Token | , |
Token ID | 11 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
badges | +0.060 |
ows | +0.059 |
tricks | +0.057 |
diapers | +0.056 |
names | +0.056 |
Neg logit contributions | |
least | -0.035 |
Little | -0.031 |
ime | -0.029 |
illy | -0.028 |
ately | -0.026 |
Token | sauce |
Token ID | 10746 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
badges | +0.012 |
ows | +0.012 |
tricks | +0.011 |
names | +0.011 |
diapers | +0.011 |
Neg logit contributions | |
least | -0.015 |
Little | -0.014 |
ime | -0.013 |
illy | -0.013 |
ately | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | +7.14e-01 |
Pos logit contributions | |
badges | +0.123 |
ows | +0.121 |
tricks | +0.117 |
names | +0.116 |
diapers | +0.115 |
Neg logit contributions | |
least | -0.068 |
Little | -0.060 |
ime | -0.056 |
illy | -0.053 |
Maria | -0.049 |
Token | cheese |
Token ID | 9891 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
badges | +0.026 |
ows | +0.025 |
tricks | +0.024 |
names | +0.024 |
chips | +0.024 |
Neg logit contributions | |
least | -0.024 |
Little | -0.022 |
ime | -0.022 |
illy | -0.020 |
Maria | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
badges | +0.148 |
ows | +0.142 |
tricks | +0.138 |
cuts | +0.138 |
diapers | +0.138 |
Neg logit contributions | |
least | -0.053 |
Little | -0.042 |
ime | -0.040 |
illy | -0.039 |
ill | -0.035 |
Token | topp |
Token ID | 23126 |
Feature activation | +8.18e+00 |
Pos logit contributions | |
ows | +0.034 |
badges | +0.034 |
tricks | +0.032 |
orts | +0.031 |
names | +0.031 |
Neg logit contributions | |
least | -0.041 |
Little | -0.037 |
ime | -0.035 |
illy | -0.034 |
Maria | -0.033 |
Token | ings |
Token ID | 654 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
badges | +0.253 |
trained | +0.230 |
facts | +0.221 |
homework | +0.217 |
cuts | +0.215 |
Neg logit contributions | |
ime | -0.289 |
ill | -0.273 |
illy | -0.266 |
oh | -0.264 |
least | -0.259 |
Token | . |
Token ID | 13 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
badges | +0.152 |
ows | +0.145 |
tricks | +0.145 |
hides | +0.144 |
cks | +0.142 |
Neg logit contributions | |
least | -0.090 |
Little | -0.085 |
illy | -0.082 |
ime | -0.079 |
Little | -0.078 |
Token | She |
Token ID | 1375 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
badges | +0.017 |
ows | +0.016 |
tricks | +0.016 |
aches | +0.016 |
cks | +0.016 |
Neg logit contributions | |
least | -0.018 |
Little | -0.017 |
Little | -0.016 |
illy | -0.016 |
ime | -0.016 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
badges | +0.027 |
ows | +0.026 |
tricks | +0.025 |
names | +0.025 |
diapers | +0.025 |
Neg logit contributions | |
least | -0.023 |
Little | -0.021 |
ime | -0.020 |
illy | -0.019 |
ately | -0.018 |
Token | them |
Token ID | 606 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
least | +0.034 |
Little | +0.032 |
illy | +0.031 |
ime | +0.030 |
Little | +0.030 |
Neg logit contributions | |
ows | -0.041 |
badges | -0.041 |
tricks | -0.039 |
gag | -0.038 |
cks | -0.038 |
Token | asks |
Token ID | 7893 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
badges | +0.035 |
ows | +0.033 |
diapers | +0.031 |
tricks | +0.031 |
orts | +0.030 |
Neg logit contributions | |
least | -0.048 |
Little | -0.044 |
ime | -0.042 |
illy | -0.041 |
Maria | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
badges | +0.066 |
ows | +0.065 |
tricks | +0.061 |
names | +0.061 |
diapers | +0.061 |
Neg logit contributions | |
least | -0.067 |
Little | -0.061 |
ime | -0.058 |
illy | -0.055 |
Maria | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
badges | +0.114 |
ows | +0.112 |
names | +0.109 |
tricks | +0.107 |
diapers | +0.107 |
Neg logit contributions | |
least | -0.073 |
Little | -0.061 |
ime | -0.061 |
illy | -0.057 |
ately | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
badges | +0.046 |
ows | +0.046 |
orts | +0.046 |
names | +0.045 |
tricks | +0.044 |
Neg logit contributions | |
least | -0.039 |
ime | -0.031 |
ately | -0.031 |
illy | -0.030 |
Little | -0.030 |
Token | Max |
Token ID | 11518 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
badges | +0.046 |
ows | +0.045 |
tricks | +0.044 |
names | +0.043 |
diapers | +0.043 |
Neg logit contributions | |
least | -0.031 |
Little | -0.027 |
ime | -0.026 |
illy | -0.025 |
ately | -0.023 |
Token | bar |
Token ID | 2318 |
Feature activation | +8.09e+00 |
Pos logit contributions | |
bones | +0.044 |
ours | +0.043 |
badges | +0.043 |
aches | +0.042 |
diapers | +0.042 |
Neg logit contributions | |
least | -0.052 |
Little | -0.045 |
Little | -0.042 |
mid | -0.042 |
ill | -0.041 |
Token | ks |
Token ID | 591 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
badges | +0.156 |
diapers | +0.135 |
tricks | +0.130 |
chips | +0.128 |
ows | +0.128 |
Neg logit contributions | |
least | -0.393 |
ime | -0.372 |
ill | -0.368 |
Little | -0.358 |
ately | -0.355 |
Token | and |
Token ID | 290 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
badges | +0.234 |
cuts | +0.226 |
ows | +0.224 |
aches | +0.220 |
diapers | +0.219 |
Neg logit contributions | |
least | -0.080 |
Little | -0.076 |
ime | -0.072 |
ill | -0.067 |
mid | -0.065 |
Token | nods |
Token ID | 34833 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
ages | +0.058 |
ows | +0.057 |
uries | +0.052 |
aches | +0.052 |
ves | +0.051 |
Neg logit contributions | |
least | -0.052 |
M | -0.044 |
mid | -0.043 |
Little | -0.043 |
tw | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
badges | +0.066 |
ows | +0.064 |
diapers | +0.061 |
tricks | +0.061 |
names | +0.061 |
Neg logit contributions | |
least | -0.046 |
Little | -0.042 |
ime | -0.039 |
illy | -0.038 |
Maria | -0.036 |
Token | He |
Token ID | 679 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
badges | +0.058 |
ows | +0.057 |
aches | +0.055 |
cks | +0.054 |
tricks | +0.054 |
Neg logit contributions | |
least | -0.045 |
Little | -0.043 |
ime | -0.039 |
Little | -0.039 |
Maria | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
ows | +0.109 |
badges | +0.108 |
tricks | +0.101 |
diapers | +0.100 |
names | +0.100 |
Neg logit contributions | |
least | -0.090 |
Little | -0.085 |
ime | -0.078 |
illy | -0.076 |
Maria | -0.074 |
Token | princess |
Token ID | 21752 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
badges | +0.108 |
ows | +0.107 |
tricks | +0.101 |
diapers | +0.100 |
names | +0.100 |
Neg logit contributions | |
least | -0.085 |
Little | -0.076 |
ime | -0.073 |
illy | -0.069 |
ately | -0.066 |
Token | says |
Token ID | 1139 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
badges | +0.138 |
ows | +0.131 |
diapers | +0.129 |
tricks | +0.127 |
names | +0.125 |
Neg logit contributions | |
least | -0.118 |
Little | -0.104 |
ime | -0.102 |
illy | -0.097 |
ately | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
badges | +0.080 |
ows | +0.076 |
diapers | +0.074 |
tricks | +0.074 |
names | +0.073 |
Neg logit contributions | |
least | -0.057 |
Little | -0.051 |
ime | -0.049 |
illy | -0.047 |
ately | -0.045 |
Token | She |
Token ID | 1375 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
badges | +0.144 |
ows | +0.142 |
names | +0.136 |
tricks | +0.135 |
diapers | +0.134 |
Neg logit contributions | |
least | -0.111 |
Little | -0.098 |
ime | -0.095 |
illy | -0.090 |
ately | -0.086 |
Token | pret |
Token ID | 2181 |
Feature activation | +7.92e+00 |
Pos logit contributions | |
badges | +0.070 |
ows | +0.068 |
diapers | +0.068 |
orts | +0.066 |
bones | +0.064 |
Neg logit contributions | |
least | -0.115 |
Little | -0.101 |
ime | -0.095 |
illy | -0.094 |
ill | -0.091 |
Token | ends |
Token ID | 2412 |
Feature activation | +4.99e+00 |
Pos logit contributions | |
diapers | +0.175 |
preferences | +0.173 |
facts | +0.166 |
choices | +0.164 |
bows | +0.162 |
Neg logit contributions | |
least | -0.316 |
oh | -0.315 |
Little | -0.296 |
M | -0.291 |
ately | -0.291 |
Token | to |
Token ID | 284 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
badges | +0.236 |
ows | +0.234 |
orts | +0.222 |
cuts | +0.221 |
diapers | +0.220 |
Neg logit contributions | |
least | -0.100 |
Little | -0.081 |
ime | -0.081 |
Maria | -0.073 |
illy | -0.071 |
Token | wear |
Token ID | 5806 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
ows | +0.053 |
badges | +0.053 |
cuts | +0.050 |
diapers | +0.049 |
orts | +0.048 |
Neg logit contributions | |
least | -0.045 |
Little | -0.037 |
ime | -0.037 |
ill | -0.035 |
mid | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | +7.29e-02 |
Pos logit contributions | |
gag | +0.059 |
ows | +0.058 |
trained | +0.058 |
orts | +0.056 |
ashes | +0.056 |
Neg logit contributions | |
least | -0.034 |
Little | -0.032 |
Little | -0.030 |
M | -0.029 |
illy | -0.028 |
Token | hat |
Token ID | 6877 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
ows | +0.002 |
badges | +0.002 |
orts | +0.002 |
cks | +0.002 |
diapers | +0.002 |
Neg logit contributions | |
least | -0.003 |
Little | -0.002 |
ime | -0.002 |
illy | -0.002 |
3 | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
badges | +0.076 |
ows | +0.076 |
names | +0.072 |
diapers | +0.072 |
tricks | +0.071 |
Neg logit contributions | |
least | -0.061 |
Little | -0.053 |
ime | -0.052 |
illy | -0.049 |
ately | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
ows | +0.016 |
orts | +0.015 |
badges | +0.015 |
names | +0.014 |
tricks | +0.014 |
Neg logit contributions | |
least | -0.015 |
ime | -0.012 |
illy | -0.011 |
bul | -0.011 |
ately | -0.011 |
Token | " |
Token ID | 1 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
badges | +0.020 |
ows | +0.019 |
tricks | +0.019 |
names | +0.019 |
diapers | +0.018 |
Neg logit contributions | |
least | -0.014 |
Little | -0.013 |
ime | -0.012 |
illy | -0.012 |
Maria | -0.011 |
Token | O |
Token ID | 46 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
badges | +0.048 |
ows | +0.046 |
tricks | +0.045 |
diapers | +0.045 |
names | +0.045 |
Neg logit contributions | |
least | -0.036 |
Little | -0.031 |
ime | -0.031 |
illy | -0.030 |
ately | -0.028 |
Token | st |
Token ID | 301 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
least | +0.004 |
Little | +0.004 |
ime | +0.003 |
illy | +0.003 |
ately | +0.003 |
Neg logit contributions | |
badges | -0.004 |
names | -0.004 |
tricks | -0.004 |
diapers | -0.004 |
ows | -0.004 |
Token | ric |
Token ID | 1173 |
Feature activation | +7.88e+00 |
Pos logit contributions | |
badges | +0.013 |
diapers | +0.012 |
helmets | +0.012 |
facts | +0.012 |
tricks | +0.012 |
Neg logit contributions | |
least | -0.012 |
ime | -0.012 |
Little | -0.010 |
ately | -0.010 |
illy | -0.010 |
Token | hes |
Token ID | 956 |
Feature activation | +3.40e+00 |
Pos logit contributions | |
badges | +0.147 |
diapers | +0.121 |
chips | +0.110 |
tricks | +0.109 |
names | +0.109 |
Neg logit contributions | |
least | -0.401 |
ime | -0.384 |
Little | -0.383 |
illy | -0.358 |
Maria | -0.358 |
Token | do |
Token ID | 466 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
badges | +0.109 |
ows | +0.103 |
diapers | +0.099 |
tricks | +0.099 |
names | +0.098 |
Neg logit contributions | |
least | -0.129 |
Little | -0.118 |
ime | -0.116 |
illy | -0.110 |
ately | -0.106 |
Token | not |
Token ID | 407 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
badges | +0.064 |
diapers | +0.058 |
ows | +0.056 |
gag | +0.056 |
irms | +0.056 |
Neg logit contributions | |
least | -0.079 |
Little | -0.074 |
ime | -0.067 |
oh | -0.062 |
' | -0.062 |
Token | like |
Token ID | 588 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
ows | +0.055 |
badges | +0.054 |
diapers | +0.052 |
bes | +0.051 |
orts | +0.050 |
Neg logit contributions | |
least | -0.058 |
Little | -0.053 |
ime | -0.052 |
illy | -0.046 |
ful | -0.046 |
Token | people |
Token ID | 661 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
badges | +0.089 |
ows | +0.087 |
tricks | +0.083 |
names | +0.082 |
diapers | +0.082 |
Neg logit contributions | |
least | -0.079 |
Little | -0.072 |
ime | -0.068 |
illy | -0.066 |
ately | -0.063 |
Token | us |
Token ID | 514 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
ows | +0.075 |
badges | +0.071 |
bows | +0.069 |
names | +0.068 |
tricks | +0.068 |
Neg logit contributions | |
least | -0.033 |
Little | -0.029 |
ime | -0.025 |
illy | -0.024 |
Little | -0.021 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
badges | +0.008 |
ows | +0.008 |
tricks | +0.008 |
names | +0.008 |
diapers | +0.008 |
Neg logit contributions | |
least | -0.007 |
Little | -0.007 |
ime | -0.006 |
illy | -0.006 |
Maria | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +7.93e-01 |
Pos logit contributions | |
badges | +0.118 |
ows | +0.116 |
tricks | +0.112 |
names | +0.112 |
diapers | +0.112 |
Neg logit contributions | |
least | -0.043 |
Little | -0.039 |
ime | -0.034 |
illy | -0.031 |
Maria | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +7.33e-01 |
Pos logit contributions | |
ows | +0.034 |
badges | +0.034 |
tricks | +0.033 |
names | +0.033 |
orts | +0.033 |
Neg logit contributions | |
least | -0.021 |
Little | -0.017 |
ime | -0.017 |
illy | -0.016 |
ately | -0.016 |
Token | Spot |
Token ID | 32565 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
badges | +0.032 |
ows | +0.031 |
tricks | +0.030 |
names | +0.030 |
diapers | +0.030 |
Neg logit contributions | |
least | -0.019 |
Little | -0.018 |
ime | -0.016 |
illy | -0.015 |
Maria | -0.014 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.73e+00 |
Pos logit contributions | |
bones | +0.112 |
ages | +0.108 |
aches | +0.106 |
ours | +0.105 |
badges | +0.105 |
Neg logit contributions | |
least | -0.101 |
and | -0.084 |
again | -0.084 |
Bear | -0.083 |
Little | -0.081 |
Token | ks |
Token ID | 591 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
badges | +0.175 |
diapers | +0.156 |
bes | +0.155 |
chips | +0.150 |
helmets | +0.147 |
Neg logit contributions | |
least | -0.342 |
ill | -0.335 |
ime | -0.327 |
in | -0.315 |
ately | -0.309 |
Token | . |
Token ID | 13 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
badges | +0.213 |
ows | +0.204 |
cuts | +0.203 |
diapers | +0.198 |
bows | +0.197 |
Neg logit contributions | |
least | -0.095 |
Little | -0.091 |
ime | -0.086 |
mid | -0.077 |
ill | -0.077 |
Token | He |
Token ID | 679 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
aches | +0.067 |
cks | +0.066 |
badges | +0.066 |
ows | +0.064 |
facts | +0.063 |
Neg logit contributions | |
Little | -0.034 |
Little | -0.034 |
least | -0.033 |
3 | -0.029 |
ime | -0.028 |
Token | wants |
Token ID | 3382 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ages | +0.074 |
bones | +0.070 |
diapers | +0.067 |
ows | +0.067 |
ometers | +0.066 |
Neg logit contributions | |
least | -0.071 |
tw | -0.056 |
andy | -0.054 |
knew | -0.054 |
and | -0.053 |
Token | to |
Token ID | 284 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
ows | +0.066 |
cks | +0.061 |
cuts | +0.059 |
bows | +0.058 |
bes | +0.058 |
Neg logit contributions | |
least | -0.015 |
3 | -0.010 |
Little | -0.009 |
B | -0.009 |
Peter | -0.009 |
Token | sad |
Token ID | 6507 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
ows | +0.094 |
badges | +0.093 |
cuts | +0.087 |
orts | +0.084 |
bes | +0.084 |
Neg logit contributions | |
least | -0.075 |
ime | -0.055 |
Little | -0.054 |
since | -0.054 |
3 | -0.053 |
Token | ." |
Token ID | 526 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
badges | +0.102 |
ows | +0.098 |
diapers | +0.090 |
orts | +0.090 |
labels | +0.090 |
Neg logit contributions | |
least | -0.085 |
Little | -0.074 |
ime | -0.073 |
illy | -0.068 |
ately | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | +9.64e-01 |
Pos logit contributions | |
badges | +0.154 |
ows | +0.152 |
tricks | +0.147 |
names | +0.146 |
diapers | +0.145 |
Neg logit contributions | |
least | -0.071 |
Little | -0.061 |
ime | -0.057 |
illy | -0.053 |
Maria | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | +8.11e-01 |
Pos logit contributions | |
badges | +0.042 |
ows | +0.042 |
tricks | +0.040 |
orts | +0.040 |
names | +0.040 |
Neg logit contributions | |
least | -0.026 |
Little | -0.021 |
ime | -0.020 |
illy | -0.020 |
ately | -0.019 |
Token | Tom |
Token ID | 13787 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
badges | +0.030 |
ows | +0.029 |
tricks | +0.028 |
names | +0.028 |
diapers | +0.028 |
Neg logit contributions | |
least | -0.027 |
Little | -0.024 |
ime | -0.023 |
illy | -0.022 |
Maria | -0.021 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.73e+00 |
Pos logit contributions | |
badges | +0.027 |
diapers | +0.026 |
bones | +0.026 |
gag | +0.025 |
tricks | +0.025 |
Neg logit contributions | |
least | -0.026 |
Little | -0.022 |
ill | -0.021 |
ime | -0.020 |
illy | -0.020 |
Token | ks |
Token ID | 591 |
Feature activation | +4.73e+00 |
Pos logit contributions | |
badges | +0.151 |
diapers | +0.132 |
ows | +0.126 |
tricks | +0.125 |
chips | +0.123 |
Neg logit contributions | |
least | -0.381 |
ime | -0.357 |
ill | -0.347 |
Little | -0.346 |
ately | -0.342 |
Token | yes |
Token ID | 3763 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
badges | +0.235 |
ows | +0.230 |
cuts | +0.229 |
bows | +0.225 |
aches | +0.225 |
Neg logit contributions | |
least | -0.077 |
Little | -0.069 |
ime | -0.062 |
again | -0.062 |
of | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
badges | +0.059 |
ows | +0.057 |
tricks | +0.055 |
names | +0.055 |
diapers | +0.054 |
Neg logit contributions | |
least | -0.043 |
Little | -0.039 |
ime | -0.037 |
illy | -0.035 |
ately | -0.033 |
Token | He |
Token ID | 679 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
badges | +0.043 |
ows | +0.042 |
aches | +0.042 |
cks | +0.040 |
tricks | +0.040 |
Neg logit contributions | |
least | -0.029 |
Little | -0.026 |
Little | -0.025 |
ime | -0.024 |
ill | -0.024 |
Token | is |
Token ID | 318 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
diapers | +0.043 |
gag | +0.043 |
bones | +0.042 |
ages | +0.042 |
ows | +0.041 |
Neg logit contributions | |
least | -0.060 |
Little | -0.046 |
andy | -0.044 |
se | -0.043 |
knew | -0.042 |
Token | says |
Token ID | 1139 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
badges | +0.039 |
ows | +0.035 |
tricks | +0.034 |
diapers | +0.034 |
helmets | +0.032 |
Neg logit contributions | |
least | -0.077 |
Little | -0.072 |
ime | -0.069 |
illy | -0.066 |
ill | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
ows | +0.062 |
badges | +0.061 |
tricks | +0.058 |
names | +0.057 |
diapers | +0.057 |
Neg logit contributions | |
least | -0.059 |
Little | -0.054 |
ime | -0.051 |
illy | -0.049 |
Maria | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
badges | +0.082 |
ows | +0.082 |
names | +0.081 |
name | +0.079 |
diapers | +0.079 |
Neg logit contributions | |
least | -0.067 |
ime | -0.056 |
Little | -0.055 |
illy | -0.053 |
ately | -0.051 |
Token |
|
Token ID | 198 |
Feature activation | +6.11e-01 |
Pos logit contributions | |
ows | +0.026 |
badges | +0.025 |
orts | +0.025 |
names | +0.025 |
tricks | +0.024 |
Neg logit contributions | |
least | -0.022 |
ime | -0.017 |
ately | -0.017 |
Little | -0.017 |
illy | -0.017 |
Token | Max |
Token ID | 11518 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
badges | +0.024 |
ows | +0.023 |
diapers | +0.022 |
tricks | +0.022 |
names | +0.022 |
Neg logit contributions | |
least | -0.019 |
ime | -0.016 |
Little | -0.015 |
illy | -0.015 |
ately | -0.015 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.71e+00 |
Pos logit contributions | |
badges | +0.032 |
bones | +0.032 |
ows | +0.031 |
diapers | +0.031 |
ages | +0.030 |
Neg logit contributions | |
least | -0.043 |
Little | -0.037 |
mid | -0.034 |
ill | -0.033 |
Little | -0.033 |
Token | ks |
Token ID | 591 |
Feature activation | +4.57e+00 |
Pos logit contributions | |
badges | +0.140 |
ows | +0.123 |
diapers | +0.122 |
tricks | +0.121 |
helmets | +0.113 |
Neg logit contributions | |
least | -0.388 |
ime | -0.360 |
Little | -0.356 |
ill | -0.348 |
ately | -0.344 |
Token | and |
Token ID | 290 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
badges | +0.230 |
cuts | +0.220 |
ows | +0.220 |
ages | +0.214 |
diapers | +0.214 |
Neg logit contributions | |
least | -0.080 |
Little | -0.070 |
ime | -0.064 |
mid | -0.062 |
again | -0.060 |
Token | nods |
Token ID | 34833 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
ows | +0.047 |
ages | +0.047 |
uries | +0.045 |
badges | +0.043 |
ves | +0.043 |
Neg logit contributions | |
least | -0.046 |
Little | -0.037 |
mid | -0.037 |
Little | -0.036 |
M | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
badges | +0.053 |
ows | +0.052 |
tricks | +0.049 |
diapers | +0.049 |
names | +0.049 |
Neg logit contributions | |
least | -0.042 |
Little | -0.038 |
ime | -0.036 |
illy | -0.035 |
ately | -0.033 |
Token | He |
Token ID | 679 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
badges | +0.039 |
ows | +0.038 |
tricks | +0.037 |
cks | +0.036 |
aches | +0.036 |
Neg logit contributions | |
least | -0.029 |
Little | -0.028 |
Little | -0.026 |
ime | -0.026 |
illy | -0.025 |
Token | ags |
Token ID | 3775 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
badges | +0.135 |
diapers | +0.129 |
names | +0.123 |
bows | +0.122 |
chips | +0.121 |
Neg logit contributions | |
least | -0.217 |
ime | -0.205 |
illy | -0.197 |
Little | -0.187 |
ill | -0.181 |
Token | his |
Token ID | 465 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
badges | +0.124 |
ows | +0.122 |
names | +0.115 |
diapers | +0.115 |
tricks | +0.114 |
Neg logit contributions | |
least | -0.092 |
Little | -0.082 |
ime | -0.079 |
Maria | -0.075 |
illy | -0.074 |
Token | tail |
Token ID | 7894 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
ows | +0.126 |
badges | +0.126 |
cuts | +0.117 |
tricks | +0.116 |
facts | +0.115 |
Neg logit contributions | |
least | -0.176 |
Little | -0.157 |
ime | -0.152 |
illy | -0.144 |
3 | -0.144 |
Token | . |
Token ID | 13 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
badges | +0.139 |
cuts | +0.131 |
diapers | +0.130 |
ows | +0.127 |
chips | +0.126 |
Neg logit contributions | |
least | -0.060 |
ime | -0.045 |
ill | -0.044 |
Little | -0.044 |
mid | -0.040 |
Token | He |
Token ID | 679 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
badges | +0.059 |
ows | +0.058 |
tricks | +0.056 |
aches | +0.056 |
cks | +0.055 |
Neg logit contributions | |
least | -0.049 |
Little | -0.046 |
ime | -0.042 |
Little | -0.042 |
illy | -0.041 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.63e+00 |
Pos logit contributions | |
ages | +0.067 |
diapers | +0.066 |
ows | +0.064 |
bones | +0.063 |
choices | +0.062 |
Neg logit contributions | |
least | -0.076 |
knew | -0.055 |
Little | -0.054 |
se | -0.052 |
and | -0.052 |
Token | ks |
Token ID | 591 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
badges | +0.153 |
diapers | +0.137 |
chips | +0.133 |
helmets | +0.129 |
tricks | +0.129 |
Neg logit contributions | |
least | -0.371 |
ill | -0.341 |
ime | -0.338 |
ately | -0.327 |
Little | -0.326 |
Token | . |
Token ID | 13 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
badges | +0.218 |
ows | +0.214 |
cuts | +0.213 |
bows | +0.211 |
aches | +0.211 |
Neg logit contributions | |
least | -0.070 |
Little | -0.066 |
of | -0.064 |
again | -0.063 |
ime | -0.061 |
Token |
|
Token ID | 198 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
badges | +0.062 |
ows | +0.061 |
tricks | +0.059 |
diapers | +0.058 |
names | +0.057 |
Neg logit contributions | |
least | -0.051 |
Little | -0.047 |
ime | -0.044 |
illy | -0.042 |
Maria | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
ows | +0.007 |
badges | +0.007 |
names | +0.006 |
tricks | +0.006 |
orts | +0.006 |
Neg logit contributions | |
least | -0.004 |
Little | -0.004 |
ime | -0.004 |
illy | -0.003 |
ately | -0.003 |
Token | " |
Token ID | 1 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
badges | +0.009 |
ows | +0.009 |
tricks | +0.009 |
names | +0.009 |
diapers | +0.009 |
Neg logit contributions | |
least | -0.005 |
Little | -0.005 |
ime | -0.005 |
illy | -0.004 |
Maria | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
badges | +0.056 |
ows | +0.054 |
names | +0.052 |
tricks | +0.052 |
diapers | +0.052 |
Neg logit contributions | |
least | -0.033 |
Little | -0.028 |
ime | -0.026 |
illy | -0.025 |
ately | -0.024 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
badges | +0.056 |
ows | +0.055 |
tricks | +0.053 |
names | +0.053 |
diapers | +0.052 |
Neg logit contributions | |
least | -0.041 |
Little | -0.037 |
ime | -0.035 |
illy | -0.033 |
ately | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
Little | +0.037 |
least | +0.036 |
ime | +0.034 |
Maria | +0.033 |
illy | +0.033 |
Neg logit contributions | |
ows | -0.061 |
badges | -0.060 |
tricks | -0.057 |
names | -0.057 |
homework | -0.057 |
Token | Mia |
Token ID | 32189 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ows | +0.064 |
facts | +0.060 |
badges | +0.059 |
names | +0.058 |
bows | +0.058 |
Neg logit contributions | |
least | -0.049 |
Little | -0.036 |
M | -0.036 |
Peter | -0.035 |
Little | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
badges | +0.042 |
ows | +0.041 |
tricks | +0.040 |
diapers | +0.039 |
names | +0.039 |
Neg logit contributions | |
least | -0.031 |
Little | -0.027 |
ime | -0.026 |
illy | -0.024 |
ately | -0.023 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.61e+00 |
Pos logit contributions | |
ows | +0.091 |
badges | +0.087 |
diapers | +0.083 |
orts | +0.083 |
tricks | +0.082 |
Neg logit contributions | |
least | -0.083 |
Little | -0.063 |
Little | -0.061 |
M | -0.060 |
Peter | -0.060 |
Token | ks |
Token ID | 591 |
Feature activation | +4.75e+00 |
Pos logit contributions | |
badges | +0.128 |
ows | +0.111 |
tricks | +0.110 |
diapers | +0.108 |
names | +0.102 |
Neg logit contributions | |
least | -0.402 |
ime | -0.370 |
Little | -0.366 |
ately | -0.361 |
illy | -0.355 |
Token | . |
Token ID | 13 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
badges | +0.210 |
ows | +0.204 |
cuts | +0.201 |
diapers | +0.198 |
preferences | +0.197 |
Neg logit contributions | |
least | -0.103 |
Little | -0.096 |
ime | -0.090 |
mid | -0.083 |
ful | -0.082 |
Token | He |
Token ID | 679 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
badges | +0.068 |
ows | +0.067 |
tricks | +0.064 |
diapers | +0.063 |
names | +0.062 |
Neg logit contributions | |
least | -0.057 |
Little | -0.053 |
ime | -0.049 |
illy | -0.047 |
Maria | -0.046 |
Token | grabs |
Token ID | 22378 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
ages | +0.068 |
diapers | +0.064 |
uggets | +0.064 |
ows | +0.064 |
ometers | +0.061 |
Neg logit contributions | |
least | -0.070 |
knew | -0.048 |
was | -0.046 |
Little | -0.046 |
tw | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
ows | +0.054 |
badges | +0.052 |
cuts | +0.051 |
tricks | +0.051 |
facts | +0.050 |
Neg logit contributions | |
least | -0.034 |
Little | -0.031 |
ime | -0.027 |
Little | -0.027 |
illy | -0.027 |
Token | noisy |
Token ID | 31210 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
cuts | +0.036 |
ows | +0.035 |
bows | +0.035 |
badges | +0.034 |
preferences | +0.034 |
Neg logit contributions | |
least | -0.022 |
nowhere | -0.020 |
3 | -0.018 |
mid | -0.018 |
ful | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
badges | +0.082 |
ows | +0.080 |
names | +0.078 |
tricks | +0.077 |
diapers | +0.077 |
Neg logit contributions | |
least | -0.032 |
Little | -0.025 |
ime | -0.025 |
illy | -0.022 |
ately | -0.020 |
Token | scary |
Token ID | 14343 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
ows | +0.040 |
badges | +0.040 |
bows | +0.039 |
aches | +0.038 |
addresses | +0.037 |
Neg logit contributions | |
least | -0.025 |
Little | -0.021 |
mid | -0.020 |
night | -0.020 |
Little | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
badges | +0.073 |
ows | +0.069 |
orts | +0.068 |
aches | +0.067 |
cuts | +0.067 |
Neg logit contributions | |
least | -0.039 |
ime | -0.033 |
Little | -0.031 |
ately | -0.029 |
ful | -0.027 |
Token | He |
Token ID | 679 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
badges | +0.081 |
ows | +0.079 |
tricks | +0.076 |
names | +0.074 |
diapers | +0.073 |
Neg logit contributions | |
least | -0.074 |
Little | -0.069 |
ime | -0.065 |
illy | -0.061 |
Maria | -0.060 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.60e+00 |
Pos logit contributions | |
bows | +0.069 |
ages | +0.068 |
ows | +0.067 |
badges | +0.067 |
choices | +0.065 |
Neg logit contributions | |
least | -0.084 |
knew | -0.064 |
se | -0.062 |
ime | -0.061 |
couldn | -0.061 |
Token | ks |
Token ID | 591 |
Feature activation | +4.58e+00 |
Pos logit contributions | |
badges | +0.174 |
diapers | +0.152 |
chips | +0.148 |
bes | +0.148 |
helmets | +0.144 |
Neg logit contributions | |
least | -0.349 |
ime | -0.323 |
ill | -0.313 |
ately | -0.311 |
ful | -0.301 |
Token | and |
Token ID | 290 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
badges | +0.230 |
ows | +0.224 |
diapers | +0.219 |
homework | +0.217 |
bows | +0.217 |
Neg logit contributions | |
least | -0.074 |
Little | -0.068 |
ime | -0.061 |
Maria | -0.049 |
mid | -0.049 |
Token | grow |
Token ID | 1663 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
ows | +0.061 |
uries | +0.054 |
ages | +0.054 |
edy | +0.053 |
cases | +0.053 |
Neg logit contributions | |
least | -0.058 |
mid | -0.049 |
M | -0.048 |
tw | -0.046 |
bur | -0.046 |
Token | ls |
Token ID | 7278 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
badges | +0.132 |
uries | +0.127 |
homework | +0.123 |
wrists | +0.122 |
receipt | +0.121 |
Neg logit contributions | |
ime | -0.165 |
Little | -0.160 |
least | -0.157 |
of | -0.147 |
in | -0.143 |
Token | and |
Token ID | 290 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
badges | +0.100 |
ows | +0.097 |
tricks | +0.094 |
diapers | +0.093 |
names | +0.093 |
Neg logit contributions | |
least | -0.061 |
Little | -0.053 |
ime | -0.050 |
illy | -0.047 |
ately | -0.046 |
Token | of |
Token ID | 286 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
badges | +0.000 |
ows | +0.000 |
tricks | +0.000 |
names | +0.000 |
diapers | +0.000 |
Neg logit contributions | |
least | -0.000 |
Little | -0.000 |
ime | -0.000 |
illy | -0.000 |
Maria | -0.000 |
Token | cheese |
Token ID | 9891 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
least | +0.012 |
ime | +0.011 |
Little | +0.010 |
night | +0.010 |
Little | +0.010 |
Neg logit contributions | |
ows | -0.009 |
orts | -0.008 |
gag | -0.008 |
badges | -0.008 |
cuts | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
badges | +0.073 |
ows | +0.072 |
tricks | +0.069 |
names | +0.069 |
diapers | +0.068 |
Neg logit contributions | |
least | -0.046 |
Little | -0.040 |
ime | -0.038 |
illy | -0.036 |
ately | -0.034 |
Token | sauce |
Token ID | 10746 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
ows | +0.037 |
badges | +0.035 |
orts | +0.034 |
cuts | +0.033 |
tricks | +0.033 |
Neg logit contributions | |
least | -0.046 |
Little | -0.041 |
ime | -0.038 |
ately | -0.037 |
illy | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
badges | +0.126 |
ows | +0.122 |
tricks | +0.118 |
names | +0.117 |
diapers | +0.116 |
Neg logit contributions | |
least | -0.101 |
Little | -0.091 |
ime | -0.087 |
illy | -0.083 |
ately | -0.079 |
Token | topp |
Token ID | 23126 |
Feature activation | +7.57e+00 |
Pos logit contributions | |
ows | +0.050 |
badges | +0.047 |
orts | +0.046 |
cuts | +0.044 |
tricks | +0.044 |
Neg logit contributions | |
least | -0.063 |
Little | -0.055 |
ime | -0.053 |
illy | -0.050 |
ately | -0.050 |
Token | ings |
Token ID | 654 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
badges | +0.232 |
trained | +0.226 |
facts | +0.211 |
bows | +0.207 |
versa | +0.205 |
Neg logit contributions | |
ime | -0.261 |
ill | -0.232 |
oh | -0.230 |
ately | -0.225 |
illy | -0.223 |
Token | . |
Token ID | 13 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
badges | +0.152 |
ows | +0.145 |
gag | +0.144 |
tricks | +0.144 |
facts | +0.138 |
Neg logit contributions | |
least | -0.103 |
Little | -0.095 |
ime | -0.088 |
illy | -0.083 |
Little | -0.082 |
Token | The |
Token ID | 383 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
badges | +0.062 |
tricks | +0.060 |
ows | +0.060 |
aches | +0.057 |
homework | +0.055 |
Neg logit contributions | |
least | -0.077 |
Little | -0.072 |
Little | -0.065 |
ime | -0.064 |
illy | -0.063 |
Token | people |
Token ID | 661 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
ows | +0.039 |
badges | +0.036 |
orts | +0.036 |
cks | +0.034 |
gag | +0.034 |
Neg logit contributions | |
least | -0.043 |
Little | -0.036 |
ime | -0.035 |
3 | -0.034 |
illy | -0.032 |
Token | who |
Token ID | 508 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
badges | +0.012 |
ows | +0.011 |
diapers | +0.011 |
tricks | +0.011 |
names | +0.011 |
Neg logit contributions | |
least | -0.011 |
ime | -0.010 |
Little | -0.010 |
ately | -0.008 |
illy | -0.008 |
Token | dog |
Token ID | 3290 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
badges | +0.018 |
ows | +0.018 |
tricks | +0.017 |
names | +0.017 |
diapers | +0.017 |
Neg logit contributions | |
least | -0.014 |
Little | -0.013 |
ime | -0.012 |
illy | -0.012 |
ately | -0.011 |
Token | watches |
Token ID | 16860 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
badges | +0.070 |
ows | +0.068 |
diapers | +0.067 |
bones | +0.066 |
orts | +0.065 |
Neg logit contributions | |
least | -0.068 |
Little | -0.054 |
ime | -0.048 |
mid | -0.047 |
st | -0.047 |
Token | them |
Token ID | 606 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
badges | +0.001 |
ows | +0.001 |
tricks | +0.001 |
names | +0.001 |
diapers | +0.001 |
Neg logit contributions | |
least | -0.001 |
Little | -0.001 |
ime | -0.001 |
illy | -0.001 |
Maria | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
badges | +0.114 |
ows | +0.112 |
tricks | +0.108 |
diapers | +0.107 |
names | +0.107 |
Neg logit contributions | |
least | -0.069 |
Little | -0.060 |
ime | -0.057 |
illy | -0.054 |
ately | -0.051 |
Token | He |
Token ID | 679 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
badges | +0.043 |
ows | +0.043 |
tricks | +0.041 |
diapers | +0.040 |
aches | +0.039 |
Neg logit contributions | |
least | -0.040 |
Little | -0.038 |
ime | -0.034 |
Little | -0.034 |
Maria | -0.033 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.57e+00 |
Pos logit contributions | |
badges | +0.052 |
ows | +0.051 |
diapers | +0.051 |
ages | +0.049 |
chips | +0.048 |
Neg logit contributions | |
least | -0.067 |
Little | -0.053 |
ime | -0.050 |
mid | -0.049 |
knew | -0.048 |
Token | ks |
Token ID | 591 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
badges | +0.155 |
diapers | +0.136 |
chips | +0.133 |
ows | +0.128 |
labels | +0.127 |
Neg logit contributions | |
least | -0.371 |
ime | -0.343 |
ill | -0.333 |
Little | -0.329 |
ately | -0.326 |
Token | and |
Token ID | 290 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
badges | +0.215 |
ows | +0.208 |
diapers | +0.204 |
cuts | +0.202 |
homework | +0.201 |
Neg logit contributions | |
least | -0.077 |
Little | -0.071 |
ime | -0.065 |
Maria | -0.055 |
illy | -0.055 |
Token | cheers |
Token ID | 34550 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
ows | +0.054 |
badges | +0.049 |
ages | +0.049 |
cuts | +0.049 |
edy | +0.049 |
Neg logit contributions | |
least | -0.054 |
mid | -0.044 |
Little | -0.042 |
M | -0.041 |
Little | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
badges | +0.084 |
ows | +0.081 |
tricks | +0.078 |
diapers | +0.078 |
names | +0.077 |
Neg logit contributions | |
least | -0.064 |
Little | -0.058 |
ime | -0.055 |
illy | -0.053 |
ately | -0.050 |
Token | He |
Token ID | 679 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
badges | +0.069 |
ows | +0.068 |
tricks | +0.066 |
diapers | +0.065 |
names | +0.064 |
Neg logit contributions | |
least | -0.047 |
Little | -0.044 |
ime | -0.040 |
illy | -0.038 |
Maria | -0.037 |
Token | l |
Token ID | 300 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
badges | +0.047 |
ows | +0.046 |
diapers | +0.045 |
orts | +0.044 |
bows | +0.044 |
Neg logit contributions | |
least | -0.030 |
Little | -0.024 |
ime | -0.023 |
ately | -0.020 |
Little | -0.020 |
Token | icks |
Token ID | 3378 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
badges | +0.158 |
diapers | +0.142 |
tricks | +0.140 |
ows | +0.138 |
cuts | +0.137 |
Neg logit contributions | |
least | -0.230 |
ime | -0.206 |
illy | -0.204 |
ill | -0.199 |
Little | -0.198 |
Token | his |
Token ID | 465 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
badges | +0.075 |
ows | +0.074 |
tricks | +0.070 |
diapers | +0.069 |
names | +0.069 |
Neg logit contributions | |
least | -0.061 |
Little | -0.057 |
ime | -0.054 |
illy | -0.052 |
Maria | -0.050 |
Token | lips |
Token ID | 11914 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
ows | +0.101 |
badges | +0.101 |
cuts | +0.097 |
bows | +0.095 |
orts | +0.092 |
Neg logit contributions | |
least | -0.113 |
ime | -0.104 |
Little | -0.102 |
3 | -0.098 |
mid | -0.096 |
Token | and |
Token ID | 290 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
badges | +0.098 |
ows | +0.093 |
tricks | +0.091 |
diapers | +0.091 |
names | +0.090 |
Neg logit contributions | |
least | -0.047 |
Little | -0.041 |
ime | -0.039 |
illy | -0.036 |
ill | -0.034 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.54e+00 |
Pos logit contributions | |
ows | +0.056 |
cuts | +0.052 |
badges | +0.052 |
orts | +0.050 |
names | +0.049 |
Neg logit contributions | |
least | -0.045 |
mid | -0.037 |
Little | -0.037 |
Little | -0.036 |
M | -0.036 |
Token | ks |
Token ID | 591 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
badges | +0.150 |
diapers | +0.126 |
tricks | +0.126 |
labels | +0.124 |
ows | +0.122 |
Neg logit contributions | |
least | -0.371 |
ime | -0.349 |
Little | -0.339 |
ately | -0.338 |
ill | -0.333 |
Token | . |
Token ID | 13 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
badges | +0.207 |
cuts | +0.199 |
ows | +0.196 |
diapers | +0.194 |
orts | +0.193 |
Neg logit contributions | |
least | -0.099 |
ime | -0.098 |
Little | -0.096 |
ful | -0.086 |
mid | -0.085 |
Token | He |
Token ID | 679 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
badges | +0.107 |
ows | +0.105 |
tricks | +0.100 |
diapers | +0.098 |
names | +0.097 |
Neg logit contributions | |
least | -0.077 |
Little | -0.072 |
ime | -0.067 |
illy | -0.063 |
Maria | -0.062 |
Token | is |
Token ID | 318 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ages | +0.075 |
ows | +0.075 |
badges | +0.072 |
diapers | +0.072 |
ours | +0.070 |
Neg logit contributions | |
least | -0.089 |
se | -0.066 |
Little | -0.066 |
knew | -0.065 |
mid | -0.064 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
cuts | +0.054 |
badges | +0.054 |
ows | +0.053 |
bows | +0.051 |
gers | +0.050 |
Neg logit contributions | |
least | -0.038 |
3 | -0.030 |
nowhere | -0.028 |
ful | -0.027 |
mid | -0.027 |
Token | dog |
Token ID | 3290 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
badges | +0.022 |
ows | +0.022 |
tricks | +0.021 |
diapers | +0.021 |
names | +0.020 |
Neg logit contributions | |
least | -0.019 |
Little | -0.017 |
ime | -0.016 |
illy | -0.015 |
ately | -0.014 |
Token | ." |
Token ID | 526 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
badges | +0.015 |
ows | +0.015 |
names | +0.014 |
diapers | +0.014 |
tricks | +0.014 |
Neg logit contributions | |
least | -0.013 |
Little | -0.012 |
ime | -0.011 |
illy | -0.011 |
ately | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +8.44e-03 |
Pos logit contributions | |
badges | +0.092 |
ows | +0.090 |
tricks | +0.087 |
names | +0.087 |
diapers | +0.086 |
Neg logit contributions | |
least | -0.049 |
Little | -0.042 |
ime | -0.041 |
illy | -0.038 |
ately | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
badges | +0.000 |
ows | +0.000 |
tricks | +0.000 |
names | +0.000 |
orts | +0.000 |
Neg logit contributions | |
least | -0.000 |
ime | -0.000 |
Little | -0.000 |
ately | -0.000 |
illy | -0.000 |
Token | Max |
Token ID | 11518 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
badges | +0.001 |
ows | +0.001 |
helmets | +0.001 |
tricks | +0.001 |
orts | +0.001 |
Neg logit contributions | |
least | -0.002 |
ime | -0.002 |
ately | -0.002 |
illy | -0.002 |
ill | -0.002 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.54e+00 |
Pos logit contributions | |
bones | +0.032 |
diapers | +0.032 |
badges | +0.032 |
ows | +0.031 |
gag | +0.031 |
Neg logit contributions | |
least | -0.033 |
Little | -0.027 |
Little | -0.025 |
mid | -0.024 |
illy | -0.024 |
Token | ks |
Token ID | 591 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
badges | +0.141 |
diapers | +0.128 |
tricks | +0.122 |
cuts | +0.117 |
ows | +0.117 |
Neg logit contributions | |
least | -0.370 |
ime | -0.342 |
ill | -0.339 |
Little | -0.332 |
ately | -0.331 |
Token | and |
Token ID | 290 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
badges | +0.226 |
ows | +0.222 |
cuts | +0.221 |
aches | +0.217 |
diapers | +0.216 |
Neg logit contributions | |
least | -0.054 |
Little | -0.048 |
ime | -0.041 |
mid | -0.038 |
Little | -0.037 |
Token | w |
Token ID | 266 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
ows | +0.042 |
ages | +0.041 |
bones | +0.041 |
preferences | +0.039 |
meanings | +0.039 |
Neg logit contributions | |
least | -0.036 |
Little | -0.030 |
M | -0.030 |
Little | -0.029 |
Bear | -0.029 |
Token | ags |
Token ID | 3775 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
badges | +0.120 |
diapers | +0.112 |
names | +0.108 |
gag | +0.104 |
hides | +0.104 |
Neg logit contributions | |
least | -0.206 |
ime | -0.196 |
illy | -0.189 |
Little | -0.185 |
ately | -0.176 |
Token | his |
Token ID | 465 |
Feature activation | +3.70e+00 |
Pos logit contributions | |
ows | +0.121 |
badges | +0.120 |
cuts | +0.113 |
diapers | +0.112 |
names | +0.112 |
Neg logit contributions | |
least | -0.079 |
Little | -0.072 |
ime | -0.068 |
Maria | -0.066 |
illy | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
badges | +0.103 |
ows | +0.096 |
names | +0.095 |
tricks | +0.095 |
diapers | +0.094 |
Neg logit contributions | |
least | -0.071 |
Little | -0.063 |
ime | -0.062 |
illy | -0.058 |
ately | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
badges | +0.139 |
ows | +0.136 |
tricks | +0.132 |
names | +0.130 |
diapers | +0.130 |
Neg logit contributions | |
least | -0.085 |
Little | -0.078 |
ime | -0.072 |
illy | -0.068 |
Maria | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
badges | +0.069 |
ows | +0.069 |
tricks | +0.066 |
names | +0.066 |
diapers | +0.065 |
Neg logit contributions | |
least | -0.036 |
Little | -0.030 |
ime | -0.029 |
illy | -0.028 |
ately | -0.026 |
Token | The |
Token ID | 464 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
badges | +0.053 |
ows | +0.052 |
tricks | +0.050 |
names | +0.049 |
diapers | +0.049 |
Neg logit contributions | |
least | -0.046 |
Little | -0.042 |
ime | -0.040 |
illy | -0.038 |
Maria | -0.036 |
Token | dog |
Token ID | 3290 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
ows | +0.067 |
badges | +0.064 |
diapers | +0.060 |
cks | +0.059 |
tricks | +0.059 |
Neg logit contributions | |
least | -0.071 |
Little | -0.065 |
ime | -0.061 |
illy | -0.059 |
3 | -0.057 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.53e+00 |
Pos logit contributions | |
ages | +0.077 |
ours | +0.075 |
bows | +0.075 |
badges | +0.075 |
ometers | +0.074 |
Neg logit contributions | |
least | -0.078 |
again | -0.066 |
mid | -0.060 |
and | -0.060 |
Little | -0.060 |
Token | ks |
Token ID | 591 |
Feature activation | +4.44e+00 |
Pos logit contributions | |
badges | +0.157 |
diapers | +0.143 |
cuts | +0.131 |
labels | +0.129 |
ours | +0.128 |
Neg logit contributions | |
least | -0.348 |
ime | -0.328 |
ill | -0.327 |
ately | -0.316 |
Little | -0.308 |
Token | and |
Token ID | 290 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
badges | +0.221 |
ows | +0.212 |
cuts | +0.211 |
diapers | +0.209 |
bows | +0.208 |
Neg logit contributions | |
least | -0.074 |
Little | -0.068 |
ime | -0.064 |
ill | -0.056 |
mid | -0.055 |
Token | runs |
Token ID | 4539 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
ows | +0.065 |
ages | +0.063 |
uries | +0.062 |
bows | +0.059 |
edy | +0.059 |
Neg logit contributions | |
least | -0.060 |
mid | -0.051 |
tw | -0.048 |
bur | -0.047 |
bubble | -0.046 |
Token | away |
Token ID | 1497 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
badges | +0.068 |
preferences | +0.067 |
aches | +0.066 |
ows | +0.065 |
cks | +0.065 |
Neg logit contributions | |
least | -0.030 |
Little | -0.024 |
B | -0.024 |
again | -0.022 |
since | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
badges | +0.103 |
cuts | +0.095 |
ows | +0.095 |
names | +0.094 |
helmets | +0.092 |
Neg logit contributions | |
least | -0.054 |
Little | -0.046 |
ime | -0.045 |
ill | -0.042 |
ately | -0.041 |
Token | name |
Token ID | 1438 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
badges | +0.027 |
ows | +0.026 |
cks | +0.026 |
bones | +0.025 |
trained | +0.024 |
Neg logit contributions | |
least | -0.080 |
Little | -0.068 |
3 | -0.068 |
illy | -0.068 |
ime | -0.067 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
badges | +0.081 |
ows | +0.080 |
tricks | +0.077 |
names | +0.076 |
diapers | +0.075 |
Neg logit contributions | |
least | -0.061 |
Little | -0.054 |
ime | -0.052 |
illy | -0.049 |
ately | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
badges | +0.103 |
ows | +0.101 |
tricks | +0.098 |
names | +0.097 |
diapers | +0.097 |
Neg logit contributions | |
least | -0.049 |
Little | -0.042 |
ime | -0.040 |
illy | -0.037 |
ately | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
badges | +0.028 |
ows | +0.028 |
tricks | +0.027 |
names | +0.027 |
orts | +0.027 |
Neg logit contributions | |
least | -0.017 |
Little | -0.013 |
ime | -0.013 |
illy | -0.013 |
ately | -0.012 |
Token | Tom |
Token ID | 13787 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
badges | +0.029 |
ows | +0.029 |
names | +0.028 |
tricks | +0.028 |
diapers | +0.028 |
Neg logit contributions | |
Little | -0.012 |
least | -0.012 |
ime | -0.010 |
Maria | -0.010 |
illy | -0.009 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.53e+00 |
Pos logit contributions | |
badges | +0.021 |
bones | +0.021 |
diapers | +0.020 |
ages | +0.020 |
ows | +0.020 |
Neg logit contributions | |
least | -0.030 |
Little | -0.026 |
mid | -0.024 |
andy | -0.024 |
ill | -0.024 |
Token | ks |
Token ID | 591 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
badges | +0.149 |
diapers | +0.130 |
ows | +0.125 |
tricks | +0.125 |
chips | +0.124 |
Neg logit contributions | |
least | -0.364 |
ime | -0.345 |
ill | -0.338 |
Little | -0.333 |
ately | -0.325 |
Token | and |
Token ID | 290 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
badges | +0.224 |
cuts | +0.216 |
ows | +0.216 |
diapers | +0.211 |
aches | +0.211 |
Neg logit contributions | |
least | -0.066 |
Little | -0.062 |
ime | -0.057 |
mid | -0.051 |
Little | -0.049 |
Token | jumps |
Token ID | 18045 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
ages | +0.040 |
ows | +0.038 |
uries | +0.037 |
bones | +0.036 |
ggies | +0.035 |
Neg logit contributions | |
least | -0.041 |
tw | -0.035 |
M | -0.035 |
mid | -0.035 |
Little | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
badges | +0.103 |
ows | +0.096 |
tricks | +0.095 |
preferences | +0.095 |
facts | +0.094 |
Neg logit contributions | |
least | -0.067 |
Little | -0.062 |
ime | -0.058 |
illy | -0.055 |
ately | -0.053 |
Token | He |
Token ID | 679 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
badges | +0.046 |
ows | +0.044 |
cks | +0.043 |
cuts | +0.042 |
aches | +0.042 |
Neg logit contributions | |
least | -0.031 |
Little | -0.030 |
Little | -0.029 |
ime | -0.028 |
Maria | -0.026 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
least | +0.011 |
Little | +0.010 |
ime | +0.009 |
illy | +0.009 |
Maria | +0.008 |
Neg logit contributions | |
badges | -0.015 |
ows | -0.015 |
tricks | -0.015 |
names | -0.015 |
diapers | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
badges | +0.084 |
ows | +0.084 |
tricks | +0.079 |
diapers | +0.079 |
names | +0.079 |
Neg logit contributions | |
least | -0.052 |
Little | -0.048 |
ime | -0.043 |
illy | -0.042 |
Maria | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
least | +0.011 |
Little | +0.009 |
ime | +0.008 |
illy | +0.008 |
ately | +0.008 |
Neg logit contributions | |
badges | -0.019 |
ows | -0.019 |
tricks | -0.018 |
names | -0.018 |
orts | -0.018 |
Token | The |
Token ID | 464 |
Feature activation | +7.56e-01 |
Pos logit contributions | |
least | +0.012 |
Little | +0.010 |
ime | +0.010 |
illy | +0.009 |
ately | +0.009 |
Neg logit contributions | |
badges | -0.022 |
ows | -0.022 |
tricks | -0.021 |
names | -0.021 |
diapers | -0.021 |
Token | dog |
Token ID | 3290 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
ows | +0.028 |
badges | +0.028 |
diapers | +0.026 |
cks | +0.025 |
names | +0.025 |
Neg logit contributions | |
least | -0.025 |
Little | -0.023 |
ime | -0.021 |
illy | -0.020 |
3 | -0.019 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.53e+00 |
Pos logit contributions | |
badges | +0.092 |
diapers | +0.089 |
ows | +0.089 |
cks | +0.087 |
bones | +0.086 |
Neg logit contributions | |
least | -0.067 |
Little | -0.060 |
Little | -0.053 |
illy | -0.051 |
ime | -0.051 |
Token | ks |
Token ID | 591 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
badges | +0.167 |
diapers | +0.151 |
ows | +0.149 |
tricks | +0.148 |
names | +0.143 |
Neg logit contributions | |
least | -0.350 |
ime | -0.328 |
Little | -0.322 |
ately | -0.316 |
ill | -0.314 |
Token | . |
Token ID | 13 |
Feature activation | +9.59e-01 |
Pos logit contributions | |
badges | +0.190 |
ows | +0.185 |
diapers | +0.178 |
tricks | +0.178 |
names | +0.177 |
Neg logit contributions | |
least | -0.109 |
Little | -0.099 |
ime | -0.094 |
illy | -0.088 |
Maria | -0.083 |
Token | The |
Token ID | 383 |
Feature activation | +9.05e-01 |
Pos logit contributions | |
badges | +0.037 |
ows | +0.036 |
diapers | +0.035 |
aches | +0.035 |
tricks | +0.035 |
Neg logit contributions | |
least | -0.028 |
Little | -0.026 |
Little | -0.024 |
ime | -0.023 |
ill | -0.023 |
Token | dog |
Token ID | 3290 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
ows | +0.033 |
badges | +0.031 |
cks | +0.030 |
diapers | +0.029 |
bows | +0.028 |
Neg logit contributions | |
least | -0.031 |
Little | -0.028 |
ime | -0.026 |
3 | -0.025 |
illy | -0.025 |
Token | is |
Token ID | 318 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
badges | +0.063 |
diapers | +0.061 |
bones | +0.060 |
ages | +0.059 |
ows | +0.058 |
Neg logit contributions | |
least | -0.060 |
Little | -0.050 |
mid | -0.046 |
Little | -0.045 |
knew | -0.045 |
Token | afraid |
Token ID | 7787 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
bows | +0.035 |
cuts | +0.035 |
facts | +0.035 |
badges | +0.034 |
orts | +0.034 |
Neg logit contributions | |
least | -0.023 |
nowhere | -0.019 |
3 | -0.018 |
ful | -0.017 |
mid | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
badges | +0.091 |
ows | +0.090 |
tricks | +0.087 |
names | +0.086 |
diapers | +0.086 |
Neg logit contributions | |
least | -0.066 |
Little | -0.058 |
ime | -0.055 |
illy | -0.054 |
Maria | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | +5.44e-01 |
Pos logit contributions | |
badges | +0.085 |
ows | +0.083 |
names | +0.080 |
diapers | +0.079 |
tricks | +0.079 |
Neg logit contributions | |
least | -0.074 |
Little | -0.065 |
ime | -0.063 |
illy | -0.062 |
ately | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
badges | +0.023 |
ows | +0.023 |
names | +0.022 |
tricks | +0.022 |
orts | +0.022 |
Neg logit contributions | |
least | -0.015 |
ime | -0.012 |
Little | -0.012 |
illy | -0.012 |
ately | -0.011 |
Token | Max |
Token ID | 11518 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
badges | +0.022 |
ows | +0.022 |
tricks | +0.021 |
names | +0.021 |
diapers | +0.021 |
Neg logit contributions | |
least | -0.015 |
Little | -0.013 |
ime | -0.013 |
illy | -0.012 |
ately | -0.012 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.48e+00 |
Pos logit contributions | |
badges | +0.028 |
facts | +0.027 |
diapers | +0.027 |
bows | +0.027 |
bones | +0.027 |
Neg logit contributions | |
least | -0.029 |
Little | -0.024 |
ime | -0.022 |
oh | -0.021 |
' | -0.021 |
Token | ks |
Token ID | 591 |
Feature activation | +4.42e+00 |
Pos logit contributions | |
badges | +0.151 |
diapers | +0.134 |
tricks | +0.127 |
chips | +0.126 |
labels | +0.125 |
Neg logit contributions | |
least | -0.355 |
ime | -0.340 |
ill | -0.324 |
ately | -0.323 |
Little | -0.319 |
Token | at |
Token ID | 379 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
badges | +0.203 |
cuts | +0.192 |
aches | +0.190 |
bows | +0.190 |
ows | +0.190 |
Neg logit contributions | |
least | -0.086 |
Little | -0.082 |
ime | -0.078 |
Little | -0.070 |
mid | -0.069 |
Token | the |
Token ID | 262 |
Feature activation | +6.14e-01 |
Pos logit contributions | |
badges | +0.024 |
diapers | +0.023 |
ows | +0.023 |
tricks | +0.022 |
names | +0.022 |
Neg logit contributions | |
least | -0.035 |
Little | -0.033 |
ime | -0.033 |
illy | -0.031 |
Em | -0.031 |
Token | van |
Token ID | 5719 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
ows | +0.022 |
badges | +0.022 |
tricks | +0.021 |
diapers | +0.020 |
names | +0.020 |
Neg logit contributions | |
least | -0.021 |
Little | -0.019 |
ime | -0.018 |
illy | -0.017 |
ately | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
least | +0.012 |
Little | +0.010 |
ime | +0.010 |
ately | +0.009 |
illy | +0.009 |
Neg logit contributions | |
badges | -0.016 |
ows | -0.016 |
diapers | -0.015 |
tricks | -0.015 |
names | -0.015 |
Token | Spot |
Token ID | 15899 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
least | +0.048 |
Little | +0.042 |
Little | +0.038 |
ime | +0.035 |
ately | +0.035 |
Neg logit contributions | |
badges | -0.076 |
ows | -0.076 |
bows | -0.073 |
tricks | -0.072 |
orts | -0.071 |
Token | is |
Token ID | 318 |
Feature activation | +5.12e-01 |
Pos logit contributions | |
badges | +0.110 |
diapers | +0.103 |
facts | +0.103 |
bones | +0.101 |
titles | +0.101 |
Neg logit contributions | |
least | -0.083 |
oh | -0.062 |
' | -0.059 |
knew | -0.058 |
Little | -0.057 |
Token | scared |
Token ID | 12008 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
badges | +0.022 |
cuts | +0.021 |
facts | +0.021 |
bows | +0.021 |
names | +0.020 |
Neg logit contributions | |
least | -0.014 |
nowhere | -0.012 |
mid | -0.011 |
3 | -0.011 |
ime | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
badges | +0.077 |
ows | +0.077 |
tricks | +0.074 |
diapers | +0.073 |
names | +0.073 |
Neg logit contributions | |
least | -0.082 |
Little | -0.074 |
ime | -0.070 |
illy | -0.068 |
ately | -0.066 |
Token | He |
Token ID | 679 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
badges | +0.086 |
tricks | +0.083 |
aches | +0.083 |
ows | +0.082 |
facts | +0.081 |
Neg logit contributions | |
Little | -0.052 |
least | -0.050 |
Little | -0.046 |
ime | -0.043 |
Maria | -0.042 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.47e+00 |
Pos logit contributions | |
bows | +0.081 |
bones | +0.074 |
ages | +0.073 |
ows | +0.071 |
choices | +0.071 |
Neg logit contributions | |
least | -0.095 |
knew | -0.076 |
Little | -0.072 |
couldn | -0.071 |
ime | -0.071 |
Token | ks |
Token ID | 591 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
badges | +0.152 |
diapers | +0.133 |
chips | +0.127 |
tricks | +0.126 |
bes | +0.126 |
Neg logit contributions | |
least | -0.359 |
ime | -0.339 |
ill | -0.326 |
ately | -0.325 |
Little | -0.322 |
Token | at |
Token ID | 379 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
badges | +0.216 |
ows | +0.210 |
bows | +0.208 |
cuts | +0.207 |
homework | +0.205 |
Neg logit contributions | |
least | -0.075 |
Little | -0.073 |
ime | -0.066 |
Little | -0.058 |
mid | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
badges | +0.038 |
ows | +0.038 |
tricks | +0.035 |
names | +0.035 |
diapers | +0.034 |
Neg logit contributions | |
least | -0.035 |
Little | -0.031 |
ime | -0.029 |
illy | -0.028 |
ately | -0.027 |
Token | truck |
Token ID | 7779 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
ows | +0.024 |
badges | +0.022 |
orts | +0.021 |
aches | +0.021 |
cks | +0.021 |
Neg logit contributions | |
least | -0.021 |
Little | -0.019 |
ime | -0.018 |
3 | -0.018 |
mid | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
badges | +0.064 |
ows | +0.061 |
diapers | +0.060 |
tricks | +0.059 |
names | +0.059 |
Neg logit contributions | |
least | -0.034 |
Little | -0.027 |
ime | -0.027 |
ately | -0.024 |
ill | -0.024 |
Token | out |
Token ID | 503 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
badges | +0.029 |
ows | +0.028 |
tricks | +0.027 |
diapers | +0.027 |
names | +0.027 |
Neg logit contributions | |
least | -0.021 |
Little | -0.019 |
ime | -0.018 |
illy | -0.017 |
ately | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ows | +0.044 |
badges | +0.043 |
tricks | +0.042 |
names | +0.042 |
diapers | +0.041 |
Neg logit contributions | |
least | -0.034 |
Little | -0.030 |
illy | -0.028 |
ime | -0.028 |
Em | -0.027 |
Token | cry |
Token ID | 3960 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
ows | +0.059 |
badges | +0.058 |
cuts | +0.057 |
oodle | +0.055 |
orts | +0.054 |
Neg logit contributions | |
least | -0.053 |
Little | -0.042 |
mid | -0.042 |
ime | -0.040 |
se | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
badges | +0.076 |
ows | +0.074 |
tricks | +0.072 |
names | +0.071 |
diapers | +0.070 |
Neg logit contributions | |
least | -0.066 |
Little | -0.059 |
ime | -0.057 |
illy | -0.054 |
ately | -0.052 |
Token | Rex |
Token ID | 17853 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
badges | +0.123 |
ows | +0.121 |
tricks | +0.117 |
diapers | +0.113 |
names | +0.112 |
Neg logit contributions | |
least | -0.082 |
Little | -0.076 |
ime | -0.070 |
illy | -0.066 |
Maria | -0.066 |
Token | bar |
Token ID | 2318 |
Feature activation | +7.46e+00 |
Pos logit contributions | |
badges | +0.124 |
bones | +0.114 |
facts | +0.113 |
receipt | +0.113 |
uries | +0.112 |
Neg logit contributions | |
least | -0.126 |
Little | -0.097 |
mid | -0.094 |
ime | -0.092 |
oh | -0.092 |
Token | ks |
Token ID | 591 |
Feature activation | +4.53e+00 |
Pos logit contributions | |
badges | +0.161 |
diapers | +0.139 |
helmets | +0.134 |
labels | +0.131 |
chips | +0.131 |
Neg logit contributions | |
least | -0.354 |
ime | -0.333 |
ill | -0.319 |
ately | -0.313 |
Little | -0.312 |
Token | at |
Token ID | 379 |
Feature activation | +9.35e-01 |
Pos logit contributions | |
badges | +0.246 |
ows | +0.237 |
bows | +0.234 |
cuts | +0.233 |
diapers | +0.231 |
Neg logit contributions | |
least | -0.058 |
Little | -0.049 |
ime | -0.043 |
mid | -0.035 |
ful | -0.033 |
Token | them |
Token ID | 606 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
badges | +0.027 |
ows | +0.026 |
tricks | +0.025 |
diapers | +0.025 |
names | +0.025 |
Neg logit contributions | |
least | -0.037 |
Little | -0.035 |
ime | -0.034 |
illy | -0.033 |
Maria | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
badges | +0.155 |
ows | +0.145 |
diapers | +0.143 |
tricks | +0.143 |
names | +0.142 |
Neg logit contributions | |
least | -0.061 |
Little | -0.048 |
ime | -0.047 |
illy | -0.040 |
ill | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
badges | +0.112 |
ows | +0.108 |
aches | +0.107 |
bows | +0.104 |
tricks | +0.104 |
Neg logit contributions | |
Little | -0.053 |
least | -0.053 |
Little | -0.047 |
ime | -0.044 |
Maria | -0.042 |
Token | aded |
Token ID | 5286 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
badges | +0.139 |
ows | +0.137 |
diapers | +0.130 |
tricks | +0.129 |
names | +0.129 |
Neg logit contributions | |
least | -0.132 |
Little | -0.118 |
ime | -0.118 |
illy | -0.113 |
ately | -0.106 |
Token | in |
Token ID | 287 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
least | +0.033 |
Little | +0.030 |
ime | +0.027 |
illy | +0.027 |
Em | +0.026 |
Neg logit contributions | |
badges | -0.039 |
ows | -0.038 |
tricks | -0.038 |
names | -0.037 |
diapers | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
enny | +0.009 |
ately | +0.009 |
least | +0.009 |
Em | +0.009 |
illy | +0.009 |
Neg logit contributions | |
names | -0.007 |
tricks | -0.007 |
labels | -0.007 |
badges | -0.007 |
helmets | -0.007 |
Token | water |
Token ID | 1660 |
Feature activation | +9.66e-02 |
Pos logit contributions | |
least | +0.029 |
ately | +0.029 |
enny | +0.026 |
Maria | +0.026 |
illy | +0.026 |
Neg logit contributions | |
badges | -0.027 |
names | -0.026 |
tricks | -0.026 |
helmets | -0.025 |
diapers | -0.025 |
Token | delic |
Token ID | 8675 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ows | +0.004 |
tricks | +0.003 |
names | +0.003 |
badges | +0.003 |
diapers | +0.003 |
Neg logit contributions | |
Little | -0.003 |
least | -0.003 |
Em | -0.003 |
Maria | -0.003 |
ime | -0.003 |
Token | ately |
Token ID | 1286 |
Feature activation | -7.47e+00 |
Pos logit contributions | |
least | +0.053 |
Little | +0.053 |
Em | +0.047 |
Maria | +0.046 |
ime | +0.043 |
Neg logit contributions | |
badges | -0.052 |
ows | -0.051 |
tricks | -0.050 |
names | -0.049 |
cks | -0.048 |
Token | with |
Token ID | 351 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
least | +0.206 |
Little | +0.182 |
ime | +0.174 |
illy | +0.165 |
ately | +0.155 |
Neg logit contributions | |
badges | -0.313 |
ows | -0.306 |
tricks | -0.294 |
names | -0.292 |
diapers | -0.292 |
Token | his |
Token ID | 465 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
badges | +0.044 |
ows | +0.042 |
tricks | +0.042 |
names | +0.041 |
diapers | +0.041 |
Neg logit contributions | |
least | -0.044 |
Little | -0.040 |
ime | -0.038 |
illy | -0.037 |
ately | -0.036 |
Token | fishing |
Token ID | 12478 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
badges | +0.039 |
tricks | +0.037 |
names | +0.037 |
ows | +0.036 |
diapers | +0.035 |
Neg logit contributions | |
least | -0.045 |
Little | -0.042 |
ime | -0.039 |
illy | -0.039 |
Maria | -0.039 |
Token | rod |
Token ID | 15299 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
badges | +0.122 |
ows | +0.119 |
tricks | +0.114 |
names | +0.112 |
diapers | +0.111 |
Neg logit contributions | |
least | -0.152 |
Little | -0.141 |
ime | -0.135 |
illy | -0.131 |
Maria | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
badges | +0.033 |
ows | +0.033 |
tricks | +0.032 |
names | +0.031 |
diapers | +0.031 |
Neg logit contributions | |
least | -0.030 |
Little | -0.028 |
ime | -0.026 |
illy | -0.025 |
Maria | -0.024 |
Token | day |
Token ID | 1110 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
least | +0.110 |
Little | +0.097 |
illy | +0.093 |
ately | +0.091 |
Daddy | +0.087 |
Neg logit contributions | |
tricks | -0.137 |
ows | -0.134 |
badges | -0.134 |
names | -0.131 |
labels | -0.130 |
Token | . |
Token ID | 13 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
least | +0.039 |
Little | +0.035 |
ime | +0.033 |
illy | +0.032 |
Maria | +0.031 |
Neg logit contributions | |
ows | -0.045 |
badges | -0.045 |
tricks | -0.043 |
names | -0.042 |
diapers | -0.042 |
Token | They |
Token ID | 1119 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
least | +0.063 |
Little | +0.058 |
ime | +0.056 |
illy | +0.054 |
ately | +0.053 |
Neg logit contributions | |
badges | -0.038 |
ows | -0.037 |
names | -0.035 |
tricks | -0.035 |
diapers | -0.034 |
Token | worked |
Token ID | 3111 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
badges | +0.038 |
ows | +0.036 |
names | +0.035 |
tricks | +0.034 |
labels | +0.034 |
Neg logit contributions | |
least | -0.032 |
Little | -0.030 |
ime | -0.030 |
illy | -0.029 |
ately | -0.027 |
Token | delic |
Token ID | 8675 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
least | +0.061 |
Little | +0.055 |
ime | +0.053 |
illy | +0.051 |
ately | +0.049 |
Neg logit contributions | |
badges | -0.062 |
ows | -0.061 |
tricks | -0.058 |
names | -0.057 |
diapers | -0.057 |
Token | ately |
Token ID | 1286 |
Feature activation | -7.18e+00 |
Pos logit contributions | |
least | +0.007 |
Little | +0.006 |
Maria | +0.005 |
Em | +0.004 |
ime | +0.004 |
Neg logit contributions | |
ows | -0.024 |
badges | -0.023 |
tricks | -0.022 |
names | -0.022 |
diapers | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -9.03e-01 |
Pos logit contributions | |
least | +0.241 |
Little | +0.224 |
illy | +0.207 |
ime | +0.203 |
Maria | +0.202 |
Neg logit contributions | |
badges | -0.254 |
ows | -0.248 |
tricks | -0.248 |
labels | -0.237 |
names | -0.236 |
Token | carefully |
Token ID | 7773 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
least | +0.026 |
Little | +0.025 |
ime | +0.024 |
illy | +0.023 |
Maria | +0.021 |
Neg logit contributions | |
badges | -0.035 |
ows | -0.033 |
tricks | -0.032 |
labels | -0.032 |
names | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
least | +0.042 |
Little | +0.038 |
illy | +0.035 |
ime | +0.035 |
Maria | +0.033 |
Neg logit contributions | |
badges | -0.053 |
ows | -0.053 |
tricks | -0.051 |
names | -0.049 |
labels | -0.049 |
Token | nothing |
Token ID | 2147 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
least | +0.018 |
Little | +0.016 |
ime | +0.015 |
illy | +0.015 |
Maria | +0.014 |
Neg logit contributions | |
badges | -0.027 |
ows | -0.026 |
tricks | -0.025 |
names | -0.025 |
diapers | -0.025 |
Token | broke |
Token ID | 6265 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
ows | +0.056 |
badges | +0.054 |
tricks | +0.053 |
names | +0.050 |
diapers | +0.050 |
Neg logit contributions | |
least | -0.047 |
Little | -0.043 |
ime | -0.040 |
illy | -0.039 |
Maria | -0.038 |
Token | Yes |
Token ID | 5297 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ows | +0.015 |
badges | +0.015 |
tricks | +0.014 |
diapers | +0.014 |
labels | +0.014 |
Neg logit contributions | |
least | -0.007 |
Little | -0.007 |
ime | -0.006 |
illy | -0.006 |
3 | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
least | +0.025 |
Little | +0.021 |
ime | +0.020 |
illy | +0.019 |
Maria | +0.018 |
Neg logit contributions | |
badges | -0.049 |
ows | -0.048 |
tricks | -0.047 |
names | -0.046 |
diapers | -0.046 |
Token | I |
Token ID | 314 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
least | +0.021 |
Little | +0.020 |
ime | +0.018 |
illy | +0.018 |
ately | +0.017 |
Neg logit contributions | |
badges | -0.030 |
ows | -0.029 |
tricks | -0.028 |
diapers | -0.028 |
names | -0.027 |
Token | am |
Token ID | 716 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
least | +0.095 |
Little | +0.088 |
ime | +0.085 |
illy | +0.082 |
ately | +0.079 |
Neg logit contributions | |
badges | -0.064 |
ows | -0.063 |
tricks | -0.059 |
names | -0.058 |
diapers | -0.058 |
Token | G |
Token ID | 402 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
badges | +0.019 |
ows | +0.018 |
names | +0.017 |
tricks | +0.017 |
diapers | +0.017 |
Neg logit contributions | |
least | -0.015 |
Little | -0.013 |
ime | -0.013 |
illy | -0.012 |
Little | -0.012 |
Token | ert |
Token ID | 861 |
Feature activation | -6.83e+00 |
Pos logit contributions | |
least | +0.057 |
Little | +0.048 |
Maria | +0.043 |
ime | +0.041 |
Em | +0.040 |
Neg logit contributions | |
ows | -0.097 |
badges | -0.090 |
tricks | -0.089 |
names | -0.088 |
cuts | -0.088 |
Token | ie |
Token ID | 494 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
least | +0.217 |
Little | +0.178 |
Maria | +0.162 |
illy | +0.160 |
Em | +0.157 |
Neg logit contributions | |
ows | -0.260 |
tricks | -0.242 |
aches | -0.235 |
cks | -0.233 |
badges | -0.230 |
Token | ! |
Token ID | 0 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
least | +0.064 |
Little | +0.056 |
ime | +0.053 |
illy | +0.048 |
ately | +0.045 |
Neg logit contributions | |
badges | -0.122 |
ows | -0.116 |
names | -0.114 |
diapers | -0.114 |
tricks | -0.113 |
Token | â |
Token ID | 22940 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
least | +0.065 |
Little | +0.059 |
ime | +0.058 |
illy | +0.055 |
ately | +0.053 |
Neg logit contributions | |
badges | -0.063 |
ows | -0.061 |
tricks | -0.059 |
names | -0.058 |
diapers | -0.058 |
Token | € |
Token ID | 26391 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
least | +0.089 |
Little | +0.082 |
ime | +0.077 |
illy | +0.075 |
Em | +0.069 |
Neg logit contributions | |
badges | -0.089 |
ows | -0.084 |
diapers | -0.082 |
names | -0.082 |
tricks | -0.082 |
Token | Mr |
Token ID | 1770 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
badges | +0.021 |
ows | +0.020 |
tricks | +0.020 |
diapers | +0.020 |
names | +0.019 |
Neg logit contributions | |
least | -0.016 |
Little | -0.015 |
ime | -0.014 |
illy | -0.013 |
Little | -0.013 |
Token | very |
Token ID | 845 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
least | +0.008 |
Little | +0.007 |
ime | +0.007 |
illy | +0.007 |
Maria | +0.006 |
Neg logit contributions | |
badges | -0.008 |
ows | -0.008 |
tricks | -0.007 |
diapers | -0.007 |
names | -0.007 |
Token | special |
Token ID | 2041 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
badges | +0.083 |
ows | +0.080 |
cks | +0.079 |
girls | +0.077 |
bows | +0.077 |
Neg logit contributions | |
least | -0.044 |
Little | -0.043 |
illy | -0.042 |
ime | -0.042 |
3 | -0.041 |
Token | toy |
Token ID | 13373 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
least | +0.060 |
Little | +0.056 |
ime | +0.054 |
illy | +0.052 |
3 | +0.049 |
Neg logit contributions | |
ows | -0.051 |
badges | -0.051 |
cks | -0.046 |
tricks | -0.046 |
names | -0.046 |
Token | for |
Token ID | 329 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
least | +0.018 |
Little | +0.016 |
ime | +0.015 |
illy | +0.015 |
ately | +0.014 |
Neg logit contributions | |
badges | -0.023 |
ows | -0.023 |
tricks | -0.022 |
names | -0.021 |
orts | -0.021 |
Token | L |
Token ID | 406 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
least | +0.001 |
Little | +0.001 |
ime | +0.001 |
Maria | +0.001 |
illy | +0.000 |
Neg logit contributions | |
badges | -0.001 |
tricks | -0.001 |
diapers | -0.001 |
labels | -0.001 |
ows | -0.001 |
Token | ott |
Token ID | 1252 |
Feature activation | -6.78e+00 |
Pos logit contributions | |
Little | +0.082 |
least | +0.082 |
Peter | +0.069 |
mid | +0.069 |
where | +0.069 |
Neg logit contributions | |
ows | -0.120 |
aches | -0.118 |
labels | -0.110 |
homework | -0.109 |
its | -0.105 |
Token | ie |
Token ID | 494 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
least | +0.132 |
Little | +0.112 |
ime | +0.099 |
illy | +0.088 |
ately | +0.084 |
Neg logit contributions | |
badges | -0.338 |
ows | -0.337 |
tricks | -0.327 |
names | -0.323 |
diapers | -0.321 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
least | +0.106 |
Little | +0.091 |
ime | +0.088 |
illy | +0.085 |
ately | +0.080 |
Neg logit contributions | |
badges | -0.149 |
ows | -0.145 |
tricks | -0.140 |
orts | -0.139 |
diapers | -0.139 |
Token | It |
Token ID | 632 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
least | +0.045 |
Little | +0.041 |
ime | +0.039 |
illy | +0.038 |
ately | +0.036 |
Neg logit contributions | |
badges | -0.047 |
ows | -0.046 |
tricks | -0.044 |
diapers | -0.043 |
names | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
badges | +0.002 |
ows | +0.002 |
tricks | +0.002 |
names | +0.002 |
diapers | +0.002 |
Neg logit contributions | |
least | -0.001 |
Little | -0.001 |
ime | -0.001 |
illy | -0.001 |
ately | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
least | +0.046 |
Little | +0.040 |
ime | +0.037 |
illy | +0.037 |
Little | +0.033 |
Neg logit contributions | |
badges | -0.075 |
ows | -0.075 |
orts | -0.072 |
tricks | -0.071 |
cuts | -0.070 |
Token | to |
Token ID | 284 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
least | +0.043 |
Little | +0.037 |
ime | +0.035 |
illy | +0.034 |
ately | +0.032 |
Neg logit contributions | |
ows | -0.065 |
badges | -0.065 |
tricks | -0.062 |
diapers | -0.062 |
names | -0.061 |
Token | wait |
Token ID | 4043 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ows | +0.020 |
badges | +0.020 |
orts | +0.018 |
diapers | +0.018 |
tricks | +0.018 |
Neg logit contributions | |
least | -0.019 |
Little | -0.017 |
ime | -0.016 |
illy | -0.016 |
ately | -0.014 |
Token | until |
Token ID | 1566 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
least | +0.036 |
Little | +0.034 |
ime | +0.032 |
Maria | +0.030 |
illy | +0.030 |
Neg logit contributions | |
ows | -0.037 |
badges | -0.036 |
tricks | -0.036 |
diapers | -0.035 |
names | -0.035 |
Token | d |
Token ID | 288 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
least | +0.087 |
Little | +0.078 |
ime | +0.074 |
illy | +0.071 |
ately | +0.067 |
Neg logit contributions | |
badges | -0.115 |
ows | -0.112 |
tricks | -0.108 |
names | -0.107 |
diapers | -0.106 |
Token | inn |
Token ID | 3732 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
badges | +0.018 |
ows | +0.018 |
tricks | +0.017 |
names | +0.017 |
diapers | +0.017 |
Neg logit contributions | |
least | -0.014 |
Little | -0.013 |
ime | -0.013 |
illy | -0.012 |
ately | -0.011 |
Token | ert |
Token ID | 861 |
Feature activation | -6.54e+00 |
Pos logit contributions | |
least | +0.040 |
Little | +0.035 |
ime | +0.033 |
illy | +0.031 |
Maria | +0.030 |
Neg logit contributions | |
badges | -0.060 |
ows | -0.059 |
tricks | -0.056 |
names | -0.056 |
diapers | -0.056 |
Token | ime |
Token ID | 524 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
least | +0.191 |
Little | +0.155 |
Peter | +0.138 |
Maria | +0.136 |
3 | +0.136 |
Neg logit contributions | |
ows | -0.273 |
aches | -0.243 |
cks | -0.231 |
badges | -0.230 |
ves | -0.229 |
Token | . |
Token ID | 13 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
least | +0.042 |
Little | +0.037 |
ime | +0.035 |
illy | +0.034 |
Maria | +0.032 |
Neg logit contributions | |
badges | -0.055 |
ows | -0.054 |
tricks | -0.052 |
names | -0.051 |
diapers | -0.051 |
Token |
|
Token ID | 198 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
least | +0.012 |
Little | +0.011 |
Little | +0.009 |
Em | +0.009 |
ime | +0.009 |
Neg logit contributions | |
badges | -0.025 |
tricks | -0.023 |
ows | -0.023 |
aches | -0.023 |
cks | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
least | +0.053 |
Little | +0.042 |
ime | +0.042 |
illy | +0.040 |
ately | +0.039 |
Neg logit contributions | |
ows | -0.090 |
badges | -0.089 |
tricks | -0.087 |
names | -0.086 |
orts | -0.085 |
Token | Anna |
Token ID | 31160 |
Feature activation | +7.10e-01 |
Pos logit contributions | |
least | +0.057 |
Little | +0.055 |
ime | +0.049 |
illy | +0.047 |
Em | +0.046 |
Neg logit contributions | |
badges | -0.084 |
ows | -0.081 |
tricks | -0.078 |
names | -0.078 |
diapers | -0.077 |
Token | little |
Token ID | 1310 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
badges | +0.011 |
ows | +0.011 |
tricks | +0.011 |
names | +0.011 |
diapers | +0.010 |
Neg logit contributions | |
least | -0.011 |
Little | -0.010 |
ime | -0.010 |
illy | -0.009 |
Maria | -0.009 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
least | +0.064 |
Little | +0.054 |
ime | +0.053 |
illy | +0.050 |
ately | +0.046 |
Neg logit contributions | |
badges | -0.112 |
ows | -0.111 |
tricks | -0.105 |
cks | -0.104 |
diapers | -0.103 |
Token | 's |
Token ID | 338 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
least | +0.060 |
ime | +0.051 |
Little | +0.051 |
ately | +0.046 |
ill | +0.045 |
Neg logit contributions | |
badges | -0.106 |
diapers | -0.098 |
cuts | -0.097 |
orts | -0.097 |
ows | -0.096 |
Token | shoulders |
Token ID | 12450 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
least | +0.031 |
Little | +0.028 |
ime | +0.027 |
illy | +0.025 |
ately | +0.025 |
Neg logit contributions | |
badges | -0.033 |
ows | -0.032 |
diapers | -0.030 |
tricks | -0.030 |
orts | -0.030 |
Token | affection |
Token ID | 17696 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
least | +0.088 |
Little | +0.081 |
ime | +0.075 |
illy | +0.075 |
Maria | +0.072 |
Neg logit contributions | |
ows | -0.094 |
badges | -0.091 |
tricks | -0.087 |
names | -0.086 |
diapers | -0.085 |
Token | ately |
Token ID | 1286 |
Feature activation | -6.44e+00 |
Pos logit contributions | |
least | +0.021 |
Little | +0.019 |
Em | +0.016 |
Maria | +0.015 |
ime | +0.015 |
Neg logit contributions | |
ows | -0.027 |
diapers | -0.024 |
badges | -0.023 |
names | -0.023 |
hats | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
least | +0.177 |
Little | +0.158 |
ime | +0.156 |
illy | +0.143 |
ately | +0.139 |
Neg logit contributions | |
badges | -0.272 |
ows | -0.258 |
names | -0.252 |
diapers | -0.250 |
tricks | -0.245 |
Token |
|
Token ID | 198 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
least | +0.042 |
ime | +0.038 |
illy | +0.035 |
Little | +0.035 |
ately | +0.033 |
Neg logit contributions | |
ows | -0.038 |
badges | -0.037 |
names | -0.036 |
name | -0.035 |
orts | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
least | +0.098 |
ately | +0.078 |
ime | +0.078 |
illy | +0.072 |
bul | +0.070 |
Neg logit contributions | |
ows | -0.086 |
badges | -0.082 |
names | -0.081 |
orts | -0.081 |
name | -0.079 |
Token | Sarah |
Token ID | 29284 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
least | +0.086 |
ime | +0.069 |
ately | +0.067 |
illy | +0.065 |
ocean | +0.061 |
Neg logit contributions | |
badges | -0.085 |
helmets | -0.080 |
ows | -0.079 |
names | -0.078 |
labels | -0.078 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
least | +0.053 |
Little | +0.047 |
ime | +0.047 |
illy | +0.045 |
ately | +0.043 |
Neg logit contributions | |
badges | -0.059 |
ows | -0.057 |
tricks | -0.056 |
names | -0.055 |
wrists | -0.053 |
Token | uff |
Token ID | 1648 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
badges | +0.057 |
ows | +0.056 |
tricks | +0.052 |
names | +0.052 |
diapers | +0.052 |
Neg logit contributions | |
least | -0.065 |
Little | -0.059 |
ime | -0.056 |
illy | -0.054 |
Maria | -0.051 |
Token | led |
Token ID | 992 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
badges | +0.041 |
ows | +0.041 |
tricks | +0.038 |
diapers | +0.037 |
names | +0.037 |
Neg logit contributions | |
least | -0.060 |
Little | -0.055 |
ime | -0.053 |
illy | -0.051 |
Maria | -0.050 |
Token | his |
Token ID | 465 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
ows | +0.070 |
badges | +0.068 |
cuts | +0.066 |
orts | +0.065 |
tricks | +0.065 |
Neg logit contributions | |
least | -0.045 |
Little | -0.041 |
ime | -0.039 |
Little | -0.038 |
3 | -0.037 |
Token | hair |
Token ID | 4190 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
badges | +0.073 |
ows | +0.073 |
tricks | +0.068 |
names | +0.067 |
diapers | +0.067 |
Neg logit contributions | |
least | -0.096 |
Little | -0.088 |
ime | -0.085 |
illy | -0.082 |
ately | -0.080 |
Token | affection |
Token ID | 17696 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
badges | +0.158 |
ows | +0.156 |
tricks | +0.150 |
diapers | +0.148 |
names | +0.147 |
Neg logit contributions | |
least | -0.089 |
Little | -0.078 |
ime | -0.073 |
illy | -0.069 |
Maria | -0.066 |
Token | ately |
Token ID | 1286 |
Feature activation | -6.42e+00 |
Pos logit contributions | |
least | +0.005 |
Little | +0.005 |
Maria | +0.004 |
Em | +0.004 |
ime | +0.004 |
Neg logit contributions | |
ows | -0.008 |
badges | -0.008 |
diapers | -0.008 |
names | -0.007 |
teeth | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -9.15e-01 |
Pos logit contributions | |
least | +0.139 |
Little | +0.122 |
ime | +0.121 |
illy | +0.111 |
ately | +0.108 |
Neg logit contributions | |
badges | -0.296 |
ows | -0.286 |
names | -0.278 |
cuts | -0.273 |
orts | -0.273 |
Token | |
Token ID | 220 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
least | +0.034 |
ime | +0.029 |
illy | +0.027 |
ately | +0.027 |
Little | +0.026 |
Neg logit contributions | |
name | -0.030 |
names | -0.030 |
orts | -0.028 |
ows | -0.028 |
badges | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
least | +0.006 |
Little | +0.005 |
ime | +0.005 |
illy | +0.004 |
ately | +0.004 |
Neg logit contributions | |
ows | -0.009 |
badges | -0.009 |
names | -0.009 |
orts | -0.008 |
tricks | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
least | +0.090 |
ately | +0.072 |
ime | +0.072 |
illy | +0.070 |
bul | +0.068 |
Neg logit contributions | |
orts | -0.059 |
ows | -0.057 |
name | -0.054 |
names | -0.053 |
ends | -0.052 |
Token | The |
Token ID | 464 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
least | +0.072 |
ately | +0.057 |
ime | +0.056 |
illy | +0.055 |
ill | +0.048 |
Neg logit contributions | |
badges | -0.071 |
helmets | -0.068 |
tricks | -0.066 |
diapers | -0.066 |
chips | -0.066 |
Token | enjoy |
Token ID | 2883 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
least | +0.039 |
Little | +0.032 |
ime | +0.031 |
ately | +0.029 |
Maria | +0.029 |
Neg logit contributions | |
ows | -0.044 |
badges | -0.044 |
orts | -0.042 |
cuts | -0.041 |
diapers | -0.040 |
Token | it |
Token ID | 340 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
least | +0.001 |
Little | +0.000 |
ime | +0.000 |
illy | +0.000 |
Maria | +0.000 |
Neg logit contributions | |
ows | -0.001 |
badges | -0.001 |
orts | -0.001 |
names | -0.001 |
diapers | -0.001 |
Token | until |
Token ID | 1566 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
least | +0.049 |
Little | +0.044 |
ime | +0.042 |
illy | +0.041 |
Maria | +0.039 |
Neg logit contributions | |
badges | -0.052 |
ows | -0.052 |
tricks | -0.050 |
names | -0.049 |
diapers | -0.049 |
Token | d |
Token ID | 288 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
least | +0.100 |
Little | +0.098 |
ime | +0.096 |
illy | +0.092 |
Maria | +0.091 |
Neg logit contributions | |
badges | -0.092 |
helmets | -0.088 |
tricks | -0.088 |
diapers | -0.086 |
names | -0.085 |
Token | inn |
Token ID | 3732 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
badges | +0.039 |
ows | +0.037 |
tricks | +0.036 |
diapers | +0.036 |
names | +0.035 |
Neg logit contributions | |
least | -0.030 |
ime | -0.027 |
Little | -0.027 |
illy | -0.025 |
ately | -0.024 |
Token | ert |
Token ID | 861 |
Feature activation | -6.40e+00 |
Pos logit contributions | |
least | +0.037 |
Little | +0.034 |
ime | +0.031 |
illy | +0.029 |
Maria | +0.029 |
Neg logit contributions | |
ows | -0.044 |
badges | -0.044 |
tricks | -0.041 |
names | -0.041 |
diapers | -0.040 |
Token | ime |
Token ID | 524 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
least | +0.173 |
Little | +0.144 |
Peter | +0.135 |
Mill | +0.132 |
Maria | +0.130 |
Neg logit contributions | |
ows | -0.275 |
aches | -0.247 |
cks | -0.239 |
ves | -0.233 |
bows | -0.232 |
Token | . |
Token ID | 13 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
least | +0.041 |
Little | +0.037 |
ime | +0.035 |
illy | +0.034 |
Maria | +0.033 |
Neg logit contributions | |
badges | -0.045 |
ows | -0.045 |
tricks | -0.043 |
names | -0.042 |
diapers | -0.042 |
Token | |
Token ID | 220 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
least | +0.025 |
Little | +0.022 |
ime | +0.022 |
illy | +0.022 |
ately | +0.021 |
Neg logit contributions | |
names | -0.030 |
badges | -0.029 |
ows | -0.029 |
tricks | -0.028 |
name | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
least | +0.002 |
Little | +0.002 |
ime | +0.001 |
illy | +0.001 |
ately | +0.001 |
Neg logit contributions | |
ows | -0.003 |
badges | -0.003 |
names | -0.003 |
tricks | -0.003 |
orts | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
least | +0.066 |
ately | +0.055 |
illy | +0.054 |
ime | +0.052 |
Little | +0.049 |
Neg logit contributions | |
ows | -0.064 |
orts | -0.063 |
tricks | -0.063 |
names | -0.063 |
badges | -0.060 |
Token | so |
Token ID | 523 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
least | +0.058 |
Little | +0.053 |
ime | +0.051 |
illy | +0.049 |
Maria | +0.046 |
Neg logit contributions | |
badges | -0.087 |
ows | -0.085 |
tricks | -0.084 |
names | -0.082 |
diapers | -0.081 |
Token | he |
Token ID | 339 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ows | +0.069 |
badges | +0.069 |
names | +0.065 |
diapers | +0.065 |
tricks | +0.064 |
Neg logit contributions | |
least | -0.047 |
Little | -0.042 |
ime | -0.039 |
illy | -0.037 |
ately | -0.036 |
Token | stepped |
Token ID | 10764 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
least | +0.038 |
Little | +0.034 |
ime | +0.033 |
illy | +0.031 |
Maria | +0.029 |
Neg logit contributions | |
badges | -0.058 |
ows | -0.057 |
tricks | -0.055 |
names | -0.054 |
diapers | -0.054 |
Token | forward |
Token ID | 2651 |
Feature activation | +9.37e-01 |
Pos logit contributions | |
least | +0.057 |
ime | +0.053 |
Little | +0.052 |
Maria | +0.049 |
Em | +0.049 |
Neg logit contributions | |
ows | -0.049 |
badges | -0.047 |
tricks | -0.045 |
names | -0.045 |
diapers | -0.044 |
Token | courage |
Token ID | 11917 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
ows | +0.029 |
tricks | +0.026 |
names | +0.024 |
badges | +0.024 |
name | +0.023 |
Neg logit contributions | |
Little | -0.037 |
least | -0.037 |
Em | -0.035 |
ime | -0.034 |
illy | -0.034 |
Token | ously |
Token ID | 3481 |
Feature activation | -6.22e+00 |
Pos logit contributions | |
least | +0.043 |
Little | +0.040 |
ime | +0.037 |
illy | +0.037 |
Maria | +0.035 |
Neg logit contributions | |
ows | -0.056 |
badges | -0.055 |
tricks | -0.053 |
helmets | -0.051 |
names | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
least | +0.170 |
Little | +0.146 |
ime | +0.143 |
ately | +0.135 |
illy | +0.134 |
Neg logit contributions | |
badges | -0.260 |
ows | -0.251 |
diapers | -0.245 |
tricks | -0.242 |
names | -0.240 |
Token |
|
Token ID | 198 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
least | +0.029 |
Little | +0.026 |
ime | +0.025 |
illy | +0.024 |
ately | +0.022 |
Neg logit contributions | |
badges | -0.045 |
ows | -0.044 |
tricks | -0.043 |
names | -0.042 |
diapers | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
least | +0.072 |
illy | +0.055 |
ime | +0.054 |
ately | +0.054 |
Little | +0.052 |
Neg logit contributions | |
ows | -0.080 |
badges | -0.076 |
orts | -0.075 |
tricks | -0.075 |
names | -0.075 |
Token | When |
Token ID | 2215 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
least | +0.051 |
Little | +0.044 |
ime | +0.042 |
illy | +0.039 |
Maria | +0.036 |
Neg logit contributions | |
badges | -0.097 |
ows | -0.095 |
tricks | -0.092 |
names | -0.091 |
diapers | -0.091 |
Token | he |
Token ID | 339 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
badges | +0.055 |
tricks | +0.051 |
helmets | +0.050 |
chips | +0.049 |
names | +0.049 |
Neg logit contributions | |
least | -0.065 |
ime | -0.060 |
Little | -0.058 |
illy | -0.058 |
ill | -0.056 |
Token | it |
Token ID | 340 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
least | +0.007 |
Little | +0.006 |
ime | +0.006 |
illy | +0.006 |
Maria | +0.005 |
Neg logit contributions | |
badges | -0.009 |
ows | -0.008 |
tricks | -0.008 |
diapers | -0.008 |
names | -0.008 |
Token | went |
Token ID | 1816 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
badges | +0.037 |
ows | +0.036 |
tricks | +0.035 |
names | +0.035 |
hides | +0.034 |
Neg logit contributions | |
least | -0.028 |
Little | -0.025 |
ime | -0.025 |
illy | -0.025 |
ately | -0.023 |
Token | on |
Token ID | 319 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
least | +0.084 |
Little | +0.076 |
ime | +0.072 |
illy | +0.069 |
ately | +0.066 |
Neg logit contributions | |
badges | -0.097 |
ows | -0.095 |
tricks | -0.091 |
names | -0.090 |
diapers | -0.090 |
Token | the |
Token ID | 262 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
least | +0.001 |
ime | +0.001 |
Little | +0.001 |
illy | +0.001 |
ately | +0.001 |
Neg logit contributions | |
badges | -0.001 |
tricks | -0.001 |
helmets | -0.001 |
homework | -0.001 |
labels | -0.001 |
Token | w |
Token ID | 266 |
Feature activation | +7.90e-01 |
Pos logit contributions | |
least | +0.037 |
ime | +0.036 |
Little | +0.035 |
illy | +0.034 |
ately | +0.034 |
Neg logit contributions | |
badges | -0.035 |
tricks | -0.034 |
helmets | -0.033 |
names | -0.033 |
ows | -0.032 |
Token | ett |
Token ID | 3087 |
Feature activation | -5.98e+00 |
Pos logit contributions | |
badges | +0.021 |
ows | +0.021 |
tricks | +0.020 |
names | +0.019 |
diapers | +0.019 |
Neg logit contributions | |
least | -0.034 |
Little | -0.032 |
Maria | -0.029 |
ime | -0.029 |
Em | -0.029 |
Token | est |
Token ID | 395 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
Little | +0.174 |
least | +0.174 |
Maria | +0.162 |
Em | +0.158 |
enny | +0.156 |
Neg logit contributions | |
tricks | -0.221 |
badges | -0.209 |
names | -0.208 |
ows | -0.205 |
bones | -0.200 |
Token | of |
Token ID | 286 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
least | +0.064 |
Little | +0.059 |
Maria | +0.055 |
Em | +0.052 |
illy | +0.052 |
Neg logit contributions | |
tricks | -0.083 |
badges | -0.083 |
names | -0.078 |
diapers | -0.076 |
helmets | -0.076 |
Token | adventures |
Token ID | 17545 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
least | +0.018 |
Little | +0.016 |
ime | +0.016 |
illy | +0.015 |
ately | +0.014 |
Neg logit contributions | |
badges | -0.018 |
ows | -0.018 |
tricks | -0.017 |
names | -0.017 |
diapers | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
least | +0.048 |
Little | +0.043 |
ime | +0.041 |
illy | +0.040 |
Maria | +0.038 |
Neg logit contributions | |
badges | -0.056 |
ows | -0.055 |
tricks | -0.053 |
names | -0.052 |
diapers | -0.052 |
Token | After |
Token ID | 2293 |
Feature activation | -4.70e-02 |
Pos logit contributions | |
least | +0.046 |
Little | +0.043 |
ime | +0.041 |
illy | +0.039 |
ately | +0.038 |
Neg logit contributions | |
badges | -0.046 |
ows | -0.044 |
tricks | -0.043 |
diapers | -0.041 |
names | -0.041 |
Token | ! |
Token ID | 0 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
least | +0.001 |
Little | +0.001 |
ime | +0.001 |
Little | +0.001 |
Peter | +0.001 |
Neg logit contributions | |
badges | -0.004 |
ows | -0.004 |
orts | -0.004 |
diapers | -0.003 |
irms | -0.003 |
Token | I |
Token ID | 314 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
least | +0.067 |
Little | +0.061 |
ime | +0.058 |
illy | +0.056 |
ately | +0.054 |
Neg logit contributions | |
badges | -0.072 |
ows | -0.070 |
tricks | -0.068 |
names | -0.067 |
diapers | -0.066 |
Token | am |
Token ID | 716 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
least | +0.062 |
Little | +0.055 |
ime | +0.052 |
illy | +0.050 |
ately | +0.047 |
Neg logit contributions | |
badges | -0.082 |
ows | -0.080 |
diapers | -0.077 |
tricks | -0.077 |
names | -0.076 |
Token | the |
Token ID | 262 |
Feature activation | -6.86e-01 |
Pos logit contributions | |
badges | +0.024 |
ows | +0.024 |
diapers | +0.022 |
names | +0.022 |
tricks | +0.022 |
Neg logit contributions | |
least | -0.018 |
Little | -0.015 |
ime | -0.014 |
illy | -0.014 |
Maria | -0.013 |
Token | fastest |
Token ID | 14162 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
least | +0.022 |
Little | +0.021 |
ime | +0.020 |
illy | +0.019 |
ately | +0.019 |
Neg logit contributions | |
badges | -0.025 |
tricks | -0.024 |
ows | -0.023 |
names | -0.023 |
diapers | -0.023 |
Token | rusher |
Token ID | 47608 |
Feature activation | -5.97e+00 |
Pos logit contributions | |
badges | +0.003 |
tricks | +0.003 |
ows | +0.003 |
names | +0.003 |
diapers | +0.003 |
Neg logit contributions | |
least | -0.003 |
Little | -0.003 |
illy | -0.002 |
ime | -0.002 |
Maria | -0.002 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
least | +0.189 |
Little | +0.170 |
ime | +0.163 |
illy | +0.156 |
ately | +0.148 |
Neg logit contributions | |
badges | -0.225 |
ows | -0.221 |
tricks | -0.211 |
names | -0.209 |
diapers | -0.209 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
badges | +0.051 |
ows | +0.050 |
tricks | +0.048 |
names | +0.048 |
diapers | +0.047 |
Neg logit contributions | |
least | -0.039 |
Little | -0.033 |
ime | -0.033 |
illy | -0.031 |
ately | -0.029 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
ows | +0.125 |
badges | +0.123 |
names | +0.121 |
tricks | +0.118 |
diapers | +0.116 |
Neg logit contributions | |
least | -0.089 |
Little | -0.072 |
illy | -0.070 |
ime | -0.068 |
ately | -0.068 |
Token | there |
Token ID | 612 |
Feature activation | +8.26e-01 |
Pos logit contributions | |
ows | +0.112 |
names | +0.104 |
badges | +0.104 |
orts | +0.103 |
tricks | +0.101 |
Neg logit contributions | |
least | -0.090 |
Little | -0.078 |
ime | -0.070 |
illy | -0.069 |
ately | -0.068 |
Token | was |
Token ID | 373 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
names | +0.026 |
ows | +0.025 |
name | +0.025 |
tricks | +0.024 |
badges | +0.024 |
Neg logit contributions | |
least | -0.027 |
illy | -0.024 |
Little | -0.024 |
ime | -0.024 |
ately | -0.023 |
Token | who |
Token ID | 508 |
Feature activation | +9.42e-01 |
Pos logit contributions | |
least | +0.084 |
Little | +0.076 |
ime | +0.073 |
illy | +0.070 |
ately | +0.067 |
Neg logit contributions | |
badges | -0.085 |
ows | -0.083 |
tricks | -0.079 |
names | -0.078 |
diapers | -0.078 |
Token | could |
Token ID | 714 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
badges | +0.039 |
ows | +0.039 |
tricks | +0.037 |
names | +0.037 |
diapers | +0.037 |
Neg logit contributions | |
least | -0.026 |
Little | -0.023 |
ime | -0.022 |
illy | -0.021 |
Maria | -0.019 |
Token | get |
Token ID | 651 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
least | +0.082 |
Little | +0.072 |
ime | +0.069 |
illy | +0.065 |
ill | +0.064 |
Neg logit contributions | |
badges | -0.085 |
ows | -0.082 |
diapers | -0.077 |
names | -0.076 |
cuts | -0.076 |
Token | the |
Token ID | 262 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
ows | +0.008 |
badges | +0.007 |
facts | +0.007 |
cuts | +0.007 |
meanings | +0.007 |
Neg logit contributions | |
least | -0.005 |
ime | -0.004 |
Little | -0.004 |
ful | -0.004 |
Little | -0.003 |
Token | w |
Token ID | 266 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
least | +0.064 |
Little | +0.054 |
ime | +0.050 |
illy | +0.048 |
3 | +0.048 |
Neg logit contributions | |
ows | -0.075 |
badges | -0.075 |
cks | -0.072 |
orts | -0.071 |
tricks | -0.069 |
Token | ett |
Token ID | 3087 |
Feature activation | -5.97e+00 |
Pos logit contributions | |
badges | +0.007 |
ows | +0.006 |
names | +0.006 |
tricks | +0.006 |
diapers | +0.006 |
Neg logit contributions | |
least | -0.015 |
Little | -0.014 |
ime | -0.014 |
illy | -0.014 |
ately | -0.013 |
Token | est |
Token ID | 395 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
least | +0.204 |
Little | +0.185 |
ime | +0.177 |
illy | +0.171 |
ately | +0.163 |
Neg logit contributions | |
badges | -0.211 |
ows | -0.207 |
tricks | -0.198 |
names | -0.195 |
diapers | -0.195 |
Token | and |
Token ID | 290 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
least | +0.068 |
Little | +0.059 |
ime | +0.058 |
illy | +0.054 |
ately | +0.052 |
Neg logit contributions | |
badges | -0.093 |
ows | -0.091 |
tricks | -0.087 |
names | -0.087 |
diapers | -0.086 |
Token | who |
Token ID | 508 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
least | +0.053 |
Little | +0.048 |
ime | +0.044 |
Little | +0.042 |
3 | +0.042 |
Neg logit contributions | |
ows | -0.057 |
badges | -0.054 |
orts | -0.053 |
cks | -0.050 |
cuts | -0.050 |
Token | could |
Token ID | 714 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
badges | +0.034 |
ows | +0.034 |
tricks | +0.033 |
names | +0.032 |
diapers | +0.032 |
Neg logit contributions | |
least | -0.022 |
Little | -0.019 |
ime | -0.018 |
illy | -0.017 |
ately | -0.016 |
Token | make |
Token ID | 787 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
least | +0.081 |
Little | +0.070 |
ime | +0.067 |
illy | +0.062 |
ill | +0.062 |
Neg logit contributions | |
badges | -0.088 |
ows | -0.085 |
diapers | -0.081 |
cuts | -0.080 |
facts | -0.079 |
Token | work |
Token ID | 670 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
least | +0.020 |
Little | +0.018 |
ime | +0.018 |
illy | +0.017 |
ately | +0.016 |
Neg logit contributions | |
badges | -0.016 |
ows | -0.015 |
tricks | -0.015 |
names | -0.015 |
labels | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
least | +0.024 |
Little | +0.022 |
ime | +0.021 |
illy | +0.021 |
Maria | +0.019 |
Neg logit contributions | |
ows | -0.033 |
badges | -0.032 |
tricks | -0.031 |
labels | -0.030 |
helmets | -0.030 |
Token | carefully |
Token ID | 7773 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
least | +0.048 |
Little | +0.044 |
ime | +0.043 |
illy | +0.041 |
ately | +0.037 |
Neg logit contributions | |
badges | -0.076 |
ows | -0.074 |
tricks | -0.071 |
labels | -0.070 |
names | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | -7.75e-01 |
Pos logit contributions | |
least | +0.056 |
Little | +0.050 |
illy | +0.047 |
ime | +0.046 |
Maria | +0.044 |
Neg logit contributions | |
badges | -0.071 |
ows | -0.068 |
tricks | -0.066 |
labels | -0.065 |
names | -0.064 |
Token | delic |
Token ID | 8675 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
least | +0.022 |
Little | +0.021 |
ime | +0.020 |
illy | +0.019 |
Maria | +0.017 |
Neg logit contributions | |
badges | -0.031 |
ows | -0.030 |
tricks | -0.029 |
labels | -0.029 |
names | -0.028 |
Token | ately |
Token ID | 1286 |
Feature activation | -5.95e+00 |
Pos logit contributions | |
Little | +0.002 |
least | +0.002 |
Em | +0.001 |
Maria | +0.001 |
ime | +0.001 |
Neg logit contributions | |
ows | -0.006 |
badges | -0.006 |
names | -0.006 |
tricks | -0.006 |
diapers | -0.006 |
Token | repairing |
Token ID | 39005 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
least | +0.186 |
Little | +0.165 |
ime | +0.154 |
illy | +0.152 |
Maria | +0.145 |
Neg logit contributions | |
badges | -0.231 |
ows | -0.224 |
tricks | -0.218 |
diapers | -0.211 |
names | -0.211 |
Token | the |
Token ID | 262 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
least | +0.005 |
Little | +0.005 |
ime | +0.004 |
illy | +0.004 |
ately | +0.004 |
Neg logit contributions | |
badges | -0.006 |
ows | -0.006 |
tricks | -0.006 |
names | -0.006 |
diapers | -0.006 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
least | +0.030 |
Little | +0.028 |
ime | +0.026 |
illy | +0.025 |
Maria | +0.024 |
Neg logit contributions | |
badges | -0.036 |
ows | -0.035 |
tricks | -0.034 |
names | -0.033 |
diapers | -0.033 |
Token | 's |
Token ID | 338 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
least | +0.071 |
Little | +0.064 |
ime | +0.061 |
illy | +0.058 |
ately | +0.055 |
Neg logit contributions | |
badges | -0.087 |
ows | -0.085 |
tricks | -0.082 |
names | -0.081 |
diapers | -0.081 |
Token | wings |
Token ID | 12098 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
least | +0.065 |
Little | +0.059 |
ime | +0.057 |
illy | +0.055 |
ately | +0.053 |
Neg logit contributions | |
badges | -0.061 |
ows | -0.059 |
tricks | -0.056 |
diapers | -0.055 |
names | -0.055 |
Token | amazed |
Token ID | 24562 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
least | +0.050 |
Little | +0.043 |
ime | +0.041 |
illy | +0.038 |
Maria | +0.036 |
Neg logit contributions | |
badges | -0.088 |
ows | -0.086 |
tricks | -0.083 |
diapers | -0.082 |
names | -0.082 |
Token | by |
Token ID | 416 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
least | +0.042 |
Little | +0.038 |
ime | +0.035 |
illy | +0.034 |
Maria | +0.032 |
Neg logit contributions | |
badges | -0.058 |
ows | -0.057 |
tricks | -0.056 |
names | -0.055 |
diapers | -0.054 |
Token | how |
Token ID | 703 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
badges | +0.010 |
ows | +0.010 |
tricks | +0.010 |
diapers | +0.010 |
names | +0.010 |
Neg logit contributions | |
least | -0.008 |
Little | -0.007 |
ime | -0.007 |
illy | -0.006 |
Maria | -0.006 |
Token | good |
Token ID | 922 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
badges | +0.106 |
ows | +0.103 |
diapers | +0.099 |
tricks | +0.096 |
orts | +0.095 |
Neg logit contributions | |
least | -0.079 |
Little | -0.078 |
ime | -0.075 |
Maria | -0.069 |
illy | -0.069 |
Token | the |
Token ID | 262 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
least | +0.011 |
Little | +0.010 |
illy | +0.009 |
ime | +0.009 |
Little | +0.009 |
Neg logit contributions | |
ows | -0.016 |
badges | -0.016 |
orts | -0.015 |
gag | -0.015 |
cuts | -0.014 |
Token | mother |
Token ID | 2802 |
Feature activation | -5.94e+00 |
Pos logit contributions | |
least | +0.071 |
Little | +0.064 |
ime | +0.060 |
illy | +0.058 |
Maria | +0.053 |
Neg logit contributions | |
badges | -0.101 |
ows | -0.101 |
diapers | -0.094 |
tricks | -0.094 |
orts | -0.092 |
Token | kept |
Token ID | 4030 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
least | +0.151 |
Little | +0.132 |
ime | +0.125 |
illy | +0.118 |
ately | +0.110 |
Neg logit contributions | |
badges | -0.262 |
ows | -0.257 |
tricks | -0.249 |
names | -0.246 |
diapers | -0.245 |
Token | it |
Token ID | 340 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
least | +0.097 |
Little | +0.085 |
ime | +0.082 |
illy | +0.079 |
ately | +0.074 |
Neg logit contributions | |
ows | -0.134 |
badges | -0.134 |
tricks | -0.125 |
diapers | -0.124 |
names | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
least | +0.081 |
Little | +0.077 |
ime | +0.072 |
illy | +0.069 |
Maria | +0.067 |
Neg logit contributions | |
ows | -0.089 |
badges | -0.087 |
tricks | -0.084 |
names | -0.083 |
diapers | -0.081 |
Token | |
Token ID | 220 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
least | +0.042 |
Little | +0.037 |
ime | +0.037 |
illy | +0.035 |
ately | +0.033 |
Neg logit contributions | |
badges | -0.046 |
ows | -0.046 |
names | -0.044 |
tricks | -0.043 |
diapers | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
least | +0.012 |
Little | +0.011 |
ime | +0.010 |
illy | +0.009 |
ately | +0.009 |
Neg logit contributions | |
badges | -0.026 |
ows | -0.025 |
tricks | -0.024 |
names | -0.024 |
diapers | -0.024 |
Token | knowing |
Token ID | 6970 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
least | +0.058 |
Little | +0.052 |
ime | +0.050 |
illy | +0.047 |
ately | +0.045 |
Neg logit contributions | |
badges | -0.081 |
ows | -0.080 |
tricks | -0.077 |
names | -0.076 |
diapers | -0.076 |
Token | that |
Token ID | 326 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
least | +0.069 |
Little | +0.064 |
ime | +0.063 |
illy | +0.057 |
Maria | +0.057 |
Neg logit contributions | |
badges | -0.060 |
tricks | -0.058 |
ows | -0.058 |
names | -0.056 |
diapers | -0.055 |
Token | even |
Token ID | 772 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
least | +0.049 |
Little | +0.044 |
ime | +0.041 |
illy | +0.039 |
Maria | +0.037 |
Neg logit contributions | |
badges | -0.069 |
ows | -0.068 |
tricks | -0.064 |
diapers | -0.064 |
names | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -9.68e-01 |
Pos logit contributions | |
least | +0.089 |
Little | +0.080 |
ime | +0.076 |
illy | +0.073 |
ately | +0.069 |
Neg logit contributions | |
badges | -0.117 |
ows | -0.114 |
tricks | -0.110 |
names | -0.108 |
diapers | -0.108 |
Token | t |
Token ID | 256 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
least | +0.028 |
Little | +0.025 |
ime | +0.024 |
illy | +0.023 |
ately | +0.022 |
Neg logit contributions | |
badges | -0.039 |
ows | -0.039 |
tricks | -0.037 |
diapers | -0.036 |
names | -0.036 |
Token | ini |
Token ID | 5362 |
Feature activation | -5.93e+00 |
Pos logit contributions | |
least | +0.070 |
Little | +0.064 |
Little | +0.058 |
mid | +0.055 |
Em | +0.055 |
Neg logit contributions | |
ows | -0.067 |
badges | -0.063 |
bows | -0.062 |
names | -0.062 |
ages | -0.062 |
Token | est |
Token ID | 395 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
least | +0.186 |
Little | +0.171 |
ime | +0.161 |
Em | +0.153 |
ately | +0.152 |
Neg logit contributions | |
ows | -0.206 |
tricks | -0.202 |
names | -0.201 |
badges | -0.198 |
bows | -0.194 |
Token | and |
Token ID | 290 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
least | +0.060 |
Little | +0.055 |
ime | +0.051 |
illy | +0.049 |
Maria | +0.048 |
Neg logit contributions | |
badges | -0.068 |
ows | -0.066 |
tricks | -0.066 |
names | -0.065 |
diapers | -0.064 |
Token | darkest |
Token ID | 35995 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
least | +0.069 |
Little | +0.061 |
ime | +0.057 |
illy | +0.055 |
Maria | +0.053 |
Neg logit contributions | |
badges | -0.069 |
ows | -0.069 |
orts | -0.064 |
tricks | -0.064 |
diapers | -0.063 |
Token | of |
Token ID | 286 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
least | +0.093 |
Little | +0.085 |
illy | +0.075 |
Em | +0.073 |
Maria | +0.073 |
Neg logit contributions | |
ows | -0.123 |
tricks | -0.123 |
badges | -0.121 |
diapers | -0.118 |
names | -0.118 |
Token | gardens |
Token ID | 25476 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
least | +0.036 |
Little | +0.034 |
ime | +0.031 |
illy | +0.030 |
ill | +0.029 |
Neg logit contributions | |
badges | -0.041 |
tricks | -0.040 |
names | -0.039 |
ows | -0.039 |
diapers | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
least | +0.018 |
Little | +0.016 |
ime | +0.015 |
illy | +0.014 |
ately | +0.013 |
Neg logit contributions | |
badges | -0.026 |
ows | -0.025 |
tricks | -0.025 |
names | -0.024 |
diapers | -0.024 |
Token | swim |
Token ID | 9422 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
least | +0.035 |
Little | +0.033 |
illy | +0.033 |
ime | +0.032 |
ately | +0.032 |
Neg logit contributions | |
badges | -0.025 |
tricks | -0.024 |
names | -0.024 |
ows | -0.023 |
diapers | -0.022 |
Token | around |
Token ID | 1088 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
badges | +0.103 |
ows | +0.097 |
diapers | +0.096 |
cuts | +0.094 |
labels | +0.093 |
Neg logit contributions | |
least | -0.074 |
Little | -0.065 |
ime | -0.063 |
illy | -0.058 |
ill | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
least | +0.025 |
Little | +0.024 |
illy | +0.023 |
ime | +0.022 |
Em | +0.022 |
Neg logit contributions | |
tricks | -0.022 |
badges | -0.020 |
names | -0.020 |
ows | -0.020 |
labels | -0.019 |
Token | p |
Token ID | 279 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
least | +0.038 |
Little | +0.035 |
ately | +0.033 |
Maria | +0.033 |
illy | +0.033 |
Neg logit contributions | |
badges | -0.046 |
tricks | -0.045 |
names | -0.044 |
ows | -0.044 |
labels | -0.042 |
Token | ail |
Token ID | 603 |
Feature activation | -5.92e+00 |
Pos logit contributions | |
least | +0.082 |
Little | +0.074 |
ime | +0.073 |
illy | +0.070 |
ill | +0.067 |
Neg logit contributions | |
badges | -0.060 |
ows | -0.057 |
diapers | -0.056 |
tricks | -0.056 |
names | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
least | +0.215 |
Little | +0.206 |
illy | +0.189 |
Maria | +0.187 |
Em | +0.187 |
Neg logit contributions | |
ows | -0.192 |
tricks | -0.182 |
badges | -0.175 |
names | -0.169 |
chips | -0.167 |
Token | The |
Token ID | 383 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
least | +0.039 |
Little | +0.035 |
ime | +0.034 |
illy | +0.032 |
Maria | +0.031 |
Neg logit contributions | |
badges | -0.038 |
ows | -0.037 |
tricks | -0.036 |
diapers | -0.035 |
names | -0.035 |
Token | fisherman |
Token ID | 48750 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
least | +0.007 |
Little | +0.006 |
ime | +0.006 |
illy | +0.006 |
Maria | +0.006 |
Neg logit contributions | |
badges | -0.007 |
ows | -0.007 |
tricks | -0.007 |
names | -0.007 |
diapers | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
tricks | +0.024 |
badges | +0.023 |
ows | +0.023 |
names | +0.022 |
labels | +0.021 |
Neg logit contributions | |
least | -0.022 |
Little | -0.021 |
illy | -0.021 |
ime | -0.020 |
Em | -0.019 |
Token | so |
Token ID | 523 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
least | +0.030 |
Little | +0.028 |
ime | +0.027 |
illy | +0.026 |
ately | +0.025 |
Neg logit contributions | |
badges | -0.048 |
ows | -0.048 |
tricks | -0.047 |
names | -0.046 |
diapers | -0.045 |
Token | hopped |
Token ID | 45230 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
least | +0.016 |
Little | +0.015 |
illy | +0.014 |
ime | +0.014 |
Em | +0.013 |
Neg logit contributions | |
ows | -0.018 |
badges | -0.018 |
tricks | -0.017 |
aches | -0.016 |
helmets | -0.016 |
Token | back |
Token ID | 736 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
least | +0.039 |
Little | +0.036 |
ime | +0.034 |
illy | +0.033 |
ately | +0.032 |
Neg logit contributions | |
badges | -0.040 |
ows | -0.039 |
tricks | -0.037 |
names | -0.037 |
diapers | -0.036 |
Token | onto |
Token ID | 4291 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
least | +0.022 |
Little | +0.020 |
ime | +0.019 |
illy | +0.018 |
Maria | +0.017 |
Neg logit contributions | |
badges | -0.031 |
ows | -0.030 |
tricks | -0.029 |
names | -0.029 |
diapers | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
least | +0.059 |
Little | +0.057 |
ime | +0.056 |
illy | +0.054 |
enny | +0.054 |
Neg logit contributions | |
badges | -0.057 |
tricks | -0.055 |
names | -0.054 |
diapers | -0.053 |
labels | -0.053 |
Token | windows |
Token ID | 9168 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
least | +0.064 |
ime | +0.061 |
ill | +0.059 |
Little | +0.058 |
illy | +0.058 |
Neg logit contributions | |
badges | -0.062 |
tricks | -0.059 |
helmets | -0.059 |
diapers | -0.058 |
ows | -0.058 |
Token | ill |
Token ID | 359 |
Feature activation | -5.90e+00 |
Pos logit contributions | |
Little | +0.069 |
least | +0.067 |
Em | +0.063 |
Quick | +0.060 |
illy | +0.059 |
Neg logit contributions | |
ows | -0.063 |
bows | -0.049 |
names | -0.049 |
badges | -0.049 |
tricks | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
least | +0.221 |
Little | +0.202 |
illy | +0.182 |
Maria | +0.181 |
ime | +0.178 |
Neg logit contributions | |
ows | -0.195 |
badges | -0.178 |
tricks | -0.178 |
names | -0.177 |
diapers | -0.168 |
Token | The |
Token ID | 383 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
least | +0.072 |
Little | +0.067 |
ime | +0.066 |
illy | +0.064 |
ately | +0.061 |
Neg logit contributions | |
ows | -0.038 |
names | -0.037 |
badges | -0.036 |
orts | -0.035 |
tricks | -0.034 |
Token | pr |
Token ID | 778 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
least | +0.009 |
Little | +0.009 |
ime | +0.009 |
illy | +0.008 |
Em | +0.008 |
Neg logit contributions | |
badges | -0.009 |
names | -0.009 |
tricks | -0.009 |
labels | -0.008 |
ows | -0.008 |
Token | une |
Token ID | 1726 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
ows | +0.042 |
badges | +0.039 |
names | +0.037 |
labels | +0.036 |
orts | +0.036 |
Neg logit contributions | |
least | -0.051 |
Little | -0.050 |
Em | -0.046 |
ime | -0.045 |
Peter | -0.044 |
Token | whispered |
Token ID | 25029 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
ows | +0.002 |
badges | +0.002 |
tricks | +0.002 |
aches | +0.002 |
helmets | +0.002 |
Neg logit contributions | |
least | -0.002 |
Little | -0.001 |
illy | -0.001 |
ime | -0.001 |
Maria | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +7.81e-01 |
Pos logit contributions | |
badges | +0.000 |
ows | +0.000 |
tricks | +0.000 |
names | +0.000 |
diapers | +0.000 |
Neg logit contributions | |
least | -0.000 |
Little | -0.000 |
ime | -0.000 |
illy | -0.000 |
ately | -0.000 |
Token | little |
Token ID | 1310 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
badges | +0.030 |
tricks | +0.028 |
names | +0.027 |
chips | +0.027 |
ows | +0.027 |
Neg logit contributions | |
least | -0.025 |
Little | -0.022 |
illy | -0.021 |
ately | -0.020 |
ime | -0.020 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
least | +0.085 |
Little | +0.076 |
ime | +0.071 |
illy | +0.069 |
Maria | +0.066 |
Neg logit contributions | |
badges | -0.124 |
ows | -0.118 |
tricks | -0.117 |
names | -0.115 |
diapers | -0.114 |
Token | named |
Token ID | 3706 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
least | +0.073 |
Little | +0.065 |
ime | +0.061 |
illy | +0.059 |
ately | +0.056 |
Neg logit contributions | |
badges | -0.094 |
ows | -0.092 |
tricks | -0.089 |
names | -0.088 |
diapers | -0.087 |
Token | G |
Token ID | 402 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
badges | +0.069 |
ows | +0.068 |
tricks | +0.066 |
names | +0.063 |
diapers | +0.063 |
Neg logit contributions | |
least | -0.052 |
Little | -0.048 |
ime | -0.045 |
illy | -0.045 |
Maria | -0.042 |
Token | ert |
Token ID | 861 |
Feature activation | -5.89e+00 |
Pos logit contributions | |
least | +0.041 |
Little | +0.037 |
Maria | +0.033 |
ime | +0.033 |
3 | +0.031 |
Neg logit contributions | |
ows | -0.063 |
badges | -0.060 |
tricks | -0.058 |
orts | -0.057 |
names | -0.057 |
Token | ie |
Token ID | 494 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
least | +0.178 |
Little | +0.149 |
Maria | +0.133 |
Peter | +0.130 |
illy | +0.128 |
Neg logit contributions | |
ows | -0.223 |
aches | -0.208 |
tricks | -0.203 |
bones | -0.201 |
badges | -0.198 |
Token | , |
Token ID | 11 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
least | +0.036 |
Little | +0.031 |
ime | +0.030 |
illy | +0.028 |
ately | +0.026 |
Neg logit contributions | |
badges | -0.073 |
ows | -0.071 |
tricks | -0.069 |
names | -0.068 |
diapers | -0.068 |
Token | stopped |
Token ID | 5025 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
badges | +0.011 |
ows | +0.010 |
names | +0.010 |
tricks | +0.010 |
diapers | +0.010 |
Neg logit contributions | |
least | -0.008 |
Little | -0.007 |
ime | -0.007 |
illy | -0.007 |
ately | -0.006 |
Token | by |
Token ID | 416 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
least | +0.034 |
Little | +0.032 |
ime | +0.029 |
illy | +0.029 |
Maria | +0.028 |
Neg logit contributions | |
badges | -0.038 |
tricks | -0.037 |
ows | -0.036 |
names | -0.035 |
diapers | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
badges | +0.024 |
tricks | +0.023 |
ows | +0.023 |
names | +0.023 |
diapers | +0.023 |
Neg logit contributions | |
least | -0.021 |
Little | -0.019 |
ime | -0.019 |
illy | -0.018 |
Maria | -0.017 |
Token | que |
Token ID | 4188 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
least | +0.058 |
Little | +0.052 |
Maria | +0.048 |
ime | +0.046 |
Daddy | +0.045 |
Neg logit contributions | |
ows | -0.083 |
badges | -0.079 |
tricks | -0.076 |
names | -0.075 |
orts | -0.074 |
Token | aky |
Token ID | 15492 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
badges | +0.077 |
ows | +0.077 |
tricks | +0.073 |
names | +0.072 |
diapers | +0.071 |
Neg logit contributions | |
least | -0.060 |
Little | -0.055 |
ime | -0.049 |
illy | -0.049 |
Maria | -0.046 |
Token | in |
Token ID | 287 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
least | +0.018 |
Little | +0.016 |
ime | +0.016 |
illy | +0.014 |
ill | +0.014 |
Neg logit contributions | |
badges | -0.027 |
ows | -0.026 |
tricks | -0.025 |
orts | -0.025 |
cuts | -0.025 |
Token | its |
Token ID | 663 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
badges | +0.039 |
ows | +0.037 |
tricks | +0.036 |
names | +0.036 |
diapers | +0.036 |
Neg logit contributions | |
least | -0.036 |
Little | -0.032 |
ime | -0.031 |
illy | -0.031 |
ately | -0.029 |
Token | sharp |
Token ID | 7786 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
badges | +0.042 |
ows | +0.040 |
tricks | +0.039 |
names | +0.039 |
diapers | +0.037 |
Neg logit contributions | |
least | -0.044 |
Little | -0.041 |
illy | -0.039 |
ime | -0.039 |
Em | -0.037 |
Token | tal |
Token ID | 3305 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
least | +0.046 |
Little | +0.043 |
illy | +0.039 |
ime | +0.039 |
Em | +0.038 |
Neg logit contributions | |
badges | -0.033 |
ows | -0.032 |
tricks | -0.031 |
diapers | -0.030 |
names | -0.030 |
Token | ons |
Token ID | 684 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
least | +0.219 |
Little | +0.188 |
3 | +0.173 |
ime | +0.172 |
ately | +0.171 |
Neg logit contributions | |
badges | -0.196 |
ows | -0.190 |
tricks | -0.187 |
cks | -0.186 |
aches | -0.179 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
least | +0.032 |
Little | +0.029 |
ime | +0.028 |
illy | +0.026 |
Maria | +0.025 |
Neg logit contributions | |
badges | -0.039 |
ows | -0.039 |
tricks | -0.037 |
names | -0.037 |
diapers | -0.036 |
Token | |
Token ID | 220 |
Feature activation | -7.95e-01 |
Pos logit contributions | |
least | +0.039 |
Little | +0.035 |
ime | +0.032 |
illy | +0.031 |
ately | +0.030 |
Neg logit contributions | |
badges | -0.062 |
ows | -0.060 |
tricks | -0.059 |
diapers | -0.058 |
homework | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
least | +0.019 |
Little | +0.016 |
ime | +0.015 |
illy | +0.014 |
ately | +0.013 |
Neg logit contributions | |
ows | -0.036 |
badges | -0.036 |
tricks | -0.035 |
names | -0.035 |
diapers | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
least | +0.069 |
Little | +0.057 |
ime | +0.055 |
illy | +0.052 |
ately | +0.049 |
Neg logit contributions | |
badges | -0.125 |
ows | -0.124 |
tricks | -0.120 |
names | -0.119 |
orts | -0.118 |
Token | the |
Token ID | 262 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
Little | +0.017 |
enny | +0.017 |
Em | +0.017 |
ime | +0.016 |
Maria | +0.015 |
Neg logit contributions | |
tricks | -0.012 |
labels | -0.012 |
names | -0.012 |
homework | -0.012 |
diapers | -0.012 |
Token | top |
Token ID | 1353 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
Em | +0.046 |
enny | +0.045 |
ime | +0.044 |
Little | +0.044 |
Maria | +0.043 |
Neg logit contributions | |
names | -0.046 |
tricks | -0.044 |
chips | -0.042 |
name | -0.042 |
diapers | -0.041 |
Token | of |
Token ID | 286 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
least | +0.095 |
Little | +0.093 |
illy | +0.092 |
Peter | +0.090 |
Em | +0.090 |
Neg logit contributions | |
ows | -0.078 |
tricks | -0.074 |
chips | -0.070 |
bones | -0.068 |
ashes | -0.067 |
Token | the |
Token ID | 262 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
enny | +0.035 |
Em | +0.032 |
ime | +0.030 |
illy | +0.029 |
enna | +0.029 |
Neg logit contributions | |
helmets | -0.017 |
badges | -0.016 |
chips | -0.016 |
labels | -0.016 |
names | -0.016 |
Token | anth |
Token ID | 26794 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
Em | +0.057 |
enny | +0.057 |
enna | +0.054 |
bul | +0.054 |
Peter | +0.054 |
Neg logit contributions | |
chips | -0.051 |
names | -0.050 |
ashes | -0.049 |
tricks | -0.049 |
diapers | -0.048 |
Token | ill |
Token ID | 359 |
Feature activation | -5.86e+00 |
Pos logit contributions | |
least | +0.066 |
Little | +0.066 |
Em | +0.051 |
Peter | +0.051 |
ime | +0.046 |
Neg logit contributions | |
ows | -0.155 |
ashes | -0.142 |
bones | -0.140 |
bows | -0.137 |
oodle | -0.137 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
least | +0.222 |
Little | +0.208 |
Em | +0.193 |
Peter | +0.193 |
Maria | +0.192 |
Neg logit contributions | |
ows | -0.176 |
tricks | -0.169 |
names | -0.156 |
aches | -0.156 |
chips | -0.155 |
Token | But |
Token ID | 887 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
least | +0.060 |
Little | +0.055 |
ime | +0.054 |
illy | +0.052 |
ately | +0.049 |
Neg logit contributions | |
badges | -0.047 |
ows | -0.047 |
names | -0.045 |
tricks | -0.044 |
diapers | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
ime | +0.086 |
Little | +0.082 |
illy | +0.081 |
enny | +0.081 |
least | +0.079 |
Neg logit contributions | |
tricks | -0.084 |
hides | -0.081 |
helmets | -0.080 |
badges | -0.079 |
ows | -0.079 |
Token | ant |
Token ID | 1885 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
names | +0.003 |
tricks | +0.003 |
chips | +0.003 |
badges | +0.003 |
diapers | +0.003 |
Neg logit contributions | |
least | -0.002 |
Em | -0.002 |
ime | -0.002 |
enny | -0.002 |
Little | -0.002 |
Token | wouldn |
Token ID | 3636 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
least | +0.056 |
Little | +0.050 |
illy | +0.049 |
ime | +0.048 |
Maria | +0.045 |
Neg logit contributions | |
ows | -0.064 |
tricks | -0.064 |
badges | -0.062 |
aches | -0.059 |
hides | -0.059 |
Token | There |
Token ID | 1318 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
badges | +0.059 |
tricks | +0.055 |
ows | +0.055 |
aches | +0.053 |
labels | +0.052 |
Neg logit contributions | |
least | -0.045 |
Little | -0.043 |
ime | -0.039 |
Maria | -0.038 |
Little | -0.038 |
Token | were |
Token ID | 547 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
least | +0.024 |
Little | +0.021 |
ime | +0.020 |
illy | +0.019 |
Em | +0.018 |
Neg logit contributions | |
ows | -0.034 |
badges | -0.033 |
names | -0.032 |
tricks | -0.032 |
labels | -0.031 |
Token | many |
Token ID | 867 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
least | +0.047 |
Little | +0.041 |
ime | +0.039 |
Maria | +0.037 |
illy | +0.037 |
Neg logit contributions | |
badges | -0.052 |
ows | -0.051 |
orts | -0.050 |
tricks | -0.048 |
cuts | -0.048 |
Token | people |
Token ID | 661 |
Feature activation | -8.59e-01 |
Pos logit contributions | |
least | +0.016 |
Little | +0.015 |
ime | +0.015 |
illy | +0.014 |
Maria | +0.013 |
Neg logit contributions | |
ows | -0.014 |
badges | -0.013 |
orts | -0.013 |
gag | -0.013 |
names | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
least | +0.023 |
Little | +0.020 |
ime | +0.020 |
illy | +0.018 |
ately | +0.017 |
Neg logit contributions | |
badges | -0.037 |
ows | -0.035 |
names | -0.035 |
diapers | -0.035 |
tricks | -0.035 |
Token | clown |
Token ID | 25573 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
ows | +0.048 |
orts | +0.045 |
badges | +0.044 |
cks | +0.042 |
cuts | +0.042 |
Neg logit contributions | |
least | -0.053 |
Little | -0.049 |
ime | -0.047 |
Maria | -0.044 |
illy | -0.044 |
Token | s |
Token ID | 82 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
badges | +0.162 |
ows | +0.156 |
tricks | +0.154 |
diapers | +0.154 |
names | +0.151 |
Neg logit contributions | |
least | -0.170 |
Little | -0.156 |
ime | -0.153 |
illy | -0.144 |
ately | -0.142 |
Token | and |
Token ID | 290 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
badges | +0.088 |
ows | +0.085 |
tricks | +0.083 |
diapers | +0.082 |
names | +0.082 |
Neg logit contributions | |
least | -0.051 |
Little | -0.044 |
ime | -0.042 |
illy | -0.039 |
ately | -0.037 |
Token | animals |
Token ID | 4695 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
ows | +0.075 |
badges | +0.069 |
orts | +0.069 |
cuts | +0.067 |
tricks | +0.065 |
Neg logit contributions | |
least | -0.072 |
Little | -0.062 |
ime | -0.060 |
Maria | -0.057 |
illy | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
badges | +0.136 |
ows | +0.132 |
diapers | +0.129 |
orts | +0.128 |
tricks | +0.128 |
Neg logit contributions | |
least | -0.083 |
Little | -0.069 |
ime | -0.069 |
illy | -0.061 |
ately | -0.061 |
Token | Anna |
Token ID | 11735 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
badges | +0.079 |
aches | +0.076 |
tricks | +0.073 |
cuts | +0.072 |
ows | +0.072 |
Neg logit contributions | |
least | -0.069 |
Little | -0.065 |
Little | -0.063 |
ill | -0.062 |
Maria | -0.061 |
Token | can |
Token ID | 460 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
badges | +0.016 |
ows | +0.015 |
diapers | +0.015 |
tricks | +0.015 |
facts | +0.015 |
Neg logit contributions | |
least | -0.013 |
Little | -0.011 |
ime | -0.010 |
ately | -0.009 |
illy | -0.009 |
Token | 't |
Token ID | 470 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
least | +0.020 |
Little | +0.017 |
ime | +0.016 |
illy | +0.014 |
ately | +0.014 |
Neg logit contributions | |
badges | -0.029 |
ows | -0.029 |
diapers | -0.028 |
facts | -0.027 |
tricks | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
least | +0.011 |
Little | +0.010 |
ime | +0.009 |
illy | +0.009 |
Maria | +0.009 |
Neg logit contributions | |
badges | -0.011 |
ows | -0.010 |
tricks | -0.010 |
names | -0.010 |
diapers | -0.010 |
Token | You |
Token ID | 921 |
Feature activation | +7.20e-01 |
Pos logit contributions | |
least | +0.005 |
Little | +0.004 |
ime | +0.004 |
Little | +0.004 |
ately | +0.004 |
Neg logit contributions | |
badges | -0.007 |
ows | -0.007 |
tricks | -0.007 |
aches | -0.007 |
chips | -0.007 |
Token | are |
Token ID | 389 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
badges | +0.025 |
ows | +0.024 |
tricks | +0.023 |
diapers | +0.023 |
facts | +0.023 |
Neg logit contributions | |
least | -0.026 |
Little | -0.022 |
ime | -0.022 |
ately | -0.020 |
illy | -0.020 |
Token | too |
Token ID | 1165 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
ows | +0.033 |
badges | +0.032 |
orts | +0.031 |
bows | +0.031 |
facts | +0.030 |
Neg logit contributions | |
least | -0.023 |
nowhere | -0.016 |
3 | -0.015 |
mid | -0.015 |
ime | -0.014 |
Token | little |
Token ID | 1310 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
badges | +0.198 |
ows | +0.195 |
tricks | +0.187 |
diapers | +0.183 |
names | +0.183 |
Neg logit contributions | |
least | -0.162 |
Little | -0.153 |
ime | -0.148 |
illy | -0.133 |
ately | -0.127 |
Token | ." |
Token ID | 526 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
least | +0.060 |
Little | +0.051 |
ime | +0.050 |
illy | +0.048 |
ately | +0.046 |
Neg logit contributions | |
badges | -0.069 |
ows | -0.068 |
tricks | -0.064 |
names | -0.064 |
orts | -0.063 |
Token | Tim |
Token ID | 5045 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
badges | +0.048 |
ows | +0.047 |
tricks | +0.045 |
names | +0.045 |
diapers | +0.044 |
Neg logit contributions | |
least | -0.040 |
Little | -0.036 |
ime | -0.034 |
illy | -0.033 |
ately | -0.031 |
Token | was |
Token ID | 373 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
least | +0.017 |
Little | +0.016 |
ime | +0.015 |
illy | +0.014 |
Maria | +0.014 |
Neg logit contributions | |
badges | -0.021 |
ows | -0.021 |
tricks | -0.020 |
names | -0.019 |
diapers | -0.019 |
Token | sad |
Token ID | 6507 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
badges | +0.019 |
facts | +0.019 |
cuts | +0.018 |
bows | +0.018 |
ows | +0.018 |
Neg logit contributions | |
least | -0.011 |
3 | -0.009 |
nowhere | -0.009 |
mid | -0.007 |
three | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
least | +0.001 |
Little | +0.001 |
ime | +0.001 |
illy | +0.001 |
ately | +0.001 |
Neg logit contributions | |
ows | -0.001 |
badges | -0.001 |
tricks | -0.001 |
orts | -0.001 |
names | -0.001 |
Token | pen |
Token ID | 3112 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
badges | +0.043 |
ows | +0.042 |
tricks | +0.040 |
names | +0.040 |
diapers | +0.040 |
Neg logit contributions | |
least | -0.040 |
Little | -0.036 |
ime | -0.034 |
illy | -0.033 |
ately | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
badges | +0.015 |
ows | +0.015 |
tricks | +0.014 |
names | +0.014 |
diapers | +0.014 |
Neg logit contributions | |
least | -0.011 |
Little | -0.010 |
ime | -0.010 |
illy | -0.009 |
ately | -0.009 |
Token | He |
Token ID | 679 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
least | +0.022 |
Little | +0.019 |
ime | +0.019 |
illy | +0.018 |
ately | +0.017 |
Neg logit contributions | |
badges | -0.025 |
ows | -0.024 |
tricks | -0.023 |
names | -0.023 |
diapers | -0.023 |
Token | wrote |
Token ID | 2630 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
badges | +0.020 |
ows | +0.019 |
tricks | +0.018 |
names | +0.018 |
diapers | +0.018 |
Neg logit contributions | |
least | -0.016 |
Little | -0.014 |
ime | -0.014 |
illy | -0.013 |
Maria | -0.012 |
Token | something |
Token ID | 1223 |
Feature activation | +4.64e+00 |
Pos logit contributions | |
badges | +0.007 |
ows | +0.006 |
hides | +0.006 |
orts | +0.006 |
diapers | +0.006 |
Neg logit contributions | |
Little | -0.004 |
least | -0.004 |
ime | -0.004 |
illy | -0.004 |
Little | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
badges | +0.174 |
ows | +0.167 |
diapers | +0.159 |
orts | +0.158 |
names | +0.157 |
Neg logit contributions | |
least | -0.148 |
Little | -0.135 |
ime | -0.128 |
illy | -0.123 |
ately | -0.115 |
Token | put |
Token ID | 1234 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
least | +0.010 |
Little | +0.009 |
ime | +0.009 |
illy | +0.008 |
ately | +0.008 |
Neg logit contributions | |
badges | -0.013 |
ows | -0.013 |
tricks | -0.013 |
names | -0.012 |
diapers | -0.012 |
Token | it |
Token ID | 340 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
least | +0.021 |
Little | +0.019 |
ime | +0.018 |
illy | +0.017 |
Little | +0.016 |
Neg logit contributions | |
badges | -0.029 |
ows | -0.029 |
tricks | -0.028 |
gag | -0.028 |
hides | -0.027 |
Token | in |
Token ID | 287 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
least | +0.005 |
Little | +0.004 |
ime | +0.004 |
illy | +0.004 |
ately | +0.004 |
Neg logit contributions | |
badges | -0.006 |
ows | -0.006 |
tricks | -0.005 |
names | -0.005 |
diapers | -0.005 |
Token | an |
Token ID | 281 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
least | +0.003 |
Little | +0.002 |
ime | +0.002 |
illy | +0.002 |
ately | +0.002 |
Neg logit contributions | |
badges | -0.003 |
names | -0.003 |
diapers | -0.003 |
labels | -0.003 |
tricks | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
least | +0.014 |
Little | +0.012 |
ime | +0.012 |
ill | +0.011 |
illy | +0.011 |
Neg logit contributions | |
badges | -0.029 |
diapers | -0.028 |
helmets | -0.027 |
gag | -0.027 |
ows | -0.027 |
Token | for |
Token ID | 329 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
badges | +0.039 |
ows | +0.038 |
tricks | +0.036 |
diapers | +0.036 |
names | +0.035 |
Neg logit contributions | |
least | -0.033 |
Little | -0.029 |
ime | -0.027 |
illy | -0.027 |
ately | -0.025 |
Token | helping |
Token ID | 5742 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
badges | +0.064 |
ows | +0.061 |
tricks | +0.061 |
names | +0.059 |
diapers | +0.059 |
Neg logit contributions | |
least | -0.048 |
Little | -0.043 |
ime | -0.042 |
illy | -0.039 |
ately | -0.038 |
Token | us |
Token ID | 514 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
badges | +0.007 |
ows | +0.007 |
tricks | +0.007 |
names | +0.007 |
diapers | +0.006 |
Neg logit contributions | |
least | -0.004 |
Little | -0.004 |
ime | -0.004 |
ately | -0.003 |
illy | -0.003 |
Token | ," |
Token ID | 553 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
badges | +0.023 |
ows | +0.023 |
tricks | +0.022 |
names | +0.022 |
diapers | +0.021 |
Neg logit contributions | |
least | -0.022 |
Little | -0.020 |
ime | -0.020 |
illy | -0.019 |
ately | -0.018 |
Token | their |
Token ID | 511 |
Feature activation | +6.05e+00 |
Pos logit contributions | |
badges | +0.095 |
ows | +0.093 |
tricks | +0.088 |
names | +0.087 |
diapers | +0.086 |
Neg logit contributions | |
least | -0.117 |
Little | -0.108 |
ime | -0.104 |
illy | -0.100 |
ately | -0.096 |
Token | dad |
Token ID | 9955 |
Feature activation | +4.43e-01 |
Pos logit contributions | |
ows | +0.225 |
badges | +0.222 |
tricks | +0.207 |
diapers | +0.206 |
names | +0.203 |
Neg logit contributions | |
least | -0.194 |
Little | -0.171 |
ime | -0.168 |
illy | -0.160 |
ately | -0.154 |
Token | said |
Token ID | 531 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
badges | +0.014 |
ows | +0.013 |
tricks | +0.012 |
names | +0.012 |
labels | +0.012 |
Neg logit contributions | |
least | -0.018 |
Little | -0.016 |
ime | -0.016 |
illy | -0.015 |
ately | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
badges | +0.052 |
ows | +0.048 |
names | +0.048 |
tricks | +0.048 |
diapers | +0.048 |
Neg logit contributions | |
least | -0.035 |
Little | -0.032 |
ime | -0.031 |
illy | -0.029 |
ately | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +9.94e-01 |
Pos logit contributions | |
badges | +0.095 |
ows | +0.092 |
tricks | +0.088 |
names | +0.087 |
diapers | +0.086 |
Neg logit contributions | |
least | -0.110 |
Little | -0.101 |
ime | -0.098 |
illy | -0.094 |
ately | -0.090 |
Token | nice |
Token ID | 3621 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
badges | +0.035 |
ows | +0.034 |
tricks | +0.033 |
names | +0.032 |
diapers | +0.032 |
Neg logit contributions | |
least | -0.034 |
Little | -0.031 |
ime | -0.030 |
illy | -0.029 |
ately | -0.027 |
Token | ." |
Token ID | 526 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
badges | +0.134 |
ows | +0.130 |
tricks | +0.125 |
names | +0.123 |
diapers | +0.123 |
Neg logit contributions | |
least | -0.110 |
Little | -0.095 |
ime | -0.088 |
illy | -0.087 |
ately | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
badges | +0.125 |
ows | +0.121 |
tricks | +0.118 |
diapers | +0.117 |
cuts | +0.115 |
Neg logit contributions | |
least | -0.064 |
Little | -0.059 |
ime | -0.052 |
Little | -0.049 |
illy | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
badges | +0.034 |
ows | +0.034 |
tricks | +0.033 |
names | +0.033 |
orts | +0.032 |
Neg logit contributions | |
least | -0.021 |
Little | -0.017 |
ime | -0.017 |
illy | -0.016 |
ately | -0.016 |
Token | The |
Token ID | 464 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
badges | +0.034 |
ows | +0.034 |
tricks | +0.033 |
names | +0.032 |
diapers | +0.032 |
Neg logit contributions | |
least | -0.018 |
Little | -0.016 |
ime | -0.015 |
illy | -0.014 |
Maria | -0.013 |
Token | ice |
Token ID | 4771 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
ows | +0.064 |
badges | +0.057 |
diapers | +0.055 |
bows | +0.054 |
gag | +0.053 |
Neg logit contributions | |
least | -0.059 |
Little | -0.054 |
ime | -0.049 |
illy | -0.046 |
3 | -0.046 |
Token | cream |
Token ID | 8566 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
badges | +0.096 |
versa | +0.093 |
meanings | +0.087 |
uries | +0.087 |
diapers | +0.084 |
Neg logit contributions | |
least | -0.121 |
in | -0.113 |
ime | -0.109 |
ill | -0.107 |
oh | -0.106 |
Token | man |
Token ID | 582 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
badges | +0.144 |
ows | +0.142 |
preferences | +0.135 |
diapers | +0.134 |
names | +0.134 |
Neg logit contributions | |
least | -0.152 |
Little | -0.135 |
ime | -0.128 |
mid | -0.121 |
ill | -0.118 |
Token | gives |
Token ID | 3607 |
Feature activation | +8.01e-02 |
Pos logit contributions | |
ages | +0.047 |
diapers | +0.046 |
badges | +0.045 |
gag | +0.044 |
uries | +0.044 |
Neg logit contributions | |
least | -0.056 |
again | -0.046 |
Little | -0.045 |
mid | -0.045 |
ill | -0.044 |
Token | them |
Token ID | 606 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
badges | +0.003 |
ows | +0.003 |
tricks | +0.003 |
diapers | +0.002 |
gag | +0.002 |
Neg logit contributions | |
least | -0.003 |
Little | -0.003 |
ime | -0.003 |
illy | -0.002 |
ill | -0.002 |
Token | ice |
Token ID | 4771 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
ows | +0.105 |
badges | +0.102 |
tricks | +0.101 |
cks | +0.101 |
cuts | +0.099 |
Neg logit contributions | |
least | -0.069 |
Little | -0.058 |
illy | -0.052 |
ime | -0.050 |
ill | -0.050 |
Token | cream |
Token ID | 8566 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
badges | +0.051 |
meanings | +0.045 |
ows | +0.045 |
facts | +0.044 |
tricks | +0.044 |
Neg logit contributions | |
least | -0.094 |
ime | -0.085 |
Little | -0.083 |
illy | -0.083 |
ately | -0.082 |
Token | bug |
Token ID | 5434 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
badges | +0.006 |
tricks | +0.006 |
names | +0.006 |
chips | +0.006 |
labels | +0.006 |
Neg logit contributions | |
least | -0.007 |
ime | -0.007 |
Little | -0.007 |
Maria | -0.006 |
Em | -0.006 |
Token | ," |
Token ID | 553 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
least | +0.011 |
ime | +0.009 |
Little | +0.009 |
illy | +0.009 |
ill | +0.008 |
Neg logit contributions | |
badges | -0.017 |
names | -0.016 |
diapers | -0.016 |
helmets | -0.016 |
ows | -0.016 |
Token | Lily |
Token ID | 20037 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
ows | +0.091 |
badges | +0.091 |
tricks | +0.085 |
names | +0.083 |
diapers | +0.083 |
Neg logit contributions | |
least | -0.102 |
Little | -0.094 |
ime | -0.088 |
illy | -0.086 |
Maria | -0.082 |
Token | said |
Token ID | 531 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
badges | +0.030 |
ows | +0.029 |
tricks | +0.028 |
names | +0.028 |
diapers | +0.027 |
Neg logit contributions | |
least | -0.024 |
Little | -0.021 |
ime | -0.020 |
illy | -0.019 |
ately | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
badges | +0.049 |
ows | +0.047 |
tricks | +0.046 |
names | +0.045 |
diapers | +0.045 |
Neg logit contributions | |
least | -0.040 |
Little | -0.036 |
ime | -0.035 |
illy | -0.033 |
ately | -0.032 |
Token | " |
Token ID | 366 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
badges | +0.089 |
ows | +0.088 |
tricks | +0.083 |
names | +0.083 |
diapers | +0.082 |
Neg logit contributions | |
least | -0.095 |
Little | -0.086 |
ime | -0.084 |
illy | -0.080 |
ately | -0.076 |
Token | We |
Token ID | 1135 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
badges | +0.084 |
ows | +0.082 |
tricks | +0.080 |
names | +0.079 |
diapers | +0.078 |
Neg logit contributions | |
least | -0.057 |
Little | -0.051 |
ime | -0.048 |
illy | -0.046 |
ately | -0.043 |
Token | like |
Token ID | 588 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
bones | +0.012 |
cks | +0.012 |
diapers | +0.012 |
badges | +0.012 |
orts | +0.012 |
Neg logit contributions | |
least | -0.009 |
Little | -0.006 |
illy | -0.006 |
ill | -0.006 |
ime | -0.006 |
Token | bugs |
Token ID | 11316 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
ows | +0.131 |
badges | +0.128 |
diapers | +0.117 |
orts | +0.117 |
cuts | +0.116 |
Neg logit contributions | |
least | -0.086 |
Little | -0.071 |
ime | -0.070 |
illy | -0.067 |
Maria | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
least | +0.016 |
Little | +0.013 |
ime | +0.013 |
illy | +0.012 |
ill | +0.011 |
Neg logit contributions | |
badges | -0.028 |
ows | -0.026 |
names | -0.026 |
tricks | -0.026 |
diapers | -0.026 |
Token | flowers |
Token ID | 12734 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ows | +0.061 |
badges | +0.060 |
diapers | +0.054 |
orts | +0.054 |
cuts | +0.054 |
Neg logit contributions | |
least | -0.059 |
Little | -0.051 |
ime | -0.049 |
illy | -0.047 |
Maria | -0.045 |
Token | He |
Token ID | 679 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
least | +0.020 |
Little | +0.018 |
ime | +0.017 |
illy | +0.016 |
Maria | +0.015 |
Neg logit contributions | |
badges | -0.029 |
ows | -0.028 |
tricks | -0.027 |
diapers | -0.027 |
names | -0.027 |
Token | loved |
Token ID | 6151 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
badges | +0.006 |
ows | +0.005 |
tricks | +0.005 |
names | +0.005 |
diapers | +0.005 |
Neg logit contributions | |
least | -0.005 |
Little | -0.005 |
ime | -0.005 |
illy | -0.005 |
ately | -0.004 |
Token | playing |
Token ID | 2712 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
least | +0.018 |
Little | +0.016 |
ime | +0.015 |
illy | +0.014 |
Maria | +0.013 |
Neg logit contributions | |
badges | -0.027 |
ows | -0.027 |
tricks | -0.026 |
names | -0.026 |
diapers | -0.025 |
Token | with |
Token ID | 351 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
badges | +0.073 |
ows | +0.072 |
names | +0.067 |
diapers | +0.067 |
orts | +0.066 |
Neg logit contributions | |
least | -0.071 |
Little | -0.066 |
ime | -0.063 |
illy | -0.060 |
Maria | -0.057 |
Token | his |
Token ID | 465 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
ows | +0.014 |
badges | +0.012 |
orts | +0.012 |
cuts | +0.012 |
bows | +0.012 |
Neg logit contributions | |
least | -0.008 |
Little | -0.007 |
ime | -0.006 |
illy | -0.006 |
Little | -0.006 |
Token | toys |
Token ID | 14958 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
badges | +0.026 |
tricks | +0.024 |
ows | +0.024 |
names | +0.024 |
diapers | +0.023 |
Neg logit contributions | |
least | -0.051 |
Little | -0.048 |
ime | -0.047 |
illy | -0.046 |
Maria | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
ows | +0.025 |
badges | +0.025 |
tricks | +0.025 |
labels | +0.024 |
helmets | +0.024 |
Neg logit contributions | |
Little | -0.022 |
least | -0.021 |
ime | -0.021 |
Maria | -0.021 |
illy | -0.020 |
Token | eating |
Token ID | 6600 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
least | +0.013 |
Little | +0.012 |
ime | +0.011 |
illy | +0.010 |
Maria | +0.010 |
Neg logit contributions | |
ows | -0.017 |
badges | -0.016 |
orts | -0.016 |
diapers | -0.015 |
cuts | -0.015 |
Token | candy |
Token ID | 18550 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
ows | +0.097 |
trained | +0.093 |
ages | +0.091 |
orts | +0.091 |
cuts | +0.091 |
Neg logit contributions | |
least | -0.081 |
Little | -0.080 |
ime | -0.073 |
se | -0.073 |
illy | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
badges | +0.122 |
orts | +0.116 |
gag | +0.115 |
names | +0.114 |
ows | +0.113 |
Neg logit contributions | |
least | -0.066 |
Little | -0.054 |
illy | -0.051 |
ime | -0.050 |
st | -0.049 |
Token | |
Token ID | 220 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
least | +0.017 |
Little | +0.015 |
ime | +0.014 |
illy | +0.014 |
ately | +0.013 |
Neg logit contributions | |
badges | -0.029 |
ows | -0.029 |
tricks | -0.028 |
names | -0.028 |
diapers | -0.027 |
Token | for |
Token ID | 329 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
badges | +0.058 |
ows | +0.056 |
tricks | +0.054 |
diapers | +0.053 |
names | +0.053 |
Neg logit contributions | |
least | -0.053 |
Little | -0.047 |
ime | -0.044 |
illy | -0.043 |
ately | -0.042 |
Token | it |
Token ID | 340 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
badges | +0.021 |
tricks | +0.020 |
ows | +0.020 |
names | +0.020 |
diapers | +0.019 |
Neg logit contributions | |
least | -0.020 |
Little | -0.019 |
ime | -0.018 |
illy | -0.017 |
Maria | -0.017 |
Token | ." |
Token ID | 526 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
least | +0.011 |
Little | +0.010 |
ime | +0.010 |
illy | +0.010 |
Maria | +0.009 |
Neg logit contributions | |
ows | -0.009 |
badges | -0.009 |
tricks | -0.009 |
names | -0.009 |
diapers | -0.009 |
Token | Mia |
Token ID | 32189 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
badges | +0.122 |
ows | +0.119 |
tricks | +0.114 |
diapers | +0.112 |
names | +0.112 |
Neg logit contributions | |
least | -0.121 |
Little | -0.110 |
ime | -0.106 |
illy | -0.101 |
Maria | -0.097 |
Token | says |
Token ID | 1139 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
badges | +0.027 |
bones | +0.026 |
diapers | +0.025 |
ages | +0.025 |
tricks | +0.024 |
Neg logit contributions | |
least | -0.046 |
Little | -0.041 |
ill | -0.040 |
ime | -0.039 |
illy | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
badges | +0.079 |
diapers | +0.073 |
tricks | +0.073 |
names | +0.072 |
cuts | +0.072 |
Neg logit contributions | |
least | -0.048 |
ime | -0.041 |
Little | -0.041 |
ately | -0.040 |
illy | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
ows | +0.095 |
badges | +0.094 |
names | +0.092 |
tricks | +0.089 |
diapers | +0.089 |
Neg logit contributions | |
least | -0.070 |
Little | -0.060 |
ime | -0.058 |
illy | -0.055 |
ately | -0.051 |
Token |
|
Token ID | 198 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
ows | +0.032 |
orts | +0.031 |
badges | +0.031 |
names | +0.030 |
tricks | +0.030 |
Neg logit contributions | |
least | -0.028 |
ime | -0.022 |
illy | -0.022 |
Little | -0.022 |
ately | -0.022 |
Token | They |
Token ID | 2990 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
badges | +0.026 |
ows | +0.025 |
tricks | +0.024 |
names | +0.023 |
diapers | +0.023 |
Neg logit contributions | |
least | -0.033 |
ime | -0.029 |
illy | -0.028 |
Little | -0.028 |
ately | -0.028 |
Token | look |
Token ID | 804 |
Feature activation | +2.92e+00 |
Pos logit contributions | |
badges | +0.121 |
diapers | +0.118 |
ows | +0.117 |
orts | +0.116 |
tricks | +0.115 |
Neg logit contributions | |
least | -0.118 |
Little | -0.097 |
ime | -0.092 |
ill | -0.088 |
ately | -0.087 |
Token | around |
Token ID | 1088 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
bows | +0.131 |
preferences | +0.128 |
ages | +0.125 |
badges | +0.125 |
aches | +0.125 |
Neg logit contributions | |
least | -0.066 |
Little | -0.052 |
Little | -0.051 |
again | -0.049 |
Jackie | -0.049 |
Token | |
Token ID | 220 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
least | +0.032 |
ime | +0.027 |
Little | +0.027 |
illy | +0.026 |
ately | +0.024 |
Neg logit contributions | |
badges | -0.032 |
ows | -0.032 |
names | -0.030 |
diapers | -0.030 |
labels | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
least | +0.013 |
Little | +0.011 |
ime | +0.010 |
illy | +0.010 |
ately | +0.009 |
Neg logit contributions | |
badges | -0.026 |
ows | -0.026 |
tricks | -0.025 |
names | -0.025 |
diapers | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
least | +0.081 |
ime | +0.062 |
bul | +0.061 |
ately | +0.061 |
illy | +0.059 |
Neg logit contributions | |
ows | -0.074 |
badges | -0.072 |
orts | -0.071 |
names | -0.070 |
tricks | -0.067 |
Token | John |
Token ID | 7554 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
least | +0.064 |
ime | +0.049 |
ately | +0.047 |
illy | +0.047 |
Little | +0.045 |
Neg logit contributions | |
badges | -0.089 |
ows | -0.085 |
tricks | -0.083 |
helmets | -0.082 |
diapers | -0.082 |
Token | walked |
Token ID | 6807 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
least | +0.013 |
Little | +0.012 |
ime | +0.011 |
illy | +0.011 |
Maria | +0.010 |
Neg logit contributions | |
badges | -0.016 |
ows | -0.015 |
tricks | -0.015 |
names | -0.015 |
diapers | -0.014 |
Token | up |
Token ID | 510 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
least | +0.035 |
Little | +0.032 |
ime | +0.029 |
illy | +0.029 |
Maria | +0.028 |
Neg logit contributions | |
badges | -0.046 |
ows | -0.045 |
tricks | -0.043 |
diapers | -0.043 |
names | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | -9.52e-02 |
Pos logit contributions | |
badges | +0.102 |
labels | +0.096 |
tricks | +0.096 |
ows | +0.096 |
chips | +0.096 |
Neg logit contributions | |
least | -0.104 |
Little | -0.093 |
ime | -0.091 |
illy | -0.088 |
ately | -0.079 |
Token | Jane |
Token ID | 12091 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
least | +0.003 |
Little | +0.003 |
ime | +0.002 |
illy | +0.002 |
ately | +0.002 |
Neg logit contributions | |
badges | -0.004 |
ows | -0.004 |
tricks | -0.003 |
names | -0.003 |
diapers | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
least | +0.053 |
Little | +0.047 |
ime | +0.043 |
illy | +0.041 |
Maria | +0.038 |
Neg logit contributions | |
badges | -0.099 |
ows | -0.098 |
tricks | -0.094 |
names | -0.094 |
diapers | -0.093 |
Token | said |
Token ID | 531 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
least | +0.004 |
Little | +0.004 |
ime | +0.004 |
illy | +0.003 |
ately | +0.003 |
Neg logit contributions | |
badges | -0.004 |
ows | -0.004 |
tricks | -0.003 |
names | -0.003 |
diapers | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
least | +0.017 |
Little | +0.015 |
ime | +0.014 |
illy | +0.013 |
ful | +0.011 |
Neg logit contributions | |
badges | -0.070 |
ows | -0.067 |
cks | -0.066 |
diapers | -0.065 |
labels | -0.064 |
Token | scan |
Token ID | 9367 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
badges | +0.031 |
ows | +0.031 |
tricks | +0.029 |
diapers | +0.029 |
facts | +0.029 |
Neg logit contributions | |
least | -0.025 |
Little | -0.021 |
ime | -0.019 |
illy | -0.017 |
ately | -0.017 |
Token | our |
Token ID | 674 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
badges | +0.013 |
ows | +0.013 |
tricks | +0.013 |
diapers | +0.012 |
names | +0.012 |
Neg logit contributions | |
least | -0.011 |
Little | -0.010 |
ime | -0.009 |
illy | -0.009 |
ately | -0.009 |
Token | tickets |
Token ID | 8587 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
ows | +0.076 |
badges | +0.074 |
orts | +0.069 |
tricks | +0.067 |
diapers | +0.066 |
Neg logit contributions | |
least | -0.119 |
Little | -0.107 |
ime | -0.106 |
ately | -0.100 |
illy | -0.097 |
Token | and |
Token ID | 290 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
badges | +0.063 |
ows | +0.061 |
tricks | +0.060 |
diapers | +0.059 |
names | +0.059 |
Neg logit contributions | |
least | -0.036 |
Little | -0.031 |
ime | -0.029 |
illy | -0.027 |
ately | -0.026 |
Token | our |
Token ID | 674 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
ows | +0.044 |
orts | +0.041 |
badges | +0.041 |
tricks | +0.039 |
cuts | +0.039 |
Neg logit contributions | |
least | -0.045 |
Little | -0.038 |
ime | -0.035 |
ately | -0.034 |
Little | -0.033 |
Token | bodies |
Token ID | 5920 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
ows | +0.100 |
badges | +0.091 |
orts | +0.090 |
diapers | +0.085 |
bows | +0.084 |
Neg logit contributions | |
least | -0.153 |
Little | -0.133 |
ime | -0.132 |
ately | -0.126 |
3 | -0.125 |
Token | . |
Token ID | 13 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
badges | +0.078 |
tricks | +0.072 |
ows | +0.072 |
diapers | +0.072 |
names | +0.071 |
Neg logit contributions | |
least | -0.058 |
Little | -0.049 |
ime | -0.048 |
ately | -0.044 |
illy | -0.044 |
Token | They |
Token ID | 1119 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
badges | +0.051 |
ows | +0.050 |
tricks | +0.048 |
diapers | +0.046 |
names | +0.046 |
Neg logit contributions | |
least | -0.062 |
Little | -0.057 |
ime | -0.054 |
illy | -0.052 |
ately | -0.050 |
Token | will |
Token ID | 481 |
Feature activation | +8.27e-01 |
Pos logit contributions | |
badges | +0.101 |
diapers | +0.098 |
orts | +0.097 |
ows | +0.097 |
tricks | +0.095 |
Neg logit contributions | |
least | -0.102 |
Little | -0.081 |
ime | -0.077 |
ately | -0.072 |
3 | -0.068 |
Token | make |
Token ID | 787 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
ows | +0.033 |
badges | +0.033 |
diapers | +0.031 |
orts | +0.031 |
facts | +0.031 |
Neg logit contributions | |
least | -0.026 |
Little | -0.021 |
ime | -0.019 |
mid | -0.017 |
illy | -0.017 |
Token | sure |
Token ID | 1654 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
badges | +0.008 |
ows | +0.008 |
bows | +0.007 |
orts | +0.007 |
diapers | +0.007 |
Neg logit contributions | |
least | -0.006 |
Little | -0.006 |
ime | -0.005 |
Little | -0.005 |
illy | -0.005 |
Token | would |
Token ID | 561 |
Feature activation | -5.75e-01 |
Pos logit contributions | |
least | +0.056 |
Little | +0.051 |
ime | +0.049 |
illy | +0.048 |
Maria | +0.045 |
Neg logit contributions | |
badges | -0.062 |
ows | -0.061 |
tricks | -0.060 |
names | -0.059 |
diapers | -0.057 |
Token | struggle |
Token ID | 6531 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
least | +0.020 |
Little | +0.018 |
ime | +0.018 |
illy | +0.017 |
Maria | +0.016 |
Neg logit contributions | |
badges | -0.020 |
ows | -0.019 |
diapers | -0.018 |
tricks | -0.018 |
names | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
badges | +0.015 |
ows | +0.014 |
tricks | +0.014 |
names | +0.014 |
diapers | +0.014 |
Neg logit contributions | |
least | -0.010 |
Little | -0.009 |
ime | -0.008 |
illy | -0.008 |
ately | -0.007 |
Token | practice |
Token ID | 3357 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
least | +0.019 |
Little | +0.017 |
ime | +0.016 |
illy | +0.015 |
Maria | +0.015 |
Neg logit contributions | |
badges | -0.017 |
ows | -0.017 |
diapers | -0.016 |
names | -0.016 |
tricks | -0.016 |
Token | so |
Token ID | 523 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
badges | +0.010 |
ows | +0.009 |
diapers | +0.009 |
names | +0.009 |
tricks | +0.009 |
Neg logit contributions | |
least | -0.007 |
Little | -0.006 |
ime | -0.005 |
illy | -0.005 |
ately | -0.005 |
Token | he |
Token ID | 339 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
badges | +0.031 |
tricks | +0.030 |
ows | +0.030 |
diapers | +0.028 |
helmets | +0.028 |
Neg logit contributions | |
least | -0.033 |
Little | -0.030 |
ime | -0.029 |
illy | -0.028 |
Maria | -0.027 |
Token | could |
Token ID | 714 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
least | +0.057 |
Little | +0.051 |
ime | +0.048 |
illy | +0.046 |
Maria | +0.043 |
Neg logit contributions | |
badges | -0.089 |
ows | -0.088 |
tricks | -0.085 |
names | -0.084 |
diapers | -0.083 |
Token | be |
Token ID | 307 |
Feature activation | -9.88e-01 |
Pos logit contributions | |
least | +0.074 |
Little | +0.065 |
ime | +0.062 |
illy | +0.058 |
ill | +0.057 |
Neg logit contributions | |
badges | -0.077 |
ows | -0.073 |
diapers | -0.071 |
labels | -0.069 |
names | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
least | +0.032 |
Little | +0.029 |
ime | +0.028 |
illy | +0.026 |
Maria | +0.025 |
Neg logit contributions | |
badges | -0.037 |
ows | -0.036 |
diapers | -0.034 |
names | -0.034 |
tricks | -0.034 |
Token | best |
Token ID | 1266 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
least | +0.034 |
Little | +0.032 |
ime | +0.031 |
illy | +0.030 |
ately | +0.029 |
Neg logit contributions | |
badges | -0.034 |
ows | -0.033 |
tricks | -0.033 |
names | -0.032 |
diapers | -0.031 |
Token | musician |
Token ID | 21623 |
Feature activation | -9.01e-01 |
Pos logit contributions | |
least | +0.057 |
Little | +0.052 |
ime | +0.049 |
illy | +0.047 |
ately | +0.045 |
Neg logit contributions | |
badges | -0.072 |
ows | -0.070 |
tricks | -0.068 |
names | -0.067 |
diapers | -0.067 |
Token | said |
Token ID | 531 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
badges | +0.061 |
ows | +0.060 |
tricks | +0.058 |
diapers | +0.057 |
names | +0.057 |
Neg logit contributions | |
least | -0.059 |
Little | -0.053 |
ime | -0.050 |
illy | -0.049 |
Maria | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
badges | +0.033 |
ows | +0.031 |
names | +0.030 |
tricks | +0.030 |
diapers | +0.030 |
Neg logit contributions | |
least | -0.022 |
Little | -0.020 |
ime | -0.019 |
illy | -0.018 |
ately | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
badges | +0.106 |
ows | +0.104 |
tricks | +0.101 |
names | +0.099 |
diapers | +0.099 |
Neg logit contributions | |
least | -0.067 |
Little | -0.058 |
ime | -0.056 |
illy | -0.053 |
ately | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
ows | +0.023 |
badges | +0.022 |
tricks | +0.022 |
names | +0.021 |
orts | +0.021 |
Neg logit contributions | |
least | -0.015 |
Little | -0.012 |
ime | -0.012 |
ately | -0.011 |
illy | -0.011 |
Token | " |
Token ID | 1 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
badges | +0.017 |
ows | +0.017 |
tricks | +0.016 |
orts | +0.016 |
names | +0.016 |
Neg logit contributions | |
least | -0.017 |
ime | -0.015 |
illy | -0.014 |
ately | -0.014 |
Little | -0.014 |
Token | I |
Token ID | 40 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
badges | +0.051 |
tricks | +0.048 |
ows | +0.048 |
names | +0.048 |
diapers | +0.047 |
Neg logit contributions | |
least | -0.045 |
ime | -0.040 |
Little | -0.039 |
illy | -0.037 |
ately | -0.036 |
Token | love |
Token ID | 1842 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
least | +0.030 |
Little | +0.027 |
ime | +0.026 |
illy | +0.025 |
Maria | +0.023 |
Neg logit contributions | |
badges | -0.031 |
ows | -0.030 |
tricks | -0.029 |
diapers | -0.028 |
names | -0.028 |
Token | you |
Token ID | 345 |
Feature activation | -9.89e-01 |
Pos logit contributions | |
least | +0.003 |
Little | +0.002 |
illy | +0.002 |
ime | +0.002 |
Maria | +0.002 |
Neg logit contributions | |
ows | -0.005 |
badges | -0.005 |
tricks | -0.005 |
names | -0.005 |
diapers | -0.005 |
Token | too |
Token ID | 1165 |
Feature activation | +4.53e+00 |
Pos logit contributions | |
least | +0.032 |
Little | +0.029 |
ime | +0.028 |
illy | +0.027 |
ately | +0.026 |
Neg logit contributions | |
badges | -0.037 |
ows | -0.036 |
tricks | -0.035 |
names | -0.034 |
diapers | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
badges | +0.186 |
ows | +0.186 |
tricks | +0.177 |
names | +0.176 |
diapers | +0.176 |
Neg logit contributions | |
least | -0.128 |
Little | -0.115 |
ime | -0.108 |
illy | -0.104 |
Maria | -0.098 |
Token | Ben |
Token ID | 3932 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
least | +0.012 |
Little | +0.012 |
ime | +0.011 |
illy | +0.011 |
ately | +0.010 |
Neg logit contributions | |
badges | -0.009 |
ows | -0.009 |
tricks | -0.009 |
names | -0.009 |
diapers | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -7.23e-01 |
Pos logit contributions | |
badges | +0.034 |
ows | +0.033 |
tricks | +0.032 |
diapers | +0.032 |
names | +0.031 |
Neg logit contributions | |
least | -0.020 |
Little | -0.017 |
ime | -0.016 |
illy | -0.016 |
ately | -0.015 |
Token | explore |
Token ID | 7301 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
least | +0.026 |
Little | +0.023 |
ime | +0.023 |
illy | +0.022 |
Maria | +0.020 |
Neg logit contributions | |
badges | -0.025 |
ows | -0.025 |
orts | -0.023 |
diapers | -0.023 |
cuts | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -8.57e-01 |
Pos logit contributions | |
least | +0.011 |
Little | +0.009 |
ime | +0.009 |
ately | +0.009 |
illy | +0.009 |
Neg logit contributions | |
tricks | -0.011 |
badges | -0.011 |
ows | -0.011 |
names | -0.010 |
helmets | -0.010 |
Token | One |
Token ID | 1881 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
least | +0.040 |
Little | +0.037 |
ime | +0.037 |
ately | +0.036 |
bul | +0.035 |
Neg logit contributions | |
names | -0.019 |
ows | -0.017 |
name | -0.017 |
diapers | -0.015 |
badges | -0.015 |
Token | day |
Token ID | 1110 |
Feature activation | +8.81e-01 |
Pos logit contributions | |
names | +0.001 |
name | +0.001 |
tricks | +0.001 |
ows | +0.001 |
badges | +0.001 |
Neg logit contributions | |
least | -0.001 |
ately | -0.001 |
Little | -0.001 |
ime | -0.001 |
Em | -0.001 |
Token | she |
Token ID | 673 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
badges | +0.036 |
ows | +0.034 |
tricks | +0.034 |
names | +0.034 |
diapers | +0.034 |
Neg logit contributions | |
least | -0.026 |
Little | -0.021 |
ime | -0.021 |
ately | -0.019 |
illy | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
least | +0.044 |
Little | +0.040 |
ime | +0.038 |
illy | +0.037 |
ately | +0.034 |
Neg logit contributions | |
badges | -0.048 |
names | -0.046 |
hides | -0.045 |
ows | -0.044 |
tricks | -0.044 |
Token | playing |
Token ID | 2712 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
least | +0.035 |
Little | +0.032 |
ime | +0.030 |
illy | +0.029 |
ately | +0.028 |
Neg logit contributions | |
badges | -0.041 |
ows | -0.041 |
tricks | -0.039 |
names | -0.039 |
diapers | -0.038 |
Token | in |
Token ID | 287 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
badges | +0.091 |
ows | +0.089 |
tricks | +0.087 |
names | +0.085 |
diapers | +0.085 |
Neg logit contributions | |
least | -0.073 |
Little | -0.066 |
ime | -0.063 |
illy | -0.060 |
ately | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
least | +0.031 |
ately | +0.029 |
illy | +0.029 |
Little | +0.028 |
ime | +0.028 |
Neg logit contributions | |
badges | -0.026 |
labels | -0.025 |
diapers | -0.025 |
tricks | -0.025 |
helmets | -0.025 |
Token | backyard |
Token ID | 24296 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ately | +0.047 |
least | +0.046 |
ime | +0.043 |
Little | +0.041 |
illy | +0.040 |
Neg logit contributions | |
badges | -0.051 |
names | -0.050 |
tricks | -0.050 |
diapers | -0.049 |
helmets | -0.049 |
Token | listen |
Token ID | 6004 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
least | +0.001 |
Little | +0.001 |
ime | +0.001 |
illy | +0.001 |
ill | +0.001 |
Neg logit contributions | |
badges | -0.001 |
ows | -0.001 |
diapers | -0.001 |
cks | -0.001 |
helmets | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -7.61e-01 |
Pos logit contributions | |
badges | +0.019 |
ows | +0.018 |
orts | +0.017 |
diapers | +0.017 |
names | +0.017 |
Neg logit contributions | |
least | -0.011 |
Little | -0.010 |
ime | -0.010 |
illy | -0.009 |
ately | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
least | +0.023 |
ime | +0.020 |
Little | +0.020 |
illy | +0.019 |
ately | +0.018 |
Neg logit contributions | |
badges | -0.030 |
ows | -0.030 |
names | -0.028 |
diapers | -0.028 |
tricks | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
least | +0.070 |
ately | +0.053 |
ime | +0.052 |
illy | +0.051 |
bul | +0.050 |
Neg logit contributions | |
ows | -0.071 |
orts | -0.066 |
names | -0.065 |
tricks | -0.065 |
badges | -0.065 |
Token | Tim |
Token ID | 14967 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
least | +0.053 |
Little | +0.047 |
ime | +0.044 |
illy | +0.042 |
ately | +0.039 |
Neg logit contributions | |
badges | -0.084 |
ows | -0.082 |
tricks | -0.080 |
names | -0.079 |
diapers | -0.078 |
Token | my |
Token ID | 1820 |
Feature activation | -8.57e-01 |
Pos logit contributions | |
least | +0.065 |
Little | +0.059 |
ime | +0.058 |
ately | +0.055 |
Jess | +0.054 |
Neg logit contributions | |
badges | -0.085 |
ows | -0.083 |
names | -0.079 |
hides | -0.078 |
homework | -0.078 |
Token | went |
Token ID | 1816 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
least | +0.029 |
ime | +0.026 |
Little | +0.026 |
illy | +0.024 |
Maria | +0.023 |
Neg logit contributions | |
badges | -0.030 |
ows | -0.029 |
tricks | -0.028 |
names | -0.027 |
hides | -0.027 |
Token | outside |
Token ID | 2354 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
least | +0.042 |
Little | +0.037 |
ime | +0.035 |
illy | +0.033 |
ately | +0.031 |
Neg logit contributions | |
badges | -0.070 |
ows | -0.069 |
tricks | -0.067 |
names | -0.066 |
diapers | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | -8.52e-01 |
Pos logit contributions | |
tricks | +0.014 |
homework | +0.013 |
hides | +0.013 |
pets | +0.012 |
ashes | +0.012 |
Neg logit contributions | |
Little | -0.027 |
illy | -0.027 |
least | -0.026 |
ime | -0.026 |
enna | -0.024 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
least | +0.022 |
Little | +0.020 |
ime | +0.019 |
illy | +0.018 |
ately | +0.017 |
Neg logit contributions | |
badges | -0.037 |
ows | -0.036 |
tricks | -0.035 |
names | -0.035 |
diapers | -0.035 |
Token | his |
Token ID | 465 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
least | +0.052 |
Little | +0.048 |
ime | +0.046 |
illy | +0.044 |
ately | +0.043 |
Neg logit contributions | |
badges | -0.057 |
ows | -0.055 |
tricks | -0.054 |
names | -0.053 |
diapers | -0.053 |
Token | creative |
Token ID | 7325 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
least | +0.037 |
Little | +0.034 |
ime | +0.033 |
illy | +0.032 |
ately | +0.031 |
Neg logit contributions | |
badges | -0.035 |
ows | -0.034 |
tricks | -0.033 |
names | -0.033 |
diapers | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
least | +0.008 |
Little | +0.007 |
ime | +0.006 |
illy | +0.006 |
ately | +0.006 |
Neg logit contributions | |
badges | -0.016 |
ows | -0.015 |
diapers | -0.015 |
names | -0.015 |
tricks | -0.015 |
Token | draw |
Token ID | 3197 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
least | +0.052 |
Little | +0.046 |
ime | +0.044 |
illy | +0.042 |
Maria | +0.041 |
Neg logit contributions | |
badges | -0.051 |
ows | -0.050 |
diapers | -0.047 |
names | -0.046 |
orts | -0.046 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -7.92e-01 |
Pos logit contributions | |
least | +0.006 |
Little | +0.005 |
ime | +0.005 |
illy | +0.005 |
Maria | +0.005 |
Neg logit contributions | |
badges | -0.006 |
ows | -0.006 |
diapers | -0.005 |
tricks | -0.005 |
names | -0.005 |
Token | pictures |
Token ID | 5986 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
least | +0.030 |
Little | +0.028 |
ime | +0.028 |
illy | +0.026 |
Little | +0.025 |
Neg logit contributions | |
ows | -0.024 |
badges | -0.024 |
orts | -0.023 |
cks | -0.022 |
cuts | -0.021 |
Token | that |
Token ID | 326 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
badges | +0.010 |
ows | +0.009 |
tricks | +0.009 |
names | +0.009 |
diapers | +0.009 |
Neg logit contributions | |
least | -0.007 |
Little | -0.006 |
ime | -0.006 |
illy | -0.006 |
ately | -0.005 |
Token | he |
Token ID | 339 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
least | +0.053 |
Little | +0.050 |
illy | +0.047 |
ime | +0.047 |
ately | +0.045 |
Neg logit contributions | |
badges | -0.059 |
tricks | -0.058 |
names | -0.057 |
ows | -0.057 |
labels | -0.056 |
Token | shared |
Token ID | 4888 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
least | +0.056 |
Little | +0.055 |
ime | +0.052 |
illy | +0.050 |
Maria | +0.048 |
Neg logit contributions | |
tricks | -0.067 |
badges | -0.065 |
ows | -0.065 |
names | -0.062 |
labels | -0.062 |
Token | with |
Token ID | 351 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ows | +0.051 |
badges | +0.051 |
orts | +0.048 |
cuts | +0.048 |
gag | +0.048 |
Neg logit contributions | |
least | -0.031 |
Little | -0.027 |
ime | -0.027 |
illy | -0.025 |
ill | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
badges | +0.042 |
tricks | +0.041 |
ows | +0.039 |
names | +0.039 |
diapers | +0.039 |
Neg logit contributions | |
least | -0.044 |
Little | -0.041 |
ime | -0.039 |
illy | -0.037 |
ately | -0.036 |
Token | world |
Token ID | 995 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
least | +0.012 |
Little | +0.011 |
ime | +0.011 |
illy | +0.010 |
ately | +0.009 |
Neg logit contributions | |
badges | -0.014 |
ows | -0.014 |
diapers | -0.013 |
tricks | -0.013 |
orts | -0.013 |
Token | The |
Token ID | 464 |
Feature activation | +6.56e-02 |
Pos logit contributions | |
least | +0.057 |
ime | +0.046 |
ately | +0.045 |
illy | +0.044 |
Little | +0.041 |
Neg logit contributions | |
badges | -0.093 |
tricks | -0.091 |
ows | -0.090 |
names | -0.088 |
helmets | -0.087 |
Token | sun |
Token ID | 4252 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
badges | +0.003 |
ows | +0.003 |
tricks | +0.003 |
diapers | +0.002 |
names | +0.002 |
Neg logit contributions | |
least | -0.002 |
Little | -0.002 |
ime | -0.002 |
illy | -0.002 |
ately | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
least | +0.023 |
Little | +0.019 |
ime | +0.019 |
illy | +0.019 |
enny | +0.018 |
Neg logit contributions | |
ows | -0.031 |
badges | -0.031 |
names | -0.029 |
tricks | -0.029 |
diapers | -0.029 |
Token | setting |
Token ID | 4634 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
least | +0.080 |
enny | +0.074 |
ately | +0.072 |
Little | +0.071 |
ime | +0.071 |
Neg logit contributions | |
badges | -0.080 |
names | -0.077 |
ows | -0.077 |
tricks | -0.076 |
helmets | -0.076 |
Token | again |
Token ID | 757 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
least | +0.033 |
ime | +0.028 |
Little | +0.028 |
illy | +0.027 |
enny | +0.027 |
Neg logit contributions | |
ows | -0.040 |
badges | -0.038 |
helmets | -0.036 |
tricks | -0.036 |
names | -0.036 |
Token | but |
Token ID | 475 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
least | +0.099 |
enny | +0.091 |
Maria | +0.090 |
Little | +0.090 |
ime | +0.090 |
Neg logit contributions | |
ows | -0.097 |
tricks | -0.091 |
bows | -0.090 |
badges | -0.090 |
names | -0.089 |
Token | it |
Token ID | 340 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
least | +0.114 |
ime | +0.114 |
Little | +0.109 |
illy | +0.108 |
enny | +0.105 |
Neg logit contributions | |
badges | -0.115 |
ows | -0.113 |
tricks | -0.113 |
names | -0.112 |
diapers | -0.110 |
Token | was |
Token ID | 373 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
ows | +0.008 |
tricks | +0.007 |
badges | +0.007 |
names | +0.007 |
diapers | +0.007 |
Neg logit contributions | |
least | -0.006 |
illy | -0.005 |
ime | -0.005 |
Little | -0.005 |
enny | -0.005 |
Token | okay |
Token ID | 8788 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
least | +0.072 |
Little | +0.066 |
ime | +0.065 |
ately | +0.065 |
illy | +0.064 |
Neg logit contributions | |
badges | -0.072 |
ows | -0.071 |
names | -0.070 |
tricks | -0.069 |
diapers | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
least | +0.051 |
Little | +0.048 |
ime | +0.045 |
illy | +0.043 |
Em | +0.042 |
Neg logit contributions | |
ows | -0.047 |
badges | -0.045 |
tricks | -0.045 |
names | -0.045 |
diapers | -0.043 |
Token | The |
Token ID | 383 |
Feature activation | -7.33e-02 |
Pos logit contributions | |
least | +0.052 |
Little | +0.048 |
ime | +0.047 |
illy | +0.046 |
ately | +0.043 |
Neg logit contributions | |
ows | -0.033 |
badges | -0.033 |
names | -0.033 |
diapers | -0.030 |
name | -0.030 |
Token | more |
Token ID | 517 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
least | +0.032 |
Little | +0.028 |
ime | +0.027 |
illy | +0.025 |
Maria | +0.024 |
Neg logit contributions | |
badges | -0.041 |
ows | -0.040 |
diapers | -0.038 |
orts | -0.038 |
tricks | -0.038 |
Token | than |
Token ID | 621 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
badges | +0.034 |
ows | +0.033 |
tricks | +0.032 |
names | +0.031 |
diapers | +0.031 |
Neg logit contributions | |
least | -0.045 |
Little | -0.041 |
ime | -0.040 |
illy | -0.039 |
ately | -0.037 |
Token | anything |
Token ID | 1997 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
least | +0.018 |
Little | +0.018 |
ime | +0.016 |
illy | +0.016 |
Em | +0.016 |
Neg logit contributions | |
badges | -0.019 |
tricks | -0.018 |
ows | -0.018 |
names | -0.018 |
diapers | -0.018 |
Token | else |
Token ID | 2073 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
ows | +0.098 |
badges | +0.097 |
tricks | +0.094 |
names | +0.092 |
diapers | +0.091 |
Neg logit contributions | |
least | -0.082 |
Little | -0.074 |
ime | -0.070 |
illy | -0.068 |
Maria | -0.065 |
Token | she |
Token ID | 673 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
badges | +0.016 |
ows | +0.016 |
tricks | +0.015 |
names | +0.015 |
diapers | +0.015 |
Neg logit contributions | |
least | -0.013 |
Little | -0.012 |
ime | -0.011 |
illy | -0.011 |
Maria | -0.010 |
Token | had |
Token ID | 550 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
least | +0.117 |
Little | +0.108 |
ime | +0.102 |
illy | +0.100 |
Maria | +0.094 |
Neg logit contributions | |
badges | -0.140 |
ows | -0.137 |
tricks | -0.133 |
names | -0.132 |
diapers | -0.129 |
Token | ever |
Token ID | 1683 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
least | +0.114 |
Little | +0.111 |
ime | +0.100 |
illy | +0.097 |
Maria | +0.093 |
Neg logit contributions | |
ows | -0.150 |
badges | -0.149 |
tricks | -0.148 |
names | -0.146 |
labels | -0.141 |
Token | carried |
Token ID | 5281 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
least | +0.085 |
Little | +0.083 |
illy | +0.075 |
ime | +0.073 |
Em | +0.072 |
Neg logit contributions | |
badges | -0.096 |
ows | -0.094 |
tricks | -0.092 |
names | -0.092 |
facts | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
least | +0.097 |
Little | +0.087 |
ime | +0.083 |
illy | +0.080 |
ately | +0.076 |
Neg logit contributions | |
badges | -0.128 |
ows | -0.125 |
tricks | -0.120 |
names | -0.119 |
diapers | -0.119 |
Token | |
Token ID | 220 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
least | +0.061 |
ime | +0.056 |
Little | +0.056 |
illy | +0.053 |
ately | +0.050 |
Neg logit contributions | |
ows | -0.055 |
names | -0.054 |
badges | -0.053 |
orts | -0.051 |
name | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
least | +0.016 |
Little | +0.013 |
ime | +0.012 |
illy | +0.012 |
ately | +0.011 |
Neg logit contributions | |
ows | -0.027 |
badges | -0.026 |
names | -0.025 |
tricks | -0.025 |
orts | -0.025 |
Token | have |
Token ID | 423 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
least | +0.002 |
Little | +0.002 |
ime | +0.002 |
illy | +0.002 |
ately | +0.002 |
Neg logit contributions | |
badges | -0.001 |
names | -0.001 |
tricks | -0.001 |
name | -0.001 |
ows | -0.001 |
Token | such |
Token ID | 884 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
least | +0.065 |
Little | +0.059 |
ime | +0.057 |
illy | +0.054 |
ately | +0.053 |
Neg logit contributions | |
badges | -0.073 |
tricks | -0.068 |
ows | -0.068 |
names | -0.066 |
labels | -0.065 |
Token | a |
Token ID | 257 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
ows | +0.109 |
orts | +0.104 |
badges | +0.102 |
cks | +0.100 |
cuts | +0.100 |
Neg logit contributions | |
least | -0.060 |
Little | -0.051 |
illy | -0.044 |
3 | -0.044 |
Little | -0.043 |
Token | cool |
Token ID | 3608 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
least | +0.029 |
ime | +0.027 |
Little | +0.027 |
illy | +0.026 |
ately | +0.026 |
Neg logit contributions | |
badges | -0.022 |
tricks | -0.021 |
ows | -0.020 |
names | -0.020 |
diapers | -0.019 |
Token | souven |
Token ID | 45043 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
least | +0.063 |
Little | +0.056 |
ime | +0.053 |
illy | +0.053 |
3 | +0.047 |
Neg logit contributions | |
badges | -0.093 |
ows | -0.092 |
cks | -0.089 |
tricks | -0.088 |
names | -0.086 |
Token | ir |
Token ID | 343 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
ows | +0.023 |
cks | +0.022 |
orts | +0.022 |
aches | +0.021 |
badges | +0.021 |
Neg logit contributions | |
least | -0.017 |
Little | -0.016 |
3 | -0.015 |
Little | -0.014 |
Giant | -0.014 |
Token | from |
Token ID | 422 |
Feature activation | -3.87e-02 |
Pos logit contributions | |
least | +0.094 |
Little | +0.082 |
ime | +0.078 |
illy | +0.073 |
ately | +0.067 |
Neg logit contributions | |
badges | -0.192 |
ows | -0.187 |
tricks | -0.181 |
diapers | -0.180 |
names | -0.180 |
Token | their |
Token ID | 511 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
least | +0.001 |
Little | +0.001 |
ime | +0.001 |
illy | +0.001 |
ately | +0.001 |
Neg logit contributions | |
badges | -0.002 |
ows | -0.002 |
tricks | -0.001 |
names | -0.001 |
diapers | -0.001 |
Token | exciting |
Token ID | 7895 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
badges | +0.050 |
ows | +0.047 |
tricks | +0.047 |
names | +0.047 |
diapers | +0.046 |
Neg logit contributions | |
least | -0.044 |
Little | -0.041 |
ime | -0.039 |
illy | -0.038 |
Maria | -0.036 |
Token | adventure |
Token ID | 8855 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
least | +0.010 |
Little | +0.009 |
ime | +0.009 |
illy | +0.008 |
ately | +0.007 |
Neg logit contributions | |
badges | -0.018 |
ows | -0.018 |
tricks | -0.017 |
diapers | -0.017 |
names | -0.017 |
Token | ! |
Token ID | 0 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
least | +0.070 |
Little | +0.064 |
ime | +0.060 |
illy | +0.058 |
Maria | +0.055 |
Neg logit contributions | |
badges | -0.085 |
ows | -0.084 |
tricks | -0.081 |
names | -0.079 |
diapers | -0.079 |
Token | He |
Token ID | 679 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
least | +0.014 |
Little | +0.013 |
ime | +0.012 |
illy | +0.012 |
ately | +0.011 |
Neg logit contributions | |
badges | -0.014 |
ows | -0.014 |
tricks | -0.013 |
names | -0.013 |
diapers | -0.013 |
Token | lived |
Token ID | 5615 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
badges | +0.007 |
ows | +0.007 |
tricks | +0.006 |
names | +0.006 |
diapers | +0.006 |
Neg logit contributions | |
least | -0.006 |
Little | -0.006 |
ime | -0.006 |
illy | -0.005 |
ately | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
least | +0.005 |
Little | +0.005 |
ime | +0.004 |
illy | +0.004 |
Maria | +0.004 |
Neg logit contributions | |
badges | -0.006 |
ows | -0.006 |
tricks | -0.006 |
names | -0.005 |
diapers | -0.005 |
Token | happy |
Token ID | 3772 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
tricks | +0.009 |
badges | +0.009 |
name | +0.009 |
names | +0.008 |
diapers | +0.008 |
Neg logit contributions | |
least | -0.010 |
illy | -0.009 |
Little | -0.009 |
ill | -0.009 |
ime | -0.009 |
Token | life |
Token ID | 1204 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
least | +0.026 |
Little | +0.024 |
Maria | +0.024 |
ill | +0.024 |
ately | +0.023 |
Neg logit contributions | |
name | -0.023 |
tricks | -0.021 |
teeth | -0.021 |
badges | -0.021 |
badge | -0.020 |
Token | but |
Token ID | 475 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
least | +0.042 |
Little | +0.037 |
ime | +0.036 |
illy | +0.034 |
ately | +0.032 |
Neg logit contributions | |
badges | -0.056 |
ows | -0.054 |
tricks | -0.052 |
names | -0.052 |
diapers | -0.052 |
Token | something |
Token ID | 1223 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
least | +0.108 |
Little | +0.095 |
ime | +0.090 |
illy | +0.085 |
Maria | +0.080 |
Neg logit contributions | |
badges | -0.148 |
ows | -0.146 |
names | -0.136 |
tricks | -0.136 |
diapers | -0.136 |
Token | was |
Token ID | 373 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
badges | +0.154 |
ows | +0.148 |
orts | +0.143 |
names | +0.141 |
diapers | +0.138 |
Neg logit contributions | |
least | -0.128 |
ime | -0.114 |
Little | -0.114 |
illy | -0.103 |
ately | -0.097 |
Token | missing |
Token ID | 4814 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
least | +0.096 |
ime | +0.078 |
Little | +0.077 |
illy | +0.076 |
nowhere | +0.074 |
Neg logit contributions | |
badges | -0.105 |
ows | -0.103 |
cuts | -0.098 |
names | -0.097 |
orts | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
least | +0.023 |
Little | +0.022 |
Maria | +0.021 |
ime | +0.020 |
illy | +0.020 |
Neg logit contributions | |
ows | -0.018 |
badges | -0.017 |
tricks | -0.017 |
names | -0.016 |
diapers | -0.016 |
Token | He |
Token ID | 679 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
least | +0.026 |
Little | +0.024 |
ime | +0.023 |
illy | +0.022 |
ately | +0.021 |
Neg logit contributions | |
badges | -0.023 |
ows | -0.023 |
tricks | -0.022 |
names | -0.022 |
diapers | -0.021 |
Token | who |
Token ID | 508 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
least | +0.003 |
ime | +0.003 |
Little | +0.003 |
illy | +0.003 |
ately | +0.003 |
Neg logit contributions | |
badges | -0.003 |
ows | -0.003 |
tricks | -0.003 |
diapers | -0.003 |
helmets | -0.003 |
Token | knew |
Token ID | 2993 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
badges | +0.068 |
ows | +0.067 |
tricks | +0.065 |
names | +0.064 |
diapers | +0.063 |
Neg logit contributions | |
least | -0.043 |
Little | -0.037 |
ime | -0.035 |
illy | -0.034 |
ately | -0.032 |
Token | how |
Token ID | 703 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
least | +0.104 |
Little | +0.096 |
ime | +0.094 |
illy | +0.089 |
Maria | +0.087 |
Neg logit contributions | |
ows | -0.092 |
badges | -0.092 |
tricks | -0.090 |
names | -0.086 |
helmets | -0.084 |
Token | disappointed |
Token ID | 11679 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
badges | +0.126 |
tricks | +0.125 |
ows | +0.123 |
names | +0.120 |
diapers | +0.118 |
Neg logit contributions | |
least | -0.104 |
Little | -0.091 |
ime | -0.085 |
illy | -0.085 |
ately | -0.080 |
Token | Ted |
Token ID | 11396 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
badges | +0.039 |
ows | +0.038 |
orts | +0.036 |
names | +0.036 |
diapers | +0.036 |
Neg logit contributions | |
least | -0.025 |
Little | -0.022 |
ime | -0.021 |
illy | -0.020 |
Little | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | -3.75e+00 |
Pos logit contributions | |
least | +0.074 |
Little | +0.065 |
ime | +0.061 |
illy | +0.058 |
ately | +0.054 |
Neg logit contributions | |
badges | -0.127 |
ows | -0.124 |
tricks | -0.120 |
names | -0.119 |
diapers | -0.118 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
least | +0.124 |
Little | +0.116 |
ime | +0.109 |
illy | +0.106 |
Maria | +0.104 |
Neg logit contributions | |
badges | -0.132 |
ows | -0.130 |
tricks | -0.129 |
names | -0.125 |
diapers | -0.121 |
Token | " |
Token ID | 366 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
least | +0.047 |
ime | +0.042 |
Little | +0.041 |
illy | +0.040 |
ately | +0.037 |
Neg logit contributions | |
ows | -0.047 |
badges | -0.046 |
names | -0.046 |
orts | -0.044 |
tricks | -0.044 |
Token | We |
Token ID | 1135 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
least | +0.038 |
ime | +0.033 |
illy | +0.033 |
Little | +0.031 |
ately | +0.031 |
Neg logit contributions | |
badges | -0.044 |
names | -0.044 |
tricks | -0.043 |
ows | -0.043 |
orts | -0.043 |
Token | can |
Token ID | 460 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
least | +0.053 |
Little | +0.047 |
ime | +0.045 |
illy | +0.043 |
ately | +0.041 |
Neg logit contributions | |
badges | -0.060 |
ows | -0.058 |
tricks | -0.056 |
diapers | -0.055 |
names | -0.054 |
Token | try |
Token ID | 1949 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
least | +0.046 |
Little | +0.042 |
ime | +0.041 |
illy | +0.039 |
Maria | +0.038 |
Neg logit contributions | |
badges | -0.045 |
ows | -0.044 |
tricks | -0.042 |
names | -0.042 |
diapers | -0.041 |