Token | the |
Token ID | 262 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
ited | +0.007 |
eling | +0.007 |
irl | +0.007 |
eking | +0.007 |
uch | +0.007 |
Neg logit contributions | |
He | -0.005 |
They | -0.005 |
automobile | -0.004 |
sidewalk | -0.004 |
The | -0.004 |
Token | cars |
Token ID | 5006 |
Feature activation | +4.77e-01 |
Pos logit contributions | |
He | +0.038 |
They | +0.034 |
The | +0.030 |
automobile | +0.029 |
She | +0.029 |
Neg logit contributions | |
ited | -0.049 |
eling | -0.048 |
irl | -0.048 |
uch | -0.047 |
eking | -0.047 |
Token | stopped |
Token ID | 5025 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
He | +0.011 |
They | +0.010 |
automobile | +0.009 |
sidewalk | +0.009 |
The | +0.009 |
Neg logit contributions | |
ited | -0.012 |
eling | -0.012 |
irl | -0.012 |
eking | -0.011 |
if | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
ited | +0.050 |
eling | +0.050 |
irl | +0.049 |
eking | +0.048 |
if | +0.048 |
Neg logit contributions | |
He | -0.030 |
They | -0.026 |
automobile | -0.024 |
sidewalk | -0.023 |
The | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
He | +0.018 |
They | +0.016 |
automobile | +0.016 |
sidewalk | +0.015 |
The | +0.014 |
Neg logit contributions | |
eling | -0.019 |
ited | -0.019 |
irl | -0.018 |
eking | -0.018 |
re | -0.017 |
Token | pedestrian |
Token ID | 22382 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
He | +0.037 |
They | +0.033 |
The | +0.029 |
She | +0.028 |
der | +0.028 |
Neg logit contributions | |
ited | -0.052 |
eling | -0.051 |
irl | -0.051 |
eking | -0.050 |
uch | -0.050 |
Token | light |
Token ID | 1657 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
He | +0.229 |
They | +0.207 |
automobile | +0.191 |
sidewalk | +0.190 |
The | +0.187 |
Neg logit contributions | |
ited | -0.309 |
eling | -0.300 |
irl | -0.295 |
eking | -0.288 |
if | -0.287 |
Token | turned |
Token ID | 2900 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
He | +0.079 |
They | +0.070 |
automobile | +0.063 |
The | +0.063 |
sidewalk | +0.062 |
Neg logit contributions | |
ited | -0.123 |
eling | -0.121 |
irl | -0.120 |
eking | -0.116 |
uch | -0.115 |
Token | green |
Token ID | 4077 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
He | +0.085 |
They | +0.077 |
automobile | +0.070 |
sidewalk | +0.069 |
The | +0.069 |
Neg logit contributions | |
ited | -0.116 |
eling | -0.114 |
irl | -0.113 |
eking | -0.110 |
uch | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | +5.65e+00 |
Pos logit contributions | |
He | +0.063 |
They | +0.055 |
The | +0.052 |
automobile | +0.051 |
tr | +0.050 |
Neg logit contributions | |
ited | -0.070 |
eling | -0.068 |
re | -0.067 |
irl | -0.067 |
eking | -0.065 |
Token | Then |
Token ID | 3244 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
automobile | +0.082 |
sidewalk | +0.076 |
arrow | +0.076 |
He | +0.075 |
X | +0.075 |
Neg logit contributions | |
re | -0.164 |
irl | -0.164 |
ited | -0.161 |
if | -0.158 |
ins | -0.152 |
Token | wave |
Token ID | 6769 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
He | +0.027 |
They | +0.024 |
The | +0.022 |
automobile | +0.021 |
sidewalk | +0.021 |
Neg logit contributions | |
ited | -0.045 |
eling | -0.044 |
irl | -0.043 |
eking | -0.042 |
uch | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
He | +0.077 |
They | +0.071 |
automobile | +0.068 |
sidewalk | +0.068 |
tr | +0.066 |
Neg logit contributions | |
ited | -0.122 |
irl | -0.118 |
eling | -0.118 |
eking | -0.114 |
uch | -0.112 |
Token | spear |
Token ID | 17514 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
He | +0.017 |
They | +0.015 |
sidewalk | +0.014 |
automobile | +0.014 |
The | +0.013 |
Neg logit contributions | |
ited | -0.032 |
irl | -0.032 |
eling | -0.032 |
eking | -0.031 |
uch | -0.031 |
Token | in |
Token ID | 287 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
He | +0.044 |
They | +0.040 |
automobile | +0.037 |
sidewalk | +0.037 |
The | +0.036 |
Neg logit contributions | |
ited | -0.065 |
eling | -0.063 |
irl | -0.063 |
eking | -0.061 |
uch | -0.061 |
Token | an |
Token ID | 281 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
He | +0.016 |
They | +0.013 |
The | +0.012 |
It | +0.010 |
Tim | +0.010 |
Neg logit contributions | |
uch | -0.043 |
ins | -0.041 |
eking | -0.041 |
ited | -0.041 |
inations | -0.041 |
Token | attempt |
Token ID | 2230 |
Feature activation | +9.68e+00 |
Pos logit contributions | |
He | +0.036 |
They | +0.034 |
automobile | +0.031 |
The | +0.031 |
tr | +0.030 |
Neg logit contributions | |
ited | -0.036 |
eling | -0.035 |
irl | -0.034 |
eking | -0.033 |
uch | -0.032 |
Token | to |
Token ID | 284 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
He | +0.139 |
They | +0.122 |
automobile | +0.109 |
sidewalk | +0.105 |
The | +0.103 |
Neg logit contributions | |
ited | -0.328 |
eling | -0.325 |
irl | -0.322 |
eking | -0.316 |
if | -0.313 |
Token | scare |
Token ID | 19437 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
He | +0.035 |
They | +0.030 |
The | +0.028 |
sidewalk | +0.028 |
arrow | +0.026 |
Neg logit contributions | |
ited | -0.063 |
if | -0.060 |
ins | -0.060 |
eling | -0.060 |
uch | -0.060 |
Token | it |
Token ID | 340 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
He | +0.093 |
They | +0.084 |
automobile | +0.077 |
sidewalk | +0.076 |
The | +0.075 |
Neg logit contributions | |
ited | -0.129 |
eling | -0.128 |
irl | -0.126 |
eking | -0.122 |
uch | -0.121 |
Token | away |
Token ID | 1497 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
He | +0.013 |
They | +0.012 |
The | +0.010 |
automobile | +0.010 |
sidewalk | +0.010 |
Neg logit contributions | |
ited | -0.027 |
eling | -0.026 |
irl | -0.026 |
uch | -0.025 |
eking | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +6.06e+00 |
Pos logit contributions | |
He | +0.026 |
automobile | +0.025 |
They | +0.024 |
sidewalk | +0.023 |
tr | +0.022 |
Neg logit contributions | |
eling | -0.049 |
eking | -0.048 |
ited | -0.048 |
if | -0.048 |
irl | -0.048 |
Token | he |
Token ID | 339 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
ited | +0.051 |
irl | +0.050 |
eling | +0.050 |
uch | +0.049 |
eking | +0.048 |
Neg logit contributions | |
He | -0.030 |
They | -0.026 |
The | -0.023 |
automobile | -0.023 |
sidewalk | -0.023 |
Token | went |
Token ID | 1816 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
He | +0.039 |
They | +0.034 |
The | +0.031 |
automobile | +0.030 |
cop | +0.029 |
Neg logit contributions | |
ited | -0.050 |
irl | -0.050 |
eling | -0.048 |
eking | -0.048 |
uch | -0.047 |
Token | out |
Token ID | 503 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
He | +0.004 |
They | +0.003 |
sidewalk | +0.003 |
der | +0.003 |
tr | +0.003 |
Neg logit contributions | |
irl | -0.006 |
ited | -0.006 |
eling | -0.006 |
eking | -0.006 |
re | -0.006 |
Token | on |
Token ID | 319 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
He | +0.028 |
They | +0.024 |
The | +0.021 |
automobile | +0.021 |
sidewalk | +0.020 |
Neg logit contributions | |
ited | -0.058 |
eling | -0.057 |
irl | -0.057 |
uch | -0.056 |
eking | -0.055 |
Token | his |
Token ID | 465 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
He | +0.018 |
They | +0.016 |
The | +0.014 |
automobile | +0.014 |
sidewalk | +0.014 |
Neg logit contributions | |
ited | -0.032 |
eling | -0.032 |
irl | -0.031 |
uch | -0.031 |
eking | -0.030 |
Token | patrol |
Token ID | 13969 |
Feature activation | +9.40e+00 |
Pos logit contributions | |
He | +0.080 |
They | +0.072 |
automobile | +0.069 |
sidewalk | +0.067 |
The | +0.066 |
Neg logit contributions | |
ited | -0.089 |
eling | -0.086 |
irl | -0.085 |
uch | -0.082 |
eking | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | +6.71e+00 |
Pos logit contributions | |
He | +0.156 |
They | +0.138 |
automobile | +0.122 |
sidewalk | +0.122 |
tr | +0.121 |
Neg logit contributions | |
ited | -0.296 |
eling | -0.291 |
irl | -0.288 |
if | -0.280 |
eking | -0.278 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
automobile | +0.061 |
sidewalk | +0.059 |
tr | +0.056 |
arrow | +0.053 |
der | +0.052 |
Neg logit contributions | |
re | -0.225 |
if | -0.220 |
ited | -0.219 |
irl | -0.214 |
inations | -0.212 |
Token | , |
Token ID | 11 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
He | +0.059 |
They | +0.053 |
automobile | +0.048 |
sidewalk | +0.048 |
The | +0.047 |
Neg logit contributions | |
ited | -0.091 |
eling | -0.090 |
irl | -0.089 |
eking | -0.086 |
uch | -0.086 |
Token | he |
Token ID | 339 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
He | +0.005 |
They | +0.004 |
automobile | +0.004 |
sidewalk | +0.004 |
tr | +0.004 |
Neg logit contributions | |
ited | -0.009 |
eling | -0.009 |
irl | -0.009 |
eking | -0.009 |
if | -0.008 |
Token | heard |
Token ID | 2982 |
Feature activation | +4.96e+00 |
Pos logit contributions | |
He | +0.065 |
They | +0.055 |
The | +0.053 |
arrow | +0.051 |
tr | +0.051 |
Neg logit contributions | |
irl | -0.063 |
eling | -0.063 |
ited | -0.063 |
uch | -0.059 |
ins | -0.059 |
Token | ™ |
Token ID | 8151 |
Feature activation | -4.85e+00 |
Pos logit contributions | |
ited | +0.040 |
eling | +0.039 |
eking | +0.035 |
sie | +0.035 |
irl | +0.034 |
Neg logit contributions | |
He | -0.051 |
They | -0.049 |
The | -0.047 |
It | -0.045 |
sidewalk | -0.044 |
Token | s |
Token ID | 82 |
Feature activation | +3.16e+00 |
Pos logit contributions | |
eling | +0.112 |
irl | +0.110 |
ited | +0.109 |
eking | +0.106 |
inations | +0.100 |
Neg logit contributions | |
He | -0.137 |
They | -0.127 |
The | -0.118 |
She | -0.113 |
Tim | -0.111 |
Token | wing |
Token ID | 8539 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
He | +0.072 |
They | +0.067 |
The | +0.061 |
automobile | +0.060 |
sidewalk | +0.059 |
Neg logit contributions | |
ited | -0.087 |
eling | -0.083 |
irl | -0.081 |
uch | -0.079 |
eking | -0.079 |
Token | began |
Token ID | 2540 |
Feature activation | +6.22e+00 |
Pos logit contributions | |
He | +0.064 |
They | +0.058 |
The | +0.053 |
automobile | +0.052 |
sidewalk | +0.051 |
Neg logit contributions | |
ited | -0.078 |
eling | -0.077 |
irl | -0.076 |
eking | -0.073 |
inations | -0.072 |
Token | to |
Token ID | 284 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
He | +0.067 |
They | +0.054 |
The | +0.045 |
She | +0.038 |
sidewalk | +0.038 |
Neg logit contributions | |
ited | -0.247 |
eling | -0.243 |
irl | -0.240 |
uch | -0.237 |
inations | -0.229 |
Token | open |
Token ID | 1280 |
Feature activation | +9.27e+00 |
Pos logit contributions | |
He | +0.061 |
They | +0.056 |
automobile | +0.051 |
The | +0.051 |
sidewalk | +0.050 |
Neg logit contributions | |
ited | -0.076 |
eling | -0.074 |
irl | -0.073 |
eking | -0.071 |
uch | -0.071 |
Token | up |
Token ID | 510 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
He | +0.165 |
They | +0.146 |
automobile | +0.131 |
sidewalk | +0.131 |
The | +0.128 |
Neg logit contributions | |
ited | -0.288 |
eling | -0.284 |
irl | -0.282 |
eking | -0.273 |
if | -0.272 |
Token | . |
Token ID | 13 |
Feature activation | +8.13e+00 |
Pos logit contributions | |
He | +0.073 |
They | +0.063 |
The | +0.055 |
automobile | +0.054 |
sidewalk | +0.054 |
Neg logit contributions | |
ited | -0.151 |
eling | -0.148 |
irl | -0.147 |
uch | -0.144 |
eking | -0.143 |
Token |
|
Token ID | 198 |
Feature activation | -4.36e+00 |
Pos logit contributions | |
He | +0.115 |
automobile | +0.107 |
They | +0.103 |
sidewalk | +0.101 |
tr | +0.099 |
Neg logit contributions | |
ited | -0.256 |
irl | -0.255 |
eling | -0.254 |
re | -0.250 |
if | -0.250 |
Token |
|
Token ID | 198 |
Feature activation | -3.77e+00 |
Pos logit contributions | |
ited | +0.151 |
eling | +0.149 |
eking | +0.145 |
irl | +0.142 |
uch | +0.141 |
Neg logit contributions | |
He | -0.074 |
They | -0.064 |
The | -0.057 |
It | -0.052 |
She | -0.051 |
Token | The |
Token ID | 464 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
ited | +0.092 |
eling | +0.091 |
irl | +0.088 |
eking | +0.088 |
uch | +0.086 |
Neg logit contributions | |
He | -0.096 |
They | -0.088 |
The | -0.081 |
automobile | -0.079 |
sidewalk | -0.078 |
Token | â |
Token ID | 22940 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
He | +0.127 |
They | +0.110 |
automobile | +0.106 |
sidewalk | +0.105 |
tr | +0.098 |
Neg logit contributions | |
ited | -0.269 |
eling | -0.264 |
irl | -0.264 |
if | -0.257 |
re | -0.257 |
Token | € |
Token ID | 26391 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
He | +0.055 |
They | +0.051 |
automobile | +0.048 |
sidewalk | +0.047 |
The | +0.047 |
Neg logit contributions | |
ited | -0.043 |
eling | -0.041 |
irl | -0.040 |
uch | -0.039 |
eking | -0.039 |
Token | Baby |
Token ID | 14801 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
ited | +0.014 |
eling | +0.014 |
eking | +0.013 |
irl | +0.013 |
sie | +0.013 |
Neg logit contributions | |
He | -0.012 |
They | -0.011 |
The | -0.010 |
It | -0.010 |
She | -0.010 |
Token | wanted |
Token ID | 2227 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
ited | +0.027 |
irl | +0.027 |
eling | +0.027 |
eking | +0.026 |
inations | +0.026 |
Neg logit contributions | |
He | -0.025 |
They | -0.023 |
The | -0.021 |
automobile | -0.020 |
She | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
He | +0.029 |
They | +0.023 |
The | +0.021 |
Tim | +0.018 |
It | +0.018 |
Neg logit contributions | |
ited | -0.098 |
eling | -0.095 |
irl | -0.095 |
kins | -0.094 |
eking | -0.093 |
Token | open |
Token ID | 1280 |
Feature activation | +8.96e+00 |
Pos logit contributions | |
He | +0.039 |
They | +0.035 |
automobile | +0.032 |
sidewalk | +0.031 |
The | +0.031 |
Neg logit contributions | |
irl | -0.054 |
ited | -0.054 |
eling | -0.053 |
uch | -0.051 |
eking | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | +3.73e+00 |
Pos logit contributions | |
He | +0.173 |
They | +0.154 |
automobile | +0.139 |
The | +0.139 |
sidewalk | +0.136 |
Neg logit contributions | |
ited | -0.271 |
eling | -0.264 |
irl | -0.262 |
uch | -0.256 |
eking | -0.254 |
Token | box |
Token ID | 3091 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
He | +0.076 |
They | +0.070 |
The | +0.062 |
automobile | +0.061 |
sidewalk | +0.060 |
Neg logit contributions | |
ited | -0.103 |
eling | -0.102 |
irl | -0.101 |
uch | -0.099 |
eking | -0.099 |
Token | so |
Token ID | 523 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
He | +0.022 |
They | +0.020 |
automobile | +0.019 |
sidewalk | +0.019 |
The | +0.018 |
Neg logit contributions | |
ited | -0.032 |
eling | -0.031 |
irl | -0.031 |
re | -0.031 |
eking | -0.031 |
Token | badly |
Token ID | 11234 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
irl | +0.087 |
eling | +0.086 |
ited | +0.086 |
uch | +0.083 |
eking | +0.082 |
Neg logit contributions | |
He | -0.057 |
They | -0.051 |
automobile | -0.049 |
sidewalk | -0.048 |
tr | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
He | +0.012 |
They | +0.010 |
automobile | +0.010 |
tr | +0.010 |
sidewalk | +0.010 |
Neg logit contributions | |
re | -0.016 |
eling | -0.016 |
ited | -0.016 |
if | -0.016 |
irl | -0.016 |
Token | sudden |
Token ID | 4802 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
He | +0.117 |
They | +0.110 |
automobile | +0.108 |
sidewalk | +0.107 |
cop | +0.105 |
Neg logit contributions | |
eling | -0.039 |
ited | -0.037 |
irl | -0.037 |
eking | -0.036 |
uch | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
ited | +0.030 |
eling | +0.029 |
irl | +0.029 |
eking | +0.028 |
uch | +0.028 |
Neg logit contributions | |
He | -0.020 |
They | -0.018 |
The | -0.016 |
automobile | -0.016 |
sidewalk | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
eling | +0.004 |
ited | +0.004 |
irl | +0.004 |
eking | +0.003 |
re | +0.003 |
Neg logit contributions | |
He | -0.002 |
They | -0.002 |
automobile | -0.002 |
sidewalk | -0.002 |
tr | -0.002 |
Token | bright |
Token ID | 6016 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
He | +0.061 |
They | +0.056 |
automobile | +0.051 |
The | +0.051 |
sidewalk | +0.051 |
Neg logit contributions | |
ited | -0.068 |
eling | -0.067 |
irl | -0.066 |
eking | -0.064 |
uch | -0.064 |
Token | light |
Token ID | 1657 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
He | +0.055 |
They | +0.051 |
sidewalk | +0.045 |
The | +0.045 |
automobile | +0.044 |
Neg logit contributions | |
ited | -0.070 |
irl | -0.068 |
eling | -0.068 |
eking | -0.065 |
uch | -0.064 |
Token | flashed |
Token ID | 30050 |
Feature activation | +8.94e+00 |
Pos logit contributions | |
He | +0.104 |
They | +0.096 |
The | +0.089 |
automobile | +0.088 |
sidewalk | +0.088 |
Neg logit contributions | |
ited | -0.092 |
eling | -0.090 |
irl | -0.089 |
eking | -0.086 |
uch | -0.085 |
Token | and |
Token ID | 290 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
He | +0.177 |
They | +0.159 |
sidewalk | +0.144 |
automobile | +0.143 |
The | +0.143 |
Neg logit contributions | |
ited | -0.256 |
irl | -0.247 |
eling | -0.247 |
eking | -0.244 |
if | -0.240 |
Token | the |
Token ID | 262 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
He | +0.032 |
They | +0.029 |
automobile | +0.027 |
sidewalk | +0.026 |
The | +0.026 |
Neg logit contributions | |
ited | -0.040 |
eling | -0.039 |
irl | -0.039 |
eking | -0.038 |
uch | -0.037 |
Token | machine |
Token ID | 4572 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
He | +0.050 |
They | +0.045 |
The | +0.040 |
She | +0.039 |
automobile | +0.038 |
Neg logit contributions | |
ited | -0.065 |
eling | -0.064 |
irl | -0.063 |
uch | -0.062 |
eking | -0.062 |
Token | began |
Token ID | 2540 |
Feature activation | +5.21e+00 |
Pos logit contributions | |
He | +0.070 |
They | +0.064 |
automobile | +0.058 |
The | +0.058 |
sidewalk | +0.058 |
Neg logit contributions | |
ited | -0.082 |
eling | -0.081 |
irl | -0.080 |
if | -0.077 |
uch | -0.076 |
Token | to |
Token ID | 284 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
He | +0.068 |
They | +0.057 |
The | +0.048 |
automobile | +0.047 |
sidewalk | +0.046 |
Neg logit contributions | |
ited | -0.191 |
eling | -0.188 |
irl | -0.187 |
uch | -0.183 |
eking | -0.180 |
Token | one |
Token ID | 530 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
He | +0.018 |
They | +0.016 |
The | +0.015 |
She | +0.014 |
automobile | +0.014 |
Neg logit contributions | |
ited | -0.036 |
eling | -0.035 |
ins | -0.035 |
inations | -0.034 |
irl | -0.034 |
Token | day |
Token ID | 1110 |
Feature activation | -9.90e-01 |
Pos logit contributions | |
He | +0.013 |
They | +0.011 |
The | +0.010 |
automobile | +0.010 |
She | +0.010 |
Neg logit contributions | |
eking | -0.018 |
eling | -0.018 |
irl | -0.017 |
ited | -0.017 |
inations | -0.017 |
Token | she |
Token ID | 673 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
ited | +0.032 |
eling | +0.031 |
irl | +0.030 |
eking | +0.030 |
if | +0.029 |
Neg logit contributions | |
He | -0.018 |
They | -0.016 |
The | -0.015 |
She | -0.014 |
automobile | -0.014 |
Token | decided |
Token ID | 3066 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
He | +0.063 |
They | +0.056 |
automobile | +0.050 |
The | +0.050 |
sidewalk | +0.049 |
Neg logit contributions | |
eling | -0.079 |
ited | -0.079 |
eking | -0.078 |
irl | -0.077 |
if | -0.077 |
Token | to |
Token ID | 284 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
He | +0.080 |
They | +0.072 |
automobile | +0.065 |
tr | +0.064 |
sidewalk | +0.062 |
Neg logit contributions | |
eling | -0.133 |
ited | -0.132 |
eking | -0.131 |
irl | -0.131 |
if | -0.130 |
Token | open |
Token ID | 1280 |
Feature activation | +8.92e+00 |
Pos logit contributions | |
He | +0.036 |
They | +0.032 |
automobile | +0.029 |
sidewalk | +0.028 |
The | +0.028 |
Neg logit contributions | |
irl | -0.058 |
eling | -0.057 |
ited | -0.057 |
uch | -0.055 |
eking | -0.054 |
Token | a |
Token ID | 257 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
He | +0.169 |
They | +0.151 |
automobile | +0.136 |
The | +0.136 |
sidewalk | +0.134 |
Neg logit contributions | |
ited | -0.271 |
eling | -0.265 |
irl | -0.263 |
uch | -0.256 |
eking | -0.256 |
Token | drawer |
Token ID | 33451 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
He | +0.085 |
They | +0.078 |
The | +0.071 |
automobile | +0.071 |
sidewalk | +0.070 |
Neg logit contributions | |
ited | -0.094 |
eling | -0.093 |
irl | -0.092 |
uch | -0.088 |
eking | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | +6.80e+00 |
Pos logit contributions | |
He | +0.043 |
They | +0.039 |
The | +0.035 |
automobile | +0.035 |
sidewalk | +0.034 |
Neg logit contributions | |
ited | -0.071 |
eking | -0.069 |
eling | -0.068 |
kins | -0.068 |
re | -0.067 |
Token | She |
Token ID | 1375 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
der | +0.064 |
automobile | +0.062 |
He | +0.062 |
tr | +0.058 |
cop | +0.058 |
Neg logit contributions | |
re | -0.240 |
if | -0.228 |
ited | -0.225 |
irl | -0.221 |
eling | -0.218 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
He | +0.071 |
They | +0.065 |
automobile | +0.060 |
The | +0.060 |
sidewalk | +0.059 |
Neg logit contributions | |
ited | -0.075 |
eling | -0.074 |
irl | -0.073 |
uch | -0.070 |
eking | -0.070 |
Token | ™ |
Token ID | 8151 |
Feature activation | -5.32e+00 |
Pos logit contributions | |
eling | +0.037 |
ited | +0.035 |
eking | +0.034 |
irl | +0.033 |
ins | +0.033 |
Neg logit contributions | |
He | -0.041 |
They | -0.040 |
The | -0.038 |
It | -0.036 |
But | -0.035 |
Token | s |
Token ID | 82 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
eling | +0.125 |
irl | +0.122 |
ited | +0.119 |
eking | +0.115 |
inations | +0.110 |
Neg logit contributions | |
He | -0.152 |
They | -0.140 |
The | -0.134 |
Tim | -0.127 |
She | -0.126 |
Token | part |
Token ID | 636 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
He | +0.038 |
They | +0.034 |
automobile | +0.030 |
The | +0.030 |
sidewalk | +0.029 |
Neg logit contributions | |
ited | -0.073 |
eling | -0.072 |
irl | -0.071 |
eking | -0.069 |
uch | -0.068 |
Token | this |
Token ID | 428 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
He | +0.091 |
They | +0.081 |
automobile | +0.072 |
The | +0.072 |
sidewalk | +0.071 |
Neg logit contributions | |
ited | -0.152 |
eling | -0.149 |
irl | -0.148 |
eking | -0.143 |
uch | -0.143 |
Token | crack |
Token ID | 8469 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
He | +0.057 |
They | +0.052 |
The | +0.047 |
sidewalk | +0.046 |
It | +0.045 |
Neg logit contributions | |
ited | -0.075 |
eling | -0.073 |
irl | -0.071 |
eking | -0.069 |
ins | -0.068 |
Token | open |
Token ID | 1280 |
Feature activation | +8.90e+00 |
Pos logit contributions | |
He | +0.057 |
They | +0.050 |
automobile | +0.044 |
The | +0.044 |
sidewalk | +0.044 |
Neg logit contributions | |
ited | -0.102 |
eling | -0.100 |
irl | -0.098 |
eking | -0.094 |
if | -0.094 |
Token | and |
Token ID | 290 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
He | +0.170 |
They | +0.152 |
automobile | +0.136 |
sidewalk | +0.135 |
The | +0.134 |
Neg logit contributions | |
ited | -0.266 |
eling | -0.261 |
irl | -0.259 |
eking | -0.252 |
if | -0.251 |
Token | see |
Token ID | 766 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
He | +0.049 |
They | +0.044 |
automobile | +0.041 |
sidewalk | +0.040 |
tr | +0.039 |
Neg logit contributions | |
eling | -0.065 |
ited | -0.065 |
irl | -0.065 |
eking | -0.062 |
uch | -0.062 |
Token | what |
Token ID | 644 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
He | +0.082 |
They | +0.074 |
sidewalk | +0.068 |
automobile | +0.067 |
tr | +0.065 |
Neg logit contributions | |
eling | -0.107 |
ited | -0.106 |
irl | -0.105 |
eking | -0.103 |
if | -0.102 |
Token | â |
Token ID | 22940 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
He | +0.034 |
They | +0.032 |
The | +0.029 |
It | +0.028 |
She | +0.025 |
Neg logit contributions | |
ited | -0.057 |
eking | -0.053 |
eling | -0.052 |
rim | -0.051 |
irl | -0.050 |
Token | € |
Token ID | 26391 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
He | +0.013 |
They | +0.013 |
automobile | +0.012 |
The | +0.012 |
sidewalk | +0.012 |
Neg logit contributions | |
ited | -0.010 |
eling | -0.010 |
irl | -0.009 |
uch | -0.009 |
eking | -0.009 |
Token | " |
Token ID | 366 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
ited | +0.077 |
eling | +0.077 |
irl | +0.077 |
eking | +0.076 |
if | +0.076 |
Neg logit contributions | |
He | -0.020 |
They | -0.016 |
automobile | -0.016 |
sidewalk | -0.014 |
tr | -0.013 |
Token | I |
Token ID | 40 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
eling | +0.011 |
irl | +0.010 |
ited | +0.010 |
eking | +0.010 |
inations | +0.010 |
Neg logit contributions | |
He | -0.008 |
They | -0.007 |
The | -0.006 |
She | -0.006 |
automobile | -0.006 |
Token | made |
Token ID | 925 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
eking | +0.013 |
if | +0.012 |
eling | +0.012 |
kins | +0.012 |
ins | +0.012 |
Neg logit contributions | |
He | -0.010 |
They | -0.009 |
Tim | -0.008 |
But | -0.008 |
The | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
He | +0.037 |
They | +0.034 |
automobile | +0.031 |
sidewalk | +0.030 |
The | +0.030 |
Neg logit contributions | |
ited | -0.050 |
eling | -0.050 |
irl | -0.050 |
eking | -0.048 |
uch | -0.047 |
Token | good |
Token ID | 922 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
He | +0.057 |
They | +0.052 |
The | +0.048 |
automobile | +0.048 |
sidewalk | +0.048 |
Neg logit contributions | |
ited | -0.054 |
eling | -0.053 |
irl | -0.053 |
uch | -0.050 |
eking | -0.050 |
Token | attempt |
Token ID | 2230 |
Feature activation | +8.88e+00 |
Pos logit contributions | |
irl | +0.039 |
ited | +0.038 |
eling | +0.037 |
uch | +0.036 |
eking | +0.036 |
Neg logit contributions | |
He | -0.041 |
They | -0.037 |
The | -0.035 |
automobile | -0.034 |
sidewalk | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
He | +0.185 |
They | +0.168 |
automobile | +0.160 |
sidewalk | +0.155 |
The | +0.153 |
Neg logit contributions | |
eling | -0.233 |
irl | -0.233 |
ited | -0.232 |
re | -0.228 |
uch | -0.227 |
Token | disappear |
Token ID | 10921 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
He | +0.026 |
They | +0.023 |
automobile | +0.021 |
tr | +0.021 |
der | +0.021 |
Neg logit contributions | |
irl | -0.035 |
eling | -0.034 |
ited | -0.033 |
uch | -0.032 |
eking | -0.032 |
Token | !". |
Token ID | 48220 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
He | +0.079 |
They | +0.072 |
automobile | +0.065 |
The | +0.065 |
sidewalk | +0.065 |
Neg logit contributions | |
ited | -0.105 |
eling | -0.103 |
irl | -0.102 |
eking | -0.099 |
uch | -0.098 |
Token | |
Token ID | 220 |
Feature activation | -8.29e+00 |
Pos logit contributions | |
He | +0.074 |
automobile | +0.073 |
sidewalk | +0.069 |
der | +0.068 |
tr | +0.066 |
Neg logit contributions | |
re | -0.127 |
irl | -0.125 |
ited | -0.124 |
eling | -0.121 |
uch | -0.119 |
Token | Everyone |
Token ID | 11075 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
irl | +0.235 |
eling | +0.233 |
ited | +0.231 |
re | +0.227 |
uch | +0.226 |
Neg logit contributions | |
He | -0.156 |
They | -0.139 |
automobile | -0.138 |
sidewalk | -0.133 |
tr | -0.126 |
Token | spotted |
Token ID | 13489 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
He | +0.050 |
They | +0.040 |
The | +0.039 |
Tim | +0.037 |
It | +0.035 |
Neg logit contributions | |
irl | -0.066 |
ins | -0.065 |
eling | -0.064 |
inations | -0.063 |
if | -0.063 |
Token | a |
Token ID | 257 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
He | +0.098 |
They | +0.091 |
automobile | +0.084 |
sidewalk | +0.082 |
tr | +0.081 |
Neg logit contributions | |
eling | -0.143 |
ited | -0.142 |
irl | -0.139 |
eking | -0.137 |
if | -0.136 |
Token | branch |
Token ID | 8478 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
He | +0.071 |
They | +0.064 |
automobile | +0.060 |
sidewalk | +0.059 |
The | +0.059 |
Neg logit contributions | |
ited | -0.079 |
eling | -0.079 |
irl | -0.077 |
eking | -0.076 |
uch | -0.075 |
Token | over |
Token ID | 625 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
He | +0.059 |
They | +0.057 |
sidewalk | +0.051 |
automobile | +0.050 |
The | +0.050 |
Neg logit contributions | |
ited | -0.087 |
eking | -0.084 |
eling | -0.083 |
if | -0.081 |
irl | -0.080 |
Token | h |
Token ID | 71 |
Feature activation | -3.39e+00 |
Pos logit contributions | |
He | +0.042 |
They | +0.036 |
The | +0.033 |
automobile | +0.031 |
sidewalk | +0.031 |
Neg logit contributions | |
ited | -0.074 |
irl | -0.073 |
eling | -0.073 |
uch | -0.072 |
inations | -0.071 |
Token | anging |
Token ID | 4924 |
Feature activation | +8.80e+00 |
Pos logit contributions | |
ited | +0.082 |
irl | +0.079 |
eling | +0.079 |
uch | +0.075 |
eking | +0.074 |
Neg logit contributions | |
He | -0.086 |
They | -0.078 |
The | -0.073 |
sidewalk | -0.072 |
automobile | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | +3.34e+00 |
Pos logit contributions | |
He | +0.161 |
They | +0.150 |
tr | +0.139 |
automobile | +0.137 |
sidewalk | +0.136 |
Neg logit contributions | |
eling | -0.261 |
ited | -0.261 |
if | -0.249 |
eking | -0.247 |
irl | -0.246 |
Token | pond |
Token ID | 16723 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
He | +0.073 |
They | +0.066 |
automobile | +0.060 |
The | +0.060 |
sidewalk | +0.059 |
Neg logit contributions | |
ited | -0.091 |
eling | -0.090 |
irl | -0.089 |
eking | -0.086 |
uch | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | +7.18e+00 |
Pos logit contributions | |
He | +0.008 |
They | +0.007 |
automobile | +0.006 |
sidewalk | +0.006 |
The | +0.006 |
Neg logit contributions | |
ited | -0.018 |
eling | -0.017 |
irl | -0.017 |
eking | -0.017 |
if | -0.017 |
Token | He |
Token ID | 679 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
automobile | +0.044 |
sidewalk | +0.036 |
tr | +0.034 |
arrow | +0.024 |
pavement | +0.024 |
Neg logit contributions | |
if | -0.268 |
re | -0.268 |
ited | -0.266 |
eling | -0.262 |
irl | -0.262 |
Token | jumped |
Token ID | 11687 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
He | +0.075 |
They | +0.065 |
The | +0.061 |
Then | +0.057 |
It | +0.057 |
Neg logit contributions | |
irl | -0.091 |
ited | -0.090 |
eling | -0.090 |
uch | -0.090 |
eking | -0.087 |
Token | â |
Token ID | 22940 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
He | +0.125 |
automobile | +0.112 |
They | +0.112 |
sidewalk | +0.111 |
tr | +0.105 |
Neg logit contributions | |
ited | -0.212 |
irl | -0.209 |
eling | -0.209 |
if | -0.205 |
re | -0.205 |
Token | € |
Token ID | 26391 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
He | +0.037 |
They | +0.035 |
automobile | +0.033 |
der | +0.033 |
roof | +0.032 |
Neg logit contributions | |
ited | -0.027 |
eling | -0.026 |
uch | -0.025 |
irl | -0.025 |
eking | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -4.98e+00 |
Pos logit contributions | |
ited | +0.041 |
eling | +0.040 |
eking | +0.038 |
irl | +0.038 |
ins | +0.037 |
Neg logit contributions | |
He | -0.034 |
They | -0.031 |
The | -0.029 |
It | -0.027 |
She | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
ited | +0.160 |
eling | +0.156 |
irl | +0.152 |
eking | +0.150 |
uch | +0.150 |
Neg logit contributions | |
He | -0.091 |
They | -0.080 |
The | -0.071 |
sidewalk | -0.069 |
automobile | -0.068 |
Token | Sw |
Token ID | 10462 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
ited | +0.124 |
eling | +0.121 |
eking | +0.119 |
uch | +0.116 |
kins | +0.115 |
Neg logit contributions | |
He | -0.095 |
They | -0.084 |
The | -0.077 |
She | -0.074 |
It | -0.072 |
Token | ift |
Token ID | 2135 |
Feature activation | +8.80e+00 |
Pos logit contributions | |
ited | +0.041 |
eling | +0.040 |
inations | +0.038 |
re | +0.038 |
if | +0.038 |
Neg logit contributions | |
He | -0.065 |
They | -0.060 |
The | -0.057 |
automobile | -0.056 |
sidewalk | -0.056 |
Token | was |
Token ID | 373 |
Feature activation | -6.49e-01 |
Pos logit contributions | |
He | +0.195 |
They | +0.177 |
automobile | +0.162 |
The | +0.161 |
sidewalk | +0.160 |
Neg logit contributions | |
ited | -0.237 |
eling | -0.233 |
irl | -0.231 |
eking | -0.223 |
uch | -0.222 |
Token | puzzled |
Token ID | 36907 |
Feature activation | +4.06e+00 |
Pos logit contributions | |
ited | +0.020 |
eling | +0.020 |
irl | +0.020 |
uch | +0.019 |
eking | +0.019 |
Neg logit contributions | |
He | -0.012 |
They | -0.011 |
The | -0.010 |
Tim | -0.009 |
automobile | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +7.82e+00 |
Pos logit contributions | |
He | +0.071 |
They | +0.062 |
The | +0.055 |
automobile | +0.055 |
sidewalk | +0.054 |
Neg logit contributions | |
ited | -0.130 |
eling | -0.128 |
irl | -0.127 |
uch | -0.123 |
eking | -0.123 |
Token | He |
Token ID | 679 |
Feature activation | +3.69e+00 |
Pos logit contributions | |
automobile | +0.062 |
tr | +0.056 |
sidewalk | +0.056 |
He | +0.056 |
They | +0.047 |
Neg logit contributions | |
eling | -0.283 |
re | -0.282 |
ited | -0.282 |
if | -0.281 |
irl | -0.280 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
He | +0.086 |
They | +0.076 |
The | +0.070 |
automobile | +0.067 |
It | +0.066 |
Neg logit contributions | |
ited | -0.097 |
irl | -0.097 |
eling | -0.096 |
uch | -0.095 |
if | -0.094 |
Token | old |
Token ID | 1468 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
He | +0.065 |
They | +0.061 |
The | +0.054 |
automobile | +0.054 |
sidewalk | +0.054 |
Neg logit contributions | |
ited | -0.084 |
eling | -0.083 |
irl | -0.082 |
if | -0.079 |
eking | -0.079 |
Token | man |
Token ID | 582 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
He | +0.066 |
They | +0.062 |
tr | +0.057 |
The | +0.053 |
der | +0.053 |
Neg logit contributions | |
ited | -0.114 |
eling | -0.113 |
irl | -0.112 |
if | -0.110 |
re | -0.109 |
Token | if |
Token ID | 611 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
ited | +0.005 |
eling | +0.005 |
irl | +0.005 |
eking | +0.004 |
re | +0.004 |
Neg logit contributions | |
He | -0.003 |
They | -0.003 |
sidewalk | -0.003 |
automobile | -0.003 |
The | -0.003 |
Token | he |
Token ID | 339 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
He | +0.022 |
They | +0.017 |
The | +0.014 |
automobile | +0.013 |
sidewalk | +0.012 |
Neg logit contributions | |
ited | -0.080 |
irl | -0.078 |
eling | -0.078 |
uch | -0.076 |
eking | -0.076 |
Token | could |
Token ID | 714 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
He | +0.052 |
They | +0.045 |
The | +0.041 |
It | +0.039 |
She | +0.039 |
Neg logit contributions | |
eling | -0.069 |
irl | -0.069 |
ited | -0.068 |
eking | -0.068 |
if | -0.066 |
Token | open |
Token ID | 1280 |
Feature activation | +8.79e+00 |
Pos logit contributions | |
ited | +0.052 |
irl | +0.051 |
eling | +0.051 |
uch | +0.049 |
eking | +0.049 |
Neg logit contributions | |
He | -0.036 |
They | -0.032 |
automobile | -0.030 |
sidewalk | -0.029 |
The | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
He | +0.164 |
They | +0.146 |
automobile | +0.131 |
sidewalk | +0.130 |
The | +0.129 |
Neg logit contributions | |
ited | -0.268 |
eling | -0.264 |
irl | -0.262 |
eking | -0.253 |
uch | -0.252 |
Token | case |
Token ID | 1339 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
He | +0.079 |
They | +0.075 |
The | +0.065 |
der | +0.062 |
Tim | +0.060 |
Neg logit contributions | |
ited | -0.099 |
eling | -0.098 |
irl | -0.097 |
eking | -0.096 |
if | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
He | +0.025 |
They | +0.024 |
automobile | +0.023 |
sidewalk | +0.023 |
tr | +0.022 |
Neg logit contributions | |
ited | -0.031 |
eling | -0.030 |
irl | -0.030 |
kins | -0.030 |
re | -0.030 |
Token | but |
Token ID | 475 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
He | +0.014 |
They | +0.012 |
automobile | +0.011 |
The | +0.011 |
sidewalk | +0.011 |
Neg logit contributions | |
ited | -0.020 |
eling | -0.020 |
irl | -0.020 |
eking | -0.019 |
uch | -0.019 |
Token | he |
Token ID | 339 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
He | +0.016 |
They | +0.013 |
The | +0.011 |
Tim | +0.010 |
automobile | +0.010 |
Neg logit contributions | |
ited | -0.039 |
irl | -0.038 |
eling | -0.038 |
eking | -0.037 |
uch | -0.037 |
Token | was |
Token ID | 373 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
He | +0.017 |
They | +0.015 |
The | +0.013 |
automobile | +0.013 |
sidewalk | +0.013 |
Neg logit contributions | |
ited | -0.027 |
eling | -0.026 |
irl | -0.026 |
uch | -0.025 |
eking | -0.025 |
Token | so |
Token ID | 523 |
Feature activation | -4.95e+00 |
Pos logit contributions | |
ited | +0.075 |
eling | +0.072 |
uch | +0.071 |
irl | +0.071 |
ins | +0.069 |
Neg logit contributions | |
He | -0.059 |
The | -0.051 |
They | -0.050 |
Tim | -0.047 |
It | -0.046 |
Token | scared |
Token ID | 12008 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
ited | +0.123 |
eling | +0.120 |
uch | +0.116 |
irl | +0.116 |
if | +0.112 |
Neg logit contributions | |
He | -0.132 |
They | -0.117 |
The | -0.110 |
She | -0.105 |
Tim | -0.102 |
Token | that |
Token ID | 326 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
He | +0.033 |
tr | +0.031 |
They | +0.030 |
sidewalk | +0.029 |
automobile | +0.029 |
Neg logit contributions | |
ited | -0.061 |
eling | -0.060 |
re | -0.059 |
if | -0.058 |
eking | -0.058 |
Token | she |
Token ID | 673 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
He | +0.048 |
They | +0.042 |
automobile | +0.037 |
sidewalk | +0.036 |
The | +0.035 |
Neg logit contributions | |
ited | -0.127 |
eling | -0.126 |
irl | -0.125 |
eking | -0.121 |
if | -0.121 |
Token | dropped |
Token ID | 5710 |
Feature activation | +8.78e+00 |
Pos logit contributions | |
He | +0.053 |
They | +0.046 |
The | +0.042 |
automobile | +0.041 |
sidewalk | +0.041 |
Neg logit contributions | |
ited | -0.060 |
irl | -0.059 |
eling | -0.058 |
if | -0.057 |
ins | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
He | +0.170 |
They | +0.151 |
The | +0.136 |
automobile | +0.134 |
sidewalk | +0.132 |
Neg logit contributions | |
ited | -0.267 |
eling | -0.262 |
irl | -0.261 |
uch | -0.253 |
eking | -0.252 |
Token | bottle |
Token ID | 9294 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
He | +0.048 |
They | +0.044 |
sidewalk | +0.043 |
automobile | +0.043 |
pavement | +0.040 |
Neg logit contributions | |
ited | -0.058 |
eling | -0.056 |
irl | -0.056 |
inations | -0.055 |
uch | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | +6.53e+00 |
Pos logit contributions | |
He | +0.009 |
They | +0.008 |
sidewalk | +0.007 |
automobile | +0.007 |
The | +0.007 |
Neg logit contributions | |
ited | -0.016 |
eling | -0.015 |
irl | -0.015 |
eking | -0.015 |
re | -0.015 |
Token | |
Token ID | 220 |
Feature activation | -7.59e+00 |
Pos logit contributions | |
He | +0.135 |
automobile | +0.125 |
They | +0.124 |
sidewalk | +0.122 |
tr | +0.119 |
Neg logit contributions | |
ited | -0.169 |
eling | -0.166 |
irl | -0.164 |
re | -0.161 |
if | -0.161 |
Token |
|
Token ID | 198 |
Feature activation | -5.00e+00 |
Pos logit contributions | |
ited | +0.227 |
eling | +0.225 |
irl | +0.225 |
uch | +0.217 |
eking | +0.217 |
Neg logit contributions | |
He | -0.139 |
They | -0.124 |
automobile | -0.115 |
sidewalk | -0.113 |
The | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | +6.89e+00 |
Pos logit contributions | |
He | +0.093 |
They | +0.084 |
sidewalk | +0.083 |
arrow | +0.079 |
X | +0.078 |
Neg logit contributions | |
re | -0.112 |
kins | -0.108 |
ited | -0.108 |
if | -0.107 |
eling | -0.106 |
Token |
|
Token ID | 198 |
Feature activation | -4.97e+00 |
Pos logit contributions | |
automobile | +0.107 |
tr | +0.105 |
sidewalk | +0.100 |
He | +0.099 |
der | +0.097 |
Neg logit contributions | |
re | -0.198 |
irl | -0.195 |
ited | -0.194 |
if | -0.187 |
eling | -0.186 |
Token |
|
Token ID | 198 |
Feature activation | -4.33e+00 |
Pos logit contributions | |
ited | +0.160 |
eling | +0.158 |
irl | +0.154 |
eking | +0.152 |
uch | +0.151 |
Neg logit contributions | |
He | -0.089 |
They | -0.079 |
The | -0.070 |
sidewalk | -0.067 |
automobile | -0.067 |
Token | Jack |
Token ID | 14295 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
ited | +0.098 |
eling | +0.098 |
irl | +0.095 |
eking | +0.093 |
uch | +0.092 |
Neg logit contributions | |
He | -0.117 |
They | -0.108 |
The | -0.099 |
sidewalk | -0.098 |
automobile | -0.098 |
Token | carefully |
Token ID | 7773 |
Feature activation | +5.67e+00 |
Pos logit contributions | |
He | +0.006 |
They | +0.006 |
The | +0.005 |
automobile | +0.005 |
sidewalk | +0.005 |
Neg logit contributions | |
ited | -0.008 |
eling | -0.008 |
irl | -0.008 |
eking | -0.008 |
uch | -0.008 |
Token | connected |
Token ID | 5884 |
Feature activation | +8.78e+00 |
Pos logit contributions | |
He | +0.140 |
They | +0.127 |
automobile | +0.119 |
sidewalk | +0.117 |
The | +0.115 |
Neg logit contributions | |
ited | -0.141 |
eling | -0.137 |
irl | -0.136 |
eking | -0.130 |
if | -0.129 |
Token | the |
Token ID | 262 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
He | +0.154 |
They | +0.136 |
automobile | +0.121 |
sidewalk | +0.120 |
The | +0.120 |
Neg logit contributions | |
ited | -0.278 |
eling | -0.274 |
irl | -0.271 |
eking | -0.263 |
uch | -0.262 |
Token | vacuum |
Token ID | 17076 |
Feature activation | +5.62e+00 |
Pos logit contributions | |
He | +0.075 |
They | +0.068 |
The | +0.061 |
automobile | +0.060 |
sidewalk | +0.060 |
Neg logit contributions | |
ited | -0.096 |
eling | -0.096 |
irl | -0.095 |
eking | -0.092 |
if | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
He | +0.125 |
They | +0.114 |
sidewalk | +0.107 |
arrow | +0.105 |
automobile | +0.104 |
Neg logit contributions | |
re | -0.137 |
if | -0.135 |
eling | -0.135 |
ited | -0.134 |
kins | -0.133 |
Token | his |
Token ID | 465 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
He | +0.059 |
tr | +0.059 |
automobile | +0.056 |
They | +0.056 |
der | +0.054 |
Neg logit contributions | |
eling | -0.078 |
irl | -0.075 |
ited | -0.073 |
kins | -0.073 |
eking | -0.070 |
Token | boat |
Token ID | 8848 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
He | +0.080 |
They | +0.072 |
The | +0.065 |
sidewalk | +0.064 |
automobile | +0.064 |
Neg logit contributions | |
irl | -0.103 |
ited | -0.103 |
eling | -0.103 |
eking | -0.098 |
uch | -0.098 |
Token | € |
Token ID | 26391 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
He | +0.041 |
They | +0.038 |
der | +0.036 |
automobile | +0.036 |
roof | +0.035 |
Neg logit contributions | |
ited | -0.031 |
eling | -0.029 |
uch | -0.028 |
irl | -0.028 |
eking | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
ited | +0.032 |
eling | +0.031 |
eking | +0.030 |
irl | +0.030 |
sie | +0.029 |
Neg logit contributions | |
He | -0.027 |
They | -0.025 |
The | -0.023 |
It | -0.022 |
automobile | -0.021 |
Token | F |
Token ID | 37 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ited | +0.136 |
eling | +0.133 |
irl | +0.129 |
eking | +0.128 |
uch | +0.127 |
Neg logit contributions | |
He | -0.092 |
They | -0.083 |
The | -0.074 |
sidewalk | -0.072 |
automobile | -0.072 |
Token | uzz |
Token ID | 4715 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ited | +0.025 |
irl | +0.024 |
eling | +0.024 |
uch | +0.023 |
if | +0.022 |
Neg logit contributions | |
He | -0.034 |
They | -0.032 |
automobile | -0.030 |
sidewalk | -0.030 |
The | -0.029 |
Token | y |
Token ID | 88 |
Feature activation | +4.06e+00 |
Pos logit contributions | |
eling | +0.027 |
ited | +0.027 |
irl | +0.027 |
eking | +0.026 |
if | +0.026 |
Neg logit contributions | |
He | -0.032 |
They | -0.030 |
The | -0.028 |
automobile | -0.027 |
der | -0.027 |
Token | dropped |
Token ID | 5710 |
Feature activation | +8.77e+00 |
Pos logit contributions | |
He | +0.090 |
They | +0.082 |
The | +0.074 |
automobile | +0.073 |
sidewalk | +0.073 |
Neg logit contributions | |
ited | -0.111 |
eling | -0.109 |
irl | -0.107 |
eking | -0.105 |
uch | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
He | +0.162 |
They | +0.144 |
The | +0.128 |
automobile | +0.128 |
sidewalk | +0.125 |
Neg logit contributions | |
ited | -0.275 |
eling | -0.270 |
irl | -0.268 |
uch | -0.261 |
eking | -0.259 |
Token | stick |
Token ID | 4859 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
He | +0.055 |
They | +0.047 |
automobile | +0.047 |
sidewalk | +0.046 |
tr | +0.044 |
Neg logit contributions | |
ited | -0.081 |
eling | -0.081 |
irl | -0.079 |
eking | -0.077 |
if | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
He | +0.085 |
They | +0.076 |
automobile | +0.068 |
sidewalk | +0.067 |
The | +0.067 |
Neg logit contributions | |
ited | -0.161 |
irl | -0.156 |
eling | -0.156 |
eking | -0.153 |
if | -0.153 |
Token | started |
Token ID | 2067 |
Feature activation | +5.43e+00 |
Pos logit contributions | |
He | +0.052 |
They | +0.047 |
automobile | +0.044 |
sidewalk | +0.043 |
The | +0.043 |
Neg logit contributions | |
ited | -0.057 |
irl | -0.057 |
eling | -0.057 |
eking | -0.054 |
re | -0.054 |
Token | running |
Token ID | 2491 |
Feature activation | +5.09e+00 |
Pos logit contributions | |
He | +0.111 |
They | +0.100 |
automobile | +0.091 |
sidewalk | +0.090 |
The | +0.090 |
Neg logit contributions | |
ited | -0.155 |
eling | -0.153 |
irl | -0.152 |
eking | -0.147 |
uch | -0.145 |
Token |
|
Token ID | 198 |
Feature activation | -4.69e+00 |
Pos logit contributions | |
ited | +0.198 |
eling | +0.196 |
eking | +0.191 |
uch | +0.187 |
irl | +0.185 |
Neg logit contributions | |
He | -0.082 |
They | -0.072 |
The | -0.064 |
She | -0.057 |
It | -0.055 |
Token | One |
Token ID | 3198 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
ited | +0.109 |
eling | +0.109 |
eking | +0.106 |
uch | +0.103 |
irl | +0.102 |
Neg logit contributions | |
He | -0.122 |
They | -0.116 |
The | -0.107 |
automobile | -0.106 |
sidewalk | -0.104 |
Token | day |
Token ID | 1110 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
ited | +0.046 |
eling | +0.045 |
irl | +0.045 |
eking | +0.043 |
uch | +0.043 |
Neg logit contributions | |
He | -0.043 |
They | -0.040 |
automobile | -0.036 |
The | -0.036 |
sidewalk | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
ited | +0.016 |
eling | +0.016 |
eking | +0.015 |
irl | +0.015 |
inations | +0.015 |
Neg logit contributions | |
He | -0.008 |
They | -0.008 |
The | -0.007 |
Tim | -0.006 |
She | -0.006 |
Token | she |
Token ID | 673 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
He | +0.004 |
They | +0.004 |
The | +0.004 |
automobile | +0.003 |
Tim | +0.003 |
Neg logit contributions | |
ited | -0.010 |
eling | -0.010 |
irl | -0.009 |
eking | -0.009 |
uch | -0.009 |
Token | scattered |
Token ID | 16830 |
Feature activation | +8.76e+00 |
Pos logit contributions | |
He | +0.056 |
They | +0.054 |
The | +0.047 |
automobile | +0.044 |
She | +0.044 |
Neg logit contributions | |
eling | -0.096 |
ited | -0.095 |
eking | -0.095 |
inations | -0.092 |
if | -0.092 |
Token | her |
Token ID | 607 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
He | +0.138 |
They | +0.121 |
The | +0.112 |
automobile | +0.105 |
Tim | +0.105 |
Neg logit contributions | |
ited | -0.298 |
eling | -0.297 |
uch | -0.297 |
irl | -0.295 |
inations | -0.294 |
Token | strips |
Token ID | 22670 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
sidewalk | +0.043 |
The | +0.041 |
They | +0.041 |
automobile | +0.040 |
He | +0.039 |
Neg logit contributions | |
ited | -0.089 |
eling | -0.088 |
rim | -0.083 |
eking | -0.080 |
iri | -0.079 |
Token | outside |
Token ID | 2354 |
Feature activation | +4.84e+00 |
Pos logit contributions | |
He | +0.029 |
They | +0.026 |
automobile | +0.024 |
The | +0.023 |
sidewalk | +0.023 |
Neg logit contributions | |
ited | -0.045 |
eling | -0.044 |
irl | -0.044 |
eking | -0.042 |
uch | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +6.73e+00 |
Pos logit contributions | |
He | +0.092 |
They | +0.080 |
automobile | +0.074 |
sidewalk | +0.073 |
tr | +0.071 |
Neg logit contributions | |
ited | -0.144 |
eling | -0.142 |
irl | -0.141 |
if | -0.136 |
eking | -0.135 |
Token | She |
Token ID | 1375 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
bike | +0.078 |
automobile | +0.077 |
extingu | +0.076 |
penter | +0.075 |
arrow | +0.073 |
Neg logit contributions | |
re | -0.191 |
teaspoons | -0.187 |
time | -0.182 |
if | -0.181 |
irl | -0.179 |
Token |
|
Token ID | 198 |
Feature activation | -5.15e+00 |
Pos logit contributions | |
ited | +0.246 |
eling | +0.243 |
irl | +0.241 |
eking | +0.233 |
uch | +0.233 |
Neg logit contributions | |
He | -0.143 |
They | -0.127 |
automobile | -0.114 |
sidewalk | -0.112 |
The | -0.112 |
Token |
|
Token ID | 198 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
ited | +0.172 |
eling | +0.172 |
irl | +0.166 |
eking | +0.166 |
uch | +0.164 |
Neg logit contributions | |
He | -0.092 |
They | -0.082 |
The | -0.071 |
She | -0.067 |
automobile | -0.066 |
Token | They |
Token ID | 2990 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
ited | +0.096 |
eling | +0.093 |
irl | +0.093 |
eking | +0.088 |
if | +0.088 |
Neg logit contributions | |
He | -0.124 |
They | -0.115 |
automobile | -0.108 |
sidewalk | -0.107 |
The | -0.107 |
Token | decided |
Token ID | 3066 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
He | +0.054 |
They | +0.049 |
sidewalk | +0.046 |
automobile | +0.046 |
The | +0.045 |
Neg logit contributions | |
ited | -0.050 |
eling | -0.049 |
irl | -0.047 |
eking | -0.046 |
re | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
He | +0.082 |
tr | +0.075 |
They | +0.072 |
automobile | +0.071 |
sidewalk | +0.069 |
Neg logit contributions | |
eking | -0.118 |
re | -0.117 |
eling | -0.115 |
ited | -0.114 |
if | -0.113 |
Token | open |
Token ID | 1280 |
Feature activation | +8.76e+00 |
Pos logit contributions | |
He | +0.035 |
They | +0.030 |
der | +0.028 |
sidewalk | +0.028 |
automobile | +0.028 |
Neg logit contributions | |
ited | -0.052 |
irl | -0.052 |
eling | -0.052 |
uch | -0.049 |
eking | -0.049 |
Token | it |
Token ID | 340 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
He | +0.192 |
They | +0.172 |
tr | +0.160 |
sidewalk | +0.160 |
automobile | +0.158 |
Neg logit contributions | |
ited | -0.232 |
eling | -0.231 |
irl | -0.229 |
eking | -0.220 |
if | -0.217 |
Token | and |
Token ID | 290 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
He | +0.045 |
They | +0.041 |
automobile | +0.041 |
sidewalk | +0.040 |
tr | +0.039 |
Neg logit contributions | |
ited | -0.061 |
irl | -0.060 |
eking | -0.060 |
re | -0.059 |
kins | -0.059 |
Token | found |
Token ID | 1043 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
He | +0.051 |
der | +0.048 |
They | +0.046 |
automobile | +0.044 |
tr | +0.044 |
Neg logit contributions | |
ited | -0.055 |
eling | -0.054 |
irl | -0.054 |
eking | -0.052 |
inations | -0.051 |
Token | a |
Token ID | 257 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
He | +0.111 |
sidewalk | +0.101 |
They | +0.098 |
tr | +0.097 |
pavement | +0.097 |
Neg logit contributions | |
kins | -0.124 |
eling | -0.123 |
eking | -0.120 |
irl | -0.118 |
re | -0.116 |
Token | small |
Token ID | 1402 |
Feature activation | +4.58e+00 |
Pos logit contributions | |
He | +0.079 |
They | +0.072 |
The | +0.066 |
automobile | +0.066 |
sidewalk | +0.066 |
Neg logit contributions | |
ited | -0.084 |
eling | -0.082 |
irl | -0.082 |
eking | -0.079 |
uch | -0.078 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
ited | +0.009 |
eling | +0.009 |
irl | +0.009 |
eking | +0.008 |
uch | +0.008 |
Neg logit contributions | |
He | -0.005 |
They | -0.005 |
The | -0.004 |
automobile | -0.004 |
She | -0.004 |
Token | worked |
Token ID | 3111 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
ited | +0.020 |
eling | +0.019 |
irl | +0.019 |
uch | +0.018 |
re | +0.018 |
Neg logit contributions | |
He | -0.016 |
They | -0.014 |
automobile | -0.013 |
sidewalk | -0.013 |
The | -0.013 |
Token | hard |
Token ID | 1327 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
He | +0.091 |
They | +0.082 |
The | +0.074 |
sidewalk | +0.072 |
automobile | +0.071 |
Neg logit contributions | |
ited | -0.117 |
eling | -0.113 |
uch | -0.113 |
irl | -0.110 |
eking | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +7.36e+00 |
Pos logit contributions | |
He | +0.054 |
They | +0.047 |
automobile | +0.042 |
The | +0.041 |
sidewalk | +0.041 |
Neg logit contributions | |
ited | -0.116 |
eling | -0.115 |
irl | -0.114 |
eking | -0.111 |
if | -0.110 |
Token | She |
Token ID | 1375 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
automobile | +0.067 |
arrow | +0.064 |
He | +0.063 |
tr | +0.061 |
cop | +0.059 |
Neg logit contributions | |
re | -0.256 |
if | -0.253 |
irl | -0.246 |
ited | -0.245 |
eling | -0.241 |
Token | scattered |
Token ID | 16830 |
Feature activation | +8.74e+00 |
Pos logit contributions | |
He | +0.077 |
They | +0.071 |
The | +0.064 |
automobile | +0.064 |
sidewalk | +0.063 |
Neg logit contributions | |
ited | -0.090 |
eling | -0.089 |
irl | -0.086 |
eking | -0.085 |
uch | -0.084 |
Token | the |
Token ID | 262 |
Feature activation | +2.92e+00 |
Pos logit contributions | |
He | +0.134 |
They | +0.114 |
The | +0.104 |
She | +0.096 |
Tim | +0.095 |
Neg logit contributions | |
uch | -0.322 |
inations | -0.307 |
eking | -0.305 |
ited | -0.301 |
irl | -0.300 |
Token | butter |
Token ID | 9215 |
Feature activation | +6.87e-01 |
Pos logit contributions | |
He | +0.061 |
sidewalk | +0.061 |
pavement | +0.057 |
automobile | +0.057 |
roof | +0.054 |
Neg logit contributions | |
eling | -0.073 |
irl | -0.071 |
ited | -0.067 |
eking | -0.067 |
elve | -0.067 |
Token | on |
Token ID | 319 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
He | +0.013 |
They | +0.012 |
automobile | +0.011 |
The | +0.011 |
sidewalk | +0.010 |
Neg logit contributions | |
ited | -0.021 |
eling | -0.020 |
irl | -0.020 |
eking | -0.019 |
uch | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +3.81e+00 |
Pos logit contributions | |
He | +0.039 |
They | +0.035 |
automobile | +0.032 |
sidewalk | +0.031 |
tr | +0.031 |
Neg logit contributions | |
ited | -0.056 |
eling | -0.056 |
irl | -0.056 |
if | -0.053 |
eking | -0.053 |
Token | pancakes |
Token ID | 47008 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
He | +0.079 |
They | +0.072 |
The | +0.063 |
automobile | +0.062 |
It | +0.060 |
Neg logit contributions | |
ited | -0.108 |
eling | -0.107 |
irl | -0.105 |
eking | -0.103 |
if | -0.102 |
Token | the |
Token ID | 262 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
He | +0.089 |
They | +0.082 |
The | +0.074 |
automobile | +0.073 |
sidewalk | +0.071 |
Neg logit contributions | |
ited | -0.126 |
eling | -0.123 |
irl | -0.122 |
uch | -0.120 |
eking | -0.117 |
Token | fake |
Token ID | 8390 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
He | +0.064 |
They | +0.058 |
automobile | +0.054 |
sidewalk | +0.053 |
The | +0.052 |
Neg logit contributions | |
ited | -0.092 |
eling | -0.091 |
irl | -0.091 |
eking | -0.087 |
uch | -0.086 |
Token | bird |
Token ID | 6512 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
He | +0.095 |
They | +0.085 |
automobile | +0.078 |
The | +0.077 |
sidewalk | +0.077 |
Neg logit contributions | |
ited | -0.129 |
eling | -0.127 |
irl | -0.126 |
eking | -0.121 |
uch | -0.121 |
Token | c |
Token ID | 66 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
He | +0.001 |
They | +0.001 |
automobile | +0.001 |
The | +0.001 |
sidewalk | +0.001 |
Neg logit contributions | |
ited | -0.001 |
eling | -0.001 |
irl | -0.001 |
uch | -0.001 |
eking | -0.001 |
Token | age |
Token ID | 496 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
ited | +0.013 |
eling | +0.013 |
irl | +0.013 |
eking | +0.012 |
uch | +0.012 |
Neg logit contributions | |
He | -0.005 |
They | -0.005 |
automobile | -0.004 |
The | -0.004 |
sidewalk | -0.004 |
Token | open |
Token ID | 1280 |
Feature activation | +8.72e+00 |
Pos logit contributions | |
He | +0.074 |
They | +0.063 |
automobile | +0.059 |
sidewalk | +0.058 |
The | +0.056 |
Neg logit contributions | |
ited | -0.135 |
eling | -0.132 |
irl | -0.132 |
eking | -0.132 |
re | -0.131 |
Token | and |
Token ID | 290 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
He | +0.144 |
They | +0.124 |
sidewalk | +0.111 |
automobile | +0.109 |
The | +0.108 |
Neg logit contributions | |
ited | -0.278 |
eling | -0.273 |
irl | -0.272 |
eking | -0.266 |
if | -0.265 |
Token | all |
Token ID | 477 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
He | +0.052 |
They | +0.047 |
automobile | +0.043 |
sidewalk | +0.043 |
The | +0.042 |
Neg logit contributions | |
ited | -0.062 |
eling | -0.061 |
irl | -0.060 |
eking | -0.058 |
uch | -0.057 |
Token | of |
Token ID | 286 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
He | +0.022 |
They | +0.020 |
The | +0.018 |
automobile | +0.018 |
sidewalk | +0.017 |
Neg logit contributions | |
ited | -0.032 |
irl | -0.031 |
eling | -0.031 |
uch | -0.031 |
eking | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
He | +0.011 |
They | +0.010 |
The | +0.009 |
Tim | +0.009 |
automobile | +0.009 |
Neg logit contributions | |
ited | -0.019 |
uch | -0.018 |
eling | -0.018 |
irl | -0.018 |
eking | -0.017 |
Token | birds |
Token ID | 10087 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
automobile | +0.045 |
He | +0.045 |
sidewalk | +0.044 |
They | +0.042 |
arrow | +0.041 |
Neg logit contributions | |
irl | -0.074 |
ited | -0.071 |
uch | -0.071 |
eling | -0.069 |
kins | -0.066 |
Token | the |
Token ID | 262 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
He | +0.076 |
They | +0.069 |
tr | +0.068 |
automobile | +0.067 |
sidewalk | +0.065 |
Neg logit contributions | |
eling | -0.101 |
irl | -0.101 |
ited | -0.101 |
if | -0.098 |
kins | -0.095 |
Token | shelf |
Token ID | 18316 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
He | +0.068 |
They | +0.063 |
automobile | +0.056 |
The | +0.055 |
tr | +0.054 |
Neg logit contributions | |
eling | -0.085 |
ited | -0.085 |
kins | -0.083 |
if | -0.083 |
sie | -0.082 |
Token | in |
Token ID | 287 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
He | +0.035 |
They | +0.032 |
automobile | +0.031 |
sidewalk | +0.030 |
tr | +0.029 |
Neg logit contributions | |
ited | -0.057 |
kins | -0.054 |
if | -0.054 |
re | -0.054 |
irl | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
He | +0.035 |
They | +0.031 |
The | +0.027 |
It | +0.024 |
automobile | +0.024 |
Neg logit contributions | |
ited | -0.071 |
eling | -0.071 |
uch | -0.070 |
irl | -0.070 |
eking | -0.069 |
Token | pant |
Token ID | 15857 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
He | +0.083 |
They | +0.081 |
der | +0.077 |
automobile | +0.076 |
Tim | +0.073 |
Neg logit contributions | |
ited | -0.089 |
eling | -0.088 |
sie | -0.083 |
years | -0.083 |
if | -0.082 |
Token | ry |
Token ID | 563 |
Feature activation | +8.71e+00 |
Pos logit contributions | |
He | +0.045 |
automobile | +0.044 |
They | +0.042 |
cop | +0.042 |
tr | +0.042 |
Neg logit contributions | |
re | -0.031 |
kins | -0.030 |
ins | -0.030 |
ited | -0.030 |
eling | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
He | +0.155 |
They | +0.141 |
automobile | +0.136 |
tr | +0.132 |
sidewalk | +0.130 |
Neg logit contributions | |
ited | -0.254 |
eling | -0.247 |
if | -0.245 |
kins | -0.242 |
re | -0.241 |
Token | started |
Token ID | 2067 |
Feature activation | +5.04e+00 |
Pos logit contributions | |
He | +0.055 |
They | +0.051 |
automobile | +0.049 |
sidewalk | +0.047 |
tr | +0.047 |
Neg logit contributions | |
ited | -0.059 |
eling | -0.058 |
irl | -0.057 |
eking | -0.055 |
if | -0.054 |
Token | organizing |
Token ID | 16924 |
Feature activation | +6.28e+00 |
Pos logit contributions | |
He | +0.087 |
They | +0.076 |
automobile | +0.072 |
sidewalk | +0.069 |
tr | +0.068 |
Neg logit contributions | |
ited | -0.156 |
eling | -0.156 |
irl | -0.156 |
eking | -0.151 |
if | -0.150 |
Token | them |
Token ID | 606 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
He | +0.120 |
They | +0.107 |
The | +0.094 |
automobile | +0.091 |
sidewalk | +0.090 |
Neg logit contributions | |
ited | -0.197 |
uch | -0.192 |
eling | -0.190 |
irl | -0.188 |
eking | -0.185 |
Token | . |
Token ID | 13 |
Feature activation | +6.66e+00 |
Pos logit contributions | |
He | +0.005 |
They | +0.005 |
The | +0.004 |
sidewalk | +0.004 |
automobile | +0.004 |
Neg logit contributions | |
ited | -0.013 |
eling | -0.013 |
irl | -0.013 |
uch | -0.012 |
eking | -0.012 |
Token | soon |
Token ID | 2582 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
eling | +0.118 |
ited | +0.116 |
ins | +0.115 |
kins | +0.112 |
uch | +0.112 |
Neg logit contributions | |
He | -0.072 |
The | -0.064 |
They | -0.064 |
Tim | -0.061 |
She | -0.058 |
Token | nothing |
Token ID | 2147 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
ited | +0.054 |
irl | +0.053 |
eling | +0.053 |
uch | +0.052 |
inations | +0.050 |
Neg logit contributions | |
He | -0.043 |
They | -0.039 |
sidewalk | -0.036 |
The | -0.036 |
automobile | -0.036 |
Token | but |
Token ID | 475 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
irl | +0.070 |
eling | +0.069 |
ited | +0.069 |
uch | +0.066 |
eking | +0.065 |
Neg logit contributions | |
He | -0.054 |
They | -0.049 |
The | -0.045 |
Tim | -0.043 |
der | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ited | +0.047 |
eling | +0.046 |
uch | +0.045 |
irl | +0.045 |
eking | +0.045 |
Neg logit contributions | |
He | -0.025 |
They | -0.022 |
The | -0.021 |
It | -0.019 |
She | -0.019 |
Token | small |
Token ID | 1402 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
He | +0.024 |
They | +0.021 |
automobile | +0.021 |
sidewalk | +0.020 |
The | +0.020 |
Neg logit contributions | |
eling | -0.028 |
ited | -0.028 |
eking | -0.027 |
irl | -0.027 |
uch | -0.026 |
Token | spe |
Token ID | 693 |
Feature activation | -1.76e+01 |
Pos logit contributions | |
He | +0.043 |
They | +0.039 |
automobile | +0.036 |
sidewalk | +0.036 |
The | +0.035 |
Neg logit contributions | |
ited | -0.055 |
eling | -0.053 |
irl | -0.053 |
eking | -0.051 |
uch | -0.051 |
Token | ck |
Token ID | 694 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
re | +0.328 |
irl | +0.325 |
kins | +0.323 |
ins | +0.322 |
eling | +0.316 |
Neg logit contributions | |
He | -0.468 |
automobile | -0.453 |
They | -0.433 |
The | -0.425 |
arrow | -0.422 |
Token | in |
Token ID | 287 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
He | +0.018 |
They | +0.016 |
automobile | +0.015 |
The | +0.015 |
sidewalk | +0.015 |
Neg logit contributions | |
ited | -0.019 |
irl | -0.019 |
eling | -0.019 |
uch | -0.018 |
eking | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
uch | +0.027 |
eling | +0.027 |
eking | +0.027 |
ited | +0.027 |
inations | +0.027 |
Neg logit contributions | |
He | -0.012 |
They | -0.011 |
The | -0.010 |
automobile | -0.009 |
It | -0.009 |
Token | sky |
Token ID | 6766 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
He | +0.022 |
They | +0.020 |
automobile | +0.019 |
The | +0.018 |
sidewalk | +0.018 |
Neg logit contributions | |
ited | -0.028 |
eling | -0.027 |
irl | -0.027 |
eking | -0.026 |
uch | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
ited | +0.061 |
eling | +0.060 |
irl | +0.060 |
uch | +0.058 |
eking | +0.058 |
Neg logit contributions | |
He | -0.024 |
They | -0.020 |
The | -0.017 |
automobile | -0.017 |
sidewalk | -0.017 |
Token | The |
Token ID | 383 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
He | +0.069 |
automobile | +0.065 |
sidewalk | +0.064 |
tr | +0.061 |
pavement | +0.058 |
Neg logit contributions | |
irl | -0.184 |
ited | -0.179 |
eling | -0.178 |
if | -0.175 |
re | -0.175 |
Token | eggs |
Token ID | 9653 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
He | +0.026 |
They | +0.023 |
automobile | +0.022 |
sidewalk | +0.021 |
The | +0.021 |
Neg logit contributions | |
ited | -0.044 |
eling | -0.043 |
irl | -0.043 |
eking | -0.041 |
uch | -0.041 |
Token | were |
Token ID | 547 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
ited | +0.055 |
irl | +0.054 |
eling | +0.053 |
inations | +0.052 |
iley | +0.051 |
Neg logit contributions | |
He | -0.038 |
They | -0.035 |
The | -0.032 |
roof | -0.031 |
sidewalk | -0.030 |
Token | blue |
Token ID | 4171 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
ited | +0.118 |
eling | +0.111 |
uch | +0.111 |
re | +0.110 |
irl | +0.110 |
Neg logit contributions | |
He | -0.071 |
They | -0.063 |
for | -0.061 |
Tim | -0.058 |
The | -0.057 |
Token | and |
Token ID | 290 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ited | +0.026 |
eling | +0.026 |
irl | +0.025 |
eking | +0.025 |
uch | +0.025 |
Neg logit contributions | |
He | -0.012 |
They | -0.010 |
automobile | -0.009 |
sidewalk | -0.009 |
The | -0.009 |
Token | spe |
Token ID | 693 |
Feature activation | -1.68e+01 |
Pos logit contributions | |
ited | +0.038 |
irl | +0.036 |
eling | +0.036 |
uch | +0.035 |
inations | +0.035 |
Neg logit contributions | |
He | -0.026 |
They | -0.025 |
The | -0.023 |
Tim | -0.021 |
It | -0.021 |
Token | ck |
Token ID | 694 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
irl | +0.326 |
inations | +0.318 |
re | +0.310 |
ited | +0.298 |
kins | +0.292 |
Neg logit contributions | |
automobile | -0.428 |
He | -0.428 |
They | -0.411 |
sidewalk | -0.402 |
The | -0.392 |
Token | led |
Token ID | 992 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
He | +0.027 |
They | +0.025 |
The | +0.022 |
automobile | +0.022 |
sidewalk | +0.021 |
Neg logit contributions | |
ited | -0.040 |
irl | -0.039 |
eling | -0.039 |
uch | -0.038 |
eking | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +5.93e+00 |
Pos logit contributions | |
He | +0.055 |
They | +0.049 |
automobile | +0.044 |
sidewalk | +0.044 |
The | +0.043 |
Neg logit contributions | |
ited | -0.103 |
eling | -0.102 |
irl | -0.101 |
eking | -0.099 |
if | -0.097 |
Token | They |
Token ID | 1119 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
He | +0.065 |
automobile | +0.062 |
sidewalk | +0.061 |
tr | +0.058 |
pavement | +0.056 |
Neg logit contributions | |
irl | -0.199 |
if | -0.192 |
re | -0.191 |
ited | -0.190 |
eling | -0.189 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
He | +0.031 |
They | +0.029 |
The | +0.027 |
Tim | +0.025 |
arrow | +0.025 |
Neg logit contributions | |
ited | -0.039 |
if | -0.038 |
irl | -0.038 |
uch | -0.038 |
eling | -0.038 |
Token | away |
Token ID | 1497 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
He | +0.057 |
They | +0.053 |
sidewalk | +0.051 |
automobile | +0.051 |
tr | +0.049 |
Neg logit contributions | |
ited | -0.059 |
irl | -0.056 |
eling | -0.056 |
eking | -0.055 |
if | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | -7.42e-01 |
Pos logit contributions | |
He | +0.020 |
They | +0.018 |
automobile | +0.016 |
sidewalk | +0.016 |
The | +0.016 |
Neg logit contributions | |
ited | -0.037 |
eling | -0.036 |
irl | -0.036 |
eking | -0.036 |
if | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
ited | +0.020 |
eling | +0.020 |
irl | +0.020 |
uch | +0.019 |
eking | +0.019 |
Neg logit contributions | |
He | -0.016 |
They | -0.015 |
The | -0.014 |
automobile | -0.013 |
sidewalk | -0.013 |
Token | bright |
Token ID | 6016 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
He | +0.059 |
They | +0.052 |
automobile | +0.051 |
sidewalk | +0.050 |
The | +0.049 |
Neg logit contributions | |
eling | -0.068 |
ited | -0.066 |
irl | -0.065 |
eking | -0.064 |
ins | -0.063 |
Token | blue |
Token ID | 4171 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
He | +0.052 |
They | +0.046 |
automobile | +0.044 |
The | +0.043 |
sidewalk | +0.042 |
Neg logit contributions | |
eling | -0.067 |
ited | -0.066 |
irl | -0.065 |
uch | -0.064 |
eking | -0.064 |
Token | spe |
Token ID | 693 |
Feature activation | -1.68e+01 |
Pos logit contributions | |
He | +0.033 |
They | +0.030 |
The | +0.027 |
sidewalk | +0.027 |
automobile | +0.027 |
Neg logit contributions | |
ited | -0.035 |
irl | -0.033 |
eling | -0.033 |
uch | -0.032 |
if | -0.032 |
Token | ck |
Token ID | 694 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
re | +0.336 |
irl | +0.333 |
eling | +0.333 |
kins | +0.327 |
ins | +0.317 |
Neg logit contributions | |
He | -0.412 |
automobile | -0.411 |
arrow | -0.382 |
They | -0.376 |
sidewalk | -0.375 |
Token | in |
Token ID | 287 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
He | +0.062 |
They | +0.056 |
automobile | +0.052 |
sidewalk | +0.052 |
The | +0.051 |
Neg logit contributions | |
ited | -0.072 |
eling | -0.071 |
irl | -0.070 |
eking | -0.068 |
uch | -0.067 |
Token | the |
Token ID | 262 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
He | +0.009 |
They | +0.008 |
The | +0.007 |
automobile | +0.007 |
sidewalk | +0.007 |
Neg logit contributions | |
ited | -0.016 |
eling | -0.016 |
irl | -0.016 |
uch | -0.015 |
eking | -0.015 |
Token | sky |
Token ID | 6766 |
Feature activation | +8.81e-02 |
Pos logit contributions | |
He | +0.058 |
They | +0.053 |
automobile | +0.047 |
The | +0.047 |
sidewalk | +0.047 |
Neg logit contributions | |
ited | -0.069 |
eling | -0.066 |
irl | -0.066 |
uch | -0.064 |
eking | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +6.59e+00 |
Pos logit contributions | |
He | +0.001 |
They | +0.001 |
automobile | +0.001 |
sidewalk | +0.001 |
The | +0.001 |
Neg logit contributions | |
ited | -0.003 |
eling | -0.003 |
irl | -0.003 |
eking | -0.003 |
if | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
He | +0.016 |
They | +0.013 |
automobile | +0.012 |
sidewalk | +0.012 |
The | +0.012 |
Neg logit contributions | |
ited | -0.037 |
eling | -0.036 |
irl | -0.036 |
eking | -0.036 |
if | -0.035 |
Token | They |
Token ID | 1119 |
Feature activation | +7.47e-01 |
Pos logit contributions | |
automobile | +0.090 |
tr | +0.086 |
bike | +0.080 |
arrow | +0.079 |
cop | +0.078 |
Neg logit contributions | |
re | -0.157 |
irl | -0.155 |
ins | -0.150 |
ona | -0.149 |
kins | -0.149 |
Token | were |
Token ID | 547 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
He | +0.017 |
They | +0.016 |
The | +0.014 |
automobile | +0.014 |
sidewalk | +0.014 |
Neg logit contributions | |
ited | -0.020 |
eling | -0.020 |
irl | -0.019 |
uch | -0.019 |
eking | -0.019 |
Token | blue |
Token ID | 4171 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ited | +0.136 |
uch | +0.126 |
eling | +0.124 |
irl | +0.123 |
if | +0.119 |
Neg logit contributions | |
He | -0.103 |
They | -0.097 |
The | -0.088 |
sidewalk | -0.085 |
She | -0.085 |
Token | and |
Token ID | 290 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ited | +0.033 |
eling | +0.032 |
irl | +0.032 |
eking | +0.031 |
if | +0.031 |
Neg logit contributions | |
He | -0.022 |
They | -0.019 |
automobile | -0.018 |
sidewalk | -0.017 |
The | -0.017 |
Token | spe |
Token ID | 693 |
Feature activation | -1.65e+01 |
Pos logit contributions | |
ited | +0.047 |
eling | +0.045 |
irl | +0.045 |
uch | +0.044 |
eking | +0.044 |
Neg logit contributions | |
He | -0.037 |
They | -0.034 |
The | -0.031 |
automobile | -0.029 |
sidewalk | -0.029 |
Token | ck |
Token ID | 694 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
inations | +0.329 |
irl | +0.314 |
re | +0.310 |
kins | +0.301 |
uch | +0.298 |
Neg logit contributions | |
automobile | -0.419 |
He | -0.404 |
They | -0.398 |
The | -0.383 |
sidewalk | -0.380 |
Token | led |
Token ID | 992 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
He | +0.007 |
They | +0.007 |
The | +0.006 |
automobile | +0.006 |
sidewalk | +0.006 |
Neg logit contributions | |
ited | -0.011 |
irl | -0.011 |
eling | -0.011 |
uch | -0.010 |
eking | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +5.13e+00 |
Pos logit contributions | |
He | +0.039 |
They | +0.035 |
automobile | +0.032 |
sidewalk | +0.032 |
The | +0.031 |
Neg logit contributions | |
ited | -0.060 |
eling | -0.059 |
irl | -0.059 |
eking | -0.057 |
if | -0.056 |
Token | L |
Token ID | 406 |
Feature activation | -7.62e+00 |
Pos logit contributions | |
automobile | +0.077 |
He | +0.071 |
tr | +0.069 |
cop | +0.069 |
arrow | +0.068 |
Neg logit contributions | |
irl | -0.155 |
re | -0.154 |
eling | -0.146 |
ins | -0.146 |
kins | -0.144 |
Token | ila |
Token ID | 10102 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ited | +0.185 |
eling | +0.171 |
if | +0.160 |
inations | +0.158 |
eking | +0.156 |
Neg logit contributions | |
He | -0.191 |
They | -0.183 |
She | -0.167 |
It | -0.165 |
The | -0.165 |
Token | of |
Token ID | 286 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
He | +0.022 |
They | +0.019 |
automobile | +0.018 |
sidewalk | +0.018 |
tr | +0.017 |
Neg logit contributions | |
ited | -0.046 |
eling | -0.045 |
irl | -0.045 |
eking | -0.044 |
re | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
ited | +0.021 |
eling | +0.021 |
irl | +0.020 |
eking | +0.020 |
uch | +0.020 |
Neg logit contributions | |
He | -0.013 |
They | -0.011 |
automobile | -0.010 |
The | -0.010 |
sidewalk | -0.010 |
Token | frog |
Token ID | 21264 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
He | +0.017 |
They | +0.016 |
The | +0.014 |
She | +0.014 |
It | +0.014 |
Neg logit contributions | |
ited | -0.023 |
eling | -0.022 |
irl | -0.022 |
if | -0.022 |
uch | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +5.19e+00 |
Pos logit contributions | |
ited | +0.010 |
eling | +0.010 |
irl | +0.010 |
eking | +0.009 |
uch | +0.009 |
Neg logit contributions | |
He | -0.007 |
They | -0.006 |
automobile | -0.005 |
sidewalk | -0.005 |
The | -0.005 |
Token | He |
Token ID | 679 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
He | +0.060 |
automobile | +0.058 |
sidewalk | +0.055 |
tr | +0.053 |
They | +0.052 |
Neg logit contributions | |
ited | -0.171 |
irl | -0.170 |
re | -0.169 |
eling | -0.168 |
if | -0.167 |
Token | spe |
Token ID | 693 |
Feature activation | -1.65e+01 |
Pos logit contributions | |
He | +0.043 |
They | +0.038 |
The | +0.035 |
arrow | +0.034 |
Tim | +0.033 |
Neg logit contributions | |
ited | -0.053 |
eling | -0.052 |
eking | -0.051 |
irl | -0.050 |
uch | -0.050 |
Token | ared |
Token ID | 1144 |
Feature activation | +3.77e+00 |
Pos logit contributions | |
irl | +0.298 |
re | +0.284 |
kins | +0.276 |
inations | +0.275 |
uch | +0.274 |
Neg logit contributions | |
He | -0.457 |
automobile | -0.441 |
They | -0.422 |
sidewalk | -0.412 |
arrow | -0.407 |
Token | the |
Token ID | 262 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
He | +0.069 |
They | +0.061 |
automobile | +0.055 |
sidewalk | +0.054 |
The | +0.054 |
Neg logit contributions | |
ited | -0.116 |
eling | -0.115 |
irl | -0.114 |
eking | -0.110 |
uch | -0.110 |
Token | lazy |
Token ID | 16931 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
He | +0.034 |
They | +0.031 |
automobile | +0.028 |
The | +0.027 |
sidewalk | +0.027 |
Neg logit contributions | |
ited | -0.044 |
eling | -0.043 |
irl | -0.043 |
eking | -0.041 |
uch | -0.041 |
Token | frog |
Token ID | 21264 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
He | +0.067 |
They | +0.063 |
automobile | +0.058 |
tr | +0.058 |
The | +0.057 |
Neg logit contributions | |
ited | -0.071 |
eling | -0.069 |
irl | -0.068 |
uch | -0.066 |
re | -0.065 |
Token | and |
Token ID | 290 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
He | +0.011 |
They | +0.009 |
The | +0.009 |
sidewalk | +0.008 |
automobile | +0.008 |
Neg logit contributions | |
ited | -0.013 |
irl | -0.013 |
eling | -0.013 |
uch | -0.013 |
inations | -0.013 |
Token | The |
Token ID | 383 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
automobile | +0.099 |
tr | +0.098 |
cop | +0.088 |
arrow | +0.088 |
sidewalk | +0.088 |
Neg logit contributions | |
irl | -0.173 |
re | -0.172 |
eling | -0.170 |
ited | -0.166 |
ona | -0.165 |
Token | eggs |
Token ID | 9653 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
He | +0.025 |
automobile | +0.025 |
roof | +0.024 |
They | +0.022 |
sidewalk | +0.021 |
Neg logit contributions | |
ited | -0.068 |
irl | -0.066 |
eling | -0.066 |
inations | -0.064 |
eking | -0.062 |
Token | were |
Token ID | 547 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
ited | +0.053 |
eling | +0.052 |
irl | +0.052 |
inations | +0.049 |
if | +0.049 |
Neg logit contributions | |
He | -0.040 |
They | -0.038 |
The | -0.033 |
She | -0.033 |
It | -0.033 |
Token | blue |
Token ID | 4171 |
Feature activation | -6.19e-01 |
Pos logit contributions | |
ited | +0.118 |
eling | +0.110 |
uch | +0.109 |
irl | +0.108 |
inations | +0.106 |
Neg logit contributions | |
He | -0.069 |
They | -0.062 |
The | -0.056 |
automobile | -0.054 |
She | -0.054 |
Token | and |
Token ID | 290 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
ited | +0.020 |
eling | +0.019 |
irl | +0.019 |
re | +0.019 |
eking | +0.019 |
Neg logit contributions | |
He | -0.010 |
They | -0.009 |
automobile | -0.008 |
sidewalk | -0.008 |
The | -0.008 |
Token | spe |
Token ID | 693 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
ited | +0.024 |
eling | +0.023 |
irl | +0.023 |
eking | +0.023 |
uch | +0.023 |
Neg logit contributions | |
He | -0.016 |
They | -0.015 |
The | -0.014 |
It | -0.013 |
She | -0.013 |
Token | ck |
Token ID | 694 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
inations | +0.311 |
irl | +0.303 |
re | +0.299 |
ited | +0.295 |
uch | +0.284 |
Neg logit contributions | |
automobile | -0.414 |
He | -0.404 |
They | -0.397 |
sidewalk | -0.384 |
The | -0.382 |
Token | led |
Token ID | 992 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
He | +0.028 |
They | +0.025 |
The | +0.022 |
automobile | +0.022 |
sidewalk | +0.021 |
Neg logit contributions | |
ited | -0.042 |
irl | -0.041 |
eling | -0.041 |
uch | -0.039 |
eking | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +6.03e+00 |
Pos logit contributions | |
He | +0.061 |
They | +0.055 |
automobile | +0.050 |
sidewalk | +0.050 |
The | +0.049 |
Neg logit contributions | |
ited | -0.092 |
eling | -0.090 |
irl | -0.089 |
eking | -0.087 |
if | -0.086 |
Token | L |
Token ID | 406 |
Feature activation | -6.71e+00 |
Pos logit contributions | |
automobile | +0.098 |
He | +0.095 |
tr | +0.092 |
cop | +0.088 |
sidewalk | +0.088 |
Neg logit contributions | |
irl | -0.176 |
re | -0.172 |
eling | -0.169 |
ited | -0.165 |
ins | -0.165 |
Token | ila |
Token ID | 10102 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
ited | +0.171 |
eling | +0.153 |
if | +0.146 |
inations | +0.146 |
uch | +0.145 |
Neg logit contributions | |
He | -0.158 |
They | -0.154 |
It | -0.139 |
The | -0.139 |
She | -0.137 |
Token | . |
Token ID | 13 |
Feature activation | +5.19e+00 |
Pos logit contributions | |
He | +0.011 |
They | +0.010 |
The | +0.010 |
Tim | +0.009 |
She | +0.009 |
Neg logit contributions | |
irl | -0.016 |
eling | -0.016 |
ited | -0.016 |
uch | -0.016 |
eking | -0.016 |
Token | We |
Token ID | 775 |
Feature activation | -4.36e+00 |
Pos logit contributions | |
He | +0.053 |
sidewalk | +0.049 |
automobile | +0.048 |
tr | +0.045 |
pavement | +0.045 |
Neg logit contributions | |
irl | -0.176 |
ited | -0.174 |
re | -0.173 |
eling | -0.170 |
if | -0.170 |
Token | had |
Token ID | 550 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
ited | +0.108 |
eling | +0.107 |
irl | +0.105 |
eking | +0.103 |
if | +0.103 |
Neg logit contributions | |
He | -0.106 |
They | -0.098 |
The | -0.090 |
automobile | -0.089 |
sidewalk | -0.087 |
Token | tea |
Token ID | 8887 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
ited | +0.063 |
eling | +0.062 |
irl | +0.062 |
eking | +0.060 |
sie | +0.059 |
Neg logit contributions | |
He | -0.040 |
They | -0.038 |
The | -0.035 |
automobile | -0.033 |
Tim | -0.033 |
Token | with |
Token ID | 351 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
ited | +0.142 |
eling | +0.141 |
irl | +0.134 |
eking | +0.133 |
if | +0.131 |
Neg logit contributions | |
He | -0.067 |
They | -0.063 |
tr | -0.053 |
automobile | -0.052 |
The | -0.052 |
Token | Mr |
Token ID | 1770 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
ited | +0.091 |
eling | +0.084 |
irl | +0.083 |
eking | +0.083 |
uch | +0.082 |
Neg logit contributions | |
They | -0.053 |
He | -0.053 |
The | -0.048 |
Tim | -0.048 |
automobile | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
ited | +0.521 |
eling | +0.514 |
irl | +0.509 |
uch | +0.499 |
if | +0.495 |
Neg logit contributions | |
He | -0.227 |
They | -0.195 |
sidewalk | -0.176 |
automobile | -0.175 |
The | -0.169 |
Token | Bear |
Token ID | 14732 |
Feature activation | -5.30e+00 |
Pos logit contributions | |
He | +0.041 |
They | +0.038 |
sidewalk | +0.036 |
automobile | +0.036 |
pavement | +0.035 |
Neg logit contributions | |
ited | -0.047 |
irl | -0.046 |
eling | -0.046 |
re | -0.045 |
uch | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
ited | +0.177 |
eling | +0.175 |
irl | +0.173 |
uch | +0.168 |
eking | +0.168 |
Neg logit contributions | |
He | -0.082 |
They | -0.070 |
automobile | -0.063 |
sidewalk | -0.062 |
The | -0.061 |
Token | we |
Token ID | 356 |
Feature activation | -8.11e-01 |
Pos logit contributions | |
ited | +0.048 |
eling | +0.048 |
irl | +0.047 |
uch | +0.046 |
re | +0.045 |
Neg logit contributions | |
He | -0.033 |
They | -0.029 |
sidewalk | -0.028 |
automobile | -0.028 |
The | -0.026 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.97e+00 |
Pos logit contributions | |
ited | +0.021 |
eling | +0.020 |
irl | +0.020 |
eking | +0.020 |
if | +0.019 |
Neg logit contributions | |
He | -0.019 |
They | -0.018 |
The | -0.016 |
automobile | -0.016 |
sidewalk | -0.016 |
Token | " |
Token ID | 366 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
irl | +0.071 |
eling | +0.069 |
ited | +0.068 |
eking | +0.068 |
re | +0.066 |
Neg logit contributions | |
He | -0.049 |
They | -0.044 |
tr | -0.041 |
der | -0.041 |
automobile | -0.040 |
Token | ch |
Token ID | 354 |
Feature activation | -6.52e+00 |
Pos logit contributions | |
irl | +0.056 |
eling | +0.055 |
eking | +0.054 |
ited | +0.054 |
inations | +0.053 |
Neg logit contributions | |
He | -0.040 |
They | -0.035 |
The | -0.033 |
automobile | -0.032 |
sidewalk | -0.031 |
Token | oo |
Token ID | 2238 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
ited | +0.161 |
eling | +0.153 |
irl | +0.149 |
eking | +0.145 |
if | +0.143 |
Neg logit contributions | |
He | -0.166 |
They | -0.153 |
The | -0.142 |
sidewalk | -0.138 |
automobile | -0.138 |
Token | cho |
Token ID | 1727 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
ited | +0.043 |
eling | +0.042 |
irl | +0.041 |
eking | +0.040 |
uch | +0.040 |
Neg logit contributions | |
He | -0.042 |
They | -0.038 |
automobile | -0.035 |
The | -0.035 |
sidewalk | -0.035 |
Token | o |
Token ID | 78 |
Feature activation | -7.02e+00 |
Pos logit contributions | |
ited | +0.291 |
eling | +0.285 |
irl | +0.280 |
eking | +0.263 |
inations | +0.259 |
Neg logit contributions | |
He | -0.437 |
They | -0.408 |
The | -0.383 |
automobile | -0.381 |
tr | -0.379 |
Token | cho |
Token ID | 1727 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
ited | +0.176 |
eling | +0.172 |
irl | +0.171 |
eking | +0.162 |
inations | +0.161 |
Neg logit contributions | |
He | -0.171 |
They | -0.156 |
The | -0.144 |
automobile | -0.144 |
sidewalk | -0.141 |
Token | o |
Token ID | 78 |
Feature activation | -7.44e+00 |
Pos logit contributions | |
ited | +0.304 |
eling | +0.299 |
irl | +0.293 |
eking | +0.274 |
inations | +0.273 |
Neg logit contributions | |
He | -0.451 |
They | -0.421 |
The | -0.395 |
automobile | -0.393 |
tr | -0.392 |
Token | ". |
Token ID | 1911 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
ited | +0.186 |
eling | +0.182 |
irl | +0.181 |
eking | +0.171 |
inations | +0.170 |
Neg logit contributions | |
He | -0.182 |
They | -0.166 |
The | -0.153 |
automobile | -0.152 |
sidewalk | -0.150 |
Token |
|
Token ID | 198 |
Feature activation | -7.38e+00 |
Pos logit contributions | |
He | +0.044 |
automobile | +0.040 |
sidewalk | +0.039 |
They | +0.039 |
tr | +0.038 |
Neg logit contributions | |
irl | -0.089 |
ited | -0.089 |
eling | -0.089 |
re | -0.088 |
if | -0.086 |
Token |
|
Token ID | 198 |
Feature activation | -6.67e+00 |
Pos logit contributions | |
ited | +0.243 |
eling | +0.238 |
irl | +0.234 |
eking | +0.231 |
uch | +0.229 |
Neg logit contributions | |
He | -0.128 |
They | -0.113 |
The | -0.099 |
automobile | -0.095 |
sidewalk | -0.094 |
Token | They |
Token ID | 2990 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
ited | +0.178 |
eling | +0.175 |
irl | +0.173 |
eking | +0.167 |
uch | +0.166 |
Neg logit contributions | |
He | -0.151 |
They | -0.137 |
automobile | -0.126 |
The | -0.125 |
sidewalk | -0.124 |
Token | ." |
Token ID | 526 |
Feature activation | +8.51e-01 |
Pos logit contributions | |
sidewalk | +0.081 |
He | +0.078 |
automobile | +0.077 |
arrow | +0.073 |
cop | +0.072 |
Neg logit contributions | |
uch | -0.148 |
irl | -0.146 |
if | -0.140 |
ited | -0.140 |
eling | -0.139 |
Token |
|
Token ID | 198 |
Feature activation | -7.22e+00 |
Pos logit contributions | |
He | +0.013 |
They | +0.012 |
automobile | +0.011 |
sidewalk | +0.011 |
tr | +0.010 |
Neg logit contributions | |
ited | -0.027 |
eling | -0.027 |
irl | -0.027 |
re | -0.026 |
uch | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -6.72e+00 |
Pos logit contributions | |
ited | +0.234 |
eling | +0.230 |
irl | +0.226 |
eking | +0.222 |
uch | +0.222 |
Neg logit contributions | |
He | -0.128 |
They | -0.113 |
The | -0.100 |
automobile | -0.096 |
sidewalk | -0.095 |
Token | They |
Token ID | 2990 |
Feature activation | -9.18e-01 |
Pos logit contributions | |
ited | +0.187 |
eling | +0.185 |
irl | +0.179 |
eking | +0.178 |
uch | +0.176 |
Neg logit contributions | |
He | -0.148 |
They | -0.135 |
The | -0.121 |
automobile | -0.120 |
sidewalk | -0.118 |
Token | hugged |
Token ID | 41130 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
ited | +0.023 |
eling | +0.023 |
irl | +0.022 |
eking | +0.021 |
uch | +0.021 |
Neg logit contributions | |
He | -0.022 |
They | -0.020 |
automobile | -0.019 |
sidewalk | -0.019 |
The | -0.018 |
Token | Mr |
Token ID | 1770 |
Feature activation | -1.50e+01 |
Pos logit contributions | |
years | +0.042 |
irl | +0.041 |
ited | +0.040 |
eling | +0.040 |
kins | +0.039 |
Neg logit contributions | |
He | -0.031 |
tr | -0.028 |
sidewalk | -0.027 |
curb | -0.026 |
automobile | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
kins | +0.436 |
ins | +0.431 |
re | +0.431 |
iri | +0.421 |
uch | +0.417 |
Neg logit contributions | |
sidewalk | -0.238 |
He | -0.237 |
curb | -0.226 |
pavement | -0.225 |
automobile | -0.225 |
Token | Fl |
Token ID | 1610 |
Feature activation | -8.60e+00 |
Pos logit contributions | |
He | +0.043 |
They | +0.038 |
automobile | +0.036 |
sidewalk | +0.036 |
The | +0.035 |
Neg logit contributions | |
ited | -0.058 |
irl | -0.057 |
eling | -0.057 |
re | -0.056 |
if | -0.055 |
Token | uffy |
Token ID | 15352 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
ited | +0.125 |
eling | +0.120 |
irl | +0.115 |
eking | +0.107 |
inations | +0.106 |
Neg logit contributions | |
He | -0.304 |
They | -0.287 |
automobile | -0.270 |
The | -0.268 |
der | -0.263 |
Token | and |
Token ID | 290 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
ited | +0.108 |
eling | +0.106 |
re | +0.103 |
years | +0.102 |
eking | +0.100 |
Neg logit contributions | |
He | -0.065 |
sidewalk | -0.061 |
pavement | -0.057 |
automobile | -0.057 |
They | -0.056 |
Token | went |
Token ID | 1816 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ited | +0.048 |
eling | +0.046 |
irl | +0.042 |
re | +0.041 |
sie | +0.041 |
Neg logit contributions | |
He | -0.053 |
sidewalk | -0.052 |
pavement | -0.051 |
automobile | -0.050 |
They | -0.049 |
Token | hair |
Token ID | 4190 |
Feature activation | -4.86e+00 |
Pos logit contributions | |
ited | +0.166 |
uch | +0.160 |
irl | +0.158 |
eling | +0.158 |
eking | +0.147 |
Neg logit contributions | |
He | -0.131 |
They | -0.118 |
The | -0.107 |
automobile | -0.106 |
sidewalk | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
ited | +0.167 |
eling | +0.165 |
irl | +0.163 |
eking | +0.158 |
uch | +0.158 |
Neg logit contributions | |
He | -0.072 |
They | -0.062 |
automobile | -0.054 |
The | -0.053 |
sidewalk | -0.053 |
Token | His |
Token ID | 2399 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
He | +0.035 |
sidewalk | +0.033 |
automobile | +0.033 |
tr | +0.031 |
der | +0.031 |
Neg logit contributions | |
irl | -0.118 |
ited | -0.116 |
re | -0.115 |
eling | -0.114 |
if | -0.113 |
Token | name |
Token ID | 1438 |
Feature activation | -4.76e+00 |
Pos logit contributions | |
He | +0.032 |
They | +0.029 |
automobile | +0.026 |
The | +0.026 |
sidewalk | +0.026 |
Neg logit contributions | |
ited | -0.046 |
eling | -0.045 |
irl | -0.045 |
eking | -0.043 |
uch | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
ited | +0.123 |
eling | +0.121 |
irl | +0.119 |
eking | +0.115 |
uch | +0.115 |
Neg logit contributions | |
He | -0.111 |
They | -0.101 |
automobile | -0.093 |
The | -0.092 |
sidewalk | -0.092 |
Token | Mr |
Token ID | 1770 |
Feature activation | -1.50e+01 |
Pos logit contributions | |
irl | +0.154 |
ited | +0.154 |
eling | +0.153 |
uch | +0.148 |
eking | +0.147 |
Neg logit contributions | |
He | -0.083 |
They | -0.074 |
automobile | -0.068 |
sidewalk | -0.068 |
tr | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
ited | +0.509 |
eling | +0.502 |
re | +0.493 |
irl | +0.491 |
if | +0.489 |
Neg logit contributions | |
He | -0.202 |
They | -0.172 |
sidewalk | -0.168 |
automobile | -0.165 |
tr | -0.151 |
Token | Smith |
Token ID | 4176 |
Feature activation | -4.58e+00 |
Pos logit contributions | |
He | +0.066 |
They | +0.060 |
automobile | +0.056 |
sidewalk | +0.056 |
The | +0.054 |
Neg logit contributions | |
ited | -0.082 |
eling | -0.081 |
irl | -0.081 |
if | -0.077 |
uch | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
ited | +0.153 |
eling | +0.151 |
irl | +0.149 |
if | +0.145 |
re | +0.145 |
Neg logit contributions | |
He | -0.068 |
They | -0.059 |
sidewalk | -0.054 |
automobile | -0.053 |
The | -0.051 |
Token |
|
Token ID | 198 |
Feature activation | -6.47e+00 |
Pos logit contributions | |
He | +0.070 |
They | +0.062 |
sidewalk | +0.060 |
automobile | +0.060 |
tr | +0.056 |
Neg logit contributions | |
ited | -0.156 |
irl | -0.154 |
eling | -0.153 |
if | -0.150 |
re | -0.150 |
Token |
|
Token ID | 198 |
Feature activation | -5.75e+00 |
Pos logit contributions | |
ited | +0.204 |
eling | +0.202 |
irl | +0.199 |
eking | +0.194 |
uch | +0.193 |
Neg logit contributions | |
He | -0.117 |
They | -0.103 |
The | -0.092 |
automobile | -0.091 |
sidewalk | -0.089 |
Token | a |
Token ID | 257 |
Feature activation | -8.96e-01 |
Pos logit contributions | |
ited | +0.124 |
irl | +0.122 |
eling | +0.121 |
re | +0.120 |
eking | +0.119 |
Neg logit contributions | |
He | -0.059 |
They | -0.050 |
sidewalk | -0.049 |
automobile | -0.048 |
der | -0.046 |
Token | wise |
Token ID | 10787 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
ited | +0.025 |
eling | +0.025 |
irl | +0.024 |
eking | +0.024 |
uch | +0.023 |
Neg logit contributions | |
He | -0.020 |
They | -0.018 |
automobile | -0.017 |
sidewalk | -0.016 |
The | -0.016 |
Token | old |
Token ID | 1468 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
He | +0.009 |
They | +0.008 |
tr | +0.008 |
Then | +0.008 |
sidewalk | +0.008 |
Neg logit contributions | |
ited | -0.012 |
sie | -0.012 |
years | -0.011 |
year | -0.011 |
irl | -0.011 |
Token | owl |
Token ID | 39610 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
He | +0.033 |
They | +0.030 |
Then | +0.028 |
The | +0.027 |
der | +0.027 |
Neg logit contributions | |
ited | -0.043 |
sie | -0.042 |
if | -0.042 |
irl | -0.042 |
kins | -0.042 |
Token | named |
Token ID | 3706 |
Feature activation | -5.36e+00 |
Pos logit contributions | |
ited | +0.015 |
eling | +0.015 |
irl | +0.015 |
eking | +0.015 |
uch | +0.015 |
Neg logit contributions | |
He | -0.008 |
They | -0.007 |
automobile | -0.006 |
The | -0.006 |
sidewalk | -0.006 |
Token | Mr |
Token ID | 1770 |
Feature activation | -1.50e+01 |
Pos logit contributions | |
eling | +0.156 |
ited | +0.154 |
irl | +0.153 |
re | +0.149 |
if | +0.149 |
Neg logit contributions | |
He | -0.090 |
They | -0.079 |
automobile | -0.077 |
sidewalk | -0.077 |
tr | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
ited | +0.498 |
eling | +0.495 |
re | +0.484 |
uch | +0.482 |
if | +0.481 |
Neg logit contributions | |
He | -0.216 |
They | -0.180 |
sidewalk | -0.175 |
automobile | -0.173 |
pavement | -0.159 |
Token | H |
Token ID | 367 |
Feature activation | -7.14e+00 |
Pos logit contributions | |
He | +0.065 |
They | +0.061 |
sidewalk | +0.060 |
automobile | +0.059 |
pavement | +0.058 |
Neg logit contributions | |
if | -0.076 |
re | -0.075 |
ited | -0.075 |
eling | -0.074 |
irl | -0.074 |
Token | oot |
Token ID | 1025 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ited | +0.192 |
eling | +0.188 |
re | +0.186 |
if | +0.186 |
uch | +0.185 |
Neg logit contributions | |
He | -0.156 |
They | -0.145 |
sidewalk | -0.131 |
automobile | -0.131 |
The | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | +5.45e+00 |
Pos logit contributions | |
re | +0.016 |
ited | +0.016 |
eling | +0.015 |
if | +0.015 |
eking | +0.015 |
Neg logit contributions | |
He | -0.010 |
They | -0.009 |
sidewalk | -0.009 |
tr | -0.009 |
automobile | -0.009 |
Token | Mr |
Token ID | 1770 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
sidewalk | +0.043 |
He | +0.042 |
tr | +0.042 |
automobile | +0.042 |
pavement | +0.039 |
Neg logit contributions | |
if | -0.191 |
re | -0.190 |
irl | -0.189 |
ited | -0.187 |
eling | -0.187 |
Token | their |
Token ID | 511 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
ited | +0.035 |
eling | +0.034 |
irl | +0.033 |
uch | +0.033 |
eking | +0.033 |
Neg logit contributions | |
He | -0.016 |
They | -0.015 |
The | -0.013 |
automobile | -0.012 |
Tim | -0.012 |
Token | salad |
Token ID | 20698 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
eling | +0.027 |
ited | +0.026 |
irl | +0.026 |
eking | +0.026 |
uch | +0.025 |
Neg logit contributions | |
He | -0.013 |
They | -0.012 |
The | -0.011 |
sidewalk | -0.010 |
Tim | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
eling | +0.087 |
ited | +0.087 |
irl | +0.083 |
eking | +0.082 |
uch | +0.081 |
Neg logit contributions | |
He | -0.050 |
They | -0.046 |
The | -0.039 |
sidewalk | -0.038 |
automobile | -0.038 |
Token | have |
Token ID | 423 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
irl | +0.032 |
eling | +0.031 |
ited | +0.031 |
kins | +0.030 |
inations | +0.030 |
Neg logit contributions | |
He | -0.024 |
automobile | -0.021 |
They | -0.021 |
tr | -0.020 |
curb | -0.020 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
ited | +0.018 |
eling | +0.018 |
eking | +0.017 |
irl | +0.017 |
uch | +0.017 |
Neg logit contributions | |
He | -0.013 |
They | -0.012 |
The | -0.011 |
Tim | -0.010 |
It | -0.010 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.46e+01 |
Pos logit contributions | |
ited | +0.062 |
eling | +0.060 |
irl | +0.058 |
eking | +0.057 |
uch | +0.056 |
Neg logit contributions | |
He | -0.078 |
They | -0.073 |
The | -0.067 |
sidewalk | -0.066 |
automobile | -0.066 |
Token | n |
Token ID | 77 |
Feature activation | -7.07e+00 |
Pos logit contributions | |
eling | +0.281 |
ited | +0.277 |
ins | +0.275 |
re | +0.249 |
eking | +0.245 |
Neg logit contributions | |
He | -0.402 |
They | -0.398 |
The | -0.394 |
sidewalk | -0.374 |
But | -0.364 |
Token | ics |
Token ID | 873 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ited | +0.108 |
eling | +0.099 |
irl | +0.095 |
if | +0.087 |
eking | +0.086 |
Neg logit contributions | |
He | -0.244 |
They | -0.233 |
The | -0.219 |
automobile | -0.216 |
sidewalk | -0.215 |
Token | together |
Token ID | 1978 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
ited | +0.044 |
eling | +0.044 |
irl | +0.043 |
eking | +0.042 |
uch | +0.041 |
Neg logit contributions | |
He | -0.029 |
They | -0.027 |
The | -0.023 |
automobile | -0.023 |
sidewalk | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +5.09e+00 |
Pos logit contributions | |
ited | +0.091 |
eling | +0.090 |
irl | +0.088 |
eking | +0.087 |
uch | +0.086 |
Neg logit contributions | |
He | -0.024 |
They | -0.021 |
The | -0.016 |
automobile | -0.014 |
sidewalk | -0.014 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
automobile | +0.055 |
He | +0.055 |
sidewalk | +0.052 |
curb | +0.049 |
pavement | +0.048 |
Neg logit contributions | |
irl | -0.171 |
re | -0.170 |
ited | -0.167 |
if | -0.166 |
uch | -0.164 |
Token | belongs |
Token ID | 14448 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
He | +0.038 |
They | +0.035 |
der | +0.033 |
Let | +0.032 |
driver | +0.032 |
Neg logit contributions | |
irl | -0.036 |
if | -0.036 |
eling | -0.036 |
ited | -0.035 |
uch | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
ited | +0.025 |
eling | +0.025 |
irl | +0.025 |
eking | +0.024 |
uch | +0.024 |
Neg logit contributions | |
He | -0.011 |
They | -0.009 |
automobile | -0.008 |
The | -0.008 |
sidewalk | -0.008 |
Token | our |
Token ID | 674 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
ited | +0.091 |
eling | +0.091 |
irl | +0.090 |
eking | +0.086 |
uch | +0.086 |
Neg logit contributions | |
He | -0.075 |
They | -0.068 |
automobile | -0.062 |
sidewalk | -0.062 |
The | -0.062 |
Token | neighbor |
Token ID | 4780 |
Feature activation | -9.18e-01 |
Pos logit contributions | |
ited | +0.098 |
eling | +0.097 |
irl | +0.097 |
uch | +0.093 |
eking | +0.093 |
Neg logit contributions | |
He | -0.082 |
They | -0.074 |
The | -0.067 |
sidewalk | -0.066 |
automobile | -0.066 |
Token | , |
Token ID | 11 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
ited | +0.029 |
eling | +0.028 |
irl | +0.028 |
eking | +0.027 |
uch | +0.027 |
Neg logit contributions | |
He | -0.016 |
They | -0.014 |
automobile | -0.013 |
sidewalk | -0.013 |
The | -0.013 |
Token | Mr |
Token ID | 1770 |
Feature activation | -1.45e+01 |
Pos logit contributions | |
ited | +0.129 |
eling | +0.128 |
irl | +0.127 |
eking | +0.123 |
uch | +0.122 |
Neg logit contributions | |
He | -0.074 |
They | -0.065 |
automobile | -0.058 |
sidewalk | -0.058 |
The | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
ins | +0.440 |
ited | +0.438 |
re | +0.438 |
kins | +0.434 |
eling | +0.433 |
Neg logit contributions | |
He | -0.217 |
sidewalk | -0.213 |
automobile | -0.204 |
pavement | -0.202 |
curb | -0.193 |
Token | Jenkins |
Token ID | 21835 |
Feature activation | -6.43e+00 |
Pos logit contributions | |
He | +0.049 |
sidewalk | +0.048 |
pavement | +0.046 |
They | +0.046 |
automobile | +0.044 |
Neg logit contributions | |
re | -0.065 |
if | -0.062 |
ited | -0.062 |
irl | -0.062 |
uch | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +4.43e+00 |
Pos logit contributions | |
ited | +0.189 |
eling | +0.186 |
irl | +0.185 |
eking | +0.179 |
uch | +0.178 |
Neg logit contributions | |
He | -0.126 |
They | -0.113 |
automobile | -0.102 |
The | -0.101 |
sidewalk | -0.101 |
Token | Let |
Token ID | 3914 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
sidewalk | +0.037 |
pavement | +0.035 |
tr | +0.034 |
automobile | +0.034 |
der | +0.031 |
Neg logit contributions | |
if | -0.153 |
re | -0.153 |
eling | -0.150 |
ited | -0.149 |
irl | -0.149 |
Token | 's |
Token ID | 338 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
He | +0.027 |
They | +0.025 |
The | +0.023 |
automobile | +0.023 |
sidewalk | +0.022 |
Neg logit contributions | |
ited | -0.032 |
irl | -0.032 |
eling | -0.031 |
eking | -0.030 |
inations | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
ited | +0.060 |
eling | +0.060 |
irl | +0.060 |
eking | +0.058 |
if | +0.058 |
Neg logit contributions | |
He | -0.037 |
They | -0.033 |
automobile | -0.030 |
sidewalk | -0.029 |
The | -0.029 |
Token | listening |
Token ID | 8680 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
ited | +0.064 |
eling | +0.063 |
irl | +0.062 |
eking | +0.060 |
uch | +0.060 |
Neg logit contributions | |
He | -0.045 |
They | -0.040 |
automobile | -0.036 |
The | -0.036 |
sidewalk | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +3.86e+00 |
Pos logit contributions | |
He | +0.024 |
They | +0.021 |
automobile | +0.020 |
sidewalk | +0.019 |
The | +0.018 |
Neg logit contributions | |
ited | -0.052 |
eling | -0.052 |
irl | -0.052 |
eking | -0.050 |
if | -0.050 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
He | +0.041 |
automobile | +0.040 |
sidewalk | +0.040 |
tr | +0.037 |
pavement | +0.037 |
Neg logit contributions | |
irl | -0.127 |
ited | -0.125 |
re | -0.125 |
if | -0.122 |
eling | -0.122 |
Token | M |
Token ID | 44 |
Feature activation | -6.79e+00 |
Pos logit contributions | |
ited | +0.341 |
eling | +0.334 |
eking | +0.322 |
irl | +0.321 |
inations | +0.314 |
Neg logit contributions | |
He | -0.224 |
They | -0.201 |
The | -0.183 |
It | -0.169 |
She | -0.167 |
Token | omm |
Token ID | 2002 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
eking | +0.146 |
ited | +0.145 |
eling | +0.144 |
if | +0.141 |
irl | +0.139 |
Neg logit contributions | |
He | -0.190 |
They | -0.171 |
The | -0.167 |
She | -0.159 |
Tim | -0.155 |
Token | a |
Token ID | 64 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
irl | +0.251 |
ited | +0.246 |
eling | +0.227 |
uch | +0.221 |
if | +0.218 |
Neg logit contributions | |
He | -0.455 |
They | -0.421 |
automobile | -0.395 |
The | -0.393 |
tr | -0.385 |
Token | and |
Token ID | 290 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
ited | +0.164 |
eling | +0.161 |
irl | +0.160 |
eking | +0.155 |
uch | +0.154 |
Neg logit contributions | |
He | -0.094 |
They | -0.083 |
automobile | -0.074 |
The | -0.073 |
sidewalk | -0.073 |
Token | Billy |
Token ID | 15890 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
ited | +0.094 |
eling | +0.093 |
irl | +0.092 |
eking | +0.089 |
uch | +0.088 |
Neg logit contributions | |
He | -0.056 |
They | -0.049 |
automobile | -0.045 |
sidewalk | -0.044 |
The | -0.044 |
Token | were |
Token ID | 547 |
Feature activation | -6.16e+00 |
Pos logit contributions | |
ited | +0.068 |
eling | +0.067 |
irl | +0.066 |
uch | +0.063 |
eking | +0.063 |
Neg logit contributions | |
He | -0.059 |
They | -0.053 |
automobile | -0.050 |
sidewalk | -0.049 |
The | -0.049 |
Token | walking |
Token ID | 6155 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
ited | +0.181 |
eling | +0.170 |
irl | +0.168 |
if | +0.168 |
uch | +0.168 |
Neg logit contributions | |
He | -0.135 |
They | -0.125 |
The | -0.115 |
automobile | -0.110 |
She | -0.110 |
Token | 'm |
Token ID | 1101 |
Feature activation | -6.82e+00 |
Pos logit contributions | |
eking | +0.139 |
eling | +0.135 |
if | +0.134 |
kins | +0.133 |
irl | +0.130 |
Neg logit contributions | |
He | -0.107 |
They | -0.094 |
The | -0.084 |
But | -0.084 |
It | -0.083 |
Token | digging |
Token ID | 18894 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
ited | +0.200 |
uch | +0.181 |
if | +0.178 |
irl | +0.177 |
eling | +0.176 |
Neg logit contributions | |
He | -0.155 |
They | -0.139 |
The | -0.131 |
Tim | -0.129 |
She | -0.124 |
Token | tunnels |
Token ID | 22642 |
Feature activation | -8.26e-01 |
Pos logit contributions | |
He | +0.009 |
They | +0.008 |
The | +0.007 |
sidewalk | +0.007 |
automobile | +0.007 |
Neg logit contributions | |
ited | -0.014 |
eling | -0.014 |
irl | -0.013 |
uch | -0.013 |
re | -0.013 |
Token | ," |
Token ID | 553 |
Feature activation | -5.03e+00 |
Pos logit contributions | |
ited | +0.025 |
eling | +0.025 |
irl | +0.025 |
uch | +0.024 |
eking | +0.024 |
Neg logit contributions | |
He | -0.015 |
They | -0.014 |
automobile | -0.012 |
The | -0.012 |
sidewalk | -0.012 |
Token | replied |
Token ID | 8712 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
ited | +0.162 |
eling | +0.159 |
irl | +0.158 |
eking | +0.153 |
uch | +0.152 |
Neg logit contributions | |
He | -0.084 |
They | -0.074 |
automobile | -0.067 |
sidewalk | -0.066 |
The | -0.065 |
Token | Mr |
Token ID | 1770 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
ited | +0.043 |
eling | +0.042 |
irl | +0.042 |
eking | +0.040 |
uch | +0.040 |
Neg logit contributions | |
He | -0.023 |
They | -0.021 |
automobile | -0.018 |
sidewalk | -0.018 |
The | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
re | +0.445 |
ited | +0.438 |
eling | +0.437 |
kins | +0.429 |
ins | +0.428 |
Neg logit contributions | |
He | -0.208 |
sidewalk | -0.198 |
automobile | -0.196 |
They | -0.186 |
pavement | -0.186 |
Token | Mole |
Token ID | 25726 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
He | +0.059 |
They | +0.053 |
The | +0.048 |
automobile | +0.047 |
sidewalk | +0.047 |
Neg logit contributions | |
ited | -0.080 |
eling | -0.079 |
irl | -0.078 |
eking | -0.076 |
uch | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | +4.32e+00 |
Pos logit contributions | |
ited | +0.093 |
eling | +0.091 |
irl | +0.091 |
eking | +0.088 |
uch | +0.088 |
Neg logit contributions | |
He | -0.042 |
They | -0.036 |
automobile | -0.032 |
sidewalk | -0.031 |
The | -0.031 |
Token | " |
Token ID | 366 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
He | +0.079 |
They | +0.070 |
automobile | +0.062 |
The | +0.062 |
sidewalk | +0.061 |
Neg logit contributions | |
ited | -0.134 |
eling | -0.132 |
irl | -0.131 |
eking | -0.127 |
uch | -0.127 |
Token | I |
Token ID | 40 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
eling | +0.012 |
ited | +0.012 |
irl | +0.012 |
eking | +0.011 |
inations | +0.011 |
Neg logit contributions | |
He | -0.008 |
They | -0.007 |
The | -0.006 |
She | -0.006 |
sidewalk | -0.006 |
Token | lived |
Token ID | 5615 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ited | +0.024 |
eling | +0.023 |
irl | +0.023 |
eking | +0.023 |
inations | +0.022 |
Neg logit contributions | |
He | -0.018 |
They | -0.016 |
The | -0.015 |
She | -0.014 |
Tim | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
ited | +0.058 |
eling | +0.057 |
irl | +0.055 |
eking | +0.054 |
uch | +0.054 |
Neg logit contributions | |
He | -0.026 |
They | -0.023 |
The | -0.021 |
automobile | -0.019 |
She | -0.018 |
Token | rich |
Token ID | 5527 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
ited | +0.025 |
irl | +0.025 |
eling | +0.025 |
uch | +0.024 |
eking | +0.024 |
Neg logit contributions | |
He | -0.024 |
They | -0.022 |
The | -0.020 |
automobile | -0.020 |
sidewalk | -0.020 |
Token | man |
Token ID | 582 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
ited | +0.010 |
irl | +0.010 |
years | +0.009 |
sie | +0.009 |
kins | +0.009 |
Neg logit contributions | |
He | -0.008 |
They | -0.007 |
tr | -0.006 |
Then | -0.006 |
der | -0.006 |
Token | named |
Token ID | 3706 |
Feature activation | -5.03e+00 |
Pos logit contributions | |
ited | +0.072 |
eling | +0.070 |
eking | +0.069 |
if | +0.069 |
re | +0.069 |
Neg logit contributions | |
He | -0.043 |
They | -0.039 |
sidewalk | -0.036 |
automobile | -0.035 |
pavement | -0.035 |
Token | Mr |
Token ID | 1770 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
eling | +0.150 |
ited | +0.148 |
irl | +0.146 |
re | +0.145 |
eking | +0.143 |
Neg logit contributions | |
He | -0.075 |
They | -0.068 |
automobile | -0.068 |
tr | -0.067 |
der | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
eling | +0.470 |
ited | +0.469 |
irl | +0.458 |
re | +0.457 |
if | +0.456 |
Neg logit contributions | |
He | -0.207 |
They | -0.173 |
automobile | -0.170 |
sidewalk | -0.170 |
tr | -0.156 |
Token | Stein |
Token ID | 15215 |
Feature activation | -4.59e+00 |
Pos logit contributions | |
He | +0.063 |
They | +0.059 |
sidewalk | +0.058 |
automobile | +0.057 |
tr | +0.056 |
Neg logit contributions | |
ited | -0.086 |
eling | -0.085 |
if | -0.085 |
irl | -0.085 |
re | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
ited | +0.151 |
eling | +0.149 |
if | +0.147 |
re | +0.147 |
irl | +0.146 |
Neg logit contributions | |
He | -0.066 |
They | -0.058 |
sidewalk | -0.053 |
automobile | -0.053 |
tr | -0.049 |
Token | He |
Token ID | 679 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
sidewalk | +0.040 |
tr | +0.039 |
automobile | +0.037 |
pavement | +0.035 |
He | +0.035 |
Neg logit contributions | |
if | -0.199 |
ited | -0.196 |
irl | -0.196 |
re | -0.195 |
eling | -0.194 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
He | +0.045 |
They | +0.042 |
automobile | +0.040 |
sidewalk | +0.039 |
pavement | +0.038 |
Neg logit contributions | |
ited | -0.042 |
eling | -0.042 |
irl | -0.042 |
if | -0.040 |
uch | -0.040 |
Token | neck |
Token ID | 7393 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
He | +0.083 |
They | +0.076 |
automobile | +0.069 |
The | +0.068 |
sidewalk | +0.068 |
Neg logit contributions | |
ited | -0.092 |
irl | -0.090 |
eling | -0.090 |
eking | -0.088 |
uch | -0.087 |
Token | that |
Token ID | 326 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
ited | +0.073 |
eling | +0.071 |
irl | +0.071 |
eking | +0.070 |
if | +0.070 |
Neg logit contributions | |
He | -0.030 |
They | -0.026 |
automobile | -0.023 |
The | -0.022 |
sidewalk | -0.022 |
Token | read |
Token ID | 1100 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
He | +0.079 |
They | +0.072 |
The | +0.066 |
sidewalk | +0.064 |
automobile | +0.064 |
Neg logit contributions | |
ited | -0.096 |
eling | -0.094 |
irl | -0.092 |
eking | -0.090 |
if | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | -9.07e-01 |
Pos logit contributions | |
He | +0.071 |
They | +0.062 |
automobile | +0.056 |
sidewalk | +0.055 |
The | +0.055 |
Neg logit contributions | |
ited | -0.136 |
eling | -0.133 |
irl | -0.133 |
eking | -0.129 |
if | -0.128 |
Token | " |
Token ID | 366 |
Feature activation | +7.00e-01 |
Pos logit contributions | |
ited | +0.028 |
eling | +0.027 |
irl | +0.027 |
eking | +0.026 |
if | +0.026 |
Neg logit contributions | |
He | -0.017 |
They | -0.015 |
automobile | -0.013 |
The | -0.013 |
sidewalk | -0.013 |
Token | Fore |
Token ID | 16351 |
Feature activation | -1.42e+01 |
Pos logit contributions | |
He | +0.016 |
They | +0.015 |
automobile | +0.014 |
The | +0.014 |
sidewalk | +0.013 |
Neg logit contributions | |
eling | -0.018 |
irl | -0.018 |
ited | -0.018 |
eking | -0.018 |
uch | -0.017 |
Token | ver |
Token ID | 332 |
Feature activation | -5.07e+00 |
Pos logit contributions | |
ited | +0.291 |
irl | +0.282 |
eling | +0.275 |
rim | +0.262 |
if | +0.252 |
Neg logit contributions | |
He | -0.401 |
They | -0.373 |
The | -0.368 |
tr | -0.355 |
automobile | -0.347 |
Token | Friends |
Token ID | 14213 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
eling | +0.159 |
irl | +0.157 |
ited | +0.157 |
uch | +0.150 |
eking | +0.150 |
Neg logit contributions | |
He | -0.096 |
They | -0.091 |
The | -0.081 |
automobile | -0.080 |
Tim | -0.075 |
Token | ". |
Token ID | 1911 |
Feature activation | +5.13e+00 |
Pos logit contributions | |
ited | +0.101 |
eling | +0.098 |
eking | +0.097 |
irl | +0.096 |
if | +0.093 |
Neg logit contributions | |
He | -0.073 |
They | -0.070 |
The | -0.063 |
It | -0.059 |
automobile | -0.058 |
Token | |
Token ID | 220 |
Feature activation | -8.27e+00 |
Pos logit contributions | |
He | +0.083 |
automobile | +0.071 |
They | +0.071 |
sidewalk | +0.069 |
tr | +0.068 |
Neg logit contributions | |
ited | -0.159 |
eling | -0.155 |
irl | -0.154 |
re | -0.153 |
if | -0.150 |
Token |
|
Token ID | 198 |
Feature activation | -5.61e+00 |
Pos logit contributions | |
ited | +0.259 |
eling | +0.255 |
irl | +0.252 |
eking | +0.245 |
uch | +0.244 |
Neg logit contributions | |
He | -0.149 |
They | -0.132 |
automobile | -0.117 |
The | -0.116 |
sidewalk | -0.115 |
Token | not |
Token ID | 407 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
ited | +0.068 |
eling | +0.064 |
irl | +0.063 |
re | +0.062 |
if | +0.061 |
Neg logit contributions | |
He | -0.054 |
They | -0.049 |
The | -0.045 |
Tim | -0.044 |
automobile | -0.044 |
Token | a |
Token ID | 257 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
eling | +0.149 |
ited | +0.149 |
irl | +0.147 |
eking | +0.143 |
uch | +0.142 |
Neg logit contributions | |
He | -0.118 |
They | -0.107 |
automobile | -0.098 |
sidewalk | -0.098 |
The | -0.097 |
Token | monster |
Token ID | 9234 |
Feature activation | -8.60e-01 |
Pos logit contributions | |
He | +0.002 |
They | +0.002 |
The | +0.002 |
automobile | +0.002 |
sidewalk | +0.002 |
Neg logit contributions | |
ited | -0.003 |
irl | -0.003 |
eling | -0.003 |
if | -0.003 |
uch | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
ited | +0.025 |
eling | +0.024 |
irl | +0.024 |
uch | +0.023 |
eking | +0.023 |
Neg logit contributions | |
He | -0.017 |
They | -0.016 |
automobile | -0.014 |
sidewalk | -0.014 |
The | -0.014 |
Token | my |
Token ID | 616 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
ited | +0.104 |
eling | +0.102 |
irl | +0.102 |
eking | +0.099 |
uch | +0.098 |
Neg logit contributions | |
He | -0.049 |
They | -0.043 |
automobile | -0.038 |
sidewalk | -0.037 |
The | -0.037 |
Token | de |
Token ID | 390 |
Feature activation | -1.41e+01 |
Pos logit contributions | |
irl | +0.029 |
ited | +0.029 |
eling | +0.029 |
uch | +0.028 |
if | +0.028 |
Neg logit contributions | |
He | -0.027 |
They | -0.025 |
The | -0.023 |
automobile | -0.023 |
sidewalk | -0.022 |
Token | ars |
Token ID | 945 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
ited | +0.212 |
irl | +0.203 |
eling | +0.201 |
if | +0.191 |
re | +0.191 |
Neg logit contributions | |
He | -0.463 |
They | -0.432 |
automobile | -0.412 |
The | -0.411 |
sidewalk | -0.410 |
Token | . |
Token ID | 13 |
Feature activation | +4.73e+00 |
Pos logit contributions | |
ited | +0.022 |
eling | +0.022 |
irl | +0.021 |
eking | +0.021 |
uch | +0.021 |
Neg logit contributions | |
He | -0.015 |
They | -0.013 |
automobile | -0.012 |
The | -0.012 |
sidewalk | -0.012 |
Token | That |
Token ID | 1320 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
sidewalk | +0.085 |
pavement | +0.080 |
automobile | +0.078 |
curb | +0.077 |
arrow | +0.074 |
Neg logit contributions | |
re | -0.115 |
uch | -0.113 |
ink | -0.111 |
ins | -0.110 |
ac | -0.108 |
Token | is |
Token ID | 318 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
He | +0.049 |
They | +0.047 |
sidewalk | +0.043 |
The | +0.043 |
automobile | +0.043 |
Neg logit contributions | |
irl | -0.041 |
uch | -0.041 |
eling | -0.041 |
ited | -0.041 |
if | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ited | +0.036 |
eling | +0.034 |
irl | +0.034 |
re | +0.033 |
if | +0.033 |
Neg logit contributions | |
He | -0.027 |
They | -0.025 |
The | -0.023 |
automobile | -0.022 |
sidewalk | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ited | +0.083 |
eling | +0.082 |
irl | +0.081 |
eking | +0.078 |
uch | +0.078 |
Neg logit contributions | |
He | -0.057 |
They | -0.051 |
automobile | -0.046 |
The | -0.046 |
sidewalk | -0.046 |
Token | lucky |
Token ID | 9670 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
ited | +0.037 |
eling | +0.036 |
irl | +0.036 |
uch | +0.034 |
eking | +0.034 |
Neg logit contributions | |
He | -0.037 |
They | -0.034 |
automobile | -0.031 |
The | -0.031 |
sidewalk | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +5.18e+00 |
Pos logit contributions | |
ited | +0.024 |
eling | +0.024 |
irl | +0.024 |
eking | +0.023 |
if | +0.023 |
Neg logit contributions | |
He | -0.011 |
They | -0.009 |
automobile | -0.008 |
sidewalk | -0.008 |
The | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -7.55e+00 |
Pos logit contributions | |
He | +0.087 |
They | +0.078 |
The | +0.068 |
It | +0.062 |
automobile | +0.062 |
Neg logit contributions | |
ited | -0.180 |
eling | -0.176 |
irl | -0.172 |
eking | -0.170 |
uch | -0.166 |
Token |
|
Token ID | 198 |
Feature activation | -6.96e+00 |
Pos logit contributions | |
irl | +0.182 |
eling | +0.177 |
ited | +0.177 |
if | +0.173 |
uch | +0.171 |
Neg logit contributions | |
He | -0.180 |
They | -0.163 |
automobile | -0.155 |
sidewalk | -0.154 |
der | -0.150 |
Token | Fore |
Token ID | 16351 |
Feature activation | -1.41e+01 |
Pos logit contributions | |
ited | +0.179 |
irl | +0.176 |
eling | +0.176 |
if | +0.169 |
uch | +0.168 |
Neg logit contributions | |
He | -0.162 |
They | -0.147 |
automobile | -0.135 |
sidewalk | -0.135 |
The | -0.134 |
Token | shadow |
Token ID | 19106 |
Feature activation | -3.30e+00 |
Pos logit contributions | |
ited | +0.284 |
eling | +0.270 |
irl | +0.269 |
rim | +0.248 |
if | +0.248 |
Neg logit contributions | |
He | -0.392 |
They | -0.373 |
The | -0.366 |
tr | -0.361 |
automobile | -0.358 |
Token | ing |
Token ID | 278 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
ited | +0.097 |
eling | +0.095 |
irl | +0.095 |
inations | +0.090 |
eking | +0.089 |
Neg logit contributions | |
He | -0.069 |
They | -0.064 |
The | -0.059 |
automobile | -0.054 |
She | -0.053 |
Token | or |
Token ID | 393 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
He | +0.049 |
They | +0.044 |
The | +0.039 |
automobile | +0.036 |
sidewalk | +0.036 |
Neg logit contributions | |
ited | -0.090 |
eling | -0.087 |
irl | -0.086 |
uch | -0.084 |
eking | -0.083 |
Token | setup |
Token ID | 9058 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
irl | +0.060 |
ited | +0.059 |
uch | +0.058 |
rim | +0.056 |
inations | +0.055 |
Neg logit contributions | |
He | -0.065 |
They | -0.062 |
The | -0.057 |
sidewalk | -0.057 |
der | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
ited | +0.059 |
eling | +0.058 |
irl | +0.057 |
uch | +0.055 |
eking | +0.055 |
Neg logit contributions | |
He | -0.022 |
They | -0.019 |
The | -0.016 |
sidewalk | -0.015 |
automobile | -0.015 |
Token | cakes |
Token ID | 30849 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
ited | +0.107 |
eling | +0.104 |
irl | +0.103 |
uch | +0.100 |
eking | +0.099 |
Neg logit contributions | |
He | -0.081 |
They | -0.074 |
The | -0.066 |
sidewalk | -0.066 |
automobile | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ited | +0.073 |
eling | +0.072 |
irl | +0.071 |
eking | +0.069 |
uch | +0.069 |
Neg logit contributions | |
He | -0.042 |
They | -0.038 |
automobile | -0.033 |
The | -0.033 |
sidewalk | -0.033 |
Token | cookies |
Token ID | 14746 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ited | +0.037 |
eling | +0.037 |
irl | +0.036 |
eking | +0.035 |
uch | +0.035 |
Neg logit contributions | |
He | -0.036 |
They | -0.033 |
automobile | -0.030 |
The | -0.030 |
sidewalk | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
ited | +0.053 |
eling | +0.052 |
irl | +0.052 |
eking | +0.050 |
uch | +0.050 |
Neg logit contributions | |
He | -0.028 |
They | -0.024 |
automobile | -0.022 |
The | -0.021 |
sidewalk | -0.021 |
Token | ch |
Token ID | 442 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
ited | +0.023 |
eling | +0.022 |
irl | +0.022 |
eking | +0.021 |
if | +0.021 |
Neg logit contributions | |
He | -0.022 |
They | -0.020 |
automobile | -0.018 |
The | -0.018 |
sidewalk | -0.018 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.40e+01 |
Pos logit contributions | |
uch | +0.076 |
ited | +0.072 |
if | +0.071 |
irl | +0.070 |
eling | +0.068 |
Neg logit contributions | |
They | -0.137 |
He | -0.136 |
The | -0.123 |
der | -0.119 |
It | -0.118 |
Token | ates |
Token ID | 689 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
ited | +0.280 |
eling | +0.257 |
irl | +0.250 |
inations | +0.225 |
sie | +0.209 |
Neg logit contributions | |
They | -0.391 |
He | -0.390 |
tr | -0.387 |
Then | -0.358 |
der | -0.357 |
Token | and |
Token ID | 290 |
Feature activation | -9.43e-01 |
Pos logit contributions | |
ited | +0.074 |
eling | +0.073 |
irl | +0.073 |
eking | +0.071 |
uch | +0.070 |
Neg logit contributions | |
He | -0.030 |
They | -0.025 |
automobile | -0.022 |
The | -0.021 |
sidewalk | -0.021 |
Token | ice |
Token ID | 4771 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
eling | +0.023 |
ited | +0.023 |
irl | +0.023 |
eking | +0.022 |
inations | +0.022 |
Neg logit contributions | |
He | -0.022 |
They | -0.020 |
automobile | -0.019 |
The | -0.018 |
sidewalk | -0.018 |
Token | cream |
Token ID | 8566 |
Feature activation | +9.09e-03 |
Pos logit contributions | |
ited | +0.108 |
eling | +0.105 |
irl | +0.104 |
uch | +0.099 |
eking | +0.099 |
Neg logit contributions | |
He | -0.117 |
They | -0.109 |
automobile | -0.100 |
sidewalk | -0.100 |
The | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | +5.07e+00 |
Pos logit contributions | |
He | +0.000 |
They | +0.000 |
automobile | +0.000 |
The | +0.000 |
sidewalk | +0.000 |
Neg logit contributions | |
ited | -0.000 |
re | -0.000 |
kins | -0.000 |
eling | -0.000 |
eking | -0.000 |
Token | wearing |
Token ID | 5762 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
ited | +0.063 |
uch | +0.058 |
irl | +0.058 |
inations | +0.057 |
eling | +0.057 |
Neg logit contributions | |
He | -0.045 |
The | -0.039 |
They | -0.038 |
She | -0.037 |
automobile | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
He | +0.036 |
They | +0.030 |
The | +0.028 |
automobile | +0.026 |
It | +0.025 |
Neg logit contributions | |
ited | -0.095 |
eling | -0.093 |
irl | -0.091 |
uch | -0.091 |
re | -0.090 |
Token | very |
Token ID | 845 |
Feature activation | -5.17e+00 |
Pos logit contributions | |
He | +0.048 |
automobile | +0.043 |
tr | +0.041 |
arrow | +0.040 |
They | +0.040 |
Neg logit contributions | |
eling | -0.071 |
ited | -0.068 |
ins | -0.065 |
re | -0.065 |
eking | -0.063 |
Token | fine |
Token ID | 3734 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
eling | +0.150 |
ited | +0.145 |
re | +0.145 |
ins | +0.141 |
if | +0.140 |
Neg logit contributions | |
He | -0.110 |
sidewalk | -0.090 |
They | -0.089 |
automobile | -0.087 |
arrow | -0.086 |
Token | dress |
Token ID | 6576 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
He | +0.071 |
automobile | +0.064 |
They | +0.061 |
tr | +0.059 |
sidewalk | +0.058 |
Neg logit contributions | |
eling | -0.104 |
ited | -0.102 |
ins | -0.098 |
irl | -0.096 |
eking | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | +6.14e+00 |
Pos logit contributions | |
He | +0.015 |
They | +0.013 |
The | +0.011 |
sidewalk | +0.011 |
automobile | +0.011 |
Neg logit contributions | |
ited | -0.028 |
eking | -0.028 |
if | -0.027 |
re | -0.027 |
irl | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -5.59e+00 |
Pos logit contributions | |
He | +0.079 |
automobile | +0.078 |
tr | +0.074 |
sidewalk | +0.073 |
pavement | +0.067 |
Neg logit contributions | |
ited | -0.194 |
re | -0.190 |
irl | -0.189 |
eling | -0.189 |
if | -0.188 |
Token |
|
Token ID | 198 |
Feature activation | -4.92e+00 |
Pos logit contributions | |
ited | +0.188 |
eling | +0.186 |
irl | +0.180 |
eking | +0.179 |
uch | +0.178 |
Neg logit contributions | |
He | -0.095 |
They | -0.083 |
The | -0.074 |
sidewalk | -0.068 |
She | -0.068 |
Token | The |
Token ID | 464 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
ited | +0.133 |
eling | +0.131 |
irl | +0.130 |
eking | +0.126 |
uch | +0.125 |
Neg logit contributions | |
He | -0.109 |
They | -0.099 |
automobile | -0.091 |
The | -0.090 |
sidewalk | -0.090 |
Token | fairy |
Token ID | 25607 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
He | +0.040 |
They | +0.035 |
automobile | +0.034 |
sidewalk | +0.034 |
The | +0.033 |
Neg logit contributions | |
ited | -0.052 |
eling | -0.051 |
irl | -0.050 |
re | -0.049 |
uch | -0.049 |
Token | said |
Token ID | 531 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
He | +0.045 |
They | +0.039 |
The | +0.038 |
automobile | +0.036 |
arrow | +0.035 |
Neg logit contributions | |
eling | -0.056 |
ited | -0.056 |
irl | -0.056 |
uch | -0.055 |
eking | -0.053 |
Token | have |
Token ID | 423 |
Feature activation | +9.58e-01 |
Pos logit contributions | |
He | +0.016 |
They | +0.015 |
automobile | +0.014 |
sidewalk | +0.014 |
The | +0.013 |
Neg logit contributions | |
ited | -0.018 |
eling | -0.018 |
irl | -0.018 |
eking | -0.017 |
uch | -0.017 |
Token | seen |
Token ID | 1775 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
He | +0.017 |
They | +0.015 |
The | +0.015 |
Tim | +0.013 |
It | +0.013 |
Neg logit contributions | |
ited | -0.032 |
eling | -0.031 |
eking | -0.030 |
irl | -0.029 |
uch | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +7.92e-01 |
Pos logit contributions | |
He | +0.034 |
They | +0.030 |
The | +0.028 |
automobile | +0.027 |
sidewalk | +0.026 |
Neg logit contributions | |
ited | -0.055 |
eling | -0.053 |
irl | -0.053 |
uch | -0.052 |
eking | -0.051 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
He | +0.015 |
They | +0.013 |
automobile | +0.013 |
sidewalk | +0.013 |
The | +0.013 |
Neg logit contributions | |
ited | -0.024 |
eling | -0.024 |
irl | -0.023 |
eking | -0.022 |
re | -0.022 |
Token | landscape |
Token ID | 10747 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
He | +0.002 |
They | +0.002 |
automobile | +0.001 |
The | +0.001 |
sidewalk | +0.001 |
Neg logit contributions | |
ited | -0.003 |
eling | -0.003 |
irl | -0.003 |
eking | -0.003 |
uch | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +5.72e+00 |
Pos logit contributions | |
ited | +0.021 |
eling | +0.020 |
irl | +0.020 |
eking | +0.020 |
if | +0.020 |
Neg logit contributions | |
He | -0.012 |
They | -0.011 |
automobile | -0.010 |
sidewalk | -0.010 |
The | -0.009 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.20e+00 |
Pos logit contributions | |
automobile | +0.119 |
sidewalk | +0.118 |
cop | +0.115 |
pavement | +0.114 |
curb | +0.114 |
Neg logit contributions | |
re | -0.129 |
irl | -0.127 |
if | -0.127 |
ited | -0.127 |
eling | -0.121 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
ited | +0.299 |
eling | +0.295 |
irl | +0.289 |
eking | +0.284 |
uch | +0.280 |
Neg logit contributions | |
He | -0.162 |
They | -0.143 |
The | -0.127 |
automobile | -0.121 |
sidewalk | -0.120 |
Token | upon |
Token ID | 2402 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
ited | +0.200 |
eling | +0.199 |
eking | +0.186 |
irl | +0.181 |
uch | +0.181 |
Neg logit contributions | |
He | -0.343 |
They | -0.318 |
The | -0.308 |
Tim | -0.299 |
She | -0.295 |
Token | a |
Token ID | 257 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
ited | +0.102 |
irl | +0.101 |
eling | +0.101 |
if | +0.097 |
uch | +0.097 |
Neg logit contributions | |
He | -0.055 |
They | -0.049 |
automobile | -0.045 |
sidewalk | -0.045 |
The | -0.042 |
Token | time |
Token ID | 640 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
He | +0.028 |
They | +0.028 |
The | +0.026 |
But | +0.026 |
And | +0.025 |
Neg logit contributions | |
eling | -0.016 |
ins | -0.015 |
eking | -0.015 |
ubb | -0.015 |
kins | -0.015 |
Token | juice |
Token ID | 13135 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
ited | +0.028 |
irl | +0.028 |
eling | +0.027 |
uch | +0.027 |
if | +0.027 |
Neg logit contributions | |
He | -0.021 |
They | -0.019 |
automobile | -0.018 |
sidewalk | -0.017 |
The | -0.017 |
Token | recipe |
Token ID | 8364 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
He | +0.010 |
They | +0.009 |
The | +0.008 |
automobile | +0.008 |
sidewalk | +0.008 |
Neg logit contributions | |
ited | -0.013 |
uch | -0.012 |
irl | -0.012 |
re | -0.012 |
eling | -0.012 |
Token | ," |
Token ID | 553 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
ited | +0.029 |
eling | +0.028 |
irl | +0.028 |
eking | +0.027 |
if | +0.026 |
Neg logit contributions | |
He | -0.013 |
They | -0.012 |
automobile | -0.010 |
sidewalk | -0.010 |
The | -0.010 |
Token | said |
Token ID | 531 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
ited | +0.020 |
irl | +0.020 |
eling | +0.019 |
kins | +0.017 |
re | +0.017 |
Neg logit contributions | |
tr | -0.030 |
curb | -0.029 |
He | -0.029 |
der | -0.028 |
sidewalk | -0.028 |
Token | Max |
Token ID | 5436 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
ited | +0.011 |
irl | +0.011 |
eling | +0.010 |
uch | +0.010 |
kins | +0.010 |
Neg logit contributions | |
He | -0.007 |
sidewalk | -0.007 |
They | -0.007 |
tr | -0.007 |
curb | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +5.13e+00 |
Pos logit contributions | |
ited | +0.061 |
re | +0.058 |
eling | +0.058 |
kins | +0.057 |
eking | +0.057 |
Neg logit contributions | |
sidewalk | -0.035 |
automobile | -0.035 |
He | -0.034 |
They | -0.033 |
pavement | -0.032 |
Token | Polly |
Token ID | 36898 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
automobile | +0.124 |
tr | +0.121 |
roof | +0.118 |
pavement | +0.117 |
sidewalk | +0.117 |
Neg logit contributions | |
re | -0.098 |
irl | -0.094 |
ited | -0.093 |
kins | -0.089 |
if | -0.089 |
Token | took |
Token ID | 1718 |
Feature activation | +8.64e-02 |
Pos logit contributions | |
ited | +0.046 |
eling | +0.045 |
eking | +0.045 |
irl | +0.044 |
uch | +0.044 |
Neg logit contributions | |
He | -0.040 |
They | -0.037 |
The | -0.034 |
automobile | -0.031 |
She | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
He | +0.002 |
They | +0.001 |
automobile | +0.001 |
sidewalk | +0.001 |
tr | +0.001 |
Neg logit contributions | |
ited | -0.003 |
irl | -0.003 |
eling | -0.003 |
uch | -0.002 |
eking | -0.002 |
Token | sip |
Token ID | 31145 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
He | +0.022 |
They | +0.020 |
The | +0.019 |
automobile | +0.018 |
It | +0.018 |
Neg logit contributions | |
eling | -0.026 |
eking | -0.025 |
irl | -0.024 |
ited | -0.024 |
if | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -8.26e-01 |
Pos logit contributions | |
He | +0.029 |
They | +0.026 |
automobile | +0.025 |
sidewalk | +0.024 |
tr | +0.023 |
Neg logit contributions | |
ited | -0.053 |
eling | -0.051 |
irl | -0.051 |
if | -0.049 |
re | -0.049 |
Token | their |
Token ID | 511 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
He | +0.029 |
tr | +0.027 |
automobile | +0.026 |
They | +0.026 |
der | +0.026 |
Neg logit contributions | |
eling | -0.033 |
irl | -0.032 |
ited | -0.031 |
years | -0.030 |
time | -0.029 |
Token | house |
Token ID | 2156 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
He | +0.038 |
They | +0.034 |
The | +0.030 |
automobile | +0.029 |
sidewalk | +0.029 |
Neg logit contributions | |
ited | -0.054 |
eling | -0.053 |
irl | -0.053 |
eking | -0.051 |
if | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
He | +0.018 |
They | +0.016 |
automobile | +0.014 |
sidewalk | +0.014 |
The | +0.014 |
Neg logit contributions | |
ited | -0.036 |
eling | -0.035 |
irl | -0.035 |
eking | -0.034 |
if | -0.034 |
Token | said |
Token ID | 531 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
He | +0.012 |
They | +0.011 |
automobile | +0.011 |
sidewalk | +0.010 |
The | +0.010 |
Neg logit contributions | |
ited | -0.011 |
eling | -0.011 |
irl | -0.010 |
eking | -0.010 |
kins | -0.010 |
Token | goodbye |
Token ID | 24829 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
irl | +0.012 |
eling | +0.012 |
ited | +0.012 |
eking | +0.011 |
re | +0.011 |
Neg logit contributions | |
He | -0.010 |
They | -0.009 |
automobile | -0.009 |
sidewalk | -0.009 |
tr | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +5.18e+00 |
Pos logit contributions | |
years | +0.061 |
if | +0.060 |
re | +0.059 |
eking | +0.059 |
kins | +0.058 |
Neg logit contributions | |
automobile | -0.050 |
He | -0.048 |
sidewalk | -0.046 |
pavement | -0.046 |
curb | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -7.00e+00 |
Pos logit contributions | |
automobile | +0.101 |
tr | +0.097 |
pavement | +0.094 |
sidewalk | +0.094 |
arrow | +0.090 |
Neg logit contributions | |
irl | -0.119 |
re | -0.117 |
if | -0.117 |
ins | -0.115 |
years | -0.114 |
Token |
|
Token ID | 198 |
Feature activation | -6.33e+00 |
Pos logit contributions | |
irl | +0.189 |
ited | +0.184 |
eling | +0.183 |
if | +0.180 |
re | +0.177 |
Neg logit contributions | |
He | -0.139 |
automobile | -0.126 |
They | -0.124 |
tr | -0.121 |
sidewalk | -0.121 |
Token | Tom |
Token ID | 13787 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
ited | +0.148 |
if | +0.144 |
irl | +0.143 |
eling | +0.143 |
ins | +0.137 |
Neg logit contributions | |
He | -0.152 |
They | -0.140 |
automobile | -0.139 |
sidewalk | -0.134 |
der | -0.134 |
Token | and |
Token ID | 290 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
eling | +0.089 |
irl | +0.089 |
ited | +0.088 |
eking | +0.086 |
inations | +0.084 |
Neg logit contributions | |
He | -0.036 |
They | -0.030 |
The | -0.027 |
automobile | -0.025 |
sidewalk | -0.023 |
Token | Anna |
Token ID | 11735 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
ited | +0.019 |
eling | +0.018 |
irl | +0.018 |
sie | +0.017 |
kins | +0.017 |
Neg logit contributions | |
He | -0.014 |
sidewalk | -0.013 |
automobile | -0.013 |
They | -0.012 |
tr | -0.012 |
Token | enough |
Token ID | 1576 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
He | +0.015 |
They | +0.014 |
automobile | +0.012 |
sidewalk | +0.012 |
The | +0.012 |
Neg logit contributions | |
ited | -0.026 |
eling | -0.025 |
irl | -0.025 |
eking | -0.024 |
if | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
irl | +0.019 |
ited | +0.018 |
eling | +0.018 |
uch | +0.018 |
ins | +0.018 |
Neg logit contributions | |
He | -0.006 |
They | -0.004 |
The | -0.004 |
sidewalk | -0.003 |
She | -0.003 |
Token | keep |
Token ID | 1394 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
He | +0.015 |
They | +0.013 |
automobile | +0.012 |
The | +0.012 |
sidewalk | +0.012 |
Neg logit contributions | |
ited | -0.022 |
eling | -0.021 |
irl | -0.021 |
eking | -0.021 |
uch | -0.021 |
Token | going |
Token ID | 1016 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
ited | +0.042 |
uch | +0.039 |
rim | +0.038 |
inations | +0.038 |
ins | +0.038 |
Neg logit contributions | |
He | -0.019 |
The | -0.016 |
They | -0.015 |
Tim | -0.015 |
Then | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
He | +0.000 |
They | +0.000 |
tr | +0.000 |
automobile | +0.000 |
der | +0.000 |
Neg logit contributions | |
eking | -0.001 |
eling | -0.001 |
ited | -0.001 |
irl | -0.001 |
re | -0.001 |
Token | complete |
Token ID | 1844 |
Feature activation | +4.50e+00 |
Pos logit contributions | |
He | +0.017 |
They | +0.016 |
automobile | +0.014 |
The | +0.014 |
sidewalk | +0.014 |
Neg logit contributions | |
ited | -0.024 |
eling | -0.024 |
irl | -0.023 |
eking | -0.023 |
uch | -0.022 |
Token | her |
Token ID | 607 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
He | +0.079 |
They | +0.070 |
The | +0.063 |
sidewalk | +0.060 |
automobile | +0.060 |
Neg logit contributions | |
ited | -0.147 |
eling | -0.140 |
uch | -0.140 |
irl | -0.138 |
re | -0.135 |
Token | adventure |
Token ID | 8855 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
He | +0.063 |
They | +0.056 |
The | +0.054 |
sidewalk | +0.054 |
automobile | +0.054 |
Neg logit contributions | |
ited | -0.068 |
ins | -0.066 |
eling | -0.064 |
rim | -0.064 |
eking | -0.061 |
Token | ! |
Token ID | 0 |
Feature activation | +6.18e+00 |
Pos logit contributions | |
He | +0.008 |
They | +0.007 |
automobile | +0.006 |
The | +0.006 |
sidewalk | +0.006 |
Neg logit contributions | |
ited | -0.016 |
eling | -0.016 |
irl | -0.016 |
eking | -0.015 |
if | -0.015 |
Token | |
Token ID | 220 |
Feature activation | -8.03e+00 |
Pos logit contributions | |
He | +0.130 |
automobile | +0.118 |
They | +0.116 |
sidewalk | +0.115 |
tr | +0.111 |
Neg logit contributions | |
ited | -0.160 |
irl | -0.156 |
eling | -0.155 |
re | -0.155 |
if | -0.152 |
Token |
|
Token ID | 198 |
Feature activation | -5.89e+00 |
Pos logit contributions | |
irl | +0.226 |
eling | +0.223 |
ited | +0.221 |
re | +0.220 |
if | +0.218 |
Neg logit contributions | |
He | -0.153 |
automobile | -0.135 |
They | -0.135 |
sidewalk | -0.130 |
cop | -0.123 |
Token | flew |
Token ID | 13112 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
He | +0.019 |
They | +0.017 |
The | +0.016 |
automobile | +0.016 |
sidewalk | +0.015 |
Neg logit contributions | |
ited | -0.021 |
irl | -0.021 |
eling | -0.021 |
uch | -0.020 |
eking | -0.020 |
Token | over |
Token ID | 625 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
He | +0.066 |
They | +0.060 |
automobile | +0.054 |
sidewalk | +0.054 |
The | +0.053 |
Neg logit contributions | |
ited | -0.103 |
eling | -0.101 |
irl | -0.101 |
eking | -0.098 |
uch | -0.097 |
Token | and |
Token ID | 290 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
He | +0.026 |
They | +0.023 |
automobile | +0.020 |
sidewalk | +0.019 |
The | +0.019 |
Neg logit contributions | |
ited | -0.056 |
eling | -0.055 |
irl | -0.055 |
eking | -0.053 |
uch | -0.053 |
Token | said |
Token ID | 531 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
He | +0.054 |
They | +0.049 |
automobile | +0.046 |
sidewalk | +0.046 |
The | +0.045 |
Neg logit contributions | |
ited | -0.050 |
eling | -0.049 |
irl | -0.048 |
eking | -0.046 |
uch | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
He | +0.032 |
sidewalk | +0.031 |
pavement | +0.030 |
They | +0.030 |
automobile | +0.030 |
Neg logit contributions | |
irl | -0.032 |
re | -0.031 |
ited | -0.031 |
eling | -0.031 |
kins | -0.031 |
Token | " |
Token ID | 366 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
if | +0.022 |
re | +0.021 |
eking | +0.021 |
irl | +0.021 |
eling | +0.021 |
Neg logit contributions | |
automobile | -0.007 |
He | -0.007 |
sidewalk | -0.007 |
pavement | -0.006 |
They | -0.006 |
Token | Let |
Token ID | 5756 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
He | +0.011 |
They | +0.010 |
sidewalk | +0.009 |
automobile | +0.009 |
The | +0.009 |
Neg logit contributions | |
ited | -0.013 |
eling | -0.013 |
irl | -0.013 |
if | -0.013 |
uch | -0.012 |
Token | me |
Token ID | 502 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
He | +0.055 |
They | +0.049 |
The | +0.045 |
automobile | +0.043 |
Tim | +0.043 |
Neg logit contributions | |
ited | -0.086 |
eling | -0.084 |
irl | -0.084 |
inations | -0.084 |
kins | -0.083 |
Token | play |
Token ID | 711 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
ited | +0.070 |
irl | +0.069 |
eling | +0.069 |
uch | +0.067 |
eking | +0.066 |
Neg logit contributions | |
He | -0.047 |
They | -0.042 |
automobile | -0.038 |
sidewalk | -0.038 |
The | -0.037 |
Token | too |
Token ID | 1165 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
He | +0.043 |
They | +0.038 |
sidewalk | +0.036 |
automobile | +0.036 |
tr | +0.035 |
Neg logit contributions | |
irl | -0.053 |
eling | -0.050 |
ited | -0.049 |
uch | -0.048 |
inations | -0.048 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
irl | +0.097 |
ited | +0.095 |
re | +0.091 |
eling | +0.091 |
if | +0.090 |
Neg logit contributions | |
He | -0.110 |
They | -0.100 |
tr | -0.098 |
automobile | -0.098 |
sidewalk | -0.098 |
Token | One |
Token ID | 1881 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
der | +0.085 |
automobile | +0.079 |
pavement | +0.077 |
sidewalk | +0.077 |
He | +0.076 |
Neg logit contributions | |
irl | -0.165 |
re | -0.159 |
ited | -0.156 |
uch | -0.155 |
eling | -0.153 |
Token | day |
Token ID | 1110 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
ited | +0.017 |
irl | +0.016 |
uch | +0.015 |
eling | +0.015 |
if | +0.015 |
Neg logit contributions | |
He | -0.020 |
They | -0.019 |
sidewalk | -0.018 |
der | -0.018 |
automobile | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
ited | +0.020 |
irl | +0.020 |
eling | +0.020 |
uch | +0.020 |
re | +0.020 |
Neg logit contributions | |
He | -0.019 |
They | -0.017 |
automobile | -0.016 |
sidewalk | -0.016 |
pavement | -0.016 |
Token | Mark |
Token ID | 2940 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
irl | +0.004 |
eling | +0.004 |
ited | +0.004 |
uch | +0.004 |
re | +0.004 |
Neg logit contributions | |
He | -0.002 |
der | -0.002 |
They | -0.002 |
sidewalk | -0.002 |
automobile | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
eling | +0.035 |
ited | +0.035 |
irl | +0.035 |
eking | +0.034 |
inations | +0.033 |
Neg logit contributions | |
He | -0.025 |
They | -0.022 |
The | -0.020 |
automobile | -0.019 |
sidewalk | -0.019 |
Token | running |
Token ID | 2491 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
ited | +0.068 |
eling | +0.063 |
kins | +0.062 |
inations | +0.061 |
if | +0.061 |
Neg logit contributions | |
He | -0.055 |
They | -0.049 |
The | -0.048 |
cop | -0.045 |
automobile | -0.045 |
Token | through |
Token ID | 832 |
Feature activation | +3.16e+00 |
Pos logit contributions | |
He | +0.082 |
They | +0.076 |
automobile | +0.073 |
tr | +0.071 |
sidewalk | +0.070 |
Neg logit contributions | |
ited | -0.099 |
irl | -0.099 |
eling | -0.098 |
eking | -0.094 |
re | -0.093 |
Token | a |
Token ID | 257 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
He | +0.062 |
They | +0.056 |
automobile | +0.051 |
tr | +0.051 |
sidewalk | +0.050 |
Neg logit contributions | |
eling | -0.091 |
ited | -0.091 |
irl | -0.089 |
if | -0.086 |
eking | -0.086 |
Token | field |
Token ID | 2214 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
He | +0.042 |
They | +0.038 |
The | +0.035 |
automobile | +0.034 |
sidewalk | +0.034 |
Neg logit contributions | |
ited | -0.048 |
eling | -0.047 |
irl | -0.047 |
uch | -0.044 |
if | -0.044 |
Token | when |
Token ID | 618 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
He | +0.034 |
They | +0.032 |
automobile | +0.031 |
sidewalk | +0.029 |
tr | +0.029 |
Neg logit contributions | |
ited | -0.048 |
irl | -0.047 |
eling | -0.046 |
if | -0.046 |
eking | -0.045 |
Token | he |
Token ID | 339 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
He | +0.022 |
They | +0.020 |
automobile | +0.018 |
sidewalk | +0.017 |
tr | +0.017 |
Neg logit contributions | |
irl | -0.057 |
eling | -0.056 |
ited | -0.056 |
uch | -0.055 |
re | -0.054 |
Token | day |
Token ID | 1110 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
He | +0.023 |
They | +0.021 |
der | +0.020 |
The | +0.020 |
Then | +0.019 |
Neg logit contributions | |
irl | -0.016 |
ited | -0.015 |
time | -0.015 |
year | -0.015 |
years | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
ited | +0.056 |
irl | +0.055 |
eling | +0.055 |
re | +0.054 |
if | +0.054 |
Neg logit contributions | |
He | -0.031 |
They | -0.027 |
sidewalk | -0.025 |
automobile | -0.025 |
tr | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
He | +0.029 |
They | +0.025 |
automobile | +0.023 |
sidewalk | +0.023 |
The | +0.023 |
Neg logit contributions | |
ited | -0.048 |
eling | -0.047 |
irl | -0.046 |
eking | -0.045 |
uch | -0.045 |
Token | park |
Token ID | 3952 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
He | +0.077 |
der | +0.074 |
They | +0.071 |
driver | +0.067 |
tr | +0.065 |
Neg logit contributions | |
irl | -0.074 |
ited | -0.073 |
eling | -0.071 |
years | -0.071 |
if | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +6.01e+00 |
Pos logit contributions | |
He | +0.040 |
tr | +0.036 |
sidewalk | +0.036 |
automobile | +0.035 |
curb | +0.035 |
Neg logit contributions | |
re | -0.066 |
years | -0.066 |
irl | -0.065 |
ited | -0.065 |
if | -0.064 |
Token | All |
Token ID | 1439 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
der | +0.074 |
tr | +0.074 |
automobile | +0.065 |
extingu | +0.061 |
arrow | +0.061 |
Neg logit contributions | |
re | -0.183 |
time | -0.181 |
ited | -0.175 |
years | -0.175 |
irl | -0.170 |
Token | the |
Token ID | 262 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
He | +0.038 |
They | +0.033 |
The | +0.030 |
automobile | +0.028 |
Tim | +0.027 |
Neg logit contributions | |
ited | -0.064 |
eking | -0.063 |
irl | -0.063 |
eling | -0.063 |
uch | -0.062 |
Token | children |
Token ID | 1751 |
Feature activation | -7.95e-01 |
Pos logit contributions | |
He | +0.044 |
sidewalk | +0.040 |
They | +0.040 |
automobile | +0.038 |
cop | +0.035 |
Neg logit contributions | |
ited | -0.077 |
eling | -0.074 |
uch | -0.074 |
irl | -0.074 |
eking | -0.073 |
Token | were |
Token ID | 547 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
ited | +0.023 |
eling | +0.022 |
irl | +0.022 |
eking | +0.021 |
uch | +0.021 |
Neg logit contributions | |
He | -0.017 |
They | -0.015 |
The | -0.014 |
automobile | -0.014 |
sidewalk | -0.013 |
Token | playing |
Token ID | 2712 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
ited | +0.105 |
eling | +0.097 |
uch | +0.095 |
inations | +0.094 |
irl | +0.093 |
Neg logit contributions | |
He | -0.079 |
They | -0.076 |
The | -0.069 |
Tim | -0.066 |
automobile | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
He | +0.097 |
They | +0.085 |
tr | +0.084 |
automobile | +0.083 |
sidewalk | +0.082 |
Neg logit contributions | |
irl | -0.111 |
eling | -0.103 |
kins | -0.103 |
ited | -0.103 |
eking | -0.100 |
Token | other |
Token ID | 584 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
ited | +0.045 |
eling | +0.045 |
irl | +0.043 |
re | +0.042 |
if | +0.042 |
Neg logit contributions | |
He | -0.056 |
They | -0.052 |
sidewalk | -0.049 |
tr | -0.048 |
automobile | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ited | +0.063 |
eling | +0.062 |
irl | +0.061 |
if | +0.060 |
eking | +0.060 |
Neg logit contributions | |
He | -0.032 |
They | -0.027 |
sidewalk | -0.025 |
automobile | -0.025 |
tr | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | -7.42e-01 |
Pos logit contributions | |
eling | +0.027 |
kins | +0.026 |
re | +0.026 |
years | +0.026 |
done | +0.026 |
Neg logit contributions | |
He | -0.028 |
pavement | -0.025 |
sidewalk | -0.025 |
curb | -0.025 |
automobile | -0.024 |
Token | their |
Token ID | 511 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
irl | +0.021 |
eling | +0.020 |
eking | +0.019 |
years | +0.018 |
uch | +0.018 |
Neg logit contributions | |
He | -0.015 |
tr | -0.012 |
der | -0.012 |
pavement | -0.012 |
curb | -0.012 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
He | +0.018 |
They | +0.016 |
The | +0.015 |
sidewalk | +0.014 |
She | +0.014 |
Neg logit contributions | |
irl | -0.019 |
eling | -0.019 |
ited | -0.019 |
eking | -0.018 |
uch | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
ited | +0.115 |
eling | +0.114 |
irl | +0.112 |
eking | +0.110 |
if | +0.109 |
Neg logit contributions | |
He | -0.044 |
They | -0.037 |
sidewalk | -0.033 |
automobile | -0.033 |
The | -0.031 |
Token | They |
Token ID | 1119 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
pavement | +0.053 |
automobile | +0.052 |
sidewalk | +0.049 |
tr | +0.047 |
curb | +0.046 |
Neg logit contributions | |
irl | -0.119 |
re | -0.118 |
if | -0.114 |
kins | -0.114 |
years | -0.113 |
Token | promised |
Token ID | 8072 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
ited | +0.005 |
eling | +0.005 |
irl | +0.004 |
re | +0.004 |
uch | +0.004 |
Neg logit contributions | |
He | -0.007 |
sidewalk | -0.007 |
pavement | -0.007 |
They | -0.007 |
automobile | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
re | +0.099 |
eking | +0.098 |
years | +0.096 |
irl | +0.096 |
if | +0.096 |
Neg logit contributions | |
He | -0.079 |
tr | -0.068 |
pavement | -0.067 |
curb | -0.067 |
automobile | -0.067 |
Token | be |
Token ID | 307 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
irl | +0.059 |
eling | +0.056 |
ited | +0.055 |
uch | +0.055 |
eking | +0.054 |
Neg logit contributions | |
He | -0.057 |
They | -0.050 |
der | -0.049 |
automobile | -0.047 |
The | -0.047 |
Token | nice |
Token ID | 3621 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
irl | +0.114 |
eling | +0.114 |
ited | +0.111 |
eking | +0.109 |
ins | +0.108 |
Neg logit contributions | |
He | -0.147 |
They | -0.136 |
sidewalk | -0.135 |
pavement | -0.132 |
automobile | -0.132 |
Token | minded |
Token ID | 35661 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
ited | +0.007 |
eling | +0.007 |
irl | +0.007 |
eking | +0.007 |
uch | +0.007 |
Neg logit contributions | |
He | -0.007 |
They | -0.006 |
automobile | -0.006 |
The | -0.006 |
sidewalk | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
He | +0.054 |
They | +0.046 |
automobile | +0.043 |
sidewalk | +0.043 |
tr | +0.041 |
Neg logit contributions | |
eling | -0.086 |
ited | -0.085 |
irl | -0.084 |
eking | -0.082 |
if | -0.080 |
Token | table |
Token ID | 3084 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
ited | +0.020 |
eling | +0.019 |
irl | +0.019 |
eking | +0.018 |
uch | +0.018 |
Neg logit contributions | |
He | -0.015 |
They | -0.014 |
automobile | -0.013 |
sidewalk | -0.013 |
The | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
ited | +0.001 |
eling | +0.001 |
irl | +0.001 |
eking | +0.001 |
uch | +0.001 |
Neg logit contributions | |
He | -0.001 |
They | -0.001 |
automobile | -0.000 |
The | -0.000 |
sidewalk | -0.000 |
Token | set |
Token ID | 900 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
ited | +0.018 |
eling | +0.018 |
irl | +0.018 |
kins | +0.017 |
eking | +0.017 |
Neg logit contributions | |
He | -0.018 |
They | -0.016 |
automobile | -0.016 |
sidewalk | -0.015 |
The | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
irl | +0.019 |
eling | +0.018 |
eking | +0.018 |
ited | +0.018 |
kins | +0.017 |
Neg logit contributions | |
He | -0.017 |
They | -0.014 |
automobile | -0.014 |
tr | -0.014 |
sidewalk | -0.013 |
Token | plates |
Token ID | 13375 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
He | +0.003 |
They | +0.003 |
automobile | +0.003 |
The | +0.003 |
sidewalk | +0.003 |
Neg logit contributions | |
ited | -0.004 |
eling | -0.004 |
irl | -0.004 |
eking | -0.004 |
uch | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
He | +0.003 |
They | +0.002 |
automobile | +0.002 |
sidewalk | +0.002 |
The | +0.002 |
Neg logit contributions | |
ited | -0.005 |
irl | -0.005 |
kins | -0.005 |
eling | -0.005 |
re | -0.005 |
Token | nap |
Token ID | 25422 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
ited | +0.037 |
eling | +0.036 |
irl | +0.036 |
eking | +0.034 |
uch | +0.034 |
Neg logit contributions | |
He | -0.036 |
They | -0.033 |
automobile | -0.031 |
sidewalk | -0.030 |
The | -0.030 |
Token | kins |
Token ID | 5331 |
Feature activation | -6.90e-01 |
Pos logit contributions | |
ited | +0.063 |
eling | +0.060 |
uch | +0.058 |
irl | +0.058 |
if | +0.056 |
Neg logit contributions | |
He | -0.052 |
They | -0.050 |
The | -0.043 |
sidewalk | -0.042 |
der | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
ited | +0.023 |
irl | +0.023 |
eling | +0.023 |
eking | +0.022 |
re | +0.022 |
Neg logit contributions | |
He | -0.010 |
They | -0.009 |
automobile | -0.008 |
sidewalk | -0.008 |
The | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
ited | +0.074 |
eling | +0.074 |
irl | +0.073 |
eking | +0.070 |
if | +0.069 |
Neg logit contributions | |
He | -0.042 |
They | -0.038 |
The | -0.033 |
der | -0.031 |
Tim | -0.031 |
Token | bird |
Token ID | 6512 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
ited | +0.050 |
eling | +0.049 |
irl | +0.048 |
eking | +0.047 |
uch | +0.047 |
Neg logit contributions | |
He | -0.036 |
They | -0.033 |
automobile | -0.030 |
The | -0.029 |
sidewalk | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | -6.05e+00 |
Pos logit contributions | |
ited | +0.126 |
irl | +0.124 |
eling | +0.123 |
uch | +0.118 |
eking | +0.118 |
Neg logit contributions | |
He | -0.093 |
They | -0.084 |
The | -0.076 |
automobile | -0.075 |
sidewalk | -0.073 |
Token | being |
Token ID | 852 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
ited | +0.168 |
eling | +0.159 |
irl | +0.153 |
uch | +0.151 |
sie | +0.150 |
Neg logit contributions | |
He | -0.139 |
They | -0.121 |
The | -0.119 |
Tim | -0.111 |
automobile | -0.111 |
Token | so |
Token ID | 523 |
Feature activation | -7.70e+00 |
Pos logit contributions | |
ited | +0.128 |
irl | +0.120 |
eling | +0.117 |
inations | +0.112 |
uch | +0.111 |
Neg logit contributions | |
He | -0.106 |
They | -0.093 |
The | -0.093 |
Tim | -0.085 |
automobile | -0.083 |
Token | nos |
Token ID | 43630 |
Feature activation | -5.48e+00 |
Pos logit contributions | |
ited | +0.189 |
irl | +0.180 |
eling | +0.179 |
uch | +0.173 |
inations | +0.172 |
Neg logit contributions | |
He | -0.200 |
They | -0.181 |
The | -0.169 |
She | -0.160 |
sidewalk | -0.160 |
Token | y |
Token ID | 88 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ited | +0.112 |
eling | +0.109 |
irl | +0.107 |
eking | +0.102 |
uch | +0.102 |
Neg logit contributions | |
He | -0.158 |
They | -0.147 |
automobile | -0.138 |
sidewalk | -0.137 |
The | -0.137 |
Token | . |
Token ID | 13 |
Feature activation | +5.20e+00 |
Pos logit contributions | |
He | +0.017 |
They | +0.015 |
automobile | +0.014 |
sidewalk | +0.014 |
tr | +0.013 |
Neg logit contributions | |
ited | -0.032 |
eling | -0.032 |
irl | -0.031 |
if | -0.031 |
eking | -0.030 |
Token | He |
Token ID | 679 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
He | +0.057 |
automobile | +0.051 |
tr | +0.050 |
sidewalk | +0.050 |
They | +0.048 |
Neg logit contributions | |
ited | -0.175 |
irl | -0.174 |
eling | -0.172 |
if | -0.172 |
re | -0.171 |
Token | hopped |
Token ID | 45230 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
He | +0.038 |
They | +0.034 |
The | +0.031 |
automobile | +0.031 |
sidewalk | +0.030 |
Neg logit contributions | |
ited | -0.042 |
eling | -0.041 |
irl | -0.040 |
uch | -0.040 |
eking | -0.039 |
Token | over |
Token ID | 625 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
He | +0.039 |
They | +0.036 |
automobile | +0.036 |
sidewalk | +0.035 |
tr | +0.035 |
Neg logit contributions | |
ited | -0.051 |
irl | -0.050 |
eling | -0.048 |
eking | -0.047 |
re | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | +8.20e-01 |
Pos logit contributions | |
ited | +0.035 |
eling | +0.035 |
irl | +0.034 |
eking | +0.034 |
if | +0.033 |
Neg logit contributions | |
He | -0.026 |
They | -0.024 |
automobile | -0.023 |
sidewalk | -0.022 |
tr | -0.022 |
Token | bunny |
Token ID | 44915 |
Feature activation | -9.31e-01 |
Pos logit contributions | |
He | +0.018 |
They | +0.016 |
automobile | +0.015 |
The | +0.015 |
sidewalk | +0.015 |
Neg logit contributions | |
ited | -0.022 |
eling | -0.022 |
irl | -0.022 |
if | -0.021 |
eking | -0.021 |
Token | was |
Token ID | 373 |
Feature activation | -3.30e+00 |
Pos logit contributions | |
ited | +0.024 |
eling | +0.024 |
irl | +0.023 |
eking | +0.023 |
uch | +0.023 |
Neg logit contributions | |
He | -0.022 |
They | -0.020 |
The | -0.018 |
automobile | -0.018 |
sidewalk | -0.018 |
Token | alive |
Token ID | 6776 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
ited | +0.097 |
uch | +0.093 |
eling | +0.093 |
irl | +0.090 |
if | +0.088 |
Neg logit contributions | |
He | -0.075 |
They | -0.065 |
Tim | -0.063 |
The | -0.063 |
She | -0.059 |
Token | ! |
Token ID | 0 |
Feature activation | +5.40e+00 |
Pos logit contributions | |
He | +0.068 |
They | +0.064 |
sidewalk | +0.060 |
automobile | +0.057 |
The | +0.056 |
Neg logit contributions | |
ited | -0.109 |
irl | -0.107 |
eling | -0.106 |
if | -0.105 |
years | -0.104 |
Token | |
Token ID | 220 |
Feature activation | -8.85e+00 |
Pos logit contributions | |
extingu | +0.096 |
automobile | +0.091 |
der | +0.090 |
tr | +0.090 |
roof | +0.089 |
Neg logit contributions | |
re | -0.138 |
ited | -0.131 |
irl | -0.130 |
time | -0.126 |
uch | -0.122 |
Token |
|
Token ID | 198 |
Feature activation | -6.53e+00 |
Pos logit contributions | |
irl | +0.241 |
eling | +0.234 |
re | +0.233 |
ited | +0.232 |
if | +0.232 |
Neg logit contributions | |
He | -0.169 |
automobile | -0.156 |
They | -0.154 |
sidewalk | -0.152 |
pavement | -0.145 |
Token |
|
Token ID | 198 |
Feature activation | -5.89e+00 |
Pos logit contributions | |
ited | +0.205 |
eling | +0.202 |
irl | +0.200 |
eking | +0.195 |
uch | +0.194 |
Neg logit contributions | |
He | -0.117 |
They | -0.104 |
automobile | -0.092 |
The | -0.092 |
sidewalk | -0.091 |
Token | Sam |
Token ID | 16305 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
ited | +0.133 |
irl | +0.129 |
eling | +0.122 |
if | +0.122 |
uch | +0.118 |
Neg logit contributions | |
der | -0.138 |
He | -0.137 |
automobile | -0.135 |
They | -0.132 |
tr | -0.131 |
Token | and |
Token ID | 290 |
Feature activation | +4.87e-02 |
Pos logit contributions | |
He | +0.005 |
They | +0.005 |
The | +0.004 |
automobile | +0.004 |
sidewalk | +0.004 |
Neg logit contributions | |
ited | -0.006 |
eling | -0.005 |
irl | -0.005 |
eking | -0.005 |
uch | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
He | +0.001 |
tr | +0.001 |
They | +0.001 |
der | +0.001 |
sidewalk | +0.001 |
Neg logit contributions | |
eling | -0.001 |
ited | -0.001 |
irl | -0.001 |
if | -0.001 |
years | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
He | +0.042 |
They | +0.038 |
automobile | +0.034 |
sidewalk | +0.034 |
tr | +0.033 |
Neg logit contributions | |
ited | -0.076 |
eling | -0.073 |
irl | -0.073 |
eking | -0.072 |
if | -0.072 |
Token | continued |
Token ID | 3767 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
He | +0.052 |
They | +0.048 |
automobile | +0.047 |
sidewalk | +0.046 |
der | +0.046 |
Neg logit contributions | |
ited | -0.040 |
eling | -0.039 |
irl | -0.038 |
re | -0.037 |
uch | -0.036 |
Token | painting |
Token ID | 12036 |
Feature activation | +5.06e+00 |
Pos logit contributions | |
He | +0.052 |
They | +0.046 |
tr | +0.045 |
automobile | +0.045 |
der | +0.044 |
Neg logit contributions | |
irl | -0.081 |
eling | -0.079 |
ited | -0.077 |
re | -0.076 |
eking | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | +6.25e+00 |
Pos logit contributions | |
He | +0.092 |
They | +0.084 |
automobile | +0.078 |
sidewalk | +0.076 |
tr | +0.076 |
Neg logit contributions | |
irl | -0.147 |
eling | -0.146 |
ited | -0.146 |
if | -0.140 |
eking | -0.140 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.84e+00 |
Pos logit contributions | |
He | +0.098 |
automobile | +0.092 |
sidewalk | +0.088 |
They | +0.087 |
tr | +0.084 |
Neg logit contributions | |
ited | -0.189 |
eling | -0.186 |
irl | -0.186 |
re | -0.183 |
if | -0.182 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.00e+01 |
Pos logit contributions | |
ited | +0.296 |
eling | +0.292 |
irl | +0.283 |
eking | +0.282 |
uch | +0.275 |
Neg logit contributions | |
He | -0.155 |
They | -0.137 |
The | -0.121 |
sidewalk | -0.110 |
She | -0.110 |
Token | upon |
Token ID | 2402 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
ited | +0.196 |
eling | +0.195 |
eking | +0.182 |
uch | +0.177 |
irl | +0.177 |
Neg logit contributions | |
He | -0.333 |
They | -0.303 |
The | -0.294 |
Tim | -0.286 |
She | -0.279 |
Token | a |
Token ID | 257 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
ited | +0.082 |
irl | +0.082 |
eling | +0.081 |
if | +0.079 |
uch | +0.079 |
Neg logit contributions | |
He | -0.050 |
They | -0.045 |
automobile | -0.044 |
sidewalk | -0.043 |
tr | -0.041 |
Token | time |
Token ID | 640 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
He | +0.050 |
They | +0.049 |
The | +0.046 |
And | +0.045 |
But | +0.045 |
Neg logit contributions | |
eling | -0.031 |
eking | -0.029 |
ubb | -0.028 |
kins | -0.028 |
ins | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
ited | +0.148 |
eling | +0.148 |
eking | +0.135 |
ins | +0.135 |
inations | +0.134 |
Neg logit contributions | |
He | -0.044 |
Tim | -0.037 |
They | -0.035 |
The | -0.033 |
She | -0.029 |
Token | there |
Token ID | 612 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ited | +0.022 |
eling | +0.022 |
irl | +0.021 |
eking | +0.020 |
uch | +0.020 |
Neg logit contributions | |
He | -0.034 |
They | -0.032 |
The | -0.030 |
automobile | -0.030 |
sidewalk | -0.029 |
Token | â |
Token ID | 22940 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
sidewalk | +0.097 |
He | +0.095 |
automobile | +0.092 |
der | +0.092 |
pavement | +0.089 |
Neg logit contributions | |
ited | -0.173 |
re | -0.168 |
irl | -0.166 |
if | -0.165 |
eling | -0.164 |
Token | € |
Token ID | 26391 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
ited | +0.012 |
eling | +0.012 |
irl | +0.011 |
uch | +0.011 |
eking | +0.011 |
Neg logit contributions | |
He | -0.016 |
They | -0.015 |
automobile | -0.014 |
sidewalk | -0.014 |
The | -0.014 |
Token | he |
Token ID | 339 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
ited | +0.086 |
eling | +0.084 |
eking | +0.079 |
irl | +0.078 |
sie | +0.077 |
Neg logit contributions | |
He | -0.076 |
They | -0.072 |
The | -0.068 |
It | -0.064 |
sidewalk | -0.062 |
Token | exclaimed |
Token ID | 33552 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
He | +0.046 |
They | +0.041 |
The | +0.038 |
It | +0.036 |
automobile | +0.036 |
Neg logit contributions | |
ited | -0.052 |
eling | -0.052 |
irl | -0.051 |
eking | -0.051 |
kins | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +6.61e+00 |
Pos logit contributions | |
ited | +0.019 |
eling | +0.019 |
irl | +0.019 |
eking | +0.019 |
if | +0.019 |
Neg logit contributions | |
He | -0.012 |
They | -0.010 |
automobile | -0.010 |
sidewalk | -0.010 |
tr | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -5.17e+00 |
Pos logit contributions | |
He | +0.101 |
They | +0.090 |
automobile | +0.089 |
sidewalk | +0.089 |
tr | +0.083 |
Neg logit contributions | |
ited | -0.204 |
irl | -0.200 |
eling | -0.200 |
re | -0.197 |
if | -0.197 |
Token |
|
Token ID | 198 |
Feature activation | -4.51e+00 |
Pos logit contributions | |
ited | +0.169 |
eling | +0.166 |
irl | +0.162 |
eking | +0.161 |
uch | +0.159 |
Neg logit contributions | |
He | -0.092 |
They | -0.080 |
The | -0.071 |
sidewalk | -0.067 |
automobile | -0.067 |
Token | Mom |
Token ID | 29252 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
ited | +0.110 |
eling | +0.109 |
irl | +0.106 |
eking | +0.105 |
uch | +0.103 |
Neg logit contributions | |
He | -0.115 |
They | -0.105 |
The | -0.096 |
automobile | -0.095 |
sidewalk | -0.094 |
Token | smiled |
Token ID | 13541 |
Feature activation | +9.50e-01 |
Pos logit contributions | |
ited | +0.042 |
eling | +0.041 |
irl | +0.041 |
eking | +0.039 |
uch | +0.039 |
Neg logit contributions | |
He | -0.042 |
They | -0.038 |
automobile | -0.035 |
The | -0.035 |
sidewalk | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +6.72e+00 |
Pos logit contributions | |
sidewalk | +0.019 |
tr | +0.018 |
pavement | +0.018 |
automobile | +0.017 |
He | +0.017 |
Neg logit contributions | |
ited | -0.024 |
if | -0.023 |
re | -0.023 |
years | -0.023 |
ins | -0.023 |
Token | � |
Token ID | 6184 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
automobile | +0.062 |
tr | +0.059 |
sidewalk | +0.057 |
He | +0.056 |
arrow | +0.055 |
Neg logit contributions | |
re | -0.231 |
ited | -0.227 |
if | -0.227 |
irl | -0.222 |
eling | -0.222 |
Token | good |
Token ID | 922 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
He | +0.065 |
They | +0.060 |
automobile | +0.057 |
sidewalk | +0.056 |
cop | +0.055 |
Neg logit contributions | |
ited | -0.053 |
eling | -0.053 |
irl | -0.050 |
eking | -0.050 |
ins | -0.049 |
Token | hiding |
Token ID | 11816 |
Feature activation | +4.14e+00 |
Pos logit contributions | |
irl | +0.035 |
ited | +0.032 |
re | +0.031 |
eling | +0.030 |
time | +0.030 |
Neg logit contributions | |
He | -0.048 |
They | -0.045 |
The | -0.044 |
sidewalk | -0.043 |
automobile | -0.042 |
Token | spot |
Token ID | 4136 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
He | +0.094 |
They | +0.086 |
The | +0.078 |
automobile | +0.078 |
sidewalk | +0.078 |
Neg logit contributions | |
ited | -0.109 |
irl | -0.107 |
eling | -0.107 |
eking | -0.103 |
uch | -0.102 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
ited | +0.013 |
eling | +0.013 |
irl | +0.013 |
eking | +0.013 |
uch | +0.013 |
Neg logit contributions | |
He | -0.008 |
They | -0.007 |
automobile | -0.006 |
The | -0.006 |
sidewalk | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
X | +0.050 |
automobile | +0.049 |
He | +0.049 |
sidewalk | +0.046 |
tr | +0.046 |
Neg logit contributions | |
re | -0.100 |
irl | -0.093 |
ited | -0.091 |
if | -0.091 |
years | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | -5.08e+00 |
Pos logit contributions | |
irl | +0.154 |
ited | +0.145 |
eling | +0.145 |
if | +0.144 |
inations | +0.144 |
Neg logit contributions | |
He | -0.124 |
They | -0.111 |
automobile | -0.108 |
sidewalk | -0.108 |
der | -0.107 |
Token | From |
Token ID | 4863 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ited | +0.125 |
irl | +0.125 |
eling | +0.122 |
if | +0.120 |
uch | +0.117 |
Neg logit contributions | |
He | -0.123 |
They | -0.112 |
automobile | -0.103 |
The | -0.103 |
sidewalk | -0.103 |
Token | that |
Token ID | 326 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
irl | +0.043 |
eling | +0.043 |
eking | +0.043 |
ited | +0.042 |
years | +0.041 |
Neg logit contributions | |
He | -0.021 |
They | -0.018 |
tr | -0.017 |
sidewalk | -0.015 |
Let | -0.015 |
Token | day |
Token ID | 1110 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
He | +0.071 |
They | +0.064 |
The | +0.060 |
automobile | +0.055 |
But | +0.054 |
Neg logit contributions | |
eling | -0.123 |
ins | -0.120 |
eking | -0.119 |
ited | -0.119 |
irl | -0.116 |
Token | on |
Token ID | 319 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
ited | +0.034 |
eling | +0.034 |
irl | +0.033 |
eking | +0.032 |
uch | +0.032 |
Neg logit contributions | |
He | -0.033 |
They | -0.030 |
automobile | -0.027 |
The | -0.027 |
sidewalk | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +8.57e-02 |
Pos logit contributions | |
He | +0.001 |
They | +0.001 |
The | +0.001 |
automobile | +0.001 |
sidewalk | +0.001 |
Neg logit contributions | |
ited | -0.002 |
eling | -0.002 |
irl | -0.002 |
eking | -0.001 |
uch | -0.001 |
Token | room |
Token ID | 2119 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
He | +0.026 |
They | +0.024 |
automobile | +0.022 |
sidewalk | +0.022 |
The | +0.022 |
Neg logit contributions | |
ited | -0.031 |
eling | -0.030 |
irl | -0.029 |
eking | -0.029 |
uch | -0.028 |
Token | stayed |
Token ID | 9658 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
He | +0.026 |
They | +0.024 |
The | +0.022 |
automobile | +0.022 |
sidewalk | +0.021 |
Neg logit contributions | |
ited | -0.034 |
eling | -0.034 |
irl | -0.033 |
eking | -0.032 |
uch | -0.032 |
Token | organized |
Token ID | 8389 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
ited | +0.070 |
eling | +0.069 |
uch | +0.068 |
re | +0.066 |
irl | +0.066 |
Neg logit contributions | |
He | -0.051 |
They | -0.046 |
automobile | -0.042 |
The | -0.042 |
sidewalk | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +5.13e+00 |
Pos logit contributions | |
He | +0.044 |
They | +0.038 |
automobile | +0.033 |
The | +0.033 |
sidewalk | +0.032 |
Neg logit contributions | |
ited | -0.097 |
eling | -0.096 |
irl | -0.096 |
eking | -0.093 |
uch | -0.092 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
automobile | +0.092 |
sidewalk | +0.089 |
pavement | +0.087 |
arrow | +0.087 |
X | +0.087 |
Neg logit contributions | |
re | -0.126 |
irl | -0.125 |
ited | -0.120 |
if | -0.120 |
time | -0.116 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.14e+01 |
Pos logit contributions | |
ited | +0.347 |
eling | +0.342 |
irl | +0.332 |
eking | +0.330 |
uch | +0.322 |
Neg logit contributions | |
He | -0.173 |
They | -0.151 |
The | -0.135 |
sidewalk | -0.121 |
It | -0.121 |
Token | upon |
Token ID | 2402 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
ited | +0.199 |
eling | +0.197 |
irl | +0.182 |
eking | +0.180 |
uch | +0.180 |
Neg logit contributions | |
He | -0.388 |
They | -0.359 |
The | -0.348 |
Tim | -0.336 |
She | -0.334 |
Token | a |
Token ID | 257 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
irl | +0.112 |
ited | +0.112 |
eling | +0.111 |
if | +0.108 |
uch | +0.108 |
Neg logit contributions | |
He | -0.070 |
They | -0.063 |
automobile | -0.061 |
sidewalk | -0.061 |
tr | -0.057 |
Token | time |
Token ID | 640 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
He | +0.007 |
They | +0.006 |
The | +0.006 |
And | +0.006 |
But | +0.006 |
Neg logit contributions | |
eling | -0.004 |
eking | -0.004 |
kins | -0.004 |
ubb | -0.004 |
ins | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
eling | +0.195 |
ited | +0.195 |
rim | +0.179 |
eking | +0.178 |
inations | +0.176 |
Neg logit contributions | |
He | -0.058 |
Tim | -0.048 |
They | -0.047 |
The | -0.044 |
She | -0.040 |
Token | there |
Token ID | 612 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
ited | +0.044 |
eling | +0.043 |
irl | +0.042 |
eking | +0.040 |
uch | +0.040 |
Neg logit contributions | |
He | -0.063 |
They | -0.059 |
The | -0.055 |
automobile | -0.055 |
sidewalk | -0.054 |
Token | her |
Token ID | 607 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
ited | +0.038 |
uch | +0.037 |
eling | +0.037 |
irl | +0.037 |
eking | +0.036 |
Neg logit contributions | |
He | -0.020 |
They | -0.019 |
The | -0.017 |
Tim | -0.016 |
automobile | -0.016 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
ited | +0.012 |
eling | +0.012 |
irl | +0.012 |
eking | +0.011 |
uch | +0.011 |
Neg logit contributions | |
He | -0.007 |
They | -0.006 |
automobile | -0.006 |
The | -0.006 |
sidewalk | -0.006 |
Token | my |
Token ID | 1820 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
irl | +0.118 |
eling | +0.116 |
ited | +0.116 |
uch | +0.111 |
eking | +0.109 |
Neg logit contributions | |
He | -0.093 |
They | -0.084 |
automobile | -0.077 |
The | -0.076 |
She | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | +5.38e+00 |
Pos logit contributions | |
ited | +0.086 |
eling | +0.085 |
irl | +0.084 |
eking | +0.082 |
uch | +0.082 |
Neg logit contributions | |
He | -0.033 |
They | -0.028 |
automobile | -0.024 |
The | -0.024 |
sidewalk | -0.023 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.49e+00 |
Pos logit contributions | |
He | +0.106 |
automobile | +0.102 |
sidewalk | +0.100 |
cop | +0.100 |
tr | +0.099 |
Neg logit contributions | |
re | -0.136 |
ited | -0.133 |
irl | -0.130 |
if | -0.128 |
eling | -0.124 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.11e+01 |
Pos logit contributions | |
ited | +0.275 |
eling | +0.270 |
irl | +0.267 |
eking | +0.259 |
uch | +0.257 |
Neg logit contributions | |
He | -0.195 |
They | -0.176 |
The | -0.158 |
automobile | -0.157 |
sidewalk | -0.155 |
Token | upon |
Token ID | 2402 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
eling | +0.215 |
ited | +0.214 |
eking | +0.205 |
uch | +0.201 |
irl | +0.198 |
Neg logit contributions | |
He | -0.361 |
They | -0.334 |
The | -0.323 |
Tim | -0.314 |
She | -0.313 |
Token | a |
Token ID | 257 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
ited | +0.122 |
irl | +0.120 |
eling | +0.120 |
if | +0.116 |
uch | +0.115 |
Neg logit contributions | |
He | -0.067 |
They | -0.060 |
automobile | -0.056 |
sidewalk | -0.055 |
tr | -0.052 |
Token | time |
Token ID | 640 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
He | +0.005 |
They | +0.004 |
The | +0.004 |
But | +0.004 |
And | +0.004 |
Neg logit contributions | |
eling | -0.003 |
eking | -0.003 |
ubb | -0.003 |
ins | -0.003 |
kins | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
ited | +0.192 |
eling | +0.192 |
eking | +0.179 |
rim | +0.176 |
inations | +0.175 |
Neg logit contributions | |
He | -0.060 |
Tim | -0.051 |
They | -0.050 |
The | -0.046 |
She | -0.043 |
Token | there |
Token ID | 612 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
ited | +0.049 |
eling | +0.049 |
irl | +0.046 |
eking | +0.045 |
inations | +0.045 |
Neg logit contributions | |
He | -0.070 |
They | -0.066 |
The | -0.062 |
automobile | -0.059 |
Tim | -0.059 |
Token | mom |
Token ID | 1995 |
Feature activation | -6.95e+00 |
Pos logit contributions | |
ited | +0.081 |
eling | +0.079 |
irl | +0.078 |
eking | +0.076 |
uch | +0.076 |
Neg logit contributions | |
He | -0.070 |
They | -0.064 |
automobile | -0.059 |
sidewalk | -0.058 |
The | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
ited | +0.204 |
eling | +0.200 |
irl | +0.198 |
eking | +0.192 |
uch | +0.192 |
Neg logit contributions | |
He | -0.137 |
They | -0.123 |
automobile | -0.111 |
sidewalk | -0.110 |
The | -0.110 |
Token | She |
Token ID | 1375 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
sidewalk | +0.021 |
tr | +0.021 |
automobile | +0.021 |
der | +0.020 |
He | +0.020 |
Neg logit contributions | |
irl | -0.109 |
re | -0.108 |
ited | -0.107 |
if | -0.106 |
eling | -0.106 |
Token | was |
Token ID | 373 |
Feature activation | -4.51e+00 |
Pos logit contributions | |
ited | +0.027 |
eling | +0.026 |
irl | +0.026 |
eking | +0.025 |
uch | +0.025 |
Neg logit contributions | |
He | -0.024 |
They | -0.022 |
The | -0.020 |
automobile | -0.020 |
sidewalk | -0.019 |
Token | wearing |
Token ID | 5762 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
ited | +0.129 |
uch | +0.119 |
irl | +0.118 |
if | +0.117 |
eling | +0.116 |
Neg logit contributions | |
He | -0.103 |
They | -0.092 |
The | -0.089 |
Tim | -0.085 |
Then | -0.083 |
Token | L |
Token ID | 406 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
ited | +0.021 |
eling | +0.020 |
irl | +0.020 |
uch | +0.020 |
eking | +0.019 |
Neg logit contributions | |
He | -0.010 |
They | -0.009 |
automobile | -0.008 |
The | -0.008 |
sidewalk | -0.008 |
Token | ila |
Token ID | 10102 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
ited | +0.222 |
eling | +0.202 |
insepar | +0.199 |
if | +0.198 |
sudden | +0.192 |
Neg logit contributions | |
He | -0.302 |
They | -0.285 |
The | -0.273 |
She | -0.271 |
Tim | -0.269 |
Token | 's |
Token ID | 338 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
eling | +0.127 |
ited | +0.125 |
irl | +0.124 |
rim | +0.118 |
inations | +0.117 |
Neg logit contributions | |
He | -0.073 |
They | -0.066 |
The | -0.062 |
automobile | -0.061 |
She | -0.057 |
Token | most |
Token ID | 749 |
Feature activation | -6.03e+00 |
Pos logit contributions | |
ited | +0.004 |
eling | +0.004 |
irl | +0.004 |
eking | +0.004 |
uch | +0.004 |
Neg logit contributions | |
He | -0.003 |
They | -0.002 |
automobile | -0.002 |
The | -0.002 |
sidewalk | -0.002 |
Token | fashion |
Token ID | 6977 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
eling | +0.160 |
ited | +0.152 |
uch | +0.152 |
eking | +0.150 |
sie | +0.145 |
Neg logit contributions | |
He | -0.132 |
automobile | -0.129 |
sidewalk | -0.117 |
They | -0.117 |
cop | -0.116 |
Token | dress |
Token ID | 6576 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
ited | +0.040 |
eling | +0.040 |
irl | +0.039 |
eking | +0.038 |
uch | +0.038 |
Neg logit contributions | |
He | -0.023 |
They | -0.020 |
The | -0.018 |
automobile | -0.018 |
sidewalk | -0.018 |
Token | their |
Token ID | 511 |
Feature activation | -6.52e-01 |
Pos logit contributions | |
He | +0.026 |
They | +0.024 |
The | +0.021 |
sidewalk | +0.020 |
automobile | +0.020 |
Neg logit contributions | |
ited | -0.045 |
uch | -0.043 |
eling | -0.042 |
irl | -0.041 |
eking | -0.041 |
Token | special |
Token ID | 2041 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
ited | +0.018 |
eling | +0.017 |
eking | +0.016 |
ins | +0.016 |
uch | +0.016 |
Neg logit contributions | |
He | -0.014 |
They | -0.013 |
sidewalk | -0.013 |
automobile | -0.013 |
The | -0.012 |
Token | friendship |
Token ID | 14738 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
He | +0.029 |
They | +0.026 |
automobile | +0.025 |
sidewalk | +0.025 |
The | +0.024 |
Neg logit contributions | |
eling | -0.037 |
ited | -0.036 |
eking | -0.035 |
uch | -0.034 |
irl | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +5.02e+00 |
Pos logit contributions | |
ited | +0.021 |
eling | +0.021 |
irl | +0.021 |
uch | +0.020 |
eking | +0.020 |
Neg logit contributions | |
He | -0.007 |
They | -0.006 |
automobile | -0.005 |
The | -0.005 |
sidewalk | -0.005 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.68e+00 |
Pos logit contributions | |
He | +0.087 |
automobile | +0.079 |
sidewalk | +0.078 |
They | +0.076 |
tr | +0.075 |
Neg logit contributions | |
ited | -0.144 |
irl | -0.143 |
eling | -0.141 |
re | -0.139 |
if | -0.138 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.14e+01 |
Pos logit contributions | |
ited | +0.321 |
eling | +0.318 |
eking | +0.307 |
irl | +0.306 |
inations | +0.300 |
Neg logit contributions | |
He | -0.181 |
They | -0.160 |
The | -0.144 |
She | -0.131 |
It | -0.131 |
Token | upon |
Token ID | 2402 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
ited | +0.219 |
eling | +0.219 |
uch | +0.202 |
irl | +0.201 |
eking | +0.200 |
Neg logit contributions | |
He | -0.377 |
They | -0.340 |
The | -0.336 |
Tim | -0.325 |
She | -0.321 |
Token | a |
Token ID | 257 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
ited | +0.124 |
irl | +0.123 |
eling | +0.122 |
if | +0.119 |
uch | +0.118 |
Neg logit contributions | |
He | -0.073 |
They | -0.066 |
automobile | -0.063 |
sidewalk | -0.062 |
tr | -0.058 |
Token | time |
Token ID | 640 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
He | +0.004 |
They | +0.004 |
The | +0.004 |
And | +0.004 |
But | +0.004 |
Neg logit contributions | |
eling | -0.003 |
eking | -0.002 |
kins | -0.002 |
ins | -0.002 |
ubb | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
eling | +0.205 |
ited | +0.205 |
inations | +0.186 |
ins | +0.185 |
irl | +0.185 |
Neg logit contributions | |
He | -0.060 |
Tim | -0.051 |
They | -0.048 |
The | -0.045 |
She | -0.042 |
Token | there |
Token ID | 612 |
Feature activation | +8.61e-01 |
Pos logit contributions | |
ited | +0.043 |
eling | +0.042 |
irl | +0.041 |
eking | +0.039 |
uch | +0.039 |
Neg logit contributions | |
He | -0.064 |
They | -0.059 |
The | -0.056 |
automobile | -0.055 |
sidewalk | -0.054 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
ited | +0.050 |
eling | +0.049 |
irl | +0.049 |
uch | +0.047 |
eking | +0.046 |
Neg logit contributions | |
He | -0.050 |
They | -0.046 |
The | -0.042 |
sidewalk | -0.041 |
automobile | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
re | +0.054 |
if | +0.052 |
irl | +0.052 |
years | +0.051 |
eling | +0.051 |
Neg logit contributions | |
He | -0.033 |
automobile | -0.030 |
tr | -0.029 |
They | -0.028 |
curb | -0.027 |
Token | safe |
Token ID | 3338 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
ited | +0.007 |
eling | +0.007 |
irl | +0.007 |
eking | +0.007 |
uch | +0.007 |
Neg logit contributions | |
He | -0.010 |
They | -0.010 |
automobile | -0.009 |
The | -0.009 |
sidewalk | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +5.26e+00 |
Pos logit contributions | |
He | +0.004 |
They | +0.004 |
automobile | +0.004 |
The | +0.003 |
tr | +0.003 |
Neg logit contributions | |
eling | -0.007 |
re | -0.007 |
irl | -0.007 |
ited | -0.007 |
eking | -0.007 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.03e+01 |
Pos logit contributions | |
automobile | +0.126 |
arrow | +0.122 |
pavement | +0.122 |
X | +0.121 |
sidewalk | +0.120 |
Neg logit contributions | |
re | -0.100 |
if | -0.098 |
irl | -0.095 |
time | -0.092 |
ins | -0.090 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
ited | +0.343 |
eling | +0.338 |
irl | +0.330 |
eking | +0.325 |
uch | +0.320 |
Neg logit contributions | |
He | -0.177 |
They | -0.155 |
The | -0.138 |
sidewalk | -0.128 |
automobile | -0.128 |
Token | upon |
Token ID | 2402 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
ited | +0.206 |
eling | +0.202 |
irl | +0.190 |
eking | +0.188 |
uch | +0.184 |
Neg logit contributions | |
He | -0.378 |
They | -0.352 |
The | -0.335 |
automobile | -0.322 |
Tim | -0.321 |
Token | a |
Token ID | 257 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
ited | +0.123 |
irl | +0.123 |
eling | +0.122 |
if | +0.119 |
uch | +0.118 |
Neg logit contributions | |
He | -0.075 |
They | -0.067 |
automobile | -0.065 |
sidewalk | -0.064 |
tr | -0.060 |
Token | time |
Token ID | 640 |
Feature activation | -4.93e+00 |
Pos logit contributions | |
eling | +0.006 |
eking | +0.005 |
ubb | +0.005 |
ins | +0.005 |
kins | +0.005 |
Neg logit contributions | |
He | -0.011 |
They | -0.011 |
The | -0.010 |
But | -0.010 |
And | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
ited | +0.211 |
eling | +0.209 |
eking | +0.193 |
rim | +0.193 |
inations | +0.191 |
Neg logit contributions | |
He | -0.064 |
Tim | -0.053 |
They | -0.053 |
The | -0.050 |
She | -0.042 |
Token | there |
Token ID | 612 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
ited | +0.048 |
eling | +0.047 |
irl | +0.046 |
eking | +0.044 |
uch | +0.043 |
Neg logit contributions | |
He | -0.080 |
They | -0.075 |
The | -0.070 |
automobile | -0.070 |
sidewalk | -0.069 |