Token | could |
Token ID | 714 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
ver | +0.008 |
hand | +0.007 |
chief | +0.007 |
Even | +0.007 |
department | +0.007 |
Neg logit contributions | |
fru | -0.007 |
skirts | -0.007 |
ey | -0.007 |
skirt | -0.007 |
makeup | -0.006 |
Token | see |
Token ID | 766 |
Feature activation | -6.25e-01 |
Pos logit contributions | |
ver | +0.019 |
hand | +0.018 |
chief | +0.017 |
department | +0.017 |
rolled | +0.016 |
Neg logit contributions | |
ey | -0.013 |
fru | -0.012 |
skirts | -0.012 |
skirt | -0.011 |
makeup | -0.011 |
Token | lots |
Token ID | 6041 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
ey | +0.021 |
skirts | +0.021 |
fru | +0.020 |
skirt | +0.018 |
ol | +0.018 |
Neg logit contributions | |
ver | -0.021 |
hand | -0.020 |
chief | -0.019 |
rolled | -0.019 |
department | -0.018 |
Token | of |
Token ID | 286 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ver | +0.031 |
hand | +0.025 |
chief | +0.025 |
rolled | +0.025 |
department | +0.022 |
Neg logit contributions | |
ey | -0.084 |
skirts | -0.083 |
fru | -0.082 |
skirt | -0.076 |
makeup | -0.075 |
Token | av |
Token ID | 1196 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
ver | +0.050 |
happily | +0.048 |
chief | +0.047 |
hand | +0.047 |
Even | +0.046 |
Neg logit contributions | |
ey | -0.033 |
skirts | -0.032 |
fru | -0.029 |
ems | -0.029 |
skirt | -0.028 |
Token | oc |
Token ID | 420 |
Feature activation | +5.57e+00 |
Pos logit contributions | |
ver | +0.076 |
hand | +0.068 |
chief | +0.065 |
department | +0.062 |
rolled | +0.061 |
Neg logit contributions | |
ey | -0.130 |
skirts | -0.126 |
fru | -0.125 |
skirt | -0.114 |
ems | -0.112 |
Token | ados |
Token ID | 22484 |
Feature activation | +8.16e-02 |
Pos logit contributions | |
ver | +0.310 |
hand | +0.275 |
tow | +0.274 |
sadly | +0.273 |
attendant | +0.273 |
Neg logit contributions | |
ey | -0.084 |
fru | -0.074 |
ol | -0.071 |
skirts | -0.057 |
brace | -0.051 |
Token | that |
Token ID | 326 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ver | +0.003 |
hand | +0.002 |
chief | +0.002 |
rolled | +0.002 |
happily | +0.002 |
Neg logit contributions | |
ey | -0.003 |
skirts | -0.003 |
fru | -0.003 |
skirt | -0.003 |
ems | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -8.24e-01 |
Pos logit contributions | |
ey | +0.006 |
skirts | +0.006 |
fru | +0.006 |
skirt | +0.006 |
ems | +0.005 |
Neg logit contributions | |
ver | -0.005 |
hand | -0.005 |
chief | -0.005 |
happily | -0.005 |
rolled | -0.004 |
Token | supplier |
Token ID | 22693 |
Feature activation | +9.25e-02 |
Pos logit contributions | |
ey | +0.026 |
skirts | +0.025 |
fru | +0.024 |
skirt | +0.022 |
makeup | +0.021 |
Neg logit contributions | |
ver | -0.032 |
hand | -0.029 |
rolled | -0.028 |
chief | -0.028 |
department | -0.027 |
Token | had |
Token ID | 550 |
Feature activation | -8.19e-01 |
Pos logit contributions | |
ver | +0.003 |
hand | +0.003 |
chief | +0.003 |
department | +0.003 |
Even | +0.003 |
Neg logit contributions | |
ey | -0.003 |
fru | -0.003 |
skirts | -0.003 |
skirt | -0.003 |
ems | -0.003 |
Token | they |
Token ID | 484 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
ver | +0.040 |
hand | +0.037 |
chief | +0.035 |
department | +0.034 |
rolled | +0.034 |
Neg logit contributions | |
skirts | -0.039 |
fru | -0.037 |
ey | -0.037 |
--- | -0.034 |
ol | -0.034 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
ver | +0.016 |
hand | +0.016 |
chief | +0.015 |
rolled | +0.015 |
Even | +0.015 |
Neg logit contributions | |
skirts | -0.014 |
ey | -0.014 |
fru | -0.013 |
skirt | -0.012 |
ems | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
skirts | +0.008 |
ey | +0.008 |
fru | +0.008 |
ol | +0.007 |
da | +0.007 |
Neg logit contributions | |
ver | -0.010 |
hand | -0.008 |
chief | -0.008 |
rolled | -0.008 |
Even | -0.008 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
ver | +0.016 |
hand | +0.015 |
rolled | +0.014 |
chief | +0.014 |
uttered | +0.014 |
Neg logit contributions | |
ey | -0.011 |
fru | -0.011 |
skirts | -0.010 |
skirt | -0.010 |
da | -0.009 |
Token | in |
Token ID | 259 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ver | +0.093 |
rolled | +0.091 |
Even | +0.091 |
hand | +0.090 |
sadly | +0.085 |
Neg logit contributions | |
ey | -0.092 |
skirts | -0.087 |
fru | -0.087 |
da | -0.082 |
robat | -0.079 |
Token | oc |
Token ID | 420 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
ver | +0.042 |
uttered | +0.036 |
hand | +0.035 |
rolled | +0.035 |
Even | +0.034 |
Neg logit contributions | |
ey | -0.042 |
fru | -0.041 |
da | -0.039 |
skirts | -0.038 |
ol | -0.038 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ver | +0.193 |
Even | +0.169 |
case | +0.160 |
happily | +0.157 |
sadly | +0.155 |
Neg logit contributions | |
ey | -0.146 |
fru | -0.142 |
ution | -0.125 |
ol | -0.123 |
da | -0.122 |
Token | . |
Token ID | 13 |
Feature activation | +9.57e-02 |
Pos logit contributions | |
ver | +0.042 |
hand | +0.038 |
Even | +0.038 |
rolled | +0.038 |
chief | +0.037 |
Neg logit contributions | |
skirts | -0.034 |
ey | -0.032 |
fru | -0.031 |
skirt | -0.030 |
makeup | -0.029 |
Token | They |
Token ID | 1119 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
ver | +0.003 |
hand | +0.002 |
chief | +0.002 |
happily | +0.002 |
rolled | +0.002 |
Neg logit contributions | |
ey | -0.004 |
skirts | -0.004 |
fru | -0.004 |
skirt | -0.004 |
ems | -0.004 |
Token | tell |
Token ID | 1560 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
hand | +0.008 |
ver | +0.007 |
chief | +0.007 |
rolled | +0.007 |
department | +0.007 |
Neg logit contributions | |
ey | -0.005 |
skirts | -0.005 |
fru | -0.005 |
skirt | -0.005 |
ems | -0.004 |
Token | their |
Token ID | 511 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
skirts | +0.003 |
fru | +0.003 |
ey | +0.003 |
da | +0.003 |
ol | +0.003 |
Neg logit contributions | |
ver | -0.005 |
hand | -0.004 |
case | -0.004 |
rolled | -0.004 |
department | -0.004 |
Token | he |
Token ID | 339 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
ver | +0.006 |
hand | +0.006 |
department | +0.006 |
rolled | +0.006 |
chief | +0.006 |
Neg logit contributions | |
ey | -0.003 |
fru | -0.003 |
skirts | -0.003 |
--- | -0.003 |
ol | -0.003 |
Token | saw |
Token ID | 2497 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
ver | +0.004 |
hand | +0.004 |
chief | +0.004 |
rolled | +0.004 |
Even | +0.004 |
Neg logit contributions | |
ey | -0.004 |
skirts | -0.004 |
fru | -0.004 |
skirt | -0.004 |
ems | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
ey | +0.012 |
fru | +0.011 |
jam | +0.011 |
da | +0.011 |
ol | +0.011 |
Neg logit contributions | |
ver | -0.016 |
rolled | -0.015 |
hand | -0.014 |
uttered | -0.013 |
department | -0.013 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
ver | +0.025 |
hand | +0.023 |
rolled | +0.021 |
chief | +0.021 |
Even | +0.021 |
Neg logit contributions | |
ey | -0.019 |
fru | -0.018 |
skirts | -0.018 |
skirt | -0.017 |
makeup | -0.016 |
Token | in |
Token ID | 259 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
rolled | +0.098 |
ver | +0.095 |
hand | +0.094 |
Even | +0.091 |
Together | +0.088 |
Neg logit contributions | |
ey | -0.091 |
fru | -0.081 |
da | -0.080 |
jam | -0.079 |
skirts | -0.078 |
Token | oc |
Token ID | 420 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
ver | +0.057 |
hand | +0.049 |
uttered | +0.048 |
rolled | +0.048 |
Even | +0.045 |
Neg logit contributions | |
ey | -0.051 |
fru | -0.049 |
da | -0.046 |
ol | -0.045 |
jam | -0.044 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ver | +0.198 |
Even | +0.163 |
rolled | +0.158 |
case | +0.152 |
Hours | +0.152 |
Neg logit contributions | |
ey | -0.154 |
fru | -0.142 |
ol | -0.129 |
da | -0.127 |
jam | -0.119 |
Token | ! |
Token ID | 0 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
ver | +0.046 |
Even | +0.040 |
hand | +0.040 |
chief | +0.040 |
rolled | +0.039 |
Neg logit contributions | |
ey | -0.032 |
skirts | -0.030 |
fru | -0.028 |
skirt | -0.028 |
doing | -0.027 |
Token | It |
Token ID | 632 |
Feature activation | +9.40e-01 |
Pos logit contributions | |
ey | +0.009 |
fru | +0.009 |
skirts | +0.008 |
--- | +0.008 |
ol | +0.008 |
Neg logit contributions | |
ver | -0.010 |
hand | -0.009 |
chief | -0.009 |
department | -0.009 |
rolled | -0.009 |
Token | looked |
Token ID | 3114 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
ver | +0.032 |
hand | +0.029 |
chief | +0.028 |
department | +0.027 |
rolled | +0.026 |
Neg logit contributions | |
fru | -0.032 |
ey | -0.031 |
skirts | -0.030 |
skirt | -0.028 |
--- | -0.026 |
Token | very |
Token ID | 845 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
ey | +0.017 |
skirts | +0.017 |
fru | +0.017 |
skirt | +0.015 |
makeup | +0.014 |
Neg logit contributions | |
ver | -0.017 |
hand | -0.016 |
chief | -0.015 |
rolled | -0.015 |
department | -0.015 |
Token | couldn |
Token ID | 3521 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
ey | +0.011 |
skirts | +0.010 |
fru | +0.010 |
skirt | +0.009 |
ems | +0.009 |
Neg logit contributions | |
ver | -0.009 |
hand | -0.008 |
chief | -0.008 |
happily | -0.008 |
sadly | -0.008 |
Token | 't |
Token ID | 470 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
ey | +0.004 |
--- | +0.004 |
doing | +0.004 |
makeup | +0.004 |
skirts | +0.004 |
Neg logit contributions | |
chief | -0.003 |
department | -0.003 |
ver | -0.003 |
hand | -0.003 |
year | -0.003 |
Token | find |
Token ID | 1064 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
ey | +0.001 |
skirts | +0.001 |
fru | +0.001 |
skirt | +0.001 |
makeup | +0.001 |
Neg logit contributions | |
ver | -0.001 |
hand | -0.001 |
chief | -0.001 |
department | -0.001 |
rolled | -0.001 |
Token | any |
Token ID | 597 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
skirts | +0.013 |
ey | +0.013 |
fru | +0.012 |
skirt | +0.011 |
makeup | +0.011 |
Neg logit contributions | |
ver | -0.017 |
chief | -0.016 |
hand | -0.016 |
rolled | -0.015 |
department | -0.015 |
Token | av |
Token ID | 1196 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
ver | +0.012 |
hand | +0.011 |
chief | +0.011 |
rolled | +0.011 |
Even | +0.010 |
Neg logit contributions | |
ey | -0.011 |
skirts | -0.010 |
fru | -0.010 |
skirt | -0.009 |
makeup | -0.009 |
Token | oc |
Token ID | 420 |
Feature activation | +4.80e+00 |
Pos logit contributions | |
ver | +0.054 |
hand | +0.047 |
chief | +0.045 |
happily | +0.042 |
rolled | +0.042 |
Neg logit contributions | |
ey | -0.109 |
skirts | -0.107 |
fru | -0.104 |
skirt | -0.098 |
ems | -0.095 |
Token | ados |
Token ID | 22484 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
ver | +0.250 |
hand | +0.228 |
Even | +0.223 |
tow | +0.222 |
department | +0.222 |
Neg logit contributions | |
ey | -0.082 |
fru | -0.073 |
ol | -0.065 |
skirts | -0.061 |
makeup | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
ey | +0.012 |
skirts | +0.012 |
fru | +0.011 |
skirt | +0.010 |
ems | +0.010 |
Neg logit contributions | |
ver | -0.010 |
hand | -0.009 |
chief | -0.008 |
happily | -0.008 |
rolled | -0.008 |
Token | Finally |
Token ID | 9461 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
ey | +0.022 |
fru | +0.021 |
skirts | +0.021 |
makeup | +0.019 |
skirt | +0.019 |
Neg logit contributions | |
ver | -0.019 |
chief | -0.017 |
hand | -0.017 |
department | -0.016 |
rolled | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
ey | +0.013 |
skirts | +0.013 |
fru | +0.013 |
skirt | +0.012 |
ems | +0.011 |
Neg logit contributions | |
ver | -0.010 |
hand | -0.009 |
chief | -0.009 |
rolled | -0.009 |
happily | -0.009 |
Token | she |
Token ID | 673 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
skirts | +0.011 |
ey | +0.011 |
fru | +0.011 |
skirt | +0.010 |
makeup | +0.010 |
Neg logit contributions | |
ver | -0.018 |
hand | -0.017 |
chief | -0.017 |
department | -0.017 |
rolled | -0.016 |
Token | saw |
Token ID | 2497 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
ey | +0.003 |
skirts | +0.003 |
fru | +0.003 |
skirt | +0.003 |
ems | +0.003 |
Neg logit contributions | |
ver | -0.006 |
hand | -0.006 |
chief | -0.006 |
Even | -0.006 |
department | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
skirts | +0.010 |
fru | +0.010 |
ey | +0.010 |
sunglasses | +0.009 |
da | +0.009 |
Neg logit contributions | |
ver | -0.012 |
rolled | -0.010 |
hand | -0.010 |
uttered | -0.009 |
chief | -0.009 |
Token | big |
Token ID | 1263 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
ver | +0.019 |
hand | +0.017 |
chief | +0.016 |
rolled | +0.016 |
Even | +0.016 |
Neg logit contributions | |
ey | -0.015 |
fru | -0.015 |
skirts | -0.015 |
skirt | -0.013 |
makeup | -0.013 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
ver | +0.021 |
hand | +0.019 |
chief | +0.019 |
rolled | +0.018 |
happily | +0.018 |
Neg logit contributions | |
ey | -0.022 |
skirts | -0.022 |
fru | -0.021 |
skirt | -0.020 |
makeup | -0.019 |
Token | in |
Token ID | 259 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
rolled | +0.097 |
ver | +0.095 |
hand | +0.095 |
Even | +0.093 |
Together | +0.090 |
Neg logit contributions | |
ey | -0.091 |
skirts | -0.083 |
fru | -0.082 |
da | -0.081 |
jam | -0.079 |
Token | oc |
Token ID | 420 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
ver | +0.050 |
hand | +0.042 |
uttered | +0.042 |
rolled | +0.041 |
Even | +0.039 |
Neg logit contributions | |
ey | -0.045 |
fru | -0.042 |
da | -0.041 |
ol | -0.040 |
jam | -0.039 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ver | +0.186 |
Even | +0.157 |
sadly | +0.147 |
rolled | +0.147 |
happily | +0.145 |
Neg logit contributions | |
ey | -0.152 |
fru | -0.140 |
ol | -0.126 |
da | -0.125 |
ution | -0.122 |
Token | . |
Token ID | 13 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
ver | +0.048 |
Even | +0.041 |
hand | +0.041 |
department | +0.041 |
chief | +0.040 |
Neg logit contributions | |
ey | -0.032 |
skirts | -0.031 |
doing | -0.029 |
skirt | -0.029 |
makeup | -0.028 |
Token | The |
Token ID | 383 |
Feature activation | +9.21e-02 |
Pos logit contributions | |
ver | +0.002 |
hand | +0.002 |
chief | +0.001 |
uttered | +0.001 |
safely | +0.001 |
Neg logit contributions | |
ey | -0.002 |
fru | -0.002 |
skirts | -0.001 |
ol | -0.001 |
makeup | -0.001 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
ver | +0.003 |
hand | +0.002 |
department | +0.002 |
chief | +0.002 |
rolled | +0.002 |
Neg logit contributions | |
ey | -0.004 |
fru | -0.004 |
skirts | -0.003 |
prett | -0.003 |
skirt | -0.003 |
Token | in |
Token ID | 259 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
ver | +0.059 |
hand | +0.059 |
rolled | +0.056 |
Even | +0.053 |
department | +0.052 |
Neg logit contributions | |
ey | -0.077 |
fru | -0.073 |
skirts | -0.071 |
makeup | -0.065 |
ems | -0.065 |
Token | they |
Token ID | 484 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
ver | +0.034 |
hand | +0.031 |
chief | +0.030 |
happily | +0.029 |
sadly | +0.028 |
Neg logit contributions | |
ey | -0.060 |
skirts | -0.058 |
fru | -0.056 |
skirt | -0.053 |
ems | -0.052 |
Token | could |
Token ID | 714 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
hand | +0.023 |
ver | +0.022 |
rolled | +0.021 |
chief | +0.021 |
Even | +0.021 |
Neg logit contributions | |
fru | -0.017 |
skirts | -0.017 |
ey | -0.016 |
ems | -0.015 |
skirt | -0.015 |
Token | grow |
Token ID | 1663 |
Feature activation | -7.53e-01 |
Pos logit contributions | |
ver | +0.024 |
hand | +0.022 |
chief | +0.022 |
rolled | +0.021 |
department | +0.021 |
Neg logit contributions | |
ey | -0.019 |
skirts | -0.018 |
fru | -0.018 |
skirt | -0.016 |
makeup | -0.016 |
Token | more |
Token ID | 517 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
ey | +0.031 |
skirts | +0.030 |
fru | +0.030 |
skirt | +0.028 |
ems | +0.027 |
Neg logit contributions | |
ver | -0.021 |
hand | -0.018 |
chief | -0.018 |
happily | -0.018 |
sadly | -0.017 |
Token | av |
Token ID | 1196 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
happily | +0.015 |
quickly | +0.014 |
safely | +0.014 |
ver | +0.014 |
chief | +0.013 |
Neg logit contributions | |
ey | -0.013 |
skirts | -0.013 |
ems | -0.012 |
fru | -0.011 |
skirt | -0.011 |
Token | oc |
Token ID | 420 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
ver | +0.056 |
hand | +0.049 |
chief | +0.047 |
happily | +0.044 |
Even | +0.043 |
Neg logit contributions | |
ey | -0.123 |
skirts | -0.121 |
fru | -0.118 |
skirt | -0.112 |
makeup | -0.108 |
Token | ados |
Token ID | 22484 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
ver | +0.263 |
tow | +0.236 |
Even | +0.236 |
case | +0.235 |
sadly | +0.234 |
Neg logit contributions | |
ey | -0.066 |
fru | -0.065 |
ol | -0.058 |
skirts | -0.047 |
brace | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
ey | +0.014 |
skirts | +0.013 |
fru | +0.013 |
skirt | +0.012 |
makeup | +0.012 |
Neg logit contributions | |
ver | -0.010 |
hand | -0.009 |
chief | -0.008 |
happily | -0.008 |
department | -0.008 |
Token | share |
Token ID | 2648 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
ver | +0.006 |
hand | +0.005 |
chief | +0.005 |
happily | +0.005 |
Even | +0.005 |
Neg logit contributions | |
ey | -0.006 |
skirts | -0.006 |
fru | -0.005 |
skirt | -0.005 |
ems | -0.005 |
Token | them |
Token ID | 606 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ey | +0.048 |
skirts | +0.047 |
fru | +0.045 |
skirt | +0.042 |
ems | +0.041 |
Neg logit contributions | |
ver | -0.033 |
hand | -0.029 |
chief | -0.029 |
happily | -0.027 |
Even | -0.027 |
Token | with |
Token ID | 351 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
skirts | +0.058 |
fru | +0.056 |
ey | +0.056 |
skirt | +0.052 |
makeup | +0.050 |
Neg logit contributions | |
ver | -0.048 |
hand | -0.044 |
chief | -0.043 |
rolled | -0.042 |
department | -0.042 |
Token | saw |
Token ID | 2497 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
ver | +0.012 |
hand | +0.011 |
chief | +0.011 |
department | +0.011 |
Even | +0.011 |
Neg logit contributions | |
ey | -0.009 |
skirts | -0.009 |
fru | -0.008 |
skirt | -0.007 |
makeup | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
skirts | +0.014 |
ey | +0.013 |
fru | +0.013 |
da | +0.013 |
skirt | +0.012 |
Neg logit contributions | |
ver | -0.012 |
hand | -0.010 |
chief | -0.010 |
rolled | -0.010 |
department | -0.009 |
Token | big |
Token ID | 1263 |
Feature activation | +6.87e-01 |
Pos logit contributions | |
ver | +0.025 |
hand | +0.023 |
chief | +0.022 |
rolled | +0.022 |
Even | +0.022 |
Neg logit contributions | |
ey | -0.021 |
skirts | -0.020 |
fru | -0.020 |
skirt | -0.018 |
makeup | -0.017 |
Token | rh |
Token ID | 9529 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
ver | +0.022 |
hand | +0.020 |
chief | +0.020 |
happily | +0.019 |
Even | +0.019 |
Neg logit contributions | |
ey | -0.025 |
skirts | -0.024 |
fru | -0.023 |
skirt | -0.022 |
ems | -0.021 |
Token | in |
Token ID | 259 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
hand | +0.108 |
Even | +0.107 |
rolled | +0.106 |
ver | +0.106 |
Together | +0.104 |
Neg logit contributions | |
ey | -0.087 |
skirts | -0.082 |
fru | -0.081 |
da | -0.081 |
P | -0.078 |
Token | oc |
Token ID | 420 |
Feature activation | +4.74e+00 |
Pos logit contributions | |
ver | +0.045 |
hand | +0.039 |
rolled | +0.036 |
uttered | +0.036 |
department | +0.036 |
Neg logit contributions | |
ey | -0.053 |
fru | -0.051 |
skirts | -0.049 |
da | -0.048 |
skirt | -0.047 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ver | +0.194 |
Even | +0.166 |
Hours | +0.156 |
case | +0.155 |
sadly | +0.153 |
Neg logit contributions | |
ey | -0.139 |
fru | -0.130 |
da | -0.121 |
ol | -0.115 |
brace | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
ver | +0.044 |
hand | +0.039 |
chief | +0.038 |
Even | +0.038 |
department | +0.038 |
Neg logit contributions | |
ey | -0.032 |
skirts | -0.032 |
skirt | -0.030 |
fru | -0.029 |
makeup | -0.029 |
Token | It |
Token ID | 632 |
Feature activation | +6.65e-01 |
Pos logit contributions | |
ey | +0.004 |
skirts | +0.004 |
fru | +0.004 |
skirt | +0.004 |
makeup | +0.004 |
Neg logit contributions | |
ver | -0.003 |
hand | -0.003 |
chief | -0.003 |
department | -0.002 |
rolled | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
ver | +0.021 |
hand | +0.019 |
chief | +0.018 |
department | +0.017 |
rolled | +0.017 |
Neg logit contributions | |
ey | -0.025 |
skirts | -0.025 |
fru | -0.025 |
skirt | -0.022 |
makeup | -0.021 |
Token | dirty |
Token ID | 11841 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
ver | +0.005 |
hand | +0.004 |
chief | +0.004 |
happily | +0.004 |
Even | +0.004 |
Neg logit contributions | |
ey | -0.006 |
skirts | -0.005 |
fru | -0.005 |
skirt | -0.005 |
ems | -0.005 |
Token | at |
Token ID | 379 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
ver | +0.011 |
sadly | +0.010 |
happily | +0.010 |
chief | +0.010 |
hand | +0.010 |
Neg logit contributions | |
ey | -0.016 |
skirts | -0.015 |
fru | -0.014 |
skirt | -0.013 |
ems | -0.013 |
Token | that |
Token ID | 326 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ver | +0.008 |
hand | +0.007 |
department | +0.007 |
chief | +0.007 |
rolled | +0.007 |
Neg logit contributions | |
ey | -0.007 |
skirts | -0.007 |
fru | -0.007 |
makeup | -0.006 |
da | -0.006 |
Token | big |
Token ID | 1263 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
ver | +0.046 |
hand | +0.044 |
department | +0.041 |
uttered | +0.041 |
rolled | +0.041 |
Neg logit contributions | |
fru | -0.051 |
ey | -0.050 |
skirts | -0.048 |
ems | -0.045 |
prett | -0.045 |
Token | rh |
Token ID | 9529 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
ver | +0.020 |
hand | +0.018 |
chief | +0.018 |
happily | +0.017 |
Even | +0.017 |
Neg logit contributions | |
ey | -0.025 |
skirts | -0.025 |
fru | -0.024 |
skirt | -0.022 |
ems | -0.021 |
Token | in |
Token ID | 259 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
ver | +0.102 |
hand | +0.101 |
Even | +0.099 |
rolled | +0.094 |
department | +0.093 |
Neg logit contributions | |
ey | -0.117 |
skirts | -0.110 |
fru | -0.108 |
da | -0.101 |
ems | -0.100 |
Token | oc |
Token ID | 420 |
Feature activation | +4.74e+00 |
Pos logit contributions | |
ver | +0.056 |
hand | +0.052 |
Even | +0.049 |
department | +0.048 |
happily | +0.048 |
Neg logit contributions | |
ey | -0.071 |
fru | -0.068 |
skirts | -0.067 |
ems | -0.064 |
skirt | -0.062 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ver | +0.191 |
case | +0.168 |
Even | +0.165 |
tow | +0.162 |
happily | +0.159 |
Neg logit contributions | |
ey | -0.137 |
fru | -0.127 |
ol | -0.123 |
da | -0.111 |
jewelry | -0.111 |
Token | !" |
Token ID | 2474 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
ver | +0.055 |
hand | +0.053 |
Even | +0.050 |
chief | +0.050 |
department | +0.049 |
Neg logit contributions | |
ey | -0.036 |
skirts | -0.035 |
fru | -0.034 |
skirt | -0.031 |
ems | -0.031 |
Token | Billy |
Token ID | 15890 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
ver | +0.015 |
hand | +0.014 |
chief | +0.013 |
department | +0.013 |
happily | +0.012 |
Neg logit contributions | |
ey | -0.014 |
skirts | -0.013 |
fru | -0.013 |
skirt | -0.012 |
makeup | -0.011 |
Token | said |
Token ID | 531 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
ver | +0.034 |
hand | +0.032 |
chief | +0.032 |
department | +0.031 |
rolled | +0.030 |
Neg logit contributions | |
skirts | -0.022 |
ey | -0.021 |
fru | -0.021 |
skirt | -0.019 |
makeup | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
hand | +0.026 |
ver | +0.026 |
department | +0.026 |
Even | +0.026 |
chief | +0.025 |
Neg logit contributions | |
fru | -0.017 |
skirts | -0.015 |
makeup | -0.015 |
skirt | -0.015 |
ey | -0.014 |
Token | suddenly |
Token ID | 6451 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
ver | +0.001 |
hand | +0.001 |
chief | +0.001 |
happily | +0.001 |
rolled | +0.001 |
Neg logit contributions | |
ey | -0.001 |
skirts | -0.001 |
fru | -0.001 |
skirt | -0.000 |
ems | -0.000 |
Token | saw |
Token ID | 2497 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
ey | +0.015 |
skirts | +0.015 |
fru | +0.015 |
skirt | +0.014 |
ems | +0.013 |
Neg logit contributions | |
ver | -0.016 |
hand | -0.015 |
chief | -0.014 |
department | -0.014 |
Even | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
ey | +0.016 |
skirts | +0.016 |
fru | +0.015 |
makeup | +0.015 |
skirt | +0.015 |
Neg logit contributions | |
ver | -0.016 |
chief | -0.014 |
hand | -0.014 |
department | -0.014 |
rolled | -0.014 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
ver | +0.024 |
hand | +0.022 |
chief | +0.021 |
rolled | +0.021 |
Even | +0.020 |
Neg logit contributions | |
ey | -0.018 |
skirts | -0.017 |
fru | -0.017 |
skirt | -0.015 |
makeup | -0.014 |
Token | in |
Token ID | 259 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
rolled | +0.102 |
ver | +0.100 |
Even | +0.100 |
Hours | +0.098 |
hand | +0.098 |
Neg logit contributions | |
ey | -0.079 |
da | -0.070 |
fru | -0.070 |
P | -0.068 |
jam | -0.066 |
Token | oc |
Token ID | 420 |
Feature activation | +4.73e+00 |
Pos logit contributions | |
ver | +0.047 |
hand | +0.039 |
uttered | +0.039 |
department | +0.038 |
rolled | +0.037 |
Neg logit contributions | |
ey | -0.044 |
fru | -0.042 |
da | -0.040 |
ol | -0.039 |
jam | -0.038 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ver | +0.196 |
Even | +0.159 |
case | +0.152 |
Hours | +0.151 |
Together | +0.151 |
Neg logit contributions | |
ey | -0.145 |
fru | -0.134 |
da | -0.120 |
jewelry | -0.118 |
brace | -0.114 |
Token | in |
Token ID | 287 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
ver | +0.041 |
hand | +0.036 |
Even | +0.035 |
chief | +0.035 |
department | +0.035 |
Neg logit contributions | |
ey | -0.030 |
skirts | -0.029 |
skirt | -0.027 |
fru | -0.027 |
makeup | -0.026 |
Token | front |
Token ID | 2166 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
ey | +0.015 |
skirts | +0.014 |
fru | +0.014 |
skirt | +0.013 |
makeup | +0.013 |
Neg logit contributions | |
ver | -0.016 |
hand | -0.014 |
chief | -0.014 |
department | -0.013 |
rolled | -0.013 |
Token | of |
Token ID | 286 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
ver | +0.006 |
hand | +0.005 |
chief | +0.005 |
department | +0.004 |
rolled | +0.004 |
Neg logit contributions | |
ey | -0.018 |
skirts | -0.017 |
fru | -0.017 |
skirt | -0.016 |
makeup | -0.016 |
Token | him |
Token ID | 683 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ver | +0.013 |
chief | +0.012 |
hand | +0.011 |
department | +0.011 |
Even | +0.011 |
Neg logit contributions | |
ey | -0.017 |
skirts | -0.017 |
fru | -0.016 |
skirt | -0.015 |
makeup | -0.015 |
Token | didn |
Token ID | 1422 |
Feature activation | +6.14e-02 |
Pos logit contributions | |
ver | +0.008 |
hand | +0.008 |
chief | +0.008 |
department | +0.007 |
rolled | +0.007 |
Neg logit contributions | |
skirts | -0.007 |
ey | -0.007 |
fru | -0.007 |
skirt | -0.006 |
ems | -0.006 |
Token | 't |
Token ID | 470 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
chief | +0.002 |
hand | +0.002 |
department | +0.002 |
brigade | +0.001 |
ver | +0.001 |
Neg logit contributions | |
ey | -0.002 |
--- | -0.002 |
skirts | -0.002 |
makeup | -0.002 |
rys | -0.002 |
Token | like |
Token ID | 588 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
ver | +0.002 |
happily | +0.002 |
Even | +0.002 |
quickly | +0.001 |
safely | +0.001 |
Neg logit contributions | |
ey | -0.004 |
fru | -0.004 |
skirts | -0.004 |
eking | -0.003 |
skirt | -0.003 |
Token | wet |
Token ID | 9583 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
ey | +0.010 |
skirts | +0.009 |
fru | +0.009 |
ems | +0.009 |
skirt | +0.009 |
Neg logit contributions | |
ver | -0.008 |
hand | -0.007 |
happily | -0.007 |
chief | -0.007 |
Even | -0.007 |
Token | av |
Token ID | 1196 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
happily | +0.006 |
sadly | +0.006 |
quickly | +0.006 |
hand | +0.005 |
Even | +0.005 |
Neg logit contributions | |
ey | -0.005 |
skirts | -0.005 |
fru | -0.005 |
ems | -0.005 |
skirt | -0.005 |
Token | oc |
Token ID | 420 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
ver | +0.052 |
happily | +0.042 |
hand | +0.041 |
Even | +0.040 |
chief | +0.039 |
Neg logit contributions | |
ey | -0.139 |
skirts | -0.137 |
fru | -0.132 |
skirt | -0.128 |
makeup | -0.123 |
Token | ados |
Token ID | 22484 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
ver | +0.244 |
hand | +0.227 |
chief | +0.225 |
department | +0.219 |
tow | +0.217 |
Neg logit contributions | |
ey | -0.082 |
fru | -0.075 |
skirts | -0.066 |
ol | -0.065 |
makeup | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
ver | +0.000 |
happily | +0.000 |
quickly | +0.000 |
sadly | +0.000 |
hand | +0.000 |
Neg logit contributions | |
ey | -0.001 |
fru | -0.001 |
skirts | -0.001 |
skirt | -0.001 |
ems | -0.001 |
Token | so |
Token ID | 523 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ey | +0.026 |
fru | +0.024 |
reath | +0.023 |
skirts | +0.023 |
ems | +0.023 |
Neg logit contributions | |
quickly | -0.015 |
sadly | -0.015 |
Even | -0.014 |
ver | -0.013 |
happily | -0.013 |
Token | he |
Token ID | 339 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
ey | +0.045 |
skirts | +0.042 |
fru | +0.041 |
ems | +0.038 |
skirt | +0.037 |
Neg logit contributions | |
ver | -0.029 |
sadly | -0.027 |
happily | -0.027 |
hand | -0.026 |
quickly | -0.026 |
Token | decided |
Token ID | 3066 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
ver | +0.013 |
hand | +0.012 |
chief | +0.011 |
rolled | +0.011 |
department | +0.011 |
Neg logit contributions | |
skirts | -0.012 |
ey | -0.012 |
fru | -0.012 |
skirt | -0.011 |
makeup | -0.011 |
Token | so |
Token ID | 523 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
ver | +0.006 |
hand | +0.006 |
department | +0.006 |
chief | +0.005 |
rolled | +0.005 |
Neg logit contributions | |
fru | -0.004 |
skirts | -0.004 |
doing | -0.004 |
ey | -0.004 |
prett | -0.003 |
Token | glad |
Token ID | 9675 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
fru | +0.018 |
ey | +0.017 |
skirts | +0.015 |
prett | +0.014 |
doing | +0.014 |
Neg logit contributions | |
hand | -0.031 |
ver | -0.031 |
department | -0.031 |
brigade | -0.029 |
chief | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
ver | +0.015 |
hand | +0.014 |
chief | +0.013 |
happily | +0.012 |
rolled | +0.012 |
Neg logit contributions | |
ey | -0.019 |
skirts | -0.019 |
fru | -0.018 |
skirt | -0.017 |
ems | -0.016 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
ver | +0.004 |
hand | +0.004 |
rolled | +0.004 |
uttered | +0.003 |
Even | +0.003 |
Neg logit contributions | |
fru | -0.004 |
ey | -0.004 |
skirts | -0.004 |
prett | -0.003 |
makeup | -0.003 |
Token | in |
Token ID | 259 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
hand | +0.093 |
rolled | +0.092 |
Together | +0.091 |
Even | +0.090 |
ver | +0.086 |
Neg logit contributions | |
ey | -0.070 |
fru | -0.069 |
b | -0.064 |
eking | -0.061 |
ol | -0.060 |
Token | oc |
Token ID | 420 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
ver | +0.035 |
hand | +0.031 |
rolled | +0.031 |
uttered | +0.031 |
Even | +0.030 |
Neg logit contributions | |
fru | -0.039 |
ey | -0.038 |
skirts | -0.034 |
da | -0.033 |
ems | -0.033 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ver | +0.194 |
Even | +0.166 |
case | +0.164 |
rolled | +0.156 |
Together | +0.155 |
Neg logit contributions | |
ey | -0.129 |
fru | -0.128 |
ol | -0.109 |
ution | -0.107 |
brace | -0.106 |
Token | went |
Token ID | 1816 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
ver | +0.037 |
hand | +0.035 |
rolled | +0.035 |
Even | +0.034 |
chief | +0.033 |
Neg logit contributions | |
fru | -0.029 |
skirts | -0.028 |
ey | -0.027 |
skirt | -0.026 |
ems | -0.025 |
Token | away |
Token ID | 1497 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
ver | +0.010 |
chief | +0.008 |
hand | +0.008 |
department | +0.008 |
Even | +0.008 |
Neg logit contributions | |
ey | -0.011 |
skirts | -0.011 |
fru | -0.010 |
skirt | -0.010 |
makeup | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
ver | +0.014 |
hand | +0.013 |
chief | +0.013 |
rolled | +0.012 |
department | +0.012 |
Neg logit contributions | |
skirts | -0.012 |
ey | -0.012 |
fru | -0.011 |
skirt | -0.011 |
makeup | -0.010 |
Token | He |
Token ID | 679 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
ey | +0.013 |
fru | +0.012 |
skirts | +0.011 |
--- | +0.010 |
ems | +0.010 |
Neg logit contributions | |
ver | -0.011 |
chief | -0.011 |
hand | -0.011 |
department | -0.010 |
brigade | -0.010 |
Token | into |
Token ID | 656 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ver | +0.029 |
hand | +0.026 |
chief | +0.025 |
department | +0.024 |
brigade | +0.023 |
Neg logit contributions | |
skirts | -0.029 |
ey | -0.029 |
fru | -0.027 |
skirt | -0.026 |
makeup | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
ey | +0.007 |
skirts | +0.007 |
fru | +0.007 |
skirt | +0.006 |
makeup | +0.006 |
Neg logit contributions | |
ver | -0.006 |
hand | -0.005 |
chief | -0.005 |
department | -0.005 |
rolled | -0.005 |
Token | big |
Token ID | 1263 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
ver | +0.004 |
hand | +0.004 |
chief | +0.004 |
happily | +0.004 |
Even | +0.004 |
Neg logit contributions | |
ey | -0.004 |
skirts | -0.004 |
fru | -0.004 |
skirt | -0.004 |
ems | -0.004 |
Token | rh |
Token ID | 9529 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
ver | +0.005 |
chief | +0.005 |
happily | +0.005 |
hand | +0.005 |
sadly | +0.005 |
Neg logit contributions | |
skirts | -0.007 |
ey | -0.007 |
fru | -0.007 |
ems | -0.006 |
skirt | -0.006 |
Token | in |
Token ID | 259 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
ver | +0.101 |
hand | +0.098 |
rolled | +0.096 |
Even | +0.090 |
department | +0.088 |
Neg logit contributions | |
ey | -0.107 |
skirts | -0.097 |
skirt | -0.094 |
fru | -0.093 |
da | -0.092 |
Token | oc |
Token ID | 420 |
Feature activation | +4.67e+00 |
Pos logit contributions | |
ver | +0.053 |
hand | +0.045 |
rolled | +0.044 |
uttered | +0.044 |
chief | +0.042 |
Neg logit contributions | |
ey | -0.060 |
fru | -0.055 |
skirts | -0.055 |
skirt | -0.054 |
da | -0.052 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ver | +0.182 |
Even | +0.143 |
rolled | +0.143 |
hand | +0.139 |
happily | +0.136 |
Neg logit contributions | |
ey | -0.165 |
fru | -0.146 |
ol | -0.134 |
da | -0.132 |
skirts | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
ver | +0.045 |
hand | +0.040 |
chief | +0.039 |
rolled | +0.038 |
Even | +0.038 |
Neg logit contributions | |
ey | -0.034 |
skirts | -0.033 |
skirt | -0.031 |
fru | -0.030 |
makeup | -0.029 |
Token | The |
Token ID | 383 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
ey | +0.011 |
fru | +0.011 |
skirts | +0.010 |
skirt | +0.010 |
makeup | +0.009 |
Neg logit contributions | |
ver | -0.009 |
hand | -0.008 |
chief | -0.008 |
department | -0.008 |
brigade | -0.007 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
ey | +0.005 |
fru | +0.005 |
skirts | +0.005 |
ol | +0.004 |
skirt | +0.004 |
Neg logit contributions | |
ver | -0.004 |
hand | -0.004 |
rolled | -0.003 |
uttered | -0.003 |
chief | -0.003 |
Token | in |
Token ID | 259 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
hand | +0.066 |
ver | +0.065 |
rolled | +0.062 |
department | +0.058 |
Even | +0.058 |
Neg logit contributions | |
ey | -0.078 |
fru | -0.072 |
skirts | -0.072 |
skirt | -0.066 |
ol | -0.065 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ey | +0.064 |
skirts | +0.063 |
fru | +0.060 |
skirt | +0.054 |
ems | +0.052 |
Neg logit contributions | |
ver | -0.097 |
hand | -0.092 |
chief | -0.090 |
rolled | -0.087 |
happily | -0.087 |
Token | The |
Token ID | 383 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
ver | +0.028 |
hand | +0.025 |
chief | +0.025 |
rolled | +0.023 |
department | +0.023 |
Neg logit contributions | |
ey | -0.046 |
skirts | -0.045 |
fru | -0.044 |
skirt | -0.041 |
ems | -0.040 |
Token | huge |
Token ID | 3236 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
ver | +0.024 |
hand | +0.021 |
rolled | +0.020 |
Even | +0.020 |
chief | +0.020 |
Neg logit contributions | |
ey | -0.019 |
skirts | -0.018 |
fru | -0.018 |
skirt | -0.016 |
makeup | -0.015 |
Token | rh |
Token ID | 9529 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
ver | +0.010 |
hand | +0.009 |
chief | +0.009 |
rolled | +0.008 |
Even | +0.008 |
Neg logit contributions | |
ey | -0.010 |
skirts | -0.009 |
fru | -0.009 |
skirt | -0.008 |
makeup | -0.008 |
Token | in |
Token ID | 259 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
rolled | +0.106 |
hand | +0.106 |
ver | +0.103 |
Even | +0.102 |
Together | +0.101 |
Neg logit contributions | |
ey | -0.095 |
fru | -0.089 |
skirts | -0.082 |
da | -0.080 |
jam | -0.080 |
Token | oc |
Token ID | 420 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
ver | +0.050 |
hand | +0.046 |
department | +0.043 |
rolled | +0.043 |
chief | +0.043 |
Neg logit contributions | |
ey | -0.046 |
fru | -0.045 |
skirts | -0.042 |
skirt | -0.040 |
ems | -0.039 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
ver | +0.188 |
Even | +0.159 |
case | +0.153 |
rolled | +0.151 |
Together | +0.150 |
Neg logit contributions | |
ey | -0.149 |
fru | -0.137 |
ol | -0.124 |
da | -0.117 |
mel | -0.114 |
Token | thought |
Token ID | 1807 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
ver | +0.053 |
hand | +0.050 |
rolled | +0.047 |
chief | +0.047 |
department | +0.047 |
Neg logit contributions | |
ey | -0.032 |
fru | -0.032 |
skirts | -0.031 |
skirt | -0.028 |
ems | -0.027 |
Token | for |
Token ID | 329 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
ver | +0.005 |
hand | +0.005 |
chief | +0.005 |
rolled | +0.005 |
happily | +0.004 |
Neg logit contributions | |
ey | -0.007 |
skirts | -0.007 |
fru | -0.007 |
skirt | -0.006 |
ems | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
ver | +0.017 |
hand | +0.015 |
chief | +0.015 |
rolled | +0.014 |
Even | +0.014 |
Neg logit contributions | |
ey | -0.025 |
skirts | -0.025 |
fru | -0.025 |
skirt | -0.022 |
makeup | -0.022 |
Token | moment |
Token ID | 2589 |
Feature activation | -6.65e-02 |
Pos logit contributions | |
ey | +0.000 |
skirts | +0.000 |
fru | +0.000 |
skirt | +0.000 |
ems | +0.000 |
Neg logit contributions | |
ver | -0.000 |
hand | -0.000 |
chief | -0.000 |
happily | -0.000 |
department | -0.000 |
Token | that |
Token ID | 326 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ver | +0.003 |
hand | +0.003 |
chief | +0.003 |
rolled | +0.003 |
department | +0.003 |
Neg logit contributions | |
ey | -0.003 |
skirts | -0.003 |
fru | -0.003 |
skirt | -0.003 |
makeup | -0.003 |
Token | boss |
Token ID | 6478 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
ver | +0.039 |
hand | +0.037 |
chief | +0.035 |
rolled | +0.035 |
department | +0.034 |
Neg logit contributions | |
ey | -0.034 |
skirts | -0.034 |
fru | -0.034 |
skirt | -0.030 |
makeup | -0.029 |
Token | y |
Token ID | 88 |
Feature activation | -8.84e-02 |
Pos logit contributions | |
ey | +0.034 |
skirts | +0.033 |
fru | +0.032 |
ems | +0.030 |
skirt | +0.030 |
Neg logit contributions | |
ver | -0.032 |
hand | -0.031 |
chief | -0.030 |
department | -0.028 |
rolled | -0.028 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
ey | +0.003 |
skirts | +0.003 |
fru | +0.003 |
skirt | +0.003 |
ems | +0.003 |
Neg logit contributions | |
ver | -0.003 |
hand | -0.003 |
chief | -0.003 |
rolled | -0.003 |
Even | -0.003 |
Token | in |
Token ID | 259 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
hand | +0.094 |
Together | +0.092 |
Even | +0.091 |
ver | +0.090 |
rolled | +0.090 |
Neg logit contributions | |
ey | -0.091 |
fru | -0.085 |
skirts | -0.083 |
eking | -0.080 |
ems | -0.077 |
Token | oc |
Token ID | 420 |
Feature activation | +4.64e+00 |
Pos logit contributions | |
ver | +0.050 |
hand | +0.046 |
rolled | +0.046 |
Even | +0.043 |
chief | +0.043 |
Neg logit contributions | |
ey | -0.059 |
fru | -0.058 |
skirts | -0.057 |
ems | -0.055 |
skirt | -0.052 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ver | +0.180 |
case | +0.164 |
Even | +0.155 |
tow | +0.151 |
sadly | +0.149 |
Neg logit contributions | |
ey | -0.132 |
fru | -0.128 |
ol | -0.119 |
ution | -0.119 |
brace | -0.115 |
Token | . |
Token ID | 13 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
ver | +0.040 |
hand | +0.038 |
rolled | +0.036 |
Even | +0.036 |
chief | +0.036 |
Neg logit contributions | |
skirts | -0.040 |
ey | -0.039 |
fru | -0.038 |
ems | -0.036 |
skirt | -0.036 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +6.14e-01 |
Pos logit contributions | |
ey | +0.010 |
fru | +0.009 |
skirts | +0.008 |
--- | +0.008 |
ol | +0.008 |
Neg logit contributions | |
ver | -0.009 |
chief | -0.009 |
hand | -0.009 |
department | -0.009 |
quickly | -0.008 |
Token | D |
Token ID | 35 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
ver | +0.023 |
chief | +0.021 |
hand | +0.021 |
department | +0.020 |
sadly | +0.020 |
Neg logit contributions | |
ey | -0.019 |
skirts | -0.017 |
fru | -0.017 |
skirt | -0.016 |
--- | -0.016 |
Token | ee |
Token ID | 1453 |
Feature activation | +8.44e-01 |
Pos logit contributions | |
ver | +0.014 |
chief | +0.013 |
hand | +0.013 |
department | +0.013 |
sadly | +0.012 |
Neg logit contributions | |
ey | -0.016 |
skirts | -0.015 |
fru | -0.014 |
skirt | -0.014 |
--- | -0.013 |
Token | they |
Token ID | 484 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
skirts | +0.012 |
ey | +0.012 |
skirt | +0.011 |
fru | +0.011 |
doing | +0.011 |
Neg logit contributions | |
ver | -0.018 |
department | -0.015 |
ittle | -0.015 |
chief | -0.015 |
itt | -0.015 |
Token | saw |
Token ID | 2497 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
ver | +0.029 |
hand | +0.027 |
chief | +0.027 |
department | +0.026 |
Even | +0.026 |
Neg logit contributions | |
ey | -0.022 |
skirts | -0.022 |
fru | -0.021 |
skirt | -0.019 |
ems | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
skirts | +0.011 |
ey | +0.010 |
fru | +0.010 |
makeup | +0.010 |
da | +0.010 |
Neg logit contributions | |
ver | -0.012 |
chief | -0.010 |
hand | -0.010 |
department | -0.010 |
rolled | -0.010 |
Token | rh |
Token ID | 9529 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
ver | +0.028 |
hand | +0.025 |
rolled | +0.024 |
chief | +0.024 |
Even | +0.024 |
Neg logit contributions | |
ey | -0.020 |
skirts | -0.019 |
fru | -0.019 |
skirt | -0.018 |
makeup | -0.017 |
Token | in |
Token ID | 259 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
Even | +0.110 |
ver | +0.109 |
hand | +0.108 |
Hours | +0.106 |
rolled | +0.105 |
Neg logit contributions | |
ey | -0.081 |
P | -0.079 |
da | -0.077 |
skirts | -0.075 |
fru | -0.073 |
Token | oc |
Token ID | 420 |
Feature activation | +4.62e+00 |
Pos logit contributions | |
ver | +0.050 |
uttered | +0.041 |
hand | +0.041 |
rolled | +0.040 |
department | +0.040 |
Neg logit contributions | |
ey | -0.048 |
fru | -0.045 |
da | -0.044 |
ol | -0.043 |
skirts | -0.042 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ver | +0.206 |
Even | +0.174 |
case | +0.166 |
Together | +0.165 |
Hours | +0.163 |
Neg logit contributions | |
ey | -0.120 |
fru | -0.105 |
ol | -0.099 |
da | -0.096 |
jewelry | -0.095 |
Token | in |
Token ID | 287 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
ver | +0.045 |
Even | +0.039 |
hand | +0.039 |
chief | +0.039 |
department | +0.039 |
Neg logit contributions | |
ey | -0.032 |
skirts | -0.032 |
skirt | -0.029 |
fru | -0.028 |
makeup | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
ey | +0.011 |
skirts | +0.011 |
fru | +0.010 |
skirt | +0.010 |
makeup | +0.009 |
Neg logit contributions | |
ver | -0.012 |
hand | -0.010 |
chief | -0.010 |
department | -0.010 |
rolled | -0.009 |
Token | field |
Token ID | 2214 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
ey | +0.019 |
skirts | +0.017 |
fru | +0.017 |
skirt | +0.016 |
prett | +0.015 |
Neg logit contributions | |
ver | -0.036 |
uttered | -0.030 |
hand | -0.030 |
rolled | -0.030 |
chief | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
ver | +0.002 |
hand | +0.001 |
chief | +0.001 |
department | +0.001 |
Even | +0.001 |
Neg logit contributions | |
ey | -0.001 |
skirts | -0.001 |
fru | -0.001 |
skirt | -0.001 |
makeup | -0.001 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
ey | +0.015 |
skirts | +0.015 |
fru | +0.014 |
skirt | +0.013 |
ems | +0.012 |
Neg logit contributions | |
ver | -0.015 |
hand | -0.013 |
chief | -0.013 |
department | -0.012 |
rolled | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
skirts | +0.011 |
ey | +0.011 |
fru | +0.011 |
skirt | +0.010 |
makeup | +0.010 |
Neg logit contributions | |
ver | -0.008 |
hand | -0.007 |
chief | -0.007 |
rolled | -0.007 |
department | -0.007 |
Token | big |
Token ID | 1263 |
Feature activation | +7.00e-01 |
Pos logit contributions | |
ver | +0.023 |
hand | +0.021 |
chief | +0.021 |
happily | +0.020 |
department | +0.020 |
Neg logit contributions | |
ey | -0.021 |
skirts | -0.021 |
fru | -0.020 |
ems | -0.018 |
skirt | -0.018 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
ver | +0.020 |
chief | +0.020 |
happily | +0.019 |
hand | +0.019 |
quickly | +0.019 |
Neg logit contributions | |
skirts | -0.025 |
ey | -0.024 |
fru | -0.024 |
ems | -0.023 |
--- | -0.021 |
Token | in |
Token ID | 259 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
ver | +0.087 |
hand | +0.085 |
rolled | +0.083 |
Even | +0.080 |
department | +0.077 |
Neg logit contributions | |
ey | -0.102 |
skirts | -0.097 |
fru | -0.093 |
skirt | -0.092 |
da | -0.089 |
Token | oc |
Token ID | 420 |
Feature activation | +4.61e+00 |
Pos logit contributions | |
ver | +0.045 |
hand | +0.039 |
rolled | +0.037 |
uttered | +0.036 |
chief | +0.035 |
Neg logit contributions | |
ey | -0.055 |
fru | -0.052 |
skirts | -0.050 |
da | -0.048 |
skirt | -0.048 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ver | +0.182 |
Even | +0.150 |
rolled | +0.144 |
hand | +0.144 |
Together | +0.143 |
Neg logit contributions | |
ey | -0.149 |
fru | -0.136 |
da | -0.124 |
ol | -0.123 |
jewelry | -0.116 |
Token | at |
Token ID | 379 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
ver | +0.045 |
hand | +0.040 |
chief | +0.039 |
Even | +0.038 |
rolled | +0.038 |
Neg logit contributions | |
ey | -0.043 |
skirts | -0.041 |
fru | -0.039 |
skirt | -0.038 |
makeup | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
ver | +0.004 |
hand | +0.004 |
chief | +0.004 |
happily | +0.003 |
rolled | +0.003 |
Neg logit contributions | |
ey | -0.005 |
skirts | -0.005 |
fru | -0.005 |
skirt | -0.004 |
ems | -0.004 |
Token | park |
Token ID | 3952 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
ey | +0.000 |
skirts | +0.000 |
fru | +0.000 |
prett | +0.000 |
skirt | +0.000 |
Neg logit contributions | |
ver | -0.001 |
uttered | -0.001 |
hand | -0.001 |
sadly | -0.001 |
rolled | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -9.57e-02 |
Pos logit contributions | |
ey | +0.016 |
skirts | +0.016 |
fru | +0.015 |
skirt | +0.014 |
makeup | +0.014 |
Neg logit contributions | |
ver | -0.018 |
hand | -0.016 |
chief | -0.016 |
department | -0.015 |
rolled | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
ey | +0.003 |
skirts | +0.003 |
fru | +0.003 |
skirt | +0.002 |
makeup | +0.002 |
Neg logit contributions | |
ver | -0.003 |
hand | -0.003 |
chief | -0.002 |
rolled | -0.002 |
department | -0.002 |
Token | very |
Token ID | 845 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
ver | +0.021 |
hand | +0.020 |
rolled | +0.020 |
uttered | +0.019 |
chief | +0.018 |
Neg logit contributions | |
fru | -0.012 |
skirts | -0.011 |
ey | -0.011 |
skirt | -0.011 |
prett | -0.010 |
Token | clever |
Token ID | 14169 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
fru | +0.007 |
ey | +0.007 |
skirts | +0.007 |
skirt | +0.006 |
makeup | +0.006 |
Neg logit contributions | |
ver | -0.010 |
hand | -0.010 |
rolled | -0.009 |
chief | -0.009 |
department | -0.009 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
ver | +0.012 |
rolled | +0.011 |
hand | +0.011 |
uttered | +0.010 |
alled | +0.010 |
Neg logit contributions | |
skirts | -0.008 |
skirt | -0.008 |
fru | -0.008 |
ey | -0.008 |
sunglasses | -0.007 |
Token | in |
Token ID | 259 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
rolled | +0.101 |
Even | +0.096 |
hand | +0.096 |
Together | +0.092 |
ver | +0.089 |
Neg logit contributions | |
fru | -0.076 |
skirts | -0.075 |
ey | -0.074 |
ems | -0.072 |
P | -0.071 |
Token | oc |
Token ID | 420 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
ver | +0.034 |
rolled | +0.032 |
hand | +0.032 |
uttered | +0.031 |
Even | +0.029 |
Neg logit contributions | |
fru | -0.039 |
skirts | -0.038 |
ey | -0.037 |
ems | -0.036 |
skirt | -0.035 |
Token | eros |
Token ID | 27498 |
Feature activation | +9.91e-01 |
Pos logit contributions | |
ver | +0.167 |
Even | +0.157 |
rolled | +0.149 |
case | +0.146 |
sadly | +0.145 |
Neg logit contributions | |
fru | -0.129 |
ey | -0.128 |
ol | -0.120 |
ution | -0.112 |
brace | -0.112 |
Token | ." |
Token ID | 526 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
ver | +0.037 |
rolled | +0.037 |
hand | +0.036 |
Even | +0.035 |
chief | +0.034 |
Neg logit contributions | |
skirts | -0.027 |
fru | -0.026 |
ems | -0.024 |
skirt | -0.024 |
ey | -0.023 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
ver | +0.026 |
chief | +0.024 |
hand | +0.023 |
rolled | +0.022 |
department | +0.022 |
Neg logit contributions | |
ey | -0.026 |
skirts | -0.026 |
fru | -0.025 |
skirt | -0.023 |
ol | -0.023 |
Token | Once |
Token ID | 7454 |
Feature activation | -8.59e-01 |
Pos logit contributions | |
ver | +0.029 |
hand | +0.027 |
chief | +0.027 |
department | +0.025 |
rolled | +0.025 |
Neg logit contributions | |
ey | -0.034 |
skirts | -0.033 |
fru | -0.032 |
skirt | -0.030 |
makeup | -0.029 |
Token | upon |
Token ID | 2402 |
Feature activation | -9.71e-01 |
Pos logit contributions | |
ey | +0.021 |
skirts | +0.021 |
fru | +0.020 |
skirt | +0.018 |
ems | +0.017 |
Neg logit contributions | |
ver | -0.037 |
hand | -0.035 |
chief | -0.035 |
rolled | -0.033 |
happily | -0.033 |
Token | Unicorn |
Token ID | 39481 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
ey | +0.001 |
skirts | +0.001 |
fru | +0.001 |
skirt | +0.001 |
makeup | +0.001 |
Neg logit contributions | |
ver | -0.001 |
hand | -0.001 |
rolled | -0.001 |
chief | -0.001 |
uttered | -0.001 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
ver | +0.030 |
hand | +0.029 |
chief | +0.028 |
rolled | +0.028 |
Even | +0.028 |
Neg logit contributions | |
skirts | -0.025 |
fru | -0.024 |
ey | -0.023 |
ems | -0.023 |
skirt | -0.023 |
Token | The |
Token ID | 383 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
ey | +0.001 |
skirts | +0.001 |
fru | +0.001 |
skirt | +0.001 |
makeup | +0.001 |
Neg logit contributions | |
ver | -0.001 |
hand | -0.001 |
chief | -0.001 |
department | -0.001 |
rolled | -0.001 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
ver | +0.025 |
hand | +0.023 |
chief | +0.022 |
rolled | +0.021 |
Even | +0.021 |
Neg logit contributions | |
ey | -0.019 |
skirts | -0.019 |
fru | -0.018 |
skirt | -0.017 |
ems | -0.016 |
Token | in |
Token ID | 259 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
hand | +0.097 |
ver | +0.097 |
rolled | +0.093 |
department | +0.091 |
Even | +0.091 |
Neg logit contributions | |
ey | -0.091 |
fru | -0.089 |
skirts | -0.081 |
da | -0.080 |
ol | -0.078 |
Token | oc |
Token ID | 420 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
ver | +0.053 |
hand | +0.047 |
department | +0.045 |
uttered | +0.044 |
chief | +0.043 |
Neg logit contributions | |
fru | -0.054 |
ey | -0.051 |
skirts | -0.048 |
scar | -0.047 |
da | -0.047 |
Token | eros |
Token ID | 27498 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
ver | +0.182 |
case | +0.152 |
Even | +0.149 |
tow | +0.142 |
rolled | +0.141 |
Neg logit contributions | |
ey | -0.151 |
fru | -0.143 |
ol | -0.128 |
da | -0.125 |
jewelry | -0.121 |
Token | said |
Token ID | 531 |
Feature activation | +8.45e-01 |
Pos logit contributions | |
ver | +0.064 |
hand | +0.062 |
rolled | +0.060 |
tow | +0.059 |
department | +0.059 |
Neg logit contributions | |
fru | -0.022 |
skirts | -0.019 |
ey | -0.018 |
da | -0.017 |
skirt | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
ver | +0.031 |
hand | +0.030 |
Even | +0.029 |
department | +0.029 |
chief | +0.029 |
Neg logit contributions | |
fru | -0.023 |
skirts | -0.021 |
skirt | -0.020 |
makeup | -0.019 |
brace | -0.018 |
Token | feeling |
Token ID | 4203 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
ver | +0.002 |
hand | +0.002 |
chief | +0.002 |
department | +0.002 |
rolled | +0.002 |
Neg logit contributions | |
ey | -0.002 |
fru | -0.002 |
skirts | -0.002 |
skirt | -0.001 |
makeup | -0.001 |
Token | much |
Token ID | 881 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
fru | +0.023 |
skirts | +0.022 |
ey | +0.021 |
skirt | +0.020 |
makeup | +0.019 |
Neg logit contributions | |
ver | -0.024 |
hand | -0.023 |
department | -0.022 |
rolled | -0.022 |
chief | -0.022 |
Token | It |
Token ID | 632 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
ey | +0.005 |
skirts | +0.004 |
fru | +0.004 |
skirt | +0.004 |
makeup | +0.004 |
Neg logit contributions | |
ver | -0.004 |
hand | -0.003 |
chief | -0.003 |
department | -0.003 |
rolled | -0.003 |
Token | is |
Token ID | 318 |
Feature activation | -7.98e-02 |
Pos logit contributions | |
ver | +0.026 |
hand | +0.024 |
chief | +0.023 |
rolled | +0.023 |
department | +0.022 |
Neg logit contributions | |
ey | -0.020 |
skirts | -0.019 |
fru | -0.019 |
skirt | -0.017 |
ems | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
ey | +0.003 |
skirts | +0.003 |
fru | +0.003 |
skirt | +0.003 |
ems | +0.003 |
Neg logit contributions | |
ver | -0.002 |
hand | -0.002 |
chief | -0.002 |
happily | -0.002 |
department | -0.002 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
ver | +0.010 |
hand | +0.009 |
chief | +0.009 |
rolled | +0.009 |
Even | +0.009 |
Neg logit contributions | |
ey | -0.009 |
skirts | -0.009 |
fru | -0.009 |
skirt | -0.008 |
makeup | -0.008 |
Token | in |
Token ID | 259 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
ver | +0.080 |
rolled | +0.079 |
hand | +0.078 |
Even | +0.076 |
Together | +0.072 |
Neg logit contributions | |
ey | -0.088 |
skirts | -0.083 |
fru | -0.081 |
skirt | -0.077 |
da | -0.076 |
Token | oc |
Token ID | 420 |
Feature activation | +4.59e+00 |
Pos logit contributions | |
ver | +0.014 |
hand | +0.013 |
rolled | +0.012 |
chief | +0.012 |
Even | +0.012 |
Neg logit contributions | |
ey | -0.020 |
fru | -0.019 |
skirts | -0.019 |
skirt | -0.018 |
da | -0.017 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
ver | +0.177 |
Even | +0.151 |
rolled | +0.143 |
happily | +0.143 |
sadly | +0.142 |
Neg logit contributions | |
ey | -0.150 |
fru | -0.140 |
ol | -0.122 |
da | -0.119 |
brace | -0.117 |
Token | . |
Token ID | 13 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
ver | +0.018 |
hand | +0.017 |
rolled | +0.016 |
chief | +0.016 |
Even | +0.016 |
Neg logit contributions | |
ey | -0.014 |
skirts | -0.014 |
fru | -0.013 |
skirt | -0.013 |
makeup | -0.012 |
Token | It |
Token ID | 632 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
ey | +0.008 |
skirts | +0.007 |
fru | +0.007 |
skirt | +0.007 |
makeup | +0.006 |
Neg logit contributions | |
ver | -0.006 |
hand | -0.006 |
chief | -0.006 |
department | -0.005 |
rolled | -0.005 |
Token | is |
Token ID | 318 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
ver | +0.031 |
hand | +0.028 |
chief | +0.027 |
rolled | +0.027 |
department | +0.026 |
Neg logit contributions | |
ey | -0.023 |
skirts | -0.022 |
fru | -0.022 |
skirt | -0.019 |
ems | -0.018 |
Token | bright |
Token ID | 6016 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
ey | +0.001 |
skirts | +0.001 |
fru | +0.001 |
skirt | +0.001 |
ems | +0.001 |
Neg logit contributions | |
ver | -0.001 |
hand | -0.000 |
chief | -0.000 |
happily | -0.000 |
sadly | -0.000 |
Token | she |
Token ID | 673 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
ey | +0.018 |
skirts | +0.017 |
fru | +0.017 |
skirt | +0.015 |
ems | +0.015 |
Neg logit contributions | |
ver | -0.021 |
hand | -0.019 |
chief | -0.019 |
happily | -0.018 |
department | -0.018 |
Token | could |
Token ID | 714 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
hand | +0.024 |
rolled | +0.023 |
chief | +0.023 |
ver | +0.022 |
attendant | +0.022 |
Neg logit contributions | |
fru | -0.016 |
skirts | -0.014 |
ey | -0.014 |
skirt | -0.013 |
makeup | -0.013 |
Token | request |
Token ID | 2581 |
Feature activation | -5.35e-02 |
Pos logit contributions | |
ver | +0.047 |
hand | +0.046 |
chief | +0.045 |
department | +0.044 |
uttered | +0.044 |
Neg logit contributions | |
ey | -0.029 |
makeup | -0.029 |
fru | -0.028 |
skirts | -0.028 |
wear | -0.026 |
Token | some |
Token ID | 617 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
skirts | +0.002 |
fru | +0.002 |
ey | +0.002 |
makeup | +0.002 |
ol | +0.002 |
Neg logit contributions | |
ver | -0.002 |
hand | -0.002 |
chief | -0.002 |
rolled | -0.002 |
department | -0.002 |
Token | av |
Token ID | 1196 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
ver | +0.026 |
hand | +0.024 |
chief | +0.023 |
happily | +0.023 |
Even | +0.022 |
Neg logit contributions | |
ey | -0.023 |
skirts | -0.022 |
fru | -0.022 |
skirt | -0.020 |
ems | -0.019 |
Token | oc |
Token ID | 420 |
Feature activation | +4.57e+00 |
Pos logit contributions | |
ver | +0.061 |
hand | +0.055 |
chief | +0.053 |
rolled | +0.050 |
department | +0.050 |
Neg logit contributions | |
ey | -0.114 |
skirts | -0.111 |
fru | -0.110 |
skirt | -0.102 |
ems | -0.100 |
Token | ados |
Token ID | 22484 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
ver | +0.244 |
hand | +0.224 |
Even | +0.220 |
tow | +0.218 |
rolled | +0.216 |
Neg logit contributions | |
ey | -0.069 |
fru | -0.066 |
ol | -0.058 |
skirts | -0.055 |
makeup | -0.048 |
Token | from |
Token ID | 422 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ver | +0.009 |
hand | +0.008 |
chief | +0.008 |
happily | +0.007 |
department | +0.007 |
Neg logit contributions | |
ey | -0.013 |
skirts | -0.012 |
fru | -0.012 |
skirt | -0.011 |
ems | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ver | +0.043 |
hand | +0.040 |
chief | +0.039 |
rolled | +0.037 |
department | +0.037 |
Neg logit contributions | |
ey | -0.034 |
skirts | -0.034 |
fru | -0.032 |
skirt | -0.030 |
makeup | -0.028 |
Token | store |
Token ID | 3650 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
ey | +0.007 |
skirts | +0.007 |
fru | +0.006 |
ems | +0.006 |
skirt | +0.006 |
Neg logit contributions | |
chief | -0.005 |
ver | -0.004 |
hand | -0.004 |
mayor | -0.004 |
department | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.80e-03 |
Pos logit contributions | |
skirts | +0.001 |
ey | +0.001 |
fru | +0.001 |
skirt | +0.001 |
makeup | +0.001 |
Neg logit contributions | |
ver | -0.002 |
hand | -0.001 |
chief | -0.001 |
rolled | -0.001 |
Even | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | -7.21e-01 |
Pos logit contributions | |
ey | +0.023 |
skirts | +0.023 |
fru | +0.022 |
skirt | +0.020 |
ems | +0.020 |
Neg logit contributions | |
ver | -0.018 |
hand | -0.017 |
chief | -0.016 |
rolled | -0.015 |
department | -0.015 |
Token | shiny |
Token ID | 22441 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
skirts | +0.023 |
ey | +0.023 |
fru | +0.021 |
ems | +0.021 |
--- | +0.019 |
Neg logit contributions | |
ver | -0.024 |
happily | -0.023 |
quickly | -0.023 |
sadly | -0.023 |
tow | -0.022 |
Token | thing |
Token ID | 1517 |
Feature activation | +6.55e-03 |
Pos logit contributions | |
skirts | +0.022 |
ey | +0.022 |
fru | +0.021 |
ems | +0.021 |
--- | +0.020 |
Neg logit contributions | |
happily | -0.028 |
sadly | -0.027 |
ver | -0.027 |
rolled | -0.027 |
quickly | -0.026 |
Token | in |
Token ID | 287 |
Feature activation | -5.74e-01 |
Pos logit contributions | |
ver | +0.000 |
hand | +0.000 |
chief | +0.000 |
happily | +0.000 |
rolled | +0.000 |
Neg logit contributions | |
ey | -0.000 |
skirts | -0.000 |
fru | -0.000 |
skirt | -0.000 |
ems | -0.000 |
Token | its |
Token ID | 663 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
ey | +0.021 |
skirts | +0.020 |
fru | +0.020 |
skirt | +0.018 |
ems | +0.018 |
Neg logit contributions | |
ver | -0.019 |
hand | -0.017 |
chief | -0.017 |
happily | -0.016 |
rolled | -0.016 |
Token | be |
Token ID | 307 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
skirts | +0.013 |
ey | +0.013 |
fru | +0.012 |
ems | +0.012 |
makeup | +0.011 |
Neg logit contributions | |
ver | -0.012 |
hand | -0.011 |
happily | -0.011 |
rolled | -0.011 |
safely | -0.011 |
Token | ak |
Token ID | 461 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
ey | +0.107 |
fru | +0.102 |
skirts | +0.102 |
skirt | +0.091 |
ems | +0.086 |
Neg logit contributions | |
ver | -0.147 |
hand | -0.132 |
happily | -0.132 |
chief | -0.129 |
safely | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | +2.59e-03 |
Pos logit contributions | |
ey | +0.061 |
skirts | +0.061 |
fru | +0.057 |
skirt | +0.054 |
ems | +0.051 |
Neg logit contributions | |
ver | -0.068 |
hand | -0.062 |
chief | -0.062 |
department | -0.059 |
Even | -0.059 |
Token | It |
Token ID | 632 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
ver | +0.000 |
hand | +0.000 |
chief | +0.000 |
Even | +0.000 |
rolled | +0.000 |
Neg logit contributions | |
ey | -0.000 |
skirts | -0.000 |
fru | -0.000 |
skirt | -0.000 |
ems | -0.000 |
Token | looked |
Token ID | 3114 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
ver | +0.024 |
hand | +0.022 |
chief | +0.021 |
department | +0.020 |
rolled | +0.020 |
Neg logit contributions | |
ey | -0.026 |
skirts | -0.025 |
fru | -0.025 |
skirt | -0.023 |
ems | -0.022 |
Token | like |
Token ID | 588 |
Feature activation | -6.02e-01 |
Pos logit contributions | |
ey | +0.009 |
skirts | +0.009 |
fru | +0.008 |
ems | +0.008 |
skirt | +0.008 |
Neg logit contributions | |
happily | -0.007 |
ver | -0.007 |
sadly | -0.007 |
quickly | -0.007 |
chief | -0.007 |
Token | piece |
Token ID | 3704 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ey | +0.012 |
skirts | +0.011 |
fru | +0.011 |
ems | +0.010 |
--- | +0.009 |
Neg logit contributions | |
ver | -0.015 |
chief | -0.014 |
hand | -0.014 |
happily | -0.014 |
sadly | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
ver | +0.024 |
hand | +0.021 |
chief | +0.020 |
rolled | +0.019 |
department | +0.018 |
Neg logit contributions | |
ey | -0.054 |
skirts | -0.054 |
fru | -0.051 |
skirt | -0.051 |
makeup | -0.049 |
Token | meat |
Token ID | 6174 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
ver | +0.005 |
happily | +0.005 |
hand | +0.005 |
tow | +0.005 |
quickly | +0.005 |
Neg logit contributions | |
ey | -0.005 |
skirts | -0.005 |
ems | -0.004 |
--- | -0.004 |
fru | -0.004 |
Token | in |
Token ID | 287 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
ver | +0.036 |
hand | +0.033 |
chief | +0.032 |
Even | +0.031 |
rolled | +0.030 |
Neg logit contributions | |
ey | -0.037 |
skirts | -0.037 |
fru | -0.035 |
skirt | -0.033 |
makeup | -0.032 |
Token | its |
Token ID | 663 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
ey | +0.028 |
skirts | +0.027 |
fru | +0.027 |
skirt | +0.025 |
ems | +0.025 |
Neg logit contributions | |
ver | -0.022 |
hand | -0.021 |
chief | -0.020 |
happily | -0.019 |
Even | -0.019 |
Token | be |
Token ID | 307 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
ey | +0.007 |
skirts | +0.007 |
fru | +0.007 |
ems | +0.006 |
skirt | +0.006 |
Neg logit contributions | |
ver | -0.008 |
hand | -0.007 |
chief | -0.007 |
happily | -0.007 |
Even | -0.007 |
Token | ak |
Token ID | 461 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
ey | +0.123 |
fru | +0.112 |
skirts | +0.111 |
skirt | +0.101 |
ems | +0.098 |
Neg logit contributions | |
ver | -0.130 |
happily | -0.122 |
safely | -0.115 |
quickly | -0.114 |
te | -0.114 |
Token | . |
Token ID | 13 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
ey | +0.070 |
skirts | +0.069 |
fru | +0.066 |
skirt | +0.061 |
ems | +0.059 |
Neg logit contributions | |
ver | -0.069 |
hand | -0.063 |
chief | -0.062 |
Even | -0.059 |
department | -0.058 |
Token | She |
Token ID | 1375 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
ey | +0.004 |
skirts | +0.004 |
fru | +0.004 |
skirt | +0.004 |
ems | +0.003 |
Neg logit contributions | |
ver | -0.004 |
hand | -0.003 |
chief | -0.003 |
rolled | -0.003 |
Even | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
ver | +0.013 |
hand | +0.013 |
chief | +0.012 |
department | +0.012 |
attendant | +0.012 |
Neg logit contributions | |
skirts | -0.008 |
fru | -0.008 |
ey | -0.007 |
skirt | -0.007 |
da | -0.007 |
Token | very |
Token ID | 845 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
ver | +0.015 |
hand | +0.015 |
department | +0.014 |
chief | +0.014 |
rolled | +0.013 |
Neg logit contributions | |
ey | -0.012 |
skirts | -0.012 |
fru | -0.012 |
skirt | -0.011 |
makeup | -0.010 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
ey | +0.018 |
fru | +0.017 |
skirts | +0.017 |
skirt | +0.016 |
makeup | +0.014 |
Neg logit contributions | |
ver | -0.029 |
hand | -0.027 |
chief | -0.026 |
rolled | -0.026 |
department | -0.025 |
Token | white |
Token ID | 2330 |
Feature activation | -5.53e-01 |
Pos logit contributions | |
ey | +0.013 |
skirts | +0.013 |
fru | +0.012 |
ems | +0.012 |
--- | +0.010 |
Neg logit contributions | |
ver | -0.015 |
happily | -0.015 |
hand | -0.015 |
sadly | -0.014 |
rolled | -0.014 |
Token | feather |
Token ID | 17972 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
ey | +0.014 |
skirts | +0.014 |
fru | +0.013 |
skirt | +0.011 |
ems | +0.011 |
Neg logit contributions | |
ver | -0.023 |
hand | -0.022 |
chief | -0.022 |
happily | -0.021 |
rolled | -0.021 |
Token | in |
Token ID | 287 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
skirts | +0.015 |
ey | +0.015 |
fru | +0.014 |
skirt | +0.014 |
makeup | +0.013 |
Neg logit contributions | |
ver | -0.014 |
hand | -0.013 |
chief | -0.013 |
Even | -0.013 |
rolled | -0.013 |
Token | its |
Token ID | 663 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
ey | +0.015 |
skirts | +0.015 |
fru | +0.014 |
ems | +0.013 |
skirt | +0.013 |
Neg logit contributions | |
ver | -0.016 |
hand | -0.015 |
chief | -0.014 |
happily | -0.014 |
Even | -0.014 |
Token | be |
Token ID | 307 |
Feature activation | -3.67e+00 |
Pos logit contributions | |
ey | +0.011 |
skirts | +0.011 |
fru | +0.011 |
ems | +0.010 |
skirt | +0.010 |
Neg logit contributions | |
ver | -0.011 |
hand | -0.011 |
happily | -0.011 |
chief | -0.010 |
Even | -0.010 |
Token | ak |
Token ID | 461 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
ey | +0.113 |
fru | +0.103 |
skirts | +0.102 |
skirt | +0.090 |
ems | +0.089 |
Neg logit contributions | |
ver | -0.141 |
happily | -0.131 |
hand | -0.129 |
safely | -0.125 |
rolled | -0.124 |
Token | . |
Token ID | 13 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
skirts | +0.063 |
ey | +0.062 |
fru | +0.059 |
skirt | +0.056 |
makeup | +0.053 |
Neg logit contributions | |
ver | -0.070 |
chief | -0.064 |
hand | -0.063 |
Even | -0.061 |
rolled | -0.060 |
Token |
|
Token ID | 198 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
ey | +0.009 |
skirts | +0.009 |
fru | +0.009 |
skirt | +0.008 |
makeup | +0.008 |
Neg logit contributions | |
ver | -0.006 |
chief | -0.006 |
hand | -0.006 |
department | -0.005 |
rolled | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -8.24e-02 |
Pos logit contributions | |
ey | +0.002 |
skirts | +0.001 |
fru | +0.001 |
skirt | +0.001 |
--- | +0.001 |
Neg logit contributions | |
ver | -0.001 |
chief | -0.001 |
department | -0.001 |
hand | -0.001 |
brigade | -0.001 |
Token | The |
Token ID | 464 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
ey | +0.003 |
skirts | +0.003 |
fru | +0.003 |
skirt | +0.002 |
makeup | +0.002 |
Neg logit contributions | |
ver | -0.003 |
hand | -0.003 |
chief | -0.003 |
department | -0.002 |
rolled | -0.002 |
Token | yellow |
Token ID | 7872 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
ey | +0.008 |
skirts | +0.008 |
fru | +0.007 |
ems | +0.007 |
skirt | +0.007 |
Neg logit contributions | |
ver | -0.008 |
chief | -0.008 |
hand | -0.008 |
happily | -0.008 |
sadly | -0.007 |
Token | corn |
Token ID | 11676 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
ey | +0.014 |
skirts | +0.013 |
fru | +0.013 |
skirt | +0.012 |
makeup | +0.011 |
Neg logit contributions | |
ver | -0.017 |
hand | -0.016 |
chief | -0.015 |
department | -0.015 |
rolled | -0.015 |
Token | cob |
Token ID | 22843 |
Feature activation | +8.65e-01 |
Pos logit contributions | |
ey | +0.008 |
skirts | +0.007 |
fru | +0.007 |
skirt | +0.007 |
ems | +0.006 |
Neg logit contributions | |
ver | -0.007 |
hand | -0.007 |
chief | -0.007 |
Even | -0.006 |
rolled | -0.006 |
Token | in |
Token ID | 287 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
ver | +0.034 |
hand | +0.031 |
chief | +0.030 |
Even | +0.030 |
rolled | +0.030 |
Neg logit contributions | |
ey | -0.026 |
skirts | -0.025 |
fru | -0.025 |
skirt | -0.022 |
ems | -0.021 |
Token | his |
Token ID | 465 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
ey | +0.024 |
skirts | +0.023 |
fru | +0.023 |
skirt | +0.021 |
makeup | +0.020 |
Neg logit contributions | |
ver | -0.029 |
hand | -0.026 |
chief | -0.026 |
department | -0.025 |
rolled | -0.024 |
Token | be |
Token ID | 307 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
ey | +0.018 |
skirts | +0.018 |
fru | +0.017 |
ems | +0.016 |
skirt | +0.016 |
Neg logit contributions | |
ver | -0.017 |
hand | -0.015 |
chief | -0.015 |
happily | -0.015 |
rolled | -0.015 |
Token | ak |
Token ID | 461 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
ey | +0.109 |
skirts | +0.102 |
fru | +0.101 |
skirt | +0.090 |
ems | +0.087 |
Neg logit contributions | |
ver | -0.142 |
happily | -0.132 |
hand | -0.130 |
safely | -0.126 |
quickly | -0.125 |
Token | ! |
Token ID | 0 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
skirts | +0.061 |
ey | +0.060 |
fru | +0.057 |
skirt | +0.054 |
makeup | +0.051 |
Neg logit contributions | |
ver | -0.077 |
hand | -0.070 |
chief | -0.070 |
Even | -0.067 |
rolled | -0.067 |
Token | He |
Token ID | 679 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
ey | +0.014 |
fru | +0.014 |
skirts | +0.013 |
ol | +0.012 |
makeup | +0.012 |
Neg logit contributions | |
ver | -0.014 |
hand | -0.013 |
chief | -0.013 |
department | -0.012 |
rolled | -0.012 |
Token | dropped |
Token ID | 5710 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
ver | +0.012 |
hand | +0.012 |
chief | +0.011 |
department | +0.011 |
mayor | +0.010 |
Neg logit contributions | |
fru | -0.009 |
skirts | -0.008 |
ey | -0.008 |
skirt | -0.007 |
da | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
ver | +0.018 |
hand | +0.015 |
chief | +0.015 |
rolled | +0.015 |
case | +0.015 |
Neg logit contributions | |
ey | -0.011 |
fru | -0.010 |
skirts | -0.010 |
ol | -0.009 |
jam | -0.009 |
Token | bright |
Token ID | 6016 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
ey | +0.014 |
skirts | +0.013 |
fru | +0.013 |
skirt | +0.012 |
makeup | +0.011 |
Neg logit contributions | |
ver | -0.018 |
hand | -0.017 |
chief | -0.017 |
rolled | -0.016 |
Even | -0.016 |
Token | green |
Token ID | 4077 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
skirts | +0.007 |
ey | +0.007 |
fru | +0.007 |
ems | +0.006 |
--- | +0.006 |
Neg logit contributions | |
ver | -0.009 |
happily | -0.009 |
hand | -0.008 |
sadly | -0.008 |
chief | -0.008 |
Token | olive |
Token ID | 19450 |
Feature activation | -8.85e-02 |
Pos logit contributions | |
ey | +0.010 |
skirts | +0.010 |
fru | +0.009 |
skirt | +0.008 |
ems | +0.008 |
Neg logit contributions | |
ver | -0.015 |
hand | -0.014 |
chief | -0.013 |
happily | -0.013 |
rolled | -0.013 |
Token | in |
Token ID | 287 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
ey | +0.003 |
skirts | +0.003 |
fru | +0.003 |
skirt | +0.003 |
makeup | +0.003 |
Neg logit contributions | |
ver | -0.003 |
hand | -0.003 |
chief | -0.003 |
rolled | -0.003 |
Even | -0.003 |
Token | its |
Token ID | 663 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
ey | +0.015 |
skirts | +0.015 |
fru | +0.014 |
ems | +0.013 |
skirt | +0.013 |
Neg logit contributions | |
ver | -0.014 |
hand | -0.014 |
happily | -0.013 |
chief | -0.013 |
Even | -0.013 |
Token | be |
Token ID | 307 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
skirts | +0.008 |
ey | +0.008 |
fru | +0.008 |
ems | +0.007 |
skirt | +0.007 |
Neg logit contributions | |
ver | -0.009 |
hand | -0.008 |
happily | -0.008 |
chief | -0.008 |
Even | -0.008 |
Token | ak |
Token ID | 461 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
ey | +0.109 |
fru | +0.101 |
skirts | +0.101 |
skirt | +0.089 |
ems | +0.087 |
Neg logit contributions | |
ver | -0.141 |
happily | -0.129 |
hand | -0.127 |
safely | -0.124 |
chief | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | -9.04e-02 |
Pos logit contributions | |
skirts | +0.063 |
ey | +0.062 |
fru | +0.060 |
skirt | +0.056 |
makeup | +0.054 |
Neg logit contributions | |
ver | -0.067 |
hand | -0.061 |
chief | -0.061 |
Even | -0.059 |
rolled | -0.058 |
Token | It |
Token ID | 632 |
Feature activation | +8.45e-01 |
Pos logit contributions | |
ey | +0.003 |
fru | +0.003 |
skirts | +0.003 |
skirt | +0.003 |
makeup | +0.003 |
Neg logit contributions | |
ver | -0.003 |
chief | -0.003 |
hand | -0.003 |
department | -0.003 |
te | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
ver | +0.030 |
hand | +0.027 |
chief | +0.026 |
rolled | +0.025 |
department | +0.025 |
Neg logit contributions | |
ey | -0.028 |
fru | -0.028 |
skirts | -0.027 |
skirt | -0.025 |
ems | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
ver | +0.006 |
hand | +0.006 |
chief | +0.006 |
rolled | +0.005 |
happily | +0.005 |
Neg logit contributions | |
ey | -0.007 |
skirts | -0.007 |
fru | -0.007 |
skirt | -0.006 |
ems | -0.006 |
Token | had |
Token ID | 550 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
ver | +0.020 |
hand | +0.019 |
chief | +0.018 |
rolled | +0.017 |
Even | +0.017 |
Neg logit contributions | |
ey | -0.025 |
skirts | -0.025 |
fru | -0.025 |
skirt | -0.023 |
ems | -0.022 |
Token | Pip |
Token ID | 25149 |
Feature activation | -9.43e-01 |
Pos logit contributions | |
ey | +0.026 |
skirts | +0.026 |
fru | +0.025 |
skirt | +0.023 |
ems | +0.022 |
Neg logit contributions | |
ver | -0.023 |
hand | -0.021 |
chief | -0.020 |
rolled | -0.019 |
happily | -0.019 |
Token | o |
Token ID | 78 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ey | +0.037 |
skirts | +0.037 |
fru | +0.036 |
skirt | +0.033 |
ems | +0.032 |
Neg logit contributions | |
ver | -0.027 |
hand | -0.025 |
chief | -0.024 |
rolled | -0.023 |
Even | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | -8.82e-01 |
Pos logit contributions | |
skirts | +0.043 |
ey | +0.041 |
fru | +0.040 |
skirt | +0.038 |
ems | +0.037 |
Neg logit contributions | |
ver | -0.045 |
hand | -0.041 |
chief | -0.041 |
Even | -0.039 |
department | -0.039 |
Token | its |
Token ID | 663 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
ey | +0.032 |
skirts | +0.031 |
fru | +0.031 |
skirt | +0.028 |
ems | +0.028 |
Neg logit contributions | |
ver | -0.028 |
hand | -0.026 |
chief | -0.025 |
happily | -0.025 |
Even | -0.024 |
Token | be |
Token ID | 307 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
skirts | +0.027 |
ey | +0.027 |
ems | +0.026 |
fru | +0.026 |
alsa | +0.024 |
Neg logit contributions | |
happily | -0.021 |
sadly | -0.020 |
tow | -0.020 |
rolled | -0.020 |
safely | -0.019 |
Token | ak |
Token ID | 461 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
ey | +0.111 |
skirts | +0.104 |
fru | +0.103 |
skirt | +0.093 |
ems | +0.090 |
Neg logit contributions | |
ver | -0.136 |
happily | -0.125 |
hand | -0.122 |
rolled | -0.120 |
safely | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
skirts | +0.062 |
ey | +0.061 |
fru | +0.057 |
skirt | +0.056 |
makeup | +0.053 |
Neg logit contributions | |
ver | -0.068 |
hand | -0.061 |
chief | -0.060 |
Even | -0.058 |
department | -0.057 |
Token | It |
Token ID | 632 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
ey | +0.005 |
skirts | +0.005 |
fru | +0.005 |
skirt | +0.005 |
ems | +0.005 |
Neg logit contributions | |
ver | -0.003 |
hand | -0.003 |
chief | -0.003 |
Even | -0.003 |
happily | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
ver | +0.024 |
hand | +0.022 |
chief | +0.021 |
rolled | +0.020 |
department | +0.020 |
Neg logit contributions | |
ey | -0.026 |
skirts | -0.025 |
fru | -0.025 |
skirt | -0.023 |
ems | -0.022 |
Token | flying |
Token ID | 7348 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
ver | +0.002 |
happily | +0.002 |
chief | +0.002 |
sadly | +0.002 |
hand | +0.002 |
Neg logit contributions | |
ey | -0.003 |
skirts | -0.003 |
fru | -0.003 |
skirt | -0.003 |
ems | -0.003 |
Token | red |
Token ID | 2266 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
ey | +0.044 |
skirts | +0.039 |
fru | +0.036 |
ems | +0.035 |
skirt | +0.035 |
Neg logit contributions | |
ver | -0.045 |
happily | -0.042 |
quickly | -0.042 |
sadly | -0.040 |
Even | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
skirts | +0.017 |
ey | +0.016 |
fru | +0.015 |
skirt | +0.015 |
makeup | +0.014 |
Neg logit contributions | |
ver | -0.021 |
rolled | -0.019 |
hand | -0.019 |
chief | -0.019 |
brigade | -0.017 |
Token | had |
Token ID | 550 |
Feature activation | -7.26e-01 |
Pos logit contributions | |
ey | +0.038 |
skirts | +0.035 |
fru | +0.033 |
skirt | +0.032 |
ems | +0.031 |
Neg logit contributions | |
ver | -0.038 |
happily | -0.034 |
chief | -0.033 |
hand | -0.033 |
quickly | -0.033 |
Token | a |
Token ID | 257 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
ey | +0.028 |
skirts | +0.027 |
fru | +0.027 |
skirt | +0.025 |
ems | +0.024 |
Neg logit contributions | |
ver | -0.022 |
hand | -0.020 |
chief | -0.019 |
rolled | -0.018 |
happily | -0.018 |
Token | yellow |
Token ID | 7872 |
Feature activation | -8.49e-01 |
Pos logit contributions | |
ey | +0.020 |
skirts | +0.019 |
fru | +0.018 |
ems | +0.017 |
skirt | +0.016 |
Neg logit contributions | |
ver | -0.027 |
happily | -0.025 |
hand | -0.025 |
chief | -0.025 |
Even | -0.025 |
Token | be |
Token ID | 307 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
ey | +0.024 |
skirts | +0.023 |
fru | +0.022 |
ems | +0.020 |
skirt | +0.019 |
Neg logit contributions | |
ver | -0.032 |
hand | -0.031 |
happily | -0.031 |
Even | -0.030 |
rolled | -0.030 |
Token | ak |
Token ID | 461 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
ey | +0.108 |
fru | +0.099 |
skirts | +0.098 |
skirt | +0.088 |
ems | +0.084 |
Neg logit contributions | |
ver | -0.142 |
hand | -0.129 |
happily | -0.127 |
chief | -0.126 |
rolled | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
skirts | +0.053 |
ey | +0.052 |
fru | +0.050 |
skirt | +0.047 |
makeup | +0.045 |
Neg logit contributions | |
ver | -0.057 |
hand | -0.052 |
chief | -0.051 |
Even | -0.050 |
rolled | -0.049 |
Token | He |
Token ID | 679 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
ey | +0.014 |
skirts | +0.014 |
fru | +0.014 |
eking | +0.013 |
skirt | +0.013 |
Neg logit contributions | |
ver | -0.009 |
Even | -0.009 |
happily | -0.009 |
hand | -0.008 |
Together | -0.008 |
Token | could |
Token ID | 714 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
ver | +0.002 |
hand | +0.002 |
chief | +0.002 |
department | +0.002 |
rolled | +0.002 |
Neg logit contributions | |
skirts | -0.001 |
fru | -0.001 |
ey | -0.001 |
skirt | -0.001 |
da | -0.001 |
Token | talk |
Token ID | 1561 |
Feature activation | +8.28e-01 |
Pos logit contributions | |
ey | +0.004 |
skirts | +0.004 |
fru | +0.003 |
skirt | +0.003 |
makeup | +0.003 |
Neg logit contributions | |
ver | -0.004 |
hand | -0.004 |
chief | -0.004 |
department | -0.004 |
rolled | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
ey | +0.003 |
skirts | +0.003 |
fru | +0.003 |
skirt | +0.002 |
ems | +0.002 |
Neg logit contributions | |
ver | -0.002 |
hand | -0.002 |
chief | -0.002 |
happily | -0.002 |
sadly | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
ey | +0.029 |
skirts | +0.028 |
fru | +0.028 |
skirt | +0.026 |
ems | +0.025 |
Neg logit contributions | |
ver | -0.019 |
hand | -0.017 |
chief | -0.017 |
happily | -0.016 |
rolled | -0.016 |
Token | feather |
Token ID | 17972 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
ey | +0.025 |
skirts | +0.024 |
fru | +0.023 |
ems | +0.021 |
skirt | +0.020 |
Neg logit contributions | |
ver | -0.030 |
hand | -0.028 |
chief | -0.028 |
happily | -0.027 |
sadly | -0.027 |
Token | in |
Token ID | 287 |
Feature activation | -8.35e-01 |
Pos logit contributions | |
ey | +0.018 |
skirts | +0.017 |
fru | +0.017 |
skirt | +0.016 |
ems | +0.015 |
Neg logit contributions | |
ver | -0.015 |
hand | -0.014 |
chief | -0.013 |
Even | -0.013 |
rolled | -0.013 |
Token | its |
Token ID | 663 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
ey | +0.030 |
skirts | +0.029 |
fru | +0.029 |
skirt | +0.026 |
ems | +0.026 |
Neg logit contributions | |
ver | -0.027 |
hand | -0.025 |
chief | -0.024 |
happily | -0.024 |
Even | -0.023 |
Token | be |
Token ID | 307 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
skirts | +0.019 |
ey | +0.019 |
fru | +0.018 |
ems | +0.017 |
--- | +0.016 |
Neg logit contributions | |
happily | -0.015 |
ver | -0.015 |
sadly | -0.014 |
Even | -0.014 |
safely | -0.014 |
Token | ak |
Token ID | 461 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ey | +0.117 |
fru | +0.108 |
skirts | +0.105 |
ems | +0.094 |
skirt | +0.094 |
Neg logit contributions | |
ver | -0.125 |
happily | -0.119 |
safely | -0.112 |
rolled | -0.110 |
quickly | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
skirts | +0.061 |
ey | +0.060 |
fru | +0.057 |
skirt | +0.055 |
makeup | +0.052 |
Neg logit contributions | |
ver | -0.066 |
hand | -0.060 |
chief | -0.060 |
Even | -0.056 |
rolled | -0.056 |
Token | It |
Token ID | 632 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
ey | +0.008 |
skirts | +0.008 |
fru | +0.008 |
skirt | +0.007 |
ems | +0.007 |
Neg logit contributions | |
ver | -0.005 |
hand | -0.004 |
chief | -0.004 |
Even | -0.004 |
happily | -0.004 |
Token | flew |
Token ID | 13112 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
ver | +0.028 |
hand | +0.026 |
chief | +0.025 |
rolled | +0.024 |
department | +0.024 |
Neg logit contributions | |
ey | -0.026 |
skirts | -0.025 |
fru | -0.025 |
skirt | -0.023 |
ems | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
ver | +0.007 |
chief | +0.007 |
hand | +0.006 |
Even | +0.006 |
itt | +0.006 |
Neg logit contributions | |
skirts | -0.008 |
ey | -0.007 |
fru | -0.007 |
skirt | -0.007 |
prett | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -9.83e-01 |
Pos logit contributions | |
ey | +0.012 |
skirts | +0.012 |
fru | +0.011 |
skirt | +0.010 |
makeup | +0.010 |
Neg logit contributions | |
ver | -0.011 |
hand | -0.009 |
chief | -0.009 |
rolled | -0.009 |
department | -0.009 |
Token | white |
Token ID | 2330 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
ey | +0.031 |
skirts | +0.029 |
ems | +0.028 |
skirt | +0.026 |
reath | +0.026 |
Neg logit contributions | |
quickly | -0.034 |
happily | -0.033 |
sadly | -0.032 |
ver | -0.032 |
chief | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -9.95e-01 |
Pos logit contributions | |
ey | +0.005 |
skirts | +0.005 |
fru | +0.005 |
skirt | +0.005 |
makeup | +0.004 |
Neg logit contributions | |
ver | -0.005 |
hand | -0.005 |
chief | -0.005 |
rolled | -0.005 |
department | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
ey | +0.036 |
skirts | +0.034 |
ems | +0.031 |
skirt | +0.030 |
fru | +0.030 |
Neg logit contributions | |
ver | -0.031 |
quickly | -0.030 |
happily | -0.030 |
chief | -0.029 |
sadly | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
ey | +0.022 |
skirts | +0.022 |
fru | +0.021 |
skirt | +0.019 |
makeup | +0.018 |
Neg logit contributions | |
ver | -0.027 |
hand | -0.025 |
chief | -0.024 |
rolled | -0.023 |
sadly | -0.023 |
Token | be |
Token ID | 307 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
ey | +0.018 |
skirts | +0.017 |
fru | +0.017 |
skirt | +0.015 |
makeup | +0.014 |
Neg logit contributions | |
ver | -0.025 |
hand | -0.023 |
chief | -0.023 |
rolled | -0.022 |
happily | -0.022 |
Token | ak |
Token ID | 461 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ey | +0.103 |
skirts | +0.097 |
fru | +0.096 |
skirt | +0.087 |
ems | +0.083 |
Neg logit contributions | |
ver | -0.141 |
hand | -0.128 |
happily | -0.128 |
chief | -0.126 |
rolled | -0.125 |
Token | and |
Token ID | 290 |
Feature activation | -8.72e-01 |
Pos logit contributions | |
ey | +0.042 |
skirts | +0.042 |
fru | +0.040 |
skirt | +0.037 |
makeup | +0.036 |
Neg logit contributions | |
ver | -0.046 |
hand | -0.042 |
chief | -0.041 |
Even | -0.040 |
rolled | -0.039 |
Token | fl |
Token ID | 781 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
ey | +0.032 |
skirts | +0.031 |
fru | +0.030 |
skirt | +0.028 |
makeup | +0.027 |
Neg logit contributions | |
ver | -0.029 |
hand | -0.026 |
chief | -0.025 |
rolled | -0.024 |
department | -0.024 |
Token | ippers |
Token ID | 16415 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
ey | +0.024 |
skirts | +0.021 |
fru | +0.021 |
eking | +0.020 |
ems | +0.019 |
Neg logit contributions | |
hand | -0.021 |
chief | -0.020 |
ver | -0.019 |
mayor | -0.019 |
department | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
skirts | +0.025 |
ey | +0.025 |
fru | +0.024 |
skirt | +0.022 |
makeup | +0.021 |
Neg logit contributions | |
ver | -0.027 |
hand | -0.025 |
chief | -0.025 |
department | -0.024 |
Even | -0.024 |
Token | it |
Token ID | 340 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
ey | +0.010 |
skirts | +0.010 |
fru | +0.010 |
skirt | +0.009 |
da | +0.009 |
Neg logit contributions | |
ver | -0.012 |
hand | -0.011 |
chief | -0.011 |
rolled | -0.011 |
department | -0.011 |
Token | had |
Token ID | 550 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
ey | +0.026 |
fru | +0.025 |
skirts | +0.025 |
skirt | +0.023 |
ems | +0.022 |
Neg logit contributions | |
ver | -0.026 |
hand | -0.024 |
chief | -0.023 |
rolled | -0.022 |
department | -0.022 |
Token | something |
Token ID | 1223 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
ey | +0.039 |
skirts | +0.039 |
fru | +0.038 |
skirt | +0.035 |
makeup | +0.033 |
Neg logit contributions | |
ver | -0.036 |
hand | -0.032 |
chief | -0.032 |
rolled | -0.030 |
department | -0.030 |
Token | in |
Token ID | 287 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
ey | +0.012 |
skirts | +0.011 |
fru | +0.011 |
ems | +0.010 |
skirt | +0.010 |
Neg logit contributions | |
ver | -0.011 |
hand | -0.010 |
chief | -0.010 |
happily | -0.010 |
sadly | -0.010 |
Token | its |
Token ID | 663 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
ey | +0.026 |
skirts | +0.026 |
fru | +0.025 |
skirt | +0.023 |
ems | +0.022 |
Neg logit contributions | |
ver | -0.022 |
hand | -0.020 |
chief | -0.020 |
happily | -0.019 |
Even | -0.019 |
Token | be |
Token ID | 307 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
skirts | +0.001 |
ey | +0.001 |
fru | +0.001 |
ems | +0.001 |
skirt | +0.001 |
Neg logit contributions | |
ver | -0.001 |
happily | -0.000 |
Even | -0.000 |
sadly | -0.000 |
hand | -0.000 |
Token | ak |
Token ID | 461 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
ey | +0.107 |
skirts | +0.102 |
fru | +0.101 |
skirt | +0.090 |
ems | +0.086 |
Neg logit contributions | |
ver | -0.135 |
happily | -0.123 |
hand | -0.122 |
safely | -0.119 |
chief | -0.117 |
Token | . |
Token ID | 13 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
ey | +0.060 |
skirts | +0.059 |
fru | +0.057 |
skirt | +0.053 |
ems | +0.051 |
Neg logit contributions | |
ver | -0.060 |
hand | -0.055 |
chief | -0.054 |
Even | -0.052 |
rolled | -0.052 |
Token | |
Token ID | 220 |
Feature activation | +8.79e-02 |
Pos logit contributions | |
ey | +0.010 |
fru | +0.010 |
skirts | +0.009 |
ol | +0.008 |
--- | +0.008 |
Neg logit contributions | |
ver | -0.012 |
department | -0.011 |
chief | -0.011 |
hand | -0.011 |
te | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
chief | +0.003 |
hand | +0.003 |
department | +0.003 |
mayor | +0.003 |
ver | +0.003 |
Neg logit contributions | |
ey | -0.003 |
fru | -0.002 |
skirts | -0.002 |
--- | -0.002 |
ems | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
ey | +0.005 |
fru | +0.005 |
skirts | +0.005 |
--- | +0.004 |
ems | +0.004 |
Neg logit contributions | |
department | -0.005 |
chief | -0.005 |
ver | -0.005 |
hand | -0.005 |
brigade | -0.005 |
Token | holding |
Token ID | 4769 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
ver | +0.017 |
hand | +0.016 |
happily | +0.015 |
chief | +0.015 |
sadly | +0.015 |
Neg logit contributions | |
ey | -0.021 |
skirts | -0.021 |
fru | -0.021 |
skirt | -0.018 |
ems | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
fru | +0.016 |
skirts | +0.016 |
makeup | +0.015 |
ey | +0.015 |
skirt | +0.015 |
Neg logit contributions | |
ver | -0.026 |
department | -0.025 |
chief | -0.024 |
hand | -0.023 |
brigade | -0.023 |
Token | carrot |
Token ID | 40562 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
ey | +0.018 |
skirts | +0.018 |
fru | +0.016 |
ems | +0.016 |
--- | +0.015 |
Neg logit contributions | |
ver | -0.018 |
hand | -0.017 |
happily | -0.017 |
chief | -0.017 |
sadly | -0.017 |
Token | in |
Token ID | 287 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
ver | +0.006 |
hand | +0.006 |
chief | +0.006 |
Even | +0.006 |
rolled | +0.006 |
Neg logit contributions | |
skirts | -0.005 |
ey | -0.005 |
skirt | -0.005 |
fru | -0.005 |
makeup | -0.004 |
Token | its |
Token ID | 663 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
ey | +0.017 |
skirts | +0.016 |
fru | +0.016 |
skirt | +0.015 |
ems | +0.014 |
Neg logit contributions | |
ver | -0.016 |
hand | -0.015 |
chief | -0.015 |
rolled | -0.014 |
happily | -0.014 |
Token | be |
Token ID | 307 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
skirts | +0.006 |
ey | +0.006 |
fru | +0.005 |
ems | +0.005 |
skirt | +0.005 |
Neg logit contributions | |
ver | -0.005 |
happily | -0.005 |
sadly | -0.005 |
hand | -0.005 |
Even | -0.005 |
Token | ak |
Token ID | 461 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
ey | +0.118 |
fru | +0.110 |
skirts | +0.110 |
ems | +0.099 |
skirt | +0.098 |
Neg logit contributions | |
ver | -0.123 |
happily | -0.117 |
safely | -0.112 |
quickly | -0.110 |
hand | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
skirts | +0.058 |
ey | +0.057 |
fru | +0.055 |
skirt | +0.052 |
makeup | +0.049 |
Neg logit contributions | |
ver | -0.066 |
hand | -0.061 |
chief | -0.061 |
Even | -0.058 |
department | -0.057 |
Token | Benny |
Token ID | 44275 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
fru | +0.011 |
skirts | +0.010 |
ey | +0.010 |
skirt | +0.009 |
makeup | +0.009 |
Neg logit contributions | |
ver | -0.010 |
hand | -0.010 |
chief | -0.009 |
te | -0.009 |
mayor | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
fru | +0.012 |
skirts | +0.011 |
ey | +0.011 |
skirt | +0.010 |
da | +0.010 |
Neg logit contributions | |
ver | -0.019 |
hand | -0.018 |
chief | -0.017 |
department | -0.017 |
tow | -0.017 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
ey | +0.033 |
fru | +0.030 |
skirts | +0.030 |
eking | +0.029 |
reath | +0.028 |
Neg logit contributions | |
happily | -0.020 |
sadly | -0.019 |
quickly | -0.019 |
Even | -0.018 |
ver | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
ver | +0.023 |
hand | +0.021 |
chief | +0.021 |
department | +0.020 |
rolled | +0.019 |
Neg logit contributions | |
ey | -0.031 |
fru | -0.030 |
skirts | -0.030 |
skirt | -0.027 |
makeup | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
ver | +0.004 |
chief | +0.004 |
hand | +0.004 |
department | +0.004 |
brigade | +0.004 |
Neg logit contributions | |
ey | -0.005 |
skirts | -0.005 |
fru | -0.005 |
--- | -0.004 |
skirt | -0.004 |
Token | It |
Token ID | 1026 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
ver | +0.001 |
hand | +0.001 |
chief | +0.001 |
happily | +0.001 |
rolled | +0.001 |
Neg logit contributions | |
ey | -0.001 |
skirts | -0.001 |
fru | -0.001 |
skirt | -0.001 |
ems | -0.001 |
Token | put |
Token ID | 1234 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
ver | +0.014 |
hand | +0.013 |
mayor | +0.012 |
chief | +0.012 |
itt | +0.012 |
Neg logit contributions | |
fru | -0.011 |
skirts | -0.010 |
ey | -0.009 |
prett | -0.009 |
skirt | -0.009 |
Token | its |
Token ID | 663 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
ey | +0.027 |
skirts | +0.026 |
fru | +0.025 |
skirt | +0.024 |
ems | +0.023 |
Neg logit contributions | |
ver | -0.021 |
hand | -0.019 |
chief | -0.019 |
happily | -0.018 |
rolled | -0.018 |
Token | be |
Token ID | 307 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
skirts | +0.005 |
ey | +0.005 |
fru | +0.004 |
ems | +0.004 |
skirt | +0.004 |
Neg logit contributions | |
ver | -0.005 |
chief | -0.005 |
happily | -0.005 |
hand | -0.004 |
safely | -0.004 |
Token | ak |
Token ID | 461 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
ey | +0.111 |
fru | +0.102 |
skirts | +0.100 |
skirt | +0.090 |
ems | +0.090 |
Neg logit contributions | |
ver | -0.135 |
happily | -0.126 |
safely | -0.123 |
hand | -0.122 |
quickly | -0.120 |
Token | into |
Token ID | 656 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
skirts | +0.061 |
ey | +0.056 |
fru | +0.055 |
skirt | +0.054 |
makeup | +0.052 |
Neg logit contributions | |
ver | -0.060 |
Even | -0.057 |
chief | -0.057 |
hand | -0.056 |
mayor | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
ey | +0.018 |
skirts | +0.018 |
fru | +0.017 |
skirt | +0.016 |
ems | +0.016 |
Neg logit contributions | |
ver | -0.012 |
hand | -0.011 |
chief | -0.011 |
happily | -0.011 |
Even | -0.010 |
Token | crack |
Token ID | 8469 |
Feature activation | -9.05e-01 |
Pos logit contributions | |
skirts | +0.011 |
ey | +0.010 |
ems | +0.010 |
fru | +0.009 |
skirt | +0.009 |
Neg logit contributions | |
chief | -0.008 |
hand | -0.008 |
happily | -0.007 |
safely | -0.007 |
ver | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +8.01e-02 |
Pos logit contributions | |
ey | +0.033 |
skirts | +0.032 |
fru | +0.031 |
skirt | +0.029 |
makeup | +0.027 |
Neg logit contributions | |
ver | -0.030 |
hand | -0.027 |
chief | -0.026 |
rolled | -0.026 |
Even | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
ey | +0.024 |
skirts | +0.024 |
fru | +0.024 |
skirt | +0.022 |
makeup | +0.021 |
Neg logit contributions | |
ver | -0.017 |
hand | -0.016 |
chief | -0.015 |
department | -0.014 |
rolled | -0.014 |
Token | little |
Token ID | 1310 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
ey | +0.028 |
skirts | +0.028 |
fru | +0.026 |
ems | +0.024 |
skirt | +0.023 |
Neg logit contributions | |
ver | -0.030 |
chief | -0.027 |
happily | -0.027 |
hand | -0.027 |
sadly | -0.027 |
Token | envelope |
Token ID | 22878 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
ey | +0.015 |
skirts | +0.014 |
fru | +0.014 |
skirt | +0.013 |
makeup | +0.012 |
Neg logit contributions | |
ver | -0.018 |
hand | -0.016 |
chief | -0.016 |
rolled | -0.015 |
Even | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
ver | +0.014 |
hand | +0.013 |
chief | +0.012 |
rolled | +0.012 |
Even | +0.012 |
Neg logit contributions | |
ey | -0.017 |
skirts | -0.017 |
fru | -0.016 |
skirt | -0.015 |
ems | -0.015 |
Token | its |
Token ID | 663 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
ey | +0.030 |
skirts | +0.029 |
fru | +0.028 |
skirt | +0.026 |
ems | +0.025 |
Neg logit contributions | |
ver | -0.026 |
hand | -0.024 |
chief | -0.023 |
happily | -0.022 |
rolled | -0.022 |
Token | be |
Token ID | 307 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
ey | +0.017 |
skirts | +0.017 |
fru | +0.016 |
ems | +0.015 |
skirt | +0.015 |
Neg logit contributions | |
ver | -0.014 |
happily | -0.013 |
sadly | -0.013 |
hand | -0.013 |
Even | -0.013 |
Token | ak |
Token ID | 461 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
ey | +0.115 |
skirts | +0.105 |
fru | +0.104 |
skirt | +0.095 |
ems | +0.091 |
Neg logit contributions | |
ver | -0.131 |
happily | -0.120 |
rolled | -0.115 |
hand | -0.114 |
safely | -0.114 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
skirts | +0.065 |
ey | +0.065 |
fru | +0.062 |
skirt | +0.058 |
makeup | +0.056 |
Neg logit contributions | |
ver | -0.065 |
hand | -0.060 |
chief | -0.059 |
Even | -0.056 |
department | -0.056 |
Token | Tim |
Token ID | 5045 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
ey | +0.008 |
fru | +0.008 |
skirts | +0.008 |
ol | +0.007 |
makeup | +0.007 |
Neg logit contributions | |
ver | -0.007 |
hand | -0.007 |
chief | -0.006 |
rolled | -0.006 |
happily | -0.006 |
Token | opened |
Token ID | 4721 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
fru | +0.018 |
skirts | +0.018 |
ey | +0.018 |
skirt | +0.016 |
da | +0.016 |
Neg logit contributions | |
hand | -0.026 |
ver | -0.025 |
chief | -0.023 |
tow | -0.023 |
department | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
ey | +0.017 |
skirts | +0.016 |
fru | +0.015 |
skirt | +0.014 |
da | +0.013 |
Neg logit contributions | |
ver | -0.026 |
hand | -0.024 |
chief | -0.024 |
department | -0.024 |
brigade | -0.023 |
Token | dropped |
Token ID | 5710 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
ver | +0.031 |
hand | +0.028 |
chief | +0.028 |
happily | +0.028 |
sadly | +0.027 |
Neg logit contributions | |
ey | -0.034 |
skirts | -0.033 |
fru | -0.032 |
skirt | -0.029 |
ems | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
ver | +0.014 |
rolled | +0.013 |
hand | +0.012 |
chief | +0.012 |
case | +0.012 |
Neg logit contributions | |
ey | -0.007 |
skirts | -0.007 |
fru | -0.007 |
ol | -0.006 |
skirt | -0.006 |
Token | car |
Token ID | 1097 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
ey | +0.016 |
skirts | +0.016 |
fru | +0.015 |
skirt | +0.014 |
makeup | +0.013 |
Neg logit contributions | |
ver | -0.016 |
hand | -0.015 |
chief | -0.014 |
Even | -0.014 |
rolled | -0.014 |
Token | from |
Token ID | 422 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
ver | +0.019 |
hand | +0.017 |
chief | +0.017 |
rolled | +0.017 |
Even | +0.017 |
Neg logit contributions | |
skirts | -0.015 |
ey | -0.015 |
skirt | -0.013 |
fru | -0.013 |
makeup | -0.013 |
Token | its |
Token ID | 663 |
Feature activation | -8.27e-02 |
Pos logit contributions | |
ver | +0.026 |
hand | +0.023 |
chief | +0.022 |
rolled | +0.022 |
department | +0.022 |
Neg logit contributions | |
ey | -0.023 |
skirts | -0.022 |
fru | -0.022 |
skirt | -0.020 |
makeup | -0.019 |
Token | be |
Token ID | 307 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
ey | +0.003 |
skirts | +0.002 |
fru | +0.002 |
skirt | +0.002 |
ems | +0.002 |
Neg logit contributions | |
ver | -0.003 |
hand | -0.003 |
chief | -0.003 |
happily | -0.003 |
rolled | -0.003 |
Token | ak |
Token ID | 461 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
ey | +0.100 |
skirts | +0.094 |
fru | +0.094 |
skirt | +0.083 |
ems | +0.080 |
Neg logit contributions | |
ver | -0.143 |
hand | -0.131 |
happily | -0.130 |
chief | -0.128 |
rolled | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
skirts | +0.056 |
ey | +0.055 |
skirt | +0.050 |
fru | +0.050 |
makeup | +0.048 |
Neg logit contributions | |
ver | -0.073 |
chief | -0.067 |
hand | -0.067 |
rolled | -0.064 |
mayor | -0.064 |
Token | The |
Token ID | 383 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
ver | +0.001 |
hand | +0.001 |
chief | +0.001 |
department | +0.001 |
rolled | +0.001 |
Neg logit contributions | |
ey | -0.002 |
fru | -0.002 |
skirts | -0.002 |
skirt | -0.002 |
ems | -0.001 |
Token | car |
Token ID | 1097 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
ver | +0.001 |
hand | +0.001 |
rolled | +0.001 |
Even | +0.001 |
chief | +0.001 |
Neg logit contributions | |
ey | -0.001 |
skirts | -0.001 |
fru | -0.001 |
skirt | -0.001 |
makeup | -0.001 |
Token | fell |
Token ID | 3214 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
ver | +0.032 |
hand | +0.030 |
Even | +0.028 |
chief | +0.028 |
rolled | +0.027 |
Neg logit contributions | |
skirts | -0.023 |
ey | -0.023 |
fru | -0.022 |
skirt | -0.021 |
brace | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
skirts | +0.006 |
fru | +0.006 |
ey | +0.005 |
makeup | +0.005 |
ol | +0.005 |
Neg logit contributions | |
ver | -0.006 |
hand | -0.006 |
chief | -0.006 |
case | -0.006 |
Even | -0.006 |
Token | needle |
Token ID | 17598 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
ey | +0.001 |
skirts | +0.001 |
fru | +0.001 |
disl | +0.001 |
ems | +0.001 |
Neg logit contributions | |
ver | -0.001 |
't | -0.001 |
happily | -0.001 |
safely | -0.001 |
sadly | -0.001 |
Token | out |
Token ID | 503 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
ver | +0.015 |
hand | +0.014 |
chief | +0.013 |
rolled | +0.013 |
Even | +0.013 |
Neg logit contributions | |
skirts | -0.017 |
ey | -0.017 |
fru | -0.016 |
skirt | -0.015 |
makeup | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
ver | +0.011 |
hand | +0.009 |
chief | +0.008 |
happily | +0.008 |
Even | +0.007 |
Neg logit contributions | |
ey | -0.032 |
skirts | -0.031 |
fru | -0.031 |
skirt | -0.029 |
ems | -0.028 |
Token | its |
Token ID | 663 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
ey | +0.007 |
skirts | +0.006 |
fru | +0.006 |
ems | +0.006 |
skirt | +0.006 |
Neg logit contributions | |
ver | -0.005 |
hand | -0.005 |
happily | -0.005 |
chief | -0.005 |
sadly | -0.005 |
Token | be |
Token ID | 307 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
happily | +0.011 |
ver | +0.011 |
sadly | +0.011 |
safely | +0.010 |
rolled | +0.010 |
Neg logit contributions | |
ey | -0.014 |
skirts | -0.013 |
fru | -0.013 |
ems | -0.012 |
--- | -0.012 |
Token | ak |
Token ID | 461 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
ey | +0.124 |
fru | +0.114 |
skirts | +0.110 |
skirt | +0.101 |
ems | +0.095 |
Neg logit contributions | |
ver | -0.126 |
happily | -0.116 |
safely | -0.113 |
quickly | -0.111 |
sadly | -0.108 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
ey | +0.060 |
skirts | +0.060 |
fru | +0.058 |
skirt | +0.054 |
makeup | +0.052 |
Neg logit contributions | |
ver | -0.057 |
hand | -0.053 |
chief | -0.052 |
rolled | -0.049 |
Even | -0.049 |
Token | After |
Token ID | 2293 |
Feature activation | -7.85e-01 |
Pos logit contributions | |
ey | +0.004 |
fru | +0.004 |
skirts | +0.004 |
ems | +0.003 |
skirt | +0.003 |
Neg logit contributions | |
ver | -0.003 |
hand | -0.003 |
chief | -0.003 |
rolled | -0.003 |
department | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
ey | +0.029 |
skirts | +0.029 |
fru | +0.028 |
makeup | +0.026 |
ol | +0.026 |
Neg logit contributions | |
ver | -0.024 |
hand | -0.021 |
chief | -0.021 |
department | -0.020 |
rolled | -0.020 |
Token | few |
Token ID | 1178 |
Feature activation | -7.69e-01 |
Pos logit contributions | |
ey | +0.026 |
skirts | +0.025 |
fru | +0.023 |
ems | +0.022 |
skirt | +0.022 |
Neg logit contributions | |
ver | -0.017 |
hand | -0.015 |
chief | -0.015 |
happily | -0.015 |
quickly | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
ver | +0.004 |
hand | +0.004 |
chief | +0.004 |
happily | +0.004 |
Even | +0.003 |
Neg logit contributions | |
ey | -0.005 |
skirts | -0.005 |
fru | -0.005 |
skirt | -0.004 |
ems | -0.004 |
Token | It |
Token ID | 632 |
Feature activation | +4.85e-01 |
Pos logit contributions | |
ey | +0.001 |
skirts | +0.001 |
fru | +0.000 |
skirt | +0.000 |
ems | +0.000 |
Neg logit contributions | |
ver | -0.000 |
hand | -0.000 |
chief | -0.000 |
Even | -0.000 |
happily | -0.000 |
Token | had |
Token ID | 550 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
ver | +0.014 |
hand | +0.012 |
chief | +0.012 |
happily | +0.011 |
rolled | +0.011 |
Neg logit contributions | |
ey | -0.020 |
skirts | -0.019 |
fru | -0.019 |
skirt | -0.018 |
ems | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | -7.00e-01 |
Pos logit contributions | |
ey | +0.017 |
skirts | +0.017 |
fru | +0.017 |
skirt | +0.015 |
makeup | +0.015 |
Neg logit contributions | |
ver | -0.016 |
hand | -0.014 |
chief | -0.014 |
department | -0.013 |
rolled | -0.013 |
Token | long |
Token ID | 890 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
ey | +0.022 |
skirts | +0.022 |
fru | +0.020 |
ems | +0.019 |
skirt | +0.018 |
Neg logit contributions | |
ver | -0.025 |
chief | -0.023 |
hand | -0.023 |
happily | -0.023 |
sadly | -0.023 |
Token | be |
Token ID | 307 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
skirts | +0.055 |
ey | +0.055 |
fru | +0.053 |
ems | +0.048 |
ol | +0.046 |
Neg logit contributions | |
ver | -0.056 |
happily | -0.055 |
Even | -0.054 |
sadly | -0.054 |
rolled | -0.053 |
Token | ak |
Token ID | 461 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ey | +0.097 |
skirts | +0.092 |
fru | +0.091 |
skirt | +0.081 |
ems | +0.077 |
Neg logit contributions | |
ver | -0.145 |
hand | -0.132 |
happily | -0.130 |
chief | -0.129 |
rolled | -0.128 |
Token | and |
Token ID | 290 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
ey | +0.055 |
skirts | +0.053 |
fru | +0.051 |
skirt | +0.048 |
makeup | +0.046 |
Neg logit contributions | |
ver | -0.048 |
hand | -0.044 |
chief | -0.043 |
department | -0.041 |
happily | -0.040 |
Token | shiny |
Token ID | 22441 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
ey | +0.046 |
skirts | +0.045 |
fru | +0.044 |
skirt | +0.041 |
makeup | +0.040 |
Neg logit contributions | |
ver | -0.033 |
hand | -0.030 |
chief | -0.029 |
department | -0.027 |
rolled | -0.027 |
Token | eyes |
Token ID | 2951 |
Feature activation | +4.88e-03 |
Pos logit contributions | |
ey | +0.006 |
skirts | +0.006 |
fru | +0.005 |
ems | +0.005 |
--- | +0.005 |
Neg logit contributions | |
ver | -0.007 |
happily | -0.007 |
sadly | -0.007 |
hand | -0.007 |
rolled | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
ver | +0.000 |
chief | +0.000 |
hand | +0.000 |
department | +0.000 |
Even | +0.000 |
Neg logit contributions | |
skirts | -0.000 |
ey | -0.000 |
fru | -0.000 |
skirt | -0.000 |
makeup | -0.000 |
Token | had |
Token ID | 550 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
ver | +0.019 |
hand | +0.017 |
chief | +0.017 |
happily | +0.016 |
Even | +0.016 |
Neg logit contributions | |
ey | -0.025 |
skirts | -0.024 |
fru | -0.024 |
skirt | -0.022 |
ems | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
ey | +0.026 |
skirts | +0.025 |
fru | +0.025 |
skirt | +0.023 |
makeup | +0.022 |
Neg logit contributions | |
ver | -0.018 |
hand | -0.016 |
chief | -0.015 |
department | -0.014 |
rolled | -0.014 |
Token | loop |
Token ID | 9052 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
ey | +0.023 |
skirts | +0.023 |
fru | +0.022 |
ems | +0.021 |
skirt | +0.019 |
Neg logit contributions | |
ver | -0.023 |
chief | -0.021 |
happily | -0.021 |
hand | -0.020 |
sadly | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
ey | +0.009 |
skirts | +0.009 |
fru | +0.008 |
skirt | +0.008 |
makeup | +0.008 |
Neg logit contributions | |
ver | -0.008 |
hand | -0.007 |
chief | -0.007 |
department | -0.007 |
Even | -0.007 |
Token | its |
Token ID | 663 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
ey | +0.025 |
skirts | +0.025 |
fru | +0.024 |
skirt | +0.022 |
makeup | +0.021 |
Neg logit contributions | |
ver | -0.026 |
hand | -0.023 |
chief | -0.023 |
department | -0.022 |
rolled | -0.022 |
Token | be |
Token ID | 307 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
ey | +0.019 |
skirts | +0.019 |
fru | +0.018 |
ems | +0.017 |
skirt | +0.016 |
Neg logit contributions | |
ver | -0.017 |
happily | -0.016 |
hand | -0.016 |
Even | -0.016 |
rolled | -0.016 |
Token | ak |
Token ID | 461 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
ey | +0.119 |
fru | +0.112 |
skirts | +0.105 |
skirt | +0.097 |
ems | +0.094 |
Neg logit contributions | |
ver | -0.128 |
happily | -0.117 |
safely | -0.114 |
rolled | -0.112 |
hand | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
skirts | +0.067 |
ey | +0.065 |
fru | +0.061 |
skirt | +0.059 |
makeup | +0.056 |
Neg logit contributions | |
ver | -0.075 |
hand | -0.070 |
chief | -0.069 |
department | -0.066 |
Even | -0.065 |
Token | The |
Token ID | 383 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
ey | +0.015 |
skirts | +0.014 |
fru | +0.014 |
skirt | +0.013 |
ems | +0.013 |
Neg logit contributions | |
ver | -0.010 |
hand | -0.010 |
chief | -0.009 |
department | -0.009 |
rolled | -0.009 |
Token | bird |
Token ID | 6512 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
ey | +0.000 |
fru | +0.000 |
skirts | +0.000 |
ems | +0.000 |
makeup | +0.000 |
Neg logit contributions | |
ver | -0.000 |
happily | -0.000 |
hand | -0.000 |
chief | -0.000 |
rolled | -0.000 |
Token | started |
Token ID | 2067 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
ey | +0.001 |
skirts | +0.001 |
fru | +0.001 |
skirt | +0.001 |
makeup | +0.001 |
Neg logit contributions | |
ver | -0.001 |
hand | -0.001 |
chief | -0.001 |
happily | -0.001 |
rolled | -0.001 |
Token | it |
Token ID | 340 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
ey | +0.011 |
skirts | +0.011 |
fru | +0.011 |
skirt | +0.010 |
ems | +0.009 |
Neg logit contributions | |
ver | -0.010 |
hand | -0.009 |
chief | -0.009 |
happily | -0.008 |
rolled | -0.008 |
Token | hurt |
Token ID | 5938 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
ey | +0.027 |
fru | +0.026 |
skirts | +0.026 |
skirt | +0.024 |
ol | +0.022 |
Neg logit contributions | |
ver | -0.028 |
hand | -0.027 |
chief | -0.025 |
rolled | -0.024 |
Even | -0.023 |
Token | Twe |
Token ID | 24205 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
skirts | +0.023 |
ey | +0.022 |
fru | +0.022 |
makeup | +0.021 |
skirt | +0.021 |
Neg logit contributions | |
ver | -0.026 |
chief | -0.025 |
hand | -0.023 |
department | -0.022 |
mayor | -0.022 |
Token | ety |
Token ID | 2963 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
ver | +0.066 |
hand | +0.062 |
happily | +0.059 |
chief | +0.059 |
rolled | +0.059 |
Neg logit contributions | |
ey | -0.037 |
skirts | -0.035 |
fru | -0.034 |
skirt | -0.029 |
makeup | -0.028 |
Token | 's |
Token ID | 338 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
ey | +0.049 |
skirts | +0.048 |
fru | +0.047 |
skirt | +0.044 |
ems | +0.042 |
Neg logit contributions | |
ver | -0.044 |
hand | -0.040 |
chief | -0.039 |
rolled | -0.037 |
Even | -0.037 |
Token | be |
Token ID | 307 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ey | +0.008 |
skirts | +0.008 |
fru | +0.008 |
ems | +0.007 |
skirt | +0.007 |
Neg logit contributions | |
ver | -0.008 |
hand | -0.007 |
happily | -0.007 |
chief | -0.007 |
sadly | -0.007 |
Token | ak |
Token ID | 461 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
ey | +0.100 |
skirts | +0.093 |
fru | +0.090 |
skirt | +0.080 |
ems | +0.078 |
Neg logit contributions | |
ver | -0.144 |
happily | -0.133 |
hand | -0.131 |
sadly | -0.128 |
quickly | -0.127 |
Token | . |
Token ID | 13 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
skirts | +0.056 |
skirt | +0.052 |
ey | +0.050 |
fru | +0.049 |
makeup | +0.047 |
Neg logit contributions | |
ver | -0.079 |
chief | -0.075 |
hand | -0.073 |
Even | -0.072 |
rolled | -0.072 |
Token | Twe |
Token ID | 24205 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
ey | +0.015 |
fru | +0.015 |
skirts | +0.014 |
makeup | +0.013 |
skirt | +0.013 |
Neg logit contributions | |
ver | -0.014 |
hand | -0.012 |
chief | -0.012 |
department | -0.012 |
brigade | -0.011 |
Token | ety |
Token ID | 2963 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
ver | +0.064 |
hand | +0.059 |
happily | +0.058 |
rolled | +0.057 |
Even | +0.057 |
Neg logit contributions | |
ey | -0.040 |
skirts | -0.039 |
fru | -0.037 |
skirt | -0.033 |
makeup | -0.031 |
Token | learned |
Token ID | 4499 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
fru | +0.018 |
ey | +0.018 |
skirts | +0.018 |
skirt | +0.017 |
ems | +0.015 |
Neg logit contributions | |
ver | -0.023 |
hand | -0.022 |
chief | -0.021 |
tow | -0.020 |
department | -0.020 |
Token | pipe |
Token ID | 12656 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
skirts | +0.024 |
fru | +0.023 |
ey | +0.022 |
ems | +0.020 |
skirt | +0.019 |
Neg logit contributions | |
chief | -0.025 |
department | -0.024 |
quickly | -0.024 |
ver | -0.024 |
hand | -0.023 |
Token | got |
Token ID | 1392 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
ey | +0.012 |
skirts | +0.012 |
fru | +0.011 |
skirt | +0.011 |
makeup | +0.010 |
Neg logit contributions | |
ver | -0.011 |
hand | -0.011 |
chief | -0.010 |
rolled | -0.010 |
Even | -0.010 |
Token | stuck |
Token ID | 7819 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
skirts | +0.007 |
ey | +0.007 |
fru | +0.007 |
prett | +0.007 |
ol | +0.006 |
Neg logit contributions | |
ver | -0.008 |
hand | -0.008 |
chief | -0.008 |
Even | -0.007 |
rolled | -0.007 |
Token | on |
Token ID | 319 |
Feature activation | +5.95e-01 |
Pos logit contributions | |
skirts | +0.022 |
ey | +0.021 |
fru | +0.019 |
skirt | +0.019 |
makeup | +0.018 |
Neg logit contributions | |
ver | -0.025 |
chief | -0.024 |
hand | -0.024 |
Even | -0.023 |
rolled | -0.023 |
Token | his |
Token ID | 465 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
ver | +0.026 |
chief | +0.023 |
hand | +0.022 |
ittle | +0.022 |
uttered | +0.022 |
Neg logit contributions | |
ey | -0.014 |
skirts | -0.014 |
ol | -0.014 |
fru | -0.014 |
makeup | -0.013 |
Token | be |
Token ID | 307 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ey | +0.009 |
skirts | +0.009 |
fru | +0.009 |
ems | +0.008 |
skirt | +0.008 |
Neg logit contributions | |
ver | -0.009 |
hand | -0.008 |
chief | -0.008 |
happily | -0.008 |
rolled | -0.008 |
Token | ak |
Token ID | 461 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
ey | +0.098 |
fru | +0.091 |
skirts | +0.091 |
skirt | +0.079 |
ems | +0.076 |
Neg logit contributions | |
ver | -0.148 |
hand | -0.133 |
happily | -0.132 |
chief | -0.130 |
rolled | -0.128 |
Token | ! |
Token ID | 0 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
skirts | +0.058 |
ey | +0.055 |
skirt | +0.052 |
fru | +0.051 |
makeup | +0.049 |
Neg logit contributions | |
ver | -0.072 |
chief | -0.070 |
hand | -0.069 |
mayor | -0.065 |
rolled | -0.065 |
Token | |
Token ID | 220 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
ey | +0.018 |
fru | +0.017 |
skirts | +0.016 |
ol | +0.015 |
makeup | +0.015 |
Neg logit contributions | |
ver | -0.018 |
hand | -0.017 |
chief | -0.017 |
department | -0.016 |
brigade | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
chief | +0.005 |
hand | +0.005 |
department | +0.004 |
ver | +0.004 |
mayor | +0.004 |
Neg logit contributions | |
ey | -0.005 |
skirts | -0.004 |
fru | -0.004 |
--- | -0.004 |
ems | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
ey | +0.007 |
fru | +0.007 |
skirts | +0.007 |
--- | +0.006 |
skirt | +0.006 |
Neg logit contributions | |
ver | -0.006 |
chief | -0.006 |
hand | -0.006 |
department | -0.005 |
brigade | -0.005 |
Token | had |
Token ID | 550 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
ver | +0.012 |
hand | +0.011 |
chief | +0.011 |
Even | +0.010 |
department | +0.010 |
Neg logit contributions | |
ey | -0.013 |
skirts | -0.013 |
fru | -0.013 |
skirt | -0.012 |
ems | -0.011 |
Token | got |
Token ID | 1392 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
ey | +0.025 |
skirts | +0.024 |
fru | +0.024 |
skirt | +0.022 |
ems | +0.021 |
Neg logit contributions | |
ver | -0.022 |
hand | -0.020 |
chief | -0.020 |
happily | -0.019 |
rolled | -0.019 |
Token | stuck |
Token ID | 7819 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
skirts | +0.006 |
ey | +0.006 |
fru | +0.006 |
makeup | +0.005 |
skirt | +0.005 |
Neg logit contributions | |
ver | -0.006 |
hand | -0.005 |
chief | -0.005 |
Even | -0.005 |
rolled | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
skirts | +0.014 |
ey | +0.014 |
fru | +0.013 |
skirt | +0.013 |
makeup | +0.012 |
Neg logit contributions | |
ver | -0.014 |
hand | -0.013 |
chief | -0.013 |
Even | -0.013 |
rolled | -0.012 |
Token | its |
Token ID | 663 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
ey | +0.011 |
skirts | +0.011 |
fru | +0.011 |
skirt | +0.010 |
makeup | +0.010 |
Neg logit contributions | |
ver | -0.010 |
hand | -0.009 |
chief | -0.008 |
rolled | -0.008 |
department | -0.008 |
Token | be |
Token ID | 307 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
ver | +0.004 |
hand | +0.004 |
chief | +0.003 |
happily | +0.003 |
Even | +0.003 |
Neg logit contributions | |
ey | -0.004 |
skirts | -0.004 |
fru | -0.004 |
skirt | -0.004 |
ems | -0.003 |
Token | ak |
Token ID | 461 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
ey | +0.094 |
skirts | +0.088 |
fru | +0.087 |
skirt | +0.077 |
ems | +0.073 |
Neg logit contributions | |
ver | -0.149 |
hand | -0.136 |
happily | -0.133 |
chief | -0.133 |
rolled | -0.130 |
Token | ! |
Token ID | 0 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
skirts | +0.055 |
ey | +0.054 |
fru | +0.050 |
skirt | +0.049 |
makeup | +0.047 |
Neg logit contributions | |
ver | -0.062 |
hand | -0.058 |
chief | -0.058 |
rolled | -0.056 |
Even | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | -7.04e-02 |
Pos logit contributions | |
ey | +0.013 |
fru | +0.012 |
skirts | +0.011 |
makeup | +0.011 |
ol | +0.010 |
Neg logit contributions | |
ver | -0.011 |
hand | -0.010 |
chief | -0.010 |
department | -0.010 |
brigade | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
ey | +0.003 |
skirts | +0.003 |
fru | +0.003 |
skirt | +0.002 |
--- | +0.002 |
Neg logit contributions | |
ver | -0.002 |
chief | -0.002 |
hand | -0.002 |
department | -0.002 |
rolled | -0.002 |
Token | The |
Token ID | 464 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
ey | +0.006 |
skirts | +0.006 |
fru | +0.006 |
skirt | +0.005 |
ems | +0.005 |
Neg logit contributions | |
ver | -0.005 |
hand | -0.004 |
chief | -0.004 |
happily | -0.004 |
Even | -0.004 |
Token | played |
Token ID | 2826 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
hand | +0.013 |
ver | +0.013 |
chief | +0.012 |
department | +0.012 |
itt | +0.012 |
Neg logit contributions | |
skirts | -0.010 |
skirt | -0.009 |
fru | -0.009 |
sunglasses | -0.009 |
ems | -0.009 |
Token | with |
Token ID | 351 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
ver | +0.017 |
hand | +0.015 |
chief | +0.015 |
department | +0.014 |
rolled | +0.014 |
Neg logit contributions | |
skirts | -0.037 |
ey | -0.037 |
fru | -0.036 |
skirt | -0.034 |
makeup | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
skirts | +0.012 |
fru | +0.012 |
da | +0.011 |
makeup | +0.011 |
ol | +0.011 |
Neg logit contributions | |
ver | -0.023 |
hand | -0.021 |
rolled | -0.021 |
chief | -0.020 |
department | -0.020 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
ey | +0.002 |
fru | +0.002 |
skirts | +0.002 |
skirt | +0.001 |
makeup | +0.001 |
Neg logit contributions | |
ver | -0.002 |
hand | -0.001 |
chief | -0.001 |
rolled | -0.001 |
Even | -0.001 |
Token | in |
Token ID | 259 |
Feature activation | +9.83e-01 |
Pos logit contributions | |
hand | +0.075 |
ver | +0.074 |
Even | +0.074 |
Together | +0.073 |
rolled | +0.071 |
Neg logit contributions | |
ey | -0.067 |
fru | -0.063 |
skirts | -0.061 |
b | -0.059 |
eking | -0.058 |
Token | oc |
Token ID | 420 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
ver | +0.036 |
uttered | +0.032 |
hand | +0.031 |
rolled | +0.030 |
Even | +0.029 |
Neg logit contributions | |
fru | -0.028 |
da | -0.027 |
ey | -0.027 |
ol | -0.026 |
skirts | -0.026 |
Token | eros |
Token ID | 27498 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
ver | +0.163 |
case | +0.134 |
Even | +0.130 |
happily | +0.125 |
Together | +0.124 |
Neg logit contributions | |
ey | -0.123 |
fru | -0.114 |
ol | -0.099 |
ution | -0.098 |
brace | -0.097 |
Token | for |
Token ID | 329 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
ver | +0.020 |
rolled | +0.019 |
hand | +0.019 |
Even | +0.018 |
chief | +0.018 |
Neg logit contributions | |
skirts | -0.018 |
skirt | -0.016 |
da | -0.016 |
ol | -0.015 |
fru | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
ey | +0.012 |
skirts | +0.012 |
fru | +0.012 |
skirt | +0.011 |
makeup | +0.010 |
Neg logit contributions | |
ver | -0.009 |
hand | -0.008 |
chief | -0.008 |
rolled | -0.007 |
department | -0.007 |
Token | while |
Token ID | 981 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
ey | +0.011 |
skirts | +0.011 |
fru | +0.011 |
skirt | +0.010 |
makeup | +0.009 |
Neg logit contributions | |
ver | -0.013 |
hand | -0.012 |
chief | -0.012 |
rolled | -0.011 |
Even | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
ver | +0.001 |
hand | +0.001 |
chief | +0.001 |
rolled | +0.001 |
department | +0.001 |
Neg logit contributions | |
skirts | -0.001 |
ey | -0.001 |
fru | -0.001 |
skirt | -0.001 |
makeup | -0.001 |
Token | big |
Token ID | 1263 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
ver | +0.013 |
hand | +0.013 |
chief | +0.012 |
rolled | +0.012 |
Even | +0.012 |
Neg logit contributions | |
ey | -0.012 |
skirts | -0.012 |
fru | -0.012 |
skirt | -0.011 |
makeup | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
ver | +0.013 |
chief | +0.012 |
hand | +0.012 |
happily | +0.012 |
Even | +0.012 |
Neg logit contributions | |
ey | -0.014 |
skirts | -0.014 |
fru | -0.014 |
skirt | -0.012 |
ems | -0.012 |
Token | gray |
Token ID | 12768 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
ver | +0.008 |
chief | +0.007 |
happily | +0.007 |
hand | +0.007 |
quickly | +0.007 |
Neg logit contributions | |
ey | -0.007 |
skirts | -0.007 |
fru | -0.007 |
ems | -0.006 |
skirt | -0.006 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
ver | +0.026 |
hand | +0.023 |
rolled | +0.021 |
chief | +0.021 |
Even | +0.021 |
Neg logit contributions | |
ey | -0.022 |
skirts | -0.021 |
fru | -0.020 |
skirt | -0.020 |
makeup | -0.019 |
Token | in |
Token ID | 259 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
hand | +0.101 |
ver | +0.098 |
rolled | +0.097 |
department | +0.092 |
Even | +0.091 |
Neg logit contributions | |
ey | -0.098 |
skirts | -0.092 |
fru | -0.088 |
skirt | -0.085 |
da | -0.085 |
Token | oc |
Token ID | 420 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
ver | +0.033 |
hand | +0.030 |
rolled | +0.028 |
department | +0.027 |
uttered | +0.027 |
Neg logit contributions | |
ey | -0.042 |
skirts | -0.041 |
fru | -0.041 |
skirt | -0.039 |
da | -0.038 |
Token | eros |
Token ID | 27498 |
Feature activation | +7.79e-01 |
Pos logit contributions | |
ver | +0.174 |
Even | +0.144 |
sadly | +0.140 |
rolled | +0.139 |
hand | +0.138 |
Neg logit contributions | |
ey | -0.140 |
fru | -0.129 |
ol | -0.118 |
da | -0.118 |
jewelry | -0.114 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
ver | +0.028 |
hand | +0.026 |
department | +0.025 |
chief | +0.024 |
rolled | +0.024 |
Neg logit contributions | |
skirts | -0.024 |
ey | -0.024 |
fru | -0.023 |
skirt | -0.022 |
makeup | -0.022 |
Token | He |
Token ID | 679 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
ey | +0.006 |
skirts | +0.006 |
fru | +0.006 |
skirt | +0.006 |
makeup | +0.005 |
Neg logit contributions | |
ver | -0.004 |
hand | -0.004 |
chief | -0.004 |
department | -0.004 |
sadly | -0.004 |
Token | lived |
Token ID | 5615 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ver | +0.001 |
chief | +0.001 |
hand | +0.001 |
happily | +0.001 |
rolled | +0.001 |
Neg logit contributions | |
ey | -0.001 |
skirts | -0.001 |
fru | -0.001 |
skirt | -0.001 |
ems | -0.001 |
Token | in |
Token ID | 287 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
hand | +0.041 |
ver | +0.041 |
Even | +0.037 |
chief | +0.037 |
department | +0.037 |
Neg logit contributions | |
ey | -0.033 |
skirts | -0.033 |
fru | -0.031 |
skirt | -0.029 |
prett | -0.028 |
Token | Daddy |
Token ID | 26926 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
ey | +0.008 |
skirts | +0.008 |
fru | +0.008 |
skirt | +0.007 |
makeup | +0.007 |
Neg logit contributions | |
ver | -0.007 |
hand | -0.006 |
chief | -0.006 |
rolled | -0.006 |
Even | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
ey | +0.014 |
skirts | +0.014 |
fru | +0.014 |
skirt | +0.013 |
makeup | +0.012 |
Neg logit contributions | |
ver | -0.012 |
hand | -0.011 |
chief | -0.011 |
rolled | -0.010 |
department | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
ey | +0.014 |
skirts | +0.014 |
fru | +0.014 |
skirt | +0.013 |
ems | +0.012 |
Neg logit contributions | |
ver | -0.013 |
hand | -0.012 |
chief | -0.011 |
rolled | -0.011 |
happily | -0.011 |
Token | three |
Token ID | 1115 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
ey | +0.038 |
skirts | +0.035 |
fru | +0.034 |
skirt | +0.032 |
ems | +0.032 |
Neg logit contributions | |
ver | -0.026 |
hand | -0.023 |
happily | -0.023 |
chief | -0.023 |
sadly | -0.022 |
Token | - |
Token ID | 12 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ver | +0.022 |
hand | +0.020 |
chief | +0.019 |
rolled | +0.019 |
department | +0.018 |
Neg logit contributions | |
skirts | -0.020 |
ey | -0.019 |
fru | -0.019 |
skirt | -0.017 |
makeup | -0.017 |
Token | year |
Token ID | 1941 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
ver | +0.027 |
Even | +0.023 |
rolled | +0.022 |
sadly | +0.022 |
hand | +0.021 |
Neg logit contributions | |
skirts | -0.056 |
fru | -0.055 |
ol | -0.053 |
makeup | -0.052 |
ey | -0.052 |
Token | - |
Token ID | 12 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
ver | +0.136 |
hand | +0.118 |
department | +0.114 |
Even | +0.112 |
surely | +0.112 |
Neg logit contributions | |
fru | -0.098 |
skirts | -0.095 |
ey | -0.092 |
ol | -0.092 |
da | -0.092 |
Token | old |
Token ID | 727 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
ver | +0.013 |
Even | +0.011 |
sadly | +0.011 |
rolled | +0.011 |
department | +0.011 |
Neg logit contributions | |
skirts | -0.025 |
fru | -0.025 |
ol | -0.024 |
ey | -0.024 |
makeup | -0.023 |
Token | little |
Token ID | 1310 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
ver | +0.037 |
hand | +0.033 |
rolled | +0.033 |
Even | +0.032 |
chief | +0.032 |
Neg logit contributions | |
skirts | -0.028 |
fru | -0.027 |
ey | -0.026 |
skirt | -0.024 |
ol | -0.024 |
Token | Sus |
Token ID | 8932 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
ver | +0.020 |
hand | +0.018 |
rolled | +0.017 |
chief | +0.017 |
Even | +0.017 |
Neg logit contributions | |
skirts | -0.013 |
ey | -0.013 |
fru | -0.012 |
skirt | -0.011 |
makeup | -0.011 |
Token | ie |
Token ID | 494 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
ver | +0.011 |
hand | +0.010 |
chief | +0.010 |
happily | +0.010 |
rolled | +0.009 |
Neg logit contributions | |
ey | -0.009 |
skirts | -0.008 |
fru | -0.008 |
skirt | -0.007 |
ems | -0.007 |
Token | trapped |
Token ID | 13640 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
ver | +0.004 |
hand | +0.004 |
rolled | +0.004 |
Even | +0.004 |
chief | +0.004 |
Neg logit contributions | |
skirts | -0.003 |
ey | -0.003 |
fru | -0.003 |
prett | -0.003 |
doing | -0.003 |
Token | with |
Token ID | 351 |
Feature activation | -6.11e-01 |
Pos logit contributions | |
ver | +0.006 |
Even | +0.005 |
rolled | +0.005 |
hand | +0.005 |
chief | +0.005 |
Neg logit contributions | |
skirts | -0.004 |
ey | -0.004 |
fru | -0.003 |
doing | -0.003 |
makeup | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
skirts | +0.015 |
fru | +0.014 |
clothing | +0.013 |
sunglasses | +0.013 |
ey | +0.013 |
Neg logit contributions | |
ver | -0.026 |
Together | -0.023 |
department | -0.023 |
hand | -0.023 |
Even | -0.023 |
Token | rh |
Token ID | 9529 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
ey | +0.004 |
fru | +0.004 |
skirts | +0.004 |
skirt | +0.003 |
prett | +0.003 |
Neg logit contributions | |
ver | -0.004 |
hand | -0.004 |
rolled | -0.003 |
Even | -0.003 |
happily | -0.003 |
Token | in |
Token ID | 259 |
Feature activation | +8.63e-01 |
Pos logit contributions | |
Even | +0.066 |
hand | +0.065 |
Together | +0.064 |
ver | +0.063 |
rolled | +0.063 |
Neg logit contributions | |
ey | -0.058 |
fru | -0.054 |
skirts | -0.052 |
eking | -0.051 |
b | -0.050 |
Token | oc |
Token ID | 420 |
Feature activation | +3.93e+00 |
Pos logit contributions | |
ver | +0.030 |
hand | +0.027 |
uttered | +0.026 |
rolled | +0.026 |
Even | +0.026 |
Neg logit contributions | |
ey | -0.025 |
fru | -0.025 |
da | -0.025 |
skirts | -0.023 |
ol | -0.023 |
Token | eros |
Token ID | 27498 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
ver | +0.156 |
Even | +0.129 |
case | +0.128 |
happily | +0.125 |
sadly | +0.122 |
Neg logit contributions | |
ey | -0.125 |
fru | -0.116 |
ution | -0.103 |
ol | -0.103 |
eking | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
rolled | +0.019 |
ver | +0.019 |
Even | +0.018 |
hand | +0.018 |
uttered | +0.017 |
Neg logit contributions | |
skirts | -0.012 |
skirt | -0.010 |
da | -0.010 |
fru | -0.010 |
but | -0.010 |
Token | They |
Token ID | 1119 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
fru | +0.004 |
ey | +0.004 |
skirts | +0.004 |
--- | +0.004 |
makeup | +0.004 |
Neg logit contributions | |
ver | -0.004 |
hand | -0.003 |
chief | -0.003 |
rolled | -0.003 |
department | -0.003 |
Token | looked |
Token ID | 3114 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
ver | +0.002 |
hand | +0.001 |
chief | +0.001 |
department | +0.001 |
rolled | +0.001 |
Neg logit contributions | |
ey | -0.001 |
skirts | -0.001 |
fru | -0.001 |
skirt | -0.001 |
makeup | -0.001 |
Token | for |
Token ID | 329 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ey | +0.019 |
skirts | +0.019 |
fru | +0.018 |
skirt | +0.017 |
ems | +0.016 |
Neg logit contributions | |
ver | -0.016 |
hand | -0.015 |
chief | -0.014 |
rolled | -0.014 |
Even | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +8.05e-01 |
Pos logit contributions | |
ver | +0.043 |
hand | +0.039 |
chief | +0.038 |
department | +0.037 |
rolled | +0.035 |
Neg logit contributions | |
ey | -0.064 |
skirts | -0.062 |
fru | -0.062 |
skirt | -0.057 |
makeup | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | +4.77e-01 |
Pos logit contributions | |
ver | +0.027 |
chief | +0.026 |
department | +0.026 |
hand | +0.024 |
brigade | +0.024 |
Neg logit contributions | |
ey | -0.026 |
fru | -0.025 |
skirts | -0.024 |
--- | -0.023 |
makeup | -0.022 |
Token | The |
Token ID | 464 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
ver | +0.018 |
department | +0.017 |
chief | +0.017 |
uttered | +0.016 |
hand | +0.016 |
Neg logit contributions | |
ey | -0.013 |
skirts | -0.013 |
fru | -0.013 |
--- | -0.012 |
skirt | -0.012 |
Token | rh |
Token ID | 9529 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
ver | +0.016 |
hand | +0.013 |
uttered | +0.013 |
rolled | +0.013 |
department | +0.013 |
Neg logit contributions | |
ey | -0.015 |
fru | -0.015 |
skirts | -0.014 |
da | -0.013 |
skirt | -0.013 |
Token | in |
Token ID | 259 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ver | +0.076 |
hand | +0.075 |
rolled | +0.072 |
Even | +0.069 |
department | +0.068 |
Neg logit contributions | |
ey | -0.099 |
fru | -0.092 |
skirts | -0.090 |
ems | -0.082 |
skirt | -0.082 |
Token | oc |
Token ID | 420 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
ver | +0.035 |
hand | +0.031 |
uttered | +0.030 |
department | +0.030 |
rolled | +0.029 |
Neg logit contributions | |
fru | -0.039 |
ey | -0.038 |
skirts | -0.033 |
da | -0.033 |
scar | -0.032 |
Token | eros |
Token ID | 27498 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
ver | +0.162 |
Even | +0.135 |
case | +0.132 |
rolled | +0.129 |
hand | +0.129 |
Neg logit contributions | |
ey | -0.151 |
fru | -0.140 |
ol | -0.123 |
skirts | -0.121 |
ution | -0.120 |
Token | nodded |
Token ID | 14464 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
ver | +0.040 |
hand | +0.038 |
rolled | +0.037 |
department | +0.036 |
tow | +0.036 |
Neg logit contributions | |
fru | -0.026 |
ey | -0.024 |
skirts | -0.024 |
skirt | -0.022 |
makeup | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
ver | +0.005 |
hand | +0.004 |
chief | +0.004 |
rolled | +0.004 |
Even | +0.004 |
Neg logit contributions | |
ey | -0.006 |
skirts | -0.006 |
fru | -0.006 |
skirt | -0.005 |
ems | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
ver | +0.001 |
hand | +0.001 |
chief | +0.001 |
rolled | +0.001 |
department | +0.001 |
Neg logit contributions | |
fru | -0.001 |
skirts | -0.001 |
ey | -0.001 |
skirt | -0.000 |
da | -0.000 |
Token | little |
Token ID | 1310 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
ey | +0.024 |
fru | +0.023 |
skirts | +0.023 |
skirt | +0.021 |
makeup | +0.020 |
Neg logit contributions | |
ver | -0.026 |
hand | -0.022 |
rolled | -0.021 |
chief | -0.021 |
uttered | -0.021 |
Token | all |
Token ID | 477 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
ver | +0.014 |
hand | +0.014 |
chief | +0.013 |
department | +0.012 |
Even | +0.012 |
Neg logit contributions | |
skirts | -0.017 |
ey | -0.016 |
fru | -0.016 |
skirt | -0.016 |
sunglasses | -0.015 |
Token | o |
Token ID | 267 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
ver | +0.011 |
hand | +0.010 |
chief | +0.010 |
department | +0.009 |
rolled | +0.009 |
Neg logit contributions | |
ey | -0.011 |
skirts | -0.011 |
fru | -0.011 |
skirt | -0.010 |
ems | -0.010 |
Token | o |
Token ID | 78 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
ver | +0.003 |
chief | +0.003 |
hand | +0.003 |
sadly | +0.003 |
department | +0.003 |
Neg logit contributions | |
ey | -0.002 |
skirts | -0.002 |
fru | -0.002 |
ems | -0.002 |
skirt | -0.002 |
Token | hed |
Token ID | 704 |
Feature activation | +8.87e-01 |
Pos logit contributions | |
ver | +0.009 |
chief | +0.008 |
department | +0.008 |
hand | +0.008 |
mayor | +0.008 |
Neg logit contributions | |
skirts | -0.009 |
fru | -0.009 |
ey | -0.008 |
da | -0.008 |
skirt | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
ver | +0.030 |
hand | +0.027 |
chief | +0.026 |
rolled | +0.025 |
Even | +0.025 |
Neg logit contributions | |
skirts | -0.032 |
ey | -0.031 |
fru | -0.031 |
skirt | -0.029 |
makeup | -0.027 |
Token | ah |
Token ID | 29042 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
skirts | +0.005 |
ey | +0.005 |
fru | +0.004 |
skirt | +0.004 |
makeup | +0.004 |
Neg logit contributions | |
ver | -0.005 |
hand | -0.005 |
chief | -0.005 |
department | -0.004 |
Even | -0.004 |
Token | hed |
Token ID | 704 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
ver | +0.028 |
hand | +0.024 |
chief | +0.024 |
department | +0.022 |
rolled | +0.021 |
Neg logit contributions | |
ey | -0.063 |
skirts | -0.062 |
fru | -0.061 |
skirt | -0.057 |
ems | -0.056 |
Token | as |
Token ID | 355 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
ver | +0.027 |
hand | +0.024 |
chief | +0.023 |
rolled | +0.022 |
Even | +0.022 |
Neg logit contributions | |
skirts | -0.030 |
ey | -0.029 |
fru | -0.029 |
skirt | -0.027 |
makeup | -0.026 |
Token | they |
Token ID | 484 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
skirts | +0.014 |
fru | +0.013 |
makeup | +0.012 |
skirt | +0.012 |
ol | +0.011 |
Neg logit contributions | |
ver | -0.016 |
rolled | -0.015 |
chief | -0.015 |
hand | -0.015 |
Even | -0.015 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
ver | +0.025 |
hand | +0.023 |
chief | +0.023 |
Even | +0.022 |
department | +0.022 |
Neg logit contributions | |
skirts | -0.023 |
ey | -0.022 |
fru | -0.022 |
skirt | -0.021 |
ems | -0.020 |
Token | at |
Token ID | 379 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
ver | +0.008 |
hand | +0.007 |
chief | +0.007 |
rolled | +0.007 |
Even | +0.007 |
Neg logit contributions | |
ey | -0.009 |
skirts | -0.009 |
fru | -0.009 |
skirt | -0.008 |
ems | -0.008 |
Token | park |
Token ID | 3952 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
skirts | +0.028 |
fru | +0.028 |
ey | +0.028 |
ems | +0.025 |
skirt | +0.024 |
Neg logit contributions | |
ver | -0.031 |
hand | -0.030 |
chief | -0.029 |
Even | -0.029 |
happily | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
skirts | +0.030 |
ey | +0.030 |
fru | +0.029 |
skirt | +0.026 |
makeup | +0.025 |
Neg logit contributions | |
ver | -0.033 |
hand | -0.031 |
chief | -0.029 |
department | -0.029 |
rolled | -0.028 |
Token | Every |
Token ID | 3887 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
ey | +0.006 |
skirts | +0.005 |
fru | +0.005 |
skirt | +0.005 |
makeup | +0.005 |
Neg logit contributions | |
ver | -0.004 |
hand | -0.003 |
chief | -0.003 |
department | -0.003 |
rolled | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
ver | +0.002 |
hand | +0.001 |
chief | +0.001 |
department | +0.001 |
rolled | +0.001 |
Neg logit contributions | |
ey | -0.001 |
skirts | -0.001 |
fru | -0.001 |
skirt | -0.001 |
makeup | -0.001 |
Token | he |
Token ID | 339 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
ey | +0.009 |
skirts | +0.009 |
fru | +0.009 |
ol | +0.008 |
skirt | +0.008 |
Neg logit contributions | |
ver | -0.011 |
hand | -0.010 |
chief | -0.010 |
department | -0.010 |
rolled | -0.010 |
Token | would |
Token ID | 561 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ver | +0.024 |
hand | +0.023 |
chief | +0.022 |
Even | +0.021 |
department | +0.021 |
Neg logit contributions | |
skirts | -0.022 |
ey | -0.022 |
fru | -0.021 |
skirt | -0.019 |
ems | -0.019 |
Token | slide |
Token ID | 10649 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
ver | +0.051 |
hand | +0.049 |
chief | +0.047 |
department | +0.046 |
brigade | +0.044 |
Neg logit contributions | |
ey | -0.034 |
skirts | -0.033 |
fru | -0.030 |
skirt | -0.030 |
makeup | -0.027 |
Token | down |
Token ID | 866 |
Feature activation | +8.89e-01 |
Pos logit contributions | |
ver | +0.017 |
hand | +0.015 |
happily | +0.014 |
chief | +0.014 |
rolled | +0.014 |
Neg logit contributions | |
ey | -0.024 |
skirts | -0.023 |
fru | -0.023 |
skirt | -0.021 |
ems | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
ver | +0.039 |
chief | +0.035 |
hand | +0.035 |
department | +0.035 |
brigade | +0.035 |
Neg logit contributions | |
ey | -0.023 |
skirts | -0.020 |
fru | -0.019 |
ol | -0.018 |
doing | -0.017 |
Token | faster |
Token ID | 5443 |
Feature activation | +5.23e-03 |
Pos logit contributions | |
ver | +0.011 |
hand | +0.010 |
chief | +0.010 |
rolled | +0.009 |
department | +0.009 |
Neg logit contributions | |
ey | -0.009 |
skirts | -0.009 |
fru | -0.009 |
skirt | -0.008 |
ems | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
happily | +0.000 |
quickly | +0.000 |
safely | +0.000 |
ver | +0.000 |
hand | +0.000 |
Neg logit contributions | |
fru | -0.000 |
ey | -0.000 |
skirts | -0.000 |
skirt | -0.000 |
ems | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
skirts | +0.004 |
clothing | +0.004 |
fru | +0.003 |
makeup | +0.003 |
sunglasses | +0.003 |
Neg logit contributions | |
ver | -0.010 |
chief | -0.009 |
department | -0.009 |
ittle | -0.009 |
case | -0.008 |
Token | wagon |
Token ID | 31860 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
ver | +0.002 |
hand | +0.001 |
chief | +0.001 |
Even | +0.001 |
te | +0.001 |
Neg logit contributions | |
ey | -0.001 |
skirts | -0.001 |
fru | -0.001 |
skirt | -0.001 |
sunglasses | -0.001 |
Token | as |
Token ID | 355 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
ver | +0.021 |
ittle | +0.020 |
Thief | +0.020 |
department | +0.020 |
hand | +0.019 |
Neg logit contributions | |
but | -0.016 |
skirts | -0.015 |
prett | -0.015 |
skirt | -0.014 |
than | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
skirts | +0.010 |
pretty | +0.008 |
ol | +0.008 |
fru | +0.008 |
doing | +0.008 |
Neg logit contributions | |
Even | -0.015 |
hand | -0.015 |
ishly | -0.015 |
chief | -0.015 |
Sadly | -0.015 |
Token | drove |
Token ID | 10357 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
chief | +0.015 |
hand | +0.015 |
Even | +0.015 |
ver | +0.015 |
mayor | +0.014 |
Neg logit contributions | |
skirts | -0.013 |
fru | -0.013 |
skirt | -0.012 |
ey | -0.012 |
prett | -0.011 |
Token | around |
Token ID | 1088 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ver | +0.063 |
chief | +0.058 |
Even | +0.056 |
itt | +0.056 |
department | +0.055 |
Neg logit contributions | |
skirts | -0.052 |
ey | -0.046 |
fru | -0.045 |
--- | -0.045 |
skirt | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
ver | +0.048 |
hand | +0.042 |
chief | +0.042 |
department | +0.041 |
Even | +0.041 |
Neg logit contributions | |
skirts | -0.035 |
ey | -0.035 |
fru | -0.033 |
skirt | -0.031 |
makeup | -0.030 |
Token | garden |
Token ID | 11376 |
Feature activation | -5.07e-01 |
Pos logit contributions | |
ver | +0.011 |
uttered | +0.009 |
hand | +0.009 |
te | +0.009 |
chief | +0.009 |
Neg logit contributions | |
ey | -0.006 |
skirts | -0.006 |
skirt | -0.005 |
fru | -0.005 |
makeup | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
skirts | +0.014 |
ey | +0.013 |
fru | +0.013 |
skirt | +0.012 |
makeup | +0.011 |
Neg logit contributions | |
ver | -0.020 |
hand | -0.018 |
chief | -0.018 |
Even | -0.018 |
rolled | -0.018 |
Token | The |
Token ID | 383 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
ey | +0.008 |
skirts | +0.008 |
fru | +0.007 |
--- | +0.007 |
makeup | +0.007 |
Neg logit contributions | |
chief | -0.010 |
ver | -0.010 |
department | -0.010 |
hand | -0.009 |
brigade | -0.009 |
Token | tough |
Token ID | 5802 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
skirts | +0.003 |
ey | +0.003 |
sunglasses | +0.003 |
skirt | +0.003 |
fru | +0.003 |
Neg logit contributions | |
ver | -0.004 |
hand | -0.004 |
uttered | -0.003 |
te | -0.003 |
chief | -0.003 |
Token | had |
Token ID | 550 |
Feature activation | -7.44e-02 |
Pos logit contributions | |
ver | +0.003 |
hand | +0.003 |
chief | +0.003 |
rolled | +0.003 |
department | +0.003 |
Neg logit contributions | |
skirts | -0.003 |
ey | -0.003 |
fru | -0.003 |
skirt | -0.003 |
ems | -0.002 |
Token | made |
Token ID | 925 |
Feature activation | -9.19e-01 |
Pos logit contributions | |
ey | +0.003 |
skirts | +0.003 |
fru | +0.003 |
skirt | +0.003 |
makeup | +0.002 |
Neg logit contributions | |
ver | -0.002 |
hand | -0.002 |
chief | -0.002 |
rolled | -0.002 |
Even | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
skirts | +0.031 |
ey | +0.030 |
fru | +0.030 |
makeup | +0.028 |
skirt | +0.028 |
Neg logit contributions | |
ver | -0.032 |
chief | -0.029 |
hand | -0.028 |
rolled | -0.028 |
department | -0.028 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
ver | +0.006 |
hand | +0.006 |
chief | +0.006 |
Even | +0.005 |
rolled | +0.005 |
Neg logit contributions | |
ey | -0.006 |
skirts | -0.006 |
fru | -0.006 |
skirt | -0.005 |
makeup | -0.005 |
Token | wait |
Token ID | 4043 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ey | +0.017 |
skirts | +0.017 |
fru | +0.016 |
ems | +0.015 |
skirt | +0.015 |
Neg logit contributions | |
ver | -0.013 |
hand | -0.012 |
chief | -0.012 |
happily | -0.012 |
sadly | -0.012 |
Token | too |
Token ID | 1165 |
Feature activation | +9.38e-01 |
Pos logit contributions | |
ey | +0.053 |
skirts | +0.050 |
fru | +0.050 |
skirt | +0.046 |
ems | +0.045 |
Neg logit contributions | |
ver | -0.042 |
hand | -0.038 |
chief | -0.037 |
happily | -0.037 |
sadly | -0.036 |
Token | long |
Token ID | 890 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
ver | +0.036 |
chief | +0.033 |
hand | +0.032 |
department | +0.032 |
rolled | +0.032 |
Neg logit contributions | |
skirts | -0.028 |
ey | -0.027 |
fru | -0.026 |
skirt | -0.024 |
makeup | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
skirts | +0.007 |
ey | +0.007 |
fru | +0.006 |
skirt | +0.006 |
ems | +0.006 |
Neg logit contributions | |
ver | -0.006 |
hand | -0.006 |
chief | -0.006 |
Even | -0.005 |
rolled | -0.005 |
Token | She |
Token ID | 1375 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
skirts | +0.002 |
fru | +0.002 |
ey | +0.002 |
skirt | +0.002 |
makeup | +0.002 |
Neg logit contributions | |
ver | -0.003 |
chief | -0.003 |
hand | -0.003 |
rolled | -0.003 |
department | -0.003 |
Token | felt |
Token ID | 2936 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
ver | +0.019 |
hand | +0.017 |
chief | +0.017 |
department | +0.017 |
uttered | +0.016 |
Neg logit contributions | |
skirts | -0.012 |
fru | -0.012 |
skirt | -0.011 |
ey | -0.011 |
da | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
ey | +0.008 |
fru | +0.008 |
skirts | +0.008 |
skirt | +0.007 |
makeup | +0.007 |
Neg logit contributions | |
ver | -0.009 |
hand | -0.008 |
department | -0.008 |
chief | -0.008 |
rolled | -0.008 |
Token | Ben |
Token ID | 3932 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
ver | +0.003 |
happily | +0.003 |
sadly | +0.003 |
Even | +0.003 |
quickly | +0.003 |
Neg logit contributions | |
ey | -0.004 |
skirts | -0.003 |
fru | -0.003 |
eking | -0.003 |
reath | -0.003 |
Token | shouted |
Token ID | 17293 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
skirts | +0.017 |
ey | +0.017 |
fru | +0.017 |
skirt | +0.015 |
ems | +0.014 |
Neg logit contributions | |
ver | -0.026 |
hand | -0.024 |
chief | -0.024 |
rolled | -0.023 |
department | -0.023 |
Token | as |
Token ID | 355 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
skirts | +0.012 |
skirt | +0.011 |
makeup | +0.011 |
fru | +0.011 |
but | +0.010 |
Neg logit contributions | |
ver | -0.025 |
rolled | -0.024 |
hand | -0.023 |
itt | -0.023 |
chief | -0.023 |
Token | he |
Token ID | 339 |
Feature activation | +7.46e-01 |
Pos logit contributions | |
ey | +0.001 |
fru | +0.001 |
skirts | +0.000 |
skirt | +0.000 |
ol | +0.000 |
Neg logit contributions | |
hand | -0.002 |
rolled | -0.002 |
ver | -0.002 |
department | -0.001 |
chief | -0.001 |
Token | jumped |
Token ID | 11687 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
ver | +0.027 |
hand | +0.026 |
chief | +0.026 |
Even | +0.024 |
rolled | +0.024 |
Neg logit contributions | |
ey | -0.023 |
skirts | -0.023 |
fru | -0.022 |
skirt | -0.020 |
ems | -0.020 |
Token | out |
Token ID | 503 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ey | +0.017 |
skirts | +0.017 |
skirt | +0.016 |
fru | +0.016 |
makeup | +0.015 |
Neg logit contributions | |
ver | -0.018 |
hand | -0.016 |
chief | -0.016 |
rolled | -0.016 |
department | -0.015 |
Token | from |
Token ID | 422 |
Feature activation | +9.29e-01 |
Pos logit contributions | |
ver | +0.024 |
hand | +0.020 |
chief | +0.019 |
rolled | +0.019 |
Even | +0.017 |
Neg logit contributions | |
ey | -0.052 |
skirts | -0.050 |
fru | -0.049 |
skirt | -0.047 |
makeup | -0.045 |
Token | behind |
Token ID | 2157 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
ver | +0.031 |
hand | +0.030 |
chief | +0.029 |
happily | +0.028 |
sadly | +0.027 |
Neg logit contributions | |
ey | -0.032 |
skirts | -0.031 |
fru | -0.031 |
skirt | -0.028 |
ems | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
ey | +0.011 |
skirts | +0.010 |
fru | +0.010 |
skirt | +0.009 |
makeup | +0.009 |
Neg logit contributions | |
ver | -0.011 |
hand | -0.010 |
chief | -0.010 |
rolled | -0.009 |
department | -0.009 |
Token | tree |
Token ID | 5509 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
ver | +0.006 |
hand | +0.005 |
chief | +0.005 |
happily | +0.005 |
Even | +0.005 |
Neg logit contributions | |
ey | -0.007 |
skirts | -0.007 |
fru | -0.007 |
skirt | -0.006 |
ems | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
ver | +0.013 |
hand | +0.012 |
rolled | +0.011 |
chief | +0.011 |
Even | +0.011 |
Neg logit contributions | |
ey | -0.009 |
skirts | -0.009 |
fru | -0.008 |
skirt | -0.008 |
makeup | -0.007 |
Token | about |
Token ID | 546 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ver | +0.023 |
hand | +0.022 |
rolled | +0.021 |
chief | +0.020 |
department | +0.019 |
Neg logit contributions | |
skirts | -0.025 |
ey | -0.025 |
fru | -0.024 |
ems | -0.022 |
skirt | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
ver | +0.047 |
hand | +0.043 |
chief | +0.042 |
happily | +0.040 |
rolled | +0.040 |
Neg logit contributions | |
ey | -0.048 |
skirts | -0.046 |
fru | -0.045 |
skirt | -0.042 |
ems | -0.040 |
Token | colors |
Token ID | 7577 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
ver | +0.024 |
hand | +0.023 |
chief | +0.023 |
happily | +0.022 |
department | +0.022 |
Neg logit contributions | |
ey | -0.019 |
skirts | -0.019 |
fru | -0.018 |
ems | -0.017 |
skirt | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
ver | +0.009 |
hand | +0.008 |
chief | +0.008 |
rolled | +0.008 |
Even | +0.008 |
Neg logit contributions | |
ey | -0.011 |
skirts | -0.011 |
fru | -0.011 |
skirt | -0.010 |
makeup | -0.010 |
Token | We |
Token ID | 775 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
ver | +0.008 |
hand | +0.008 |
chief | +0.008 |
rolled | +0.008 |
happily | +0.007 |
Neg logit contributions | |
ey | -0.008 |
skirts | -0.008 |
fru | -0.008 |
ems | -0.007 |
makeup | -0.007 |
Token | just |
Token ID | 655 |
Feature activation | +4.98e-01 |
Pos logit contributions | |
hand | +0.044 |
ver | +0.042 |
rolled | +0.039 |
chief | +0.039 |
department | +0.039 |
Neg logit contributions | |
fru | -0.029 |
skirts | -0.028 |
ey | -0.027 |
ems | -0.026 |
makeup | -0.025 |
Token | want |
Token ID | 765 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
ver | +0.018 |
hand | +0.016 |
chief | +0.016 |
rolled | +0.015 |
department | +0.015 |
Neg logit contributions | |
ey | -0.016 |
skirts | -0.016 |
fru | -0.016 |
skirt | -0.014 |
ems | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
skirts | +0.013 |
ey | +0.013 |
fru | +0.013 |
skirt | +0.012 |
makeup | +0.012 |
Neg logit contributions | |
ver | -0.013 |
hand | -0.011 |
chief | -0.011 |
rolled | -0.011 |
Even | -0.010 |
Token | eat |
Token ID | 4483 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
ey | +0.006 |
skirts | +0.006 |
fru | +0.006 |
skirt | +0.006 |
makeup | +0.006 |
Neg logit contributions | |
ver | -0.008 |
hand | -0.007 |
chief | -0.007 |
department | -0.007 |
uttered | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
ey | +0.015 |
skirts | +0.015 |
fru | +0.014 |
skirt | +0.013 |
ems | +0.013 |
Neg logit contributions | |
ver | -0.012 |
hand | -0.011 |
chief | -0.011 |
happily | -0.010 |
sadly | -0.010 |
Token | ice |
Token ID | 4771 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
ver | +0.011 |
hand | +0.011 |
chief | +0.010 |
happily | +0.010 |
rolled | +0.010 |
Neg logit contributions | |
ey | -0.013 |
skirts | -0.012 |
ems | -0.011 |
fru | -0.011 |
skirt | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
ey | +0.023 |
skirts | +0.022 |
fru | +0.021 |
skirt | +0.020 |
makeup | +0.019 |
Neg logit contributions | |
ver | -0.035 |
hand | -0.032 |
department | -0.032 |
chief | -0.031 |
uttered | -0.030 |
Token | friend |
Token ID | 1545 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
ver | +0.012 |
hand | +0.011 |
rolled | +0.010 |
Even | +0.010 |
uttered | +0.010 |
Neg logit contributions | |
skirts | -0.008 |
ey | -0.007 |
fru | -0.007 |
skirt | -0.007 |
makeup | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
ver | +0.006 |
hand | +0.005 |
happily | +0.005 |
chief | +0.005 |
sadly | +0.005 |
Neg logit contributions | |
ey | -0.010 |
fru | -0.010 |
skirts | -0.010 |
skirt | -0.009 |
ems | -0.008 |
Token | But |
Token ID | 887 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
ver | +0.003 |
hand | +0.003 |
chief | +0.003 |
sadly | +0.003 |
happily | +0.003 |
Neg logit contributions | |
ey | -0.004 |
fru | -0.004 |
skirts | -0.004 |
skirt | -0.003 |
makeup | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
ver | +0.012 |
hand | +0.012 |
chief | +0.011 |
department | +0.011 |
Even | +0.011 |
Neg logit contributions | |
ey | -0.010 |
skirts | -0.010 |
fru | -0.009 |
skirt | -0.009 |
makeup | -0.008 |
Token | balloon |
Token ID | 21190 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
skirts | +0.028 |
ey | +0.027 |
fru | +0.027 |
skirt | +0.027 |
sunglasses | +0.025 |
Neg logit contributions | |
ver | -0.031 |
hand | -0.028 |
Even | -0.027 |
rolled | -0.026 |
chief | -0.026 |
Token | was |
Token ID | 373 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
ver | +0.012 |
hand | +0.012 |
Even | +0.011 |
chief | +0.011 |
rolled | +0.011 |
Neg logit contributions | |
fru | -0.012 |
ey | -0.012 |
skirts | -0.012 |
skirt | -0.011 |
makeup | -0.010 |
Token | so |
Token ID | 523 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
ver | +0.009 |
hand | +0.009 |
rolled | +0.008 |
chief | +0.008 |
Even | +0.008 |
Neg logit contributions | |
skirts | -0.009 |
ey | -0.009 |
fru | -0.009 |
skirt | -0.008 |
makeup | -0.008 |
Token | big |
Token ID | 1263 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
fru | +0.021 |
ey | +0.020 |
skirts | +0.020 |
pretty | +0.017 |
prett | +0.017 |
Neg logit contributions | |
hand | -0.035 |
ver | -0.035 |
chief | -0.033 |
department | -0.032 |
Even | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
ver | +0.020 |
hand | +0.017 |
chief | +0.017 |
happily | +0.017 |
Even | +0.016 |
Neg logit contributions | |
ey | -0.035 |
skirts | -0.033 |
fru | -0.033 |
skirt | -0.031 |
ems | -0.029 |
Token | it |
Token ID | 340 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
ey | +0.022 |
skirts | +0.021 |
fru | +0.020 |
skirt | +0.018 |
ems | +0.018 |
Neg logit contributions | |
ver | -0.024 |
hand | -0.022 |
chief | -0.021 |
happily | -0.021 |
sadly | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
rolled | +0.014 |
hand | +0.014 |
ver | +0.014 |
department | +0.014 |
uttered | +0.014 |
Neg logit contributions | |
skirts | -0.005 |
but | -0.004 |
ey | -0.004 |
fru | -0.004 |
than | -0.004 |
Token | She |
Token ID | 1375 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
ver | +0.003 |
hand | +0.003 |
chief | +0.003 |
Even | +0.003 |
happily | +0.003 |
Neg logit contributions | |
ey | -0.005 |
skirts | -0.005 |
fru | -0.005 |
skirt | -0.004 |
ems | -0.004 |
Token | climbs |
Token ID | 34332 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
ver | +0.016 |
hand | +0.015 |
rolled | +0.014 |
chief | +0.014 |
te | +0.014 |
Neg logit contributions | |
ey | -0.009 |
skirts | -0.008 |
fru | -0.008 |
skirt | -0.007 |
scar | -0.007 |
Token | out |
Token ID | 503 |
Feature activation | +9.02e-01 |
Pos logit contributions | |
ver | +0.010 |
chief | +0.009 |
Even | +0.008 |
department | +0.008 |
rolled | +0.008 |
Neg logit contributions | |
skirts | -0.009 |
ey | -0.008 |
fru | -0.008 |
jam | -0.008 |
ol | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
ver | +0.024 |
chief | +0.021 |
hand | +0.021 |
rolled | +0.020 |
Even | +0.019 |
Neg logit contributions | |
ey | -0.038 |
skirts | -0.038 |
fru | -0.036 |
skirt | -0.035 |
makeup | -0.033 |
Token | She |
Token ID | 1375 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
ey | +0.010 |
skirts | +0.010 |
fru | +0.009 |
skirt | +0.009 |
makeup | +0.008 |
Neg logit contributions | |
ver | -0.009 |
hand | -0.008 |
chief | -0.008 |
rolled | -0.007 |
department | -0.007 |
Token | is |
Token ID | 318 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
ver | +0.010 |
hand | +0.010 |
chief | +0.009 |
rolled | +0.009 |
te | +0.009 |
Neg logit contributions | |
ey | -0.005 |
skirts | -0.005 |
fru | -0.005 |
scar | -0.004 |
jam | -0.004 |
Token | wet |
Token ID | 9583 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
ver | +0.003 |
hand | +0.003 |
chief | +0.002 |
Even | +0.002 |
rolled | +0.002 |
Neg logit contributions | |
ey | -0.003 |
skirts | -0.003 |
fru | -0.003 |
skirt | -0.002 |
makeup | -0.002 |
Token | too |
Token ID | 1165 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
ey | +0.009 |
skirts | +0.009 |
fru | +0.009 |
skirt | +0.008 |
makeup | +0.008 |
Neg logit contributions | |
ver | -0.009 |
hand | -0.008 |
rolled | -0.008 |
chief | -0.008 |
Even | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
hand | +0.002 |
ver | +0.002 |
rolled | +0.002 |
chief | +0.001 |
department | +0.001 |
Neg logit contributions | |
skirts | -0.001 |
fru | -0.001 |
ey | -0.001 |
skirt | -0.001 |
jam | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ey | +0.007 |
skirts | +0.006 |
fru | +0.006 |
skirt | +0.006 |
ems | +0.006 |
Neg logit contributions | |
ver | -0.004 |
Even | -0.003 |
hand | -0.003 |
chief | -0.003 |
happily | -0.003 |
Token | young |
Token ID | 1862 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
ver | +0.011 |
hand | +0.010 |
Even | +0.010 |
rolled | +0.010 |
chief | +0.010 |
Neg logit contributions | |
ey | -0.009 |
fru | -0.009 |
skirts | -0.008 |
skirt | -0.008 |
makeup | -0.007 |
Token | boy |
Token ID | 2933 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
ver | +0.020 |
hand | +0.019 |
rolled | +0.019 |
uttered | +0.018 |
Even | +0.018 |
Neg logit contributions | |
fru | -0.013 |
ey | -0.013 |
skirts | -0.013 |
skirt | -0.012 |
da | -0.011 |
Token | in |
Token ID | 287 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
ver | +0.023 |
hand | +0.020 |
chief | +0.020 |
Even | +0.020 |
department | +0.019 |
Neg logit contributions | |
ey | -0.021 |
skirts | -0.021 |
fru | -0.019 |
skirt | -0.019 |
makeup | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
ey | +0.001 |
skirts | +0.001 |
fru | +0.001 |
skirt | +0.001 |
makeup | +0.001 |
Neg logit contributions | |
ver | -0.001 |
hand | -0.001 |
chief | -0.001 |
rolled | -0.001 |
department | -0.001 |
Token | ocean |
Token ID | 9151 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
ey | +0.010 |
skirts | +0.010 |
fru | +0.010 |
ems | +0.009 |
skirt | +0.009 |
Neg logit contributions | |
ver | -0.011 |
hand | -0.010 |
chief | -0.010 |
happily | -0.010 |
Even | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
ey | +0.020 |
skirts | +0.020 |
fru | +0.019 |
skirt | +0.018 |
makeup | +0.017 |
Neg logit contributions | |
ver | -0.021 |
hand | -0.019 |
chief | -0.018 |
Even | -0.018 |
rolled | -0.018 |
Token | The |
Token ID | 383 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
ey | +0.007 |
fru | +0.006 |
skirts | +0.006 |
ol | +0.005 |
--- | +0.005 |
Neg logit contributions | |
ver | -0.007 |
hand | -0.007 |
chief | -0.007 |
te | -0.006 |
department | -0.006 |
Token | sailor |
Token ID | 43272 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
ver | +0.006 |
hand | +0.006 |
rolled | +0.006 |
Even | +0.006 |
uttered | +0.006 |
Neg logit contributions | |
ey | -0.004 |
skirts | -0.004 |
fru | -0.004 |
skirt | -0.004 |
prett | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
fru | +0.001 |
skirts | +0.000 |
but | +0.000 |
doing | +0.000 |
da | +0.000 |
Neg logit contributions | |
hand | -0.001 |
ver | -0.001 |
itt | -0.001 |
uggle | -0.001 |
Even | -0.001 |
Token | in |
Token ID | 287 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
ver | +0.018 |
hand | +0.018 |
department | +0.016 |
brigade | +0.016 |
Even | +0.016 |
Neg logit contributions | |
doing | -0.012 |
skirts | -0.011 |
fru | -0.011 |
ey | -0.011 |
prett | -0.010 |
Token | danger |
Token ID | 3514 |
Feature activation | -1.27e-03 |
Pos logit contributions | |
ver | +0.001 |
hand | +0.001 |
chief | +0.001 |
rolled | +0.001 |
department | +0.001 |
Neg logit contributions | |
ey | -0.001 |
skirts | -0.001 |
fru | -0.001 |
skirt | -0.001 |
makeup | -0.001 |
Token | girl |
Token ID | 2576 |
Feature activation | +6.78e-03 |
Pos logit contributions | |
ver | +0.011 |
hand | +0.011 |
rolled | +0.010 |
chief | +0.010 |
Even | +0.010 |
Neg logit contributions | |
ey | -0.006 |
skirts | -0.006 |
fru | -0.006 |
skirt | -0.005 |
makeup | -0.005 |
Token | named |
Token ID | 3706 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
ver | +0.000 |
hand | +0.000 |
chief | +0.000 |
department | +0.000 |
rolled | +0.000 |
Neg logit contributions | |
ey | -0.000 |
skirts | -0.000 |
fru | -0.000 |
skirt | -0.000 |
makeup | -0.000 |
Token | Amy |
Token ID | 14235 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
ver | +0.015 |
hand | +0.014 |
chief | +0.014 |
department | +0.013 |
rolled | +0.013 |
Neg logit contributions | |
ey | -0.009 |
skirts | -0.009 |
fru | -0.009 |
skirt | -0.008 |
makeup | -0.008 |
Token | went |
Token ID | 1816 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ey | +0.008 |
skirts | +0.008 |
fru | +0.008 |
skirt | +0.007 |
ems | +0.007 |
Neg logit contributions | |
ver | -0.005 |
hand | -0.004 |
happily | -0.004 |
chief | -0.004 |
Even | -0.004 |
Token | for |
Token ID | 329 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
ver | +0.005 |
hand | +0.004 |
chief | +0.004 |
Even | +0.004 |
rolled | +0.004 |
Neg logit contributions | |
ey | -0.006 |
skirts | -0.006 |
fru | -0.006 |
skirt | -0.006 |
makeup | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
ey | +0.011 |
fru | +0.010 |
skirts | +0.010 |
ems | +0.010 |
skirt | +0.009 |
Neg logit contributions | |
ver | -0.003 |
hand | -0.003 |
happily | -0.003 |
quickly | -0.002 |
safely | -0.002 |
Token | walk |
Token ID | 2513 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
ver | +0.002 |
hand | +0.002 |
chief | +0.002 |
rolled | +0.002 |
Even | +0.002 |
Neg logit contributions | |
ey | -0.002 |
skirts | -0.002 |
fru | -0.002 |
skirt | -0.001 |
ems | -0.001 |
Token | with |
Token ID | 351 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
skirts | +0.017 |
ey | +0.016 |
fru | +0.016 |
skirt | +0.015 |
makeup | +0.015 |
Neg logit contributions | |
ver | -0.018 |
hand | -0.016 |
chief | -0.016 |
department | -0.016 |
Even | -0.015 |
Token | her |
Token ID | 607 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
ey | +0.014 |
fru | +0.013 |
skirts | +0.013 |
ems | +0.012 |
skirt | +0.011 |
Neg logit contributions | |
ver | -0.020 |
hand | -0.019 |
happily | -0.019 |
chief | -0.019 |
Even | -0.019 |
Token | mom |
Token ID | 1995 |
Feature activation | -9.33e-01 |
Pos logit contributions | |
ey | +0.008 |
skirts | +0.008 |
fru | +0.008 |
skirt | +0.007 |
ems | +0.007 |
Neg logit contributions | |
ver | -0.004 |
hand | -0.003 |
chief | -0.003 |
happily | -0.003 |
rolled | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ey | +0.045 |
fru | +0.043 |
skirts | +0.040 |
disl | +0.038 |
ems | +0.037 |
Neg logit contributions | |
ver | -0.022 |
happily | -0.019 |
hand | -0.019 |
quickly | -0.018 |
chief | -0.018 |
Token | they |
Token ID | 484 |
Feature activation | +8.16e-01 |
Pos logit contributions | |
ey | +0.015 |
skirts | +0.014 |
fru | +0.014 |
skirt | +0.013 |
ems | +0.012 |
Neg logit contributions | |
ver | -0.017 |
hand | -0.016 |
chief | -0.016 |
happily | -0.015 |
Even | -0.015 |
Token | could |
Token ID | 714 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
hand | +0.035 |
ver | +0.034 |
chief | +0.034 |
rolled | +0.033 |
Even | +0.033 |
Neg logit contributions | |
skirts | -0.019 |
fru | -0.016 |
--- | -0.015 |
skirt | -0.015 |
ey | -0.015 |
Token | play |
Token ID | 711 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
ver | +0.032 |
year | +0.031 |
hand | +0.030 |
chief | +0.030 |
uttered | +0.029 |
Neg logit contributions | |
wear | -0.013 |
skirts | -0.013 |
skirt | -0.012 |
fru | -0.012 |
ey | -0.012 |
Token | together |
Token ID | 1978 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
ver | +0.008 |
hand | +0.007 |
chief | +0.007 |
happily | +0.007 |
sadly | +0.006 |
Neg logit contributions | |
ey | -0.014 |
skirts | -0.013 |
fru | -0.013 |
skirt | -0.012 |
ems | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
ver | +0.004 |
hand | +0.004 |
rolled | +0.003 |
chief | +0.003 |
department | +0.003 |
Neg logit contributions | |
skirts | -0.004 |
ey | -0.004 |
fru | -0.003 |
skirt | -0.003 |
makeup | -0.003 |
Token | be |
Token ID | 307 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
ver | +0.007 |
hand | +0.007 |
chief | +0.007 |
happily | +0.007 |
sadly | +0.006 |
Neg logit contributions | |
ey | -0.008 |
skirts | -0.008 |
fru | -0.008 |
skirt | -0.007 |
ems | -0.007 |
Token | happy |
Token ID | 3772 |
Feature activation | -6.93e-03 |
Pos logit contributions | |
ey | +0.083 |
skirts | +0.078 |
fru | +0.076 |
skirt | +0.070 |
ems | +0.068 |
Neg logit contributions | |
ver | -0.079 |
hand | -0.072 |
chief | -0.071 |
happily | -0.071 |
sadly | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +6.48e-03 |
Pos logit contributions | |
ey | +0.000 |
skirts | +0.000 |
fru | +0.000 |
skirt | +0.000 |
makeup | +0.000 |
Neg logit contributions | |
ver | -0.000 |
hand | -0.000 |
chief | -0.000 |
rolled | -0.000 |
Even | -0.000 |
Token | She |
Token ID | 1375 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
ver | +0.000 |
hand | +0.000 |
Even | +0.000 |
chief | +0.000 |
happily | +0.000 |
Neg logit contributions | |
ey | -0.000 |
skirts | -0.000 |
fru | -0.000 |
skirt | -0.000 |
ems | -0.000 |
Token | hugged |
Token ID | 41130 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
ver | +0.005 |
hand | +0.004 |
chief | +0.004 |
rolled | +0.004 |
department | +0.004 |
Neg logit contributions | |
ey | -0.004 |
skirts | -0.004 |
fru | -0.004 |
skirt | -0.003 |
ems | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
ver | +0.015 |
hand | +0.014 |
chief | +0.014 |
rolled | +0.014 |
department | +0.013 |
Neg logit contributions | |
skirts | -0.011 |
ey | -0.011 |
fru | -0.010 |
skirt | -0.010 |
makeup | -0.009 |
Token | bald |
Token ID | 28273 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
ey | +0.006 |
fru | +0.006 |
skirts | +0.006 |
skirt | +0.005 |
--- | +0.005 |
Neg logit contributions | |
ver | -0.004 |
happily | -0.004 |
hand | -0.004 |
chief | -0.004 |
quickly | -0.003 |
Token | old |
Token ID | 1468 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
ey | +0.007 |
skirts | +0.007 |
fru | +0.007 |
skirt | +0.006 |
ems | +0.006 |
Neg logit contributions | |
ver | -0.009 |
hand | -0.008 |
chief | -0.008 |
rolled | -0.008 |
happily | -0.008 |
Token | men |
Token ID | 1450 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
ey | +0.023 |
skirts | +0.022 |
fru | +0.021 |
skirt | +0.019 |
ems | +0.019 |
Neg logit contributions | |
ver | -0.025 |
hand | -0.023 |
chief | -0.022 |
happily | -0.022 |
Even | -0.021 |
Token | should |
Token ID | 815 |
Feature activation | +7.52e-01 |
Pos logit contributions | |
ey | +0.013 |
skirts | +0.012 |
fru | +0.012 |
skirt | +0.011 |
ems | +0.011 |
Neg logit contributions | |
ver | -0.012 |
hand | -0.011 |
chief | -0.011 |
rolled | -0.011 |
Even | -0.010 |
Token | not |
Token ID | 407 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
ver | +0.027 |
hand | +0.026 |
chief | +0.025 |
department | +0.025 |
rolled | +0.024 |
Neg logit contributions | |
ey | -0.023 |
skirts | -0.023 |
fru | -0.021 |
skirt | -0.021 |
makeup | -0.020 |
Token | be |
Token ID | 307 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
ver | +0.014 |
hand | +0.013 |
chief | +0.012 |
rolled | +0.012 |
department | +0.012 |
Neg logit contributions | |
ey | -0.011 |
skirts | -0.011 |
fru | -0.010 |
skirt | -0.010 |
makeup | -0.009 |
Token | afraid |
Token ID | 7787 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
ey | +0.067 |
skirts | +0.065 |
fru | +0.063 |
skirt | +0.058 |
ems | +0.056 |
Neg logit contributions | |
ver | -0.066 |
hand | -0.060 |
chief | -0.059 |
happily | -0.056 |
rolled | -0.056 |
Token | to |
Token ID | 284 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
ver | +0.011 |
hand | +0.010 |
chief | +0.010 |
rolled | +0.009 |
Even | +0.009 |
Neg logit contributions | |
skirts | -0.020 |
ey | -0.020 |
fru | -0.019 |
skirt | -0.018 |
makeup | -0.018 |
Token | go |
Token ID | 467 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
ey | +0.017 |
skirts | +0.016 |
fru | +0.016 |
skirt | +0.014 |
ems | +0.014 |
Neg logit contributions | |
ver | -0.017 |
hand | -0.015 |
chief | -0.015 |
happily | -0.014 |
Even | -0.014 |
Token | out |
Token ID | 503 |
Feature activation | +7.50e-01 |
Pos logit contributions | |
ey | +0.006 |
skirts | +0.006 |
fru | +0.006 |
skirt | +0.005 |
makeup | +0.005 |
Neg logit contributions | |
ver | -0.005 |
hand | -0.004 |
chief | -0.004 |
Even | -0.004 |
rolled | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +8.86e-01 |
Pos logit contributions | |
ver | +0.022 |
hand | +0.020 |
chief | +0.019 |
rolled | +0.018 |
Even | +0.018 |
Neg logit contributions | |
ey | -0.029 |
skirts | -0.029 |
fru | -0.028 |
skirt | -0.026 |
ems | -0.025 |
Token | ted |
Token ID | 1513 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
skirts | +0.006 |
fru | +0.006 |
ey | +0.006 |
skirt | +0.005 |
sunglasses | +0.005 |
Neg logit contributions | |
ver | -0.007 |
chief | -0.007 |
case | -0.007 |
department | -0.007 |
Even | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
ver | +0.063 |
hand | +0.057 |
department | +0.057 |
chief | +0.056 |
rolled | +0.055 |
Neg logit contributions | |
skirts | -0.045 |
ey | -0.044 |
fru | -0.043 |
skirt | -0.040 |
ol | -0.039 |
Token | rabbit |
Token ID | 22746 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
ver | +0.009 |
hand | +0.008 |
chief | +0.008 |
happily | +0.008 |
department | +0.008 |
Neg logit contributions | |
ey | -0.012 |
skirts | -0.011 |
fru | -0.011 |
ems | -0.010 |
skirt | -0.010 |
Token | softly |
Token ID | 26625 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
ver | +0.012 |
hand | +0.011 |
chief | +0.010 |
rolled | +0.010 |
department | +0.010 |
Neg logit contributions | |
skirts | -0.012 |
ey | -0.011 |
fru | -0.011 |
skirt | -0.011 |
makeup | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -2.32e-03 |
Pos logit contributions | |
ver | +0.027 |
hand | +0.025 |
chief | +0.024 |
happily | +0.024 |
sadly | +0.023 |
Neg logit contributions | |
ey | -0.036 |
skirts | -0.035 |
fru | -0.034 |
skirt | -0.031 |
ems | -0.030 |
Token | To |
Token ID | 1675 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
fru | +0.000 |
ey | +0.000 |
skirts | +0.000 |
ol | +0.000 |
makeup | +0.000 |
Neg logit contributions | |
ver | -0.000 |
chief | -0.000 |
hand | -0.000 |
te | -0.000 |
department | -0.000 |
Token | his |
Token ID | 465 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
ey | +0.048 |
skirts | +0.047 |
fru | +0.046 |
skirt | +0.042 |
ems | +0.040 |
Neg logit contributions | |
ver | -0.053 |
hand | -0.050 |
chief | -0.048 |
department | -0.046 |
rolled | -0.046 |
Token | surprise |
Token ID | 5975 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
ver | +0.027 |
chief | +0.024 |
hand | +0.023 |
rolled | +0.023 |
department | +0.022 |
Neg logit contributions | |
ey | -0.018 |
skirts | -0.017 |
skirt | -0.016 |
fru | -0.016 |
makeup | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
ey | +0.053 |
skirts | +0.052 |
fru | +0.050 |
skirt | +0.047 |
ems | +0.046 |
Neg logit contributions | |
ver | -0.035 |
hand | -0.031 |
chief | -0.030 |
happily | -0.029 |
sadly | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
hand | +0.009 |
ver | +0.009 |
chief | +0.009 |
rolled | +0.009 |
department | +0.009 |
Neg logit contributions | |
fru | -0.006 |
ey | -0.006 |
skirts | -0.005 |
skirt | -0.005 |
ol | -0.005 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
ey | +0.016 |
skirts | +0.015 |
fru | +0.015 |
skirt | +0.013 |
ems | +0.013 |
Neg logit contributions | |
ver | -0.014 |
hand | -0.013 |
chief | -0.013 |
happily | -0.012 |
Even | -0.012 |
Token | he |
Token ID | 339 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
ver | +0.003 |
hand | +0.003 |
rolled | +0.003 |
chief | +0.003 |
Even | +0.003 |
Neg logit contributions | |
skirts | -0.003 |
fru | -0.003 |
ey | -0.003 |
skirt | -0.003 |
ems | -0.003 |
Token | had |
Token ID | 550 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
fru | +0.008 |
skirts | +0.007 |
prett | +0.006 |
skirt | +0.006 |
sunglasses | +0.006 |
Neg logit contributions | |
hand | -0.016 |
ver | -0.015 |
itt | -0.015 |
rolled | -0.014 |
Even | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
skirts | +0.023 |
ey | +0.023 |
fru | +0.023 |
skirt | +0.021 |
makeup | +0.021 |
Neg logit contributions | |
ver | -0.017 |
hand | -0.015 |
chief | -0.015 |
rolled | -0.014 |
department | -0.014 |
Token | leave |
Token ID | 2666 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
ey | +0.036 |
skirts | +0.035 |
fru | +0.034 |
skirt | +0.032 |
makeup | +0.031 |
Neg logit contributions | |
ver | -0.040 |
hand | -0.037 |
chief | -0.036 |
department | -0.035 |
rolled | -0.035 |
Token | it |
Token ID | 340 |
Feature activation | -7.73e-02 |
Pos logit contributions | |
ver | +0.015 |
hand | +0.014 |
chief | +0.013 |
rolled | +0.013 |
department | +0.013 |
Neg logit contributions | |
ey | -0.014 |
skirts | -0.014 |
fru | -0.014 |
skirt | -0.012 |
makeup | -0.012 |
Token | be |
Token ID | 307 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
ey | +0.003 |
skirts | +0.002 |
fru | +0.002 |
skirt | +0.002 |
makeup | +0.002 |
Neg logit contributions | |
ver | -0.003 |
hand | -0.003 |
chief | -0.002 |
rolled | -0.002 |
Even | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
ey | +0.097 |
skirts | +0.089 |
fru | +0.087 |
skirt | +0.081 |
ems | +0.079 |
Neg logit contributions | |
ver | -0.079 |
happily | -0.072 |
hand | -0.071 |
chief | -0.070 |
sadly | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
ey | +0.002 |
skirts | +0.002 |
fru | +0.002 |
skirt | +0.002 |
ems | +0.001 |
Neg logit contributions | |
ver | -0.001 |
hand | -0.001 |
chief | -0.001 |
department | -0.001 |
happily | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
ey | +0.007 |
skirts | +0.006 |
fru | +0.006 |
skirt | +0.006 |
ems | +0.006 |
Neg logit contributions | |
ver | -0.005 |
hand | -0.004 |
chief | -0.004 |
department | -0.004 |
rolled | -0.004 |
Token | That |
Token ID | 2504 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
ey | +0.007 |
skirts | +0.007 |
fru | +0.007 |
skirt | +0.006 |
makeup | +0.006 |
Neg logit contributions | |
ver | -0.006 |
hand | -0.005 |
chief | -0.005 |
Even | -0.005 |
rolled | -0.005 |
Token | night |
Token ID | 1755 |
Feature activation | -7.91e-01 |
Pos logit contributions | |
ver | +0.005 |
hand | +0.004 |
rolled | +0.004 |
chief | +0.004 |
department | +0.004 |
Neg logit contributions | |
skirts | -0.003 |
ey | -0.003 |
fru | -0.003 |
skirt | -0.003 |
ems | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | -7.00e-01 |
Pos logit contributions | |
ver | +0.037 |
happily | +0.033 |
chief | +0.033 |
hand | +0.033 |
Even | +0.032 |
Neg logit contributions | |
ey | -0.037 |
skirts | -0.036 |
fru | -0.034 |
ems | -0.030 |
skirt | -0.030 |
Token | on |
Token ID | 319 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ey | +0.024 |
skirts | +0.023 |
fru | +0.022 |
skirt | +0.021 |
ems | +0.020 |
Neg logit contributions | |
ver | -0.024 |
hand | -0.022 |
chief | -0.022 |
rolled | -0.021 |
department | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
ver | +0.031 |
hand | +0.028 |
chief | +0.027 |
happily | +0.026 |
Even | +0.025 |
Neg logit contributions | |
ey | -0.048 |
skirts | -0.047 |
fru | -0.046 |
skirt | -0.043 |
ems | -0.042 |
Token | Tim |
Token ID | 5045 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
ver | +0.004 |
hand | +0.004 |
chief | +0.004 |
rolled | +0.003 |
department | +0.003 |
Neg logit contributions | |
skirts | -0.004 |
fru | -0.004 |
ey | -0.004 |
ol | -0.003 |
makeup | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
ey | +0.018 |
skirts | +0.017 |
fru | +0.017 |
skirt | +0.016 |
ems | +0.015 |
Neg logit contributions | |
ver | -0.016 |
hand | -0.015 |
chief | -0.014 |
rolled | -0.014 |
department | -0.013 |
Token | Sue |
Token ID | 26089 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ey | +0.029 |
skirts | +0.028 |
fru | +0.027 |
skirt | +0.025 |
ems | +0.024 |
Neg logit contributions | |
ver | -0.025 |
hand | -0.023 |
chief | -0.022 |
happily | -0.021 |
Even | -0.021 |
Token | would |
Token ID | 561 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
skirts | +0.052 |
ey | +0.050 |
fru | +0.049 |
skirt | +0.046 |
ems | +0.046 |
Neg logit contributions | |
hand | -0.059 |
ver | -0.058 |
chief | -0.052 |
department | -0.051 |
rolled | -0.051 |
Token | meet |
Token ID | 1826 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
ver | +0.037 |
hand | +0.035 |
chief | +0.032 |
keeper | +0.031 |
department | +0.031 |
Neg logit contributions | |
ey | -0.027 |
skirts | -0.027 |
fru | -0.024 |
skirt | -0.024 |
ems | -0.023 |
Token | every |
Token ID | 790 |
Feature activation | +7.24e-03 |
Pos logit contributions | |
skirts | +0.007 |
ey | +0.007 |
fru | +0.007 |
skirt | +0.007 |
da | +0.006 |
Neg logit contributions | |
ver | -0.007 |
hand | -0.007 |
rolled | -0.007 |
chief | -0.006 |
tow | -0.006 |
Token | night |
Token ID | 1755 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
ver | +0.000 |
hand | +0.000 |
rolled | +0.000 |
Even | +0.000 |
chief | +0.000 |
Neg logit contributions | |
ey | -0.000 |
skirts | -0.000 |
fru | -0.000 |
skirt | -0.000 |
makeup | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
ver | +0.023 |
hand | +0.021 |
chief | +0.020 |
happily | +0.019 |
rolled | +0.019 |
Neg logit contributions | |
ey | -0.030 |
skirts | -0.029 |
fru | -0.028 |
skirt | -0.026 |
ems | -0.025 |