Token | in |
Token ID | 287 |
Feature activation | -4.28e-03 |
Pos logit contributions | |
Lum | +0.019 |
Beam | +0.018 |
Lau | +0.017 |
Layer | +0.017 |
Meng | +0.016 |
Neg logit contributions | |
Augustus | -0.021 |
Cyrus | -0.019 |
Elias | -0.019 |
Jace | -0.018 |
rabbi | -0.018 |
Token | foreign |
Token ID | 3215 |
Feature activation | -7.99e-03 |
Pos logit contributions | |
Lum | +0.004 |
Beam | +0.004 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.004 |
Token | policy |
Token ID | 2450 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.007 |
Lau | +0.007 |
Layer | +0.007 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.008 |
Cyrus | -0.007 |
Elias | -0.007 |
Jace | -0.007 |
rabbi | -0.007 |
Token | are |
Token ID | 389 |
Feature activation | -1.78e-03 |
Pos logit contributions | |
Augustus | +0.036 |
Cyrus | +0.033 |
Elias | +0.032 |
Jace | +0.031 |
rabbi | +0.031 |
Neg logit contributions | |
Lum | -0.032 |
Beam | -0.031 |
Lau | -0.029 |
Layer | -0.029 |
Meng | -0.028 |
Token | lux |
Token ID | 10632 |
Feature activation | +4.20e-03 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | uries |
Token ID | 4740 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
Augustus | +0.004 |
Cyrus | +0.003 |
Elias | +0.003 |
Jace | +0.003 |
rabbi | +0.003 |
Neg logit contributions | |
Lum | -0.004 |
Beam | -0.004 |
Lau | -0.004 |
Layer | -0.004 |
Meng | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
Augustus | +0.303 |
Cyrus | +0.274 |
Elias | +0.270 |
Jace | +0.264 |
rabbi | +0.261 |
Neg logit contributions | |
Lum | -0.305 |
Beam | -0.293 |
Lau | -0.277 |
Layer | -0.272 |
Meng | -0.263 |
Token | can |
Token ID | 460 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
Augustus | +0.023 |
Cyrus | +0.021 |
Elias | +0.021 |
Jace | +0.020 |
rabbi | +0.020 |
Neg logit contributions | |
Lum | -0.020 |
Beam | -0.019 |
Lau | -0.018 |
Layer | -0.018 |
Meng | -0.017 |
Token | no |
Token ID | 645 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
Augustus | +0.027 |
Cyrus | +0.024 |
Elias | +0.024 |
Jace | +0.023 |
rabbi | +0.023 |
Neg logit contributions | |
Lum | -0.024 |
Beam | -0.023 |
Lau | -0.021 |
Layer | -0.021 |
Meng | -0.020 |
Token | longer |
Token ID | 2392 |
Feature activation | +7.93e-02 |
Pos logit contributions | |
Augustus | +0.012 |
Cyrus | +0.011 |
Elias | +0.011 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.009 |
Beam | -0.009 |
Lau | -0.008 |
Layer | -0.008 |
Meng | -0.008 |
Token | afford |
Token ID | 5368 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
Augustus | +0.082 |
Cyrus | +0.075 |
Elias | +0.074 |
Jace | +0.072 |
rabbi | +0.072 |
Neg logit contributions | |
Lum | -0.068 |
Beam | -0.065 |
Lau | -0.061 |
Layer | -0.060 |
Meng | -0.058 |
Token | ️ |
Token ID | 37929 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Augustus | +0.048 |
Cyrus | +0.044 |
Elias | +0.043 |
Jace | +0.042 |
rabbi | +0.042 |
Neg logit contributions | |
Lum | -0.044 |
Beam | -0.042 |
Lau | -0.039 |
Layer | -0.038 |
Meng | -0.037 |
Token | Be |
Token ID | 1355 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
Augustus | +0.106 |
Cyrus | +0.097 |
Elias | +0.096 |
Jace | +0.093 |
rabbi | +0.093 |
Neg logit contributions | |
Lum | -0.087 |
Beam | -0.083 |
Lau | -0.078 |
Layer | -0.076 |
Meng | -0.073 |
Token | safe |
Token ID | 3338 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
Augustus | +0.030 |
Cyrus | +0.027 |
Elias | +0.027 |
Jace | +0.026 |
rabbi | +0.026 |
Neg logit contributions | |
Lum | -0.029 |
Beam | -0.028 |
Lau | -0.026 |
Layer | -0.026 |
Meng | -0.025 |
Token | pic |
Token ID | 8301 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
Augustus | +0.025 |
Cyrus | +0.023 |
Elias | +0.022 |
Jace | +0.022 |
rabbi | +0.022 |
Neg logit contributions | |
Lum | -0.022 |
Beam | -0.021 |
Lau | -0.019 |
Layer | -0.019 |
Meng | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -4.89e-03 |
Pos logit contributions | |
Lum | +0.035 |
Beam | +0.034 |
Lau | +0.032 |
Layer | +0.031 |
Meng | +0.030 |
Neg logit contributions | |
Augustus | -0.037 |
Cyrus | -0.034 |
Elias | -0.033 |
rabbi | -0.033 |
Jace | -0.033 |
Token | twitter |
Token ID | 6956 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
Lum | +0.006 |
Beam | +0.006 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.003 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -6.00e-03 |
Pos logit contributions | |
Augustus | +0.281 |
Cyrus | +0.254 |
Elias | +0.251 |
Jace | +0.244 |
rabbi | +0.243 |
Neg logit contributions | |
Lum | -0.281 |
Beam | -0.270 |
Lau | -0.255 |
Layer | -0.250 |
Meng | -0.242 |
Token | com |
Token ID | 785 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
Lum | +0.006 |
Beam | +0.006 |
Lau | +0.006 |
Layer | +0.005 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.005 |
Cyrus | -0.005 |
Elias | -0.005 |
Jace | -0.005 |
rabbi | -0.004 |
Token | / |
Token ID | 14 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
Lum | +0.049 |
Beam | +0.047 |
Lau | +0.044 |
Layer | +0.043 |
Meng | +0.042 |
Neg logit contributions | |
Augustus | -0.047 |
Cyrus | -0.042 |
Elias | -0.042 |
Jace | -0.040 |
rabbi | -0.040 |
Token | 74 |
Token ID | 4524 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
Lum | +0.014 |
Beam | +0.013 |
Lau | +0.013 |
Layer | +0.012 |
Meng | +0.012 |
Neg logit contributions | |
Augustus | -0.014 |
Cyrus | -0.012 |
Elias | -0.012 |
Jace | -0.012 |
rabbi | -0.012 |
Token | E |
Token ID | 36 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
Augustus | +0.022 |
Cyrus | +0.020 |
Elias | +0.020 |
Jace | +0.019 |
rabbi | +0.019 |
Neg logit contributions | |
Lum | -0.023 |
Beam | -0.022 |
Lau | -0.021 |
Layer | -0.020 |
Meng | -0.019 |
Token | � |
Token ID | 250 |
Feature activation | -4.03e-03 |
Pos logit contributions | |
Lum | +0.012 |
Beam | +0.012 |
Lau | +0.011 |
Layer | +0.011 |
Meng | +0.011 |
Neg logit contributions | |
Augustus | -0.009 |
Cyrus | -0.008 |
Elias | -0.008 |
Jace | -0.008 |
rabbi | -0.008 |
Token | The |
Token ID | 464 |
Feature activation | +8.33e-03 |
Pos logit contributions | |
Lum | +0.004 |
Beam | +0.003 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.003 |
Token | big |
Token ID | 1263 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
Augustus | +0.008 |
Cyrus | +0.008 |
Elias | +0.008 |
Jace | +0.007 |
rabbi | +0.007 |
Neg logit contributions | |
Lum | -0.007 |
Beam | -0.007 |
Lau | -0.007 |
Layer | -0.007 |
Meng | -0.006 |
Token | ( |
Token ID | 357 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
Augustus | +0.018 |
Cyrus | +0.016 |
Elias | +0.016 |
Jace | +0.016 |
rabbi | +0.016 |
Neg logit contributions | |
Lum | -0.014 |
Beam | -0.013 |
Lau | -0.012 |
Layer | -0.012 |
Meng | -0.012 |
Token | diff |
Token ID | 26069 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
Lum | +0.032 |
Beam | +0.030 |
Lau | +0.029 |
Layer | +0.028 |
Meng | +0.027 |
Neg logit contributions | |
Augustus | -0.031 |
Cyrus | -0.028 |
Elias | -0.028 |
Jace | -0.027 |
rabbi | -0.027 |
Token | erence |
Token ID | 1945 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
Lum | +0.028 |
Beam | +0.027 |
Lau | +0.026 |
Layer | +0.025 |
Meng | +0.024 |
Neg logit contributions | |
Augustus | -0.024 |
Cyrus | -0.022 |
Elias | -0.021 |
Jace | -0.021 |
rabbi | -0.020 |
Token | ) |
Token ID | 8 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
Augustus | +0.294 |
Cyrus | +0.267 |
Elias | +0.263 |
Jace | +0.257 |
rabbi | +0.255 |
Neg logit contributions | |
Lum | -0.266 |
Beam | -0.254 |
Lau | -0.239 |
Layer | -0.234 |
Meng | -0.226 |
Token | is |
Token ID | 318 |
Feature activation | -8.01e-03 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.010 |
Beam | -0.010 |
Lau | -0.009 |
Layer | -0.009 |
Meng | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
Lum | +0.008 |
Beam | +0.007 |
Lau | +0.007 |
Layer | +0.007 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.007 |
Cyrus | -0.007 |
Elias | -0.007 |
Jace | -0.006 |
rabbi | -0.006 |
Token | coming |
Token ID | 2406 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
Lum | +0.014 |
Beam | +0.013 |
Lau | +0.013 |
Layer | +0.012 |
Meng | +0.012 |
Neg logit contributions | |
Augustus | -0.015 |
Cyrus | -0.013 |
Elias | -0.013 |
Jace | -0.013 |
rabbi | -0.013 |
Token | back |
Token ID | 736 |
Feature activation | -3.53e-03 |
Pos logit contributions | |
Lum | +0.012 |
Beam | +0.011 |
Lau | +0.011 |
Layer | +0.011 |
Meng | +0.010 |
Neg logit contributions | |
Augustus | -0.013 |
Cyrus | -0.011 |
Elias | -0.011 |
Jace | -0.011 |
rabbi | -0.011 |
Token | 8 |
Token ID | 23 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
Lum | +0.089 |
Beam | +0.085 |
Lau | +0.081 |
Layer | +0.079 |
Meng | +0.077 |
Neg logit contributions | |
Augustus | -0.087 |
Cyrus | -0.078 |
Elias | -0.077 |
Jace | -0.075 |
rabbi | -0.074 |
Token | b |
Token ID | 65 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
Augustus | +0.013 |
Cyrus | +0.012 |
Elias | +0.011 |
Jace | +0.011 |
rabbi | +0.011 |
Neg logit contributions | |
Lum | -0.013 |
Beam | -0.013 |
Lau | -0.012 |
Layer | -0.012 |
Meng | -0.011 |
Token | ET |
Token ID | 2767 |
Feature activation | -6.96e-02 |
Pos logit contributions | |
Augustus | +0.015 |
Cyrus | +0.013 |
Elias | +0.013 |
Jace | +0.013 |
rabbi | +0.013 |
Neg logit contributions | |
Lum | -0.015 |
Beam | -0.015 |
Lau | -0.014 |
Layer | -0.013 |
Meng | -0.013 |
Token | pic |
Token ID | 8301 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
Lum | +0.058 |
Beam | +0.055 |
Lau | +0.052 |
Layer | +0.050 |
Meng | +0.049 |
Neg logit contributions | |
Augustus | -0.074 |
Cyrus | -0.068 |
Elias | -0.067 |
Jace | -0.066 |
rabbi | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
Lum | +0.052 |
Beam | +0.050 |
Lau | +0.047 |
Layer | +0.046 |
Meng | +0.045 |
Neg logit contributions | |
Augustus | -0.055 |
Cyrus | -0.050 |
Elias | -0.049 |
rabbi | -0.048 |
Jace | -0.048 |
Token | twitter |
Token ID | 6956 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
Lum | +0.020 |
Beam | +0.019 |
Lau | +0.019 |
Layer | +0.018 |
Meng | +0.018 |
Neg logit contributions | |
Augustus | -0.009 |
Cyrus | -0.008 |
Elias | -0.008 |
Jace | -0.007 |
rabbi | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
Augustus | +0.279 |
Cyrus | +0.253 |
Elias | +0.249 |
Jace | +0.242 |
rabbi | +0.241 |
Neg logit contributions | |
Lum | -0.279 |
Beam | -0.267 |
Lau | -0.253 |
Layer | -0.248 |
Meng | -0.240 |
Token | com |
Token ID | 785 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
Lum | +0.015 |
Beam | +0.014 |
Lau | +0.014 |
Layer | +0.013 |
Meng | +0.013 |
Neg logit contributions | |
Augustus | -0.013 |
Cyrus | -0.012 |
Elias | -0.011 |
Jace | -0.011 |
rabbi | -0.011 |
Token | / |
Token ID | 14 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
Lum | +0.058 |
Beam | +0.056 |
Lau | +0.053 |
Layer | +0.052 |
Meng | +0.050 |
Neg logit contributions | |
Augustus | -0.056 |
Cyrus | -0.050 |
Elias | -0.049 |
Jace | -0.048 |
rabbi | -0.048 |
Token | W |
Token ID | 54 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
Lum | +0.013 |
Beam | +0.012 |
Lau | +0.011 |
Layer | +0.011 |
Meng | +0.011 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.011 |
Jace | -0.011 |
rabbi | -0.011 |
Token | 8 |
Token ID | 23 |
Feature activation | +1.73e-03 |
Pos logit contributions | |
Lum | +0.027 |
Beam | +0.026 |
Lau | +0.025 |
Layer | +0.024 |
Meng | +0.024 |
Neg logit contributions | |
Augustus | -0.027 |
Cyrus | -0.024 |
Elias | -0.024 |
Jace | -0.023 |
rabbi | -0.023 |
Token | never |
Token ID | 1239 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
Augustus | +0.007 |
Cyrus | +0.006 |
Elias | +0.006 |
Jace | +0.006 |
rabbi | +0.006 |
Neg logit contributions | |
Lum | -0.006 |
Beam | -0.006 |
Lau | -0.006 |
Layer | -0.006 |
Meng | -0.005 |
Token | called |
Token ID | 1444 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.018 |
Elias | +0.017 |
Jace | +0.017 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.016 |
Meng | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +8.60e-03 |
Pos logit contributions | |
Augustus | +0.016 |
Cyrus | +0.014 |
Elias | +0.014 |
Jace | +0.014 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.014 |
Beam | -0.014 |
Lau | -0.013 |
Layer | -0.013 |
Meng | -0.012 |
Token | pic |
Token ID | 8301 |
Feature activation | -5.28e-02 |
Pos logit contributions | |
Augustus | +0.009 |
Cyrus | +0.008 |
Elias | +0.008 |
Jace | +0.008 |
rabbi | +0.008 |
Neg logit contributions | |
Lum | -0.008 |
Beam | -0.007 |
Lau | -0.007 |
Layer | -0.007 |
Meng | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
Lum | +0.049 |
Beam | +0.047 |
Lau | +0.044 |
Layer | +0.043 |
Meng | +0.042 |
Neg logit contributions | |
Augustus | -0.051 |
Cyrus | -0.047 |
Elias | -0.046 |
rabbi | -0.045 |
Jace | -0.045 |
Token | twitter |
Token ID | 6956 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
Lum | +0.032 |
Beam | +0.031 |
Lau | +0.030 |
Layer | +0.029 |
Meng | +0.028 |
Neg logit contributions | |
Augustus | -0.015 |
Cyrus | -0.013 |
Elias | -0.013 |
rabbi | -0.012 |
Jace | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Augustus | +0.275 |
Cyrus | +0.249 |
Elias | +0.246 |
Jace | +0.239 |
rabbi | +0.238 |
Neg logit contributions | |
Lum | -0.274 |
Beam | -0.263 |
Lau | -0.249 |
Layer | -0.244 |
Meng | -0.236 |
Token | com |
Token ID | 785 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
Lum | +0.013 |
Beam | +0.013 |
Lau | +0.012 |
Layer | +0.012 |
Meng | +0.012 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.011 |
rabbi | -0.010 |
Jace | -0.010 |
Token | / |
Token ID | 14 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
Lum | +0.060 |
Beam | +0.058 |
Lau | +0.055 |
Layer | +0.054 |
Meng | +0.052 |
Neg logit contributions | |
Augustus | -0.058 |
Cyrus | -0.053 |
Elias | -0.052 |
Jace | -0.051 |
rabbi | -0.051 |
Token | CS |
Token ID | 7902 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
Lum | +0.013 |
Beam | +0.012 |
Lau | +0.012 |
Layer | +0.011 |
Meng | +0.011 |
Neg logit contributions | |
Augustus | -0.013 |
Cyrus | -0.011 |
Elias | -0.011 |
Jace | -0.011 |
rabbi | -0.011 |
Token | 3 |
Token ID | 18 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
Augustus | +0.024 |
Cyrus | +0.022 |
Elias | +0.021 |
Jace | +0.021 |
rabbi | +0.021 |
Neg logit contributions | |
Lum | -0.025 |
Beam | -0.024 |
Lau | -0.023 |
Layer | -0.022 |
Meng | -0.022 |
Token | GO |
Token ID | 10351 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Augustus | +0.003 |
Cyrus | +0.002 |
Elias | +0.002 |
Jace | +0.002 |
rabbi | +0.002 |
Neg logit contributions | |
Lum | -0.002 |
Beam | -0.002 |
Lau | -0.002 |
Layer | -0.002 |
Meng | -0.002 |
Token | AL |
Token ID | 1847 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
Lum | +0.013 |
Beam | +0.012 |
Lau | +0.012 |
Layer | +0.012 |
Meng | +0.011 |
Neg logit contributions | |
Augustus | -0.014 |
Cyrus | -0.013 |
Elias | -0.013 |
Jace | -0.012 |
rabbi | -0.012 |
Token | ! |
Token ID | 0 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
Lum | +0.017 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.015 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.018 |
Cyrus | -0.016 |
Elias | -0.016 |
Jace | -0.016 |
rabbi | -0.016 |
Token | pic |
Token ID | 8301 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
Lum | +0.046 |
Beam | +0.044 |
Lau | +0.042 |
Layer | +0.041 |
Meng | +0.039 |
Neg logit contributions | |
Augustus | -0.056 |
Cyrus | -0.051 |
Elias | -0.050 |
Jace | -0.049 |
rabbi | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
Lum | +0.052 |
Beam | +0.050 |
Lau | +0.047 |
Layer | +0.046 |
Meng | +0.044 |
Neg logit contributions | |
Augustus | -0.055 |
Cyrus | -0.050 |
Elias | -0.050 |
rabbi | -0.049 |
Jace | -0.048 |
Token | twitter |
Token ID | 6956 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
Lum | +0.043 |
Beam | +0.042 |
Lau | +0.040 |
Layer | +0.039 |
Meng | +0.039 |
Neg logit contributions | |
Augustus | -0.021 |
Cyrus | -0.018 |
Elias | -0.017 |
rabbi | -0.017 |
Jace | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
Augustus | +0.272 |
Cyrus | +0.246 |
Elias | +0.242 |
Jace | +0.236 |
rabbi | +0.235 |
Neg logit contributions | |
Lum | -0.272 |
Beam | -0.261 |
Lau | -0.247 |
Layer | -0.242 |
Meng | -0.234 |
Token | com |
Token ID | 785 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
Lum | +0.027 |
Beam | +0.026 |
Lau | +0.025 |
Layer | +0.024 |
Meng | +0.024 |
Neg logit contributions | |
Augustus | -0.024 |
Cyrus | -0.021 |
Elias | -0.021 |
Jace | -0.020 |
rabbi | -0.020 |
Token | / |
Token ID | 14 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
Lum | +0.052 |
Beam | +0.050 |
Lau | +0.047 |
Layer | +0.046 |
Meng | +0.045 |
Neg logit contributions | |
Augustus | -0.049 |
Cyrus | -0.044 |
Elias | -0.044 |
Jace | -0.043 |
rabbi | -0.042 |
Token | E |
Token ID | 36 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
Lum | +0.021 |
Beam | +0.020 |
Lau | +0.019 |
Layer | +0.018 |
Meng | +0.018 |
Neg logit contributions | |
Augustus | -0.020 |
Cyrus | -0.018 |
Elias | -0.018 |
Jace | -0.017 |
rabbi | -0.017 |
Token | q |
Token ID | 80 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
Lum | +0.044 |
Beam | +0.043 |
Lau | +0.040 |
Layer | +0.040 |
Meng | +0.038 |
Neg logit contributions | |
Augustus | -0.041 |
Cyrus | -0.037 |
Elias | -0.036 |
Jace | -0.035 |
rabbi | -0.035 |
Token | them |
Token ID | 606 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.011 |
Jace | -0.010 |
rabbi | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
Augustus | +0.015 |
Cyrus | +0.014 |
Elias | +0.014 |
Jace | +0.013 |
rabbi | +0.013 |
Neg logit contributions | |
Lum | -0.015 |
Beam | -0.014 |
Lau | -0.014 |
Layer | -0.013 |
Meng | -0.013 |
Token | Also |
Token ID | 4418 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.017 |
Jace | +0.017 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.016 |
Beam | -0.016 |
Lau | -0.015 |
Layer | -0.014 |
Meng | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -2.59e-03 |
Pos logit contributions | |
Augustus | +0.021 |
Cyrus | +0.019 |
Elias | +0.018 |
Jace | +0.018 |
rabbi | +0.018 |
Neg logit contributions | |
Lum | -0.020 |
Beam | -0.019 |
Lau | -0.018 |
Layer | -0.018 |
Meng | -0.017 |
Token | Jub |
Token ID | 38075 |
Feature activation | -6.31e-03 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.003 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | ilee |
Token ID | 40626 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
Lum | +0.008 |
Beam | +0.008 |
Lau | +0.008 |
Layer | +0.008 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.003 |
Elias | -0.003 |
Jace | -0.003 |
rabbi | -0.003 |
Token | actually |
Token ID | 1682 |
Feature activation | -9.27e-04 |
Pos logit contributions | |
Augustus | +0.278 |
Cyrus | +0.252 |
Elias | +0.248 |
Jace | +0.243 |
rabbi | +0.240 |
Neg logit contributions | |
Lum | -0.263 |
Beam | -0.252 |
Lau | -0.238 |
Layer | -0.233 |
Meng | -0.225 |
Token | gets |
Token ID | 3011 |
Feature activation | +6.43e-03 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
Augustus | +0.006 |
Cyrus | +0.006 |
Elias | +0.006 |
Jace | +0.006 |
rabbi | +0.006 |
Neg logit contributions | |
Lum | -0.006 |
Beam | -0.005 |
Lau | -0.005 |
Layer | -0.005 |
Meng | -0.005 |
Token | use |
Token ID | 779 |
Feature activation | +3.62e-03 |
Pos logit contributions | |
Lum | +0.012 |
Beam | +0.012 |
Lau | +0.011 |
Layer | +0.011 |
Meng | +0.010 |
Neg logit contributions | |
Augustus | -0.013 |
Cyrus | -0.012 |
Elias | -0.012 |
Jace | -0.012 |
rabbi | -0.012 |
Token | her |
Token ID | 607 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
Augustus | +0.004 |
Cyrus | +0.003 |
Elias | +0.003 |
Jace | +0.003 |
rabbi | +0.003 |
Neg logit contributions | |
Lum | -0.003 |
Beam | -0.003 |
Lau | -0.003 |
Layer | -0.003 |
Meng | -0.003 |
Token | ju |
Token ID | 7544 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
Lum | +0.004 |
Beam | +0.003 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.003 |
Elias | -0.003 |
Jace | -0.003 |
rabbi | -0.003 |
Token | illet |
Token ID | 32512 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
Augustus | +0.013 |
Cyrus | +0.012 |
Elias | +0.012 |
Jace | +0.011 |
rabbi | +0.011 |
Neg logit contributions | |
Lum | -0.015 |
Beam | -0.014 |
Lau | -0.014 |
Layer | -0.013 |
Meng | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
Augustus | +0.038 |
Cyrus | +0.035 |
Elias | +0.034 |
Jace | +0.034 |
rabbi | +0.033 |
Neg logit contributions | |
Lum | -0.035 |
Beam | -0.033 |
Lau | -0.032 |
Layer | -0.031 |
Meng | -0.030 |
Token | pic |
Token ID | 8301 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
Augustus | +0.022 |
Cyrus | +0.020 |
Elias | +0.020 |
Jace | +0.019 |
rabbi | +0.019 |
Neg logit contributions | |
Lum | -0.016 |
Beam | -0.016 |
Lau | -0.015 |
Layer | -0.014 |
Meng | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -5.67e-03 |
Pos logit contributions | |
Lum | +0.027 |
Beam | +0.025 |
Lau | +0.024 |
Layer | +0.023 |
Meng | +0.023 |
Neg logit contributions | |
Augustus | -0.028 |
Cyrus | -0.026 |
Elias | -0.025 |
rabbi | -0.025 |
Jace | -0.025 |
Token | twitter |
Token ID | 6956 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.007 |
Lau | +0.007 |
Layer | +0.007 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.003 |
Cyrus | -0.003 |
Elias | -0.003 |
rabbi | -0.003 |
Jace | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +4.00e-03 |
Pos logit contributions | |
Augustus | +0.273 |
Cyrus | +0.247 |
Elias | +0.244 |
Jace | +0.237 |
rabbi | +0.237 |
Neg logit contributions | |
Lum | -0.268 |
Beam | -0.257 |
Lau | -0.243 |
Layer | -0.238 |
Meng | -0.230 |
Token | com |
Token ID | 785 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
Augustus | +0.004 |
Cyrus | +0.003 |
Elias | +0.003 |
Jace | +0.003 |
rabbi | +0.003 |
Neg logit contributions | |
Lum | -0.004 |
Beam | -0.004 |
Lau | -0.004 |
Layer | -0.004 |
Meng | -0.003 |
Token | / |
Token ID | 14 |
Feature activation | +5.19e-03 |
Pos logit contributions | |
Lum | +0.044 |
Beam | +0.042 |
Lau | +0.040 |
Layer | +0.039 |
Meng | +0.038 |
Neg logit contributions | |
Augustus | -0.043 |
Cyrus | -0.039 |
Elias | -0.038 |
Jace | -0.037 |
rabbi | -0.037 |
Token | u |
Token ID | 84 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
Augustus | +0.005 |
Cyrus | +0.004 |
Elias | +0.004 |
Jace | +0.004 |
rabbi | +0.004 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.005 |
Lau | -0.005 |
Layer | -0.004 |
Meng | -0.004 |
Token | I |
Token ID | 40 |
Feature activation | +3.09e-03 |
Pos logit contributions | |
Augustus | +0.035 |
Cyrus | +0.031 |
Elias | +0.031 |
Jace | +0.030 |
rabbi | +0.030 |
Neg logit contributions | |
Lum | -0.036 |
Beam | -0.035 |
Lau | -0.033 |
Layer | -0.032 |
Meng | -0.031 |
Token | Deb |
Token ID | 16587 |
Feature activation | -5.69e-03 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | ates |
Token ID | 689 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
Lum | +0.006 |
Beam | +0.006 |
Lau | +0.005 |
Layer | +0.005 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.005 |
Cyrus | -0.005 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.004 |
Token | 2016 |
Token ID | 5304 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
Augustus | +0.053 |
Cyrus | +0.048 |
Elias | +0.048 |
Jace | +0.046 |
rabbi | +0.046 |
Neg logit contributions | |
Lum | -0.048 |
Beam | -0.045 |
Lau | -0.043 |
Layer | -0.042 |
Meng | -0.041 |
Token | pic |
Token ID | 8301 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
Augustus | +0.027 |
Cyrus | +0.025 |
Elias | +0.025 |
Jace | +0.024 |
rabbi | +0.024 |
Neg logit contributions | |
Lum | -0.023 |
Beam | -0.022 |
Lau | -0.020 |
Layer | -0.020 |
Meng | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
Lum | +0.033 |
Beam | +0.032 |
Lau | +0.030 |
Layer | +0.029 |
Meng | +0.028 |
Neg logit contributions | |
Augustus | -0.035 |
Cyrus | -0.032 |
Elias | -0.032 |
rabbi | -0.031 |
Jace | -0.031 |
Token | twitter |
Token ID | 6956 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
Lum | +0.017 |
Beam | +0.017 |
Lau | +0.016 |
Layer | +0.016 |
Meng | +0.015 |
Neg logit contributions | |
Augustus | -0.008 |
Cyrus | -0.007 |
Elias | -0.007 |
rabbi | -0.007 |
Jace | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -2.69e-03 |
Pos logit contributions | |
Augustus | +0.271 |
Cyrus | +0.246 |
Elias | +0.242 |
Jace | +0.236 |
rabbi | +0.235 |
Neg logit contributions | |
Lum | -0.267 |
Beam | -0.256 |
Lau | -0.242 |
Layer | -0.237 |
Meng | -0.229 |
Token | com |
Token ID | 785 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | / |
Token ID | 14 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
Lum | +0.047 |
Beam | +0.045 |
Lau | +0.043 |
Layer | +0.042 |
Meng | +0.041 |
Neg logit contributions | |
Augustus | -0.046 |
Cyrus | -0.042 |
Elias | -0.041 |
Jace | -0.040 |
rabbi | -0.040 |
Token | j |
Token ID | 73 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
Lum | +0.013 |
Beam | +0.013 |
Lau | +0.012 |
Layer | +0.012 |
Meng | +0.012 |
Neg logit contributions | |
Augustus | -0.013 |
Cyrus | -0.012 |
Elias | -0.012 |
Jace | -0.011 |
rabbi | -0.011 |
Token | Ax |
Token ID | 31554 |
Feature activation | +1.94e-03 |
Pos logit contributions | |
Lum | +0.021 |
Beam | +0.020 |
Lau | +0.019 |
Layer | +0.019 |
Meng | +0.018 |
Neg logit contributions | |
Augustus | -0.021 |
Cyrus | -0.019 |
Elias | -0.019 |
Jace | -0.018 |
rabbi | -0.018 |
Token | In |
Token ID | 554 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.006 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.007 |
Cyrus | -0.006 |
Elias | -0.006 |
Jace | -0.006 |
rabbi | -0.006 |
Token | January |
Token ID | 3269 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
Lum | +0.027 |
Beam | +0.026 |
Lau | +0.025 |
Layer | +0.024 |
Meng | +0.023 |
Neg logit contributions | |
Augustus | -0.029 |
Cyrus | -0.026 |
Elias | -0.026 |
Jace | -0.025 |
rabbi | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
Lum | +0.040 |
Beam | +0.038 |
Lau | +0.036 |
Layer | +0.035 |
Meng | +0.034 |
Neg logit contributions | |
Augustus | -0.044 |
Cyrus | -0.040 |
Elias | -0.039 |
Jace | -0.039 |
rabbi | -0.038 |
Token | North |
Token ID | 2258 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | umb |
Token ID | 2178 |
Feature activation | -6.08e-02 |
Pos logit contributions | |
Lum | +0.020 |
Beam | +0.019 |
Lau | +0.016 |
Layer | +0.016 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.066 |
Cyrus | -0.062 |
Elias | -0.062 |
Jace | -0.061 |
rabbi | -0.060 |
Token | rian |
Token ID | 4484 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
Lum | +0.082 |
Beam | +0.079 |
Lau | +0.077 |
Layer | +0.075 |
Meng | +0.074 |
Neg logit contributions | |
Augustus | -0.032 |
Cyrus | -0.027 |
Elias | -0.027 |
Jace | -0.026 |
rabbi | -0.025 |
Token | rebels |
Token ID | 12669 |
Feature activation | -9.83e-03 |
Pos logit contributions | |
Augustus | +0.253 |
Cyrus | +0.227 |
Elias | +0.223 |
Jace | +0.218 |
rabbi | +0.216 |
Neg logit contributions | |
Lum | -0.285 |
Beam | -0.273 |
Lau | -0.260 |
Layer | -0.254 |
Meng | -0.247 |
Token | conver |
Token ID | 6718 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
Lum | +0.009 |
Beam | +0.009 |
Lau | +0.008 |
Layer | +0.008 |
Meng | +0.008 |
Neg logit contributions | |
Augustus | -0.010 |
Cyrus | -0.009 |
Elias | -0.009 |
Jace | -0.008 |
rabbi | -0.008 |
Token | ged |
Token ID | 2004 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
Lum | +0.041 |
Beam | +0.038 |
Lau | +0.036 |
Layer | +0.035 |
Meng | +0.034 |
Neg logit contributions | |
Augustus | -0.067 |
Cyrus | -0.062 |
Elias | -0.061 |
rabbi | -0.060 |
Jace | -0.060 |
Token | on |
Token ID | 319 |
Feature activation | +9.21e-04 |
Pos logit contributions | |
Lum | +0.040 |
Beam | +0.039 |
Lau | +0.037 |
Layer | +0.036 |
Meng | +0.035 |
Neg logit contributions | |
Augustus | -0.041 |
Cyrus | -0.037 |
Elias | -0.036 |
Jace | -0.035 |
rabbi | -0.035 |
Token | Durham |
Token ID | 30928 |
Feature activation | +1.32e-03 |
Pos logit contributions | |
Augustus | +0.001 |
Cyrus | +0.001 |
Elias | +0.001 |
Jace | +0.001 |
rabbi | +0.001 |
Neg logit contributions | |
Lum | -0.001 |
Beam | -0.001 |
Lau | -0.001 |
Layer | -0.001 |
Meng | -0.001 |
Token | fin |
Token ID | 957 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
Augustus | +0.046 |
Cyrus | +0.043 |
Elias | +0.042 |
Jace | +0.042 |
rabbi | +0.041 |
Neg logit contributions | |
Lum | -0.025 |
Beam | -0.024 |
Lau | -0.022 |
Layer | -0.021 |
Meng | -0.020 |
Token | isher |
Token ID | 4828 |
Feature activation | +6.23e-02 |
Pos logit contributions | |
Lum | +0.032 |
Beam | +0.031 |
Lau | +0.029 |
Layer | +0.029 |
Meng | +0.028 |
Neg logit contributions | |
Augustus | -0.030 |
Cyrus | -0.027 |
Elias | -0.026 |
Jace | -0.026 |
rabbi | -0.026 |
Token | " |
Token ID | 1 |
Feature activation | +3.69e-03 |
Pos logit contributions | |
Augustus | +0.061 |
Cyrus | +0.056 |
Elias | +0.055 |
Jace | +0.054 |
rabbi | +0.053 |
Neg logit contributions | |
Lum | -0.057 |
Beam | -0.055 |
Lau | -0.051 |
Layer | -0.050 |
Meng | -0.049 |
Token | pic |
Token ID | 8301 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
Augustus | +0.004 |
Cyrus | +0.004 |
Elias | +0.004 |
Jace | +0.004 |
rabbi | +0.003 |
Neg logit contributions | |
Lum | -0.003 |
Beam | -0.003 |
Lau | -0.003 |
Layer | -0.003 |
Meng | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e-03 |
Pos logit contributions | |
Lum | +0.038 |
Beam | +0.037 |
Lau | +0.035 |
Layer | +0.034 |
Meng | +0.033 |
Neg logit contributions | |
Augustus | -0.041 |
Cyrus | -0.037 |
Elias | -0.036 |
rabbi | -0.036 |
Jace | -0.036 |
Token | twitter |
Token ID | 6956 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
rabbi | -0.000 |
Jace | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -1.89e-03 |
Pos logit contributions | |
Augustus | +0.269 |
Cyrus | +0.244 |
Elias | +0.240 |
Jace | +0.234 |
rabbi | +0.233 |
Neg logit contributions | |
Lum | -0.266 |
Beam | -0.255 |
Lau | -0.242 |
Layer | -0.237 |
Meng | -0.229 |
Token | com |
Token ID | 785 |
Feature activation | -6.32e-02 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | / |
Token ID | 14 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
Lum | +0.061 |
Beam | +0.058 |
Lau | +0.055 |
Layer | +0.054 |
Meng | +0.053 |
Neg logit contributions | |
Augustus | -0.059 |
Cyrus | -0.053 |
Elias | -0.053 |
Jace | -0.051 |
rabbi | -0.051 |
Token | Eh |
Token ID | 43894 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
Lum | +0.021 |
Beam | +0.020 |
Lau | +0.019 |
Layer | +0.018 |
Meng | +0.018 |
Neg logit contributions | |
Augustus | -0.020 |
Cyrus | -0.018 |
Elias | -0.018 |
Jace | -0.017 |
rabbi | -0.017 |
Token | U |
Token ID | 52 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
Augustus | +0.027 |
Cyrus | +0.025 |
Elias | +0.024 |
Jace | +0.024 |
rabbi | +0.024 |
Neg logit contributions | |
Lum | -0.028 |
Beam | -0.027 |
Lau | -0.026 |
Layer | -0.025 |
Meng | -0.024 |
Token | media |
Token ID | 2056 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
Augustus | +0.031 |
Cyrus | +0.028 |
Elias | +0.027 |
Jace | +0.027 |
rabbi | +0.027 |
Neg logit contributions | |
Lum | -0.026 |
Beam | -0.025 |
Lau | -0.024 |
Layer | -0.023 |
Meng | -0.022 |
Token | bus |
Token ID | 1323 |
Feature activation | +6.24e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.017 |
Jace | +0.017 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.016 |
Beam | -0.015 |
Lau | -0.014 |
Layer | -0.014 |
Meng | -0.013 |
Token | ) |
Token ID | 8 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
Augustus | +0.063 |
Cyrus | +0.057 |
Elias | +0.056 |
Jace | +0.055 |
rabbi | +0.054 |
Neg logit contributions | |
Lum | -0.056 |
Beam | -0.054 |
Lau | -0.050 |
Layer | -0.049 |
Meng | -0.048 |
Token | pic |
Token ID | 8301 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
Augustus | +0.017 |
Cyrus | +0.016 |
Elias | +0.016 |
Jace | +0.015 |
rabbi | +0.015 |
Neg logit contributions | |
Lum | -0.012 |
Beam | -0.011 |
Lau | -0.011 |
Layer | -0.010 |
Meng | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
Lum | +0.040 |
Beam | +0.038 |
Lau | +0.036 |
Layer | +0.035 |
Meng | +0.034 |
Neg logit contributions | |
Augustus | -0.042 |
Cyrus | -0.038 |
Elias | -0.038 |
rabbi | -0.037 |
Jace | -0.037 |
Token | twitter |
Token ID | 6956 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
Lum | +0.042 |
Beam | +0.041 |
Lau | +0.039 |
Layer | +0.039 |
Meng | +0.038 |
Neg logit contributions | |
Augustus | -0.020 |
Cyrus | -0.017 |
Elias | -0.017 |
rabbi | -0.016 |
Jace | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
Augustus | +0.269 |
Cyrus | +0.244 |
Elias | +0.240 |
Jace | +0.234 |
rabbi | +0.233 |
Neg logit contributions | |
Lum | -0.265 |
Beam | -0.255 |
Lau | -0.241 |
Layer | -0.236 |
Meng | -0.228 |
Token | com |
Token ID | 785 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
Lum | +0.026 |
Beam | +0.025 |
Lau | +0.024 |
Layer | +0.023 |
Meng | +0.023 |
Neg logit contributions | |
Augustus | -0.023 |
Cyrus | -0.021 |
Elias | -0.020 |
Jace | -0.020 |
rabbi | -0.020 |
Token | / |
Token ID | 14 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
Lum | +0.044 |
Beam | +0.042 |
Lau | +0.040 |
Layer | +0.039 |
Meng | +0.038 |
Neg logit contributions | |
Augustus | -0.043 |
Cyrus | -0.039 |
Elias | -0.038 |
Jace | -0.037 |
rabbi | -0.037 |
Token | z |
Token ID | 89 |
Feature activation | -2.02e-03 |
Pos logit contributions | |
Lum | +0.017 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.015 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.016 |
Cyrus | -0.015 |
Elias | -0.014 |
Jace | -0.014 |
rabbi | -0.014 |
Token | J |
Token ID | 41 |
Feature activation | -9.43e-03 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +1.32e-03 |
Pos logit contributions | |
Augustus | +0.017 |
Cyrus | +0.015 |
Elias | +0.015 |
Jace | +0.015 |
rabbi | +0.015 |
Neg logit contributions | |
Lum | -0.017 |
Beam | -0.016 |
Lau | -0.015 |
Layer | -0.015 |
Meng | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -7.26e-04 |
Pos logit contributions | |
Augustus | +0.001 |
Cyrus | +0.001 |
Elias | +0.001 |
Jace | +0.001 |
rabbi | +0.001 |
Neg logit contributions | |
Lum | -0.001 |
Beam | -0.001 |
Lau | -0.001 |
Layer | -0.001 |
Meng | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | The |
Token ID | 464 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
Augustus | +0.010 |
Cyrus | +0.009 |
Elias | +0.009 |
Jace | +0.008 |
rabbi | +0.008 |
Neg logit contributions | |
Lum | -0.010 |
Beam | -0.010 |
Lau | -0.009 |
Layer | -0.009 |
Meng | -0.009 |
Token | Dram |
Token ID | 36736 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
Augustus | +0.012 |
Cyrus | +0.011 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.010 |
Beam | -0.009 |
Lau | -0.009 |
Layer | -0.008 |
Meng | -0.008 |
Token | atically |
Token ID | 4142 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.009 |
Elias | +0.009 |
Jace | +0.009 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.017 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.016 |
Meng | -0.015 |
Token | Different |
Token ID | 20615 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
Augustus | +0.313 |
Cyrus | +0.288 |
Elias | +0.284 |
Jace | +0.279 |
rabbi | +0.275 |
Neg logit contributions | |
Lum | -0.220 |
Beam | -0.209 |
Lau | -0.195 |
Layer | -0.190 |
Meng | -0.183 |
Token | Attention |
Token ID | 47406 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
Lum | +0.016 |
Beam | +0.015 |
Lau | +0.014 |
Layer | +0.014 |
Meng | +0.013 |
Neg logit contributions | |
Augustus | -0.020 |
Cyrus | -0.018 |
Elias | -0.018 |
Jace | -0.018 |
rabbi | -0.017 |
Token | Economy |
Token ID | 18493 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
Lum | +0.029 |
Beam | +0.027 |
Lau | +0.026 |
Layer | +0.025 |
Meng | +0.024 |
Neg logit contributions | |
Augustus | -0.035 |
Cyrus | -0.032 |
Elias | -0.032 |
Jace | -0.031 |
rabbi | -0.031 |
Token | Of |
Token ID | 3226 |
Feature activation | +6.78e-03 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.009 |
Lau | +0.009 |
Layer | +0.008 |
Meng | +0.008 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | The |
Token ID | 383 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
Augustus | +0.007 |
Cyrus | +0.006 |
Elias | +0.006 |
Jace | +0.006 |
rabbi | +0.006 |
Neg logit contributions | |
Lum | -0.006 |
Beam | -0.006 |
Lau | -0.005 |
Layer | -0.005 |
Meng | -0.005 |
Token | c |
Token ID | 66 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
Augustus | +0.083 |
Cyrus | +0.075 |
Elias | +0.074 |
Jace | +0.072 |
rabbi | +0.072 |
Neg logit contributions | |
Lum | -0.084 |
Beam | -0.080 |
Lau | -0.076 |
Layer | -0.074 |
Meng | -0.072 |
Token | rawler |
Token ID | 39464 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
Augustus | +0.029 |
Cyrus | +0.026 |
Elias | +0.026 |
Jace | +0.025 |
rabbi | +0.025 |
Neg logit contributions | |
Lum | -0.029 |
Beam | -0.028 |
Lau | -0.027 |
Layer | -0.026 |
Meng | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +6.41e-03 |
Pos logit contributions | |
Augustus | +0.012 |
Cyrus | +0.011 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.012 |
Beam | -0.011 |
Lau | -0.011 |
Layer | -0.011 |
Meng | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
Augustus | +0.006 |
Cyrus | +0.006 |
Elias | +0.006 |
Jace | +0.006 |
rabbi | +0.006 |
Neg logit contributions | |
Lum | -0.006 |
Beam | -0.005 |
Lau | -0.005 |
Layer | -0.005 |
Meng | -0.005 |
Token | Jub |
Token ID | 38075 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.011 |
Jace | -0.011 |
rabbi | -0.010 |
Token | ilee |
Token ID | 40626 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
Lum | +0.019 |
Beam | +0.018 |
Lau | +0.017 |
Layer | +0.017 |
Meng | +0.017 |
Neg logit contributions | |
Augustus | -0.008 |
Cyrus | -0.007 |
Elias | -0.006 |
Jace | -0.006 |
rabbi | -0.006 |
Token | hanging |
Token ID | 10938 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
Augustus | +0.281 |
Cyrus | +0.256 |
Elias | +0.252 |
Jace | +0.246 |
rabbi | +0.243 |
Neg logit contributions | |
Lum | -0.251 |
Beam | -0.240 |
Lau | -0.226 |
Layer | -0.222 |
Meng | -0.214 |
Token | out |
Token ID | 503 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.009 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.009 |
Beam | -0.009 |
Lau | -0.008 |
Layer | -0.008 |
Meng | -0.008 |
Token | in |
Token ID | 287 |
Feature activation | -7.38e-03 |
Pos logit contributions | |
Augustus | +0.012 |
Cyrus | +0.011 |
Elias | +0.011 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.012 |
Beam | -0.011 |
Lau | -0.011 |
Layer | -0.010 |
Meng | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -6.19e-03 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.006 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.007 |
Cyrus | -0.007 |
Elias | -0.007 |
Jace | -0.007 |
rabbi | -0.006 |
Token | mall |
Token ID | 17374 |
Feature activation | +2.50e-03 |
Pos logit contributions | |
Lum | +0.006 |
Beam | +0.005 |
Lau | +0.005 |
Layer | +0.005 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.006 |
Cyrus | -0.006 |
Elias | -0.006 |
Jace | -0.005 |
rabbi | -0.005 |
Token | ed |
Token ID | 276 |
Feature activation | +5.42e-03 |
Pos logit contributions | |
Augustus | +0.004 |
Cyrus | +0.004 |
Elias | +0.004 |
Jace | +0.004 |
rabbi | +0.004 |
Neg logit contributions | |
Lum | -0.004 |
Beam | -0.004 |
Lau | -0.004 |
Layer | -0.004 |
Meng | -0.004 |
Token | Out |
Token ID | 7975 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
Augustus | +0.005 |
Cyrus | +0.005 |
Elias | +0.005 |
Jace | +0.005 |
rabbi | +0.005 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.005 |
Lau | -0.005 |
Layer | -0.004 |
Meng | -0.004 |
Token | Atlanta |
Token ID | 43482 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
Augustus | +0.016 |
Cyrus | +0.014 |
Elias | +0.014 |
Jace | +0.014 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.012 |
Beam | -0.012 |
Lau | -0.011 |
Layer | -0.011 |
Meng | -0.010 |
Token | pic |
Token ID | 8301 |
Feature activation | -6.39e-02 |
Pos logit contributions | |
Augustus | +0.017 |
Cyrus | +0.016 |
Elias | +0.015 |
Jace | +0.015 |
rabbi | +0.015 |
Neg logit contributions | |
Lum | -0.013 |
Beam | -0.012 |
Lau | -0.011 |
Layer | -0.011 |
Meng | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
Lum | +0.060 |
Beam | +0.057 |
Lau | +0.054 |
Layer | +0.053 |
Meng | +0.051 |
Neg logit contributions | |
Augustus | -0.062 |
Cyrus | -0.056 |
Elias | -0.055 |
rabbi | -0.054 |
Jace | -0.054 |
Token | twitter |
Token ID | 6956 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
Lum | +0.044 |
Beam | +0.043 |
Lau | +0.041 |
Layer | +0.041 |
Meng | +0.040 |
Neg logit contributions | |
Augustus | -0.020 |
Cyrus | -0.017 |
Elias | -0.017 |
rabbi | -0.016 |
Jace | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
Augustus | +0.261 |
Cyrus | +0.236 |
Elias | +0.232 |
Jace | +0.226 |
rabbi | +0.224 |
Neg logit contributions | |
Lum | -0.266 |
Beam | -0.255 |
Lau | -0.241 |
Layer | -0.237 |
Meng | -0.229 |
Token | com |
Token ID | 785 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
Lum | +0.021 |
Beam | +0.021 |
Lau | +0.020 |
Layer | +0.019 |
Meng | +0.019 |
Neg logit contributions | |
Augustus | -0.018 |
Cyrus | -0.016 |
Elias | -0.016 |
Jace | -0.016 |
rabbi | -0.016 |
Token | / |
Token ID | 14 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
Lum | +0.061 |
Beam | +0.059 |
Lau | +0.056 |
Layer | +0.054 |
Meng | +0.053 |
Neg logit contributions | |
Augustus | -0.058 |
Cyrus | -0.052 |
Elias | -0.052 |
Jace | -0.050 |
rabbi | -0.050 |
Token | xx |
Token ID | 5324 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
Lum | +0.015 |
Beam | +0.014 |
Lau | +0.013 |
Layer | +0.013 |
Meng | +0.013 |
Neg logit contributions | |
Augustus | -0.014 |
Cyrus | -0.013 |
Elias | -0.013 |
Jace | -0.012 |
rabbi | -0.012 |
Token | A |
Token ID | 32 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.010 |
Cyrus | -0.009 |
Elias | -0.009 |
Jace | -0.009 |
rabbi | -0.008 |
Token | Deb |
Token ID | 16587 |
Feature activation | -8.12e-04 |
Pos logit contributions | |
Lum | +0.011 |
Beam | +0.011 |
Lau | +0.010 |
Layer | +0.010 |
Meng | +0.010 |
Neg logit contributions | |
Augustus | -0.010 |
Cyrus | -0.009 |
Elias | -0.009 |
Jace | -0.009 |
rabbi | -0.009 |
Token | ate |
Token ID | 378 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | 2016 |
Token ID | 5304 |
Feature activation | +9.54e-03 |
Pos logit contributions | |
Augustus | +0.018 |
Cyrus | +0.016 |
Elias | +0.016 |
Jace | +0.016 |
rabbi | +0.016 |
Neg logit contributions | |
Lum | -0.022 |
Beam | -0.021 |
Lau | -0.020 |
Layer | -0.020 |
Meng | -0.019 |
Token | pic |
Token ID | 8301 |
Feature activation | -6.20e-02 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.007 |
Beam | -0.007 |
Lau | -0.007 |
Layer | -0.006 |
Meng | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
Lum | +0.057 |
Beam | +0.055 |
Lau | +0.052 |
Layer | +0.051 |
Meng | +0.049 |
Neg logit contributions | |
Augustus | -0.060 |
Cyrus | -0.055 |
Elias | -0.054 |
rabbi | -0.053 |
Jace | -0.053 |
Token | twitter |
Token ID | 6956 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
Lum | +0.065 |
Beam | +0.064 |
Lau | +0.061 |
Layer | +0.060 |
Meng | +0.059 |
Neg logit contributions | |
Augustus | -0.031 |
Cyrus | -0.026 |
Elias | -0.026 |
rabbi | -0.024 |
Jace | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
Augustus | +0.259 |
Cyrus | +0.235 |
Elias | +0.231 |
Jace | +0.225 |
rabbi | +0.225 |
Neg logit contributions | |
Lum | -0.255 |
Beam | -0.244 |
Lau | -0.231 |
Layer | -0.226 |
Meng | -0.219 |
Token | com |
Token ID | 785 |
Feature activation | -6.59e-02 |
Pos logit contributions | |
Lum | +0.036 |
Beam | +0.035 |
Lau | +0.033 |
Layer | +0.033 |
Meng | +0.032 |
Neg logit contributions | |
Augustus | -0.032 |
Cyrus | -0.029 |
Elias | -0.028 |
Jace | -0.027 |
rabbi | -0.027 |
Token | / |
Token ID | 14 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
Lum | +0.063 |
Beam | +0.061 |
Lau | +0.058 |
Layer | +0.056 |
Meng | +0.055 |
Neg logit contributions | |
Augustus | -0.062 |
Cyrus | -0.056 |
Elias | -0.055 |
Jace | -0.053 |
rabbi | -0.053 |
Token | F |
Token ID | 37 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
Lum | +0.033 |
Beam | +0.032 |
Lau | +0.030 |
Layer | +0.030 |
Meng | +0.029 |
Neg logit contributions | |
Augustus | -0.032 |
Cyrus | -0.029 |
Elias | -0.029 |
Jace | -0.028 |
rabbi | -0.028 |
Token | c |
Token ID | 66 |
Feature activation | -1.28e-03 |
Pos logit contributions | |
Lum | +0.039 |
Beam | +0.037 |
Lau | +0.035 |
Layer | +0.035 |
Meng | +0.033 |
Neg logit contributions | |
Augustus | -0.037 |
Cyrus | -0.034 |
Elias | -0.033 |
Jace | -0.032 |
rabbi | -0.032 |
Token | perks |
Token ID | 28429 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
Augustus | +0.030 |
Cyrus | +0.027 |
Elias | +0.027 |
Jace | +0.026 |
rabbi | +0.026 |
Neg logit contributions | |
Lum | -0.029 |
Beam | -0.028 |
Lau | -0.026 |
Layer | -0.025 |
Meng | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
Lum | +0.024 |
Beam | +0.023 |
Lau | +0.021 |
Layer | +0.021 |
Meng | +0.020 |
Neg logit contributions | |
Augustus | -0.025 |
Cyrus | -0.023 |
Elias | -0.022 |
Jace | -0.022 |
rabbi | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
Augustus | +0.022 |
Cyrus | +0.020 |
Elias | +0.020 |
Jace | +0.019 |
rabbi | +0.019 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.016 |
Meng | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.017 |
Jace | +0.017 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.014 |
Beam | -0.013 |
Lau | -0.012 |
Layer | -0.012 |
Meng | -0.012 |
Token | Integ |
Token ID | 34500 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
Augustus | +0.017 |
Cyrus | +0.015 |
Elias | +0.015 |
Jace | +0.014 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.017 |
Beam | -0.016 |
Lau | -0.015 |
Layer | -0.015 |
Meng | -0.014 |
Token | rity |
Token ID | 10138 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
Augustus | +0.023 |
Cyrus | +0.021 |
Elias | +0.020 |
Jace | +0.019 |
rabbi | +0.019 |
Neg logit contributions | |
Lum | -0.035 |
Beam | -0.034 |
Lau | -0.032 |
Layer | -0.032 |
Meng | -0.031 |
Token | of |
Token ID | 286 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
Augustus | +0.266 |
Cyrus | +0.242 |
Elias | +0.238 |
Jace | +0.232 |
rabbi | +0.230 |
Neg logit contributions | |
Lum | -0.248 |
Beam | -0.237 |
Lau | -0.224 |
Layer | -0.219 |
Meng | -0.212 |
Token | a |
Token ID | 257 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
Augustus | +0.041 |
Cyrus | +0.037 |
Elias | +0.037 |
Jace | +0.036 |
rabbi | +0.035 |
Neg logit contributions | |
Lum | -0.039 |
Beam | -0.037 |
Lau | -0.035 |
Layer | -0.035 |
Meng | -0.033 |
Token | tournament |
Token ID | 7756 |
Feature activation | -1.56e-03 |
Pos logit contributions | |
Augustus | +0.031 |
Cyrus | +0.028 |
Elias | +0.028 |
Jace | +0.027 |
rabbi | +0.027 |
Neg logit contributions | |
Lum | -0.029 |
Beam | -0.027 |
Lau | -0.026 |
Layer | -0.025 |
Meng | -0.024 |
Token | should |
Token ID | 815 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | always |
Token ID | 1464 |
Feature activation | -1.99e-03 |
Pos logit contributions | |
Augustus | +0.024 |
Cyrus | +0.022 |
Elias | +0.021 |
Jace | +0.021 |
rabbi | +0.021 |
Neg logit contributions | |
Lum | -0.021 |
Beam | -0.020 |
Lau | -0.019 |
Layer | -0.019 |
Meng | -0.018 |
Token | The |
Token ID | 464 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
Lum | +0.068 |
Beam | +0.065 |
Lau | +0.062 |
Layer | +0.060 |
Meng | +0.058 |
Neg logit contributions | |
Augustus | -0.063 |
Cyrus | -0.057 |
Elias | -0.056 |
Jace | -0.054 |
rabbi | -0.054 |
Token | Deb |
Token ID | 16587 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.017 |
Jace | +0.017 |
rabbi | +0.016 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.017 |
Lau | -0.017 |
Layer | -0.016 |
Meng | -0.016 |
Token | ates |
Token ID | 689 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.017 |
Jace | +0.017 |
rabbi | +0.016 |
Neg logit contributions | |
Lum | -0.023 |
Beam | -0.022 |
Lau | -0.021 |
Layer | -0.020 |
Meng | -0.020 |
Token | pic |
Token ID | 8301 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
Augustus | +0.053 |
Cyrus | +0.049 |
Elias | +0.048 |
Jace | +0.047 |
rabbi | +0.047 |
Neg logit contributions | |
Lum | -0.032 |
Beam | -0.030 |
Lau | -0.028 |
Layer | -0.027 |
Meng | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
Lum | +0.059 |
Beam | +0.057 |
Lau | +0.054 |
Layer | +0.052 |
Meng | +0.051 |
Neg logit contributions | |
Augustus | -0.064 |
Cyrus | -0.058 |
Elias | -0.057 |
rabbi | -0.056 |
Jace | -0.055 |
Token | twitter |
Token ID | 6956 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
Lum | +0.063 |
Beam | +0.062 |
Lau | +0.059 |
Layer | +0.058 |
Meng | +0.057 |
Neg logit contributions | |
Augustus | -0.030 |
Cyrus | -0.025 |
Elias | -0.025 |
rabbi | -0.024 |
Jace | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
Augustus | +0.258 |
Cyrus | +0.234 |
Elias | +0.231 |
Jace | +0.225 |
rabbi | +0.224 |
Neg logit contributions | |
Lum | -0.254 |
Beam | -0.244 |
Lau | -0.231 |
Layer | -0.226 |
Meng | -0.219 |
Token | com |
Token ID | 785 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
Lum | +0.043 |
Beam | +0.041 |
Lau | +0.039 |
Layer | +0.038 |
Meng | +0.037 |
Neg logit contributions | |
Augustus | -0.038 |
Cyrus | -0.034 |
Elias | -0.033 |
rabbi | -0.032 |
Jace | -0.032 |
Token | / |
Token ID | 14 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
Lum | +0.054 |
Beam | +0.051 |
Lau | +0.049 |
Layer | +0.048 |
Meng | +0.046 |
Neg logit contributions | |
Augustus | -0.053 |
Cyrus | -0.048 |
Elias | -0.047 |
Jace | -0.046 |
rabbi | -0.046 |
Token | O |
Token ID | 46 |
Feature activation | +9.58e-03 |
Pos logit contributions | |
Lum | +0.023 |
Beam | +0.022 |
Lau | +0.021 |
Layer | +0.021 |
Meng | +0.020 |
Neg logit contributions | |
Augustus | -0.023 |
Cyrus | -0.020 |
Elias | -0.020 |
Jace | -0.020 |
rabbi | -0.019 |
Token | JV |
Token ID | 41697 |
Feature activation | +9.69e-02 |
Pos logit contributions | |
Augustus | +0.009 |
Cyrus | +0.008 |
Elias | +0.008 |
Jace | +0.008 |
rabbi | +0.007 |
Neg logit contributions | |
Lum | -0.009 |
Beam | -0.009 |
Lau | -0.009 |
Layer | -0.008 |
Meng | -0.008 |
Token | future |
Token ID | 2003 |
Feature activation | -2.44e-03 |
Pos logit contributions | |
Lum | +0.004 |
Beam | +0.004 |
Lau | +0.004 |
Layer | +0.004 |
Meng | +0.004 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.003 |
rabbi | -0.003 |
Token | ? |
Token ID | 30 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | � |
Token ID | 447 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
Augustus | +0.035 |
Cyrus | +0.032 |
Elias | +0.032 |
Jace | +0.031 |
rabbi | +0.031 |
Neg logit contributions | |
Lum | -0.027 |
Beam | -0.026 |
Lau | -0.024 |
Layer | -0.024 |
Meng | -0.023 |
Token | � |
Token ID | 251 |
Feature activation | -8.43e-04 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.016 |
Jace | +0.016 |
rabbi | +0.016 |
Neg logit contributions | |
Lum | -0.025 |
Beam | -0.024 |
Lau | -0.023 |
Layer | -0.022 |
Meng | -0.022 |
Token | Techn |
Token ID | 5429 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | ically |
Token ID | 1146 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
Augustus | +0.009 |
Cyrus | +0.008 |
Elias | +0.008 |
Jace | +0.008 |
rabbi | +0.008 |
Neg logit contributions | |
Lum | -0.014 |
Beam | -0.013 |
Lau | -0.012 |
Layer | -0.012 |
Meng | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
Augustus | +0.255 |
Cyrus | +0.231 |
Elias | +0.227 |
Jace | +0.222 |
rabbi | +0.220 |
Neg logit contributions | |
Lum | -0.253 |
Beam | -0.243 |
Lau | -0.229 |
Layer | -0.225 |
Meng | -0.218 |
Token | you |
Token ID | 345 |
Feature activation | -1.50e-03 |
Pos logit contributions | |
Augustus | +0.016 |
Cyrus | +0.015 |
Elias | +0.015 |
Jace | +0.014 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.017 |
Beam | -0.016 |
Lau | -0.015 |
Layer | -0.015 |
Meng | -0.014 |
Token | can |
Token ID | 460 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | � |
Token ID | 447 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
Augustus | +0.026 |
Cyrus | +0.023 |
Elias | +0.023 |
Jace | +0.022 |
rabbi | +0.022 |
Neg logit contributions | |
Lum | -0.025 |
Beam | -0.024 |
Lau | -0.022 |
Layer | -0.022 |
Meng | -0.021 |
Token | � |
Token ID | 247 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
Augustus | +0.036 |
Cyrus | +0.033 |
Elias | +0.032 |
Jace | +0.031 |
rabbi | +0.031 |
Neg logit contributions | |
Lum | -0.036 |
Beam | -0.035 |
Lau | -0.033 |
Layer | -0.032 |
Meng | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
Augustus | +0.005 |
Cyrus | +0.005 |
Elias | +0.005 |
Jace | +0.005 |
rabbi | +0.004 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.005 |
Lau | -0.005 |
Layer | -0.005 |
Meng | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -5.78e-03 |
Pos logit contributions | |
Augustus | +0.022 |
Cyrus | +0.020 |
Elias | +0.020 |
Jace | +0.020 |
rabbi | +0.019 |
Neg logit contributions | |
Lum | -0.022 |
Beam | -0.021 |
Lau | -0.020 |
Layer | -0.019 |
Meng | -0.019 |
Token | new |
Token ID | 649 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.005 |
Lau | +0.005 |
Layer | +0.005 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.006 |
Cyrus | -0.005 |
Elias | -0.005 |
Jace | -0.005 |
rabbi | -0.005 |
Token | North |
Token ID | 2258 |
Feature activation | -5.40e-02 |
Pos logit contributions | |
Lum | +0.016 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.014 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.018 |
Cyrus | -0.016 |
Elias | -0.016 |
Jace | -0.016 |
rabbi | -0.016 |
Token | umb |
Token ID | 2178 |
Feature activation | -7.52e-02 |
Pos logit contributions | |
Lum | +0.028 |
Beam | +0.026 |
Lau | +0.023 |
Layer | +0.022 |
Meng | +0.021 |
Neg logit contributions | |
Augustus | -0.075 |
Cyrus | -0.070 |
Elias | -0.069 |
Jace | -0.068 |
rabbi | -0.067 |
Token | rian |
Token ID | 4484 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
Lum | +0.094 |
Beam | +0.091 |
Lau | +0.088 |
Layer | +0.086 |
Meng | +0.084 |
Neg logit contributions | |
Augustus | -0.047 |
Cyrus | -0.041 |
Elias | -0.041 |
Jace | -0.039 |
Reyes | -0.038 |
Token | uprising |
Token ID | 26312 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
Augustus | +0.241 |
Cyrus | +0.217 |
Elias | +0.214 |
Jace | +0.208 |
rabbi | +0.206 |
Neg logit contributions | |
Lum | -0.265 |
Beam | -0.254 |
Lau | -0.241 |
Layer | -0.236 |
Meng | -0.229 |
Token | in |
Token ID | 287 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
Augustus | +0.023 |
Cyrus | +0.021 |
Elias | +0.021 |
Jace | +0.020 |
rabbi | +0.020 |
Neg logit contributions | |
Lum | -0.023 |
Beam | -0.022 |
Lau | -0.021 |
Layer | -0.020 |
Meng | -0.020 |
Token | early |
Token ID | 1903 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
Lum | +0.023 |
Beam | +0.022 |
Lau | +0.021 |
Layer | +0.020 |
Meng | +0.020 |
Neg logit contributions | |
Augustus | -0.023 |
Cyrus | -0.021 |
Elias | -0.021 |
Jace | -0.020 |
rabbi | -0.020 |
Token | 10 |
Token ID | 838 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
Lum | +0.015 |
Beam | +0.014 |
Lau | +0.013 |
Layer | +0.013 |
Meng | +0.013 |
Neg logit contributions | |
Augustus | -0.014 |
Cyrus | -0.013 |
Elias | -0.013 |
Jace | -0.012 |
rabbi | -0.012 |
Token | 69 |
Token ID | 3388 |
Feature activation | -8.84e-04 |
Pos logit contributions | |
Lum | +0.011 |
Beam | +0.010 |
Lau | +0.010 |
Layer | +0.010 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | spons |
Token ID | 20587 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | ors |
Token ID | 669 |
Feature activation | +2.33e-03 |
Pos logit contributions | |
Lum | +0.018 |
Beam | +0.018 |
Lau | +0.017 |
Layer | +0.017 |
Meng | +0.016 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.011 |
Jace | -0.010 |
rabbi | -0.010 |
Token | ", |
Token ID | 1600 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
Augustus | +0.002 |
Cyrus | +0.002 |
Elias | +0.002 |
Jace | +0.002 |
rabbi | +0.002 |
Neg logit contributions | |
Lum | -0.002 |
Beam | -0.002 |
Lau | -0.002 |
Layer | -0.002 |
Meng | -0.002 |
Token | " |
Token ID | 366 |
Feature activation | +3.07e-03 |
Pos logit contributions | |
Augustus | +0.065 |
Cyrus | +0.059 |
Elias | +0.058 |
Jace | +0.057 |
rabbi | +0.057 |
Neg logit contributions | |
Lum | -0.056 |
Beam | -0.054 |
Lau | -0.051 |
Layer | -0.050 |
Meng | -0.048 |
Token | sum |
Token ID | 16345 |
Feature activation | +3.08e-03 |
Pos logit contributions | |
Augustus | +0.003 |
Cyrus | +0.003 |
Elias | +0.003 |
Jace | +0.003 |
rabbi | +0.002 |
Neg logit contributions | |
Lum | -0.003 |
Beam | -0.003 |
Lau | -0.003 |
Layer | -0.003 |
Meng | -0.003 |
Token | mer |
Token ID | 647 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
Augustus | +0.005 |
Cyrus | +0.004 |
Elias | +0.004 |
Jace | +0.004 |
rabbi | +0.004 |
Neg logit contributions | |
Lum | -0.001 |
Beam | -0.001 |
Lau | -0.001 |
Layer | -0.001 |
Meng | -0.001 |
Token | " |
Token ID | 1 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
Lum | +0.267 |
Beam | +0.255 |
Lau | +0.240 |
Layer | +0.235 |
Meng | +0.227 |
Neg logit contributions | |
Augustus | -0.299 |
Cyrus | -0.273 |
Elias | -0.269 |
Jace | -0.263 |
rabbi | -0.260 |
Token | and |
Token ID | 290 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
Augustus | +0.010 |
Cyrus | +0.009 |
Elias | +0.009 |
Jace | +0.009 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.009 |
Beam | -0.009 |
Lau | -0.008 |
Layer | -0.008 |
Meng | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
Augustus | +0.026 |
Cyrus | +0.024 |
Elias | +0.023 |
Jace | +0.023 |
rabbi | +0.023 |
Neg logit contributions | |
Lum | -0.022 |
Beam | -0.021 |
Lau | -0.020 |
Layer | -0.019 |
Meng | -0.019 |
Token | London |
Token ID | 23421 |
Feature activation | -1.30e-03 |
Pos logit contributions | |
Augustus | +0.039 |
Cyrus | +0.035 |
Elias | +0.034 |
Jace | +0.034 |
rabbi | +0.033 |
Neg logit contributions | |
Lum | -0.038 |
Beam | -0.037 |
Lau | -0.035 |
Layer | -0.034 |
Meng | -0.033 |
Token | " |
Token ID | 1 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.005 |
Lau | +0.005 |
Layer | +0.005 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.005 |
Cyrus | -0.005 |
Elias | -0.005 |
Jace | -0.005 |
rabbi | -0.005 |
Token | Lonely |
Token ID | 46501 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
Augustus | +0.023 |
Cyrus | +0.021 |
Elias | +0.020 |
Jace | +0.020 |
rabbi | +0.020 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.016 |
Meng | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
Lum | +0.013 |
Beam | +0.013 |
Lau | +0.012 |
Layer | +0.012 |
Meng | +0.011 |
Neg logit contributions | |
Augustus | -0.014 |
Cyrus | -0.012 |
Elias | -0.012 |
Jace | -0.012 |
rabbi | -0.012 |
Token | Mis |
Token ID | 14136 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
Augustus | +0.030 |
Cyrus | +0.027 |
Elias | +0.027 |
Jace | +0.026 |
rabbi | +0.026 |
Neg logit contributions | |
Lum | -0.023 |
Beam | -0.022 |
Lau | -0.021 |
Layer | -0.020 |
Meng | -0.020 |
Token | under |
Token ID | 4625 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
Lum | +0.040 |
Beam | +0.038 |
Lau | +0.036 |
Layer | +0.035 |
Meng | +0.034 |
Neg logit contributions | |
Augustus | -0.048 |
Cyrus | -0.044 |
Elias | -0.044 |
Jace | -0.043 |
rabbi | -0.042 |
Token | stood |
Token ID | 6501 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
Lum | +0.019 |
Beam | +0.017 |
Lau | +0.015 |
Layer | +0.014 |
Meng | +0.012 |
Neg logit contributions | |
Augustus | -0.077 |
Cyrus | -0.072 |
Elias | -0.072 |
Jace | -0.071 |
rabbi | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | -8.41e-03 |
Pos logit contributions | |
Lum | +0.260 |
Beam | +0.247 |
Lau | +0.233 |
Layer | +0.228 |
Meng | +0.221 |
Neg logit contributions | |
Augustus | -0.291 |
Cyrus | -0.265 |
Elias | -0.261 |
Jace | -0.255 |
rabbi | -0.252 |
Token |
|
Token ID | 198 |
Feature activation | -4.47e-03 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.006 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.009 |
Cyrus | -0.008 |
Elias | -0.008 |
Jace | -0.008 |
rabbi | -0.008 |
Token | Why |
Token ID | 5195 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.004 |
Lau | +0.004 |
Layer | +0.004 |
Meng | +0.004 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.003 |
Elias | -0.003 |
Jace | -0.003 |
rabbi | -0.003 |
Token | 1954 |
Token ID | 24718 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
Augustus | +0.010 |
Cyrus | +0.009 |
Elias | +0.008 |
Jace | +0.008 |
rabbi | +0.008 |
Neg logit contributions | |
Lum | -0.012 |
Beam | -0.012 |
Lau | -0.011 |
Layer | -0.011 |
Meng | -0.011 |
Token | ? |
Token ID | 30 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
Lum | +0.016 |
Beam | +0.015 |
Lau | +0.014 |
Layer | +0.014 |
Meng | +0.013 |
Neg logit contributions | |
Augustus | -0.017 |
Cyrus | -0.015 |
Elias | -0.015 |
Jace | -0.015 |
rabbi | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.007 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.007 |
Cyrus | -0.006 |
Elias | -0.006 |
Jace | -0.006 |
rabbi | -0.006 |
Token | Christ |
Token ID | 1951 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
Lum | +0.039 |
Beam | +0.037 |
Lau | +0.035 |
Layer | +0.034 |
Meng | +0.033 |
Neg logit contributions | |
Augustus | -0.035 |
Cyrus | -0.032 |
Elias | -0.031 |
Jace | -0.031 |
rabbi | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
Lum | +0.033 |
Beam | +0.031 |
Lau | +0.030 |
Layer | +0.029 |
Meng | +0.028 |
Neg logit contributions | |
Augustus | -0.032 |
Cyrus | -0.029 |
Elias | -0.028 |
Jace | -0.028 |
rabbi | -0.027 |
Token | Sac |
Token ID | 9295 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
Lum | +0.032 |
Beam | +0.030 |
Lau | +0.029 |
Layer | +0.028 |
Meng | +0.027 |
Neg logit contributions | |
Augustus | -0.035 |
Cyrus | -0.031 |
Elias | -0.031 |
Jace | -0.030 |
rabbi | -0.030 |
Token | ram |
Token ID | 859 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
Lum | +0.025 |
Beam | +0.024 |
Lau | +0.023 |
Layer | +0.023 |
Meng | +0.022 |
Neg logit contributions | |
Augustus | -0.014 |
Cyrus | -0.012 |
Elias | -0.012 |
Jace | -0.012 |
rabbi | -0.012 |
Token | ents |
Token ID | 658 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.010 |
Cyrus | -0.009 |
Elias | -0.009 |
Jace | -0.009 |
rabbi | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
Lum | +0.269 |
Beam | +0.258 |
Lau | +0.243 |
Layer | +0.239 |
Meng | +0.231 |
Neg logit contributions | |
Augustus | -0.269 |
Cyrus | -0.244 |
Elias | -0.240 |
Jace | -0.234 |
rabbi | -0.232 |
Token | Bapt |
Token ID | 18226 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
Lum | +0.020 |
Beam | +0.019 |
Lau | +0.018 |
Layer | +0.018 |
Meng | +0.017 |
Neg logit contributions | |
Augustus | -0.022 |
Cyrus | -0.020 |
Elias | -0.020 |
Jace | -0.020 |
rabbi | -0.020 |
Token | ism |
Token ID | 1042 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
Lum | +0.032 |
Beam | +0.031 |
Lau | +0.029 |
Layer | +0.029 |
Meng | +0.028 |
Neg logit contributions | |
Augustus | -0.034 |
Cyrus | -0.031 |
Elias | -0.030 |
Jace | -0.030 |
rabbi | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
Lum | +0.017 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.015 |
Meng | +0.015 |
Neg logit contributions | |
Augustus | -0.017 |
Cyrus | -0.015 |
Elias | -0.015 |
Jace | -0.015 |
rabbi | -0.015 |
Token | Conf |
Token ID | 7326 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
Lum | +0.023 |
Beam | +0.022 |
Lau | +0.021 |
Layer | +0.020 |
Meng | +0.019 |
Neg logit contributions | |
Augustus | -0.026 |
Cyrus | -0.024 |
Elias | -0.023 |
Jace | -0.023 |
rabbi | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
Augustus | +0.014 |
Cyrus | +0.013 |
Elias | +0.013 |
Jace | +0.012 |
rabbi | +0.012 |
Neg logit contributions | |
Lum | -0.010 |
Beam | -0.010 |
Lau | -0.009 |
Layer | -0.009 |
Meng | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.018 |
Elias | +0.017 |
Jace | +0.017 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.014 |
Beam | -0.013 |
Lau | -0.013 |
Layer | -0.012 |
Meng | -0.012 |
Token | � |
Token ID | 447 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
Augustus | +0.020 |
Cyrus | +0.019 |
Elias | +0.019 |
Jace | +0.018 |
rabbi | +0.018 |
Neg logit contributions | |
Lum | -0.016 |
Beam | -0.015 |
Lau | -0.014 |
Layer | -0.014 |
Meng | -0.013 |
Token | � |
Token ID | 250 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
Augustus | +0.014 |
Cyrus | +0.013 |
Elias | +0.012 |
Jace | +0.012 |
rabbi | +0.012 |
Neg logit contributions | |
Lum | -0.017 |
Beam | -0.016 |
Lau | -0.016 |
Layer | -0.015 |
Meng | -0.015 |
Token | Ke |
Token ID | 8896 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
Augustus | +0.017 |
Cyrus | +0.015 |
Elias | +0.015 |
Jace | +0.014 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.016 |
Meng | -0.015 |
Token | pler |
Token ID | 20053 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
Augustus | +0.034 |
Cyrus | +0.032 |
Elias | +0.032 |
Jace | +0.031 |
rabbi | +0.031 |
Neg logit contributions | |
Lum | -0.009 |
Beam | -0.008 |
Lau | -0.007 |
Layer | -0.007 |
Meng | -0.006 |
Token | - |
Token ID | 12 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
Lum | +0.234 |
Beam | +0.223 |
Lau | +0.210 |
Layer | +0.205 |
Meng | +0.198 |
Neg logit contributions | |
Augustus | -0.286 |
Cyrus | -0.261 |
Elias | -0.257 |
Jace | -0.252 |
rabbi | -0.249 |
Token | 10 |
Token ID | 940 |
Feature activation | -9.21e-03 |
Pos logit contributions | |
Augustus | +0.012 |
Cyrus | +0.011 |
Elias | +0.011 |
Jace | +0.011 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.017 |
Beam | -0.016 |
Lau | -0.015 |
Layer | -0.015 |
Meng | -0.015 |
Token | c |
Token ID | 66 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
Lum | +0.009 |
Beam | +0.009 |
Lau | +0.008 |
Layer | +0.008 |
Meng | +0.008 |
Neg logit contributions | |
Augustus | -0.008 |
Cyrus | -0.008 |
Elias | -0.008 |
Jace | -0.007 |
rabbi | -0.007 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.010 |
Beam | -0.010 |
Lau | -0.009 |
Layer | -0.009 |
Meng | -0.009 |
Token | � |
Token ID | 447 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
Augustus | +0.016 |
Cyrus | +0.015 |
Elias | +0.015 |
Jace | +0.014 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.014 |
Beam | -0.014 |
Lau | -0.013 |
Layer | -0.013 |
Meng | -0.012 |
Token | at |
Token ID | 379 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
Augustus | +0.051 |
Cyrus | +0.046 |
Elias | +0.045 |
Jace | +0.044 |
rabbi | +0.044 |
Neg logit contributions | |
Lum | -0.047 |
Beam | -0.045 |
Lau | -0.043 |
Layer | -0.042 |
Meng | -0.041 |
Token | 19 |
Token ID | 678 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
Augustus | +0.059 |
Cyrus | +0.054 |
Elias | +0.053 |
Jace | +0.052 |
rabbi | +0.051 |
Neg logit contributions | |
Lum | -0.056 |
Beam | -0.053 |
Lau | -0.050 |
Layer | -0.049 |
Meng | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
Lum | +0.022 |
Beam | +0.021 |
Lau | +0.020 |
Layer | +0.019 |
Meng | +0.019 |
Neg logit contributions | |
Augustus | -0.024 |
Cyrus | -0.022 |
Elias | -0.021 |
Jace | -0.021 |
rabbi | -0.021 |
Token | 7 |
Token ID | 22 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
Augustus | +0.025 |
Cyrus | +0.023 |
Elias | +0.022 |
Jace | +0.022 |
rabbi | +0.022 |
Neg logit contributions | |
Lum | -0.025 |
Beam | -0.024 |
Lau | -0.022 |
Layer | -0.022 |
Meng | -0.021 |
Token | per |
Token ID | 583 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
Lum | +0.019 |
Beam | +0.018 |
Lau | +0.017 |
Layer | +0.017 |
Meng | +0.016 |
Neg logit contributions | |
Augustus | -0.025 |
Cyrus | -0.023 |
Elias | -0.023 |
Jace | -0.022 |
rabbi | -0.022 |
Token | cent |
Token ID | 1247 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
Lum | +0.006 |
Beam | +0.006 |
Lau | +0.005 |
Layer | +0.004 |
Meng | +0.004 |
Neg logit contributions | |
Augustus | -0.035 |
Cyrus | -0.033 |
Elias | -0.032 |
Jace | -0.032 |
rabbi | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
Lum | +0.252 |
Beam | +0.241 |
Lau | +0.228 |
Layer | +0.223 |
Meng | +0.216 |
Neg logit contributions | |
Augustus | -0.259 |
Cyrus | -0.235 |
Elias | -0.231 |
Jace | -0.225 |
rabbi | -0.224 |
Token |
|
Token ID | 198 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
Augustus | +0.027 |
Cyrus | +0.025 |
Elias | +0.025 |
Jace | +0.024 |
rabbi | +0.024 |
Neg logit contributions | |
Lum | -0.024 |
Beam | -0.023 |
Lau | -0.021 |
Layer | -0.021 |
Meng | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.017 |
Jace | +0.017 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.014 |
Beam | -0.013 |
Lau | -0.012 |
Layer | -0.012 |
Meng | -0.012 |
Token | Australian |
Token ID | 38036 |
Feature activation | -2.38e-03 |
Pos logit contributions | |
Augustus | +0.016 |
Cyrus | +0.015 |
Elias | +0.014 |
Jace | +0.014 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.016 |
Beam | -0.015 |
Lau | -0.014 |
Layer | -0.014 |
Meng | -0.014 |
Token | clinical |
Token ID | 8668 |
Feature activation | -8.33e-03 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | showed |
Token ID | 3751 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
Augustus | +0.040 |
Cyrus | +0.037 |
Elias | +0.036 |
Jace | +0.035 |
rabbi | +0.035 |
Neg logit contributions | |
Lum | -0.031 |
Beam | -0.029 |
Lau | -0.027 |
Layer | -0.027 |
Meng | -0.026 |
Token | that |
Token ID | 326 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
Lum | +0.020 |
Beam | +0.019 |
Lau | +0.018 |
Layer | +0.018 |
Meng | +0.017 |
Neg logit contributions | |
Augustus | -0.021 |
Cyrus | -0.019 |
Elias | -0.019 |
Jace | -0.018 |
rabbi | -0.018 |
Token | about |
Token ID | 546 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
Augustus | +0.018 |
Cyrus | +0.017 |
Elias | +0.016 |
Jace | +0.016 |
rabbi | +0.016 |
Neg logit contributions | |
Lum | -0.016 |
Beam | -0.016 |
Lau | -0.015 |
Layer | -0.015 |
Meng | -0.014 |
Token | 14 |
Token ID | 1478 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
Augustus | +0.017 |
Cyrus | +0.016 |
Elias | +0.016 |
Jace | +0.015 |
rabbi | +0.015 |
Neg logit contributions | |
Lum | -0.013 |
Beam | -0.012 |
Lau | -0.011 |
Layer | -0.011 |
Meng | -0.011 |
Token | per |
Token ID | 583 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
Lum | +0.024 |
Beam | +0.023 |
Lau | +0.021 |
Layer | +0.021 |
Meng | +0.020 |
Neg logit contributions | |
Augustus | -0.036 |
Cyrus | -0.033 |
Elias | -0.032 |
Jace | -0.032 |
rabbi | -0.032 |
Token | cent |
Token ID | 1247 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
Lum | +0.009 |
Beam | +0.008 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.050 |
Cyrus | -0.048 |
Elias | -0.047 |
Jace | -0.046 |
rabbi | -0.046 |
Token | of |
Token ID | 286 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
Lum | +0.270 |
Beam | +0.261 |
Lau | +0.247 |
Layer | +0.244 |
Meng | +0.236 |
Neg logit contributions | |
Augustus | -0.231 |
Cyrus | -0.207 |
Elias | -0.204 |
Jace | -0.198 |
rabbi | -0.196 |
Token | families |
Token ID | 4172 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
Augustus | +0.040 |
Cyrus | +0.036 |
Elias | +0.036 |
Jace | +0.035 |
rabbi | +0.035 |
Neg logit contributions | |
Lum | -0.038 |
Beam | -0.036 |
Lau | -0.034 |
Layer | -0.034 |
Meng | -0.032 |
Token | in |
Token ID | 287 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
Augustus | +0.017 |
Cyrus | +0.015 |
Elias | +0.015 |
Jace | +0.015 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.014 |
Beam | -0.014 |
Lau | -0.013 |
Layer | -0.013 |
Meng | -0.012 |
Token | traditionally |
Token ID | 16083 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
Augustus | +0.013 |
Cyrus | +0.012 |
Elias | +0.012 |
Jace | +0.012 |
rabbi | +0.011 |
Neg logit contributions | |
Lum | -0.011 |
Beam | -0.011 |
Lau | -0.010 |
Layer | -0.010 |
Meng | -0.010 |
Token | middle |
Token ID | 3504 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
Lum | +0.022 |
Beam | +0.021 |
Lau | +0.020 |
Layer | +0.019 |
Meng | +0.019 |
Neg logit contributions | |
Augustus | -0.027 |
Cyrus | -0.025 |
Elias | -0.024 |
Jace | -0.024 |
rabbi | -0.024 |
Token | of |
Token ID | 286 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
Lum | +0.066 |
Beam | +0.064 |
Lau | +0.060 |
Layer | +0.059 |
Meng | +0.057 |
Neg logit contributions | |
Augustus | -0.063 |
Cyrus | -0.056 |
Elias | -0.056 |
Jace | -0.054 |
rabbi | -0.054 |
Token | 8 |
Token ID | 807 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
Lum | +0.032 |
Beam | +0.030 |
Lau | +0.029 |
Layer | +0.028 |
Meng | +0.027 |
Neg logit contributions | |
Augustus | -0.038 |
Cyrus | -0.034 |
Elias | -0.034 |
Jace | -0.033 |
rabbi | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -2.35e-04 |
Pos logit contributions | |
Lum | +0.029 |
Beam | +0.028 |
Lau | +0.026 |
Layer | +0.026 |
Meng | +0.025 |
Neg logit contributions | |
Augustus | -0.029 |
Cyrus | -0.027 |
Elias | -0.026 |
Jace | -0.026 |
rabbi | -0.025 |
Token | 9 |
Token ID | 24 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
Lum | +0.000 |
Beam | +0.000 |
Lau | +0.000 |
Layer | +0.000 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.000 |
Cyrus | -0.000 |
Elias | -0.000 |
Jace | -0.000 |
rabbi | -0.000 |
Token | per |
Token ID | 583 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
Lum | +0.032 |
Beam | +0.030 |
Lau | +0.028 |
Layer | +0.028 |
Meng | +0.027 |
Neg logit contributions | |
Augustus | -0.044 |
Cyrus | -0.040 |
Elias | -0.040 |
Jace | -0.039 |
rabbi | -0.039 |
Token | cent |
Token ID | 1247 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.008 |
Lau | +0.007 |
Layer | +0.006 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.055 |
Cyrus | -0.052 |
Elias | -0.051 |
Jace | -0.051 |
rabbi | -0.050 |
Token | of |
Token ID | 286 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
Lum | +0.264 |
Beam | +0.253 |
Lau | +0.240 |
Layer | +0.235 |
Meng | +0.228 |
Neg logit contributions | |
Augustus | -0.239 |
Cyrus | -0.215 |
Elias | -0.211 |
Jace | -0.206 |
rabbi | -0.204 |
Token | respondent |
Token ID | 31282 |
Feature activation | -3.98e-04 |
Pos logit contributions | |
Augustus | +0.036 |
Cyrus | +0.033 |
Elias | +0.032 |
Jace | +0.032 |
rabbi | +0.031 |
Neg logit contributions | |
Lum | -0.034 |
Beam | -0.033 |
Lau | -0.031 |
Layer | -0.030 |
Meng | -0.029 |
Token | households |
Token ID | 12503 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
Lum | +0.000 |
Beam | +0.000 |
Lau | +0.000 |
Layer | +0.000 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.000 |
Cyrus | -0.000 |
Elias | -0.000 |
Jace | -0.000 |
rabbi | -0.000 |
Token | reported |
Token ID | 2098 |
Feature activation | -7.80e-03 |
Pos logit contributions | |
Lum | +0.011 |
Beam | +0.010 |
Lau | +0.010 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.011 |
Jace | -0.010 |
rabbi | -0.010 |
Token | that |
Token ID | 326 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.007 |
Lau | +0.007 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.007 |
Cyrus | -0.007 |
Elias | -0.007 |
Jace | -0.007 |
rabbi | -0.006 |
Token | various |
Token ID | 2972 |
Feature activation | -2.10e-03 |
Pos logit contributions | |
Lum | +0.015 |
Beam | +0.014 |
Lau | +0.013 |
Layer | +0.013 |
Meng | +0.013 |
Neg logit contributions | |
Augustus | -0.018 |
Cyrus | -0.017 |
Elias | -0.017 |
Jace | -0.016 |
rabbi | -0.016 |
Token | � |
Token ID | 564 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | � |
Token ID | 247 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
Lum | +0.022 |
Beam | +0.021 |
Lau | +0.020 |
Layer | +0.020 |
Meng | +0.019 |
Neg logit contributions | |
Augustus | -0.019 |
Cyrus | -0.017 |
Elias | -0.016 |
Jace | -0.016 |
rabbi | -0.016 |
Token | jack |
Token ID | 19650 |
Feature activation | +6.33e-02 |
Pos logit contributions | |
Lum | +0.016 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.014 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.017 |
Cyrus | -0.015 |
Elias | -0.015 |
Jace | -0.014 |
rabbi | -0.014 |
Token | marsh |
Token ID | 22397 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
Augustus | +0.069 |
Cyrus | +0.063 |
Elias | +0.063 |
Jace | +0.061 |
rabbi | +0.061 |
Neg logit contributions | |
Lum | -0.051 |
Beam | -0.048 |
Lau | -0.045 |
Layer | -0.044 |
Meng | -0.043 |
Token | als |
Token ID | 874 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.014 |
Cyrus | -0.013 |
Elias | -0.013 |
Jace | -0.013 |
rabbi | -0.013 |
Token | available |
Token ID | 1695 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
Lum | +0.188 |
Beam | +0.177 |
Lau | +0.165 |
Layer | +0.160 |
Meng | +0.153 |
Neg logit contributions | |
Augustus | -0.304 |
Cyrus | -0.280 |
Elias | -0.277 |
Jace | -0.272 |
rabbi | -0.269 |
Token | to |
Token ID | 284 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
Augustus | +0.015 |
Cyrus | +0.014 |
Elias | +0.014 |
Jace | +0.013 |
rabbi | +0.013 |
Neg logit contributions | |
Lum | -0.016 |
Beam | -0.015 |
Lau | -0.014 |
Layer | -0.014 |
Meng | -0.014 |
Token | Merc |
Token ID | 12185 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
Lum | +0.011 |
Beam | +0.010 |
Lau | +0.010 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.010 |
Cyrus | -0.009 |
Elias | -0.009 |
Jace | -0.009 |
rabbi | -0.009 |
Token | enaries |
Token ID | 30216 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
Lum | +0.025 |
Beam | +0.024 |
Lau | +0.023 |
Layer | +0.022 |
Meng | +0.022 |
Neg logit contributions | |
Augustus | -0.019 |
Cyrus | -0.017 |
Elias | -0.016 |
Jace | -0.016 |
rabbi | -0.016 |
Token | due |
Token ID | 2233 |
Feature activation | -7.24e-03 |
Pos logit contributions | |
Lum | +0.011 |
Beam | +0.010 |
Lau | +0.010 |
Layer | +0.010 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | pay |
Token ID | 1414 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
Lum | +0.009 |
Beam | +0.008 |
Lau | +0.008 |
Layer | +0.008 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.009 |
Cyrus | -0.008 |
Elias | -0.008 |
Jace | -0.008 |
rabbi | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -3.67e-03 |
Pos logit contributions | |
Lum | +0.017 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.015 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.019 |
Cyrus | -0.017 |
Elias | -0.017 |
Jace | -0.016 |
rabbi | -0.016 |
Token | additional |
Token ID | 3224 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.003 |
Elias | -0.003 |
Jace | -0.003 |
rabbi | -0.003 |
Token | 30 |
Token ID | 1542 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
Lum | +0.050 |
Beam | +0.048 |
Lau | +0.045 |
Layer | +0.044 |
Meng | +0.043 |
Neg logit contributions | |
Augustus | -0.049 |
Cyrus | -0.044 |
Elias | -0.043 |
Jace | -0.042 |
rabbi | -0.042 |
Token | per |
Token ID | 583 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
Lum | +0.021 |
Beam | +0.020 |
Lau | +0.019 |
Layer | +0.019 |
Meng | +0.018 |
Neg logit contributions | |
Augustus | -0.024 |
Cyrus | -0.022 |
Elias | -0.021 |
Jace | -0.021 |
rabbi | -0.021 |
Token | cent |
Token ID | 1247 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
Lum | +0.006 |
Beam | +0.005 |
Lau | +0.004 |
Layer | +0.004 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.032 |
Cyrus | -0.030 |
Elias | -0.030 |
Jace | -0.030 |
rabbi | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -1.57e-03 |
Pos logit contributions | |
Lum | +0.241 |
Beam | +0.231 |
Lau | +0.218 |
Layer | +0.215 |
Meng | +0.207 |
Neg logit contributions | |
Augustus | -0.244 |
Cyrus | -0.222 |
Elias | -0.217 |
Jace | -0.212 |
rabbi | -0.210 |
Token | roughly |
Token ID | 7323 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | one |
Token ID | 530 |
Feature activation | +2.82e-04 |
Pos logit contributions | |
Lum | +0.035 |
Beam | +0.033 |
Lau | +0.031 |
Layer | +0.031 |
Meng | +0.030 |
Neg logit contributions | |
Augustus | -0.039 |
Cyrus | -0.036 |
Elias | -0.035 |
Jace | -0.034 |
rabbi | -0.034 |
Token | - |
Token ID | 12 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
Augustus | +0.000 |
Cyrus | +0.000 |
Elias | +0.000 |
Jace | +0.000 |
rabbi | +0.000 |
Neg logit contributions | |
Lum | -0.000 |
Beam | -0.000 |
Lau | -0.000 |
Layer | -0.000 |
Meng | -0.000 |
Token | third |
Token ID | 17089 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
Lum | +0.024 |
Beam | +0.023 |
Lau | +0.022 |
Layer | +0.021 |
Meng | +0.020 |
Neg logit contributions | |
Augustus | -0.026 |
Cyrus | -0.024 |
Elias | -0.023 |
Jace | -0.023 |
rabbi | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | +4.63e-03 |
Pos logit contributions | |
Lum | +0.014 |
Beam | +0.013 |
Lau | +0.013 |
Layer | +0.012 |
Meng | +0.012 |
Neg logit contributions | |
Augustus | -0.016 |
Cyrus | -0.014 |
Elias | -0.014 |
Jace | -0.014 |
rabbi | -0.014 |
Token | 2 |
Token ID | 362 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
Augustus | +0.005 |
Cyrus | +0.004 |
Elias | +0.004 |
Jace | +0.004 |
rabbi | +0.004 |
Neg logit contributions | |
Lum | -0.004 |
Beam | -0.004 |
Lau | -0.004 |
Layer | -0.004 |
Meng | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +1.78e-03 |
Pos logit contributions | |
Augustus | +0.021 |
Cyrus | +0.019 |
Elias | +0.019 |
Jace | +0.019 |
rabbi | +0.019 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.016 |
Meng | -0.015 |
Token | 5 |
Token ID | 20 |
Feature activation | +7.19e-04 |
Pos logit contributions | |
Augustus | +0.002 |
Cyrus | +0.002 |
Elias | +0.002 |
Jace | +0.001 |
rabbi | +0.001 |
Neg logit contributions | |
Lum | -0.002 |
Beam | -0.002 |
Lau | -0.002 |
Layer | -0.002 |
Meng | -0.001 |
Token | per |
Token ID | 583 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
Augustus | +0.001 |
Cyrus | +0.001 |
Elias | +0.001 |
Jace | +0.001 |
rabbi | +0.001 |
Neg logit contributions | |
Lum | -0.001 |
Beam | -0.001 |
Lau | -0.000 |
Layer | -0.000 |
Meng | -0.000 |
Token | cent |
Token ID | 1247 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
Lum | +0.008 |
Beam | +0.007 |
Lau | +0.006 |
Layer | +0.005 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.042 |
Cyrus | -0.039 |
Elias | -0.039 |
Jace | -0.039 |
rabbi | -0.038 |
Token | increase |
Token ID | 2620 |
Feature activation | -1.51e-03 |
Pos logit contributions | |
Lum | +0.236 |
Beam | +0.226 |
Lau | +0.213 |
Layer | +0.209 |
Meng | +0.202 |
Neg logit contributions | |
Augustus | -0.245 |
Cyrus | -0.222 |
Elias | -0.219 |
Jace | -0.213 |
rabbi | -0.212 |
Token | in |
Token ID | 287 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +1.62e-04 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.017 |
Jace | +0.017 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.019 |
Beam | -0.018 |
Lau | -0.017 |
Layer | -0.017 |
Meng | -0.017 |
Token | first |
Token ID | 717 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
Augustus | +0.000 |
Cyrus | +0.000 |
Elias | +0.000 |
Jace | +0.000 |
rabbi | +0.000 |
Neg logit contributions | |
Lum | -0.000 |
Beam | -0.000 |
Lau | -0.000 |
Layer | -0.000 |
Meng | -0.000 |
Token | year |
Token ID | 614 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
Lum | +0.017 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.015 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.018 |
Cyrus | -0.016 |
Elias | -0.016 |
Jace | -0.015 |
rabbi | -0.015 |
Token | that |
Token ID | 326 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
Lum | +0.055 |
Beam | +0.052 |
Lau | +0.050 |
Layer | +0.049 |
Meng | +0.047 |
Neg logit contributions | |
Augustus | -0.048 |
Cyrus | -0.043 |
Elias | -0.043 |
Jace | -0.041 |
rabbi | -0.041 |
Token | would |
Token ID | 561 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
Augustus | +0.021 |
Cyrus | +0.019 |
Elias | +0.019 |
Jace | +0.018 |
rabbi | +0.018 |
Neg logit contributions | |
Lum | -0.019 |
Beam | -0.018 |
Lau | -0.017 |
Layer | -0.016 |
Meng | -0.016 |
Token | have |
Token ID | 423 |
Feature activation | +2.27e-03 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.010 |
Beam | -0.009 |
Lau | -0.009 |
Layer | -0.009 |
Meng | -0.008 |
Token | set |
Token ID | 900 |
Feature activation | +3.55e-02 |
Pos logit contributions | |
Augustus | +0.002 |
Cyrus | +0.002 |
Elias | +0.002 |
Jace | +0.002 |
rabbi | +0.002 |
Neg logit contributions | |
Lum | -0.002 |
Beam | -0.002 |
Lau | -0.002 |
Layer | -0.002 |
Meng | -0.002 |
Token | preced |
Token ID | 8555 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Augustus | +0.036 |
Cyrus | +0.032 |
Elias | +0.032 |
Jace | +0.031 |
rabbi | +0.031 |
Neg logit contributions | |
Lum | -0.032 |
Beam | -0.031 |
Lau | -0.029 |
Layer | -0.028 |
Meng | -0.027 |
Token | ents |
Token ID | 658 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
Augustus | +0.113 |
Cyrus | +0.105 |
Elias | +0.103 |
Jace | +0.101 |
rabbi | +0.100 |
Neg logit contributions | |
Lum | -0.079 |
Beam | -0.076 |
Lau | -0.071 |
Layer | -0.069 |
Meng | -0.066 |
Token | , |
Token ID | 11 |
Feature activation | +8.48e-04 |
Pos logit contributions | |
Lum | +0.250 |
Beam | +0.241 |
Lau | +0.228 |
Layer | +0.224 |
Meng | +0.217 |
Neg logit contributions | |
Augustus | -0.227 |
Cyrus | -0.204 |
Elias | -0.201 |
Jace | -0.196 |
rabbi | -0.194 |
Token | raised |
Token ID | 4376 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
Augustus | +0.001 |
Cyrus | +0.001 |
Elias | +0.001 |
Jace | +0.001 |
rabbi | +0.001 |
Neg logit contributions | |
Lum | -0.001 |
Beam | -0.001 |
Lau | -0.001 |
Layer | -0.001 |
Meng | -0.001 |
Token | expectations |
Token ID | 9027 |
Feature activation | +3.65e-02 |
Pos logit contributions | |
Lum | +0.080 |
Beam | +0.077 |
Lau | +0.072 |
Layer | +0.071 |
Meng | +0.068 |
Neg logit contributions | |
Augustus | -0.088 |
Cyrus | -0.080 |
Elias | -0.079 |
Jace | -0.077 |
rabbi | -0.076 |
Token | and |
Token ID | 290 |
Feature activation | -5.92e-04 |
Pos logit contributions | |
Augustus | +0.035 |
Cyrus | +0.032 |
Elias | +0.031 |
Jace | +0.031 |
rabbi | +0.030 |
Neg logit contributions | |
Lum | -0.034 |
Beam | -0.033 |
Lau | -0.031 |
Layer | -0.030 |
Meng | -0.029 |
Token | elevated |
Token ID | 15321 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
Lum | +0.000 |
Beam | +0.000 |
Lau | +0.000 |
Layer | +0.000 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
Lum | +0.012 |
Beam | +0.012 |
Lau | +0.011 |
Layer | +0.011 |
Meng | +0.010 |
Neg logit contributions | |
Augustus | -0.014 |
Cyrus | -0.012 |
Elias | -0.012 |
Jace | -0.012 |
rabbi | -0.012 |
Token | 0 |
Token ID | 657 |
Feature activation | +3.86e-03 |
Pos logit contributions | |
Augustus | +0.021 |
Cyrus | +0.019 |
Elias | +0.019 |
Jace | +0.018 |
rabbi | +0.018 |
Neg logit contributions | |
Lum | -0.017 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.015 |
Meng | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +6.72e-03 |
Pos logit contributions | |
Augustus | +0.004 |
Cyrus | +0.003 |
Elias | +0.003 |
Jace | +0.003 |
rabbi | +0.003 |
Neg logit contributions | |
Lum | -0.004 |
Beam | -0.003 |
Lau | -0.003 |
Layer | -0.003 |
Meng | -0.003 |
Token | 5 |
Token ID | 20 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
Augustus | +0.006 |
Cyrus | +0.006 |
Elias | +0.006 |
Jace | +0.006 |
rabbi | +0.006 |
Neg logit contributions | |
Lum | -0.006 |
Beam | -0.006 |
Lau | -0.006 |
Layer | -0.006 |
Meng | -0.005 |
Token | per |
Token ID | 583 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
Augustus | +0.023 |
Cyrus | +0.021 |
Elias | +0.021 |
Jace | +0.020 |
rabbi | +0.020 |
Neg logit contributions | |
Lum | -0.016 |
Beam | -0.016 |
Lau | -0.015 |
Layer | -0.014 |
Meng | -0.014 |
Token | cent |
Token ID | 1247 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.018 |
Cyrus | -0.017 |
Elias | -0.017 |
Jace | -0.016 |
rabbi | -0.016 |
Token | for |
Token ID | 329 |
Feature activation | -4.55e-03 |
Pos logit contributions | |
Lum | +0.246 |
Beam | +0.236 |
Lau | +0.223 |
Layer | +0.219 |
Meng | +0.212 |
Neg logit contributions | |
Augustus | -0.231 |
Cyrus | -0.209 |
Elias | -0.205 |
Jace | -0.200 |
rabbi | -0.198 |
Token | this |
Token ID | 428 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
Lum | +0.004 |
Beam | +0.004 |
Lau | +0.004 |
Layer | +0.004 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.005 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.004 |
Token | year |
Token ID | 614 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.017 |
Jace | +0.016 |
rabbi | +0.016 |
Neg logit contributions | |
Lum | -0.020 |
Beam | -0.019 |
Lau | -0.018 |
Layer | -0.018 |
Meng | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
Lum | +0.015 |
Beam | +0.014 |
Lau | +0.013 |
Layer | +0.013 |
Meng | +0.013 |
Neg logit contributions | |
Augustus | -0.017 |
Cyrus | -0.015 |
Elias | -0.015 |
Jace | -0.015 |
rabbi | -0.015 |
Token | MPs |
Token ID | 14952 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
Lum | +0.013 |
Beam | +0.012 |
Lau | +0.011 |
Layer | +0.011 |
Meng | +0.011 |
Neg logit contributions | |
Augustus | -0.015 |
Cyrus | -0.014 |
Elias | -0.013 |
Jace | -0.013 |
rabbi | -0.013 |
Token | as |
Token ID | 355 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.011 |
Beam | -0.010 |
Lau | -0.010 |
Layer | -0.010 |
Meng | -0.009 |
Token | much |
Token ID | 881 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
Augustus | +0.023 |
Cyrus | +0.021 |
Elias | +0.021 |
Jace | +0.020 |
rabbi | +0.020 |
Neg logit contributions | |
Lum | -0.016 |
Beam | -0.015 |
Lau | -0.014 |
Layer | -0.014 |
Meng | -0.013 |
Token | as |
Token ID | 355 |
Feature activation | +1.20e-03 |
Pos logit contributions | |
Lum | +0.024 |
Beam | +0.023 |
Lau | +0.021 |
Layer | +0.021 |
Meng | +0.020 |
Neg logit contributions | |
Augustus | -0.031 |
Cyrus | -0.028 |
Elias | -0.028 |
Jace | -0.027 |
rabbi | -0.027 |
Token | 20 |
Token ID | 1160 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
Augustus | +0.001 |
Cyrus | +0.001 |
Elias | +0.001 |
Jace | +0.001 |
rabbi | +0.001 |
Neg logit contributions | |
Lum | -0.001 |
Beam | -0.001 |
Lau | -0.001 |
Layer | -0.001 |
Meng | -0.001 |
Token | per |
Token ID | 583 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
Lum | +0.009 |
Beam | +0.008 |
Lau | +0.008 |
Layer | +0.008 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.011 |
Jace | -0.011 |
rabbi | -0.011 |
Token | cent |
Token ID | 1247 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Lum | +0.006 |
Beam | +0.005 |
Lau | +0.004 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.033 |
Cyrus | -0.031 |
Elias | -0.031 |
Jace | -0.030 |
rabbi | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
Lum | +0.233 |
Beam | +0.223 |
Lau | +0.211 |
Layer | +0.207 |
Meng | +0.200 |
Neg logit contributions | |
Augustus | -0.240 |
Cyrus | -0.218 |
Elias | -0.214 |
Jace | -0.209 |
rabbi | -0.208 |
Token |
|
Token ID | 198 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
Augustus | +0.015 |
Cyrus | +0.014 |
Elias | +0.014 |
Jace | +0.014 |
rabbi | +0.013 |
Neg logit contributions | |
Lum | -0.013 |
Beam | -0.012 |
Lau | -0.011 |
Layer | -0.011 |
Meng | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
Augustus | +0.026 |
Cyrus | +0.023 |
Elias | +0.023 |
Jace | +0.023 |
rabbi | +0.022 |
Neg logit contributions | |
Lum | -0.019 |
Beam | -0.018 |
Lau | -0.017 |
Layer | -0.017 |
Meng | -0.016 |
Token | Ad |
Token ID | 2782 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
Augustus | +0.023 |
Cyrus | +0.021 |
Elias | +0.021 |
Jace | +0.020 |
rabbi | +0.020 |
Neg logit contributions | |
Lum | -0.023 |
Beam | -0.022 |
Lau | -0.021 |
Layer | -0.020 |
Meng | -0.020 |
Token | vertising |
Token ID | 31809 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
Lum | +0.023 |
Beam | +0.022 |
Lau | +0.021 |
Layer | +0.021 |
Meng | +0.020 |
Neg logit contributions | |
Augustus | -0.020 |
Cyrus | -0.018 |
Elias | -0.018 |
Jace | -0.018 |
rabbi | -0.017 |
Token | ices |
Token ID | 1063 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
Lum | +0.103 |
Beam | +0.100 |
Lau | +0.096 |
Layer | +0.095 |
Meng | +0.093 |
Neg logit contributions | |
Augustus | -0.043 |
Cyrus | -0.036 |
Elias | -0.035 |
Jace | -0.034 |
rabbi | -0.033 |
Token | is |
Token ID | 318 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
Augustus | +0.149 |
Cyrus | +0.135 |
Elias | +0.133 |
Jace | +0.130 |
rabbi | +0.129 |
Neg logit contributions | |
Lum | -0.146 |
Beam | -0.140 |
Lau | -0.132 |
Layer | -0.129 |
Meng | -0.125 |
Token | called |
Token ID | 1444 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
Lum | +0.012 |
Beam | +0.012 |
Lau | +0.011 |
Layer | +0.011 |
Meng | +0.011 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | normal |
Token ID | 3487 |
Feature activation | -1.55e-03 |
Pos logit contributions | |
Lum | +0.017 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.015 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.022 |
Cyrus | -0.020 |
Elias | -0.020 |
Jace | -0.020 |
rabbi | -0.019 |
Token | extr |
Token ID | 22820 |
Feature activation | +1.84e-04 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | usion |
Token ID | 4241 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Augustus | +0.000 |
Cyrus | +0.000 |
Elias | +0.000 |
Jace | +0.000 |
rabbi | +0.000 |
Neg logit contributions | |
Lum | -0.000 |
Beam | -0.000 |
Lau | -0.000 |
Layer | -0.000 |
Meng | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -2.86e-03 |
Pos logit contributions | |
Lum | +0.236 |
Beam | +0.226 |
Lau | +0.214 |
Layer | +0.210 |
Meng | +0.203 |
Neg logit contributions | |
Augustus | -0.237 |
Cyrus | -0.214 |
Elias | -0.211 |
Jace | -0.206 |
rabbi | -0.204 |
Token | and |
Token ID | 290 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.003 |
Cyrus | -0.003 |
Elias | -0.003 |
Jace | -0.002 |
rabbi | -0.002 |
Token | is |
Token ID | 318 |
Feature activation | -1.52e-03 |
Pos logit contributions | |
Lum | +0.012 |
Beam | +0.011 |
Lau | +0.011 |
Layer | +0.010 |
Meng | +0.010 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.011 |
Jace | -0.011 |
rabbi | -0.011 |
Token | discussed |
Token ID | 6693 |
Feature activation | -1.85e-03 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | intensive |
Token ID | 18590 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.001 |
rabbi | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
Lum | +0.023 |
Beam | +0.022 |
Lau | +0.021 |
Layer | +0.020 |
Meng | +0.020 |
Neg logit contributions | |
Augustus | -0.029 |
Cyrus | -0.026 |
Elias | -0.026 |
Jace | -0.025 |
rabbi | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
Lum | +0.019 |
Beam | +0.018 |
Lau | +0.017 |
Layer | +0.016 |
Meng | +0.016 |
Neg logit contributions | |
Augustus | -0.025 |
Cyrus | -0.023 |
Elias | -0.023 |
Jace | -0.022 |
rabbi | -0.022 |
Token | The |
Token ID | 464 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.009 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.008 |
Neg logit contributions | |
Augustus | -0.009 |
Cyrus | -0.008 |
Elias | -0.008 |
Jace | -0.008 |
rabbi | -0.008 |
Token | normal |
Token ID | 3487 |
Feature activation | -6.31e-03 |
Pos logit contributions | |
Lum | +0.017 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.015 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.019 |
Cyrus | -0.017 |
Elias | -0.017 |
Jace | -0.017 |
rabbi | -0.017 |
Token | extr |
Token ID | 22820 |
Feature activation | +1.07e-03 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.005 |
Lau | +0.005 |
Layer | +0.005 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.006 |
Cyrus | -0.006 |
Elias | -0.006 |
Jace | -0.006 |
rabbi | -0.006 |
Token | usion |
Token ID | 4241 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
Augustus | +0.001 |
Cyrus | +0.001 |
Elias | +0.001 |
Jace | +0.001 |
rabbi | +0.001 |
Neg logit contributions | |
Lum | -0.001 |
Beam | -0.001 |
Lau | -0.000 |
Layer | -0.000 |
Meng | -0.000 |
Token | done |
Token ID | 1760 |
Feature activation | -4.23e-04 |
Pos logit contributions | |
Lum | +0.225 |
Beam | +0.215 |
Lau | +0.202 |
Layer | +0.199 |
Meng | +0.192 |
Neg logit contributions | |
Augustus | -0.246 |
Cyrus | -0.223 |
Elias | -0.220 |
Jace | -0.215 |
rabbi | -0.213 |
Token | in |
Token ID | 287 |
Feature activation | -4.50e-03 |
Pos logit contributions | |
Lum | +0.000 |
Beam | +0.000 |
Lau | +0.000 |
Layer | +0.000 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.000 |
Cyrus | -0.000 |
Elias | -0.000 |
Jace | -0.000 |
rabbi | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | -8.57e-03 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.005 |
Cyrus | -0.005 |
Elias | -0.005 |
Jace | -0.005 |
rabbi | -0.005 |
Token | previous |
Token ID | 2180 |
Feature activation | -2.53e-03 |
Pos logit contributions | |
Lum | +0.008 |
Beam | +0.008 |
Lau | +0.008 |
Layer | +0.007 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.008 |
Cyrus | -0.007 |
Elias | -0.007 |
Jace | -0.007 |
rabbi | -0.007 |
Token | section |
Token ID | 2665 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | that |
Token ID | 326 |
Feature activation | +4.09e-03 |
Pos logit contributions | |
Augustus | +0.046 |
Cyrus | +0.042 |
Elias | +0.041 |
Jace | +0.040 |
rabbi | +0.040 |
Neg logit contributions | |
Lum | -0.048 |
Beam | -0.046 |
Lau | -0.044 |
Layer | -0.043 |
Meng | -0.041 |
Token | we |
Token ID | 356 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
Augustus | +0.004 |
Cyrus | +0.004 |
Elias | +0.004 |
Jace | +0.004 |
rabbi | +0.004 |
Neg logit contributions | |
Lum | -0.004 |
Beam | -0.004 |
Lau | -0.003 |
Layer | -0.003 |
Meng | -0.003 |
Token | want |
Token ID | 765 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
Augustus | +0.025 |
Cyrus | +0.023 |
Elias | +0.023 |
Jace | +0.022 |
rabbi | +0.022 |
Neg logit contributions | |
Lum | -0.022 |
Beam | -0.021 |
Lau | -0.020 |
Layer | -0.019 |
Meng | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -8.30e-03 |
Pos logit contributions | |
Lum | +0.025 |
Beam | +0.024 |
Lau | +0.023 |
Layer | +0.022 |
Meng | +0.022 |
Neg logit contributions | |
Augustus | -0.027 |
Cyrus | -0.024 |
Elias | -0.024 |
Jace | -0.023 |
rabbi | -0.023 |
Token | extr |
Token ID | 22820 |
Feature activation | -6.39e-03 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.007 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.009 |
Cyrus | -0.008 |
Elias | -0.008 |
Jace | -0.008 |
rabbi | -0.008 |
Token | usion |
Token ID | 4241 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.009 |
Cyrus | -0.008 |
Elias | -0.008 |
Jace | -0.008 |
rabbi | -0.008 |
Token | value |
Token ID | 1988 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
Lum | +0.236 |
Beam | +0.226 |
Lau | +0.214 |
Layer | +0.210 |
Meng | +0.203 |
Neg logit contributions | |
Augustus | -0.231 |
Cyrus | -0.209 |
Elias | -0.205 |
Jace | -0.200 |
rabbi | -0.198 |
Token | to |
Token ID | 284 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
Lum | +0.028 |
Beam | +0.027 |
Lau | +0.025 |
Layer | +0.025 |
Meng | +0.024 |
Neg logit contributions | |
Augustus | -0.027 |
Cyrus | -0.024 |
Elias | -0.024 |
Jace | -0.023 |
rabbi | -0.023 |
Token | go |
Token ID | 467 |
Feature activation | +1.81e-03 |
Pos logit contributions | |
Augustus | +0.016 |
Cyrus | +0.014 |
Elias | +0.014 |
Jace | +0.014 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.013 |
Beam | -0.013 |
Lau | -0.012 |
Layer | -0.012 |
Meng | -0.011 |
Token | from |
Token ID | 422 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
Augustus | +0.002 |
Cyrus | +0.002 |
Elias | +0.002 |
Jace | +0.002 |
rabbi | +0.002 |
Neg logit contributions | |
Lum | -0.002 |
Beam | -0.002 |
Lau | -0.001 |
Layer | -0.001 |
Meng | -0.001 |
Token | o |
Token ID | 267 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
Lum | +0.024 |
Beam | +0.023 |
Lau | +0.021 |
Layer | +0.021 |
Meng | +0.020 |
Neg logit contributions | |
Augustus | -0.032 |
Cyrus | -0.029 |
Elias | -0.028 |
Jace | -0.028 |
rabbi | -0.028 |
Token | home |
Token ID | 1363 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.003 |
Cyrus | -0.003 |
Elias | -0.003 |
Jace | -0.003 |
rabbi | -0.003 |
Token | habits |
Token ID | 13870 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
Augustus | +0.029 |
Cyrus | +0.026 |
Elias | +0.026 |
Jace | +0.025 |
rabbi | +0.025 |
Neg logit contributions | |
Lum | -0.028 |
Beam | -0.027 |
Lau | -0.025 |
Layer | -0.025 |
Meng | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.011 |
Beam | -0.011 |
Lau | -0.010 |
Layer | -0.010 |
Meng | -0.009 |
Token | 42 |
Token ID | 5433 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
Augustus | +0.013 |
Cyrus | +0.012 |
Elias | +0.012 |
Jace | +0.012 |
rabbi | +0.011 |
Neg logit contributions | |
Lum | -0.011 |
Beam | -0.011 |
Lau | -0.010 |
Layer | -0.010 |
Meng | -0.009 |
Token | per |
Token ID | 583 |
Feature activation | -7.76e-04 |
Pos logit contributions | |
Augustus | +0.026 |
Cyrus | +0.024 |
Elias | +0.024 |
Jace | +0.023 |
rabbi | +0.023 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.016 |
Meng | -0.015 |
Token | cent |
Token ID | 1247 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
Lum | +0.000 |
Beam | +0.000 |
Lau | +0.000 |
Layer | +0.000 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | of |
Token ID | 286 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
Lum | +0.247 |
Beam | +0.237 |
Lau | +0.225 |
Layer | +0.221 |
Meng | +0.214 |
Neg logit contributions | |
Augustus | -0.220 |
Cyrus | -0.198 |
Elias | -0.194 |
Jace | -0.189 |
rabbi | -0.187 |
Token | workers |
Token ID | 3259 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
Augustus | +0.042 |
Cyrus | +0.038 |
Elias | +0.038 |
Jace | +0.037 |
rabbi | +0.036 |
Neg logit contributions | |
Lum | -0.044 |
Beam | -0.042 |
Lau | -0.040 |
Layer | -0.039 |
Meng | -0.038 |
Token | revealed |
Token ID | 4602 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.012 |
Beam | -0.011 |
Lau | -0.010 |
Layer | -0.010 |
Meng | -0.010 |
Token | that |
Token ID | 326 |
Feature activation | +4.91e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.010 |
Cyrus | -0.009 |
Elias | -0.009 |
Jace | -0.009 |
rabbi | -0.009 |
Token | they |
Token ID | 484 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
Augustus | +0.051 |
Cyrus | +0.046 |
Elias | +0.046 |
Jace | +0.045 |
rabbi | +0.044 |
Neg logit contributions | |
Lum | -0.042 |
Beam | -0.040 |
Lau | -0.038 |
Layer | -0.037 |
Meng | -0.036 |
Token | increased |
Token ID | 3220 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.004 |
Token | more |
Token ID | 517 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
Lum | +0.019 |
Beam | +0.018 |
Lau | +0.017 |
Layer | +0.017 |
Meng | +0.016 |
Neg logit contributions | |
Augustus | -0.021 |
Cyrus | -0.019 |
Elias | -0.019 |
Jace | -0.018 |
rabbi | -0.018 |
Token | than |
Token ID | 621 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
Augustus | +0.005 |
Cyrus | +0.005 |
Elias | +0.005 |
Jace | +0.005 |
rabbi | +0.005 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.005 |
Lau | -0.005 |
Layer | -0.005 |
Meng | -0.004 |
Token | 100 |
Token ID | 1802 |
Feature activation | +3.49e-03 |
Pos logit contributions | |
Lum | +0.033 |
Beam | +0.031 |
Lau | +0.029 |
Layer | +0.029 |
Meng | +0.028 |
Neg logit contributions | |
Augustus | -0.044 |
Cyrus | -0.040 |
Elias | -0.040 |
Jace | -0.039 |
rabbi | -0.038 |
Token | per |
Token ID | 583 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
Augustus | +0.004 |
Cyrus | +0.004 |
Elias | +0.004 |
Jace | +0.004 |
rabbi | +0.003 |
Neg logit contributions | |
Lum | -0.003 |
Beam | -0.003 |
Lau | -0.002 |
Layer | -0.002 |
Meng | -0.002 |
Token | cent |
Token ID | 1247 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.018 |
Cyrus | -0.017 |
Elias | -0.017 |
Jace | -0.016 |
rabbi | -0.016 |
Token | in |
Token ID | 287 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
Lum | +0.221 |
Beam | +0.211 |
Lau | +0.199 |
Layer | +0.194 |
Meng | +0.188 |
Neg logit contributions | |
Augustus | -0.246 |
Cyrus | -0.224 |
Elias | -0.220 |
Jace | -0.215 |
rabbi | -0.214 |
Token | the |
Token ID | 262 |
Feature activation | +8.86e-03 |
Pos logit contributions | |
Augustus | +0.018 |
Cyrus | +0.016 |
Elias | +0.016 |
Jace | +0.016 |
rabbi | +0.016 |
Neg logit contributions | |
Lum | -0.017 |
Beam | -0.016 |
Lau | -0.015 |
Layer | -0.015 |
Meng | -0.014 |
Token | last |
Token ID | 938 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
Augustus | +0.007 |
Cyrus | +0.007 |
Elias | +0.006 |
Jace | +0.006 |
rabbi | +0.006 |
Neg logit contributions | |
Lum | -0.009 |
Beam | -0.009 |
Lau | -0.009 |
Layer | -0.009 |
Meng | -0.008 |
Token | year |
Token ID | 614 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
Lum | +0.014 |
Beam | +0.014 |
Lau | +0.013 |
Layer | +0.013 |
Meng | +0.012 |
Neg logit contributions | |
Augustus | -0.016 |
Cyrus | -0.014 |
Elias | -0.014 |
Jace | -0.014 |
rabbi | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +4.40e-03 |
Pos logit contributions | |
Lum | +0.024 |
Beam | +0.023 |
Lau | +0.022 |
Layer | +0.021 |
Meng | +0.021 |
Neg logit contributions | |
Augustus | -0.025 |
Cyrus | -0.022 |
Elias | -0.022 |
Jace | -0.021 |
rabbi | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
Augustus | +0.035 |
Cyrus | +0.032 |
Elias | +0.031 |
Jace | +0.031 |
rabbi | +0.030 |
Neg logit contributions | |
Lum | -0.031 |
Beam | -0.030 |
Lau | -0.028 |
Layer | -0.027 |
Meng | -0.027 |
Token | mess |
Token ID | 2085 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
Augustus | +0.021 |
Cyrus | +0.019 |
Elias | +0.019 |
Jace | +0.019 |
rabbi | +0.019 |
Neg logit contributions | |
Lum | -0.019 |
Beam | -0.018 |
Lau | -0.017 |
Layer | -0.017 |
Meng | -0.016 |
Token | of |
Token ID | 286 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
Augustus | +0.031 |
Cyrus | +0.028 |
Elias | +0.028 |
Jace | +0.027 |
rabbi | +0.027 |
Neg logit contributions | |
Lum | -0.031 |
Beam | -0.030 |
Lau | -0.028 |
Layer | -0.028 |
Meng | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.017 |
Jace | +0.017 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.016 |
Meng | -0.015 |
Token | inco |
Token ID | 44237 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
Augustus | +0.029 |
Cyrus | +0.027 |
Elias | +0.026 |
Jace | +0.026 |
rabbi | +0.025 |
Neg logit contributions | |
Lum | -0.027 |
Beam | -0.026 |
Lau | -0.024 |
Layer | -0.024 |
Meng | -0.023 |
Token | herent |
Token ID | 8334 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
Augustus | +0.055 |
Cyrus | +0.051 |
Elias | +0.051 |
Jace | +0.050 |
rabbi | +0.049 |
Neg logit contributions | |
Lum | -0.024 |
Beam | -0.023 |
Lau | -0.020 |
Layer | -0.020 |
Meng | -0.019 |
Token | politics |
Token ID | 4819 |
Feature activation | +6.72e-02 |
Pos logit contributions | |
Lum | +0.223 |
Beam | +0.214 |
Lau | +0.202 |
Layer | +0.198 |
Meng | +0.191 |
Neg logit contributions | |
Augustus | -0.242 |
Cyrus | -0.220 |
Elias | -0.217 |
Jace | -0.212 |
rabbi | -0.210 |
Token | that |
Token ID | 326 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
Augustus | +0.062 |
Cyrus | +0.056 |
Elias | +0.055 |
Jace | +0.053 |
rabbi | +0.053 |
Neg logit contributions | |
Lum | -0.066 |
Beam | -0.063 |
Lau | -0.060 |
Layer | -0.059 |
Meng | -0.057 |
Token | is |
Token ID | 318 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
Augustus | +0.028 |
Cyrus | +0.026 |
Elias | +0.025 |
Jace | +0.025 |
rabbi | +0.025 |
Neg logit contributions | |
Lum | -0.025 |
Beam | -0.024 |
Lau | -0.023 |
Layer | -0.022 |
Meng | -0.021 |
Token | has |
Token ID | 468 |
Feature activation | +1.06e-03 |
Pos logit contributions | |
Augustus | +0.025 |
Cyrus | +0.023 |
Elias | +0.023 |
Jace | +0.022 |
rabbi | +0.022 |
Neg logit contributions | |
Lum | -0.023 |
Beam | -0.022 |
Lau | -0.021 |
Layer | -0.021 |
Meng | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | +9.30e-03 |
Pos logit contributions | |
Augustus | +0.001 |
Cyrus | +0.001 |
Elias | +0.001 |
Jace | +0.001 |
rabbi | +0.001 |
Neg logit contributions | |
Lum | -0.001 |
Beam | -0.001 |
Lau | -0.001 |
Layer | -0.001 |
Meng | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +5.86e-03 |
Pos logit contributions | |
Lum | +0.015 |
Beam | +0.014 |
Lau | +0.013 |
Layer | +0.013 |
Meng | +0.013 |
Neg logit contributions | |
Augustus | -0.015 |
Cyrus | -0.014 |
Elias | -0.014 |
Jace | -0.013 |
rabbi | -0.013 |
Token | or |
Token ID | 393 |
Feature activation | +5.90e-03 |
Pos logit contributions | |
Augustus | +0.006 |
Cyrus | +0.005 |
Elias | +0.005 |
Jace | +0.005 |
rabbi | +0.005 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.005 |
Lau | -0.005 |
Layer | -0.005 |
Meng | -0.004 |
Token | about |
Token ID | 546 |
Feature activation | -3.84e-03 |
Pos logit contributions | |
Augustus | +0.006 |
Cyrus | +0.005 |
Elias | +0.005 |
Jace | +0.005 |
rabbi | +0.005 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.005 |
Lau | -0.005 |
Layer | -0.005 |
Meng | -0.005 |
Token | three |
Token ID | 1115 |
Feature activation | -3.99e-03 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.004 |
Token | per |
Token ID | 583 |
Feature activation | -9.69e-03 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.004 |
Token | cent |
Token ID | 1247 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.015 |
Cyrus | -0.015 |
Elias | -0.014 |
Jace | -0.014 |
rabbi | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
Lum | +0.249 |
Beam | +0.239 |
Lau | +0.227 |
Layer | +0.223 |
Meng | +0.216 |
Neg logit contributions | |
Augustus | -0.215 |
Cyrus | -0.193 |
Elias | -0.190 |
Jace | -0.185 |
rabbi | -0.183 |
Token | the |
Token ID | 262 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
Augustus | +0.034 |
Cyrus | +0.031 |
Elias | +0.030 |
Jace | +0.030 |
rabbi | +0.029 |
Neg logit contributions | |
Lum | -0.031 |
Beam | -0.029 |
Lau | -0.028 |
Layer | -0.027 |
Meng | -0.026 |
Token | 9 |
Token ID | 860 |
Feature activation | -4.22e-03 |
Pos logit contributions | |
Augustus | +0.035 |
Cyrus | +0.031 |
Elias | +0.031 |
Jace | +0.030 |
rabbi | +0.030 |
Neg logit contributions | |
Lum | -0.033 |
Beam | -0.031 |
Lau | -0.029 |
Layer | -0.029 |
Meng | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +7.48e-03 |
Pos logit contributions | |
Lum | +0.004 |
Beam | +0.004 |
Lau | +0.004 |
Layer | +0.004 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.003 |
rabbi | -0.003 |
Token | 204 |
Token ID | 18638 |
Feature activation | +6.09e-04 |
Pos logit contributions | |
Augustus | +0.007 |
Cyrus | +0.006 |
Elias | +0.006 |
Jace | +0.006 |
rabbi | +0.006 |
Neg logit contributions | |
Lum | -0.007 |
Beam | -0.007 |
Lau | -0.007 |
Layer | -0.007 |
Meng | -0.006 |
Token | wrote |
Token ID | 2630 |
Feature activation | +5.98e-03 |
Pos logit contributions | |
Lum | +0.118 |
Beam | +0.113 |
Lau | +0.108 |
Layer | +0.105 |
Meng | +0.103 |
Neg logit contributions | |
Augustus | -0.097 |
Cyrus | -0.087 |
Elias | -0.085 |
Jace | -0.083 |
rabbi | -0.082 |
Token | : |
Token ID | 25 |
Feature activation | -8.46e-02 |
Pos logit contributions | |
Augustus | +0.006 |
Cyrus | +0.006 |
Elias | +0.006 |
Jace | +0.006 |
rabbi | +0.006 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.005 |
Lau | -0.004 |
Layer | -0.004 |
Meng | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -3.55e-03 |
Pos logit contributions | |
Lum | +0.068 |
Beam | +0.064 |
Lau | +0.060 |
Layer | +0.059 |
Meng | +0.056 |
Neg logit contributions | |
Augustus | -0.093 |
Cyrus | -0.085 |
Elias | -0.084 |
Jace | -0.082 |
rabbi | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | -6.61e-03 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.003 |
rabbi | -0.003 |
Token | Cl |
Token ID | 2601 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
Lum | +0.006 |
Beam | +0.006 |
Lau | +0.006 |
Layer | +0.005 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.006 |
Cyrus | -0.006 |
Elias | -0.006 |
Jace | -0.006 |
rabbi | -0.006 |
Token | arity |
Token ID | 6806 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
Augustus | +0.041 |
Cyrus | +0.037 |
Elias | +0.036 |
Jace | +0.035 |
rabbi | +0.035 |
Neg logit contributions | |
Lum | -0.045 |
Beam | -0.043 |
Lau | -0.041 |
Layer | -0.040 |
Meng | -0.039 |
Token | Gaming |
Token ID | 14426 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
Augustus | +0.167 |
Cyrus | +0.152 |
Elias | +0.150 |
Jace | +0.146 |
rabbi | +0.145 |
Neg logit contributions | |
Lum | -0.152 |
Beam | -0.145 |
Lau | -0.137 |
Layer | -0.134 |
Meng | -0.129 |
Token | will |
Token ID | 481 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
Augustus | +0.042 |
Cyrus | +0.038 |
Elias | +0.037 |
Jace | +0.037 |
rabbi | +0.036 |
Neg logit contributions | |
Lum | -0.038 |
Beam | -0.036 |
Lau | -0.034 |
Layer | -0.034 |
Meng | -0.033 |
Token | also |
Token ID | 635 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
Augustus | +0.022 |
Cyrus | +0.020 |
Elias | +0.019 |
Jace | +0.019 |
rabbi | +0.019 |
Neg logit contributions | |
Lum | -0.020 |
Beam | -0.019 |
Lau | -0.018 |
Layer | -0.017 |
Meng | -0.017 |
Token | be |
Token ID | 307 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
Augustus | +0.027 |
Cyrus | +0.024 |
Elias | +0.024 |
Jace | +0.023 |
rabbi | +0.023 |
Neg logit contributions | |
Lum | -0.024 |
Beam | -0.023 |
Lau | -0.022 |
Layer | -0.021 |
Meng | -0.020 |
Token | releasing |
Token ID | 13011 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
Lum | +0.032 |
Beam | +0.030 |
Lau | +0.029 |
Layer | +0.028 |
Meng | +0.027 |
Neg logit contributions | |
Augustus | -0.030 |
Cyrus | -0.027 |
Elias | -0.026 |
Jace | -0.026 |
rabbi | -0.026 |
Token | statue |
Token ID | 15207 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
Lum | +0.040 |
Beam | +0.038 |
Lau | +0.036 |
Layer | +0.035 |
Meng | +0.034 |
Neg logit contributions | |
Augustus | -0.045 |
Cyrus | -0.041 |
Elias | -0.040 |
Jace | -0.039 |
rabbi | -0.039 |
Token | in |
Token ID | 287 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
Lum | +0.022 |
Beam | +0.021 |
Lau | +0.020 |
Layer | +0.019 |
Meng | +0.019 |
Neg logit contributions | |
Augustus | -0.024 |
Cyrus | -0.022 |
Elias | -0.021 |
Jace | -0.021 |
rabbi | -0.021 |
Token | London |
Token ID | 3576 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.009 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.008 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.009 |
Jace | -0.009 |
rabbi | -0.009 |
Token | 's |
Token ID | 338 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
Augustus | +0.010 |
Cyrus | +0.009 |
Elias | +0.009 |
Jace | +0.009 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.009 |
Beam | -0.009 |
Lau | -0.008 |
Layer | -0.008 |
Meng | -0.008 |
Token | Kens |
Token ID | 29018 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.010 |
Beam | -0.010 |
Lau | -0.009 |
Layer | -0.009 |
Meng | -0.009 |
Token | ington |
Token ID | 9557 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
Lum | +0.044 |
Beam | +0.042 |
Lau | +0.040 |
Layer | +0.040 |
Meng | +0.039 |
Neg logit contributions | |
Augustus | -0.022 |
Cyrus | -0.019 |
Elias | -0.019 |
Jace | -0.018 |
rabbi | -0.018 |
Token | Gardens |
Token ID | 29330 |
Feature activation | +9.97e-03 |
Pos logit contributions | |
Augustus | +0.181 |
Cyrus | +0.165 |
Elias | +0.162 |
Jace | +0.159 |
rabbi | +0.157 |
Neg logit contributions | |
Lum | -0.161 |
Beam | -0.154 |
Lau | -0.145 |
Layer | -0.142 |
Meng | -0.137 |
Token | has |
Token ID | 468 |
Feature activation | -9.61e-03 |
Pos logit contributions | |
Augustus | +0.010 |
Cyrus | +0.009 |
Elias | +0.009 |
Jace | +0.009 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.009 |
Beam | -0.008 |
Lau | -0.008 |
Layer | -0.008 |
Meng | -0.008 |
Token | been |
Token ID | 587 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
Lum | +0.008 |
Beam | +0.008 |
Lau | +0.007 |
Layer | +0.007 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.010 |
Cyrus | -0.009 |
Elias | -0.009 |
Jace | -0.009 |
rabbi | -0.009 |
Token | at |
Token ID | 379 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
Lum | +0.035 |
Beam | +0.033 |
Lau | +0.031 |
Layer | +0.031 |
Meng | +0.030 |
Neg logit contributions | |
Augustus | -0.041 |
Cyrus | -0.037 |
Elias | -0.036 |
Jace | -0.036 |
rabbi | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -3.26e-03 |
Pos logit contributions | |
Augustus | +0.025 |
Cyrus | +0.023 |
Elias | +0.022 |
Jace | +0.022 |
rabbi | +0.022 |
Neg logit contributions | |
Lum | -0.022 |
Beam | -0.021 |
Lau | -0.020 |
Layer | -0.020 |
Meng | -0.019 |
Token | little |
Token ID | 1310 |
Feature activation | -5.21e-05 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.006 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.007 |
Cyrus | -0.007 |
Elias | -0.007 |
Jace | -0.006 |
rabbi | -0.006 |
Token | pressing |
Token ID | 12273 |
Feature activation | +7.49e-03 |
Pos logit contributions | |
Lum | +0.000 |
Beam | +0.000 |
Lau | +0.000 |
Layer | +0.000 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.000 |
Cyrus | -0.000 |
Elias | -0.000 |
Jace | -0.000 |
rabbi | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +9.71e-03 |
Pos logit contributions | |
Augustus | +0.007 |
Cyrus | +0.007 |
Elias | +0.006 |
Jace | +0.006 |
rabbi | +0.006 |
Neg logit contributions | |
Lum | -0.007 |
Beam | -0.007 |
Lau | -0.006 |
Layer | -0.006 |
Meng | -0.006 |
Token | few |
Token ID | 1178 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
Augustus | +0.010 |
Cyrus | +0.009 |
Elias | +0.009 |
Jace | +0.009 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.008 |
Beam | -0.008 |
Lau | -0.008 |
Layer | -0.007 |
Meng | -0.007 |
Token | penet |
Token ID | 11958 |
Feature activation | -2.83e-03 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.013 |
Cyrus | -0.012 |
Elias | -0.012 |
Jace | -0.011 |
rabbi | -0.011 |
Token | rative |
Token ID | 13260 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.003 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | runs |
Token ID | 4539 |
Feature activation | -1.17e-03 |
Pos logit contributions | |
Augustus | +0.213 |
Cyrus | +0.193 |
Elias | +0.191 |
Jace | +0.186 |
rabbi | +0.184 |
Neg logit contributions | |
Lum | -0.200 |
Beam | -0.191 |
Lau | -0.180 |
Layer | -0.177 |
Meng | -0.171 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-03 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
Lum | +0.022 |
Beam | +0.021 |
Lau | +0.020 |
Layer | +0.019 |
Meng | +0.019 |
Neg logit contributions | |
Augustus | -0.030 |
Cyrus | -0.028 |
Elias | -0.028 |
Jace | -0.027 |
rabbi | -0.027 |
Token | The |
Token ID | 464 |
Feature activation | +4.81e-03 |
Pos logit contributions | |
Lum | +0.021 |
Beam | +0.020 |
Lau | +0.019 |
Layer | +0.019 |
Meng | +0.018 |
Neg logit contributions | |
Augustus | -0.022 |
Cyrus | -0.020 |
Elias | -0.019 |
Jace | -0.019 |
rabbi | -0.019 |
Token | Man |
Token ID | 1869 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
Lum | +0.031 |
Beam | +0.029 |
Lau | +0.026 |
Layer | +0.025 |
Meng | +0.023 |
Neg logit contributions | |
Augustus | -0.085 |
Cyrus | -0.080 |
Elias | -0.079 |
Jace | -0.078 |
rabbi | -0.078 |
Token | agers |
Token ID | 10321 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
Augustus | +0.042 |
Cyrus | +0.039 |
Elias | +0.038 |
rabbi | +0.037 |
Jace | +0.037 |
Neg logit contributions | |
Lum | -0.029 |
Beam | -0.028 |
Lau | -0.026 |
Layer | -0.025 |
Meng | -0.024 |
Token | � |
Token ID | 447 |
Feature activation | +4.57e-03 |
Pos logit contributions | |
Lum | +0.049 |
Beam | +0.046 |
Lau | +0.043 |
Layer | +0.042 |
Meng | +0.041 |
Neg logit contributions | |
Augustus | -0.067 |
Cyrus | -0.061 |
Elias | -0.060 |
Jace | -0.059 |
rabbi | -0.059 |
Token | � |
Token ID | 247 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
Augustus | +0.004 |
Cyrus | +0.004 |
Elias | +0.004 |
Jace | +0.004 |
rabbi | +0.004 |
Neg logit contributions | |
Lum | -0.004 |
Beam | -0.004 |
Lau | -0.004 |
Layer | -0.004 |
Meng | -0.004 |
Token | Ind |
Token ID | 1423 |
Feature activation | +3.73e-05 |
Pos logit contributions | |
Lum | +0.018 |
Beam | +0.018 |
Lau | +0.017 |
Layer | +0.016 |
Meng | +0.016 |
Neg logit contributions | |
Augustus | -0.019 |
Cyrus | -0.017 |
Elias | -0.017 |
Jace | -0.016 |
rabbi | -0.016 |
Token | ices |
Token ID | 1063 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
Augustus | +0.000 |
Cyrus | +0.000 |
Elias | +0.000 |
Jace | +0.000 |
rabbi | +0.000 |
Neg logit contributions | |
Lum | -0.000 |
Beam | -0.000 |
Lau | -0.000 |
Layer | -0.000 |
Meng | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +5.79e-03 |
Pos logit contributions | |
Augustus | +0.200 |
Cyrus | +0.182 |
Elias | +0.179 |
Jace | +0.174 |
rabbi | +0.172 |
Neg logit contributions | |
Lum | -0.190 |
Beam | -0.181 |
Lau | -0.172 |
Layer | -0.168 |
Meng | -0.163 |
Token |
|
Token ID | 198 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
Augustus | +0.006 |
Cyrus | +0.005 |
Elias | +0.005 |
Jace | +0.005 |
rabbi | +0.005 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.005 |
Lau | -0.004 |
Layer | -0.004 |
Meng | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.018 |
Elias | +0.017 |
Jace | +0.017 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.014 |
Beam | -0.014 |
Lau | -0.013 |
Layer | -0.012 |
Meng | -0.012 |
Token | � |
Token ID | 447 |
Feature activation | -5.83e-03 |
Pos logit contributions | |
Augustus | +0.020 |
Cyrus | +0.018 |
Elias | +0.018 |
Jace | +0.018 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.018 |
Lau | -0.017 |
Layer | -0.016 |
Meng | -0.016 |
Token | � |
Token ID | 250 |
Feature activation | -6.36e-03 |
Pos logit contributions | |
Lum | +0.006 |
Beam | +0.006 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.005 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.004 |
Token | becomes |
Token ID | 4329 |
Feature activation | +7.88e-03 |
Pos logit contributions | |
Lum | +0.014 |
Beam | +0.014 |
Lau | +0.013 |
Layer | +0.013 |
Meng | +0.012 |
Neg logit contributions | |
Augustus | -0.017 |
Cyrus | -0.015 |
Elias | -0.015 |
Jace | -0.015 |
rabbi | -0.015 |
Token | older |
Token ID | 4697 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
Augustus | +0.008 |
Cyrus | +0.007 |
Elias | +0.007 |
Jace | +0.007 |
rabbi | +0.007 |
Neg logit contributions | |
Lum | -0.007 |
Beam | -0.007 |
Lau | -0.006 |
Layer | -0.006 |
Meng | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
Lum | +0.015 |
Beam | +0.014 |
Lau | +0.014 |
Layer | +0.013 |
Meng | +0.013 |
Neg logit contributions | |
Augustus | -0.015 |
Cyrus | -0.014 |
Elias | -0.014 |
Jace | -0.013 |
rabbi | -0.013 |
Token | more |
Token ID | 517 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.009 |
Lau | +0.009 |
Layer | +0.008 |
Meng | +0.008 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.009 |
Token | ethn |
Token ID | 33961 |
Feature activation | -5.72e-02 |
Pos logit contributions | |
Lum | +0.020 |
Beam | +0.019 |
Lau | +0.018 |
Layer | +0.018 |
Meng | +0.017 |
Neg logit contributions | |
Augustus | -0.024 |
Cyrus | -0.022 |
Elias | -0.021 |
Jace | -0.021 |
rabbi | -0.021 |
Token | ically |
Token ID | 1146 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
Lum | +0.064 |
Beam | +0.062 |
Lau | +0.059 |
Layer | +0.058 |
Meng | +0.057 |
Neg logit contributions | |
Augustus | -0.044 |
Cyrus | -0.039 |
Elias | -0.039 |
Jace | -0.037 |
rabbi | -0.037 |
Token | diverse |
Token ID | 10084 |
Feature activation | -1.32e-03 |
Pos logit contributions | |
Augustus | +0.198 |
Cyrus | +0.182 |
Elias | +0.180 |
Jace | +0.175 |
rabbi | +0.174 |
Neg logit contributions | |
Lum | -0.149 |
Beam | -0.142 |
Lau | -0.133 |
Layer | -0.130 |
Meng | -0.125 |
Token | , |
Token ID | 11 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -9.37e-03 |
Pos logit contributions | |
Lum | +0.028 |
Beam | +0.027 |
Lau | +0.025 |
Layer | +0.025 |
Meng | +0.024 |
Neg logit contributions | |
Augustus | -0.031 |
Cyrus | -0.028 |
Elias | -0.028 |
Jace | -0.027 |
rabbi | -0.027 |
Token | additional |
Token ID | 3224 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
Lum | +0.008 |
Beam | +0.008 |
Lau | +0.007 |
Layer | +0.007 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.010 |
Cyrus | -0.009 |
Elias | -0.009 |
Jace | -0.008 |
rabbi | -0.008 |
Token | challenges |
Token ID | 6459 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
Lum | +0.036 |
Beam | +0.034 |
Lau | +0.032 |
Layer | +0.032 |
Meng | +0.030 |
Neg logit contributions | |
Augustus | -0.045 |
Cyrus | -0.041 |
Elias | -0.041 |
Jace | -0.040 |
rabbi | -0.039 |
Token | be |
Token ID | 307 |
Feature activation | +9.66e-03 |
Pos logit contributions | |
Augustus | +0.026 |
Cyrus | +0.024 |
Elias | +0.024 |
Jace | +0.023 |
rabbi | +0.023 |
Neg logit contributions | |
Lum | -0.023 |
Beam | -0.022 |
Lau | -0.020 |
Layer | -0.020 |
Meng | -0.019 |
Token | flowing |
Token ID | 17609 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
Augustus | +0.010 |
Cyrus | +0.009 |
Elias | +0.009 |
Jace | +0.009 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.008 |
Beam | -0.008 |
Lau | -0.008 |
Layer | -0.007 |
Meng | -0.007 |
Token | freely |
Token ID | 12748 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
Lum | +0.017 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.015 |
Meng | +0.015 |
Neg logit contributions | |
Augustus | -0.018 |
Cyrus | -0.016 |
Elias | -0.016 |
Jace | -0.015 |
rabbi | -0.015 |
Token | during |
Token ID | 1141 |
Feature activation | -8.14e-03 |
Pos logit contributions | |
Lum | +0.024 |
Beam | +0.023 |
Lau | +0.022 |
Layer | +0.022 |
Meng | +0.021 |
Neg logit contributions | |
Augustus | -0.025 |
Cyrus | -0.022 |
Elias | -0.022 |
Jace | -0.021 |
rabbi | -0.021 |
Token | these |
Token ID | 777 |
Feature activation | +7.90e-03 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.007 |
Lau | +0.007 |
Layer | +0.007 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.008 |
Cyrus | -0.007 |
Elias | -0.007 |
Jace | -0.007 |
rabbi | -0.007 |
Token | events |
Token ID | 2995 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
Augustus | +0.007 |
Cyrus | +0.007 |
Elias | +0.007 |
Jace | +0.006 |
rabbi | +0.006 |
Neg logit contributions | |
Lum | -0.008 |
Beam | -0.007 |
Lau | -0.007 |
Layer | -0.007 |
Meng | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
Augustus | +0.056 |
Cyrus | +0.051 |
Elias | +0.050 |
Jace | +0.049 |
rabbi | +0.048 |
Neg logit contributions | |
Lum | -0.055 |
Beam | -0.052 |
Lau | -0.049 |
Layer | -0.049 |
Meng | -0.047 |
Token | prep |
Token ID | 3143 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -3.01e-03 |
Pos logit contributions | |
Augustus | +0.031 |
Cyrus | +0.028 |
Elias | +0.028 |
Jace | +0.027 |
rabbi | +0.027 |
Neg logit contributions | |
Lum | -0.032 |
Beam | -0.030 |
Lau | -0.029 |
Layer | -0.028 |
Meng | -0.027 |
Token | party |
Token ID | 2151 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.003 |
Cyrus | -0.003 |
Elias | -0.003 |
Jace | -0.003 |
rabbi | -0.003 |
Token | � |
Token ID | 447 |
Feature activation | +7.95e-04 |
Pos logit contributions | |
Augustus | +0.031 |
Cyrus | +0.028 |
Elias | +0.027 |
Jace | +0.027 |
rabbi | +0.026 |
Neg logit contributions | |
Lum | -0.033 |
Beam | -0.031 |
Lau | -0.030 |
Layer | -0.029 |
Meng | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.009 |
Lau | +0.009 |
Layer | +0.008 |
Meng | +0.008 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.009 |
rabbi | -0.009 |
Token | one |
Token ID | 530 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
Augustus | +0.016 |
Cyrus | +0.014 |
Elias | +0.014 |
Jace | +0.014 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.015 |
Beam | -0.014 |
Lau | -0.013 |
Layer | -0.013 |
Meng | -0.012 |
Token | of |
Token ID | 286 |
Feature activation | +9.12e-03 |
Pos logit contributions | |
Lum | +0.019 |
Beam | +0.018 |
Lau | +0.017 |
Layer | +0.017 |
Meng | +0.016 |
Neg logit contributions | |
Augustus | -0.017 |
Cyrus | -0.015 |
Elias | -0.015 |
Jace | -0.015 |
rabbi | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
Augustus | +0.009 |
Cyrus | +0.008 |
Elias | +0.008 |
Jace | +0.008 |
rabbi | +0.008 |
Neg logit contributions | |
Lum | -0.008 |
Beam | -0.008 |
Lau | -0.007 |
Layer | -0.007 |
Meng | -0.007 |
Token | few |
Token ID | 1178 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.010 |
Beam | -0.009 |
Lau | -0.009 |
Layer | -0.009 |
Meng | -0.008 |
Token | fun |
Token ID | 1257 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
Lum | +0.039 |
Beam | +0.037 |
Lau | +0.035 |
Layer | +0.034 |
Meng | +0.033 |
Neg logit contributions | |
Augustus | -0.041 |
Cyrus | -0.037 |
Elias | -0.037 |
Jace | -0.036 |
rabbi | -0.036 |
Token | plays |
Token ID | 5341 |
Feature activation | +1.74e-03 |
Pos logit contributions | |
Augustus | +0.033 |
Cyrus | +0.030 |
Elias | +0.030 |
Jace | +0.029 |
rabbi | +0.029 |
Neg logit contributions | |
Lum | -0.030 |
Beam | -0.029 |
Lau | -0.027 |
Layer | -0.026 |
Meng | -0.026 |
Token | from |
Token ID | 422 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
Augustus | +0.002 |
Cyrus | +0.001 |
Elias | +0.001 |
Jace | +0.001 |
rabbi | +0.001 |
Neg logit contributions | |
Lum | -0.002 |
Beam | -0.002 |
Lau | -0.002 |
Layer | -0.002 |
Meng | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
Lum | +0.044 |
Beam | +0.042 |
Lau | +0.040 |
Layer | +0.039 |
Meng | +0.037 |
Neg logit contributions | |
Augustus | -0.047 |
Cyrus | -0.043 |
Elias | -0.042 |
Jace | -0.041 |
rabbi | -0.041 |
Token | 2016 |
Token ID | 1584 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
Lum | +0.012 |
Beam | +0.011 |
Lau | +0.011 |
Layer | +0.011 |
Meng | +0.010 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.011 |
Jace | -0.010 |
rabbi | -0.010 |
Token | Jets |
Token ID | 14728 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
Lum | +0.016 |
Beam | +0.015 |
Lau | +0.014 |
Layer | +0.014 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.014 |
Cyrus | -0.012 |
Elias | -0.012 |
Jace | -0.012 |
rabbi | -0.012 |
Token | Cup |
Token ID | 5454 |
Feature activation | -9.37e-02 |
Pos logit contributions | |
Lum | +0.024 |
Beam | +0.021 |
Lau | +0.017 |
Layer | +0.016 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.128 |
Cyrus | -0.121 |
Elias | -0.120 |
Jace | -0.118 |
rabbi | -0.117 |
Token | , |
Token ID | 11 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
Lum | +0.083 |
Beam | +0.079 |
Lau | +0.075 |
Layer | +0.073 |
Meng | +0.070 |
Neg logit contributions | |
Augustus | -0.095 |
Cyrus | -0.086 |
Elias | -0.085 |
Jace | -0.083 |
rabbi | -0.082 |
Token | then |
Token ID | 788 |
Feature activation | +7.08e-02 |
Pos logit contributions | |
Augustus | +0.047 |
Cyrus | +0.043 |
Elias | +0.042 |
Jace | +0.041 |
rabbi | +0.041 |
Neg logit contributions | |
Lum | -0.043 |
Beam | -0.041 |
Lau | -0.039 |
Layer | -0.038 |
Meng | -0.036 |
Token | certainly |
Token ID | 3729 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
Augustus | +0.068 |
Cyrus | +0.062 |
Elias | +0.061 |
Jace | +0.059 |
rabbi | +0.059 |
Neg logit contributions | |
Lum | -0.066 |
Beam | -0.064 |
Lau | -0.060 |
Layer | -0.059 |
Meng | -0.057 |
Token | ( |
Token ID | 357 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
Augustus | +0.033 |
Cyrus | +0.030 |
Elias | +0.030 |
Jace | +0.029 |
rabbi | +0.029 |
Neg logit contributions | |
Lum | -0.031 |
Beam | -0.030 |
Lau | -0.028 |
Layer | -0.028 |
Meng | -0.027 |
Token | the |
Token ID | 1169 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
Augustus | +0.031 |
Cyrus | +0.028 |
Elias | +0.027 |
Jace | +0.027 |
rabbi | +0.026 |
Neg logit contributions | |
Lum | -0.034 |
Beam | -0.033 |
Lau | -0.031 |
Layer | -0.030 |
Meng | -0.030 |
Token | trade |
Token ID | 3292 |
Feature activation | -3.29e-03 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.011 |
Beam | -0.010 |
Lau | -0.010 |
Layer | -0.009 |
Meng | -0.009 |
Token | ) |
Token ID | 8 |
Feature activation | +6.24e-02 |
Pos logit contributions | |
Lum | +0.003 |
Beam | +0.003 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.003 |
Cyrus | -0.003 |
Elias | -0.003 |
Jace | -0.003 |
rabbi | -0.003 |
Token | is |
Token ID | 318 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
Augustus | +0.062 |
Cyrus | +0.057 |
Elias | +0.056 |
Jace | +0.055 |
rabbi | +0.054 |
Neg logit contributions | |
Lum | -0.056 |
Beam | -0.054 |
Lau | -0.051 |
Layer | -0.050 |
Meng | -0.048 |
Token | difficult |
Token ID | 2408 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
Augustus | +0.042 |
Cyrus | +0.038 |
Elias | +0.038 |
Jace | +0.037 |
rabbi | +0.037 |
Neg logit contributions | |
Lum | -0.037 |
Beam | -0.036 |
Lau | -0.034 |
Layer | -0.033 |
Meng | -0.032 |
Token | ," |
Token ID | 553 |
Feature activation | +1.23e-03 |
Pos logit contributions | |
Augustus | +0.037 |
Cyrus | +0.033 |
Elias | +0.033 |
Jace | +0.032 |
rabbi | +0.032 |
Neg logit contributions | |
Lum | -0.035 |
Beam | -0.034 |
Lau | -0.032 |
Layer | -0.031 |
Meng | -0.030 |
Token | � |
Token ID | 447 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
Augustus | +0.015 |
Cyrus | +0.014 |
Elias | +0.014 |
Jace | +0.013 |
rabbi | +0.013 |
Neg logit contributions | |
Lum | -0.012 |
Beam | -0.012 |
Lau | -0.011 |
Layer | -0.011 |
Meng | -0.011 |
Token | � |
Token ID | 247 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
Augustus | +0.028 |
Cyrus | +0.025 |
Elias | +0.025 |
Jace | +0.024 |
rabbi | +0.024 |
Neg logit contributions | |
Lum | -0.030 |
Beam | -0.029 |
Lau | -0.027 |
Layer | -0.027 |
Meng | -0.026 |
Token | s |
Token ID | 82 |
Feature activation | +5.24e-03 |
Pos logit contributions | |
Augustus | +0.020 |
Cyrus | +0.018 |
Elias | +0.018 |
Jace | +0.017 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.016 |
Meng | -0.015 |
Token | such |
Token ID | 884 |
Feature activation | -5.32e-03 |
Pos logit contributions | |
Augustus | +0.005 |
Cyrus | +0.005 |
Elias | +0.005 |
Jace | +0.005 |
rabbi | +0.005 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.005 |
Lau | -0.004 |
Layer | -0.004 |
Meng | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -4.47e-03 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.004 |
Lau | +0.004 |
Layer | +0.004 |
Meng | +0.004 |
Neg logit contributions | |
Augustus | -0.005 |
Cyrus | -0.005 |
Elias | -0.005 |
Jace | -0.005 |
rabbi | -0.005 |
Token | positive |
Token ID | 3967 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
Lum | +0.004 |
Beam | +0.004 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.005 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.004 |
Token | person |
Token ID | 1048 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
Augustus | +0.026 |
Cyrus | +0.024 |
Elias | +0.023 |
Jace | +0.023 |
rabbi | +0.022 |
Neg logit contributions | |
Lum | -0.021 |
Beam | -0.020 |
Lau | -0.019 |
Layer | -0.018 |
Meng | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -9.72e-03 |
Pos logit contributions | |
Lum | +0.018 |
Beam | +0.017 |
Lau | +0.016 |
Layer | +0.016 |
Meng | +0.015 |
Neg logit contributions | |
Augustus | -0.018 |
Cyrus | -0.016 |
Elias | -0.016 |
Jace | -0.016 |
rabbi | -0.016 |
Token | He |
Token ID | 679 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.007 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | would |
Token ID | 561 |
Feature activation | -5.95e-03 |
Pos logit contributions | |
Lum | +0.012 |
Beam | +0.012 |
Lau | +0.011 |
Layer | +0.011 |
Meng | +0.010 |
Neg logit contributions | |
Augustus | -0.015 |
Cyrus | -0.014 |
Elias | -0.014 |
Jace | -0.013 |
rabbi | -0.013 |
Token | do |
Token ID | 466 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.005 |
Lau | +0.005 |
Layer | +0.005 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.006 |
Cyrus | -0.005 |
Elias | -0.005 |
Jace | -0.005 |
rabbi | -0.005 |
Token | Knight |
Token ID | 6700 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
Augustus | +0.023 |
Cyrus | +0.021 |
Elias | +0.021 |
Jace | +0.020 |
rabbi | +0.020 |
Neg logit contributions | |
Lum | -0.016 |
Beam | -0.016 |
Lau | -0.014 |
Layer | -0.014 |
Meng | -0.014 |
Token | : |
Token ID | 25 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
Lum | +0.011 |
Beam | +0.010 |
Lau | +0.010 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | an |
Token ID | 281 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.009 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.008 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.009 |
Token | original |
Token ID | 2656 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | story |
Token ID | 1621 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
Augustus | +0.013 |
Cyrus | +0.012 |
Elias | +0.011 |
Jace | +0.011 |
rabbi | +0.011 |
Neg logit contributions | |
Lum | -0.012 |
Beam | -0.012 |
Lau | -0.011 |
Layer | -0.011 |
Meng | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +8.63e-03 |
Pos logit contributions | |
Lum | +0.031 |
Beam | +0.029 |
Lau | +0.028 |
Layer | +0.027 |
Meng | +0.026 |
Neg logit contributions | |
Augustus | -0.032 |
Cyrus | -0.029 |
Elias | -0.029 |
Jace | -0.028 |
rabbi | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | +1.75e-04 |
Pos logit contributions | |
Augustus | +0.009 |
Cyrus | +0.008 |
Elias | +0.008 |
Jace | +0.008 |
rabbi | +0.008 |
Neg logit contributions | |
Lum | -0.008 |
Beam | -0.007 |
Lau | -0.007 |
Layer | -0.007 |
Meng | -0.007 |
Token | darker |
Token ID | 18646 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
Augustus | +0.000 |
Cyrus | +0.000 |
Elias | +0.000 |
Jace | +0.000 |
rabbi | +0.000 |
Neg logit contributions | |
Lum | -0.000 |
Beam | -0.000 |
Lau | -0.000 |
Layer | -0.000 |
Meng | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
Lum | +0.034 |
Beam | +0.032 |
Lau | +0.030 |
Layer | +0.029 |
Meng | +0.028 |
Neg logit contributions | |
Augustus | -0.045 |
Cyrus | -0.041 |
Elias | -0.041 |
Jace | -0.040 |
rabbi | -0.040 |
Token | more |
Token ID | 517 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
Augustus | +0.020 |
Cyrus | +0.018 |
Elias | +0.018 |
Jace | +0.017 |
rabbi | +0.017 |
Neg logit contributions | |
Lum | -0.015 |
Beam | -0.014 |
Lau | -0.013 |
Layer | -0.013 |
Meng | -0.013 |
Token | realistic |
Token ID | 12653 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
Lum | +0.009 |
Beam | +0.009 |
Lau | +0.008 |
Layer | +0.008 |
Meng | +0.008 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.009 |
rabbi | -0.009 |
Token | up |
Token ID | 510 |
Feature activation | -1.17e-03 |
Pos logit contributions | |
Lum | +0.018 |
Beam | +0.017 |
Lau | +0.016 |
Layer | +0.016 |
Meng | +0.015 |
Neg logit contributions | |
Augustus | -0.025 |
Cyrus | -0.023 |
Elias | -0.023 |
Jace | -0.023 |
rabbi | -0.022 |
Token | works |
Token ID | 2499 |
Feature activation | -4.98e-04 |
Pos logit contributions | |
Lum | +0.001 |
Beam | +0.001 |
Lau | +0.001 |
Layer | +0.001 |
Meng | +0.001 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.001 |
Elias | -0.001 |
Jace | -0.001 |
rabbi | -0.001 |
Token | really |
Token ID | 1107 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
Lum | +0.000 |
Beam | +0.000 |
Lau | +0.000 |
Layer | +0.000 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.001 |
Cyrus | -0.000 |
Elias | -0.000 |
Jace | -0.000 |
rabbi | -0.000 |
Token | really |
Token ID | 1107 |
Feature activation | +2.62e-03 |
Pos logit contributions | |
Lum | +0.008 |
Beam | +0.008 |
Lau | +0.007 |
Layer | +0.007 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.014 |
Cyrus | -0.013 |
Elias | -0.013 |
Jace | -0.012 |
rabbi | -0.012 |
Token | well |
Token ID | 880 |
Feature activation | +5.00e-03 |
Pos logit contributions | |
Augustus | +0.003 |
Cyrus | +0.003 |
Elias | +0.003 |
Jace | +0.003 |
rabbi | +0.003 |
Neg logit contributions | |
Lum | -0.002 |
Beam | -0.002 |
Lau | -0.002 |
Layer | -0.002 |
Meng | -0.001 |
Token | . |
Token ID | 764 |
Feature activation | -8.28e-03 |
Pos logit contributions | |
Augustus | +0.005 |
Cyrus | +0.004 |
Elias | +0.004 |
Jace | +0.004 |
rabbi | +0.004 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.004 |
Lau | -0.004 |
Layer | -0.004 |
Meng | -0.004 |
Token | Tape |
Token ID | 40849 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
Lum | +0.008 |
Beam | +0.007 |
Lau | +0.007 |
Layer | +0.007 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.008 |
Cyrus | -0.007 |
Elias | -0.007 |
Jace | -0.007 |
rabbi | -0.007 |
Token | is |
Token ID | 318 |
Feature activation | +9.98e-03 |
Pos logit contributions | |
Augustus | +0.018 |
Cyrus | +0.016 |
Elias | +0.016 |
Jace | +0.015 |
rabbi | +0.015 |
Neg logit contributions | |
Lum | -0.017 |
Beam | -0.016 |
Lau | -0.015 |
Layer | -0.015 |
Meng | -0.015 |
Token | great |
Token ID | 1049 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
Augustus | +0.010 |
Cyrus | +0.009 |
Elias | +0.009 |
Jace | +0.009 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.009 |
Beam | -0.008 |
Lau | -0.008 |
Layer | -0.008 |
Meng | -0.007 |
Token | because |
Token ID | 780 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.010 |
rabbi | +0.010 |
Neg logit contributions | |
Lum | -0.012 |
Beam | -0.011 |
Lau | -0.010 |
Layer | -0.010 |
Meng | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
Augustus | +0.044 |
Cyrus | +0.040 |
Elias | +0.039 |
Jace | +0.038 |
rabbi | +0.038 |
Neg logit contributions | |
Lum | -0.042 |
Beam | -0.040 |
Lau | -0.038 |
Layer | -0.037 |
Meng | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | -5.62e-03 |
Pos logit contributions | |
Augustus | +0.014 |
Cyrus | +0.013 |
Elias | +0.013 |
Jace | +0.013 |
rabbi | +0.013 |
Neg logit contributions | |
Lum | -0.012 |
Beam | -0.012 |
Lau | -0.011 |
Layer | -0.011 |
Meng | -0.011 |
Token | Hust |
Token ID | 42941 |
Feature activation | -6.53e-03 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.005 |
Lau | +0.004 |
Layer | +0.004 |
Meng | +0.004 |
Neg logit contributions | |
Augustus | -0.006 |
Cyrus | -0.005 |
Elias | -0.005 |
Jace | -0.005 |
rabbi | -0.005 |
Token | la |
Token ID | 5031 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.005 |
Lau | +0.004 |
Layer | +0.004 |
Meng | +0.004 |
Neg logit contributions | |
Augustus | -0.007 |
Cyrus | -0.007 |
Elias | -0.007 |
Jace | -0.007 |
rabbi | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
Lum | +0.038 |
Beam | +0.036 |
Lau | +0.034 |
Layer | +0.033 |
Meng | +0.031 |
Neg logit contributions | |
Augustus | -0.066 |
Cyrus | -0.061 |
Elias | -0.061 |
Jace | -0.059 |
rabbi | -0.059 |
Token | � |
Token ID | 447 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
Augustus | +0.015 |
Cyrus | +0.014 |
Elias | +0.014 |
Jace | +0.014 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.010 |
Beam | -0.010 |
Lau | -0.009 |
Layer | -0.009 |
Meng | -0.009 |
Token | � |
Token ID | 251 |
Feature activation | -6.85e-03 |
Pos logit contributions | |
Augustus | +0.014 |
Cyrus | +0.012 |
Elias | +0.012 |
Jace | +0.012 |
rabbi | +0.012 |
Neg logit contributions | |
Lum | -0.016 |
Beam | -0.015 |
Lau | -0.015 |
Layer | -0.014 |
Meng | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -8.12e-03 |
Pos logit contributions | |
Lum | +0.006 |
Beam | +0.006 |
Lau | +0.005 |
Layer | +0.005 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.007 |
Cyrus | -0.006 |
Elias | -0.006 |
Jace | -0.006 |
rabbi | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -9.92e-05 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.006 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.009 |
Cyrus | -0.008 |
Elias | -0.008 |
Jace | -0.008 |
rabbi | -0.008 |
Token | Newsletter |
Token ID | 33031 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
Lum | +0.000 |
Beam | +0.000 |
Lau | +0.000 |
Layer | +0.000 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.000 |
Cyrus | -0.000 |
Elias | -0.000 |
Jace | -0.000 |
rabbi | -0.000 |
Token | Sign |
Token ID | 5865 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
Lum | +0.033 |
Beam | +0.031 |
Lau | +0.029 |
Layer | +0.028 |
Meng | +0.027 |
Neg logit contributions | |
Augustus | -0.046 |
Cyrus | -0.042 |
Elias | -0.042 |
Jace | -0.041 |
rabbi | -0.041 |
Token | Up |
Token ID | 3205 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
Lum | +0.036 |
Beam | +0.034 |
Lau | +0.031 |
Layer | +0.030 |
Meng | +0.029 |
Neg logit contributions | |
Augustus | -0.066 |
Cyrus | -0.061 |
Elias | -0.060 |
Jace | -0.059 |
rabbi | -0.059 |
Token | C |
Token ID | 327 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
Augustus | +0.021 |
Cyrus | +0.019 |
Elias | +0.019 |
Jace | +0.018 |
rabbi | +0.018 |
Neg logit contributions | |
Lum | -0.019 |
Beam | -0.018 |
Lau | -0.017 |
Layer | -0.016 |
Meng | -0.016 |
Token | OUNT |
Token ID | 28270 |
Feature activation | +1.97e-03 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | ( |
Token ID | 357 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
Augustus | +0.002 |
Cyrus | +0.002 |
Elias | +0.002 |
Jace | +0.002 |
rabbi | +0.002 |
Neg logit contributions | |
Lum | -0.002 |
Beam | -0.002 |
Lau | -0.002 |
Layer | -0.001 |
Meng | -0.001 |
Token | * |
Token ID | 1635 |
Feature activation | -4.77e-02 |
Pos logit contributions | |
Augustus | +0.038 |
Cyrus | +0.034 |
Elias | +0.034 |
Jace | +0.033 |
rabbi | +0.033 |
Neg logit contributions | |
Lum | -0.033 |
Beam | -0.032 |
Lau | -0.030 |
Layer | -0.029 |
Meng | -0.028 |
Token | ) |
Token ID | 1267 |
Feature activation | +4.44e-02 |
Pos logit contributions | |
Lum | +0.045 |
Beam | +0.043 |
Lau | +0.041 |
Layer | +0.040 |
Meng | +0.039 |
Neg logit contributions | |
Augustus | -0.045 |
Cyrus | -0.041 |
Elias | -0.040 |
Jace | -0.039 |
rabbi | -0.039 |
Token | AS |
Token ID | 7054 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
Augustus | +0.038 |
Cyrus | +0.034 |
Elias | +0.033 |
Jace | +0.033 |
rabbi | +0.032 |
Neg logit contributions | |
Lum | -0.046 |
Beam | -0.044 |
Lau | -0.042 |
Layer | -0.042 |
Meng | -0.040 |
Token | count |
Token ID | 954 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
Lum | +0.028 |
Beam | +0.027 |
Lau | +0.025 |
Layer | +0.025 |
Meng | +0.024 |
Neg logit contributions | |
Augustus | -0.031 |
Cyrus | -0.028 |
Elias | -0.028 |
Jace | -0.027 |
rabbi | -0.027 |
Token | FROM |
Token ID | 16034 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
Lum | +0.028 |
Beam | +0.027 |
Lau | +0.026 |
Layer | +0.025 |
Meng | +0.025 |
Neg logit contributions | |
Augustus | -0.026 |
Cyrus | -0.023 |
Elias | -0.023 |
Jace | -0.022 |
rabbi | -0.022 |
Token | ( |
Token ID | 357 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
Lum | +0.035 |
Beam | +0.033 |
Lau | +0.031 |
Layer | +0.030 |
Meng | +0.029 |
Neg logit contributions | |
Augustus | -0.050 |
Cyrus | -0.046 |
Elias | -0.045 |
Jace | -0.044 |
rabbi | -0.044 |
Token | SELECT |
Token ID | 33493 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
Augustus | +0.036 |
Cyrus | +0.033 |
Elias | +0.032 |
Jace | +0.032 |
rabbi | +0.031 |
Neg logit contributions | |
Lum | -0.031 |
Beam | -0.030 |
Lau | -0.028 |
Layer | -0.028 |
Meng | -0.027 |
Token | SPL |
Token ID | 46341 |
Feature activation | -1.68e-03 |
Pos logit contributions | |
Augustus | +0.022 |
Cyrus | +0.020 |
Elias | +0.020 |
Jace | +0.019 |
rabbi | +0.019 |
Neg logit contributions | |
Lum | -0.021 |
Beam | -0.020 |
Lau | -0.019 |
Layer | -0.018 |
Meng | -0.018 |
Token | 2002 |
Token ID | 6244 |
Feature activation | -1.43e-04 |
Pos logit contributions | |
Lum | +0.000 |
Beam | +0.000 |
Lau | +0.000 |
Layer | +0.000 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.000 |
Cyrus | -0.000 |
Elias | -0.000 |
Jace | -0.000 |
rabbi | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
Lum | +0.000 |
Beam | +0.000 |
Lau | +0.000 |
Layer | +0.000 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.000 |
Cyrus | -0.000 |
Elias | -0.000 |
Jace | -0.000 |
rabbi | -0.000 |
Token | scholar |
Token ID | 15606 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
Augustus | +0.031 |
Cyrus | +0.028 |
Elias | +0.027 |
Jace | +0.027 |
rabbi | +0.026 |
Neg logit contributions | |
Lum | -0.029 |
Beam | -0.028 |
Lau | -0.026 |
Layer | -0.026 |
Meng | -0.025 |
Token | Mole |
Token ID | 25726 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
Lum | +0.032 |
Beam | +0.030 |
Lau | +0.029 |
Layer | +0.028 |
Meng | +0.027 |
Neg logit contributions | |
Augustus | -0.029 |
Cyrus | -0.026 |
Elias | -0.026 |
Jace | -0.025 |
rabbi | -0.025 |
Token | fi |
Token ID | 12463 |
Feature activation | +4.03e-04 |
Pos logit contributions | |
Lum | +0.046 |
Beam | +0.044 |
Lau | +0.042 |
Layer | +0.041 |
Meng | +0.039 |
Neg logit contributions | |
Augustus | -0.047 |
Cyrus | -0.043 |
Elias | -0.042 |
Jace | -0.041 |
rabbi | -0.041 |
Token | K |
Token ID | 509 |
Feature activation | -4.96e-03 |
Pos logit contributions | |
Augustus | +0.000 |
Cyrus | +0.000 |
Elias | +0.000 |
Jace | +0.000 |
rabbi | +0.000 |
Neg logit contributions | |
Lum | -0.000 |
Beam | -0.000 |
Lau | -0.000 |
Layer | -0.000 |
Meng | -0.000 |
Token | ete |
Token ID | 14471 |
Feature activation | -7.48e-02 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.005 |
Lau | +0.004 |
Layer | +0.004 |
Meng | +0.004 |
Neg logit contributions | |
Augustus | -0.005 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.004 |
Token | As |
Token ID | 1081 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
Lum | +0.072 |
Beam | +0.069 |
Lau | +0.066 |
Layer | +0.064 |
Meng | +0.062 |
Neg logit contributions | |
Augustus | -0.070 |
Cyrus | -0.063 |
Elias | -0.062 |
Jace | -0.061 |
rabbi | -0.060 |
Token | ante |
Token ID | 12427 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
Lum | +0.016 |
Beam | +0.015 |
Lau | +0.014 |
Layer | +0.014 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.015 |
Cyrus | -0.013 |
Elias | -0.013 |
Jace | -0.013 |
rabbi | -0.013 |
Token | listed |
Token ID | 5610 |
Feature activation | +3.48e-03 |
Pos logit contributions | |
Augustus | +0.033 |
Cyrus | +0.030 |
Elias | +0.030 |
Jace | +0.029 |
rabbi | +0.029 |
Neg logit contributions | |
Lum | -0.035 |
Beam | -0.034 |
Lau | -0.032 |
Layer | -0.031 |
Meng | -0.030 |
Token | Mary |
Token ID | 5335 |
Feature activation | -6.08e-02 |
Pos logit contributions | |
Augustus | +0.003 |
Cyrus | +0.003 |
Elias | +0.003 |
Jace | +0.003 |
rabbi | +0.003 |
Neg logit contributions | |
Lum | -0.003 |
Beam | -0.003 |
Lau | -0.003 |
Layer | -0.003 |
Meng | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -6.19e-03 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.007 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.009 |
Cyrus | -0.008 |
Elias | -0.008 |
Jace | -0.008 |
rabbi | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -2.00e-03 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.005 |
Lau | +0.004 |
Layer | +0.004 |
Meng | +0.004 |
Neg logit contributions | |
Augustus | -0.007 |
Cyrus | -0.006 |
Elias | -0.006 |
Jace | -0.006 |
rabbi | -0.006 |
Token | Make |
Token ID | 12050 |
Feature activation | -1.90e-03 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | it |
Token ID | 340 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | about |
Token ID | 546 |
Feature activation | +5.88e-03 |
Pos logit contributions | |
Lum | +0.009 |
Beam | +0.008 |
Lau | +0.008 |
Layer | +0.008 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.009 |
Token | THEM |
Token ID | 44788 |
Feature activation | -9.20e-02 |
Pos logit contributions | |
Augustus | +0.006 |
Cyrus | +0.005 |
Elias | +0.005 |
Jace | +0.005 |
rabbi | +0.005 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.005 |
Lau | -0.005 |
Layer | -0.005 |
Meng | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +1.72e-03 |
Pos logit contributions | |
Lum | +0.082 |
Beam | +0.078 |
Lau | +0.073 |
Layer | +0.072 |
Meng | +0.069 |
Neg logit contributions | |
Augustus | -0.093 |
Cyrus | -0.085 |
Elias | -0.084 |
Jace | -0.082 |
rabbi | -0.081 |
Token | Whenever |
Token ID | 21326 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
Augustus | +0.002 |
Cyrus | +0.002 |
Elias | +0.002 |
Jace | +0.002 |
rabbi | +0.002 |
Neg logit contributions | |
Lum | -0.002 |
Beam | -0.001 |
Lau | -0.001 |
Layer | -0.001 |
Meng | -0.001 |
Token | possible |
Token ID | 1744 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
Augustus | +0.026 |
Cyrus | +0.023 |
Elias | +0.023 |
Jace | +0.023 |
rabbi | +0.022 |
Neg logit contributions | |
Lum | -0.027 |
Beam | -0.025 |
Lau | -0.024 |
Layer | -0.024 |
Meng | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
Augustus | +0.023 |
Cyrus | +0.021 |
Elias | +0.021 |
Jace | +0.020 |
rabbi | +0.020 |
Neg logit contributions | |
Lum | -0.023 |
Beam | -0.022 |
Lau | -0.021 |
Layer | -0.020 |
Meng | -0.020 |
Token | make |
Token ID | 787 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | data |
Token ID | 1366 |
Feature activation | -1.98e-03 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.017 |
Jace | +0.016 |
rabbi | +0.016 |
Neg logit contributions | |
Lum | -0.017 |
Beam | -0.016 |
Lau | -0.015 |
Layer | -0.015 |
Meng | -0.015 |
Token | structures |
Token ID | 8573 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
Lum | +0.002 |
Beam | +0.002 |
Lau | +0.002 |
Layer | +0.002 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.002 |
Cyrus | -0.002 |
Elias | -0.002 |
Jace | -0.002 |
rabbi | -0.002 |
Token | / |
Token ID | 14 |
Feature activation | +9.46e-03 |
Pos logit contributions | |
Augustus | +0.011 |
Cyrus | +0.010 |
Elias | +0.010 |
Jace | +0.009 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.010 |
Beam | -0.009 |
Lau | -0.009 |
Layer | -0.009 |
Meng | -0.008 |
Token | al |
Token ID | 282 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
Augustus | +0.009 |
Cyrus | +0.008 |
Elias | +0.008 |
Jace | +0.008 |
rabbi | +0.008 |
Neg logit contributions | |
Lum | -0.009 |
Beam | -0.009 |
Lau | -0.008 |
Layer | -0.008 |
Meng | -0.008 |
Token | gorith |
Token ID | 7727 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
Augustus | +0.015 |
Cyrus | +0.013 |
Elias | +0.012 |
Jace | +0.012 |
rabbi | +0.012 |
Neg logit contributions | |
Lum | -0.033 |
Beam | -0.032 |
Lau | -0.031 |
Layer | -0.031 |
Meng | -0.030 |
Token | ms |
Token ID | 907 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
Augustus | +0.043 |
Cyrus | +0.039 |
Elias | +0.039 |
Jace | +0.038 |
rabbi | +0.038 |
Neg logit contributions | |
Lum | -0.024 |
Beam | -0.023 |
Lau | -0.021 |
Layer | -0.021 |
Meng | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
Lum | +0.163 |
Beam | +0.156 |
Lau | +0.147 |
Layer | +0.145 |
Meng | +0.140 |
Neg logit contributions | |
Augustus | -0.173 |
Cyrus | -0.157 |
Elias | -0.155 |
Jace | -0.151 |
rabbi | -0.150 |
Token | more |
Token ID | 517 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
Lum | +0.017 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.015 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.020 |
Cyrus | -0.019 |
Elias | -0.018 |
Jace | -0.018 |
rabbi | -0.018 |
Token | about |
Token ID | 546 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
Lum | +0.028 |
Beam | +0.027 |
Lau | +0.026 |
Layer | +0.025 |
Meng | +0.024 |
Neg logit contributions | |
Augustus | -0.027 |
Cyrus | -0.024 |
Elias | -0.024 |
Jace | -0.023 |
rabbi | -0.023 |
Token | web |
Token ID | 3992 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
Lum | +0.012 |
Beam | +0.012 |
Lau | +0.011 |
Layer | +0.011 |
Meng | +0.010 |
Neg logit contributions | |
Augustus | -0.014 |
Cyrus | -0.012 |
Elias | -0.012 |
Jace | -0.012 |
rabbi | -0.012 |
Token | development |
Token ID | 2478 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
Lum | +0.034 |
Beam | +0.032 |
Lau | +0.031 |
Layer | +0.030 |
Meng | +0.029 |
Neg logit contributions | |
Augustus | -0.033 |
Cyrus | -0.030 |
Elias | -0.030 |
Jace | -0.029 |
rabbi | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
Augustus | +0.028 |
Cyrus | +0.026 |
Elias | +0.025 |
Jace | +0.025 |
rabbi | +0.025 |
Neg logit contributions | |
Lum | -0.027 |
Beam | -0.026 |
Lau | -0.024 |
Layer | -0.024 |
Meng | -0.023 |
Token | past |
Token ID | 1613 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
Augustus | +0.037 |
Cyrus | +0.034 |
Elias | +0.033 |
Jace | +0.032 |
rabbi | +0.032 |
Neg logit contributions | |
Lum | -0.037 |
Beam | -0.036 |
Lau | -0.034 |
Layer | -0.033 |
Meng | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
Augustus | +0.019 |
Cyrus | +0.017 |
Elias | +0.017 |
Jace | +0.016 |
rabbi | +0.016 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.016 |
Meng | -0.015 |
Token | which |
Token ID | 543 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
Augustus | +0.027 |
Cyrus | +0.025 |
Elias | +0.024 |
Jace | +0.024 |
rabbi | +0.024 |
Neg logit contributions | |
Lum | -0.025 |
Beam | -0.024 |
Lau | -0.023 |
Layer | -0.022 |
Meng | -0.022 |
Token | Progress |
Token ID | 18387 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
Augustus | +0.020 |
Cyrus | +0.019 |
Elias | +0.018 |
Jace | +0.018 |
rabbi | +0.018 |
Neg logit contributions | |
Lum | -0.018 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.016 |
Meng | -0.015 |
Token | ives |
Token ID | 1083 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
Augustus | +0.069 |
Cyrus | +0.064 |
Elias | +0.063 |
Jace | +0.062 |
rabbi | +0.062 |
Neg logit contributions | |
Lum | -0.035 |
Beam | -0.033 |
Lau | -0.030 |
Layer | -0.029 |
Meng | -0.028 |
Token | have |
Token ID | 423 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
Lum | +0.155 |
Beam | +0.148 |
Lau | +0.140 |
Layer | +0.137 |
Meng | +0.132 |
Neg logit contributions | |
Augustus | -0.174 |
Cyrus | -0.159 |
Elias | -0.156 |
Jace | -0.153 |
rabbi | -0.151 |
Token | spent |
Token ID | 3377 |
Feature activation | -4.31e-04 |
Pos logit contributions | |
Augustus | +0.034 |
Cyrus | +0.031 |
Elias | +0.031 |
Jace | +0.030 |
rabbi | +0.030 |
Neg logit contributions | |
Lum | -0.028 |
Beam | -0.027 |
Lau | -0.025 |
Layer | -0.025 |
Meng | -0.024 |
Token | time |
Token ID | 640 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
Lum | +0.000 |
Beam | +0.000 |
Lau | +0.000 |
Layer | +0.000 |
Meng | +0.000 |
Neg logit contributions | |
Augustus | -0.000 |
Cyrus | -0.000 |
Elias | -0.000 |
Jace | -0.000 |
rabbi | -0.000 |
Token | fighting |
Token ID | 4330 |
Feature activation | -8.29e-03 |
Pos logit contributions | |
Augustus | +0.015 |
Cyrus | +0.014 |
Elias | +0.014 |
Jace | +0.013 |
rabbi | +0.013 |
Neg logit contributions | |
Lum | -0.015 |
Beam | -0.014 |
Lau | -0.013 |
Layer | -0.013 |
Meng | -0.013 |
Token | back |
Token ID | 736 |
Feature activation | +6.04e-03 |
Pos logit contributions | |
Lum | +0.008 |
Beam | +0.007 |
Lau | +0.007 |
Layer | +0.007 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.008 |
Cyrus | -0.007 |
Elias | -0.007 |
Jace | -0.007 |
rabbi | -0.007 |
Token | ost |
Token ID | 455 |
Feature activation | -4.37e-02 |
Pos logit contributions | |
Augustus | +0.043 |
Cyrus | +0.039 |
Elias | +0.039 |
Jace | +0.038 |
rabbi | +0.038 |
Neg logit contributions | |
Lum | -0.031 |
Beam | -0.029 |
Lau | -0.027 |
Layer | -0.026 |
Meng | -0.025 |
Token | ino |
Token ID | 2879 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
Lum | +0.044 |
Beam | +0.042 |
Lau | +0.040 |
Layer | +0.039 |
Meng | +0.038 |
Neg logit contributions | |
Augustus | -0.039 |
Cyrus | -0.035 |
Elias | -0.034 |
Jace | -0.034 |
rabbi | -0.033 |
Token | says |
Token ID | 1139 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
Augustus | +0.016 |
Cyrus | +0.014 |
Elias | +0.014 |
Jace | +0.014 |
rabbi | +0.014 |
Neg logit contributions | |
Lum | -0.015 |
Beam | -0.015 |
Lau | -0.014 |
Layer | -0.014 |
Meng | -0.013 |
Token | 14 |
Token ID | 1478 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
Augustus | +0.010 |
Cyrus | +0.009 |
Elias | +0.009 |
Jace | +0.009 |
rabbi | +0.009 |
Neg logit contributions | |
Lum | -0.009 |
Beam | -0.009 |
Lau | -0.008 |
Layer | -0.008 |
Meng | -0.008 |
Token | per |
Token ID | 583 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
Lum | +0.026 |
Beam | +0.025 |
Lau | +0.023 |
Layer | +0.022 |
Meng | +0.021 |
Neg logit contributions | |
Augustus | -0.039 |
Cyrus | -0.036 |
Elias | -0.035 |
Jace | -0.035 |
rabbi | -0.034 |
Token | cent |
Token ID | 1247 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
Lum | +0.004 |
Beam | +0.004 |
Lau | +0.003 |
Layer | +0.003 |
Meng | +0.002 |
Neg logit contributions | |
Augustus | -0.024 |
Cyrus | -0.022 |
Elias | -0.022 |
Jace | -0.022 |
rabbi | -0.022 |
Token | of |
Token ID | 286 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
Lum | +0.238 |
Beam | +0.228 |
Lau | +0.216 |
Layer | +0.212 |
Meng | +0.206 |
Neg logit contributions | |
Augustus | -0.220 |
Cyrus | -0.198 |
Elias | -0.195 |
Jace | -0.190 |
rabbi | -0.188 |
Token | families |
Token ID | 4172 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
Augustus | +0.036 |
Cyrus | +0.032 |
Elias | +0.032 |
Jace | +0.031 |
rabbi | +0.031 |
Neg logit contributions | |
Lum | -0.038 |
Beam | -0.037 |
Lau | -0.035 |
Layer | -0.034 |
Meng | -0.033 |
Token | struggling |
Token ID | 9648 |
Feature activation | -5.38e-03 |
Pos logit contributions | |
Augustus | +0.014 |
Cyrus | +0.013 |
Elias | +0.013 |
Jace | +0.012 |
rabbi | +0.012 |
Neg logit contributions | |
Lum | -0.013 |
Beam | -0.013 |
Lau | -0.012 |
Layer | -0.012 |
Meng | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.005 |
Lau | +0.005 |
Layer | +0.005 |
Meng | +0.005 |
Neg logit contributions | |
Augustus | -0.005 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.004 |
Token | afford |
Token ID | 5368 |
Feature activation | +3.02e-03 |
Pos logit contributions | |
Augustus | +0.020 |
Cyrus | +0.018 |
Elias | +0.018 |
Jace | +0.018 |
rabbi | +0.018 |
Neg logit contributions | |
Lum | -0.019 |
Beam | -0.019 |
Lau | -0.017 |
Layer | -0.017 |
Meng | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
Augustus | +0.020 |
Cyrus | +0.018 |
Elias | +0.018 |
Jace | +0.018 |
rabbi | +0.018 |
Neg logit contributions | |
Lum | -0.017 |
Beam | -0.017 |
Lau | -0.016 |
Layer | -0.015 |
Meng | -0.015 |
Token | hes |
Token ID | 956 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
Lum | +0.017 |
Beam | +0.016 |
Lau | +0.015 |
Layer | +0.014 |
Meng | +0.014 |
Neg logit contributions | |
Augustus | -0.020 |
Cyrus | -0.019 |
Elias | -0.018 |
Jace | -0.018 |
rabbi | -0.018 |
Token | in |
Token ID | 287 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
Lum | +0.027 |
Beam | +0.026 |
Lau | +0.025 |
Layer | +0.024 |
Meng | +0.023 |
Neg logit contributions | |
Augustus | -0.027 |
Cyrus | -0.024 |
Elias | -0.024 |
Jace | -0.023 |
rabbi | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
Lum | +0.012 |
Beam | +0.011 |
Lau | +0.011 |
Layer | +0.011 |
Meng | +0.010 |
Neg logit contributions | |
Augustus | -0.011 |
Cyrus | -0.010 |
Elias | -0.010 |
Jace | -0.009 |
rabbi | -0.009 |
Token | Moj |
Token ID | 40120 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
Lum | +0.025 |
Beam | +0.024 |
Lau | +0.023 |
Layer | +0.022 |
Meng | +0.022 |
Neg logit contributions | |
Augustus | -0.024 |
Cyrus | -0.021 |
Elias | -0.021 |
Jace | -0.021 |
rabbi | -0.020 |
Token | ave |
Token ID | 1015 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
Lum | +0.011 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.008 |
Neg logit contributions | |
Augustus | -0.029 |
Cyrus | -0.027 |
Elias | -0.027 |
Jace | -0.026 |
rabbi | -0.026 |
Token | Desert |
Token ID | 18692 |
Feature activation | +7.52e-03 |
Pos logit contributions | |
Lum | +0.176 |
Beam | +0.169 |
Lau | +0.161 |
Layer | +0.158 |
Meng | +0.154 |
Neg logit contributions | |
Augustus | -0.146 |
Cyrus | -0.131 |
Elias | -0.128 |
Jace | -0.125 |
rabbi | -0.124 |
Token | to |
Token ID | 284 |
Feature activation | -9.53e-03 |
Pos logit contributions | |
Augustus | +0.007 |
Cyrus | +0.006 |
Elias | +0.006 |
Jace | +0.006 |
rabbi | +0.006 |
Neg logit contributions | |
Lum | -0.007 |
Beam | -0.007 |
Lau | -0.007 |
Layer | -0.006 |
Meng | -0.006 |
Token | Mo |
Token ID | 4270 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
Lum | +0.008 |
Beam | +0.008 |
Lau | +0.007 |
Layer | +0.007 |
Meng | +0.007 |
Neg logit contributions | |
Augustus | -0.010 |
Cyrus | -0.009 |
Elias | -0.009 |
Jace | -0.009 |
rabbi | -0.009 |
Token | ab |
Token ID | 397 |
Feature activation | +8.78e-03 |
Pos logit contributions | |
Lum | +0.010 |
Beam | +0.010 |
Lau | +0.009 |
Layer | +0.009 |
Meng | +0.009 |
Neg logit contributions | |
Augustus | -0.012 |
Cyrus | -0.011 |
Elias | -0.010 |
Jace | -0.010 |
rabbi | -0.010 |
Token | 's |
Token ID | 338 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
Augustus | +0.009 |
Cyrus | +0.008 |
Elias | +0.008 |
Jace | +0.008 |
rabbi | +0.007 |
Neg logit contributions | |
Lum | -0.008 |
Beam | -0.008 |
Lau | -0.007 |
Layer | -0.007 |
Meng | -0.007 |
Token | that |
Token ID | 326 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
Augustus | +0.044 |
Cyrus | +0.040 |
Elias | +0.040 |
Jace | +0.039 |
rabbi | +0.038 |
Neg logit contributions | |
Lum | -0.040 |
Beam | -0.039 |
Lau | -0.036 |
Layer | -0.036 |
Meng | -0.034 |
Token | participate |
Token ID | 8277 |
Feature activation | +5.03e-03 |
Pos logit contributions | |
Augustus | +0.031 |
Cyrus | +0.028 |
Elias | +0.028 |
Jace | +0.027 |
rabbi | +0.027 |
Neg logit contributions | |
Lum | -0.029 |
Beam | -0.028 |
Lau | -0.026 |
Layer | -0.025 |
Meng | -0.025 |
Token | in |
Token ID | 287 |
Feature activation | -4.37e-03 |
Pos logit contributions | |
Augustus | +0.005 |
Cyrus | +0.004 |
Elias | +0.004 |
Jace | +0.004 |
rabbi | +0.004 |
Neg logit contributions | |
Lum | -0.005 |
Beam | -0.005 |
Lau | -0.005 |
Layer | -0.004 |
Meng | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -6.88e-03 |
Pos logit contributions | |
Lum | +0.004 |
Beam | +0.004 |
Lau | +0.004 |
Layer | +0.003 |
Meng | +0.003 |
Neg logit contributions | |
Augustus | -0.004 |
Cyrus | -0.004 |
Elias | -0.004 |
Jace | -0.004 |
rabbi | -0.004 |
Token | Medicare |
Token ID | 13771 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
Lum | +0.005 |
Beam | +0.004 |
Lau | +0.004 |
Layer | +0.004 |
Meng | +0.004 |
Neg logit contributions | |
Augustus | -0.008 |
Cyrus | -0.008 |
Elias | -0.008 |
Jace | -0.007 |
rabbi | -0.007 |
Token | Advantage |
Token ID | 45318 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
Lum | +0.013 |
Beam | +0.012 |
Lau | +0.011 |
Layer | +0.010 |
Meng | +0.010 |
Neg logit contributions | |
Augustus | -0.037 |
Cyrus | -0.034 |
Elias | -0.034 |
Jace | -0.033 |
rabbi | -0.033 |
Token | program |
Token ID | 1430 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
Lum | +0.158 |
Beam | +0.151 |
Lau | +0.142 |
Layer | +0.139 |
Meng | +0.135 |
Neg logit contributions | |
Augustus | -0.181 |
Cyrus | -0.164 |
Elias | -0.162 |
Jace | -0.158 |
rabbi | -0.157 |
Token | devote |
Token ID | 24103 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
Lum | +0.018 |
Beam | +0.017 |
Lau | +0.016 |
Layer | +0.016 |
Meng | +0.015 |
Neg logit contributions | |
Augustus | -0.020 |
Cyrus | -0.018 |
Elias | -0.018 |
Jace | -0.017 |
rabbi | -0.017 |
Token | big |
Token ID | 1263 |
Feature activation | -7.96e-03 |
Pos logit contributions | |
Lum | +0.020 |
Beam | +0.019 |
Lau | +0.018 |
Layer | +0.018 |
Meng | +0.017 |
Neg logit contributions | |
Augustus | -0.023 |
Cyrus | -0.021 |
Elias | -0.021 |
Jace | -0.021 |
rabbi | -0.020 |
Token | chunks |
Token ID | 22716 |
Feature activation | -6.04e-02 |
Pos logit contributions | |
Lum | +0.007 |
Beam | +0.007 |
Lau | +0.006 |
Layer | +0.006 |
Meng | +0.006 |
Neg logit contributions | |
Augustus | -0.008 |
Cyrus | -0.008 |
Elias | -0.007 |
Jace | -0.007 |
rabbi | -0.007 |
Token | of |
Token ID | 286 |
Feature activation | -4.57e-03 |
Pos logit contributions | |
Lum | +0.061 |
Beam | +0.058 |
Lau | +0.055 |
Layer | +0.054 |
Meng | +0.053 |
Neg logit contributions | |
Augustus | -0.054 |
Cyrus | -0.049 |
Elias | -0.048 |
Jace | -0.047 |
rabbi | -0.046 |