Token | way |
Token ID | 835 |
Feature activation | +8.69e-02 |
Pos logit contributions | |
BOX | +0.007 |
icut | +0.007 |
defe | +0.007 |
Fram | +0.007 |
KEN | +0.007 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.007 |
cars | -0.007 |
Math | -0.006 |
pet | -0.006 |
Token | of |
Token ID | 286 |
Feature activation | +5.61e-03 |
Pos logit contributions | |
BOX | +0.059 |
icut | +0.057 |
defe | +0.057 |
Fram | +0.056 |
KEN | +0.056 |
Neg logit contributions | |
rals | -0.059 |
heimer | -0.055 |
cars | -0.054 |
Math | -0.051 |
pet | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
BOX | +0.004 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.004 |
cars | -0.004 |
Math | -0.003 |
pet | -0.003 |
Token | � |
Token ID | 564 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
BOX | +0.006 |
icut | +0.006 |
defe | +0.006 |
Fram | +0.006 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.006 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.006 |
pet | -0.006 |
Token | � |
Token ID | 250 |
Feature activation | +4.62e-03 |
Pos logit contributions | |
BOX | +0.008 |
icut | +0.008 |
defe | +0.008 |
Fram | +0.008 |
KEN | +0.008 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.006 |
pet | -0.006 |
Token | jay |
Token ID | 33708 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.003 |
pet | -0.003 |
Math | -0.003 |
Token | ve |
Token ID | 303 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
BOX | +0.293 |
icut | +0.286 |
defe | +0.282 |
Fram | +0.278 |
KEN | +0.277 |
Neg logit contributions | |
rals | -0.311 |
heimer | -0.290 |
cars | -0.286 |
pet | -0.269 |
Math | -0.269 |
Token | es |
Token ID | 274 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
rals | +0.015 |
heimer | +0.014 |
cars | +0.014 |
Math | +0.013 |
pet | +0.013 |
Neg logit contributions | |
BOX | -0.017 |
icut | -0.017 |
defe | -0.016 |
Fram | -0.016 |
KEN | -0.016 |
Token | � |
Token ID | 447 |
Feature activation | +3.51e-03 |
Pos logit contributions | |
BOX | +0.028 |
icut | +0.028 |
defe | +0.027 |
Fram | +0.027 |
KEN | +0.027 |
Neg logit contributions | |
rals | -0.030 |
heimer | -0.028 |
cars | -0.027 |
Math | -0.026 |
pet | -0.026 |
Token | � |
Token ID | 251 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.002 |
pet | -0.002 |
Token | that |
Token ID | 326 |
Feature activation | +6.22e-03 |
Pos logit contributions | |
BOX | +0.016 |
icut | +0.015 |
defe | +0.015 |
Fram | +0.015 |
KEN | +0.015 |
Neg logit contributions | |
rals | -0.016 |
heimer | -0.015 |
cars | -0.015 |
Math | -0.014 |
pet | -0.014 |
Token | auc |
Token ID | 14272 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
rals | +0.067 |
heimer | +0.061 |
cars | +0.060 |
Math | +0.056 |
pet | +0.056 |
Neg logit contributions | |
BOX | -0.087 |
icut | -0.086 |
defe | -0.085 |
Fram | -0.084 |
KEN | -0.083 |
Token | ously |
Token ID | 3481 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
rals | +0.117 |
heimer | +0.110 |
cars | +0.109 |
pet | +0.104 |
Math | +0.103 |
Neg logit contributions | |
BOX | -0.092 |
icut | -0.090 |
defe | -0.089 |
Fram | -0.088 |
KEN | -0.087 |
Token | celebrated |
Token ID | 13943 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
BOX | +0.090 |
icut | +0.088 |
defe | +0.086 |
Fram | +0.086 |
KEN | +0.086 |
Neg logit contributions | |
rals | -0.095 |
heimer | -0.089 |
cars | -0.087 |
Math | -0.082 |
pet | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +9.86e-03 |
Pos logit contributions | |
rals | +0.039 |
heimer | +0.036 |
cars | +0.036 |
Math | +0.034 |
pet | +0.034 |
Neg logit contributions | |
BOX | -0.038 |
icut | -0.037 |
defe | -0.036 |
Fram | -0.036 |
KEN | -0.036 |
Token | dism |
Token ID | 18099 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
BOX | +0.007 |
icut | +0.007 |
defe | +0.006 |
Fram | +0.006 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.006 |
pet | -0.006 |
Token | ember |
Token ID | 1491 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
BOX | +0.016 |
icut | +0.013 |
defe | +0.013 |
Fram | +0.012 |
KEN | +0.012 |
Neg logit contributions | |
rals | -0.121 |
heimer | -0.116 |
cars | -0.115 |
pet | -0.112 |
Math | -0.112 |
Token | ment |
Token ID | 434 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
BOX | +0.212 |
icut | +0.207 |
defe | +0.202 |
Fram | +0.199 |
KEN | +0.196 |
Neg logit contributions | |
rals | -0.379 |
heimer | -0.359 |
cars | -0.355 |
Math | -0.338 |
pet | -0.338 |
Token | of |
Token ID | 286 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
BOX | +0.091 |
icut | +0.089 |
defe | +0.088 |
Fram | +0.087 |
KEN | +0.087 |
Neg logit contributions | |
rals | -0.091 |
heimer | -0.085 |
cars | -0.083 |
Math | -0.079 |
pet | -0.079 |
Token | four |
Token ID | 1440 |
Feature activation | -5.62e-02 |
Pos logit contributions | |
BOX | +0.007 |
icut | +0.007 |
defe | +0.007 |
Fram | +0.007 |
KEN | +0.007 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.007 |
cars | -0.007 |
Math | -0.006 |
pet | -0.006 |
Token | rabb |
Token ID | 19368 |
Feature activation | -7.72e-02 |
Pos logit contributions | |
rals | +0.038 |
heimer | +0.036 |
cars | +0.035 |
Math | +0.033 |
pet | +0.033 |
Neg logit contributions | |
BOX | -0.038 |
icut | -0.037 |
defe | -0.037 |
Fram | -0.036 |
KEN | -0.036 |
Token | is |
Token ID | 271 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
rals | +0.064 |
heimer | +0.061 |
cars | +0.060 |
Math | +0.057 |
pet | +0.057 |
Neg logit contributions | |
BOX | -0.040 |
icut | -0.040 |
defe | -0.039 |
Fram | -0.038 |
KEN | -0.038 |
Token | - |
Token ID | 12 |
Feature activation | +1.61e-03 |
Pos logit contributions | |
rals | +0.151 |
heimer | +0.141 |
cars | +0.139 |
Math | +0.131 |
pet | +0.131 |
Neg logit contributions | |
BOX | -0.151 |
icut | -0.147 |
defe | -0.145 |
Fram | -0.144 |
KEN | -0.143 |
Token | desktop |
Token ID | 41375 |
Feature activation | -9.82e-02 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
KEN | +0.001 |
Fram | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
pet | -0.001 |
Math | -0.001 |
Token | - |
Token ID | 12 |
Feature activation | -2.68e-03 |
Pos logit contributions | |
rals | +0.066 |
heimer | +0.062 |
cars | +0.061 |
Math | +0.057 |
pet | +0.057 |
Neg logit contributions | |
BOX | -0.067 |
icut | -0.066 |
defe | -0.065 |
Fram | -0.064 |
KEN | -0.064 |
Token | se |
Token ID | 325 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
pet | +0.002 |
Math | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
KEN | -0.002 |
Fram | -0.002 |
Token | pt |
Token ID | 457 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
rals | +0.050 |
heimer | +0.046 |
cars | +0.045 |
Math | +0.042 |
pet | +0.042 |
Neg logit contributions | |
BOX | -0.057 |
icut | -0.056 |
defe | -0.055 |
Fram | -0.055 |
KEN | -0.054 |
Token | ember |
Token ID | 1491 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
BOX | +0.065 |
defe | +0.060 |
icut | +0.059 |
KEN | +0.057 |
Fram | +0.056 |
Neg logit contributions | |
rals | -0.279 |
heimer | -0.267 |
cars | -0.265 |
Math | -0.256 |
pet | -0.255 |
Token | - |
Token ID | 12 |
Feature activation | +4.98e-03 |
Pos logit contributions | |
BOX | +0.289 |
icut | +0.282 |
defe | +0.279 |
Fram | +0.276 |
KEN | +0.275 |
Neg logit contributions | |
rals | -0.289 |
heimer | -0.269 |
cars | -0.265 |
Math | -0.249 |
pet | -0.249 |
Token | 2017 |
Token ID | 5539 |
Feature activation | -5.58e-02 |
Pos logit contributions | |
BOX | +0.004 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.003 |
pet | -0.002 |
Math | -0.002 |
Token | - |
Token ID | 12 |
Feature activation | +4.09e-03 |
Pos logit contributions | |
rals | +0.038 |
heimer | +0.036 |
cars | +0.035 |
Math | +0.033 |
pet | +0.033 |
Neg logit contributions | |
BOX | -0.037 |
icut | -0.036 |
defe | -0.036 |
Fram | -0.035 |
KEN | -0.035 |
Token | feature |
Token ID | 30053 |
Feature activation | -8.54e-02 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
KEN | +0.003 |
Fram | +0.003 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.002 |
pet | -0.002 |
Math | -0.002 |
Token | - |
Token ID | 12 |
Feature activation | -1.33e-03 |
Pos logit contributions | |
rals | +0.057 |
heimer | +0.053 |
cars | +0.053 |
pet | +0.049 |
Math | +0.049 |
Neg logit contributions | |
BOX | -0.059 |
icut | -0.058 |
defe | -0.057 |
Fram | -0.056 |
KEN | -0.056 |
Token | game |
Token ID | 983 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
BOX | +0.027 |
icut | +0.026 |
defe | +0.026 |
Fram | +0.025 |
KEN | +0.025 |
Neg logit contributions | |
rals | -0.026 |
heimer | -0.024 |
cars | -0.024 |
Math | -0.022 |
pet | -0.022 |
Token | in |
Token ID | 287 |
Feature activation | -3.39e-03 |
Pos logit contributions | |
rals | +0.014 |
heimer | +0.013 |
cars | +0.013 |
Math | +0.012 |
pet | +0.012 |
Neg logit contributions | |
BOX | -0.013 |
icut | -0.013 |
defe | -0.013 |
Fram | -0.013 |
KEN | -0.013 |
Token | charge |
Token ID | 3877 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | at |
Token ID | 379 |
Feature activation | +8.00e-03 |
Pos logit contributions | |
BOX | +0.035 |
icut | +0.034 |
defe | +0.034 |
Fram | +0.033 |
KEN | +0.033 |
Neg logit contributions | |
rals | -0.037 |
heimer | -0.034 |
cars | -0.034 |
Math | -0.032 |
pet | -0.032 |
Token | Old |
Token ID | 5706 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
BOX | +0.006 |
icut | +0.006 |
defe | +0.006 |
Fram | +0.006 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.005 |
cars | -0.005 |
Math | -0.004 |
pet | -0.004 |
Token | Trafford |
Token ID | 41107 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
rals | +0.070 |
heimer | +0.067 |
cars | +0.066 |
Math | +0.065 |
pet | +0.064 |
Neg logit contributions | |
BOX | -0.009 |
icut | -0.008 |
defe | -0.008 |
Fram | -0.007 |
KEN | -0.007 |
Token | against |
Token ID | 1028 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
BOX | +0.277 |
icut | +0.271 |
defe | +0.267 |
Fram | +0.264 |
KEN | +0.262 |
Neg logit contributions | |
rals | -0.299 |
heimer | -0.279 |
cars | -0.275 |
Math | -0.260 |
pet | -0.259 |
Token | Swansea |
Token ID | 36019 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
BOX | +0.020 |
icut | +0.019 |
defe | +0.019 |
Fram | +0.019 |
KEN | +0.019 |
Neg logit contributions | |
rals | -0.022 |
heimer | -0.020 |
cars | -0.020 |
Math | -0.019 |
pet | -0.019 |
Token | on |
Token ID | 319 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
rals | +0.034 |
heimer | +0.031 |
cars | +0.031 |
Math | +0.029 |
pet | +0.029 |
Neg logit contributions | |
BOX | -0.031 |
icut | -0.031 |
defe | -0.030 |
Fram | -0.030 |
KEN | -0.030 |
Token | Sunday |
Token ID | 3502 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
BOX | +0.011 |
icut | +0.011 |
defe | +0.011 |
Fram | +0.010 |
KEN | +0.010 |
Neg logit contributions | |
rals | -0.010 |
heimer | -0.009 |
cars | -0.009 |
Math | -0.008 |
pet | -0.008 |
Token | by |
Token ID | 416 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
rals | +0.023 |
heimer | +0.021 |
cars | +0.021 |
Math | +0.020 |
pet | +0.020 |
Neg logit contributions | |
BOX | -0.022 |
icut | -0.022 |
defe | -0.022 |
Fram | -0.022 |
KEN | -0.021 |
Token | match |
Token ID | 2872 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
BOX | +0.022 |
icut | +0.022 |
defe | +0.021 |
Fram | +0.021 |
KEN | +0.021 |
Neg logit contributions | |
rals | -0.020 |
heimer | -0.019 |
cars | -0.018 |
Math | -0.017 |
pet | -0.017 |
Token | in |
Token ID | 287 |
Feature activation | -6.50e-03 |
Pos logit contributions | |
rals | +0.087 |
heimer | +0.081 |
cars | +0.080 |
Math | +0.076 |
pet | +0.076 |
Neg logit contributions | |
BOX | -0.084 |
icut | -0.082 |
defe | -0.082 |
Fram | -0.080 |
KEN | -0.080 |
Token | charge |
Token ID | 3877 |
Feature activation | +4.99e-02 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.004 |
cars | +0.004 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.004 |
icut | -0.004 |
defe | -0.004 |
Fram | -0.004 |
KEN | -0.004 |
Token | at |
Token ID | 379 |
Feature activation | -2.01e-03 |
Pos logit contributions | |
BOX | +0.035 |
icut | +0.034 |
defe | +0.034 |
Fram | +0.033 |
KEN | +0.033 |
Neg logit contributions | |
rals | -0.033 |
heimer | -0.030 |
cars | -0.030 |
Math | -0.028 |
pet | -0.028 |
Token | Old |
Token ID | 5706 |
Feature activation | -6.18e-02 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | Trafford |
Token ID | 41107 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
rals | +0.072 |
heimer | +0.069 |
cars | +0.069 |
Math | +0.067 |
pet | +0.067 |
Neg logit contributions | |
BOX | -0.012 |
icut | -0.011 |
defe | -0.011 |
KEN | -0.010 |
Fram | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -2.17e-03 |
Pos logit contributions | |
BOX | +0.298 |
icut | +0.292 |
defe | +0.288 |
Fram | +0.285 |
KEN | +0.283 |
Neg logit contributions | |
rals | -0.269 |
heimer | -0.250 |
cars | -0.246 |
Math | -0.232 |
pet | -0.231 |
Token |
|
Token ID | 198 |
Feature activation | -9.97e-03 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | The |
Token ID | 464 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.006 |
cars | +0.006 |
pet | +0.006 |
Math | +0.006 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.007 |
Fram | -0.006 |
KEN | -0.006 |
Token | class |
Token ID | 1398 |
Feature activation | -5.89e-02 |
Pos logit contributions | |
rals | +0.022 |
heimer | +0.021 |
cars | +0.021 |
Math | +0.019 |
pet | +0.019 |
Neg logit contributions | |
BOX | -0.023 |
icut | -0.023 |
defe | -0.022 |
Fram | -0.022 |
KEN | -0.022 |
Token | of |
Token ID | 286 |
Feature activation | -6.03e-04 |
Pos logit contributions | |
rals | +0.040 |
heimer | +0.037 |
cars | +0.037 |
Math | +0.035 |
pet | +0.035 |
Neg logit contributions | |
BOX | -0.040 |
icut | -0.039 |
defe | -0.039 |
KEN | -0.038 |
Fram | -0.038 |
Token | Morse |
Token ID | 44049 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
BOX | +0.011 |
icut | +0.011 |
defe | +0.011 |
Fram | +0.011 |
KEN | +0.011 |
Neg logit contributions | |
rals | -0.009 |
heimer | -0.008 |
cars | -0.008 |
pet | -0.008 |
Math | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
rals | +0.028 |
heimer | +0.026 |
cars | +0.025 |
Math | +0.024 |
pet | +0.024 |
Neg logit contributions | |
BOX | -0.028 |
icut | -0.027 |
defe | -0.027 |
Fram | -0.026 |
KEN | -0.026 |
Token | tie |
Token ID | 9839 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
BOX | +0.009 |
icut | +0.009 |
Fram | +0.009 |
defe | +0.009 |
KEN | +0.009 |
Neg logit contributions | |
rals | -0.008 |
heimer | -0.008 |
cars | -0.007 |
Math | -0.007 |
pet | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
rals | +0.068 |
heimer | +0.063 |
cars | +0.062 |
Math | +0.059 |
pet | +0.059 |
Neg logit contributions | |
BOX | -0.066 |
icut | -0.064 |
defe | -0.063 |
Fram | -0.063 |
KEN | -0.063 |
Token | Ian |
Token ID | 12930 |
Feature activation | -1.82e-03 |
Pos logit contributions | |
BOX | +0.009 |
icut | +0.009 |
Fram | +0.009 |
defe | +0.009 |
KEN | +0.009 |
Neg logit contributions | |
rals | -0.009 |
heimer | -0.008 |
cars | -0.008 |
Math | -0.008 |
pet | -0.008 |
Token | Desmond |
Token ID | 41045 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
defe | -0.001 |
icut | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +9.14e-03 |
Pos logit contributions | |
BOX | +0.290 |
icut | +0.283 |
defe | +0.279 |
Fram | +0.278 |
KEN | +0.276 |
Neg logit contributions | |
rals | -0.276 |
heimer | -0.257 |
cars | -0.253 |
Math | -0.241 |
pet | -0.238 |
Token | take |
Token ID | 1011 |
Feature activation | +6.47e-02 |
Pos logit contributions | |
BOX | +0.006 |
icut | +0.006 |
defe | +0.006 |
Fram | +0.006 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.006 |
heimer | -0.006 |
cars | -0.005 |
Math | -0.005 |
pet | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +9.31e-03 |
Pos logit contributions | |
BOX | +0.043 |
icut | +0.042 |
defe | +0.041 |
Fram | +0.041 |
KEN | +0.040 |
Neg logit contributions | |
rals | -0.045 |
heimer | -0.042 |
cars | -0.042 |
Math | -0.039 |
pet | -0.039 |
Token | lead |
Token ID | 1085 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
BOX | +0.007 |
icut | +0.007 |
defe | +0.007 |
Fram | +0.007 |
KEN | +0.007 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.005 |
cars | -0.005 |
Math | -0.004 |
pet | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +8.17e-03 |
Pos logit contributions | |
rals | +0.069 |
heimer | +0.065 |
cars | +0.064 |
Math | +0.060 |
pet | +0.060 |
Neg logit contributions | |
BOX | -0.073 |
icut | -0.071 |
defe | -0.070 |
Fram | -0.069 |
KEN | -0.069 |
Token | warm |
Token ID | 5814 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
BOX | +0.038 |
icut | +0.037 |
defe | +0.037 |
Fram | +0.036 |
KEN | +0.036 |
Neg logit contributions | |
rals | -0.039 |
heimer | -0.036 |
cars | -0.036 |
Math | -0.034 |
pet | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -6.30e-03 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.007 |
cars | +0.007 |
Math | +0.006 |
pet | +0.006 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.007 |
Fram | -0.007 |
KEN | -0.007 |
Token | stubborn |
Token ID | 24444 |
Feature activation | +7.55e-02 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.004 |
cars | +0.004 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.004 |
icut | -0.004 |
defe | -0.004 |
Fram | -0.004 |
KEN | -0.004 |
Token | sonic |
Token ID | 36220 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
BOX | +0.051 |
icut | +0.049 |
defe | +0.048 |
Fram | +0.048 |
KEN | +0.048 |
Neg logit contributions | |
rals | -0.052 |
heimer | -0.048 |
cars | -0.048 |
Math | -0.045 |
pet | -0.045 |
Token | imp |
Token ID | 848 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
BOX | +0.124 |
icut | +0.121 |
defe | +0.118 |
KEN | +0.117 |
Fram | +0.117 |
Neg logit contributions | |
rals | -0.130 |
heimer | -0.120 |
cars | -0.119 |
Math | -0.113 |
pet | -0.112 |
Token | urities |
Token ID | 10886 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
BOX | +0.059 |
icut | +0.057 |
defe | +0.056 |
Fram | +0.056 |
KEN | +0.055 |
Neg logit contributions | |
rals | -0.079 |
heimer | -0.074 |
cars | -0.073 |
pet | -0.070 |
Math | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | -5.09e-03 |
Pos logit contributions | |
BOX | +0.283 |
icut | +0.277 |
defe | +0.272 |
Fram | +0.270 |
KEN | +0.269 |
Neg logit contributions | |
rals | -0.282 |
heimer | -0.263 |
cars | -0.258 |
Math | -0.244 |
pet | -0.243 |
Token | having |
Token ID | 1719 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.004 |
icut | -0.003 |
defe | -0.003 |
KEN | -0.003 |
Fram | -0.003 |
Token | been |
Token ID | 587 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
rals | +0.030 |
heimer | +0.028 |
cars | +0.028 |
Math | +0.026 |
pet | +0.026 |
Neg logit contributions | |
BOX | -0.031 |
icut | -0.030 |
KEN | -0.030 |
defe | -0.030 |
Fram | -0.029 |
Token | encoded |
Token ID | 30240 |
Feature activation | -7.30e-02 |
Pos logit contributions | |
rals | +0.019 |
heimer | +0.018 |
cars | +0.017 |
Math | +0.017 |
pet | +0.016 |
Neg logit contributions | |
BOX | -0.017 |
icut | -0.017 |
defe | -0.017 |
KEN | -0.017 |
Fram | -0.016 |
Token | in |
Token ID | 287 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
rals | +0.049 |
heimer | +0.046 |
cars | +0.045 |
Math | +0.043 |
pet | +0.042 |
Neg logit contributions | |
BOX | -0.050 |
icut | -0.048 |
KEN | -0.047 |
Fram | -0.047 |
defe | -0.047 |
Token | Jets |
Token ID | 14728 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | � |
Token ID | 447 |
Feature activation | +4.48e-03 |
Pos logit contributions | |
rals | +0.012 |
heimer | +0.011 |
cars | +0.011 |
Math | +0.010 |
pet | +0.010 |
Neg logit contributions | |
BOX | -0.014 |
icut | -0.014 |
defe | -0.013 |
Fram | -0.013 |
KEN | -0.013 |
Token | � |
Token ID | 247 |
Feature activation | +2.01e-03 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.003 |
Math | -0.003 |
pet | -0.003 |
Token | un |
Token ID | 555 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
pet | -0.001 |
Math | -0.001 |
Token | quest |
Token ID | 6138 |
Feature activation | -9.05e-02 |
Pos logit contributions | |
rals | +0.017 |
heimer | +0.016 |
cars | +0.016 |
pet | +0.015 |
Math | +0.014 |
Neg logit contributions | |
BOX | -0.022 |
icut | -0.022 |
defe | -0.022 |
Fram | -0.021 |
KEN | -0.021 |
Token | ioned |
Token ID | 14994 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
rals | +0.109 |
heimer | +0.105 |
cars | +0.104 |
pet | +0.101 |
Math | +0.101 |
Neg logit contributions | |
BOX | -0.014 |
icut | -0.012 |
defe | -0.012 |
Fram | -0.011 |
KEN | -0.011 |
Token | best |
Token ID | 1266 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
BOX | +0.260 |
icut | +0.254 |
defe | +0.249 |
Fram | +0.248 |
KEN | +0.246 |
Neg logit contributions | |
rals | -0.302 |
heimer | -0.283 |
cars | -0.279 |
Math | -0.266 |
pet | -0.265 |
Token | sniper |
Token ID | 26160 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
BOX | +0.034 |
icut | +0.033 |
defe | +0.033 |
Fram | +0.032 |
KEN | +0.032 |
Neg logit contributions | |
rals | -0.038 |
heimer | -0.036 |
cars | -0.035 |
Math | -0.033 |
pet | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | +6.32e-03 |
Pos logit contributions | |
BOX | +0.042 |
icut | +0.041 |
defe | +0.040 |
Fram | +0.040 |
KEN | +0.040 |
Neg logit contributions | |
rals | -0.044 |
heimer | -0.041 |
cars | -0.040 |
Math | -0.038 |
pet | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -3.70e-03 |
Pos logit contributions | |
BOX | +0.004 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.004 |
cars | -0.004 |
Math | -0.004 |
pet | -0.004 |
Token | he |
Token ID | 339 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | Memorial |
Token ID | 14861 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
BOX | +0.045 |
icut | +0.044 |
defe | +0.043 |
Fram | +0.043 |
KEN | +0.043 |
Neg logit contributions | |
rals | -0.045 |
heimer | -0.041 |
cars | -0.041 |
Math | -0.038 |
pet | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | -8.12e-03 |
Pos logit contributions | |
BOX | +0.031 |
icut | +0.030 |
defe | +0.029 |
Fram | +0.029 |
KEN | +0.029 |
Neg logit contributions | |
rals | -0.030 |
heimer | -0.027 |
cars | -0.027 |
Math | -0.025 |
pet | -0.025 |
Token | by |
Token ID | 416 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.006 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | artist |
Token ID | 6802 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
BOX | +0.009 |
icut | +0.008 |
defe | +0.008 |
Fram | +0.008 |
KEN | +0.008 |
Neg logit contributions | |
rals | -0.009 |
heimer | -0.008 |
cars | -0.008 |
Math | -0.008 |
pet | -0.008 |
Token | Robert |
Token ID | 5199 |
Feature activation | +5.49e-02 |
Pos logit contributions | |
BOX | +0.110 |
icut | +0.107 |
defe | +0.106 |
Fram | +0.105 |
KEN | +0.104 |
Neg logit contributions | |
rals | -0.118 |
heimer | -0.110 |
cars | -0.108 |
Math | -0.102 |
pet | -0.102 |
Token | Berks |
Token ID | 45167 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
BOX | +0.038 |
icut | +0.037 |
defe | +0.036 |
Fram | +0.036 |
KEN | +0.036 |
Neg logit contributions | |
rals | -0.037 |
heimer | -0.034 |
cars | -0.034 |
pet | -0.032 |
Math | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | -2.79e-04 |
Pos logit contributions | |
BOX | +0.298 |
icut | +0.292 |
defe | +0.287 |
Fram | +0.284 |
KEN | +0.283 |
Neg logit contributions | |
rals | -0.262 |
heimer | -0.242 |
cars | -0.239 |
Math | -0.224 |
pet | -0.223 |
Token | was |
Token ID | 373 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
rals | +0.000 |
heimer | +0.000 |
cars | +0.000 |
Math | +0.000 |
pet | +0.000 |
Neg logit contributions | |
BOX | -0.000 |
icut | -0.000 |
defe | -0.000 |
Fram | -0.000 |
KEN | -0.000 |
Token | erected |
Token ID | 28415 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
BOX | +0.012 |
icut | +0.012 |
defe | +0.012 |
Fram | +0.012 |
KEN | +0.012 |
Neg logit contributions | |
rals | -0.011 |
heimer | -0.011 |
cars | -0.010 |
Math | -0.010 |
pet | -0.010 |
Token | in |
Token ID | 287 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
BOX | +0.026 |
icut | +0.025 |
defe | +0.025 |
KEN | +0.025 |
Fram | +0.025 |
Neg logit contributions | |
rals | -0.026 |
heimer | -0.024 |
cars | -0.024 |
Math | -0.023 |
pet | -0.022 |
Token | her |
Token ID | 607 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
rals | +0.010 |
heimer | +0.009 |
cars | +0.009 |
Math | +0.009 |
pet | +0.009 |
Neg logit contributions | |
BOX | -0.008 |
icut | -0.008 |
defe | -0.008 |
KEN | -0.008 |
Fram | -0.008 |
Token | on |
Token ID | 319 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
BOX | +0.015 |
icut | +0.014 |
defe | +0.014 |
Fram | +0.014 |
KEN | +0.014 |
Neg logit contributions | |
rals | -0.016 |
heimer | -0.015 |
cars | -0.014 |
Math | -0.014 |
pet | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
BOX | +0.020 |
icut | +0.020 |
defe | +0.020 |
Fram | +0.019 |
KEN | +0.019 |
Neg logit contributions | |
rals | -0.021 |
heimer | -0.020 |
cars | -0.019 |
Math | -0.018 |
pet | -0.018 |
Token | pure |
Token ID | 5899 |
Feature activation | +8.95e-02 |
Pos logit contributions | |
BOX | +0.008 |
icut | +0.008 |
defe | +0.008 |
Fram | +0.008 |
KEN | +0.008 |
Neg logit contributions | |
rals | -0.008 |
heimer | -0.008 |
cars | -0.007 |
Math | -0.007 |
pet | -0.007 |
Token | unf |
Token ID | 3684 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
BOX | +0.063 |
icut | +0.062 |
defe | +0.061 |
Fram | +0.060 |
KEN | +0.060 |
Neg logit contributions | |
rals | -0.058 |
heimer | -0.054 |
cars | -0.053 |
Math | -0.050 |
pet | -0.050 |
Token | il |
Token ID | 346 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
BOX | +0.009 |
icut | +0.009 |
defe | +0.009 |
Fram | +0.009 |
KEN | +0.009 |
Neg logit contributions | |
rals | -0.006 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.005 |
pet | -0.005 |
Token | tered |
Token ID | 4400 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
rals | +0.295 |
heimer | +0.283 |
cars | +0.281 |
pet | +0.273 |
Math | +0.272 |
Neg logit contributions | |
BOX | -0.046 |
defe | -0.041 |
icut | -0.040 |
Fram | -0.036 |
KEN | -0.036 |
Token | visual |
Token ID | 5874 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
BOX | +0.289 |
icut | +0.282 |
defe | +0.279 |
Fram | +0.276 |
KEN | +0.275 |
Neg logit contributions | |
rals | -0.266 |
heimer | -0.247 |
cars | -0.243 |
Math | -0.229 |
pet | -0.228 |
Token | emotions |
Token ID | 10825 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
BOX | +0.028 |
icut | +0.027 |
defe | +0.027 |
KEN | +0.027 |
Fram | +0.027 |
Neg logit contributions | |
rals | -0.026 |
heimer | -0.024 |
cars | -0.024 |
Math | -0.022 |
pet | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
BOX | +0.014 |
icut | +0.013 |
defe | +0.013 |
Fram | +0.013 |
KEN | +0.013 |
Neg logit contributions | |
rals | -0.014 |
heimer | -0.013 |
cars | -0.013 |
Math | -0.012 |
pet | -0.012 |
Token | Tak |
Token ID | 15804 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
BOX | +0.013 |
icut | +0.013 |
defe | +0.013 |
Fram | +0.013 |
KEN | +0.013 |
Neg logit contributions | |
rals | -0.014 |
heimer | -0.013 |
cars | -0.013 |
Math | -0.012 |
pet | -0.012 |
Token | ara |
Token ID | 3301 |
Feature activation | -8.39e-03 |
Pos logit contributions | |
rals | +0.018 |
heimer | +0.017 |
cars | +0.016 |
Math | +0.016 |
pet | +0.016 |
Neg logit contributions | |
BOX | -0.014 |
icut | -0.014 |
defe | -0.013 |
Fram | -0.013 |
KEN | -0.013 |
Token | other |
Token ID | 584 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
BOX | +0.047 |
icut | +0.046 |
defe | +0.045 |
Fram | +0.045 |
KEN | +0.045 |
Neg logit contributions | |
rals | -0.048 |
heimer | -0.045 |
cars | -0.044 |
Math | -0.042 |
pet | -0.041 |
Token | income |
Token ID | 3739 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
rals | +0.010 |
heimer | +0.009 |
cars | +0.009 |
Math | +0.009 |
pet | +0.009 |
Neg logit contributions | |
BOX | -0.010 |
icut | -0.009 |
defe | -0.009 |
Fram | -0.009 |
KEN | -0.009 |
Token | — |
Token ID | 960 |
Feature activation | +9.47e-02 |
Pos logit contributions | |
BOX | +0.036 |
icut | +0.035 |
defe | +0.034 |
Fram | +0.034 |
KEN | +0.034 |
Neg logit contributions | |
rals | -0.034 |
heimer | -0.032 |
cars | -0.031 |
Math | -0.030 |
pet | -0.030 |
Token | chief |
Token ID | 17351 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
BOX | +0.068 |
icut | +0.067 |
defe | +0.066 |
Fram | +0.065 |
KEN | +0.065 |
Neg logit contributions | |
rals | -0.060 |
heimer | -0.056 |
cars | -0.055 |
pet | -0.051 |
Math | -0.051 |
Token | ly |
Token ID | 306 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
BOX | +0.030 |
icut | +0.029 |
defe | +0.029 |
Fram | +0.028 |
KEN | +0.028 |
Neg logit contributions | |
rals | -0.043 |
heimer | -0.041 |
cars | -0.040 |
Math | -0.038 |
pet | -0.038 |
Token | golf |
Token ID | 13126 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
BOX | +0.092 |
icut | +0.090 |
defe | +0.088 |
Fram | +0.087 |
KEN | +0.087 |
Neg logit contributions | |
rals | -0.092 |
heimer | -0.086 |
cars | -0.085 |
Math | -0.080 |
pet | -0.080 |
Token | course |
Token ID | 1781 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
BOX | +0.296 |
icut | +0.290 |
defe | +0.286 |
Fram | +0.283 |
KEN | +0.282 |
Neg logit contributions | |
rals | -0.258 |
heimer | -0.239 |
cars | -0.235 |
Math | -0.220 |
pet | -0.220 |
Token | revenues |
Token ID | 13089 |
Feature activation | -7.11e-02 |
Pos logit contributions | |
rals | +0.126 |
heimer | +0.118 |
cars | +0.116 |
Math | +0.110 |
pet | +0.110 |
Neg logit contributions | |
BOX | -0.111 |
icut | -0.109 |
defe | -0.107 |
Fram | -0.106 |
KEN | -0.105 |
Token | and |
Token ID | 290 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
rals | +0.049 |
heimer | +0.046 |
cars | +0.045 |
Math | +0.043 |
pet | +0.042 |
Neg logit contributions | |
BOX | -0.048 |
icut | -0.047 |
defe | -0.046 |
Fram | -0.045 |
KEN | -0.045 |
Token | condo |
Token ID | 32119 |
Feature activation | -7.69e-02 |
Pos logit contributions | |
BOX | +0.010 |
icut | +0.010 |
defe | +0.009 |
Fram | +0.009 |
KEN | +0.009 |
Neg logit contributions | |
rals | -0.010 |
heimer | -0.010 |
cars | -0.009 |
Math | -0.009 |
pet | -0.009 |
Token | sales |
Token ID | 4200 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
rals | +0.054 |
heimer | +0.051 |
cars | +0.050 |
Math | +0.047 |
pet | +0.047 |
Neg logit contributions | |
BOX | -0.050 |
icut | -0.049 |
defe | -0.048 |
Fram | -0.048 |
KEN | -0.047 |
Token | known |
Token ID | 1900 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
BOX | +0.000 |
icut | +0.000 |
defe | +0.000 |
Fram | +0.000 |
KEN | +0.000 |
Neg logit contributions | |
rals | -0.000 |
heimer | -0.000 |
cars | -0.000 |
Math | -0.000 |
pet | -0.000 |
Token | historically |
Token ID | 15074 |
Feature activation | +9.80e-02 |
Pos logit contributions | |
rals | +0.013 |
heimer | +0.012 |
cars | +0.012 |
Math | +0.012 |
pet | +0.012 |
Neg logit contributions | |
BOX | -0.013 |
icut | -0.013 |
defe | -0.013 |
Fram | -0.013 |
KEN | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +2.27e-03 |
Pos logit contributions | |
BOX | +0.067 |
icut | +0.066 |
defe | +0.065 |
Fram | +0.064 |
KEN | +0.064 |
Neg logit contributions | |
rals | -0.066 |
heimer | -0.061 |
cars | -0.060 |
Math | -0.057 |
pet | -0.057 |
Token | The |
Token ID | 383 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | fluct |
Token ID | 19180 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
BOX | +0.009 |
icut | +0.008 |
defe | +0.008 |
Fram | +0.008 |
KEN | +0.008 |
Neg logit contributions | |
rals | -0.009 |
heimer | -0.008 |
cars | -0.008 |
Math | -0.008 |
pet | -0.008 |
Token | uation |
Token ID | 2288 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
BOX | +0.023 |
icut | +0.022 |
defe | +0.021 |
Fram | +0.020 |
KEN | +0.020 |
Neg logit contributions | |
rals | -0.083 |
heimer | -0.079 |
cars | -0.079 |
pet | -0.076 |
Math | -0.076 |
Token | of |
Token ID | 286 |
Feature activation | +7.60e-03 |
Pos logit contributions | |
BOX | +0.276 |
icut | +0.270 |
defe | +0.266 |
Fram | +0.263 |
KEN | +0.262 |
Neg logit contributions | |
rals | -0.275 |
heimer | -0.257 |
cars | -0.253 |
Math | -0.239 |
pet | -0.238 |
Token | the |
Token ID | 262 |
Feature activation | +2.21e-03 |
Pos logit contributions | |
BOX | +0.005 |
icut | +0.005 |
defe | +0.005 |
Fram | +0.005 |
KEN | +0.005 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.005 |
cars | -0.005 |
Math | -0.004 |
pet | -0.004 |
Token | amount |
Token ID | 2033 |
Feature activation | +9.86e-04 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | of |
Token ID | 286 |
Feature activation | +1.04e-03 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | C |
Token ID | 327 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | mass |
Token ID | 2347 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.002 |
pet | -0.002 |
Token | of |
Token ID | 286 |
Feature activation | +7.21e-04 |
Pos logit contributions | |
rals | +0.029 |
heimer | +0.027 |
cars | +0.026 |
pet | +0.025 |
Math | +0.025 |
Neg logit contributions | |
BOX | -0.027 |
icut | -0.027 |
defe | -0.026 |
KEN | -0.026 |
Fram | -0.026 |
Token | lies |
Token ID | 7363 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
BOX | +0.000 |
icut | +0.000 |
KEN | +0.000 |
defe | +0.000 |
Fram | +0.000 |
Neg logit contributions | |
rals | -0.000 |
heimer | -0.000 |
cars | -0.000 |
Math | -0.000 |
pet | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | +3.73e-03 |
Pos logit contributions | |
BOX | +0.029 |
icut | +0.028 |
defe | +0.028 |
KEN | +0.027 |
Fram | +0.027 |
Neg logit contributions | |
rals | -0.029 |
heimer | -0.027 |
cars | -0.026 |
pet | -0.025 |
Math | -0.025 |
Token | ev |
Token ID | 819 |
Feature activation | +4.59e-03 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
KEN | +0.002 |
defe | +0.002 |
Fram | +0.002 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.002 |
cars | -0.002 |
pet | -0.002 |
Math | -0.002 |
Token | asions |
Token ID | 39327 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
BOX | +0.001 |
defe | +0.001 |
icut | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.005 |
cars | -0.005 |
pet | -0.005 |
Math | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +7.04e-04 |
Pos logit contributions | |
BOX | +0.271 |
icut | +0.265 |
defe | +0.260 |
Fram | +0.257 |
KEN | +0.257 |
Neg logit contributions | |
rals | -0.279 |
heimer | -0.261 |
cars | -0.257 |
Math | -0.242 |
pet | -0.242 |
Token | folly |
Token ID | 43411 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
BOX | +0.000 |
icut | +0.000 |
defe | +0.000 |
KEN | +0.000 |
Fram | +0.000 |
Neg logit contributions | |
rals | -0.000 |
heimer | -0.000 |
cars | -0.000 |
Math | -0.000 |
pet | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
BOX | +0.033 |
icut | +0.033 |
defe | +0.032 |
Fram | +0.032 |
KEN | +0.032 |
Neg logit contributions | |
rals | -0.034 |
heimer | -0.032 |
cars | -0.032 |
pet | -0.030 |
Math | -0.030 |
Token | hatred |
Token ID | 13763 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
BOX | +0.009 |
icut | +0.008 |
defe | +0.008 |
Fram | +0.008 |
KEN | +0.008 |
Neg logit contributions | |
rals | -0.009 |
heimer | -0.008 |
cars | -0.008 |
Math | -0.008 |
pet | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +5.28e-03 |
Pos logit contributions | |
BOX | +0.034 |
icut | +0.034 |
defe | +0.033 |
Fram | +0.033 |
KEN | +0.033 |
Neg logit contributions | |
rals | -0.036 |
heimer | -0.034 |
cars | -0.033 |
Math | -0.031 |
pet | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | -2.05e-03 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | most |
Token ID | 749 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
KEN | -0.001 |
Fram | -0.001 |
Token | recent |
Token ID | 2274 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
BOX | +0.020 |
icut | +0.020 |
defe | +0.020 |
Fram | +0.019 |
KEN | +0.019 |
Neg logit contributions | |
rals | -0.019 |
heimer | -0.017 |
cars | -0.017 |
Math | -0.016 |
pet | -0.016 |
Token | accounts |
Token ID | 5504 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
BOX | +0.027 |
icut | +0.026 |
defe | +0.026 |
Fram | +0.025 |
KEN | +0.025 |
Neg logit contributions | |
rals | -0.027 |
heimer | -0.025 |
cars | -0.025 |
Math | -0.024 |
pet | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | -3.65e-03 |
Pos logit contributions | |
BOX | +0.017 |
icut | +0.016 |
defe | +0.016 |
Fram | +0.016 |
KEN | +0.016 |
Neg logit contributions | |
rals | -0.016 |
heimer | -0.015 |
cars | -0.015 |
Math | -0.014 |
pet | -0.014 |
Token | Desmond |
Token ID | 41045 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | Morris |
Token ID | 14433 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
BOX | +0.272 |
icut | +0.266 |
defe | +0.262 |
Fram | +0.259 |
KEN | +0.258 |
Neg logit contributions | |
rals | -0.277 |
heimer | -0.259 |
cars | -0.255 |
Math | -0.240 |
pet | -0.240 |
Token | 's |
Token ID | 338 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
BOX | +0.079 |
icut | +0.078 |
defe | +0.077 |
Fram | +0.076 |
KEN | +0.076 |
Neg logit contributions | |
rals | -0.072 |
heimer | -0.067 |
cars | -0.066 |
Math | -0.062 |
pet | -0.062 |
Token | Cat |
Token ID | 5181 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
rals | +0.009 |
heimer | +0.008 |
cars | +0.008 |
Math | +0.008 |
pet | +0.008 |
Neg logit contributions | |
BOX | -0.009 |
icut | -0.009 |
defe | -0.009 |
Fram | -0.009 |
KEN | -0.009 |
Token | World |
Token ID | 2159 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
rals | +0.069 |
heimer | +0.065 |
cars | +0.064 |
pet | +0.061 |
Math | +0.060 |
Neg logit contributions | |
BOX | -0.070 |
icut | -0.069 |
defe | -0.068 |
KEN | -0.067 |
Fram | -0.067 |
Token | , |
Token ID | 11 |
Feature activation | -4.18e-03 |
Pos logit contributions | |
rals | +0.060 |
heimer | +0.056 |
cars | +0.056 |
pet | +0.052 |
Math | +0.052 |
Neg logit contributions | |
BOX | -0.061 |
icut | -0.060 |
defe | -0.059 |
Fram | -0.058 |
KEN | -0.058 |
Token | .[ |
Token ID | 3693 |
Feature activation | +5.08e-02 |
Pos logit contributions | |
rals | +0.023 |
heimer | +0.022 |
cars | +0.022 |
Math | +0.020 |
pet | +0.020 |
Neg logit contributions | |
BOX | -0.024 |
icut | -0.023 |
KEN | -0.023 |
defe | -0.023 |
Fram | -0.023 |
Token | 117 |
Token ID | 17657 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
BOX | +0.039 |
icut | +0.039 |
defe | +0.038 |
Fram | +0.038 |
KEN | +0.038 |
Neg logit contributions | |
rals | -0.029 |
heimer | -0.027 |
cars | -0.027 |
pet | -0.025 |
Math | -0.025 |
Token | ] |
Token ID | 60 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
rals | +0.048 |
heimer | +0.045 |
cars | +0.045 |
Math | +0.042 |
pet | +0.042 |
Neg logit contributions | |
BOX | -0.044 |
icut | -0.043 |
defe | -0.042 |
Fram | -0.042 |
KEN | -0.042 |
Token | A |
Token ID | 317 |
Feature activation | +2.69e-03 |
Pos logit contributions | |
BOX | +0.018 |
icut | +0.018 |
defe | +0.017 |
Fram | +0.017 |
KEN | +0.017 |
Neg logit contributions | |
rals | -0.018 |
heimer | -0.017 |
cars | -0.017 |
Math | -0.016 |
pet | -0.016 |
Token | br |
Token ID | 1671 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.002 |
pet | -0.002 |
Token | asions |
Token ID | 39327 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
BOX | +0.065 |
icut | +0.063 |
defe | +0.062 |
Fram | +0.061 |
KEN | +0.061 |
Neg logit contributions | |
rals | -0.081 |
heimer | -0.076 |
cars | -0.075 |
Math | -0.071 |
pet | -0.071 |
Token | around |
Token ID | 1088 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
BOX | +0.275 |
icut | +0.268 |
defe | +0.264 |
Fram | +0.262 |
KEN | +0.261 |
Neg logit contributions | |
rals | -0.274 |
heimer | -0.257 |
cars | -0.252 |
pet | -0.238 |
Math | -0.237 |
Token | the |
Token ID | 262 |
Feature activation | -5.20e-03 |
Pos logit contributions | |
rals | +0.035 |
heimer | +0.033 |
cars | +0.032 |
Math | +0.031 |
pet | +0.031 |
Neg logit contributions | |
BOX | -0.034 |
icut | -0.034 |
defe | -0.033 |
Fram | -0.033 |
KEN | -0.033 |
Token | eyes |
Token ID | 2951 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -2.94e-04 |
Pos logit contributions | |
BOX | +0.009 |
icut | +0.009 |
defe | +0.009 |
Fram | +0.008 |
KEN | +0.008 |
Neg logit contributions | |
rals | -0.009 |
heimer | -0.008 |
cars | -0.008 |
pet | -0.008 |
Math | -0.008 |
Token | nose |
Token ID | 9686 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
rals | +0.000 |
heimer | +0.000 |
cars | +0.000 |
Math | +0.000 |
pet | +0.000 |
Neg logit contributions | |
BOX | -0.000 |
icut | -0.000 |
defe | -0.000 |
KEN | -0.000 |
Fram | -0.000 |
Token | Toronto |
Token ID | 6586 |
Feature activation | -6.58e-02 |
Pos logit contributions | |
rals | +0.021 |
heimer | +0.020 |
cars | +0.019 |
Math | +0.018 |
pet | +0.018 |
Neg logit contributions | |
BOX | -0.021 |
icut | -0.021 |
defe | -0.020 |
KEN | -0.020 |
Fram | -0.020 |
Token | has |
Token ID | 468 |
Feature activation | -7.52e-03 |
Pos logit contributions | |
rals | +0.045 |
heimer | +0.042 |
cars | +0.042 |
Math | +0.039 |
pet | +0.039 |
Neg logit contributions | |
BOX | -0.044 |
icut | -0.043 |
defe | -0.042 |
Fram | -0.042 |
KEN | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | -1.97e-03 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
KEN | -0.005 |
Fram | -0.005 |
Token | contemporary |
Token ID | 11811 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | |
Token ID | 220 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
rals | +0.012 |
heimer | +0.011 |
cars | +0.011 |
Math | +0.010 |
pet | +0.010 |
Neg logit contributions | |
BOX | -0.012 |
icut | -0.012 |
defe | -0.012 |
KEN | -0.012 |
Fram | -0.012 |
Token | vern |
Token ID | 933 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
BOX | +0.057 |
icut | +0.049 |
defe | +0.049 |
Fram | +0.047 |
KEN | +0.047 |
Neg logit contributions | |
rals | -0.314 |
heimer | -0.305 |
cars | -0.298 |
pet | -0.290 |
Math | -0.288 |
Token | acular |
Token ID | 12754 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
BOX | +0.227 |
icut | +0.221 |
defe | +0.217 |
Fram | +0.214 |
KEN | +0.213 |
Neg logit contributions | |
rals | -0.322 |
heimer | -0.303 |
cars | -0.299 |
Math | -0.284 |
pet | -0.284 |
Token | when |
Token ID | 618 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
BOX | +0.052 |
icut | +0.050 |
defe | +0.049 |
KEN | +0.049 |
Fram | +0.049 |
Neg logit contributions | |
rals | -0.047 |
heimer | -0.043 |
cars | -0.043 |
Math | -0.040 |
pet | -0.040 |
Token | it |
Token ID | 340 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
rals | +0.026 |
heimer | +0.025 |
cars | +0.024 |
Math | +0.023 |
pet | +0.023 |
Neg logit contributions | |
BOX | -0.028 |
icut | -0.028 |
defe | -0.027 |
KEN | -0.027 |
Fram | -0.027 |
Token | comes |
Token ID | 2058 |
Feature activation | +8.94e-02 |
Pos logit contributions | |
rals | +0.009 |
heimer | +0.008 |
cars | +0.008 |
Math | +0.008 |
pet | +0.008 |
Neg logit contributions | |
BOX | -0.006 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
BOX | +0.060 |
icut | +0.058 |
defe | +0.058 |
Fram | +0.057 |
KEN | +0.057 |
Neg logit contributions | |
rals | -0.062 |
heimer | -0.058 |
cars | -0.057 |
Math | -0.053 |
pet | -0.053 |
Token | of |
Token ID | 286 |
Feature activation | -6.78e-03 |
Pos logit contributions | |
BOX | +0.024 |
icut | +0.024 |
defe | +0.023 |
Fram | +0.023 |
KEN | +0.023 |
Neg logit contributions | |
rals | -0.025 |
heimer | -0.023 |
cars | -0.023 |
Math | -0.021 |
pet | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -4.47e-03 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.004 |
cars | +0.004 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.004 |
defe | -0.004 |
KEN | -0.004 |
Fram | -0.004 |
Token | � |
Token ID | 564 |
Feature activation | +8.73e-03 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
KEN | -0.003 |
Fram | -0.003 |
defe | -0.003 |
Token | � |
Token ID | 250 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
BOX | +0.006 |
icut | +0.006 |
defe | +0.006 |
Fram | +0.006 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.006 |
heimer | -0.005 |
cars | -0.005 |
Math | -0.005 |
pet | -0.005 |
Token | great |
Token ID | 18223 |
Feature activation | -9.45e-02 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.007 |
cars | +0.007 |
pet | +0.006 |
Math | +0.006 |
Neg logit contributions | |
BOX | -0.009 |
icut | -0.009 |
defe | -0.008 |
KEN | -0.008 |
Fram | -0.008 |
Token | capitals |
Token ID | 44590 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
rals | +0.068 |
heimer | +0.062 |
cars | +0.062 |
Math | +0.059 |
pet | +0.058 |
Neg logit contributions | |
BOX | -0.062 |
icut | -0.059 |
defe | -0.058 |
Fram | -0.058 |
KEN | -0.058 |
Token | of |
Token ID | 286 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
BOX | +0.264 |
icut | +0.257 |
defe | +0.253 |
Fram | +0.250 |
KEN | +0.249 |
Neg logit contributions | |
rals | -0.283 |
heimer | -0.264 |
cars | -0.260 |
Math | -0.246 |
pet | -0.246 |
Token | Islamic |
Token ID | 5533 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
rals | +0.008 |
heimer | +0.008 |
cars | +0.008 |
Math | +0.007 |
pet | +0.007 |
Neg logit contributions | |
BOX | -0.008 |
icut | -0.008 |
defe | -0.008 |
Fram | -0.008 |
KEN | -0.008 |
Token | finance |
Token ID | 9604 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
BOX | +0.130 |
icut | +0.126 |
defe | +0.124 |
KEN | +0.124 |
Fram | +0.122 |
Neg logit contributions | |
rals | -0.139 |
heimer | -0.130 |
cars | -0.128 |
Math | -0.123 |
pet | -0.121 |
Token | � |
Token ID | 447 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
BOX | +0.092 |
icut | +0.090 |
defe | +0.089 |
KEN | +0.088 |
Fram | +0.088 |
Neg logit contributions | |
rals | -0.091 |
heimer | -0.085 |
cars | -0.084 |
Math | -0.079 |
pet | -0.079 |
Token | � |
Token ID | 251 |
Feature activation | +8.49e-03 |
Pos logit contributions | |
rals | +0.008 |
heimer | +0.007 |
cars | +0.007 |
Math | +0.007 |
pet | +0.007 |
Neg logit contributions | |
BOX | -0.008 |
icut | -0.008 |
defe | -0.008 |
Fram | -0.008 |
KEN | -0.008 |
Token | his |
Token ID | 465 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
BOX | +0.062 |
icut | +0.061 |
defe | +0.060 |
Fram | +0.059 |
KEN | +0.059 |
Neg logit contributions | |
rals | -0.063 |
heimer | -0.059 |
cars | -0.058 |
Math | -0.055 |
pet | -0.054 |
Token | first |
Token ID | 717 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
rals | +0.014 |
heimer | +0.013 |
cars | +0.013 |
Math | +0.012 |
pet | +0.012 |
Neg logit contributions | |
BOX | -0.014 |
icut | -0.013 |
defe | -0.013 |
Fram | -0.013 |
KEN | -0.013 |
Token | photography |
Token ID | 16223 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
BOX | +0.008 |
icut | +0.008 |
defe | +0.008 |
Fram | +0.007 |
KEN | +0.007 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.006 |
pet | -0.006 |
Token | book |
Token ID | 1492 |
Feature activation | +6.74e-02 |
Pos logit contributions | |
rals | +0.112 |
cars | +0.104 |
heimer | +0.103 |
Math | +0.098 |
pet | +0.097 |
Neg logit contributions | |
BOX | -0.121 |
icut | -0.119 |
defe | -0.117 |
KEN | -0.116 |
Fram | -0.116 |
Token | by |
Token ID | 416 |
Feature activation | +2.70e-03 |
Pos logit contributions | |
BOX | +0.045 |
icut | +0.044 |
defe | +0.043 |
Fram | +0.043 |
KEN | +0.042 |
Neg logit contributions | |
rals | -0.047 |
heimer | -0.043 |
cars | -0.043 |
Math | -0.040 |
pet | -0.040 |
Token | Wee |
Token ID | 48643 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.002 |
pet | -0.002 |
Token | gee |
Token ID | 29622 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
BOX | +0.268 |
icut | +0.261 |
defe | +0.258 |
Fram | +0.255 |
KEN | +0.254 |
Neg logit contributions | |
rals | -0.274 |
heimer | -0.256 |
cars | -0.252 |
Math | -0.237 |
pet | -0.237 |
Token | , |
Token ID | 11 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
rals | +0.079 |
heimer | +0.074 |
cars | +0.073 |
Math | +0.069 |
pet | +0.068 |
Neg logit contributions | |
BOX | -0.081 |
icut | -0.079 |
defe | -0.078 |
Fram | -0.077 |
KEN | -0.077 |
Token | who |
Token ID | 508 |
Feature activation | +1.28e-03 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.006 |
cars | +0.006 |
Math | +0.006 |
pet | +0.006 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.006 |
Fram | -0.006 |
KEN | -0.006 |
Token | is |
Token ID | 318 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | famous |
Token ID | 5863 |
Feature activation | -2.24e-03 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.007 |
cars | +0.007 |
Math | +0.006 |
pet | +0.006 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.007 |
Fram | -0.007 |
KEN | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +2.06e-03 |
Pos logit contributions | |
BOX | +0.058 |
icut | +0.057 |
defe | +0.056 |
Fram | +0.055 |
KEN | +0.055 |
Neg logit contributions | |
rals | -0.057 |
heimer | -0.053 |
cars | -0.052 |
Math | -0.049 |
pet | -0.049 |
Token | forms |
Token ID | 5107 |
Feature activation | -8.22e-02 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | of |
Token ID | 286 |
Feature activation | +5.48e-04 |
Pos logit contributions | |
rals | +0.054 |
heimer | +0.051 |
cars | +0.050 |
pet | +0.047 |
Math | +0.047 |
Neg logit contributions | |
BOX | -0.057 |
icut | -0.056 |
defe | -0.055 |
KEN | -0.054 |
Fram | -0.054 |
Token | trans |
Token ID | 1007 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
BOX | +0.000 |
icut | +0.000 |
defe | +0.000 |
Fram | +0.000 |
KEN | +0.000 |
Neg logit contributions | |
rals | -0.000 |
heimer | -0.000 |
cars | -0.000 |
Math | -0.000 |
pet | -0.000 |
Token | val |
Token ID | 2100 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
rals | +0.154 |
heimer | +0.145 |
cars | +0.143 |
pet | +0.137 |
Math | +0.137 |
Neg logit contributions | |
BOX | -0.095 |
icut | -0.091 |
defe | -0.090 |
Fram | -0.089 |
KEN | -0.088 |
Token | uation |
Token ID | 2288 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
BOX | +0.025 |
icut | +0.021 |
defe | +0.020 |
Fram | +0.019 |
KEN | +0.018 |
Neg logit contributions | |
rals | -0.222 |
heimer | -0.213 |
cars | -0.211 |
pet | -0.205 |
Math | -0.205 |
Token | manifested |
Token ID | 34300 |
Feature activation | -2.66e-03 |
Pos logit contributions | |
BOX | +0.273 |
icut | +0.266 |
defe | +0.262 |
Fram | +0.260 |
KEN | +0.259 |
Neg logit contributions | |
rals | -0.269 |
heimer | -0.251 |
cars | -0.247 |
pet | -0.232 |
Math | -0.232 |
Token | by |
Token ID | 416 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +1.77e-03 |
Pos logit contributions | |
BOX | +0.012 |
icut | +0.012 |
defe | +0.011 |
Fram | +0.011 |
KEN | +0.011 |
Neg logit contributions | |
rals | -0.012 |
heimer | -0.011 |
cars | -0.011 |
Math | -0.011 |
pet | -0.011 |
Token | reversal |
Token ID | 27138 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | of |
Token ID | 286 |
Feature activation | +2.91e-03 |
Pos logit contributions | |
BOX | +0.087 |
icut | +0.085 |
defe | +0.084 |
Fram | +0.083 |
KEN | +0.082 |
Neg logit contributions | |
rals | -0.088 |
heimer | -0.082 |
cars | -0.081 |
Math | -0.076 |
pet | -0.076 |
Token | personal |
Token ID | 2614 |
Feature activation | +1.04e-03 |
Pos logit contributions | |
BOX | +0.023 |
icut | +0.022 |
defe | +0.022 |
Fram | +0.022 |
KEN | +0.022 |
Neg logit contributions | |
rals | -0.022 |
heimer | -0.020 |
cars | -0.020 |
Math | -0.019 |
pet | -0.019 |
Token | use |
Token ID | 779 |
Feature activation | -9.08e-02 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
KEN | +0.001 |
Fram | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | apps |
Token ID | 6725 |
Feature activation | +5.67e-03 |
Pos logit contributions | |
rals | +0.062 |
heimer | +0.058 |
cars | +0.057 |
Math | +0.054 |
pet | +0.054 |
Neg logit contributions | |
BOX | -0.061 |
icut | -0.060 |
defe | -0.059 |
Fram | -0.058 |
KEN | -0.058 |
Token | like |
Token ID | 588 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
BOX | +0.004 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.004 |
cars | -0.004 |
Math | -0.003 |
pet | -0.003 |
Token | E |
Token ID | 412 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
rals | +0.023 |
heimer | +0.021 |
cars | +0.021 |
Math | +0.019 |
pet | +0.019 |
Neg logit contributions | |
BOX | -0.026 |
icut | -0.026 |
defe | -0.025 |
Fram | -0.025 |
KEN | -0.025 |
Token | vern |
Token ID | 933 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
rals | +0.261 |
heimer | +0.250 |
cars | +0.248 |
pet | +0.240 |
Math | +0.239 |
Neg logit contributions | |
BOX | -0.063 |
icut | -0.060 |
defe | -0.058 |
Fram | -0.055 |
KEN | -0.055 |
Token | ote |
Token ID | 1258 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
BOX | +0.257 |
icut | +0.251 |
defe | +0.247 |
Fram | +0.244 |
KEN | +0.242 |
Neg logit contributions | |
rals | -0.283 |
heimer | -0.264 |
cars | -0.261 |
Math | -0.246 |
pet | -0.246 |
Token | and |
Token ID | 290 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
BOX | +0.011 |
icut | +0.011 |
defe | +0.011 |
Fram | +0.011 |
KEN | +0.011 |
Neg logit contributions | |
rals | -0.011 |
heimer | -0.011 |
cars | -0.011 |
Math | -0.010 |
pet | -0.010 |
Token | Dropbox |
Token ID | 38930 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.007 |
cars | +0.007 |
Math | +0.006 |
pet | +0.006 |
Neg logit contributions | |
BOX | -0.008 |
icut | -0.008 |
defe | -0.008 |
Fram | -0.008 |
KEN | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -7.52e-03 |
Pos logit contributions | |
BOX | +0.011 |
icut | +0.011 |
defe | +0.011 |
Fram | +0.011 |
KEN | +0.011 |
Neg logit contributions | |
rals | -0.011 |
heimer | -0.011 |
cars | -0.010 |
Math | -0.010 |
pet | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
KEN | -0.005 |
Fram | -0.005 |
Token | at |
Token ID | 379 |
Feature activation | +9.84e-03 |
Pos logit contributions | |
BOX | +0.036 |
icut | +0.035 |
defe | +0.035 |
Fram | +0.034 |
KEN | +0.034 |
Neg logit contributions | |
rals | -0.033 |
heimer | -0.031 |
cars | -0.030 |
Math | -0.028 |
pet | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +4.09e-03 |
Pos logit contributions | |
BOX | +0.007 |
icut | +0.007 |
defe | +0.006 |
Fram | +0.006 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.006 |
pet | -0.006 |
Token | corner |
Token ID | 5228 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.003 |
Math | -0.002 |
pet | -0.002 |
Token | of |
Token ID | 286 |
Feature activation | -2.29e-03 |
Pos logit contributions | |
BOX | +0.022 |
icut | +0.022 |
defe | +0.022 |
Fram | +0.021 |
KEN | +0.021 |
Neg logit contributions | |
rals | -0.024 |
heimer | -0.022 |
cars | -0.022 |
pet | -0.021 |
Math | -0.020 |
Token | West |
Token ID | 2688 |
Feature activation | +8.51e-03 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
pet | +0.001 |
Math | +0.001 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | heimer |
Token ID | 16288 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
BOX | +0.008 |
icut | +0.008 |
defe | +0.008 |
Fram | +0.008 |
KEN | +0.008 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.003 |
Math | -0.002 |
pet | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -9.63e-04 |
Pos logit contributions | |
rals | +0.347 |
heimer | +0.324 |
cars | +0.320 |
Math | +0.302 |
pet | +0.301 |
Neg logit contributions | |
BOX | -0.324 |
icut | -0.316 |
defe | -0.311 |
Fram | -0.307 |
KEN | -0.306 |
Token | Mont |
Token ID | 5575 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
pet | +0.001 |
Math | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | rose |
Token ID | 13698 |
Feature activation | -6.11e-02 |
Pos logit contributions | |
rals | +0.140 |
heimer | +0.129 |
cars | +0.126 |
Math | +0.118 |
pet | +0.117 |
Neg logit contributions | |
BOX | -0.193 |
icut | -0.189 |
defe | -0.187 |
Fram | -0.185 |
KEN | -0.184 |
Token | . |
Token ID | 13 |
Feature activation | -9.79e-03 |
Pos logit contributions | |
rals | +0.041 |
heimer | +0.038 |
cars | +0.037 |
Math | +0.035 |
pet | +0.035 |
Neg logit contributions | |
BOX | -0.042 |
icut | -0.041 |
defe | -0.041 |
Fram | -0.040 |
KEN | -0.040 |
Token | I |
Token ID | 314 |
Feature activation | -7.91e-03 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.006 |
cars | +0.006 |
Math | +0.006 |
pet | +0.006 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.006 |
Fram | -0.006 |
KEN | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +2.16e-03 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
pet | +0.001 |
Math | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -3.76e-03 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | 19 |
Token ID | 1129 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
pet | +0.002 |
Math | +0.002 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | 15 |
Token ID | 1314 |
Feature activation | -3.71e-03 |
Pos logit contributions | |
rals | +0.027 |
heimer | +0.025 |
cars | +0.025 |
Math | +0.023 |
pet | +0.023 |
Neg logit contributions | |
BOX | -0.033 |
icut | -0.032 |
defe | -0.031 |
Fram | -0.031 |
KEN | -0.031 |
Token | West |
Token ID | 2688 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.002 |
KEN | -0.002 |
Token | heimer |
Token ID | 16288 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
BOX | +0.010 |
icut | +0.009 |
defe | +0.009 |
Fram | +0.009 |
KEN | +0.009 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.007 |
cars | -0.007 |
Math | -0.006 |
pet | -0.006 |
Token | Today |
Token ID | 6288 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
rals | +0.327 |
heimer | +0.309 |
cars | +0.302 |
Math | +0.285 |
pet | +0.284 |
Neg logit contributions | |
BOX | -0.334 |
icut | -0.326 |
defe | -0.325 |
Fram | -0.319 |
KEN | -0.318 |
Token | : |
Token ID | 25 |
Feature activation | -7.93e-03 |
Pos logit contributions | |
rals | +0.014 |
heimer | +0.013 |
cars | +0.013 |
Math | +0.012 |
pet | +0.012 |
Neg logit contributions | |
BOX | -0.015 |
icut | -0.015 |
defe | -0.014 |
Fram | -0.014 |
KEN | -0.014 |
Token | Fir |
Token ID | 36425 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | kin |
Token ID | 5116 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
rals | +0.107 |
heimer | +0.100 |
cars | +0.098 |
pet | +0.092 |
Math | +0.092 |
Neg logit contributions | |
BOX | -0.117 |
icut | -0.114 |
defe | -0.113 |
Fram | -0.112 |
KEN | -0.111 |
Token | and |
Token ID | 290 |
Feature activation | -7.19e-03 |
Pos logit contributions | |
rals | +0.132 |
heimer | +0.124 |
cars | +0.121 |
Math | +0.114 |
pet | +0.114 |
Neg logit contributions | |
BOX | -0.133 |
icut | -0.130 |
defe | -0.128 |
Fram | -0.126 |
KEN | -0.126 |
Token | A |
Token ID | 32 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.005 |
cars | +0.005 |
pet | +0.004 |
Math | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | video |
Token ID | 2008 |
Feature activation | +4.91e-02 |
Pos logit contributions | |
BOX | +0.014 |
icut | +0.014 |
defe | +0.013 |
Fram | +0.013 |
KEN | +0.013 |
Neg logit contributions | |
rals | -0.014 |
heimer | -0.013 |
cars | -0.013 |
Math | -0.012 |
pet | -0.012 |
Token | posted |
Token ID | 4481 |
Feature activation | +6.64e-03 |
Pos logit contributions | |
BOX | +0.034 |
icut | +0.033 |
defe | +0.033 |
Fram | +0.033 |
KEN | +0.032 |
Neg logit contributions | |
rals | -0.032 |
heimer | -0.030 |
cars | -0.030 |
Math | -0.028 |
pet | -0.028 |
Token | by |
Token ID | 416 |
Feature activation | +9.17e-04 |
Pos logit contributions | |
BOX | +0.004 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.004 |
cars | -0.004 |
Math | -0.004 |
pet | -0.004 |
Token | Robert |
Token ID | 5199 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | Math |
Token ID | 16320 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
BOX | +0.034 |
icut | +0.033 |
defe | +0.033 |
Fram | +0.032 |
KEN | +0.032 |
Neg logit contributions | |
rals | -0.034 |
heimer | -0.031 |
cars | -0.031 |
Math | -0.029 |
pet | -0.029 |
Token | is |
Token ID | 271 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
rals | +0.333 |
heimer | +0.313 |
cars | +0.308 |
Math | +0.291 |
pet | +0.291 |
Neg logit contributions | |
BOX | -0.296 |
defe | -0.286 |
icut | -0.285 |
Fram | -0.282 |
KEN | -0.281 |
Token | (@ |
Token ID | 4275 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
BOX | +0.071 |
icut | +0.069 |
defe | +0.068 |
Fram | +0.068 |
KEN | +0.067 |
Neg logit contributions | |
rals | -0.068 |
heimer | -0.063 |
cars | -0.062 |
Math | -0.059 |
pet | -0.058 |
Token | r |
Token ID | 81 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
rals | +0.012 |
heimer | +0.011 |
cars | +0.011 |
Math | +0.010 |
pet | +0.010 |
Neg logit contributions | |
BOX | -0.015 |
icut | -0.014 |
defe | -0.014 |
Fram | -0.014 |
KEN | -0.014 |
Token | math |
Token ID | 11018 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
rals | +0.144 |
heimer | +0.132 |
cars | +0.130 |
Math | +0.120 |
pet | +0.120 |
Neg logit contributions | |
BOX | -0.194 |
icut | -0.190 |
defe | -0.187 |
Fram | -0.186 |
KEN | -0.185 |
Token | is |
Token ID | 271 |
Feature activation | +9.41e-02 |
Pos logit contributions | |
rals | +0.220 |
heimer | +0.208 |
cars | +0.205 |
Math | +0.195 |
pet | +0.195 |
Neg logit contributions | |
BOX | -0.143 |
icut | -0.139 |
defe | -0.137 |
Fram | -0.134 |
KEN | -0.134 |
Token | 18 |
Token ID | 1507 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
pet | +0.001 |
Math | +0.001 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
KEN | -0.002 |
Fram | -0.002 |
Token | campaign |
Token ID | 1923 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
rals | +0.015 |
heimer | +0.014 |
cars | +0.014 |
Math | +0.013 |
pet | +0.013 |
Neg logit contributions | |
BOX | -0.014 |
icut | -0.014 |
defe | -0.013 |
Fram | -0.013 |
KEN | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
rals | +0.013 |
heimer | +0.012 |
cars | +0.012 |
Math | +0.011 |
pet | +0.011 |
Neg logit contributions | |
BOX | -0.013 |
icut | -0.013 |
defe | -0.012 |
Fram | -0.012 |
KEN | -0.012 |
Token | on |
Token ID | 319 |
Feature activation | +5.10e-03 |
Pos logit contributions | |
rals | +0.008 |
heimer | +0.007 |
cars | +0.007 |
Math | +0.007 |
pet | +0.007 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.007 |
Fram | -0.007 |
KEN | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.003 |
cars | -0.003 |
pet | -0.003 |
Math | -0.003 |
Token | Bagg |
Token ID | 49329 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
rals | +0.008 |
heimer | +0.007 |
cars | +0.007 |
pet | +0.007 |
Math | +0.007 |
Neg logit contributions | |
BOX | -0.008 |
icut | -0.008 |
defe | -0.008 |
Fram | -0.008 |
KEN | -0.008 |
Token | ies |
Token ID | 444 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
rals | +0.205 |
heimer | +0.185 |
cars | +0.182 |
pet | +0.167 |
Math | +0.166 |
Neg logit contributions | |
BOX | -0.380 |
icut | -0.375 |
defe | -0.371 |
KEN | -0.366 |
Fram | -0.366 |
Token | ' |
Token ID | 6 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
rals | +0.101 |
heimer | +0.093 |
cars | +0.091 |
pet | +0.085 |
Math | +0.085 |
Neg logit contributions | |
BOX | -0.118 |
icut | -0.115 |
defe | -0.114 |
Fram | -0.113 |
KEN | -0.112 |
Token | training |
Token ID | 3047 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
rals | +0.085 |
heimer | +0.079 |
cars | +0.078 |
pet | +0.073 |
Math | +0.073 |
Neg logit contributions | |
BOX | -0.086 |
icut | -0.084 |
defe | -0.083 |
Fram | -0.082 |
KEN | -0.082 |
Token | wear |
Token ID | 13927 |
Feature activation | -9.07e-02 |
Pos logit contributions | |
BOX | +0.011 |
icut | +0.011 |
defe | +0.011 |
Fram | +0.011 |
KEN | +0.011 |
Neg logit contributions | |
rals | -0.009 |
heimer | -0.008 |
cars | -0.008 |
pet | -0.007 |
Math | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
rals | +0.063 |
heimer | +0.059 |
cars | +0.057 |
Math | +0.054 |
pet | +0.054 |
Neg logit contributions | |
BOX | -0.060 |
icut | -0.059 |
defe | -0.058 |
Fram | -0.057 |
KEN | -0.057 |
Token | prevent |
Token ID | 2948 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
BOX | +0.004 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.004 |
cars | -0.004 |
Math | -0.004 |
pet | -0.004 |
Token | Ibrahim |
Token ID | 30917 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
BOX | +0.098 |
icut | +0.095 |
defe | +0.094 |
Fram | +0.093 |
KEN | +0.093 |
Neg logit contributions | |
rals | -0.100 |
heimer | -0.093 |
cars | -0.092 |
Math | -0.087 |
pet | -0.086 |
Token | 's |
Token ID | 338 |
Feature activation | -4.94e-03 |
Pos logit contributions | |
BOX | +0.025 |
icut | +0.024 |
defe | +0.024 |
Fram | +0.023 |
KEN | +0.023 |
Neg logit contributions | |
rals | -0.024 |
heimer | -0.022 |
cars | -0.022 |
Math | -0.021 |
pet | -0.021 |
Token | daughter |
Token ID | 4957 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -3.02e-04 |
Pos logit contributions | |
BOX | +0.054 |
icut | +0.052 |
defe | +0.052 |
Fram | +0.051 |
KEN | +0.051 |
Neg logit contributions | |
rals | -0.053 |
heimer | -0.049 |
cars | -0.049 |
Math | -0.046 |
pet | -0.046 |
Token | Rai |
Token ID | 42954 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
rals | +0.000 |
heimer | +0.000 |
cars | +0.000 |
Math | +0.000 |
pet | +0.000 |
Neg logit contributions | |
BOX | -0.000 |
icut | -0.000 |
defe | -0.000 |
Fram | -0.000 |
KEN | -0.000 |
Token | han |
Token ID | 7637 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
rals | +0.303 |
heimer | +0.283 |
cars | +0.279 |
Math | +0.263 |
pet | +0.263 |
Neg logit contributions | |
BOX | -0.283 |
icut | -0.276 |
defe | -0.272 |
Fram | -0.269 |
KEN | -0.267 |
Token | Must |
Token ID | 12039 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
BOX | +0.018 |
icut | +0.017 |
defe | +0.017 |
Fram | +0.017 |
KEN | +0.017 |
Neg logit contributions | |
rals | -0.020 |
heimer | -0.019 |
cars | -0.018 |
Math | -0.017 |
pet | -0.017 |
Token | afa |
Token ID | 28485 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
BOX | +0.009 |
icut | +0.009 |
defe | +0.009 |
Fram | +0.009 |
KEN | +0.009 |
Neg logit contributions | |
rals | -0.015 |
heimer | -0.014 |
cars | -0.014 |
Math | -0.013 |
pet | -0.013 |
Token | Kam |
Token ID | 12670 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
BOX | +0.113 |
icut | +0.110 |
defe | +0.108 |
Fram | +0.107 |
KEN | +0.107 |
Neg logit contributions | |
rals | -0.118 |
heimer | -0.110 |
cars | -0.108 |
pet | -0.102 |
Math | -0.102 |
Token | al |
Token ID | 282 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
BOX | +0.217 |
icut | +0.212 |
defe | +0.210 |
Fram | +0.208 |
KEN | +0.207 |
Neg logit contributions | |
rals | -0.171 |
heimer | -0.158 |
cars | -0.155 |
Math | -0.145 |
pet | -0.144 |
Token | re |
Token ID | 260 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.007 |
cars | +0.007 |
pet | +0.006 |
Math | +0.006 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.007 |
Fram | -0.007 |
KEN | -0.007 |
Token | excited |
Token ID | 6568 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
rals | +0.016 |
heimer | +0.015 |
cars | +0.014 |
Math | +0.014 |
pet | +0.014 |
Neg logit contributions | |
BOX | -0.016 |
icut | -0.016 |
defe | -0.015 |
Fram | -0.015 |
KEN | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
rals | +0.048 |
heimer | +0.044 |
cars | +0.044 |
Math | +0.041 |
pet | +0.041 |
Neg logit contributions | |
BOX | -0.047 |
icut | -0.046 |
defe | -0.046 |
Fram | -0.045 |
KEN | -0.045 |
Token | announce |
Token ID | 5453 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
rals | +0.009 |
heimer | +0.008 |
cars | +0.008 |
Math | +0.007 |
pet | +0.007 |
Neg logit contributions | |
BOX | -0.011 |
icut | -0.011 |
defe | -0.011 |
Fram | -0.011 |
KEN | -0.010 |
Token | Shop |
Token ID | 13705 |
Feature activation | -7.99e-02 |
Pos logit contributions | |
rals | +0.034 |
heimer | +0.031 |
cars | +0.031 |
Math | +0.029 |
pet | +0.029 |
Neg logit contributions | |
BOX | -0.033 |
icut | -0.032 |
defe | -0.031 |
Fram | -0.031 |
KEN | -0.031 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
rals | +0.036 |
heimer | +0.032 |
cars | +0.031 |
Math | +0.028 |
pet | +0.028 |
Neg logit contributions | |
BOX | -0.073 |
icut | -0.072 |
defe | -0.071 |
Fram | -0.070 |
KEN | -0.070 |
Token | POS |
Token ID | 28069 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
rals | +0.287 |
heimer | +0.267 |
cars | +0.263 |
Math | +0.247 |
pet | +0.247 |
Neg logit contributions | |
BOX | -0.295 |
icut | -0.289 |
defe | -0.285 |
Fram | -0.281 |
KEN | -0.280 |
Token | , |
Token ID | 11 |
Feature activation | -7.31e-03 |
Pos logit contributions | |
rals | +0.097 |
heimer | +0.090 |
cars | +0.088 |
Math | +0.084 |
pet | +0.083 |
Neg logit contributions | |
BOX | -0.099 |
icut | -0.097 |
defe | -0.096 |
Fram | -0.095 |
KEN | -0.094 |
Token | a |
Token ID | 257 |
Feature activation | -9.16e-03 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | brand |
Token ID | 4508 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.006 |
cars | +0.006 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.006 |
icut | -0.006 |
defe | -0.006 |
Fram | -0.006 |
KEN | -0.006 |
Token | new |
Token ID | 649 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
BOX | +0.108 |
icut | +0.105 |
defe | +0.104 |
Fram | +0.103 |
KEN | +0.103 |
Neg logit contributions | |
rals | -0.090 |
heimer | -0.083 |
cars | -0.082 |
Math | -0.077 |
pet | -0.077 |
Token | Rohingya |
Token ID | 34016 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
BOX | +0.015 |
icut | +0.015 |
defe | +0.015 |
Fram | +0.014 |
KEN | +0.014 |
Neg logit contributions | |
rals | -0.010 |
heimer | -0.009 |
cars | -0.009 |
Math | -0.009 |
pet | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -7.96e-03 |
Pos logit contributions | |
rals | +0.160 |
heimer | +0.149 |
cars | +0.147 |
Math | +0.138 |
pet | +0.138 |
Neg logit contributions | |
BOX | -0.169 |
icut | -0.166 |
defe | -0.163 |
Fram | -0.162 |
KEN | -0.161 |
Token |
|
Token ID | 198 |
Feature activation | -7.91e-03 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -4.48e-04 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | Fort |
Token ID | 21926 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
rals | +0.000 |
heimer | +0.000 |
cars | +0.000 |
pet | +0.000 |
Math | +0.000 |
Neg logit contributions | |
BOX | -0.000 |
icut | -0.000 |
defe | -0.000 |
Fram | -0.000 |
KEN | -0.000 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
rals | +0.096 |
heimer | +0.089 |
cars | +0.088 |
Math | +0.083 |
pet | +0.083 |
Neg logit contributions | |
BOX | -0.089 |
icut | -0.087 |
defe | -0.085 |
Fram | -0.084 |
KEN | -0.084 |
Token | Rights |
Token ID | 6923 |
Feature activation | +8.48e-02 |
Pos logit contributions | |
rals | +0.284 |
heimer | +0.263 |
cars | +0.259 |
Math | +0.246 |
pet | +0.246 |
Neg logit contributions | |
BOX | -0.296 |
icut | -0.290 |
defe | -0.286 |
KEN | -0.282 |
Fram | -0.282 |
Token | said |
Token ID | 531 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
BOX | +0.057 |
icut | +0.056 |
defe | +0.055 |
Fram | +0.055 |
KEN | +0.054 |
Neg logit contributions | |
rals | -0.058 |
heimer | -0.054 |
cars | -0.053 |
Math | -0.050 |
pet | -0.050 |
Token | it |
Token ID | 340 |
Feature activation | -5.51e-03 |
Pos logit contributions | |
BOX | +0.008 |
icut | +0.008 |
defe | +0.008 |
Fram | +0.008 |
KEN | +0.007 |
Neg logit contributions | |
rals | -0.008 |
heimer | -0.007 |
cars | -0.007 |
Math | -0.007 |
pet | -0.007 |
Token | had |
Token ID | 550 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.004 |
icut | -0.004 |
defe | -0.004 |
KEN | -0.004 |
Fram | -0.004 |
Token | obtained |
Token ID | 6492 |
Feature activation | +6.24e-02 |
Pos logit contributions | |
rals | +0.011 |
heimer | +0.010 |
cars | +0.010 |
Math | +0.009 |
pet | +0.009 |
Neg logit contributions | |
BOX | -0.010 |
icut | -0.010 |
defe | -0.010 |
Fram | -0.010 |
KEN | -0.010 |
Token | retail |
Token ID | 6308 |
Feature activation | +8.17e-02 |
Pos logit contributions | |
BOX | +0.019 |
icut | +0.019 |
defe | +0.018 |
Fram | +0.018 |
KEN | +0.018 |
Neg logit contributions | |
rals | -0.021 |
heimer | -0.019 |
cars | -0.019 |
Math | -0.018 |
pet | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -7.03e-03 |
Pos logit contributions | |
BOX | +0.056 |
icut | +0.054 |
defe | +0.054 |
Fram | +0.053 |
KEN | +0.053 |
Neg logit contributions | |
rals | -0.055 |
heimer | -0.051 |
cars | -0.051 |
Math | -0.048 |
pet | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -7.15e-04 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.004 |
cars | +0.004 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -4.66e-03 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.000 |
cars | +0.000 |
Math | +0.000 |
pet | +0.000 |
Neg logit contributions | |
BOX | -0.000 |
icut | -0.000 |
defe | -0.000 |
Fram | -0.000 |
KEN | -0.000 |
Token | Shop |
Token ID | 29917 |
Feature activation | -7.18e-03 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.003 |
cars | +0.003 |
pet | +0.003 |
Math | +0.003 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.008 |
icut | -0.008 |
defe | -0.008 |
Fram | -0.008 |
KEN | -0.008 |
Token | POS |
Token ID | 28069 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
rals | +0.227 |
heimer | +0.207 |
cars | +0.203 |
Math | +0.188 |
pet | +0.188 |
Neg logit contributions | |
BOX | -0.350 |
icut | -0.344 |
defe | -0.340 |
Fram | -0.336 |
KEN | -0.335 |
Token | delivers |
Token ID | 16316 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
rals | +0.097 |
heimer | +0.090 |
cars | +0.089 |
Math | +0.084 |
pet | +0.084 |
Neg logit contributions | |
BOX | -0.097 |
icut | -0.095 |
defe | -0.093 |
Fram | -0.092 |
KEN | -0.092 |
Token | in |
Token ID | 287 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
BOX | +0.056 |
icut | +0.054 |
defe | +0.054 |
Fram | +0.053 |
KEN | +0.053 |
Neg logit contributions | |
rals | -0.056 |
heimer | -0.052 |
cars | -0.052 |
Math | -0.049 |
pet | -0.049 |
Token | - |
Token ID | 12 |
Feature activation | -6.85e-03 |
Pos logit contributions | |
rals | +0.011 |
heimer | +0.010 |
cars | +0.010 |
Math | +0.009 |
pet | +0.009 |
Neg logit contributions | |
BOX | -0.011 |
icut | -0.011 |
defe | -0.010 |
Fram | -0.010 |
KEN | -0.010 |
Token | store |
Token ID | 8095 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.003 |
cars | +0.003 |
pet | +0.003 |
Math | +0.003 |
Neg logit contributions | |
BOX | -0.006 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | optics |
Token ID | 36237 |
Feature activation | +6.78e-02 |
Pos logit contributions | |
BOX | +0.012 |
icut | +0.011 |
defe | +0.011 |
KEN | +0.011 |
Fram | +0.011 |
Neg logit contributions | |
rals | -0.011 |
heimer | -0.011 |
cars | -0.011 |
Math | -0.010 |
pet | -0.010 |
Token | are |
Token ID | 389 |
Feature activation | -6.72e-03 |
Pos logit contributions | |
BOX | +0.046 |
icut | +0.045 |
defe | +0.044 |
Fram | +0.044 |
KEN | +0.044 |
Neg logit contributions | |
rals | -0.046 |
heimer | -0.043 |
cars | -0.042 |
Math | -0.040 |
pet | -0.039 |
Token | used |
Token ID | 973 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.004 |
cars | +0.004 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.004 |
defe | -0.004 |
KEN | -0.004 |
Fram | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -3.76e-03 |
Pos logit contributions | |
rals | +0.015 |
heimer | +0.014 |
cars | +0.014 |
Math | +0.013 |
pet | +0.013 |
Neg logit contributions | |
BOX | -0.015 |
icut | -0.014 |
defe | -0.014 |
Fram | -0.014 |
KEN | -0.014 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
rals | +0.032 |
heimer | +0.024 |
cars | +0.022 |
Math | +0.016 |
pet | +0.015 |
Neg logit contributions | |
BOX | -0.207 |
icut | -0.204 |
defe | -0.203 |
Fram | -0.201 |
KEN | -0.201 |
Token | the |
Token ID | 262 |
Feature activation | -3.97e-05 |
Pos logit contributions | |
rals | +0.288 |
heimer | +0.267 |
cars | +0.265 |
Math | +0.252 |
pet | +0.248 |
Neg logit contributions | |
BOX | -0.286 |
icut | -0.280 |
defe | -0.277 |
Fram | -0.274 |
KEN | -0.274 |
Token | image |
Token ID | 2939 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
rals | +0.000 |
heimer | +0.000 |
cars | +0.000 |
Math | +0.000 |
pet | +0.000 |
Neg logit contributions | |
BOX | -0.000 |
icut | -0.000 |
defe | -0.000 |
Fram | -0.000 |
KEN | -0.000 |
Token | though |
Token ID | 996 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
rals | +0.018 |
heimer | +0.016 |
cars | +0.016 |
Math | +0.015 |
pet | +0.015 |
Neg logit contributions | |
BOX | -0.018 |
icut | -0.017 |
defe | -0.017 |
KEN | -0.017 |
Fram | -0.017 |
Token | if |
Token ID | 611 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
rals | +0.024 |
heimer | +0.022 |
cars | +0.022 |
Math | +0.021 |
pet | +0.021 |
Neg logit contributions | |
BOX | -0.024 |
icut | -0.024 |
defe | -0.023 |
KEN | -0.023 |
Fram | -0.023 |
Token | you |
Token ID | 345 |
Feature activation | -2.52e-03 |
Pos logit contributions | |
rals | +0.015 |
heimer | +0.014 |
cars | +0.014 |
Math | +0.013 |
pet | +0.013 |
Neg logit contributions | |
BOX | -0.016 |
icut | -0.015 |
defe | -0.015 |
KEN | -0.015 |
Fram | -0.015 |
Token | cause |
Token ID | 2728 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | heavy |
Token ID | 4334 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
rals | +0.023 |
heimer | +0.022 |
cars | +0.021 |
Math | +0.020 |
pet | +0.020 |
Neg logit contributions | |
BOX | -0.025 |
icut | -0.024 |
defe | -0.024 |
Fram | -0.024 |
KEN | -0.023 |
Token | damage |
Token ID | 2465 |
Feature activation | -9.60e-02 |
Pos logit contributions | |
rals | +0.027 |
heimer | +0.025 |
cars | +0.025 |
Math | +0.023 |
pet | +0.023 |
Neg logit contributions | |
BOX | -0.033 |
icut | -0.033 |
defe | -0.032 |
Fram | -0.032 |
KEN | -0.032 |
Token | or |
Token ID | 393 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
rals | +0.065 |
heimer | +0.060 |
cars | +0.059 |
Math | +0.056 |
pet | +0.055 |
Neg logit contributions | |
BOX | -0.066 |
icut | -0.064 |
defe | -0.064 |
KEN | -0.063 |
Fram | -0.063 |
Token | null |
Token ID | 9242 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
rals | +0.020 |
heimer | +0.019 |
cars | +0.018 |
Math | +0.017 |
pet | +0.017 |
Neg logit contributions | |
BOX | -0.021 |
icut | -0.020 |
defe | -0.020 |
Fram | -0.020 |
KEN | -0.020 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
rals | +0.027 |
heimer | +0.020 |
cars | +0.019 |
Math | +0.015 |
pet | +0.015 |
Neg logit contributions | |
BOX | -0.158 |
icut | -0.155 |
defe | -0.155 |
Fram | -0.154 |
KEN | -0.154 |
Token | enemies |
Token ID | 5775 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
rals | +0.276 |
heimer | +0.255 |
cars | +0.254 |
Math | +0.242 |
pet | +0.237 |
Neg logit contributions | |
BOX | -0.291 |
icut | -0.284 |
Fram | -0.279 |
KEN | -0.279 |
defe | -0.278 |
Token | � |
Token ID | 447 |
Feature activation | -9.21e-04 |
Pos logit contributions | |
BOX | +0.031 |
icut | +0.031 |
defe | +0.030 |
Fram | +0.030 |
KEN | +0.030 |
Neg logit contributions | |
rals | -0.025 |
heimer | -0.023 |
cars | -0.022 |
Math | -0.021 |
pet | -0.021 |
Token | � |
Token ID | 247 |
Feature activation | +1.33e-03 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | powers |
Token ID | 5635 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
KEN | +0.001 |
Fram | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +3.65e-03 |
Pos logit contributions | |
BOX | +0.040 |
icut | +0.039 |
defe | +0.039 |
Fram | +0.038 |
KEN | +0.038 |
Neg logit contributions | |
rals | -0.040 |
heimer | -0.037 |
cars | -0.036 |
pet | -0.034 |
Math | -0.034 |
Token | cache |
Token ID | 12940 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | your |
Token ID | 534 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
BOX | +0.085 |
icut | +0.082 |
defe | +0.081 |
Fram | +0.081 |
KEN | +0.080 |
Neg logit contributions | |
rals | -0.093 |
heimer | -0.087 |
cars | -0.086 |
Math | -0.081 |
pet | -0.081 |
Token | food |
Token ID | 2057 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
BOX | +0.035 |
icut | +0.034 |
defe | +0.034 |
Fram | +0.034 |
KEN | +0.033 |
Neg logit contributions | |
rals | -0.037 |
heimer | -0.035 |
cars | -0.034 |
Math | -0.032 |
pet | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | +7.38e-03 |
Pos logit contributions | |
BOX | +0.026 |
icut | +0.025 |
defe | +0.025 |
Fram | +0.025 |
KEN | +0.025 |
Neg logit contributions | |
rals | -0.027 |
heimer | -0.025 |
cars | -0.024 |
pet | -0.023 |
Math | -0.023 |
Token | divers |
Token ID | 15070 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
BOX | +0.005 |
icut | +0.005 |
defe | +0.005 |
KEN | +0.005 |
Fram | +0.005 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.005 |
cars | -0.005 |
Math | -0.004 |
pet | -0.004 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
rals | +0.036 |
heimer | +0.027 |
cars | +0.025 |
Math | +0.018 |
pet | +0.018 |
Neg logit contributions | |
BOX | -0.224 |
icut | -0.221 |
defe | -0.220 |
Fram | -0.218 |
KEN | -0.218 |
Token | your |
Token ID | 534 |
Feature activation | +4.88e-02 |
Pos logit contributions | |
rals | +0.307 |
heimer | +0.287 |
cars | +0.282 |
Math | +0.268 |
pet | +0.267 |
Neg logit contributions | |
BOX | -0.260 |
icut | -0.253 |
defe | -0.248 |
Fram | -0.247 |
KEN | -0.246 |
Token | crops |
Token ID | 14450 |
Feature activation | +7.64e-02 |
Pos logit contributions | |
BOX | +0.033 |
icut | +0.032 |
defe | +0.032 |
Fram | +0.032 |
KEN | +0.032 |
Neg logit contributions | |
rals | -0.033 |
heimer | -0.031 |
cars | -0.030 |
Math | -0.029 |
pet | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -4.79e-03 |
Pos logit contributions | |
BOX | +0.052 |
icut | +0.050 |
defe | +0.050 |
Fram | +0.049 |
KEN | +0.049 |
Neg logit contributions | |
rals | -0.052 |
heimer | -0.048 |
cars | -0.048 |
pet | -0.045 |
Math | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +9.92e-03 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | provide |
Token ID | 2148 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
BOX | +0.007 |
icut | +0.006 |
defe | +0.006 |
Fram | +0.006 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.006 |
pet | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -3.09e-04 |
Pos logit contributions | |
rals | +0.110 |
heimer | +0.102 |
cars | +0.101 |
Math | +0.095 |
pet | +0.095 |
Neg logit contributions | |
BOX | -0.111 |
icut | -0.108 |
defe | -0.107 |
Fram | -0.105 |
KEN | -0.105 |
Token |
|
Token ID | 198 |
Feature activation | -9.14e-04 |
Pos logit contributions | |
rals | +0.000 |
heimer | +0.000 |
cars | +0.000 |
Math | +0.000 |
pet | +0.000 |
Neg logit contributions | |
BOX | -0.000 |
icut | -0.000 |
defe | -0.000 |
Fram | -0.000 |
KEN | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +5.80e-03 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
pet | +0.001 |
Math | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | With |
Token ID | 3152 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
BOX | +0.004 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.004 |
cars | -0.004 |
pet | -0.003 |
Math | -0.003 |
Token | Shop |
Token ID | 13705 |
Feature activation | -6.62e-02 |
Pos logit contributions | |
BOX | +0.021 |
icut | +0.020 |
defe | +0.020 |
Fram | +0.020 |
KEN | +0.020 |
Neg logit contributions | |
rals | -0.018 |
heimer | -0.016 |
cars | -0.016 |
Math | -0.015 |
pet | -0.015 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
rals | +0.013 |
heimer | +0.010 |
cars | +0.009 |
Math | +0.007 |
pet | +0.007 |
Neg logit contributions | |
BOX | -0.077 |
icut | -0.076 |
defe | -0.075 |
Fram | -0.075 |
KEN | -0.075 |
Token | Payments |
Token ID | 41318 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
rals | +0.322 |
heimer | +0.303 |
cars | +0.299 |
pet | +0.283 |
Math | +0.283 |
Neg logit contributions | |
BOX | -0.243 |
icut | -0.236 |
defe | -0.233 |
Fram | -0.230 |
KEN | -0.229 |
Token | you |
Token ID | 345 |
Feature activation | -2.95e-03 |
Pos logit contributions | |
BOX | +0.050 |
icut | +0.049 |
defe | +0.048 |
Fram | +0.048 |
KEN | +0.048 |
Neg logit contributions | |
rals | -0.052 |
heimer | -0.049 |
cars | -0.048 |
pet | -0.045 |
Math | -0.045 |
Token | get |
Token ID | 651 |
Feature activation | +4.31e-03 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | low |
Token ID | 1877 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.002 |
Math | -0.002 |
pet | -0.002 |
Token | credit |
Token ID | 3884 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
rals | +0.028 |
heimer | +0.026 |
cars | +0.025 |
Math | +0.024 |
pet | +0.024 |
Neg logit contributions | |
BOX | -0.024 |
icut | -0.024 |
defe | -0.023 |
Fram | -0.023 |
KEN | -0.023 |
Token | arrest |
Token ID | 3251 |
Feature activation | -7.42e-02 |
Pos logit contributions | |
rals | +0.055 |
heimer | +0.051 |
cars | +0.050 |
Math | +0.047 |
pet | +0.047 |
Neg logit contributions | |
BOX | -0.071 |
icut | -0.069 |
defe | -0.068 |
Fram | -0.068 |
KEN | -0.068 |
Token | warrant |
Token ID | 8084 |
Feature activation | +8.29e-02 |
Pos logit contributions | |
rals | +0.059 |
heimer | +0.056 |
cars | +0.055 |
Math | +0.052 |
pet | +0.052 |
Neg logit contributions | |
BOX | -0.042 |
icut | -0.040 |
defe | -0.040 |
Fram | -0.039 |
KEN | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
BOX | +0.057 |
icut | +0.055 |
defe | +0.055 |
Fram | +0.054 |
KEN | +0.054 |
Neg logit contributions | |
rals | -0.056 |
heimer | -0.052 |
cars | -0.051 |
Math | -0.048 |
pet | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -7.46e-03 |
Pos logit contributions | |
rals | +0.014 |
heimer | +0.013 |
cars | +0.013 |
pet | +0.012 |
Math | +0.012 |
Neg logit contributions | |
BOX | -0.012 |
icut | -0.012 |
defe | -0.012 |
Fram | -0.012 |
KEN | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | Wiki |
Token ID | 32603 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
rals | +0.008 |
heimer | +0.007 |
cars | +0.007 |
pet | +0.007 |
Math | +0.007 |
Neg logit contributions | |
BOX | -0.009 |
icut | -0.008 |
defe | -0.008 |
Fram | -0.008 |
KEN | -0.008 |
Token | Leaks |
Token ID | 17874 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
rals | +0.132 |
heimer | +0.115 |
cars | +0.109 |
pet | +0.095 |
Math | +0.094 |
Neg logit contributions | |
BOX | -0.430 |
icut | -0.425 |
defe | -0.423 |
Fram | -0.419 |
KEN | -0.417 |
Token | pr |
Token ID | 778 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
rals | +0.190 |
heimer | +0.178 |
cars | +0.175 |
Math | +0.165 |
pet | +0.165 |
Neg logit contributions | |
BOX | -0.186 |
icut | -0.182 |
defe | -0.179 |
Fram | -0.177 |
KEN | -0.176 |
Token | ides |
Token ID | 1460 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
BOX | +0.045 |
icut | +0.044 |
defe | +0.044 |
Fram | +0.044 |
KEN | +0.043 |
Neg logit contributions | |
rals | -0.017 |
heimer | -0.015 |
cars | -0.014 |
Math | -0.013 |
pet | -0.013 |
Token | itself |
Token ID | 2346 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
rals | +0.119 |
heimer | +0.109 |
cars | +0.107 |
Math | +0.099 |
pet | +0.099 |
Neg logit contributions | |
BOX | -0.180 |
icut | -0.176 |
defe | -0.174 |
Fram | -0.173 |
KEN | -0.172 |
Token | as |
Token ID | 355 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
rals | +0.096 |
heimer | +0.090 |
cars | +0.089 |
Math | +0.084 |
pet | +0.084 |
Neg logit contributions | |
BOX | -0.084 |
icut | -0.082 |
defe | -0.081 |
Fram | -0.080 |
KEN | -0.079 |
Token | start |
Token ID | 923 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.002 |
Math | -0.002 |
pet | -0.002 |
Token | selling |
Token ID | 6301 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
BOX | +0.074 |
icut | +0.073 |
defe | +0.072 |
Fram | +0.071 |
KEN | +0.071 |
Neg logit contributions | |
rals | -0.074 |
heimer | -0.069 |
cars | -0.068 |
Math | -0.064 |
pet | -0.064 |
Token | more |
Token ID | 517 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
BOX | +0.028 |
icut | +0.027 |
defe | +0.027 |
Fram | +0.026 |
KEN | +0.026 |
Neg logit contributions | |
rals | -0.025 |
heimer | -0.023 |
cars | -0.023 |
Math | -0.022 |
pet | -0.021 |
Token | with |
Token ID | 351 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
rals | +0.012 |
heimer | +0.011 |
cars | +0.011 |
Math | +0.011 |
pet | +0.010 |
Neg logit contributions | |
BOX | -0.014 |
icut | -0.014 |
defe | -0.013 |
Fram | -0.013 |
KEN | -0.013 |
Token | Shop |
Token ID | 13705 |
Feature activation | -8.13e-02 |
Pos logit contributions | |
BOX | +0.011 |
icut | +0.011 |
defe | +0.011 |
Fram | +0.011 |
KEN | +0.011 |
Neg logit contributions | |
rals | -0.009 |
heimer | -0.008 |
cars | -0.008 |
Math | -0.007 |
pet | -0.007 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
rals | +0.016 |
heimer | +0.012 |
cars | +0.011 |
Math | +0.008 |
pet | +0.008 |
Neg logit contributions | |
BOX | -0.095 |
icut | -0.093 |
defe | -0.092 |
Fram | -0.092 |
KEN | -0.092 |
Token | POS |
Token ID | 28069 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
rals | +0.224 |
heimer | +0.205 |
cars | +0.201 |
Math | +0.185 |
pet | +0.185 |
Neg logit contributions | |
BOX | -0.339 |
icut | -0.332 |
defe | -0.328 |
Fram | -0.325 |
KEN | -0.324 |
Token | . |
Token ID | 13 |
Feature activation | -5.03e-03 |
Pos logit contributions | |
rals | +0.094 |
heimer | +0.087 |
cars | +0.086 |
Math | +0.081 |
pet | +0.081 |
Neg logit contributions | |
BOX | -0.096 |
icut | -0.094 |
defe | -0.093 |
Fram | -0.092 |
KEN | -0.091 |
Token |
|
Token ID | 198 |
Feature activation | -2.99e-03 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -6.67e-03 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | Man |
Token ID | 5124 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.004 |
cars | +0.004 |
pet | +0.004 |
Math | +0.004 |
Neg logit contributions | |
BOX | -0.004 |
icut | -0.004 |
defe | -0.004 |
Fram | -0.004 |
KEN | -0.004 |
Token | program |
Token ID | 1430 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
BOX | +0.025 |
icut | +0.024 |
defe | +0.024 |
Fram | +0.024 |
KEN | +0.024 |
Neg logit contributions | |
rals | -0.023 |
heimer | -0.021 |
cars | -0.021 |
Math | -0.020 |
pet | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | -9.30e-03 |
Pos logit contributions | |
rals | +0.012 |
heimer | +0.011 |
cars | +0.011 |
Math | +0.010 |
pet | +0.010 |
Neg logit contributions | |
BOX | -0.012 |
icut | -0.011 |
defe | -0.011 |
Fram | -0.011 |
KEN | -0.011 |
Token | invis |
Token ID | 44837 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.006 |
cars | +0.006 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.006 |
icut | -0.006 |
defe | -0.006 |
Fram | -0.006 |
KEN | -0.006 |
Token | ibly |
Token ID | 3193 |
Feature activation | -8.48e-03 |
Pos logit contributions | |
rals | +0.079 |
heimer | +0.074 |
cars | +0.072 |
Math | +0.068 |
pet | +0.068 |
Neg logit contributions | |
BOX | -0.091 |
icut | -0.088 |
defe | -0.088 |
Fram | -0.087 |
KEN | -0.086 |
Token | fals |
Token ID | 27807 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.006 |
icut | -0.006 |
defe | -0.006 |
Fram | -0.006 |
KEN | -0.006 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
rals | +0.053 |
heimer | +0.040 |
cars | +0.038 |
Math | +0.028 |
pet | +0.027 |
Neg logit contributions | |
BOX | -0.320 |
icut | -0.316 |
defe | -0.313 |
Fram | -0.311 |
KEN | -0.310 |
Token | the |
Token ID | 262 |
Feature activation | -1.57e-03 |
Pos logit contributions | |
rals | +0.294 |
heimer | +0.274 |
cars | +0.271 |
Math | +0.257 |
pet | +0.255 |
Neg logit contributions | |
BOX | -0.268 |
icut | -0.262 |
defe | -0.257 |
Fram | -0.255 |
KEN | -0.254 |
Token | outcomes |
Token ID | 10906 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -3.41e-03 |
Pos logit contributions | |
BOX | +0.030 |
icut | +0.029 |
defe | +0.028 |
Fram | +0.028 |
KEN | +0.028 |
Neg logit contributions | |
rals | -0.030 |
heimer | -0.028 |
cars | -0.028 |
Math | -0.026 |
pet | -0.026 |
Token | Ad |
Token ID | 1215 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | Policy |
Token ID | 7820 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
BOX | +0.019 |
icut | +0.018 |
defe | +0.018 |
KEN | +0.018 |
Fram | +0.018 |
Neg logit contributions | |
rals | -0.018 |
heimer | -0.017 |
cars | -0.016 |
pet | -0.015 |
Math | -0.015 |
Token | Hardware |
Token ID | 28715 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
BOX | +0.027 |
icut | +0.026 |
defe | +0.026 |
Fram | +0.026 |
KEN | +0.025 |
Neg logit contributions | |
rals | -0.031 |
heimer | -0.029 |
cars | -0.029 |
Math | -0.027 |
pet | -0.027 |
Token | Kit |
Token ID | 10897 |
Feature activation | -2.89e-03 |
Pos logit contributions | |
rals | +0.141 |
heimer | +0.131 |
cars | +0.129 |
Math | +0.122 |
pet | +0.121 |
Neg logit contributions | |
BOX | -0.151 |
icut | -0.147 |
defe | -0.146 |
Fram | -0.144 |
KEN | -0.143 |
Token |
|
Token ID | 198 |
Feature activation | -8.73e-04 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +5.85e-03 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | Shop |
Token ID | 29917 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
BOX | +0.004 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.004 |
cars | -0.004 |
pet | -0.003 |
Math | -0.003 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
BOX | +0.012 |
icut | +0.012 |
defe | +0.012 |
Fram | +0.012 |
KEN | +0.012 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | POS |
Token ID | 28069 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
rals | +0.254 |
heimer | +0.235 |
cars | +0.231 |
Math | +0.217 |
pet | +0.216 |
Neg logit contributions | |
BOX | -0.308 |
icut | -0.302 |
defe | -0.298 |
Fram | -0.294 |
KEN | -0.293 |
Token | runs |
Token ID | 4539 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
rals | +0.085 |
heimer | +0.079 |
cars | +0.078 |
Math | +0.074 |
pet | +0.074 |
Neg logit contributions | |
BOX | -0.082 |
icut | -0.081 |
defe | -0.080 |
Fram | -0.078 |
KEN | -0.078 |
Token | on |
Token ID | 319 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
rals | +0.023 |
heimer | +0.022 |
cars | +0.022 |
Math | +0.020 |
pet | +0.020 |
Neg logit contributions | |
BOX | -0.022 |
icut | -0.021 |
defe | -0.021 |
Fram | -0.021 |
KEN | -0.021 |
Token | an |
Token ID | 281 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
BOX | +0.010 |
icut | +0.009 |
defe | +0.009 |
Fram | +0.009 |
KEN | +0.009 |
Neg logit contributions | |
rals | -0.010 |
heimer | -0.009 |
cars | -0.009 |
Math | -0.008 |
pet | -0.008 |
Token | iPad |
Token ID | 13133 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.007 |
cars | +0.007 |
Math | +0.006 |
pet | +0.006 |
Neg logit contributions | |
BOX | -0.008 |
icut | -0.008 |
defe | -0.008 |
Fram | -0.008 |
KEN | -0.008 |
Token | likely |
Token ID | 1884 |
Feature activation | -3.30e-03 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.007 |
cars | +0.007 |
Math | +0.006 |
pet | +0.006 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.007 |
Fram | -0.007 |
KEN | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +4.86e-03 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | self |
Token ID | 2116 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.003 |
Math | -0.003 |
pet | -0.003 |
Token | - |
Token ID | 12 |
Feature activation | +2.60e-03 |
Pos logit contributions | |
rals | +0.030 |
heimer | +0.028 |
cars | +0.027 |
pet | +0.026 |
Math | +0.026 |
Neg logit contributions | |
BOX | -0.030 |
icut | -0.029 |
defe | -0.028 |
Fram | -0.028 |
KEN | -0.028 |
Token | ident |
Token ID | 738 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
pet | -0.002 |
Math | -0.001 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
BOX | +0.025 |
icut | +0.024 |
defe | +0.024 |
Fram | +0.024 |
KEN | +0.024 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.003 |
cars | -0.003 |
Math | -0.002 |
pet | -0.002 |
Token | as |
Token ID | 355 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
rals | +0.296 |
heimer | +0.276 |
cars | +0.274 |
Math | +0.259 |
pet | +0.257 |
Neg logit contributions | |
BOX | -0.266 |
icut | -0.258 |
defe | -0.256 |
Fram | -0.253 |
KEN | -0.253 |
Token | heterosexual |
Token ID | 24026 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
BOX | +0.020 |
icut | +0.020 |
defe | +0.019 |
Fram | +0.019 |
KEN | +0.019 |
Neg logit contributions | |
rals | -0.016 |
heimer | -0.015 |
cars | -0.015 |
Math | -0.014 |
pet | -0.014 |
Token | s |
Token ID | 82 |
Feature activation | +7.85e-03 |
Pos logit contributions | |
rals | +0.033 |
heimer | +0.031 |
cars | +0.031 |
Math | +0.029 |
pet | +0.029 |
Neg logit contributions | |
BOX | -0.034 |
icut | -0.033 |
defe | -0.033 |
Fram | -0.033 |
KEN | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -3.90e-03 |
Pos logit contributions | |
BOX | +0.005 |
icut | +0.005 |
defe | +0.005 |
Fram | +0.005 |
KEN | +0.005 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.005 |
cars | -0.005 |
Math | -0.005 |
pet | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +3.44e-03 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.002 |
KEN | -0.002 |
Token | as |
Token ID | 355 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
rals | +0.036 |
heimer | +0.033 |
cars | +0.033 |
Math | +0.031 |
pet | +0.031 |
Neg logit contributions | |
BOX | -0.034 |
icut | -0.033 |
defe | -0.032 |
Fram | -0.032 |
KEN | -0.032 |
Token | it |
Token ID | 340 |
Feature activation | -8.95e-03 |
Pos logit contributions | |
BOX | +0.017 |
icut | +0.017 |
defe | +0.017 |
Fram | +0.017 |
KEN | +0.017 |
Neg logit contributions | |
rals | -0.018 |
heimer | -0.016 |
cars | -0.016 |
Math | -0.015 |
pet | -0.015 |
Token | continues |
Token ID | 4477 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.006 |
cars | +0.006 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.006 |
icut | -0.006 |
defe | -0.006 |
Fram | -0.006 |
KEN | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | -8.26e-03 |
Pos logit contributions | |
BOX | +0.035 |
icut | +0.034 |
defe | +0.033 |
Fram | +0.033 |
KEN | +0.033 |
Neg logit contributions | |
rals | -0.037 |
heimer | -0.035 |
cars | -0.034 |
Math | -0.032 |
pet | -0.032 |
Token | dens |
Token ID | 29509 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.006 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
BOX | +0.129 |
icut | +0.128 |
defe | +0.126 |
Fram | +0.126 |
KEN | +0.125 |
Neg logit contributions | |
rals | -0.026 |
heimer | -0.020 |
cars | -0.019 |
pet | -0.014 |
Math | -0.014 |
Token | its |
Token ID | 663 |
Feature activation | +6.68e-02 |
Pos logit contributions | |
rals | +0.294 |
heimer | +0.276 |
cars | +0.272 |
Math | +0.257 |
pet | +0.256 |
Neg logit contributions | |
BOX | -0.267 |
icut | -0.261 |
defe | -0.257 |
Fram | -0.254 |
KEN | -0.253 |
Token | network |
Token ID | 3127 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
BOX | +0.043 |
icut | +0.042 |
defe | +0.042 |
Fram | +0.041 |
KEN | +0.041 |
Neg logit contributions | |
rals | -0.047 |
heimer | -0.044 |
cars | -0.044 |
Math | -0.041 |
pet | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | -1.81e-03 |
Pos logit contributions | |
BOX | +0.071 |
icut | +0.069 |
defe | +0.068 |
Fram | +0.067 |
KEN | +0.067 |
Neg logit contributions | |
rals | -0.073 |
heimer | -0.068 |
cars | -0.067 |
Math | -0.063 |
pet | -0.063 |
Token | C |
Token ID | 327 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | TO |
Token ID | 10468 |
Feature activation | -7.60e-04 |
Pos logit contributions | |
rals | +0.100 |
heimer | +0.093 |
cars | +0.092 |
Math | +0.086 |
pet | +0.086 |
Neg logit contributions | |
BOX | -0.106 |
icut | -0.103 |
defe | -0.102 |
Fram | -0.101 |
KEN | -0.100 |
Token | Processing |
Token ID | 28403 |
Feature activation | -7.97e-02 |
Pos logit contributions | |
BOX | +0.024 |
icut | +0.023 |
defe | +0.023 |
Fram | +0.023 |
KEN | +0.023 |
Neg logit contributions | |
rals | -0.021 |
heimer | -0.020 |
cars | -0.019 |
pet | -0.018 |
Math | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -5.29e-03 |
Pos logit contributions | |
rals | +0.055 |
heimer | +0.051 |
cars | +0.051 |
Math | +0.048 |
pet | +0.048 |
Neg logit contributions | |
BOX | -0.053 |
icut | -0.052 |
defe | -0.051 |
Fram | -0.050 |
KEN | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | +2.98e-03 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.004 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | The |
Token ID | 464 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
pet | -0.002 |
Math | -0.002 |
Token | Shop |
Token ID | 13705 |
Feature activation | -7.37e-02 |
Pos logit contributions | |
rals | +0.017 |
heimer | +0.016 |
cars | +0.016 |
Math | +0.015 |
pet | +0.015 |
Neg logit contributions | |
BOX | -0.019 |
icut | -0.019 |
defe | -0.018 |
Fram | -0.018 |
KEN | -0.018 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
rals | +0.014 |
heimer | +0.011 |
cars | +0.010 |
Math | +0.007 |
pet | +0.007 |
Neg logit contributions | |
BOX | -0.086 |
icut | -0.085 |
defe | -0.084 |
Fram | -0.084 |
KEN | -0.083 |
Token | card |
Token ID | 2657 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
rals | +0.216 |
heimer | +0.197 |
cars | +0.193 |
Math | +0.178 |
pet | +0.178 |
Neg logit contributions | |
BOX | -0.346 |
icut | -0.339 |
defe | -0.335 |
Fram | -0.332 |
KEN | -0.331 |
Token | reader |
Token ID | 9173 |
Feature activation | +9.52e-03 |
Pos logit contributions | |
BOX | +0.010 |
icut | +0.009 |
defe | +0.009 |
Fram | +0.009 |
KEN | +0.009 |
Neg logit contributions | |
rals | -0.008 |
heimer | -0.007 |
cars | -0.007 |
Math | -0.007 |
pet | -0.007 |
Token | lets |
Token ID | 8781 |
Feature activation | +9.62e-02 |
Pos logit contributions | |
BOX | +0.006 |
icut | +0.006 |
defe | +0.006 |
Fram | +0.006 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.006 |
pet | -0.006 |
Token | you |
Token ID | 345 |
Feature activation | -5.44e-03 |
Pos logit contributions | |
BOX | +0.066 |
icut | +0.065 |
defe | +0.064 |
Fram | +0.063 |
KEN | +0.063 |
Neg logit contributions | |
rals | -0.064 |
heimer | -0.060 |
cars | -0.059 |
Math | -0.056 |
pet | -0.055 |
Token | take |
Token ID | 1011 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.004 |
icut | -0.004 |
defe | -0.004 |
Fram | -0.004 |
KEN | -0.004 |
Token | Dash |
Token ID | 16189 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
rals | +0.019 |
heimer | +0.018 |
cars | +0.017 |
Math | +0.016 |
pet | +0.016 |
Neg logit contributions | |
BOX | -0.022 |
icut | -0.021 |
defe | -0.021 |
Fram | -0.021 |
KEN | -0.021 |
Token | board |
Token ID | 3526 |
Feature activation | -7.46e-02 |
Pos logit contributions | |
rals | +0.028 |
heimer | +0.026 |
cars | +0.025 |
Math | +0.024 |
pet | +0.024 |
Neg logit contributions | |
BOX | -0.033 |
icut | -0.032 |
defe | -0.032 |
Fram | -0.031 |
KEN | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +8.48e-04 |
Pos logit contributions | |
rals | +0.052 |
heimer | +0.049 |
cars | +0.048 |
Math | +0.045 |
pet | +0.045 |
Neg logit contributions | |
BOX | -0.049 |
icut | -0.048 |
defe | -0.047 |
Fram | -0.047 |
KEN | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | +2.75e-03 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | Shop |
Token ID | 29917 |
Feature activation | +1.28e-03 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
pet | -0.002 |
Math | -0.002 |
Token | ify |
Token ID | 1958 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.000 |
heimer | -0.000 |
cars | -0.000 |
Math | -0.000 |
pet | -0.000 |
Token | POS |
Token ID | 28069 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
rals | +0.214 |
heimer | +0.195 |
cars | +0.191 |
Math | +0.176 |
pet | +0.176 |
Neg logit contributions | |
BOX | -0.345 |
icut | -0.339 |
defe | -0.335 |
Fram | -0.332 |
KEN | -0.330 |
Token | lets |
Token ID | 8781 |
Feature activation | +9.82e-02 |
Pos logit contributions | |
rals | +0.088 |
heimer | +0.082 |
cars | +0.081 |
Math | +0.077 |
pet | +0.076 |
Neg logit contributions | |
BOX | -0.088 |
icut | -0.086 |
defe | -0.084 |
Fram | -0.084 |
KEN | -0.083 |
Token | you |
Token ID | 345 |
Feature activation | +4.79e-03 |
Pos logit contributions | |
BOX | +0.068 |
icut | +0.066 |
defe | +0.065 |
Fram | +0.065 |
KEN | +0.064 |
Neg logit contributions | |
rals | -0.066 |
heimer | -0.061 |
cars | -0.060 |
Math | -0.057 |
pet | -0.056 |
Token | easily |
Token ID | 3538 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.003 |
Math | -0.003 |
pet | -0.003 |
Token | manage |
Token ID | 6687 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
BOX | +0.147 |
icut | +0.144 |
defe | +0.142 |
Fram | +0.141 |
KEN | +0.140 |
Neg logit contributions | |
rals | -0.138 |
heimer | -0.128 |
cars | -0.126 |
Math | -0.119 |
pet | -0.118 |
Token | the |
Token ID | 262 |
Feature activation | -4.08e-03 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | time |
Token ID | 640 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | P |
Token ID | 350 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
rals | +0.015 |
heimer | +0.014 |
cars | +0.014 |
Math | +0.013 |
pet | +0.013 |
Neg logit contributions | |
BOX | -0.014 |
icut | -0.014 |
defe | -0.013 |
Fram | -0.013 |
KEN | -0.013 |
Token | eder |
Token ID | 5702 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
BOX | +0.047 |
icut | +0.045 |
defe | +0.045 |
Fram | +0.044 |
KEN | +0.044 |
Neg logit contributions | |
rals | -0.050 |
heimer | -0.047 |
cars | -0.046 |
pet | -0.043 |
Math | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
BOX | +0.022 |
icut | +0.022 |
defe | +0.021 |
Fram | +0.021 |
KEN | +0.021 |
Neg logit contributions | |
rals | -0.024 |
heimer | -0.022 |
cars | -0.022 |
Math | -0.021 |
pet | -0.021 |
Token | charged |
Token ID | 5047 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
BOX | +0.014 |
icut | +0.013 |
defe | +0.013 |
Fram | +0.013 |
KEN | +0.013 |
Neg logit contributions | |
rals | -0.013 |
heimer | -0.012 |
cars | -0.012 |
Math | -0.011 |
pet | -0.011 |
Token | for |
Token ID | 329 |
Feature activation | +9.26e-03 |
Pos logit contributions | |
BOX | +0.156 |
icut | +0.152 |
defe | +0.151 |
Fram | +0.149 |
KEN | +0.148 |
Neg logit contributions | |
rals | -0.162 |
heimer | -0.151 |
cars | -0.149 |
Math | -0.141 |
pet | -0.140 |
Token | not |
Token ID | 407 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
BOX | +0.007 |
icut | +0.006 |
defe | +0.006 |
Fram | +0.006 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.006 |
heimer | -0.006 |
cars | -0.005 |
Math | -0.005 |
pet | -0.005 |
Token | using |
Token ID | 1262 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
rals | +0.009 |
heimer | +0.008 |
cars | +0.008 |
Math | +0.008 |
pet | +0.008 |
Neg logit contributions | |
BOX | -0.009 |
icut | -0.009 |
defe | -0.009 |
KEN | -0.009 |
Fram | -0.009 |
Token | an |
Token ID | 281 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
BOX | +0.025 |
icut | +0.024 |
defe | +0.024 |
Fram | +0.023 |
KEN | +0.023 |
Neg logit contributions | |
rals | -0.025 |
heimer | -0.024 |
cars | -0.023 |
Math | -0.022 |
pet | -0.022 |
Token | inter |
Token ID | 987 |
Feature activation | -9.07e-02 |
Pos logit contributions | |
rals | +0.009 |
heimer | +0.009 |
cars | +0.008 |
pet | +0.008 |
Math | +0.008 |
Neg logit contributions | |
BOX | -0.011 |
icut | -0.011 |
defe | -0.011 |
Fram | -0.010 |
KEN | -0.010 |
Token | lives |
Token ID | 3160 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
rals | +0.013 |
heimer | +0.012 |
cars | +0.012 |
Math | +0.011 |
pet | +0.011 |
Neg logit contributions | |
BOX | -0.017 |
icut | -0.017 |
defe | -0.016 |
Fram | -0.016 |
KEN | -0.016 |
Token | ! |
Token ID | 0 |
Feature activation | -9.94e-02 |
Pos logit contributions | |
rals | +0.051 |
heimer | +0.047 |
cars | +0.047 |
Math | +0.044 |
pet | +0.044 |
Neg logit contributions | |
BOX | -0.053 |
icut | -0.052 |
defe | -0.051 |
Fram | -0.051 |
KEN | -0.051 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
rals | +0.069 |
heimer | +0.065 |
cars | +0.064 |
Math | +0.061 |
pet | +0.060 |
Neg logit contributions | |
BOX | -0.065 |
icut | -0.064 |
defe | -0.064 |
Fram | -0.062 |
KEN | -0.062 |
Token | Revolution |
Token ID | 50237 |
Feature activation | +7.24e-02 |
Pos logit contributions | |
BOX | +0.029 |
icut | +0.029 |
defe | +0.028 |
Fram | +0.028 |
KEN | +0.028 |
Neg logit contributions | |
rals | -0.029 |
heimer | -0.027 |
cars | -0.027 |
pet | -0.025 |
Math | -0.025 |
Token | ary |
Token ID | 560 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
BOX | +0.043 |
icut | +0.042 |
defe | +0.042 |
Fram | +0.041 |
KEN | +0.041 |
Neg logit contributions | |
rals | -0.055 |
heimer | -0.052 |
cars | -0.051 |
Math | -0.048 |
pet | -0.048 |
Token | Rac |
Token ID | 24746 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
BOX | +0.085 |
icut | +0.083 |
defe | +0.082 |
Fram | +0.081 |
KEN | +0.080 |
Neg logit contributions | |
rals | -0.085 |
heimer | -0.079 |
cars | -0.078 |
Math | -0.074 |
pet | -0.073 |
Token | ism |
Token ID | 1042 |
Feature activation | +6.51e-03 |
Pos logit contributions | |
BOX | +0.187 |
icut | +0.182 |
defe | +0.179 |
Fram | +0.177 |
KEN | +0.177 |
Neg logit contributions | |
rals | -0.207 |
heimer | -0.193 |
cars | -0.189 |
pet | -0.179 |
Math | -0.179 |
Token | : |
Token ID | 25 |
Feature activation | +6.71e-03 |
Pos logit contributions | |
BOX | +0.004 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.004 |
cars | -0.004 |
Math | -0.004 |
pet | -0.004 |
Token | Af |
Token ID | 2483 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
BOX | +0.005 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.004 |
cars | -0.004 |
Math | -0.004 |
pet | -0.004 |
Token | ro |
Token ID | 305 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
BOX | +0.020 |
icut | +0.019 |
defe | +0.019 |
Fram | +0.019 |
KEN | +0.019 |
Neg logit contributions | |
rals | -0.023 |
heimer | -0.021 |
cars | -0.021 |
Math | -0.020 |
pet | -0.020 |
Token | ‑ |
Token ID | 20977 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
BOX | +0.093 |
icut | +0.090 |
defe | +0.089 |
Fram | +0.088 |
KEN | +0.088 |
Neg logit contributions | |
rals | -0.093 |
heimer | -0.087 |
cars | -0.085 |
Math | -0.080 |
pet | -0.080 |
Token | s |
Token ID | 82 |
Feature activation | +3.79e-03 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.002 |
pet | -0.002 |
Token | an |
Token ID | 281 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.002 |
pet | -0.002 |
Token | extraordinary |
Token ID | 11359 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
rals | +0.013 |
heimer | +0.012 |
cars | +0.012 |
Math | +0.011 |
pet | +0.011 |
Neg logit contributions | |
BOX | -0.013 |
icut | -0.012 |
defe | -0.012 |
Fram | -0.012 |
KEN | -0.012 |
Token | story |
Token ID | 1621 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
rals | +0.024 |
heimer | +0.022 |
cars | +0.022 |
Math | +0.021 |
pet | +0.020 |
Neg logit contributions | |
BOX | -0.025 |
icut | -0.024 |
defe | -0.024 |
Fram | -0.024 |
KEN | -0.024 |
Token | of |
Token ID | 286 |
Feature activation | +2.35e-03 |
Pos logit contributions | |
rals | +0.044 |
heimer | +0.041 |
cars | +0.040 |
Math | +0.038 |
pet | +0.038 |
Neg logit contributions | |
BOX | -0.044 |
icut | -0.044 |
defe | -0.043 |
Fram | -0.042 |
KEN | -0.042 |
Token | human |
Token ID | 1692 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | relationships |
Token ID | 6958 |
Feature activation | +9.62e-02 |
Pos logit contributions | |
BOX | +0.165 |
icut | +0.160 |
defe | +0.157 |
KEN | +0.157 |
Fram | +0.157 |
Neg logit contributions | |
rals | -0.171 |
heimer | -0.159 |
cars | -0.158 |
Math | -0.150 |
pet | -0.149 |
Token | and |
Token ID | 290 |
Feature activation | -1.86e-03 |
Pos logit contributions | |
BOX | +0.066 |
icut | +0.064 |
defe | +0.063 |
KEN | +0.062 |
Fram | +0.062 |
Neg logit contributions | |
rals | -0.065 |
heimer | -0.061 |
cars | -0.060 |
Math | -0.056 |
pet | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | +2.48e-03 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
defe | -0.001 |
Token | sexual |
Token ID | 3206 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
KEN | +0.002 |
Fram | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.001 |
pet | -0.001 |
Token | dynamics |
Token ID | 17262 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
rals | +0.039 |
heimer | +0.037 |
cars | +0.036 |
Math | +0.035 |
pet | +0.034 |
Neg logit contributions | |
BOX | -0.039 |
icut | -0.038 |
defe | -0.038 |
KEN | -0.038 |
Fram | -0.037 |
Token | Federal |
Token ID | 5618 |
Feature activation | -8.81e-02 |
Pos logit contributions | |
rals | +0.013 |
heimer | +0.012 |
cars | +0.012 |
Math | +0.011 |
pet | +0.011 |
Neg logit contributions | |
BOX | -0.020 |
icut | -0.019 |
defe | -0.019 |
Fram | -0.019 |
KEN | -0.019 |
Token | Way |
Token ID | 6378 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
rals | +0.055 |
heimer | +0.051 |
cars | +0.050 |
Math | +0.047 |
pet | +0.047 |
Neg logit contributions | |
BOX | -0.065 |
icut | -0.063 |
defe | -0.063 |
Fram | -0.062 |
KEN | -0.062 |
Token | teens |
Token ID | 15508 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
BOX | +0.106 |
icut | +0.104 |
defe | +0.102 |
Fram | +0.102 |
KEN | +0.101 |
Neg logit contributions | |
rals | -0.065 |
heimer | -0.059 |
cars | -0.058 |
pet | -0.053 |
Math | -0.053 |
Token | , |
Token ID | 11 |
Feature activation | +7.86e-03 |
Pos logit contributions | |
rals | +0.121 |
heimer | +0.113 |
cars | +0.112 |
Math | +0.106 |
pet | +0.105 |
Neg logit contributions | |
BOX | -0.123 |
icut | -0.120 |
defe | -0.119 |
Fram | -0.117 |
KEN | -0.117 |
Token | is |
Token ID | 318 |
Feature activation | -4.31e-04 |
Pos logit contributions | |
BOX | +0.005 |
icut | +0.005 |
defe | +0.005 |
Fram | +0.005 |
KEN | +0.005 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.005 |
cars | -0.005 |
Math | -0.005 |
pet | -0.005 |
Token | charged |
Token ID | 5047 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
rals | +0.000 |
heimer | +0.000 |
cars | +0.000 |
Math | +0.000 |
pet | +0.000 |
Neg logit contributions | |
BOX | -0.000 |
icut | -0.000 |
defe | -0.000 |
KEN | -0.000 |
Fram | -0.000 |
Token | for |
Token ID | 329 |
Feature activation | +9.80e-03 |
Pos logit contributions | |
BOX | +0.153 |
icut | +0.149 |
defe | +0.147 |
Fram | +0.145 |
KEN | +0.145 |
Neg logit contributions | |
rals | -0.169 |
heimer | -0.158 |
cars | -0.156 |
Math | -0.147 |
pet | -0.147 |
Token | driving |
Token ID | 5059 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
BOX | +0.007 |
icut | +0.007 |
defe | +0.007 |
Fram | +0.007 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.006 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.006 |
pet | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | +3.85e-03 |
Pos logit contributions | |
rals | +0.032 |
heimer | +0.030 |
cars | +0.030 |
Math | +0.028 |
pet | +0.028 |
Neg logit contributions | |
BOX | -0.034 |
icut | -0.032 |
defe | -0.032 |
Fram | -0.032 |
KEN | -0.032 |
Token | vehicle |
Token ID | 4038 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.002 |
pet | -0.002 |
Token | without |
Token ID | 1231 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
rals | +0.097 |
heimer | +0.091 |
cars | +0.090 |
Math | +0.085 |
pet | +0.084 |
Neg logit contributions | |
BOX | -0.095 |
icut | -0.092 |
defe | -0.091 |
Fram | -0.090 |
KEN | -0.090 |
Token | . |
Token ID | 13 |
Feature activation | -4.02e-03 |
Pos logit contributions | |
BOX | +0.052 |
icut | +0.051 |
defe | +0.050 |
Fram | +0.050 |
KEN | +0.050 |
Neg logit contributions | |
rals | -0.052 |
heimer | -0.049 |
cars | -0.048 |
Math | -0.045 |
pet | -0.045 |
Token | Clarke |
Token ID | 19635 |
Feature activation | +8.22e-02 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.003 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | used |
Token ID | 973 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
BOX | +0.057 |
icut | +0.055 |
defe | +0.055 |
KEN | +0.054 |
Fram | +0.054 |
Neg logit contributions | |
rals | -0.055 |
heimer | -0.051 |
cars | -0.050 |
Math | -0.048 |
pet | -0.047 |
Token | this |
Token ID | 428 |
Feature activation | -9.79e-03 |
Pos logit contributions | |
rals | +0.013 |
heimer | +0.012 |
cars | +0.012 |
Math | +0.011 |
pet | +0.011 |
Neg logit contributions | |
BOX | -0.013 |
icut | -0.012 |
defe | -0.012 |
Fram | -0.012 |
KEN | -0.012 |
Token | biological |
Token ID | 10685 |
Feature activation | +6.73e-03 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.006 |
cars | +0.006 |
Math | +0.006 |
pet | +0.006 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.007 |
KEN | -0.006 |
Fram | -0.006 |
Token | angle |
Token ID | 9848 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
BOX | +0.005 |
icut | +0.004 |
KEN | +0.004 |
defe | +0.004 |
Fram | +0.004 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.004 |
cars | -0.004 |
Math | -0.004 |
pet | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -1.73e-03 |
Pos logit contributions | |
BOX | +0.189 |
icut | +0.184 |
defe | +0.182 |
KEN | +0.180 |
Fram | +0.179 |
Neg logit contributions | |
rals | -0.193 |
heimer | -0.180 |
cars | -0.177 |
Math | -0.167 |
pet | -0.167 |
Token | argue |
Token ID | 7267 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | against |
Token ID | 1028 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
BOX | +0.070 |
icut | +0.068 |
defe | +0.067 |
Fram | +0.067 |
KEN | +0.066 |
Neg logit contributions | |
rals | -0.069 |
heimer | -0.064 |
cars | -0.063 |
Math | -0.059 |
pet | -0.059 |
Token | co |
Token ID | 763 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
BOX | +0.022 |
icut | +0.021 |
defe | +0.021 |
Fram | +0.021 |
KEN | +0.021 |
Neg logit contributions | |
rals | -0.022 |
heimer | -0.021 |
cars | -0.020 |
Math | -0.019 |
pet | -0.019 |
Token | - |
Token ID | 12 |
Feature activation | +6.47e-03 |
Pos logit contributions | |
rals | +0.076 |
heimer | +0.071 |
cars | +0.069 |
pet | +0.065 |
Math | +0.065 |
Neg logit contributions | |
BOX | -0.078 |
icut | -0.076 |
defe | -0.075 |
Fram | -0.074 |
KEN | -0.074 |
Token | a |
Token ID | 257 |
Feature activation | -4.20e-03 |
Pos logit contributions | |
rals | +0.009 |
heimer | +0.008 |
cars | +0.008 |
Math | +0.007 |
pet | +0.007 |
Neg logit contributions | |
BOX | -0.009 |
icut | -0.008 |
defe | -0.008 |
Fram | -0.008 |
KEN | -0.008 |
Token | phone |
Token ID | 3072 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | interview |
Token ID | 2720 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
BOX | +0.097 |
icut | +0.095 |
defe | +0.094 |
Fram | +0.093 |
KEN | +0.092 |
Neg logit contributions | |
rals | -0.078 |
heimer | -0.072 |
cars | -0.071 |
Math | -0.066 |
pet | -0.066 |
Token | on |
Token ID | 319 |
Feature activation | +6.81e-03 |
Pos logit contributions | |
rals | +0.043 |
heimer | +0.041 |
cars | +0.040 |
Math | +0.037 |
pet | +0.037 |
Neg logit contributions | |
BOX | -0.046 |
icut | -0.045 |
defe | -0.045 |
Fram | -0.044 |
KEN | -0.044 |
Token | Friday |
Token ID | 3217 |
Feature activation | -7.16e-02 |
Pos logit contributions | |
BOX | +0.005 |
icut | +0.005 |
defe | +0.005 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.004 |
cars | -0.004 |
Math | -0.004 |
pet | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.72e-03 |
Pos logit contributions | |
rals | +0.047 |
heimer | +0.044 |
cars | +0.043 |
Math | +0.041 |
pet | +0.041 |
Neg logit contributions | |
BOX | -0.050 |
icut | -0.049 |
defe | -0.048 |
KEN | -0.048 |
Fram | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | +4.48e-04 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -3.41e-03 |
Pos logit contributions | |
BOX | +0.000 |
icut | +0.000 |
defe | +0.000 |
Fram | +0.000 |
KEN | +0.000 |
Neg logit contributions | |
rals | -0.000 |
heimer | -0.000 |
cars | -0.000 |
Math | -0.000 |
pet | -0.000 |
Token | Currently |
Token ID | 21327 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
rals | +0.002 |
heimer | +0.002 |
cars | +0.002 |
pet | +0.002 |
Math | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +4.44e-03 |
Pos logit contributions | |
BOX | +0.029 |
icut | +0.028 |
defe | +0.028 |
Fram | +0.028 |
KEN | +0.028 |
Neg logit contributions | |
rals | -0.029 |
heimer | -0.027 |
cars | -0.027 |
Math | -0.025 |
pet | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | -9.17e-03 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.003 |
Math | -0.003 |
pet | -0.003 |
Token | ™ |
Token ID | 8151 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
BOX | +0.051 |
icut | +0.049 |
defe | +0.049 |
Fram | +0.048 |
KEN | +0.048 |
Neg logit contributions | |
rals | -0.052 |
heimer | -0.049 |
cars | -0.048 |
Math | -0.045 |
pet | -0.045 |
Token | that |
Token ID | 326 |
Feature activation | -3.65e-03 |
Pos logit contributions | |
rals | +0.203 |
heimer | +0.189 |
cars | +0.188 |
Math | +0.177 |
pet | +0.177 |
Neg logit contributions | |
BOX | -0.201 |
icut | -0.198 |
defe | -0.195 |
Fram | -0.192 |
KEN | -0.191 |
Token | enables |
Token ID | 13536 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.002 |
cars | +0.002 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.002 |
icut | -0.002 |
defe | -0.002 |
Fram | -0.002 |
KEN | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +6.04e-03 |
Pos logit contributions | |
rals | +0.016 |
heimer | +0.015 |
cars | +0.015 |
Math | +0.014 |
pet | +0.014 |
Neg logit contributions | |
BOX | -0.018 |
icut | -0.018 |
defe | -0.018 |
Fram | -0.018 |
KEN | -0.017 |
Token | driver |
Token ID | 4639 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
BOX | +0.004 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.004 |
cars | -0.004 |
Math | -0.003 |
pet | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -1.14e-03 |
Pos logit contributions | |
BOX | +0.034 |
icut | +0.033 |
defe | +0.033 |
Fram | +0.032 |
KEN | +0.032 |
Neg logit contributions | |
rals | -0.037 |
heimer | -0.034 |
cars | -0.034 |
Math | -0.032 |
pet | -0.032 |
Token | select |
Token ID | 2922 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | modes |
Token ID | 12881 |
Feature activation | +9.23e-02 |
Pos logit contributions | |
rals | +0.017 |
heimer | +0.015 |
cars | +0.015 |
Math | +0.014 |
pet | +0.014 |
Neg logit contributions | |
BOX | -0.016 |
icut | -0.016 |
defe | -0.015 |
Fram | -0.015 |
KEN | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
BOX | +0.062 |
icut | +0.061 |
defe | +0.060 |
Fram | +0.059 |
KEN | +0.059 |
Neg logit contributions | |
rals | -0.063 |
heimer | -0.059 |
cars | -0.058 |
Math | -0.055 |
pet | -0.055 |
Token | optimize |
Token ID | 27183 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
rals | +0.008 |
heimer | +0.008 |
cars | +0.008 |
Math | +0.007 |
pet | +0.007 |
Neg logit contributions | |
BOX | -0.008 |
icut | -0.008 |
defe | -0.007 |
Fram | -0.007 |
KEN | -0.007 |
Token | driving |
Token ID | 5059 |
Feature activation | -5.73e-02 |
Pos logit contributions | |
BOX | +0.029 |
icut | +0.028 |
defe | +0.028 |
Fram | +0.028 |
KEN | +0.028 |
Neg logit contributions | |
rals | -0.030 |
heimer | -0.028 |
cars | -0.028 |
Math | -0.026 |
pet | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | -9.43e-03 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.006 |
cars | +0.006 |
Math | +0.006 |
pet | +0.006 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.006 |
Fram | -0.006 |
KEN | -0.006 |
Token | States |
Token ID | 1829 |
Feature activation | -7.06e-02 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.007 |
Fram | -0.007 |
KEN | -0.007 |
Token | at |
Token ID | 379 |
Feature activation | +7.30e-04 |
Pos logit contributions | |
rals | +0.048 |
heimer | +0.045 |
cars | +0.044 |
Math | +0.042 |
pet | +0.042 |
Neg logit contributions | |
BOX | -0.048 |
icut | -0.047 |
defe | -0.046 |
Fram | -0.045 |
KEN | -0.045 |
Token | Screen |
Token ID | 15216 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
BOX | +0.000 |
icut | +0.000 |
defe | +0.000 |
Fram | +0.000 |
KEN | +0.000 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.000 |
cars | -0.000 |
pet | -0.000 |
Math | -0.000 |
Token | Gems |
Token ID | 30196 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
BOX | +0.077 |
icut | +0.076 |
defe | +0.075 |
Fram | +0.074 |
KEN | +0.073 |
Neg logit contributions | |
rals | -0.072 |
heimer | -0.067 |
cars | -0.066 |
Math | -0.062 |
pet | -0.062 |
Token | , |
Token ID | 11 |
Feature activation | -1.77e-03 |
Pos logit contributions | |
rals | +0.067 |
heimer | +0.062 |
cars | +0.061 |
Math | +0.058 |
pet | +0.058 |
Neg logit contributions | |
BOX | -0.067 |
icut | -0.065 |
defe | -0.064 |
Fram | -0.064 |
KEN | -0.063 |
Token | we |
Token ID | 356 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | knew |
Token ID | 2993 |
Feature activation | -9.73e-02 |
Pos logit contributions | |
rals | +0.010 |
heimer | +0.009 |
cars | +0.009 |
Math | +0.009 |
pet | +0.009 |
Neg logit contributions | |
BOX | -0.011 |
icut | -0.011 |
defe | -0.010 |
Fram | -0.010 |
KEN | -0.010 |
Token | that |
Token ID | 326 |
Feature activation | -6.88e-03 |
Pos logit contributions | |
rals | +0.066 |
heimer | +0.062 |
cars | +0.061 |
Math | +0.057 |
pet | +0.057 |
Neg logit contributions | |
BOX | -0.066 |
icut | -0.064 |
defe | -0.064 |
Fram | -0.063 |
KEN | -0.062 |
Token | was |
Token ID | 373 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.004 |
cars | +0.004 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.004 |
KEN | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -3.50e-04 |
Pos logit contributions | |
BOX | +0.016 |
icut | +0.015 |
defe | +0.015 |
Fram | +0.015 |
KEN | +0.015 |
Neg logit contributions | |
rals | -0.016 |
heimer | -0.015 |
cars | -0.015 |
Math | -0.014 |
pet | -0.014 |
Token | once |
Token ID | 1752 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
BOX | +0.007 |
icut | +0.007 |
defe | +0.007 |
Fram | +0.007 |
KEN | +0.007 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.005 |
cars | -0.005 |
Math | -0.004 |
pet | -0.004 |
Token | relentlessly |
Token ID | 37952 |
Feature activation | +6.54e-02 |
Pos logit contributions | |
rals | +0.098 |
heimer | +0.091 |
cars | +0.090 |
Math | +0.085 |
pet | +0.084 |
Neg logit contributions | |
BOX | -0.096 |
icut | -0.094 |
defe | -0.093 |
Fram | -0.092 |
KEN | -0.091 |
Token | cynical |
Token ID | 28858 |
Feature activation | -9.53e-02 |
Pos logit contributions | |
BOX | +0.045 |
icut | +0.044 |
defe | +0.043 |
Fram | +0.043 |
KEN | +0.042 |
Neg logit contributions | |
rals | -0.044 |
heimer | -0.041 |
cars | -0.040 |
Math | -0.038 |
pet | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | +2.14e-03 |
Pos logit contributions | |
rals | +0.065 |
heimer | +0.061 |
cars | +0.060 |
Math | +0.057 |
pet | +0.056 |
Neg logit contributions | |
BOX | -0.064 |
icut | -0.063 |
defe | -0.062 |
Fram | -0.061 |
KEN | -0.061 |
Token | outward |
Token ID | 23537 |
Feature activation | +6.14e-02 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | ly |
Token ID | 306 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
BOX | +0.033 |
icut | +0.032 |
defe | +0.032 |
Fram | +0.031 |
KEN | +0.031 |
Neg logit contributions | |
rals | -0.050 |
heimer | -0.047 |
cars | -0.047 |
Math | -0.044 |
pet | -0.044 |
Token | optimistic |
Token ID | 16915 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
BOX | +0.090 |
icut | +0.088 |
defe | +0.087 |
Fram | +0.086 |
KEN | +0.086 |
Neg logit contributions | |
rals | -0.082 |
heimer | -0.076 |
cars | -0.075 |
Math | -0.071 |
pet | -0.071 |
Token | about |
Token ID | 546 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
rals | +0.015 |
heimer | +0.014 |
cars | +0.014 |
Math | +0.013 |
pet | +0.013 |
Neg logit contributions | |
BOX | -0.016 |
icut | -0.015 |
defe | -0.015 |
KEN | -0.015 |
Fram | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +9.61e-03 |
Pos logit contributions | |
rals | +0.010 |
heimer | +0.010 |
cars | +0.010 |
Math | +0.009 |
pet | +0.009 |
Neg logit contributions | |
BOX | -0.010 |
icut | -0.010 |
defe | -0.010 |
KEN | -0.010 |
Fram | -0.010 |
Token | conce |
Token ID | 8571 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
BOX | +0.006 |
icut | +0.006 |
defe | +0.006 |
Fram | +0.006 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.006 |
pet | -0.006 |
Token | it |
Token ID | 270 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
BOX | +0.198 |
icut | +0.195 |
defe | +0.192 |
Fram | +0.190 |
KEN | +0.190 |
Neg logit contributions | |
rals | -0.110 |
heimer | -0.100 |
cars | -0.097 |
Math | -0.089 |
pet | -0.089 |
Token | cliché |
Token ID | 43973 |
Feature activation | +9.27e-02 |
Pos logit contributions | |
rals | +0.089 |
heimer | +0.083 |
cars | +0.081 |
Math | +0.077 |
pet | +0.077 |
Neg logit contributions | |
BOX | -0.090 |
icut | -0.087 |
defe | -0.086 |
Fram | -0.085 |
KEN | -0.085 |
Token | bo |
Token ID | 1489 |
Feature activation | +4.62e-02 |
Pos logit contributions | |
BOX | +0.062 |
icut | +0.061 |
defe | +0.060 |
Fram | +0.059 |
KEN | +0.059 |
Neg logit contributions | |
rals | -0.064 |
heimer | -0.059 |
cars | -0.058 |
Math | -0.055 |
pet | -0.055 |
Token | hem |
Token ID | 4411 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
BOX | +0.026 |
icut | +0.026 |
defe | +0.025 |
Fram | +0.025 |
KEN | +0.025 |
Neg logit contributions | |
rals | -0.036 |
heimer | -0.034 |
cars | -0.034 |
Math | -0.032 |
pet | -0.032 |
Token | ian |
Token ID | 666 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
BOX | +0.135 |
icut | +0.131 |
defe | +0.128 |
Fram | +0.126 |
KEN | +0.126 |
Neg logit contributions | |
rals | -0.225 |
heimer | -0.213 |
cars | -0.210 |
Math | -0.201 |
pet | -0.200 |
Token | ideals |
Token ID | 22247 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
BOX | +0.084 |
icut | +0.082 |
defe | +0.080 |
Fram | +0.080 |
KEN | +0.079 |
Neg logit contributions | |
rals | -0.089 |
heimer | -0.083 |
cars | -0.082 |
Math | -0.078 |
pet | -0.077 |
Token | that |
Token ID | 326 |
Feature activation | +8.82e-04 |
Pos logit contributions | |
BOX | +0.054 |
icut | +0.053 |
defe | +0.052 |
KEN | +0.051 |
Fram | +0.051 |
Neg logit contributions | |
rals | -0.068 |
heimer | -0.064 |
cars | -0.063 |
Math | -0.060 |
pet | -0.060 |
Token | have |
Token ID | 423 |
Feature activation | +2.42e-03 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | been |
Token ID | 587 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token | exhausted |
Token ID | 19064 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
rals | +0.010 |
heimer | +0.009 |
cars | +0.009 |
Math | +0.009 |
pet | +0.009 |
Neg logit contributions | |
BOX | -0.010 |
icut | -0.009 |
defe | -0.009 |
Fram | -0.009 |
KEN | -0.009 |
Token | for |
Token ID | 329 |
Feature activation | +7.80e-03 |
Pos logit contributions | |
rals | +0.032 |
heimer | +0.030 |
cars | +0.030 |
Math | +0.028 |
pet | +0.028 |
Neg logit contributions | |
BOX | -0.030 |
icut | -0.030 |
defe | -0.029 |
KEN | -0.029 |
Fram | -0.029 |
Token | years |
Token ID | 812 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
BOX | +0.005 |
icut | +0.005 |
defe | +0.005 |
Fram | +0.005 |
KEN | +0.005 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.005 |
cars | -0.005 |
Math | -0.005 |
pet | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -4.88e-03 |
Pos logit contributions | |
rals | +0.019 |
heimer | +0.018 |
cars | +0.018 |
Math | +0.017 |
pet | +0.017 |
Neg logit contributions | |
BOX | -0.019 |
icut | -0.019 |
defe | -0.019 |
Fram | -0.018 |
KEN | -0.018 |
Token | my |
Token ID | 616 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | family |
Token ID | 1641 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
rals | +0.011 |
heimer | +0.010 |
cars | +0.010 |
Math | +0.010 |
pet | +0.010 |
Neg logit contributions | |
BOX | -0.010 |
icut | -0.010 |
defe | -0.009 |
Fram | -0.009 |
KEN | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +2.43e-03 |
Pos logit contributions | |
BOX | +0.021 |
icut | +0.020 |
defe | +0.020 |
KEN | +0.020 |
Fram | +0.020 |
Neg logit contributions | |
rals | -0.025 |
heimer | -0.023 |
cars | -0.023 |
Math | -0.022 |
pet | -0.022 |
Token | I |
Token ID | 314 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.001 |
pet | -0.001 |
Token | established |
Token ID | 4920 |
Feature activation | -7.51e-02 |
Pos logit contributions | |
rals | +0.008 |
heimer | +0.007 |
cars | +0.007 |
Math | +0.007 |
pet | +0.007 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.007 |
KEN | -0.007 |
Fram | -0.007 |
Token | ourselves |
Token ID | 6731 |
Feature activation | +8.38e-02 |
Pos logit contributions | |
rals | +0.051 |
cars | +0.047 |
heimer | +0.047 |
Math | +0.044 |
pet | +0.044 |
Neg logit contributions | |
BOX | -0.051 |
icut | -0.050 |
defe | -0.049 |
KEN | -0.049 |
Fram | -0.049 |
Token | by |
Token ID | 416 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
BOX | +0.057 |
icut | +0.055 |
KEN | +0.055 |
defe | +0.055 |
Fram | +0.054 |
Neg logit contributions | |
rals | -0.056 |
heimer | -0.052 |
cars | -0.052 |
Math | -0.049 |
pet | -0.049 |
Token | un |
Token ID | 555 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
BOX | +0.015 |
icut | +0.015 |
defe | +0.014 |
KEN | +0.014 |
Fram | +0.014 |
Neg logit contributions | |
rals | -0.016 |
heimer | -0.015 |
cars | -0.015 |
Math | -0.014 |
pet | -0.014 |
Token | sw |
Token ID | 2032 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
rals | +0.017 |
heimer | +0.015 |
cars | +0.015 |
pet | +0.014 |
Math | +0.014 |
Neg logit contributions | |
BOX | -0.023 |
icut | -0.022 |
defe | -0.022 |
Fram | -0.022 |
KEN | -0.022 |
Token | erv |
Token ID | 712 |
Feature activation | +8.64e-03 |
Pos logit contributions | |
rals | +0.164 |
heimer | +0.154 |
cars | +0.152 |
pet | +0.145 |
Math | +0.145 |
Neg logit contributions | |
BOX | -0.125 |
icut | -0.121 |
defe | -0.119 |
Fram | -0.118 |
KEN | -0.117 |
Token | him |
Token ID | 683 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
rals | +0.015 |
heimer | +0.014 |
cars | +0.014 |
Math | +0.013 |
pet | +0.013 |
Neg logit contributions | |
BOX | -0.015 |
icut | -0.015 |
defe | -0.015 |
Fram | -0.015 |
KEN | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
BOX | +0.011 |
icut | +0.010 |
defe | +0.010 |
KEN | +0.010 |
Fram | +0.010 |
Neg logit contributions | |
rals | -0.010 |
heimer | -0.009 |
cars | -0.009 |
Math | -0.009 |
pet | -0.009 |
Token | Instead |
Token ID | 5455 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
rals | +0.008 |
heimer | +0.007 |
cars | +0.007 |
Math | +0.007 |
pet | +0.007 |
Neg logit contributions | |
BOX | -0.008 |
icut | -0.007 |
defe | -0.007 |
Fram | -0.007 |
KEN | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -5.09e-03 |
Pos logit contributions | |
rals | +0.009 |
heimer | +0.008 |
cars | +0.008 |
Math | +0.008 |
pet | +0.008 |
Neg logit contributions | |
BOX | -0.009 |
icut | -0.009 |
defe | -0.009 |
Fram | -0.008 |
KEN | -0.008 |
Token | he |
Token ID | 339 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | chose |
Token ID | 7690 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
BOX | +0.011 |
icut | +0.011 |
defe | +0.011 |
Fram | +0.011 |
KEN | +0.011 |
Neg logit contributions | |
rals | -0.011 |
heimer | -0.010 |
cars | -0.010 |
Math | -0.009 |
pet | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -6.87e-03 |
Pos logit contributions | |
rals | +0.031 |
heimer | +0.029 |
cars | +0.029 |
Math | +0.027 |
pet | +0.027 |
Neg logit contributions | |
BOX | -0.030 |
icut | -0.029 |
defe | -0.029 |
Fram | -0.029 |
KEN | -0.028 |
Token | stay |
Token ID | 2652 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.004 |
cars | +0.004 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
rals | +0.078 |
heimer | +0.072 |
cars | +0.071 |
Math | +0.068 |
pet | +0.067 |
Neg logit contributions | |
BOX | -0.076 |
icut | -0.074 |
defe | -0.073 |
Fram | -0.072 |
KEN | -0.072 |
Token | Europe |
Token ID | 2031 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
rals | +0.009 |
heimer | +0.008 |
cars | +0.008 |
Math | +0.008 |
pet | +0.008 |
Neg logit contributions | |
BOX | -0.009 |
icut | -0.009 |
defe | -0.009 |
Fram | -0.008 |
KEN | -0.008 |
Token | even |
Token ID | 772 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
BOX | +0.018 |
icut | +0.017 |
defe | +0.017 |
KEN | +0.017 |
Fram | +0.017 |
Neg logit contributions | |
rals | -0.017 |
heimer | -0.016 |
cars | -0.016 |
Math | -0.015 |
pet | -0.015 |
Token | himself |
Token ID | 2241 |
Feature activation | -8.74e-02 |
Pos logit contributions | |
BOX | +0.032 |
icut | +0.032 |
defe | +0.031 |
Fram | +0.031 |
KEN | +0.031 |
Neg logit contributions | |
rals | -0.026 |
heimer | -0.024 |
cars | -0.023 |
Math | -0.022 |
pet | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
rals | +0.059 |
heimer | +0.055 |
cars | +0.054 |
Math | +0.051 |
pet | +0.051 |
Neg logit contributions | |
BOX | -0.060 |
icut | -0.058 |
defe | -0.057 |
Fram | -0.057 |
KEN | -0.057 |
Token | Sy |
Token ID | 1632 |
Feature activation | +6.60e-02 |
Pos logit contributions | |
rals | +0.007 |
heimer | +0.007 |
cars | +0.007 |
pet | +0.006 |
Math | +0.006 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.006 |
defe | -0.006 |
Fram | -0.006 |
KEN | -0.006 |
Token | told |
Token ID | 1297 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
BOX | +0.050 |
icut | +0.049 |
defe | +0.049 |
Fram | +0.048 |
KEN | +0.048 |
Neg logit contributions | |
rals | -0.039 |
heimer | -0.036 |
cars | -0.036 |
Math | -0.033 |
pet | -0.033 |
Token | Action |
Token ID | 7561 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
rals | +0.023 |
heimer | +0.021 |
cars | +0.021 |
pet | +0.020 |
Math | +0.020 |
Neg logit contributions | |
BOX | -0.020 |
icut | -0.019 |
defe | -0.019 |
Fram | -0.019 |
KEN | -0.019 |
Token | News |
Token ID | 3000 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
BOX | +0.098 |
icut | +0.095 |
defe | +0.094 |
Fram | +0.093 |
KEN | +0.093 |
Neg logit contributions | |
rals | -0.059 |
heimer | -0.053 |
cars | -0.053 |
pet | -0.048 |
Math | -0.047 |
Token | she |
Token ID | 673 |
Feature activation | -4.99e-03 |
Pos logit contributions | |
rals | +0.101 |
heimer | +0.094 |
cars | +0.092 |
Math | +0.087 |
pet | +0.087 |
Neg logit contributions | |
BOX | -0.099 |
icut | -0.097 |
defe | -0.095 |
Fram | -0.094 |
KEN | -0.094 |
Token | was |
Token ID | 373 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | aware |
Token ID | 3910 |
Feature activation | -9.22e-02 |
Pos logit contributions | |
BOX | +0.010 |
icut | +0.010 |
defe | +0.010 |
KEN | +0.009 |
Fram | +0.009 |
Neg logit contributions | |
rals | -0.009 |
heimer | -0.008 |
cars | -0.008 |
Math | -0.007 |
pet | -0.007 |
Token | of |
Token ID | 286 |
Feature activation | -8.64e-03 |
Pos logit contributions | |
rals | +0.063 |
heimer | +0.059 |
cars | +0.058 |
Math | +0.055 |
pet | +0.055 |
Neg logit contributions | |
BOX | -0.062 |
icut | -0.060 |
defe | -0.060 |
Fram | -0.059 |
KEN | -0.059 |
Token | other |
Token ID | 584 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.006 |
cars | +0.006 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.006 |
icut | -0.006 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
BOX | +0.006 |
icut | +0.005 |
defe | +0.005 |
Fram | +0.005 |
KEN | +0.005 |
Neg logit contributions | |
rals | -0.006 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.005 |
pet | -0.005 |
Token | Trump |
Token ID | 6170 |
Feature activation | -5.79e-03 |
Pos logit contributions | |
BOX | +0.012 |
icut | +0.012 |
defe | +0.011 |
Fram | +0.011 |
KEN | +0.011 |
Neg logit contributions | |
rals | -0.011 |
heimer | -0.011 |
cars | -0.011 |
pet | -0.010 |
Math | -0.010 |
Token | is |
Token ID | 318 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.004 |
icut | -0.004 |
defe | -0.004 |
Fram | -0.004 |
KEN | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
BOX | +0.009 |
icut | +0.008 |
defe | +0.008 |
Fram | +0.008 |
KEN | +0.008 |
Neg logit contributions | |
rals | -0.008 |
heimer | -0.008 |
cars | -0.008 |
Math | -0.007 |
pet | -0.007 |
Token | light |
Token ID | 1657 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
BOX | +0.015 |
icut | +0.014 |
defe | +0.014 |
Fram | +0.014 |
KEN | +0.014 |
Neg logit contributions | |
rals | -0.014 |
heimer | -0.013 |
cars | -0.013 |
Math | -0.012 |
pet | -0.012 |
Token | ening |
Token ID | 3101 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
rals | +0.012 |
heimer | +0.012 |
cars | +0.011 |
Math | +0.011 |
pet | +0.011 |
Neg logit contributions | |
BOX | -0.013 |
icut | -0.012 |
defe | -0.012 |
Fram | -0.012 |
KEN | -0.012 |
Token | rod |
Token ID | 15299 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
rals | +0.013 |
heimer | +0.012 |
cars | +0.011 |
Math | +0.010 |
pet | +0.010 |
Neg logit contributions | |
BOX | -0.027 |
icut | -0.027 |
defe | -0.027 |
Fram | -0.026 |
KEN | -0.026 |
Token | for |
Token ID | 329 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
rals | +0.117 |
heimer | +0.109 |
cars | +0.108 |
Math | +0.102 |
pet | +0.102 |
Neg logit contributions | |
BOX | -0.117 |
icut | -0.114 |
defe | -0.113 |
Fram | -0.111 |
KEN | -0.111 |
Token | hate |
Token ID | 5465 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
BOX | +0.012 |
icut | +0.011 |
defe | +0.011 |
Fram | +0.011 |
KEN | +0.011 |
Neg logit contributions | |
rals | -0.011 |
heimer | -0.011 |
cars | -0.011 |
Math | -0.010 |
pet | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
rals | +0.028 |
heimer | +0.027 |
cars | +0.026 |
Math | +0.025 |
pet | +0.025 |
Neg logit contributions | |
BOX | -0.028 |
icut | -0.027 |
defe | -0.027 |
Fram | -0.026 |
KEN | -0.026 |
Token | He |
Token ID | 679 |
Feature activation | +8.59e-03 |
Pos logit contributions | |
BOX | +0.011 |
icut | +0.011 |
defe | +0.010 |
Fram | +0.010 |
KEN | +0.010 |
Neg logit contributions | |
rals | -0.011 |
heimer | -0.010 |
cars | -0.010 |
Math | -0.009 |
pet | -0.009 |
Token | t |
Token ID | 83 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.004 |
cars | +0.004 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.004 |
icut | -0.004 |
defe | -0.004 |
Fram | -0.004 |
KEN | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -5.81e-03 |
Pos logit contributions | |
rals | +0.011 |
heimer | +0.010 |
cars | +0.010 |
pet | +0.010 |
Math | +0.010 |
Neg logit contributions | |
BOX | -0.011 |
icut | -0.011 |
defe | -0.011 |
Fram | -0.011 |
KEN | -0.011 |
Token | A |
Token ID | 317 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.004 |
cars | +0.004 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.004 |
icut | -0.004 |
defe | -0.004 |
Fram | -0.004 |
KEN | -0.004 |
Token | US |
Token ID | 2937 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
BOX | +0.007 |
icut | +0.007 |
defe | +0.006 |
Fram | +0.006 |
KEN | +0.006 |
Neg logit contributions | |
rals | -0.008 |
heimer | -0.008 |
cars | -0.008 |
Math | -0.007 |
pet | -0.007 |
Token | 16 |
Token ID | 1467 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
BOX | +0.037 |
icut | +0.036 |
defe | +0.036 |
Fram | +0.035 |
KEN | +0.035 |
Neg logit contributions | |
rals | -0.036 |
heimer | -0.033 |
cars | -0.033 |
Math | -0.031 |
pet | -0.031 |
Token | Steele |
Token ID | 28549 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
rals | +0.011 |
heimer | +0.010 |
cars | +0.010 |
Math | +0.010 |
pet | +0.009 |
Neg logit contributions | |
BOX | -0.011 |
icut | -0.011 |
defe | -0.011 |
Fram | -0.011 |
KEN | -0.011 |
Token | Von |
Token ID | 26985 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
rals | +0.205 |
heimer | +0.197 |
cars | +0.193 |
Math | +0.188 |
pet | +0.186 |
Neg logit contributions | |
BOX | -0.098 |
defe | -0.093 |
icut | -0.093 |
Fram | -0.092 |
KEN | -0.090 |
Token | Hoff |
Token ID | 21890 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
BOX | +0.228 |
icut | +0.227 |
Fram | +0.219 |
defe | +0.219 |
KEN | +0.218 |
Neg logit contributions | |
rals | -0.150 |
heimer | -0.140 |
cars | -0.134 |
Math | -0.131 |
pet | -0.126 |
Token | ( |
Token ID | 357 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
rals | +0.045 |
heimer | +0.042 |
cars | +0.042 |
Math | +0.039 |
pet | +0.039 |
Neg logit contributions | |
BOX | -0.040 |
icut | -0.039 |
defe | -0.038 |
Fram | -0.038 |
KEN | -0.038 |
Token | Team |
Token ID | 15592 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
BOX | +0.015 |
icut | +0.015 |
defe | +0.015 |
Fram | +0.014 |
KEN | +0.014 |
Neg logit contributions | |
rals | -0.015 |
heimer | -0.014 |
cars | -0.014 |
pet | -0.013 |
Math | -0.013 |
Token | Chip |
Token ID | 17869 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
BOX | +0.013 |
icut | +0.013 |
defe | +0.013 |
Fram | +0.013 |
KEN | +0.013 |
Neg logit contributions | |
rals | -0.012 |
heimer | -0.012 |
cars | -0.011 |
pet | -0.011 |
Math | -0.011 |
Token | installer |
Token ID | 29124 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.006 |
cars | +0.006 |
Math | +0.006 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.006 |
defe | -0.006 |
Fram | -0.006 |
KEN | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
rals | +0.025 |
heimer | +0.024 |
cars | +0.023 |
Math | +0.022 |
pet | +0.022 |
Neg logit contributions | |
BOX | -0.026 |
icut | -0.026 |
defe | -0.025 |
KEN | -0.025 |
Fram | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +4.04e-03 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
Math | -0.001 |
pet | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +5.01e-03 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.003 |
Math | -0.002 |
pet | -0.002 |
Token | [ |
Token ID | 58 |
Feature activation | -9.59e-03 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
KEN | +0.003 |
Fram | +0.003 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.003 |
pet | -0.003 |
Math | -0.003 |
Token | Math |
Token ID | 37372 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.009 |
icut | -0.009 |
defe | -0.009 |
Fram | -0.009 |
KEN | -0.009 |
Token | ieu |
Token ID | 22304 |
Feature activation | +6.73e-03 |
Pos logit contributions | |
rals | +0.259 |
heimer | +0.242 |
cars | +0.236 |
Math | +0.223 |
pet | +0.222 |
Neg logit contributions | |
BOX | -0.287 |
icut | -0.279 |
defe | -0.277 |
Fram | -0.273 |
KEN | -0.273 |
Token | ] |
Token ID | 60 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
BOX | +0.004 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.005 |
cars | -0.005 |
Math | -0.004 |
pet | -0.004 |
Token | What |
Token ID | 1867 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
BOX | +0.021 |
icut | +0.021 |
defe | +0.020 |
Fram | +0.020 |
KEN | +0.020 |
Neg logit contributions | |
rals | -0.022 |
heimer | -0.020 |
cars | -0.020 |
Math | -0.019 |
pet | -0.019 |
Token | have |
Token ID | 423 |
Feature activation | +2.49e-03 |
Pos logit contributions | |
BOX | +0.024 |
icut | +0.024 |
defe | +0.023 |
Fram | +0.023 |
KEN | +0.023 |
Neg logit contributions | |
rals | -0.024 |
heimer | -0.023 |
cars | -0.022 |
Math | -0.021 |
pet | -0.021 |
Token | you |
Token ID | 345 |
Feature activation | -4.83e-03 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.001 |
pet | -0.001 |
Token | 2016 |
Token ID | 1584 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
BOX | +0.005 |
icut | +0.004 |
defe | +0.004 |
Fram | +0.004 |
KEN | +0.004 |
Neg logit contributions | |
rals | -0.005 |
heimer | -0.004 |
cars | -0.004 |
Math | -0.004 |
pet | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.47e-04 |
Pos logit contributions | |
BOX | +0.017 |
icut | +0.017 |
defe | +0.017 |
Fram | +0.016 |
KEN | +0.016 |
Neg logit contributions | |
rals | -0.017 |
heimer | -0.016 |
cars | -0.016 |
Math | -0.015 |
pet | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +4.75e-03 |
Pos logit contributions | |
rals | +0.000 |
heimer | +0.000 |
cars | +0.000 |
Math | +0.000 |
pet | +0.000 |
Neg logit contributions | |
BOX | -0.000 |
icut | -0.000 |
defe | -0.000 |
Fram | -0.000 |
KEN | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +1.55e-03 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.003 |
heimer | -0.003 |
cars | -0.003 |
Math | -0.003 |
pet | -0.003 |
Token | Media |
Token ID | 13152 |
Feature activation | +9.65e-02 |
Pos logit contributions | |
BOX | +0.001 |
icut | +0.001 |
defe | +0.001 |
Fram | +0.001 |
KEN | +0.001 |
Neg logit contributions | |
rals | -0.001 |
heimer | -0.001 |
cars | -0.001 |
pet | -0.001 |
Math | -0.001 |
Token | works |
Token ID | 5225 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
BOX | +0.065 |
icut | +0.064 |
defe | +0.063 |
KEN | +0.062 |
Fram | +0.062 |
Neg logit contributions | |
rals | -0.065 |
heimer | -0.061 |
cars | -0.060 |
Math | -0.057 |
pet | -0.057 |
Token | ' |
Token ID | 6 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
rals | +0.212 |
heimer | +0.198 |
cars | +0.195 |
Math | +0.184 |
pet | +0.183 |
Neg logit contributions | |
BOX | -0.218 |
icut | -0.213 |
defe | -0.211 |
Fram | -0.208 |
KEN | -0.207 |
Token | other |
Token ID | 584 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
rals | +0.074 |
heimer | +0.069 |
cars | +0.068 |
Math | +0.064 |
pet | +0.064 |
Neg logit contributions | |
BOX | -0.076 |
icut | -0.074 |
defe | -0.073 |
Fram | -0.072 |
KEN | -0.072 |
Token | reality |
Token ID | 3950 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
rals | +0.022 |
heimer | +0.020 |
cars | +0.020 |
Math | +0.019 |
pet | +0.018 |
Neg logit contributions | |
BOX | -0.023 |
icut | -0.022 |
defe | -0.022 |
Fram | -0.022 |
KEN | -0.021 |
Token | offerings |
Token ID | 18369 |
Feature activation | -7.35e-03 |
Pos logit contributions | |
BOX | +0.016 |
icut | +0.016 |
defe | +0.016 |
Fram | +0.015 |
KEN | +0.015 |
Neg logit contributions | |
rals | -0.017 |
heimer | -0.016 |
cars | -0.015 |
Math | -0.015 |
pet | -0.014 |
Token | Grand |
Token ID | 5675 |
Feature activation | -7.68e-02 |
Pos logit contributions | |
rals | +0.005 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
BOX | +0.031 |
icut | +0.030 |
defe | +0.030 |
Fram | +0.030 |
KEN | +0.029 |
Neg logit contributions | |
rals | -0.030 |
heimer | -0.028 |
cars | -0.028 |
Math | -0.026 |
pet | -0.026 |
Token | stay |
Token ID | 2652 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
rals | +0.009 |
heimer | +0.009 |
cars | +0.008 |
Math | +0.008 |
pet | +0.008 |
Neg logit contributions | |
BOX | -0.009 |
icut | -0.009 |
defe | -0.009 |
Fram | -0.009 |
KEN | -0.009 |
Token | calm |
Token ID | 9480 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
rals | +0.069 |
heimer | +0.064 |
cars | +0.062 |
pet | +0.058 |
Math | +0.058 |
Neg logit contributions | |
BOX | -0.087 |
icut | -0.085 |
defe | -0.085 |
Fram | -0.083 |
KEN | -0.083 |
Token | and |
Token ID | 290 |
Feature activation | -6.30e-03 |
Pos logit contributions | |
rals | +0.079 |
heimer | +0.073 |
cars | +0.072 |
Math | +0.068 |
pet | +0.067 |
Neg logit contributions | |
BOX | -0.093 |
icut | -0.091 |
defe | -0.089 |
Fram | -0.089 |
KEN | -0.088 |
Token | not |
Token ID | 407 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.004 |
cars | +0.004 |
Math | +0.004 |
pet | +0.004 |
Neg logit contributions | |
BOX | -0.004 |
icut | -0.004 |
defe | -0.004 |
Fram | -0.004 |
KEN | -0.004 |
Token | alarm |
Token ID | 10436 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
rals | +0.009 |
heimer | +0.009 |
cars | +0.008 |
Math | +0.008 |
pet | +0.008 |
Neg logit contributions | |
BOX | -0.010 |
icut | -0.009 |
defe | -0.009 |
Fram | -0.009 |
KEN | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -3.98e-03 |
Pos logit contributions | |
rals | +0.177 |
heimer | +0.165 |
cars | +0.162 |
Math | +0.153 |
pet | +0.153 |
Neg logit contributions | |
BOX | -0.183 |
icut | -0.179 |
defe | -0.176 |
Fram | -0.174 |
KEN | -0.173 |
Token | children |
Token ID | 1751 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.003 |
cars | +0.002 |
pet | +0.002 |
Math | +0.002 |
Neg logit contributions | |
BOX | -0.003 |
icut | -0.003 |
defe | -0.003 |
Fram | -0.003 |
KEN | -0.003 |
Token | .' |
Token ID | 2637 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
BOX | +0.008 |
icut | +0.008 |
defe | +0.008 |
Fram | +0.008 |
KEN | +0.008 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.006 |
cars | -0.006 |
Math | -0.006 |
pet | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
rals | +0.148 |
heimer | +0.138 |
cars | +0.136 |
pet | +0.128 |
Math | +0.128 |
Neg logit contributions | |
BOX | -0.134 |
icut | -0.130 |
defe | -0.128 |
Fram | -0.127 |
KEN | -0.126 |
Token |
|
Token ID | 198 |
Feature activation | -2.17e-03 |
Pos logit contributions | |
rals | +0.008 |
heimer | +0.007 |
cars | +0.007 |
pet | +0.007 |
Math | +0.007 |
Neg logit contributions | |
BOX | -0.007 |
icut | -0.007 |
defe | -0.007 |
Fram | -0.007 |
KEN | -0.007 |
Token | president |
Token ID | 1893 |
Feature activation | +8.50e-02 |
Pos logit contributions | |
BOX | +0.043 |
icut | +0.042 |
defe | +0.041 |
KEN | +0.041 |
Fram | +0.041 |
Neg logit contributions | |
rals | -0.048 |
heimer | -0.045 |
cars | -0.044 |
Math | -0.043 |
pet | -0.042 |
Token | after |
Token ID | 706 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
BOX | +0.058 |
icut | +0.057 |
defe | +0.056 |
KEN | +0.055 |
Fram | +0.055 |
Neg logit contributions | |
rals | -0.057 |
heimer | -0.054 |
cars | -0.053 |
Math | -0.050 |
pet | -0.049 |
Token | eight |
Token ID | 3624 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
BOX | +0.011 |
icut | +0.011 |
defe | +0.011 |
KEN | +0.011 |
Fram | +0.011 |
Neg logit contributions | |
rals | -0.011 |
heimer | -0.011 |
cars | -0.011 |
Math | -0.010 |
pet | -0.010 |
Token | years |
Token ID | 812 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
rals | +0.014 |
heimer | +0.013 |
cars | +0.013 |
Math | +0.013 |
pet | +0.013 |
Neg logit contributions | |
BOX | -0.013 |
icut | -0.013 |
defe | -0.013 |
Fram | -0.013 |
KEN | -0.013 |
Token | of |
Token ID | 286 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
BOX | +0.030 |
icut | +0.029 |
defe | +0.029 |
KEN | +0.029 |
Fram | +0.029 |
Neg logit contributions | |
rals | -0.031 |
heimer | -0.029 |
cars | -0.028 |
Math | -0.027 |
pet | -0.026 |
Token | Nixon |
Token ID | 16642 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
BOX | +0.007 |
icut | +0.007 |
defe | +0.007 |
Fram | +0.007 |
KEN | +0.007 |
Neg logit contributions | |
rals | -0.007 |
heimer | -0.007 |
cars | -0.007 |
Math | -0.006 |
pet | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +3.46e-03 |
Pos logit contributions | |
rals | +0.216 |
heimer | +0.201 |
cars | +0.198 |
Math | +0.186 |
pet | +0.186 |
Neg logit contributions | |
BOX | -0.211 |
icut | -0.206 |
defe | -0.203 |
Fram | -0.201 |
KEN | -0.201 |
Token | Ford |
Token ID | 8092 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.002 |
pet | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
rals | +0.172 |
heimer | +0.161 |
cars | +0.158 |
Math | +0.149 |
pet | +0.148 |
Neg logit contributions | |
BOX | -0.173 |
icut | -0.169 |
defe | -0.166 |
KEN | -0.164 |
Fram | -0.164 |
Token |
|
Token ID | 198 |
Feature activation | +5.22e-03 |
Pos logit contributions | |
BOX | +0.002 |
icut | +0.002 |
defe | +0.002 |
Fram | +0.002 |
KEN | +0.002 |
Neg logit contributions | |
rals | -0.002 |
heimer | -0.002 |
cars | -0.002 |
Math | -0.002 |
pet | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +6.62e-03 |
Pos logit contributions | |
BOX | +0.003 |
icut | +0.003 |
defe | +0.003 |
Fram | +0.003 |
KEN | +0.003 |
Neg logit contributions | |
rals | -0.004 |
heimer | -0.003 |
cars | -0.003 |
Math | -0.003 |
pet | -0.003 |
Token | mentioned |
Token ID | 4750 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
rals | +0.006 |
heimer | +0.005 |
cars | +0.005 |
Math | +0.005 |
pet | +0.005 |
Neg logit contributions | |
BOX | -0.005 |
icut | -0.005 |
defe | -0.005 |
Fram | -0.005 |
KEN | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | -5.77e-03 |
Pos logit contributions | |
rals | +0.021 |
heimer | +0.020 |
cars | +0.019 |
Math | +0.018 |
pet | +0.018 |
Neg logit contributions | |
BOX | -0.021 |
icut | -0.021 |
defe | -0.020 |
KEN | -0.020 |
Fram | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | -1.53e-03 |
Pos logit contributions | |
rals | +0.004 |
heimer | +0.004 |
cars | +0.004 |
Math | +0.003 |
pet | +0.003 |
Neg logit contributions | |
BOX | -0.004 |
icut | -0.004 |
defe | -0.004 |
Fram | -0.004 |
KEN | -0.004 |
Token | Sunni |
Token ID | 18592 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | had |
Token ID | 550 |
Feature activation | -9.42e-03 |
Pos logit contributions | |
rals | +0.052 |
heimer | +0.049 |
cars | +0.048 |
Math | +0.045 |
pet | +0.045 |
Neg logit contributions | |
BOX | -0.052 |
icut | -0.050 |
defe | -0.050 |
Fram | -0.049 |
KEN | -0.049 |
Token | ith |
Token ID | 342 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
rals | +0.003 |
heimer | +0.003 |
cars | +0.003 |
Math | +0.002 |
pet | +0.002 |
Neg logit contributions | |
BOX | -0.009 |
icut | -0.009 |
defe | -0.009 |
Fram | -0.009 |
KEN | -0.009 |
Token | collections |
Token ID | 17268 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
rals | +0.196 |
heimer | +0.183 |
cars | +0.180 |
Math | +0.172 |
pet | +0.171 |
Neg logit contributions | |
BOX | -0.198 |
icut | -0.193 |
defe | -0.189 |
KEN | -0.188 |
Fram | -0.188 |
Token | , |
Token ID | 11 |
Feature activation | -2.18e-03 |
Pos logit contributions | |
rals | +0.014 |
heimer | +0.013 |
cars | +0.012 |
Math | +0.012 |
pet | +0.012 |
Neg logit contributions | |
BOX | -0.013 |
icut | -0.013 |
defe | -0.013 |
Fram | -0.013 |
KEN | -0.013 |
Token | stating |
Token ID | 12316 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |
Token | : |
Token ID | 25 |
Feature activation | -1.42e-03 |
Pos logit contributions | |
BOX | +0.026 |
icut | +0.025 |
defe | +0.025 |
Fram | +0.024 |
KEN | +0.024 |
Neg logit contributions | |
rals | -0.025 |
heimer | -0.023 |
cars | -0.023 |
Math | -0.022 |
pet | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -2.93e-03 |
Pos logit contributions | |
rals | +0.001 |
heimer | +0.001 |
cars | +0.001 |
Math | +0.001 |
pet | +0.001 |
Neg logit contributions | |
BOX | -0.001 |
icut | -0.001 |
defe | -0.001 |
Fram | -0.001 |
KEN | -0.001 |