | 284 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
penny | +0.001 |
flavor | +0.001 |
ornament | +0.001 |
peanut | +0.001 |
cube | +0.001 |
Neg logit contributions | |
ows | -0.003 |
pering | -0.003 |
arl | -0.003 |
obb | -0.003 |
iving | -0.002 |
Token | play |
Token ID | 711 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
penny | +0.014 |
flavor | +0.013 |
ornament | +0.013 |
dime | +0.013 |
cube | +0.012 |
Neg logit contributions | |
pering | -0.011 |
realize | -0.011 |
oms | -0.011 |
ows | -0.011 |
arl | -0.011 |
Token | on |
Token ID | 319 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
ows | +0.038 |
pering | +0.037 |
obb | +0.036 |
arl | +0.035 |
iving | +0.034 |
Neg logit contributions | |
penny | -0.049 |
flavor | -0.048 |
ornament | -0.047 |
peanut | -0.046 |
cube | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
penny | +0.083 |
flavor | +0.083 |
ornament | +0.081 |
cube | +0.080 |
peanut | +0.079 |
Neg logit contributions | |
pering | -0.056 |
ows | -0.055 |
obb | -0.052 |
oms | -0.050 |
arl | -0.050 |
Token | lawn |
Token ID | 21133 |
Feature activation | -9.93e-01 |
Pos logit contributions | |
ows | +0.017 |
pering | +0.017 |
arl | +0.016 |
obb | +0.016 |
oms | +0.016 |
Neg logit contributions | |
penny | -0.017 |
flavor | -0.017 |
ornament | -0.017 |
cube | -0.017 |
peanut | -0.017 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
ows | +0.007 |
pering | +0.006 |
obb | +0.006 |
arl | +0.006 |
iving | +0.006 |
Neg logit contributions | |
flavor | -0.005 |
penny | -0.005 |
ornament | -0.005 |
cube | -0.004 |
peanut | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.93e-03 |
Pos logit contributions | |
penny | +0.004 |
flavor | +0.004 |
ornament | +0.003 |
peanut | +0.003 |
cube | +0.003 |
Neg logit contributions | |
ows | -0.004 |
pering | -0.004 |
obb | -0.003 |
arl | -0.003 |
adders | -0.003 |
Token | flew |
Token ID | 13112 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
flavor | +0.000 |
ornament | +0.000 |
penny | +0.000 |
cube | +0.000 |
peanut | +0.000 |
Neg logit contributions | |
ows | -0.000 |
pering | -0.000 |
obb | -0.000 |
arl | -0.000 |
iving | -0.000 |
Token | away |
Token ID | 1497 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
ows | +0.139 |
pering | +0.137 |
obb | +0.133 |
arl | +0.131 |
oms | +0.129 |
Neg logit contributions | |
penny | -0.090 |
flavor | -0.087 |
ornament | -0.086 |
peanut | -0.083 |
cube | -0.083 |
Token | , |
Token ID | 11 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
penny | +0.014 |
flavor | +0.014 |
ornament | +0.014 |
cube | +0.013 |
peanut | +0.013 |
Neg logit contributions | |
ows | -0.018 |
pering | -0.018 |
arl | -0.017 |
obb | -0.017 |
iving | -0.017 |
Token | happy |
Token ID | 3772 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
pering | +0.041 |
ows | +0.040 |
obb | +0.037 |
arl | +0.037 |
iving | +0.036 |
Neg logit contributions | |
flavor | -0.046 |
penny | -0.045 |
ornament | -0.044 |
cube | -0.043 |
peanut | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
flavor | +0.013 |
penny | +0.013 |
ornament | +0.013 |
cube | +0.012 |
peanut | +0.012 |
Neg logit contributions | |
pering | -0.016 |
ows | -0.016 |
arl | -0.015 |
obb | -0.015 |
iving | -0.015 |
Token | help |
Token ID | 1037 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
pering | +0.039 |
arl | +0.037 |
realize | +0.036 |
ows | +0.034 |
obb | +0.034 |
Neg logit contributions | |
flavor | -0.047 |
penny | -0.047 |
ornament | -0.046 |
cube | -0.044 |
peanut | -0.043 |
Token | his |
Token ID | 465 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
pering | +0.049 |
ows | +0.048 |
arl | +0.044 |
iving | +0.044 |
obb | +0.043 |
Neg logit contributions | |
penny | -0.046 |
flavor | -0.045 |
cube | -0.044 |
ornament | -0.044 |
peanut | -0.042 |
Token | friend |
Token ID | 1545 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
pering | +0.001 |
ows | +0.001 |
arl | +0.001 |
sons | +0.001 |
iving | +0.001 |
Neg logit contributions | |
flavor | -0.001 |
penny | -0.001 |
peanut | -0.001 |
cube | -0.001 |
ornament | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
obb | +0.048 |
arl | +0.047 |
pering | +0.044 |
oms | +0.044 |
announcements | +0.043 |
Neg logit contributions | |
ornament | -0.030 |
flavor | -0.029 |
penny | -0.029 |
cube | -0.029 |
peanut | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
penny | +0.142 |
flavor | +0.137 |
ornament | +0.137 |
cube | +0.136 |
peanut | +0.132 |
Neg logit contributions | |
pering | -0.153 |
obb | -0.149 |
ows | -0.147 |
arl | -0.144 |
announcements | -0.143 |
Token | says |
Token ID | 1139 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
flavor | +0.025 |
penny | +0.024 |
peanut | +0.024 |
ornament | +0.023 |
pencil | +0.023 |
Neg logit contributions | |
pering | -0.033 |
arl | -0.032 |
ows | -0.031 |
obb | -0.030 |
iving | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
penny | +0.020 |
flavor | +0.020 |
ornament | +0.020 |
cube | +0.019 |
peanut | +0.019 |
Neg logit contributions | |
ows | -0.018 |
oms | -0.017 |
pering | -0.017 |
adders | -0.017 |
obb | -0.017 |
Token | " |
Token ID | 366 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ows | +0.006 |
pering | +0.006 |
obb | +0.005 |
arl | +0.005 |
iving | +0.005 |
Neg logit contributions | |
penny | -0.005 |
flavor | -0.005 |
ornament | -0.005 |
peanut | -0.005 |
cube | -0.005 |
Token | This |
Token ID | 1212 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
ornament | +0.031 |
penny | +0.031 |
cube | +0.031 |
nut | +0.030 |
magnet | +0.029 |
Neg logit contributions | |
aults | -0.026 |
Run | -0.025 |
iving | -0.025 |
stations | -0.025 |
pering | -0.024 |
Token | can |
Token ID | 460 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
flavor | +0.059 |
penny | +0.059 |
peanut | +0.058 |
cube | +0.056 |
ornament | +0.056 |
Neg logit contributions | |
ows | -0.075 |
obb | -0.068 |
pering | -0.068 |
oms | -0.067 |
arl | -0.067 |
Token | be |
Token ID | 307 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
penny | +0.012 |
flavor | +0.012 |
peanut | +0.011 |
ornament | +0.011 |
cube | +0.011 |
Neg logit contributions | |
ows | -0.014 |
pering | -0.013 |
obb | -0.013 |
oms | -0.012 |
arl | -0.012 |
Token | our |
Token ID | 674 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
penny | +0.046 |
flavor | +0.045 |
ornament | +0.044 |
cube | +0.043 |
peanut | +0.042 |
Neg logit contributions | |
ows | -0.050 |
pering | -0.048 |
obb | -0.046 |
arl | -0.045 |
adders | -0.044 |
Token | roof |
Token ID | 9753 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
flavor | +0.045 |
penny | +0.045 |
ornament | +0.043 |
peanut | +0.042 |
cube | +0.042 |
Neg logit contributions | |
ows | -0.074 |
pering | -0.073 |
obb | -0.070 |
arl | -0.070 |
iving | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
flavor | +0.015 |
ornament | +0.015 |
cube | +0.015 |
penny | +0.014 |
color | +0.014 |
Neg logit contributions | |
pering | -0.017 |
obb | -0.016 |
announcements | -0.016 |
arl | -0.016 |
ighters | -0.015 |
Token | We |
Token ID | 775 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
penny | +0.014 |
flavor | +0.014 |
ornament | +0.014 |
peanut | +0.013 |
cube | +0.013 |
Neg logit contributions | |
ows | -0.018 |
pering | -0.018 |
arl | -0.017 |
obb | -0.017 |
iving | -0.017 |
Token | They |
Token ID | 1119 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
ows | +0.012 |
pering | +0.012 |
obb | +0.011 |
arl | +0.011 |
iving | +0.011 |
Neg logit contributions | |
flavor | -0.012 |
penny | -0.012 |
ornament | -0.012 |
peanut | -0.012 |
cube | -0.012 |
Token | wished |
Token ID | 16555 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
pering | +0.089 |
arl | +0.086 |
ows | +0.084 |
iving | +0.084 |
oms | +0.083 |
Neg logit contributions | |
penny | -0.055 |
flavor | -0.054 |
peanut | -0.053 |
ornament | -0.051 |
cube | -0.050 |
Token | they |
Token ID | 484 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
pering | +0.038 |
ows | +0.038 |
arl | +0.037 |
obb | +0.037 |
iving | +0.035 |
Neg logit contributions | |
flavor | -0.041 |
penny | -0.041 |
ornament | -0.039 |
cube | -0.039 |
peanut | -0.039 |
Token | had |
Token ID | 550 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
ows | +0.158 |
pering | +0.157 |
obb | +0.152 |
arl | +0.151 |
iving | +0.148 |
Neg logit contributions | |
penny | -0.067 |
flavor | -0.067 |
ornament | -0.064 |
peanut | -0.061 |
cube | -0.061 |
Token | not |
Token ID | 407 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
ornament | +0.058 |
penny | +0.055 |
cube | +0.055 |
orn | +0.053 |
magnet | +0.052 |
Neg logit contributions | |
ows | -0.054 |
pering | -0.051 |
obb | -0.051 |
announcements | -0.050 |
adders | -0.050 |
Token | gone |
Token ID | 3750 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
ows | +0.023 |
pering | +0.022 |
adders | +0.022 |
iving | +0.021 |
obb | +0.021 |
Neg logit contributions | |
ornament | -0.021 |
penny | -0.020 |
flavor | -0.019 |
cube | -0.019 |
magnet | -0.018 |
Token | in |
Token ID | 287 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
penny | +0.007 |
flavor | +0.007 |
ornament | +0.007 |
peanut | +0.006 |
cube | +0.006 |
Neg logit contributions | |
ows | -0.011 |
pering | -0.011 |
arl | -0.010 |
obb | -0.010 |
iving | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
penny | +0.011 |
flavor | +0.011 |
ornament | +0.010 |
peanut | +0.010 |
cube | +0.010 |
Neg logit contributions | |
ows | -0.014 |
pering | -0.013 |
arl | -0.013 |
obb | -0.013 |
iving | -0.013 |
Token | chamber |
Token ID | 11847 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
pering | +0.041 |
obb | +0.041 |
ows | +0.040 |
arl | +0.039 |
iving | +0.037 |
Neg logit contributions | |
flavor | -0.042 |
penny | -0.041 |
peanut | -0.039 |
ornament | -0.039 |
cube | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
arl | +0.065 |
obb | +0.063 |
pering | +0.062 |
ows | +0.061 |
announcements | +0.060 |
Neg logit contributions | |
penny | -0.036 |
ornament | -0.036 |
cube | -0.036 |
flavor | -0.035 |
peanut | -0.032 |
Token | They |
Token ID | 1119 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
ows | +0.019 |
pering | +0.019 |
obb | +0.018 |
arl | +0.018 |
iving | +0.017 |
Neg logit contributions | |
flavor | -0.018 |
penny | -0.017 |
ornament | -0.017 |
peanut | -0.017 |
cube | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
ows | +0.011 |
pering | +0.011 |
obb | +0.011 |
arl | +0.011 |
iving | +0.010 |
Neg logit contributions | |
penny | -0.011 |
flavor | -0.011 |
ornament | -0.011 |
peanut | -0.011 |
cube | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
flavor | +0.004 |
penny | +0.004 |
ornament | +0.004 |
cube | +0.004 |
peanut | +0.004 |
Neg logit contributions | |
ows | -0.008 |
pering | -0.008 |
obb | -0.007 |
arl | -0.007 |
iving | -0.007 |
Token | Now |
Token ID | 3844 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
flavor | +0.008 |
penny | +0.008 |
cube | +0.008 |
ornament | +0.007 |
peanut | +0.007 |
Neg logit contributions | |
ows | -0.010 |
pering | -0.009 |
obb | -0.009 |
arl | -0.009 |
oms | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +6.70e-01 |
Pos logit contributions | |
pering | +0.084 |
arl | +0.083 |
ows | +0.081 |
obb | +0.080 |
announcements | +0.077 |
Neg logit contributions | |
penny | -0.041 |
peanut | -0.039 |
flavor | -0.039 |
cube | -0.036 |
ornament | -0.034 |
Token | Tim |
Token ID | 5045 |
Feature activation | -4.65e-01 |
Pos logit contributions | |
pering | +0.026 |
ows | +0.026 |
obb | +0.025 |
arl | +0.025 |
oms | +0.024 |
Neg logit contributions | |
flavor | -0.017 |
penny | -0.017 |
peanut | -0.016 |
cube | -0.016 |
ornament | -0.016 |
Token | 's |
Token ID | 338 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
penny | +0.011 |
cube | +0.010 |
peanut | +0.010 |
ornament | +0.010 |
flavor | +0.010 |
Neg logit contributions | |
pering | -0.019 |
arl | -0.019 |
obb | -0.018 |
announcements | -0.018 |
ows | -0.018 |
Token | room |
Token ID | 2119 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
penny | +0.103 |
ornament | +0.102 |
flavor | +0.098 |
cube | +0.093 |
peanut | +0.089 |
Neg logit contributions | |
pering | -0.175 |
ows | -0.173 |
arl | -0.164 |
obb | -0.162 |
announcements | -0.160 |
Token | was |
Token ID | 373 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
flavor | +0.034 |
penny | +0.034 |
cube | +0.032 |
ornament | +0.031 |
peanut | +0.029 |
Neg logit contributions | |
pering | -0.140 |
arl | -0.135 |
ows | -0.131 |
iving | -0.130 |
obb | -0.129 |
Token | clean |
Token ID | 3424 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ows | +0.038 |
pering | +0.038 |
obb | +0.036 |
arl | +0.036 |
iving | +0.035 |
Neg logit contributions | |
penny | -0.030 |
flavor | -0.030 |
ornament | -0.029 |
peanut | -0.028 |
cube | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
ornament | +0.031 |
flavor | +0.030 |
cube | +0.030 |
penny | +0.030 |
magnet | +0.028 |
Neg logit contributions | |
arl | -0.054 |
pering | -0.053 |
adders | -0.053 |
ighters | -0.052 |
oms | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
ows | +0.061 |
pering | +0.061 |
arl | +0.059 |
obb | +0.058 |
oms | +0.057 |
Neg logit contributions | |
penny | -0.061 |
flavor | -0.061 |
ornament | -0.060 |
peanut | -0.059 |
cube | -0.059 |
Token | He |
Token ID | 679 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
flavor | +0.005 |
peanut | +0.005 |
penny | +0.005 |
ornament | +0.005 |
cube | +0.005 |
Neg logit contributions | |
ows | -0.010 |
obb | -0.010 |
pering | -0.009 |
arl | -0.009 |
iving | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
arl | +0.031 |
pering | +0.031 |
iving | +0.030 |
ows | +0.029 |
obb | +0.029 |
Neg logit contributions | |
flavor | -0.013 |
penny | -0.013 |
peanut | -0.012 |
cube | -0.012 |
ornament | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +7.78e-01 |
Pos logit contributions | |
flavor | +0.084 |
penny | +0.082 |
peanut | +0.079 |
ornament | +0.076 |
cube | +0.075 |
Neg logit contributions | |
ows | -0.135 |
pering | -0.135 |
arl | -0.132 |
iving | -0.129 |
obb | -0.128 |
Token | big |
Token ID | 1263 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
ows | +0.024 |
pering | +0.024 |
obb | +0.023 |
arl | +0.023 |
oms | +0.022 |
Neg logit contributions | |
flavor | -0.023 |
ornament | -0.023 |
penny | -0.023 |
cube | -0.022 |
peanut | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
flavor | +0.019 |
penny | +0.019 |
ornament | +0.019 |
peanut | +0.018 |
cube | +0.018 |
Neg logit contributions | |
ows | -0.025 |
pering | -0.025 |
obb | -0.024 |
arl | -0.024 |
oms | -0.023 |
Token | red |
Token ID | 2266 |
Feature activation | -5.83e+00 |
Pos logit contributions | |
ows | +0.034 |
pering | +0.033 |
obb | +0.032 |
arl | +0.031 |
announcements | +0.030 |
Neg logit contributions | |
penny | -0.028 |
flavor | -0.028 |
ornament | -0.027 |
cube | -0.026 |
peanut | -0.026 |
Token | ball |
Token ID | 2613 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
flavor | +0.144 |
penny | +0.143 |
ornament | +0.139 |
peanut | +0.133 |
cube | +0.133 |
Neg logit contributions | |
ows | -0.207 |
pering | -0.205 |
arl | -0.202 |
obb | -0.199 |
oms | -0.195 |
Token | . |
Token ID | 13 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
penny | +0.082 |
cube | +0.080 |
flavor | +0.080 |
ornament | +0.079 |
peanut | +0.078 |
Neg logit contributions | |
pering | -0.092 |
ows | -0.091 |
obb | -0.090 |
arl | -0.088 |
announcements | -0.088 |
Token | He |
Token ID | 679 |
Feature activation | +9.27e-01 |
Pos logit contributions | |
flavor | +0.009 |
penny | +0.009 |
cube | +0.009 |
peanut | +0.009 |
ornament | +0.008 |
Neg logit contributions | |
pering | -0.016 |
ows | -0.016 |
obb | -0.016 |
announcements | -0.015 |
aults | -0.015 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
pering | +0.037 |
ows | +0.037 |
arl | +0.036 |
obb | +0.035 |
iving | +0.034 |
Neg logit contributions | |
flavor | -0.021 |
penny | -0.021 |
ornament | -0.020 |
peanut | -0.019 |
cube | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
pering | +0.006 |
ows | +0.005 |
arl | +0.005 |
obb | +0.005 |
oms | +0.005 |
Neg logit contributions | |
penny | -0.004 |
flavor | -0.004 |
ornament | -0.004 |
cube | -0.004 |
peanut | -0.004 |
Token | gave |
Token ID | 2921 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
possession | +0.058 |
penny | +0.056 |
flavor | +0.054 |
ornament | +0.054 |
ruby | +0.054 |
Neg logit contributions | |
ows | -0.046 |
oms | -0.039 |
amused | -0.038 |
chers | -0.037 |
obb | -0.037 |
Token | her |
Token ID | 607 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
flavor | +0.068 |
penny | +0.067 |
cube | +0.065 |
ornament | +0.062 |
peanut | +0.061 |
Neg logit contributions | |
ows | -0.073 |
arl | -0.070 |
pering | -0.069 |
obb | -0.068 |
oms | -0.067 |
Token | a |
Token ID | 257 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
penny | +0.049 |
flavor | +0.049 |
ornament | +0.047 |
cube | +0.046 |
peanut | +0.046 |
Neg logit contributions | |
ows | -0.051 |
pering | -0.050 |
obb | -0.048 |
arl | -0.048 |
iving | -0.046 |
Token | big |
Token ID | 1263 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
arl | +0.003 |
ows | +0.003 |
obb | +0.003 |
loud | +0.003 |
realize | +0.003 |
Neg logit contributions | |
mar | -0.004 |
ci | -0.004 |
ake | -0.004 |
possession | -0.004 |
mattered | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
mar | +0.031 |
flavor | +0.030 |
mattered | +0.030 |
shr | +0.029 |
penny | +0.029 |
Neg logit contributions | |
arl | -0.039 |
obb | -0.037 |
oms | -0.037 |
ows | -0.037 |
iving | -0.035 |
Token | juicy |
Token ID | 35075 |
Feature activation | -4.97e+00 |
Pos logit contributions | |
pering | +0.014 |
obb | +0.014 |
ows | +0.014 |
arl | +0.013 |
deter | +0.012 |
Neg logit contributions | |
flavor | -0.013 |
penny | -0.013 |
ornament | -0.013 |
cube | -0.013 |
mar | -0.013 |
Token | apple |
Token ID | 17180 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
ci | +0.134 |
mattered | +0.131 |
Ell | +0.127 |
mar | +0.127 |
shr | +0.121 |
Neg logit contributions | |
obb | -0.144 |
arl | -0.137 |
loud | -0.135 |
realize | -0.133 |
men | -0.130 |
Token | . |
Token ID | 13 |
Feature activation | -7.03e-01 |
Pos logit contributions | |
flavor | +0.152 |
shr | +0.148 |
mattered | +0.148 |
penny | +0.145 |
possession | +0.145 |
Neg logit contributions | |
ows | -0.088 |
pering | -0.074 |
arl | -0.074 |
iving | -0.070 |
adders | -0.068 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
penny | +0.020 |
flavor | +0.020 |
ornament | +0.019 |
cube | +0.019 |
peanut | +0.018 |
Neg logit contributions | |
ows | -0.024 |
pering | -0.023 |
arl | -0.023 |
obb | -0.022 |
iving | -0.022 |
Token | took |
Token ID | 1718 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
penny | +0.038 |
possession | +0.038 |
gem | +0.038 |
ornament | +0.038 |
ruby | +0.037 |
Neg logit contributions | |
ows | -0.035 |
obb | -0.031 |
sure | -0.030 |
screams | -0.030 |
realize | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | +8.87e-01 |
Pos logit contributions | |
obb | +0.011 |
ows | +0.010 |
pering | +0.010 |
adders | +0.009 |
oms | +0.009 |
Neg logit contributions | |
cube | -0.018 |
flavor | -0.017 |
penny | -0.017 |
ornament | -0.016 |
possession | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
obb | +0.050 |
pering | +0.048 |
arl | +0.048 |
ighters | +0.047 |
uits | +0.047 |
Neg logit contributions | |
cube | -0.021 |
penny | -0.021 |
flavor | -0.020 |
peanut | -0.020 |
ornament | -0.020 |
Token | Where |
Token ID | 6350 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
penny | +0.009 |
flavor | +0.009 |
ornament | +0.009 |
cube | +0.009 |
peanut | +0.009 |
Neg logit contributions | |
ows | -0.013 |
pering | -0.013 |
obb | -0.013 |
arl | -0.013 |
iving | -0.012 |
Token | is |
Token ID | 318 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
pering | +0.048 |
arl | +0.043 |
ows | +0.042 |
obb | +0.042 |
announcements | +0.042 |
Neg logit contributions | |
flavor | -0.032 |
cube | -0.030 |
penny | -0.030 |
ornament | -0.030 |
peanut | -0.027 |
Token | his |
Token ID | 465 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
flavor | +0.037 |
penny | +0.037 |
ornament | +0.036 |
cube | +0.035 |
peanut | +0.034 |
Neg logit contributions | |
pering | -0.040 |
ows | -0.039 |
obb | -0.038 |
arl | -0.037 |
oms | -0.037 |
Token | blue |
Token ID | 4171 |
Feature activation | -5.29e+00 |
Pos logit contributions | |
penny | +0.029 |
flavor | +0.029 |
ornament | +0.029 |
peanut | +0.028 |
cube | +0.027 |
Neg logit contributions | |
ows | -0.039 |
pering | -0.039 |
obb | -0.038 |
arl | -0.037 |
oms | -0.036 |
Token | bowl |
Token ID | 9396 |
Feature activation | -4.12e+00 |
Pos logit contributions | |
penny | +0.127 |
ornament | +0.124 |
flavor | +0.121 |
peanut | +0.118 |
pol | +0.113 |
Neg logit contributions | |
ows | -0.180 |
arl | -0.176 |
pering | -0.173 |
oms | -0.172 |
obb | -0.170 |
Token | ? |
Token ID | 30 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
flavor | +0.122 |
penny | +0.121 |
ornament | +0.118 |
cube | +0.118 |
peanut | +0.115 |
Neg logit contributions | |
pering | -0.137 |
ows | -0.135 |
obb | -0.128 |
announcements | -0.128 |
arl | -0.126 |
Token |
|
Token ID | 198 |
Feature activation | -6.56e-01 |
Pos logit contributions | |
penny | +0.010 |
flavor | +0.010 |
ornament | +0.010 |
peanut | +0.009 |
cube | +0.009 |
Neg logit contributions | |
ows | -0.012 |
pering | -0.012 |
arl | -0.011 |
obb | -0.011 |
oms | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -6.64e-01 |
Pos logit contributions | |
flavor | +0.014 |
penny | +0.013 |
ornament | +0.013 |
cube | +0.013 |
peanut | +0.013 |
Neg logit contributions | |
ows | -0.027 |
pering | -0.027 |
obb | -0.026 |
iving | -0.026 |
arl | -0.025 |
Token | He |
Token ID | 1544 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
penny | +0.019 |
flavor | +0.019 |
ornament | +0.018 |
cube | +0.018 |
peanut | +0.018 |
Neg logit contributions | |
ows | -0.023 |
pering | -0.022 |
obb | -0.021 |
oms | -0.021 |
arl | -0.020 |
Token | asks |
Token ID | 7893 |
Feature activation | -8.51e-01 |
Pos logit contributions | |
pering | +0.048 |
arl | +0.048 |
ows | +0.047 |
iving | +0.046 |
adders | +0.043 |
Neg logit contributions | |
peanut | -0.039 |
penny | -0.039 |
cube | -0.038 |
flavor | -0.037 |
ornament | -0.037 |
Token | guitar |
Token ID | 10047 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
pering | +0.013 |
ows | +0.013 |
arl | +0.013 |
obb | +0.012 |
announcements | +0.012 |
Neg logit contributions | |
penny | -0.008 |
ornament | -0.008 |
flavor | -0.008 |
cube | -0.007 |
peanut | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
ornament | +0.046 |
flavor | +0.043 |
cube | +0.043 |
penny | +0.042 |
color | +0.040 |
Neg logit contributions | |
arl | -0.083 |
pering | -0.083 |
obb | -0.079 |
ows | -0.076 |
announcements | -0.075 |
Token | He |
Token ID | 679 |
Feature activation | -6.58e-02 |
Pos logit contributions | |
penny | +0.072 |
flavor | +0.071 |
ornament | +0.069 |
peanut | +0.068 |
cube | +0.067 |
Neg logit contributions | |
ows | -0.070 |
pering | -0.069 |
arl | -0.066 |
obb | -0.065 |
iving | -0.064 |
Token | had |
Token ID | 550 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
flavor | +0.001 |
penny | +0.001 |
ornament | +0.001 |
cube | +0.001 |
peanut | +0.001 |
Neg logit contributions | |
pering | -0.003 |
ows | -0.003 |
arl | -0.003 |
obb | -0.003 |
iving | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
flavor | +0.121 |
penny | +0.121 |
ornament | +0.117 |
peanut | +0.115 |
cube | +0.114 |
Neg logit contributions | |
ows | -0.131 |
pering | -0.131 |
obb | -0.125 |
arl | -0.124 |
iving | -0.120 |
Token | nice |
Token ID | 3621 |
Feature activation | -4.11e+00 |
Pos logit contributions | |
arl | +0.004 |
obb | +0.004 |
pering | +0.004 |
ows | +0.004 |
iving | +0.004 |
Neg logit contributions | |
flavor | -0.006 |
seed | -0.006 |
ornament | -0.006 |
penny | -0.006 |
pol | -0.006 |
Token | guitar |
Token ID | 10047 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
penny | +0.112 |
flavor | +0.111 |
ornament | +0.109 |
peanut | +0.108 |
cube | +0.106 |
Neg logit contributions | |
ows | -0.131 |
arl | -0.124 |
iving | -0.123 |
pering | -0.122 |
obb | -0.122 |
Token | with |
Token ID | 351 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
penny | +0.063 |
ornament | +0.061 |
flavor | +0.061 |
cube | +0.061 |
peanut | +0.060 |
Neg logit contributions | |
pering | -0.091 |
ows | -0.088 |
obb | -0.087 |
arl | -0.086 |
announcements | -0.084 |
Token | red |
Token ID | 2266 |
Feature activation | -6.21e+00 |
Pos logit contributions | |
flavor | +0.066 |
penny | +0.066 |
ornament | +0.063 |
peanut | +0.063 |
cube | +0.062 |
Neg logit contributions | |
ows | -0.089 |
pering | -0.088 |
obb | -0.084 |
arl | -0.084 |
iving | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
penny | +0.195 |
flavor | +0.191 |
ornament | +0.191 |
cube | +0.188 |
peanut | +0.186 |
Neg logit contributions | |
ows | -0.182 |
arl | -0.172 |
pering | -0.172 |
oms | -0.170 |
iving | -0.167 |
Token | blue |
Token ID | 4171 |
Feature activation | -5.17e+00 |
Pos logit contributions | |
penny | +0.019 |
flavor | +0.019 |
ornament | +0.019 |
cube | +0.018 |
peanut | +0.018 |
Neg logit contributions | |
ows | -0.022 |
pering | -0.022 |
obb | -0.021 |
arl | -0.021 |
iving | -0.020 |