Token | was |
Token ID | 373 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
arl | +0.057 |
purpose | +0.056 |
Hawaii | +0.052 |
Ireland | +0.050 |
This | +0.050 |
Neg logit contributions | |
forgot | -0.056 |
felt | -0.048 |
laughed | -0.045 |
got | -0.044 |
knew | -0.044 |
Token | all |
Token ID | 477 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
forgot | +0.046 |
felt | +0.039 |
laughed | +0.037 |
got | +0.036 |
knew | +0.036 |
Neg logit contributions | |
arl | -0.046 |
purpose | -0.045 |
Hawaii | -0.042 |
Ireland | -0.041 |
This | -0.040 |
Token | gone |
Token ID | 3750 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
forgot | +0.060 |
felt | +0.050 |
laughed | +0.048 |
ppa | +0.047 |
knew | +0.046 |
Neg logit contributions | |
purpose | -0.057 |
arl | -0.057 |
Hawaii | -0.052 |
This | -0.051 |
Ireland | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
arl | +0.054 |
purpose | +0.053 |
Hawaii | +0.050 |
Ireland | +0.048 |
This | +0.048 |
Neg logit contributions | |
forgot | -0.050 |
felt | -0.043 |
laughed | -0.040 |
got | -0.039 |
knew | -0.039 |
Token | El |
Token ID | 2574 |
Feature activation | +7.56e+00 |
Pos logit contributions | |
arl | +0.003 |
purpose | +0.003 |
Hawaii | +0.003 |
outer | +0.003 |
Ireland | +0.003 |
Neg logit contributions | |
forgot | -0.005 |
felt | -0.004 |
laughed | -0.004 |
my | -0.004 |
knew | -0.004 |
Token | iza |
Token ID | 23638 |
Feature activation | +1.39e+01 |
Pos logit contributions | |
forgot | +0.137 |
my | +0.103 |
felt | +0.101 |
ppa | +0.098 |
got | +0.097 |
Neg logit contributions | |
arl | -0.170 |
purpose | -0.169 |
Hawaii | -0.168 |
outer | -0.167 |
time | -0.165 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
ppy | +0.155 |
forgot | +0.154 |
ppa | +0.151 |
ba | +0.144 |
anut | +0.132 |
Neg logit contributions | |
time | -0.315 |
This | -0.301 |
purpose | -0.293 |
A | -0.288 |
Can | -0.275 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
forgot | +0.004 |
felt | +0.004 |
laughed | +0.003 |
knew | +0.003 |
got | +0.003 |
Neg logit contributions | |
arl | -0.004 |
purpose | -0.004 |
Hawaii | -0.003 |
Ireland | -0.003 |
This | -0.003 |
Token | my |
Token ID | 1820 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
forgot | +0.083 |
felt | +0.068 |
laughed | +0.065 |
struggled | +0.063 |
knew | +0.062 |
Neg logit contributions | |
arl | -0.089 |
purpose | -0.087 |
Hawaii | -0.082 |
This | -0.081 |
Ireland | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
forgot | +0.104 |
felt | +0.090 |
laughed | +0.086 |
got | +0.084 |
knew | +0.084 |
Neg logit contributions | |
arl | -0.080 |
purpose | -0.079 |
Hawaii | -0.072 |
This | -0.070 |
Ireland | -0.069 |
Token | ' |
Token ID | 705 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
forgot | +0.080 |
felt | +0.071 |
laughed | +0.069 |
got | +0.067 |
knew | +0.067 |
Neg logit contributions | |
arl | -0.036 |
purpose | -0.035 |
Hawaii | -0.031 |
Ireland | -0.029 |
This | -0.029 |
Token | angry |
Token ID | 7954 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
forgot | +0.025 |
felt | +0.020 |
ppa | +0.019 |
pretended | +0.019 |
ppy | +0.019 |
Neg logit contributions | |
purpose | -0.022 |
arl | -0.021 |
This | -0.020 |
Hawaii | -0.020 |
sight | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
arl | +0.037 |
purpose | +0.037 |
Hawaii | +0.034 |
This | +0.033 |
Ireland | +0.033 |
Neg logit contributions | |
forgot | -0.028 |
felt | -0.023 |
laughed | -0.021 |
knew | -0.021 |
got | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
arl | +0.001 |
purpose | +0.001 |
Hawaii | +0.001 |
Ireland | +0.001 |
outer | +0.001 |
Neg logit contributions | |
forgot | -0.001 |
felt | -0.001 |
laughed | -0.001 |
got | -0.001 |
knew | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -9.18e-01 |
Pos logit contributions | |
arl | +0.025 |
purpose | +0.025 |
Hawaii | +0.023 |
Ireland | +0.023 |
outer | +0.022 |
Neg logit contributions | |
forgot | -0.036 |
felt | -0.032 |
knew | -0.030 |
my | -0.030 |
laughed | -0.030 |
Token | The |
Token ID | 464 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
arl | +0.018 |
purpose | +0.018 |
Hawaii | +0.016 |
This | +0.015 |
Ireland | +0.015 |
Neg logit contributions | |
forgot | -0.023 |
felt | -0.020 |
laughed | -0.019 |
got | -0.018 |
knew | -0.018 |
Token | kid |
Token ID | 5141 |
Feature activation | +1.38e+01 |
Pos logit contributions | |
forgot | +0.044 |
felt | +0.037 |
laughed | +0.035 |
got | +0.034 |
ppa | +0.034 |
Neg logit contributions | |
purpose | -0.044 |
arl | -0.044 |
Hawaii | -0.040 |
Ireland | -0.038 |
sight | -0.038 |
Token | still |
Token ID | 991 |
Feature activation | +7.22e+00 |
Pos logit contributions | |
ppa | +0.178 |
forgot | +0.154 |
ppy | +0.151 |
anut | +0.151 |
any | +0.146 |
Neg logit contributions | |
time | -0.314 |
A | -0.283 |
This | -0.280 |
Please | -0.277 |
purpose | -0.273 |
Token | wanted |
Token ID | 2227 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
ppa | +0.073 |
forgot | +0.068 |
ppy | +0.057 |
ie | +0.054 |
mie | +0.054 |
Neg logit contributions | |
purpose | -0.194 |
arl | -0.187 |
This | -0.184 |
sight | -0.180 |
Can | -0.179 |
Token | to |
Token ID | 284 |
Feature activation | -8.50e-01 |
Pos logit contributions | |
arl | +0.050 |
purpose | +0.046 |
Hawaii | +0.043 |
Ireland | +0.041 |
outer | +0.040 |
Neg logit contributions | |
forgot | -0.077 |
felt | -0.068 |
laughed | -0.066 |
got | -0.064 |
knew | -0.064 |
Token | go |
Token ID | 467 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
purpose | +0.019 |
arl | +0.018 |
Hawaii | +0.017 |
This | +0.016 |
time | +0.016 |
Neg logit contributions | |
forgot | -0.019 |
felt | -0.016 |
laughed | -0.015 |
ppa | -0.015 |
knew | -0.015 |
Token | outside |
Token ID | 2354 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
arl | +0.051 |
purpose | +0.050 |
Hawaii | +0.046 |
This | +0.044 |
Ireland | +0.043 |
Neg logit contributions | |
forgot | -0.080 |
felt | -0.069 |
laughed | -0.067 |
knew | -0.065 |
struggled | -0.063 |
Token | it |
Token ID | 340 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
arl | +0.077 |
purpose | +0.069 |
ay | +0.065 |
This | +0.064 |
Hawaii | +0.063 |
Neg logit contributions | |
forgot | -0.061 |
felt | -0.052 |
laughed | -0.052 |
had | -0.048 |
's | -0.048 |
Token | was |
Token ID | 373 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
forgot | +0.029 |
ppa | +0.023 |
felt | +0.021 |
laughed | +0.019 |
ppy | +0.019 |
Neg logit contributions | |
purpose | -0.067 |
arl | -0.062 |
time | -0.062 |
This | -0.062 |
Hawaii | -0.060 |
Token | yours |
Token ID | 12431 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
arl | +0.010 |
purpose | +0.009 |
Hawaii | +0.009 |
This | +0.008 |
ay | +0.008 |
Neg logit contributions | |
forgot | -0.012 |
felt | -0.010 |
laughed | -0.010 |
got | -0.010 |
had | -0.010 |
Token | ." |
Token ID | 526 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
arl | +0.021 |
purpose | +0.020 |
Hawaii | +0.019 |
Ireland | +0.018 |
outer | +0.018 |
Neg logit contributions | |
forgot | -0.024 |
felt | -0.021 |
laughed | -0.020 |
got | -0.019 |
knew | -0.019 |
Token | The |
Token ID | 383 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
arl | +0.013 |
purpose | +0.012 |
Hawaii | +0.011 |
Ireland | +0.011 |
This | +0.010 |
Neg logit contributions | |
forgot | -0.022 |
felt | -0.020 |
laughed | -0.019 |
got | -0.018 |
knew | -0.018 |
Token | man |
Token ID | 582 |
Feature activation | +1.38e+01 |
Pos logit contributions | |
forgot | +0.037 |
felt | +0.031 |
laughed | +0.029 |
got | +0.028 |
knew | +0.028 |
Neg logit contributions | |
arl | -0.043 |
purpose | -0.042 |
Hawaii | -0.039 |
Ireland | -0.038 |
outer | -0.037 |
Token | said |
Token ID | 531 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
ppa | +0.212 |
forgot | +0.207 |
ppy | +0.187 |
pads | +0.185 |
any | +0.172 |
Neg logit contributions | |
A | -0.268 |
hello | -0.266 |
time | -0.260 |
This | -0.252 |
The | -0.241 |
Token | , |
Token ID | 11 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
forgot | +0.005 |
felt | +0.004 |
laughed | +0.004 |
got | +0.004 |
knew | +0.004 |
Neg logit contributions | |
arl | -0.005 |
purpose | -0.004 |
Hawaii | -0.004 |
Ireland | -0.004 |
outer | -0.004 |
Token | " |
Token ID | 366 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
forgot | +0.074 |
felt | +0.066 |
laughed | +0.064 |
got | +0.062 |
knew | +0.062 |
Neg logit contributions | |
arl | -0.030 |
purpose | -0.029 |
Hawaii | -0.026 |
This | -0.025 |
Ireland | -0.024 |
Token | It |
Token ID | 1026 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
arl | +0.016 |
purpose | +0.016 |
time | +0.014 |
This | +0.014 |
Can | +0.014 |
Neg logit contributions | |
forgot | -0.026 |
felt | -0.023 |
laughed | -0.022 |
got | -0.021 |
ppa | -0.021 |
Token | 's |
Token ID | 338 |
Feature activation | -9.41e-02 |
Pos logit contributions | |
ppa | +0.062 |
forgot | +0.055 |
ppy | +0.046 |
anut | +0.044 |
ilia | +0.043 |
Neg logit contributions | |
This | -0.096 |
Sometimes | -0.091 |
Can | -0.090 |
purpose | -0.089 |
our | -0.089 |
Token | Mom |
Token ID | 11254 |
Feature activation | +5.66e+00 |
Pos logit contributions | |
arl | +0.024 |
purpose | +0.022 |
Hawaii | +0.020 |
Ireland | +0.020 |
outer | +0.019 |
Neg logit contributions | |
forgot | -0.025 |
felt | -0.023 |
laughed | -0.021 |
knew | -0.021 |
got | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
forgot | +0.071 |
felt | +0.042 |
struggled | +0.041 |
pretended | +0.038 |
swung | +0.035 |
Neg logit contributions | |
arl | -0.178 |
purpose | -0.172 |
This | -0.171 |
ay | -0.162 |
time | -0.161 |
Token |
|
Token ID | 198 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
arl | +0.026 |
purpose | +0.025 |
Hawaii | +0.023 |
Ireland | +0.023 |
outer | +0.022 |
Neg logit contributions | |
forgot | -0.030 |
felt | -0.028 |
knew | -0.027 |
got | -0.026 |
laughed | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
arl | +0.046 |
purpose | +0.045 |
Ireland | +0.041 |
Hawaii | +0.041 |
outer | +0.039 |
Neg logit contributions | |
forgot | -0.055 |
felt | -0.051 |
knew | -0.049 |
laughed | -0.047 |
got | -0.047 |
Token | Mar |
Token ID | 7676 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
arl | +0.040 |
purpose | +0.039 |
Hawaii | +0.035 |
Ireland | +0.034 |
This | +0.034 |
Neg logit contributions | |
forgot | -0.050 |
felt | -0.044 |
laughed | -0.042 |
knew | -0.042 |
got | -0.041 |
Token | iah |
Token ID | 9520 |
Feature activation | +1.37e+01 |
Pos logit contributions | |
forgot | +0.075 |
felt | +0.060 |
laughed | +0.058 |
pretended | +0.055 |
got | +0.054 |
Neg logit contributions | |
arl | -0.095 |
purpose | -0.091 |
Hawaii | -0.088 |
This | -0.086 |
outer | -0.084 |
Token | searched |
Token ID | 16499 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
forgot | +0.175 |
ppa | +0.141 |
ba | +0.115 |
pretended | +0.114 |
ppy | +0.114 |
Neg logit contributions | |
This | -0.322 |
time | -0.319 |
purpose | -0.310 |
A | -0.296 |
Earth | -0.286 |
Token | and |
Token ID | 290 |
Feature activation | +4.82e+00 |
Pos logit contributions | |
arl | +0.026 |
purpose | +0.026 |
Hawaii | +0.024 |
sight | +0.024 |
Ireland | +0.023 |
Neg logit contributions | |
forgot | -0.034 |
felt | -0.028 |
ppa | -0.027 |
laughed | -0.027 |
got | -0.026 |
Token | searched |
Token ID | 16499 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
forgot | +0.069 |
ppa | +0.057 |
ppy | +0.050 |
laughed | +0.044 |
pper | +0.043 |
Neg logit contributions | |
outer | -0.128 |
sight | -0.123 |
Hawaii | -0.122 |
This | -0.121 |
Ireland | -0.120 |
Token | , |
Token ID | 11 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
arl | +0.041 |
purpose | +0.041 |
Hawaii | +0.038 |
Ireland | +0.037 |
outer | +0.036 |
Neg logit contributions | |
forgot | -0.050 |
felt | -0.042 |
laughed | -0.040 |
ppa | -0.040 |
got | -0.039 |
Token | but |
Token ID | 475 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
forgot | +0.049 |
ppa | +0.043 |
ppy | +0.043 |
felt | +0.041 |
ie | +0.041 |
Neg logit contributions | |
sight | -0.030 |
purpose | -0.030 |
This | -0.029 |
Earth | -0.028 |
outer | -0.027 |
Token | anymore |
Token ID | 7471 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
arl | +0.007 |
purpose | +0.006 |
Hawaii | +0.006 |
Ireland | +0.006 |
This | +0.006 |
Neg logit contributions | |
forgot | -0.009 |
felt | -0.008 |
laughed | -0.007 |
my | -0.007 |
got | -0.007 |
Token | ." |
Token ID | 526 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
arl | +0.041 |
purpose | +0.038 |
Hawaii | +0.036 |
outer | +0.034 |
Ireland | +0.034 |
Neg logit contributions | |
forgot | -0.053 |
felt | -0.046 |
laughed | -0.044 |
got | -0.043 |
knew | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
arl | +0.020 |
purpose | +0.020 |
Hawaii | +0.018 |
outer | +0.017 |
Ireland | +0.017 |
Neg logit contributions | |
forgot | -0.029 |
felt | -0.026 |
laughed | -0.025 |
got | -0.024 |
knew | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
arl | +0.043 |
purpose | +0.042 |
Hawaii | +0.038 |
Ireland | +0.038 |
outer | +0.037 |
Neg logit contributions | |
forgot | -0.055 |
felt | -0.050 |
my | -0.047 |
knew | -0.047 |
laughed | -0.047 |
Token | L |
Token ID | 43 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
arl | +0.032 |
purpose | +0.032 |
Hawaii | +0.029 |
Ireland | +0.028 |
This | +0.028 |
Neg logit contributions | |
forgot | -0.043 |
felt | -0.038 |
laughed | -0.036 |
got | -0.035 |
knew | -0.035 |
Token | ola |
Token ID | 5708 |
Feature activation | +1.37e+01 |
Pos logit contributions | |
forgot | +0.069 |
felt | +0.060 |
laughed | +0.057 |
got | +0.055 |
started | +0.055 |
Neg logit contributions | |
arl | -0.048 |
purpose | -0.044 |
Hawaii | -0.042 |
Ireland | -0.040 |
This | -0.040 |
Token | felt |
Token ID | 2936 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
ppa | +0.181 |
forgot | +0.169 |
my | +0.161 |
anut | +0.156 |
ie | +0.149 |
Neg logit contributions | |
time | -0.293 |
This | -0.285 |
A | -0.280 |
purpose | -0.272 |
Earth | -0.271 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
forgot | +0.022 |
felt | +0.019 |
laughed | +0.018 |
got | +0.017 |
knew | +0.017 |
Neg logit contributions | |
arl | -0.022 |
purpose | -0.022 |
Hawaii | -0.021 |
Ireland | -0.020 |
This | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
arl | +0.030 |
purpose | +0.030 |
Hawaii | +0.028 |
Ireland | +0.027 |
outer | +0.027 |
Neg logit contributions | |
forgot | -0.028 |
felt | -0.024 |
laughed | -0.022 |
knew | -0.022 |
got | -0.022 |
Token | She |
Token ID | 1375 |
Feature activation | +9.23e+00 |
Pos logit contributions | |
purpose | +0.039 |
arl | +0.038 |
Hawaii | +0.036 |
outer | +0.036 |
Ireland | +0.034 |
Neg logit contributions | |
forgot | -0.039 |
felt | -0.036 |
knew | -0.033 |
my | -0.033 |
got | -0.032 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
ppa | +0.134 |
ie | +0.113 |
ppers | +0.112 |
anut | +0.110 |
oise | +0.107 |
Neg logit contributions | |
Can | -0.214 |
A | -0.203 |
Please | -0.196 |
hello | -0.196 |
It | -0.193 |
Token | way |
Token ID | 835 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
arl | +0.021 |
purpose | +0.019 |
Hawaii | +0.018 |
This | +0.018 |
Ireland | +0.017 |
Neg logit contributions | |
forgot | -0.031 |
felt | -0.027 |
got | -0.026 |
knew | -0.026 |
laughed | -0.026 |
Token | she |
Token ID | 673 |
Feature activation | +6.00e+00 |
Pos logit contributions | |
arl | +0.062 |
purpose | +0.060 |
Hawaii | +0.056 |
Ireland | +0.055 |
This | +0.053 |
Neg logit contributions | |
forgot | -0.065 |
felt | -0.056 |
laughed | -0.052 |
got | -0.051 |
knew | -0.051 |
Token | could |
Token ID | 714 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
forgot | +0.081 |
ppa | +0.078 |
my | +0.072 |
ie | +0.069 |
laughed | +0.068 |
Neg logit contributions | |
This | -0.138 |
Can | -0.137 |
Earth | -0.134 |
purpose | -0.133 |
Hawaii | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
purpose | +0.044 |
arl | +0.041 |
Hawaii | +0.039 |
This | +0.039 |
time | +0.038 |
Neg logit contributions | |
forgot | -0.037 |
felt | -0.030 |
laughed | -0.030 |
ppa | -0.029 |
my | -0.028 |
Token | Mar |
Token ID | 1526 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
arl | +0.042 |
purpose | +0.042 |
Ireland | +0.038 |
Hawaii | +0.037 |
outer | +0.036 |
Neg logit contributions | |
forgot | -0.043 |
felt | -0.041 |
knew | -0.039 |
got | -0.038 |
my | -0.037 |
Token | iah |
Token ID | 9520 |
Feature activation | +1.37e+01 |
Pos logit contributions | |
forgot | +0.081 |
laughed | +0.066 |
felt | +0.062 |
pretended | +0.061 |
swung | +0.057 |
Neg logit contributions | |
arl | -0.094 |
purpose | -0.087 |
This | -0.086 |
Hawaii | -0.086 |
Can | -0.082 |
Token | found |
Token ID | 1043 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
forgot | +0.176 |
ppa | +0.153 |
ppy | +0.142 |
my | +0.124 |
pretended | +0.123 |
Neg logit contributions | |
time | -0.311 |
purpose | -0.301 |
This | -0.291 |
Sale | -0.281 |
Earth | -0.276 |
Token | many |
Token ID | 867 |
Feature activation | -8.17e-01 |
Pos logit contributions | |
arl | +0.023 |
purpose | +0.023 |
Hawaii | +0.021 |
Ireland | +0.020 |
This | +0.020 |
Neg logit contributions | |
forgot | -0.039 |
felt | -0.034 |
laughed | -0.033 |
got | -0.032 |
knew | -0.032 |
Token | ways |
Token ID | 2842 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
purpose | +0.013 |
arl | +0.012 |
Hawaii | +0.011 |
Ireland | +0.011 |
outer | +0.011 |
Neg logit contributions | |
forgot | -0.023 |
felt | -0.020 |
laughed | -0.019 |
ppa | -0.019 |
my | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
purpose | +0.012 |
arl | +0.012 |
Hawaii | +0.011 |
This | +0.010 |
Ireland | +0.010 |
Neg logit contributions | |
forgot | -0.016 |
felt | -0.014 |
laughed | -0.013 |
got | -0.013 |
knew | -0.013 |
Token | help |
Token ID | 1037 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
purpose | +0.022 |
arl | +0.021 |
Hawaii | +0.020 |
This | +0.019 |
Ireland | +0.019 |
Neg logit contributions | |
forgot | -0.023 |
felt | -0.019 |
laughed | -0.019 |
my | -0.018 |
knew | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
arl | +0.023 |
purpose | +0.022 |
Hawaii | +0.022 |
Ireland | +0.021 |
outer | +0.020 |
Neg logit contributions | |
forgot | -0.023 |
felt | -0.020 |
laughed | -0.019 |
got | -0.019 |
knew | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
arl | +0.015 |
purpose | +0.015 |
Hawaii | +0.014 |
outer | +0.013 |
Ireland | +0.013 |
Neg logit contributions | |
forgot | -0.019 |
felt | -0.017 |
my | -0.016 |
laughed | -0.016 |
knew | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -9.58e-01 |
Pos logit contributions | |
arl | +0.024 |
purpose | +0.023 |
Hawaii | +0.021 |
Ireland | +0.020 |
outer | +0.020 |
Neg logit contributions | |
forgot | -0.037 |
felt | -0.032 |
laughed | -0.031 |
knew | -0.031 |
got | -0.030 |
Token | The |
Token ID | 464 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
arl | +0.018 |
purpose | +0.018 |
Hawaii | +0.016 |
This | +0.015 |
Ireland | +0.015 |
Neg logit contributions | |
forgot | -0.025 |
felt | -0.021 |
laughed | -0.020 |
got | -0.020 |
knew | -0.020 |
Token | strong |
Token ID | 1913 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
forgot | +0.062 |
felt | +0.053 |
laughed | +0.051 |
got | +0.050 |
knew | +0.049 |
Neg logit contributions | |
arl | -0.051 |
purpose | -0.051 |
Hawaii | -0.047 |
Ireland | -0.045 |
This | -0.044 |
Token | kid |
Token ID | 5141 |
Feature activation | +1.37e+01 |
Pos logit contributions | |
forgot | +0.066 |
felt | +0.055 |
laughed | +0.053 |
ppa | +0.053 |
pretended | +0.051 |
Neg logit contributions | |
purpose | -0.056 |
arl | -0.054 |
Hawaii | -0.049 |
sight | -0.048 |
This | -0.048 |
Token | came |
Token ID | 1625 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
ppa | +0.153 |
forgot | +0.153 |
ppy | +0.132 |
anut | +0.125 |
ba | +0.111 |
Neg logit contributions | |
time | -0.334 |
This | -0.314 |
purpose | -0.312 |
Earth | -0.306 |
Can | -0.302 |
Token | back |
Token ID | 736 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
forgot | +0.012 |
felt | +0.010 |
laughed | +0.010 |
ppa | +0.009 |
got | +0.009 |
Neg logit contributions | |
arl | -0.008 |
purpose | -0.008 |
Hawaii | -0.008 |
This | -0.007 |
Ireland | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
arl | +0.004 |
purpose | +0.004 |
Hawaii | +0.004 |
Ireland | +0.004 |
This | +0.004 |
Neg logit contributions | |
forgot | -0.004 |
felt | -0.003 |
laughed | -0.003 |
got | -0.003 |
knew | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
arl | +0.009 |
purpose | +0.009 |
Hawaii | +0.008 |
Ireland | +0.008 |
ay | +0.008 |
Neg logit contributions | |
forgot | -0.011 |
felt | -0.010 |
laughed | -0.009 |
knew | -0.009 |
got | -0.009 |
Token | beach |
Token ID | 10481 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
forgot | +0.028 |
felt | +0.025 |
laughed | +0.023 |
got | +0.023 |
knew | +0.023 |
Neg logit contributions | |
arl | -0.024 |
purpose | -0.023 |
Hawaii | -0.022 |
Ireland | -0.021 |
This | -0.021 |
Token | sadness |
Token ID | 25303 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
forgot | +0.014 |
felt | +0.012 |
laughed | +0.012 |
got | +0.011 |
knew | +0.011 |
Neg logit contributions | |
arl | -0.009 |
purpose | -0.009 |
Hawaii | -0.008 |
Ireland | -0.008 |
This | -0.008 |
Token | ." |
Token ID | 526 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
forgot | +0.009 |
felt | +0.008 |
laughed | +0.007 |
got | +0.007 |
knew | +0.007 |
Neg logit contributions | |
arl | -0.007 |
purpose | -0.007 |
Hawaii | -0.007 |
Ireland | -0.007 |
This | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
arl | +0.026 |
purpose | +0.026 |
Hawaii | +0.024 |
Ireland | +0.023 |
outer | +0.022 |
Neg logit contributions | |
forgot | -0.039 |
felt | -0.035 |
laughed | -0.033 |
knew | -0.032 |
my | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
arl | +0.044 |
purpose | +0.043 |
Hawaii | +0.040 |
Ireland | +0.038 |
outer | +0.037 |
Neg logit contributions | |
forgot | -0.072 |
felt | -0.064 |
laughed | -0.061 |
knew | -0.060 |
got | -0.060 |
Token | D |
Token ID | 35 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
purpose | +0.036 |
arl | +0.035 |
Hawaii | +0.032 |
This | +0.031 |
time | +0.030 |
Neg logit contributions | |
forgot | -0.051 |
felt | -0.044 |
laughed | -0.043 |
pretended | -0.041 |
got | -0.041 |
Token | iana |
Token ID | 7484 |
Feature activation | +1.37e+01 |
Pos logit contributions | |
forgot | +0.062 |
felt | +0.050 |
laughed | +0.050 |
pretended | +0.046 |
got | +0.045 |
Neg logit contributions | |
arl | -0.080 |
purpose | -0.077 |
Hawaii | -0.072 |
This | -0.071 |
Ireland | -0.069 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
forgot | +0.183 |
ppa | +0.165 |
ppy | +0.150 |
ppers | +0.148 |
anut | +0.146 |
Neg logit contributions | |
purpose | -0.292 |
This | -0.291 |
time | -0.290 |
A | -0.277 |
Earth | -0.268 |
Token | so |
Token ID | 523 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
forgot | +0.022 |
felt | +0.019 |
laughed | +0.017 |
knew | +0.017 |
got | +0.017 |
Neg logit contributions | |
arl | -0.024 |
purpose | -0.023 |
Hawaii | -0.022 |
Ireland | -0.021 |
This | -0.021 |
Token | bad |
Token ID | 2089 |
Feature activation | -8.80e-01 |
Pos logit contributions | |
forgot | +0.040 |
felt | +0.034 |
laughed | +0.032 |
got | +0.031 |
knew | +0.031 |
Neg logit contributions | |
arl | -0.043 |
purpose | -0.043 |
Hawaii | -0.040 |
This | -0.039 |
Ireland | -0.039 |
Token | for |
Token ID | 329 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
arl | +0.018 |
purpose | +0.018 |
Hawaii | +0.017 |
Ireland | +0.016 |
outer | +0.016 |
Neg logit contributions | |
forgot | -0.021 |
felt | -0.018 |
laughed | -0.017 |
got | -0.017 |
knew | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
arl | +0.031 |
purpose | +0.030 |
Hawaii | +0.028 |
ay | +0.027 |
This | +0.027 |
Neg logit contributions | |
forgot | -0.033 |
felt | -0.030 |
laughed | -0.028 |
knew | -0.028 |
got | -0.027 |
Token | helping |
Token ID | 5742 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
forgot | +0.013 |
felt | +0.011 |
laughed | +0.010 |
knew | +0.010 |
got | +0.010 |
Neg logit contributions | |
arl | -0.012 |
purpose | -0.011 |
Hawaii | -0.011 |
time | -0.010 |
ay | -0.010 |
Token | each |
Token ID | 1123 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
arl | +0.093 |
purpose | +0.084 |
Hawaii | +0.082 |
Ireland | +0.080 |
outer | +0.080 |
Neg logit contributions | |
forgot | -0.100 |
felt | -0.090 |
got | -0.086 |
my | -0.086 |
's | -0.084 |
Token | other |
Token ID | 584 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
forgot | +0.002 |
felt | +0.002 |
laughed | +0.002 |
knew | +0.002 |
got | +0.002 |
Neg logit contributions | |
arl | -0.003 |
purpose | -0.003 |
Hawaii | -0.003 |
Ireland | -0.003 |
This | -0.003 |
Token | ." |
Token ID | 526 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
arl | +0.011 |
purpose | +0.010 |
Hawaii | +0.010 |
Ireland | +0.009 |
outer | +0.009 |
Neg logit contributions | |
forgot | -0.014 |
felt | -0.013 |
got | -0.012 |
my | -0.012 |
knew | -0.012 |
Token | Big |
Token ID | 4403 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
arl | +0.036 |
purpose | +0.034 |
Hawaii | +0.031 |
Ireland | +0.030 |
outer | +0.030 |
Neg logit contributions | |
forgot | -0.059 |
felt | -0.054 |
knew | -0.050 |
got | -0.050 |
laughed | -0.050 |
Token | gie |
Token ID | 22699 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
forgot | +0.039 |
felt | +0.032 |
laughed | +0.030 |
got | +0.029 |
knew | +0.029 |
Neg logit contributions | |
arl | -0.049 |
purpose | -0.048 |
Hawaii | -0.045 |
This | -0.044 |
Ireland | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
forgot | +0.221 |
ppa | +0.186 |
ppy | +0.181 |
ilia | +0.175 |
pretended | +0.172 |
Neg logit contributions | |
This | -0.273 |
A | -0.269 |
The | -0.259 |
purpose | -0.255 |
It | -0.243 |
Token | his |
Token ID | 465 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
forgot | +0.086 |
felt | +0.073 |
laughed | +0.072 |
ppa | +0.072 |
my | +0.070 |
Neg logit contributions | |
arl | -0.050 |
purpose | -0.049 |
Hawaii | -0.044 |
Ireland | -0.042 |
This | -0.042 |
Token | friends |
Token ID | 2460 |
Feature activation | +9.68e+00 |
Pos logit contributions | |
forgot | +0.071 |
felt | +0.058 |
laughed | +0.056 |
got | +0.055 |
my | +0.054 |
Neg logit contributions | |
purpose | -0.074 |
arl | -0.074 |
Hawaii | -0.066 |
sight | -0.065 |
Ireland | -0.064 |
Token | listened |
Token ID | 16399 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
ppa | +0.134 |
ppy | +0.123 |
anut | +0.121 |
nie | +0.119 |
mo | +0.117 |
Neg logit contributions | |
hello | -0.196 |
time | -0.189 |
east | -0.187 |
Earth | -0.181 |
sight | -0.180 |
Token | to |
Token ID | 284 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
arl | +0.069 |
purpose | +0.067 |
Hawaii | +0.063 |
Ireland | +0.060 |
outer | +0.058 |
Neg logit contributions | |
forgot | -0.087 |
felt | -0.077 |
laughed | -0.073 |
got | -0.072 |
knew | -0.072 |
Token | as |
Token ID | 355 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
arl | +0.009 |
purpose | +0.009 |
Hawaii | +0.008 |
Ireland | +0.008 |
This | +0.008 |
Neg logit contributions | |
forgot | -0.012 |
felt | -0.010 |
laughed | -0.010 |
got | -0.010 |
knew | -0.010 |
Token | Hel |
Token ID | 5053 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
arl | +0.028 |
purpose | +0.027 |
Hawaii | +0.024 |
This | +0.023 |
Ireland | +0.023 |
Neg logit contributions | |
forgot | -0.034 |
felt | -0.030 |
laughed | -0.029 |
knew | -0.029 |
had | -0.028 |
Token | io |
Token ID | 952 |
Feature activation | +8.66e+00 |
Pos logit contributions | |
arl | +0.007 |
purpose | +0.007 |
Hawaii | +0.007 |
outer | +0.007 |
This | +0.007 |
Neg logit contributions | |
forgot | -0.005 |
felt | -0.004 |
laughed | -0.004 |
got | -0.004 |
knew | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
forgot | +0.154 |
felt | +0.115 |
ppa | +0.110 |
laughed | +0.106 |
struggled | +0.106 |
Neg logit contributions | |
purpose | -0.200 |
arl | -0.200 |
Hawaii | -0.196 |
This | -0.195 |
ay | -0.192 |
Token | Hel |
Token ID | 5053 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
purpose | +0.032 |
arl | +0.031 |
Hawaii | +0.028 |
Ireland | +0.028 |
outer | +0.027 |
Neg logit contributions | |
forgot | -0.035 |
felt | -0.032 |
knew | -0.030 |
got | -0.030 |
my | -0.030 |
Token | io |
Token ID | 952 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
forgot | +0.023 |
felt | +0.019 |
laughed | +0.018 |
my | +0.017 |
got | +0.017 |
Neg logit contributions | |
arl | -0.024 |
purpose | -0.024 |
Hawaii | -0.022 |
This | -0.022 |
Ireland | -0.022 |
Token | explained |
Token ID | 4893 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
forgot | +0.191 |
ppa | +0.141 |
struggled | +0.111 |
punished | +0.111 |
pretended | +0.109 |
Neg logit contributions | |
ay | -0.337 |
purpose | -0.316 |
This | -0.312 |
Earth | -0.303 |
Home | -0.299 |
Token | that |
Token ID | 326 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
arl | +0.044 |
purpose | +0.042 |
Hawaii | +0.039 |
Ireland | +0.038 |
outer | +0.036 |
Neg logit contributions | |
forgot | -0.055 |
felt | -0.050 |
knew | -0.048 |
laughed | -0.048 |
got | -0.047 |
Token | their |
Token ID | 511 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
forgot | +0.006 |
felt | +0.005 |
laughed | +0.005 |
got | +0.005 |
knew | +0.005 |
Neg logit contributions | |
arl | -0.006 |
purpose | -0.006 |
Hawaii | -0.006 |
Ireland | -0.006 |
This | -0.006 |
Token | oxygen |
Token ID | 11863 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
forgot | +0.040 |
felt | +0.036 |
knew | +0.034 |
laughed | +0.034 |
had | +0.034 |
Neg logit contributions | |
arl | -0.030 |
purpose | -0.027 |
Hawaii | -0.025 |
This | -0.025 |
Ireland | -0.025 |
Token | was |
Token ID | 373 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
forgot | +0.047 |
ppa | +0.037 |
felt | +0.034 |
laughed | +0.033 |
my | +0.033 |
Neg logit contributions | |
purpose | -0.082 |
arl | -0.077 |
This | -0.075 |
sight | -0.074 |
Hawaii | -0.073 |
Token | ys |
Token ID | 893 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
forgot | +0.055 |
felt | +0.046 |
laughed | +0.043 |
got | +0.041 |
knew | +0.041 |
Neg logit contributions | |
arl | -0.074 |
purpose | -0.074 |
Hawaii | -0.069 |
Ireland | -0.067 |
This | -0.067 |
Token | !" |
Token ID | 2474 |
Feature activation | -7.75e-01 |
Pos logit contributions | |
forgot | +0.037 |
felt | +0.032 |
laughed | +0.031 |
got | +0.031 |
my | +0.031 |
Neg logit contributions | |
arl | -0.033 |
purpose | -0.032 |
Hawaii | -0.030 |
Ireland | -0.029 |
outer | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
arl | +0.014 |
purpose | +0.013 |
Hawaii | +0.012 |
outer | +0.011 |
Ireland | +0.011 |
Neg logit contributions | |
forgot | -0.021 |
felt | -0.018 |
laughed | -0.018 |
got | -0.017 |
knew | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
arl | +0.053 |
purpose | +0.052 |
Hawaii | +0.046 |
Ireland | +0.045 |
outer | +0.045 |
Neg logit contributions | |
forgot | -0.074 |
felt | -0.066 |
laughed | -0.063 |
got | -0.062 |
knew | -0.062 |
Token | The |
Token ID | 464 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
arl | +0.043 |
purpose | +0.043 |
Hawaii | +0.039 |
Ireland | +0.037 |
This | +0.037 |
Neg logit contributions | |
forgot | -0.058 |
felt | -0.051 |
laughed | -0.048 |
got | -0.047 |
knew | -0.047 |
Token | kid |
Token ID | 5141 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
forgot | +0.037 |
felt | +0.032 |
laughed | +0.030 |
got | +0.029 |
knew | +0.029 |
Neg logit contributions | |
arl | -0.036 |
purpose | -0.035 |
Hawaii | -0.033 |
Ireland | -0.032 |
This | -0.031 |
Token | had |
Token ID | 550 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
ppa | +0.182 |
forgot | +0.162 |
ppy | +0.142 |
my | +0.140 |
any | +0.136 |
Neg logit contributions | |
time | -0.309 |
Earth | -0.305 |
This | -0.300 |
A | -0.297 |
hello | -0.296 |
Token | a |
Token ID | 257 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
forgot | +0.011 |
felt | +0.009 |
laughed | +0.009 |
my | +0.009 |
knew | +0.009 |
Neg logit contributions | |
purpose | -0.008 |
arl | -0.008 |
Hawaii | -0.007 |
Ireland | -0.007 |
This | -0.007 |
Token | grand |
Token ID | 4490 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
forgot | +0.047 |
felt | +0.041 |
laughed | +0.040 |
my | +0.039 |
ppa | +0.038 |
Neg logit contributions | |
purpose | -0.027 |
arl | -0.026 |
Hawaii | -0.024 |
Ireland | -0.023 |
This | -0.022 |
Token | old |
Token ID | 1468 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
forgot | +0.044 |
felt | +0.037 |
laughed | +0.035 |
got | +0.034 |
knew | +0.034 |
Neg logit contributions | |
purpose | -0.036 |
arl | -0.036 |
Hawaii | -0.033 |
Ireland | -0.032 |
This | -0.031 |
Token | time |
Token ID | 640 |
Feature activation | -8.55e-01 |
Pos logit contributions | |
forgot | +0.044 |
felt | +0.037 |
laughed | +0.035 |
got | +0.034 |
knew | +0.033 |
Neg logit contributions | |
purpose | -0.046 |
arl | -0.045 |
Hawaii | -0.042 |
Ireland | -0.041 |
This | -0.040 |
Token | above |
Token ID | 2029 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
arl | +0.018 |
purpose | +0.016 |
Hawaii | +0.015 |
This | +0.015 |
Ireland | +0.015 |
Neg logit contributions | |
forgot | -0.019 |
felt | -0.017 |
laughed | -0.016 |
got | -0.016 |
had | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
arl | +0.033 |
purpose | +0.029 |
Hawaii | +0.027 |
Ireland | +0.027 |
This | +0.027 |
Neg logit contributions | |
forgot | -0.028 |
felt | -0.026 |
had | -0.024 |
laughed | -0.024 |
got | -0.024 |
Token | Even |
Token ID | 3412 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
arl | +0.009 |
purpose | +0.009 |
Hawaii | +0.008 |
Ireland | +0.007 |
outer | +0.007 |
Neg logit contributions | |
forgot | -0.011 |
felt | -0.010 |
laughed | -0.009 |
had | -0.009 |
knew | -0.009 |
Token | though |
Token ID | 996 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
forgot | +0.040 |
felt | +0.034 |
laughed | +0.032 |
got | +0.032 |
knew | +0.032 |
Neg logit contributions | |
arl | -0.036 |
purpose | -0.036 |
Hawaii | -0.033 |
Ireland | -0.032 |
outer | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
arl | +0.010 |
purpose | +0.009 |
Hawaii | +0.009 |
This | +0.009 |
Ireland | +0.009 |
Neg logit contributions | |
forgot | -0.010 |
felt | -0.009 |
laughed | -0.009 |
got | -0.008 |
knew | -0.008 |
Token | kid |
Token ID | 5141 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
forgot | +0.061 |
felt | +0.054 |
laughed | +0.051 |
knew | +0.050 |
got | +0.050 |
Neg logit contributions | |
arl | -0.044 |
purpose | -0.043 |
Hawaii | -0.039 |
This | -0.038 |
Ireland | -0.037 |
Token | had |
Token ID | 550 |
Feature activation | +9.66e-01 |
Pos logit contributions | |
ppa | +0.150 |
forgot | +0.148 |
ppy | +0.123 |
anut | +0.108 |
my | +0.105 |
Neg logit contributions | |
time | -0.341 |
purpose | -0.341 |
This | -0.329 |
sight | -0.326 |
Earth | -0.322 |
Token | never |
Token ID | 1239 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
forgot | +0.020 |
felt | +0.016 |
laughed | +0.016 |
my | +0.015 |
knew | +0.015 |
Neg logit contributions | |
purpose | -0.022 |
arl | -0.022 |
Hawaii | -0.021 |
Ireland | -0.020 |
This | -0.020 |
Token | seen |
Token ID | 1775 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
forgot | +0.046 |
ppa | +0.043 |
ppy | +0.037 |
mie | +0.035 |
my | +0.035 |
Neg logit contributions | |
purpose | -0.090 |
sight | -0.084 |
Hawaii | -0.082 |
outer | -0.081 |
arl | -0.081 |
Token | anything |
Token ID | 1997 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
arl | +0.038 |
purpose | +0.033 |
Hawaii | +0.030 |
This | +0.029 |
Ireland | +0.029 |
Neg logit contributions | |
forgot | -0.058 |
felt | -0.053 |
laughed | -0.051 |
knew | -0.049 |
got | -0.049 |
Token | so |
Token ID | 523 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
forgot | +0.012 |
felt | +0.010 |
laughed | +0.010 |
got | +0.009 |
knew | +0.009 |
Neg logit contributions | |
arl | -0.011 |
purpose | -0.011 |
Hawaii | -0.010 |
Ireland | -0.010 |
This | -0.010 |
Token | from |
Token ID | 422 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
forgot | +0.022 |
felt | +0.018 |
laughed | +0.018 |
ppa | +0.018 |
my | +0.017 |
Neg logit contributions | |
purpose | -0.019 |
arl | -0.018 |
Hawaii | -0.017 |
This | -0.017 |
sight | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
arl | +0.028 |
purpose | +0.026 |
This | +0.024 |
Hawaii | +0.024 |
Ireland | +0.023 |
Neg logit contributions | |
forgot | -0.033 |
felt | -0.029 |
laughed | -0.028 |
knew | -0.028 |
got | -0.027 |
Token | tank |
Token ID | 6873 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
arl | +0.017 |
purpose | +0.015 |
Hawaii | +0.014 |
This | +0.014 |
Ireland | +0.013 |
Neg logit contributions | |
forgot | -0.022 |
felt | -0.020 |
laughed | -0.019 |
had | -0.019 |
knew | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
forgot | +0.012 |
felt | +0.010 |
laughed | +0.010 |
ppa | +0.010 |
pretended | +0.010 |
Neg logit contributions | |
purpose | -0.010 |
arl | -0.009 |
This | -0.009 |
Hawaii | -0.009 |
time | -0.008 |
Token | Hel |
Token ID | 5053 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
purpose | +0.038 |
arl | +0.036 |
Hawaii | +0.033 |
Ireland | +0.033 |
outer | +0.032 |
Neg logit contributions | |
forgot | -0.046 |
felt | -0.042 |
knew | -0.039 |
my | -0.039 |
got | -0.039 |
Token | io |
Token ID | 952 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
forgot | +0.011 |
felt | +0.010 |
laughed | +0.009 |
my | +0.009 |
got | +0.009 |
Neg logit contributions | |
arl | -0.012 |
purpose | -0.011 |
Hawaii | -0.011 |
Ireland | -0.011 |
This | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
forgot | +0.196 |
ppa | +0.146 |
struggled | +0.120 |
pretended | +0.120 |
ppy | +0.113 |
Neg logit contributions | |
ay | -0.327 |
purpose | -0.321 |
This | -0.319 |
Earth | -0.297 |
time | -0.296 |
Token | thankful |
Token ID | 25535 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
forgot | +0.050 |
felt | +0.041 |
laughed | +0.039 |
my | +0.038 |
ppa | +0.038 |
Neg logit contributions | |
purpose | -0.050 |
arl | -0.049 |
Hawaii | -0.047 |
This | -0.045 |
Ireland | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +4.87e+00 |
Pos logit contributions | |
arl | +0.039 |
purpose | +0.038 |
Hawaii | +0.035 |
This | +0.034 |
Ireland | +0.033 |
Neg logit contributions | |
forgot | -0.049 |
felt | -0.042 |
laughed | -0.040 |
got | -0.039 |
knew | -0.039 |
Token | thanked |
Token ID | 26280 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
forgot | +0.072 |
ppy | +0.061 |
ppa | +0.060 |
my | +0.059 |
ie | +0.052 |
Neg logit contributions | |
purpose | -0.122 |
This | -0.121 |
arl | -0.120 |
pleasure | -0.118 |
Hawaii | -0.118 |
Token | T |
Token ID | 309 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
arl | +0.059 |
purpose | +0.057 |
Hawaii | +0.052 |
Ireland | +0.051 |
This | +0.049 |
Neg logit contributions | |
forgot | -0.074 |
felt | -0.066 |
laughed | -0.062 |
knew | -0.062 |
got | -0.061 |
Token | it |
Token ID | 340 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
arl | +0.100 |
purpose | +0.091 |
This | +0.083 |
Ireland | +0.082 |
Hawaii | +0.081 |
Neg logit contributions | |
forgot | -0.096 |
felt | -0.088 |
got | -0.083 |
decided | -0.082 |
had | -0.082 |
Token | ." |
Token ID | 526 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
arl | +0.030 |
purpose | +0.029 |
Hawaii | +0.027 |
Ireland | +0.026 |
outer | +0.025 |
Neg logit contributions | |
forgot | -0.041 |
felt | -0.036 |
got | -0.034 |
my | -0.034 |
laughed | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
arl | +0.021 |
purpose | +0.020 |
Hawaii | +0.018 |
This | +0.018 |
Ireland | +0.017 |
Neg logit contributions | |
forgot | -0.033 |
felt | -0.029 |
laughed | -0.028 |
got | -0.027 |
knew | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
arl | +0.042 |
purpose | +0.041 |
Hawaii | +0.037 |
Ireland | +0.036 |
outer | +0.035 |
Neg logit contributions | |
forgot | -0.066 |
felt | -0.059 |
knew | -0.056 |
laughed | -0.055 |
got | -0.055 |
Token | G |
Token ID | 38 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
purpose | +0.036 |
arl | +0.036 |
Hawaii | +0.033 |
Ireland | +0.031 |
This | +0.031 |
Neg logit contributions | |
forgot | -0.051 |
felt | -0.044 |
laughed | -0.043 |
got | -0.041 |
pretended | -0.041 |
Token | avin |
Token ID | 20637 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
forgot | +0.052 |
laughed | +0.046 |
felt | +0.044 |
pretended | +0.043 |
got | +0.041 |
Neg logit contributions | |
arl | -0.049 |
purpose | -0.045 |
Hawaii | -0.043 |
This | -0.041 |
ay | -0.040 |
Token | thanked |
Token ID | 26280 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
forgot | +0.166 |
ppa | +0.143 |
ppy | +0.110 |
anut | +0.108 |
tasted | +0.102 |
Neg logit contributions | |
purpose | -0.306 |
time | -0.298 |
This | -0.289 |
Earth | -0.280 |
Home | -0.279 |
Token | the |
Token ID | 262 |
Feature activation | +6.76e-01 |
Pos logit contributions | |
arl | +0.031 |
purpose | +0.030 |
Hawaii | +0.027 |
Ireland | +0.027 |
This | +0.026 |
Neg logit contributions | |
forgot | -0.039 |
felt | -0.035 |
laughed | -0.033 |
knew | -0.033 |
got | -0.032 |
Token | woman |
Token ID | 2415 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
forgot | +0.014 |
felt | +0.011 |
laughed | +0.011 |
got | +0.010 |
knew | +0.010 |
Neg logit contributions | |
arl | -0.016 |
purpose | -0.016 |
Hawaii | -0.015 |
Ireland | -0.015 |
This | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +6.60e+00 |
Pos logit contributions | |
forgot | +0.104 |
felt | +0.080 |
struggled | +0.079 |
pretended | +0.076 |
laughed | +0.075 |
Neg logit contributions | |
arl | -0.088 |
purpose | -0.088 |
This | -0.087 |
Hawaii | -0.085 |
Earth | -0.083 |
Token | drove |
Token ID | 10357 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
forgot | +0.089 |
ppy | +0.076 |
my | +0.072 |
ppa | +0.071 |
aky | +0.066 |
Neg logit contributions | |
Hawaii | -0.162 |
hello | -0.154 |
Earth | -0.152 |
sight | -0.150 |
east | -0.149 |
Token | !" |
Token ID | 2474 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
arl | +0.082 |
Hawaii | +0.076 |
purpose | +0.076 |
Ireland | +0.072 |
outer | +0.070 |
Neg logit contributions | |
forgot | -0.096 |
felt | -0.086 |
my | -0.081 |
got | -0.080 |
laughed | -0.079 |
Token | |
Token ID | 220 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
purpose | +0.014 |
arl | +0.013 |
This | +0.013 |
Hawaii | +0.012 |
Ireland | +0.012 |
Neg logit contributions | |
forgot | -0.020 |
felt | -0.017 |
laughed | -0.017 |
got | -0.016 |
knew | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
arl | +0.044 |
purpose | +0.043 |
Hawaii | +0.040 |
Ireland | +0.038 |
outer | +0.038 |
Neg logit contributions | |
forgot | -0.069 |
felt | -0.062 |
laughed | -0.058 |
got | -0.058 |
knew | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
arl | +0.054 |
purpose | +0.053 |
Hawaii | +0.049 |
Ireland | +0.047 |
outer | +0.046 |
Neg logit contributions | |
forgot | -0.074 |
felt | -0.067 |
laughed | -0.063 |
knew | -0.063 |
my | -0.062 |
Token | The |
Token ID | 464 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
arl | +0.042 |
purpose | +0.041 |
Hawaii | +0.038 |
Ireland | +0.036 |
This | +0.036 |
Neg logit contributions | |
forgot | -0.064 |
felt | -0.056 |
laughed | -0.053 |
got | -0.052 |
knew | -0.052 |
Token | painter |
Token ID | 34537 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
forgot | +0.025 |
felt | +0.021 |
laughed | +0.020 |
pretended | +0.020 |
got | +0.020 |
Neg logit contributions | |
purpose | -0.021 |
arl | -0.020 |
Hawaii | -0.019 |
sight | -0.019 |
Ireland | -0.018 |
Token | said |
Token ID | 531 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
ppa | +0.187 |
forgot | +0.178 |
ppy | +0.170 |
ba | +0.156 |
anut | +0.155 |
Neg logit contributions | |
time | -0.290 |
purpose | -0.278 |
A | -0.262 |
This | -0.259 |
sight | -0.257 |
Token | , |
Token ID | 11 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
arl | +0.016 |
purpose | +0.015 |
Hawaii | +0.015 |
Ireland | +0.014 |
outer | +0.014 |
Neg logit contributions | |
forgot | -0.013 |
felt | -0.012 |
laughed | -0.012 |
got | -0.011 |
knew | -0.011 |
Token | " |
Token ID | 366 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
forgot | +0.034 |
felt | +0.031 |
laughed | +0.030 |
got | +0.029 |
knew | +0.029 |
Neg logit contributions | |
arl | -0.018 |
purpose | -0.018 |
Hawaii | -0.016 |
Ireland | -0.015 |
outer | -0.014 |
Token | Thank |
Token ID | 10449 |
Feature activation | -4.76e+00 |
Pos logit contributions | |
purpose | +0.021 |
arl | +0.021 |
Can | +0.019 |
This | +0.018 |
Hawaii | +0.018 |
Neg logit contributions | |
forgot | -0.035 |
felt | -0.031 |
laughed | -0.030 |
got | -0.029 |
pretended | -0.028 |
Token | you |
Token ID | 345 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
arl | +0.096 |
purpose | +0.094 |
Hawaii | +0.087 |
Ireland | +0.084 |
outer | +0.082 |
Neg logit contributions | |
forgot | -0.107 |
felt | -0.100 |
laughed | -0.098 |
my | -0.096 |
ie | -0.095 |
Token | colorful |
Token ID | 20239 |
Feature activation | -8.11e-01 |
Pos logit contributions | |
forgot | +0.004 |
felt | +0.003 |
laughed | +0.003 |
knew | +0.003 |
got | +0.003 |
Neg logit contributions | |
arl | -0.002 |
purpose | -0.002 |
Hawaii | -0.002 |
This | -0.002 |
Ireland | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
arl | +0.015 |
purpose | +0.014 |
Hawaii | +0.013 |
Ireland | +0.013 |
This | +0.013 |
Neg logit contributions | |
forgot | -0.021 |
felt | -0.018 |
laughed | -0.018 |
got | -0.017 |
knew | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -9.87e-01 |
Pos logit contributions | |
arl | +0.020 |
purpose | +0.020 |
Hawaii | +0.018 |
Ireland | +0.018 |
This | +0.017 |
Neg logit contributions | |
forgot | -0.023 |
felt | -0.019 |
laughed | -0.018 |
knew | -0.018 |
got | -0.018 |
Token | Everywhere |
Token ID | 48201 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
arl | +0.016 |
purpose | +0.016 |
Hawaii | +0.014 |
outer | +0.014 |
Ireland | +0.013 |
Neg logit contributions | |
forgot | -0.026 |
felt | -0.024 |
my | -0.023 |
knew | -0.023 |
got | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
forgot | +0.008 |
felt | +0.007 |
laughed | +0.006 |
got | +0.006 |
knew | +0.006 |
Neg logit contributions | |
arl | -0.006 |
purpose | -0.006 |
Hawaii | -0.006 |
Ireland | -0.006 |
This | -0.006 |
Token | person |
Token ID | 1048 |
Feature activation | +1.35e+01 |
Pos logit contributions | |
forgot | +0.049 |
felt | +0.041 |
laughed | +0.040 |
got | +0.038 |
my | +0.038 |
Neg logit contributions | |
purpose | -0.041 |
arl | -0.041 |
Hawaii | -0.038 |
Ireland | -0.037 |
This | -0.036 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
ppa | +0.164 |
forgot | +0.151 |
ppy | +0.139 |
anut | +0.127 |
ppers | +0.120 |
Neg logit contributions | |
Earth | -0.327 |
time | -0.314 |
Hawaii | -0.312 |
life | -0.307 |
history | -0.302 |
Token | , |
Token ID | 11 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
forgot | +0.005 |
felt | +0.004 |
laughed | +0.004 |
got | +0.004 |
knew | +0.004 |
Neg logit contributions | |
arl | -0.004 |
purpose | -0.004 |
Hawaii | -0.004 |
Ireland | -0.004 |
This | -0.004 |
Token | there |
Token ID | 612 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
forgot | +0.037 |
felt | +0.032 |
laughed | +0.030 |
ppa | +0.030 |
knew | +0.030 |
Neg logit contributions | |
purpose | -0.028 |
arl | -0.028 |
Hawaii | -0.026 |
Ireland | -0.025 |
This | -0.025 |
Token | were |
Token ID | 547 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
forgot | +0.033 |
ppa | +0.027 |
ppy | +0.023 |
my | +0.022 |
felt | +0.021 |
Neg logit contributions | |
purpose | -0.062 |
time | -0.059 |
Hawaii | -0.059 |
This | -0.058 |
Earth | -0.057 |
Token | bright |
Token ID | 6016 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
forgot | +0.022 |
felt | +0.019 |
laughed | +0.018 |
got | +0.018 |
knew | +0.018 |
Neg logit contributions | |
purpose | -0.018 |
arl | -0.018 |
Hawaii | -0.016 |
Ireland | -0.016 |
This | -0.015 |
Token | tight |
Token ID | 5381 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
forgot | +0.031 |
felt | +0.025 |
laughed | +0.025 |
struggled | +0.024 |
knew | +0.024 |
Neg logit contributions | |
purpose | -0.028 |
arl | -0.027 |
time | -0.025 |
This | -0.024 |
Hawaii | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
arl | +0.015 |
purpose | +0.015 |
Hawaii | +0.014 |
Ireland | +0.014 |
This | +0.014 |
Neg logit contributions | |
forgot | -0.013 |
felt | -0.011 |
laughed | -0.010 |
got | -0.010 |
knew | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
purpose | +0.004 |
arl | +0.004 |
Hawaii | +0.004 |
outer | +0.004 |
Ireland | +0.004 |
Neg logit contributions | |
forgot | -0.006 |
felt | -0.005 |
my | -0.005 |
laughed | -0.005 |
had | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -8.03e-01 |
Pos logit contributions | |
purpose | +0.016 |
arl | +0.016 |
Hawaii | +0.014 |
Ireland | +0.014 |
outer | +0.013 |
Neg logit contributions | |
forgot | -0.023 |
felt | -0.020 |
laughed | -0.019 |
knew | -0.019 |
my | -0.019 |
Token | M |
Token ID | 44 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
arl | +0.016 |
purpose | +0.015 |
Hawaii | +0.014 |
Ireland | +0.014 |
This | +0.014 |
Neg logit contributions | |
forgot | -0.020 |
felt | -0.017 |
laughed | -0.016 |
got | -0.016 |
knew | -0.016 |
Token | addy |
Token ID | 13218 |
Feature activation | +1.35e+01 |
Pos logit contributions | |
forgot | +0.071 |
felt | +0.063 |
laughed | +0.060 |
pretended | +0.059 |
got | +0.059 |
Neg logit contributions | |
arl | -0.040 |
purpose | -0.038 |
Hawaii | -0.033 |
time | -0.032 |
ay | -0.032 |
Token | never |
Token ID | 1239 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
forgot | +0.150 |
ppa | +0.135 |
ppy | +0.131 |
ba | +0.121 |
anut | +0.117 |
Neg logit contributions | |
time | -0.313 |
This | -0.311 |
purpose | -0.303 |
A | -0.298 |
Home | -0.292 |
Token | worried |
Token ID | 7960 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
ppa | +0.048 |
forgot | +0.037 |
ba | +0.037 |
anut | +0.033 |
mie | +0.029 |
Neg logit contributions | |
purpose | -0.129 |
arl | -0.127 |
sight | -0.123 |
pleasure | -0.121 |
Hawaii | -0.120 |
Token | about |
Token ID | 546 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
arl | +0.032 |
purpose | +0.031 |
Hawaii | +0.029 |
Ireland | +0.028 |
outer | +0.027 |
Neg logit contributions | |
forgot | -0.046 |
felt | -0.041 |
laughed | -0.040 |
got | -0.038 |
knew | -0.038 |
Token | losing |
Token ID | 6078 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
arl | +0.016 |
purpose | +0.015 |
Hawaii | +0.014 |
Ireland | +0.013 |
This | +0.013 |
Neg logit contributions | |
forgot | -0.022 |
felt | -0.019 |
laughed | -0.018 |
knew | -0.018 |
got | -0.018 |
Token | her |
Token ID | 607 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
arl | +0.036 |
purpose | +0.035 |
Hawaii | +0.032 |
Ireland | +0.031 |
This | +0.030 |
Neg logit contributions | |
forgot | -0.052 |
felt | -0.046 |
laughed | -0.044 |
got | -0.042 |
knew | -0.042 |
Token | fun |
Token ID | 1257 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
purpose | +0.006 |
arl | +0.006 |
Hawaii | +0.005 |
Ireland | +0.005 |
This | +0.005 |
Neg logit contributions | |
forgot | -0.008 |
felt | -0.007 |
laughed | -0.007 |
knew | -0.007 |
got | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
arl | +0.062 |
purpose | +0.061 |
Hawaii | +0.057 |
Ireland | +0.055 |
This | +0.054 |
Neg logit contributions | |
forgot | -0.067 |
felt | -0.058 |
laughed | -0.055 |
got | -0.054 |
knew | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
arl | +0.027 |
purpose | +0.026 |
Hawaii | +0.024 |
Ireland | +0.023 |
outer | +0.023 |
Neg logit contributions | |
forgot | -0.034 |
felt | -0.029 |
laughed | -0.028 |
knew | -0.027 |
got | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
arl | +0.029 |
purpose | +0.028 |
Hawaii | +0.025 |
Ireland | +0.024 |
This | +0.024 |
Neg logit contributions | |
forgot | -0.053 |
felt | -0.047 |
laughed | -0.045 |
got | -0.044 |
knew | -0.044 |
Token | L |
Token ID | 43 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
purpose | +0.020 |
arl | +0.019 |
time | +0.018 |
Hawaii | +0.017 |
This | +0.017 |
Neg logit contributions | |
forgot | -0.026 |
felt | -0.022 |
laughed | -0.021 |
pretended | -0.021 |
got | -0.020 |
Token | ola |
Token ID | 5708 |
Feature activation | +1.35e+01 |
Pos logit contributions | |
forgot | +0.074 |
felt | +0.065 |
laughed | +0.061 |
got | +0.061 |
pretended | +0.060 |
Neg logit contributions | |
arl | -0.046 |
purpose | -0.043 |
Hawaii | -0.039 |
ires | -0.039 |
time | -0.038 |
Token | shook |
Token ID | 14682 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
ppa | +0.162 |
forgot | +0.146 |
my | +0.142 |
ppy | +0.140 |
anut | +0.139 |
Neg logit contributions | |
time | -0.299 |
A | -0.270 |
This | -0.270 |
purpose | -0.260 |
Sunday | -0.258 |
Token | her |
Token ID | 607 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
arl | +0.036 |
purpose | +0.036 |
Hawaii | +0.034 |
This | +0.032 |
Ireland | +0.032 |
Neg logit contributions | |
forgot | -0.038 |
felt | -0.032 |
laughed | -0.031 |
got | -0.030 |
knew | -0.030 |
Token | head |
Token ID | 1182 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
forgot | +0.039 |
felt | +0.029 |
got | +0.027 |
laughed | +0.027 |
ppa | +0.027 |
Neg logit contributions | |
purpose | -0.068 |
arl | -0.066 |
sight | -0.061 |
Hawaii | -0.060 |
Earth | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +6.35e+00 |
Pos logit contributions | |
forgot | +0.032 |
felt | +0.026 |
laughed | +0.024 |
ppa | +0.024 |
knew | +0.024 |
Neg logit contributions | |
arl | -0.037 |
purpose | -0.036 |
This | -0.033 |
Hawaii | -0.033 |
time | -0.033 |
Token | said |
Token ID | 531 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
ppa | +0.075 |
forgot | +0.071 |
ppy | +0.070 |
nie | +0.069 |
aky | +0.068 |
Neg logit contributions | |
hello | -0.147 |
purpose | -0.144 |
sight | -0.143 |
This | -0.141 |
something | -0.140 |
Token | said |
Token ID | 531 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
forgot | +0.177 |
ppa | +0.148 |
felt | +0.133 |
my | +0.131 |
struggled | +0.130 |
Neg logit contributions | |
arl | -0.223 |
purpose | -0.214 |
This | -0.208 |
time | -0.199 |
Hawaii | -0.195 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
forgot | +0.038 |
felt | +0.033 |
laughed | +0.031 |
got | +0.030 |
knew | +0.030 |
Neg logit contributions | |
arl | -0.033 |
purpose | -0.033 |
Hawaii | -0.031 |
Ireland | -0.029 |
This | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
purpose | +0.023 |
arl | +0.021 |
outer | +0.020 |
Hawaii | +0.020 |
sight | +0.020 |
Neg logit contributions | |
forgot | -0.026 |
felt | -0.023 |
laughed | -0.022 |
had | -0.022 |
got | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
purpose | +0.020 |
arl | +0.020 |
Ireland | +0.018 |
outer | +0.018 |
Hawaii | +0.018 |
Neg logit contributions | |
forgot | -0.021 |
felt | -0.019 |
my | -0.019 |
had | -0.018 |
knew | -0.018 |
Token | M |
Token ID | 44 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
purpose | +0.017 |
arl | +0.017 |
Hawaii | +0.016 |
sight | +0.015 |
Ireland | +0.015 |
Neg logit contributions | |
forgot | -0.018 |
felt | -0.015 |
laughed | -0.015 |
got | -0.015 |
my | -0.015 |
Token | ira |
Token ID | 8704 |
Feature activation | +1.35e+01 |
Pos logit contributions | |
forgot | +0.053 |
felt | +0.046 |
laughed | +0.044 |
got | +0.043 |
knew | +0.042 |
Neg logit contributions | |
arl | -0.043 |
purpose | -0.042 |
Hawaii | -0.038 |
Ireland | -0.037 |
This | -0.036 |
Token | thought |
Token ID | 1807 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
forgot | +0.212 |
ppa | +0.197 |
any | +0.170 |
my | +0.165 |
anut | +0.153 |
Neg logit contributions | |
time | -0.253 |
hello | -0.248 |
A | -0.247 |
This | -0.246 |
Earth | -0.245 |
Token | about |
Token ID | 546 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
arl | +0.048 |
purpose | +0.047 |
Hawaii | +0.043 |
Ireland | +0.041 |
This | +0.041 |
Neg logit contributions | |
forgot | -0.071 |
felt | -0.062 |
laughed | -0.059 |
got | -0.058 |
knew | -0.057 |
Token | Ben |
Token ID | 3932 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
arl | +0.020 |
purpose | +0.018 |
ay | +0.016 |
Hawaii | +0.016 |
Ireland | +0.015 |
Neg logit contributions | |
forgot | -0.022 |
felt | -0.021 |
laughed | -0.020 |
knew | -0.019 |
had | -0.019 |
Token | 's |
Token ID | 338 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
forgot | +0.081 |
felt | +0.066 |
laughed | +0.064 |
struggled | +0.062 |
ppa | +0.061 |
Neg logit contributions | |
purpose | -0.072 |
arl | -0.071 |
This | -0.067 |
Hawaii | -0.066 |
time | -0.064 |
Token | idea |
Token ID | 2126 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
forgot | +0.046 |
felt | +0.041 |
laughed | +0.040 |
knew | +0.039 |
my | +0.039 |
Neg logit contributions | |
arl | -0.035 |
purpose | -0.032 |
Hawaii | -0.032 |
Ireland | -0.030 |
outer | -0.030 |
Token | Thank |
Token ID | 10449 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
arl | +0.008 |
purpose | +0.008 |
Hawaii | +0.007 |
Ireland | +0.007 |
This | +0.007 |
Neg logit contributions | |
forgot | -0.012 |
felt | -0.011 |
laughed | -0.010 |
knew | -0.010 |
got | -0.010 |
Token | you |
Token ID | 345 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
purpose | +0.084 |
arl | +0.084 |
Hawaii | +0.076 |
Ireland | +0.073 |
outer | +0.073 |
Neg logit contributions | |
forgot | -0.100 |
felt | -0.092 |
laughed | -0.092 |
had | -0.089 |
got | -0.089 |
Token | again |
Token ID | 757 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
purpose | +0.022 |
arl | +0.022 |
Hawaii | +0.020 |
outer | +0.019 |
Ireland | +0.019 |
Neg logit contributions | |
forgot | -0.024 |
felt | -0.021 |
laughed | -0.020 |
my | -0.020 |
got | -0.020 |
Token | ," |
Token ID | 553 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
arl | +0.087 |
purpose | +0.083 |
Hawaii | +0.078 |
Ireland | +0.075 |
This | +0.074 |
Neg logit contributions | |
forgot | -0.099 |
felt | -0.085 |
laughed | -0.081 |
knew | -0.080 |
got | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
forgot | +0.044 |
felt | +0.036 |
ppa | +0.034 |
laughed | +0.034 |
got | +0.034 |
Neg logit contributions | |
purpose | -0.034 |
arl | -0.033 |
Hawaii | -0.032 |
This | -0.032 |
Ireland | -0.031 |
Token | boy |
Token ID | 2933 |
Feature activation | +1.35e+01 |
Pos logit contributions | |
forgot | +0.036 |
felt | +0.029 |
laughed | +0.027 |
got | +0.026 |
knew | +0.026 |
Neg logit contributions | |
arl | -0.050 |
purpose | -0.050 |
Hawaii | -0.046 |
Ireland | -0.045 |
This | -0.045 |
Token | said |
Token ID | 531 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
forgot | +0.232 |
ppy | +0.208 |
ppa | +0.205 |
'd | +0.200 |
eling | +0.199 |
Neg logit contributions | |
A | -0.254 |
The | -0.235 |
This | -0.229 |
hello | -0.226 |
Mom | -0.213 |
Token | . |
Token ID | 13 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
forgot | +0.026 |
felt | +0.022 |
laughed | +0.021 |
knew | +0.020 |
got | +0.020 |
Neg logit contributions | |
arl | -0.019 |
purpose | -0.019 |
This | -0.017 |
Hawaii | -0.017 |
Ireland | -0.016 |
Token | " |
Token ID | 366 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
purpose | +0.053 |
arl | +0.050 |
Hawaii | +0.047 |
outer | +0.047 |
Ireland | +0.045 |
Neg logit contributions | |
forgot | -0.062 |
felt | -0.058 |
knew | -0.054 |
laughed | -0.054 |
got | -0.053 |
Token | You |
Token ID | 1639 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
arl | +0.011 |
purpose | +0.010 |
Hawaii | +0.010 |
Ireland | +0.009 |
This | +0.009 |
Neg logit contributions | |
forgot | -0.013 |
felt | -0.012 |
laughed | -0.011 |
got | -0.011 |
knew | -0.011 |
Token | are |
Token ID | 389 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
forgot | +0.052 |
ppa | +0.049 |
ie | +0.045 |
i | +0.045 |
felt | +0.043 |
Neg logit contributions | |
arl | -0.044 |
purpose | -0.043 |
pleasure | -0.043 |
This | -0.042 |
Earth | -0.041 |
Token | ummy |
Token ID | 13513 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
purpose | +0.057 |
arl | +0.056 |
Hawaii | +0.055 |
outer | +0.054 |
Ireland | +0.053 |
Neg logit contributions | |
forgot | -0.084 |
felt | -0.073 |
got | -0.070 |
knew | -0.070 |
laughed | -0.069 |
Token | , |
Token ID | 11 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
forgot | +0.049 |
felt | +0.042 |
laughed | +0.040 |
got | +0.039 |
knew | +0.039 |
Neg logit contributions | |
arl | -0.041 |
purpose | -0.041 |
Hawaii | -0.038 |
Ireland | -0.036 |
This | -0.036 |
Token | why |
Token ID | 1521 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
arl | +0.037 |
purpose | +0.037 |
Hawaii | +0.033 |
Ireland | +0.032 |
This | +0.031 |
Neg logit contributions | |
forgot | -0.054 |
felt | -0.048 |
laughed | -0.046 |
got | -0.045 |
knew | -0.045 |
Token | is |
Token ID | 318 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
arl | +0.066 |
purpose | +0.066 |
Hawaii | +0.060 |
This | +0.059 |
Ireland | +0.058 |
Neg logit contributions | |
forgot | -0.069 |
felt | -0.059 |
laughed | -0.056 |
ppa | -0.055 |
knew | -0.054 |
Token | it |
Token ID | 340 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
arl | +0.070 |
purpose | +0.064 |
ay | +0.057 |
Hawaii | +0.057 |
ags | +0.055 |
Neg logit contributions | |
forgot | -0.080 |
felt | -0.071 |
laughed | -0.071 |
got | -0.069 |
had | -0.068 |
Token | hanging |
Token ID | 10938 |
Feature activation | -9.04e+00 |
Pos logit contributions | |
arl | +0.031 |
purpose | +0.031 |
Hawaii | +0.029 |
Ireland | +0.028 |
This | +0.027 |
Neg logit contributions | |
forgot | -0.032 |
felt | -0.027 |
laughed | -0.026 |
got | -0.025 |
knew | -0.025 |
Token | there |
Token ID | 612 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
arl | +0.216 |
purpose | +0.211 |
pleasure | +0.208 |
common | +0.197 |
ocol | +0.197 |
Neg logit contributions | |
my | -0.153 |
felt | -0.137 |
got | -0.137 |
had | -0.128 |
started | -0.125 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
arl | +0.084 |
purpose | +0.073 |
Hawaii | +0.070 |
Ireland | +0.069 |
outer | +0.067 |
Neg logit contributions | |
forgot | -0.103 |
felt | -0.097 |
my | -0.096 |
got | -0.092 |
's | -0.089 |
Token |
|
Token ID | 198 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
arl | +0.054 |
purpose | +0.053 |
Hawaii | +0.049 |
Ireland | +0.046 |
outer | +0.045 |
Neg logit contributions | |
forgot | -0.082 |
felt | -0.073 |
laughed | -0.070 |
knew | -0.068 |
got | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
arl | +0.089 |
purpose | +0.087 |
Hawaii | +0.081 |
Ireland | +0.079 |
outer | +0.078 |
Neg logit contributions | |
forgot | -0.106 |
felt | -0.095 |
knew | -0.091 |
laughed | -0.090 |
got | -0.089 |
Token | M |
Token ID | 44 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
arl | +0.082 |
purpose | +0.081 |
Hawaii | +0.075 |
This | +0.073 |
Ireland | +0.073 |
Neg logit contributions | |
forgot | -0.081 |
felt | -0.069 |
laughed | -0.065 |
got | -0.063 |
knew | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
forgot | +0.038 |
felt | +0.032 |
laughed | +0.031 |
got | +0.030 |
knew | +0.030 |
Neg logit contributions | |
arl | -0.030 |
purpose | -0.029 |
Hawaii | -0.027 |
Ireland | -0.026 |
This | -0.026 |
Token | " |
Token ID | 366 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
forgot | +0.031 |
felt | +0.028 |
knew | +0.027 |
laughed | +0.027 |
got | +0.027 |
Neg logit contributions | |
purpose | -0.023 |
arl | -0.022 |
Hawaii | -0.020 |
Ireland | -0.019 |
outer | -0.019 |
Token | Can |
Token ID | 6090 |
Feature activation | -4.58e+00 |
Pos logit contributions | |
arl | +0.036 |
purpose | +0.035 |
Hawaii | +0.033 |
Ireland | +0.030 |
Can | +0.030 |
Neg logit contributions | |
forgot | -0.089 |
felt | -0.079 |
laughed | -0.077 |
knew | -0.074 |
got | -0.074 |
Token | I |
Token ID | 314 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
arl | +0.070 |
purpose | +0.066 |
Hawaii | +0.058 |
Ireland | +0.055 |
outer | +0.054 |
Neg logit contributions | |
forgot | -0.137 |
felt | -0.122 |
laughed | -0.118 |
got | -0.117 |
knew | -0.115 |
Token | go |
Token ID | 467 |
Feature activation | -6.14e+00 |
Pos logit contributions | |
arl | +0.056 |
purpose | +0.056 |
Hawaii | +0.052 |
Ireland | +0.050 |
This | +0.050 |
Neg logit contributions | |
forgot | -0.054 |
felt | -0.046 |
laughed | -0.044 |
got | -0.043 |
knew | -0.043 |
Token | sailing |
Token ID | 28499 |
Feature activation | -8.93e+00 |
Pos logit contributions | |
purpose | +0.127 |
arl | +0.125 |
Hawaii | +0.110 |
Ireland | +0.108 |
This | +0.106 |
Neg logit contributions | |
forgot | -0.146 |
felt | -0.128 |
got | -0.122 |
laughed | -0.121 |
my | -0.120 |
Token | on |
Token ID | 319 |
Feature activation | -6.87e+00 |
Pos logit contributions | |
arl | +0.236 |
purpose | +0.222 |
umm | +0.194 |
ocol | +0.194 |
tornado | +0.193 |
Neg logit contributions | |
forgot | -0.151 |
felt | -0.146 |
got | -0.141 |
my | -0.138 |
laughed | -0.134 |
Token | a |
Token ID | 257 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
arl | +0.207 |
umm | +0.174 |
purpose | +0.167 |
ay | +0.161 |
Sorry | +0.160 |
Neg logit contributions | |
felt | -0.093 |
's | -0.091 |
sc | -0.088 |
got | -0.087 |
tried | -0.086 |
Token | yacht |
Token ID | 43207 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
arl | +0.070 |
purpose | +0.067 |
Hawaii | +0.062 |
This | +0.061 |
Ireland | +0.060 |
Neg logit contributions | |
forgot | -0.073 |
felt | -0.063 |
laughed | -0.059 |
got | -0.059 |
knew | -0.058 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
arl | +0.067 |
purpose | +0.067 |
Hawaii | +0.059 |
Ireland | +0.056 |
This | +0.055 |
Neg logit contributions | |
forgot | -0.114 |
felt | -0.102 |
got | -0.098 |
laughed | -0.097 |
my | -0.097 |
Token |
|
Token ID | 198 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
arl | +0.054 |
purpose | +0.054 |
Hawaii | +0.049 |
This | +0.047 |
Ireland | +0.047 |
Neg logit contributions | |
forgot | -0.078 |
felt | -0.068 |
laughed | -0.065 |
got | -0.063 |
knew | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +8.20e-01 |
Pos logit contributions | |
forgot | +0.036 |
felt | +0.031 |
laughed | +0.030 |
my | +0.029 |
got | +0.029 |
Neg logit contributions | |
arl | -0.034 |
purpose | -0.033 |
Hawaii | -0.031 |
Ireland | -0.029 |
outer | -0.029 |
Token | " |
Token ID | 366 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
forgot | +0.020 |
felt | +0.018 |
knew | +0.018 |
got | +0.017 |
took | +0.017 |
Neg logit contributions | |
purpose | -0.017 |
arl | -0.016 |
Hawaii | -0.015 |
outer | -0.014 |
Ireland | -0.014 |
Token | Can |
Token ID | 6090 |
Feature activation | -4.87e+00 |
Pos logit contributions | |
arl | +0.040 |
purpose | +0.040 |
Hawaii | +0.036 |
Ireland | +0.034 |
Can | +0.034 |
Neg logit contributions | |
forgot | -0.083 |
felt | -0.073 |
laughed | -0.072 |
knew | -0.068 |
struggled | -0.068 |
Token | we |
Token ID | 356 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
arl | +0.083 |
purpose | +0.077 |
Hawaii | +0.070 |
outer | +0.067 |
Ireland | +0.067 |
Neg logit contributions | |
forgot | -0.139 |
felt | -0.123 |
got | -0.118 |
laughed | -0.117 |
had | -0.116 |
Token | go |
Token ID | 467 |
Feature activation | -7.11e+00 |
Pos logit contributions | |
arl | +0.063 |
purpose | +0.062 |
Hawaii | +0.058 |
Ireland | +0.057 |
outer | +0.056 |
Neg logit contributions | |
forgot | -0.051 |
felt | -0.042 |
got | -0.041 |
laughed | -0.039 |
took | -0.038 |
Token | fishing |
Token ID | 12478 |
Feature activation | -8.91e+00 |
Pos logit contributions | |
purpose | +0.132 |
arl | +0.131 |
Hawaii | +0.111 |
Ireland | +0.110 |
This | +0.107 |
Neg logit contributions | |
forgot | -0.185 |
felt | -0.165 |
got | -0.159 |
laughed | -0.156 |
my | -0.156 |
Token | tomorrow |
Token ID | 9439 |
Feature activation | -6.04e+00 |
Pos logit contributions | |
arl | +0.211 |
purpose | +0.194 |
Hawaii | +0.178 |
outer | +0.177 |
This | +0.174 |
Neg logit contributions | |
forgot | -0.186 |
felt | -0.169 |
got | -0.164 |
laughed | -0.158 |
took | -0.153 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
arl | +0.126 |
purpose | +0.119 |
outer | +0.110 |
ocol | +0.109 |
Ireland | +0.109 |
Neg logit contributions | |
forgot | -0.144 |
felt | -0.128 |
my | -0.125 |
got | -0.125 |
had | -0.117 |
Token |
|
Token ID | 198 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
arl | +0.055 |
purpose | +0.053 |
Hawaii | +0.049 |
Ireland | +0.047 |
outer | +0.046 |
Neg logit contributions | |
forgot | -0.076 |
felt | -0.067 |
laughed | -0.064 |
got | -0.063 |
knew | -0.062 |
Token |
|
Token ID | 198 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
arl | +0.081 |
purpose | +0.078 |
Hawaii | +0.072 |
Ireland | +0.071 |
outer | +0.069 |
Neg logit contributions | |
forgot | -0.103 |
felt | -0.093 |
my | -0.089 |
got | -0.089 |
knew | -0.087 |
Token | His |
Token ID | 6653 |
Feature activation | +8.02e-01 |
Pos logit contributions | |
arl | +0.078 |
purpose | +0.077 |
Hawaii | +0.072 |
Ireland | +0.069 |
This | +0.069 |
Neg logit contributions | |
forgot | -0.079 |
felt | -0.067 |
laughed | -0.064 |
got | -0.062 |
knew | -0.062 |
Token | , |
Token ID | 11 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
forgot | +0.035 |
felt | +0.028 |
laughed | +0.026 |
got | +0.025 |
knew | +0.025 |
Neg logit contributions | |
arl | -0.049 |
purpose | -0.048 |
Hawaii | -0.045 |
Ireland | -0.044 |
This | -0.044 |
Token | " |
Token ID | 366 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
forgot | +0.026 |
felt | +0.024 |
knew | +0.023 |
laughed | +0.023 |
got | +0.023 |
Neg logit contributions | |
purpose | -0.019 |
arl | -0.018 |
Hawaii | -0.017 |
Ireland | -0.016 |
outer | -0.016 |
Token | How |
Token ID | 2437 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
arl | +0.035 |
purpose | +0.034 |
Hawaii | +0.031 |
Ireland | +0.029 |
Can | +0.028 |
Neg logit contributions | |
forgot | -0.081 |
felt | -0.071 |
laughed | -0.069 |
knew | -0.067 |
got | -0.066 |
Token | can |
Token ID | 460 |
Feature activation | -6.17e+00 |
Pos logit contributions | |
arl | +0.069 |
purpose | +0.069 |
Hawaii | +0.064 |
sight | +0.061 |
Ireland | +0.060 |
Neg logit contributions | |
forgot | -0.064 |
felt | -0.056 |
knew | -0.053 |
got | -0.053 |
had | -0.052 |
Token | I |
Token ID | 314 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
arl | +0.094 |
purpose | +0.086 |
Hawaii | +0.078 |
Ireland | +0.074 |
sight | +0.070 |
Neg logit contributions | |
forgot | -0.184 |
felt | -0.166 |
laughed | -0.160 |
got | -0.157 |
knew | -0.155 |
Token | build |
Token ID | 1382 |
Feature activation | -8.78e+00 |
Pos logit contributions | |
arl | +0.065 |
purpose | +0.064 |
Hawaii | +0.060 |
Ireland | +0.058 |
This | +0.057 |
Neg logit contributions | |
forgot | -0.063 |
felt | -0.054 |
laughed | -0.050 |
got | -0.050 |
knew | -0.049 |
Token | an |
Token ID | 281 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
arl | +0.233 |
purpose | +0.175 |
ifying | +0.172 |
pleasure | +0.172 |
Hawaii | +0.169 |
Neg logit contributions | |
felt | -0.165 |
got | -0.163 |
forgot | -0.162 |
's | -0.159 |
my | -0.157 |
Token | angel |
Token ID | 18304 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
purpose | +0.065 |
arl | +0.063 |
Hawaii | +0.059 |
Ireland | +0.057 |
This | +0.056 |
Neg logit contributions | |
forgot | -0.103 |
felt | -0.089 |
laughed | -0.086 |
got | -0.082 |
knew | -0.082 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
arl | +0.037 |
purpose | +0.036 |
Hawaii | +0.032 |
Ireland | +0.030 |
This | +0.030 |
Neg logit contributions | |
forgot | -0.073 |
felt | -0.065 |
laughed | -0.062 |
got | -0.061 |
knew | -0.061 |
Token |
|
Token ID | 198 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
arl | +0.055 |
purpose | +0.054 |
Hawaii | +0.050 |
Ireland | +0.047 |
This | +0.047 |
Neg logit contributions | |
forgot | -0.081 |
felt | -0.071 |
laughed | -0.068 |
got | -0.066 |
knew | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
arl | +0.093 |
purpose | +0.088 |
Hawaii | +0.082 |
Ireland | +0.081 |
outer | +0.078 |
Neg logit contributions | |
forgot | -0.103 |
felt | -0.095 |
my | -0.091 |
got | -0.090 |
knew | -0.090 |
Token | , |
Token ID | 11 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
arl | +0.035 |
purpose | +0.034 |
Hawaii | +0.031 |
Ireland | +0.030 |
This | +0.030 |
Neg logit contributions | |
forgot | -0.041 |
felt | -0.036 |
laughed | -0.034 |
got | -0.033 |
knew | -0.033 |
Token | it |
Token ID | 340 |
Feature activation | +6.64e+00 |
Pos logit contributions | |
arl | +0.003 |
purpose | +0.003 |
Hawaii | +0.003 |
Ireland | +0.003 |
This | +0.003 |
Neg logit contributions | |
forgot | -0.003 |
felt | -0.003 |
laughed | -0.003 |
got | -0.003 |
knew | -0.002 |
Token | kept |
Token ID | 4030 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
ppa | +0.093 |
my | +0.090 |
nie | +0.086 |
any | +0.086 |
ilia | +0.084 |
Neg logit contributions | |
sight | -0.143 |
time | -0.142 |
purpose | -0.138 |
brightly | -0.131 |
Ireland | -0.130 |
Token | on |
Token ID | 319 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
arl | +0.056 |
purpose | +0.054 |
Hawaii | +0.050 |
This | +0.049 |
Ireland | +0.048 |
Neg logit contributions | |
forgot | -0.060 |
felt | -0.052 |
laughed | -0.050 |
got | -0.048 |
knew | -0.048 |
Token | flying |
Token ID | 7348 |
Feature activation | -5.03e+00 |
Pos logit contributions | |
arl | +0.075 |
purpose | +0.072 |
Hawaii | +0.068 |
Ireland | +0.065 |
This | +0.065 |
Neg logit contributions | |
forgot | -0.083 |
felt | -0.073 |
laughed | -0.069 |
knew | -0.067 |
got | -0.067 |
Token | err |
Token ID | 11454 |
Feature activation | -8.72e+00 |
Pos logit contributions | |
arl | +0.117 |
purpose | +0.115 |
Hawaii | +0.107 |
Ireland | +0.103 |
This | +0.103 |
Neg logit contributions | |
forgot | -0.105 |
felt | -0.090 |
laughed | -0.085 |
got | -0.083 |
knew | -0.082 |
Token | atically |
Token ID | 4142 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
arl | +0.134 |
purpose | +0.134 |
Hawaii | +0.120 |
This | +0.114 |
Ireland | +0.114 |
Neg logit contributions | |
forgot | -0.244 |
felt | -0.215 |
laughed | -0.208 |
got | -0.204 |
knew | -0.204 |
Token | with |
Token ID | 351 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
purpose | +0.033 |
arl | +0.033 |
Hawaii | +0.030 |
This | +0.029 |
Ireland | +0.029 |
Neg logit contributions | |
forgot | -0.034 |
felt | -0.028 |
laughed | -0.027 |
knew | -0.026 |
got | -0.026 |
Token | inexplicable |
Token ID | 49602 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
arl | +0.030 |
purpose | +0.028 |
Hawaii | +0.026 |
This | +0.025 |
Ireland | +0.025 |
Neg logit contributions | |
forgot | -0.041 |
felt | -0.037 |
laughed | -0.036 |
knew | -0.035 |
got | -0.035 |
Token | speed |
Token ID | 2866 |
Feature activation | -8.40e-01 |
Pos logit contributions | |
forgot | +0.005 |
felt | +0.004 |
laughed | +0.004 |
got | +0.004 |
knew | +0.004 |
Neg logit contributions | |
arl | -0.004 |
purpose | -0.004 |
Hawaii | -0.004 |
Ireland | -0.003 |
This | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
arl | +0.017 |
purpose | +0.017 |
Hawaii | +0.015 |
This | +0.015 |
Ireland | +0.014 |
Neg logit contributions | |
forgot | -0.020 |
felt | -0.017 |
laughed | -0.016 |
ppa | -0.016 |
knew | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
forgot | +0.023 |
felt | +0.016 |
laughed | +0.015 |
knew | +0.013 |
got | +0.013 |
Neg logit contributions | |
arl | -0.062 |
purpose | -0.061 |
Hawaii | -0.058 |
This | -0.057 |
Ireland | -0.057 |
Token | " |
Token ID | 366 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
forgot | +0.035 |
felt | +0.032 |
knew | +0.031 |
laughed | +0.031 |
my | +0.030 |
Neg logit contributions | |
purpose | -0.027 |
arl | -0.025 |
Hawaii | -0.024 |
Ireland | -0.022 |
outer | -0.022 |
Token | Can |
Token ID | 6090 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
arl | +0.042 |
purpose | +0.041 |
Hawaii | +0.037 |
Ireland | +0.035 |
This | +0.035 |
Neg logit contributions | |
forgot | -0.079 |
felt | -0.069 |
laughed | -0.067 |
knew | -0.065 |
got | -0.064 |
Token | I |
Token ID | 314 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
arl | +0.080 |
purpose | +0.073 |
Hawaii | +0.067 |
outer | +0.064 |
sight | +0.063 |
Neg logit contributions | |
forgot | -0.129 |
felt | -0.116 |
got | -0.111 |
laughed | -0.110 |
had | -0.109 |
Token | start |
Token ID | 923 |
Feature activation | -6.67e+00 |
Pos logit contributions | |
arl | +0.057 |
purpose | +0.057 |
Hawaii | +0.053 |
This | +0.051 |
Ireland | +0.051 |
Neg logit contributions | |
forgot | -0.051 |
felt | -0.043 |
laughed | -0.040 |
knew | -0.039 |
got | -0.039 |
Token | drawing |
Token ID | 8263 |
Feature activation | -8.71e+00 |
Pos logit contributions | |
arl | +0.165 |
purpose | +0.144 |
Hawaii | +0.137 |
pleasure | +0.133 |
tornado | +0.130 |
Neg logit contributions | |
forgot | -0.133 |
my | -0.126 |
felt | -0.122 |
got | -0.118 |
laughed | -0.118 |
Token | now |
Token ID | 783 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
arl | +0.234 |
purpose | +0.201 |
pleasure | +0.194 |
Hawaii | +0.194 |
ifying | +0.189 |
Neg logit contributions | |
my | -0.150 |
felt | -0.144 |
's | -0.140 |
got | -0.136 |
laughed | -0.136 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
arl | +0.077 |
purpose | +0.075 |
Hawaii | +0.074 |
Ireland | +0.071 |
outer | +0.071 |
Neg logit contributions | |
forgot | -0.088 |
felt | -0.078 |
my | -0.075 |
got | -0.074 |
took | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
arl | +0.053 |
purpose | +0.052 |
Hawaii | +0.048 |
Ireland | +0.046 |
outer | +0.046 |
Neg logit contributions | |
forgot | -0.067 |
felt | -0.059 |
laughed | -0.056 |
got | -0.055 |
knew | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
arl | +0.074 |
purpose | +0.073 |
Hawaii | +0.068 |
Ireland | +0.067 |
outer | +0.066 |
Neg logit contributions | |
forgot | -0.100 |
felt | -0.090 |
my | -0.087 |
got | -0.086 |
knew | -0.086 |
Token | Her |
Token ID | 9360 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
arl | +0.065 |
purpose | +0.065 |
Hawaii | +0.060 |
Ireland | +0.057 |
This | +0.057 |
Neg logit contributions | |
forgot | -0.079 |
felt | -0.068 |
laughed | -0.065 |
got | -0.063 |
knew | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
arl | +0.094 |
purpose | +0.092 |
Hawaii | +0.084 |
This | +0.082 |
outer | +0.082 |
Neg logit contributions | |
forgot | -0.091 |
felt | -0.078 |
laughed | -0.075 |
got | -0.073 |
knew | -0.072 |
Token | Do |
Token ID | 2141 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
purpose | +0.077 |
arl | +0.066 |
sight | +0.064 |
time | +0.058 |
igsaw | +0.058 |
Neg logit contributions | |
forgot | -0.112 |
had | -0.104 |
my | -0.103 |
's | -0.103 |
took | -0.102 |
Token | you |
Token ID | 345 |
Feature activation | -7.25e-01 |
Pos logit contributions | |
arl | +0.050 |
purpose | +0.048 |
Hawaii | +0.043 |
Ireland | +0.040 |
outer | +0.040 |
Neg logit contributions | |
forgot | -0.094 |
felt | -0.084 |
laughed | -0.081 |
got | -0.080 |
knew | -0.079 |
Token | want |
Token ID | 765 |
Feature activation | -6.75e+00 |
Pos logit contributions | |
arl | +0.014 |
purpose | +0.013 |
Hawaii | +0.013 |
Ireland | +0.013 |
This | +0.012 |
Neg logit contributions | |
forgot | -0.017 |
felt | -0.015 |
ppa | -0.014 |
struggled | -0.014 |
laughed | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -5.17e+00 |
Pos logit contributions | |
arl | +0.174 |
purpose | +0.142 |
pleasure | +0.139 |
umm | +0.139 |
ires | +0.134 |
Neg logit contributions | |
got | -0.130 |
forgot | -0.129 |
said | -0.125 |
felt | -0.124 |
's | -0.124 |
Token | read |
Token ID | 1100 |
Feature activation | -8.68e+00 |
Pos logit contributions | |
arl | +0.137 |
purpose | +0.129 |
ires | +0.119 |
time | +0.117 |
Hawaii | +0.117 |
Neg logit contributions | |
forgot | -0.093 |
felt | -0.078 |
got | -0.078 |
laughed | -0.073 |
my | -0.073 |
Token | it |
Token ID | 340 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
arl | +0.217 |
umm | +0.185 |
purpose | +0.184 |
pleasure | +0.180 |
igsaw | +0.163 |
Neg logit contributions | |
's | -0.164 |
forgot | -0.163 |
my | -0.162 |
felt | -0.161 |
got | -0.161 |
Token | with |
Token ID | 351 |
Feature activation | -4.83e+00 |
Pos logit contributions | |
arl | +0.105 |
purpose | +0.097 |
pleasure | +0.088 |
Hawaii | +0.087 |
outer | +0.086 |
Neg logit contributions | |
forgot | -0.114 |
felt | -0.108 |
my | -0.106 |
got | -0.104 |
laughed | -0.100 |
Token | me |
Token ID | 502 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
arl | +0.123 |
purpose | +0.107 |
outer | +0.102 |
Hawaii | +0.101 |
Today | +0.099 |
Neg logit contributions | |
forgot | -0.093 |
laughed | -0.081 |
felt | -0.081 |
got | -0.079 |
my | -0.077 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
arl | +0.057 |
purpose | +0.055 |
Hawaii | +0.047 |
sight | +0.046 |
umm | +0.046 |
Neg logit contributions | |
forgot | -0.071 |
felt | -0.066 |
my | -0.066 |
got | -0.064 |
laughed | -0.062 |
Token |
|
Token ID | 198 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
arl | +0.028 |
purpose | +0.027 |
Hawaii | +0.025 |
Ireland | +0.024 |
outer | +0.023 |
Neg logit contributions | |
forgot | -0.046 |
felt | -0.041 |
laughed | -0.039 |
got | -0.038 |
knew | -0.038 |
Token | Me |
Token ID | 5308 |
Feature activation | -8.50e-01 |
Pos logit contributions | |
arl | +0.044 |
purpose | +0.044 |
Hawaii | +0.039 |
This | +0.038 |
Can | +0.037 |
Neg logit contributions | |
forgot | -0.073 |
felt | -0.064 |
laughed | -0.061 |
got | -0.060 |
knew | -0.060 |
Token | too |
Token ID | 1165 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
arl | +0.022 |
purpose | +0.022 |
Hawaii | +0.021 |
outer | +0.020 |
Ireland | +0.019 |
Neg logit contributions | |
forgot | -0.016 |
felt | -0.014 |
laughed | -0.014 |
got | -0.013 |
my | -0.013 |
Token | ! |
Token ID | 0 |
Feature activation | -4.56e+00 |
Pos logit contributions | |
arl | +0.058 |
Hawaii | +0.054 |
purpose | +0.054 |
Ireland | +0.051 |
outer | +0.051 |
Neg logit contributions | |
forgot | -0.058 |
felt | -0.049 |
laughed | -0.048 |
knew | -0.047 |
got | -0.047 |
Token | Want |
Token ID | 16168 |
Feature activation | -7.04e+00 |
Pos logit contributions | |
purpose | +0.085 |
arl | +0.076 |
sight | +0.073 |
outer | +0.072 |
Hawaii | +0.071 |
Neg logit contributions | |
forgot | -0.115 |
my | -0.103 |
felt | -0.102 |
laughed | -0.101 |
had | -0.101 |
Token | to |
Token ID | 284 |
Feature activation | -4.66e+00 |
Pos logit contributions | |
arl | +0.186 |
purpose | +0.144 |
ocol | +0.143 |
ires | +0.141 |
ags | +0.140 |
Neg logit contributions | |
's | -0.144 |
forgot | -0.142 |
felt | -0.139 |
got | -0.138 |
my | -0.137 |
Token | play |
Token ID | 711 |
Feature activation | -8.65e+00 |
Pos logit contributions | |
arl | +0.122 |
purpose | +0.116 |
Ireland | +0.109 |
ires | +0.108 |
Hawaii | +0.107 |
Neg logit contributions | |
forgot | -0.085 |
felt | -0.072 |
got | -0.070 |
my | -0.070 |
laughed | -0.067 |
Token | hide |
Token ID | 7808 |
Feature activation | -8.01e+00 |
Pos logit contributions | |
arl | +0.244 |
purpose | +0.217 |
ocol | +0.200 |
Hawaii | +0.198 |
pleasure | +0.198 |
Neg logit contributions | |
forgot | -0.125 |
got | -0.124 |
's | -0.123 |
pretended | -0.121 |
felt | -0.120 |
Token | and |
Token ID | 290 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
arl | +0.185 |
purpose | +0.181 |
Hawaii | +0.166 |
This | +0.165 |
pleasure | +0.162 |
Neg logit contributions | |
forgot | -0.162 |
felt | -0.142 |
laughed | -0.136 |
got | -0.135 |
my | -0.135 |
Token | seek |
Token ID | 5380 |
Feature activation | -5.71e+00 |
Pos logit contributions | |
arl | +0.036 |
purpose | +0.035 |
Hawaii | +0.033 |
Ireland | +0.032 |
This | +0.032 |
Neg logit contributions | |
forgot | -0.034 |
felt | -0.029 |
laughed | -0.027 |
got | -0.026 |
knew | -0.026 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
arl | +0.122 |
purpose | +0.116 |
Hawaii | +0.104 |
This | +0.099 |
Ireland | +0.099 |
Neg logit contributions | |
forgot | -0.128 |
felt | -0.117 |
my | -0.117 |
laughed | -0.113 |
got | -0.112 |
Token | |
Token ID | 220 |
Feature activation | -4.61e+00 |
Pos logit contributions | |
This | +0.063 |
arl | +0.062 |
purpose | +0.062 |
Hawaii | +0.057 |
Sometimes | +0.055 |
Neg logit contributions | |
forgot | -0.084 |
ppa | -0.069 |
felt | -0.069 |
laughed | -0.066 |
knew | -0.066 |
Token | you |
Token ID | 345 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
arl | +0.056 |
purpose | +0.055 |
Hawaii | +0.048 |
Ireland | +0.045 |
This | +0.045 |
Neg logit contributions | |
forgot | -0.119 |
felt | -0.106 |
laughed | -0.102 |
got | -0.100 |
knew | -0.099 |
Token | sure |
Token ID | 1654 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
arl | +0.012 |
purpose | +0.012 |
Hawaii | +0.011 |
Ireland | +0.011 |
This | +0.011 |
Neg logit contributions | |
forgot | -0.013 |
felt | -0.011 |
laughed | -0.010 |
knew | -0.010 |
got | -0.010 |
Token | you |
Token ID | 345 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
arl | +0.059 |
purpose | +0.056 |
Hawaii | +0.049 |
This | +0.048 |
Ireland | +0.048 |
Neg logit contributions | |
forgot | -0.078 |
felt | -0.068 |
laughed | -0.065 |
knew | -0.064 |
got | -0.064 |
Token | want |
Token ID | 765 |
Feature activation | -6.23e+00 |
Pos logit contributions | |
forgot | +0.021 |
ppa | +0.019 |
felt | +0.017 |
ie | +0.017 |
rushed | +0.016 |
Neg logit contributions | |
Ireland | -0.025 |
arl | -0.025 |
purpose | -0.025 |
This | -0.025 |
common | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
arl | +0.166 |
umm | +0.129 |
purpose | +0.126 |
ocol | +0.122 |
Hawaii | +0.121 |
Neg logit contributions | |
forgot | -0.121 |
's | -0.120 |
got | -0.117 |
felt | -0.116 |
laughed | -0.112 |
Token | read |
Token ID | 1100 |
Feature activation | -8.63e+00 |
Pos logit contributions | |
arl | +0.148 |
purpose | +0.137 |
ires | +0.131 |
Ireland | +0.131 |
Hawaii | +0.131 |
Neg logit contributions | |
forgot | -0.055 |
felt | -0.041 |
got | -0.040 |
my | -0.039 |
laughed | -0.037 |
Token | this |
Token ID | 428 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
arl | +0.228 |
umm | +0.189 |
purpose | +0.180 |
pleasure | +0.171 |
uttering | +0.170 |
Neg logit contributions | |
's | -0.161 |
my | -0.154 |
forgot | -0.154 |
felt | -0.153 |
got | -0.150 |
Token | book |
Token ID | 1492 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
arl | +0.050 |
purpose | +0.046 |
Hawaii | +0.042 |
yourselves | +0.041 |
ocol | +0.041 |
Neg logit contributions | |
forgot | -0.039 |
felt | -0.035 |
my | -0.035 |
got | -0.034 |
knew | -0.034 |
Token | ? |
Token ID | 30 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
arl | +0.102 |
purpose | +0.097 |
Hawaii | +0.089 |
outer | +0.085 |
ocol | +0.084 |
Neg logit contributions | |
forgot | -0.128 |
felt | -0.115 |
my | -0.111 |
got | -0.111 |
had | -0.108 |
Token | It |
Token ID | 632 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
purpose | +0.062 |
arl | +0.059 |
sight | +0.052 |
Hawaii | +0.046 |
time | +0.046 |
Neg logit contributions | |
forgot | -0.104 |
felt | -0.096 |
my | -0.096 |
's | -0.096 |
got | -0.095 |
Token | might |
Token ID | 1244 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
forgot | +0.028 |
ppa | +0.024 |
felt | +0.022 |
laughed | +0.021 |
struggled | +0.021 |
Neg logit contributions | |
This | -0.029 |
purpose | -0.029 |
Earth | -0.028 |
Hawaii | -0.028 |
Ireland | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
arl | +0.065 |
purpose | +0.064 |
Hawaii | +0.059 |
outer | +0.056 |
Ireland | +0.056 |
Neg logit contributions | |
forgot | -0.092 |
felt | -0.083 |
laughed | -0.078 |
got | -0.078 |
knew | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
arl | +0.061 |
purpose | +0.061 |
Hawaii | +0.056 |
Ireland | +0.053 |
outer | +0.052 |
Neg logit contributions | |
forgot | -0.097 |
felt | -0.086 |
laughed | -0.082 |
got | -0.081 |
knew | -0.081 |
Token | " |
Token ID | 1 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
arl | +0.048 |
purpose | +0.047 |
Hawaii | +0.042 |
This | +0.041 |
time | +0.041 |
Neg logit contributions | |
forgot | -0.084 |
felt | -0.073 |
laughed | -0.071 |
ppa | -0.070 |
knew | -0.068 |
Token | Can |
Token ID | 6090 |
Feature activation | -5.23e+00 |
Pos logit contributions | |
arl | +0.024 |
purpose | +0.024 |
Hawaii | +0.022 |
outer | +0.021 |
Ireland | +0.021 |
Neg logit contributions | |
forgot | -0.027 |
felt | -0.023 |
laughed | -0.022 |
knew | -0.022 |
decided | -0.022 |
Token | we |
Token ID | 356 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
arl | +0.108 |
purpose | +0.102 |
Hawaii | +0.092 |
outer | +0.091 |
sight | +0.089 |
Neg logit contributions | |
forgot | -0.131 |
felt | -0.116 |
had | -0.110 |
got | -0.110 |
laughed | -0.110 |
Token | drink |
Token ID | 4144 |
Feature activation | -8.61e+00 |
Pos logit contributions | |
arl | +0.054 |
purpose | +0.054 |
Hawaii | +0.050 |
This | +0.049 |
Ireland | +0.048 |
Neg logit contributions | |
forgot | -0.048 |
felt | -0.041 |
laughed | -0.038 |
knew | -0.037 |
got | -0.036 |
Token | it |
Token ID | 340 |
Feature activation | -4.90e+00 |
Pos logit contributions | |
arl | +0.198 |
purpose | +0.177 |
Hawaii | +0.171 |
outer | +0.167 |
Ireland | +0.165 |
Neg logit contributions | |
forgot | -0.176 |
felt | -0.173 |
got | -0.171 |
my | -0.166 |
took | -0.161 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
arl | +0.106 |
purpose | +0.097 |
Hawaii | +0.095 |
Ireland | +0.089 |
outer | +0.088 |
Neg logit contributions | |
forgot | -0.112 |
felt | -0.105 |
got | -0.101 |
my | -0.100 |
started | -0.095 |
Token | Mia |
Token ID | 32189 |
Feature activation | +9.54e+00 |
Pos logit contributions | |
purpose | +0.027 |
arl | +0.026 |
This | +0.025 |
Hawaii | +0.024 |
Ireland | +0.024 |
Neg logit contributions | |
forgot | -0.051 |
ppa | -0.045 |
felt | -0.043 |
laughed | -0.042 |
my | -0.042 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
forgot | +0.158 |
ppa | +0.108 |
managed | +0.105 |
felt | +0.099 |
wouldn | +0.097 |
Neg logit contributions | |
purpose | -0.226 |
This | -0.225 |
Today | -0.219 |
Can | -0.219 |
Spanish | -0.213 |
Token | . |
Token ID | 13 |
Feature activation | -4.04e+00 |
Pos logit contributions | |
arl | +0.035 |
purpose | +0.035 |
Hawaii | +0.032 |
This | +0.031 |
Ireland | +0.031 |
Neg logit contributions | |
forgot | -0.043 |
felt | -0.036 |
laughed | -0.034 |
knew | -0.034 |
got | -0.034 |
Token | How |
Token ID | 2437 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
arl | +0.042 |
purpose | +0.041 |
Hawaii | +0.036 |
Ireland | +0.034 |
This | +0.034 |
Neg logit contributions | |
forgot | -0.081 |
felt | -0.072 |
laughed | -0.069 |
got | -0.068 |
knew | -0.068 |
Token | do |
Token ID | 466 |
Feature activation | -7.14e+00 |
Pos logit contributions | |
arl | +0.060 |
purpose | +0.059 |
Hawaii | +0.053 |
sight | +0.052 |
outer | +0.051 |
Neg logit contributions | |
forgot | -0.053 |
felt | -0.046 |
got | -0.043 |
knew | -0.043 |
had | -0.043 |
Token | I |
Token ID | 314 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
arl | +0.144 |
purpose | +0.124 |
sight | +0.107 |
Hawaii | +0.107 |
ags | +0.106 |
Neg logit contributions | |
forgot | -0.184 |
felt | -0.164 |
got | -0.161 |
laughed | -0.158 |
had | -0.157 |
Token | manage |
Token ID | 6687 |
Feature activation | -6.45e+00 |
Pos logit contributions | |
purpose | +0.017 |
arl | +0.017 |
Hawaii | +0.016 |
This | +0.016 |
Ireland | +0.016 |
Neg logit contributions | |
forgot | -0.017 |
felt | -0.014 |
laughed | -0.014 |
ppa | -0.013 |
my | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
arl | +0.145 |
purpose | +0.119 |
umm | +0.116 |
Hawaii | +0.115 |
Sorry | +0.113 |
Neg logit contributions | |
forgot | -0.139 |
felt | -0.130 |
got | -0.128 |
my | -0.127 |
had | -0.123 |
Token | read |
Token ID | 1100 |
Feature activation | -8.57e+00 |
Pos logit contributions | |
arl | +0.089 |
purpose | +0.084 |
Hawaii | +0.080 |
Ireland | +0.078 |
This | +0.077 |
Neg logit contributions | |
forgot | -0.068 |
felt | -0.057 |
laughed | -0.053 |
got | -0.053 |
knew | -0.052 |
Token | all |
Token ID | 477 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
arl | +0.212 |
purpose | +0.172 |
pleasure | +0.159 |
umm | +0.157 |
uttering | +0.155 |
Neg logit contributions | |
forgot | -0.166 |
my | -0.165 |
felt | -0.163 |
's | -0.161 |
got | -0.160 |
Token | of |
Token ID | 286 |
Feature activation | -4.56e+00 |
Pos logit contributions | |
arl | +0.055 |
purpose | +0.051 |
Hawaii | +0.046 |
Ireland | +0.045 |
sight | +0.042 |
Neg logit contributions | |
forgot | -0.075 |
felt | -0.068 |
my | -0.068 |
laughed | -0.065 |
got | -0.065 |
Token | these |
Token ID | 777 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
arl | +0.077 |
purpose | +0.063 |
Hawaii | +0.056 |
This | +0.056 |
time | +0.055 |
Neg logit contributions | |
forgot | -0.124 |
felt | -0.112 |
laughed | -0.110 |
my | -0.109 |
got | -0.107 |
Token | books |
Token ID | 3835 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
arl | +0.046 |
purpose | +0.044 |
Hawaii | +0.040 |
Ireland | +0.038 |
This | +0.038 |
Neg logit contributions | |
forgot | -0.048 |
felt | -0.043 |
laughed | -0.040 |
my | -0.040 |
got | -0.040 |
Token | ?". |
Token ID | 43634 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
arl | +0.064 |
purpose | +0.063 |
Hawaii | +0.057 |
Ireland | +0.054 |
outer | +0.053 |
Neg logit contributions | |
forgot | -0.098 |
felt | -0.088 |
got | -0.084 |
laughed | -0.083 |
my | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | -5.26e+00 |
Pos logit contributions | |
arl | +0.164 |
purpose | +0.153 |
outer | +0.141 |
Hawaii | +0.141 |
Ireland | +0.140 |
Neg logit contributions | |
forgot | -0.151 |
felt | -0.135 |
got | -0.130 |
my | -0.127 |
took | -0.126 |
Token | Do |
Token ID | 2141 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
purpose | +0.105 |
sight | +0.096 |
arl | +0.093 |
outer | +0.090 |
common | +0.084 |
Neg logit contributions | |
forgot | -0.121 |
's | -0.110 |
had | -0.106 |
took | -0.106 |
felt | -0.105 |
Token | you |
Token ID | 345 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
arl | +0.055 |
purpose | +0.052 |
Hawaii | +0.046 |
outer | +0.046 |
Ireland | +0.045 |
Neg logit contributions | |
forgot | -0.095 |
felt | -0.083 |
laughed | -0.080 |
got | -0.079 |
knew | -0.078 |
Token | want |
Token ID | 765 |
Feature activation | -6.85e+00 |
Pos logit contributions | |
arl | +0.008 |
purpose | +0.008 |
Hawaii | +0.008 |
Ireland | +0.007 |
This | +0.007 |
Neg logit contributions | |
forgot | -0.007 |
felt | -0.006 |
laughed | -0.006 |
ppa | -0.006 |
got | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
arl | +0.175 |
purpose | +0.140 |
Hawaii | +0.137 |
Ireland | +0.137 |
ires | +0.132 |
Neg logit contributions | |
forgot | -0.140 |
felt | -0.133 |
got | -0.133 |
's | -0.129 |
my | -0.128 |
Token | read |
Token ID | 1100 |
Feature activation | -8.51e+00 |
Pos logit contributions | |
arl | +0.113 |
purpose | +0.107 |
Hawaii | +0.101 |
Ireland | +0.101 |
outer | +0.099 |
Neg logit contributions | |
forgot | -0.087 |
felt | -0.073 |
got | -0.070 |
laughed | -0.068 |
took | -0.066 |
Token | a |
Token ID | 257 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
arl | +0.224 |
purpose | +0.184 |
umm | +0.182 |
pleasure | +0.176 |
uttering | +0.170 |
Neg logit contributions | |
's | -0.157 |
my | -0.155 |
got | -0.152 |
felt | -0.152 |
forgot | -0.151 |
Token | book |
Token ID | 1492 |
Feature activation | -6.90e+00 |
Pos logit contributions | |
arl | +0.060 |
purpose | +0.058 |
Hawaii | +0.053 |
yourselves | +0.052 |
This | +0.052 |
Neg logit contributions | |
forgot | -0.071 |
felt | -0.062 |
laughed | -0.061 |
got | -0.061 |
knew | -0.060 |
Token | while |
Token ID | 981 |
Feature activation | -5.26e+00 |
Pos logit contributions | |
arl | +0.137 |
purpose | +0.137 |
pleasure | +0.122 |
Hawaii | +0.120 |
sight | +0.119 |
Neg logit contributions | |
forgot | -0.160 |
felt | -0.145 |
got | -0.143 |
my | -0.140 |
had | -0.136 |
Token | we |
Token ID | 356 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
arl | +0.106 |
purpose | +0.100 |
outer | +0.088 |
Hawaii | +0.088 |
ay | +0.086 |
Neg logit contributions | |
forgot | -0.129 |
felt | -0.113 |
got | -0.110 |
had | -0.108 |
laughed | -0.107 |
Token | wait |
Token ID | 4043 |
Feature activation | -6.28e+00 |
Pos logit contributions | |
forgot | +0.030 |
felt | +0.025 |
laughed | +0.025 |
ppa | +0.024 |
knew | +0.023 |
Neg logit contributions | |
arl | -0.030 |
purpose | -0.029 |
Hawaii | -0.028 |
This | -0.027 |
Ireland | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -4.74e+00 |
Pos logit contributions | |
arl | +0.058 |
purpose | +0.057 |
Hawaii | +0.052 |
Ireland | +0.050 |
This | +0.050 |
Neg logit contributions | |
forgot | -0.069 |
felt | -0.060 |
laughed | -0.057 |
got | -0.056 |
knew | -0.055 |
Token | Do |
Token ID | 2141 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
purpose | +0.084 |
sight | +0.075 |
arl | +0.070 |
outer | +0.064 |
time | +0.064 |
Neg logit contributions | |
forgot | -0.123 |
had | -0.111 |
my | -0.111 |
's | -0.110 |
felt | -0.109 |
Token | you |
Token ID | 345 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
arl | +0.044 |
purpose | +0.043 |
Hawaii | +0.038 |
Ireland | +0.036 |
This | +0.035 |
Neg logit contributions | |
forgot | -0.101 |
felt | -0.090 |
laughed | -0.087 |
got | -0.085 |
knew | -0.085 |
Token | want |
Token ID | 765 |
Feature activation | -6.66e+00 |
Pos logit contributions | |
arl | +0.014 |
purpose | +0.014 |
Hawaii | +0.013 |
Ireland | +0.013 |
This | +0.013 |
Neg logit contributions | |
forgot | -0.015 |
felt | -0.013 |
laughed | -0.012 |
ppa | -0.012 |
struggled | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -5.13e+00 |
Pos logit contributions | |
arl | +0.160 |
purpose | +0.130 |
Hawaii | +0.123 |
pleasure | +0.120 |
Ireland | +0.118 |
Neg logit contributions | |
forgot | -0.144 |
felt | -0.136 |
got | -0.135 |
laughed | -0.130 |
my | -0.129 |
Token | read |
Token ID | 1100 |
Feature activation | -8.48e+00 |
Pos logit contributions | |
arl | +0.135 |
purpose | +0.124 |
Hawaii | +0.118 |
ires | +0.117 |
Ireland | +0.116 |
Neg logit contributions | |
forgot | -0.093 |
felt | -0.079 |
got | -0.076 |
laughed | -0.073 |
knew | -0.071 |
Token | it |
Token ID | 340 |
Feature activation | -4.81e+00 |
Pos logit contributions | |
arl | +0.207 |
purpose | +0.173 |
umm | +0.166 |
pleasure | +0.165 |
tornado | +0.156 |
Neg logit contributions | |
forgot | -0.167 |
my | -0.164 |
felt | -0.163 |
's | -0.160 |
got | -0.159 |
Token | with |
Token ID | 351 |
Feature activation | -4.77e+00 |
Pos logit contributions | |
arl | +0.103 |
purpose | +0.097 |
Hawaii | +0.089 |
outer | +0.086 |
pleasure | +0.086 |
Neg logit contributions | |
forgot | -0.107 |
felt | -0.101 |
my | -0.099 |
got | -0.097 |
laughed | -0.093 |
Token | me |
Token ID | 502 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
arl | +0.123 |
purpose | +0.108 |
Hawaii | +0.105 |
Ireland | +0.101 |
outer | +0.101 |
Neg logit contributions | |
forgot | -0.092 |
felt | -0.081 |
laughed | -0.078 |
my | -0.077 |
got | -0.075 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
arl | +0.056 |
purpose | +0.055 |
Hawaii | +0.048 |
sight | +0.047 |
tornado | +0.046 |
Neg logit contributions | |
forgot | -0.073 |
felt | -0.067 |
my | -0.066 |
got | -0.064 |
laughed | -0.063 |
Token | Anna |
Token ID | 11735 |
Feature activation | +9.37e+00 |
Pos logit contributions | |
arl | +0.045 |
purpose | +0.043 |
Hawaii | +0.040 |
outer | +0.038 |
Ireland | +0.038 |
Neg logit contributions | |
forgot | -0.057 |
felt | -0.051 |
laughed | -0.048 |
got | -0.047 |
knew | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | -3.23e+00 |
Pos logit contributions | |
arl | +0.053 |
purpose | +0.052 |
outer | +0.050 |
Hawaii | +0.049 |
Ireland | +0.047 |
Neg logit contributions | |
forgot | -0.069 |
felt | -0.061 |
got | -0.057 |
laughed | -0.056 |
knew | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
arl | +0.062 |
purpose | +0.062 |
Hawaii | +0.056 |
Ireland | +0.056 |
outer | +0.055 |
Neg logit contributions | |
forgot | -0.079 |
felt | -0.071 |
got | -0.066 |
my | -0.066 |
knew | -0.066 |
Token | " |
Token ID | 1 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
arl | +0.051 |
purpose | +0.051 |
Hawaii | +0.046 |
This | +0.045 |
Ireland | +0.044 |
Neg logit contributions | |
forgot | -0.065 |
felt | -0.056 |
laughed | -0.054 |
knew | -0.052 |
got | -0.052 |
Token | Can |
Token ID | 6090 |
Feature activation | -5.29e+00 |
Pos logit contributions | |
arl | +0.035 |
purpose | +0.033 |
Hawaii | +0.030 |
Ireland | +0.028 |
This | +0.028 |
Neg logit contributions | |
forgot | -0.056 |
felt | -0.049 |
laughed | -0.048 |
got | -0.047 |
knew | -0.046 |
Token | we |
Token ID | 356 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
arl | +0.112 |
purpose | +0.100 |
sight | +0.090 |
outer | +0.090 |
Hawaii | +0.090 |
Neg logit contributions | |
forgot | -0.133 |
felt | -0.120 |
laughed | -0.115 |
had | -0.114 |
got | -0.113 |
Token | read |
Token ID | 1100 |
Feature activation | -8.47e+00 |
Pos logit contributions | |
purpose | +0.047 |
arl | +0.047 |
Hawaii | +0.044 |
This | +0.044 |
Ireland | +0.043 |
Neg logit contributions | |
forgot | -0.041 |
felt | -0.034 |
laughed | -0.032 |
knew | -0.031 |
ppa | -0.031 |
Token | this |
Token ID | 428 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
arl | +0.223 |
purpose | +0.184 |
umm | +0.184 |
pleasure | +0.166 |
ocol | +0.165 |
Neg logit contributions | |
forgot | -0.162 |
laughed | -0.162 |
felt | -0.160 |
's | -0.159 |
my | -0.155 |
Token | book |
Token ID | 1492 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
arl | +0.028 |
purpose | +0.025 |
Hawaii | +0.023 |
pleasure | +0.022 |
ocol | +0.022 |
Neg logit contributions | |
forgot | -0.023 |
felt | -0.021 |
laughed | -0.021 |
my | -0.020 |
got | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
arl | +0.104 |
purpose | +0.098 |
Hawaii | +0.089 |
outer | +0.085 |
Ireland | +0.083 |
Neg logit contributions | |
forgot | -0.130 |
felt | -0.118 |
got | -0.112 |
my | -0.111 |
laughed | -0.110 |
Token | please |
Token ID | 3387 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
arl | +0.057 |
purpose | +0.054 |
outer | +0.049 |
Hawaii | +0.047 |
sight | +0.045 |
Neg logit contributions | |
forgot | -0.077 |
felt | -0.069 |
laughed | -0.068 |
said | -0.065 |
got | -0.064 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
arl | +0.096 |
purpose | +0.087 |
ifying | +0.077 |
umm | +0.076 |
sight | +0.076 |
Neg logit contributions | |
forgot | -0.101 |
my | -0.090 |
said | -0.089 |
got | -0.089 |
took | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -5.16e+00 |
Pos logit contributions | |
arl | +0.096 |
purpose | +0.096 |
Hawaii | +0.089 |
Ireland | +0.085 |
This | +0.084 |
Neg logit contributions | |
forgot | -0.086 |
felt | -0.074 |
laughed | -0.069 |
got | -0.068 |
my | -0.067 |
Token | Do |
Token ID | 2141 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
purpose | +0.093 |
sight | +0.082 |
arl | +0.075 |
time | +0.072 |
common | +0.071 |
Neg logit contributions | |
forgot | -0.125 |
my | -0.125 |
's | -0.122 |
took | -0.121 |
had | -0.120 |
Token | you |
Token ID | 345 |
Feature activation | -5.84e-01 |
Pos logit contributions | |
arl | +0.056 |
purpose | +0.053 |
Hawaii | +0.048 |
outer | +0.046 |
sight | +0.045 |
Neg logit contributions | |
forgot | -0.098 |
felt | -0.087 |
laughed | -0.084 |
got | -0.083 |
knew | -0.082 |
Token | want |
Token ID | 765 |
Feature activation | -7.17e+00 |
Pos logit contributions | |
arl | +0.011 |
purpose | +0.010 |
Hawaii | +0.010 |
Ireland | +0.010 |
This | +0.009 |
Neg logit contributions | |
forgot | -0.014 |
felt | -0.013 |
ppa | -0.012 |
laughed | -0.012 |
struggled | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -5.03e+00 |
Pos logit contributions | |
arl | +0.182 |
Hawaii | +0.148 |
purpose | +0.146 |
umm | +0.146 |
tornado | +0.142 |
Neg logit contributions | |
got | -0.140 |
forgot | -0.140 |
felt | -0.138 |
my | -0.132 |
's | -0.131 |
Token | try |
Token ID | 1949 |
Feature activation | -8.46e+00 |
Pos logit contributions | |
arl | +0.128 |
purpose | +0.121 |
Hawaii | +0.118 |
Ireland | +0.113 |
time | +0.112 |
Neg logit contributions | |
forgot | -0.096 |
felt | -0.081 |
got | -0.078 |
laughed | -0.074 |
knew | -0.073 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
arl | +0.193 |
purpose | +0.172 |
umm | +0.161 |
pleasure | +0.160 |
tornado | +0.159 |
Neg logit contributions | |
forgot | -0.181 |
my | -0.171 |
felt | -0.169 |
got | -0.163 |
's | -0.162 |
Token |
|
Token ID | 198 |
Feature activation | -4.48e+00 |
Pos logit contributions | |
arl | +0.057 |
purpose | +0.053 |
outer | +0.053 |
Hawaii | +0.052 |
Ireland | +0.050 |
Neg logit contributions | |
forgot | -0.062 |
felt | -0.057 |
got | -0.053 |
laughed | -0.053 |
my | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
purpose | +0.104 |
arl | +0.103 |
Ireland | +0.102 |
outer | +0.101 |
Hawaii | +0.099 |
Neg logit contributions | |
forgot | -0.087 |
felt | -0.081 |
said | -0.078 |
my | -0.077 |
got | -0.076 |
Token | Tom |
Token ID | 13787 |
Feature activation | +8.58e+00 |
Pos logit contributions | |
arl | +0.076 |
purpose | +0.075 |
Hawaii | +0.069 |
Ireland | +0.067 |
outer | +0.066 |
Neg logit contributions | |
forgot | -0.090 |
felt | -0.078 |
laughed | -0.074 |
got | -0.073 |
knew | -0.072 |
Token | says |
Token ID | 1139 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
forgot | +0.217 |
ppa | +0.203 |
managed | +0.168 |
ba | +0.167 |
struggled | +0.167 |
Neg logit contributions | |
This | -0.142 |
Sometimes | -0.136 |
arl | -0.132 |
Today | -0.129 |
purpose | -0.128 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
ppa | +0.040 |
forgot | +0.040 |
ppy | +0.038 |
ba | +0.037 |
ll | +0.033 |
Neg logit contributions | |
This | -0.099 |
purpose | -0.097 |
Can | -0.096 |
hello | -0.095 |
A | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
arl | +0.032 |
purpose | +0.029 |
Hawaii | +0.028 |
Ireland | +0.027 |
outer | +0.027 |
Neg logit contributions | |
forgot | -0.026 |
felt | -0.024 |
laughed | -0.022 |
got | -0.022 |
knew | -0.022 |
Token | " |
Token ID | 366 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
forgot | +0.032 |
felt | +0.029 |
laughed | +0.028 |
knew | +0.027 |
got | +0.027 |
Neg logit contributions | |
purpose | -0.017 |
arl | -0.017 |
Hawaii | -0.015 |
Ireland | -0.014 |
outer | -0.014 |
Token | Can |
Token ID | 6090 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
arl | +0.029 |
purpose | +0.029 |
Hawaii | +0.026 |
Can | +0.025 |
Ireland | +0.024 |
Neg logit contributions | |
forgot | -0.072 |
felt | -0.063 |
laughed | -0.062 |
knew | -0.060 |
got | -0.059 |
Token | I |
Token ID | 314 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
arl | +0.069 |
purpose | +0.066 |
Hawaii | +0.059 |
Ireland | +0.056 |
outer | +0.055 |
Neg logit contributions | |
forgot | -0.120 |
felt | -0.107 |
laughed | -0.102 |
got | -0.102 |
knew | -0.100 |
Token | drink |
Token ID | 4144 |
Feature activation | -8.45e+00 |
Pos logit contributions | |
arl | +0.050 |
purpose | +0.049 |
Hawaii | +0.046 |
This | +0.044 |
Ireland | +0.044 |
Neg logit contributions | |
forgot | -0.053 |
felt | -0.045 |
laughed | -0.043 |
knew | -0.042 |
got | -0.041 |
Token | it |
Token ID | 340 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
arl | +0.199 |
purpose | +0.187 |
Hawaii | +0.186 |
Ireland | +0.178 |
outer | +0.177 |
Neg logit contributions | |
got | -0.160 |
felt | -0.157 |
forgot | -0.157 |
took | -0.150 |
my | -0.149 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
arl | +0.092 |
purpose | +0.088 |
Hawaii | +0.087 |
Ireland | +0.082 |
outer | +0.080 |
Neg logit contributions | |
forgot | -0.096 |
felt | -0.088 |
got | -0.085 |
my | -0.083 |
started | -0.080 |
Token | The |
Token ID | 383 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
purpose | +0.030 |
This | +0.027 |
arl | +0.027 |
Hawaii | +0.024 |
time | +0.022 |
Neg logit contributions | |
forgot | -0.089 |
felt | -0.077 |
ppa | -0.076 |
laughed | -0.075 |
ppy | -0.075 |
Token | driver |
Token ID | 4639 |
Feature activation | +9.71e+00 |
Pos logit contributions | |
arl | +0.004 |
purpose | +0.004 |
Hawaii | +0.003 |
Ireland | +0.003 |
This | +0.003 |
Neg logit contributions | |
forgot | -0.003 |
felt | -0.002 |
laughed | -0.002 |
knew | -0.002 |
got | -0.002 |
Token | laughed |
Token ID | 13818 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
forgot | +0.147 |
ppa | +0.132 |
ilia | +0.107 |
ppy | +0.106 |
anut | +0.104 |
Neg logit contributions | |
time | -0.220 |
A | -0.200 |
This | -0.200 |
ay | -0.197 |
birthday | -0.193 |
Token | " |
Token ID | 1 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
arl | +0.039 |
purpose | +0.039 |
Hawaii | +0.036 |
Ireland | +0.034 |
This | +0.034 |
Neg logit contributions | |
forgot | -0.051 |
felt | -0.044 |
laughed | -0.042 |
got | -0.041 |
knew | -0.041 |
Token | Can |
Token ID | 6090 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
purpose | +0.018 |
arl | +0.018 |
This | +0.016 |
Hawaii | +0.016 |
outer | +0.016 |
Neg logit contributions | |
forgot | -0.026 |
felt | -0.023 |
laughed | -0.022 |
took | -0.022 |
had | -0.022 |
Token | we |
Token ID | 356 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
arl | +0.097 |
purpose | +0.092 |
outer | +0.084 |
sight | +0.082 |
Hawaii | +0.082 |
Neg logit contributions | |
forgot | -0.112 |
felt | -0.097 |
got | -0.093 |
had | -0.092 |
laughed | -0.092 |
Token | try |
Token ID | 1949 |
Feature activation | -7.30e+00 |
Pos logit contributions | |
arl | +0.031 |
purpose | +0.031 |
Hawaii | +0.029 |
This | +0.028 |
Ireland | +0.028 |
Neg logit contributions | |
forgot | -0.028 |
felt | -0.024 |
laughed | -0.022 |
knew | -0.022 |
got | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
arl | +0.173 |
purpose | +0.149 |
ay | +0.143 |
Hawaii | +0.140 |
Ireland | +0.138 |
Neg logit contributions | |
forgot | -0.156 |
my | -0.142 |
felt | -0.142 |
's | -0.138 |
got | -0.135 |
Token | whistle |
Token ID | 16121 |
Feature activation | -8.43e+00 |
Pos logit contributions | |
arl | +0.100 |
purpose | +0.097 |
Hawaii | +0.092 |
Ireland | +0.090 |
This | +0.088 |
Neg logit contributions | |
forgot | -0.073 |
felt | -0.061 |
laughed | -0.056 |
got | -0.056 |
knew | -0.054 |
Token | now |
Token ID | 783 |
Feature activation | -4.24e+00 |
Pos logit contributions | |
arl | +0.222 |
purpose | +0.195 |
Hawaii | +0.188 |
umm | +0.184 |
outer | +0.181 |
Neg logit contributions | |
forgot | -0.154 |
felt | -0.153 |
laughed | -0.141 |
got | -0.139 |
my | -0.138 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
arl | +0.088 |
purpose | +0.083 |
outer | +0.078 |
Ireland | +0.078 |
Hawaii | +0.078 |
Neg logit contributions | |
forgot | -0.098 |
felt | -0.087 |
got | -0.084 |
said | -0.081 |
took | -0.080 |
Token | Tom |
Token ID | 4186 |
Feature activation | +8.77e+00 |
Pos logit contributions | |
arl | +0.021 |
purpose | +0.020 |
Hawaii | +0.018 |
outer | +0.017 |
Ireland | +0.017 |
Neg logit contributions | |
forgot | -0.025 |
felt | -0.022 |
laughed | -0.021 |
got | -0.021 |
knew | -0.021 |
Token | asks |
Token ID | 7893 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
forgot | +0.215 |
felt | +0.172 |
laughed | +0.168 |
managed | +0.168 |
got | +0.164 |
Neg logit contributions | |
arl | -0.166 |
purpose | -0.162 |
Today | -0.156 |
This | -0.154 |
Sometimes | -0.154 |
Token | . |
Token ID | 13 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
arl | +0.022 |
purpose | +0.022 |
Hawaii | +0.020 |
Ireland | +0.020 |
This | +0.019 |
Neg logit contributions | |
forgot | -0.022 |
felt | -0.019 |
laughed | -0.018 |
got | -0.018 |
knew | -0.017 |
Token | Can |
Token ID | 6090 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
arl | +0.053 |
purpose | +0.052 |
Hawaii | +0.048 |
Ireland | +0.045 |
This | +0.045 |
Neg logit contributions | |
forgot | -0.082 |
felt | -0.072 |
laughed | -0.069 |
got | -0.067 |
knew | -0.067 |
Token | you |
Token ID | 345 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
arl | +0.064 |
purpose | +0.060 |
Hawaii | +0.053 |
Ireland | +0.050 |
outer | +0.049 |
Neg logit contributions | |
forgot | -0.132 |
felt | -0.119 |
laughed | -0.114 |
got | -0.113 |
had | -0.112 |
Token | teach |
Token ID | 4545 |
Feature activation | -6.51e+00 |
Pos logit contributions | |
purpose | +0.050 |
arl | +0.050 |
Hawaii | +0.046 |
This | +0.045 |
Ireland | +0.044 |
Neg logit contributions | |
forgot | -0.053 |
felt | -0.045 |
laughed | -0.043 |
knew | -0.041 |
got | -0.041 |
Token | me |
Token ID | 502 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
arl | +0.177 |
Hawaii | +0.154 |
purpose | +0.152 |
Ireland | +0.148 |
outer | +0.144 |
Neg logit contributions | |
forgot | -0.113 |
felt | -0.104 |
my | -0.102 |
laughed | -0.102 |
's | -0.095 |
Token | to |
Token ID | 284 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
arl | +0.052 |
purpose | +0.048 |
umm | +0.045 |
Hawaii | +0.044 |
sight | +0.043 |
Neg logit contributions | |
forgot | -0.044 |
felt | -0.041 |
my | -0.041 |
got | -0.040 |
laughed | -0.039 |
Token | read |
Token ID | 1100 |
Feature activation | -8.41e+00 |
Pos logit contributions | |
arl | +0.082 |
purpose | +0.079 |
Hawaii | +0.075 |
Ireland | +0.073 |
This | +0.072 |
Neg logit contributions | |
forgot | -0.066 |
felt | -0.055 |
laughed | -0.052 |
got | -0.051 |
knew | -0.050 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
arl | +0.202 |
umm | +0.158 |
purpose | +0.158 |
uttering | +0.154 |
pleasure | +0.151 |
Neg logit contributions | |
my | -0.170 |
's | -0.166 |
forgot | -0.164 |
felt | -0.164 |
laughed | -0.159 |
Token |
|
Token ID | 198 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
arl | +0.048 |
purpose | +0.046 |
Hawaii | +0.042 |
Ireland | +0.041 |
outer | +0.040 |
Neg logit contributions | |
forgot | -0.066 |
felt | -0.059 |
laughed | -0.056 |
knew | -0.055 |
got | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
arl | +0.080 |
purpose | +0.076 |
Hawaii | +0.071 |
Ireland | +0.071 |
outer | +0.068 |
Neg logit contributions | |
forgot | -0.090 |
felt | -0.083 |
my | -0.080 |
knew | -0.080 |
laughed | -0.079 |
Token | The |
Token ID | 464 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
arl | +0.068 |
purpose | +0.065 |
Hawaii | +0.062 |
Ireland | +0.059 |
outer | +0.058 |
Neg logit contributions | |
forgot | -0.073 |
felt | -0.063 |
laughed | -0.060 |
knew | -0.059 |
got | -0.059 |
Token | tutor |
Token ID | 48680 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
forgot | +0.006 |
felt | +0.005 |
laughed | +0.005 |
knew | +0.005 |
got | +0.005 |
Neg logit contributions | |
arl | -0.003 |
purpose | -0.003 |
Hawaii | -0.003 |
Ireland | -0.003 |
This | -0.003 |
Token | " |
Token ID | 366 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
forgot | +0.014 |
felt | +0.013 |
laughed | +0.012 |
knew | +0.012 |
my | +0.012 |
Neg logit contributions | |
arl | -0.011 |
purpose | -0.011 |
Hawaii | -0.009 |
ay | -0.009 |
outer | -0.009 |
Token | Do |
Token ID | 5211 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
arl | +0.034 |
purpose | +0.034 |
Hawaii | +0.031 |
Can | +0.028 |
Ireland | +0.028 |
Neg logit contributions | |
forgot | -0.097 |
felt | -0.086 |
laughed | -0.084 |
knew | -0.081 |
struggled | -0.080 |
Token | you |
Token ID | 345 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
arl | +0.059 |
purpose | +0.056 |
Hawaii | +0.051 |
Ireland | +0.048 |
outer | +0.048 |
Neg logit contributions | |
forgot | -0.104 |
felt | -0.092 |
laughed | -0.089 |
got | -0.088 |
knew | -0.086 |
Token | want |
Token ID | 765 |
Feature activation | -6.99e+00 |
Pos logit contributions | |
arl | +0.028 |
purpose | +0.028 |
This | +0.026 |
Ireland | +0.025 |
Today | +0.025 |
Neg logit contributions | |
forgot | -0.033 |
ppa | -0.029 |
felt | -0.028 |
mie | -0.026 |
laughed | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -4.91e+00 |
Pos logit contributions | |
arl | +0.194 |
umm | +0.162 |
ocol | +0.162 |
ires | +0.162 |
purpose | +0.158 |
Neg logit contributions | |
got | -0.122 |
forgot | -0.118 |
's | -0.117 |
felt | -0.117 |
my | -0.116 |
Token | play |
Token ID | 711 |
Feature activation | -8.41e+00 |
Pos logit contributions | |
arl | +0.133 |
purpose | +0.123 |
Ireland | +0.118 |
Hawaii | +0.117 |
ires | +0.117 |
Neg logit contributions | |
forgot | -0.087 |
felt | -0.071 |
got | -0.070 |
my | -0.067 |
laughed | -0.066 |
Token | with |
Token ID | 351 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
arl | +0.227 |
purpose | +0.216 |
umm | +0.207 |
ifying | +0.199 |
pleasure | +0.198 |
Neg logit contributions | |
's | -0.128 |
felt | -0.125 |
got | -0.124 |
forgot | -0.123 |
pretended | -0.123 |
Token | me |
Token ID | 502 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
arl | +0.159 |
umm | +0.138 |
pleasure | +0.137 |
purpose | +0.137 |
Today | +0.130 |
Neg logit contributions | |
's | -0.098 |
forgot | -0.095 |
laughed | -0.093 |
got | -0.092 |
felt | -0.090 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
arl | +0.063 |
purpose | +0.061 |
Hawaii | +0.055 |
outer | +0.052 |
Ireland | +0.052 |
Neg logit contributions | |
forgot | -0.084 |
felt | -0.077 |
my | -0.077 |
got | -0.074 |
laughed | -0.073 |
Token |
|
Token ID | 198 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
arl | +0.053 |
purpose | +0.050 |
Hawaii | +0.047 |
Ireland | +0.045 |
outer | +0.044 |
Neg logit contributions | |
forgot | -0.075 |
felt | -0.068 |
laughed | -0.065 |
got | -0.064 |
knew | -0.063 |
Token |
|
Token ID | 198 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
arl | +0.100 |
purpose | +0.095 |
Hawaii | +0.089 |
Ireland | +0.088 |
outer | +0.086 |
Neg logit contributions | |
forgot | -0.101 |
felt | -0.094 |
my | -0.090 |
got | -0.089 |
had | -0.087 |
Token | needed |
Token ID | 2622 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
forgot | +0.075 |
ppa | +0.071 |
ppy | +0.062 |
ie | +0.056 |
i | +0.055 |
Neg logit contributions | |
This | -0.125 |
ay | -0.123 |
arl | -0.122 |
Can | -0.122 |
pleasure | -0.121 |
Token | to |
Token ID | 284 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
arl | +0.095 |
purpose | +0.077 |
ires | +0.075 |
Ireland | +0.073 |
ay | +0.073 |
Neg logit contributions | |
forgot | -0.089 |
got | -0.083 |
felt | -0.081 |
took | -0.080 |
had | -0.079 |
Token | go |
Token ID | 467 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
arl | +0.053 |
purpose | +0.052 |
Hawaii | +0.049 |
Ireland | +0.047 |
This | +0.047 |
Neg logit contributions | |
forgot | -0.046 |
felt | -0.038 |
laughed | -0.036 |
got | -0.035 |
knew | -0.035 |
Token | do |
Token ID | 466 |
Feature activation | -5.85e+00 |
Pos logit contributions | |
arl | +0.089 |
purpose | +0.088 |
Hawaii | +0.079 |
Ireland | +0.077 |
This | +0.075 |
Neg logit contributions | |
forgot | -0.115 |
felt | -0.101 |
got | -0.096 |
laughed | -0.095 |
my | -0.093 |
Token | some |
Token ID | 617 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
arl | +0.101 |
purpose | +0.094 |
Hawaii | +0.087 |
Ireland | +0.083 |
This | +0.081 |
Neg logit contributions | |
forgot | -0.161 |
felt | -0.143 |
laughed | -0.136 |
got | -0.134 |
knew | -0.133 |
Token | err |
Token ID | 11454 |
Feature activation | -8.41e+00 |
Pos logit contributions | |
arl | +0.064 |
purpose | +0.058 |
Hawaii | +0.054 |
Ireland | +0.053 |
This | +0.052 |
Neg logit contributions | |
forgot | -0.089 |
felt | -0.080 |
got | -0.076 |
laughed | -0.076 |
started | -0.074 |
Token | ands |
Token ID | 1746 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
arl | +0.148 |
purpose | +0.146 |
Hawaii | +0.133 |
Ireland | +0.127 |
This | +0.126 |
Neg logit contributions | |
forgot | -0.221 |
felt | -0.192 |
laughed | -0.185 |
got | -0.181 |
knew | -0.180 |
Token | . |
Token ID | 13 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
arl | +0.058 |
purpose | +0.058 |
Hawaii | +0.053 |
This | +0.051 |
Ireland | +0.051 |
Neg logit contributions | |
forgot | -0.069 |
felt | -0.059 |
laughed | -0.056 |
knew | -0.054 |
got | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
arl | +0.049 |
purpose | +0.047 |
Hawaii | +0.043 |
Ireland | +0.042 |
outer | +0.041 |
Neg logit contributions | |
forgot | -0.067 |
felt | -0.061 |
got | -0.058 |
my | -0.057 |
started | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
arl | +0.049 |
purpose | +0.048 |
Hawaii | +0.045 |
Ireland | +0.043 |
outer | +0.042 |
Neg logit contributions | |
forgot | -0.067 |
felt | -0.059 |
laughed | -0.056 |
got | -0.056 |
my | -0.055 |
Token | Mom |
Token ID | 29252 |
Feature activation | +6.70e+00 |
Pos logit contributions | |
arl | +0.042 |
purpose | +0.042 |
Hawaii | +0.039 |
Ireland | +0.037 |
This | +0.037 |
Neg logit contributions | |
forgot | -0.055 |
felt | -0.048 |
laughed | -0.046 |
knew | -0.045 |
got | -0.045 |
Token | Yes |
Token ID | 5297 |
Feature activation | -4.04e+00 |
Pos logit contributions | |
arl | +0.032 |
purpose | +0.031 |
Hawaii | +0.029 |
Ireland | +0.028 |
outer | +0.027 |
Neg logit contributions | |
forgot | -0.036 |
felt | -0.031 |
laughed | -0.030 |
knew | -0.030 |
got | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
arl | +0.095 |
purpose | +0.093 |
Hawaii | +0.091 |
outer | +0.089 |
Ireland | +0.088 |
Neg logit contributions | |
forgot | -0.082 |
felt | -0.070 |
laughed | -0.069 |
got | -0.068 |
knew | -0.066 |
Token | please |
Token ID | 3387 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
purpose | +0.058 |
arl | +0.058 |
Hawaii | +0.050 |
outer | +0.048 |
Ireland | +0.047 |
Neg logit contributions | |
forgot | -0.094 |
felt | -0.083 |
laughed | -0.083 |
knew | -0.080 |
got | -0.080 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
arl | +0.043 |
purpose | +0.041 |
Hawaii | +0.038 |
Ireland | +0.037 |
outer | +0.036 |
Neg logit contributions | |
forgot | -0.058 |
felt | -0.051 |
laughed | -0.049 |
got | -0.049 |
knew | -0.048 |
Token | Tim |
Token ID | 5045 |
Feature activation | +7.97e+00 |
Pos logit contributions | |
purpose | +0.019 |
This | +0.019 |
arl | +0.017 |
Hawaii | +0.017 |
Can | +0.016 |
Neg logit contributions | |
forgot | -0.043 |
ppa | -0.038 |
ppy | -0.036 |
felt | -0.036 |
my | -0.035 |
Token | my |
Token ID | 1820 |
Feature activation | +9.49e+00 |
Pos logit contributions | |
forgot | +0.105 |
struggled | +0.070 |
dragged | +0.067 |
managed | +0.065 |
ppa | +0.064 |
Neg logit contributions | |
purpose | -0.223 |
arl | -0.213 |
This | -0.212 |
ay | -0.210 |
Today | -0.209 |
Token | replied |
Token ID | 8712 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
forgot | +0.146 |
ppa | +0.121 |
eling | +0.118 |
ppy | +0.109 |
my | +0.102 |
Neg logit contributions | |
This | -0.231 |
A | -0.218 |
purpose | -0.215 |
Can | -0.201 |
time | -0.200 |
Token | . |
Token ID | 13 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
arl | +0.034 |
purpose | +0.034 |
Hawaii | +0.031 |
This | +0.030 |
Ireland | +0.030 |
Neg logit contributions | |
forgot | -0.041 |
felt | -0.035 |
laughed | -0.033 |
knew | -0.033 |
got | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -4.11e+00 |
Pos logit contributions | |
purpose | +0.082 |
arl | +0.081 |
Hawaii | +0.076 |
outer | +0.075 |
Ireland | +0.072 |
Neg logit contributions | |
forgot | -0.100 |
felt | -0.090 |
laughed | -0.087 |
got | -0.086 |
took | -0.084 |
Token |
|
Token ID | 198 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
purpose | +0.075 |
arl | +0.075 |
Hawaii | +0.069 |
Ireland | +0.067 |
outer | +0.066 |
Neg logit contributions | |
forgot | -0.105 |
felt | -0.094 |
laughed | -0.089 |
got | -0.089 |
my | -0.088 |
Token | Bob |
Token ID | 18861 |
Feature activation | +8.43e+00 |
Pos logit contributions | |
arl | +0.067 |
purpose | +0.066 |
Hawaii | +0.060 |
This | +0.059 |
Ireland | +0.058 |
Neg logit contributions | |
forgot | -0.090 |
felt | -0.078 |
laughed | -0.074 |
knew | -0.072 |
got | -0.072 |
Token | ." |
Token ID | 526 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
arl | +0.040 |
purpose | +0.040 |
Hawaii | +0.037 |
Ireland | +0.035 |
outer | +0.035 |
Neg logit contributions | |
forgot | -0.038 |
felt | -0.033 |
got | -0.032 |
laughed | -0.031 |
had | -0.030 |
Token | |
Token ID | 220 |
Feature activation | -3.84e+00 |
Pos logit contributions | |
arl | +0.059 |
purpose | +0.058 |
Hawaii | +0.054 |
This | +0.052 |
Ireland | +0.052 |
Neg logit contributions | |
forgot | -0.061 |
felt | -0.052 |
laughed | -0.049 |
got | -0.048 |
knew | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -3.67e+00 |
Pos logit contributions | |
purpose | +0.077 |
arl | +0.077 |
outer | +0.072 |
Hawaii | +0.071 |
Ireland | +0.070 |
Neg logit contributions | |
forgot | -0.088 |
felt | -0.078 |
my | -0.076 |
got | -0.072 |
knew | -0.072 |
Token |
|
Token ID | 198 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
arl | +0.062 |
purpose | +0.061 |
Hawaii | +0.055 |
Ireland | +0.053 |
outer | +0.052 |
Neg logit contributions | |
forgot | -0.100 |
felt | -0.089 |
laughed | -0.085 |
knew | -0.083 |
got | -0.083 |
Token | The |
Token ID | 464 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
purpose | +0.048 |
arl | +0.046 |
Hawaii | +0.042 |
This | +0.041 |
time | +0.040 |
Neg logit contributions | |
forgot | -0.086 |
felt | -0.076 |
laughed | -0.072 |
got | -0.070 |
knew | -0.070 |
Token | baby |
Token ID | 5156 |
Feature activation | +9.20e+00 |
Pos logit contributions | |
forgot | +0.016 |
felt | +0.013 |
laughed | +0.013 |
knew | +0.012 |
got | +0.012 |
Neg logit contributions | |
arl | -0.016 |
purpose | -0.015 |
Hawaii | -0.014 |
This | -0.014 |
Ireland | -0.014 |
Token | gigg |
Token ID | 30442 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
forgot | +0.113 |
ppa | +0.103 |
ppy | +0.090 |
ge | +0.085 |
struggled | +0.080 |
Neg logit contributions | |
purpose | -0.217 |
time | -0.217 |
sight | -0.211 |
Home | -0.202 |
Sunday | -0.201 |
Token | led |
Token ID | 992 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
forgot | +0.019 |
felt | +0.014 |
laughed | +0.014 |
my | +0.014 |
knew | +0.014 |
Neg logit contributions | |
arl | -0.030 |
purpose | -0.029 |
time | -0.027 |
ires | -0.027 |
ay | -0.027 |
Token | with |
Token ID | 351 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
arl | +0.069 |
purpose | +0.067 |
Hawaii | +0.064 |
Ireland | +0.062 |
This | +0.061 |
Neg logit contributions | |
forgot | -0.033 |
felt | -0.027 |
laughed | -0.025 |
got | -0.024 |
knew | -0.024 |
Token | excitement |
Token ID | 14067 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
forgot | +0.010 |
felt | +0.009 |
laughed | +0.009 |
knew | +0.009 |
got | +0.008 |
Neg logit contributions | |
arl | -0.008 |
purpose | -0.007 |
Hawaii | -0.007 |
Ireland | -0.007 |
outer | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
arl | +0.025 |
purpose | +0.025 |
Hawaii | +0.023 |
This | +0.022 |
Ireland | +0.022 |
Neg logit contributions | |
forgot | -0.026 |
felt | -0.022 |
laughed | -0.021 |
knew | -0.020 |
got | -0.020 |
Token | pretty |
Token ID | 2495 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
forgot | +0.024 |
felt | +0.021 |
laughed | +0.020 |
knew | +0.020 |
got | +0.020 |
Neg logit contributions | |
arl | -0.017 |
purpose | -0.016 |
Hawaii | -0.015 |
This | -0.014 |
Ireland | -0.014 |
Token | calendar |
Token ID | 11845 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
forgot | +0.000 |
felt | +0.000 |
laughed | +0.000 |
got | +0.000 |
knew | +0.000 |
Neg logit contributions | |
arl | -0.000 |
purpose | -0.000 |
Hawaii | -0.000 |
Ireland | -0.000 |
This | -0.000 |
Token | she |
Token ID | 673 |
Feature activation | +6.42e+00 |
Pos logit contributions | |
forgot | +0.023 |
felt | +0.019 |
knew | +0.018 |
laughed | +0.018 |
ppa | +0.018 |
Neg logit contributions | |
purpose | -0.016 |
arl | -0.015 |
Hawaii | -0.015 |
This | -0.014 |
time | -0.014 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
forgot | +0.059 |
ppa | +0.040 |
felt | +0.038 |
my | +0.037 |
struggled | +0.036 |
Neg logit contributions | |
purpose | -0.192 |
Can | -0.192 |
This | -0.191 |
arl | -0.188 |
time | -0.186 |
Token | ! |
Token ID | 0 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
arl | +0.073 |
purpose | +0.067 |
Hawaii | +0.063 |
Ireland | +0.061 |
outer | +0.059 |
Neg logit contributions | |
forgot | -0.082 |
felt | -0.073 |
laughed | -0.069 |
got | -0.069 |
had | -0.068 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
arl | +0.011 |
purpose | +0.011 |
Hawaii | +0.010 |
Ireland | +0.010 |
outer | +0.010 |
Neg logit contributions | |
forgot | -0.015 |
felt | -0.013 |
laughed | -0.012 |
got | -0.012 |
knew | -0.012 |
Token | was |
Token ID | 373 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
ppa | +0.166 |
ppy | +0.159 |
forgot | +0.155 |
pper | +0.153 |
anut | +0.151 |
Neg logit contributions | |
This | -0.200 |
A | -0.186 |
New | -0.180 |
The | -0.177 |
purpose | -0.176 |
Token | so |
Token ID | 523 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
forgot | +0.056 |
felt | +0.046 |
ppy | +0.045 |
ppa | +0.044 |
laughed | +0.044 |
Neg logit contributions | |
purpose | -0.048 |
arl | -0.045 |
Hawaii | -0.044 |
This | -0.043 |
Ireland | -0.043 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
forgot | +0.029 |
felt | +0.024 |
laughed | +0.022 |
my | +0.022 |
ppa | +0.022 |
Neg logit contributions | |
arl | -0.032 |
purpose | -0.032 |
Hawaii | -0.030 |
This | -0.029 |
Ireland | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
purpose | +0.021 |
arl | +0.021 |
Hawaii | +0.019 |
This | +0.019 |
Ireland | +0.018 |
Neg logit contributions | |
forgot | -0.024 |
felt | -0.020 |
laughed | -0.019 |
knew | -0.019 |
got | -0.019 |
Token | said |
Token ID | 531 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
forgot | +0.066 |
ppy | +0.065 |
my | +0.061 |
ppa | +0.059 |
ie | +0.056 |
Neg logit contributions | |
This | -0.136 |
purpose | -0.136 |
Can | -0.131 |
arl | -0.128 |
sight | -0.127 |
Token | down |
Token ID | 866 |
Feature activation | -3.88e+00 |
Pos logit contributions | |
arl | +0.052 |
purpose | +0.052 |
Hawaii | +0.049 |
Ireland | +0.047 |
This | +0.046 |
Neg logit contributions | |
forgot | -0.053 |
felt | -0.047 |
laughed | -0.044 |
got | -0.044 |
knew | -0.043 |
Token | on |
Token ID | 319 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
arl | +0.096 |
purpose | +0.092 |
pleasure | +0.088 |
Hawaii | +0.085 |
This | +0.082 |
Neg logit contributions | |
forgot | -0.074 |
felt | -0.069 |
got | -0.066 |
started | -0.065 |
had | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
arl | +0.105 |
pleasure | +0.085 |
ires | +0.083 |
Can | +0.081 |
Hawaii | +0.080 |
Neg logit contributions | |
forgot | -0.075 |
felt | -0.074 |
got | -0.070 |
started | -0.069 |
had | -0.068 |
Token | beach |
Token ID | 10481 |
Feature activation | -7.20e-01 |
Pos logit contributions | |
arl | +0.043 |
purpose | +0.042 |
Hawaii | +0.039 |
Ireland | +0.038 |
This | +0.038 |
Neg logit contributions | |
forgot | -0.044 |
felt | -0.037 |
laughed | -0.035 |
got | -0.034 |
knew | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
purpose | +0.012 |
arl | +0.011 |
This | +0.011 |
Hawaii | +0.011 |
time | +0.011 |
Neg logit contributions | |
forgot | -0.019 |
felt | -0.016 |
ppa | -0.016 |
laughed | -0.016 |
ppy | -0.016 |
Token | Jane |
Token ID | 12091 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
arl | +0.061 |
purpose | +0.056 |
outer | +0.053 |
Hawaii | +0.053 |
Ireland | +0.050 |
Neg logit contributions | |
forgot | -0.063 |
felt | -0.058 |
got | -0.054 |
laughed | -0.054 |
had | -0.053 |
Token | was |
Token ID | 373 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
ppa | +0.185 |
anut | +0.166 |
ppers | +0.151 |
ilia | +0.150 |
ppy | +0.148 |
Neg logit contributions | |
This | -0.162 |
Can | -0.159 |
A | -0.157 |
time | -0.157 |
It | -0.155 |
Token | scared |
Token ID | 12008 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
forgot | +0.030 |
ppa | +0.024 |
felt | +0.023 |
ppy | +0.023 |
ie | +0.023 |
Neg logit contributions | |
purpose | -0.032 |
arl | -0.030 |
This | -0.029 |
Hawaii | -0.029 |
sight | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
arl | +0.104 |
purpose | +0.103 |
Hawaii | +0.099 |
Ireland | +0.096 |
outer | +0.094 |
Neg logit contributions | |
forgot | -0.067 |
felt | -0.056 |
laughed | -0.051 |
my | -0.051 |
knew | -0.051 |
Token | but |
Token ID | 475 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
forgot | +0.073 |
ppa | +0.061 |
ppy | +0.059 |
felt | +0.059 |
laughed | +0.058 |
Neg logit contributions | |
purpose | -0.055 |
arl | -0.053 |
This | -0.052 |
sight | -0.050 |
Can | -0.049 |
Token | her |
Token ID | 607 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
forgot | +0.081 |
ppa | +0.071 |
ppy | +0.069 |
my | +0.066 |
felt | +0.066 |
Neg logit contributions | |
purpose | -0.064 |
Hawaii | -0.062 |
Ireland | -0.061 |
Earth | -0.061 |
sight | -0.061 |
Token | Ben |
Token ID | 11696 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
arl | +0.040 |
purpose | +0.039 |
Hawaii | +0.037 |
Ireland | +0.035 |
This | +0.035 |
Neg logit contributions | |
forgot | -0.048 |
felt | -0.041 |
laughed | -0.039 |
got | -0.038 |
started | -0.037 |
Token | nods |
Token ID | 34833 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
forgot | +0.251 |
ppa | +0.214 |
managed | +0.194 |
got | +0.192 |
struggled | +0.191 |
Neg logit contributions | |
arl | -0.185 |
purpose | -0.182 |
This | -0.175 |
Sometimes | -0.175 |
Today | -0.174 |
Token | and |
Token ID | 290 |
Feature activation | +5.95e+00 |
Pos logit contributions | |
arl | +0.004 |
purpose | +0.004 |
Hawaii | +0.004 |
Ireland | +0.004 |
This | +0.004 |
Neg logit contributions | |
forgot | -0.003 |
felt | -0.002 |
laughed | -0.002 |
knew | -0.002 |
got | -0.002 |
Token | smiles |
Token ID | 21845 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
forgot | +0.133 |
ppa | +0.113 |
felt | +0.107 |
my | +0.104 |
ppy | +0.103 |
Neg logit contributions | |
arl | -0.121 |
Hawaii | -0.116 |
purpose | -0.115 |
This | -0.111 |
Ireland | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
forgot | +0.007 |
felt | +0.006 |
got | +0.006 |
knew | +0.006 |
laughed | +0.006 |
Neg logit contributions | |
purpose | -0.009 |
arl | -0.009 |
Hawaii | -0.009 |
Ireland | -0.008 |
outer | -0.008 |
Token | He |
Token ID | 679 |
Feature activation | +8.30e+00 |
Pos logit contributions | |
purpose | +0.053 |
arl | +0.050 |
Hawaii | +0.049 |
Ireland | +0.049 |
outer | +0.049 |
Neg logit contributions | |
forgot | -0.037 |
felt | -0.035 |
knew | -0.033 |
had | -0.032 |
took | -0.031 |
Token | likes |
Token ID | 7832 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
forgot | +0.189 |
ppa | +0.171 |
ie | +0.157 |
my | +0.155 |
struggled | +0.151 |
Neg logit contributions | |
arl | -0.151 |
Hawaii | -0.143 |
Sometimes | -0.142 |
This | -0.140 |
purpose | -0.138 |
Token | to |
Token ID | 284 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
arl | +0.054 |
purpose | +0.049 |
ay | +0.046 |
Ireland | +0.045 |
Hawaii | +0.045 |
Neg logit contributions | |
forgot | -0.057 |
felt | -0.052 |
laughed | -0.051 |
knew | -0.049 |
got | -0.048 |
Token | jog |
Token ID | 48342 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
arl | +0.037 |
purpose | +0.036 |
Hawaii | +0.034 |
Ireland | +0.033 |
This | +0.033 |
Neg logit contributions | |
forgot | -0.034 |
felt | -0.029 |
laughed | -0.027 |
got | -0.026 |
knew | -0.026 |
Token | with |
Token ID | 351 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
arl | +0.107 |
purpose | +0.105 |
Hawaii | +0.097 |
Ireland | +0.096 |
pleasure | +0.092 |
Neg logit contributions | |
forgot | -0.075 |
felt | -0.069 |
got | -0.063 |
laughed | -0.061 |
had | -0.060 |
Token | his |
Token ID | 465 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
arl | +0.036 |
purpose | +0.034 |
Hawaii | +0.032 |
This | +0.031 |
Ireland | +0.031 |
Neg logit contributions | |
forgot | -0.032 |
felt | -0.027 |
laughed | -0.026 |
knew | -0.025 |
got | -0.025 |
Token | At |
Token ID | 2953 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
arl | +0.030 |
purpose | +0.029 |
Hawaii | +0.027 |
This | +0.026 |
Ireland | +0.026 |
Neg logit contributions | |
forgot | -0.038 |
felt | -0.033 |
laughed | -0.032 |
knew | -0.031 |
got | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
arl | +0.018 |
purpose | +0.018 |
Can | +0.016 |
This | +0.016 |
Ireland | +0.016 |
Neg logit contributions | |
forgot | -0.020 |
felt | -0.018 |
got | -0.017 |
my | -0.017 |
laughed | -0.017 |
Token | vet |
Token ID | 20202 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
forgot | +0.006 |
felt | +0.005 |
got | +0.005 |
knew | +0.005 |
my | +0.005 |
Neg logit contributions | |
arl | -0.006 |
purpose | -0.006 |
Hawaii | -0.005 |
Ireland | -0.005 |
This | -0.005 |
Token | 's |
Token ID | 338 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
forgot | +0.078 |
felt | +0.062 |
laughed | +0.062 |
ppa | +0.061 |
struggled | +0.058 |
Neg logit contributions | |
purpose | -0.086 |
This | -0.086 |
arl | -0.083 |
Hawaii | -0.082 |
time | -0.081 |
Token | office |
Token ID | 2607 |
Feature activation | +8.18e-01 |
Pos logit contributions | |
forgot | +0.014 |
felt | +0.013 |
got | +0.012 |
my | +0.012 |
knew | +0.012 |
Neg logit contributions | |
arl | -0.013 |
purpose | -0.012 |
Hawaii | -0.011 |
Ireland | -0.011 |
outer | -0.011 |
Token | Mia |
Token ID | 32189 |
Feature activation | +7.64e+00 |
Pos logit contributions | |
forgot | +0.020 |
felt | +0.017 |
laughed | +0.017 |
knew | +0.016 |
ppa | +0.016 |
Neg logit contributions | |
purpose | -0.015 |
arl | -0.015 |
Hawaii | -0.014 |
This | -0.014 |
time | -0.013 |
Token | found |
Token ID | 1043 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
forgot | +0.097 |
laughed | +0.069 |
ppa | +0.066 |
felt | +0.061 |
pretended | +0.059 |
Neg logit contributions | |
This | -0.215 |
purpose | -0.201 |
arl | -0.201 |
time | -0.200 |
Home | -0.198 |
Token | out |
Token ID | 503 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
arl | +0.030 |
purpose | +0.029 |
Hawaii | +0.026 |
Ireland | +0.025 |
This | +0.025 |
Neg logit contributions | |
forgot | -0.041 |
felt | -0.037 |
laughed | -0.035 |
got | -0.034 |
knew | -0.034 |
Token | why |
Token ID | 1521 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
arl | +0.017 |
purpose | +0.017 |
Hawaii | +0.016 |
Ireland | +0.015 |
This | +0.015 |
Neg logit contributions | |
forgot | -0.021 |
felt | -0.019 |
laughed | -0.018 |
got | -0.017 |
knew | -0.017 |
Token | she |
Token ID | 673 |
Feature activation | +6.63e+00 |
Pos logit contributions | |
arl | +0.043 |
purpose | +0.041 |
Hawaii | +0.039 |
Ireland | +0.038 |
outer | +0.036 |
Neg logit contributions | |
forgot | -0.042 |
felt | -0.036 |
got | -0.034 |
laughed | -0.034 |
knew | -0.033 |
Token | was |
Token ID | 373 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
forgot | +0.075 |
ppa | +0.073 |
ppy | +0.059 |
ba | +0.053 |
my | +0.053 |
Neg logit contributions | |
purpose | -0.167 |
This | -0.162 |
time | -0.162 |
Can | -0.155 |
history | -0.155 |
Token | excited |
Token ID | 6568 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
forgot | +0.002 |
felt | +0.001 |
laughed | +0.001 |
got | +0.001 |
knew | +0.001 |
Neg logit contributions | |
arl | -0.002 |
purpose | -0.002 |
Hawaii | -0.002 |
Ireland | -0.002 |
This | -0.002 |
Token | that |
Token ID | 326 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
arl | +0.045 |
purpose | +0.044 |
Hawaii | +0.040 |
This | +0.038 |
Ireland | +0.038 |
Neg logit contributions | |
forgot | -0.074 |
felt | -0.065 |
laughed | -0.062 |
knew | -0.061 |
got | -0.061 |
Token | he |
Token ID | 339 |
Feature activation | +7.90e+00 |
Pos logit contributions | |
arl | +0.015 |
purpose | +0.015 |
Hawaii | +0.014 |
Ireland | +0.013 |
This | +0.013 |
Neg logit contributions | |
forgot | -0.010 |
felt | -0.008 |
laughed | -0.008 |
got | -0.008 |
knew | -0.008 |
Token | ran |
Token ID | 4966 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
my | +0.084 |
ppa | +0.083 |
ie | +0.068 |
any | +0.068 |
ppy | +0.068 |
Neg logit contributions | |
Can | -0.198 |
time | -0.187 |
purpose | -0.184 |
our | -0.182 |
Earth | -0.180 |
Token | around |
Token ID | 1088 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
arl | +0.050 |
purpose | +0.050 |
Hawaii | +0.046 |
This | +0.045 |
Ireland | +0.044 |
Neg logit contributions | |
forgot | -0.060 |
felt | -0.052 |
laughed | -0.049 |
knew | -0.048 |
got | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
arl | +0.065 |
purpose | +0.062 |
Hawaii | +0.058 |
Ireland | +0.056 |
This | +0.055 |
Neg logit contributions | |
forgot | -0.061 |
felt | -0.053 |
laughed | -0.051 |
got | -0.049 |
knew | -0.049 |
Token | around |
Token ID | 1088 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
ppa | +0.053 |
ppy | +0.047 |
forgot | +0.047 |
ie | +0.044 |
i | +0.042 |
Neg logit contributions | |
This | -0.083 |
Do | -0.078 |
Can | -0.078 |
Earth | -0.076 |
sight | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
arl | +0.048 |
purpose | +0.043 |
Hawaii | +0.041 |
Ireland | +0.040 |
ay | +0.039 |
Neg logit contributions | |
forgot | -0.041 |
felt | -0.037 |
laughed | -0.035 |
knew | -0.034 |
got | -0.033 |
Token | He |
Token ID | 679 |
Feature activation | +8.03e+00 |
Pos logit contributions | |
arl | +0.066 |
purpose | +0.063 |
Hawaii | +0.060 |
outer | +0.058 |
Ireland | +0.056 |
Neg logit contributions | |
forgot | -0.068 |
felt | -0.062 |
got | -0.059 |
laughed | -0.058 |
started | -0.058 |
Token | ran |
Token ID | 4966 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
ppa | +0.126 |
ppers | +0.105 |
ppy | +0.102 |
pper | +0.099 |
anut | +0.096 |
Neg logit contributions | |
Can | -0.158 |
Earth | -0.157 |
Sometimes | -0.155 |
time | -0.153 |
Home | -0.153 |
Token | so |
Token ID | 523 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
purpose | +0.052 |
arl | +0.052 |
Hawaii | +0.048 |
This | +0.048 |
sight | +0.047 |
Neg logit contributions | |
forgot | -0.059 |
felt | -0.050 |
ppa | -0.048 |
laughed | -0.047 |
knew | -0.046 |
Token | of |
Token ID | 286 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
arl | +0.111 |
purpose | +0.107 |
Hawaii | +0.099 |
outer | +0.097 |
pleasure | +0.096 |
Neg logit contributions | |
forgot | -0.095 |
felt | -0.089 |
got | -0.085 |
had | -0.084 |
knew | -0.083 |
Token | the |
Token ID | 262 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
arl | +0.003 |
purpose | +0.003 |
Hawaii | +0.003 |
Ireland | +0.003 |
This | +0.003 |
Neg logit contributions | |
forgot | -0.005 |
felt | -0.004 |
laughed | -0.004 |
knew | -0.004 |
got | -0.004 |
Token | story |
Token ID | 1621 |
Feature activation | -8.32e-01 |
Pos logit contributions | |
forgot | +0.005 |
felt | +0.004 |
laughed | +0.004 |
had | +0.004 |
knew | +0.004 |
Neg logit contributions | |
arl | -0.004 |
purpose | -0.003 |
outer | -0.003 |
Hawaii | -0.003 |
Ireland | -0.003 |
Token | is |
Token ID | 318 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
arl | +0.021 |
outer | +0.021 |
purpose | +0.021 |
ocol | +0.021 |
pleasure | +0.020 |
Neg logit contributions | |
forgot | -0.014 |
felt | -0.013 |
had | -0.012 |
knew | -0.012 |
got | -0.012 |
Token | : |
Token ID | 25 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
arl | +0.033 |
purpose | +0.033 |
outer | +0.032 |
Hawaii | +0.032 |
pleasure | +0.031 |
Neg logit contributions | |
forgot | -0.029 |
knew | -0.025 |
felt | -0.025 |
had | -0.024 |
got | -0.024 |
Token | We |
Token ID | 775 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
arl | +0.027 |
purpose | +0.025 |
sight | +0.023 |
Hawaii | +0.022 |
outer | +0.022 |
Neg logit contributions | |
forgot | -0.052 |
felt | -0.048 |
's | -0.047 |
my | -0.047 |
knew | -0.047 |
Token | should |
Token ID | 815 |
Feature activation | -9.57e-01 |
Pos logit contributions | |
forgot | +0.070 |
ppa | +0.060 |
felt | +0.060 |
laughed | +0.059 |
ie | +0.055 |
Neg logit contributions | |
purpose | -0.062 |
arl | -0.059 |
This | -0.058 |
Earth | -0.055 |
Hawaii | -0.055 |
Token | be |
Token ID | 307 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
purpose | +0.022 |
arl | +0.022 |
Hawaii | +0.020 |
This | +0.020 |
Ireland | +0.020 |
Neg logit contributions | |
forgot | -0.020 |
felt | -0.017 |
laughed | -0.016 |
got | -0.015 |
knew | -0.015 |
Token | kind |
Token ID | 1611 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
arl | +0.024 |
purpose | +0.024 |
Hawaii | +0.023 |
This | +0.022 |
Ireland | +0.022 |
Neg logit contributions | |
forgot | -0.027 |
felt | -0.023 |
laughed | -0.022 |
knew | -0.021 |
got | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
arl | +0.102 |
purpose | +0.098 |
Hawaii | +0.095 |
outer | +0.092 |
Ireland | +0.092 |
Neg logit contributions | |
forgot | -0.089 |
felt | -0.077 |
laughed | -0.071 |
knew | -0.071 |
got | -0.071 |
Token | respectful |
Token ID | 25923 |
Feature activation | -4.94e+00 |
Pos logit contributions | |
forgot | +0.013 |
felt | +0.011 |
laughed | +0.010 |
knew | +0.010 |
got | +0.010 |
Neg logit contributions | |
arl | -0.012 |
purpose | -0.012 |
Hawaii | -0.011 |
This | -0.011 |
Ireland | -0.011 |
Token | there |
Token ID | 612 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
purpose | +0.025 |
arl | +0.024 |
Hawaii | +0.022 |
Ireland | +0.021 |
sight | +0.020 |
Neg logit contributions | |
forgot | -0.028 |
felt | -0.025 |
had | -0.024 |
my | -0.023 |
laughed | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
forgot | +0.018 |
felt | +0.012 |
laughed | +0.011 |
ppa | +0.010 |
got | +0.010 |
Neg logit contributions | |
purpose | -0.052 |
arl | -0.051 |
Hawaii | -0.049 |
This | -0.048 |
Ireland | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | -6.05e-01 |
Pos logit contributions | |
arl | +0.054 |
purpose | +0.048 |
Hawaii | +0.044 |
Ireland | +0.041 |
This | +0.041 |
Neg logit contributions | |
forgot | -0.056 |
felt | -0.052 |
laughed | -0.051 |
had | -0.050 |
my | -0.049 |
Token | kind |
Token ID | 1611 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
arl | +0.012 |
purpose | +0.011 |
Hawaii | +0.011 |
Ireland | +0.010 |
This | +0.010 |
Neg logit contributions | |
forgot | -0.015 |
felt | -0.013 |
laughed | -0.012 |
got | -0.012 |
struggled | -0.012 |
Token | taxi |
Token ID | 17536 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
arl | +0.039 |
purpose | +0.038 |
Hawaii | +0.036 |
Ireland | +0.034 |
This | +0.034 |
Neg logit contributions | |
forgot | -0.037 |
felt | -0.032 |
laughed | -0.030 |
knew | -0.029 |
got | -0.029 |
Token | driver |
Token ID | 4639 |
Feature activation | +4.57e+00 |
Pos logit contributions | |
forgot | +0.055 |
felt | +0.046 |
laughed | +0.043 |
knew | +0.041 |
struggled | +0.041 |
Neg logit contributions | |
arl | -0.059 |
purpose | -0.058 |
Hawaii | -0.055 |
This | -0.054 |
Ireland | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
forgot | +0.105 |
felt | +0.084 |
struggled | +0.081 |
laughed | +0.079 |
ppa | +0.079 |
Neg logit contributions | |
This | -0.090 |
purpose | -0.089 |
arl | -0.089 |
Hawaii | -0.087 |
Can | -0.084 |
Token | He |
Token ID | 679 |
Feature activation | +7.25e+00 |
Pos logit contributions | |
arl | +0.065 |
purpose | +0.064 |
ires | +0.056 |
uttering | +0.056 |
ags | +0.056 |
Neg logit contributions | |
forgot | -0.052 |
got | -0.050 |
had | -0.050 |
felt | -0.050 |
took | -0.049 |
Token | would |
Token ID | 561 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
forgot | +0.104 |
ppy | +0.083 |
ppa | +0.083 |
i | +0.073 |
my | +0.073 |
Neg logit contributions | |
This | -0.185 |
purpose | -0.183 |
Hawaii | -0.182 |
Can | -0.181 |
time | -0.180 |
Token | always |
Token ID | 1464 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
arl | +0.008 |
purpose | +0.007 |
Hawaii | +0.007 |
This | +0.007 |
Ireland | +0.007 |
Neg logit contributions | |
forgot | -0.007 |
felt | -0.006 |
ppa | -0.006 |
laughed | -0.006 |
struggled | -0.006 |
Token | have |
Token ID | 423 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
forgot | +0.051 |
felt | +0.043 |
ppa | +0.041 |
laughed | +0.041 |
ppy | +0.039 |
Neg logit contributions | |
purpose | -0.052 |
arl | -0.051 |
Hawaii | -0.049 |
outer | -0.047 |
This | -0.047 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.03e+00 |
Pos logit contributions | |
purpose | +0.083 |
arl | +0.079 |
ocol | +0.076 |
outer | +0.074 |
tornado | +0.071 |
Neg logit contributions | |
forgot | -0.064 |
had | -0.059 |
felt | -0.058 |
my | -0.054 |
took | -0.053 |
Token | Tom |
Token ID | 13787 |
Feature activation | +6.17e+00 |
Pos logit contributions | |
purpose | +0.110 |
arl | +0.103 |
pleasure | +0.095 |
Ireland | +0.092 |
Hawaii | +0.091 |
Neg logit contributions | |
my | -0.103 |
had | -0.103 |
felt | -0.102 |
got | -0.097 |
's | -0.096 |
Token | and |
Token ID | 290 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
forgot | +0.134 |
laughed | +0.110 |
ppa | +0.108 |
felt | +0.108 |
struggled | +0.105 |
Neg logit contributions | |
arl | -0.130 |
purpose | -0.122 |
This | -0.119 |
Hawaii | -0.118 |
Sometimes | -0.113 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.05e+00 |
Pos logit contributions | |
forgot | +0.044 |
felt | +0.037 |
laughed | +0.036 |
knew | +0.035 |
got | +0.035 |
Neg logit contributions | |
purpose | -0.033 |
arl | -0.033 |
Hawaii | -0.031 |
This | -0.030 |
Ireland | -0.029 |
Token | were |
Token ID | 547 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
forgot | +0.142 |
ppa | +0.130 |
felt | +0.109 |
laughed | +0.109 |
struggled | +0.107 |
Neg logit contributions | |
Hawaii | -0.180 |
arl | -0.177 |
Earth | -0.176 |
This | -0.176 |
purpose | -0.170 |
Token | twins |
Token ID | 20345 |
Feature activation | +5.58e+00 |
Pos logit contributions | |
arl | +0.009 |
purpose | +0.009 |
Hawaii | +0.008 |
This | +0.008 |
Ireland | +0.008 |
Neg logit contributions | |
forgot | -0.011 |
felt | -0.010 |
laughed | -0.009 |
got | -0.009 |
knew | -0.009 |
Token | who |
Token ID | 508 |
Feature activation | +5.73e+00 |
Pos logit contributions | |
forgot | +0.133 |
felt | +0.111 |
struggled | +0.107 |
laughed | +0.105 |
managed | +0.105 |
Neg logit contributions | |
purpose | -0.105 |
arl | -0.099 |
This | -0.097 |
Hawaii | -0.095 |
time | -0.093 |
Token | liked |
Token ID | 8288 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
forgot | +0.094 |
ppa | +0.092 |
ppy | +0.085 |
struggled | +0.079 |
ie | +0.076 |
Neg logit contributions | |
This | -0.132 |
time | -0.124 |
common | -0.123 |
purpose | -0.122 |
history | -0.119 |
Token | to |
Token ID | 284 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
arl | +0.091 |
purpose | +0.080 |
ay | +0.073 |
Hawaii | +0.073 |
Ireland | +0.072 |
Neg logit contributions | |
forgot | -0.101 |
felt | -0.093 |
laughed | -0.090 |
got | -0.089 |
took | -0.087 |
Token | go |
Token ID | 467 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
arl | +0.054 |
purpose | +0.054 |
Hawaii | +0.050 |
This | +0.048 |
Ireland | +0.048 |
Neg logit contributions | |
forgot | -0.053 |
felt | -0.045 |
laughed | -0.042 |
knew | -0.041 |
got | -0.041 |
Token | on |
Token ID | 319 |
Feature activation | -5.85e+00 |
Pos logit contributions | |
purpose | +0.113 |
arl | +0.111 |
Ireland | +0.096 |
Hawaii | +0.096 |
pleasure | +0.094 |
Neg logit contributions | |
forgot | -0.110 |
felt | -0.100 |
got | -0.096 |
my | -0.095 |
laughed | -0.093 |
Token | finally |
Token ID | 3443 |
Feature activation | +3.73e+00 |
Pos logit contributions | |
forgot | +0.093 |
felt | +0.074 |
ppa | +0.073 |
laughed | +0.069 |
my | +0.069 |
Neg logit contributions | |
purpose | -0.126 |
arl | -0.123 |
Hawaii | -0.118 |
Ireland | -0.114 |
This | -0.114 |
Token | , |
Token ID | 11 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
forgot | +0.050 |
ppa | +0.042 |
ppy | +0.037 |
felt | +0.037 |
my | +0.037 |
Neg logit contributions | |
purpose | -0.102 |
arl | -0.098 |
This | -0.097 |
Hawaii | -0.096 |
sight | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
forgot | +0.054 |
felt | +0.047 |
laughed | +0.045 |
knew | +0.044 |
got | +0.044 |
Neg logit contributions | |
purpose | -0.040 |
arl | -0.040 |
Hawaii | -0.037 |
This | -0.036 |
Ireland | -0.036 |
Token | license |
Token ID | 5964 |
Feature activation | +6.04e+00 |
Pos logit contributions | |
forgot | +0.063 |
felt | +0.055 |
laughed | +0.052 |
got | +0.051 |
knew | +0.051 |
Neg logit contributions | |
arl | -0.051 |
purpose | -0.051 |
Hawaii | -0.046 |
Ireland | -0.044 |
This | -0.044 |
Token | arrived |
Token ID | 5284 |
Feature activation | -5.75e-01 |
Pos logit contributions | |
forgot | +0.081 |
ppa | +0.069 |
ppy | +0.065 |
my | +0.059 |
i | +0.056 |
Neg logit contributions | |
purpose | -0.168 |
time | -0.159 |
Hawaii | -0.157 |
arl | -0.155 |
This | -0.154 |
Token | in |
Token ID | 287 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
arl | +0.013 |
purpose | +0.013 |
Hawaii | +0.012 |
Ireland | +0.012 |
This | +0.012 |
Neg logit contributions | |
forgot | -0.012 |
felt | -0.010 |
laughed | -0.010 |
got | -0.010 |
knew | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +7.91e-01 |
Pos logit contributions | |
forgot | +0.006 |
felt | +0.006 |
laughed | +0.006 |
had | +0.006 |
got | +0.005 |
Neg logit contributions | |
arl | -0.005 |
purpose | -0.005 |
Hawaii | -0.004 |
Ireland | -0.004 |
This | -0.004 |
Token | mail |
Token ID | 6920 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
forgot | +0.020 |
felt | +0.018 |
laughed | +0.017 |
got | +0.017 |
knew | +0.016 |
Neg logit contributions | |
arl | -0.015 |
purpose | -0.014 |
Hawaii | -0.013 |
Ireland | -0.013 |
This | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
arl | +0.008 |
purpose | +0.008 |
Hawaii | +0.007 |
This | +0.007 |
Ireland | +0.007 |
Neg logit contributions | |
forgot | -0.009 |
felt | -0.007 |
laughed | -0.007 |
knew | -0.007 |
got | -0.007 |
Token | It |
Token ID | 632 |
Feature activation | +6.50e+00 |
Pos logit contributions | |
arl | +0.005 |
purpose | +0.005 |
Hawaii | +0.004 |
outer | +0.004 |
Ireland | +0.004 |
Neg logit contributions | |
forgot | -0.007 |
felt | -0.006 |
laughed | -0.006 |
got | -0.006 |
my | -0.006 |
Token | felt |
Token ID | 2936 |
Feature activation | -4.30e-03 |
Pos logit contributions | |
ppa | +0.096 |
forgot | +0.088 |
ppy | +0.087 |
i | +0.072 |
anut | +0.070 |
Neg logit contributions | |
purpose | -0.150 |
time | -0.148 |
sight | -0.147 |
This | -0.145 |
Earth | -0.144 |
Token | Tim |
Token ID | 5045 |
Feature activation | +6.13e+00 |
Pos logit contributions | |
arl | +0.057 |
purpose | +0.054 |
Hawaii | +0.052 |
Ireland | +0.049 |
outer | +0.048 |
Neg logit contributions | |
forgot | -0.055 |
felt | -0.049 |
got | -0.046 |
laughed | -0.046 |
started | -0.044 |
Token | my |
Token ID | 1820 |
Feature activation | +8.00e+00 |
Pos logit contributions | |
forgot | +0.088 |
ppa | +0.081 |
laughed | +0.078 |
screamed | +0.074 |
mun | +0.073 |
Neg logit contributions | |
purpose | -0.128 |
ay | -0.126 |
Sunday | -0.125 |
Sometimes | -0.123 |
time | -0.122 |
Token | 's |
Token ID | 338 |
Feature activation | -7.77e-02 |
Pos logit contributions | |
ppa | +0.118 |
pper | +0.104 |
eling | +0.103 |
hetically | +0.101 |
ppy | +0.100 |
Neg logit contributions | |
time | -0.174 |
purpose | -0.170 |
This | -0.169 |
Sunday | -0.168 |
Can | -0.165 |
Token | mom |
Token ID | 1995 |
Feature activation | +7.24e+00 |
Pos logit contributions | |
arl | +0.002 |
purpose | +0.002 |
Hawaii | +0.002 |
Ireland | +0.002 |
outer | +0.002 |
Neg logit contributions | |
forgot | -0.002 |
felt | -0.001 |
laughed | -0.001 |
got | -0.001 |
knew | -0.001 |
Token | told |
Token ID | 1297 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
mo | +0.096 |
ppy | +0.096 |
ede | +0.095 |
eling | +0.095 |
anut | +0.094 |
Neg logit contributions | |
ay | -0.142 |
time | -0.140 |
Can | -0.133 |
here | -0.133 |
This | -0.132 |
Token | him |
Token ID | 683 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
arl | +0.069 |
purpose | +0.068 |
Hawaii | +0.063 |
Ireland | +0.060 |
This | +0.059 |
Neg logit contributions | |
forgot | -0.089 |
felt | -0.078 |
laughed | -0.074 |
got | -0.073 |
knew | -0.072 |
Token | to |
Token ID | 284 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
forgot | +0.008 |
felt | +0.007 |
laughed | +0.007 |
got | +0.007 |
knew | +0.007 |
Neg logit contributions | |
arl | -0.006 |
purpose | -0.006 |
Hawaii | -0.005 |
Ireland | -0.005 |
outer | -0.005 |
Token | clean |
Token ID | 3424 |
Feature activation | -5.17e+00 |
Pos logit contributions | |
purpose | +0.024 |
arl | +0.024 |
Hawaii | +0.022 |
This | +0.021 |
time | +0.021 |
Neg logit contributions | |
forgot | -0.023 |
felt | -0.019 |
laughed | -0.018 |
knew | -0.018 |
ppa | -0.018 |
Token | up |
Token ID | 510 |
Feature activation | -4.33e+00 |
Pos logit contributions | |
arl | +0.117 |
purpose | +0.110 |
Hawaii | +0.105 |
Ireland | +0.102 |
This | +0.099 |
Neg logit contributions | |
forgot | -0.112 |
felt | -0.098 |
laughed | -0.092 |
got | -0.091 |
my | -0.090 |
Token | his |
Token ID | 465 |
Feature activation | +9.65e-01 |
Pos logit contributions | |
arl | +0.099 |
purpose | +0.093 |
Hawaii | +0.091 |
Ireland | +0.088 |
pleasure | +0.085 |
Neg logit contributions | |
forgot | -0.092 |
felt | -0.081 |
laughed | -0.075 |
got | -0.075 |
knew | -0.072 |
Token | toys |
Token ID | 14958 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
forgot | +0.022 |
felt | +0.018 |
laughed | +0.017 |
knew | +0.017 |
got | +0.017 |
Neg logit contributions | |
purpose | -0.021 |
arl | -0.021 |
Hawaii | -0.019 |
This | -0.019 |
Ireland | -0.019 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
arl | +0.001 |
purpose | +0.001 |
Hawaii | +0.001 |
Ireland | +0.001 |
This | +0.001 |
Neg logit contributions | |
forgot | -0.001 |
felt | -0.001 |
laughed | -0.001 |
got | -0.001 |
knew | -0.001 |
Token | bracelet |
Token ID | 42893 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
forgot | +0.012 |
felt | +0.010 |
laughed | +0.010 |
got | +0.010 |
knew | +0.010 |
Neg logit contributions | |
arl | -0.013 |
purpose | -0.012 |
Hawaii | -0.011 |
Ireland | -0.011 |
This | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
forgot | +0.054 |
ppa | +0.047 |
ppy | +0.044 |
my | +0.043 |
felt | +0.043 |
Neg logit contributions | |
This | -0.038 |
time | -0.038 |
purpose | -0.038 |
ay | -0.036 |
Can | -0.035 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.62e+00 |
Pos logit contributions | |
arl | +0.052 |
purpose | +0.051 |
Hawaii | +0.049 |
outer | +0.048 |
Ireland | +0.046 |
Neg logit contributions | |
forgot | -0.061 |
felt | -0.057 |
got | -0.055 |
had | -0.055 |
took | -0.053 |
Token | loved |
Token ID | 6151 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
ppa | +0.151 |
mo | +0.150 |
ppy | +0.143 |
pper | +0.140 |
forgot | +0.134 |
Neg logit contributions | |
This | -0.171 |
Can | -0.168 |
A | -0.158 |
purpose | -0.158 |
time | -0.152 |
Token | it |
Token ID | 340 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
arl | +0.066 |
purpose | +0.061 |
Hawaii | +0.057 |
Ireland | +0.055 |
This | +0.053 |
Neg logit contributions | |
forgot | -0.082 |
felt | -0.074 |
laughed | -0.070 |
knew | -0.069 |
got | -0.069 |
Token | so |
Token ID | 523 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
purpose | +0.012 |
arl | +0.011 |
This | +0.011 |
time | +0.010 |
sight | +0.010 |
Neg logit contributions | |
forgot | -0.012 |
felt | -0.010 |
ppa | -0.010 |
laughed | -0.009 |
struggled | -0.009 |
Token | much |
Token ID | 881 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
arl | +0.013 |
purpose | +0.013 |
Hawaii | +0.011 |
This | +0.011 |
Ireland | +0.011 |
Neg logit contributions | |
forgot | -0.031 |
felt | -0.028 |
laughed | -0.027 |
got | -0.026 |
knew | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
arl | +0.068 |
purpose | +0.064 |
Hawaii | +0.062 |
Ireland | +0.060 |
outer | +0.058 |
Neg logit contributions | |
forgot | -0.069 |
felt | -0.059 |
got | -0.057 |
knew | -0.056 |
laughed | -0.056 |
Token | she |
Token ID | 673 |
Feature activation | +8.18e+00 |
Pos logit contributions | |
forgot | +0.008 |
felt | +0.007 |
struggled | +0.007 |
laughed | +0.007 |
ppa | +0.007 |
Neg logit contributions | |
purpose | -0.008 |
arl | -0.008 |
This | -0.008 |
Can | -0.008 |
sight | -0.007 |
Token | wore |
Token ID | 12408 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
mo | +0.105 |
hetically | +0.103 |
my | +0.099 |
ppy | +0.097 |
any | +0.096 |
Neg logit contributions | |
Can | -0.179 |
your | -0.161 |
time | -0.159 |
birthday | -0.158 |
Yes | -0.158 |
Token | and |
Token ID | 290 |
Feature activation | +5.11e+00 |
Pos logit contributions | |
purpose | +0.016 |
arl | +0.015 |
sight | +0.014 |
Hawaii | +0.014 |
This | +0.014 |
Neg logit contributions | |
forgot | -0.020 |
my | -0.016 |
felt | -0.016 |
ppa | -0.016 |
ppy | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
forgot | +0.088 |
ppa | +0.082 |
ppy | +0.082 |
my | +0.078 |
nie | +0.071 |
Neg logit contributions | |
purpose | -0.104 |
sight | -0.103 |
This | -0.101 |
Hawaii | -0.101 |
Earth | -0.100 |
Token | bear |
Token ID | 6842 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
forgot | +0.037 |
felt | +0.030 |
laughed | +0.029 |
got | +0.028 |
ppa | +0.028 |
Neg logit contributions | |
purpose | -0.039 |
arl | -0.039 |
sight | -0.036 |
Hawaii | -0.036 |
Ireland | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
ppa | +0.132 |
forgot | +0.130 |
ppy | +0.125 |
my | +0.123 |
nie | +0.115 |
Neg logit contributions | |
time | -0.270 |
purpose | -0.263 |
sight | -0.255 |
This | -0.254 |
Earth | -0.243 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
forgot | +0.064 |
felt | +0.053 |
pretended | +0.051 |
laughed | +0.051 |
ppa | +0.050 |
Neg logit contributions | |
purpose | -0.057 |
arl | -0.056 |
Hawaii | -0.054 |
sight | -0.051 |
This | -0.051 |
Token | to |
Token ID | 284 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
purpose | +0.003 |
arl | +0.003 |
Hawaii | +0.002 |
This | +0.002 |
Ireland | +0.002 |
Neg logit contributions | |
forgot | -0.004 |
felt | -0.003 |
laughed | -0.003 |
got | -0.003 |
ppa | -0.003 |
Token | have |
Token ID | 423 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
forgot | +0.012 |
felt | +0.010 |
laughed | +0.009 |
got | +0.009 |
knew | +0.009 |
Neg logit contributions | |
arl | -0.014 |
purpose | -0.014 |
Hawaii | -0.013 |
Ireland | -0.013 |
This | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
purpose | +0.036 |
arl | +0.035 |
Hawaii | +0.034 |
This | +0.032 |
Ireland | +0.032 |
Neg logit contributions | |
forgot | -0.030 |
felt | -0.025 |
laughed | -0.024 |
my | -0.023 |
knew | -0.023 |
Token | shampoo |
Token ID | 47259 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
forgot | +0.010 |
felt | +0.009 |
laughed | +0.008 |
got | +0.008 |
knew | +0.008 |
Neg logit contributions | |
arl | -0.007 |
purpose | -0.007 |
Hawaii | -0.006 |
Ireland | -0.006 |
This | -0.006 |
Token | all |
Token ID | 477 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
forgot | +0.033 |
felt | +0.027 |
laughed | +0.026 |
ppa | +0.026 |
struggled | +0.026 |
Neg logit contributions | |
purpose | -0.026 |
arl | -0.025 |
Hawaii | -0.023 |
time | -0.023 |
This | -0.023 |
Token | for |
Token ID | 329 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
forgot | +0.041 |
felt | +0.035 |
ppa | +0.034 |
laughed | +0.033 |
pretended | +0.033 |
Neg logit contributions | |
purpose | -0.029 |
arl | -0.028 |
Hawaii | -0.027 |
This | -0.027 |
time | -0.026 |
Token | Tim |
Token ID | 14967 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
purpose | +0.049 |
arl | +0.049 |
Hawaii | +0.044 |
This | +0.043 |
time | +0.043 |
Neg logit contributions | |
forgot | -0.057 |
felt | -0.049 |
laughed | -0.046 |
got | -0.045 |
knew | -0.045 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.37e+00 |
Pos logit contributions | |
ge | +0.164 |
inge | +0.164 |
eling | +0.163 |
andon | +0.160 |
ppa | +0.159 |
Neg logit contributions | |
The | -0.180 |
A | -0.175 |
It | -0.161 |
This | -0.161 |
Mom | -0.155 |
Token | to |
Token ID | 284 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
arl | +0.053 |
purpose | +0.051 |
Hawaii | +0.046 |
Ireland | +0.044 |
This | +0.043 |
Neg logit contributions | |
forgot | -0.097 |
felt | -0.086 |
laughed | -0.083 |
got | -0.081 |
knew | -0.081 |
Token | help |
Token ID | 1037 |
Feature activation | -3.30e+00 |
Pos logit contributions | |
purpose | +0.025 |
arl | +0.024 |
This | +0.022 |
Hawaii | +0.021 |
Ireland | +0.021 |
Neg logit contributions | |
forgot | -0.036 |
felt | -0.031 |
ppa | -0.030 |
laughed | -0.030 |
struggled | -0.029 |
Token | his |
Token ID | 465 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
purpose | +0.059 |
arl | +0.054 |
This | +0.050 |
Hawaii | +0.049 |
time | +0.047 |
Neg logit contributions | |
forgot | -0.091 |
felt | -0.076 |
laughed | -0.074 |
struggled | -0.073 |
knew | -0.071 |
Token | friend |
Token ID | 1545 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
forgot | +0.039 |
felt | +0.033 |
laughed | +0.032 |
got | +0.031 |
ppa | +0.031 |
Neg logit contributions | |
purpose | -0.033 |
arl | -0.031 |
Hawaii | -0.029 |
Ireland | -0.028 |
sight | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
forgot | +0.061 |
ppa | +0.052 |
felt | +0.051 |
struggled | +0.050 |
knew | +0.050 |
Neg logit contributions | |
purpose | -0.032 |
This | -0.031 |
birthday | -0.027 |
Hawaii | -0.026 |
time | -0.026 |
Token | so |
Token ID | 523 |
Feature activation | +8.64e-01 |
Pos logit contributions | |
forgot | +0.060 |
ppa | +0.052 |
felt | +0.049 |
struggled | +0.049 |
ppy | +0.048 |
Neg logit contributions | |
purpose | -0.035 |
This | -0.035 |
Do | -0.034 |
Can | -0.033 |
pleasure | -0.032 |
Token | he |
Token ID | 339 |
Feature activation | +9.29e+00 |
Pos logit contributions | |
forgot | +0.019 |
felt | +0.016 |
ppa | +0.016 |
ppy | +0.015 |
struggled | +0.015 |
Neg logit contributions | |
purpose | -0.017 |
arl | -0.016 |
time | -0.016 |
This | -0.015 |
sight | -0.015 |
Token | had |
Token ID | 550 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
eling | +0.121 |
any | +0.115 |
mun | +0.115 |
inge | +0.111 |
ge | +0.110 |
Neg logit contributions | |
A | -0.181 |
Can | -0.178 |
Bang | -0.176 |
Please | -0.172 |
Inside | -0.170 |
Token | an |
Token ID | 281 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
arl | +0.000 |
purpose | +0.000 |
Hawaii | +0.000 |
Ireland | +0.000 |
ay | +0.000 |
Neg logit contributions | |
forgot | -0.001 |
felt | -0.001 |
laughed | -0.001 |
got | -0.001 |
knew | -0.001 |
Token | so |
Token ID | 523 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
purpose | +0.012 |
time | +0.011 |
This | +0.011 |
ay | +0.011 |
arl | +0.010 |
Neg logit contributions | |
forgot | -0.015 |
struggled | -0.012 |
laughed | -0.012 |
screamed | -0.012 |
pretended | -0.012 |
Token | it |
Token ID | 340 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
arl | +0.039 |
purpose | +0.038 |
Hawaii | +0.036 |
Ireland | +0.035 |
This | +0.033 |
Neg logit contributions | |
forgot | -0.046 |
felt | -0.041 |
knew | -0.038 |
laughed | -0.038 |
got | -0.037 |
Token | wouldn |
Token ID | 3636 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
ppa | +0.041 |
forgot | +0.039 |
ppy | +0.034 |
laughed | +0.033 |
my | +0.033 |
Neg logit contributions | |
time | -0.089 |
purpose | -0.088 |
Earth | -0.082 |
This | -0.080 |
Sometimes | -0.079 |
Token | 't |
Token ID | 470 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
arl | +0.044 |
purpose | +0.044 |
This | +0.038 |
Hawaii | +0.038 |
ay | +0.035 |
Neg logit contributions | |
forgot | -0.097 |
felt | -0.085 |
laughed | -0.084 |
knew | -0.081 |
pretended | -0.081 |
Token | become |
Token ID | 1716 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
purpose | +0.000 |
arl | +0.000 |
Hawaii | +0.000 |
birthday | +0.000 |
sight | +0.000 |
Neg logit contributions | |
forgot | -0.000 |
laughed | -0.000 |
felt | -0.000 |
ppa | -0.000 |
pretended | -0.000 |
Token | dull |
Token ID | 19222 |
Feature activation | -4.09e+00 |
Pos logit contributions | |
arl | +0.050 |
purpose | +0.050 |
Hawaii | +0.045 |
Ireland | +0.043 |
This | +0.043 |
Neg logit contributions | |
forgot | -0.079 |
felt | -0.069 |
laughed | -0.066 |
got | -0.065 |
knew | -0.065 |
Token | like |
Token ID | 588 |
Feature activation | -3.77e+00 |
Pos logit contributions | |
arl | +0.087 |
purpose | +0.085 |
Hawaii | +0.079 |
Ireland | +0.076 |
This | +0.076 |
Neg logit contributions | |
forgot | -0.094 |
felt | -0.081 |
laughed | -0.077 |
got | -0.075 |
knew | -0.074 |
Token | her |
Token ID | 607 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
arl | +0.076 |
purpose | +0.071 |
Hawaii | +0.067 |
Ireland | +0.065 |
This | +0.064 |
Neg logit contributions | |
forgot | -0.090 |
felt | -0.080 |
laughed | -0.076 |
got | -0.075 |
knew | -0.075 |
Token | old |
Token ID | 1468 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
forgot | +0.009 |
felt | +0.008 |
laughed | +0.007 |
knew | +0.007 |
got | +0.007 |
Neg logit contributions | |
arl | -0.008 |
purpose | -0.008 |
Hawaii | -0.007 |
Ireland | -0.007 |
This | -0.007 |
Token | one |
Token ID | 530 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
forgot | +0.003 |
felt | +0.002 |
laughed | +0.002 |
got | +0.002 |
knew | +0.002 |
Neg logit contributions | |
purpose | -0.004 |
arl | -0.003 |
Hawaii | -0.003 |
sight | -0.003 |
This | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
forgot | +0.049 |
felt | +0.042 |
ppa | +0.040 |
laughed | +0.040 |
pretended | +0.038 |
Neg logit contributions | |
purpose | -0.043 |
arl | -0.040 |
time | -0.039 |
This | -0.039 |
Hawaii | -0.038 |
Token | wonderful |
Token ID | 7932 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
purpose | +0.031 |
arl | +0.028 |
Hawaii | +0.028 |
sight | +0.026 |
Ireland | +0.026 |
Neg logit contributions | |
forgot | -0.058 |
felt | -0.049 |
ppy | -0.048 |
ppa | -0.047 |
laughed | -0.047 |
Token | toys |
Token ID | 14958 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
purpose | +0.017 |
arl | +0.016 |
Hawaii | +0.015 |
Ireland | +0.014 |
sight | +0.013 |
Neg logit contributions | |
forgot | -0.039 |
felt | -0.034 |
laughed | -0.033 |
knew | -0.033 |
ppa | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
purpose | +0.025 |
arl | +0.025 |
Hawaii | +0.024 |
This | +0.023 |
sight | +0.023 |
Neg logit contributions | |
forgot | -0.031 |
felt | -0.026 |
ppa | -0.025 |
laughed | -0.024 |
knew | -0.024 |
Token | sweets |
Token ID | 42402 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
forgot | +0.050 |
ppy | +0.047 |
ppa | +0.047 |
felt | +0.042 |
struggled | +0.042 |
Neg logit contributions | |
purpose | -0.040 |
Hawaii | -0.035 |
birthday | -0.035 |
history | -0.035 |
outer | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
purpose | +0.010 |
arl | +0.010 |
Hawaii | +0.009 |
sight | +0.009 |
time | +0.009 |
Neg logit contributions | |
forgot | -0.015 |
felt | -0.013 |
ppa | -0.013 |
laughed | -0.012 |
struggled | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
arl | +0.070 |
purpose | +0.067 |
Hawaii | +0.061 |
outer | +0.060 |
Ireland | +0.059 |
Neg logit contributions | |
forgot | -0.064 |
felt | -0.056 |
had | -0.054 |
laughed | -0.052 |
got | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
arl | +0.065 |
purpose | +0.064 |
Hawaii | +0.057 |
Ireland | +0.056 |
outer | +0.054 |
Neg logit contributions | |
forgot | -0.087 |
felt | -0.078 |
laughed | -0.074 |
knew | -0.073 |
got | -0.073 |
Token | Jack |
Token ID | 14295 |
Feature activation | +9.04e+00 |
Pos logit contributions | |
arl | +0.056 |
purpose | +0.055 |
Hawaii | +0.051 |
Ireland | +0.049 |
This | +0.048 |
Neg logit contributions | |
forgot | -0.070 |
felt | -0.061 |
laughed | -0.058 |
got | -0.056 |
knew | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
forgot | +0.148 |
ppa | +0.139 |
ppy | +0.125 |
ba | +0.121 |
struggled | +0.118 |
Neg logit contributions | |
This | -0.175 |
purpose | -0.165 |
ay | -0.161 |
ummy | -0.160 |
Earth | -0.159 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +9.51e+00 |
Pos logit contributions | |
forgot | +0.065 |
felt | +0.056 |
ppa | +0.055 |
laughed | +0.054 |
knew | +0.053 |
Neg logit contributions | |
purpose | -0.029 |
arl | -0.027 |
Hawaii | -0.025 |
This | -0.025 |
Earth | -0.023 |
Token | were |
Token ID | 547 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
ppa | +0.144 |
ches | +0.132 |
ba | +0.127 |
forgot | +0.125 |
ppy | +0.125 |
Neg logit contributions | |
New | -0.178 |
birthday | -0.176 |
This | -0.175 |
pleasure | -0.173 |
Earth | -0.167 |
Token | hop |
Token ID | 1725 |
Feature activation | -5.58e+00 |
Pos logit contributions | |
purpose | +0.046 |
arl | +0.046 |
Hawaii | +0.042 |
This | +0.041 |
Ireland | +0.041 |
Neg logit contributions | |
forgot | -0.044 |
felt | -0.037 |
laughed | -0.035 |
knew | -0.034 |
got | -0.034 |
Token | around |
Token ID | 1088 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
arl | +0.136 |
purpose | +0.133 |
Hawaii | +0.126 |
Ireland | +0.121 |
This | +0.119 |
Neg logit contributions | |
forgot | -0.111 |
felt | -0.095 |
laughed | -0.089 |
got | -0.088 |
knew | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
arl | +0.100 |
purpose | +0.093 |
Hawaii | +0.090 |
Ireland | +0.087 |
This | +0.084 |
Neg logit contributions | |
forgot | -0.065 |
felt | -0.058 |
laughed | -0.053 |
got | -0.051 |
knew | -0.051 |
Token | garden |
Token ID | 11376 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
arl | +0.043 |
purpose | +0.040 |
Hawaii | +0.038 |
Ireland | +0.037 |
outer | +0.036 |
Neg logit contributions | |
forgot | -0.042 |
felt | -0.038 |
my | -0.035 |
got | -0.035 |
knew | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
purpose | +0.042 |
arl | +0.040 |
This | +0.038 |
Hawaii | +0.037 |
time | +0.037 |
Neg logit contributions | |
forgot | -0.057 |
felt | -0.047 |
ppa | -0.046 |
laughed | -0.045 |
knew | -0.044 |
Token | |
Token ID | 220 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
arl | +0.052 |
purpose | +0.051 |
Hawaii | +0.047 |
Ireland | +0.046 |
outer | +0.045 |
Neg logit contributions | |
forgot | -0.067 |
felt | -0.061 |
my | -0.059 |
got | -0.057 |
laughed | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
purpose | +0.079 |
arl | +0.078 |
Hawaii | +0.072 |
Ireland | +0.070 |
outer | +0.069 |
Neg logit contributions | |
forgot | -0.103 |
felt | -0.093 |
my | -0.090 |
knew | -0.086 |
got | -0.086 |
Token |
|
Token ID | 198 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
arl | +0.055 |
purpose | +0.054 |
Hawaii | +0.049 |
Ireland | +0.047 |
outer | +0.046 |
Neg logit contributions | |
forgot | -0.080 |
felt | -0.072 |
my | -0.068 |
laughed | -0.068 |
got | -0.067 |
Token | One |
Token ID | 3198 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
arl | +0.043 |
purpose | +0.043 |
Hawaii | +0.038 |
Ireland | +0.036 |
This | +0.036 |
Neg logit contributions | |
forgot | -0.082 |
felt | -0.073 |
laughed | -0.070 |
got | -0.068 |
knew | -0.068 |
Token | day |
Token ID | 1110 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
arl | +0.003 |
purpose | +0.003 |
Hawaii | +0.003 |
This | +0.003 |
Ireland | +0.003 |
Neg logit contributions | |
forgot | -0.004 |
felt | -0.004 |
laughed | -0.004 |
got | -0.004 |
knew | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
arl | +0.053 |
purpose | +0.052 |
Hawaii | +0.049 |
Ireland | +0.047 |
outer | +0.047 |
Neg logit contributions | |
forgot | -0.048 |
felt | -0.042 |
got | -0.040 |
laughed | -0.039 |
started | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
arl | +0.069 |
purpose | +0.067 |
Hawaii | +0.062 |
Ireland | +0.062 |
outer | +0.059 |
Neg logit contributions | |
forgot | -0.083 |
felt | -0.074 |
knew | -0.070 |
my | -0.070 |
got | -0.069 |
Token | Little |
Token ID | 22253 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
arl | +0.056 |
purpose | +0.055 |
Hawaii | +0.051 |
Ireland | +0.049 |
This | +0.048 |
Neg logit contributions | |
forgot | -0.068 |
felt | -0.059 |
laughed | -0.056 |
got | -0.055 |
knew | -0.055 |
Token | Johnny |
Token ID | 15470 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
forgot | +0.037 |
felt | +0.032 |
laughed | +0.030 |
knew | +0.030 |
ppa | +0.029 |
Neg logit contributions | |
purpose | -0.031 |
arl | -0.031 |
This | -0.028 |
Hawaii | -0.028 |
Ireland | -0.027 |
Token | wanted |
Token ID | 2227 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
ppa | +0.147 |
forgot | +0.132 |
ppy | +0.117 |
anut | +0.115 |
ba | +0.106 |
Neg logit contributions | |
Earth | -0.238 |
purpose | -0.237 |
This | -0.234 |
time | -0.233 |
Home | -0.222 |
Token | to |
Token ID | 284 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
arl | +0.082 |
purpose | +0.071 |
Hawaii | +0.067 |
Ireland | +0.064 |
ay | +0.062 |
Neg logit contributions | |
forgot | -0.111 |
felt | -0.100 |
laughed | -0.096 |
got | -0.094 |
knew | -0.094 |
Token | make |
Token ID | 787 |
Feature activation | -4.65e+00 |
Pos logit contributions | |
purpose | +0.058 |
arl | +0.056 |
Hawaii | +0.053 |
This | +0.051 |
Ireland | +0.050 |
Neg logit contributions | |
forgot | -0.057 |
felt | -0.048 |
laughed | -0.045 |
knew | -0.044 |
got | -0.044 |
Token | it |
Token ID | 340 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
arl | +0.083 |
purpose | +0.077 |
Hawaii | +0.071 |
Ireland | +0.068 |
This | +0.066 |
Neg logit contributions | |
forgot | -0.125 |
felt | -0.111 |
laughed | -0.106 |
got | -0.104 |
knew | -0.104 |
Token | even |
Token ID | 772 |
Feature activation | +6.95e-01 |
Pos logit contributions | |
arl | +0.038 |
purpose | +0.038 |
Hawaii | +0.035 |
Ireland | +0.034 |
This | +0.034 |
Neg logit contributions | |
forgot | -0.048 |
felt | -0.041 |
laughed | -0.039 |
knew | -0.038 |
got | -0.038 |
Token | better |
Token ID | 1365 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
forgot | +0.017 |
felt | +0.014 |
laughed | +0.014 |
ppa | +0.014 |
knew | +0.014 |
Neg logit contributions | |
purpose | -0.013 |
arl | -0.013 |
Hawaii | -0.012 |
Ireland | -0.012 |
outer | -0.012 |
Token | by |
Token ID | 416 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
arl | +0.074 |
purpose | +0.072 |
Hawaii | +0.067 |
Ireland | +0.064 |
This | +0.064 |
Neg logit contributions | |
forgot | -0.078 |
felt | -0.067 |
laughed | -0.063 |
got | -0.062 |
knew | -0.062 |
Token | flying |
Token ID | 7348 |
Feature activation | -4.98e+00 |
Pos logit contributions | |
arl | +0.054 |
purpose | +0.051 |
Hawaii | +0.049 |
Ireland | +0.048 |
ay | +0.047 |
Neg logit contributions | |
forgot | -0.049 |
felt | -0.042 |
laughed | -0.041 |
knew | -0.039 |
got | -0.039 |
Token | she |
Token ID | 673 |
Feature activation | +5.89e+00 |
Pos logit contributions | |
arl | +0.032 |
purpose | +0.030 |
Hawaii | +0.028 |
Ireland | +0.027 |
This | +0.027 |
Neg logit contributions | |
forgot | -0.032 |
felt | -0.027 |
laughed | -0.026 |
knew | -0.026 |
got | -0.025 |
Token | had |
Token ID | 550 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
ppa | +0.090 |
forgot | +0.074 |
shouted | +0.073 |
anut | +0.072 |
ppy | +0.071 |
Neg logit contributions | |
Can | -0.114 |
time | -0.110 |
pleasure | -0.108 |
Earth | -0.108 |
history | -0.107 |
Token | done |
Token ID | 1760 |
Feature activation | -5.67e+00 |
Pos logit contributions | |
purpose | +0.036 |
arl | +0.035 |
Hawaii | +0.033 |
This | +0.032 |
Ireland | +0.032 |
Neg logit contributions | |
forgot | -0.025 |
felt | -0.020 |
laughed | -0.019 |
my | -0.018 |
ppa | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
arl | +0.115 |
purpose | +0.089 |
Hawaii | +0.088 |
Ireland | +0.084 |
ay | +0.083 |
Neg logit contributions | |
forgot | -0.153 |
felt | -0.141 |
laughed | -0.133 |
knew | -0.131 |
got | -0.131 |
Token | good |
Token ID | 922 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
purpose | +0.013 |
arl | +0.013 |
Hawaii | +0.011 |
Ireland | +0.011 |
This | +0.011 |
Neg logit contributions | |
forgot | -0.030 |
felt | -0.026 |
laughed | -0.026 |
got | -0.025 |
my | -0.025 |
Token | job |
Token ID | 1693 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
purpose | +0.005 |
arl | +0.004 |
Hawaii | +0.004 |
This | +0.003 |
Ireland | +0.003 |
Neg logit contributions | |
forgot | -0.016 |
felt | -0.014 |
laughed | -0.014 |
got | -0.014 |
knew | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
arl | +0.078 |
purpose | +0.075 |
Hawaii | +0.070 |
Ireland | +0.068 |
This | +0.067 |
Neg logit contributions | |
forgot | -0.085 |
felt | -0.073 |
laughed | -0.069 |
got | -0.067 |
knew | -0.067 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
purpose | +0.059 |
arl | +0.058 |
Hawaii | +0.055 |
outer | +0.055 |
Ireland | +0.054 |
Neg logit contributions | |
forgot | -0.067 |
felt | -0.061 |
knew | -0.059 |
my | -0.058 |
got | -0.056 |
Token | M |
Token ID | 44 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
arl | +0.105 |
purpose | +0.103 |
Ireland | +0.093 |
Hawaii | +0.091 |
outer | +0.090 |
Neg logit contributions | |
forgot | -0.094 |
felt | -0.092 |
had | -0.090 |
my | -0.087 |
got | -0.085 |
Token | ama |
Token ID | 1689 |
Feature activation | +7.09e+00 |
Pos logit contributions | |
arl | +0.004 |
purpose | +0.004 |
Hawaii | +0.003 |
Ireland | +0.003 |
This | +0.003 |
Neg logit contributions | |
forgot | -0.007 |
felt | -0.006 |
laughed | -0.005 |
got | -0.005 |
knew | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
forgot | +0.136 |
ppa | +0.107 |
struggled | +0.102 |
felt | +0.099 |
laughed | +0.095 |
Neg logit contributions | |
This | -0.150 |
Can | -0.145 |
arl | -0.143 |
Earth | -0.139 |
ay | -0.139 |