Token | band |
Token ID | 4097 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
made | +0.017 |
cleaned | +0.017 |
put | +0.017 |
never | +0.017 |
fixed | +0.017 |
Neg logit contributions | |
arry | -0.033 |
oles | -0.033 |
Today | -0.033 |
elly | -0.032 |
owers | -0.032 |
Token | - |
Token ID | 12 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
oles | +0.058 |
arry | +0.058 |
Today | +0.057 |
elly | +0.056 |
arl | +0.056 |
Neg logit contributions | |
never | -0.065 |
made | -0.063 |
put | -0.062 |
punished | -0.061 |
cleaned | -0.061 |
Token | aid |
Token ID | 1698 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
elly | +0.052 |
arry | +0.051 |
arl | +0.051 |
oles | +0.050 |
ute | +0.049 |
Neg logit contributions | |
never | -0.037 |
made | -0.036 |
put | -0.035 |
pretended | -0.034 |
learned | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
arry | +0.098 |
oles | +0.098 |
Today | +0.097 |
elly | +0.095 |
arl | +0.094 |
Neg logit contributions | |
never | -0.066 |
made | -0.064 |
put | -0.062 |
learned | -0.062 |
cleaned | -0.061 |
Token | They |
Token ID | 1119 |
Feature activation | +8.97e-02 |
Pos logit contributions | |
oles | +0.044 |
arry | +0.042 |
ute | +0.041 |
owers | +0.040 |
ower | +0.039 |
Neg logit contributions | |
never | -0.050 |
made | -0.049 |
learned | -0.049 |
put | -0.048 |
cleaned | -0.048 |
Token | also |
Token ID | 635 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
permission | +0.095 |
himself | +0.085 |
pretending | +0.084 |
finding | +0.083 |
rid | +0.082 |
Neg logit contributions | |
� | -0.183 |
olly | -0.181 |
owers | -0.180 |
arry | -0.179 |
uddle | -0.175 |
Token | helped |
Token ID | 4193 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
himself | +0.091 |
never | +0.089 |
permission | +0.077 |
care | +0.077 |
supposed | +0.074 |
Neg logit contributions | |
elly | -0.217 |
arry | -0.206 |
owers | -0.206 |
oms | -0.202 |
arl | -0.202 |
Token | to |
Token ID | 284 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
never | +0.011 |
learned | +0.010 |
saved | +0.010 |
punished | +0.010 |
made | +0.010 |
Neg logit contributions | |
oles | -0.024 |
Today | -0.024 |
elly | -0.024 |
ummy | -0.024 |
arl | -0.024 |
Token | wash |
Token ID | 13502 |
Feature activation | -7.32e-03 |
Pos logit contributions | |
pretended | +0.057 |
learned | +0.057 |
remembered | +0.057 |
wished | +0.055 |
made | +0.054 |
Neg logit contributions | |
arl | -0.139 |
ute | -0.133 |
Today | -0.132 |
� | -0.131 |
arry | -0.131 |
Token | the |
Token ID | 262 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
oles | +0.020 |
Today | +0.020 |
arry | +0.020 |
elly | +0.020 |
arl | +0.020 |
Neg logit contributions | |
never | -0.008 |
made | -0.008 |
learned | -0.007 |
punished | -0.007 |
fixed | -0.007 |
Token | salt |
Token ID | 8268 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
oles | +0.036 |
elly | +0.036 |
ute | +0.035 |
Today | +0.035 |
arl | +0.034 |
Neg logit contributions | |
never | -0.030 |
fixed | -0.027 |
learned | -0.027 |
pretended | -0.027 |
punished | -0.026 |
Token | hugs |
Token ID | 40657 |
Feature activation | -5.75e-03 |
Pos logit contributions | |
never | +0.003 |
punished | +0.003 |
forgot | +0.002 |
made | +0.002 |
saved | +0.002 |
Neg logit contributions | |
elly | -0.004 |
Today | -0.004 |
ute | -0.004 |
oles | -0.004 |
aris | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +6.06e-02 |
Pos logit contributions | |
arry | +0.016 |
Today | +0.016 |
oles | +0.016 |
elly | +0.016 |
arl | +0.015 |
Neg logit contributions | |
never | -0.006 |
made | -0.006 |
punished | -0.006 |
cleaned | -0.006 |
learned | -0.006 |
Token | kisses |
Token ID | 36544 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
never | +0.086 |
punished | +0.081 |
cleaned | +0.074 |
fixed | +0.073 |
built | +0.071 |
Neg logit contributions | |
oles | -0.135 |
elly | -0.134 |
Today | -0.132 |
arry | -0.130 |
owers | -0.125 |
Token | . |
Token ID | 13 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
oles | +0.041 |
arry | +0.041 |
Today | +0.041 |
elly | +0.040 |
arl | +0.040 |
Neg logit contributions | |
never | -0.025 |
made | -0.024 |
learned | -0.024 |
punished | -0.024 |
cleaned | -0.024 |
Token | They |
Token ID | 1119 |
Feature activation | +9.16e-02 |
Pos logit contributions | |
oles | +0.050 |
ute | +0.049 |
arry | +0.049 |
owers | +0.048 |
ower | +0.048 |
Neg logit contributions | |
put | -0.042 |
made | -0.042 |
learned | -0.042 |
never | -0.041 |
cleaned | -0.040 |
Token | also |
Token ID | 635 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
punished | +0.103 |
never | +0.101 |
permission | +0.098 |
himself | +0.092 |
hurting | +0.090 |
Neg logit contributions | |
arry | -0.202 |
owers | -0.199 |
oms | -0.189 |
oles | -0.188 |
ane | -0.187 |
Token | gave |
Token ID | 2921 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
never | +0.088 |
himself | +0.084 |
punished | +0.074 |
permission | +0.069 |
hurting | +0.068 |
Neg logit contributions | |
arry | -0.239 |
oms | -0.238 |
owers | -0.233 |
elly | -0.230 |
Today | -0.229 |
Token | them |
Token ID | 606 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
never | +0.026 |
punished | +0.024 |
learned | +0.023 |
fixed | +0.023 |
cleaned | +0.023 |
Neg logit contributions | |
oles | -0.050 |
elly | -0.049 |
Today | -0.049 |
ummy | -0.049 |
arl | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
elly | +0.032 |
oles | +0.032 |
aris | +0.032 |
arry | +0.032 |
Today | +0.032 |
Neg logit contributions | |
never | -0.018 |
punished | -0.018 |
made | -0.016 |
tried | -0.016 |
fixed | -0.016 |
Token | y |
Token ID | 331 |
Feature activation | -8.62e-02 |
Pos logit contributions | |
made | +0.028 |
never | +0.027 |
put | +0.027 |
fixed | +0.026 |
cleaned | +0.026 |
Neg logit contributions | |
oles | -0.042 |
arry | -0.042 |
elly | -0.041 |
Today | -0.041 |
owers | -0.040 |
Token | ummy |
Token ID | 13513 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
oles | +0.040 |
arry | +0.036 |
Today | +0.035 |
owers | +0.031 |
arl | +0.030 |
Neg logit contributions | |
made | -0.293 |
never | -0.293 |
put | -0.291 |
cleaned | -0.286 |
fixed | -0.283 |
Token | they |
Token ID | 484 |
Feature activation | +7.39e-02 |
Pos logit contributions | |
never | +0.019 |
made | +0.018 |
put | +0.017 |
learned | +0.017 |
cleaned | +0.017 |
Neg logit contributions | |
arry | -0.025 |
oles | -0.025 |
Today | -0.025 |
elly | -0.025 |
arl | -0.024 |
Token | loved |
Token ID | 6151 |
Feature activation | +9.42e-03 |
Pos logit contributions | |
punished | +0.118 |
pretended | +0.110 |
his | +0.107 |
put | +0.106 |
buried | +0.106 |
Neg logit contributions | |
Today | -0.135 |
owers | -0.133 |
� | -0.128 |
arry | -0.126 |
elly | -0.126 |
Token | them |
Token ID | 606 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
punished | +0.011 |
never | +0.011 |
saved | +0.011 |
fixed | +0.011 |
learned | +0.011 |
Neg logit contributions | |
Today | -0.024 |
arl | -0.024 |
elly | -0.024 |
ummy | -0.024 |
oles | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
Today | +0.053 |
arry | +0.052 |
oles | +0.051 |
elly | +0.051 |
arl | +0.050 |
Neg logit contributions | |
never | -0.035 |
made | -0.034 |
put | -0.033 |
cleaned | -0.033 |
punished | -0.033 |
Token | They |
Token ID | 1119 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
oles | +0.055 |
arry | +0.053 |
ute | +0.051 |
owers | +0.051 |
andel | +0.050 |
Neg logit contributions | |
never | -0.046 |
learned | -0.045 |
made | -0.045 |
put | -0.045 |
cleaned | -0.043 |
Token | also |
Token ID | 635 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
permission | +0.109 |
ts | +0.104 |
rid | +0.103 |
punished | +0.102 |
ps | +0.098 |
Neg logit contributions | |
arry | -0.178 |
olly | -0.178 |
Today | -0.172 |
elly | -0.169 |
uffy | -0.167 |
Token | said |
Token ID | 531 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
permission | +0.087 |
himself | +0.087 |
rid | +0.083 |
ts | +0.083 |
never | +0.072 |
Neg logit contributions | |
elly | -0.222 |
Today | -0.218 |
arry | -0.207 |
arl | -0.202 |
olly | -0.199 |
Token | they |
Token ID | 484 |
Feature activation | +8.14e-02 |
Pos logit contributions | |
never | +0.022 |
made | +0.021 |
put | +0.020 |
learned | +0.020 |
cleaned | +0.020 |
Neg logit contributions | |
arry | -0.029 |
oles | -0.029 |
Today | -0.029 |
elly | -0.028 |
arl | -0.028 |
Token | liked |
Token ID | 8288 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
punished | +0.123 |
his | +0.118 |
permission | +0.117 |
ts | +0.114 |
asking | +0.113 |
Neg logit contributions | |
owers | -0.145 |
Today | -0.141 |
elly | -0.139 |
ane | -0.137 |
op | -0.133 |
Token | the |
Token ID | 262 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
punished | +0.035 |
never | +0.035 |
learned | +0.035 |
fixed | +0.034 |
found | +0.034 |
Neg logit contributions | |
Today | -0.046 |
elly | -0.046 |
oles | -0.046 |
ummy | -0.045 |
arl | -0.044 |
Token | movie |
Token ID | 3807 |
Feature activation | -5.96e-02 |
Pos logit contributions | |
Today | +0.021 |
oles | +0.020 |
elly | +0.020 |
arry | +0.020 |
ute | +0.020 |
Neg logit contributions | |
never | -0.020 |
made | -0.018 |
learned | -0.018 |
put | -0.018 |
cleaned | -0.018 |
Token | squeezed |
Token ID | 29650 |
Feature activation | +2.03e-03 |
Pos logit contributions | |
never | +0.085 |
punished | +0.082 |
learned | +0.073 |
cleaned | +0.072 |
buried | +0.072 |
Neg logit contributions | |
oles | -0.151 |
ew | -0.144 |
Today | -0.142 |
ower | -0.141 |
ay | -0.139 |
Token | their |
Token ID | 511 |
Feature activation | -5.09e-04 |
Pos logit contributions | |
never | +0.002 |
made | +0.002 |
put | +0.002 |
learned | +0.002 |
cleaned | +0.002 |
Neg logit contributions | |
arry | -0.006 |
oles | -0.006 |
Today | -0.006 |
elly | -0.006 |
arl | -0.006 |
Token | balloons |
Token ID | 35485 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
oles | +0.001 |
arry | +0.001 |
Today | +0.001 |
elly | +0.001 |
arl | +0.001 |
Neg logit contributions | |
never | -0.001 |
made | -0.001 |
learned | -0.001 |
cleaned | -0.001 |
put | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
oles | +0.094 |
Today | +0.093 |
arry | +0.093 |
ute | +0.091 |
elly | +0.090 |
Neg logit contributions | |
never | -0.077 |
learned | -0.074 |
made | -0.073 |
punished | -0.073 |
cleaned | -0.072 |
Token | They |
Token ID | 1119 |
Feature activation | +8.61e-02 |
Pos logit contributions | |
oles | +0.054 |
arry | +0.053 |
ute | +0.050 |
elly | +0.049 |
owers | +0.049 |
Neg logit contributions | |
put | -0.050 |
made | -0.050 |
never | -0.050 |
cleaned | -0.049 |
learned | -0.049 |
Token | also |
Token ID | 635 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
permission | +0.102 |
punished | +0.089 |
himself | +0.078 |
never | +0.077 |
least | +0.073 |
Neg logit contributions | |
oles | -0.193 |
op | -0.184 |
ute | -0.178 |
ew | -0.177 |
ane | -0.176 |
Token | made |
Token ID | 925 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
permission | +0.091 |
himself | +0.090 |
never | +0.086 |
herself | +0.078 |
punished | +0.068 |
Neg logit contributions | |
oms | -0.232 |
ew | -0.232 |
oles | -0.226 |
elly | -0.226 |
ack | -0.219 |
Token | a |
Token ID | 257 |
Feature activation | +8.26e-03 |
Pos logit contributions | |
never | +0.034 |
punished | +0.032 |
cleaned | +0.032 |
learned | +0.032 |
remembered | +0.032 |
Neg logit contributions | |
oles | -0.062 |
Today | -0.060 |
ute | -0.059 |
ay | -0.058 |
arry | -0.058 |
Token | loud |
Token ID | 7812 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
never | +0.015 |
made | +0.015 |
put | +0.014 |
learned | +0.014 |
cleaned | +0.014 |
Neg logit contributions | |
arry | -0.018 |
oles | -0.018 |
Today | -0.017 |
elly | -0.017 |
arl | -0.017 |
Token | sound |
Token ID | 2128 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
arry | +0.049 |
oles | +0.049 |
Today | +0.048 |
elly | +0.048 |
arl | +0.048 |
Neg logit contributions | |
never | -0.031 |
made | -0.030 |
put | -0.029 |
learned | -0.029 |
cleaned | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
oles | +0.037 |
arry | +0.036 |
Today | +0.036 |
elly | +0.036 |
arl | +0.035 |
Neg logit contributions | |
never | -0.028 |
made | -0.027 |
learned | -0.027 |
cleaned | -0.027 |
punished | -0.027 |
Token | they |
Token ID | 484 |
Feature activation | +7.86e-02 |
Pos logit contributions | |
never | +0.011 |
made | +0.011 |
learned | +0.011 |
put | +0.011 |
cleaned | +0.010 |
Neg logit contributions | |
arry | -0.019 |
elly | -0.018 |
Today | -0.018 |
oles | -0.018 |
ute | -0.018 |
Token | loved |
Token ID | 6151 |
Feature activation | +8.15e-03 |
Pos logit contributions | |
punished | +0.131 |
forth | +0.126 |
his | +0.124 |
permission | +0.123 |
asking | +0.120 |
Neg logit contributions | |
owers | -0.126 |
Today | -0.113 |
elly | -0.113 |
oles | -0.112 |
� | -0.112 |
Token | them |
Token ID | 606 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
punished | +0.010 |
never | +0.009 |
cleaned | +0.009 |
fixed | +0.009 |
saved | +0.009 |
Neg logit contributions | |
oles | -0.021 |
elly | -0.020 |
ummy | -0.020 |
arl | -0.020 |
Today | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
Today | +0.044 |
arry | +0.044 |
oles | +0.043 |
elly | +0.043 |
arl | +0.042 |
Neg logit contributions | |
never | -0.033 |
made | -0.032 |
put | -0.031 |
cleaned | -0.031 |
punished | -0.031 |
Token | They |
Token ID | 1119 |
Feature activation | +8.89e-02 |
Pos logit contributions | |
oles | +0.056 |
arry | +0.054 |
ute | +0.053 |
owers | +0.052 |
ane | +0.052 |
Neg logit contributions | |
learned | -0.049 |
made | -0.049 |
never | -0.048 |
put | -0.047 |
found | -0.045 |
Token | also |
Token ID | 635 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
permission | +0.121 |
punished | +0.108 |
rid | +0.108 |
never | +0.104 |
himself | +0.104 |
Neg logit contributions | |
arry | -0.164 |
owers | -0.162 |
olly | -0.157 |
elly | -0.154 |
� | -0.153 |
Token | said |
Token ID | 531 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
permission | +0.093 |
himself | +0.091 |
rid | +0.084 |
never | +0.084 |
ts | +0.081 |
Neg logit contributions | |
arry | -0.200 |
elly | -0.198 |
owers | -0.195 |
utter | -0.186 |
arl | -0.185 |
Token | they |
Token ID | 484 |
Feature activation | +8.18e-02 |
Pos logit contributions | |
never | +0.023 |
made | +0.022 |
put | +0.021 |
learned | +0.021 |
cleaned | +0.021 |
Neg logit contributions | |
arry | -0.031 |
oles | -0.031 |
Today | -0.031 |
elly | -0.030 |
arl | -0.030 |
Token | were |
Token ID | 547 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
punished | +0.131 |
forth | +0.130 |
permission | +0.128 |
his | +0.127 |
asking | +0.127 |
Neg logit contributions | |
owers | -0.135 |
elly | -0.120 |
ane | -0.118 |
� | -0.117 |
Today | -0.115 |
Token | proud |
Token ID | 6613 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
pretended | +0.073 |
tried | +0.072 |
wed | +0.071 |
made | +0.071 |
imagined | +0.070 |
Neg logit contributions | |
arl | -0.068 |
apping | -0.062 |
ummy | -0.062 |
owers | -0.061 |
olly | -0.061 |
Token | of |
Token ID | 286 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
elly | +0.029 |
oles | +0.028 |
Today | +0.028 |
arry | +0.028 |
arl | +0.027 |
Neg logit contributions | |
never | -0.052 |
punished | -0.050 |
cleaned | -0.049 |
fixed | -0.049 |
made | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | -9.21e-03 |
Pos logit contributions | |
punished | +0.035 |
pretended | +0.035 |
made | +0.034 |
cleaned | +0.033 |
never | +0.033 |
Neg logit contributions | |
oles | -0.060 |
Today | -0.057 |
ummy | -0.056 |
arry | -0.056 |
ay | -0.056 |
Token | robot |
Token ID | 9379 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
Today | +0.019 |
ummy | +0.019 |
ay | +0.019 |
owers | +0.018 |
igsaw | +0.018 |
Neg logit contributions | |
never | -0.017 |
remembered | -0.015 |
hoped | -0.015 |
punished | -0.015 |
learned | -0.015 |
Token | 's |
Token ID | 338 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
oles | +0.041 |
Today | +0.040 |
arry | +0.040 |
elly | +0.039 |
ummy | +0.039 |
Neg logit contributions | |
punished | -0.033 |
never | -0.033 |
cleaned | -0.032 |
made | -0.031 |
learned | -0.031 |
Token | creations |
Token ID | 28443 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
never | +0.022 |
made | +0.020 |
punished | +0.020 |
learned | +0.020 |
pretended | +0.019 |
Neg logit contributions | |
Today | -0.032 |
oles | -0.032 |
arry | -0.031 |
owers | -0.031 |
elly | -0.031 |
Token | that |
Token ID | 326 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
oles | +0.030 |
owers | +0.029 |
Today | +0.028 |
arry | +0.028 |
ummy | +0.028 |
Neg logit contributions | |
punished | -0.029 |
never | -0.028 |
cleaned | -0.027 |
pretended | -0.027 |
learned | -0.027 |
Token | they |
Token ID | 484 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
punished | +0.085 |
never | +0.079 |
cleaned | +0.079 |
learned | +0.078 |
put | +0.076 |
Neg logit contributions | |
oles | -0.126 |
owers | -0.118 |
Today | -0.116 |
ane | -0.115 |
arry | -0.113 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
permission | +0.106 |
himself | +0.101 |
wounds | +0.099 |
cast | +0.090 |
wed | +0.090 |
Neg logit contributions | |
owers | -0.220 |
oles | -0.211 |
arry | -0.209 |
aris | -0.202 |
� | -0.201 |
Token | it |
Token ID | 340 |
Feature activation | +2.85e-03 |
Pos logit contributions | |
punished | +0.016 |
never | +0.016 |
pretended | +0.015 |
cleaned | +0.015 |
put | +0.015 |
Neg logit contributions | |
oles | -0.036 |
owers | -0.035 |
arry | -0.035 |
ay | -0.035 |
Today | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
punished | +0.004 |
never | +0.004 |
pretended | +0.004 |
tried | +0.004 |
learned | +0.004 |
Neg logit contributions | |
oles | -0.006 |
elly | -0.006 |
Today | -0.006 |
arry | -0.006 |
ute | -0.006 |
Token | big |
Token ID | 1263 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
never | +0.038 |
made | +0.037 |
put | +0.036 |
learned | +0.036 |
cleaned | +0.036 |
Neg logit contributions | |
arry | -0.054 |
oles | -0.054 |
Today | -0.053 |
elly | -0.053 |
arl | -0.052 |
Token | hug |
Token ID | 16225 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
oles | +0.044 |
arry | +0.044 |
Today | +0.044 |
elly | +0.043 |
arl | +0.043 |
Neg logit contributions | |
never | -0.023 |
made | -0.022 |
put | -0.021 |
learned | -0.021 |
cleaned | -0.021 |
Token | them |
Token ID | 606 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
never | +0.016 |
punished | +0.015 |
learned | +0.015 |
cleaned | +0.015 |
built | +0.015 |
Neg logit contributions | |
oles | -0.023 |
Today | -0.023 |
owers | -0.022 |
ay | -0.022 |
arl | -0.022 |
Token | more |
Token ID | 517 |
Feature activation | -8.38e-03 |
Pos logit contributions | |
Today | +0.040 |
elly | +0.040 |
arry | +0.040 |
oles | +0.040 |
arl | +0.039 |
Neg logit contributions | |
never | -0.022 |
punished | -0.020 |
made | -0.019 |
tried | -0.019 |
remembered | -0.019 |
Token | food |
Token ID | 2057 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
oles | +0.020 |
arry | +0.020 |
Today | +0.020 |
elly | +0.020 |
arl | +0.020 |
Neg logit contributions | |
never | -0.013 |
made | -0.012 |
put | -0.012 |
learned | -0.012 |
cleaned | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
oles | +0.102 |
arry | +0.101 |
Today | +0.100 |
elly | +0.100 |
arl | +0.099 |
Neg logit contributions | |
never | -0.082 |
made | -0.079 |
punished | -0.077 |
learned | -0.077 |
put | -0.077 |
Token | They |
Token ID | 1119 |
Feature activation | +8.97e-02 |
Pos logit contributions | |
oles | +0.059 |
arry | +0.057 |
ute | +0.056 |
ower | +0.054 |
elly | +0.054 |
Neg logit contributions | |
never | -0.057 |
learned | -0.056 |
made | -0.056 |
put | -0.055 |
cleaned | -0.055 |
Token | also |
Token ID | 635 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
permission | +0.096 |
punished | +0.089 |
himself | +0.083 |
never | +0.080 |
asking | +0.076 |
Neg logit contributions | |
owers | -0.198 |
oles | -0.195 |
� | -0.190 |
arry | -0.190 |
utter | -0.179 |
Token | thanked |
Token ID | 26280 |
Feature activation | +2.22e-03 |
Pos logit contributions | |
himself | +0.076 |
never | +0.075 |
permission | +0.069 |
punished | +0.059 |
hurting | +0.056 |
Neg logit contributions | |
owers | -0.231 |
arry | -0.230 |
oles | -0.225 |
elly | -0.224 |
� | -0.220 |
Token | their |
Token ID | 511 |
Feature activation | -6.90e-03 |
Pos logit contributions | |
never | +0.003 |
punished | +0.003 |
made | +0.003 |
buried | +0.003 |
learned | +0.003 |
Neg logit contributions | |
Today | -0.006 |
arry | -0.005 |
arl | -0.005 |
ummy | -0.005 |
oles | -0.005 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
oles | +0.015 |
Today | +0.015 |
arry | +0.015 |
arl | +0.015 |
elly | +0.015 |
Neg logit contributions | |
never | -0.011 |
made | -0.011 |
learned | -0.011 |
punished | -0.011 |
cleaned | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
arry | +0.061 |
oles | +0.061 |
Today | +0.061 |
elly | +0.060 |
arl | +0.059 |
Neg logit contributions | |
never | -0.041 |
made | -0.040 |
put | -0.038 |
learned | -0.038 |
cleaned | -0.038 |
Token | said |
Token ID | 531 |
Feature activation | +5.83e-03 |
Pos logit contributions | |
never | +0.057 |
punished | +0.053 |
buried | +0.049 |
built | +0.048 |
pretended | +0.048 |
Neg logit contributions | |
arl | -0.114 |
arry | -0.114 |
oles | -0.110 |
owers | -0.108 |
Today | -0.107 |
Token | on |
Token ID | 319 |
Feature activation | +6.21e-03 |
Pos logit contributions | |
oles | +0.060 |
ute | +0.059 |
arl | +0.058 |
elly | +0.058 |
arry | +0.058 |
Neg logit contributions | |
never | -0.051 |
punished | -0.049 |
learned | -0.049 |
made | -0.048 |
pretended | -0.048 |
Token | his |
Token ID | 465 |
Feature activation | +9.47e-03 |
Pos logit contributions | |
never | +0.005 |
made | +0.005 |
learned | +0.005 |
cleaned | +0.005 |
punished | +0.005 |
Neg logit contributions | |
oles | -0.019 |
Today | -0.019 |
arry | -0.018 |
elly | -0.018 |
arl | -0.018 |
Token | paw |
Token ID | 41827 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
never | +0.016 |
made | +0.016 |
learned | +0.015 |
punished | +0.015 |
cleaned | +0.015 |
Neg logit contributions | |
ute | -0.020 |
oles | -0.020 |
Today | -0.020 |
arry | -0.019 |
ower | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
oles | +0.082 |
arry | +0.081 |
Today | +0.080 |
elly | +0.079 |
arl | +0.078 |
Neg logit contributions | |
never | -0.062 |
made | -0.060 |
learned | -0.059 |
put | -0.058 |
cleaned | -0.058 |
Token | They |
Token ID | 1119 |
Feature activation | +8.57e-02 |
Pos logit contributions | |
oles | +0.064 |
ute | +0.063 |
arry | +0.062 |
ower | +0.062 |
elly | +0.061 |
Neg logit contributions | |
never | -0.054 |
put | -0.053 |
learned | -0.053 |
made | -0.053 |
cleaned | -0.051 |
Token | also |
Token ID | 635 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
permission | +0.081 |
hurting | +0.078 |
punished | +0.078 |
himself | +0.077 |
never | +0.074 |
Neg logit contributions | |
oles | -0.180 |
op | -0.170 |
utter | -0.170 |
owers | -0.169 |
arry | -0.169 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
himself | +0.078 |
never | +0.076 |
permission | +0.067 |
herself | +0.060 |
hurting | +0.060 |
Neg logit contributions | |
oles | -0.214 |
arry | -0.212 |
elly | -0.211 |
oms | -0.208 |
rig | -0.208 |
Token | him |
Token ID | 683 |
Feature activation | -3.57e-03 |
Pos logit contributions | |
never | +0.020 |
learned | +0.019 |
punished | +0.019 |
cleaned | +0.018 |
fixed | +0.018 |
Neg logit contributions | |
oles | -0.048 |
ummy | -0.047 |
ay | -0.046 |
Today | -0.046 |
arry | -0.046 |
Token | a |
Token ID | 257 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
elly | +0.008 |
Today | +0.008 |
arry | +0.008 |
oles | +0.008 |
ay | +0.008 |
Neg logit contributions | |
never | -0.005 |
punished | -0.004 |
made | -0.004 |
tried | -0.004 |
pretended | -0.004 |
Token | delicious |
Token ID | 12625 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
never | +0.023 |
made | +0.023 |
put | +0.022 |
cleaned | +0.022 |
fixed | +0.022 |
Neg logit contributions | |
oles | -0.038 |
arry | -0.038 |
Today | -0.037 |
elly | -0.037 |
owers | -0.036 |
Token | bowl |
Token ID | 9396 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
Today | +0.066 |
oles | +0.066 |
arry | +0.065 |
elly | +0.064 |
arl | +0.064 |
Neg logit contributions | |
never | -0.055 |
learned | -0.051 |
made | -0.051 |
punished | -0.051 |
cleaned | -0.049 |
Token | kissed |
Token ID | 27946 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
never | +0.077 |
punished | +0.067 |
built | +0.064 |
buried | +0.061 |
saved | +0.061 |
Neg logit contributions | |
oles | -0.111 |
elly | -0.110 |
Today | -0.109 |
arl | -0.108 |
arry | -0.108 |
Token | their |
Token ID | 511 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
oles | +0.050 |
arry | +0.050 |
Today | +0.050 |
elly | +0.049 |
arl | +0.049 |
Neg logit contributions | |
never | -0.023 |
made | -0.022 |
learned | -0.021 |
put | -0.021 |
cleaned | -0.021 |
Token | dolls |
Token ID | 36062 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
Today | +0.038 |
oles | +0.037 |
arl | +0.037 |
arry | +0.037 |
ute | +0.036 |
Neg logit contributions | |
never | -0.023 |
made | -0.022 |
learned | -0.021 |
punished | -0.021 |
cleaned | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
Today | +0.110 |
oles | +0.110 |
arry | +0.110 |
elly | +0.107 |
arl | +0.107 |
Neg logit contributions | |
never | -0.086 |
made | -0.083 |
punished | -0.081 |
learned | -0.080 |
put | -0.080 |
Token | They |
Token ID | 1119 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
oles | +0.056 |
arry | +0.054 |
ane | +0.053 |
owers | +0.052 |
ute | +0.052 |
Neg logit contributions | |
made | -0.045 |
put | -0.044 |
never | -0.043 |
learned | -0.042 |
cleaned | -0.042 |
Token | also |
Token ID | 635 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
permission | +0.115 |
punished | +0.099 |
himself | +0.099 |
never | +0.097 |
finding | +0.094 |
Neg logit contributions | |
owers | -0.176 |
arry | -0.170 |
� | -0.169 |
uffy | -0.164 |
oles | -0.164 |
Token | hugged |
Token ID | 41130 |
Feature activation | +2.49e-03 |
Pos logit contributions | |
himself | +0.095 |
never | +0.093 |
permission | +0.093 |
rid | +0.078 |
punished | +0.072 |
Neg logit contributions | |
elly | -0.207 |
owers | -0.202 |
arry | -0.195 |
Today | -0.192 |
arl | -0.190 |
Token | and |
Token ID | 290 |
Feature activation | +5.83e-02 |
Pos logit contributions | |
never | +0.004 |
made | +0.004 |
learned | +0.004 |
punished | +0.004 |
cleaned | +0.004 |
Neg logit contributions | |
Today | -0.006 |
oles | -0.006 |
arry | -0.006 |
elly | -0.006 |
arl | -0.006 |
Token | kissed |
Token ID | 27946 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
never | +0.083 |
punished | +0.070 |
built | +0.067 |
found | +0.065 |
buried | +0.065 |
Neg logit contributions | |
oles | -0.114 |
Today | -0.113 |
arl | -0.113 |
elly | -0.113 |
owers | -0.113 |
Token | each |
Token ID | 1123 |
Feature activation | +7.93e-03 |
Pos logit contributions | |
oles | +0.050 |
Today | +0.050 |
arry | +0.049 |
elly | +0.049 |
arl | +0.048 |
Neg logit contributions | |
never | -0.023 |
made | -0.022 |
punished | -0.021 |
learned | -0.021 |
cleaned | -0.021 |
Token | other |
Token ID | 584 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
never | +0.008 |
punished | +0.008 |
made | +0.007 |
cleaned | +0.007 |
fixed | +0.007 |
Neg logit contributions | |
Today | -0.023 |
elly | -0.022 |
arl | -0.022 |
arry | -0.022 |
oles | -0.022 |
Token | on |
Token ID | 319 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
never | +0.038 |
made | +0.038 |
put | +0.035 |
learned | +0.034 |
cleaned | +0.033 |
Neg logit contributions | |
oles | -0.069 |
arry | -0.068 |
elly | -0.067 |
arl | -0.065 |
Today | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | -7.70e-03 |
Pos logit contributions | |
arry | +0.027 |
elly | +0.027 |
arl | +0.027 |
oles | +0.027 |
Today | +0.027 |
Neg logit contributions | |
never | -0.028 |
made | -0.027 |
put | -0.026 |
learned | -0.026 |
cleaned | -0.025 |
Token | Sally |
Token ID | 25737 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
arry | +0.019 |
oles | +0.019 |
Today | +0.019 |
elly | +0.019 |
arl | +0.019 |
Neg logit contributions | |
never | -0.011 |
made | -0.011 |
put | -0.010 |
learned | -0.010 |
cleaned | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
punished | +0.060 |
permission | +0.056 |
cleaned | +0.054 |
buried | +0.054 |
never | +0.054 |
Neg logit contributions | |
arry | -0.110 |
owers | -0.109 |
elly | -0.108 |
oles | -0.108 |
ane | -0.106 |
Token | Jimmy |
Token ID | 12963 |
Feature activation | +6.26e-02 |
Pos logit contributions | |
never | +0.039 |
made | +0.038 |
put | +0.037 |
learned | +0.036 |
cleaned | +0.036 |
Neg logit contributions | |
oles | -0.059 |
arry | -0.059 |
Today | -0.058 |
elly | -0.057 |
arl | -0.057 |
Token | always |
Token ID | 1464 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
punished | +0.073 |
permission | +0.063 |
seen | +0.063 |
buried | +0.062 |
cleaned | +0.062 |
Neg logit contributions | |
owers | -0.144 |
arry | -0.139 |
� | -0.136 |
Today | -0.136 |
oles | -0.134 |
Token | weighed |
Token ID | 20261 |
Feature activation | +1.61e-03 |
Pos logit contributions | |
never | +0.071 |
herself | +0.069 |
permission | +0.063 |
punished | +0.063 |
rid | +0.062 |
Neg logit contributions | |
arry | -0.234 |
Today | -0.223 |
� | -0.219 |
owers | -0.210 |
ay | -0.210 |
Token | presents |
Token ID | 10969 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
never | +0.002 |
cleaned | +0.002 |
punished | +0.002 |
learned | +0.002 |
made | +0.002 |
Neg logit contributions | |
oles | -0.004 |
arry | -0.004 |
elly | -0.004 |
Today | -0.004 |
ummy | -0.004 |
Token | before |
Token ID | 878 |
Feature activation | +5.79e-03 |
Pos logit contributions | |
oles | +0.080 |
arry | +0.080 |
elly | +0.078 |
Today | +0.078 |
ummy | +0.076 |
Neg logit contributions | |
never | -0.062 |
punished | -0.061 |
learned | -0.061 |
cleaned | -0.060 |
made | -0.059 |
Token | giving |
Token ID | 3501 |
Feature activation | -5.09e-04 |
Pos logit contributions | |
never | +0.009 |
made | +0.008 |
put | +0.008 |
learned | +0.008 |
cleaned | +0.008 |
Neg logit contributions | |
arry | -0.014 |
oles | -0.014 |
Today | -0.014 |
elly | -0.014 |
arl | -0.014 |
Token | them |
Token ID | 606 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
arry | +0.001 |
oles | +0.001 |
Today | +0.001 |
elly | +0.001 |
ummy | +0.001 |
Neg logit contributions | |
never | -0.001 |
cleaned | -0.001 |
made | -0.001 |
learned | -0.001 |
punished | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
oles | +0.026 |
arry | +0.026 |
Today | +0.026 |
elly | +0.026 |
arl | +0.025 |
Neg logit contributions | |
never | -0.020 |
made | -0.019 |
learned | -0.019 |
cleaned | -0.019 |
put | -0.019 |
Token | their |
Token ID | 511 |
Feature activation | +1.33e-03 |
Pos logit contributions | |
punished | +0.039 |
cleaned | +0.038 |
fixed | +0.037 |
learned | +0.037 |
never | +0.037 |
Neg logit contributions | |
Today | -0.089 |
oles | -0.087 |
arry | -0.084 |
ummy | -0.083 |
arl | -0.083 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.30e-03 |
Pos logit contributions | |
never | +0.002 |
made | +0.002 |
learned | +0.002 |
put | +0.002 |
cleaned | +0.002 |
Neg logit contributions | |
oles | -0.003 |
arry | -0.003 |
Today | -0.003 |
elly | -0.003 |
arl | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
oles | +0.005 |
Today | +0.005 |
arry | +0.005 |
elly | +0.005 |
arl | +0.005 |
Neg logit contributions | |
never | -0.004 |
made | -0.004 |
learned | -0.004 |
cleaned | -0.004 |
put | -0.004 |
Token | They |
Token ID | 1119 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
ute | +0.065 |
arry | +0.064 |
arl | +0.064 |
ogging | +0.063 |
elly | +0.062 |
Neg logit contributions | |
never | -0.066 |
learned | -0.063 |
made | -0.061 |
put | -0.058 |
tried | -0.057 |
Token | also |
Token ID | 635 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
permission | +0.117 |
himself | +0.102 |
yours | +0.093 |
wed | +0.092 |
his | +0.089 |
Neg logit contributions | |
� | -0.165 |
oles | -0.157 |
arry | -0.154 |
uddy | -0.149 |
ho | -0.148 |
Token | did |
Token ID | 750 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
himself | +0.097 |
permission | +0.088 |
herself | +0.082 |
ps | +0.068 |
never | +0.068 |
Neg logit contributions | |
arry | -0.208 |
oke | -0.202 |
ew | -0.201 |
� | -0.199 |
elly | -0.199 |
Token | not |
Token ID | 407 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
cleaned | +0.041 |
buried | +0.038 |
made | +0.038 |
found | +0.038 |
pretended | +0.037 |
Neg logit contributions | |
Today | -0.096 |
oles | -0.093 |
arry | -0.090 |
� | -0.089 |
apping | -0.089 |
Token | see |
Token ID | 766 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
never | +0.068 |
saved | +0.065 |
buried | +0.064 |
hoped | +0.064 |
cleaned | +0.062 |
Neg logit contributions | |
arl | -0.110 |
arry | -0.110 |
Today | -0.107 |
elly | -0.105 |
apping | -0.103 |
Token | that |
Token ID | 326 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
never | +0.032 |
cleaned | +0.031 |
made | +0.031 |
learned | +0.031 |
punished | +0.030 |
Neg logit contributions | |
oles | -0.081 |
Today | -0.080 |
elly | -0.080 |
arry | -0.079 |
arl | -0.078 |
Token | Ted |
Token ID | 11396 |
Feature activation | -4.84e-03 |
Pos logit contributions | |
never | +0.035 |
cleaned | +0.032 |
punished | +0.032 |
made | +0.032 |
put | +0.032 |
Neg logit contributions | |
oles | -0.067 |
Today | -0.067 |
arry | -0.065 |
elly | -0.065 |
ute | -0.063 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
punished | +0.031 |
cleaned | +0.031 |
fixed | +0.030 |
made | +0.029 |
never | +0.029 |
Neg logit contributions | |
Today | -0.043 |
arry | -0.042 |
oles | -0.042 |
arl | -0.041 |
ummy | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
arry | +0.069 |
oles | +0.069 |
Today | +0.069 |
elly | +0.067 |
arl | +0.066 |
Neg logit contributions | |
never | -0.059 |
made | -0.057 |
cleaned | -0.055 |
punished | -0.055 |
learned | -0.055 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
punished | +0.052 |
never | +0.050 |
cleaned | +0.049 |
fixed | +0.048 |
learned | +0.045 |
Neg logit contributions | |
arry | -0.102 |
oles | -0.102 |
Today | -0.100 |
arl | -0.099 |
olly | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
oles | +0.051 |
arry | +0.050 |
Today | +0.049 |
arl | +0.049 |
elly | +0.048 |
Neg logit contributions | |
never | -0.029 |
made | -0.029 |
put | -0.027 |
learned | -0.027 |
fixed | -0.027 |
Token | They |
Token ID | 1119 |
Feature activation | +8.95e-02 |
Pos logit contributions | |
andel | +0.057 |
oles | +0.056 |
arry | +0.054 |
owers | +0.054 |
igsaw | +0.053 |
Neg logit contributions | |
made | -0.046 |
learned | -0.046 |
never | -0.046 |
put | -0.044 |
his | -0.043 |
Token | also |
Token ID | 635 |
Feature activation | +9.99e-02 |
Pos logit contributions | |
permission | +0.109 |
punished | +0.107 |
worrying | +0.092 |
never | +0.092 |
hurting | +0.091 |
Neg logit contributions | |
owers | -0.188 |
olly | -0.186 |
oles | -0.185 |
arry | -0.181 |
� | -0.181 |
Token | said |
Token ID | 531 |
Feature activation | +9.77e-03 |
Pos logit contributions | |
permission | +0.100 |
never | +0.088 |
himself | +0.087 |
punished | +0.084 |
rid | +0.080 |
Neg logit contributions | |
owers | -0.209 |
elly | -0.205 |
arry | -0.203 |
� | -0.200 |
oles | -0.199 |
Token | sorry |
Token ID | 7926 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
never | +0.017 |
made | +0.017 |
learned | +0.016 |
put | +0.016 |
cleaned | +0.016 |
Neg logit contributions | |
arry | -0.022 |
Today | -0.021 |
elly | -0.021 |
oles | -0.021 |
arl | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
Today | +0.027 |
arry | +0.027 |
oles | +0.027 |
elly | +0.027 |
arl | +0.026 |
Neg logit contributions | |
never | -0.020 |
made | -0.019 |
punished | -0.019 |
cleaned | -0.019 |
learned | -0.019 |
Token | each |
Token ID | 1123 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
never | +0.035 |
cleaned | +0.035 |
punished | +0.035 |
made | +0.035 |
fixed | +0.035 |
Neg logit contributions | |
Today | -0.058 |
oles | -0.057 |
arry | -0.056 |
arl | -0.055 |
elly | -0.054 |
Token | other |
Token ID | 584 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
never | +0.023 |
cleaned | +0.022 |
punished | +0.022 |
fixed | +0.021 |
made | +0.020 |
Neg logit contributions | |
oles | -0.058 |
elly | -0.058 |
Today | -0.058 |
arry | -0.057 |
ute | -0.056 |
Token | sorts |
Token ID | 10524 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
never | +0.026 |
punished | +0.024 |
learned | +0.023 |
cleaned | +0.023 |
fixed | +0.023 |
Neg logit contributions | |
oles | -0.035 |
Today | -0.035 |
arry | -0.034 |
elly | -0.033 |
owers | -0.033 |
Token | of |
Token ID | 286 |
Feature activation | +3.50e-03 |
Pos logit contributions | |
oles | +0.021 |
arry | +0.021 |
elly | +0.020 |
ummy | +0.020 |
apping | +0.020 |
Neg logit contributions | |
never | -0.045 |
punished | -0.045 |
made | -0.043 |
learned | -0.043 |
remembered | -0.043 |
Token | clothes |
Token ID | 8242 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
never | +0.007 |
remembered | +0.006 |
punished | +0.006 |
learned | +0.006 |
managed | +0.006 |
Neg logit contributions | |
Today | -0.007 |
elly | -0.007 |
oles | -0.007 |
andel | -0.007 |
oms | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
oles | +0.061 |
arry | +0.059 |
Today | +0.059 |
elly | +0.059 |
owers | +0.057 |
Neg logit contributions | |
never | -0.060 |
punished | -0.058 |
learned | -0.058 |
cleaned | -0.057 |
saved | -0.055 |
Token | They |
Token ID | 1119 |
Feature activation | +8.75e-02 |
Pos logit contributions | |
ute | +0.084 |
arl | +0.083 |
elly | +0.082 |
andel | +0.082 |
oles | +0.081 |
Neg logit contributions | |
made | -0.063 |
put | -0.063 |
never | -0.062 |
learned | -0.062 |
tried | -0.061 |
Token | also |
Token ID | 635 |
Feature activation | +9.98e-02 |
Pos logit contributions | |
punished | +0.128 |
never | +0.116 |
rid | +0.111 |
saved | +0.103 |
hurting | +0.102 |
Neg logit contributions | |
oles | -0.189 |
arry | -0.184 |
owers | -0.171 |
apping | -0.170 |
olly | -0.169 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
punished | +0.099 |
never | +0.099 |
rid | +0.080 |
care | +0.075 |
hurting | +0.075 |
Neg logit contributions | |
arry | -0.242 |
oles | -0.237 |
apping | -0.230 |
oms | -0.226 |
ay | -0.226 |
Token | some |
Token ID | 617 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
never | +0.056 |
punished | +0.053 |
cleaned | +0.051 |
learned | +0.050 |
made | +0.050 |
Neg logit contributions | |
oles | -0.104 |
elly | -0.101 |
arry | -0.100 |
Today | -0.099 |
ummy | -0.098 |
Token | toys |
Token ID | 14958 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
never | +0.027 |
punished | +0.024 |
learned | +0.024 |
fixed | +0.023 |
cleaned | +0.023 |
Neg logit contributions | |
oles | -0.035 |
elly | -0.034 |
arry | -0.033 |
Today | -0.032 |
owers | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
oles | +0.067 |
arry | +0.065 |
Today | +0.064 |
elly | +0.063 |
owers | +0.062 |
Neg logit contributions | |
never | -0.065 |
punished | -0.064 |
learned | -0.063 |
cleaned | -0.061 |
fixed | -0.060 |
Token | too |
Token ID | 1165 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
never | +0.039 |
punished | +0.037 |
remembered | +0.035 |
cleaned | +0.034 |
learned | +0.034 |
Neg logit contributions | |
Today | -0.052 |
oles | -0.052 |
elly | -0.051 |
arry | -0.049 |
aris | -0.049 |
Token | in |
Token ID | 287 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
Today | +0.071 |
ute | +0.071 |
arl | +0.071 |
arry | +0.071 |
oles | +0.070 |
Neg logit contributions | |
never | -0.054 |
learned | -0.052 |
made | -0.051 |
punished | -0.051 |
cleaned | -0.049 |
Token | her |
Token ID | 607 |
Feature activation | +4.44e-03 |
Pos logit contributions | |
never | +0.015 |
punished | +0.015 |
saved | +0.014 |
learned | +0.014 |
cleaned | +0.014 |
Neg logit contributions | |
Today | -0.039 |
ay | -0.036 |
arl | -0.036 |
arry | -0.035 |
ummy | -0.035 |
Token | bag |
Token ID | 6131 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
never | +0.008 |
made | +0.008 |
learned | +0.008 |
put | +0.008 |
cleaned | +0.008 |
Neg logit contributions | |
Today | -0.009 |
arry | -0.009 |
oles | -0.009 |
elly | -0.009 |
arl | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
arry | +0.104 |
oles | +0.104 |
Today | +0.103 |
elly | +0.102 |
arl | +0.100 |
Neg logit contributions | |
never | -0.081 |
made | -0.078 |
learned | -0.076 |
put | -0.075 |
cleaned | -0.075 |
Token | She |
Token ID | 1375 |
Feature activation | +9.08e-02 |
Pos logit contributions | |
oles | +0.068 |
ute | +0.063 |
arry | +0.063 |
oing | +0.062 |
owers | +0.062 |
Neg logit contributions | |
put | -0.049 |
made | -0.048 |
never | -0.047 |
learned | -0.045 |
cleaned | -0.045 |
Token | also |
Token ID | 635 |
Feature activation | +9.98e-02 |
Pos logit contributions | |
punished | +0.085 |
permission | +0.084 |
never | +0.081 |
remind | +0.077 |
hurting | +0.076 |
Neg logit contributions | |
arry | -0.215 |
op | -0.212 |
oles | -0.208 |
� | -0.208 |
olly | -0.208 |
Token | put |
Token ID | 1234 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
never | +0.103 |
permission | +0.086 |
herself | +0.082 |
punished | +0.081 |
himself | +0.081 |
Neg logit contributions | |
Today | -0.219 |
elly | -0.214 |
suits | -0.213 |
arry | -0.210 |
� | -0.209 |
Token | her |
Token ID | 607 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
never | +0.056 |
learned | +0.053 |
punished | +0.053 |
forgot | +0.053 |
pretended | +0.050 |
Neg logit contributions | |
ay | -0.073 |
arl | -0.073 |
Today | -0.072 |
ute | -0.069 |
owers | -0.068 |
Token | doll |
Token ID | 3654 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
never | +0.018 |
learned | +0.017 |
punished | +0.017 |
made | +0.017 |
saved | +0.016 |
Neg logit contributions | |
Today | -0.022 |
ute | -0.022 |
arl | -0.022 |
ay | -0.021 |
uffy | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +6.96e-02 |
Pos logit contributions | |
Today | +0.069 |
ute | +0.069 |
arry | +0.068 |
oles | +0.068 |
elly | +0.066 |
Neg logit contributions | |
never | -0.054 |
learned | -0.051 |
made | -0.051 |
punished | -0.050 |
cleaned | -0.049 |
Token | a |
Token ID | 257 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
never | +0.108 |
punished | +0.099 |
least | +0.094 |
saved | +0.094 |
learned | +0.093 |
Neg logit contributions | |
Today | -0.124 |
uffy | -0.119 |
oles | -0.116 |
owers | -0.115 |
ower | -0.114 |
Token | the |
Token ID | 262 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
oles | +0.013 |
arry | +0.013 |
Today | +0.013 |
owers | +0.013 |
elly | +0.013 |
Neg logit contributions | |
cleaned | -0.005 |
never | -0.005 |
punished | -0.005 |
made | -0.005 |
put | -0.005 |
Token | people |
Token ID | 661 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
Today | +0.046 |
arry | +0.046 |
oles | +0.045 |
elly | +0.044 |
ute | +0.044 |
Neg logit contributions | |
never | -0.045 |
punished | -0.043 |
made | -0.043 |
cleaned | -0.042 |
put | -0.041 |
Token | of |
Token ID | 286 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
punished | +0.039 |
cleaned | +0.038 |
pretended | +0.037 |
never | +0.037 |
fixed | +0.037 |
Neg logit contributions | |
oles | -0.077 |
ane | -0.076 |
arry | -0.076 |
ute | -0.075 |
owers | -0.075 |
Token | the |
Token ID | 262 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
never | +0.030 |
made | +0.027 |
cleaned | +0.027 |
punished | +0.027 |
learned | +0.026 |
Neg logit contributions | |
oles | -0.086 |
arry | -0.085 |
Today | -0.085 |
elly | -0.084 |
arl | -0.083 |
Token | town |
Token ID | 3240 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
arry | +0.035 |
Today | +0.034 |
oles | +0.033 |
elly | +0.033 |
ute | +0.032 |
Neg logit contributions | |
never | -0.041 |
made | -0.040 |
punished | -0.039 |
cleaned | -0.039 |
put | -0.038 |
Token | always |
Token ID | 1464 |
Feature activation | +9.95e-02 |
Pos logit contributions | |
punished | +0.003 |
cleaned | +0.003 |
never | +0.003 |
put | +0.003 |
buried | +0.003 |
Neg logit contributions | |
oles | -0.008 |
arry | -0.008 |
ane | -0.008 |
apping | -0.008 |
owers | -0.008 |
Token | nodded |
Token ID | 14464 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
never | +0.068 |
punished | +0.060 |
buried | +0.049 |
pretended | +0.047 |
herself | +0.047 |
Neg logit contributions | |
arry | -0.258 |
Today | -0.250 |
apping | -0.249 |
owers | -0.241 |
elly | -0.240 |
Token | when |
Token ID | 618 |
Feature activation | +8.02e-03 |
Pos logit contributions | |
never | +0.021 |
punished | +0.020 |
cleaned | +0.020 |
made | +0.020 |
learned | +0.020 |
Neg logit contributions | |
Today | -0.035 |
oles | -0.035 |
arry | -0.035 |
elly | -0.034 |
arl | -0.033 |
Token | they |
Token ID | 484 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
cleaned | +0.014 |
punished | +0.014 |
fixed | +0.014 |
never | +0.013 |
made | +0.013 |
Neg logit contributions | |
oles | -0.017 |
arry | -0.016 |
Today | -0.016 |
owers | -0.015 |
ummy | -0.015 |
Token | passed |
Token ID | 3804 |
Feature activation | +6.45e-03 |
Pos logit contributions | |
never | +0.071 |
punished | +0.069 |
cleaned | +0.064 |
pretended | +0.062 |
fixed | +0.061 |
Neg logit contributions | |
arry | -0.191 |
owers | -0.186 |
oles | -0.186 |
Today | -0.184 |
ane | -0.183 |
Token | by |
Token ID | 416 |
Feature activation | -9.59e-03 |
Pos logit contributions | |
never | +0.008 |
made | +0.008 |
cleaned | +0.008 |
put | +0.008 |
learned | +0.008 |
Neg logit contributions | |
arry | -0.016 |
oles | -0.016 |
Today | -0.016 |
elly | -0.016 |
ute | -0.015 |
Token | their |
Token ID | 511 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
never | +0.088 |
punished | +0.087 |
cleaned | +0.086 |
fixed | +0.083 |
learned | +0.078 |
Neg logit contributions | |
oles | -0.126 |
Today | -0.117 |
aris | -0.117 |
elly | -0.115 |
arry | -0.115 |
Token | bright |
Token ID | 6016 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
never | +0.043 |
made | +0.041 |
punished | +0.041 |
cleaned | +0.041 |
learned | +0.040 |
Neg logit contributions | |
oles | -0.059 |
arry | -0.058 |
Today | -0.058 |
ute | -0.057 |
arl | -0.057 |
Token | sails |
Token ID | 48370 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
arry | +0.022 |
oles | +0.022 |
Today | +0.022 |
elly | +0.021 |
arl | +0.021 |
Neg logit contributions | |
never | -0.019 |
made | -0.018 |
learned | -0.018 |
cleaned | -0.018 |
punished | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
oles | +0.046 |
arry | +0.045 |
Today | +0.045 |
ute | +0.044 |
elly | +0.044 |
Neg logit contributions | |
never | -0.038 |
punished | -0.036 |
learned | -0.035 |
cleaned | -0.035 |
made | -0.035 |
Token | They |
Token ID | 1119 |
Feature activation | +8.35e-02 |
Pos logit contributions | |
oles | +0.081 |
arry | +0.078 |
ower | +0.076 |
owers | +0.076 |
ute | +0.074 |
Neg logit contributions | |
learned | -0.075 |
made | -0.075 |
put | -0.073 |
never | -0.073 |
found | -0.070 |
Token | also |
Token ID | 635 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
punished | +0.117 |
never | +0.095 |
cleaned | +0.092 |
permission | +0.091 |
fixed | +0.089 |
Neg logit contributions | |
oles | -0.182 |
arry | -0.176 |
aris | -0.171 |
oms | -0.170 |
ane | -0.170 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
punished | +0.090 |
never | +0.079 |
permission | +0.074 |
cleaned | +0.072 |
himself | +0.066 |
Neg logit contributions | |
arry | -0.241 |
Today | -0.231 |
oles | -0.231 |
aris | -0.230 |
oms | -0.227 |
Token | a |
Token ID | 257 |
Feature activation | +1.03e-03 |
Pos logit contributions | |
never | +0.056 |
punished | +0.054 |
cleaned | +0.054 |
fixed | +0.052 |
made | +0.051 |
Neg logit contributions | |
oles | -0.101 |
elly | -0.097 |
arry | -0.097 |
Today | -0.097 |
ute | -0.095 |
Token | tug |
Token ID | 27762 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
never | +0.002 |
made | +0.002 |
put | +0.002 |
learned | +0.002 |
fixed | +0.002 |
Neg logit contributions | |
arry | -0.002 |
Today | -0.002 |
oles | -0.002 |
elly | -0.002 |
arl | -0.002 |
Token | boat |
Token ID | 24482 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
oles | +0.071 |
arry | +0.069 |
elly | +0.069 |
Today | +0.069 |
ute | +0.067 |
Neg logit contributions | |
never | -0.065 |
learned | -0.062 |
made | -0.062 |
cleaned | -0.062 |
punished | -0.061 |
Token | which |
Token ID | 543 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
oles | +0.063 |
arry | +0.063 |
Today | +0.062 |
elly | +0.061 |
ute | +0.061 |
Neg logit contributions | |
never | -0.048 |
made | -0.046 |
learned | -0.045 |
cleaned | -0.045 |
put | -0.045 |
Token | back |
Token ID | 736 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
arl | +0.083 |
elly | +0.082 |
arry | +0.082 |
Today | +0.082 |
oles | +0.081 |
Neg logit contributions | |
never | -0.075 |
punished | -0.072 |
made | -0.070 |
pretended | -0.070 |
remembered | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
oles | +0.026 |
arry | +0.026 |
elly | +0.025 |
Today | +0.025 |
arl | +0.025 |
Neg logit contributions | |
never | -0.022 |
made | -0.021 |
punished | -0.021 |
learned | -0.020 |
put | -0.020 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
never | +0.083 |
supposed | +0.079 |
always | +0.070 |
eling | +0.069 |
punished | +0.069 |
Neg logit contributions | |
uddle | -0.108 |
ite | -0.107 |
oms | -0.107 |
oles | -0.106 |
ogging | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
oles | +0.052 |
arry | +0.052 |
Today | +0.051 |
elly | +0.050 |
owers | +0.050 |
Neg logit contributions | |
never | -0.033 |
made | -0.032 |
put | -0.031 |
learned | -0.031 |
fixed | -0.031 |
Token | They |
Token ID | 1119 |
Feature activation | +8.74e-02 |
Pos logit contributions | |
oles | +0.062 |
arry | +0.060 |
owers | +0.059 |
ane | +0.059 |
ower | +0.058 |
Neg logit contributions | |
learned | -0.050 |
made | -0.050 |
put | -0.049 |
never | -0.049 |
cleaned | -0.046 |
Token | also |
Token ID | 635 |
Feature activation | +9.90e-02 |
Pos logit contributions | |
permission | +0.121 |
asking | +0.108 |
supposed | +0.106 |
worrying | +0.106 |
pretending | +0.103 |
Neg logit contributions | |
� | -0.161 |
utter | -0.155 |
olly | -0.154 |
arry | -0.153 |
owers | -0.152 |
Token | offered |
Token ID | 4438 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
permission | +0.092 |
himself | +0.088 |
never | +0.085 |
rid | +0.079 |
supposed | +0.079 |
Neg logit contributions | |
elly | -0.198 |
arry | -0.197 |
� | -0.192 |
owers | -0.189 |
utter | -0.187 |
Token | to |
Token ID | 284 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
punished | +0.034 |
never | +0.033 |
fixed | +0.033 |
saved | +0.032 |
cleaned | +0.032 |
Neg logit contributions | |
oles | -0.043 |
elly | -0.043 |
Today | -0.042 |
ummy | -0.042 |
arl | -0.040 |
Token | share |
Token ID | 2648 |
Feature activation | +3.00e-03 |
Pos logit contributions | |
remembered | +0.049 |
wished | +0.049 |
punished | +0.049 |
pretended | +0.047 |
learned | +0.045 |
Neg logit contributions | |
arl | -0.084 |
� | -0.082 |
ay | -0.080 |
ute | -0.079 |
Today | -0.078 |
Token | their |
Token ID | 511 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
never | +0.004 |
punished | +0.004 |
cleaned | +0.004 |
made | +0.004 |
saved | +0.004 |
Neg logit contributions | |
Today | -0.007 |
elly | -0.007 |
oles | -0.007 |
arl | -0.007 |
ute | -0.006 |
Token | new |
Token ID | 649 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
Today | +0.030 |
arl | +0.028 |
arry | +0.028 |
oles | +0.028 |
elly | +0.028 |
Neg logit contributions | |
never | -0.014 |
made | -0.013 |
learned | -0.012 |
punished | -0.012 |
cleaned | -0.012 |
Token | in |
Token ID | 287 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
oles | +0.050 |
elly | +0.050 |
Today | +0.050 |
ute | +0.050 |
arry | +0.050 |
Neg logit contributions | |
never | -0.037 |
punished | -0.037 |
learned | -0.037 |
cleaned | -0.035 |
made | -0.035 |
Token | shiny |
Token ID | 22441 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
punished | +0.021 |
never | +0.020 |
learned | +0.020 |
cleaned | +0.019 |
saved | +0.019 |
Neg logit contributions | |
Today | -0.034 |
ay | -0.032 |
ummy | -0.031 |
arry | -0.031 |
arl | -0.031 |
Token | paper |
Token ID | 3348 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
Today | +0.042 |
arry | +0.042 |
elly | +0.042 |
oles | +0.041 |
ute | +0.041 |
Neg logit contributions | |
never | -0.031 |
punished | -0.029 |
learned | -0.029 |
made | -0.028 |
cleaned | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
arry | +0.154 |
oles | +0.154 |
Today | +0.153 |
elly | +0.150 |
arl | +0.149 |
Neg logit contributions | |
never | -0.114 |
made | -0.110 |
put | -0.106 |
learned | -0.106 |
cleaned | -0.106 |
Token | He |
Token ID | 679 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
oles | +0.061 |
arry | +0.059 |
ute | +0.057 |
owers | +0.057 |
elly | +0.057 |
Neg logit contributions | |
never | -0.048 |
made | -0.048 |
put | -0.048 |
fixed | -0.046 |
cleaned | -0.046 |
Token | also |
Token ID | 635 |
Feature activation | +9.89e-02 |
Pos logit contributions | |
permission | +0.097 |
care | +0.096 |
punished | +0.092 |
never | +0.086 |
hurting | +0.076 |
Neg logit contributions | |
oles | -0.200 |
arry | -0.193 |
elly | -0.190 |
olly | -0.186 |
uffy | -0.186 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
never | +0.102 |
permission | +0.097 |
care | +0.097 |
punished | +0.096 |
himself | +0.091 |
Neg logit contributions | |
oms | -0.213 |
arry | -0.211 |
oles | -0.208 |
elly | -0.207 |
owers | -0.201 |
Token | them |
Token ID | 606 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
learned | +0.025 |
never | +0.024 |
punished | +0.023 |
cleaned | +0.023 |
pretended | +0.023 |
Neg logit contributions | |
ay | -0.035 |
Today | -0.035 |
ummy | -0.035 |
arl | -0.035 |
elly | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
arry | +0.049 |
Today | +0.048 |
elly | +0.048 |
oles | +0.048 |
ay | +0.047 |
Neg logit contributions | |
never | -0.026 |
punished | -0.026 |
learned | -0.024 |
tried | -0.023 |
made | -0.023 |
Token | hug |
Token ID | 16225 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
made | +0.023 |
never | +0.023 |
put | +0.023 |
fixed | +0.022 |
cleaned | +0.022 |
Neg logit contributions | |
oles | -0.036 |
arry | -0.035 |
elly | -0.035 |
Today | -0.034 |
owers | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
Today | +0.071 |
arry | +0.071 |
oles | +0.071 |
elly | +0.069 |
arl | +0.068 |
Neg logit contributions | |
never | -0.054 |
made | -0.052 |
learned | -0.051 |
punished | -0.051 |
put | -0.051 |
Token | their |
Token ID | 511 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
cleaned | +0.036 |
fixed | +0.036 |
pretended | +0.035 |
made | +0.034 |
punished | +0.034 |
Neg logit contributions | |
oles | -0.061 |
Today | -0.058 |
ummy | -0.058 |
arl | -0.058 |
ute | -0.058 |
Token | elderly |
Token ID | 13830 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
never | +0.023 |
made | +0.022 |
cleaned | +0.022 |
put | +0.021 |
fixed | +0.021 |
Neg logit contributions | |
arry | -0.028 |
Today | -0.028 |
oles | -0.028 |
elly | -0.027 |
arl | -0.027 |
Token | companion |
Token ID | 15185 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
oles | +0.116 |
arry | +0.113 |
elly | +0.111 |
ummy | +0.111 |
Today | +0.110 |
Neg logit contributions | |
never | -0.083 |
cleaned | -0.083 |
punished | -0.082 |
fixed | -0.082 |
made | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
oles | +0.049 |
arry | +0.049 |
Today | +0.048 |
elly | +0.047 |
ute | +0.046 |
Neg logit contributions | |
never | -0.040 |
made | -0.039 |
cleaned | -0.039 |
punished | -0.039 |
put | -0.039 |
Token | They |
Token ID | 1119 |
Feature activation | +8.96e-02 |
Pos logit contributions | |
oles | +0.074 |
ower | +0.073 |
arl | +0.073 |
ute | +0.071 |
arry | +0.071 |
Neg logit contributions | |
never | -0.076 |
learned | -0.074 |
made | -0.072 |
put | -0.069 |
fixed | -0.068 |
Token | also |
Token ID | 635 |
Feature activation | +9.88e-02 |
Pos logit contributions | |
punished | +0.115 |
permission | +0.091 |
cleaned | +0.091 |
fixed | +0.090 |
tightened | +0.089 |
Neg logit contributions | |
arry | -0.201 |
owers | -0.199 |
oles | -0.199 |
olly | -0.189 |
utter | -0.188 |
Token | found |
Token ID | 1043 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
punished | +0.068 |
tightened | +0.058 |
ts | +0.057 |
cleaned | +0.057 |
fixed | +0.055 |
Neg logit contributions | |
arry | -0.253 |
oles | -0.244 |
owers | -0.241 |
Today | -0.240 |
apping | -0.230 |
Token | a |
Token ID | 257 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
never | +0.052 |
made | +0.049 |
punished | +0.049 |
cleaned | +0.049 |
put | +0.048 |
Neg logit contributions | |
oles | -0.107 |
arry | -0.105 |
elly | -0.104 |
Today | -0.102 |
owers | -0.102 |
Token | bone |
Token ID | 9970 |
Feature activation | -5.25e-02 |
Pos logit contributions | |
never | +0.022 |
made | +0.021 |
learned | +0.021 |
put | +0.021 |
cleaned | +0.021 |
Neg logit contributions | |
oles | -0.028 |
arry | -0.028 |
Today | -0.028 |
elly | -0.027 |
arl | -0.027 |
Token | near |
Token ID | 1474 |
Feature activation | +3.01e-04 |
Pos logit contributions | |
oles | +0.118 |
arry | +0.115 |
elly | +0.114 |
Today | +0.114 |
ute | +0.112 |
Neg logit contributions | |
never | -0.083 |
punished | -0.082 |
learned | -0.081 |
cleaned | -0.080 |
made | -0.078 |
Token | the |
Token ID | 262 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
never | +0.000 |
made | +0.000 |
cleaned | +0.000 |
learned | +0.000 |
punished | +0.000 |
Neg logit contributions | |
oles | -0.001 |
arry | -0.001 |
Today | -0.001 |
elly | -0.001 |
arl | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +4.79e-02 |
Pos logit contributions | |
oles | +0.067 |
Today | +0.064 |
arry | +0.064 |
elly | +0.063 |
owers | +0.061 |
Neg logit contributions | |
never | -0.054 |
punished | -0.053 |
cleaned | -0.052 |
made | -0.052 |
learned | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | -7.59e-03 |
Pos logit contributions | |
never | +0.055 |
punished | +0.054 |
cleaned | +0.054 |
fixed | +0.054 |
pretended | +0.050 |
Neg logit contributions | |
oles | -0.107 |
Today | -0.101 |
oms | -0.100 |
arl | -0.098 |
aris | -0.098 |
Token | flowers |
Token ID | 12734 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
oles | +0.016 |
arry | +0.016 |
Today | +0.016 |
elly | +0.015 |
arl | +0.015 |
Neg logit contributions | |
never | -0.014 |
made | -0.013 |
cleaned | -0.013 |
learned | -0.013 |
put | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
oles | +0.069 |
Today | +0.067 |
arry | +0.066 |
elly | +0.065 |
owers | +0.065 |
Neg logit contributions | |
never | -0.058 |
punished | -0.056 |
made | -0.055 |
cleaned | -0.055 |
learned | -0.055 |
Token | They |
Token ID | 1119 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
oles | +0.053 |
arry | +0.050 |
ute | +0.050 |
opus | +0.050 |
ogging | +0.049 |
Neg logit contributions | |
never | -0.040 |
learned | -0.039 |
made | -0.037 |
put | -0.037 |
hoped | -0.036 |
Token | also |
Token ID | 635 |
Feature activation | +9.87e-02 |
Pos logit contributions | |
permission | +0.108 |
punished | +0.104 |
rid | +0.104 |
hurting | +0.101 |
care | +0.100 |
Neg logit contributions | |
arry | -0.178 |
oles | -0.169 |
owers | -0.159 |
utter | -0.158 |
olly | -0.154 |
Token | told |
Token ID | 1297 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
never | +0.089 |
himself | +0.084 |
permission | +0.082 |
rid | +0.082 |
punished | +0.079 |
Neg logit contributions | |
arry | -0.218 |
oms | -0.203 |
elly | -0.201 |
aris | -0.198 |
owers | -0.196 |
Token | her |
Token ID | 607 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
punished | +0.022 |
made | +0.022 |
fixed | +0.022 |
never | +0.022 |
cleaned | +0.022 |
Neg logit contributions | |
oles | -0.051 |
Today | -0.051 |
ummy | -0.050 |
arry | -0.049 |
arl | -0.049 |
Token | about |
Token ID | 546 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
never | +0.024 |
made | +0.024 |
punished | +0.024 |
cleaned | +0.024 |
fixed | +0.024 |
Neg logit contributions | |
oles | -0.037 |
arry | -0.035 |
Today | -0.035 |
ute | -0.034 |
elly | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
punished | +0.033 |
cleaned | +0.032 |
learned | +0.032 |
fixed | +0.032 |
never | +0.032 |
Neg logit contributions | |
oles | -0.085 |
Today | -0.083 |
arry | -0.080 |
owers | -0.080 |
arl | -0.080 |
Token | old |
Token ID | 1468 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
Today | +0.030 |
oles | +0.030 |
arry | +0.030 |
elly | +0.029 |
arl | +0.029 |
Neg logit contributions | |
never | -0.027 |
made | -0.026 |
cleaned | -0.025 |
punished | -0.025 |
learned | -0.025 |
Token | it |
Token ID | 340 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
never | +0.040 |
punished | +0.038 |
cleaned | +0.038 |
learned | +0.036 |
made | +0.036 |
Neg logit contributions | |
oles | -0.094 |
elly | -0.091 |
Today | -0.090 |
ute | -0.089 |
arry | -0.089 |
Token | come |
Token ID | 1282 |
Feature activation | -2.90e-03 |
Pos logit contributions | |
never | +0.018 |
punished | +0.018 |
cleaned | +0.018 |
learned | +0.018 |
made | +0.017 |
Neg logit contributions | |
oles | -0.029 |
Today | -0.028 |
arry | -0.028 |
elly | -0.027 |
ute | -0.027 |
Token | out |
Token ID | 503 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
arry | +0.008 |
elly | +0.008 |
oles | +0.008 |
Today | +0.007 |
owers | +0.007 |
Neg logit contributions | |
never | -0.004 |
made | -0.004 |
put | -0.004 |
fixed | -0.004 |
found | -0.004 |
Token | of |
Token ID | 286 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
arry | +0.075 |
oles | +0.074 |
Today | +0.073 |
arl | +0.072 |
elly | +0.072 |
Neg logit contributions | |
never | -0.071 |
made | -0.070 |
put | -0.068 |
learned | -0.066 |
cleaned | -0.066 |
Token | the |
Token ID | 262 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
punished | +0.027 |
never | +0.025 |
learned | +0.025 |
forgot | +0.024 |
cleaned | +0.023 |
Neg logit contributions | |
Today | -0.051 |
arl | -0.050 |
oles | -0.049 |
ummy | -0.049 |
elly | -0.048 |
Token | seaw |
Token ID | 32085 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
arry | +0.061 |
oles | +0.061 |
Today | +0.060 |
elly | +0.059 |
arl | +0.059 |
Neg logit contributions | |
never | -0.033 |
made | -0.033 |
put | -0.031 |
learned | -0.031 |
cleaned | -0.031 |
Token | eed |
Token ID | 2308 |
Feature activation | -7.13e-02 |
Pos logit contributions | |
Today | +0.236 |
oles | +0.220 |
owers | +0.212 |
arry | +0.209 |
ane | +0.194 |
Neg logit contributions | |
put | -0.359 |
made | -0.356 |
never | -0.355 |
learned | -0.344 |
found | -0.343 |
Token | and |
Token ID | 290 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
arry | +0.166 |
oles | +0.165 |
Today | +0.164 |
elly | +0.161 |
arl | +0.160 |
Neg logit contributions | |
never | -0.115 |
made | -0.112 |
put | -0.108 |
learned | -0.107 |
cleaned | -0.106 |
Token | walk |
Token ID | 2513 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
punished | +0.071 |
never | +0.068 |
cleaned | +0.064 |
fixed | +0.064 |
repaired | +0.059 |
Neg logit contributions | |
oles | -0.123 |
Today | -0.117 |
oms | -0.109 |
owers | -0.108 |
ore | -0.107 |
Token | around |
Token ID | 1088 |
Feature activation | -3.49e-03 |
Pos logit contributions | |
oles | +0.048 |
arry | +0.048 |
Today | +0.047 |
elly | +0.047 |
arl | +0.046 |
Neg logit contributions | |
never | -0.035 |
made | -0.033 |
learned | -0.032 |
cleaned | -0.032 |
put | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
oles | +0.009 |
arry | +0.009 |
Today | +0.009 |
elly | +0.009 |
arl | +0.008 |
Neg logit contributions | |
never | -0.005 |
made | -0.005 |
learned | -0.005 |
cleaned | -0.005 |
put | -0.005 |
Token | 't |
Token ID | 470 |
Feature activation | +8.44e-03 |
Pos logit contributions | |
never | +0.029 |
made | +0.029 |
cleaned | +0.029 |
learned | +0.029 |
punished | +0.028 |
Neg logit contributions | |
oles | -0.024 |
arry | -0.023 |
Today | -0.023 |
elly | -0.022 |
ummy | -0.022 |
Token | I |
Token ID | 314 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
never | +0.016 |
made | +0.016 |
learned | +0.015 |
cleaned | +0.015 |
punished | +0.015 |
Neg logit contributions | |
oles | -0.017 |
arry | -0.017 |
Today | -0.017 |
elly | -0.016 |
arl | -0.016 |
Token | play |
Token ID | 711 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
never | +0.044 |
made | +0.042 |
learned | +0.041 |
cleaned | +0.040 |
punished | +0.040 |
Neg logit contributions | |
arry | -0.109 |
Today | -0.108 |
oles | -0.108 |
elly | -0.106 |
arl | -0.106 |
Token | with |
Token ID | 351 |
Feature activation | -6.52e-03 |
Pos logit contributions | |
arry | +0.057 |
Today | +0.057 |
oles | +0.057 |
elly | +0.056 |
ute | +0.055 |
Neg logit contributions | |
never | -0.049 |
made | -0.046 |
learned | -0.046 |
cleaned | -0.045 |
put | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
oles | +0.016 |
elly | +0.015 |
ummy | +0.015 |
arry | +0.015 |
Today | +0.015 |
Neg logit contributions | |
never | -0.009 |
learned | -0.009 |
cleaned | -0.009 |
made | -0.009 |
punished | -0.009 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
oles | +0.045 |
arry | +0.045 |
Today | +0.044 |
elly | +0.044 |
ute | +0.043 |
Neg logit contributions | |
never | -0.046 |
made | -0.044 |
cleaned | -0.044 |
learned | -0.044 |
punished | -0.043 |
Token | opus |
Token ID | 25790 |
Feature activation | -7.57e-02 |
Pos logit contributions | |
Today | +0.225 |
arl | +0.197 |
andel | +0.174 |
arry | +0.162 |
owers | +0.155 |
Neg logit contributions | |
made | -0.363 |
never | -0.357 |
found | -0.346 |
buried | -0.345 |
learned | -0.343 |
Token | , |
Token ID | 11 |
Feature activation | +2.53e-03 |
Pos logit contributions | |
arry | +0.142 |
oles | +0.142 |
Today | +0.140 |
elly | +0.137 |
arl | +0.136 |
Neg logit contributions | |
never | -0.156 |
made | -0.152 |
put | -0.148 |
learned | -0.148 |
cleaned | -0.147 |
Token | too |
Token ID | 1165 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
never | +0.004 |
made | +0.004 |
put | +0.004 |
learned | +0.004 |
cleaned | +0.004 |
Neg logit contributions | |
arry | -0.006 |
oles | -0.006 |
Today | -0.006 |
elly | -0.006 |
arl | -0.006 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
owers | +0.032 |
oles | +0.032 |
arry | +0.031 |
andel | +0.030 |
� | +0.030 |
Neg logit contributions | |
made | -0.023 |
never | -0.023 |
put | -0.022 |
fixed | -0.021 |
used | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -6.36e-02 |
Pos logit contributions | |
arl | +0.091 |
elly | +0.091 |
oles | +0.089 |
arry | +0.089 |
ute | +0.087 |
Neg logit contributions | |
never | -0.104 |
learned | -0.100 |
made | -0.099 |
put | -0.098 |
tried | -0.097 |
Token | wanted |
Token ID | 2227 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
his | +0.078 |
himself | +0.078 |
ts | +0.072 |
more | +0.070 |
tightened | +0.068 |
Neg logit contributions | |
oles | -0.103 |
owers | -0.103 |
arry | -0.102 |
elly | -0.101 |
ay | -0.100 |
Token | to |
Token ID | 284 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
learned | +0.038 |
pretended | +0.038 |
never | +0.038 |
punished | +0.038 |
remembered | +0.037 |
Neg logit contributions | |
ummy | -0.035 |
oles | -0.035 |
elly | -0.034 |
Today | -0.034 |
ay | -0.032 |
Token | play |
Token ID | 711 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
learned | +0.048 |
punished | +0.047 |
made | +0.047 |
never | +0.045 |
saved | +0.045 |
Neg logit contributions | |
Today | -0.098 |
arry | -0.097 |
oles | -0.095 |
ute | -0.095 |
arl | -0.094 |
Token | with |
Token ID | 351 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
arry | +0.074 |
oles | +0.074 |
Today | +0.074 |
elly | +0.072 |
arl | +0.071 |
Neg logit contributions | |
never | -0.051 |
made | -0.049 |
learned | -0.047 |
put | -0.047 |
cleaned | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
oles | +0.033 |
elly | +0.031 |
ummy | +0.031 |
arry | +0.031 |
owers | +0.030 |
Neg logit contributions | |
never | -0.018 |
learned | -0.017 |
made | -0.017 |
punished | -0.017 |
pretended | -0.016 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
oles | +0.055 |
Today | +0.054 |
elly | +0.054 |
arry | +0.053 |
arl | +0.053 |
Neg logit contributions | |
never | -0.037 |
made | -0.035 |
learned | -0.034 |
put | -0.034 |
cleaned | -0.033 |
Token | opus |
Token ID | 25790 |
Feature activation | -7.03e-02 |
Pos logit contributions | |
Today | +0.209 |
andel | +0.172 |
arl | +0.165 |
owers | +0.155 |
arry | +0.149 |
Neg logit contributions | |
made | -0.368 |
never | -0.361 |
fixed | -0.357 |
learned | -0.350 |
punished | -0.349 |
Token | , |
Token ID | 11 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
oles | +0.145 |
arry | +0.144 |
Today | +0.143 |
elly | +0.141 |
arl | +0.138 |
Neg logit contributions | |
never | -0.131 |
made | -0.126 |
learned | -0.123 |
put | -0.123 |
cleaned | -0.123 |
Token | but |
Token ID | 475 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
never | +0.027 |
punished | +0.026 |
made | +0.026 |
learned | +0.026 |
cleaned | +0.026 |
Neg logit contributions | |
oles | -0.035 |
Today | -0.034 |
arry | -0.034 |
elly | -0.033 |
arl | -0.033 |
Token | we |
Token ID | 356 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
punished | +0.048 |
learned | +0.047 |
cleaned | +0.046 |
wed | +0.045 |
fixed | +0.045 |
Neg logit contributions | |
oles | -0.047 |
ummy | -0.044 |
Today | -0.044 |
arry | -0.041 |
ower | -0.040 |
Token | broke |
Token ID | 6265 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
himself | +0.061 |
his | +0.059 |
more | +0.056 |
permission | +0.055 |
ts | +0.051 |
Neg logit contributions | |
oles | -0.108 |
arry | -0.105 |
Today | -0.102 |
elly | -0.102 |
owers | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
arry | +0.133 |
oles | +0.132 |
Today | +0.132 |
elly | +0.130 |
arl | +0.128 |
Neg logit contributions | |
never | -0.105 |
made | -0.101 |
learned | -0.098 |
put | -0.098 |
cleaned | -0.098 |
Token | We |
Token ID | 775 |
Feature activation | +5.10e-02 |
Pos logit contributions | |
oles | +0.079 |
elly | +0.075 |
ower | +0.074 |
ute | +0.073 |
owers | +0.073 |
Neg logit contributions | |
never | -0.079 |
made | -0.073 |
put | -0.072 |
learned | -0.071 |
fixed | -0.070 |
Token | will |
Token ID | 481 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
punished | +0.095 |
his | +0.091 |
never | +0.090 |
cleaned | +0.088 |
pretended | +0.087 |
Neg logit contributions | |
Today | -0.089 |
ay | -0.087 |
arry | -0.085 |
ute | -0.084 |
owers | -0.082 |
Token | share |
Token ID | 2648 |
Feature activation | +3.46e-03 |
Pos logit contributions | |
punished | +0.054 |
cleaned | +0.052 |
fixed | +0.049 |
saved | +0.048 |
learned | +0.048 |
Neg logit contributions | |
Today | -0.129 |
arry | -0.123 |
arl | -0.123 |
ute | -0.121 |
� | -0.119 |
Token | the |
Token ID | 262 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
never | +0.004 |
punished | +0.004 |
cleaned | +0.004 |
made | +0.004 |
put | +0.004 |
Neg logit contributions | |
oles | -0.009 |
Today | -0.009 |
arry | -0.009 |
elly | -0.008 |
ute | -0.008 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
oles | +0.038 |
arry | +0.038 |
Today | +0.038 |
elly | +0.037 |
arl | +0.037 |
Neg logit contributions | |
never | -0.036 |
made | -0.035 |
put | -0.034 |
learned | -0.034 |
cleaned | -0.034 |
Token | opus |
Token ID | 25790 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
Today | +0.209 |
arl | +0.180 |
andel | +0.152 |
arry | +0.147 |
owers | +0.141 |
Neg logit contributions | |
never | -0.367 |
made | -0.362 |
learned | -0.354 |
fixed | -0.351 |
put | -0.348 |
Token | together |
Token ID | 1978 |
Feature activation | -4.78e-03 |
Pos logit contributions | |
oles | +0.141 |
arry | +0.140 |
Today | +0.139 |
elly | +0.138 |
ute | +0.135 |
Neg logit contributions | |
never | -0.110 |
made | -0.107 |
cleaned | -0.104 |
put | -0.104 |
learned | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
oles | +0.010 |
arry | +0.010 |
Today | +0.010 |
elly | +0.010 |
arl | +0.010 |
Neg logit contributions | |
never | -0.008 |
made | -0.008 |
cleaned | -0.008 |
learned | -0.008 |
punished | -0.008 |
Token | Let |
Token ID | 3914 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
oles | +0.073 |
elly | +0.070 |
ute | +0.068 |
arry | +0.068 |
ower | +0.068 |
Neg logit contributions | |
never | -0.080 |
made | -0.075 |
put | -0.074 |
learned | -0.073 |
fixed | -0.072 |
Token | 's |
Token ID | 338 |
Feature activation | +4.84e-02 |
Pos logit contributions | |
never | +0.021 |
made | +0.020 |
learned | +0.020 |
cleaned | +0.020 |
put | +0.020 |
Neg logit contributions | |
arry | -0.041 |
oles | -0.041 |
Today | -0.041 |
elly | -0.040 |
arl | -0.039 |
Token | started |
Token ID | 2067 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
remind | +0.029 |
care | +0.028 |
fixed | +0.028 |
saved | +0.027 |
make | +0.027 |
Neg logit contributions | |
oles | -0.074 |
arry | -0.071 |
op | -0.068 |
olly | -0.067 |
apping | -0.067 |
Token | to |
Token ID | 284 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
never | +0.063 |
fixed | +0.062 |
learned | +0.060 |
made | +0.060 |
cleaned | +0.060 |
Neg logit contributions | |
oles | -0.078 |
Today | -0.076 |
arry | -0.074 |
elly | -0.074 |
ogging | -0.072 |
Token | pick |
Token ID | 2298 |
Feature activation | +8.44e-03 |
Pos logit contributions | |
never | +0.088 |
learned | +0.086 |
fixed | +0.084 |
made | +0.083 |
saved | +0.082 |
Neg logit contributions | |
Today | -0.124 |
oles | -0.119 |
arry | -0.119 |
ogging | -0.111 |
ute | -0.111 |
Token | up |
Token ID | 510 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
never | +0.013 |
punished | +0.012 |
learned | +0.012 |
cleaned | +0.012 |
fixed | +0.012 |
Neg logit contributions | |
elly | -0.020 |
oles | -0.020 |
Today | -0.020 |
arry | -0.020 |
ute | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
oles | +0.032 |
arry | +0.032 |
Today | +0.032 |
elly | +0.031 |
owers | +0.031 |
Neg logit contributions | |
never | -0.012 |
punished | -0.010 |
learned | -0.010 |
made | -0.010 |
fixed | -0.010 |
Token | seaw |
Token ID | 32085 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
oles | +0.030 |
elly | +0.030 |
ute | +0.029 |
Today | +0.029 |
arry | +0.029 |
Neg logit contributions | |
never | -0.025 |
punished | -0.024 |
learned | -0.023 |
fixed | -0.022 |
cleaned | -0.022 |
Token | eed |
Token ID | 2308 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
Today | +0.219 |
oles | +0.194 |
arry | +0.190 |
owers | +0.185 |
ane | +0.179 |
Neg logit contributions | |
put | -0.358 |
made | -0.354 |
never | -0.348 |
found | -0.339 |
cleaned | -0.338 |
Token | and |
Token ID | 290 |
Feature activation | +5.76e-02 |
Pos logit contributions | |
arry | +0.157 |
oles | +0.157 |
Today | +0.156 |
elly | +0.153 |
arl | +0.151 |
Neg logit contributions | |
never | -0.114 |
made | -0.111 |
put | -0.107 |
learned | -0.107 |
cleaned | -0.106 |
Token | put |
Token ID | 1234 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
never | +0.058 |
punished | +0.058 |
learned | +0.053 |
fixed | +0.051 |
saved | +0.050 |
Neg logit contributions | |
oles | -0.138 |
Today | -0.132 |
arry | -0.124 |
ower | -0.124 |
ute | -0.124 |
Token | it |
Token ID | 340 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
punished | +0.049 |
never | +0.047 |
learned | +0.045 |
forgot | +0.044 |
pretended | +0.043 |
Neg logit contributions | |
ay | -0.074 |
oles | -0.073 |
Today | -0.072 |
ore | -0.072 |
owers | -0.072 |
Token | in |
Token ID | 287 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
learned | +0.006 |
punished | +0.005 |
never | +0.005 |
pretended | +0.005 |
made | +0.005 |
Neg logit contributions | |
oles | -0.006 |
arl | -0.006 |
Today | -0.006 |
ute | -0.006 |
ower | -0.006 |
Token | - |
Token ID | 12 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
Today | +0.137 |
arry | +0.132 |
arl | +0.130 |
� | +0.125 |
ower | +0.121 |
Neg logit contributions | |
made | -0.144 |
put | -0.141 |
never | -0.134 |
cleaned | -0.132 |
fixed | -0.132 |
Token | war |
Token ID | 5767 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
Today | +0.056 |
oles | +0.055 |
arry | +0.055 |
arl | +0.053 |
elly | +0.052 |
Neg logit contributions | |
never | -0.084 |
made | -0.082 |
learned | -0.079 |
put | -0.079 |
cleaned | -0.078 |
Token | ," |
Token ID | 553 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
arry | +0.233 |
arl | +0.233 |
Today | +0.230 |
elly | +0.227 |
oles | +0.227 |
Neg logit contributions | |
never | -0.169 |
made | -0.168 |
put | -0.162 |
fixed | -0.157 |
learned | -0.155 |
Token | said |
Token ID | 531 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
oles | +0.054 |
arry | +0.054 |
Today | +0.054 |
elly | +0.052 |
owers | +0.051 |
Neg logit contributions | |
never | -0.026 |
punished | -0.024 |
cleaned | -0.023 |
fixed | -0.023 |
learned | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
oles | +0.040 |
arry | +0.040 |
Today | +0.040 |
elly | +0.039 |
arl | +0.038 |
Neg logit contributions | |
never | -0.016 |
made | -0.015 |
learned | -0.014 |
cleaned | -0.014 |
put | -0.014 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
arry | +0.052 |
oles | +0.052 |
Today | +0.051 |
elly | +0.050 |
arl | +0.050 |
Neg logit contributions | |
never | -0.045 |
made | -0.044 |
put | -0.043 |
learned | -0.043 |
cleaned | -0.043 |
Token | opus |
Token ID | 25790 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
Today | +0.218 |
arl | +0.172 |
andel | +0.164 |
owers | +0.160 |
arry | +0.152 |
Neg logit contributions | |
never | -0.359 |
made | -0.347 |
put | -0.337 |
tried | -0.334 |
cut | -0.332 |
Token | . |
Token ID | 13 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
oles | +0.045 |
arry | +0.044 |
Today | +0.044 |
elly | +0.042 |
ute | +0.041 |
Neg logit contributions | |
never | -0.038 |
made | -0.037 |
learned | -0.036 |
cleaned | -0.036 |
punished | -0.036 |
Token | " |
Token ID | 366 |
Feature activation | +9.26e-03 |
Pos logit contributions | |
oles | +0.099 |
ute | +0.099 |
owers | +0.098 |
ower | +0.097 |
arry | +0.097 |
Neg logit contributions | |
put | -0.082 |
made | -0.082 |
never | -0.081 |
learned | -0.079 |
tried | -0.079 |
Token | I |
Token ID | 40 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
never | +0.016 |
made | +0.016 |
put | +0.015 |
learned | +0.015 |
cleaned | +0.015 |
Neg logit contributions | |
arry | -0.020 |
oles | -0.020 |
Today | -0.020 |
elly | -0.020 |
arl | -0.020 |
Token | have |
Token ID | 423 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
never | +0.055 |
punished | +0.052 |
learned | +0.052 |
cleaned | +0.051 |
made | +0.050 |
Neg logit contributions | |
Today | -0.086 |
arry | -0.085 |
oles | -0.082 |
elly | -0.081 |
ute | -0.080 |
Token | he |
Token ID | 339 |
Feature activation | +8.03e-02 |
Pos logit contributions | |
Today | +0.027 |
oles | +0.027 |
arry | +0.026 |
elly | +0.025 |
ummy | +0.025 |
Neg logit contributions | |
never | -0.016 |
punished | -0.016 |
learned | -0.016 |
cleaned | -0.015 |
fixed | -0.015 |
Token | decided |
Token ID | 3066 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
permission | +0.076 |
herself | +0.075 |
care | +0.073 |
never | +0.069 |
her | +0.069 |
Neg logit contributions | |
oles | -0.189 |
arry | -0.189 |
ute | -0.181 |
uddle | -0.178 |
elly | -0.175 |
Token | to |
Token ID | 284 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
pretended | +0.058 |
cleaned | +0.057 |
wed | +0.056 |
punished | +0.056 |
fixed | +0.055 |
Neg logit contributions | |
arry | -0.035 |
ogging | -0.035 |
oles | -0.034 |
Today | -0.032 |
ute | -0.032 |
Token | remove |
Token ID | 4781 |
Feature activation | +8.16e-03 |
Pos logit contributions | |
learned | +0.074 |
remembered | +0.070 |
punished | +0.068 |
made | +0.067 |
never | +0.067 |
Neg logit contributions | |
arry | -0.146 |
Today | -0.145 |
ute | -0.141 |
oles | -0.140 |
arl | -0.138 |
Token | the |
Token ID | 262 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
never | +0.010 |
punished | +0.010 |
learned | +0.009 |
remembered | +0.009 |
cleaned | +0.009 |
Neg logit contributions | |
Today | -0.020 |
oles | -0.020 |
elly | -0.020 |
ay | -0.020 |
ore | -0.020 |
Token | seaw |
Token ID | 32085 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
oles | +0.041 |
elly | +0.041 |
Today | +0.040 |
ute | +0.040 |
arry | +0.039 |
Neg logit contributions | |
never | -0.037 |
punished | -0.036 |
cleaned | -0.034 |
fixed | -0.033 |
learned | -0.033 |
Token | eed |
Token ID | 2308 |
Feature activation | -6.50e-02 |
Pos logit contributions | |
Today | +0.189 |
oles | +0.178 |
owers | +0.173 |
arry | +0.173 |
elly | +0.166 |
Neg logit contributions | |
put | -0.370 |
made | -0.369 |
never | -0.365 |
learned | -0.354 |
cleaned | -0.353 |
Token | from |
Token ID | 422 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
oles | +0.137 |
arry | +0.135 |
Today | +0.133 |
elly | +0.133 |
ute | +0.131 |
Neg logit contributions | |
never | -0.117 |
learned | -0.113 |
made | -0.112 |
cleaned | -0.111 |
punished | -0.111 |
Token | the |
Token ID | 262 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
learned | +0.019 |
punished | +0.019 |
cleaned | +0.018 |
forgot | +0.018 |
never | +0.017 |
Neg logit contributions | |
Today | -0.039 |
ore | -0.038 |
ute | -0.037 |
� | -0.037 |
ummy | -0.036 |
Token | water |
Token ID | 1660 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
oles | +0.037 |
Today | +0.036 |
elly | +0.036 |
arl | +0.036 |
ute | +0.036 |
Neg logit contributions | |
never | -0.033 |
fixed | -0.032 |
punished | -0.032 |
cleaned | -0.032 |
saved | -0.031 |
Token | so |
Token ID | 523 |
Feature activation | -4.67e-03 |
Pos logit contributions | |
oles | +0.103 |
arry | +0.102 |
Today | +0.102 |
elly | +0.101 |
arl | +0.099 |
Neg logit contributions | |
never | -0.081 |
made | -0.078 |
learned | -0.077 |
cleaned | -0.077 |
punished | -0.077 |
Token | car |
Token ID | 1097 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
oles | +0.057 |
ute | +0.056 |
elly | +0.056 |
Today | +0.055 |
arry | +0.055 |
Neg logit contributions | |
never | -0.049 |
punished | -0.048 |
cleaned | -0.047 |
learned | -0.046 |
fixed | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
arry | +0.092 |
Today | +0.092 |
oles | +0.091 |
elly | +0.091 |
arl | +0.088 |
Neg logit contributions | |
never | -0.071 |
punished | -0.068 |
made | -0.067 |
learned | -0.067 |
cleaned | -0.067 |
Token | give |
Token ID | 1577 |
Feature activation | +1.52e-04 |
Pos logit contributions | |
never | +0.051 |
remembered | +0.050 |
punished | +0.049 |
learned | +0.049 |
wished | +0.049 |
Neg logit contributions | |
arl | -0.099 |
arry | -0.095 |
Today | -0.094 |
ay | -0.094 |
� | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
never | +0.000 |
pretended | +0.000 |
punished | +0.000 |
cleaned | +0.000 |
remembered | +0.000 |
Neg logit contributions | |
oles | -0.000 |
ore | -0.000 |
elly | -0.000 |
arry | -0.000 |
ummy | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
punished | +0.040 |
cleaned | +0.039 |
learned | +0.039 |
pretended | +0.038 |
never | +0.038 |
Neg logit contributions | |
ummy | -0.075 |
Today | -0.073 |
oles | -0.073 |
arry | -0.072 |
arl | -0.072 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
oles | +0.041 |
arry | +0.040 |
elly | +0.040 |
Today | +0.039 |
ute | +0.038 |
Neg logit contributions | |
never | -0.045 |
cleaned | -0.043 |
made | -0.043 |
fixed | -0.043 |
punished | -0.043 |
Token | opus |
Token ID | 25790 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
Today | +0.210 |
arl | +0.176 |
andel | +0.162 |
owers | +0.145 |
arry | +0.140 |
Neg logit contributions | |
never | -0.361 |
made | -0.349 |
learned | -0.339 |
found | -0.339 |
buried | -0.335 |
Token | . |
Token ID | 13 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
oles | +0.085 |
Today | +0.084 |
arry | +0.084 |
ute | +0.082 |
elly | +0.081 |
Neg logit contributions | |
never | -0.084 |
punished | -0.081 |
made | -0.080 |
cleaned | -0.079 |
learned | -0.077 |
Token |
|
Token ID | 198 |
Feature activation | -5.63e-02 |
Pos logit contributions | |
ute | +0.102 |
andel | +0.097 |
arl | +0.096 |
elly | +0.095 |
uttering | +0.095 |
Neg logit contributions | |
never | -0.051 |
put | -0.050 |
made | -0.050 |
fixed | -0.049 |
learned | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | -5.78e-02 |
Pos logit contributions | |
owers | +0.120 |
elly | +0.119 |
andel | +0.119 |
umm | +0.118 |
arl | +0.118 |
Neg logit contributions | |
never | -0.107 |
put | -0.099 |
learned | -0.099 |
made | -0.099 |
found | -0.095 |
Token | Tim |
Token ID | 14967 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
arry | +0.122 |
oles | +0.121 |
Today | +0.121 |
elly | +0.119 |
arl | +0.117 |
Neg logit contributions | |
never | -0.108 |
made | -0.104 |
put | -0.101 |
learned | -0.101 |
cleaned | -0.100 |
Token | Can |
Token ID | 6090 |
Feature activation | +9.24e-03 |
Pos logit contributions | |
never | +0.011 |
made | +0.010 |
learned | +0.010 |
cleaned | +0.010 |
punished | +0.010 |
Neg logit contributions | |
Today | -0.012 |
arry | -0.012 |
oles | -0.012 |
elly | -0.012 |
arl | -0.011 |
Token | we |
Token ID | 356 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
never | +0.018 |
made | +0.018 |
put | +0.017 |
learned | +0.017 |
cleaned | +0.017 |
Neg logit contributions | |
arry | -0.018 |
oles | -0.018 |
Today | -0.018 |
elly | -0.018 |
arl | -0.018 |
Token | go |
Token ID | 467 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
never | +0.054 |
learned | +0.052 |
made | +0.051 |
cleaned | +0.051 |
punished | +0.050 |
Neg logit contributions | |
arry | -0.097 |
Today | -0.097 |
oles | -0.097 |
arl | -0.094 |
ute | -0.093 |
Token | see |
Token ID | 766 |
Feature activation | +8.10e-03 |
Pos logit contributions | |
Today | +0.036 |
arry | +0.036 |
oles | +0.035 |
ute | +0.035 |
arl | +0.035 |
Neg logit contributions | |
never | -0.023 |
learned | -0.022 |
punished | -0.022 |
cleaned | -0.022 |
made | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
never | +0.009 |
made | +0.008 |
learned | +0.008 |
cleaned | +0.008 |
put | +0.008 |
Neg logit contributions | |
oles | -0.023 |
arry | -0.023 |
Today | -0.023 |
elly | -0.023 |
arl | -0.022 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
arry | +0.057 |
oles | +0.056 |
Today | +0.056 |
arl | +0.055 |
elly | +0.054 |
Neg logit contributions | |
never | -0.049 |
made | -0.048 |
put | -0.046 |
learned | -0.045 |
cleaned | -0.045 |
Token | opus |
Token ID | 25790 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
Today | +0.204 |
arl | +0.166 |
andel | +0.148 |
arry | +0.147 |
oles | +0.137 |
Neg logit contributions | |
made | -0.377 |
never | -0.367 |
learned | -0.354 |
found | -0.351 |
put | -0.351 |
Token | ?" |
Token ID | 1701 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
arry | +0.106 |
oles | +0.105 |
Today | +0.104 |
elly | +0.103 |
arl | +0.101 |
Neg logit contributions | |
never | -0.106 |
made | -0.103 |
learned | -0.100 |
put | -0.100 |
cleaned | -0.100 |
Token | His |
Token ID | 2399 |
Feature activation | -3.30e-04 |
Pos logit contributions | |
oles | +0.109 |
arry | +0.108 |
elly | +0.107 |
arl | +0.106 |
Today | +0.103 |
Neg logit contributions | |
never | -0.102 |
made | -0.101 |
put | -0.100 |
cleaned | -0.097 |
learned | -0.097 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
arry | +0.001 |
oles | +0.001 |
elly | +0.001 |
Today | +0.001 |
owers | +0.001 |
Neg logit contributions | |
never | -0.000 |
made | -0.000 |
put | -0.000 |
learned | -0.000 |
cleaned | -0.000 |
Token | answered |
Token ID | 9373 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
never | +0.041 |
punished | +0.039 |
learned | +0.037 |
saved | +0.037 |
buried | +0.036 |
Neg logit contributions | |
arry | -0.070 |
oles | -0.067 |
Today | -0.067 |
ummy | -0.064 |
elly | -0.064 |
Token | you |
Token ID | 345 |
Feature activation | -2.97e-04 |
Pos logit contributions | |
never | +0.009 |
made | +0.009 |
learned | +0.009 |
cleaned | +0.009 |
punished | +0.009 |
Neg logit contributions | |
oles | -0.013 |
arry | -0.012 |
Today | -0.012 |
elly | -0.012 |
ummy | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
Today | +0.001 |
arry | +0.001 |
oles | +0.001 |
arl | +0.001 |
elly | +0.001 |
Neg logit contributions | |
never | -0.001 |
punished | -0.001 |
made | -0.001 |
cleaned | -0.001 |
put | -0.001 |
Token | we |
Token ID | 356 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
punished | +0.065 |
fixed | +0.065 |
cleaned | +0.064 |
never | +0.063 |
saved | +0.063 |
Neg logit contributions | |
Today | -0.067 |
oles | -0.065 |
ummy | -0.064 |
arry | -0.064 |
arl | -0.063 |
Token | love |
Token ID | 1842 |
Feature activation | +4.82e-03 |
Pos logit contributions | |
ts | +0.056 |
never | +0.056 |
his | +0.055 |
buried | +0.055 |
himself | +0.054 |
Neg logit contributions | |
arry | -0.114 |
arl | -0.113 |
Today | -0.112 |
ute | -0.108 |
owers | -0.106 |
Token | the |
Token ID | 262 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
never | +0.007 |
made | +0.007 |
learned | +0.006 |
cleaned | +0.006 |
punished | +0.006 |
Neg logit contributions | |
oles | -0.012 |
Today | -0.012 |
arry | -0.012 |
elly | -0.012 |
ummy | -0.011 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
oles | +0.053 |
arry | +0.053 |
Today | +0.052 |
elly | +0.051 |
ute | +0.050 |
Neg logit contributions | |
never | -0.055 |
cleaned | -0.053 |
punished | -0.053 |
made | -0.053 |
learned | -0.052 |
Token | opus |
Token ID | 25790 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
Today | +0.208 |
arl | +0.168 |
andel | +0.153 |
arry | +0.142 |
elly | +0.132 |
Neg logit contributions | |
never | -0.355 |
made | -0.348 |
fixed | -0.332 |
learned | -0.332 |
found | -0.331 |
Token | ." |
Token ID | 526 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
oles | +0.126 |
elly | +0.125 |
Today | +0.124 |
arry | +0.122 |
ute | +0.118 |
Neg logit contributions | |
never | -0.111 |
made | -0.106 |
punished | -0.106 |
cleaned | -0.106 |
put | -0.105 |
Token |
|
Token ID | 198 |
Feature activation | -6.69e-02 |
Pos logit contributions | |
andel | +0.108 |
ute | +0.105 |
arl | +0.103 |
arry | +0.103 |
elly | +0.102 |
Neg logit contributions | |
learned | -0.080 |
never | -0.080 |
put | -0.077 |
made | -0.076 |
his | -0.075 |
Token |
|
Token ID | 198 |
Feature activation | -5.98e-02 |
Pos logit contributions | |
elly | +0.144 |
owers | +0.142 |
arry | +0.140 |
andel | +0.140 |
umm | +0.138 |
Neg logit contributions | |
never | -0.127 |
learned | -0.122 |
put | -0.122 |
fixed | -0.115 |
made | -0.114 |
Token | Mom |
Token ID | 29252 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
Today | +0.134 |
arry | +0.133 |
elly | +0.132 |
oles | +0.131 |
owers | +0.129 |
Neg logit contributions | |
never | -0.107 |
made | -0.100 |
put | -0.100 |
learned | -0.100 |
cleaned | -0.098 |
Token | tried |
Token ID | 3088 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
punished | +0.051 |
never | +0.050 |
permission | +0.043 |
learned | +0.043 |
accident | +0.043 |
Neg logit contributions | |
oles | -0.110 |
op | -0.108 |
ore | -0.107 |
aris | -0.106 |
arry | -0.106 |
Token | to |
Token ID | 284 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
punished | +0.034 |
never | +0.034 |
cleaned | +0.033 |
made | +0.033 |
learned | +0.033 |
Neg logit contributions | |
oles | -0.043 |
Today | -0.042 |
arry | -0.041 |
ute | -0.040 |
elly | -0.040 |
Token | un |
Token ID | 555 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
learned | +0.084 |
never | +0.078 |
wished | +0.077 |
saved | +0.077 |
punished | +0.076 |
Neg logit contributions | |
Today | -0.095 |
arry | -0.095 |
ute | -0.094 |
arl | -0.094 |
oles | -0.091 |
Token | pack |
Token ID | 8002 |
Feature activation | +7.26e-04 |
Pos logit contributions | |
arry | +0.085 |
Today | +0.082 |
ute | +0.081 |
oles | +0.081 |
ay | +0.080 |
Neg logit contributions | |
never | -0.069 |
learned | -0.067 |
made | -0.067 |
cleaned | -0.064 |
punished | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
punished | +0.001 |
never | +0.001 |
learned | +0.001 |
saved | +0.001 |
cleaned | +0.001 |
Neg logit contributions | |
Today | -0.002 |
ay | -0.002 |
oles | -0.002 |
arry | -0.002 |
ummy | -0.002 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
Today | +0.042 |
elly | +0.041 |
arry | +0.040 |
oles | +0.040 |
ute | +0.040 |
Neg logit contributions | |
never | -0.034 |
learned | -0.031 |
punished | -0.031 |
cleaned | -0.031 |
made | -0.030 |
Token | opus |
Token ID | 25790 |
Feature activation | -5.66e-02 |
Pos logit contributions | |
Today | +0.200 |
arl | +0.169 |
andel | +0.146 |
arry | +0.136 |
oles | +0.133 |
Neg logit contributions | |
never | -0.359 |
made | -0.358 |
fixed | -0.348 |
found | -0.346 |
learned | -0.343 |
Token | . |
Token ID | 13 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
oles | +0.118 |
Today | +0.117 |
arry | +0.116 |
elly | +0.113 |
ute | +0.112 |
Neg logit contributions | |
never | -0.104 |
learned | -0.098 |
punished | -0.097 |
made | -0.096 |
cleaned | -0.095 |
Token |
|
Token ID | 198 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
arl | +0.086 |
elly | +0.083 |
ute | +0.081 |
arry | +0.081 |
andel | +0.079 |
Neg logit contributions | |
never | -0.077 |
made | -0.075 |
put | -0.075 |
tried | -0.073 |
found | -0.072 |
Token |
|
Token ID | 198 |
Feature activation | -5.27e-02 |
Pos logit contributions | |
owers | +0.124 |
elly | +0.121 |
umm | +0.121 |
andel | +0.119 |
arl | +0.118 |
Neg logit contributions | |
never | -0.108 |
put | -0.098 |
made | -0.096 |
learned | -0.095 |
found | -0.094 |
Token | " |
Token ID | 1 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
arry | +0.104 |
Today | +0.104 |
elly | +0.103 |
arl | +0.102 |
oles | +0.102 |
Neg logit contributions | |
never | -0.107 |
made | -0.103 |
put | -0.100 |
learned | -0.099 |
cleaned | -0.097 |
Token | He |
Token ID | 679 |
Feature activation | +7.67e-02 |
Pos logit contributions | |
owers | +0.104 |
ute | +0.101 |
oles | +0.097 |
ower | +0.097 |
andel | +0.096 |
Neg logit contributions | |
never | -0.064 |
made | -0.063 |
put | -0.063 |
his | -0.062 |
tried | -0.059 |
Token | wanted |
Token ID | 2227 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
punished | +0.089 |
permission | +0.086 |
never | +0.083 |
care | +0.077 |
cleaned | +0.072 |
Neg logit contributions | |
arry | -0.180 |
oles | -0.180 |
elly | -0.173 |
apping | -0.173 |
ute | -0.172 |
Token | to |
Token ID | 284 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
punished | +0.046 |
never | +0.044 |
learned | +0.044 |
cleaned | +0.043 |
pretended | +0.043 |
Neg logit contributions | |
oles | -0.039 |
Today | -0.038 |
elly | -0.038 |
ummy | -0.036 |
ore | -0.036 |
Token | see |
Token ID | 766 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
punished | +0.054 |
cleaned | +0.051 |
learned | +0.051 |
never | +0.050 |
saved | +0.049 |
Neg logit contributions | |
Today | -0.098 |
arry | -0.094 |
oles | -0.093 |
ute | -0.093 |
ay | -0.088 |
Token | the |
Token ID | 262 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
never | +0.019 |
cleaned | +0.018 |
learned | +0.018 |
punished | +0.018 |
made | +0.017 |
Neg logit contributions | |
oles | -0.044 |
Today | -0.043 |
arry | -0.043 |
elly | -0.043 |
ute | -0.042 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
oles | +0.053 |
arry | +0.053 |
Today | +0.052 |
elly | +0.051 |
arl | +0.050 |
Neg logit contributions | |
never | -0.048 |
made | -0.047 |
cleaned | -0.046 |
learned | -0.046 |
put | -0.045 |
Token | opus |
Token ID | 25790 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
Today | +0.221 |
arl | +0.204 |
andel | +0.166 |
arry | +0.157 |
elly | +0.152 |
Neg logit contributions | |
never | -0.335 |
made | -0.333 |
found | -0.312 |
put | -0.311 |
learned | -0.310 |
Token | better |
Token ID | 1365 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
oles | +0.092 |
Today | +0.091 |
arry | +0.089 |
elly | +0.087 |
ute | +0.086 |
Neg logit contributions | |
never | -0.071 |
cleaned | -0.069 |
punished | -0.069 |
made | -0.068 |
learned | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
oles | +0.052 |
arry | +0.052 |
Today | +0.051 |
elly | +0.050 |
arl | +0.050 |
Neg logit contributions | |
never | -0.040 |
made | -0.038 |
learned | -0.037 |
put | -0.037 |
cleaned | -0.037 |
Token | He |
Token ID | 679 |
Feature activation | +7.48e-02 |
Pos logit contributions | |
owers | +0.084 |
oles | +0.082 |
ute | +0.080 |
andel | +0.080 |
ower | +0.079 |
Neg logit contributions | |
never | -0.060 |
made | -0.059 |
put | -0.059 |
his | -0.056 |
fixed | -0.055 |
Token | ran |
Token ID | 4966 |
Feature activation | -2.01e-04 |
Pos logit contributions | |
punished | +0.082 |
permission | +0.081 |
care | +0.077 |
never | +0.076 |
fixed | +0.070 |
Neg logit contributions | |
arry | -0.174 |
oles | -0.172 |
elly | -0.169 |
olly | -0.166 |
uddle | -0.160 |
Token | , |
Token ID | 11 |
Feature activation | -6.64e-03 |
Pos logit contributions | |
arl | +0.106 |
� | +0.102 |
owers | +0.101 |
ower | +0.101 |
Today | +0.100 |
Neg logit contributions | |
put | -0.070 |
made | -0.070 |
found | -0.064 |
pretended | -0.060 |
fixed | -0.060 |
Token | duck |
Token ID | 22045 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
arry | +0.016 |
oles | +0.016 |
owers | +0.015 |
Today | +0.015 |
elly | +0.015 |
Neg logit contributions | |
made | -0.011 |
put | -0.011 |
never | -0.010 |
fixed | -0.010 |
learned | -0.010 |
Token | ," |
Token ID | 553 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
Today | +0.098 |
arry | +0.097 |
oles | +0.097 |
arl | +0.095 |
elly | +0.095 |
Neg logit contributions | |
never | -0.084 |
made | -0.082 |
put | -0.080 |
learned | -0.078 |
fixed | -0.077 |
Token | said |
Token ID | 531 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
Today | +0.051 |
oles | +0.050 |
arry | +0.049 |
elly | +0.048 |
owers | +0.046 |
Neg logit contributions | |
never | -0.042 |
punished | -0.040 |
fixed | -0.040 |
cleaned | -0.039 |
made | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
arry | +0.043 |
oles | +0.043 |
Today | +0.042 |
elly | +0.042 |
arl | +0.042 |
Neg logit contributions | |
never | -0.026 |
made | -0.026 |
put | -0.025 |
learned | -0.024 |
cleaned | -0.024 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
arry | +0.065 |
oles | +0.064 |
Today | +0.063 |
elly | +0.062 |
arl | +0.062 |
Neg logit contributions | |
made | -0.058 |
never | -0.058 |
put | -0.057 |
learned | -0.055 |
cleaned | -0.054 |
Token | opus |
Token ID | 25790 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
Today | +0.219 |
arl | +0.182 |
andel | +0.168 |
owers | +0.155 |
arry | +0.154 |
Neg logit contributions | |
never | -0.351 |
made | -0.343 |
put | -0.330 |
learned | -0.326 |
found | -0.325 |
Token | . |
Token ID | 13 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
oles | +0.048 |
arry | +0.048 |
Today | +0.048 |
elly | +0.047 |
ute | +0.046 |
Neg logit contributions | |
never | -0.038 |
made | -0.037 |
cleaned | -0.036 |
learned | -0.036 |
punished | -0.036 |
Token | " |
Token ID | 366 |
Feature activation | +2.43e-03 |
Pos logit contributions | |
ute | +0.103 |
oles | +0.103 |
owers | +0.103 |
arry | +0.102 |
ower | +0.100 |
Neg logit contributions | |
made | -0.082 |
put | -0.082 |
never | -0.080 |
tried | -0.078 |
learned | -0.077 |
Token | Do |
Token ID | 5211 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
never | +0.004 |
made | +0.004 |
learned | +0.004 |
put | +0.004 |
cleaned | +0.004 |
Neg logit contributions | |
oles | -0.006 |
arry | -0.006 |
Today | -0.006 |
elly | -0.006 |
arl | -0.006 |
Token | you |
Token ID | 345 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
never | +0.044 |
made | +0.043 |
put | +0.042 |
learned | +0.041 |
cleaned | +0.041 |
Neg logit contributions | |
oles | -0.071 |
arry | -0.070 |
Today | -0.070 |
elly | -0.069 |
arl | -0.068 |
Token | he |
Token ID | 339 |
Feature activation | +7.90e-02 |
Pos logit contributions | |
cleaned | +0.047 |
punished | +0.046 |
fixed | +0.045 |
never | +0.045 |
learned | +0.044 |
Neg logit contributions | |
oles | -0.099 |
Today | -0.098 |
arry | -0.094 |
elly | -0.094 |
ummy | -0.092 |
Token | had |
Token ID | 550 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
punished | +0.115 |
permission | +0.113 |
cleaned | +0.109 |
ts | +0.109 |
glued | +0.107 |
Neg logit contributions | |
arry | -0.145 |
ane | -0.143 |
oles | -0.142 |
elly | -0.139 |
Today | -0.137 |
Token | to |
Token ID | 284 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
pretended | +0.075 |
punished | +0.071 |
ts | +0.070 |
cleaned | +0.065 |
fixed | +0.065 |
Neg logit contributions | |
aris | -0.128 |
Today | -0.126 |
elly | -0.125 |
oles | -0.125 |
apping | -0.125 |
Token | leave |
Token ID | 2666 |
Feature activation | +2.53e-03 |
Pos logit contributions | |
pretended | +0.069 |
cleaned | +0.067 |
fixed | +0.067 |
made | +0.066 |
learned | +0.064 |
Neg logit contributions | |
Today | -0.149 |
arry | -0.149 |
arl | -0.144 |
ute | -0.142 |
oles | -0.141 |
Token | the |
Token ID | 262 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
cleaned | +0.003 |
punished | +0.003 |
pretended | +0.003 |
never | +0.003 |
made | +0.003 |
Neg logit contributions | |
oles | -0.006 |
arry | -0.006 |
Today | -0.006 |
elly | -0.006 |
ummy | -0.006 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
oles | +0.079 |
elly | +0.079 |
arry | +0.079 |
Today | +0.078 |
ute | +0.076 |
Neg logit contributions | |
never | -0.072 |
made | -0.068 |
cleaned | -0.068 |
fixed | -0.067 |
learned | -0.067 |
Token | opus |
Token ID | 25790 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
Today | +0.202 |
arl | +0.152 |
andel | +0.143 |
owers | +0.133 |
arry | +0.133 |
Neg logit contributions | |
never | -0.377 |
made | -0.363 |
punished | -0.362 |
learned | -0.361 |
fixed | -0.356 |
Token | alone |
Token ID | 3436 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
oles | +0.113 |
Today | +0.112 |
arry | +0.111 |
elly | +0.110 |
ute | +0.110 |
Neg logit contributions | |
never | -0.075 |
made | -0.074 |
cleaned | -0.073 |
learned | -0.073 |
punished | -0.072 |
Token | ! |
Token ID | 0 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
oles | +0.054 |
arry | +0.053 |
Today | +0.053 |
elly | +0.052 |
arl | +0.051 |
Neg logit contributions | |
never | -0.044 |
made | -0.042 |
put | -0.041 |
cleaned | -0.041 |
learned | -0.041 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -6.99e-02 |
Pos logit contributions | |
ute | +0.079 |
oles | +0.078 |
owers | +0.077 |
ower | +0.076 |
arl | +0.075 |
Neg logit contributions | |
never | -0.080 |
learned | -0.076 |
made | -0.074 |
fixed | -0.072 |
saved | -0.071 |
Token | Once |
Token ID | 7454 |
Feature activation | -5.42e-02 |
Pos logit contributions | |
ower | +0.164 |
andel | +0.164 |
arl | +0.162 |
oles | +0.161 |
ute | +0.161 |
Neg logit contributions | |
made | -0.126 |
never | -0.123 |
found | -0.121 |
learned | -0.115 |
put | -0.115 |
Token | ily |
Token ID | 813 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
arry | +0.130 |
oles | +0.130 |
Today | +0.129 |
elly | +0.127 |
arl | +0.126 |
Neg logit contributions | |
never | -0.083 |
made | -0.080 |
put | -0.077 |
learned | -0.077 |
cleaned | -0.076 |
Token | wanted |
Token ID | 2227 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
punished | +0.044 |
never | +0.041 |
saved | +0.039 |
permission | +0.038 |
fixed | +0.038 |
Neg logit contributions | |
arry | -0.100 |
oles | -0.097 |
Today | -0.096 |
elly | -0.093 |
ute | -0.092 |
Token | to |
Token ID | 284 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
punished | +0.047 |
pretended | +0.044 |
fixed | +0.044 |
never | +0.044 |
cleaned | +0.044 |
Neg logit contributions | |
oles | -0.038 |
elly | -0.037 |
ore | -0.035 |
Today | -0.035 |
ummy | -0.034 |
Token | touch |
Token ID | 3638 |
Feature activation | -9.85e-03 |
Pos logit contributions | |
punished | +0.063 |
never | +0.062 |
cleaned | +0.061 |
learned | +0.060 |
made | +0.059 |
Neg logit contributions | |
Today | -0.102 |
arry | -0.101 |
oles | -0.100 |
ute | -0.097 |
ay | -0.093 |
Token | the |
Token ID | 262 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
oles | +0.030 |
Today | +0.030 |
arry | +0.029 |
elly | +0.029 |
ute | +0.029 |
Neg logit contributions | |
never | -0.009 |
cleaned | -0.008 |
punished | -0.008 |
made | -0.008 |
learned | -0.008 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
arry | +0.047 |
oles | +0.047 |
Today | +0.046 |
elly | +0.046 |
arl | +0.045 |
Neg logit contributions | |
never | -0.041 |
made | -0.041 |
put | -0.039 |
learned | -0.039 |
cleaned | -0.039 |
Token | opus |
Token ID | 25790 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
Today | +0.201 |
arl | +0.182 |
andel | +0.155 |
arry | +0.145 |
owers | +0.136 |
Neg logit contributions | |
made | -0.356 |
never | -0.349 |
found | -0.335 |
put | -0.333 |
buried | -0.333 |
Token | , |
Token ID | 11 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
oles | +0.091 |
arry | +0.090 |
Today | +0.090 |
elly | +0.088 |
ute | +0.086 |
Neg logit contributions | |
never | -0.092 |
made | -0.089 |
learned | -0.088 |
cleaned | -0.087 |
put | -0.087 |
Token | but |
Token ID | 475 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
never | +0.038 |
fixed | +0.037 |
punished | +0.037 |
cleaned | +0.037 |
made | +0.035 |
Neg logit contributions | |
Today | -0.047 |
oles | -0.046 |
arry | -0.044 |
elly | -0.043 |
ummy | -0.042 |
Token | her |
Token ID | 607 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
punished | +0.027 |
saved | +0.025 |
fixed | +0.024 |
repaired | +0.024 |
cleaned | +0.024 |
Neg logit contributions | |
Today | -0.033 |
oles | -0.032 |
ama | -0.030 |
ummy | -0.030 |
uffy | -0.029 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
never | +0.018 |
punished | +0.017 |
learned | +0.017 |
made | +0.017 |
cleaned | +0.017 |
Neg logit contributions | |
Today | -0.022 |
oles | -0.021 |
ummy | -0.021 |
arry | -0.021 |
ute | -0.021 |
Token | they |
Token ID | 484 |
Feature activation | +6.61e-02 |
Pos logit contributions | |
never | +0.007 |
punished | +0.007 |
cleaned | +0.007 |
fixed | +0.007 |
learned | +0.006 |
Neg logit contributions | |
oles | -0.008 |
Today | -0.008 |
arry | -0.008 |
elly | -0.008 |
ute | -0.008 |
Token | say |
Token ID | 910 |
Feature activation | +9.35e-03 |
Pos logit contributions | |
never | +0.094 |
punished | +0.092 |
cleaned | +0.089 |
fixed | +0.089 |
made | +0.087 |
Neg logit contributions | |
Today | -0.152 |
arl | -0.147 |
arry | -0.146 |
oles | -0.143 |
ute | -0.142 |
Token | goodbye |
Token ID | 24829 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
never | +0.018 |
made | +0.018 |
learned | +0.017 |
put | +0.017 |
fixed | +0.016 |
Neg logit contributions | |
arry | -0.020 |
elly | -0.020 |
oles | -0.019 |
Today | -0.019 |
owers | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
oles | +0.098 |
arry | +0.097 |
Today | +0.095 |
elly | +0.095 |
arl | +0.094 |
Neg logit contributions | |
never | -0.059 |
made | -0.056 |
put | -0.054 |
learned | -0.054 |
cleaned | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
punished | +0.025 |
never | +0.024 |
cleaned | +0.024 |
fixed | +0.023 |
made | +0.023 |
Neg logit contributions | |
Today | -0.064 |
oles | -0.063 |
arry | -0.062 |
ute | -0.062 |
arl | -0.061 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
oles | +0.037 |
arry | +0.037 |
Today | +0.036 |
elly | +0.036 |
ute | +0.035 |
Neg logit contributions | |
never | -0.041 |
punished | -0.040 |
cleaned | -0.040 |
made | -0.039 |
fixed | -0.039 |
Token | opus |
Token ID | 25790 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
Today | +0.203 |
arl | +0.161 |
andel | +0.147 |
arry | +0.134 |
owers | +0.133 |
Neg logit contributions | |
never | -0.363 |
made | -0.355 |
learned | -0.337 |
put | -0.334 |
found | -0.333 |
Token | . |
Token ID | 13 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
Today | +0.088 |
oles | +0.087 |
arry | +0.086 |
elly | +0.085 |
ute | +0.084 |
Neg logit contributions | |
never | -0.074 |
punished | -0.071 |
made | -0.071 |
cleaned | -0.071 |
learned | -0.069 |
Token | They |
Token ID | 1119 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
oles | +0.040 |
arry | +0.040 |
ower | +0.038 |
owers | +0.038 |
andel | +0.037 |
Neg logit contributions | |
made | -0.032 |
learned | -0.032 |
put | -0.031 |
never | -0.031 |
tried | -0.029 |
Token | hug |
Token ID | 16225 |
Feature activation | +7.37e-04 |
Pos logit contributions | |
punished | +0.108 |
never | +0.105 |
cleaned | +0.104 |
fixed | +0.102 |
made | +0.100 |
Neg logit contributions | |
Today | -0.163 |
arry | -0.162 |
arl | -0.160 |
ute | -0.155 |
oles | -0.151 |
Token | it |
Token ID | 340 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
never | +0.001 |
made | +0.001 |
learned | +0.001 |
cleaned | +0.001 |
put | +0.001 |
Neg logit contributions | |
oles | -0.002 |
arry | -0.002 |
Today | -0.002 |
elly | -0.002 |
arl | -0.002 |
Token | this |
Token ID | 428 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
oles | +0.061 |
arry | +0.061 |
Today | +0.061 |
elly | +0.060 |
arl | +0.060 |
Neg logit contributions | |
never | -0.028 |
made | -0.027 |
learned | -0.026 |
put | -0.026 |
cleaned | -0.026 |
Token | mess |
Token ID | 2085 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
arry | +0.060 |
oles | +0.060 |
Today | +0.059 |
elly | +0.059 |
arl | +0.058 |
Neg logit contributions | |
never | -0.028 |
made | -0.027 |
put | -0.026 |
learned | -0.026 |
cleaned | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
oles | +0.118 |
arry | +0.118 |
Today | +0.118 |
elly | +0.117 |
arl | +0.116 |
Neg logit contributions | |
never | -0.094 |
made | -0.091 |
learned | -0.089 |
put | -0.089 |
punished | -0.089 |
Token | fix |
Token ID | 4259 |
Feature activation | -6.06e-03 |
Pos logit contributions | |
pretended | +0.045 |
never | +0.043 |
learned | +0.043 |
saved | +0.042 |
made | +0.041 |
Neg logit contributions | |
arl | -0.106 |
Today | -0.104 |
ower | -0.100 |
ute | -0.100 |
arry | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
oles | +0.016 |
elly | +0.016 |
arry | +0.016 |
Today | +0.016 |
ute | +0.015 |
Neg logit contributions | |
never | -0.007 |
cleaned | -0.007 |
learned | -0.007 |
made | -0.007 |
fixed | -0.007 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
oles | +0.049 |
Today | +0.049 |
arry | +0.049 |
elly | +0.048 |
arl | +0.048 |
Neg logit contributions | |
never | -0.047 |
made | -0.045 |
cleaned | -0.044 |
learned | -0.044 |
punished | -0.043 |
Token | opus |
Token ID | 25790 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
Today | +0.194 |
arl | +0.155 |
andel | +0.145 |
arry | +0.136 |
oles | +0.134 |
Neg logit contributions | |
never | -0.355 |
made | -0.347 |
learned | -0.338 |
buried | -0.334 |
fixed | -0.333 |
Token | . |
Token ID | 13 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
oles | +0.103 |
elly | +0.103 |
Today | +0.100 |
ute | +0.099 |
arry | +0.098 |
Neg logit contributions | |
never | -0.091 |
learned | -0.087 |
made | -0.086 |
pretended | -0.084 |
cleaned | -0.084 |
Token | Maybe |
Token ID | 6674 |
Feature activation | +8.21e-03 |
Pos logit contributions | |
oles | +0.061 |
arry | +0.056 |
ute | +0.056 |
owers | +0.056 |
ower | +0.054 |
Neg logit contributions | |
never | -0.063 |
made | -0.059 |
learned | -0.059 |
put | -0.058 |
cleaned | -0.058 |
Token | you |
Token ID | 345 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
never | +0.012 |
punished | +0.011 |
learned | +0.011 |
cleaned | +0.011 |
made | +0.011 |
Neg logit contributions | |
Today | -0.020 |
oles | -0.020 |
arry | -0.020 |
elly | -0.019 |
ummy | -0.019 |
Token | can |
Token ID | 460 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
never | +0.058 |
punished | +0.057 |
pretended | +0.057 |
his | +0.056 |
made | +0.056 |
Neg logit contributions | |
Today | -0.069 |
arry | -0.069 |
elly | -0.067 |
arl | -0.067 |
ute | -0.066 |
Token | Tim |
Token ID | 14967 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
arry | +0.122 |
oles | +0.121 |
Today | +0.121 |
elly | +0.119 |
arl | +0.117 |
Neg logit contributions | |
never | -0.108 |
made | -0.104 |
put | -0.101 |
learned | -0.101 |
cleaned | -0.100 |
Token | went |
Token ID | 1816 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
punished | +0.064 |
permission | +0.063 |
rid | +0.060 |
never | +0.060 |
remind | +0.059 |
Neg logit contributions | |
arry | -0.088 |
oles | -0.081 |
Today | -0.081 |
elly | -0.080 |
olly | -0.080 |
Token | back |
Token ID | 736 |
Feature activation | -9.85e-03 |
Pos logit contributions | |
never | +0.020 |
made | +0.019 |
punished | +0.019 |
cleaned | +0.019 |
learned | +0.019 |
Neg logit contributions | |
Today | -0.026 |
arl | -0.026 |
arry | -0.026 |
elly | -0.025 |
ute | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
oles | +0.024 |
arry | +0.024 |
Today | +0.023 |
elly | +0.023 |
arl | +0.023 |
Neg logit contributions | |
never | -0.015 |
made | -0.015 |
put | -0.014 |
learned | -0.014 |
cleaned | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
punished | +0.030 |
never | +0.028 |
cleaned | +0.027 |
made | +0.026 |
fixed | +0.025 |
Neg logit contributions | |
Today | -0.058 |
ogging | -0.053 |
ore | -0.052 |
ummy | -0.052 |
arry | -0.051 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
elly | +0.047 |
Today | +0.047 |
arry | +0.047 |
oles | +0.047 |
ute | +0.046 |
Neg logit contributions | |
never | -0.054 |
made | -0.052 |
punished | -0.051 |
cleaned | -0.051 |
fixed | -0.051 |
Token | opus |
Token ID | 25790 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
Today | +0.193 |
arl | +0.156 |
andel | +0.139 |
owers | +0.128 |
arry | +0.128 |
Neg logit contributions | |
never | -0.379 |
made | -0.366 |
learned | -0.358 |
found | -0.353 |
put | -0.353 |
Token | and |
Token ID | 290 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
Today | +0.088 |
oles | +0.087 |
arry | +0.087 |
elly | +0.086 |
ute | +0.085 |
Neg logit contributions | |
never | -0.079 |
punished | -0.079 |
cleaned | -0.076 |
made | -0.076 |
fixed | -0.074 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
permission | +0.057 |
wed | +0.055 |
punished | +0.055 |
ps | +0.053 |
never | +0.053 |
Neg logit contributions | |
oles | -0.094 |
arry | -0.092 |
Pole | -0.090 |
aris | -0.090 |
ore | -0.089 |
Token | it |
Token ID | 340 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
punished | +0.016 |
forgot | +0.016 |
never | +0.015 |
cleaned | +0.015 |
learned | +0.015 |
Neg logit contributions | |
ay | -0.020 |
ore | -0.020 |
ummy | -0.020 |
oles | -0.019 |
Today | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
ay | +0.028 |
Today | +0.028 |
elly | +0.027 |
arry | +0.026 |
ore | +0.026 |
Neg logit contributions | |
punished | -0.026 |
never | -0.025 |
forgot | -0.024 |
wished | -0.023 |
remembered | -0.023 |
Token | The |
Token ID | 383 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
oles | +0.076 |
owers | +0.074 |
ower | +0.074 |
ute | +0.074 |
uttering | +0.070 |
Neg logit contributions | |
never | -0.063 |
learned | -0.062 |
put | -0.060 |
made | -0.059 |
saved | -0.058 |
Token | crab |
Token ID | 32202 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
oles | +0.036 |
elly | +0.034 |
Today | +0.034 |
arry | +0.034 |
ute | +0.033 |
Neg logit contributions | |
never | -0.038 |
cleaned | -0.037 |
fixed | -0.037 |
punished | -0.036 |
made | -0.036 |
Token | had |
Token ID | 550 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
punished | +0.019 |
cleaned | +0.018 |
never | +0.018 |
fixed | +0.017 |
made | +0.017 |
Neg logit contributions | |
oles | -0.042 |
arry | -0.040 |
Today | -0.038 |
elly | -0.038 |
olly | -0.037 |
Token | questioned |
Token ID | 11434 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
punished | +0.062 |
ts | +0.057 |
pretended | +0.056 |
cleaned | +0.056 |
forth | +0.052 |
Neg logit contributions | |
oles | -0.162 |
elly | -0.149 |
Today | -0.149 |
arry | -0.148 |
aris | -0.147 |
Token | the |
Token ID | 262 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
never | +0.012 |
made | +0.011 |
cleaned | +0.011 |
learned | +0.011 |
put | +0.011 |
Neg logit contributions | |
oles | -0.030 |
arry | -0.030 |
Today | -0.030 |
elly | -0.029 |
arl | -0.029 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
oles | +0.036 |
Today | +0.036 |
arry | +0.035 |
elly | +0.035 |
ute | +0.034 |
Neg logit contributions | |
never | -0.035 |
cleaned | -0.034 |
fixed | -0.033 |
made | -0.033 |
learned | -0.033 |
Token | opus |
Token ID | 25790 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
Today | +0.206 |
arl | +0.164 |
arry | +0.147 |
andel | +0.139 |
ay | +0.133 |
Neg logit contributions | |
never | -0.373 |
made | -0.351 |
learned | -0.348 |
saved | -0.341 |
found | -0.341 |
Token | , |
Token ID | 11 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
oles | +0.054 |
arry | +0.053 |
Today | +0.053 |
elly | +0.051 |
ute | +0.050 |
Neg logit contributions | |
never | -0.046 |
made | -0.045 |
cleaned | -0.045 |
punished | -0.044 |
put | -0.044 |
Token | but |
Token ID | 475 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
never | +0.055 |
made | +0.054 |
cleaned | +0.053 |
punished | +0.053 |
learned | +0.053 |
Neg logit contributions | |
Today | -0.044 |
oles | -0.044 |
arry | -0.044 |
elly | -0.042 |
arl | -0.041 |
Token | he |
Token ID | 339 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
cleaned | +0.032 |
punished | +0.031 |
fixed | +0.030 |
pretended | +0.030 |
never | +0.028 |
Neg logit contributions | |
oles | -0.077 |
arry | -0.074 |
Today | -0.074 |
ute | -0.072 |
apping | -0.071 |
Token | never |
Token ID | 1239 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
punished | +0.071 |
cleaned | +0.071 |
fixed | +0.071 |
pretended | +0.067 |
repaired | +0.064 |
Neg logit contributions | |
oles | -0.170 |
arry | -0.161 |
olly | -0.151 |
Today | -0.149 |
elly | -0.148 |
Token | he |
Token ID | 339 |
Feature activation | +7.97e-02 |
Pos logit contributions | |
pretended | +0.058 |
cleaned | +0.057 |
fixed | +0.055 |
punished | +0.055 |
made | +0.054 |
Neg logit contributions | |
oles | -0.057 |
arry | -0.056 |
elly | -0.055 |
Today | -0.055 |
ogging | -0.054 |
Token | wanted |
Token ID | 2227 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
cleaned | +0.082 |
punished | +0.079 |
ts | +0.079 |
fixed | +0.078 |
permission | +0.078 |
Neg logit contributions | |
oles | -0.192 |
arry | -0.186 |
elly | -0.183 |
Today | -0.176 |
uddy | -0.175 |
Token | to |
Token ID | 284 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
pretended | +0.065 |
punished | +0.062 |
never | +0.061 |
cleaned | +0.060 |
remembered | +0.060 |
Neg logit contributions | |
elly | -0.052 |
oles | -0.051 |
ummy | -0.050 |
Today | -0.049 |
ore | -0.048 |
Token | keep |
Token ID | 1394 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
cleaned | +0.061 |
pretended | +0.060 |
made | +0.059 |
learned | +0.059 |
never | +0.058 |
Neg logit contributions | |
Today | -0.139 |
arry | -0.138 |
oles | -0.137 |
ute | -0.132 |
� | -0.130 |
Token | the |
Token ID | 262 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
never | +0.043 |
cleaned | +0.043 |
pretended | +0.041 |
learned | +0.040 |
made | +0.040 |
Neg logit contributions | |
oles | -0.078 |
Today | -0.076 |
ummy | -0.075 |
ogging | -0.075 |
ute | -0.075 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
Today | +0.052 |
arry | +0.051 |
oles | +0.051 |
elly | +0.051 |
ute | +0.050 |
Neg logit contributions | |
never | -0.052 |
punished | -0.050 |
cleaned | -0.050 |
fixed | -0.049 |
learned | -0.049 |
Token | opus |
Token ID | 25790 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
Today | +0.200 |
arl | +0.177 |
andel | +0.142 |
arry | +0.141 |
elly | +0.136 |
Neg logit contributions | |
never | -0.355 |
made | -0.351 |
put | -0.338 |
fixed | -0.337 |
learned | -0.334 |
Token | anyway |
Token ID | 6949 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
oles | +0.103 |
Today | +0.100 |
arry | +0.100 |
ute | +0.099 |
elly | +0.098 |
Neg logit contributions | |
never | -0.064 |
made | -0.061 |
learned | -0.060 |
cleaned | -0.060 |
pretended | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
arry | +0.064 |
oles | +0.063 |
arl | +0.063 |
elly | +0.063 |
Today | +0.062 |
Neg logit contributions | |
never | -0.040 |
made | -0.038 |
put | -0.037 |
learned | -0.035 |
fixed | -0.035 |
Token | " |
Token ID | 366 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
oles | +0.065 |
ute | +0.061 |
arry | +0.061 |
arl | +0.060 |
owers | +0.060 |
Neg logit contributions | |
never | -0.065 |
made | -0.064 |
learned | -0.064 |
put | -0.063 |
cleaned | -0.062 |
Token | I |
Token ID | 40 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
never | +0.021 |
made | +0.021 |
cleaned | +0.020 |
put | +0.020 |
learned | +0.020 |
Neg logit contributions | |
arry | -0.023 |
oles | -0.023 |
Today | -0.023 |
elly | -0.023 |
arl | -0.022 |
Token | cone |
Token ID | 27763 |
Feature activation | -2.68e-03 |
Pos logit contributions | |
arry | +0.090 |
oles | +0.090 |
Today | +0.090 |
elly | +0.087 |
arl | +0.085 |
Neg logit contributions | |
never | -0.131 |
made | -0.127 |
learned | -0.125 |
cleaned | -0.125 |
punished | -0.124 |
Token | felt |
Token ID | 2936 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
oles | +0.007 |
arry | +0.006 |
Today | +0.006 |
ute | +0.006 |
ower | +0.006 |
Neg logit contributions | |
punished | -0.003 |
never | -0.003 |
cleaned | -0.003 |
learned | -0.003 |
fixed | -0.003 |
Token | very |
Token ID | 845 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
pretended | +0.059 |
never | +0.057 |
wrote | +0.056 |
cleaned | +0.055 |
shared | +0.055 |
Neg logit contributions | |
Today | -0.096 |
oles | -0.093 |
arl | -0.091 |
apping | -0.090 |
ore | -0.090 |
Token | weak |
Token ID | 4939 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
never | +0.036 |
made | +0.035 |
learned | +0.033 |
put | +0.033 |
cleaned | +0.033 |
Neg logit contributions | |
arry | -0.061 |
oles | -0.061 |
Today | -0.060 |
elly | -0.059 |
arl | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
Today | +0.053 |
elly | +0.052 |
oles | +0.052 |
arry | +0.052 |
arl | +0.050 |
Neg logit contributions | |
never | -0.046 |
punished | -0.044 |
learned | -0.044 |
cleaned | -0.044 |
made | -0.043 |
Token | it |
Token ID | 340 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
cleaned | +0.072 |
fixed | +0.070 |
punished | +0.070 |
pretended | +0.069 |
learned | +0.066 |
Neg logit contributions | |
oles | -0.135 |
Today | -0.131 |
arry | -0.122 |
ower | -0.121 |
ore | -0.121 |
Token | knew |
Token ID | 2993 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
permission | +0.073 |
punished | +0.067 |
fixed | +0.064 |
share | +0.063 |
remind | +0.063 |
Neg logit contributions | |
oles | -0.143 |
ute | -0.131 |
uddle | -0.131 |
Today | -0.129 |
op | -0.129 |
Token | that |
Token ID | 326 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
punished | +0.041 |
pretended | +0.040 |
cleaned | +0.039 |
learned | +0.039 |
fixed | +0.037 |
Neg logit contributions | |
oles | -0.090 |
arry | -0.086 |
ute | -0.083 |
Today | -0.082 |
ummy | -0.082 |
Token | it |
Token ID | 340 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
punished | +0.048 |
cleaned | +0.047 |
never | +0.045 |
learned | +0.045 |
fixed | +0.044 |
Neg logit contributions | |
oles | -0.124 |
Today | -0.121 |
arry | -0.117 |
elly | -0.115 |
ute | -0.115 |
Token | was |
Token ID | 373 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
punished | +0.081 |
cleaned | +0.076 |
fixed | +0.075 |
permission | +0.074 |
put | +0.072 |
Neg logit contributions | |
oles | -0.129 |
elly | -0.118 |
arry | -0.117 |
uddy | -0.117 |
Today | -0.117 |
Token | going |
Token ID | 1016 |
Feature activation | +5.15e-03 |
Pos logit contributions | |
pretended | +0.077 |
learned | +0.072 |
wrote | +0.072 |
drew | +0.069 |
cleaned | +0.068 |
Neg logit contributions | |
Today | -0.108 |
oles | -0.099 |
arl | -0.098 |
uttering | -0.095 |
apping | -0.094 |
Token | they |
Token ID | 484 |
Feature activation | +7.79e-02 |
Pos logit contributions | |
cleaned | +0.046 |
fixed | +0.046 |
punished | +0.045 |
never | +0.045 |
learned | +0.045 |
Neg logit contributions | |
oles | -0.054 |
arry | -0.052 |
Today | -0.050 |
elly | -0.049 |
ummy | -0.048 |
Token | worked |
Token ID | 3111 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
punished | +0.075 |
never | +0.070 |
his | +0.068 |
ep | +0.064 |
fixed | +0.064 |
Neg logit contributions | |
oles | -0.183 |
arry | -0.181 |
owers | -0.180 |
elly | -0.174 |
olly | -0.173 |
Token | together |
Token ID | 1978 |
Feature activation | +9.57e-03 |
Pos logit contributions | |
never | +0.020 |
learned | +0.019 |
put | +0.019 |
punished | +0.019 |
cleaned | +0.019 |
Neg logit contributions | |
arry | -0.022 |
Today | -0.022 |
oles | -0.022 |
elly | -0.021 |
ute | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
never | +0.017 |
made | +0.017 |
punished | +0.017 |
learned | +0.016 |
cleaned | +0.016 |
Neg logit contributions | |
oles | -0.020 |
arry | -0.020 |
Today | -0.020 |
elly | -0.019 |
ute | -0.019 |
Token | they |
Token ID | 484 |
Feature activation | +7.71e-02 |
Pos logit contributions | |
never | +0.021 |
cleaned | +0.021 |
fixed | +0.020 |
made | +0.020 |
punished | +0.020 |
Neg logit contributions | |
oles | -0.028 |
Today | -0.027 |
arry | -0.027 |
elly | -0.026 |
ute | -0.025 |
Token | could |
Token ID | 714 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
punished | +0.091 |
permission | +0.082 |
hurting | +0.082 |
never | +0.081 |
himself | +0.080 |
Neg logit contributions | |
oles | -0.158 |
ane | -0.158 |
arry | -0.156 |
owers | -0.153 |
Today | -0.148 |
Token | build |
Token ID | 1382 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
cleaned | +0.052 |
punished | +0.052 |
learned | +0.051 |
made | +0.050 |
pretended | +0.050 |
Neg logit contributions | |
Today | -0.116 |
arry | -0.114 |
ute | -0.113 |
oles | -0.113 |
� | -0.113 |
Token | the |
Token ID | 262 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
never | +0.021 |
punished | +0.021 |
cleaned | +0.021 |
learned | +0.020 |
remembered | +0.019 |
Neg logit contributions | |
oles | -0.039 |
Today | -0.038 |
arry | -0.037 |
ute | -0.036 |
elly | -0.036 |
Token | house |
Token ID | 2156 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
oles | +0.030 |
Today | +0.030 |
arry | +0.030 |
elly | +0.029 |
ute | +0.029 |
Neg logit contributions | |
never | -0.030 |
made | -0.029 |
cleaned | -0.028 |
learned | -0.028 |
put | -0.028 |
Token | together |
Token ID | 1978 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
Today | +0.070 |
oles | +0.069 |
arry | +0.068 |
elly | +0.066 |
apping | +0.066 |
Neg logit contributions | |
never | -0.053 |
learned | -0.052 |
punished | -0.051 |
cleaned | -0.051 |
made | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
never | +0.020 |
punished | +0.019 |
cleaned | +0.019 |
learned | +0.019 |
made | +0.019 |
Neg logit contributions | |
oles | -0.023 |
Today | -0.023 |
arry | -0.022 |
elly | -0.022 |
ute | -0.021 |
Token | that |
Token ID | 326 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
oles | +0.025 |
Today | +0.025 |
arry | +0.025 |
elly | +0.025 |
arl | +0.025 |
Neg logit contributions | |
never | -0.008 |
learned | -0.008 |
fixed | -0.008 |
made | -0.007 |
cleaned | -0.007 |
Token | they |
Token ID | 484 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
never | +0.063 |
punished | +0.062 |
learned | +0.062 |
cleaned | +0.061 |
fixed | +0.061 |
Neg logit contributions | |
oles | -0.098 |
Today | -0.093 |
owers | -0.091 |
elly | -0.091 |
arry | -0.090 |
Token | were |
Token ID | 547 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
himself | +0.112 |
permission | +0.110 |
his | +0.104 |
tightened | +0.101 |
never | +0.101 |
Neg logit contributions | |
owers | -0.168 |
oles | -0.163 |
� | -0.158 |
uffy | -0.152 |
arry | -0.151 |
Token | as |
Token ID | 355 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
pretended | +0.057 |
learned | +0.057 |
never | +0.056 |
cleaned | +0.053 |
wished | +0.051 |
Neg logit contributions | |
oles | -0.103 |
elly | -0.095 |
owers | -0.094 |
ummy | -0.094 |
ower | -0.094 |
Token | soft |
Token ID | 2705 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
never | +0.041 |
learned | +0.039 |
made | +0.039 |
punished | +0.038 |
cleaned | +0.038 |
Neg logit contributions | |
oles | -0.057 |
arry | -0.054 |
elly | -0.054 |
arl | -0.053 |
owers | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
oles | +0.045 |
arry | +0.044 |
Today | +0.044 |
elly | +0.044 |
owers | +0.043 |
Neg logit contributions | |
never | -0.033 |
made | -0.031 |
learned | -0.031 |
cleaned | -0.031 |
punished | -0.031 |
Token | sweet |
Token ID | 6029 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
never | +0.078 |
punished | +0.077 |
learned | +0.076 |
cleaned | +0.075 |
fixed | +0.073 |
Neg logit contributions | |
oles | -0.098 |
arry | -0.096 |
Today | -0.095 |
elly | -0.093 |
owers | -0.093 |
Token | as |
Token ID | 355 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
oles | +0.034 |
arry | +0.034 |
Today | +0.034 |
elly | +0.033 |
arl | +0.032 |
Neg logit contributions | |
never | -0.027 |
made | -0.025 |
learned | -0.025 |
cleaned | -0.025 |
fixed | -0.025 |
Token | she |
Token ID | 673 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
never | +0.024 |
learned | +0.024 |
punished | +0.023 |
made | +0.023 |
cleaned | +0.023 |
Neg logit contributions | |
oles | -0.030 |
arry | -0.029 |
elly | -0.029 |
Today | -0.029 |
arl | -0.028 |
Token | said |
Token ID | 531 |
Feature activation | +4.52e-03 |
Pos logit contributions | |
never | +0.063 |
punished | +0.063 |
cleaned | +0.056 |
fixed | +0.055 |
learned | +0.053 |
Neg logit contributions | |
owers | -0.168 |
arry | -0.168 |
oles | -0.167 |
elly | -0.167 |
� | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
never | +0.008 |
made | +0.007 |
put | +0.007 |
learned | +0.007 |
cleaned | +0.007 |
Neg logit contributions | |
oles | -0.010 |
arry | -0.010 |
Today | -0.010 |
elly | -0.010 |
arl | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
oles | +0.039 |
Today | +0.038 |
arry | +0.038 |
elly | +0.038 |
ute | +0.037 |
Neg logit contributions | |
never | -0.032 |
cleaned | -0.030 |
made | -0.030 |
learned | -0.030 |
punished | -0.030 |
Token | Jack |
Token ID | 3619 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
oles | +0.066 |
arry | +0.066 |
elly | +0.064 |
Today | +0.063 |
arl | +0.063 |
Neg logit contributions | |
never | -0.087 |
made | -0.084 |
learned | -0.083 |
put | -0.082 |
cleaned | -0.082 |
Token | felt |
Token ID | 2936 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
permission | +0.049 |
punished | +0.048 |
cleaned | +0.045 |
ts | +0.043 |
fixed | +0.043 |
Neg logit contributions | |
oles | -0.091 |
arry | -0.090 |
ane | -0.088 |
ute | -0.088 |
olly | -0.087 |
Token | happy |
Token ID | 3772 |
Feature activation | +7.61e-03 |
Pos logit contributions | |
cleaned | +0.052 |
never | +0.052 |
punished | +0.052 |
pretended | +0.051 |
fixed | +0.050 |
Neg logit contributions | |
Today | -0.079 |
elly | -0.076 |
apping | -0.075 |
arry | -0.074 |
oles | -0.074 |
Token | to |
Token ID | 284 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
punished | +0.015 |
cleaned | +0.014 |
fixed | +0.013 |
wed | +0.013 |
repaired | +0.013 |
Neg logit contributions | |
Today | -0.014 |
apping | -0.014 |
oles | -0.014 |
ummy | -0.014 |
ogging | -0.014 |
Token | have |
Token ID | 423 |
Feature activation | +4.84e-02 |
Pos logit contributions | |
punished | +0.055 |
never | +0.052 |
cleaned | +0.052 |
fixed | +0.050 |
made | +0.050 |
Neg logit contributions | |
Today | -0.133 |
arry | -0.125 |
oles | -0.123 |
arl | -0.123 |
ute | -0.120 |
Token | such |
Token ID | 884 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
punished | +0.045 |
never | +0.039 |
cleaned | +0.037 |
wed | +0.036 |
put | +0.036 |
Neg logit contributions | |
over | -0.121 |
apping | -0.119 |
ay | -0.118 |
elly | -0.116 |
aris | -0.116 |
Token | a |
Token ID | 257 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
never | +0.048 |
cleaned | +0.045 |
made | +0.045 |
punished | +0.044 |
pretended | +0.043 |
Neg logit contributions | |
arry | -0.078 |
oles | -0.078 |
Today | -0.078 |
ute | -0.078 |
elly | -0.077 |
Token | brave |
Token ID | 14802 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
never | +0.026 |
made | +0.025 |
learned | +0.024 |
put | +0.024 |
cleaned | +0.024 |
Neg logit contributions | |
arry | -0.042 |
Today | -0.041 |
elly | -0.040 |
ute | -0.040 |
arl | -0.040 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
Today | +0.071 |
arry | +0.070 |
elly | +0.069 |
arl | +0.068 |
oles | +0.068 |
Neg logit contributions | |
never | -0.063 |
made | -0.060 |
learned | -0.059 |
cleaned | -0.059 |
punished | -0.059 |
Token | like |
Token ID | 588 |
Feature activation | +7.47e-03 |
Pos logit contributions | |
Today | +0.025 |
elly | +0.025 |
arry | +0.025 |
oles | +0.024 |
ute | +0.023 |
Neg logit contributions | |
never | -0.019 |
cleaned | -0.018 |
punished | -0.018 |
put | -0.018 |
made | -0.018 |
Token | they |
Token ID | 484 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
punished | +0.013 |
learned | +0.013 |
never | +0.013 |
cleaned | +0.013 |
fixed | +0.013 |
Neg logit contributions | |
Today | -0.020 |
oles | -0.019 |
arry | -0.019 |
ute | -0.018 |
elly | -0.018 |
Token | are |
Token ID | 389 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
fixed | +0.080 |
cleaned | +0.077 |
punished | +0.075 |
made | +0.073 |
pretended | +0.073 |
Neg logit contributions | |
Today | -0.134 |
oles | -0.133 |
arry | -0.132 |
arl | -0.127 |
� | -0.126 |
Token | very |
Token ID | 845 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
wrote | +0.031 |
pretended | +0.030 |
fixed | +0.030 |
cleaned | +0.030 |
never | +0.030 |
Neg logit contributions | |
apping | -0.060 |
arl | -0.058 |
oles | -0.057 |
Today | -0.057 |
elly | -0.056 |
Token | impatient |
Token ID | 33440 |
Feature activation | -4.65e-02 |
Pos logit contributions | |
never | +0.017 |
made | +0.016 |
put | +0.016 |
learned | +0.015 |
cleaned | +0.015 |
Neg logit contributions | |
Today | -0.030 |
arry | -0.030 |
oles | -0.029 |
owers | -0.029 |
elly | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
arry | +0.106 |
oles | +0.106 |
Today | +0.105 |
elly | +0.104 |
arl | +0.103 |
Neg logit contributions | |
never | -0.077 |
made | -0.074 |
put | -0.072 |
learned | -0.072 |
cleaned | -0.071 |
Token | They |
Token ID | 1119 |
Feature activation | +7.19e-02 |
Pos logit contributions | |
ute | +0.065 |
oles | +0.064 |
opus | +0.064 |
olly | +0.063 |
ogging | +0.062 |
Neg logit contributions | |
put | -0.034 |
never | -0.034 |
his | -0.032 |
fixed | -0.032 |
made | -0.032 |
Token | want |
Token ID | 765 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
fixed | +0.102 |
cleaned | +0.097 |
punished | +0.097 |
never | +0.096 |
made | +0.096 |
Neg logit contributions | |
Today | -0.157 |
arry | -0.150 |
arl | -0.148 |
oles | -0.140 |
ute | -0.138 |
Token | to |
Token ID | 284 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
never | +0.041 |
punished | +0.041 |
learned | +0.041 |
pretended | +0.040 |
remembered | +0.040 |
Neg logit contributions | |
oles | -0.040 |
elly | -0.040 |
Today | -0.039 |
ummy | -0.038 |
arry | -0.038 |
Token | go |
Token ID | 467 |
Feature activation | -4.70e-03 |
Pos logit contributions | |
cleaned | +0.063 |
learned | +0.062 |
made | +0.062 |
never | +0.062 |
punished | +0.061 |
Neg logit contributions | |
Today | -0.109 |
arry | -0.108 |
oles | -0.107 |
ute | -0.104 |
arl | -0.104 |
Token | on |
Token ID | 319 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
arry | +0.011 |
oles | +0.011 |
Today | +0.011 |
elly | +0.011 |
arl | +0.011 |
Neg logit contributions | |
never | -0.007 |
made | -0.007 |
learned | -0.007 |
cleaned | -0.007 |
put | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
arry | +0.049 |
oles | +0.049 |
Today | +0.049 |
elly | +0.048 |
arl | +0.048 |
Neg logit contributions | |
never | -0.014 |
made | -0.013 |
put | -0.012 |
learned | -0.012 |
cleaned | -0.012 |
Token | An |
Token ID | 2025 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
oles | +0.117 |
Today | +0.116 |
arry | +0.116 |
arl | +0.115 |
elly | +0.114 |
Neg logit contributions | |
never | -0.126 |
learned | -0.120 |
made | -0.120 |
put | -0.119 |
cleaned | -0.117 |
Token | nie |
Token ID | 11952 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
arry | +0.104 |
oles | +0.104 |
Today | +0.103 |
elly | +0.102 |
arl | +0.101 |
Neg logit contributions | |
never | -0.054 |
made | -0.052 |
learned | -0.050 |
put | -0.050 |
cleaned | -0.050 |
Token | wasn |
Token ID | 2492 |
Feature activation | +5.78e-03 |
Pos logit contributions | |
punished | +0.036 |
never | +0.035 |
fixed | +0.034 |
made | +0.034 |
cleaned | +0.033 |
Neg logit contributions | |
oles | -0.089 |
arry | -0.089 |
Today | -0.087 |
elly | -0.084 |
apping | -0.084 |
Token | 't |
Token ID | 470 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
never | +0.011 |
put | +0.009 |
made | +0.009 |
fixed | +0.009 |
punished | +0.009 |
Neg logit contributions | |
elly | -0.014 |
arry | -0.013 |
owers | -0.013 |
oles | -0.013 |
Today | -0.013 |
Token | sure |
Token ID | 1654 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
never | +0.048 |
cleaned | +0.044 |
made | +0.044 |
learned | +0.043 |
fixed | +0.042 |
Neg logit contributions | |
arry | -0.136 |
Today | -0.136 |
oles | -0.136 |
elly | -0.132 |
arl | -0.130 |
Token | what |
Token ID | 644 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
never | +0.033 |
made | +0.032 |
learned | +0.031 |
put | +0.030 |
cleaned | +0.030 |
Neg logit contributions | |
oles | -0.055 |
arry | -0.055 |
Today | -0.054 |
elly | -0.053 |
arl | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | -7.24e-03 |
Pos logit contributions | |
never | +0.036 |
made | +0.035 |
punished | +0.034 |
put | +0.034 |
learned | +0.034 |
Neg logit contributions | |
oles | -0.059 |
Today | -0.059 |
arry | -0.058 |
elly | -0.057 |
ute | -0.056 |
Token | old |
Token ID | 1468 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
oles | +0.016 |
arry | +0.016 |
Today | +0.015 |
elly | +0.015 |
arl | +0.015 |
Neg logit contributions | |
never | -0.013 |
made | -0.013 |
put | -0.012 |
learned | -0.012 |
cleaned | -0.012 |
Token | woman |
Token ID | 2415 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
Today | +0.040 |
arry | +0.040 |
oles | +0.039 |
elly | +0.038 |
ute | +0.037 |
Neg logit contributions | |
never | -0.046 |
punished | -0.044 |
made | -0.044 |
cleaned | -0.044 |
fixed | -0.044 |
Token | meant |
Token ID | 4001 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
never | +0.029 |
punished | +0.028 |
cleaned | +0.026 |
fixed | +0.025 |
made | +0.025 |
Neg logit contributions | |
oles | -0.076 |
arry | -0.076 |
Today | -0.075 |
elly | -0.074 |
owers | -0.073 |
Token | , |
Token ID | 11 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
never | +0.041 |
made | +0.040 |
punished | +0.040 |
learned | +0.040 |
cleaned | +0.040 |
Neg logit contributions | |
Today | -0.039 |
oles | -0.039 |
arry | -0.039 |
elly | -0.038 |
arl | -0.037 |
Token | place |
Token ID | 1295 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
punished | +0.026 |
never | +0.026 |
forgot | +0.024 |
cleaned | +0.024 |
pretended | +0.024 |
Neg logit contributions | |
Today | -0.068 |
ay | -0.063 |
elly | -0.062 |
igsaw | -0.062 |
ummy | -0.062 |
Token | in |
Token ID | 287 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
oles | +0.026 |
arry | +0.026 |
Today | +0.026 |
elly | +0.025 |
arl | +0.025 |
Neg logit contributions | |
never | -0.020 |
made | -0.020 |
learned | -0.019 |
cleaned | -0.019 |
punished | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -7.85e-03 |
Pos logit contributions | |
punished | +0.024 |
never | +0.024 |
learned | +0.023 |
cleaned | +0.023 |
made | +0.023 |
Neg logit contributions | |
Today | -0.072 |
oles | -0.069 |
arry | -0.069 |
elly | -0.068 |
arl | -0.068 |
Token | museum |
Token ID | 13257 |
Feature activation | -6.06e-02 |
Pos logit contributions | |
Today | +0.015 |
arry | +0.015 |
oles | +0.015 |
elly | +0.014 |
arl | +0.014 |
Neg logit contributions | |
never | -0.016 |
made | -0.015 |
put | -0.015 |
learned | -0.015 |
cleaned | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
oles | +0.133 |
arry | +0.132 |
Today | +0.131 |
elly | +0.129 |
arl | +0.127 |
Neg logit contributions | |
never | -0.104 |
made | -0.100 |
put | -0.098 |
learned | -0.098 |
cleaned | -0.098 |
Token | excited |
Token ID | 6568 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
never | +0.021 |
cleaned | +0.021 |
made | +0.021 |
punished | +0.021 |
learned | +0.021 |
Neg logit contributions | |
oles | -0.034 |
Today | -0.034 |
arry | -0.033 |
elly | -0.033 |
ute | -0.032 |
Token | for |
Token ID | 329 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
arry | +0.033 |
oles | +0.033 |
Today | +0.033 |
arl | +0.032 |
elly | +0.032 |
Neg logit contributions | |
never | -0.023 |
made | -0.022 |
put | -0.021 |
learned | -0.021 |
cleaned | -0.021 |
Token | her |
Token ID | 607 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
punished | +0.041 |
cleaned | +0.038 |
pretended | +0.037 |
wed | +0.037 |
forgot | +0.037 |
Neg logit contributions | |
Today | -0.074 |
ummy | -0.072 |
oles | -0.071 |
igsaw | -0.070 |
ogging | -0.069 |
Token | next |
Token ID | 1306 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
never | +0.018 |
made | +0.018 |
learned | +0.017 |
cleaned | +0.017 |
put | +0.017 |
Neg logit contributions | |
oles | -0.025 |
arry | -0.025 |
Today | -0.025 |
elly | -0.024 |
ute | -0.023 |
Token | adventure |
Token ID | 8855 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
oles | +0.037 |
arry | +0.036 |
elly | +0.036 |
arl | +0.035 |
owers | +0.035 |
Neg logit contributions | |
never | -0.026 |
made | -0.025 |
put | -0.025 |
fixed | -0.024 |
cleaned | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
oles | +0.092 |
arry | +0.092 |
Today | +0.091 |
elly | +0.090 |
arl | +0.088 |
Neg logit contributions | |
never | -0.073 |
made | -0.071 |
learned | -0.069 |
put | -0.069 |
cleaned | -0.069 |
Token | his |
Token ID | 465 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
never | +0.081 |
permission | +0.079 |
wed | +0.077 |
punished | +0.074 |
supposed | +0.072 |
Neg logit contributions | |
oles | -0.134 |
ls | -0.131 |
oms | -0.126 |
arl | -0.126 |
elly | -0.122 |
Token | hat |
Token ID | 6877 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
never | +0.023 |
punished | +0.022 |
made | +0.021 |
cleaned | +0.021 |
learned | +0.021 |
Neg logit contributions | |
arl | -0.031 |
ute | -0.031 |
oles | -0.031 |
Today | -0.030 |
arry | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
oles | +0.052 |
arl | +0.051 |
Today | +0.051 |
ute | +0.051 |
ay | +0.049 |
Neg logit contributions | |
punished | -0.046 |
never | -0.045 |
cleaned | -0.044 |
learned | -0.044 |
put | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
oles | +0.061 |
owers | +0.061 |
arry | +0.060 |
ute | +0.059 |
andel | +0.058 |
Neg logit contributions | |
never | -0.046 |
made | -0.045 |
his | -0.044 |
put | -0.043 |
learned | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
owers | +0.115 |
oles | +0.107 |
andel | +0.107 |
arry | +0.106 |
ower | +0.105 |
Neg logit contributions | |
never | -0.103 |
put | -0.095 |
learned | -0.095 |
made | -0.092 |
fixed | -0.091 |
Token | Tom |
Token ID | 13787 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
oles | +0.103 |
arry | +0.102 |
Today | +0.101 |
owers | +0.100 |
elly | +0.098 |
Neg logit contributions | |
never | -0.100 |
made | -0.094 |
learned | -0.093 |
put | -0.092 |
cleaned | -0.091 |
Token | laughed |
Token ID | 13818 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
permission | +0.023 |
remind | +0.023 |
never | +0.022 |
care | +0.022 |
punished | +0.022 |
Neg logit contributions | |
oles | -0.042 |
ew | -0.042 |
elly | -0.042 |
uffy | -0.041 |
apping | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | +6.15e-02 |
Pos logit contributions | |
oles | +0.069 |
arry | +0.069 |
Today | +0.067 |
elly | +0.067 |
arl | +0.066 |
Neg logit contributions | |
never | -0.045 |
made | -0.044 |
put | -0.042 |
learned | -0.042 |
cleaned | -0.041 |
Token | walked |
Token ID | 6807 |
Feature activation | +2.54e-03 |
Pos logit contributions | |
permission | +0.081 |
never | +0.076 |
wed | +0.071 |
punished | +0.067 |
repaired | +0.066 |
Neg logit contributions | |
arl | -0.127 |
oles | -0.122 |
arry | -0.118 |
apping | -0.116 |
ew | -0.116 |
Token | to |
Token ID | 284 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
never | +0.003 |
cleaned | +0.003 |
fixed | +0.003 |
learned | +0.003 |
made | +0.003 |
Neg logit contributions | |
Today | -0.007 |
arry | -0.007 |
elly | -0.007 |
arl | -0.007 |
oles | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
punished | +0.018 |
never | +0.017 |
cleaned | +0.017 |
learned | +0.017 |
fixed | +0.017 |
Neg logit contributions | |
Today | -0.024 |
arl | -0.023 |
arry | -0.022 |
ore | -0.022 |
elly | -0.021 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
never | +0.013 |
made | +0.012 |
cleaned | +0.012 |
fixed | +0.012 |
punished | +0.012 |
Neg logit contributions | |
oles | -0.023 |
Today | -0.022 |
arry | -0.022 |
ute | -0.022 |
arl | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +6.51e-02 |
Pos logit contributions | |
never | +0.023 |
punished | +0.022 |
made | +0.022 |
cleaned | +0.022 |
fixed | +0.022 |
Neg logit contributions | |
Today | -0.024 |
oles | -0.024 |
arry | -0.024 |
ute | -0.022 |
elly | -0.022 |
Token | never |
Token ID | 1239 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
punished | +0.091 |
wy | +0.078 |
wed | +0.078 |
fixed | +0.077 |
repaired | +0.076 |
Neg logit contributions | |
oles | -0.129 |
apping | -0.118 |
ute | -0.116 |
arry | -0.114 |
Today | -0.114 |
Token | slipped |
Token ID | 18859 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
never | +0.067 |
punished | +0.061 |
learned | +0.058 |
fixed | +0.058 |
cleaned | +0.058 |
Neg logit contributions | |
arry | -0.181 |
Today | -0.179 |
oles | -0.176 |
elly | -0.176 |
arl | -0.174 |
Token | on |
Token ID | 319 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
never | +0.028 |
learned | +0.027 |
punished | +0.027 |
cleaned | +0.027 |
made | +0.026 |
Neg logit contributions | |
Today | -0.040 |
oles | -0.040 |
arl | -0.039 |
arry | -0.039 |
elly | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
punished | +0.025 |
cleaned | +0.024 |
learned | +0.023 |
pretended | +0.023 |
never | +0.023 |
Neg logit contributions | |
oles | -0.056 |
Today | -0.053 |
� | -0.053 |
owers | -0.051 |
ute | -0.051 |
Token | rock |
Token ID | 3881 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
ute | +0.032 |
oles | +0.032 |
Today | +0.032 |
arry | +0.031 |
arl | +0.031 |
Neg logit contributions | |
punished | -0.020 |
never | -0.020 |
learned | -0.018 |
shared | -0.018 |
cleaned | -0.018 |
Token | again |
Token ID | 757 |
Feature activation | -7.23e-03 |
Pos logit contributions | |
arry | +0.112 |
oles | +0.111 |
Today | +0.110 |
elly | +0.109 |
arl | +0.108 |
Neg logit contributions | |
never | -0.072 |
made | -0.070 |
put | -0.067 |
learned | -0.066 |
cleaned | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
arry | +0.018 |
oles | +0.018 |
Today | +0.017 |
arl | +0.017 |
elly | +0.017 |
Neg logit contributions | |
never | -0.011 |
made | -0.011 |
put | -0.010 |
learned | -0.010 |
fixed | -0.010 |
Token | The |
Token ID | 383 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
oles | +0.075 |
ute | +0.072 |
ower | +0.072 |
arry | +0.071 |
arl | +0.071 |
Neg logit contributions | |
never | -0.082 |
learned | -0.081 |
made | -0.079 |
fixed | -0.076 |
put | -0.075 |
Token | goat |
Token ID | 26917 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
oles | +0.031 |
Today | +0.030 |
arry | +0.030 |
arl | +0.030 |
ute | +0.029 |
Neg logit contributions | |
never | -0.027 |
cleaned | -0.026 |
fixed | -0.026 |
made | -0.025 |
punished | -0.025 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
never | +0.043 |
made | +0.042 |
cleaned | +0.042 |
fixed | +0.041 |
punished | +0.041 |
Neg logit contributions | |
Today | -0.074 |
oles | -0.072 |
arry | -0.070 |
ute | -0.070 |
elly | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
never | +0.021 |
punished | +0.021 |
cleaned | +0.021 |
made | +0.020 |
fixed | +0.020 |
Neg logit contributions | |
Today | -0.026 |
oles | -0.026 |
arry | -0.025 |
elly | -0.025 |
ummy | -0.024 |
Token | that |
Token ID | 326 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
punished | +0.065 |
pretended | +0.065 |
cleaned | +0.063 |
fixed | +0.063 |
never | +0.061 |
Neg logit contributions | |
oles | -0.116 |
ummy | -0.115 |
Today | -0.114 |
arry | -0.112 |
apping | -0.111 |
Token | was |
Token ID | 373 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
punished | +0.047 |
cleaned | +0.047 |
fixed | +0.046 |
never | +0.046 |
pretended | +0.045 |
Neg logit contributions | |
Today | -0.075 |
oles | -0.074 |
arry | -0.071 |
elly | -0.071 |
ummy | -0.071 |
Token | all |
Token ID | 477 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
pretended | +0.049 |
tried | +0.045 |
learned | +0.044 |
ts | +0.043 |
cleaned | +0.042 |
Neg logit contributions | |
Today | -0.091 |
ummy | -0.090 |
elly | -0.087 |
oles | -0.087 |
ay | -0.086 |
Token | that |
Token ID | 326 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
pretended | +0.057 |
never | +0.054 |
remembered | +0.052 |
put | +0.052 |
punished | +0.051 |
Neg logit contributions | |
elly | -0.092 |
Today | -0.091 |
ummy | -0.090 |
igsaw | -0.088 |
ay | -0.086 |
Token | mattered |
Token ID | 33343 |
Feature activation | +9.90e-03 |
Pos logit contributions | |
punished | +0.050 |
never | +0.049 |
cleaned | +0.048 |
fixed | +0.048 |
learned | +0.048 |
Neg logit contributions | |
oles | -0.079 |
Today | -0.077 |
arry | -0.075 |
owers | -0.074 |
elly | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
never | +0.017 |
punished | +0.017 |
cleaned | +0.017 |
fixed | +0.017 |
put | +0.017 |
Neg logit contributions | |
elly | -0.020 |
oles | -0.020 |
Today | -0.020 |
arry | -0.020 |
owers | -0.019 |
Token | And |
Token ID | 843 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
oles | +0.077 |
ower | +0.075 |
ute | +0.074 |
arl | +0.072 |
owers | +0.070 |
Neg logit contributions | |
never | -0.074 |
learned | -0.073 |
made | -0.068 |
put | -0.067 |
cleaned | -0.065 |
Token | so |
Token ID | 523 |
Feature activation | -2.36e-03 |
Pos logit contributions | |
never | +0.024 |
cleaned | +0.023 |
fixed | +0.023 |
made | +0.023 |
punished | +0.023 |
Neg logit contributions | |
arry | -0.036 |
oles | -0.036 |
Today | -0.036 |
elly | -0.035 |
ummy | -0.034 |
Token | she |
Token ID | 673 |
Feature activation | +7.79e-02 |
Pos logit contributions | |
arry | +0.006 |
oles | +0.006 |
Today | +0.006 |
elly | +0.006 |
arl | +0.006 |
Neg logit contributions | |
never | -0.004 |
made | -0.003 |
learned | -0.003 |
put | -0.003 |
cleaned | -0.003 |
Token | lions |
Token ID | 31420 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
arry | +0.049 |
oles | +0.049 |
Today | +0.048 |
elly | +0.047 |
arl | +0.047 |
Neg logit contributions | |
never | -0.043 |
made | -0.041 |
cleaned | -0.040 |
learned | -0.040 |
put | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
arry | +0.121 |
oles | +0.120 |
Today | +0.120 |
elly | +0.118 |
arl | +0.117 |
Neg logit contributions | |
never | -0.083 |
made | -0.080 |
put | -0.078 |
learned | -0.077 |
cleaned | -0.076 |
Token | " |
Token ID | 366 |
Feature activation | +3.01e-03 |
Pos logit contributions | |
oles | +0.091 |
ute | +0.088 |
arry | +0.087 |
ane | +0.086 |
ogging | +0.083 |
Neg logit contributions | |
put | -0.060 |
made | -0.059 |
cleaned | -0.058 |
learned | -0.058 |
never | -0.057 |
Token | Ro |
Token ID | 15450 |
Feature activation | -9.44e-02 |
Pos logit contributions | |
never | +0.006 |
made | +0.005 |
put | +0.005 |
learned | +0.005 |
pretended | +0.005 |
Neg logit contributions | |
oles | -0.007 |
arry | -0.006 |
ute | -0.006 |
elly | -0.006 |
Today | -0.006 |
Token | ar |
Token ID | 283 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
oles | +0.159 |
ute | +0.156 |
Today | +0.156 |
ower | +0.148 |
owers | +0.146 |
Neg logit contributions | |
made | -0.209 |
never | -0.208 |
put | -0.203 |
cleaned | -0.202 |
tried | -0.199 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
Today | +0.072 |
arry | +0.070 |
elly | +0.068 |
oles | +0.067 |
arl | +0.067 |
Neg logit contributions | |
never | -0.070 |
made | -0.069 |
put | -0.067 |
learned | -0.064 |
cleaned | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
Today | +0.088 |
oles | +0.088 |
arry | +0.087 |
elly | +0.086 |
arl | +0.084 |
Neg logit contributions | |
never | -0.067 |
made | -0.063 |
punished | -0.063 |
learned | -0.062 |
cleaned | -0.061 |
Token | lions |
Token ID | 31420 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
oles | +0.056 |
Today | +0.056 |
arry | +0.056 |
elly | +0.053 |
arl | +0.053 |
Neg logit contributions | |
never | -0.061 |
cleaned | -0.059 |
made | -0.058 |
punished | -0.058 |
fixed | -0.058 |
Token | said |
Token ID | 531 |
Feature activation | -1.84e-03 |
Pos logit contributions | |
oles | +0.043 |
arry | +0.041 |
owers | +0.040 |
Today | +0.039 |
apping | +0.039 |
Neg logit contributions | |
never | -0.020 |
punished | -0.019 |
cleaned | -0.018 |
fixed | -0.018 |
learned | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
arry | +0.004 |
elly | +0.004 |
oles | +0.004 |
Today | +0.004 |
arl | +0.004 |
Neg logit contributions | |
never | -0.003 |
made | -0.003 |
put | -0.003 |
cleaned | -0.003 |
fixed | -0.003 |
Token | " |
Token ID | 366 |
Feature activation | +4.29e-03 |
Pos logit contributions | |
oles | +0.095 |
arry | +0.092 |
ute | +0.091 |
ane | +0.090 |
elly | +0.089 |
Neg logit contributions | |
put | -0.063 |
made | -0.063 |
cleaned | -0.061 |
fixed | -0.061 |
learned | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
never | +0.006 |
made | +0.005 |
punished | +0.005 |
learned | +0.005 |
pretended | +0.005 |
Neg logit contributions | |
Today | -0.010 |
arl | -0.010 |
oles | -0.010 |
� | -0.010 |
ummy | -0.010 |
Token | room |
Token ID | 2119 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
Today | +0.059 |
elly | +0.055 |
arl | +0.055 |
ute | +0.055 |
ummy | +0.053 |
Neg logit contributions | |
never | -0.051 |
shared | -0.049 |
learned | -0.049 |
punished | -0.048 |
made | -0.047 |
Token | again |
Token ID | 757 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
Today | +0.074 |
elly | +0.072 |
apping | +0.071 |
arry | +0.070 |
arl | +0.070 |
Neg logit contributions | |
punished | -0.061 |
never | -0.060 |
put | -0.058 |
learned | -0.057 |
made | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
oles | +0.037 |
arry | +0.037 |
Today | +0.037 |
elly | +0.036 |
arl | +0.036 |
Neg logit contributions | |
never | -0.031 |
made | -0.030 |
put | -0.029 |
learned | -0.029 |
cleaned | -0.029 |
Token | |
Token ID | 220 |
Feature activation | -5.63e-02 |
Pos logit contributions | |
ute | +0.075 |
oles | +0.074 |
elly | +0.071 |
arry | +0.071 |
owers | +0.069 |
Neg logit contributions | |
never | -0.069 |
made | -0.067 |
fixed | -0.066 |
cleaned | -0.066 |
learned | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
andel | +0.120 |
elly | +0.118 |
oles | +0.115 |
ute | +0.114 |
arry | +0.111 |
Neg logit contributions | |
never | -0.104 |
fixed | -0.103 |
put | -0.099 |
cleaned | -0.098 |
saved | -0.097 |
Token |
|
Token ID | 198 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
arry | +0.105 |
elly | +0.105 |
oles | +0.105 |
owers | +0.103 |
ute | +0.102 |
Neg logit contributions | |
never | -0.117 |
learned | -0.111 |
made | -0.110 |
put | -0.109 |
cleaned | -0.108 |
Token | Tim |
Token ID | 14967 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
arry | +0.108 |
oles | +0.108 |
Today | +0.107 |
elly | +0.105 |
arl | +0.104 |
Neg logit contributions | |
never | -0.105 |
made | -0.104 |
put | -0.101 |
learned | -0.100 |
cleaned | -0.100 |
Token | my |
Token ID | 1820 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
never | +0.015 |
saved | +0.014 |
punished | +0.014 |
buried | +0.014 |
put | +0.014 |
Neg logit contributions | |
Today | -0.025 |
elly | -0.025 |
ay | -0.025 |
arry | -0.025 |
ummy | -0.024 |
Token | felt |
Token ID | 2936 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
punished | +0.064 |
permission | +0.060 |
rid | +0.057 |
fixed | +0.057 |
buried | +0.055 |
Neg logit contributions | |
arry | -0.131 |
Today | -0.128 |
oles | -0.126 |
olly | -0.125 |
ute | -0.124 |
Token | scared |
Token ID | 12008 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
pretended | +0.058 |
built | +0.055 |
never | +0.055 |
made | +0.054 |
cleaned | +0.052 |
Neg logit contributions | |
Today | -0.083 |
arl | -0.078 |
elly | -0.077 |
ute | -0.076 |
ore | -0.076 |
Token |
|
Token ID | 198 |
Feature activation | -5.45e-02 |
Pos logit contributions | |
ute | +0.081 |
arl | +0.080 |
elly | +0.079 |
andel | +0.077 |
uddle | +0.076 |
Neg logit contributions | |
never | -0.067 |
learned | -0.058 |
tried | -0.057 |
found | -0.057 |
put | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
elly | +0.113 |
owers | +0.110 |
arl | +0.107 |
ower | +0.105 |
umm | +0.105 |
Neg logit contributions | |
never | -0.114 |
learned | -0.102 |
put | -0.099 |
found | -0.098 |
made | -0.098 |
Token | Joe |
Token ID | 19585 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
elly | +0.098 |
arl | +0.097 |
Today | +0.096 |
arry | +0.096 |
oles | +0.095 |
Neg logit contributions | |
never | -0.105 |
made | -0.097 |
learned | -0.096 |
put | -0.094 |
cleaned | -0.092 |
Token | was |
Token ID | 373 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
never | +0.015 |
cleaned | +0.015 |
fixed | +0.015 |
made | +0.015 |
punished | +0.014 |
Neg logit contributions | |
oles | -0.034 |
arry | -0.034 |
elly | -0.033 |
Today | -0.033 |
ane | -0.031 |
Token | getting |
Token ID | 1972 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
pretended | +0.061 |
cleaned | +0.060 |
made | +0.058 |
fixed | +0.058 |
never | +0.056 |
Neg logit contributions | |
Today | -0.086 |
apping | -0.085 |
oles | -0.082 |
elly | -0.080 |
arl | -0.077 |
Token | annoyed |
Token ID | 25602 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
never | +0.031 |
pretended | +0.029 |
learned | +0.029 |
saved | +0.028 |
made | +0.027 |
Neg logit contributions | |
elly | -0.034 |
apping | -0.033 |
Today | -0.033 |
oles | -0.033 |
ore | -0.032 |
Token | by |
Token ID | 416 |
Feature activation | -1.66e-03 |
Pos logit contributions | |
oles | +0.076 |
arry | +0.075 |
arl | +0.075 |
Today | +0.074 |
ute | +0.073 |
Neg logit contributions | |
never | -0.045 |
made | -0.045 |
put | -0.043 |
learned | -0.041 |
punished | -0.040 |
Token | his |
Token ID | 465 |
Feature activation | -6.90e-03 |
Pos logit contributions | |
oles | +0.005 |
Today | +0.005 |
arry | +0.005 |
arl | +0.005 |
elly | +0.005 |
Neg logit contributions | |
never | -0.002 |
made | -0.002 |
learned | -0.001 |
cleaned | -0.001 |
fixed | -0.001 |
Token | nos |
Token ID | 43630 |
Feature activation | -7.51e-02 |
Pos logit contributions | |
arry | +0.016 |
oles | +0.016 |
elly | +0.016 |
Today | +0.016 |
arl | +0.016 |
Neg logit contributions | |
never | -0.011 |
made | -0.011 |
put | -0.010 |
learned | -0.010 |
cleaned | -0.010 |
Token | y |
Token ID | 88 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
oles | +0.118 |
arry | +0.116 |
Today | +0.116 |
ute | +0.114 |
arl | +0.112 |
Neg logit contributions | |
never | -0.174 |
made | -0.173 |
put | -0.167 |
learned | -0.165 |
cleaned | -0.165 |
Token | sister |
Token ID | 6621 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
arry | +0.068 |
oles | +0.068 |
Today | +0.068 |
elly | +0.067 |
arl | +0.066 |
Neg logit contributions | |
never | -0.050 |
made | -0.048 |
put | -0.047 |
learned | -0.047 |
cleaned | -0.047 |
Token | that |
Token ID | 326 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
never | +0.026 |
punished | +0.025 |
made | +0.025 |
cleaned | +0.025 |
learned | +0.025 |
Neg logit contributions | |
arry | -0.047 |
oles | -0.047 |
Today | -0.047 |
ute | -0.046 |
elly | -0.046 |
Token | day |
Token ID | 1110 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
punished | +0.022 |
never | +0.022 |
put | +0.021 |
cleaned | +0.021 |
made | +0.021 |
Neg logit contributions | |
Today | -0.021 |
oles | -0.020 |
arry | -0.020 |
elly | -0.020 |
ute | -0.019 |
Token | on |
Token ID | 319 |
Feature activation | -5.16e-03 |
Pos logit contributions | |
arry | +0.025 |
oles | +0.025 |
elly | +0.024 |
Today | +0.024 |
arl | +0.024 |
Neg logit contributions | |
never | -0.016 |
made | -0.015 |
learned | -0.014 |
put | -0.014 |
cleaned | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
arry | +0.010 |
elly | +0.010 |
arl | +0.010 |
oles | +0.010 |
Today | +0.010 |
Neg logit contributions | |
never | -0.011 |
made | -0.010 |
learned | -0.010 |
put | -0.010 |
cleaned | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
oles | +0.043 |
arry | +0.042 |
Today | +0.042 |
elly | +0.041 |
owers | +0.041 |
Neg logit contributions | |
never | -0.015 |
punished | -0.015 |
cleaned | -0.014 |
made | -0.014 |
learned | -0.014 |
Token | little |
Token ID | 1310 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
arry | +0.063 |
oles | +0.063 |
Today | +0.062 |
elly | +0.061 |
arl | +0.060 |
Neg logit contributions | |
never | -0.059 |
made | -0.058 |
put | -0.056 |
learned | -0.056 |
cleaned | -0.055 |
Token | bird |
Token ID | 6512 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
oles | +0.093 |
arry | +0.093 |
Today | +0.092 |
elly | +0.091 |
arl | +0.089 |
Neg logit contributions | |
never | -0.089 |
made | -0.086 |
learned | -0.084 |
put | -0.084 |
cleaned | -0.084 |
Token | ie |
Token ID | 494 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
oles | +0.022 |
elly | +0.021 |
arry | +0.021 |
ute | +0.020 |
Today | +0.020 |
Neg logit contributions | |
punished | -0.017 |
never | -0.017 |
cleaned | -0.017 |
put | -0.017 |
fixed | -0.017 |
Token | sang |
Token ID | 25889 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
punished | +0.051 |
put | +0.048 |
cleaned | +0.048 |
fixed | +0.048 |
pretended | +0.047 |
Neg logit contributions | |
oles | -0.076 |
arry | -0.073 |
owers | -0.072 |
elly | -0.072 |
ane | -0.069 |
Token | even |
Token ID | 772 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
arry | +0.040 |
Today | +0.040 |
oles | +0.040 |
ute | +0.039 |
elly | +0.039 |
Neg logit contributions | |
never | -0.025 |
punished | -0.025 |
cleaned | -0.024 |
made | -0.024 |
put | -0.023 |
Token | more |
Token ID | 517 |
Feature activation | -5.37e-03 |
Pos logit contributions | |
never | +0.049 |
cleaned | +0.043 |
made | +0.042 |
punished | +0.042 |
fixed | +0.041 |
Neg logit contributions | |
Today | -0.152 |
arry | -0.151 |
elly | -0.150 |
oles | -0.148 |
arl | -0.147 |
Token | p |
Token ID | 279 |
Feature activation | -6.90e-02 |
Pos logit contributions | |
oles | +0.108 |
arry | +0.105 |
Today | +0.104 |
elly | +0.104 |
owers | +0.102 |
Neg logit contributions | |
never | -0.090 |
punished | -0.088 |
made | -0.087 |
cleaned | -0.087 |
learned | -0.086 |
Token | uddle |
Token ID | 24500 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
Today | +0.081 |
oles | +0.077 |
arry | +0.076 |
owers | +0.074 |
ute | +0.071 |
Neg logit contributions | |
never | -0.183 |
made | -0.181 |
cleaned | -0.181 |
learned | -0.179 |
put | -0.178 |
Token | every |
Token ID | 790 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
arry | +0.139 |
oles | +0.138 |
Today | +0.137 |
elly | +0.135 |
arl | +0.134 |
Neg logit contributions | |
never | -0.104 |
made | -0.101 |
put | -0.097 |
learned | -0.097 |
cleaned | -0.096 |
Token | day |
Token ID | 1110 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
never | +0.034 |
made | +0.033 |
put | +0.032 |
learned | +0.032 |
cleaned | +0.032 |
Neg logit contributions | |
arry | -0.038 |
oles | -0.038 |
Today | -0.038 |
elly | -0.037 |
arl | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
arry | +0.031 |
oles | +0.031 |
Today | +0.031 |
arl | +0.031 |
elly | +0.031 |
Neg logit contributions | |
never | -0.022 |
made | -0.022 |
put | -0.021 |
learned | -0.021 |
cleaned | -0.020 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -7.09e-02 |
Pos logit contributions | |
arl | +0.072 |
ute | +0.071 |
elly | +0.068 |
ower | +0.068 |
oles | +0.067 |
Neg logit contributions | |
never | -0.075 |
learned | -0.071 |
made | -0.070 |
found | -0.066 |
put | -0.065 |
Token | Once |
Token ID | 7454 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
ower | +0.154 |
andel | +0.154 |
arl | +0.153 |
ute | +0.150 |
oles | +0.148 |
Neg logit contributions | |
made | -0.143 |
found | -0.137 |
never | -0.132 |
learned | -0.129 |
used | -0.129 |
Token | upon |
Token ID | 2402 |
Feature activation | -2.36e-03 |
Pos logit contributions | |
owers | +0.098 |
elly | +0.098 |
arry | +0.097 |
ower | +0.097 |
arl | +0.097 |
Neg logit contributions | |
never | -0.103 |
made | -0.100 |
found | -0.098 |
put | -0.096 |
learned | -0.094 |
Token | a |
Token ID | 257 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
Today | +0.007 |
oles | +0.007 |
arry | +0.007 |
elly | +0.006 |
ute | +0.006 |
Neg logit contributions | |
never | -0.002 |
punished | -0.002 |
made | -0.002 |
cleaned | -0.002 |
learned | -0.002 |
Token | time |
Token ID | 640 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
oles | +0.092 |
arry | +0.089 |
ane | +0.088 |
owers | +0.088 |
ute | +0.087 |
Neg logit contributions | |
never | -0.057 |
made | -0.052 |
learned | -0.050 |
put | -0.050 |
found | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
andel | +0.115 |
arl | +0.113 |
umm | +0.113 |
elly | +0.112 |
ute | +0.111 |
Neg logit contributions | |
never | -0.044 |
found | -0.040 |
made | -0.039 |
his | -0.036 |
her | -0.035 |
Token | go |
Token ID | 467 |
Feature activation | +6.85e-03 |
Pos logit contributions | |
punished | +0.068 |
learned | +0.065 |
cleaned | +0.064 |
made | +0.063 |
fixed | +0.062 |
Neg logit contributions | |
Today | -0.164 |
arry | -0.158 |
oles | -0.155 |
arl | -0.155 |
ute | -0.153 |
Token | on |
Token ID | 319 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
never | +0.010 |
learned | +0.010 |
made | +0.010 |
cleaned | +0.010 |
punished | +0.010 |
Neg logit contributions | |
Today | -0.016 |
arry | -0.016 |
oles | -0.015 |
arl | -0.015 |
ute | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
Today | +0.029 |
arry | +0.028 |
oles | +0.028 |
elly | +0.028 |
ute | +0.027 |
Neg logit contributions | |
never | -0.011 |
made | -0.011 |
cleaned | -0.011 |
put | -0.010 |
punished | -0.010 |
Token | walk |
Token ID | 2513 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
Today | +0.033 |
arry | +0.033 |
oles | +0.032 |
elly | +0.032 |
arl | +0.031 |
Neg logit contributions | |
never | -0.032 |
made | -0.031 |
learned | -0.030 |
cleaned | -0.030 |
punished | -0.030 |
Token | after |
Token ID | 706 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
oles | +0.098 |
arry | +0.098 |
Today | +0.097 |
arl | +0.096 |
elly | +0.096 |
Neg logit contributions | |
never | -0.067 |
made | -0.064 |
put | -0.062 |
learned | -0.061 |
fixed | -0.060 |
Token | N |
Token ID | 399 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
oles | +0.039 |
arry | +0.039 |
Today | +0.039 |
elly | +0.039 |
arl | +0.038 |
Neg logit contributions | |
never | -0.014 |
made | -0.014 |
learned | -0.013 |
put | -0.013 |
cleaned | -0.013 |
Token | appy |
Token ID | 7774 |
Feature activation | -9.82e-03 |
Pos logit contributions | |
Today | +0.129 |
arl | +0.123 |
ifying | +0.113 |
oing | +0.110 |
ute | +0.109 |
Neg logit contributions | |
put | -0.236 |
used | -0.233 |
never | -0.230 |
found | -0.229 |
made | -0.227 |
Token | 's |
Token ID | 338 |
Feature activation | +9.77e-03 |
Pos logit contributions | |
oles | +0.021 |
arry | +0.021 |
Today | +0.021 |
elly | +0.020 |
owers | +0.019 |
Neg logit contributions | |
never | -0.017 |
punished | -0.016 |
made | -0.016 |
cleaned | -0.016 |
fixed | -0.016 |
Token | nap |
Token ID | 25422 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
never | +0.015 |
made | +0.014 |
put | +0.014 |
learned | +0.014 |
cleaned | +0.014 |
Neg logit contributions | |
arry | -0.024 |
oles | -0.024 |
elly | -0.023 |
arl | -0.023 |
Today | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
oles | +0.081 |
arry | +0.079 |
Today | +0.079 |
elly | +0.078 |
ute | +0.076 |
Neg logit contributions | |
never | -0.071 |
made | -0.068 |
learned | -0.067 |
punished | -0.067 |
cleaned | -0.067 |
Token |
|
Token ID | 198 |
Feature activation | -5.75e-02 |
Pos logit contributions | |
oles | +0.082 |
ower | +0.077 |
elly | +0.075 |
arl | +0.074 |
arry | +0.074 |
Neg logit contributions | |
never | -0.075 |
learned | -0.075 |
put | -0.074 |
made | -0.074 |
found | -0.069 |
Token | a |
Token ID | 257 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
never | +0.031 |
punished | +0.028 |
learned | +0.028 |
made | +0.027 |
cleaned | +0.027 |
Neg logit contributions | |
Today | -0.065 |
ay | -0.065 |
oles | -0.064 |
ummy | -0.063 |
igsaw | -0.063 |
Token | bou |
Token ID | 35833 |
Feature activation | -8.97e-02 |
Pos logit contributions | |
never | +0.020 |
made | +0.019 |
put | +0.018 |
learned | +0.018 |
cleaned | +0.018 |
Neg logit contributions | |
arry | -0.025 |
oles | -0.025 |
Today | -0.025 |
elly | -0.025 |
arl | -0.025 |
Token | quet |
Token ID | 21108 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
Today | +0.119 |
oles | +0.112 |
arry | +0.110 |
elly | +0.107 |
arl | +0.105 |
Neg logit contributions | |
never | -0.237 |
made | -0.235 |
fixed | -0.231 |
cleaned | -0.231 |
put | -0.230 |
Token | of |
Token ID | 286 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
arry | +0.091 |
oles | +0.091 |
Today | +0.091 |
elly | +0.090 |
ute | +0.087 |
Neg logit contributions | |
never | -0.094 |
punished | -0.092 |
learned | -0.092 |
made | -0.091 |
cleaned | -0.090 |
Token | wild |
Token ID | 4295 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
never | +0.025 |
punished | +0.024 |
learned | +0.023 |
made | +0.022 |
cleaned | +0.022 |
Neg logit contributions | |
oles | -0.031 |
Today | -0.030 |
elly | -0.030 |
ummy | -0.030 |
ay | -0.030 |
Token | fl |
Token ID | 2704 |
Feature activation | -9.23e-02 |
Pos logit contributions | |
arry | +0.086 |
oles | +0.085 |
Today | +0.085 |
elly | +0.083 |
arl | +0.083 |
Neg logit contributions | |
never | -0.058 |
made | -0.056 |
put | -0.054 |
learned | -0.054 |
cleaned | -0.054 |
Token | owers |
Token ID | 3618 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
Today | +0.094 |
arry | +0.088 |
oles | +0.080 |
arl | +0.077 |
andel | +0.073 |
Neg logit contributions | |
made | -0.263 |
cleaned | -0.255 |
fixed | -0.253 |
built | -0.247 |
saved | -0.246 |
Token | he |
Token ID | 339 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
oles | +0.115 |
arry | +0.114 |
Today | +0.113 |
elly | +0.112 |
arl | +0.109 |
Neg logit contributions | |
never | -0.088 |
made | -0.084 |
learned | -0.083 |
put | -0.082 |
cleaned | -0.082 |
Token | had |
Token ID | 550 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
punished | +0.049 |
never | +0.045 |
cleaned | +0.044 |
fixed | +0.041 |
put | +0.040 |
Neg logit contributions | |
arry | -0.148 |
Today | -0.147 |
elly | -0.147 |
oles | -0.143 |
ay | -0.142 |
Token | picked |
Token ID | 6497 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
punished | +0.061 |
cleaned | +0.057 |
fixed | +0.054 |
learned | +0.054 |
put | +0.054 |
Neg logit contributions | |
oles | -0.147 |
elly | -0.145 |
owers | -0.143 |
arry | -0.143 |
ower | -0.140 |
Token | for |
Token ID | 329 |
Feature activation | +8.31e-03 |
Pos logit contributions | |
never | +0.016 |
made | +0.015 |
put | +0.014 |
learned | +0.014 |
fixed | +0.014 |
Neg logit contributions | |
arry | -0.024 |
oles | -0.024 |
Today | -0.024 |
arl | -0.024 |
elly | -0.023 |
Token | sweet |
Token ID | 6029 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
arry | +0.018 |
oles | +0.018 |
Today | +0.018 |
elly | +0.018 |
arl | +0.018 |
Neg logit contributions | |
never | -0.014 |
made | -0.013 |
put | -0.013 |
learned | -0.013 |
cleaned | -0.013 |
Token | taste |
Token ID | 6938 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
arry | +0.074 |
oles | +0.073 |
Today | +0.072 |
arl | +0.072 |
elly | +0.072 |
Neg logit contributions | |
never | -0.053 |
made | -0.052 |
put | -0.051 |
learned | -0.050 |
cleaned | -0.050 |
Token | of |
Token ID | 286 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
oles | +0.052 |
elly | +0.050 |
arry | +0.050 |
Today | +0.050 |
ummy | +0.048 |
Neg logit contributions | |
never | -0.064 |
punished | -0.062 |
learned | -0.062 |
made | -0.062 |
cleaned | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
punished | +0.028 |
pretended | +0.026 |
learned | +0.025 |
never | +0.025 |
remembered | +0.025 |
Neg logit contributions | |
Today | -0.055 |
ummy | -0.054 |
oles | -0.054 |
� | -0.052 |
arl | -0.052 |
Token | now |
Token ID | 783 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
oles | +0.035 |
arry | +0.035 |
Today | +0.035 |
elly | +0.034 |
owers | +0.033 |
Neg logit contributions | |
never | -0.032 |
made | -0.031 |
punished | -0.030 |
cleaned | -0.030 |
learned | -0.030 |
Token | pl |
Token ID | 458 |
Feature activation | -9.15e-02 |
Pos logit contributions | |
never | +0.042 |
punished | +0.040 |
learned | +0.039 |
made | +0.039 |
cleaned | +0.039 |
Neg logit contributions | |
Today | -0.054 |
oles | -0.054 |
arry | -0.053 |
elly | -0.053 |
ute | -0.051 |
Token | ump |
Token ID | 931 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
Today | +0.151 |
arry | +0.139 |
arl | +0.132 |
oles | +0.131 |
ute | +0.130 |
Neg logit contributions | |
never | -0.212 |
made | -0.207 |
fixed | -0.203 |
cleaned | -0.202 |
learned | -0.202 |
Token | mel |
Token ID | 7758 |
Feature activation | -9.73e-02 |
Pos logit contributions | |
oles | +0.093 |
elly | +0.086 |
owers | +0.086 |
ower | +0.084 |
arry | +0.084 |
Neg logit contributions | |
never | -0.119 |
learned | -0.114 |
punished | -0.113 |
fixed | -0.112 |
pretended | -0.111 |
Token | on |
Token ID | 261 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
Today | +0.235 |
arry | +0.231 |
arl | +0.229 |
elly | +0.226 |
oles | +0.226 |
Neg logit contributions | |
made | -0.152 |
never | -0.151 |
fixed | -0.144 |
put | -0.143 |
cleaned | -0.143 |
Token | . |
Token ID | 13 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
oles | +0.093 |
arry | +0.089 |
Today | +0.088 |
ummy | +0.087 |
owers | +0.087 |
Neg logit contributions | |
never | -0.070 |
punished | -0.068 |
learned | -0.068 |
made | -0.065 |
cleaned | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
ute | +0.083 |
arl | +0.081 |
elly | +0.079 |
andel | +0.078 |
ower | +0.078 |
Neg logit contributions | |
never | -0.080 |
learned | -0.074 |
made | -0.071 |
saved | -0.070 |
put | -0.069 |
Token | opard |
Token ID | 15478 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
Today | +0.151 |
igsaw | +0.132 |
arry | +0.129 |
ogging | +0.122 |
aris | +0.119 |
Neg logit contributions | |
never | -0.295 |
made | -0.294 |
put | -0.282 |
used | -0.277 |
learned | -0.276 |
Token | 's |
Token ID | 338 |
Feature activation | +6.36e-03 |
Pos logit contributions | |
oles | +0.090 |
arry | +0.089 |
Today | +0.089 |
elly | +0.087 |
arl | +0.086 |
Neg logit contributions | |
never | -0.064 |
made | -0.061 |
learned | -0.060 |
cleaned | -0.060 |
put | -0.059 |
Token | tail |
Token ID | 7894 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
never | +0.011 |
made | +0.011 |
learned | +0.010 |
cleaned | +0.010 |
punished | +0.010 |
Neg logit contributions | |
oles | -0.014 |
Today | -0.014 |
arry | -0.014 |
elly | -0.013 |
arl | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
oles | +0.121 |
arry | +0.120 |
Today | +0.120 |
elly | +0.118 |
arl | +0.116 |
Neg logit contributions | |
never | -0.090 |
made | -0.087 |
learned | -0.084 |
put | -0.084 |
cleaned | -0.084 |
Token | The |
Token ID | 383 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
oles | +0.069 |
ute | +0.069 |
ower | +0.068 |
arry | +0.068 |
elly | +0.067 |
Neg logit contributions | |
put | -0.047 |
never | -0.047 |
made | -0.046 |
his | -0.045 |
her | -0.044 |
Token | le |
Token ID | 443 |
Feature activation | -9.46e-02 |
Pos logit contributions | |
oles | +0.035 |
arry | +0.035 |
Today | +0.035 |
elly | +0.034 |
arl | +0.034 |
Neg logit contributions | |
never | -0.033 |
made | -0.031 |
cleaned | -0.031 |
learned | -0.031 |
punished | -0.030 |
Token | opard |
Token ID | 15478 |
Feature activation | +7.90e-03 |
Pos logit contributions | |
Today | +0.117 |
arry | +0.100 |
igsaw | +0.099 |
ummy | +0.094 |
arl | +0.094 |
Neg logit contributions | |
never | -0.264 |
made | -0.261 |
put | -0.251 |
learned | -0.249 |
saved | -0.247 |
Token | woke |
Token ID | 19092 |
Feature activation | -6.66e-03 |
Pos logit contributions | |
punished | +0.009 |
never | +0.009 |
fixed | +0.009 |
cleaned | +0.008 |
learned | +0.008 |
Neg logit contributions | |
oles | -0.021 |
ew | -0.019 |
elly | -0.018 |
arry | -0.018 |
owers | -0.018 |
Token | up |
Token ID | 510 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
oles | +0.015 |
arry | +0.015 |
Today | +0.015 |
elly | +0.015 |
arl | +0.015 |
Neg logit contributions | |
never | -0.011 |
made | -0.010 |
learned | -0.010 |
cleaned | -0.010 |
put | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
arry | +0.035 |
oles | +0.035 |
Today | +0.034 |
elly | +0.034 |
owers | +0.034 |
Neg logit contributions | |
never | -0.022 |
made | -0.021 |
put | -0.021 |
learned | -0.020 |
fixed | -0.020 |
Token | It |
Token ID | 632 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
ute | +0.069 |
oles | +0.068 |
ower | +0.067 |
arry | +0.066 |
owers | +0.065 |
Neg logit contributions | |
put | -0.046 |
made | -0.046 |
never | -0.045 |
learned | -0.044 |
his | -0.044 |
Token | dirt |
Token ID | 13647 |
Feature activation | -5.94e-02 |
Pos logit contributions | |
Today | +0.023 |
elly | +0.022 |
oles | +0.022 |
arry | +0.022 |
arl | +0.021 |
Neg logit contributions | |
never | -0.015 |
punished | -0.015 |
made | -0.014 |
fixed | -0.014 |
learned | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
oles | +0.134 |
arry | +0.134 |
Today | +0.133 |
elly | +0.131 |
arl | +0.129 |
Neg logit contributions | |
never | -0.099 |
made | -0.096 |
learned | -0.093 |
put | -0.093 |
cleaned | -0.092 |
Token | They |
Token ID | 1119 |
Feature activation | +6.86e-02 |
Pos logit contributions | |
ute | +0.074 |
arl | +0.072 |
arry | +0.072 |
oles | +0.068 |
oing | +0.068 |
Neg logit contributions | |
put | -0.053 |
made | -0.053 |
never | -0.052 |
tried | -0.050 |
found | -0.050 |
Token | use |
Token ID | 779 |
Feature activation | +4.46e-02 |
Pos logit contributions | |
punished | +0.107 |
fixed | +0.100 |
cleaned | +0.099 |
saved | +0.092 |
never | +0.090 |
Neg logit contributions | |
Today | -0.155 |
oles | -0.154 |
arry | -0.150 |
owers | -0.139 |
oms | -0.136 |
Token | their |
Token ID | 511 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
punished | +0.077 |
forgot | +0.064 |
learned | +0.063 |
cleaned | +0.062 |
never | +0.061 |
Neg logit contributions | |
oles | -0.100 |
Today | -0.095 |
oms | -0.090 |
ings | -0.089 |
uttering | -0.085 |
Token | sp |
Token ID | 599 |
Feature activation | -9.23e-02 |
Pos logit contributions | |
punished | +0.037 |
never | +0.037 |
learned | +0.035 |
fixed | +0.034 |
cleaned | +0.034 |
Neg logit contributions | |
oles | -0.044 |
Today | -0.042 |
arry | -0.040 |
owers | -0.039 |
arl | -0.039 |
Token | oons |
Token ID | 13022 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
Today | +0.155 |
arry | +0.150 |
oles | +0.146 |
arl | +0.143 |
elly | +0.142 |
Neg logit contributions | |
never | -0.205 |
made | -0.203 |
put | -0.198 |
cleaned | -0.195 |
learned | -0.192 |
Token | to |
Token ID | 284 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
oles | +0.088 |
arry | +0.085 |
Today | +0.085 |
elly | +0.084 |
arl | +0.082 |
Neg logit contributions | |
never | -0.079 |
punished | -0.077 |
learned | -0.076 |
cleaned | -0.074 |
made | -0.074 |
Token | get |
Token ID | 651 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
punished | +0.078 |
learned | +0.072 |
wished | +0.069 |
never | +0.068 |
remembered | +0.067 |
Neg logit contributions | |
Today | -0.097 |
oles | -0.096 |
ute | -0.089 |
arry | -0.089 |
utter | -0.088 |
Token | it |
Token ID | 340 |
Feature activation | -8.58e-03 |
Pos logit contributions | |
never | +0.031 |
learned | +0.030 |
punished | +0.029 |
forgot | +0.029 |
pretended | +0.028 |
Neg logit contributions | |
oles | -0.040 |
oms | -0.037 |
ore | -0.035 |
elly | -0.034 |
uttering | -0.034 |
Token | out |
Token ID | 503 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
oles | +0.019 |
Today | +0.019 |
arry | +0.018 |
elly | +0.018 |
arl | +0.018 |
Neg logit contributions | |
never | -0.015 |
learned | -0.014 |
made | -0.014 |
punished | -0.014 |
cleaned | -0.014 |