Token | on |
Token ID | 319 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
arl | +0.045 |
ay | +0.043 |
elly | +0.042 |
ane | +0.042 |
ute | +0.041 |
Neg logit contributions | |
put | -0.036 |
pretended | -0.036 |
never | -0.035 |
punished | -0.034 |
forgot | -0.034 |
Token | his |
Token ID | 465 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
put | +0.005 |
never | +0.005 |
pretended | +0.004 |
learned | +0.004 |
used | +0.004 |
Neg logit contributions | |
arl | -0.017 |
ay | -0.017 |
elly | -0.017 |
Today | -0.017 |
ane | -0.017 |
Token | paw |
Token ID | 41827 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
arl | +0.004 |
ay | +0.004 |
ute | +0.004 |
elly | +0.004 |
Today | +0.004 |
Neg logit contributions | |
pretended | -0.003 |
put | -0.003 |
never | -0.003 |
used | -0.003 |
made | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
arl | +0.083 |
ay | +0.079 |
elly | +0.079 |
Today | +0.077 |
ane | +0.076 |
Neg logit contributions | |
put | -0.062 |
never | -0.060 |
pretended | -0.059 |
used | -0.058 |
learned | -0.058 |
Token | They |
Token ID | 1119 |
Feature activation | +8.71e+00 |
Pos logit contributions | |
arl | +0.059 |
ane | +0.057 |
ute | +0.057 |
elly | +0.056 |
ay | +0.055 |
Neg logit contributions | |
put | -0.051 |
pretended | -0.048 |
learned | -0.048 |
never | -0.047 |
used | -0.046 |
Token | also |
Token ID | 635 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
punished | +0.075 |
permission | +0.074 |
eling | +0.073 |
supposed | +0.073 |
myself | +0.071 |
Neg logit contributions | |
� | -0.181 |
ay | -0.177 |
ane | -0.170 |
ifying | -0.170 |
arl | -0.170 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
himself | +0.063 |
never | +0.063 |
herself | +0.058 |
ck | +0.057 |
ts | +0.057 |
Neg logit contributions | |
arl | -0.230 |
� | -0.223 |
elly | -0.220 |
ay | -0.218 |
ute | -0.209 |
Token | him |
Token ID | 683 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
forgot | +0.019 |
put | +0.019 |
pretended | +0.019 |
never | +0.018 |
learned | +0.018 |
Neg logit contributions | |
ay | -0.043 |
arl | -0.040 |
ummy | -0.040 |
elly | -0.040 |
ore | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
ay | +0.009 |
arl | +0.009 |
elly | +0.009 |
igsaw | +0.009 |
aris | +0.009 |
Neg logit contributions | |
pretended | -0.005 |
forgot | -0.005 |
never | -0.005 |
put | -0.005 |
remembered | -0.005 |
Token | delicious |
Token ID | 12625 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
put | +0.009 |
never | +0.009 |
pretended | +0.009 |
used | +0.009 |
cleaned | +0.008 |
Neg logit contributions | |
arl | -0.014 |
elly | -0.014 |
ay | -0.014 |
ane | -0.014 |
Today | -0.013 |
Token | bowl |
Token ID | 9396 |
Feature activation | -5.59e+00 |
Pos logit contributions | |
arl | +0.082 |
ay | +0.080 |
elly | +0.078 |
Today | +0.076 |
ute | +0.075 |
Neg logit contributions | |
put | -0.062 |
never | -0.062 |
pretended | -0.061 |
learned | -0.059 |
hoped | -0.059 |
Token | band |
Token ID | 4097 |
Feature activation | -3.84e+00 |
Pos logit contributions | |
put | +0.007 |
cleaned | +0.007 |
used | +0.006 |
never | +0.006 |
made | +0.006 |
Neg logit contributions | |
arl | -0.012 |
elly | -0.012 |
ane | -0.012 |
ay | -0.012 |
Today | -0.011 |
Token | - |
Token ID | 12 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
arl | +0.066 |
ay | +0.063 |
elly | +0.062 |
ane | +0.060 |
Today | +0.059 |
Neg logit contributions | |
put | -0.086 |
never | -0.084 |
pretended | -0.084 |
used | -0.082 |
learned | -0.082 |
Token | aid |
Token ID | 1698 |
Feature activation | -4.42e+00 |
Pos logit contributions | |
arl | +0.034 |
ay | +0.033 |
elly | +0.033 |
ane | +0.032 |
ute | +0.031 |
Neg logit contributions | |
put | -0.018 |
pretended | -0.018 |
never | -0.018 |
forgot | -0.017 |
used | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
arl | +0.106 |
elly | +0.102 |
ay | +0.102 |
Today | +0.101 |
� | +0.099 |
Neg logit contributions | |
put | -0.071 |
never | -0.068 |
made | -0.066 |
used | -0.065 |
pretended | -0.065 |
Token | They |
Token ID | 1119 |
Feature activation | +9.18e+00 |
Pos logit contributions | |
ane | +0.057 |
ute | +0.056 |
arl | +0.056 |
ay | +0.056 |
owers | +0.053 |
Neg logit contributions | |
put | -0.056 |
learned | -0.053 |
never | -0.053 |
pretended | -0.052 |
hoped | -0.052 |
Token | also |
Token ID | 635 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
permission | +0.087 |
himself | +0.081 |
finding | +0.080 |
wed | +0.079 |
supposed | +0.078 |
Neg logit contributions | |
� | -0.199 |
ay | -0.186 |
owers | -0.184 |
ane | -0.183 |
arl | -0.183 |
Token | praised |
Token ID | 15342 |
Feature activation | -3.85e-01 |
Pos logit contributions | |
himself | +0.082 |
never | +0.078 |
pretended | +0.075 |
herself | +0.072 |
myself | +0.072 |
Neg logit contributions | |
arl | -0.220 |
� | -0.218 |
elly | -0.212 |
ay | -0.206 |
oms | -0.200 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
arl | +0.011 |
ay | +0.011 |
elly | +0.010 |
Today | +0.010 |
ane | +0.010 |
Neg logit contributions | |
put | -0.004 |
never | -0.004 |
pretended | -0.004 |
used | -0.004 |
learned | -0.004 |
Token | for |
Token ID | 329 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
arl | +0.032 |
ay | +0.032 |
elly | +0.032 |
Today | +0.031 |
ane | +0.030 |
Neg logit contributions | |
put | -0.027 |
pretended | -0.026 |
never | -0.026 |
punished | -0.026 |
used | -0.025 |
Token | being |
Token ID | 852 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
pretended | +0.037 |
forgot | +0.036 |
hoped | +0.032 |
counted | +0.031 |
remembered | +0.031 |
Neg logit contributions | |
arl | -0.061 |
� | -0.053 |
apping | -0.051 |
ummy | -0.051 |
ay | -0.050 |
Token | smart |
Token ID | 4451 |
Feature activation | -3.75e+00 |
Pos logit contributions | |
pretended | +0.048 |
forgot | +0.045 |
put | +0.043 |
wed | +0.042 |
imagined | +0.042 |
Neg logit contributions | |
arl | -0.061 |
elly | -0.060 |
apping | -0.058 |
ay | -0.058 |
over | -0.056 |
Token | band |
Token ID | 4097 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
put | +0.009 |
cleaned | +0.008 |
learned | +0.008 |
used | +0.008 |
made | +0.008 |
Neg logit contributions | |
arl | -0.017 |
elly | -0.017 |
ay | -0.017 |
ane | -0.017 |
Today | -0.017 |
Token | - |
Token ID | 12 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
arl | +0.048 |
ay | +0.046 |
elly | +0.045 |
ane | +0.044 |
Today | +0.043 |
Neg logit contributions | |
put | -0.052 |
never | -0.051 |
pretended | -0.050 |
used | -0.049 |
punished | -0.049 |
Token | aid |
Token ID | 1698 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
arl | +0.026 |
ay | +0.025 |
elly | +0.025 |
ane | +0.024 |
ute | +0.023 |
Neg logit contributions | |
put | -0.015 |
pretended | -0.015 |
never | -0.014 |
forgot | -0.014 |
remembered | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
arl | +0.091 |
ay | +0.088 |
elly | +0.087 |
Today | +0.086 |
ane | +0.084 |
Neg logit contributions | |
put | -0.063 |
never | -0.060 |
pretended | -0.059 |
made | -0.058 |
used | -0.058 |
Token | They |
Token ID | 1119 |
Feature activation | +8.51e+00 |
Pos logit contributions | |
ane | +0.064 |
ay | +0.063 |
ute | +0.062 |
arl | +0.062 |
owers | +0.060 |
Neg logit contributions | |
put | -0.067 |
learned | -0.065 |
pretended | -0.064 |
never | -0.063 |
hoped | -0.063 |
Token | also |
Token ID | 635 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
eling | +0.080 |
ck | +0.080 |
permission | +0.079 |
finding | +0.077 |
pretending | +0.076 |
Neg logit contributions | |
� | -0.187 |
elly | -0.171 |
owers | -0.170 |
arl | -0.169 |
uddle | -0.168 |
Token | helped |
Token ID | 4193 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
himself | +0.072 |
never | +0.072 |
myself | +0.071 |
supposed | +0.068 |
eling | +0.068 |
Neg logit contributions | |
elly | -0.216 |
arl | -0.215 |
� | -0.207 |
owers | -0.196 |
ay | -0.196 |
Token | to |
Token ID | 284 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
arl | +0.008 |
ay | +0.008 |
elly | +0.007 |
Today | +0.007 |
ummy | +0.007 |
Neg logit contributions | |
never | -0.004 |
pretended | -0.004 |
hoped | -0.003 |
learned | -0.003 |
remembered | -0.003 |
Token | wash |
Token ID | 13502 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
pretended | +0.049 |
remembered | +0.046 |
hoped | +0.044 |
forgot | +0.043 |
learned | +0.043 |
Neg logit contributions | |
arl | -0.119 |
� | -0.110 |
ay | -0.108 |
ute | -0.107 |
Today | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
arl | +0.035 |
ay | +0.034 |
elly | +0.034 |
Today | +0.033 |
ute | +0.033 |
Neg logit contributions | |
put | -0.014 |
never | -0.014 |
pretended | -0.013 |
learned | -0.013 |
used | -0.013 |
Token | salt |
Token ID | 8268 |
Feature activation | -5.17e+00 |
Pos logit contributions | |
arl | +0.051 |
elly | +0.049 |
ay | +0.048 |
ute | +0.047 |
Today | +0.046 |
Neg logit contributions | |
never | -0.036 |
pretended | -0.036 |
put | -0.035 |
hoped | -0.034 |
used | -0.034 |
Token | they |
Token ID | 484 |
Feature activation | +5.94e+00 |
Pos logit contributions | |
arl | +0.024 |
ay | +0.023 |
elly | +0.023 |
ute | +0.022 |
ane | +0.022 |
Neg logit contributions | |
put | -0.012 |
never | -0.012 |
learned | -0.011 |
made | -0.011 |
cleaned | -0.011 |
Token | loved |
Token ID | 6151 |
Feature activation | +7.07e-02 |
Pos logit contributions | |
ck | +0.108 |
forth | +0.103 |
punished | +0.098 |
eling | +0.096 |
pretended | +0.095 |
Neg logit contributions | |
owers | -0.093 |
� | -0.092 |
ane | -0.092 |
ay | -0.090 |
elly | -0.090 |
Token | them |
Token ID | 606 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
pretended | +0.001 |
punished | +0.001 |
forgot | +0.001 |
put | +0.001 |
remembered | +0.001 |
Neg logit contributions | |
arl | -0.002 |
elly | -0.002 |
ay | -0.002 |
ummy | -0.002 |
Today | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
arl | +0.040 |
ay | +0.038 |
elly | +0.038 |
Today | +0.037 |
ane | +0.037 |
Neg logit contributions | |
put | -0.029 |
pretended | -0.027 |
never | -0.027 |
used | -0.027 |
learned | -0.026 |
Token | They |
Token ID | 1119 |
Feature activation | +8.72e+00 |
Pos logit contributions | |
ane | +0.062 |
anger | +0.059 |
ay | +0.059 |
ute | +0.059 |
andel | +0.059 |
Neg logit contributions | |
put | -0.056 |
learned | -0.055 |
hoped | -0.053 |
made | -0.052 |
pretended | -0.052 |
Token | also |
Token ID | 635 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
permission | +0.106 |
eling | +0.105 |
ck | +0.101 |
ts | +0.100 |
forth | +0.095 |
Neg logit contributions | |
� | -0.166 |
owers | -0.161 |
elly | -0.160 |
ay | -0.156 |
arry | -0.155 |
Token | said |
Token ID | 531 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
eling | +0.085 |
ts | +0.084 |
permission | +0.081 |
forth | +0.080 |
ck | +0.079 |
Neg logit contributions | |
arl | -0.203 |
elly | -0.203 |
� | -0.197 |
owers | -0.191 |
arry | -0.187 |
Token | they |
Token ID | 484 |
Feature activation | +6.15e+00 |
Pos logit contributions | |
arl | +0.007 |
ay | +0.006 |
elly | +0.006 |
Today | +0.006 |
ane | +0.006 |
Neg logit contributions | |
put | -0.004 |
never | -0.004 |
pretended | -0.003 |
learned | -0.003 |
made | -0.003 |
Token | were |
Token ID | 547 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
ck | +0.111 |
forth | +0.105 |
eling | +0.102 |
punished | +0.097 |
asking | +0.096 |
Neg logit contributions | |
owers | -0.099 |
ane | -0.098 |
� | -0.095 |
ay | -0.095 |
elly | -0.094 |
Token | proud |
Token ID | 6613 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
pretended | +0.064 |
wed | +0.063 |
imagined | +0.063 |
forgot | +0.060 |
tried | +0.060 |
Neg logit contributions | |
arl | -0.062 |
ay | -0.052 |
ummy | -0.051 |
ute | -0.049 |
owers | -0.049 |
Token | of |
Token ID | 286 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
arl | +0.021 |
elly | +0.020 |
ay | +0.019 |
Today | +0.018 |
ummy | +0.018 |
Neg logit contributions | |
put | -0.037 |
pretended | -0.036 |
never | -0.036 |
punished | -0.036 |
used | -0.035 |
Token | they |
Token ID | 484 |
Feature activation | +5.52e+00 |
Pos logit contributions | |
arl | +0.011 |
elly | +0.011 |
ay | +0.011 |
Today | +0.010 |
ane | +0.010 |
Neg logit contributions | |
put | -0.006 |
never | -0.006 |
pretended | -0.006 |
learned | -0.006 |
made | -0.006 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
ck | +0.097 |
pretended | +0.088 |
punished | +0.086 |
put | +0.085 |
ts | +0.080 |
Neg logit contributions | |
� | -0.102 |
ane | -0.101 |
ay | -0.101 |
Today | -0.099 |
elly | -0.098 |
Token | them |
Token ID | 606 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
pretended | +0.002 |
forgot | +0.002 |
punished | +0.002 |
put | +0.002 |
remembered | +0.002 |
Neg logit contributions | |
arl | -0.005 |
elly | -0.005 |
ay | -0.005 |
ummy | -0.005 |
Today | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
arl | +0.050 |
ay | +0.049 |
elly | +0.048 |
Today | +0.047 |
ane | +0.047 |
Neg logit contributions | |
put | -0.033 |
pretended | -0.032 |
never | -0.031 |
used | -0.031 |
punished | -0.031 |
Token | They |
Token ID | 1119 |
Feature activation | +8.59e+00 |
Pos logit contributions | |
ane | +0.067 |
ay | +0.065 |
anger | +0.065 |
ute | +0.064 |
arl | +0.064 |
Neg logit contributions | |
put | -0.058 |
hoped | -0.055 |
learned | -0.055 |
never | -0.055 |
pretended | -0.054 |
Token | also |
Token ID | 635 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
ck | +0.107 |
ts | +0.105 |
eling | +0.098 |
permission | +0.097 |
ps | +0.094 |
Neg logit contributions | |
� | -0.172 |
ay | -0.168 |
elly | -0.168 |
arl | -0.165 |
arry | -0.163 |
Token | said |
Token ID | 531 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
ck | +0.097 |
ts | +0.091 |
forth | +0.081 |
permission | +0.080 |
eling | +0.078 |
Neg logit contributions | |
elly | -0.220 |
arl | -0.213 |
Today | -0.206 |
ay | -0.198 |
� | -0.194 |
Token | they |
Token ID | 484 |
Feature activation | +5.99e+00 |
Pos logit contributions | |
arl | +0.011 |
ay | +0.011 |
elly | +0.011 |
Today | +0.011 |
ane | +0.011 |
Neg logit contributions | |
put | -0.007 |
never | -0.006 |
pretended | -0.006 |
learned | -0.006 |
made | -0.006 |
Token | liked |
Token ID | 8288 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
ck | +0.107 |
punished | +0.088 |
forth | +0.088 |
eling | +0.088 |
ts | +0.088 |
Neg logit contributions | |
ane | -0.110 |
ay | -0.109 |
elly | -0.106 |
� | -0.105 |
owers | -0.105 |
Token | the |
Token ID | 262 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
pretended | +0.015 |
put | +0.015 |
punished | +0.015 |
forgot | +0.015 |
learned | +0.015 |
Neg logit contributions | |
arl | -0.019 |
elly | -0.018 |
ay | -0.018 |
ummy | -0.018 |
Today | -0.018 |
Token | movie |
Token ID | 3807 |
Feature activation | -5.62e+00 |
Pos logit contributions | |
arl | +0.048 |
ay | +0.046 |
elly | +0.045 |
Today | +0.044 |
ane | +0.044 |
Neg logit contributions | |
put | -0.042 |
never | -0.040 |
pretended | -0.040 |
used | -0.039 |
learned | -0.039 |
Token | make |
Token ID | 787 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
pretended | +0.063 |
hoped | +0.060 |
learned | +0.058 |
cleaned | +0.057 |
put | +0.057 |
Neg logit contributions | |
arl | -0.124 |
elly | -0.118 |
ay | -0.116 |
Today | -0.114 |
ute | -0.112 |
Token | them |
Token ID | 606 |
Feature activation | -7.94e-02 |
Pos logit contributions | |
arl | +0.008 |
ay | +0.008 |
elly | +0.008 |
ute | +0.008 |
ane | +0.008 |
Neg logit contributions | |
pretended | -0.005 |
put | -0.005 |
never | -0.005 |
hoped | -0.005 |
learned | -0.005 |
Token | angry |
Token ID | 7954 |
Feature activation | -3.30e+00 |
Pos logit contributions | |
arl | +0.002 |
ay | +0.002 |
elly | +0.002 |
ute | +0.002 |
ane | +0.002 |
Neg logit contributions | |
pretended | -0.001 |
put | -0.001 |
never | -0.001 |
used | -0.001 |
hoped | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
arl | +0.071 |
ay | +0.068 |
elly | +0.067 |
Today | +0.065 |
ane | +0.065 |
Neg logit contributions | |
put | -0.061 |
never | -0.059 |
pretended | -0.058 |
learned | -0.057 |
used | -0.057 |
Token | They |
Token ID | 1119 |
Feature activation | +8.78e+00 |
Pos logit contributions | |
arl | +0.069 |
ane | +0.068 |
ute | +0.067 |
ay | +0.066 |
owers | +0.065 |
Neg logit contributions | |
put | -0.057 |
learned | -0.057 |
never | -0.056 |
cleaned | -0.054 |
hoped | -0.054 |
Token | also |
Token ID | 635 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
ts | +0.102 |
eling | +0.095 |
pretended | +0.093 |
pretending | +0.093 |
anut | +0.090 |
Neg logit contributions | |
ay | -0.182 |
ane | -0.173 |
arry | -0.166 |
arl | -0.163 |
uttered | -0.156 |
Token | told |
Token ID | 1297 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
ts | +0.104 |
pretended | +0.092 |
himself | +0.082 |
pretending | +0.080 |
ck | +0.076 |
Neg logit contributions | |
ay | -0.218 |
arl | -0.218 |
arry | -0.212 |
ane | -0.205 |
ute | -0.202 |
Token | her |
Token ID | 607 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
pretended | +0.012 |
used | +0.012 |
remembered | +0.012 |
put | +0.012 |
hoped | +0.012 |
Neg logit contributions | |
arl | -0.030 |
ay | -0.028 |
elly | -0.028 |
ummy | -0.028 |
Today | -0.027 |
Token | that |
Token ID | 326 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
pretended | +0.020 |
put | +0.020 |
used | +0.019 |
punished | +0.019 |
hoped | +0.018 |
Neg logit contributions | |
arl | -0.028 |
ay | -0.027 |
elly | -0.027 |
ane | -0.027 |
ute | -0.027 |
Token | things |
Token ID | 1243 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
put | +0.033 |
pretended | +0.032 |
punished | +0.031 |
never | +0.031 |
learned | +0.031 |
Neg logit contributions | |
arl | -0.060 |
elly | -0.060 |
ay | -0.058 |
ane | -0.058 |
Today | -0.058 |
Token | are |
Token ID | 389 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
pretended | +0.029 |
put | +0.027 |
remembered | +0.027 |
hoped | +0.026 |
punished | +0.026 |
Neg logit contributions | |
arl | -0.038 |
ay | -0.038 |
ane | -0.036 |
elly | -0.035 |
ute | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | +7.22e+00 |
Pos logit contributions | |
arl | +0.109 |
ay | +0.104 |
elly | +0.104 |
Today | +0.102 |
ane | +0.101 |
Neg logit contributions | |
put | -0.075 |
never | -0.072 |
pretended | -0.071 |
used | -0.069 |
learned | -0.069 |
Token | a |
Token ID | 257 |
Feature activation | +9.34e-01 |
Pos logit contributions | |
never | +0.101 |
punished | +0.099 |
pretended | +0.094 |
forgot | +0.094 |
wed | +0.094 |
Neg logit contributions | |
arl | -0.153 |
ay | -0.153 |
Today | -0.148 |
elly | -0.146 |
ute | -0.144 |
Token | kiss |
Token ID | 9245 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
put | +0.012 |
used | +0.011 |
cleaned | +0.011 |
pretended | +0.011 |
never | +0.011 |
Neg logit contributions | |
arl | -0.025 |
elly | -0.025 |
ay | -0.025 |
ane | -0.024 |
Today | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
arl | +0.061 |
ay | +0.059 |
elly | +0.058 |
Today | +0.057 |
ane | +0.057 |
Neg logit contributions | |
put | -0.046 |
never | -0.044 |
pretended | -0.044 |
used | -0.043 |
learned | -0.043 |
Token | They |
Token ID | 1119 |
Feature activation | +8.56e+00 |
Pos logit contributions | |
ane | +0.069 |
arl | +0.067 |
ute | +0.067 |
ay | +0.065 |
elly | +0.063 |
Neg logit contributions | |
put | -0.055 |
learned | -0.052 |
hoped | -0.052 |
pretended | -0.052 |
never | -0.052 |
Token | also |
Token ID | 635 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
remembered | +0.101 |
never | +0.099 |
pretended | +0.099 |
punished | +0.098 |
tried | +0.095 |
Neg logit contributions | |
arl | -0.211 |
ay | -0.201 |
Today | -0.195 |
� | -0.194 |
ane | -0.193 |
Token | give |
Token ID | 1577 |
Feature activation | +9.28e-01 |
Pos logit contributions | |
never | +0.101 |
pretended | +0.097 |
forgot | +0.097 |
remembered | +0.093 |
tried | +0.089 |
Neg logit contributions | |
arl | -0.261 |
Today | -0.241 |
ay | -0.239 |
ute | -0.236 |
� | -0.234 |
Token | Sam |
Token ID | 3409 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
punished | +0.010 |
pretended | +0.010 |
put | +0.010 |
never | +0.010 |
forgot | +0.010 |
Neg logit contributions | |
ay | -0.025 |
elly | -0.024 |
arl | -0.023 |
ummy | -0.023 |
ore | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
elly | +0.013 |
ay | +0.013 |
arl | +0.013 |
ane | +0.013 |
Today | +0.012 |
Neg logit contributions | |
pretended | -0.007 |
punished | -0.007 |
put | -0.007 |
forgot | -0.007 |
never | -0.007 |
Token | hug |
Token ID | 16225 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
put | +0.008 |
never | +0.008 |
used | +0.008 |
cleaned | +0.008 |
made | +0.007 |
Neg logit contributions | |
arl | -0.017 |
elly | -0.017 |
ay | -0.017 |
Today | -0.016 |
ane | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +7.09e+00 |
Pos logit contributions | |
arl | +0.074 |
ay | +0.070 |
elly | +0.070 |
Today | +0.068 |
ane | +0.068 |
Neg logit contributions | |
put | -0.052 |
never | -0.050 |
pretended | -0.049 |
learned | -0.048 |
used | -0.048 |
Token | do |
Token ID | 466 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
pretended | +0.052 |
hoped | +0.050 |
put | +0.049 |
learned | +0.048 |
cleaned | +0.047 |
Neg logit contributions | |
arl | -0.123 |
ay | -0.118 |
elly | -0.117 |
Today | -0.117 |
ute | -0.114 |
Token | it |
Token ID | 340 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
pretended | +0.008 |
put | +0.007 |
cleaned | +0.007 |
hoped | +0.007 |
used | +0.007 |
Neg logit contributions | |
arl | -0.012 |
elly | -0.011 |
ay | -0.011 |
ane | -0.011 |
ute | -0.011 |
Token | again |
Token ID | 757 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
arl | +0.018 |
ay | +0.018 |
elly | +0.017 |
ane | +0.017 |
Today | +0.017 |
Neg logit contributions | |
put | -0.014 |
never | -0.013 |
pretended | -0.013 |
learned | -0.013 |
used | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
arl | +0.026 |
ay | +0.025 |
elly | +0.024 |
ane | +0.024 |
Today | +0.024 |
Neg logit contributions | |
put | -0.018 |
never | -0.018 |
pretended | -0.017 |
used | -0.017 |
learned | -0.017 |
Token | They |
Token ID | 1119 |
Feature activation | +8.51e+00 |
Pos logit contributions | |
ane | +0.066 |
ute | +0.066 |
andel | +0.065 |
ay | +0.065 |
owers | +0.064 |
Neg logit contributions | |
put | -0.055 |
never | -0.054 |
hoped | -0.054 |
learned | -0.053 |
pretended | -0.051 |
Token | also |
Token ID | 635 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
eling | +0.091 |
ts | +0.088 |
pretending | +0.087 |
supposed | +0.083 |
ck | +0.082 |
Neg logit contributions | |
� | -0.169 |
ay | -0.165 |
olly | -0.164 |
arl | -0.162 |
ane | -0.160 |
Token | said |
Token ID | 531 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
ts | +0.086 |
himself | +0.084 |
eling | +0.079 |
pretending | +0.077 |
myself | +0.076 |
Neg logit contributions | |
arl | -0.202 |
elly | -0.202 |
� | -0.197 |
ay | -0.193 |
arry | -0.191 |
Token | they |
Token ID | 484 |
Feature activation | +6.08e+00 |
Pos logit contributions | |
arl | +0.008 |
ay | +0.008 |
elly | +0.008 |
Today | +0.008 |
ane | +0.008 |
Neg logit contributions | |
put | -0.004 |
never | -0.004 |
pretended | -0.004 |
learned | -0.004 |
made | -0.004 |
Token | were |
Token ID | 547 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
ck | +0.103 |
pretended | +0.090 |
eling | +0.090 |
punished | +0.089 |
ts | +0.088 |
Neg logit contributions | |
ane | -0.110 |
ay | -0.109 |
� | -0.107 |
elly | -0.106 |
owers | -0.106 |
Token | sorry |
Token ID | 7926 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
pretended | +0.068 |
imagined | +0.066 |
wed | +0.064 |
forgot | +0.063 |
put | +0.062 |
Neg logit contributions | |
arl | -0.064 |
ay | -0.052 |
ummy | -0.051 |
apping | -0.051 |
elly | -0.051 |
Token | for |
Token ID | 329 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
arl | +0.038 |
ay | +0.036 |
elly | +0.036 |
ane | +0.035 |
Today | +0.035 |
Neg logit contributions | |
put | -0.024 |
never | -0.023 |
pretended | -0.023 |
learned | -0.022 |
used | -0.022 |
Token | taped |
Token ID | 36114 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
yours | +0.069 |
eling | +0.067 |
permission | +0.067 |
ck | +0.066 |
myself | +0.065 |
Neg logit contributions | |
� | -0.173 |
ane | -0.162 |
owers | -0.161 |
op | -0.159 |
arl | -0.158 |
Token | their |
Token ID | 511 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
arl | +0.018 |
ay | +0.017 |
elly | +0.017 |
Today | +0.017 |
ane | +0.017 |
Neg logit contributions | |
put | -0.008 |
pretended | -0.007 |
never | -0.007 |
learned | -0.007 |
used | -0.007 |
Token | books |
Token ID | 3835 |
Feature activation | -4.95e+00 |
Pos logit contributions | |
arl | +0.000 |
ay | +0.000 |
elly | +0.000 |
Today | +0.000 |
ane | +0.000 |
Neg logit contributions | |
put | -0.000 |
never | -0.000 |
pretended | -0.000 |
used | -0.000 |
learned | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -3.23e+00 |
Pos logit contributions | |
arl | +0.120 |
elly | +0.117 |
ay | +0.113 |
Today | +0.112 |
arry | +0.111 |
Neg logit contributions | |
put | -0.080 |
never | -0.078 |
used | -0.074 |
made | -0.074 |
pretended | -0.073 |
Token | They |
Token ID | 1119 |
Feature activation | +8.40e+00 |
Pos logit contributions | |
ane | +0.073 |
ay | +0.070 |
ute | +0.068 |
arl | +0.067 |
owers | +0.066 |
Neg logit contributions | |
put | -0.060 |
pretended | -0.057 |
never | -0.056 |
hoped | -0.056 |
learned | -0.056 |
Token | also |
Token ID | 635 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
permission | +0.101 |
yours | +0.089 |
asking | +0.088 |
finding | +0.085 |
ck | +0.083 |
Neg logit contributions | |
� | -0.173 |
he | -0.159 |
owers | -0.157 |
ane | -0.155 |
ay | -0.153 |
Token | put |
Token ID | 1234 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
himself | +0.078 |
permission | +0.067 |
ck | +0.067 |
sure | +0.066 |
myself | +0.064 |
Neg logit contributions | |
arl | -0.214 |
elly | -0.210 |
� | -0.203 |
ay | -0.199 |
oms | -0.195 |
Token | the |
Token ID | 262 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
forgot | +0.027 |
pretended | +0.027 |
put | +0.027 |
never | +0.027 |
remembered | +0.026 |
Neg logit contributions | |
arl | -0.051 |
ay | -0.051 |
elly | -0.047 |
ane | -0.047 |
� | -0.047 |
Token | par |
Token ID | 1582 |
Feature activation | -9.19e+00 |
Pos logit contributions | |
arl | +0.056 |
ay | +0.054 |
elly | +0.053 |
Today | +0.052 |
ane | +0.052 |
Neg logit contributions | |
put | -0.038 |
never | -0.037 |
pretended | -0.036 |
learned | -0.035 |
used | -0.035 |
Token | rot |
Token ID | 10599 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
arl | +0.097 |
Today | +0.096 |
elly | +0.094 |
ummy | +0.080 |
� | +0.080 |
Neg logit contributions | |
put | -0.264 |
never | -0.262 |
used | -0.255 |
pretended | -0.254 |
learned | -0.254 |
Token | in |
Token ID | 287 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
arl | +0.065 |
ay | +0.063 |
elly | +0.062 |
ane | +0.062 |
ute | +0.061 |
Neg logit contributions | |
put | -0.047 |
never | -0.046 |
punished | -0.045 |
pretended | -0.045 |
learned | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
pretended | +0.006 |
put | +0.006 |
never | +0.006 |
remembered | +0.006 |
hoped | +0.006 |
Neg logit contributions | |
arl | -0.007 |
ay | -0.007 |
elly | -0.006 |
ore | -0.006 |
Today | -0.006 |
Token | bed |
Token ID | 3996 |
Feature activation | -5.59e+00 |
Pos logit contributions | |
pretended | +0.022 |
never | +0.021 |
remembered | +0.021 |
made | +0.021 |
hoped | +0.021 |
Neg logit contributions | |
arl | -0.049 |
Today | -0.046 |
ay | -0.045 |
� | -0.045 |
elly | -0.045 |
Token | early |
Token ID | 1903 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
arl | +0.133 |
ay | +0.131 |
elly | +0.128 |
ane | +0.126 |
Today | +0.126 |
Neg logit contributions | |
put | -0.094 |
never | -0.089 |
punished | -0.085 |
learned | -0.085 |
pretended | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
arl | +0.082 |
ay | +0.079 |
elly | +0.078 |
ane | +0.076 |
arry | +0.075 |
Neg logit contributions | |
put | -0.058 |
never | -0.055 |
pretended | -0.054 |
used | -0.053 |
made | -0.053 |
Token | They |
Token ID | 1119 |
Feature activation | +9.00e+00 |
Pos logit contributions | |
ane | +0.054 |
ay | +0.053 |
ute | +0.052 |
owers | +0.052 |
anger | +0.051 |
Neg logit contributions | |
put | -0.046 |
learned | -0.044 |
hoped | -0.044 |
never | -0.044 |
pretended | -0.042 |
Token | also |
Token ID | 635 |
Feature activation | +9.98e+00 |
Pos logit contributions | |
ck | +0.106 |
ts | +0.105 |
ps | +0.104 |
eling | +0.102 |
myself | +0.101 |
Neg logit contributions | |
arl | -0.173 |
� | -0.164 |
ay | -0.162 |
arry | -0.156 |
elly | -0.154 |
Token | said |
Token ID | 531 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
himself | +0.090 |
ck | +0.086 |
ts | +0.085 |
herself | +0.081 |
myself | +0.080 |
Neg logit contributions | |
arl | -0.220 |
elly | -0.203 |
ay | -0.186 |
oky | -0.185 |
� | -0.184 |
Token | they |
Token ID | 484 |
Feature activation | +6.12e+00 |
Pos logit contributions | |
arl | +0.010 |
ay | +0.010 |
elly | +0.010 |
ute | +0.009 |
ane | +0.009 |
Neg logit contributions | |
put | -0.005 |
never | -0.005 |
pretended | -0.005 |
learned | -0.005 |
made | -0.005 |
Token | had |
Token ID | 550 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
ck | +0.104 |
pretended | +0.089 |
eling | +0.088 |
forth | +0.088 |
ts | +0.086 |
Neg logit contributions | |
ay | -0.113 |
ane | -0.111 |
elly | -0.109 |
� | -0.106 |
Today | -0.104 |
Token | to |
Token ID | 284 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
pretended | +0.063 |
ck | +0.061 |
put | +0.060 |
used | +0.057 |
hoped | +0.057 |
Neg logit contributions | |
arl | -0.059 |
elly | -0.059 |
ay | -0.057 |
Today | -0.054 |
ummy | -0.053 |
Token | be |
Token ID | 307 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
pretended | +0.056 |
hoped | +0.052 |
remembered | +0.050 |
used | +0.050 |
learned | +0.049 |
Neg logit contributions | |
arl | -0.116 |
ay | -0.109 |
Today | -0.108 |
ute | -0.105 |
� | -0.103 |
Token | clean |
Token ID | 3424 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
pretended | +0.013 |
remembered | +0.012 |
never | +0.012 |
forgot | +0.012 |
hoped | +0.012 |
Neg logit contributions | |
arl | -0.016 |
ay | -0.015 |
Today | -0.015 |
ute | -0.015 |
elly | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
arl | +0.060 |
ay | +0.059 |
elly | +0.059 |
Today | +0.057 |
ute | +0.056 |
Neg logit contributions | |
put | -0.038 |
never | -0.038 |
pretended | -0.037 |
learned | -0.036 |
used | -0.036 |
Token | mess |
Token ID | 2085 |
Feature activation | -4.77e+00 |
Pos logit contributions | |
arl | +0.073 |
ay | +0.070 |
elly | +0.070 |
Today | +0.068 |
ute | +0.067 |
Neg logit contributions | |
put | -0.058 |
never | -0.057 |
pretended | -0.057 |
hoped | -0.055 |
used | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
arl | +0.099 |
ay | +0.095 |
elly | +0.094 |
ane | +0.090 |
Today | +0.090 |
Neg logit contributions | |
put | -0.088 |
pretended | -0.087 |
never | -0.086 |
learned | -0.084 |
remembered | -0.084 |
Token | They |
Token ID | 1119 |
Feature activation | +8.91e+00 |
Pos logit contributions | |
ane | +0.050 |
arl | +0.050 |
ay | +0.050 |
igsaw | +0.049 |
� | +0.049 |
Neg logit contributions | |
put | -0.058 |
pretended | -0.055 |
hoped | -0.055 |
made | -0.055 |
never | -0.054 |
Token | also |
Token ID | 635 |
Feature activation | +9.96e+00 |
Pos logit contributions | |
pretended | +0.095 |
remembered | +0.095 |
able | +0.092 |
ck | +0.088 |
imagined | +0.088 |
Neg logit contributions | |
arl | -0.222 |
ay | -0.202 |
oms | -0.189 |
� | -0.189 |
Today | -0.186 |
Token | tell |
Token ID | 1560 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
pretended | +0.096 |
himself | +0.087 |
ck | +0.086 |
never | +0.083 |
imagined | +0.083 |
Neg logit contributions | |
arl | -0.257 |
ay | -0.225 |
oms | -0.221 |
Today | -0.218 |
elly | -0.215 |
Token | them |
Token ID | 606 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
pretended | +0.019 |
put | +0.019 |
punished | +0.019 |
used | +0.019 |
never | +0.018 |
Neg logit contributions | |
arl | -0.031 |
elly | -0.029 |
ay | -0.029 |
Today | -0.028 |
andel | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | +4.17e+00 |
Pos logit contributions | |
pretended | +0.011 |
put | +0.011 |
punished | +0.011 |
used | +0.010 |
remembered | +0.010 |
Neg logit contributions | |
arl | -0.012 |
elly | -0.012 |
ay | -0.012 |
Today | -0.012 |
ane | -0.012 |
Token | ask |
Token ID | 1265 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
pretended | +0.052 |
remembered | +0.046 |
forgot | +0.044 |
hoped | +0.044 |
used | +0.043 |
Neg logit contributions | |
arl | -0.114 |
ay | -0.104 |
ute | -0.099 |
Today | -0.098 |
� | -0.097 |
Token | before |
Token ID | 878 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
put | +0.007 |
pretended | +0.007 |
never | +0.007 |
punished | +0.007 |
used | +0.007 |
Neg logit contributions | |
arl | -0.014 |
elly | -0.014 |
ay | -0.014 |
Today | -0.013 |
ute | -0.013 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
punished | +0.018 |
pretended | +0.018 |
forgot | +0.017 |
managed | +0.017 |
remembered | +0.017 |
Neg logit contributions | |
arl | -0.030 |
ay | -0.028 |
Today | -0.028 |
ummy | -0.027 |
ute | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +5.41e+00 |
Pos logit contributions | |
arl | +0.032 |
elly | +0.030 |
ay | +0.030 |
ane | +0.029 |
Today | +0.029 |
Neg logit contributions | |
pretended | -0.029 |
put | -0.028 |
never | -0.028 |
punished | -0.028 |
used | -0.028 |
Token | dad |
Token ID | 9955 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
pretended | +0.104 |
ck | +0.104 |
ts | +0.097 |
eling | +0.094 |
never | +0.093 |
Neg logit contributions | |
arl | -0.104 |
elly | -0.087 |
ay | -0.084 |
ummy | -0.083 |
apping | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
pretended | +0.013 |
put | +0.013 |
punished | +0.013 |
used | +0.013 |
never | +0.012 |
Neg logit contributions | |
arl | -0.013 |
elly | -0.012 |
ay | -0.012 |
ane | -0.012 |
Today | -0.012 |
Token | They |
Token ID | 1119 |
Feature activation | +8.71e+00 |
Pos logit contributions | |
igsaw | +0.069 |
ane | +0.067 |
ay | +0.065 |
� | +0.064 |
arl | +0.064 |
Neg logit contributions | |
put | -0.060 |
hoped | -0.058 |
never | -0.057 |
learned | -0.057 |
made | -0.056 |
Token | also |
Token ID | 635 |
Feature activation | +9.94e+00 |
Pos logit contributions | |
pretended | +0.105 |
ck | +0.098 |
ts | +0.098 |
remembered | +0.095 |
imagined | +0.093 |
Neg logit contributions | |
arl | -0.211 |
ay | -0.195 |
arry | -0.186 |
Today | -0.185 |
� | -0.183 |
Token | say |
Token ID | 910 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
pretended | +0.105 |
forgot | +0.096 |
ck | +0.095 |
ts | +0.094 |
never | +0.091 |
Neg logit contributions | |
arl | -0.246 |
ay | -0.219 |
Today | -0.216 |
elly | -0.207 |
arry | -0.202 |
Token | they |
Token ID | 484 |
Feature activation | +5.95e+00 |
Pos logit contributions | |
elly | +0.010 |
arl | +0.010 |
ay | +0.010 |
Today | +0.010 |
arry | +0.010 |
Neg logit contributions | |
put | -0.005 |
never | -0.005 |
learned | -0.005 |
made | -0.005 |
pretended | -0.005 |
Token | love |
Token ID | 1842 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
ck | +0.096 |
pretended | +0.096 |
put | +0.086 |
his | +0.084 |
imagined | +0.084 |
Neg logit contributions | |
arl | -0.118 |
ay | -0.118 |
ane | -0.113 |
elly | -0.111 |
Today | -0.109 |
Token | them |
Token ID | 606 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
arl | +0.015 |
elly | +0.014 |
ay | +0.014 |
ummy | +0.014 |
Today | +0.013 |
Neg logit contributions | |
punished | -0.009 |
pretended | -0.009 |
forgot | -0.009 |
remembered | -0.008 |
put | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
arl | +0.055 |
ay | +0.054 |
elly | +0.053 |
Today | +0.052 |
ane | +0.052 |
Neg logit contributions | |
put | -0.038 |
pretended | -0.037 |
never | -0.036 |
punished | -0.036 |
used | -0.035 |
Token | away |
Token ID | 1497 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
arl | +0.026 |
ay | +0.025 |
elly | +0.024 |
Today | +0.024 |
ane | +0.023 |
Neg logit contributions | |
put | -0.016 |
never | -0.016 |
pretended | -0.016 |
learned | -0.016 |
hoped | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
arl | +0.055 |
ay | +0.053 |
ane | +0.052 |
elly | +0.052 |
ute | +0.051 |
Neg logit contributions | |
put | -0.034 |
never | -0.032 |
used | -0.031 |
made | -0.031 |
pretended | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
ane | +0.069 |
ute | +0.066 |
opus | +0.065 |
andel | +0.065 |
ay | +0.064 |
Neg logit contributions | |
put | -0.048 |
never | -0.045 |
hoped | -0.043 |
cleaned | -0.042 |
learned | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
arl | +0.074 |
ane | +0.074 |
owers | +0.073 |
elly | +0.072 |
umm | +0.071 |
Neg logit contributions | |
put | -0.081 |
never | -0.079 |
learned | -0.076 |
used | -0.074 |
cleaned | -0.073 |
Token | They |
Token ID | 2990 |
Feature activation | +8.22e+00 |
Pos logit contributions | |
arl | +0.063 |
ay | +0.061 |
elly | +0.060 |
ane | +0.060 |
Today | +0.059 |
Neg logit contributions | |
put | -0.047 |
never | -0.046 |
learned | -0.044 |
used | -0.043 |
made | -0.043 |
Token | also |
Token ID | 635 |
Feature activation | +9.94e+00 |
Pos logit contributions | |
remembered | +0.099 |
pretended | +0.097 |
punished | +0.092 |
hoped | +0.090 |
used | +0.090 |
Neg logit contributions | |
arl | -0.202 |
� | -0.188 |
ay | -0.188 |
Today | -0.183 |
ute | -0.173 |
Token | do |
Token ID | 466 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
never | +0.100 |
pretended | +0.099 |
ck | +0.095 |
stayed | +0.094 |
forgot | +0.094 |
Neg logit contributions | |
arl | -0.252 |
Today | -0.230 |
ay | -0.226 |
� | -0.226 |
oms | -0.219 |
Token | some |
Token ID | 617 |
Feature activation | +8.96e-01 |
Pos logit contributions | |
pretended | +0.019 |
remembered | +0.017 |
punished | +0.017 |
wished | +0.016 |
forgot | +0.016 |
Neg logit contributions | |
arl | -0.056 |
Today | -0.053 |
� | -0.052 |
ute | -0.052 |
ay | -0.052 |
Token | chores |
Token ID | 46173 |
Feature activation | -3.75e+00 |
Pos logit contributions | |
never | +0.013 |
pretended | +0.013 |
put | +0.013 |
hoped | +0.012 |
learned | +0.012 |
Neg logit contributions | |
arl | -0.022 |
ay | -0.021 |
elly | -0.021 |
Today | -0.021 |
� | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
arl | +0.085 |
ay | +0.082 |
elly | +0.081 |
ane | +0.079 |
Today | +0.078 |
Neg logit contributions | |
put | -0.064 |
pretended | -0.062 |
never | -0.061 |
learned | -0.059 |
punished | -0.059 |
Token | They |
Token ID | 1119 |
Feature activation | +8.33e+00 |
Pos logit contributions | |
ane | +0.063 |
ute | +0.060 |
arl | +0.059 |
opus | +0.059 |
ay | +0.058 |
Neg logit contributions | |
put | -0.045 |
never | -0.041 |
hoped | -0.040 |
cleaned | -0.039 |
pretended | -0.039 |
Token | belong |
Token ID | 5594 |
Feature activation | -9.67e-01 |
Pos logit contributions | |
pretended | +0.071 |
hoped | +0.067 |
remembered | +0.065 |
learned | +0.065 |
punished | +0.065 |
Neg logit contributions | |
arl | -0.115 |
ay | -0.110 |
Today | -0.105 |
arry | -0.103 |
ute | -0.103 |
Token | to |
Token ID | 284 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
arl | +0.020 |
elly | +0.020 |
ay | +0.019 |
ane | +0.019 |
Today | +0.018 |
Neg logit contributions | |
put | -0.018 |
pretended | -0.017 |
never | -0.017 |
learned | -0.016 |
punished | -0.016 |
Token | us |
Token ID | 514 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
put | +0.018 |
never | +0.017 |
pretended | +0.017 |
used | +0.016 |
learned | +0.016 |
Neg logit contributions | |
arl | -0.042 |
ay | -0.040 |
elly | -0.040 |
Today | -0.039 |
ane | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
arl | +0.009 |
ay | +0.009 |
elly | +0.009 |
ane | +0.009 |
Today | +0.009 |
Neg logit contributions | |
put | -0.007 |
never | -0.007 |
pretended | -0.007 |
used | -0.007 |
learned | -0.007 |
Token | They |
Token ID | 1119 |
Feature activation | +8.71e+00 |
Pos logit contributions | |
arl | +0.057 |
ute | +0.055 |
andel | +0.054 |
ane | +0.054 |
anger | +0.054 |
Neg logit contributions | |
never | -0.055 |
put | -0.054 |
learned | -0.053 |
pretended | -0.051 |
used | -0.051 |
Token | also |
Token ID | 635 |
Feature activation | +9.94e+00 |
Pos logit contributions | |
ts | +0.086 |
pretended | +0.084 |
ck | +0.080 |
wed | +0.079 |
tightened | +0.077 |
Neg logit contributions | |
ane | -0.190 |
ay | -0.187 |
arry | -0.184 |
Today | -0.183 |
arl | -0.180 |
Token | told |
Token ID | 1297 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ts | +0.082 |
pretended | +0.081 |
ck | +0.072 |
wed | +0.062 |
counted | +0.061 |
Neg logit contributions | |
arl | -0.250 |
Today | -0.239 |
arry | -0.237 |
� | -0.233 |
ay | -0.232 |
Token | him |
Token ID | 683 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
put | +0.013 |
pretended | +0.012 |
used | +0.012 |
never | +0.012 |
hoped | +0.012 |
Neg logit contributions | |
arl | -0.033 |
ay | -0.032 |
elly | -0.031 |
Today | -0.031 |
ummy | -0.031 |
Token | that |
Token ID | 326 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
pretended | +0.025 |
put | +0.025 |
punished | +0.024 |
used | +0.023 |
cleaned | +0.023 |
Neg logit contributions | |
arl | -0.031 |
elly | -0.030 |
ane | -0.030 |
ay | -0.030 |
Today | -0.030 |
Token | it |
Token ID | 340 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
put | +0.039 |
pretended | +0.038 |
punished | +0.037 |
never | +0.036 |
learned | +0.036 |
Neg logit contributions | |
arl | -0.075 |
elly | -0.074 |
ay | -0.073 |
Today | -0.073 |
ane | -0.073 |
Token | 's |
Token ID | 338 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
pretended | +0.052 |
put | +0.051 |
ck | +0.048 |
hoped | +0.047 |
punished | +0.047 |
Neg logit contributions | |
ane | -0.075 |
elly | -0.075 |
ay | -0.074 |
Today | -0.073 |
arl | -0.072 |
Token | for |
Token ID | 329 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
arl | +0.054 |
elly | +0.053 |
ay | +0.053 |
ane | +0.051 |
� | +0.051 |
Neg logit contributions | |
put | -0.030 |
never | -0.028 |
pretended | -0.027 |
learned | -0.027 |
used | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
forgot | +0.030 |
pretended | +0.030 |
imagined | +0.026 |
remembered | +0.026 |
counted | +0.025 |
Neg logit contributions | |
arl | -0.044 |
� | -0.037 |
ay | -0.036 |
apping | -0.036 |
uttering | -0.035 |
Token | surprise |
Token ID | 5975 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
arl | +0.017 |
ay | +0.016 |
elly | +0.016 |
ane | +0.016 |
Today | +0.016 |
Neg logit contributions | |
put | -0.012 |
never | -0.012 |
pretended | -0.011 |
used | -0.011 |
learned | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
ay | +0.072 |
arl | +0.071 |
elly | +0.070 |
Today | +0.068 |
ane | +0.068 |
Neg logit contributions | |
put | -0.074 |
pretended | -0.073 |
punished | -0.073 |
never | -0.072 |
learned | -0.071 |
Token | They |
Token ID | 1119 |
Feature activation | +8.69e+00 |
Pos logit contributions | |
ay | +0.068 |
andel | +0.067 |
ane | +0.067 |
anger | +0.066 |
ute | +0.066 |
Neg logit contributions | |
put | -0.059 |
hoped | -0.058 |
never | -0.058 |
learned | -0.055 |
pretended | -0.054 |
Token | also |
Token ID | 635 |
Feature activation | +9.92e+00 |
Pos logit contributions | |
eling | +0.086 |
ts | +0.085 |
ck | +0.082 |
permission | +0.081 |
wed | +0.079 |
Neg logit contributions | |
� | -0.185 |
ay | -0.178 |
arl | -0.176 |
arry | -0.173 |
ane | -0.168 |
Token | said |
Token ID | 531 |
Feature activation | -8.89e-01 |
Pos logit contributions | |
ts | +0.074 |
eling | +0.071 |
ck | +0.071 |
himself | +0.069 |
permission | +0.065 |
Neg logit contributions | |
arl | -0.214 |
elly | -0.206 |
ay | -0.205 |
� | -0.202 |
arry | -0.198 |
Token | they |
Token ID | 484 |
Feature activation | +5.89e+00 |
Pos logit contributions | |
arl | +0.022 |
elly | +0.022 |
ay | +0.021 |
Today | +0.021 |
arry | +0.021 |
Neg logit contributions | |
put | -0.014 |
never | -0.014 |
learned | -0.013 |
made | -0.013 |
pretended | -0.013 |
Token | were |
Token ID | 547 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
ck | +0.108 |
forth | +0.097 |
eling | +0.095 |
punished | +0.095 |
asking | +0.094 |
Neg logit contributions | |
ay | -0.096 |
� | -0.094 |
owers | -0.092 |
ane | -0.092 |
elly | -0.090 |
Token | sorry |
Token ID | 7926 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
pretended | +0.061 |
imagined | +0.059 |
wed | +0.059 |
forgot | +0.057 |
tried | +0.056 |
Neg logit contributions | |
arl | -0.059 |
ay | -0.051 |
apping | -0.049 |
ummy | -0.049 |
ane | -0.047 |
Token | for |
Token ID | 329 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
arl | +0.042 |
ay | +0.041 |
elly | +0.041 |
ane | +0.040 |
Today | +0.039 |
Neg logit contributions | |
put | -0.027 |
never | -0.026 |
pretended | -0.025 |
learned | -0.024 |
used | -0.024 |
Token | for |
Token ID | 329 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
arl | +0.028 |
ay | +0.028 |
elly | +0.028 |
� | +0.027 |
ane | +0.027 |
Neg logit contributions | |
put | -0.016 |
never | -0.015 |
learned | -0.014 |
pretended | -0.014 |
used | -0.014 |
Token | his |
Token ID | 465 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
pretended | +0.029 |
forgot | +0.028 |
imagined | +0.025 |
remembered | +0.025 |
counted | +0.024 |
Neg logit contributions | |
arl | -0.045 |
� | -0.037 |
ay | -0.036 |
apping | -0.036 |
ore | -0.036 |
Token | book |
Token ID | 1492 |
Feature activation | -6.69e+00 |
Pos logit contributions | |
put | +0.014 |
never | +0.013 |
pretended | +0.013 |
learned | +0.013 |
used | +0.013 |
Neg logit contributions | |
arl | -0.023 |
ay | -0.022 |
elly | -0.022 |
ane | -0.022 |
Today | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
arl | +0.154 |
ay | +0.148 |
elly | +0.147 |
Today | +0.143 |
ane | +0.142 |
Neg logit contributions | |
put | -0.114 |
never | -0.110 |
pretended | -0.107 |
learned | -0.106 |
used | -0.106 |
Token | They |
Token ID | 1119 |
Feature activation | +8.65e+00 |
Pos logit contributions | |
ay | +0.057 |
arl | +0.056 |
ane | +0.055 |
ute | +0.054 |
andel | +0.054 |
Neg logit contributions | |
put | -0.058 |
never | -0.057 |
hoped | -0.056 |
learned | -0.055 |
pretended | -0.055 |
Token | also |
Token ID | 635 |
Feature activation | +9.91e+00 |
Pos logit contributions | |
ck | +0.070 |
wed | +0.068 |
eling | +0.068 |
ts | +0.068 |
pretended | +0.064 |
Neg logit contributions | |
� | -0.199 |
arl | -0.198 |
arry | -0.194 |
ay | -0.188 |
ane | -0.187 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
ck | +0.064 |
ts | +0.063 |
pretended | +0.060 |
sure | +0.058 |
herself | +0.056 |
Neg logit contributions | |
arl | -0.235 |
� | -0.223 |
arry | -0.217 |
elly | -0.217 |
ay | -0.215 |
Token | to |
Token ID | 284 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
pretended | +0.030 |
forgot | +0.029 |
put | +0.029 |
remembered | +0.029 |
punished | +0.029 |
Neg logit contributions | |
arl | -0.025 |
elly | -0.025 |
ay | -0.025 |
ummy | -0.024 |
Today | -0.023 |
Token | record |
Token ID | 1700 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
pretended | +0.042 |
remembered | +0.039 |
cleaned | +0.039 |
hoped | +0.038 |
punished | +0.038 |
Neg logit contributions | |
arl | -0.098 |
ay | -0.095 |
� | -0.092 |
Today | -0.091 |
ute | -0.091 |
Token | more |
Token ID | 517 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
arl | +0.055 |
ay | +0.054 |
elly | +0.053 |
Today | +0.052 |
ane | +0.052 |
Neg logit contributions | |
put | -0.025 |
never | -0.024 |
pretended | -0.024 |
used | -0.023 |
punished | -0.023 |
Token | things |
Token ID | 1243 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
arl | +0.057 |
� | +0.054 |
elly | +0.053 |
ay | +0.053 |
ane | +0.052 |
Neg logit contributions | |
put | -0.034 |
pretended | -0.031 |
never | -0.031 |
learned | -0.031 |
made | -0.031 |
Token | while |
Token ID | 981 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
put | +0.023 |
never | +0.022 |
learned | +0.021 |
used | +0.021 |
pretended | +0.021 |
Neg logit contributions | |
arl | -0.047 |
ay | -0.045 |
elly | -0.045 |
ane | -0.044 |
Today | -0.044 |
Token | they |
Token ID | 484 |
Feature activation | +6.45e+00 |
Pos logit contributions | |
arl | +0.010 |
ay | +0.010 |
elly | +0.010 |
Today | +0.009 |
ane | +0.009 |
Neg logit contributions | |
put | -0.006 |
never | -0.006 |
pretended | -0.006 |
learned | -0.006 |
used | -0.006 |
Token | cleaned |
Token ID | 20750 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
punished | +0.053 |
ck | +0.052 |
never | +0.052 |
pretended | +0.049 |
wounds | +0.049 |
Neg logit contributions | |
ay | -0.171 |
� | -0.171 |
owers | -0.161 |
arry | -0.158 |
ane | -0.156 |
Token | . |
Token ID | 13 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
arl | +0.008 |
ay | +0.008 |
elly | +0.008 |
Today | +0.008 |
ane | +0.008 |
Neg logit contributions | |
put | -0.005 |
never | -0.004 |
pretended | -0.004 |
learned | -0.004 |
used | -0.004 |
Token | They |
Token ID | 1119 |
Feature activation | +8.60e+00 |
Pos logit contributions | |
arl | +0.053 |
ane | +0.051 |
ay | +0.050 |
ute | +0.050 |
owers | +0.049 |
Neg logit contributions | |
put | -0.056 |
never | -0.055 |
learned | -0.054 |
hoped | -0.053 |
used | -0.052 |
Token | also |
Token ID | 635 |
Feature activation | +9.91e+00 |
Pos logit contributions | |
ck | +0.099 |
eling | +0.093 |
permission | +0.092 |
wounds | +0.089 |
asking | +0.083 |
Neg logit contributions | |
� | -0.185 |
ay | -0.181 |
ane | -0.172 |
arry | -0.168 |
owers | -0.167 |
Token | learned |
Token ID | 4499 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
ck | +0.092 |
eling | +0.078 |
himself | +0.070 |
forth | +0.069 |
ts | +0.068 |
Neg logit contributions | |
arl | -0.218 |
ay | -0.215 |
elly | -0.210 |
� | -0.208 |
igsaw | -0.193 |
Token | to |
Token ID | 284 |
Feature activation | +4.80e+00 |
Pos logit contributions | |
put | +0.027 |
never | +0.026 |
pretended | +0.025 |
learned | +0.025 |
used | +0.025 |
Neg logit contributions | |
arl | -0.048 |
ay | -0.046 |
elly | -0.046 |
Today | -0.045 |
ane | -0.045 |
Token | respect |
Token ID | 2461 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
pretended | +0.060 |
hoped | +0.054 |
forgot | +0.053 |
punished | +0.052 |
remembered | +0.052 |
Neg logit contributions | |
arl | -0.131 |
Today | -0.120 |
ay | -0.120 |
arry | -0.118 |
ute | -0.116 |
Token | their |
Token ID | 511 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
arl | +0.016 |
ay | +0.016 |
elly | +0.016 |
Today | +0.015 |
ute | +0.015 |
Neg logit contributions | |
pretended | -0.007 |
put | -0.007 |
never | -0.007 |
forgot | -0.007 |
punished | -0.007 |
Token | toys |
Token ID | 14958 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
arl | +0.022 |
ay | +0.021 |
elly | +0.021 |
Today | +0.020 |
ane | +0.020 |
Neg logit contributions | |
put | -0.014 |
never | -0.013 |
pretended | -0.013 |
used | -0.013 |
learned | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
arl | +0.045 |
ay | +0.042 |
elly | +0.042 |
Today | +0.041 |
ane | +0.041 |
Neg logit contributions | |
put | -0.035 |
pretended | -0.034 |
never | -0.033 |
used | -0.033 |
learned | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
pretended | +0.018 |
put | +0.018 |
used | +0.017 |
punished | +0.017 |
cleaned | +0.017 |
Neg logit contributions | |
arl | -0.040 |
ay | -0.038 |
elly | -0.037 |
Today | -0.037 |
ummy | -0.036 |
Token | man |
Token ID | 582 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
arl | +0.038 |
ay | +0.036 |
elly | +0.036 |
ane | +0.035 |
Today | +0.035 |
Neg logit contributions | |
put | -0.033 |
never | -0.032 |
pretended | -0.031 |
learned | -0.031 |
used | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
arl | +0.049 |
ay | +0.047 |
elly | +0.047 |
Today | +0.046 |
ane | +0.046 |
Neg logit contributions | |
put | -0.038 |
pretended | -0.036 |
never | -0.036 |
used | -0.035 |
learned | -0.035 |
Token | They |
Token ID | 1119 |
Feature activation | +8.89e+00 |
Pos logit contributions | |
ay | +0.063 |
ane | +0.061 |
arl | +0.061 |
ute | +0.060 |
owers | +0.060 |
Neg logit contributions | |
never | -0.065 |
put | -0.064 |
hoped | -0.062 |
learned | -0.062 |
used | -0.059 |
Token | also |
Token ID | 635 |
Feature activation | +9.89e+00 |
Pos logit contributions | |
myself | +0.089 |
pretending | +0.084 |
forth | +0.083 |
pretended | +0.082 |
yours | +0.080 |
Neg logit contributions | |
arl | -0.184 |
arry | -0.166 |
� | -0.160 |
ay | -0.159 |
ari | -0.152 |
Token | had |
Token ID | 550 |
Feature activation | +4.79e+00 |
Pos logit contributions | |
pretended | +0.082 |
herself | +0.079 |
myself | +0.078 |
himself | +0.075 |
forth | +0.075 |
Neg logit contributions | |
arl | -0.214 |
elly | -0.196 |
arry | -0.189 |
ay | -0.180 |
ore | -0.178 |
Token | to |
Token ID | 284 |
Feature activation | +5.11e+00 |
Pos logit contributions | |
pretended | +0.081 |
ck | +0.075 |
forgot | +0.073 |
eling | +0.072 |
remember | +0.071 |
Neg logit contributions | |
arl | -0.089 |
ay | -0.083 |
elly | -0.081 |
aris | -0.077 |
� | -0.076 |
Token | work |
Token ID | 670 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
pretended | +0.068 |
forgot | +0.062 |
remembered | +0.062 |
hoped | +0.061 |
shared | +0.057 |
Neg logit contributions | |
arl | -0.136 |
ay | -0.123 |
Today | -0.121 |
ute | -0.120 |
arry | -0.117 |
Token | hard |
Token ID | 1327 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
arl | +0.043 |
ay | +0.041 |
elly | +0.041 |
Today | +0.040 |
ane | +0.040 |
Neg logit contributions | |
put | -0.028 |
never | -0.027 |
pretended | -0.027 |
learned | -0.026 |
used | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
arl | +0.057 |
ay | +0.055 |
elly | +0.054 |
ute | +0.053 |
ane | +0.053 |
Neg logit contributions | |
put | -0.035 |
never | -0.033 |
made | -0.033 |
used | -0.032 |
learned | -0.032 |
Token | five |
Token ID | 13261 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
arl | +0.029 |
ay | +0.027 |
elly | +0.027 |
Today | +0.027 |
ane | +0.026 |
Neg logit contributions | |
put | -0.028 |
never | -0.027 |
pretended | -0.027 |
learned | -0.027 |
used | -0.026 |
Token | their |
Token ID | 511 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
arl | +0.008 |
ay | +0.008 |
elly | +0.008 |
ane | +0.008 |
Today | +0.008 |
Neg logit contributions | |
put | -0.005 |
never | -0.004 |
pretended | -0.004 |
used | -0.004 |
punished | -0.004 |
Token | friends |
Token ID | 2460 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
arl | +0.026 |
elly | +0.025 |
ay | +0.024 |
ane | +0.024 |
Today | +0.024 |
Neg logit contributions | |
put | -0.032 |
never | -0.030 |
pretended | -0.030 |
learned | -0.030 |
used | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
arl | +0.005 |
ay | +0.005 |
elly | +0.005 |
ane | +0.005 |
Today | +0.005 |
Neg logit contributions | |
put | -0.004 |
never | -0.004 |
pretended | -0.004 |
used | -0.004 |
learned | -0.004 |
Token | They |
Token ID | 1119 |
Feature activation | +8.73e+00 |
Pos logit contributions | |
anger | +0.068 |
ane | +0.067 |
andel | +0.066 |
ute | +0.065 |
arl | +0.065 |
Neg logit contributions | |
learned | -0.049 |
put | -0.048 |
never | -0.046 |
hoped | -0.046 |
made | -0.045 |
Token | also |
Token ID | 635 |
Feature activation | +9.88e+00 |
Pos logit contributions | |
punished | +0.130 |
cleaned | +0.125 |
pretended | +0.121 |
remembered | +0.120 |
ck | +0.118 |
Neg logit contributions | |
ay | -0.188 |
arl | -0.186 |
ane | -0.181 |
� | -0.179 |
Today | -0.176 |
Token | hug |
Token ID | 16225 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
ck | +0.120 |
cleaned | +0.119 |
punished | +0.117 |
pretended | +0.113 |
never | +0.109 |
Neg logit contributions | |
arl | -0.226 |
ay | -0.222 |
Today | -0.216 |
ane | -0.214 |
� | -0.206 |
Token | and |
Token ID | 290 |
Feature activation | +7.31e+00 |
Pos logit contributions | |
arl | +0.011 |
elly | +0.010 |
ay | +0.010 |
ane | +0.010 |
� | +0.010 |
Neg logit contributions | |
put | -0.003 |
never | -0.002 |
pretended | -0.002 |
made | -0.002 |
learned | -0.002 |
Token | high |
Token ID | 1029 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
punished | +0.093 |
never | +0.090 |
pretended | +0.089 |
cleaned | +0.088 |
hoped | +0.088 |
Neg logit contributions | |
arl | -0.178 |
ay | -0.173 |
Today | -0.167 |
� | -0.164 |
igsaw | -0.163 |
Token | - |
Token ID | 12 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
arl | +0.098 |
ay | +0.092 |
� | +0.092 |
elly | +0.091 |
Today | +0.090 |
Neg logit contributions | |
put | -0.063 |
pretended | -0.059 |
made | -0.058 |
never | -0.057 |
used | -0.056 |
Token | five |
Token ID | 13261 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
arl | +0.034 |
elly | +0.032 |
ay | +0.032 |
Today | +0.031 |
� | +0.030 |
Neg logit contributions | |
put | -0.031 |
never | -0.031 |
pretended | -0.030 |
learned | -0.030 |
used | -0.029 |
Token | kissed |
Token ID | 27946 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
never | +0.081 |
wed | +0.076 |
ck | +0.076 |
punished | +0.072 |
eling | +0.070 |
Neg logit contributions | |
arl | -0.149 |
elly | -0.142 |
� | -0.140 |
ay | -0.136 |
Today | -0.134 |
Token | their |
Token ID | 511 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
arl | +0.047 |
elly | +0.046 |
ay | +0.045 |
ane | +0.045 |
Today | +0.044 |
Neg logit contributions | |
put | -0.019 |
never | -0.018 |
pretended | -0.017 |
learned | -0.017 |
made | -0.017 |
Token | dolls |
Token ID | 36062 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
arl | +0.055 |
ay | +0.053 |
elly | +0.051 |
Today | +0.051 |
ute | +0.050 |
Neg logit contributions | |
never | -0.033 |
put | -0.033 |
pretended | -0.032 |
used | -0.031 |
learned | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
arl | +0.083 |
ay | +0.079 |
elly | +0.078 |
Today | +0.076 |
ane | +0.076 |
Neg logit contributions | |
put | -0.066 |
never | -0.065 |
pretended | -0.064 |
punished | -0.063 |
used | -0.062 |
Token | They |
Token ID | 1119 |
Feature activation | +8.41e+00 |
Pos logit contributions | |
ane | +0.084 |
igsaw | +0.080 |
ay | +0.078 |
ute | +0.077 |
andel | +0.077 |
Neg logit contributions | |
put | -0.064 |
pretended | -0.060 |
hoped | -0.059 |
made | -0.058 |
learned | -0.057 |
Token | also |
Token ID | 635 |
Feature activation | +9.88e+00 |
Pos logit contributions | |
permission | +0.100 |
ck | +0.094 |
eling | +0.094 |
asking | +0.089 |
finding | +0.089 |
Neg logit contributions | |
� | -0.172 |
owers | -0.165 |
arl | -0.161 |
elly | -0.160 |
ay | -0.156 |
Token | hugged |
Token ID | 41130 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
permission | +0.082 |
himself | +0.080 |
never | +0.079 |
ck | +0.076 |
eling | +0.075 |
Neg logit contributions | |
elly | -0.207 |
arl | -0.203 |
owers | -0.194 |
� | -0.192 |
ay | -0.188 |
Token | and |
Token ID | 290 |
Feature activation | +6.74e+00 |
Pos logit contributions | |
arl | +0.006 |
ay | +0.006 |
elly | +0.006 |
ane | +0.006 |
Today | +0.006 |
Neg logit contributions | |
put | -0.004 |
never | -0.004 |
pretended | -0.004 |
learned | -0.003 |
used | -0.003 |
Token | kissed |
Token ID | 27946 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
never | +0.080 |
ck | +0.072 |
wed | +0.071 |
eling | +0.070 |
punished | +0.070 |
Neg logit contributions | |
arl | -0.149 |
elly | -0.139 |
� | -0.139 |
ay | -0.138 |
owers | -0.133 |
Token | each |
Token ID | 1123 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
arl | +0.049 |
ay | +0.048 |
elly | +0.048 |
Today | +0.046 |
ane | +0.046 |
Neg logit contributions | |
put | -0.021 |
never | -0.020 |
pretended | -0.020 |
learned | -0.019 |
used | -0.019 |
Token | other |
Token ID | 584 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
never | +0.006 |
pretended | +0.006 |
put | +0.006 |
punished | +0.006 |
forgot | +0.006 |
Neg logit contributions | |
arl | -0.014 |
ay | -0.014 |
elly | -0.014 |
Today | -0.014 |
ane | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
put | +0.024 |
never | +0.024 |
pretended | +0.024 |
punished | +0.023 |
forgot | +0.023 |
Neg logit contributions | |
arl | -0.055 |
ay | -0.054 |
elly | -0.054 |
ane | -0.053 |
ute | -0.052 |
Token | come |
Token ID | 1282 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
punished | +0.017 |
put | +0.017 |
pretended | +0.016 |
never | +0.016 |
learned | +0.016 |
Neg logit contributions | |
arl | -0.024 |
ay | -0.024 |
Today | -0.024 |
elly | -0.024 |
ane | -0.023 |
Token | out |
Token ID | 503 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
arl | +0.005 |
elly | +0.005 |
� | +0.005 |
ay | +0.005 |
igsaw | +0.005 |
Neg logit contributions | |
put | -0.003 |
never | -0.003 |
pretended | -0.002 |
used | -0.002 |
made | -0.002 |
Token | of |
Token ID | 286 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
arl | +0.050 |
elly | +0.047 |
ay | +0.047 |
ane | +0.046 |
� | +0.046 |
Neg logit contributions | |
put | -0.049 |
never | -0.047 |
pretended | -0.046 |
used | -0.046 |
made | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
forgot | +0.017 |
punished | +0.017 |
pretended | +0.016 |
learned | +0.016 |
never | +0.015 |
Neg logit contributions | |
arl | -0.035 |
ay | -0.033 |
Today | -0.032 |
� | -0.032 |
elly | -0.032 |
Token | seaw |
Token ID | 32085 |
Feature activation | -1.41e+01 |
Pos logit contributions | |
arl | +0.072 |
ay | +0.070 |
elly | +0.069 |
ane | +0.068 |
Today | +0.068 |
Neg logit contributions | |
put | -0.041 |
never | -0.037 |
made | -0.037 |
pretended | -0.037 |
used | -0.036 |
Token | eed |
Token ID | 2308 |
Feature activation | -6.53e+00 |
Pos logit contributions | |
Today | +0.216 |
arl | +0.209 |
ane | +0.205 |
ay | +0.200 |
owers | +0.195 |
Neg logit contributions | |
put | -0.335 |
never | -0.306 |
made | -0.304 |
learned | -0.303 |
used | -0.302 |
Token | and |
Token ID | 290 |
Feature activation | +6.43e+00 |
Pos logit contributions | |
arl | +0.159 |
ay | +0.152 |
elly | +0.150 |
Today | +0.147 |
ane | +0.145 |
Neg logit contributions | |
put | -0.106 |
never | -0.101 |
pretended | -0.099 |
made | -0.098 |
used | -0.097 |
Token | walk |
Token ID | 2513 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
punished | +0.083 |
wed | +0.077 |
never | +0.075 |
cleaned | +0.073 |
fixed | +0.067 |
Neg logit contributions | |
Today | -0.130 |
arl | -0.128 |
ore | -0.125 |
ifying | -0.124 |
oles | -0.122 |
Token | around |
Token ID | 1088 |
Feature activation | +8.02e-02 |
Pos logit contributions | |
arl | +0.068 |
ay | +0.065 |
elly | +0.065 |
Today | +0.063 |
ane | +0.063 |
Neg logit contributions | |
put | -0.047 |
never | -0.045 |
pretended | -0.044 |
used | -0.043 |
made | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
put | +0.001 |
never | +0.001 |
pretended | +0.001 |
used | +0.001 |
made | +0.001 |
Neg logit contributions | |
arl | -0.002 |
ay | -0.002 |
elly | -0.002 |
ane | -0.002 |
Today | -0.002 |
Token | started |
Token ID | 2067 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
remind | +0.040 |
ts | +0.036 |
make | +0.036 |
pretended | +0.036 |
share | +0.036 |
Neg logit contributions | |
elly | -0.087 |
arry | -0.087 |
ane | -0.086 |
ay | -0.086 |
oles | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
pretended | +0.054 |
put | +0.054 |
never | +0.053 |
learned | +0.052 |
punished | +0.052 |
Neg logit contributions | |
arl | -0.059 |
elly | -0.057 |
Today | -0.056 |
ay | -0.055 |
ute | -0.054 |
Token | pick |
Token ID | 2298 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
learned | +0.066 |
never | +0.066 |
shared | +0.065 |
pretended | +0.065 |
wed | +0.064 |
Neg logit contributions | |
Today | -0.102 |
arl | -0.101 |
� | -0.096 |
arry | -0.094 |
ute | -0.093 |
Token | up |
Token ID | 510 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
arl | +0.017 |
elly | +0.017 |
ay | +0.017 |
Today | +0.016 |
ane | +0.016 |
Neg logit contributions | |
never | -0.011 |
pretended | -0.010 |
put | -0.010 |
forgot | -0.010 |
punished | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
arl | +0.016 |
elly | +0.015 |
ay | +0.015 |
Today | +0.015 |
� | +0.015 |
Neg logit contributions | |
put | -0.005 |
never | -0.005 |
pretended | -0.005 |
punished | -0.005 |
learned | -0.005 |
Token | seaw |
Token ID | 32085 |
Feature activation | -1.38e+01 |
Pos logit contributions | |
arl | +0.052 |
elly | +0.050 |
ay | +0.049 |
ute | +0.048 |
Today | +0.047 |
Neg logit contributions | |
never | -0.038 |
punished | -0.037 |
learned | -0.036 |
put | -0.036 |
hoped | -0.036 |
Token | eed |
Token ID | 2308 |
Feature activation | -6.05e+00 |
Pos logit contributions | |
Today | +0.202 |
arl | +0.197 |
ane | +0.192 |
ay | +0.187 |
� | +0.182 |
Neg logit contributions | |
put | -0.342 |
pretended | -0.313 |
used | -0.313 |
made | -0.312 |
never | -0.310 |
Token | and |
Token ID | 290 |
Feature activation | +6.61e+00 |
Pos logit contributions | |
arl | +0.141 |
ay | +0.135 |
elly | +0.134 |
Today | +0.131 |
ane | +0.130 |
Neg logit contributions | |
put | -0.101 |
never | -0.097 |
pretended | -0.096 |
used | -0.094 |
learned | -0.093 |
Token | put |
Token ID | 1234 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
wed | +0.063 |
punished | +0.063 |
never | +0.059 |
learned | +0.056 |
ck | +0.056 |
Neg logit contributions | |
Today | -0.148 |
� | -0.145 |
ute | -0.142 |
ore | -0.142 |
uddle | -0.141 |
Token | it |
Token ID | 340 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
forgot | +0.027 |
punished | +0.026 |
pretended | +0.025 |
never | +0.025 |
remembered | +0.024 |
Neg logit contributions | |
ay | -0.041 |
arl | -0.039 |
� | -0.038 |
ore | -0.037 |
anger | -0.037 |
Token | in |
Token ID | 287 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
pretended | +0.006 |
learned | +0.006 |
punished | +0.006 |
put | +0.006 |
forgot | +0.006 |
Neg logit contributions | |
arl | -0.007 |
ay | -0.006 |
elly | -0.006 |
ute | -0.006 |
Today | -0.006 |
Token | he |
Token ID | 339 |
Feature activation | +8.07e+00 |
Pos logit contributions | |
ay | +0.018 |
Today | +0.018 |
arl | +0.018 |
� | +0.018 |
elly | +0.017 |
Neg logit contributions | |
put | -0.011 |
punished | -0.011 |
never | -0.011 |
learned | -0.011 |
pretended | -0.011 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
herself | +0.077 |
remind | +0.074 |
eling | +0.074 |
permission | +0.071 |
care | +0.068 |
Neg logit contributions | |
ute | -0.180 |
arry | -0.178 |
elly | -0.178 |
uddle | -0.177 |
ay | -0.171 |
Token | to |
Token ID | 284 |
Feature activation | +4.58e+00 |
Pos logit contributions | |
pretended | +0.028 |
wed | +0.026 |
ck | +0.026 |
put | +0.026 |
punished | +0.026 |
Neg logit contributions | |
arl | -0.017 |
elly | -0.016 |
ane | -0.016 |
ute | -0.015 |
arry | -0.015 |
Token | remove |
Token ID | 4781 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
remembered | +0.053 |
pretended | +0.053 |
learned | +0.052 |
hoped | +0.052 |
forgot | +0.050 |
Neg logit contributions | |
arl | -0.119 |
ay | -0.118 |
� | -0.114 |
Today | -0.113 |
ute | -0.113 |
Token | the |
Token ID | 262 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
ay | +0.016 |
arl | +0.016 |
elly | +0.016 |
ore | +0.015 |
andel | +0.015 |
Neg logit contributions | |
forgot | -0.008 |
pretended | -0.008 |
never | -0.008 |
remembered | -0.008 |
punished | -0.007 |
Token | seaw |
Token ID | 32085 |
Feature activation | -1.38e+01 |
Pos logit contributions | |
arl | +0.060 |
ay | +0.059 |
elly | +0.058 |
ute | +0.056 |
ane | +0.055 |
Neg logit contributions | |
never | -0.048 |
punished | -0.048 |
pretended | -0.047 |
put | -0.047 |
shared | -0.046 |
Token | eed |
Token ID | 2308 |
Feature activation | -5.99e+00 |
Pos logit contributions | |
arl | +0.183 |
Today | +0.175 |
ane | +0.170 |
elly | +0.169 |
ay | +0.167 |
Neg logit contributions | |
put | -0.362 |
pretended | -0.338 |
never | -0.336 |
made | -0.335 |
used | -0.335 |
Token | from |
Token ID | 422 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
arl | +0.130 |
ay | +0.125 |
elly | +0.124 |
ane | +0.120 |
ute | +0.120 |
Neg logit contributions | |
put | -0.105 |
pretended | -0.103 |
never | -0.103 |
learned | -0.101 |
punished | -0.100 |
Token | the |
Token ID | 262 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
forgot | +0.013 |
learned | +0.012 |
punished | +0.012 |
remembered | +0.012 |
pretended | +0.012 |
Neg logit contributions | |
arl | -0.026 |
ay | -0.025 |
� | -0.025 |
ore | -0.025 |
Today | -0.025 |
Token | water |
Token ID | 1660 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
arl | +0.053 |
ay | +0.051 |
elly | +0.050 |
ute | +0.048 |
Today | +0.048 |
Neg logit contributions | |
put | -0.043 |
pretended | -0.043 |
never | -0.042 |
punished | -0.041 |
used | -0.041 |
Token | so |
Token ID | 523 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
arl | +0.106 |
ay | +0.102 |
elly | +0.101 |
Today | +0.098 |
ane | +0.098 |
Neg logit contributions | |
put | -0.078 |
never | -0.075 |
pretended | -0.074 |
used | -0.072 |
learned | -0.072 |
Token | box |
Token ID | 3091 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
arl | +0.064 |
elly | +0.061 |
ay | +0.061 |
ane | +0.059 |
Today | +0.059 |
Neg logit contributions | |
put | -0.056 |
never | -0.053 |
pretended | -0.053 |
used | -0.052 |
learned | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +7.44e+00 |
Pos logit contributions | |
arl | +0.088 |
ay | +0.086 |
elly | +0.085 |
Today | +0.083 |
ane | +0.082 |
Neg logit contributions | |
put | -0.068 |
never | -0.066 |
pretended | -0.066 |
punished | -0.065 |
learned | -0.064 |
Token | share |
Token ID | 2648 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
punished | +0.110 |
never | +0.108 |
wed | +0.101 |
learned | +0.097 |
remembered | +0.097 |
Neg logit contributions | |
arl | -0.155 |
Today | -0.155 |
oms | -0.152 |
ay | -0.151 |
� | -0.149 |
Token | the |
Token ID | 262 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
punished | +0.002 |
never | +0.002 |
pretended | +0.002 |
forgot | +0.002 |
put | +0.002 |
Neg logit contributions | |
arl | -0.004 |
ay | -0.004 |
elly | -0.004 |
Today | -0.004 |
ane | -0.004 |
Token | ch |
Token ID | 442 |
Feature activation | -8.31e+00 |
Pos logit contributions | |
ay | +0.040 |
arl | +0.040 |
Today | +0.039 |
elly | +0.037 |
ane | +0.036 |
Neg logit contributions | |
never | -0.037 |
punished | -0.037 |
put | -0.036 |
forgot | -0.036 |
pretended | -0.035 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
arl | +0.110 |
elly | +0.102 |
ay | +0.101 |
Today | +0.098 |
ane | +0.097 |
Neg logit contributions | |
put | -0.221 |
never | -0.213 |
pretended | -0.212 |
used | -0.210 |
learned | -0.209 |
Token | ates |
Token ID | 689 |
Feature activation | -5.67e+00 |
Pos logit contributions | |
arl | +0.278 |
Today | +0.267 |
elly | +0.265 |
ay | +0.262 |
� | +0.256 |
Neg logit contributions | |
put | -0.243 |
never | -0.238 |
used | -0.234 |
pretended | -0.232 |
learned | -0.229 |
Token | . |
Token ID | 13 |
Feature activation | -3.39e+00 |
Pos logit contributions | |
arl | +0.113 |
ay | +0.111 |
elly | +0.111 |
ane | +0.107 |
Today | +0.106 |
Neg logit contributions | |
put | -0.106 |
pretended | -0.105 |
never | -0.104 |
punished | -0.104 |
cleaned | -0.102 |
Token | They |
Token ID | 1119 |
Feature activation | +8.36e+00 |
Pos logit contributions | |
ane | +0.083 |
ute | +0.077 |
arl | +0.077 |
igsaw | +0.077 |
elly | +0.076 |
Neg logit contributions | |
put | -0.063 |
hoped | -0.055 |
learned | -0.054 |
used | -0.054 |
made | -0.054 |
Token | are |
Token ID | 389 |
Feature activation | +4.40e+00 |
Pos logit contributions | |
punished | +0.127 |
remembered | +0.113 |
wed | +0.111 |
cleaned | +0.108 |
promised | +0.108 |
Neg logit contributions | |
� | -0.172 |
arl | -0.171 |
ay | -0.167 |
Today | -0.161 |
owers | -0.158 |
Token | happy |
Token ID | 3772 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
wed | +0.081 |
punished | +0.079 |
tried | +0.077 |
eling | +0.076 |
prayed | +0.074 |
Neg logit contributions | |
ay | -0.069 |
apping | -0.069 |
arl | -0.069 |
Today | -0.065 |
ummy | -0.064 |
Token | beach |
Token ID | 10481 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
arl | +0.032 |
ay | +0.030 |
elly | +0.030 |
Today | +0.029 |
ane | +0.029 |
Neg logit contributions | |
put | -0.030 |
never | -0.029 |
pretended | -0.029 |
used | -0.028 |
learned | -0.028 |
Token | was |
Token ID | 373 |
Feature activation | +3.15e+00 |
Pos logit contributions | |
elly | +0.028 |
ay | +0.028 |
ane | +0.028 |
ummy | +0.027 |
Today | +0.027 |
Neg logit contributions | |
put | -0.016 |
pretended | -0.016 |
punished | -0.015 |
ck | -0.015 |
never | -0.015 |
Token | full |
Token ID | 1336 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
pretended | +0.048 |
ck | +0.047 |
hoped | +0.046 |
put | +0.046 |
learned | +0.045 |
Neg logit contributions | |
arl | -0.066 |
ay | -0.066 |
elly | -0.066 |
Today | -0.065 |
� | -0.063 |
Token | of |
Token ID | 286 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
arl | +0.031 |
ay | +0.028 |
ane | +0.028 |
elly | +0.028 |
ute | +0.028 |
Neg logit contributions | |
put | -0.038 |
never | -0.036 |
made | -0.036 |
used | -0.036 |
cleaned | -0.035 |
Token | messy |
Token ID | 24097 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
never | +0.019 |
pretended | +0.019 |
punished | +0.019 |
forgot | +0.019 |
remembered | +0.018 |
Neg logit contributions | |
ay | -0.022 |
arl | -0.022 |
elly | -0.022 |
ummy | -0.021 |
Today | -0.021 |
Token | seaw |
Token ID | 32085 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
arl | +0.086 |
ay | +0.082 |
elly | +0.082 |
Today | +0.079 |
ane | +0.079 |
Neg logit contributions | |
put | -0.062 |
never | -0.060 |
pretended | -0.059 |
learned | -0.057 |
used | -0.057 |
Token | eed |
Token ID | 2308 |
Feature activation | -6.35e+00 |
Pos logit contributions | |
Today | +0.204 |
arl | +0.196 |
ane | +0.189 |
elly | +0.178 |
� | +0.175 |
Neg logit contributions | |
put | -0.322 |
never | -0.296 |
made | -0.294 |
used | -0.294 |
pretended | -0.292 |
Token | . |
Token ID | 13 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
arl | +0.136 |
elly | +0.132 |
ay | +0.132 |
ane | +0.127 |
Today | +0.126 |
Neg logit contributions | |
put | -0.113 |
never | -0.109 |
pretended | -0.109 |
learned | -0.107 |
punished | -0.107 |
Token | The |
Token ID | 383 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
arl | +0.098 |
ute | +0.093 |
ane | +0.087 |
ower | +0.085 |
elly | +0.085 |
Neg logit contributions | |
put | -0.070 |
learned | -0.066 |
tried | -0.063 |
used | -0.063 |
never | -0.063 |
Token | girl |
Token ID | 2576 |
Feature activation | +5.11e+00 |
Pos logit contributions | |
arl | +0.033 |
ay | +0.031 |
elly | +0.031 |
Today | +0.030 |
ane | +0.030 |
Neg logit contributions | |
put | -0.039 |
never | -0.038 |
pretended | -0.038 |
used | -0.037 |
punished | -0.037 |
Token | tried |
Token ID | 3088 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
remind | +0.070 |
ws | +0.067 |
ck | +0.064 |
permission | +0.064 |
share | +0.063 |
Neg logit contributions | |
uddle | -0.100 |
op | -0.100 |
elly | -0.097 |
oles | -0.096 |
ane | -0.096 |
Token | not |
Token ID | 407 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
cleaned | +0.001 |
pretended | +0.001 |
punished | +0.001 |
put | +0.001 |
hoped | +0.001 |
Neg logit contributions | |
Today | -0.002 |
arl | -0.002 |
ay | -0.002 |
ute | -0.002 |
elly | -0.002 |
Token | like |
Token ID | 588 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
pretended | +0.030 |
punished | +0.029 |
remembered | +0.028 |
saved | +0.028 |
hoped | +0.028 |
Neg logit contributions | |
arl | -0.062 |
ay | -0.061 |
elly | -0.060 |
Today | -0.060 |
ute | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
put | +0.005 |
pretended | +0.005 |
never | +0.005 |
punished | +0.005 |
learned | +0.005 |
Neg logit contributions | |
arl | -0.008 |
ay | -0.008 |
elly | -0.008 |
ane | -0.008 |
Today | -0.008 |
Token | share |
Token ID | 2648 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
remembered | +0.037 |
hoped | +0.037 |
saved | +0.036 |
punished | +0.035 |
managed | +0.035 |
Neg logit contributions | |
ay | -0.073 |
arl | -0.071 |
Today | -0.069 |
ute | -0.067 |
igsaw | -0.066 |
Token | his |
Token ID | 465 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
arl | +0.038 |
ay | +0.036 |
elly | +0.036 |
Today | +0.035 |
ane | +0.035 |
Neg logit contributions | |
put | -0.023 |
never | -0.023 |
pretended | -0.022 |
learned | -0.022 |
used | -0.022 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
arl | +0.043 |
ay | +0.042 |
elly | +0.041 |
Today | +0.040 |
ane | +0.040 |
Neg logit contributions | |
put | -0.024 |
never | -0.023 |
pretended | -0.022 |
punished | -0.022 |
used | -0.022 |
Token | ay |
Token ID | 323 |
Feature activation | -6.12e+00 |
Pos logit contributions | |
arl | +0.063 |
elly | +0.043 |
� | +0.041 |
Today | +0.039 |
arry | +0.027 |
Neg logit contributions | |
put | -0.452 |
never | -0.444 |
made | -0.441 |
used | -0.434 |
pretended | -0.430 |
Token | ons |
Token ID | 684 |
Feature activation | -6.88e+00 |
Pos logit contributions | |
ay | +0.146 |
arl | +0.146 |
elly | +0.140 |
ane | +0.138 |
Today | +0.136 |
Neg logit contributions | |
put | -0.096 |
never | -0.093 |
pretended | -0.092 |
punished | -0.090 |
hoped | -0.089 |
Token | with |
Token ID | 351 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
arl | +0.150 |
ay | +0.144 |
elly | +0.144 |
ane | +0.138 |
Today | +0.138 |
Neg logit contributions | |
put | -0.123 |
never | -0.119 |
pretended | -0.119 |
learned | -0.116 |
used | -0.115 |
Token | his |
Token ID | 465 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
put | +0.002 |
never | +0.002 |
pretended | +0.002 |
punished | +0.002 |
learned | +0.002 |
Neg logit contributions | |
arl | -0.005 |
ay | -0.005 |
elly | -0.005 |
ane | -0.005 |
Today | -0.005 |
Token | sister |
Token ID | 6621 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
arl | +0.022 |
ay | +0.022 |
elly | +0.021 |
Today | +0.021 |
ane | +0.021 |
Neg logit contributions | |
put | -0.013 |
never | -0.012 |
pretended | -0.012 |
punished | -0.012 |
learned | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
arl | +0.053 |
ay | +0.050 |
elly | +0.050 |
Today | +0.049 |
ane | +0.048 |
Neg logit contributions | |
put | -0.040 |
never | -0.039 |
pretended | -0.039 |
learned | -0.038 |
used | -0.038 |
Token | Can |
Token ID | 1680 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
arl | +0.074 |
ay | +0.070 |
ane | +0.069 |
ute | +0.069 |
andel | +0.069 |
Neg logit contributions | |
put | -0.078 |
never | -0.076 |
pretended | -0.074 |
made | -0.074 |
learned | -0.073 |
Token | we |
Token ID | 356 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
put | +0.016 |
pretended | +0.016 |
never | +0.015 |
used | +0.015 |
learned | +0.015 |
Neg logit contributions | |
arl | -0.017 |
ay | -0.016 |
elly | -0.016 |
Today | -0.016 |
ane | -0.016 |
Token | share |
Token ID | 2648 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
pretended | +0.032 |
hoped | +0.030 |
punished | +0.029 |
remembered | +0.028 |
learned | +0.028 |
Neg logit contributions | |
ay | -0.080 |
arl | -0.079 |
� | -0.075 |
ute | -0.074 |
Today | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | -3.71e+00 |
Pos logit contributions | |
ay | +0.055 |
arl | +0.054 |
elly | +0.053 |
Today | +0.051 |
ummy | +0.051 |
Neg logit contributions | |
pretended | -0.029 |
never | -0.029 |
punished | -0.029 |
forgot | -0.029 |
put | -0.028 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
ay | +0.091 |
arl | +0.090 |
elly | +0.087 |
Today | +0.086 |
ane | +0.085 |
Neg logit contributions | |
never | -0.054 |
put | -0.053 |
punished | -0.053 |
pretended | -0.052 |
hoped | -0.051 |
Token | ay |
Token ID | 323 |
Feature activation | -6.89e+00 |
Pos logit contributions | |
arl | +0.079 |
elly | +0.052 |
� | +0.049 |
Today | +0.046 |
arry | +0.044 |
Neg logit contributions | |
put | -0.438 |
made | -0.429 |
never | -0.425 |
used | -0.421 |
fixed | -0.418 |
Token | on |
Token ID | 261 |
Feature activation | -7.76e+00 |
Pos logit contributions | |
arl | +0.158 |
ay | +0.152 |
elly | +0.151 |
Today | +0.146 |
ane | +0.146 |
Neg logit contributions | |
put | -0.117 |
never | -0.112 |
pretended | -0.111 |
used | -0.108 |
learned | -0.108 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
arl | +0.176 |
elly | +0.165 |
ay | +0.162 |
Today | +0.160 |
� | +0.159 |
Neg logit contributions | |
put | -0.137 |
never | -0.131 |
pretended | -0.129 |
made | -0.128 |
learned | -0.127 |
Token |
|
Token ID | 198 |
Feature activation | -5.32e+00 |
Pos logit contributions | |
arl | +0.091 |
ay | +0.088 |
ane | +0.088 |
igsaw | +0.087 |
umm | +0.085 |
Neg logit contributions | |
put | -0.082 |
learned | -0.077 |
hoped | -0.074 |
made | -0.074 |
never | -0.073 |
Token |
|
Token ID | 198 |
Feature activation | -4.24e+00 |
Pos logit contributions | |
arl | +0.103 |
elly | +0.101 |
ane | +0.101 |
igsaw | +0.100 |
umm | +0.099 |
Neg logit contributions | |
put | -0.115 |
learned | -0.108 |
never | -0.107 |
used | -0.104 |
made | -0.102 |
Token | He |
Token ID | 679 |
Feature activation | +6.99e+00 |
Pos logit contributions | |
arl | +0.087 |
ute | +0.083 |
ane | +0.083 |
ay | +0.082 |
owers | +0.080 |
Neg logit contributions | |
put | -0.093 |
never | -0.089 |
learned | -0.089 |
tried | -0.087 |
made | -0.087 |
Token | tries |
Token ID | 8404 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
never | +0.106 |
pretended | +0.098 |
cleaned | +0.098 |
learned | +0.097 |
used | +0.097 |
Neg logit contributions | |
arl | -0.165 |
ay | -0.155 |
elly | -0.154 |
Today | -0.148 |
arry | -0.143 |
Token | to |
Token ID | 284 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
arl | +0.008 |
ay | +0.007 |
elly | +0.007 |
Today | +0.007 |
ane | +0.007 |
Neg logit contributions | |
put | -0.006 |
pretended | -0.006 |
never | -0.006 |
learned | -0.006 |
used | -0.006 |
Token | grab |
Token ID | 5552 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
remembered | +0.060 |
hoped | +0.060 |
learned | +0.059 |
wished | +0.057 |
pretended | +0.056 |
Neg logit contributions | |
arl | -0.091 |
ay | -0.082 |
� | -0.080 |
igsaw | -0.075 |
Today | -0.075 |
Token | the |
Token ID | 262 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
punished | +0.001 |
forgot | +0.001 |
never | +0.001 |
pretended | +0.001 |
learned | +0.001 |
Neg logit contributions | |
ay | -0.001 |
arl | -0.001 |
elly | -0.001 |
Today | -0.001 |
ane | -0.001 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
arl | +0.096 |
ay | +0.095 |
elly | +0.092 |
Today | +0.091 |
ane | +0.090 |
Neg logit contributions | |
put | -0.055 |
never | -0.054 |
pretended | -0.053 |
punished | -0.051 |
learned | -0.051 |
Token | ay |
Token ID | 323 |
Feature activation | -6.70e+00 |
Pos logit contributions | |
arl | +0.074 |
elly | +0.048 |
Today | +0.045 |
� | +0.045 |
ute | +0.042 |
Neg logit contributions | |
put | -0.432 |
made | -0.424 |
never | -0.420 |
used | -0.416 |
fixed | -0.414 |
Token | on |
Token ID | 261 |
Feature activation | -6.96e+00 |
Pos logit contributions | |
arl | +0.161 |
elly | +0.152 |
ay | +0.147 |
Today | +0.147 |
ute | +0.145 |
Neg logit contributions | |
put | -0.109 |
never | -0.103 |
pretended | -0.102 |
made | -0.100 |
used | -0.100 |
Token | from |
Token ID | 422 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
arl | +0.166 |
elly | +0.157 |
ay | +0.154 |
ane | +0.151 |
Today | +0.151 |
Neg logit contributions | |
put | -0.116 |
never | -0.109 |
pretended | -0.106 |
made | -0.106 |
used | -0.106 |
Token | Mia |
Token ID | 32189 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
punished | +0.004 |
cleaned | +0.004 |
forgot | +0.004 |
learned | +0.004 |
used | +0.004 |
Neg logit contributions | |
arl | -0.005 |
Today | -0.004 |
ay | -0.004 |
elly | -0.004 |
ummy | -0.004 |
Token | 's |
Token ID | 338 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
arl | +0.082 |
elly | +0.077 |
ay | +0.077 |
ane | +0.075 |
Today | +0.075 |
Neg logit contributions | |
put | -0.047 |
never | -0.045 |
pretended | -0.044 |
used | -0.043 |
learned | -0.043 |
Token | happy |
Token ID | 3772 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
pretended | +0.053 |
forgot | +0.051 |
wrote | +0.047 |
never | +0.046 |
counted | +0.045 |
Neg logit contributions | |
arl | -0.074 |
elly | -0.066 |
over | -0.064 |
apping | -0.062 |
Today | -0.062 |
Token | to |
Token ID | 284 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
arl | +0.007 |
elly | +0.007 |
ay | +0.007 |
Today | +0.006 |
ane | +0.006 |
Neg logit contributions | |
put | -0.005 |
pretended | -0.005 |
never | -0.005 |
learned | -0.005 |
punished | -0.005 |
Token | see |
Token ID | 766 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
pretended | +0.035 |
punished | +0.034 |
hoped | +0.034 |
cleaned | +0.033 |
learned | +0.033 |
Neg logit contributions | |
arl | -0.073 |
ay | -0.068 |
Today | -0.068 |
� | -0.066 |
elly | -0.065 |
Token | our |
Token ID | 674 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
arl | +0.002 |
ay | +0.002 |
elly | +0.002 |
Today | +0.002 |
ummy | +0.002 |
Neg logit contributions | |
pretended | -0.001 |
put | -0.001 |
never | -0.001 |
learned | -0.001 |
forgot | -0.001 |
Token | big |
Token ID | 1263 |
Feature activation | -4.85e+00 |
Pos logit contributions | |
arl | +0.025 |
elly | +0.024 |
ay | +0.024 |
ummy | +0.023 |
Today | +0.023 |
Neg logit contributions | |
put | -0.018 |
never | -0.017 |
pretended | -0.017 |
punished | -0.017 |
learned | -0.017 |
Token | cuc |
Token ID | 38421 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
arl | +0.105 |
elly | +0.101 |
ay | +0.100 |
Today | +0.098 |
ummy | +0.096 |
Neg logit contributions | |
put | -0.086 |
never | -0.086 |
pretended | -0.084 |
learned | -0.082 |
punished | -0.081 |
Token | umber |
Token ID | 4494 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
arl | +0.203 |
ay | +0.193 |
Today | +0.192 |
elly | +0.186 |
ane | +0.179 |
Neg logit contributions | |
put | -0.316 |
used | -0.303 |
never | -0.302 |
made | -0.298 |
punished | -0.298 |
Token | ." |
Token ID | 526 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
arl | +0.070 |
elly | +0.067 |
ay | +0.066 |
ane | +0.066 |
Today | +0.065 |
Neg logit contributions | |
pretended | -0.066 |
put | -0.066 |
never | -0.065 |
learned | -0.065 |
cleaned | -0.064 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
arl | +0.093 |
ay | +0.090 |
ane | +0.090 |
elly | +0.089 |
ute | +0.088 |
Neg logit contributions | |
put | -0.051 |
pretended | -0.047 |
never | -0.046 |
learned | -0.046 |
used | -0.045 |
Token | says |
Token ID | 1139 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
never | +0.070 |
put | +0.069 |
pretended | +0.067 |
punished | +0.066 |
used | +0.065 |
Neg logit contributions | |
arl | -0.120 |
elly | -0.117 |
Today | -0.116 |
ay | -0.116 |
ane | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
arl | +0.038 |
ay | +0.037 |
elly | +0.037 |
ane | +0.036 |
Today | +0.035 |
Neg logit contributions | |
put | -0.026 |
never | -0.024 |
used | -0.024 |
pretended | -0.023 |
learned | -0.023 |
Token | It |
Token ID | 632 |
Feature activation | +5.07e+00 |
Pos logit contributions | |
ute | +0.091 |
ane | +0.090 |
ower | +0.090 |
arl | +0.090 |
owers | +0.086 |
Neg logit contributions | |
put | -0.051 |
learned | -0.046 |
hoped | -0.045 |
pretended | -0.045 |
tried | -0.044 |
Token | was |
Token ID | 373 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
punished | +0.075 |
permission | +0.069 |
wed | +0.066 |
cleaned | +0.065 |
never | +0.064 |
Neg logit contributions | |
elly | -0.111 |
Today | -0.111 |
ay | -0.110 |
� | -0.108 |
ane | -0.106 |
Token | hiding |
Token ID | 11816 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
punished | +0.048 |
put | +0.048 |
learned | +0.048 |
pretended | +0.048 |
forgot | +0.047 |
Neg logit contributions | |
arl | -0.067 |
ay | -0.064 |
elly | -0.063 |
Today | -0.062 |
ummy | -0.060 |
Token | under |
Token ID | 739 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
arl | +0.052 |
ay | +0.050 |
elly | +0.050 |
ane | +0.048 |
Today | +0.048 |
Neg logit contributions | |
put | -0.035 |
never | -0.033 |
pretended | -0.033 |
learned | -0.033 |
punished | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
put | +0.013 |
never | +0.013 |
pretended | +0.013 |
punished | +0.012 |
learned | +0.012 |
Neg logit contributions | |
arl | -0.035 |
ay | -0.034 |
elly | -0.034 |
Today | -0.033 |
� | -0.033 |
Token | seaw |
Token ID | 32085 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
arl | +0.039 |
elly | +0.038 |
Today | +0.037 |
owers | +0.037 |
arry | +0.037 |
Neg logit contributions | |
put | -0.024 |
made | -0.021 |
never | -0.021 |
used | -0.021 |
pretended | -0.020 |
Token | eed |
Token ID | 2308 |
Feature activation | -6.06e+00 |
Pos logit contributions | |
arl | +0.183 |
Today | +0.183 |
ane | +0.178 |
owers | +0.174 |
� | +0.172 |
Neg logit contributions | |
put | -0.325 |
never | -0.307 |
made | -0.300 |
used | -0.297 |
pretended | -0.297 |
Token | . |
Token ID | 13 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
arl | +0.144 |
ay | +0.139 |
elly | +0.137 |
Today | +0.135 |
ane | +0.133 |
Neg logit contributions | |
put | -0.101 |
never | -0.096 |
made | -0.094 |
pretended | -0.093 |
used | -0.092 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.84e+00 |
Pos logit contributions | |
ane | +0.097 |
ute | +0.097 |
ower | +0.096 |
arl | +0.095 |
ay | +0.093 |
Neg logit contributions | |
put | -0.039 |
learned | -0.034 |
hoped | -0.032 |
tried | -0.032 |
pretended | -0.032 |
Token | said |
Token ID | 531 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
punished | +0.064 |
remind | +0.060 |
never | +0.056 |
cleaned | +0.056 |
wed | +0.056 |
Neg logit contributions | |
elly | -0.113 |
arry | -0.107 |
Today | -0.105 |
ay | -0.104 |
oles | -0.104 |
Token | , |
Token ID | 11 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
arl | +0.012 |
elly | +0.012 |
ay | +0.012 |
igsaw | +0.011 |
� | +0.011 |
Neg logit contributions | |
put | -0.011 |
never | -0.010 |
made | -0.010 |
tried | -0.010 |
used | -0.010 |
Token | still |
Token ID | 991 |
Feature activation | +6.55e+00 |
Pos logit contributions | |
pretended | +0.060 |
remembered | +0.058 |
hoped | +0.057 |
ck | +0.057 |
wed | +0.056 |
Neg logit contributions | |
arl | -0.095 |
ay | -0.095 |
ane | -0.091 |
elly | -0.090 |
owers | -0.089 |
Token | have |
Token ID | 423 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
pretended | +0.082 |
hoped | +0.078 |
remembered | +0.076 |
ck | +0.075 |
punished | +0.073 |
Neg logit contributions | |
arl | -0.153 |
elly | -0.153 |
ay | -0.152 |
Today | -0.146 |
ore | -0.145 |
Token | to |
Token ID | 284 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
pretended | +0.066 |
put | +0.061 |
ck | +0.061 |
never | +0.060 |
hoped | +0.060 |
Neg logit contributions | |
arl | -0.076 |
elly | -0.075 |
ay | -0.072 |
ummy | -0.067 |
aris | -0.067 |
Token | remove |
Token ID | 4781 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
pretended | +0.047 |
hoped | +0.045 |
forgot | +0.042 |
remembered | +0.042 |
learned | +0.041 |
Neg logit contributions | |
arl | -0.111 |
ay | -0.103 |
Today | -0.101 |
ute | -0.099 |
arry | -0.096 |
Token | the |
Token ID | 262 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
arl | +0.041 |
ay | +0.039 |
elly | +0.039 |
Today | +0.038 |
ane | +0.038 |
Neg logit contributions | |
put | -0.015 |
never | -0.014 |
pretended | -0.014 |
learned | -0.014 |
used | -0.014 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
arl | +0.105 |
elly | +0.100 |
ay | +0.099 |
Today | +0.096 |
ane | +0.096 |
Neg logit contributions | |
put | -0.078 |
never | -0.074 |
pretended | -0.074 |
used | -0.072 |
learned | -0.072 |
Token | ay |
Token ID | 323 |
Feature activation | -6.98e+00 |
Pos logit contributions | |
arl | +0.071 |
� | +0.054 |
elly | +0.052 |
Today | +0.052 |
aris | +0.043 |
Neg logit contributions | |
put | -0.424 |
made | -0.418 |
never | -0.416 |
used | -0.409 |
fixed | -0.404 |
Token | on |
Token ID | 261 |
Feature activation | -7.57e+00 |
Pos logit contributions | |
arl | +0.190 |
elly | +0.181 |
Today | +0.177 |
arry | +0.174 |
� | +0.174 |
Neg logit contributions | |
put | -0.091 |
never | -0.084 |
made | -0.084 |
pretended | -0.082 |
used | -0.081 |
Token | marks |
Token ID | 8849 |
Feature activation | -7.27e+00 |
Pos logit contributions | |
arl | +0.183 |
elly | +0.173 |
� | +0.171 |
Today | +0.170 |
ay | +0.170 |
Neg logit contributions | |
put | -0.123 |
never | -0.116 |
made | -0.112 |
used | -0.112 |
pretended | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
arl | +0.171 |
elly | +0.164 |
ay | +0.162 |
Today | +0.160 |
arry | +0.160 |
Neg logit contributions | |
put | -0.121 |
never | -0.114 |
made | -0.111 |
used | -0.111 |
pretended | -0.110 |
Token | White |
Token ID | 2635 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
arl | +0.065 |
ane | +0.064 |
ute | +0.064 |
ay | +0.062 |
elly | +0.061 |
Neg logit contributions | |
put | -0.074 |
never | -0.073 |
learned | -0.070 |
used | -0.069 |
pretended | -0.069 |
Token | brother |
Token ID | 3956 |
Feature activation | +7.65e-02 |
Pos logit contributions | |
arl | +0.039 |
elly | +0.038 |
Today | +0.037 |
ay | +0.037 |
arry | +0.037 |
Neg logit contributions | |
put | -0.021 |
never | -0.019 |
pretended | -0.019 |
made | -0.018 |
cleaned | -0.018 |
Token | accidentally |
Token ID | 14716 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
ck | +0.001 |
hoped | +0.001 |
ts | +0.001 |
never | +0.001 |
punished | +0.001 |
Neg logit contributions | |
ay | -0.001 |
elly | -0.001 |
Today | -0.001 |
ummy | -0.001 |
arl | -0.001 |
Token | knocked |
Token ID | 13642 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
learned | +0.046 |
hoped | +0.045 |
never | +0.043 |
forgot | +0.041 |
ts | +0.040 |
Neg logit contributions | |
ore | -0.072 |
anger | -0.072 |
ay | -0.072 |
ane | -0.072 |
arl | -0.071 |
Token | over |
Token ID | 625 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
arl | +0.034 |
ay | +0.033 |
elly | +0.033 |
Today | +0.032 |
ane | +0.031 |
Neg logit contributions | |
pretended | -0.018 |
put | -0.018 |
never | -0.017 |
forgot | -0.017 |
hoped | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
arl | +0.012 |
ay | +0.011 |
elly | +0.011 |
Today | +0.011 |
� | +0.011 |
Neg logit contributions | |
put | -0.004 |
never | -0.004 |
pretended | -0.004 |
forgot | -0.003 |
hoped | -0.003 |
Token | v |
Token ID | 410 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
arl | +0.094 |
arry | +0.092 |
elly | +0.089 |
Today | +0.089 |
ay | +0.088 |
Neg logit contributions | |
put | -0.066 |
never | -0.061 |
made | -0.060 |
used | -0.059 |
cleaned | -0.059 |
Token | ase |
Token ID | 589 |
Feature activation | -6.05e+00 |
Pos logit contributions | |
arl | +0.183 |
Today | +0.181 |
arry | +0.166 |
elly | +0.157 |
owers | +0.156 |
Neg logit contributions | |
never | -0.322 |
put | -0.315 |
made | -0.309 |
learned | -0.308 |
cleaned | -0.307 |
Token | and |
Token ID | 290 |
Feature activation | +6.08e+00 |
Pos logit contributions | |
arl | +0.157 |
elly | +0.149 |
arry | +0.146 |
ute | +0.146 |
ane | +0.145 |
Neg logit contributions | |
put | -0.090 |
never | -0.086 |
made | -0.085 |
used | -0.083 |
cleaned | -0.081 |
Token | it |
Token ID | 340 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
wed | +0.099 |
asured | +0.094 |
ck | +0.092 |
permission | +0.090 |
anut | +0.087 |
Neg logit contributions | |
ay | -0.101 |
time | -0.099 |
ore | -0.097 |
tornado | -0.095 |
� | -0.094 |
Token | broke |
Token ID | 6265 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
learned | +0.042 |
hoped | +0.041 |
pretended | +0.041 |
put | +0.040 |
shared | +0.040 |
Neg logit contributions | |
ay | -0.071 |
arl | -0.067 |
elly | -0.065 |
ane | -0.065 |
� | -0.065 |
Token | into |
Token ID | 656 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
arl | +0.070 |
ay | +0.068 |
elly | +0.068 |
ane | +0.065 |
Today | +0.065 |
Neg logit contributions | |
put | -0.052 |
never | -0.050 |
pretended | -0.050 |
learned | -0.049 |
punished | -0.048 |
Token | started |
Token ID | 2067 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
punished | +0.075 |
never | +0.073 |
learned | +0.066 |
wed | +0.063 |
remind | +0.061 |
Neg logit contributions | |
ay | -0.167 |
ane | -0.160 |
ore | -0.155 |
aris | -0.155 |
elly | -0.154 |
Token | eating |
Token ID | 6600 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
punished | +0.056 |
forgot | +0.056 |
learned | +0.055 |
never | +0.055 |
hoped | +0.053 |
Neg logit contributions | |
ay | -0.054 |
arl | -0.054 |
Today | -0.053 |
ore | -0.053 |
ane | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
arl | +0.016 |
ay | +0.015 |
elly | +0.015 |
Today | +0.015 |
� | +0.015 |
Neg logit contributions | |
put | -0.007 |
never | -0.007 |
pretended | -0.007 |
punished | -0.007 |
learned | -0.007 |
Token | delicious |
Token ID | 12625 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
ay | +0.048 |
arl | +0.046 |
Today | +0.045 |
elly | +0.045 |
owers | +0.043 |
Neg logit contributions | |
never | -0.038 |
forgot | -0.037 |
punished | -0.036 |
hoped | -0.036 |
remembered | -0.036 |
Token | ch |
Token ID | 442 |
Feature activation | -7.80e+00 |
Pos logit contributions | |
ay | +0.105 |
arl | +0.105 |
elly | +0.102 |
Today | +0.100 |
ummy | +0.099 |
Neg logit contributions | |
never | -0.075 |
put | -0.073 |
pretended | -0.072 |
learned | -0.072 |
hoped | -0.072 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
arl | +0.108 |
elly | +0.099 |
ay | +0.099 |
Today | +0.099 |
ane | +0.096 |
Neg logit contributions | |
put | -0.202 |
never | -0.193 |
used | -0.192 |
pretended | -0.191 |
learned | -0.190 |
Token | ates |
Token ID | 689 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
arl | +0.278 |
Today | +0.271 |
elly | +0.264 |
ay | +0.262 |
ute | +0.253 |
Neg logit contributions | |
put | -0.224 |
never | -0.224 |
used | -0.220 |
learned | -0.214 |
cleaned | -0.212 |
Token | . |
Token ID | 13 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
arl | +0.086 |
ay | +0.086 |
elly | +0.086 |
ane | +0.084 |
ummy | +0.082 |
Neg logit contributions | |
put | -0.081 |
pretended | -0.080 |
never | -0.080 |
punished | -0.080 |
learned | -0.079 |
Token | She |
Token ID | 1375 |
Feature activation | +8.13e+00 |
Pos logit contributions | |
arl | +0.087 |
elly | +0.084 |
ute | +0.083 |
ane | +0.080 |
andel | +0.079 |
Neg logit contributions | |
put | -0.067 |
never | -0.065 |
learned | -0.062 |
hoped | -0.061 |
tried | -0.060 |
Token | was |
Token ID | 373 |
Feature activation | +4.46e+00 |
Pos logit contributions | |
punished | +0.102 |
permission | +0.096 |
remind | +0.091 |
wed | +0.089 |
ck | +0.087 |
Neg logit contributions | |
ay | -0.174 |
� | -0.174 |
oles | -0.169 |
elly | -0.167 |
ummy | -0.166 |
Token | so |
Token ID | 523 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
wed | +0.076 |
pretended | +0.074 |
forgot | +0.073 |
ck | +0.072 |
ts | +0.072 |
Neg logit contributions | |
ay | -0.081 |
Today | -0.078 |
apping | -0.078 |
ummy | -0.076 |
� | -0.074 |
Token | you |
Token ID | 345 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
put | +0.040 |
pretended | +0.039 |
never | +0.038 |
used | +0.037 |
punished | +0.037 |
Neg logit contributions | |
arl | -0.061 |
ay | -0.059 |
elly | -0.058 |
Today | -0.057 |
ane | -0.056 |
Token | want |
Token ID | 765 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
never | +0.022 |
punished | +0.021 |
put | +0.021 |
pretended | +0.021 |
hoped | +0.021 |
Neg logit contributions | |
arl | -0.039 |
Today | -0.038 |
ay | -0.038 |
elly | -0.037 |
ute | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
ay | +0.009 |
ummy | +0.009 |
elly | +0.009 |
arl | +0.009 |
Today | +0.008 |
Neg logit contributions | |
pretended | -0.008 |
forgot | -0.008 |
punished | -0.008 |
hoped | -0.008 |
never | -0.008 |
Token | borrow |
Token ID | 8804 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
punished | +0.017 |
hoped | +0.016 |
pretended | +0.016 |
never | +0.016 |
forgot | +0.016 |
Neg logit contributions | |
ay | -0.040 |
arl | -0.039 |
Today | -0.038 |
ane | -0.037 |
� | -0.037 |
Token | another |
Token ID | 1194 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
ay | +0.056 |
arl | +0.054 |
elly | +0.053 |
Today | +0.052 |
ane | +0.051 |
Neg logit contributions | |
forgot | -0.025 |
pretended | -0.025 |
never | -0.025 |
put | -0.025 |
punished | -0.025 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
arl | +0.109 |
ay | +0.105 |
elly | +0.104 |
Today | +0.102 |
ane | +0.101 |
Neg logit contributions | |
put | -0.068 |
never | -0.066 |
pretended | -0.065 |
learned | -0.063 |
used | -0.063 |
Token | ay |
Token ID | 323 |
Feature activation | -6.60e+00 |
Pos logit contributions | |
arl | +0.073 |
elly | +0.048 |
� | +0.044 |
owers | +0.043 |
arry | +0.040 |
Neg logit contributions | |
put | -0.434 |
made | -0.430 |
never | -0.420 |
fixed | -0.416 |
used | -0.415 |
Token | on |
Token ID | 261 |
Feature activation | -8.20e+00 |
Pos logit contributions | |
arl | +0.142 |
ay | +0.141 |
elly | +0.136 |
ane | +0.134 |
Today | +0.132 |
Neg logit contributions | |
put | -0.119 |
never | -0.116 |
pretended | -0.115 |
punished | -0.112 |
learned | -0.112 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
arl | +0.183 |
elly | +0.170 |
� | +0.167 |
arry | +0.166 |
ay | +0.165 |
Neg logit contributions | |
put | -0.151 |
made | -0.144 |
pretended | -0.138 |
never | -0.138 |
used | -0.138 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
arl | +0.098 |
ay | +0.095 |
elly | +0.094 |
ane | +0.093 |
ute | +0.092 |
Neg logit contributions | |
put | -0.046 |
never | -0.043 |
pretended | -0.042 |
learned | -0.042 |
used | -0.041 |
Token | asks |
Token ID | 7893 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
never | +0.049 |
put | +0.047 |
punished | +0.046 |
cleaned | +0.045 |
pretended | +0.045 |
Neg logit contributions | |
arl | -0.078 |
elly | -0.077 |
Today | -0.077 |
ay | -0.076 |
ane | -0.074 |
Token | him |
Token ID | 683 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
arl | +0.012 |
ay | +0.011 |
elly | +0.011 |
Today | +0.011 |
ane | +0.011 |
Neg logit contributions | |
put | -0.004 |
never | -0.003 |
pretended | -0.003 |
learned | -0.003 |
used | -0.003 |
Token | if |
Token ID | 611 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
arl | +0.016 |
ay | +0.015 |
elly | +0.015 |
Today | +0.015 |
ane | +0.014 |
Neg logit contributions | |
put | -0.009 |
never | -0.009 |
pretended | -0.009 |
learned | -0.009 |
used | -0.009 |
Token | he |
Token ID | 339 |
Feature activation | +5.38e+00 |
Pos logit contributions | |
put | +0.020 |
pretended | +0.019 |
punished | +0.019 |
cleaned | +0.019 |
used | +0.019 |
Neg logit contributions | |
arl | -0.030 |
ane | -0.029 |
elly | -0.029 |
ay | -0.028 |
Today | -0.028 |
Token | took |
Token ID | 1718 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
punished | +0.074 |
ck | +0.068 |
eling | +0.066 |
remind | +0.066 |
permission | +0.065 |
Neg logit contributions | |
elly | -0.119 |
ane | -0.116 |
ay | -0.115 |
arry | -0.107 |
� | -0.107 |
Token | her |
Token ID | 607 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
forgot | +0.012 |
hoped | +0.011 |
never | +0.011 |
remembered | +0.011 |
pretended | +0.011 |
Neg logit contributions | |
ay | -0.022 |
arl | -0.021 |
igsaw | -0.020 |
Today | -0.020 |
elly | -0.019 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
arl | +0.085 |
elly | +0.081 |
ay | +0.079 |
ane | +0.078 |
arry | +0.078 |
Neg logit contributions | |
put | -0.041 |
pretended | -0.039 |
never | -0.038 |
used | -0.037 |
made | -0.037 |
Token | ay |
Token ID | 323 |
Feature activation | -6.42e+00 |
Pos logit contributions | |
arl | +0.057 |
elly | +0.040 |
Today | +0.036 |
� | +0.035 |
owers | +0.030 |
Neg logit contributions | |
put | -0.442 |
never | -0.433 |
made | -0.432 |
used | -0.426 |
fixed | -0.425 |
Token | ons |
Token ID | 684 |
Feature activation | -5.61e+00 |
Pos logit contributions | |
arl | +0.155 |
elly | +0.147 |
Today | +0.142 |
ay | +0.142 |
arry | +0.140 |
Neg logit contributions | |
put | -0.103 |
never | -0.098 |
pretended | -0.096 |
made | -0.095 |
used | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
arl | +0.126 |
ay | +0.121 |
elly | +0.120 |
ane | +0.116 |
Today | +0.116 |
Neg logit contributions | |
put | -0.098 |
never | -0.094 |
pretended | -0.094 |
learned | -0.091 |
used | -0.091 |
Token | He |
Token ID | 679 |
Feature activation | +7.80e+00 |
Pos logit contributions | |
andel | +0.101 |
arl | +0.096 |
ute | +0.095 |
elly | +0.094 |
ane | +0.093 |
Neg logit contributions | |
put | -0.070 |
never | -0.066 |
hoped | -0.065 |
learned | -0.064 |
tried | -0.064 |
Token | said |
Token ID | 531 |
Feature activation | -6.63e-01 |
Pos logit contributions | |
care | +0.112 |
permission | +0.110 |
eling | +0.108 |
ts | +0.102 |
supposed | +0.098 |
Neg logit contributions | |
elly | -0.136 |
arry | -0.127 |
uffy | -0.127 |
ay | -0.124 |
uddle | -0.123 |
Token | " |
Token ID | 366 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
put | +0.062 |
never | +0.059 |
used | +0.058 |
pretended | +0.058 |
made | +0.058 |
Neg logit contributions | |
arl | -0.055 |
elly | -0.053 |
ay | -0.053 |
ane | -0.051 |
ute | -0.050 |
Token | Can |
Token ID | 6090 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
arl | +0.009 |
ay | +0.008 |
elly | +0.008 |
Today | +0.008 |
ane | +0.008 |
Neg logit contributions | |
put | -0.008 |
never | -0.008 |
pretended | -0.008 |
learned | -0.007 |
used | -0.007 |
Token | I |
Token ID | 314 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
put | +0.036 |
never | +0.035 |
pretended | +0.033 |
used | +0.033 |
learned | +0.033 |
Neg logit contributions | |
arl | -0.041 |
elly | -0.039 |
ay | -0.039 |
Today | -0.037 |
ane | -0.037 |
Token | take |
Token ID | 1011 |
Feature activation | +5.74e-03 |
Pos logit contributions | |
punished | +0.034 |
hoped | +0.033 |
never | +0.033 |
pretended | +0.032 |
learned | +0.032 |
Neg logit contributions | |
ay | -0.083 |
arl | -0.079 |
Today | -0.077 |
� | -0.076 |
elly | -0.075 |
Token | these |
Token ID | 777 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
forgot | +0.000 |
hoped | +0.000 |
never | +0.000 |
pretended | +0.000 |
punished | +0.000 |
Neg logit contributions | |
ay | -0.000 |
arl | -0.000 |
elly | -0.000 |
ane | -0.000 |
ummy | -0.000 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
arl | +0.119 |
ay | +0.114 |
elly | +0.114 |
Today | +0.111 |
ane | +0.111 |
Neg logit contributions | |
put | -0.063 |
never | -0.060 |
pretended | -0.059 |
used | -0.058 |
learned | -0.057 |
Token | ay |
Token ID | 323 |
Feature activation | -6.23e+00 |
Pos logit contributions | |
arl | +0.074 |
elly | +0.053 |
Today | +0.048 |
� | +0.043 |
arry | +0.038 |
Neg logit contributions | |
put | -0.422 |
made | -0.413 |
never | -0.409 |
used | -0.406 |
fixed | -0.404 |
Token | ons |
Token ID | 684 |
Feature activation | -5.56e+00 |
Pos logit contributions | |
arl | +0.154 |
ay | +0.150 |
elly | +0.148 |
ane | +0.144 |
Today | +0.144 |
Neg logit contributions | |
put | -0.094 |
never | -0.090 |
pretended | -0.089 |
learned | -0.086 |
used | -0.086 |
Token | home |
Token ID | 1363 |
Feature activation | -5.19e+00 |
Pos logit contributions | |
arl | +0.125 |
elly | +0.119 |
ay | +0.118 |
Today | +0.115 |
ane | +0.114 |
Neg logit contributions | |
put | -0.098 |
never | -0.094 |
pretended | -0.092 |
made | -0.091 |
used | -0.091 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.72e+00 |
Pos logit contributions | |
arl | +0.121 |
elly | +0.113 |
ay | +0.110 |
arry | +0.109 |
� | +0.109 |
Neg logit contributions | |
put | -0.093 |
never | -0.087 |
made | -0.086 |
used | -0.084 |
pretended | -0.083 |
Token | Her |
Token ID | 2332 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
arl | +0.101 |
elly | +0.095 |
ay | +0.089 |
igsaw | +0.088 |
ane | +0.087 |
Neg logit contributions | |
put | -0.097 |
never | -0.092 |
tried | -0.090 |
learned | -0.090 |
pretended | -0.089 |
Token | you |
Token ID | 345 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
put | +0.032 |
never | +0.031 |
pretended | +0.030 |
used | +0.030 |
made | +0.029 |
Neg logit contributions | |
arl | -0.040 |
elly | -0.038 |
ay | -0.037 |
ane | -0.036 |
Today | -0.036 |
Token | help |
Token ID | 1037 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
hoped | +0.021 |
pretended | +0.021 |
forgot | +0.020 |
learned | +0.020 |
never | +0.020 |
Neg logit contributions | |
ay | -0.052 |
arl | -0.051 |
� | -0.049 |
Today | -0.049 |
ute | -0.048 |
Token | me |
Token ID | 502 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
ay | +0.028 |
arl | +0.028 |
ummy | +0.027 |
elly | +0.026 |
ute | +0.026 |
Neg logit contributions | |
pretended | -0.019 |
hoped | -0.019 |
learned | -0.019 |
forgot | -0.019 |
remembered | -0.018 |
Token | get |
Token ID | 651 |
Feature activation | +8.42e-01 |
Pos logit contributions | |
arl | +0.019 |
ay | +0.019 |
ute | +0.018 |
Today | +0.018 |
elly | +0.018 |
Neg logit contributions | |
pretended | -0.010 |
hoped | -0.010 |
never | -0.010 |
learned | -0.010 |
forgot | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
pretended | +0.013 |
forgot | +0.012 |
hoped | +0.012 |
learned | +0.012 |
never | +0.012 |
Neg logit contributions | |
arl | -0.020 |
ay | -0.020 |
elly | -0.019 |
ummy | -0.019 |
ore | -0.019 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
arl | +0.091 |
ay | +0.089 |
elly | +0.087 |
Today | +0.085 |
ane | +0.084 |
Neg logit contributions | |
put | -0.054 |
never | -0.052 |
pretended | -0.052 |
learned | -0.050 |
punished | -0.050 |
Token | ay |
Token ID | 323 |
Feature activation | -6.30e+00 |
Pos logit contributions | |
arl | +0.071 |
elly | +0.057 |
Today | +0.046 |
aris | +0.044 |
� | +0.043 |
Neg logit contributions | |
put | -0.420 |
made | -0.413 |
fixed | -0.404 |
used | -0.404 |
never | -0.402 |
Token | ons |
Token ID | 684 |
Feature activation | -4.98e+00 |
Pos logit contributions | |
arl | +0.155 |
elly | +0.147 |
Today | +0.142 |
ay | +0.141 |
arry | +0.140 |
Neg logit contributions | |
put | -0.099 |
never | -0.094 |
pretended | -0.092 |
made | -0.092 |
used | -0.091 |
Token | from |
Token ID | 422 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
arl | +0.110 |
elly | +0.105 |
ay | +0.103 |
Today | +0.102 |
arry | +0.100 |
Neg logit contributions | |
put | -0.091 |
never | -0.086 |
made | -0.085 |
used | -0.084 |
pretended | -0.084 |
Token | the |
Token ID | 262 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
pretended | +0.012 |
forgot | +0.012 |
learned | +0.012 |
put | +0.012 |
never | +0.011 |
Neg logit contributions | |
arl | -0.029 |
ay | -0.028 |
elly | -0.027 |
ummy | -0.027 |
Today | -0.027 |
Token | cabinet |
Token ID | 13447 |
Feature activation | -7.22e+00 |
Pos logit contributions | |
arl | +0.047 |
ay | +0.044 |
elly | +0.044 |
ane | +0.042 |
Today | +0.042 |
Neg logit contributions | |
put | -0.068 |
never | -0.066 |
pretended | -0.065 |
used | -0.064 |
learned | -0.064 |
Token | 'm |
Token ID | 1101 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
punished | +0.058 |
never | +0.058 |
ts | +0.057 |
hoped | +0.057 |
ck | +0.057 |
Neg logit contributions | |
elly | -0.063 |
ay | -0.060 |
Today | -0.060 |
� | -0.059 |
arl | -0.059 |
Token | trying |
Token ID | 2111 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
pretended | +0.035 |
learned | +0.035 |
hoped | +0.035 |
wed | +0.035 |
shared | +0.035 |
Neg logit contributions | |
arl | -0.066 |
� | -0.064 |
ay | -0.063 |
Today | -0.063 |
ute | -0.062 |
Token | to |
Token ID | 284 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
arl | +0.003 |
ay | +0.003 |
elly | +0.003 |
Today | +0.003 |
ute | +0.003 |
Neg logit contributions | |
put | -0.003 |
pretended | -0.003 |
never | -0.003 |
punished | -0.003 |
hoped | -0.003 |
Token | get |
Token ID | 651 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
hoped | +0.050 |
forgot | +0.047 |
remembered | +0.047 |
pretended | +0.047 |
learned | +0.046 |
Neg logit contributions | |
ay | -0.100 |
ute | -0.097 |
arl | -0.094 |
� | -0.090 |
arry | -0.090 |
Token | the |
Token ID | 262 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
hoped | +0.029 |
pretended | +0.029 |
forgot | +0.029 |
never | +0.028 |
remembered | +0.027 |
Neg logit contributions | |
arl | -0.042 |
ay | -0.040 |
ore | -0.040 |
elly | -0.039 |
igsaw | -0.039 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
arl | +0.084 |
ay | +0.083 |
elly | +0.079 |
Today | +0.079 |
ane | +0.078 |
Neg logit contributions | |
put | -0.064 |
never | -0.063 |
learned | -0.061 |
pretended | -0.061 |
hoped | -0.060 |
Token | ay |
Token ID | 323 |
Feature activation | -5.92e+00 |
Pos logit contributions | |
arl | +0.070 |
elly | +0.060 |
Today | +0.051 |
ute | +0.050 |
� | +0.050 |
Neg logit contributions | |
put | -0.409 |
made | -0.405 |
fixed | -0.399 |
used | -0.396 |
never | -0.394 |
Token | ons |
Token ID | 684 |
Feature activation | -4.71e+00 |
Pos logit contributions | |
arl | +0.149 |
elly | +0.142 |
Today | +0.137 |
ay | +0.136 |
ute | +0.135 |
Neg logit contributions | |
put | -0.090 |
never | -0.084 |
pretended | -0.084 |
made | -0.083 |
used | -0.082 |
Token | from |
Token ID | 422 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
arl | +0.108 |
elly | +0.103 |
ay | +0.103 |
Today | +0.100 |
ane | +0.099 |
Neg logit contributions | |
put | -0.080 |
never | -0.077 |
pretended | -0.076 |
used | -0.074 |
made | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
put | +0.007 |
pretended | +0.006 |
never | +0.006 |
learned | +0.006 |
forgot | +0.006 |
Neg logit contributions | |
arl | -0.017 |
ay | -0.017 |
elly | -0.016 |
ane | -0.016 |
Today | -0.016 |
Token | cabinet |
Token ID | 13447 |
Feature activation | -6.42e+00 |
Pos logit contributions | |
arl | +0.051 |
ay | +0.048 |
elly | +0.047 |
Today | +0.046 |
� | +0.044 |
Neg logit contributions | |
put | -0.085 |
pretended | -0.083 |
never | -0.083 |
learned | -0.081 |
used | -0.081 |
Token | a |
Token ID | 257 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
forgot | +0.009 |
pretended | +0.009 |
put | +0.009 |
never | +0.008 |
punished | +0.008 |
Neg logit contributions | |
ay | -0.011 |
arl | -0.010 |
elly | -0.010 |
Today | -0.010 |
ane | -0.010 |
Token | big |
Token ID | 1263 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
arl | +0.012 |
elly | +0.011 |
ay | +0.011 |
Today | +0.011 |
ane | +0.011 |
Neg logit contributions | |
put | -0.009 |
never | -0.008 |
pretended | -0.008 |
used | -0.008 |
made | -0.008 |
Token | box |
Token ID | 3091 |
Feature activation | -5.26e+00 |
Pos logit contributions | |
arl | +0.057 |
elly | +0.054 |
ay | +0.053 |
Today | +0.052 |
ane | +0.052 |
Neg logit contributions | |
put | -0.048 |
never | -0.045 |
pretended | -0.045 |
used | -0.044 |
learned | -0.044 |
Token | of |
Token ID | 286 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
arl | +0.095 |
ay | +0.093 |
elly | +0.091 |
ane | +0.088 |
Today | +0.087 |
Neg logit contributions | |
put | -0.110 |
pretended | -0.109 |
never | -0.108 |
punished | -0.107 |
learned | -0.107 |
Token | ch |
Token ID | 442 |
Feature activation | -7.09e+00 |
Pos logit contributions | |
ay | +0.011 |
elly | +0.010 |
ummy | +0.010 |
Today | +0.010 |
arl | +0.010 |
Neg logit contributions | |
forgot | -0.010 |
never | -0.010 |
remembered | -0.010 |
managed | -0.009 |
hoped | -0.009 |
Token | ocol |
Token ID | 4668 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
arl | +0.100 |
Today | +0.099 |
� | +0.098 |
ay | +0.089 |
elly | +0.088 |
Neg logit contributions | |
put | -0.176 |
used | -0.166 |
never | -0.165 |
learned | -0.165 |
cleaned | -0.164 |
Token | ates |
Token ID | 689 |
Feature activation | -5.78e+00 |
Pos logit contributions | |
arl | +0.267 |
Today | +0.259 |
elly | +0.254 |
ay | +0.250 |
� | +0.244 |
Neg logit contributions | |
put | -0.231 |
never | -0.229 |
used | -0.225 |
pretended | -0.223 |
cleaned | -0.220 |
Token | to |
Token ID | 284 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
arl | +0.128 |
ay | +0.124 |
elly | +0.124 |
ane | +0.121 |
Today | +0.119 |
Neg logit contributions | |
put | -0.100 |
never | -0.097 |
pretended | -0.096 |
learned | -0.094 |
punished | -0.093 |
Token | share |
Token ID | 2648 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
forgot | +0.027 |
punished | +0.026 |
remembered | +0.026 |
pretended | +0.026 |
never | +0.025 |
Neg logit contributions | |
ay | -0.056 |
arl | -0.055 |
Today | -0.054 |
� | -0.052 |
ore | -0.051 |
Token | with |
Token ID | 351 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
arl | +0.034 |
ay | +0.034 |
elly | +0.033 |
Today | +0.032 |
ane | +0.032 |
Neg logit contributions | |
put | -0.022 |
pretended | -0.022 |
never | -0.022 |
punished | -0.022 |
forgot | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
pretended | +0.002 |
put | +0.002 |
punished | +0.002 |
cleaned | +0.002 |
forgot | +0.002 |
Neg logit contributions | |
ay | -0.005 |
elly | -0.005 |
ummy | -0.004 |
ore | -0.004 |
arl | -0.004 |
Token | He |
Token ID | 679 |
Feature activation | +8.48e+00 |
Pos logit contributions | |
ane | +0.098 |
andel | +0.095 |
owers | +0.094 |
ute | +0.094 |
ay | +0.092 |
Neg logit contributions | |
put | -0.067 |
his | -0.063 |
made | -0.061 |
used | -0.061 |
tried | -0.060 |
Token | remembers |
Token ID | 18140 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
never | +0.126 |
pretended | +0.122 |
punished | +0.122 |
cleaned | +0.120 |
ts | +0.119 |
Neg logit contributions | |
arl | -0.180 |
elly | -0.171 |
Today | -0.167 |
arry | -0.157 |
uddle | -0.155 |
Token | he |
Token ID | 339 |
Feature activation | +5.42e+00 |
Pos logit contributions | |
put | +0.030 |
pretended | +0.030 |
punished | +0.029 |
never | +0.028 |
learned | +0.028 |
Neg logit contributions | |
arl | -0.058 |
elly | -0.056 |
Today | -0.053 |
ummy | -0.053 |
ay | -0.053 |
Token | borrowed |
Token ID | 22546 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
pretended | +0.063 |
punished | +0.062 |
ts | +0.061 |
never | +0.059 |
ck | +0.056 |
Neg logit contributions | |
elly | -0.125 |
arl | -0.116 |
ay | -0.115 |
ore | -0.113 |
ane | -0.113 |
Token | the |
Token ID | 262 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
ay | +0.042 |
arl | +0.041 |
elly | +0.040 |
ummy | +0.039 |
Today | +0.039 |
Neg logit contributions | |
forgot | -0.019 |
pretended | -0.018 |
never | -0.018 |
punished | -0.017 |
remembered | -0.017 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
arl | +0.089 |
ay | +0.087 |
elly | +0.085 |
Today | +0.084 |
ane | +0.083 |
Neg logit contributions | |
put | -0.066 |
never | -0.065 |
pretended | -0.064 |
punished | -0.063 |
learned | -0.062 |
Token | ay |
Token ID | 323 |
Feature activation | -6.60e+00 |
Pos logit contributions | |
arl | +0.070 |
elly | +0.048 |
� | +0.045 |
Today | +0.042 |
ute | +0.040 |
Neg logit contributions | |
put | -0.425 |
made | -0.418 |
never | -0.413 |
used | -0.411 |
fixed | -0.407 |
Token | on |
Token ID | 261 |
Feature activation | -7.26e+00 |
Pos logit contributions | |
arl | +0.154 |
elly | +0.146 |
ay | +0.145 |
Today | +0.141 |
ane | +0.141 |
Neg logit contributions | |
put | -0.111 |
never | -0.106 |
pretended | -0.105 |
used | -0.103 |
learned | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
arl | +0.168 |
elly | +0.159 |
ay | +0.159 |
Today | +0.154 |
ane | +0.154 |
Neg logit contributions | |
put | -0.124 |
never | -0.119 |
pretended | -0.117 |
used | -0.115 |
made | -0.114 |
Token |
|
Token ID | 198 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
ane | +0.094 |
owers | +0.093 |
andel | +0.093 |
ute | +0.090 |
ay | +0.088 |
Neg logit contributions | |
put | -0.066 |
his | -0.060 |
learned | -0.060 |
used | -0.060 |
hoped | -0.059 |
Token |
|
Token ID | 198 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
arl | +0.089 |
ane | +0.089 |
owers | +0.089 |
elly | +0.088 |
� | +0.087 |
Neg logit contributions | |
put | -0.100 |
never | -0.095 |
learned | -0.093 |
used | -0.091 |
made | -0.089 |
Token | they |
Token ID | 484 |
Feature activation | +6.08e+00 |
Pos logit contributions | |
put | +0.028 |
punished | +0.028 |
never | +0.027 |
cleaned | +0.027 |
pretended | +0.027 |
Neg logit contributions | |
ay | -0.031 |
elly | -0.031 |
arl | -0.030 |
Today | -0.030 |
ane | -0.029 |
Token | disagreed |
Token ID | 27884 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
put | +0.076 |
punished | +0.072 |
pretended | +0.071 |
used | +0.069 |
cleaned | +0.069 |
Neg logit contributions | |
ay | -0.144 |
arl | -0.144 |
ane | -0.141 |
elly | -0.141 |
� | -0.137 |
Token | on |
Token ID | 319 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
arl | +0.059 |
ay | +0.056 |
elly | +0.056 |
ane | +0.055 |
Today | +0.054 |
Neg logit contributions | |
put | -0.035 |
never | -0.034 |
made | -0.032 |
used | -0.032 |
pretended | -0.032 |
Token | something |
Token ID | 1223 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
put | +0.007 |
pretended | +0.006 |
never | +0.006 |
used | +0.006 |
learned | +0.006 |
Neg logit contributions | |
arl | -0.019 |
ay | -0.018 |
elly | -0.018 |
Today | -0.018 |
ane | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +6.29e+00 |
Pos logit contributions | |
arl | +0.024 |
ay | +0.023 |
elly | +0.023 |
Today | +0.023 |
ane | +0.023 |
Neg logit contributions | |
put | -0.017 |
never | -0.016 |
pretended | -0.016 |
used | -0.015 |
learned | -0.015 |
Token | they |
Token ID | 484 |
Feature activation | +7.78e+00 |
Pos logit contributions | |
wed | +0.057 |
put | +0.055 |
never | +0.055 |
cleaned | +0.054 |
pretended | +0.054 |
Neg logit contributions | |
elly | -0.148 |
arl | -0.147 |
ay | -0.145 |
Today | -0.144 |
ane | -0.144 |
Token | began |
Token ID | 2540 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
wed | +0.067 |
put | +0.065 |
ws | +0.065 |
ck | +0.064 |
cleaned | +0.063 |
Neg logit contributions | |
ay | -0.172 |
ifying | -0.168 |
ane | -0.168 |
arry | -0.165 |
arl | -0.163 |
Token | to |
Token ID | 284 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
put | +0.048 |
pretended | +0.047 |
never | +0.046 |
hoped | +0.046 |
cleaned | +0.046 |
Neg logit contributions | |
arl | -0.049 |
elly | -0.048 |
Today | -0.047 |
ay | -0.047 |
ute | -0.046 |
Token | scatter |
Token ID | 41058 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
put | +0.066 |
pretended | +0.065 |
hoped | +0.065 |
used | +0.064 |
cleaned | +0.064 |
Neg logit contributions | |
arl | -0.108 |
Today | -0.107 |
ay | -0.102 |
� | -0.101 |
ute | -0.101 |
Token | in |
Token ID | 287 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
arl | +0.030 |
ay | +0.029 |
elly | +0.029 |
Today | +0.028 |
ane | +0.027 |
Neg logit contributions | |
put | -0.015 |
never | -0.015 |
pretended | -0.015 |
learned | -0.014 |
forgot | -0.014 |
Token | different |
Token ID | 1180 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
put | +0.015 |
pretended | +0.014 |
never | +0.014 |
forgot | +0.014 |
punished | +0.014 |
Neg logit contributions | |
arl | -0.035 |
ay | -0.035 |
elly | -0.034 |
Today | -0.034 |
� | -0.034 |
Token | little |
Token ID | 1310 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
arl | +0.042 |
elly | +0.040 |
ay | +0.040 |
ane | +0.039 |
Today | +0.038 |
Neg logit contributions | |
put | -0.038 |
never | -0.036 |
pretended | -0.035 |
learned | -0.035 |
used | -0.035 |
Token | boy |
Token ID | 2933 |
Feature activation | +5.52e+00 |
Pos logit contributions | |
arl | +0.045 |
elly | +0.044 |
ay | +0.044 |
Today | +0.042 |
ane | +0.042 |
Neg logit contributions | |
put | -0.050 |
never | -0.050 |
pretended | -0.049 |
learned | -0.048 |
punished | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
punished | +0.066 |
pretended | +0.065 |
wed | +0.063 |
ck | +0.063 |
ts | +0.062 |
Neg logit contributions | |
ay | -0.128 |
ane | -0.125 |
� | -0.119 |
Today | -0.118 |
arry | -0.118 |
Token | his |
Token ID | 465 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
never | +0.024 |
put | +0.024 |
pretended | +0.024 |
punished | +0.023 |
used | +0.023 |
Neg logit contributions | |
arl | -0.094 |
ay | -0.093 |
Today | -0.091 |
ane | -0.090 |
elly | -0.089 |
Token | mom |
Token ID | 1995 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
put | +0.002 |
never | +0.002 |
pretended | +0.002 |
learned | +0.002 |
used | +0.002 |
Neg logit contributions | |
arl | -0.004 |
ay | -0.003 |
elly | -0.003 |
Today | -0.003 |
ane | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +4.41e+00 |
Pos logit contributions | |
punished | +0.050 |
ts | +0.046 |
pretended | +0.045 |
ck | +0.045 |
hoped | +0.043 |
Neg logit contributions | |
ay | -0.107 |
ane | -0.100 |
ummy | -0.095 |
arry | -0.093 |
aris | -0.092 |
Token | dad |
Token ID | 9955 |
Feature activation | +4.83e+00 |
Pos logit contributions | |
never | +0.075 |
punished | +0.073 |
used | +0.072 |
pretended | +0.071 |
forgot | +0.071 |
Neg logit contributions | |
ay | -0.090 |
arl | -0.089 |
ane | -0.086 |
Today | -0.083 |
ute | -0.082 |
Token | had |
Token ID | 550 |
Feature activation | +5.03e+00 |
Pos logit contributions | |
punished | +0.068 |
ck | +0.059 |
wounds | +0.057 |
pretended | +0.056 |
put | +0.056 |
Neg logit contributions | |
ay | -0.109 |
ane | -0.104 |
aris | -0.095 |
owers | -0.094 |
arry | -0.093 |
Token | a |
Token ID | 257 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
ck | +0.084 |
punished | +0.081 |
pretended | +0.077 |
forgot | +0.076 |
put | +0.074 |
Neg logit contributions | |
ay | -0.100 |
aris | -0.097 |
Today | -0.088 |
� | -0.086 |
igsaw | -0.085 |
Token | big |
Token ID | 1263 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
put | +0.024 |
never | +0.023 |
pretended | +0.023 |
learned | +0.022 |
used | +0.022 |
Neg logit contributions | |
arl | -0.034 |
elly | -0.033 |
ay | -0.033 |
ane | -0.032 |
Today | -0.032 |
Token | celebration |
Token ID | 14395 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
arl | +0.017 |
ay | +0.017 |
elly | +0.017 |
Today | +0.016 |
ane | +0.016 |
Neg logit contributions | |
put | -0.014 |
never | -0.014 |
pretended | -0.014 |
used | -0.013 |
learned | -0.013 |
Token | Sunday |
Token ID | 3502 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
arl | +0.063 |
ute | +0.060 |
elly | +0.059 |
ane | +0.057 |
ay | +0.056 |
Neg logit contributions | |
never | -0.055 |
learned | -0.054 |
put | -0.054 |
used | -0.051 |
hoped | -0.050 |
Token | was |
Token ID | 373 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
arl | +0.029 |
elly | +0.027 |
ay | +0.027 |
Today | +0.026 |
ane | +0.026 |
Neg logit contributions | |
put | -0.020 |
never | -0.020 |
pretended | -0.019 |
learned | -0.019 |
made | -0.019 |
Token | very |
Token ID | 845 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
put | +0.046 |
pretended | +0.045 |
learned | +0.042 |
used | +0.042 |
hoped | +0.042 |
Neg logit contributions | |
arl | -0.099 |
ay | -0.099 |
Today | -0.095 |
elly | -0.095 |
ane | -0.092 |
Token | proud |
Token ID | 6613 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
put | +0.043 |
pretended | +0.041 |
never | +0.041 |
used | +0.040 |
learned | +0.039 |
Neg logit contributions | |
arl | -0.084 |
ay | -0.080 |
elly | -0.080 |
Today | -0.078 |
ane | -0.077 |
Token | and |
Token ID | 290 |
Feature activation | +5.77e+00 |
Pos logit contributions | |
arl | +0.027 |
ay | +0.026 |
elly | +0.026 |
Today | +0.025 |
ane | +0.025 |
Neg logit contributions | |
put | -0.040 |
never | -0.039 |
pretended | -0.039 |
learned | -0.038 |
used | -0.038 |
Token | everyone |
Token ID | 2506 |
Feature activation | +5.27e+00 |
Pos logit contributions | |
pretended | +0.077 |
put | +0.076 |
wed | +0.075 |
punished | +0.074 |
cleaned | +0.072 |
Neg logit contributions | |
ay | -0.136 |
arl | -0.131 |
Today | -0.128 |
elly | -0.128 |
ummy | -0.125 |
Token | praised |
Token ID | 15342 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
pretended | +0.061 |
punished | +0.059 |
wed | +0.058 |
put | +0.058 |
ck | +0.055 |
Neg logit contributions | |
ane | -0.128 |
ore | -0.126 |
ay | -0.122 |
elly | -0.121 |
apping | -0.118 |
Token | him |
Token ID | 683 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
arl | +0.012 |
elly | +0.012 |
ane | +0.012 |
ay | +0.012 |
ute | +0.011 |
Neg logit contributions | |
put | -0.004 |
never | -0.003 |
used | -0.003 |
made | -0.003 |
learned | -0.003 |
Token | for |
Token ID | 329 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
arl | +0.006 |
ay | +0.006 |
elly | +0.006 |
Today | +0.005 |
ane | +0.005 |
Neg logit contributions | |
put | -0.003 |
never | -0.003 |
pretended | -0.003 |
used | -0.003 |
learned | -0.003 |
Token | being |
Token ID | 852 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
pretended | +0.041 |
forgot | +0.040 |
put | +0.036 |
hoped | +0.036 |
remembered | +0.035 |
Neg logit contributions | |
arl | -0.076 |
� | -0.070 |
ay | -0.070 |
apping | -0.067 |
ore | -0.066 |
Token | so |
Token ID | 523 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
pretended | +0.056 |
put | +0.055 |
forgot | +0.052 |
hoped | +0.050 |
never | +0.050 |
Neg logit contributions | |
ay | -0.091 |
arl | -0.090 |
elly | -0.089 |
apping | -0.085 |
ane | -0.084 |
Token | to |
Token ID | 284 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
arl | +0.021 |
ane | +0.021 |
elly | +0.021 |
owers | +0.020 |
ute | +0.020 |
Neg logit contributions | |
put | -0.014 |
never | -0.013 |
made | -0.013 |
used | -0.013 |
tried | -0.012 |
Token | her |
Token ID | 607 |
Feature activation | -7.51e-01 |
Pos logit contributions | |
pretended | +0.021 |
forgot | +0.020 |
punished | +0.019 |
used | +0.019 |
hoped | +0.019 |
Neg logit contributions | |
arl | -0.065 |
ay | -0.064 |
Today | -0.062 |
� | -0.061 |
ummy | -0.061 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
arl | +0.019 |
elly | +0.018 |
ay | +0.018 |
ane | +0.018 |
Today | +0.018 |
Neg logit contributions | |
put | -0.011 |
never | -0.011 |
pretended | -0.011 |
used | -0.010 |
learned | -0.010 |
Token | in |
Token ID | 287 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
arl | +0.023 |
elly | +0.022 |
ay | +0.022 |
ute | +0.022 |
ane | +0.022 |
Neg logit contributions | |
put | -0.020 |
never | -0.019 |
pretended | -0.018 |
used | -0.018 |
made | -0.018 |
Token | tears |
Token ID | 10953 |
Feature activation | -9.63e-01 |
Pos logit contributions | |
put | +0.028 |
never | +0.026 |
made | +0.025 |
pretended | +0.025 |
learned | +0.025 |
Neg logit contributions | |
arl | -0.077 |
elly | -0.074 |
ay | -0.073 |
ane | -0.073 |
ute | -0.073 |
Token | and |
Token ID | 290 |
Feature activation | +6.98e+00 |
Pos logit contributions | |
arl | +0.021 |
ay | +0.020 |
elly | +0.020 |
Today | +0.020 |
ane | +0.020 |
Neg logit contributions | |
put | -0.017 |
never | -0.017 |
pretended | -0.016 |
used | -0.016 |
learned | -0.016 |
Token | asked |
Token ID | 1965 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
wed | +0.075 |
ck | +0.067 |
ts | +0.064 |
never | +0.061 |
wy | +0.061 |
Neg logit contributions | |
ay | -0.151 |
Today | -0.150 |
� | -0.148 |
arry | -0.147 |
ummy | -0.146 |
Token | , |
Token ID | 11 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
put | +0.007 |
pretended | +0.007 |
never | +0.006 |
used | +0.006 |
punished | +0.006 |
Neg logit contributions | |
arl | -0.020 |
ay | -0.020 |
elly | -0.019 |
Today | -0.019 |
ane | -0.019 |
Token | why |
Token ID | 1521 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
put | +0.034 |
never | +0.033 |
used | +0.032 |
made | +0.031 |
learned | +0.031 |
Neg logit contributions | |
arl | -0.032 |
elly | -0.032 |
ay | -0.032 |
ane | -0.031 |
ute | -0.030 |
Token | would |
Token ID | 561 |
Feature activation | +4.17e+00 |
Pos logit contributions | |
put | +0.001 |
pretended | +0.001 |
never | +0.001 |
punished | +0.001 |
learned | +0.001 |
Neg logit contributions | |
arl | -0.001 |
ay | -0.001 |
elly | -0.001 |
Today | -0.001 |
ane | -0.001 |
Token | you |
Token ID | 345 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
pretended | +0.051 |
wed | +0.048 |
hoped | +0.045 |
learned | +0.043 |
cleaned | +0.043 |
Neg logit contributions | |
ummy | -0.100 |
� | -0.100 |
arl | -0.098 |
Today | -0.098 |
ay | -0.096 |
Token | Tim |
Token ID | 5045 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
arl | +0.053 |
elly | +0.051 |
ay | +0.050 |
ane | +0.050 |
ute | +0.050 |
Neg logit contributions | |
put | -0.027 |
pretended | -0.025 |
never | -0.025 |
used | -0.024 |
made | -0.024 |
Token | my |
Token ID | 1820 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
elly | +0.054 |
arl | +0.054 |
ay | +0.054 |
ummy | +0.052 |
Today | +0.052 |
Neg logit contributions | |
punished | -0.042 |
pretended | -0.042 |
hoped | -0.041 |
wished | -0.040 |
never | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
ay | +0.022 |
elly | +0.022 |
arl | +0.022 |
Today | +0.021 |
ummy | +0.020 |
Neg logit contributions | |
punished | -0.027 |
pretended | -0.026 |
forgot | -0.026 |
put | -0.026 |
hoped | -0.025 |
Token | Sally |
Token ID | 25737 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
wed | +0.073 |
punished | +0.072 |
ck | +0.071 |
eling | +0.068 |
put | +0.067 |
Neg logit contributions | |
Today | -0.058 |
arl | -0.057 |
arry | -0.055 |
ay | -0.054 |
igsaw | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -4.98e+00 |
Pos logit contributions | |
arl | +0.020 |
ay | +0.019 |
elly | +0.019 |
Today | +0.018 |
arry | +0.018 |
Neg logit contributions | |
put | -0.021 |
pretended | -0.021 |
punished | -0.021 |
never | -0.020 |
hoped | -0.020 |
Token | They |
Token ID | 1119 |
Feature activation | +6.37e+00 |
Pos logit contributions | |
arl | +0.097 |
ane | +0.094 |
ute | +0.093 |
elly | +0.092 |
ay | +0.091 |
Neg logit contributions | |
put | -0.105 |
pretended | -0.101 |
never | -0.099 |
used | -0.098 |
made | -0.097 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
punished | +0.096 |
wounds | +0.093 |
ck | +0.092 |
eling | +0.086 |
tightened | +0.086 |
Neg logit contributions | |
ay | -0.133 |
owers | -0.131 |
Today | -0.126 |
ane | -0.123 |
arl | -0.123 |
Token | to |
Token ID | 284 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
punished | +0.026 |
put | +0.026 |
forgot | +0.025 |
never | +0.025 |
pretended | +0.025 |
Neg logit contributions | |
arl | -0.025 |
ay | -0.025 |
elly | -0.024 |
ummy | -0.023 |
Today | -0.023 |
Token | play |
Token ID | 711 |
Feature activation | -3.76e+00 |
Pos logit contributions | |
punished | +0.035 |
never | +0.035 |
put | +0.034 |
hoped | +0.033 |
learned | +0.033 |
Neg logit contributions | |
arl | -0.075 |
ay | -0.073 |
Today | -0.073 |
� | -0.071 |
ute | -0.070 |
Token | together |
Token ID | 1978 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
arl | +0.090 |
ay | +0.086 |
elly | +0.086 |
ane | +0.083 |
Today | +0.083 |
Neg logit contributions | |
put | -0.062 |
never | -0.059 |
pretended | -0.059 |
used | -0.057 |
learned | -0.057 |
Token | every |
Token ID | 790 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
arl | +0.032 |
elly | +0.030 |
ay | +0.030 |
ane | +0.030 |
Today | +0.030 |
Neg logit contributions | |
put | -0.018 |
never | -0.018 |
pretended | -0.017 |
used | -0.017 |
made | -0.017 |
Token | you |
Token ID | 345 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
put | +0.011 |
pretended | +0.011 |
never | +0.010 |
used | +0.010 |
learned | +0.010 |
Neg logit contributions | |
arl | -0.016 |
ay | -0.016 |
elly | -0.016 |
Today | -0.015 |
ane | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
arl | +0.004 |
ay | +0.004 |
elly | +0.004 |
Today | +0.004 |
ane | +0.004 |
Neg logit contributions | |
put | -0.003 |
pretended | -0.003 |
never | -0.003 |
used | -0.003 |
punished | -0.003 |
Token | too |
Token ID | 1165 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
put | +0.012 |
never | +0.011 |
pretended | +0.011 |
used | +0.011 |
learned | +0.011 |
Neg logit contributions | |
arl | -0.018 |
ay | -0.017 |
elly | -0.017 |
Today | -0.017 |
ane | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
ane | +0.017 |
arl | +0.017 |
anger | +0.017 |
ute | +0.017 |
ay | +0.016 |
Neg logit contributions | |
put | -0.008 |
never | -0.008 |
used | -0.008 |
made | -0.007 |
learned | -0.007 |
Token | You |
Token ID | 921 |
Feature activation | +4.45e+00 |
Pos logit contributions | |
arl | +0.053 |
elly | +0.050 |
ay | +0.049 |
ute | +0.049 |
ane | +0.049 |
Neg logit contributions | |
put | -0.065 |
never | -0.065 |
learned | -0.062 |
pretended | -0.062 |
made | -0.062 |
Token | are |
Token ID | 389 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
ck | +0.099 |
eling | +0.087 |
wed | +0.086 |
asking | +0.086 |
forth | +0.085 |
Neg logit contributions | |
ay | -0.066 |
ane | -0.055 |
owers | -0.053 |
arl | -0.053 |
� | -0.053 |
Token | good |
Token ID | 922 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
wed | +0.047 |
pretended | +0.046 |
forgot | +0.045 |
tried | +0.043 |
imagined | +0.043 |
Neg logit contributions | |
arl | -0.061 |
ay | -0.060 |
ummy | -0.056 |
Today | -0.054 |
elly | -0.054 |
Token | children |
Token ID | 1751 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
arl | +0.032 |
igsaw | +0.031 |
ay | +0.031 |
� | +0.031 |
elly | +0.030 |
Neg logit contributions | |
put | -0.015 |
made | -0.014 |
never | -0.014 |
pretended | -0.014 |
learned | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
arl | +0.036 |
ay | +0.035 |
elly | +0.035 |
ane | +0.034 |
Today | +0.033 |
Neg logit contributions | |
put | -0.032 |
pretended | -0.031 |
never | -0.031 |
used | -0.030 |
punished | -0.030 |
Token | Now |
Token ID | 2735 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
arl | +0.059 |
ay | +0.054 |
ute | +0.054 |
ane | +0.054 |
elly | +0.054 |
Neg logit contributions | |
put | -0.071 |
never | -0.070 |
learned | -0.068 |
pretended | -0.067 |
made | -0.067 |
Token | , |
Token ID | 11 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
put | +0.032 |
pretended | +0.032 |
never | +0.032 |
used | +0.032 |
learned | +0.032 |
Neg logit contributions | |
arl | -0.047 |
ay | -0.046 |
Today | -0.044 |
elly | -0.044 |
ane | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
put | +0.017 |
pretended | +0.017 |
never | +0.017 |
punished | +0.017 |
learned | +0.016 |
Neg logit contributions | |
arl | -0.019 |
ay | -0.019 |
elly | -0.019 |
Today | -0.018 |
ummy | -0.018 |
Token | try |
Token ID | 1949 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
pretended | +0.048 |
hoped | +0.045 |
punished | +0.044 |
never | +0.044 |
forgot | +0.043 |
Neg logit contributions | |
arl | -0.102 |
� | -0.100 |
Today | -0.098 |
ay | -0.097 |
ummy | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
arl | +0.009 |
ay | +0.009 |
elly | +0.009 |
ane | +0.008 |
Today | +0.008 |
Neg logit contributions | |
put | -0.005 |
never | -0.005 |
pretended | -0.005 |
used | -0.005 |
learned | -0.005 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
arl | +0.056 |
ute | +0.054 |
ane | +0.053 |
elly | +0.053 |
ay | +0.053 |
Neg logit contributions | |
put | -0.050 |
never | -0.048 |
pretended | -0.047 |
learned | -0.046 |
used | -0.046 |
Token | took |
Token ID | 1718 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
ck | +0.054 |
punished | +0.053 |
permission | +0.052 |
ts | +0.050 |
remind | +0.049 |
Neg logit contributions | |
elly | -0.108 |
� | -0.106 |
Today | -0.104 |
ay | -0.104 |
ummy | -0.104 |
Token | a |
Token ID | 257 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
pretended | +0.030 |
forgot | +0.029 |
never | +0.029 |
put | +0.029 |
hoped | +0.028 |
Neg logit contributions | |
ay | -0.061 |
arl | -0.060 |
elly | -0.058 |
Today | -0.058 |
ane | -0.056 |
Token | bite |
Token ID | 13197 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
put | +0.000 |
used | +0.000 |
tried | +0.000 |
made | +0.000 |
cleaned | +0.000 |
Neg logit contributions | |
arl | -0.001 |
elly | -0.001 |
ane | -0.001 |
arry | -0.001 |
owers | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +6.71e+00 |
Pos logit contributions | |
arl | +0.099 |
ay | +0.093 |
elly | +0.093 |
ane | +0.091 |
Today | +0.091 |
Neg logit contributions | |
put | -0.083 |
never | -0.079 |
pretended | -0.078 |
made | -0.077 |
used | -0.077 |
Token | said |
Token ID | 531 |
Feature activation | -8.55e-01 |
Pos logit contributions | |
wed | +0.071 |
punished | +0.069 |
never | +0.068 |
permission | +0.067 |
ck | +0.062 |
Neg logit contributions | |
� | -0.148 |
elly | -0.140 |
ummy | -0.136 |
ore | -0.135 |
oles | -0.135 |
Token | , |
Token ID | 11 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
arl | +0.019 |
elly | +0.018 |
ay | +0.018 |
Today | +0.017 |
ane | +0.017 |
Neg logit contributions | |
put | -0.017 |
never | -0.016 |
made | -0.016 |
learned | -0.015 |
used | -0.015 |
Token | " |
Token ID | 366 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
put | +0.037 |
never | +0.036 |
made | +0.034 |
pretended | +0.034 |
tried | +0.034 |
Neg logit contributions | |
elly | -0.039 |
arl | -0.039 |
ay | -0.039 |
ane | -0.037 |
ute | -0.037 |
Token | that |
Token ID | 326 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
pretended | +0.037 |
put | +0.037 |
learned | +0.034 |
never | +0.034 |
punished | +0.034 |
Neg logit contributions | |
arl | -0.062 |
elly | -0.060 |
ay | -0.060 |
ane | -0.059 |
ute | -0.058 |
Token | even |
Token ID | 772 |
Feature activation | +4.14e+00 |
Pos logit contributions | |
pretended | +0.024 |
put | +0.023 |
punished | +0.023 |
hoped | +0.023 |
forgot | +0.022 |
Neg logit contributions | |
ay | -0.046 |
Today | -0.046 |
elly | -0.046 |
ane | -0.045 |
ummy | -0.045 |
Token | if |
Token ID | 611 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
never | +0.061 |
put | +0.060 |
pretended | +0.060 |
hoped | +0.058 |
learned | +0.057 |
Neg logit contributions | |
arl | -0.096 |
elly | -0.095 |
ay | -0.091 |
Today | -0.089 |
ane | -0.087 |
Token | he |
Token ID | 339 |
Feature activation | +5.88e+00 |
Pos logit contributions | |
pretended | +0.030 |
put | +0.029 |
hoped | +0.028 |
learned | +0.028 |
never | +0.028 |
Neg logit contributions | |
arl | -0.050 |
elly | -0.049 |
ay | -0.049 |
ummy | -0.048 |
Today | -0.048 |
Token | struggled |
Token ID | 11615 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
pretended | +0.056 |
punished | +0.054 |
put | +0.052 |
hoped | +0.052 |
never | +0.050 |
Neg logit contributions | |
elly | -0.152 |
� | -0.151 |
ane | -0.151 |
ay | -0.150 |
ore | -0.146 |
Token | , |
Token ID | 11 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
arl | +0.010 |
ay | +0.009 |
elly | +0.009 |
Today | +0.009 |
ane | +0.009 |
Neg logit contributions | |
put | -0.007 |
pretended | -0.006 |
never | -0.006 |
learned | -0.006 |
used | -0.006 |
Token | he |
Token ID | 339 |
Feature activation | +6.64e+00 |
Pos logit contributions | |
pretended | +0.022 |
put | +0.022 |
hoped | +0.021 |
punished | +0.021 |
cleaned | +0.021 |
Neg logit contributions | |
arl | -0.041 |
ay | -0.040 |
elly | -0.040 |
Today | -0.039 |
� | -0.039 |
Token | could |
Token ID | 714 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
pretended | +0.096 |
ck | +0.091 |
punished | +0.091 |
ts | +0.089 |
put | +0.089 |
Neg logit contributions | |
ane | -0.145 |
ay | -0.136 |
� | -0.132 |
ore | -0.132 |
elly | -0.132 |
Token | still |
Token ID | 991 |
Feature activation | +5.54e+00 |
Pos logit contributions | |
pretended | +0.037 |
cleaned | +0.035 |
hoped | +0.034 |
punished | +0.033 |
learned | +0.033 |
Neg logit contributions | |
� | -0.079 |
arl | -0.077 |
Today | -0.077 |
ute | -0.076 |
ay | -0.076 |
Token | do |
Token ID | 466 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
pretended | +0.067 |
cleaned | +0.064 |
punished | +0.063 |
put | +0.061 |
hoped | +0.061 |
Neg logit contributions | |
� | -0.137 |
Today | -0.136 |
arl | -0.135 |
ay | -0.135 |
ute | -0.134 |
Token | great |
Token ID | 1049 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
pretended | +0.019 |
put | +0.018 |
cleaned | +0.018 |
hoped | +0.018 |
waited | +0.017 |
Neg logit contributions | |
ay | -0.023 |
igsaw | -0.022 |
elly | -0.022 |
Today | -0.022 |
arl | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
put | +0.002 |
never | +0.002 |
used | +0.002 |
pretended | +0.002 |
made | +0.002 |
Neg logit contributions | |
arl | -0.003 |
elly | -0.003 |
ay | -0.003 |
Today | -0.003 |
arry | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
ane | +0.081 |
ute | +0.080 |
arl | +0.078 |
elly | +0.077 |
ay | +0.076 |
Neg logit contributions | |
put | -0.057 |
learned | -0.052 |
never | -0.052 |
used | -0.051 |
hoped | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
arl | +0.091 |
umm | +0.087 |
ane | +0.086 |
owers | +0.086 |
elly | +0.085 |
Neg logit contributions | |
put | -0.087 |
never | -0.085 |
learned | -0.082 |
used | -0.079 |
made | -0.076 |
Token | One |
Token ID | 3198 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
arl | +0.073 |
ay | +0.070 |
elly | +0.069 |
ane | +0.068 |
Today | +0.067 |
Neg logit contributions | |
put | -0.065 |
never | -0.063 |
learned | -0.061 |
pretended | -0.061 |
used | -0.060 |
Token | day |
Token ID | 1110 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
put | +0.003 |
never | +0.003 |
pretended | +0.003 |
used | +0.003 |
made | +0.003 |
Neg logit contributions | |
arl | -0.003 |
ay | -0.003 |
elly | -0.003 |
ane | -0.003 |
� | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
arl | +0.061 |
elly | +0.056 |
ute | +0.055 |
ay | +0.054 |
ane | +0.054 |
Neg logit contributions | |
put | -0.045 |
made | -0.043 |
tried | -0.042 |
used | -0.042 |
pretended | -0.041 |
Token | they |
Token ID | 484 |
Feature activation | +7.11e+00 |
Pos logit contributions | |
put | +0.005 |
never | +0.004 |
pretended | +0.004 |
learned | +0.004 |
used | +0.004 |
Neg logit contributions | |
arl | -0.010 |
ay | -0.010 |
elly | -0.010 |
Today | -0.009 |
ane | -0.009 |
Token | play |
Token ID | 711 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
never | +0.069 |
punished | +0.069 |
ck | +0.069 |
pretended | +0.069 |
put | +0.067 |
Neg logit contributions | |
arl | -0.187 |
ay | -0.182 |
ane | -0.180 |
elly | -0.177 |
Today | -0.177 |
Token | match |
Token ID | 2872 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
arl | +0.075 |
ay | +0.071 |
elly | +0.071 |
ane | +0.069 |
Today | +0.069 |
Neg logit contributions | |
put | -0.044 |
never | -0.041 |
pretended | -0.041 |
learned | -0.040 |
used | -0.040 |
Token | in |
Token ID | 287 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
arl | +0.065 |
ay | +0.062 |
elly | +0.062 |
Today | +0.061 |
ane | +0.060 |
Neg logit contributions | |
put | -0.036 |
never | -0.035 |
pretended | -0.034 |
used | -0.033 |
learned | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
punished | +0.016 |
forgot | +0.016 |
pretended | +0.016 |
never | +0.016 |
put | +0.016 |
Neg logit contributions | |
ay | -0.040 |
arl | -0.039 |
elly | -0.039 |
Today | -0.038 |
igsaw | -0.038 |
Token | L |
Token ID | 43 |
Feature activation | -4.23e+00 |
Pos logit contributions | |
ay | +0.087 |
arl | +0.085 |
igsaw | +0.083 |
Today | +0.083 |
ane | +0.083 |
Neg logit contributions | |
put | -0.090 |
never | -0.089 |
learned | -0.085 |
used | -0.081 |
made | -0.079 |
Token | ily |
Token ID | 813 |
Feature activation | +4.67e+00 |
Pos logit contributions | |
arl | +0.102 |
ay | +0.099 |
elly | +0.097 |
Today | +0.095 |
ane | +0.095 |
Neg logit contributions | |
put | -0.068 |
never | -0.065 |
pretended | -0.064 |
learned | -0.063 |
used | -0.062 |
Token | 's |
Token ID | 338 |
Feature activation | -7.12e-01 |
Pos logit contributions | |
punished | +0.053 |
remind | +0.052 |
permission | +0.050 |
ck | +0.048 |
never | +0.048 |
Neg logit contributions | |
elly | -0.104 |
� | -0.101 |
ane | -0.100 |
uddle | -0.099 |
olly | -0.099 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
arl | +0.014 |
ay | +0.014 |
ummy | +0.013 |
Today | +0.013 |
elly | +0.013 |
Neg logit contributions | |
pretended | -0.014 |
put | -0.014 |
never | -0.014 |
punished | -0.014 |
used | -0.014 |
Token | heard |
Token ID | 2982 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
never | +0.041 |
punished | +0.040 |
permission | +0.038 |
pretended | +0.038 |
remind | +0.037 |
Neg logit contributions | |
ay | -0.092 |
elly | -0.089 |
arl | -0.087 |
� | -0.087 |
omed | -0.087 |
Token | her |
Token ID | 607 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
put | +0.001 |
never | +0.001 |
pretended | +0.001 |
made | +0.001 |
learned | +0.001 |
Neg logit contributions | |
arl | -0.007 |
elly | -0.007 |
ay | -0.007 |
ane | -0.007 |
ute | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +6.83e+00 |
Pos logit contributions | |
arl | +0.010 |
ay | +0.010 |
elly | +0.010 |
ane | +0.009 |
� | +0.009 |
Neg logit contributions | |
put | -0.005 |
never | -0.005 |
pretended | -0.005 |
learned | -0.005 |
used | -0.005 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
wed | +0.097 |
never | +0.096 |
punished | +0.094 |
pretended | +0.089 |
remind | +0.088 |
Neg logit contributions | |
arl | -0.145 |
oms | -0.138 |
arry | -0.134 |
elly | -0.134 |
ay | -0.134 |
Token | too |
Token ID | 1165 |
Feature activation | -8.07e-01 |
Pos logit contributions | |
arl | +0.053 |
� | +0.053 |
ane | +0.052 |
ay | +0.051 |
anger | +0.051 |
Neg logit contributions | |
put | -0.026 |
made | -0.024 |
learned | -0.022 |
pretended | -0.022 |
used | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
arl | +0.023 |
ay | +0.023 |
ane | +0.022 |
ute | +0.022 |
anger | +0.022 |
Neg logit contributions | |
put | -0.011 |
never | -0.010 |
used | -0.009 |
learned | -0.009 |
made | -0.009 |
Token | She |
Token ID | 1375 |
Feature activation | +8.00e+00 |
Pos logit contributions | |
ane | +0.091 |
ute | +0.091 |
ay | +0.089 |
igsaw | +0.089 |
andel | +0.085 |
Neg logit contributions | |
put | -0.058 |
hoped | -0.055 |
learned | -0.053 |
made | -0.053 |
wished | -0.051 |
Token | butter |
Token ID | 9215 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
arl | +0.049 |
ay | +0.047 |
elly | +0.047 |
Today | +0.045 |
ane | +0.045 |
Neg logit contributions | |
put | -0.034 |
never | -0.033 |
pretended | -0.033 |
learned | -0.032 |
used | -0.032 |
Token | around |
Token ID | 1088 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
arl | +0.121 |
ay | +0.116 |
elly | +0.116 |
ane | +0.112 |
Today | +0.112 |
Neg logit contributions | |
put | -0.088 |
never | -0.085 |
pretended | -0.084 |
learned | -0.082 |
used | -0.082 |
Token | in |
Token ID | 287 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
put | +0.002 |
pretended | +0.002 |
never | +0.002 |
punished | +0.002 |
learned | +0.002 |
Neg logit contributions | |
arl | -0.006 |
ay | -0.006 |
elly | -0.006 |
Today | -0.006 |
� | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
punished | +0.025 |
pretended | +0.024 |
forgot | +0.024 |
hoped | +0.023 |
never | +0.023 |
Neg logit contributions | |
ay | -0.051 |
arl | -0.049 |
Today | -0.048 |
elly | -0.048 |
ummy | -0.048 |
Token | bowl |
Token ID | 9396 |
Feature activation | -5.43e+00 |
Pos logit contributions | |
arl | +0.037 |
ay | +0.036 |
elly | +0.035 |
Today | +0.034 |
ane | +0.034 |
Neg logit contributions | |
put | -0.031 |
never | -0.030 |
pretended | -0.030 |
used | -0.029 |
learned | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
arl | +0.116 |
ay | +0.112 |
elly | +0.111 |
Today | +0.107 |
ane | +0.107 |
Neg logit contributions | |
put | -0.099 |
never | -0.096 |
pretended | -0.095 |
learned | -0.093 |
used | -0.093 |
Token | It |
Token ID | 632 |
Feature activation | +5.28e+00 |
Pos logit contributions | |
arl | +0.073 |
elly | +0.069 |
ute | +0.069 |
ane | +0.066 |
andel | +0.064 |
Neg logit contributions | |
put | -0.060 |
never | -0.057 |
used | -0.055 |
tried | -0.055 |
pretended | -0.054 |
Token | became |
Token ID | 2627 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
punished | +0.077 |
permission | +0.076 |
never | +0.068 |
learned | +0.067 |
pretended | +0.065 |
Neg logit contributions | |
ay | -0.111 |
� | -0.108 |
elly | -0.108 |
oles | -0.104 |
owers | -0.104 |
Token | smoother |
Token ID | 32686 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
never | +0.054 |
punished | +0.053 |
pretended | +0.053 |
learned | +0.053 |
forgot | +0.053 |
Neg logit contributions | |
ay | -0.073 |
arl | -0.071 |
elly | -0.070 |
Today | -0.069 |
ummy | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | +6.34e+00 |
Pos logit contributions | |
arl | +0.066 |
elly | +0.065 |
ay | +0.065 |
Today | +0.061 |
ute | +0.061 |
Neg logit contributions | |
put | -0.056 |
never | -0.055 |
pretended | -0.055 |
learned | -0.054 |
punished | -0.054 |
Token | more |
Token ID | 517 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
punished | +0.096 |
pretended | +0.093 |
learned | +0.093 |
permission | +0.092 |
never | +0.091 |
Neg logit contributions | |
ay | -0.123 |
ummy | -0.122 |
elly | -0.117 |
umm | -0.115 |
arl | -0.114 |
Token | and |
Token ID | 290 |
Feature activation | +5.65e+00 |
Pos logit contributions | |
arl | +0.030 |
elly | +0.028 |
ay | +0.028 |
ane | +0.028 |
ute | +0.028 |
Neg logit contributions | |
put | -0.018 |
never | -0.017 |
made | -0.017 |
used | -0.016 |
pretended | -0.016 |
Token | it |
Token ID | 340 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
punished | +0.070 |
put | +0.067 |
pretended | +0.067 |
learned | +0.066 |
never | +0.065 |
Neg logit contributions | |
ay | -0.131 |
ane | -0.130 |
arl | -0.129 |
Today | -0.128 |
elly | -0.127 |
Token | started |
Token ID | 2067 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
permission | +0.046 |
punished | +0.046 |
pretended | +0.044 |
wed | +0.044 |
put | +0.043 |
Neg logit contributions | |
ay | -0.096 |
elly | -0.094 |
ane | -0.094 |
time | -0.091 |
ute | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | +4.20e+00 |
Pos logit contributions | |
put | +0.022 |
never | +0.021 |
pretended | +0.021 |
learned | +0.021 |
punished | +0.021 |
Neg logit contributions | |
arl | -0.027 |
ay | -0.026 |
elly | -0.026 |
Today | -0.025 |
ane | -0.025 |
Token | get |
Token ID | 651 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
put | +0.070 |
pretended | +0.069 |
learned | +0.069 |
punished | +0.069 |
never | +0.069 |
Neg logit contributions | |
arl | -0.092 |
Today | -0.087 |
ay | -0.086 |
� | -0.084 |
elly | -0.084 |
Token | long |
Token ID | 890 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
never | +0.043 |
put | +0.043 |
pretended | +0.043 |
learned | +0.043 |
hoped | +0.042 |
Neg logit contributions | |
arl | -0.040 |
ay | -0.039 |
elly | -0.037 |
Today | -0.037 |
ummy | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
arl | +0.048 |
ay | +0.046 |
elly | +0.046 |
Today | +0.044 |
ane | +0.044 |
Neg logit contributions | |
put | -0.038 |
never | -0.037 |
pretended | -0.036 |
learned | -0.036 |
used | -0.035 |
Token | |
Token ID | 220 |
Feature activation | -5.49e+00 |
Pos logit contributions | |
arl | +0.080 |
ute | +0.074 |
ane | +0.073 |
ay | +0.071 |
elly | +0.071 |
Neg logit contributions | |
put | -0.055 |
never | -0.053 |
tried | -0.051 |
learned | -0.050 |
used | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | -3.84e+00 |
Pos logit contributions | |
arl | +0.116 |
elly | +0.114 |
andel | +0.112 |
umm | +0.112 |
ane | +0.108 |
Neg logit contributions | |
put | -0.101 |
never | -0.099 |
used | -0.094 |
learned | -0.093 |
sure | -0.092 |
Token |
|
Token ID | 198 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
arl | +0.082 |
ane | +0.076 |
ute | +0.075 |
elly | +0.074 |
owers | +0.074 |
Neg logit contributions | |
never | -0.078 |
put | -0.075 |
learned | -0.072 |
used | -0.071 |
tried | -0.069 |
Token | When |
Token ID | 2215 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
arl | +0.069 |
ay | +0.064 |
ane | +0.064 |
elly | +0.063 |
ute | +0.062 |
Neg logit contributions | |
never | -0.055 |
put | -0.055 |
learned | -0.052 |
used | -0.051 |
pretended | -0.050 |
Token | radio |
Token ID | 5243 |
Feature activation | -4.33e+00 |
Pos logit contributions | |
arl | +0.042 |
ay | +0.040 |
elly | +0.039 |
Today | +0.039 |
ane | +0.038 |
Neg logit contributions | |
put | -0.031 |
never | -0.030 |
pretended | -0.030 |
learned | -0.030 |
used | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
arl | +0.100 |
ay | +0.096 |
elly | +0.095 |
ane | +0.093 |
Today | +0.093 |
Neg logit contributions | |
put | -0.071 |
never | -0.069 |
pretended | -0.068 |
learned | -0.067 |
used | -0.066 |
Token | call |
Token ID | 869 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
pretended | +0.037 |
remembered | +0.035 |
forgot | +0.035 |
learned | +0.035 |
hoped | +0.034 |
Neg logit contributions | |
arl | -0.104 |
ay | -0.100 |
Today | -0.096 |
ute | -0.096 |
� | -0.094 |
Token | them |
Token ID | 606 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
arl | +0.048 |
elly | +0.047 |
ay | +0.046 |
� | +0.045 |
ane | +0.045 |
Neg logit contributions | |
put | -0.020 |
never | -0.018 |
pretended | -0.017 |
made | -0.017 |
learned | -0.017 |
Token | ." |
Token ID | 526 |
Feature activation | -4.79e+00 |
Pos logit contributions | |
arl | +0.040 |
elly | +0.039 |
� | +0.039 |
ay | +0.038 |
igsaw | +0.038 |
Neg logit contributions | |
put | -0.023 |
never | -0.021 |
pretended | -0.021 |
used | -0.020 |
made | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
arl | +0.104 |
ute | +0.102 |
ane | +0.101 |
elly | +0.100 |
andel | +0.098 |
Neg logit contributions | |
put | -0.094 |
learned | -0.087 |
tried | -0.086 |
used | -0.086 |
never | -0.085 |
Token |
|
Token ID | 198 |
Feature activation | -4.06e+00 |
Pos logit contributions | |
arl | +0.100 |
elly | +0.099 |
ane | +0.098 |
owers | +0.095 |
ay | +0.093 |
Neg logit contributions | |
put | -0.116 |
never | -0.111 |
learned | -0.108 |
used | -0.106 |
tried | -0.104 |
Token | He |
Token ID | 1544 |
Feature activation | +7.46e+00 |
Pos logit contributions | |
arl | +0.082 |
ay | +0.078 |
elly | +0.077 |
arry | +0.075 |
Today | +0.074 |
Neg logit contributions | |
put | -0.078 |
pretended | -0.077 |
never | -0.075 |
made | -0.074 |
used | -0.073 |
Token | took |
Token ID | 1718 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
never | +0.063 |
punished | +0.062 |
remind | +0.056 |
care | +0.055 |
eling | +0.055 |
Neg logit contributions | |
arl | -0.194 |
ay | -0.190 |
elly | -0.189 |
arry | -0.187 |
anger | -0.185 |
Token | them |
Token ID | 606 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
forgot | +0.031 |
pretended | +0.030 |
never | +0.030 |
learned | +0.029 |
remembered | +0.029 |
Neg logit contributions | |
ay | -0.059 |
arl | -0.057 |
elly | -0.053 |
andel | -0.052 |
Today | -0.052 |
Token | by |
Token ID | 416 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
arl | +0.022 |
ay | +0.021 |
elly | +0.020 |
ane | +0.020 |
Today | +0.020 |
Neg logit contributions | |
put | -0.018 |
pretended | -0.018 |
forgot | -0.018 |
never | -0.018 |
punished | -0.017 |
Token | back |
Token ID | 736 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
arl | +0.013 |
elly | +0.013 |
� | +0.012 |
ane | +0.012 |
arry | +0.012 |
Neg logit contributions | |
put | -0.010 |
never | -0.010 |
used | -0.010 |
made | -0.009 |
learned | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
� | +0.031 |
arl | +0.031 |
elly | +0.030 |
owers | +0.029 |
ane | +0.029 |
Neg logit contributions | |
put | -0.017 |
never | -0.016 |
used | -0.015 |
made | -0.015 |
pretended | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -5.04e+00 |
Pos logit contributions | |
arl | +0.097 |
elly | +0.089 |
ifying | +0.084 |
andel | +0.084 |
ute | +0.083 |
Neg logit contributions | |
put | -0.078 |
never | -0.073 |
used | -0.071 |
made | -0.069 |
found | -0.069 |
Token |
|
Token ID | 198 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
arl | +0.118 |
umm | +0.110 |
ifying | +0.106 |
elly | +0.105 |
owers | +0.105 |
Neg logit contributions | |
put | -0.093 |
never | -0.092 |
used | -0.086 |
learned | -0.085 |
found | -0.085 |
Token | One |
Token ID | 3198 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
arl | +0.081 |
elly | +0.073 |
ay | +0.073 |
� | +0.072 |
ane | +0.072 |
Neg logit contributions | |
put | -0.077 |
never | -0.076 |
used | -0.072 |
learned | -0.071 |
made | -0.071 |
Token | day |
Token ID | 1110 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
arl | +0.018 |
� | +0.017 |
ay | +0.017 |
elly | +0.017 |
ane | +0.017 |
Neg logit contributions | |
put | -0.015 |
never | -0.014 |
made | -0.014 |
pretended | -0.014 |
used | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
arl | +0.060 |
elly | +0.055 |
ane | +0.053 |
ay | +0.053 |
ute | +0.053 |
Neg logit contributions | |
put | -0.050 |
made | -0.046 |
used | -0.046 |
pretended | -0.046 |
tried | -0.045 |
Token | Sus |
Token ID | 8932 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
ay | +0.016 |
arl | +0.016 |
elly | +0.016 |
Today | +0.015 |
� | +0.015 |
Neg logit contributions | |
put | -0.010 |
never | -0.010 |
learned | -0.010 |
punished | -0.010 |
pretended | -0.010 |
Token | ie |
Token ID | 494 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
elly | +0.006 |
ane | +0.006 |
arl | +0.006 |
ay | +0.006 |
ummy | +0.006 |
Neg logit contributions | |
put | -0.005 |
pretended | -0.005 |
never | -0.005 |
punished | -0.004 |
learned | -0.004 |
Token | wanted |
Token ID | 2227 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
put | +0.043 |
never | +0.042 |
punished | +0.042 |
pretended | +0.041 |
ck | +0.040 |
Neg logit contributions | |
elly | -0.105 |
ane | -0.101 |
ay | -0.101 |
Today | -0.101 |
arl | -0.099 |
Token | to |
Token ID | 284 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
pretended | +0.017 |
forgot | +0.016 |
put | +0.016 |
punished | +0.016 |
hoped | +0.016 |
Neg logit contributions | |
elly | -0.013 |
ay | -0.013 |
arl | -0.013 |
ane | -0.013 |
Today | -0.013 |
Token | out |
Token ID | 503 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
arl | +0.110 |
elly | +0.106 |
arry | +0.105 |
ane | +0.104 |
ay | +0.103 |
Neg logit contributions | |
put | -0.057 |
never | -0.055 |
made | -0.052 |
used | -0.052 |
pretended | -0.052 |
Token | of |
Token ID | 286 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
arl | +0.055 |
elly | +0.053 |
ay | +0.053 |
Today | +0.051 |
ane | +0.051 |
Neg logit contributions | |
put | -0.039 |
never | -0.038 |
pretended | -0.036 |
used | -0.036 |
made | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
pretended | +0.021 |
forgot | +0.021 |
learned | +0.020 |
remembered | +0.020 |
shared | +0.020 |
Neg logit contributions | |
arl | -0.038 |
ay | -0.038 |
ummy | -0.036 |
elly | -0.036 |
igsaw | -0.035 |
Token | window |
Token ID | 4324 |
Feature activation | -6.61e+00 |
Pos logit contributions | |
arl | +0.067 |
arry | +0.064 |
ane | +0.064 |
ay | +0.063 |
elly | +0.063 |
Neg logit contributions | |
put | -0.051 |
never | -0.048 |
made | -0.047 |
used | -0.046 |
cleaned | -0.046 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
arl | +0.164 |
elly | +0.154 |
ay | +0.154 |
� | +0.151 |
ute | +0.151 |
Neg logit contributions | |
put | -0.103 |
never | -0.100 |
pretended | -0.096 |
made | -0.096 |
used | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | -6.13e+00 |
Pos logit contributions | |
arl | +0.095 |
andel | +0.091 |
ute | +0.091 |
elly | +0.091 |
ane | +0.089 |
Neg logit contributions | |
put | -0.090 |
learned | -0.088 |
never | -0.086 |
tried | -0.084 |
made | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | -4.79e+00 |
Pos logit contributions | |
arl | +0.117 |
elly | +0.115 |
owers | +0.113 |
ane | +0.112 |
andel | +0.112 |
Neg logit contributions | |
put | -0.133 |
never | -0.132 |
learned | -0.128 |
cleaned | -0.120 |
used | -0.119 |
Token | Tom |
Token ID | 13787 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
arl | +0.094 |
ay | +0.090 |
elly | +0.089 |
Today | +0.086 |
ane | +0.085 |
Neg logit contributions | |
put | -0.097 |
pretended | -0.094 |
never | -0.093 |
used | -0.092 |
made | -0.090 |
Token | was |
Token ID | 373 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
remind | +0.047 |
pretended | +0.044 |
punished | +0.044 |
ts | +0.042 |
ck | +0.041 |
Neg logit contributions | |
ay | -0.092 |
arry | -0.090 |
uddle | -0.088 |
elly | -0.087 |
� | -0.087 |
Token | very |
Token ID | 845 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
pretended | +0.081 |
wed | +0.074 |
forgot | +0.074 |
wrote | +0.073 |
put | +0.073 |
Neg logit contributions | |
Today | -0.088 |
apping | -0.088 |
arl | -0.087 |
elly | -0.083 |
� | -0.083 |
Token | scared |
Token ID | 12008 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
pretended | +0.060 |
put | +0.059 |
never | +0.058 |
made | +0.055 |
used | +0.054 |
Neg logit contributions | |
arl | -0.094 |
elly | -0.090 |
ay | -0.087 |
Today | -0.086 |
ute | -0.085 |
Token | about |
Token ID | 546 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
put | +0.011 |
pretended | +0.011 |
never | +0.011 |
used | +0.010 |
punished | +0.010 |
Neg logit contributions | |
arl | -0.016 |
ay | -0.015 |
elly | -0.015 |
Today | -0.015 |
ane | -0.015 |
Token | her |
Token ID | 607 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
pretended | +0.028 |
put | +0.028 |
punished | +0.027 |
forgot | +0.026 |
cleaned | +0.026 |
Neg logit contributions | |
arl | -0.076 |
ay | -0.074 |
Today | -0.073 |
elly | -0.073 |
� | -0.072 |
Token | sailing |
Token ID | 28499 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
arl | +0.006 |
ay | +0.006 |
elly | +0.006 |
Today | +0.006 |
ane | +0.006 |
Neg logit contributions | |
put | -0.004 |
never | -0.004 |
pretended | -0.004 |
learned | -0.004 |
used | -0.004 |
Token | adventure |
Token ID | 8855 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
arl | +0.078 |
elly | +0.074 |
ay | +0.073 |
Today | +0.071 |
arry | +0.070 |
Neg logit contributions | |
put | -0.059 |
never | -0.056 |
made | -0.055 |
pretended | -0.055 |
used | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
arl | +0.083 |
elly | +0.078 |
ay | +0.077 |
Today | +0.076 |
ane | +0.075 |
Neg logit contributions | |
put | -0.059 |
never | -0.057 |
pretended | -0.055 |
used | -0.055 |
made | -0.055 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.43e+00 |
Pos logit contributions | |
arl | +0.063 |
ute | +0.062 |
ane | +0.059 |
elly | +0.058 |
andel | +0.057 |
Neg logit contributions | |
never | -0.059 |
put | -0.058 |
learned | -0.057 |
hoped | -0.055 |
used | -0.054 |
Token | One |
Token ID | 3198 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
arl | +0.197 |
andel | +0.194 |
ower | +0.181 |
ute | +0.180 |
igsaw | +0.177 |
Neg logit contributions | |
made | -0.159 |
put | -0.155 |
found | -0.154 |
used | -0.152 |
tried | -0.149 |
Token | day |
Token ID | 1110 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
arl | +0.012 |
elly | +0.011 |
ay | +0.011 |
ane | +0.011 |
Today | +0.011 |
Neg logit contributions | |
put | -0.012 |
never | -0.011 |
pretended | -0.011 |
learned | -0.011 |
used | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
arl | +0.041 |
elly | +0.038 |
ay | +0.037 |
ane | +0.037 |
ute | +0.036 |
Neg logit contributions | |
put | -0.030 |
tried | -0.028 |
made | -0.028 |
used | -0.028 |
never | -0.028 |
Token | Mrs |
Token ID | 9074 |
Feature activation | -6.15e+00 |
Pos logit contributions | |
arl | +0.071 |
elly | +0.066 |
ay | +0.066 |
ane | +0.066 |
ute | +0.064 |
Neg logit contributions | |
put | -0.019 |
never | -0.018 |
used | -0.017 |
pretended | -0.016 |
made | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -6.39e+00 |
Pos logit contributions | |
arl | +0.151 |
ay | +0.143 |
ute | +0.141 |
elly | +0.140 |
ane | +0.139 |
Neg logit contributions | |
put | -0.099 |
never | -0.092 |
pretended | -0.091 |
used | -0.091 |
made | -0.090 |
Token | he |
Token ID | 339 |
Feature activation | +8.10e+00 |
Pos logit contributions | |
wed | +0.095 |
wy | +0.087 |
ws | +0.087 |
ps | +0.087 |
ck | +0.085 |
Neg logit contributions | |
aris | -0.105 |
ay | -0.105 |
Nice | -0.104 |
ogging | -0.101 |
arry | -0.099 |
Token | laughed |
Token ID | 13818 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
permission | +0.101 |
punished | +0.087 |
remind | +0.087 |
ts | +0.085 |
eling | +0.085 |
Neg logit contributions | |
ane | -0.169 |
ute | -0.167 |
ay | -0.166 |
uddle | -0.160 |
arry | -0.160 |
Token | and |
Token ID | 290 |
Feature activation | +6.83e+00 |
Pos logit contributions | |
arl | +0.042 |
ay | +0.041 |
elly | +0.040 |
Today | +0.040 |
ane | +0.039 |
Neg logit contributions | |
put | -0.030 |
never | -0.029 |
pretended | -0.028 |
punished | -0.028 |
learned | -0.028 |
Token | laughed |
Token ID | 13818 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
wed | +0.082 |
punished | +0.079 |
wy | +0.078 |
permission | +0.074 |
eling | +0.072 |
Neg logit contributions | |
Nice | -0.141 |
Today | -0.136 |
ogging | -0.134 |
ay | -0.134 |
apping | -0.133 |
Token | . |
Token ID | 13 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
arl | +0.041 |
ay | +0.040 |
elly | +0.040 |
Today | +0.039 |
ane | +0.038 |
Neg logit contributions | |
put | -0.029 |
never | -0.028 |
pretended | -0.028 |
learned | -0.027 |
used | -0.027 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.43e+00 |
Pos logit contributions | |
arl | +0.076 |
elly | +0.072 |
ay | +0.072 |
ane | +0.069 |
ute | +0.069 |
Neg logit contributions | |
put | -0.084 |
never | -0.081 |
pretended | -0.080 |
learned | -0.079 |
used | -0.079 |
Token | Once |
Token ID | 7454 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
arl | +0.218 |
andel | +0.213 |
ute | +0.209 |
ower | +0.203 |
elly | +0.203 |
Neg logit contributions | |
put | -0.139 |
made | -0.136 |
never | -0.134 |
found | -0.131 |
used | -0.130 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
arl | +0.099 |
elly | +0.093 |
� | +0.090 |
owers | +0.089 |
ay | +0.089 |
Neg logit contributions | |
put | -0.124 |
never | -0.121 |
used | -0.117 |
made | -0.117 |
pretended | -0.116 |
Token | a |
Token ID | 257 |
Feature activation | -4.86e+00 |
Pos logit contributions | |
arl | +0.041 |
ay | +0.040 |
elly | +0.040 |
Today | +0.039 |
ane | +0.039 |
Neg logit contributions | |
put | -0.017 |
pretended | -0.016 |
never | -0.016 |
punished | -0.016 |
used | -0.015 |
Token | time |
Token ID | 640 |
Feature activation | -4.50e+00 |
Pos logit contributions | |
ane | +0.108 |
arl | +0.106 |
anger | +0.102 |
ute | +0.100 |
owers | +0.099 |
Neg logit contributions | |
put | -0.092 |
never | -0.092 |
learned | -0.085 |
made | -0.085 |
pretended | -0.085 |
Token | , |
Token ID | 11 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
arl | +0.133 |
andel | +0.130 |
umm | +0.126 |
elly | +0.125 |
ute | +0.124 |
Neg logit contributions | |
never | -0.056 |
put | -0.053 |
tried | -0.052 |
found | -0.052 |
made | -0.052 |
Token | when |
Token ID | 618 |
Feature activation | +4.53e-01 |
Pos logit contributions | |
arl | +0.057 |
ay | +0.055 |
elly | +0.054 |
Today | +0.052 |
ute | +0.052 |
Neg logit contributions | |
pretended | -0.051 |
put | -0.050 |
never | -0.050 |
learned | -0.050 |
hoped | -0.049 |
Token | he |
Token ID | 339 |
Feature activation | +4.88e+00 |
Pos logit contributions | |
put | +0.006 |
pretended | +0.006 |
never | +0.006 |
punished | +0.006 |
learned | +0.006 |
Neg logit contributions | |
arl | -0.011 |
elly | -0.011 |
ay | -0.010 |
Today | -0.010 |
ane | -0.010 |
Token | needed |
Token ID | 2622 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
put | +0.048 |
never | +0.048 |
pretended | +0.047 |
punished | +0.045 |
cleaned | +0.043 |
Neg logit contributions | |
arl | -0.135 |
ay | -0.134 |
elly | -0.134 |
ane | -0.130 |
� | -0.129 |
Token | it |
Token ID | 340 |
Feature activation | -9.24e-01 |
Pos logit contributions | |
pretended | +0.016 |
forgot | +0.015 |
put | +0.014 |
hoped | +0.014 |
punished | +0.014 |
Neg logit contributions | |
arl | -0.032 |
ay | -0.032 |
� | -0.031 |
elly | -0.031 |
Today | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
arl | +0.019 |
ay | +0.018 |
elly | +0.018 |
Today | +0.018 |
ane | +0.017 |
Neg logit contributions | |
put | -0.017 |
pretended | -0.017 |
never | -0.017 |
learned | -0.016 |
punished | -0.016 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.75e+00 |
Pos logit contributions | |
arl | +0.074 |
ute | +0.070 |
andel | +0.069 |
elly | +0.068 |
ane | +0.068 |
Neg logit contributions | |
never | -0.069 |
put | -0.068 |
learned | -0.068 |
made | -0.066 |
used | -0.066 |
Token | One |
Token ID | 3198 |
Feature activation | -9.45e-01 |
Pos logit contributions | |
arl | +0.229 |
andel | +0.227 |
ute | +0.217 |
ower | +0.217 |
umm | +0.216 |
Neg logit contributions | |
put | -0.141 |
made | -0.138 |
never | -0.132 |
used | -0.131 |
found | -0.130 |
Token | day |
Token ID | 1110 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
arl | +0.019 |
ay | +0.018 |
elly | +0.018 |
Today | +0.017 |
ane | +0.017 |
Neg logit contributions | |
put | -0.019 |
never | -0.018 |
pretended | -0.018 |
used | -0.018 |
learned | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
arl | +0.048 |
elly | +0.045 |
andel | +0.045 |
ane | +0.044 |
ay | +0.044 |
Neg logit contributions | |
put | -0.035 |
made | -0.033 |
tried | -0.033 |
used | -0.033 |
never | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
arl | +0.075 |
elly | +0.072 |
ay | +0.072 |
ane | +0.071 |
ute | +0.070 |
Neg logit contributions | |
put | -0.026 |
never | -0.024 |
pretended | -0.023 |
used | -0.023 |
made | -0.023 |
Token | boy |
Token ID | 2933 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
arl | +0.068 |
ane | +0.068 |
Today | +0.066 |
ay | +0.066 |
anger | +0.065 |
Neg logit contributions | |
put | -0.049 |
used | -0.043 |
made | -0.042 |
cleaned | -0.041 |
pretended | -0.041 |
Token | blue |
Token ID | 4171 |
Feature activation | -6.37e+00 |
Pos logit contributions | |
arl | +0.088 |
elly | +0.081 |
ay | +0.079 |
� | +0.078 |
arry | +0.078 |
Neg logit contributions | |
put | -0.072 |
never | -0.068 |
made | -0.068 |
pretended | -0.066 |
used | -0.066 |
Token | wings |
Token ID | 12098 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
arl | +0.162 |
igsaw | +0.148 |
elly | +0.146 |
aris | +0.144 |
arry | +0.143 |
Neg logit contributions | |
put | -0.105 |
made | -0.103 |
never | -0.099 |
tried | -0.095 |
used | -0.095 |
Token | of |
Token ID | 286 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
arl | +0.058 |
ay | +0.057 |
elly | +0.055 |
ane | +0.055 |
Today | +0.054 |
Neg logit contributions | |
put | -0.045 |
pretended | -0.045 |
punished | -0.044 |
never | -0.044 |
learned | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
put | +0.014 |
pretended | +0.014 |
punished | +0.014 |
never | +0.014 |
forgot | +0.014 |
Neg logit contributions | |
arl | -0.028 |
ay | -0.028 |
elly | -0.027 |
Today | -0.026 |
ane | -0.026 |
Token | bird |
Token ID | 6512 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
arl | +0.059 |
elly | +0.055 |
ay | +0.054 |
arry | +0.053 |
Today | +0.052 |
Neg logit contributions | |
put | -0.072 |
never | -0.070 |
made | -0.068 |
pretended | -0.068 |
used | -0.068 |
Token | fl |
Token ID | 781 |
Feature activation | -8.15e+00 |
Pos logit contributions | |
arl | +0.038 |
ay | +0.038 |
elly | +0.037 |
Today | +0.036 |
ane | +0.036 |
Neg logit contributions | |
put | -0.026 |
never | -0.025 |
pretended | -0.025 |
punished | -0.024 |
learned | -0.024 |
Token | apping |
Token ID | 5912 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
Today | +0.085 |
igsaw | +0.075 |
elly | +0.072 |
arl | +0.071 |
Pole | +0.066 |
Neg logit contributions | |
never | -0.219 |
put | -0.219 |
used | -0.216 |
ck | -0.213 |
made | -0.210 |
Token | around |
Token ID | 1088 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
arl | +0.064 |
ay | +0.063 |
elly | +0.061 |
Today | +0.060 |
ane | +0.060 |
Neg logit contributions | |
put | -0.038 |
pretended | -0.036 |
never | -0.036 |
punished | -0.036 |
learned | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
put | +0.007 |
never | +0.006 |
pretended | +0.006 |
used | +0.006 |
learned | +0.006 |
Neg logit contributions | |
arl | -0.010 |
ay | -0.009 |
elly | -0.009 |
ane | -0.009 |
Today | -0.009 |
Token | He |
Token ID | 679 |
Feature activation | +7.68e+00 |
Pos logit contributions | |
arl | +0.087 |
elly | +0.081 |
ute | +0.080 |
ay | +0.079 |
ane | +0.079 |
Neg logit contributions | |
put | -0.077 |
learned | -0.073 |
never | -0.073 |
used | -0.072 |
pretended | -0.072 |
Token | had |
Token ID | 550 |
Feature activation | +4.93e+00 |
Pos logit contributions | |
punished | +0.103 |
permission | +0.089 |
cleaned | +0.089 |
wed | +0.088 |
remind | +0.087 |
Neg logit contributions | |
ay | -0.170 |
arry | -0.167 |
oles | -0.166 |
elly | -0.165 |
ute | -0.164 |
Token | crown |
Token ID | 12389 |
Feature activation | -4.72e+00 |
Pos logit contributions | |
put | +0.014 |
never | +0.013 |
pretended | +0.013 |
used | +0.013 |
learned | +0.013 |
Neg logit contributions | |
arl | -0.019 |
ay | -0.018 |
elly | -0.018 |
Today | -0.018 |
ane | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +6.01e+00 |
Pos logit contributions | |
arl | +0.113 |
elly | +0.109 |
ay | +0.107 |
Today | +0.106 |
arry | +0.104 |
Neg logit contributions | |
put | -0.076 |
never | -0.074 |
pretended | -0.072 |
made | -0.071 |
used | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
punished | +0.078 |
ck | +0.076 |
wed | +0.072 |
put | +0.071 |
cleaned | +0.070 |
Neg logit contributions | |
ay | -0.141 |
ane | -0.133 |
elly | -0.130 |
Today | -0.127 |
ummy | -0.127 |
Token | swing |
Token ID | 9628 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
put | +0.024 |
never | +0.023 |
pretended | +0.023 |
used | +0.022 |
learned | +0.022 |
Neg logit contributions | |
arl | -0.034 |
ay | -0.032 |
elly | -0.032 |
Today | -0.031 |
ane | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
arl | +0.094 |
elly | +0.089 |
Today | +0.087 |
ay | +0.087 |
arry | +0.086 |
Neg logit contributions | |
put | -0.065 |
never | -0.063 |
made | -0.061 |
pretended | -0.060 |
used | -0.059 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.46e+00 |
Pos logit contributions | |
arl | +0.056 |
elly | +0.053 |
ute | +0.050 |
ay | +0.050 |
ane | +0.049 |
Neg logit contributions | |
never | -0.066 |
put | -0.066 |
learned | -0.062 |
hoped | -0.062 |
made | -0.062 |
Token | Once |
Token ID | 7454 |
Feature activation | -5.16e+00 |
Pos logit contributions | |
arl | +0.186 |
andel | +0.181 |
ower | +0.169 |
ute | +0.169 |
igsaw | +0.166 |
Neg logit contributions | |
made | -0.171 |
put | -0.168 |
found | -0.167 |
used | -0.163 |
tried | -0.161 |
Token | there |
Token ID | 612 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
arl | +0.099 |
elly | +0.093 |
� | +0.090 |
ute | +0.088 |
umm | +0.088 |
Neg logit contributions | |
put | -0.109 |
never | -0.107 |
made | -0.104 |
used | -0.102 |
forgot | -0.102 |
Token | was |
Token ID | 373 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
ay | +0.032 |
arl | +0.031 |
Today | +0.031 |
elly | +0.031 |
ane | +0.031 |
Neg logit contributions | |
put | -0.016 |
punished | -0.015 |
pretended | -0.015 |
never | -0.015 |
used | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
arl | +0.011 |
ay | +0.010 |
elly | +0.010 |
ane | +0.010 |
arry | +0.010 |
Neg logit contributions | |
put | -0.003 |
never | -0.003 |
pretended | -0.003 |
used | -0.003 |
learned | -0.003 |
Token | boy |
Token ID | 2933 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
arl | +0.053 |
ane | +0.051 |
ay | +0.051 |
owers | +0.050 |
elly | +0.049 |
Neg logit contributions | |
put | -0.042 |
used | -0.037 |
never | -0.037 |
pretended | -0.037 |
made | -0.036 |