Token | Lily |
Token ID | 20037 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
Flor | +0.017 |
grate | +0.016 |
violet | +0.015 |
cheer | +0.015 |
| +0.015 |
Neg logit contributions | |
dd | -0.030 |
ime | -0.027 |
happened | -0.027 |
ug | -0.026 |
endings | -0.026 |
Token | accidentally |
Token ID | 14716 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
Flor | +0.043 |
grate | +0.041 |
| +0.040 |
violet | +0.038 |
cheer | +0.038 |
Neg logit contributions | |
dd | -0.085 |
ime | -0.078 |
ug | -0.076 |
endings | -0.075 |
happened | -0.072 |
Token | knocked |
Token ID | 13642 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
dd | +0.075 |
ime | +0.070 |
ug | +0.067 |
endings | +0.066 |
aw | +0.060 |
Neg logit contributions | |
Flor | -0.069 |
grate | -0.065 |
cheer | -0.062 |
| -0.062 |
violet | -0.061 |
Token | over |
Token ID | 625 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Flor | +0.020 |
grate | +0.018 |
| +0.018 |
cheer | +0.017 |
violet | +0.016 |
Neg logit contributions | |
dd | -0.057 |
ime | -0.053 |
endings | -0.052 |
ug | -0.051 |
happened | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
Flor | +0.040 |
grate | +0.037 |
| +0.035 |
cheer | +0.035 |
violet | +0.034 |
Neg logit contributions | |
dd | -0.069 |
ime | -0.064 |
ug | -0.061 |
happened | -0.060 |
endings | -0.060 |
Token | te |
Token ID | 573 |
Feature activation | +5.01e+00 |
Pos logit contributions | |
Flor | +0.047 |
grate | +0.044 |
cheer | +0.042 |
| +0.041 |
violet | +0.041 |
Neg logit contributions | |
dd | -0.067 |
ime | -0.060 |
ug | -0.058 |
happened | -0.057 |
endings | -0.057 |
Token | ap |
Token ID | 499 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
| +0.172 |
refreshed | +0.160 |
closer | +0.157 |
Flor | +0.157 |
forget | +0.152 |
Neg logit contributions | |
ime | -0.174 |
ug | -0.171 |
dd | -0.165 |
aw | -0.161 |
endings | -0.161 |
Token | ot |
Token ID | 313 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
Flor | +0.119 |
grate | +0.112 |
violet | +0.107 |
| +0.106 |
cheer | +0.105 |
Neg logit contributions | |
dd | -0.123 |
ime | -0.114 |
ug | -0.106 |
happened | -0.106 |
endings | -0.104 |
Token | and |
Token ID | 290 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
dd | +0.041 |
ime | +0.038 |
endings | +0.036 |
ug | +0.036 |
happened | +0.036 |
Neg logit contributions | |
Flor | -0.027 |
grate | -0.026 |
cheer | -0.024 |
| -0.024 |
violet | -0.024 |
Token | spilled |
Token ID | 31896 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
Flor | +0.002 |
grate | +0.002 |
violet | +0.002 |
| +0.002 |
cheer | +0.002 |
Neg logit contributions | |
dd | -0.003 |
ime | -0.003 |
happened | -0.003 |
ug | -0.003 |
endings | -0.003 |
Token | tea |
Token ID | 8887 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
dd | +0.032 |
ime | +0.030 |
ug | +0.028 |
happened | +0.028 |
endings | +0.028 |
Neg logit contributions | |
Flor | -0.020 |
grate | -0.019 |
violet | -0.018 |
| -0.018 |
cheer | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
Flor | +0.016 |
| +0.014 |
grate | +0.013 |
Av | +0.013 |
violet | +0.013 |
Neg logit contributions | |
dd | -0.039 |
ime | -0.037 |
ug | -0.036 |
endings | -0.035 |
ruck | -0.033 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Flor | +0.010 |
grate | +0.009 |
| +0.009 |
cheer | +0.009 |
violet | +0.009 |
Neg logit contributions | |
dd | -0.021 |
ime | -0.020 |
ug | -0.019 |
endings | -0.019 |
happened | -0.018 |
Token | accidentally |
Token ID | 14716 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
Flor | +0.051 |
grate | +0.048 |
| +0.045 |
violet | +0.044 |
cheer | +0.044 |
Neg logit contributions | |
dd | -0.090 |
ime | -0.083 |
ug | -0.080 |
endings | -0.079 |
happened | -0.078 |
Token | dropped |
Token ID | 5710 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
dd | +0.067 |
ime | +0.061 |
ug | +0.058 |
endings | +0.058 |
happened | +0.054 |
Neg logit contributions | |
Flor | -0.068 |
grate | -0.065 |
| -0.062 |
violet | -0.062 |
cheer | -0.062 |
Token | the |
Token ID | 262 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Flor | +0.002 |
grate | +0.002 |
| +0.002 |
cheer | +0.002 |
violet | +0.002 |
Neg logit contributions | |
dd | -0.004 |
ime | -0.004 |
ug | -0.004 |
endings | -0.004 |
happened | -0.004 |
Token | te |
Token ID | 573 |
Feature activation | +4.59e+00 |
Pos logit contributions | |
grate | +0.028 |
Flor | +0.025 |
| +0.024 |
Amen | +0.024 |
Together | +0.024 |
Neg logit contributions | |
happened | -0.043 |
dd | -0.043 |
ime | -0.040 |
emergency | -0.040 |
electricity | -0.039 |
Token | ap |
Token ID | 499 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
| +0.170 |
Flor | +0.160 |
refreshed | +0.155 |
applaud | +0.151 |
closer | +0.151 |
Neg logit contributions | |
dd | -0.148 |
ime | -0.146 |
ug | -0.144 |
endings | -0.135 |
aw | -0.131 |
Token | ot |
Token ID | 313 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
Flor | +0.111 |
grate | +0.107 |
cheer | +0.102 |
violet | +0.102 |
| +0.101 |
Neg logit contributions | |
dd | -0.121 |
ime | -0.108 |
ug | -0.102 |
endings | -0.102 |
happened | -0.101 |
Token | and |
Token ID | 290 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
dd | +0.035 |
ime | +0.033 |
endings | +0.032 |
ug | +0.031 |
happened | +0.031 |
Neg logit contributions | |
Flor | -0.027 |
grate | -0.027 |
cheer | -0.026 |
violet | -0.025 |
| -0.025 |
Token | it |
Token ID | 340 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Flor | +0.005 |
grate | +0.005 |
violet | +0.005 |
cheer | +0.004 |
| +0.004 |
Neg logit contributions | |
dd | -0.006 |
ime | -0.006 |
happened | -0.005 |
ug | -0.005 |
endings | -0.005 |
Token | broke |
Token ID | 6265 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
dd | +0.037 |
ime | +0.034 |
ug | +0.032 |
endings | +0.031 |
happened | +0.031 |
Neg logit contributions | |
Flor | -0.039 |
grate | -0.037 |
| -0.035 |
cheer | -0.035 |
violet | -0.035 |
Token | He |
Token ID | 679 |
Feature activation | -8.44e-01 |
Pos logit contributions | |
Flor | +0.001 |
grate | +0.001 |
cheer | +0.001 |
violet | +0.001 |
promise | +0.001 |
Neg logit contributions | |
dd | -0.001 |
ime | -0.001 |
ug | -0.001 |
happened | -0.001 |
endings | -0.001 |
Token | went |
Token ID | 1816 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
dd | +0.037 |
ime | +0.035 |
endings | +0.033 |
ug | +0.033 |
happened | +0.031 |
Neg logit contributions | |
Flor | -0.023 |
grate | -0.021 |
| -0.021 |
cheer | -0.020 |
violet | -0.020 |
Token | up |
Token ID | 510 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
dd | +0.014 |
happened | +0.012 |
ruck | +0.012 |
ime | +0.011 |
ly | +0.011 |
Neg logit contributions | |
Flor | -0.028 |
Deal | -0.027 |
violet | -0.026 |
grate | -0.026 |
promise | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
dd | +0.034 |
ime | +0.031 |
endings | +0.030 |
ug | +0.030 |
happened | +0.029 |
Neg logit contributions | |
Flor | -0.021 |
grate | -0.019 |
| -0.018 |
cheer | -0.018 |
Av | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
Flor | +0.002 |
grate | +0.002 |
violet | +0.001 |
cheer | +0.001 |
| +0.001 |
Neg logit contributions | |
dd | -0.002 |
ime | -0.002 |
happened | -0.002 |
ug | -0.002 |
endings | -0.002 |
Token | te |
Token ID | 573 |
Feature activation | +4.58e+00 |
Pos logit contributions | |
dd | +0.005 |
ime | +0.005 |
ug | +0.005 |
happened | +0.005 |
endings | +0.005 |
Neg logit contributions | |
Flor | -0.004 |
grate | -0.003 |
cheer | -0.003 |
| -0.003 |
violet | -0.003 |
Token | ash |
Token ID | 1077 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
| +0.159 |
refreshed | +0.154 |
closer | +0.151 |
Flor | +0.151 |
better | +0.147 |
Neg logit contributions | |
ug | -0.155 |
ime | -0.153 |
aw | -0.147 |
dd | -0.144 |
endings | -0.143 |
Token | op |
Token ID | 404 |
Feature activation | -6.71e-01 |
Pos logit contributions | |
Flor | +0.025 |
grate | +0.025 |
violet | +0.023 |
cheer | +0.023 |
| +0.023 |
Neg logit contributions | |
dd | -0.022 |
ime | -0.020 |
ug | -0.019 |
happened | -0.018 |
endings | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
dd | +0.025 |
ime | +0.023 |
ug | +0.023 |
happened | +0.021 |
aw | +0.021 |
Neg logit contributions | |
Flor | -0.022 |
grate | -0.020 |
violet | -0.020 |
cheer | -0.019 |
Deal | -0.019 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
dd | +0.011 |
happened | +0.010 |
ime | +0.010 |
ug | +0.009 |
endings | +0.009 |
Neg logit contributions | |
Flor | -0.007 |
grate | -0.007 |
violet | -0.007 |
promise | -0.007 |
cheer | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
dd | +0.093 |
ime | +0.085 |
ug | +0.083 |
endings | +0.081 |
happened | +0.078 |
Neg logit contributions | |
Flor | -0.048 |
| -0.042 |
grate | -0.040 |
Av | -0.039 |
cheer | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
Flor | +0.008 |
grate | +0.008 |
violet | +0.007 |
| +0.007 |
cheer | +0.007 |
Neg logit contributions | |
dd | -0.014 |
ime | -0.013 |
ug | -0.012 |
happened | -0.012 |
endings | -0.012 |
Token | she |
Token ID | 673 |
Feature activation | +5.16e-01 |
Pos logit contributions | |
Flor | +0.007 |
grate | +0.007 |
violet | +0.006 |
cheer | +0.006 |
| +0.006 |
Neg logit contributions | |
dd | -0.016 |
ime | -0.015 |
happened | -0.014 |
ug | -0.014 |
endings | -0.014 |
Token | opened |
Token ID | 4721 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
Flor | +0.017 |
violet | +0.016 |
grate | +0.016 |
cheer | +0.015 |
Deal | +0.015 |
Neg logit contributions | |
dd | -0.019 |
happened | -0.017 |
ime | -0.017 |
ug | -0.016 |
aw | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
Flor | +0.054 |
grate | +0.053 |
violet | +0.049 |
cheer | +0.049 |
promise | +0.048 |
Neg logit contributions | |
dd | -0.068 |
ime | -0.062 |
happened | -0.059 |
ug | -0.058 |
endings | -0.056 |
Token | fl |
Token ID | 781 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
Flor | +0.035 |
grate | +0.033 |
violet | +0.031 |
| +0.031 |
cheer | +0.031 |
Neg logit contributions | |
dd | -0.045 |
ime | -0.043 |
happened | -0.041 |
endings | -0.040 |
ug | -0.040 |
Token | aps |
Token ID | 1686 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
Flor | +0.050 |
grate | +0.049 |
violet | +0.046 |
cheer | +0.046 |
| +0.045 |
Neg logit contributions | |
dd | -0.039 |
ime | -0.033 |
happened | -0.032 |
endings | -0.031 |
ug | -0.029 |
Token | of |
Token ID | 286 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
Flor | +0.142 |
grate | +0.134 |
violet | +0.127 |
| +0.127 |
cheer | +0.127 |
Neg logit contributions | |
dd | -0.166 |
ime | -0.152 |
happened | -0.143 |
ug | -0.143 |
endings | -0.140 |
Token | the |
Token ID | 262 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
grate | +0.008 |
Flor | +0.008 |
promise | +0.008 |
violet | +0.007 |
refreshed | +0.007 |
Neg logit contributions | |
dd | -0.009 |
ime | -0.009 |
happened | -0.008 |
ug | -0.008 |
endings | -0.008 |
Token | box |
Token ID | 3091 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
dd | +0.011 |
ime | +0.010 |
ug | +0.009 |
happened | +0.009 |
endings | +0.009 |
Neg logit contributions | |
Flor | -0.007 |
grate | -0.006 |
cheer | -0.006 |
| -0.006 |
violet | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Flor | +0.025 |
Deal | +0.022 |
violet | +0.022 |
Av | +0.021 |
Amen | +0.021 |
Neg logit contributions | |
ime | -0.017 |
dd | -0.016 |
happened | -0.016 |
aw | -0.015 |
ug | -0.015 |
Token | climbed |
Token ID | 19952 |
Feature activation | -9.71e-01 |
Pos logit contributions | |
Flor | +0.004 |
grate | +0.004 |
violet | +0.004 |
promise | +0.004 |
cheer | +0.004 |
Neg logit contributions | |
dd | -0.005 |
happened | -0.005 |
ime | -0.005 |
ug | -0.004 |
endings | -0.004 |
Token | She |
Token ID | 3347 |
Feature activation | +7.70e-01 |
Pos logit contributions | |
Flor | +0.014 |
grate | +0.014 |
violet | +0.013 |
| +0.013 |
cheer | +0.013 |
Neg logit contributions | |
dd | -0.016 |
ime | -0.014 |
ug | -0.013 |
endings | -0.013 |
happened | -0.013 |
Token | reached |
Token ID | 4251 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
Flor | +0.022 |
grate | +0.020 |
cheer | +0.020 |
| +0.019 |
violet | +0.019 |
Neg logit contributions | |
dd | -0.032 |
ime | -0.029 |
happened | -0.029 |
ug | -0.027 |
endings | -0.027 |
Token | for |
Token ID | 329 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ime | +0.012 |
dd | +0.012 |
happened | +0.011 |
ug | +0.011 |
endings | +0.010 |
Neg logit contributions | |
grate | -0.010 |
Flor | -0.009 |
violet | -0.009 |
cheer | -0.009 |
forget | -0.009 |
Token | her |
Token ID | 607 |
Feature activation | +6.15e-01 |
Pos logit contributions | |
dd | +0.048 |
happened | +0.043 |
ime | +0.042 |
endings | +0.040 |
electricity | +0.038 |
Neg logit contributions | |
grate | -0.040 |
Flor | -0.037 |
promise | -0.037 |
violet | -0.037 |
bug | -0.036 |
Token | toy |
Token ID | 13373 |
Feature activation | +3.71e-01 |
Pos logit contributions | |
grate | +0.019 |
refreshed | +0.018 |
Together | +0.017 |
Amen | +0.017 |
Flor | +0.017 |
Neg logit contributions | |
happened | -0.020 |
ime | -0.018 |
emergency | -0.018 |
ly | -0.017 |
- | -0.016 |
Token | te |
Token ID | 573 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
grate | +0.011 |
refreshed | +0.011 |
| +0.011 |
violet | +0.011 |
Flor | +0.011 |
Neg logit contributions | |
dd | -0.012 |
happened | -0.012 |
ug | -0.012 |
ime | -0.012 |
endings | -0.011 |
Token | ap |
Token ID | 499 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
| +0.154 |
refreshed | +0.150 |
Flor | +0.149 |
forget | +0.142 |
cheer | +0.141 |
Neg logit contributions | |
ime | -0.138 |
ug | -0.138 |
dd | -0.137 |
endings | -0.127 |
aw | -0.127 |
Token | ot |
Token ID | 313 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
Flor | +0.127 |
grate | +0.121 |
violet | +0.117 |
| +0.117 |
refreshed | +0.117 |
Neg logit contributions | |
dd | -0.098 |
ime | -0.090 |
ug | -0.084 |
happened | -0.083 |
endings | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
dd | +0.055 |
ime | +0.051 |
ug | +0.048 |
happened | +0.048 |
endings | +0.047 |
Neg logit contributions | |
Flor | -0.040 |
grate | -0.037 |
violet | -0.035 |
| -0.035 |
cheer | -0.035 |
Token | but |
Token ID | 475 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
promise | +0.011 |
Flor | +0.010 |
forget | +0.010 |
refreshed | +0.010 |
grate | +0.010 |
Neg logit contributions | |
dd | -0.012 |
happened | -0.010 |
aged | -0.010 |
aw | -0.010 |
anked | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
dd | +0.027 |
happened | +0.026 |
ime | +0.023 |
ug | +0.022 |
endings | +0.022 |
Neg logit contributions | |
grate | -0.019 |
violet | -0.018 |
Flor | -0.018 |
cheer | -0.017 |
promise | -0.017 |
Token | came |
Token ID | 1625 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
dd | +0.028 |
ime | +0.025 |
endings | +0.024 |
ug | +0.024 |
happened | +0.024 |
Neg logit contributions | |
Flor | -0.019 |
grate | -0.018 |
| -0.017 |
cheer | -0.017 |
violet | -0.016 |
Token | home |
Token ID | 1363 |
Feature activation | -9.79e-01 |
Pos logit contributions | |
dd | +0.017 |
ime | +0.016 |
happened | +0.015 |
ug | +0.015 |
endings | +0.013 |
Neg logit contributions | |
Flor | -0.018 |
violet | -0.018 |
grate | -0.018 |
promise | -0.016 |
refreshed | -0.016 |
Token | she |
Token ID | 673 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ime | +0.033 |
happened | +0.032 |
dd | +0.032 |
ug | +0.028 |
emergency | +0.027 |
Neg logit contributions | |
violet | -0.034 |
Flor | -0.031 |
grate | -0.029 |
Deal | -0.029 |
Amen | -0.028 |
Token | asked |
Token ID | 1965 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
Flor | +0.039 |
violet | +0.037 |
grate | +0.035 |
Av | +0.034 |
cheer | +0.033 |
Neg logit contributions | |
dd | -0.041 |
happened | -0.039 |
ime | -0.038 |
ug | -0.033 |
aw | -0.033 |
Token | if |
Token ID | 611 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
grate | +0.002 |
violet | +0.002 |
cheer | +0.002 |
Flor | +0.002 |
applaud | +0.002 |
Neg logit contributions | |
dd | -0.003 |
happened | -0.003 |
ime | -0.003 |
endings | -0.003 |
ug | -0.003 |
Token | Flor |
Token ID | 4432 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
grate | +0.017 |
Flor | +0.017 |
cheer | +0.016 |
violet | +0.016 |
| +0.016 |
Neg logit contributions | |
dd | -0.041 |
ime | -0.038 |
happened | -0.037 |
endings | -0.037 |
ug | -0.036 |
Token | a |
Token ID | 64 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
Flor | +0.145 |
grate | +0.141 |
cheer | +0.134 |
| +0.134 |
violet | +0.134 |
Neg logit contributions | |
dd | -0.159 |
ime | -0.146 |
ug | -0.136 |
happened | -0.135 |
endings | -0.134 |
Token | had |
Token ID | 550 |
Feature activation | +9.64e-01 |
Pos logit contributions | |
Flor | +0.043 |
grate | +0.041 |
violet | +0.038 |
cheer | +0.038 |
| +0.038 |
Neg logit contributions | |
dd | -0.059 |
ime | -0.054 |
happened | -0.051 |
ug | -0.051 |
endings | -0.050 |
Token | washed |
Token ID | 18989 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Flor | +0.027 |
grate | +0.025 |
| +0.024 |
Av | +0.024 |
cheer | +0.023 |
Neg logit contributions | |
dd | -0.042 |
ug | -0.038 |
ime | -0.038 |
endings | -0.035 |
opter | -0.035 |
Token | her |
Token ID | 607 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
Flor | +0.044 |
violet | +0.043 |
grate | +0.042 |
cheer | +0.040 |
count | +0.039 |
Neg logit contributions | |
dd | -0.063 |
ime | -0.059 |
happened | -0.055 |
ug | -0.055 |
endings | -0.054 |
Token | toys |
Token ID | 14958 |
Feature activation | +9.43e-01 |
Pos logit contributions | |
Flor | +0.033 |
grate | +0.032 |
violet | +0.030 |
| +0.030 |
cheer | +0.030 |
Neg logit contributions | |
dd | -0.032 |
ime | -0.029 |
happened | -0.027 |
ug | -0.027 |
endings | -0.027 |
Token | them |
Token ID | 606 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
Flor | +0.045 |
grate | +0.043 |
violet | +0.042 |
cheer | +0.040 |
promise | +0.040 |
Neg logit contributions | |
dd | -0.069 |
ime | -0.062 |
happened | -0.060 |
ug | -0.059 |
endings | -0.058 |
Token | tea |
Token ID | 8887 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
Flor | +0.013 |
grate | +0.012 |
violet | +0.012 |
| +0.012 |
cheer | +0.012 |
Neg logit contributions | |
dd | -0.014 |
ime | -0.013 |
happened | -0.012 |
ug | -0.012 |
endings | -0.012 |
Token | from |
Token ID | 422 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
Flor | +0.068 |
grate | +0.064 |
violet | +0.062 |
cheer | +0.061 |
| +0.061 |
Neg logit contributions | |
dd | -0.088 |
ime | -0.079 |
happened | -0.077 |
ug | -0.076 |
endings | -0.074 |
Token | her |
Token ID | 607 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
Flor | +0.019 |
grate | +0.018 |
| +0.017 |
violet | +0.017 |
cheer | +0.017 |
Neg logit contributions | |
dd | -0.031 |
ime | -0.028 |
ug | -0.027 |
happened | -0.027 |
endings | -0.027 |
Token | toy |
Token ID | 13373 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
Flor | +0.011 |
cheer | +0.010 |
grate | +0.010 |
violet | +0.010 |
| +0.010 |
Neg logit contributions | |
dd | -0.013 |
ime | -0.013 |
endings | -0.012 |
ug | -0.012 |
happened | -0.012 |
Token | te |
Token ID | 573 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
Flor | +0.016 |
grate | +0.015 |
cheer | +0.015 |
violet | +0.015 |
refreshed | +0.014 |
Neg logit contributions | |
dd | -0.018 |
ime | -0.017 |
ug | -0.017 |
endings | -0.017 |
happened | -0.017 |
Token | ap |
Token ID | 499 |
Feature activation | +3.00e+00 |
Pos logit contributions | |
Flor | +0.150 |
| +0.147 |
refreshed | +0.144 |
cheer | +0.143 |
forget | +0.140 |
Neg logit contributions | |
dd | -0.142 |
ime | -0.138 |
ug | -0.138 |
endings | -0.128 |
aw | -0.127 |
Token | ot |
Token ID | 313 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Flor | +0.111 |
grate | +0.106 |
| +0.101 |
violet | +0.101 |
cheer | +0.101 |
Neg logit contributions | |
dd | -0.102 |
ime | -0.091 |
ug | -0.086 |
happened | -0.085 |
endings | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
dd | +0.061 |
ime | +0.057 |
happened | +0.054 |
endings | +0.054 |
ug | +0.053 |
Neg logit contributions | |
Flor | -0.045 |
grate | -0.043 |
cheer | -0.040 |
| -0.040 |
violet | -0.040 |
Token | they |
Token ID | 484 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
dd | +0.010 |
happened | +0.009 |
ug | +0.008 |
ime | +0.008 |
endings | +0.008 |
Neg logit contributions | |
Flor | -0.006 |
grate | -0.006 |
Deal | -0.006 |
| -0.006 |
violet | -0.006 |
Token | all |
Token ID | 477 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
dd | +0.024 |
happened | +0.021 |
ime | +0.021 |
ug | +0.020 |
opter | +0.018 |
Neg logit contributions | |
Flor | -0.021 |
grate | -0.020 |
Deal | -0.019 |
violet | -0.019 |
Av | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +5.24e-02 |
Pos logit contributions | |
dd | +0.050 |
ime | +0.047 |
happened | +0.046 |
ug | +0.046 |
endings | +0.042 |
Neg logit contributions | |
Flor | -0.038 |
violet | -0.036 |
grate | -0.035 |
promise | -0.033 |
forget | -0.033 |
Token | She |
Token ID | 1375 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
promise | +0.002 |
grate | +0.002 |
forget | +0.002 |
refreshed | +0.002 |
Flor | +0.002 |
Neg logit contributions | |
dd | -0.002 |
ime | -0.002 |
happened | -0.002 |
ug | -0.002 |
aw | -0.002 |
Token | picked |
Token ID | 6497 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
Flor | +0.036 |
grate | +0.034 |
cheer | +0.033 |
violet | +0.033 |
| +0.032 |
Neg logit contributions | |
dd | -0.044 |
ime | -0.040 |
happened | -0.039 |
ug | -0.038 |
endings | -0.037 |
Token | up |
Token ID | 510 |
Feature activation | +7.34e-01 |
Pos logit contributions | |
Flor | +0.005 |
grate | +0.005 |
violet | +0.004 |
cheer | +0.004 |
| +0.004 |
Neg logit contributions | |
dd | -0.010 |
ime | -0.009 |
happened | -0.008 |
ug | -0.008 |
endings | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
Flor | +0.020 |
grate | +0.019 |
violet | +0.018 |
cheer | +0.018 |
| +0.018 |
Neg logit contributions | |
dd | -0.031 |
ime | -0.029 |
happened | -0.027 |
ug | -0.027 |
endings | -0.027 |
Token | doll |
Token ID | 3654 |
Feature activation | +4.25e+00 |
Pos logit contributions | |
Flor | +0.038 |
grate | +0.038 |
violet | +0.035 |
| +0.034 |
cheer | +0.034 |
Neg logit contributions | |
dd | -0.052 |
ime | -0.048 |
happened | -0.047 |
endings | -0.045 |
ug | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
Deal | +0.192 |
refreshed | +0.186 |
Flor | +0.181 |
applaud | +0.176 |
violet | +0.172 |
Neg logit contributions | |
ly | -0.080 |
aw | -0.076 |
dd | -0.071 |
ime | -0.070 |
ors | -0.068 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Flor | +0.014 |
grate | +0.013 |
violet | +0.013 |
Deal | +0.013 |
refreshed | +0.012 |
Neg logit contributions | |
dd | -0.016 |
happened | -0.016 |
ime | -0.015 |
aw | -0.014 |
ug | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
Flor | +0.038 |
violet | +0.037 |
forget | +0.034 |
count | +0.034 |
promise | +0.034 |
Neg logit contributions | |
dd | -0.035 |
happened | -0.032 |
aw | -0.031 |
ime | -0.030 |
ly | -0.030 |
Token | The |
Token ID | 383 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
grate | +0.005 |
promise | +0.005 |
Flor | +0.005 |
forget | +0.005 |
cheer | +0.005 |
Neg logit contributions | |
dd | -0.007 |
happened | -0.007 |
ime | -0.007 |
ug | -0.006 |
aw | -0.006 |
Token | doll |
Token ID | 3654 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
grate | +0.023 |
Flor | +0.023 |
violet | +0.021 |
refreshed | +0.021 |
| +0.021 |
Neg logit contributions | |
happened | -0.023 |
dd | -0.023 |
emergency | -0.022 |
endings | -0.022 |
ime | -0.022 |
Token | poured |
Token ID | 19036 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Flor | +0.009 |
grate | +0.009 |
violet | +0.008 |
cheer | +0.008 |
| +0.008 |
Neg logit contributions | |
dd | -0.012 |
ime | -0.010 |
happened | -0.010 |
ug | -0.010 |
aw | -0.009 |
Token | some |
Token ID | 617 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
Flor | +0.037 |
grate | +0.036 |
violet | +0.035 |
cheer | +0.033 |
promise | +0.033 |
Neg logit contributions | |
dd | -0.057 |
ime | -0.051 |
happened | -0.049 |
ug | -0.049 |
aw | -0.048 |
Token | water |
Token ID | 1660 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
Flor | +0.022 |
grate | +0.021 |
| +0.020 |
violet | +0.020 |
forget | +0.019 |
Neg logit contributions | |
dd | -0.041 |
ime | -0.038 |
happened | -0.036 |
ug | -0.035 |
aw | -0.035 |
Token | from |
Token ID | 422 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
dd | +0.004 |
ime | +0.004 |
ug | +0.004 |
happened | +0.004 |
endings | +0.003 |
Neg logit contributions | |
Flor | -0.003 |
grate | -0.003 |
| -0.003 |
violet | -0.003 |
cheer | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
Flor | +0.011 |
grate | +0.010 |
| +0.010 |
violet | +0.010 |
cheer | +0.010 |
Neg logit contributions | |
dd | -0.019 |
ime | -0.018 |
ug | -0.017 |
endings | -0.017 |
happened | -0.017 |
Token | te |
Token ID | 573 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
grate | +0.011 |
Flor | +0.010 |
| +0.009 |
applaud | +0.009 |
Together | +0.009 |
Neg logit contributions | |
ime | -0.016 |
dd | -0.015 |
emergency | -0.015 |
happened | -0.015 |
ly | -0.015 |
Token | ap |
Token ID | 499 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
| +0.152 |
forget | +0.143 |
refreshed | +0.143 |
Flor | +0.142 |
closer | +0.140 |
Neg logit contributions | |
ug | -0.138 |
ime | -0.135 |
aw | -0.130 |
dd | -0.130 |
endings | -0.126 |
Token | ot |
Token ID | 313 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Flor | +0.103 |
grate | +0.098 |
| +0.094 |
violet | +0.093 |
cheer | +0.093 |
Neg logit contributions | |
dd | -0.090 |
ime | -0.082 |
ug | -0.077 |
happened | -0.076 |
endings | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
dd | +0.053 |
ime | +0.049 |
endings | +0.046 |
happened | +0.046 |
ug | +0.046 |
Neg logit contributions | |
Flor | -0.040 |
grate | -0.038 |
violet | -0.036 |
cheer | -0.036 |
| -0.036 |
Token | pretended |
Token ID | 36766 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
Flor | +0.009 |
grate | +0.009 |
promise | +0.008 |
Deal | +0.008 |
| +0.008 |
Neg logit contributions | |
dd | -0.013 |
happened | -0.011 |
ime | -0.011 |
ug | -0.010 |
aw | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Flor | +0.004 |
grate | +0.004 |
| +0.004 |
cheer | +0.004 |
violet | +0.004 |
Neg logit contributions | |
dd | -0.006 |
ime | -0.005 |
ug | -0.005 |
happened | -0.005 |
endings | -0.005 |
Token | poured |
Token ID | 19036 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Flor | +0.055 |
Deal | +0.051 |
grate | +0.051 |
violet | +0.050 |
cheer | +0.049 |
Neg logit contributions | |
dd | -0.073 |
happened | -0.065 |
ime | -0.064 |
aw | -0.062 |
ug | -0.060 |
Token | some |
Token ID | 617 |
Feature activation | +9.20e-01 |
Pos logit contributions | |
Flor | +0.040 |
grate | +0.037 |
violet | +0.036 |
cheer | +0.035 |
| +0.034 |
Neg logit contributions | |
dd | -0.063 |
ime | -0.057 |
happened | -0.055 |
ug | -0.054 |
endings | -0.053 |
Token | water |
Token ID | 1660 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
Flor | +0.026 |
grate | +0.025 |
violet | +0.024 |
| +0.024 |
cheer | +0.023 |
Neg logit contributions | |
dd | -0.038 |
ime | -0.035 |
happened | -0.033 |
ug | -0.033 |
aw | -0.032 |
Token | from |
Token ID | 422 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
dd | +0.006 |
ime | +0.006 |
ug | +0.005 |
happened | +0.005 |
endings | +0.005 |
Neg logit contributions | |
Flor | -0.005 |
grate | -0.004 |
violet | -0.004 |
| -0.004 |
cheer | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
Flor | +0.012 |
grate | +0.012 |
violet | +0.011 |
cheer | +0.011 |
| +0.011 |
Neg logit contributions | |
dd | -0.014 |
ime | -0.013 |
happened | -0.012 |
ug | -0.012 |
endings | -0.012 |
Token | te |
Token ID | 573 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
grate | +0.013 |
Flor | +0.013 |
violet | +0.011 |
Together | +0.011 |
| +0.011 |
Neg logit contributions | |
ime | -0.019 |
dd | -0.019 |
happened | -0.019 |
emergency | -0.019 |
ly | -0.018 |
Token | ap |
Token ID | 499 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
| +0.145 |
Flor | +0.144 |
refreshed | +0.139 |
forget | +0.137 |
applaud | +0.136 |
Neg logit contributions | |
ug | -0.140 |
ime | -0.139 |
dd | -0.137 |
aw | -0.132 |
endings | -0.130 |
Token | ot |
Token ID | 313 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
Flor | +0.101 |
grate | +0.096 |
violet | +0.092 |
| +0.092 |
cheer | +0.091 |
Neg logit contributions | |
dd | -0.095 |
ime | -0.086 |
ug | -0.080 |
happened | -0.080 |
endings | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
dd | +0.060 |
ime | +0.057 |
endings | +0.054 |
happened | +0.054 |
ug | +0.053 |
Neg logit contributions | |
Flor | -0.037 |
grate | -0.036 |
cheer | -0.033 |
| -0.033 |
violet | -0.033 |
Token | She |
Token ID | 1375 |
Feature activation | +9.12e-01 |
Pos logit contributions | |
dd | +0.008 |
ime | +0.008 |
happened | +0.008 |
ug | +0.007 |
endings | +0.007 |
Neg logit contributions | |
Flor | -0.006 |
grate | -0.006 |
cheer | -0.006 |
promise | -0.006 |
violet | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
Flor | +0.027 |
grate | +0.025 |
| +0.025 |
cheer | +0.024 |
Deal | +0.024 |
Neg logit contributions | |
dd | -0.036 |
happened | -0.032 |
ime | -0.032 |
aw | -0.030 |
ug | -0.030 |
Token | poured |
Token ID | 19036 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Flor | +0.025 |
| +0.023 |
cheer | +0.023 |
grate | +0.022 |
violet | +0.022 |
Neg logit contributions | |
dd | -0.035 |
happened | -0.032 |
ime | -0.031 |
ug | -0.029 |
aw | -0.028 |
Token | some |
Token ID | 617 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
Flor | +0.036 |
grate | +0.034 |
violet | +0.033 |
cheer | +0.032 |
promise | +0.032 |
Neg logit contributions | |
dd | -0.057 |
ime | -0.051 |
happened | -0.049 |
ug | -0.048 |
endings | -0.047 |
Token | water |
Token ID | 1660 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
Flor | +0.023 |
grate | +0.022 |
violet | +0.022 |
| +0.021 |
cheer | +0.020 |
Neg logit contributions | |
dd | -0.033 |
ime | -0.030 |
happened | -0.029 |
ug | -0.028 |
aw | -0.027 |
Token | from |
Token ID | 422 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
dd | +0.011 |
ime | +0.010 |
ug | +0.009 |
happened | +0.009 |
endings | +0.009 |
Neg logit contributions | |
Flor | -0.008 |
grate | -0.008 |
violet | -0.007 |
| -0.007 |
cheer | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
Flor | +0.010 |
grate | +0.009 |
| +0.009 |
violet | +0.009 |
cheer | +0.009 |
Neg logit contributions | |
dd | -0.018 |
ime | -0.016 |
ug | -0.016 |
endings | -0.016 |
happened | -0.016 |
Token | te |
Token ID | 573 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
grate | +0.008 |
Flor | +0.008 |
applaud | +0.007 |
violet | +0.007 |
| +0.007 |
Neg logit contributions | |
ime | -0.013 |
dd | -0.013 |
emergency | -0.012 |
happened | -0.012 |
endings | -0.012 |
Token | ap |
Token ID | 499 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
| +0.148 |
Flor | +0.145 |
refreshed | +0.142 |
forget | +0.141 |
applaud | +0.139 |
Neg logit contributions | |
ime | -0.134 |
ug | -0.134 |
dd | -0.134 |
aw | -0.126 |
endings | -0.125 |
Token | ot |
Token ID | 313 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
Flor | +0.101 |
grate | +0.096 |
violet | +0.092 |
| +0.092 |
cheer | +0.091 |
Neg logit contributions | |
dd | -0.095 |
ime | -0.086 |
ug | -0.080 |
happened | -0.080 |
endings | -0.078 |
Token | and |
Token ID | 290 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
dd | +0.054 |
ime | +0.050 |
endings | +0.048 |
happened | +0.047 |
ug | +0.046 |
Neg logit contributions | |
Flor | -0.042 |
grate | -0.040 |
cheer | -0.038 |
| -0.038 |
violet | -0.038 |
Token | handed |
Token ID | 10158 |
Feature activation | +9.32e-01 |
Pos logit contributions | |
Flor | +0.003 |
promise | +0.003 |
grate | +0.003 |
Deal | +0.003 |
| +0.003 |
Neg logit contributions | |
dd | -0.005 |
happened | -0.005 |
ime | -0.004 |
ug | -0.004 |
aw | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
Flor | +0.028 |
grate | +0.027 |
violet | +0.026 |
cheer | +0.025 |
| +0.025 |
Neg logit contributions | |
dd | -0.037 |
ime | -0.034 |
happened | -0.032 |
ug | -0.032 |
endings | -0.032 |
Token | ily |
Token ID | 813 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Flor | +0.016 |
grate | +0.015 |
violet | +0.015 |
| +0.015 |
Av | +0.015 |
Neg logit contributions | |
dd | -0.015 |
ime | -0.013 |
happened | -0.012 |
ug | -0.012 |
endings | -0.012 |
Token | tried |
Token ID | 3088 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
Flor | +0.035 |
grate | +0.033 |
violet | +0.032 |
cheer | +0.031 |
Av | +0.030 |
Neg logit contributions | |
dd | -0.041 |
ime | -0.038 |
happened | -0.038 |
aw | -0.036 |
endings | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | +8.49e-01 |
Pos logit contributions | |
Flor | +0.053 |
grate | +0.051 |
violet | +0.050 |
cheer | +0.047 |
refreshed | +0.046 |
Neg logit contributions | |
dd | -0.051 |
happened | -0.048 |
ime | -0.047 |
endings | -0.044 |
ug | -0.041 |
Token | grab |
Token ID | 5552 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
Flor | +0.022 |
violet | +0.021 |
grate | +0.020 |
refreshed | +0.019 |
Av | +0.019 |
Neg logit contributions | |
dd | -0.036 |
ime | -0.034 |
happened | -0.032 |
ug | -0.032 |
endings | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
Flor | +0.034 |
grate | +0.032 |
violet | +0.030 |
cheer | +0.030 |
| +0.030 |
Neg logit contributions | |
dd | -0.056 |
ime | -0.051 |
ug | -0.049 |
happened | -0.048 |
endings | -0.048 |
Token | doll |
Token ID | 3654 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
Flor | +0.032 |
grate | +0.030 |
cheer | +0.029 |
violet | +0.029 |
| +0.029 |
Neg logit contributions | |
dd | -0.040 |
ime | -0.035 |
ug | -0.034 |
endings | -0.034 |
happened | -0.033 |
Token | from |
Token ID | 422 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
Flor | +0.145 |
violet | +0.139 |
refreshed | +0.136 |
grate | +0.130 |
| +0.128 |
Neg logit contributions | |
dd | -0.132 |
ime | -0.126 |
ug | -0.122 |
ly | -0.120 |
ors | -0.118 |
Token | Jake |
Token ID | 14757 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
Flor | +0.007 |
grate | +0.007 |
violet | +0.007 |
| +0.007 |
cheer | +0.007 |
Neg logit contributions | |
dd | -0.011 |
ime | -0.010 |
happened | -0.009 |
ug | -0.009 |
endings | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
Flor | +0.007 |
grate | +0.007 |
violet | +0.007 |
cheer | +0.006 |
| +0.006 |
Neg logit contributions | |
dd | -0.013 |
ime | -0.012 |
ug | -0.011 |
happened | -0.011 |
endings | -0.011 |
Token | but |
Token ID | 475 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
Flor | +0.014 |
grate | +0.013 |
violet | +0.012 |
Av | +0.012 |
cheer | +0.011 |
Neg logit contributions | |
dd | -0.022 |
ime | -0.021 |
happened | -0.020 |
ug | -0.020 |
endings | -0.019 |
Token | he |
Token ID | 339 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
dd | +0.024 |
ime | +0.021 |
happened | +0.021 |
ug | +0.020 |
endings | +0.020 |
Neg logit contributions | |
Flor | -0.018 |
grate | -0.018 |
violet | -0.017 |
cheer | -0.016 |
| -0.016 |
Token | Her |
Token ID | 2332 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
cheer | +0.004 |
grate | +0.004 |
promise | +0.004 |
count | +0.004 |
Flor | +0.004 |
Neg logit contributions | |
dd | -0.005 |
happened | -0.005 |
endings | -0.005 |
ime | -0.005 |
aw | -0.005 |
Token | mom |
Token ID | 1995 |
Feature activation | +7.07e-01 |
Pos logit contributions | |
Flor | +0.027 |
grate | +0.026 |
violet | +0.024 |
| +0.024 |
cheer | +0.023 |
Neg logit contributions | |
dd | -0.048 |
ime | -0.044 |
happened | -0.043 |
ug | -0.042 |
endings | -0.042 |
Token | picked |
Token ID | 6497 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
Flor | +0.020 |
violet | +0.018 |
grate | +0.017 |
Av | +0.016 |
refreshed | +0.016 |
Neg logit contributions | |
dd | -0.030 |
ime | -0.028 |
happened | -0.028 |
ug | -0.026 |
aw | -0.025 |
Token | up |
Token ID | 510 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
Flor | +0.017 |
grate | +0.016 |
| +0.015 |
cheer | +0.014 |
Av | +0.014 |
Neg logit contributions | |
dd | -0.041 |
ime | -0.038 |
ug | -0.037 |
endings | -0.037 |
happened | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
Flor | +0.032 |
grate | +0.030 |
violet | +0.029 |
cheer | +0.028 |
| +0.028 |
Neg logit contributions | |
dd | -0.053 |
ime | -0.049 |
happened | -0.046 |
ug | -0.046 |
endings | -0.045 |
Token | doll |
Token ID | 3654 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
Flor | +0.030 |
grate | +0.028 |
violet | +0.026 |
| +0.026 |
cheer | +0.026 |
Neg logit contributions | |
dd | -0.054 |
ime | -0.050 |
happened | -0.048 |
ug | -0.048 |
endings | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
Flor | +0.163 |
violet | +0.155 |
refreshed | +0.148 |
Deal | +0.148 |
Av | +0.143 |
Neg logit contributions | |
dd | -0.114 |
ime | -0.108 |
happened | -0.099 |
ug | -0.098 |
ly | -0.098 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Flor | +0.009 |
Deal | +0.009 |
violet | +0.009 |
grate | +0.009 |
| +0.008 |
Neg logit contributions | |
dd | -0.015 |
happened | -0.015 |
ime | -0.013 |
ug | -0.013 |
aw | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
Flor | +0.043 |
grate | +0.041 |
violet | +0.040 |
cheer | +0.039 |
| +0.038 |
Neg logit contributions | |
dd | -0.048 |
ime | -0.044 |
happened | -0.042 |
endings | -0.040 |
ug | -0.040 |
Token | back |
Token ID | 736 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
dd | +0.010 |
happened | +0.009 |
ime | +0.009 |
endings | +0.008 |
ug | +0.008 |
Neg logit contributions | |
Flor | -0.009 |
violet | -0.009 |
grate | -0.008 |
| -0.008 |
cheer | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
dd | +0.002 |
ime | +0.002 |
happened | +0.002 |
ug | +0.002 |
endings | +0.001 |
Neg logit contributions | |
Flor | -0.002 |
grate | -0.001 |
violet | -0.001 |
cheer | -0.001 |
refreshed | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
Flor | +0.023 |
grate | +0.022 |
violet | +0.021 |
| +0.020 |
cheer | +0.020 |
Neg logit contributions | |
dd | -0.038 |
ime | -0.036 |
happened | -0.035 |
endings | -0.034 |
ug | -0.033 |
Token | He |
Token ID | 679 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
Flor | +0.001 |
grate | +0.001 |
cheer | +0.001 |
violet | +0.001 |
| +0.001 |
Neg logit contributions | |
dd | -0.001 |
ime | -0.001 |
happened | -0.001 |
ug | -0.001 |
endings | -0.001 |
Token | picks |
Token ID | 11103 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
dd | +0.038 |
happened | +0.035 |
ime | +0.034 |
aw | +0.033 |
endings | +0.032 |
Neg logit contributions | |
Flor | -0.025 |
grate | -0.023 |
violet | -0.023 |
| -0.022 |
cheer | -0.022 |
Token | up |
Token ID | 510 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
Flor | +0.013 |
grate | +0.012 |
| +0.011 |
Av | +0.011 |
cheer | +0.011 |
Neg logit contributions | |
dd | -0.036 |
ime | -0.034 |
ug | -0.032 |
endings | -0.032 |
happened | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +9.75e-01 |
Pos logit contributions | |
Flor | +0.013 |
grate | +0.013 |
| +0.012 |
cheer | +0.011 |
violet | +0.011 |
Neg logit contributions | |
dd | -0.026 |
ime | -0.024 |
ug | -0.023 |
happened | -0.022 |
endings | -0.022 |
Token | doll |
Token ID | 3654 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
Flor | +0.031 |
grate | +0.029 |
cheer | +0.028 |
violet | +0.028 |
| +0.028 |
Neg logit contributions | |
dd | -0.038 |
ime | -0.035 |
ug | -0.033 |
happened | -0.032 |
endings | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
Flor | +0.163 |
refreshed | +0.160 |
Deal | +0.158 |
violet | +0.156 |
applaud | +0.150 |
Neg logit contributions | |
aw | -0.097 |
ly | -0.093 |
dd | -0.093 |
ime | -0.092 |
ors | -0.086 |
Token | the |
Token ID | 262 |
Feature activation | +5.33e-03 |
Pos logit contributions | |
happened | +0.004 |
dd | +0.004 |
aw | +0.003 |
missiles | +0.003 |
motorcycles | +0.003 |
Neg logit contributions | |
Flor | -0.003 |
grate | -0.003 |
refreshed | -0.003 |
violet | -0.003 |
| -0.003 |
Token | car |
Token ID | 1097 |
Feature activation | +7.20e-01 |
Pos logit contributions | |
Flor | +0.000 |
grate | +0.000 |
violet | +0.000 |
| +0.000 |
cheer | +0.000 |
Neg logit contributions | |
dd | -0.000 |
ime | -0.000 |
happened | -0.000 |
endings | -0.000 |
ug | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
Flor | +0.022 |
violet | +0.020 |
grate | +0.019 |
| +0.019 |
Av | +0.019 |
Neg logit contributions | |
dd | -0.026 |
ime | -0.025 |
happened | -0.025 |
aw | -0.025 |
ug | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
dd | +0.014 |
ime | +0.013 |
happened | +0.012 |
ug | +0.012 |
endings | +0.012 |
Neg logit contributions | |
Flor | -0.014 |
grate | -0.013 |
violet | -0.013 |
cheer | -0.013 |
promise | -0.012 |
Token | wave |
Token ID | 6769 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
dd | +0.011 |
ime | +0.010 |
endings | +0.009 |
ug | +0.009 |
happened | +0.009 |
Neg logit contributions | |
Flor | -0.009 |
grate | -0.009 |
cheer | -0.009 |
violet | -0.008 |
| -0.008 |
Token | comes |
Token ID | 2058 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
dd | +0.025 |
ime | +0.023 |
ug | +0.021 |
endings | +0.021 |
happened | +0.021 |
Neg logit contributions | |
Flor | -0.018 |
grate | -0.017 |
cheer | -0.016 |
| -0.016 |
violet | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +9.82e-02 |
Pos logit contributions | |
dd | +0.018 |
ime | +0.017 |
ug | +0.016 |
happened | +0.016 |
endings | +0.015 |
Neg logit contributions | |
Flor | -0.012 |
grate | -0.011 |
violet | -0.011 |
| -0.011 |
cheer | -0.010 |
Token | It |
Token ID | 632 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
Flor | +0.003 |
grate | +0.003 |
| +0.003 |
violet | +0.003 |
cheer | +0.003 |
Neg logit contributions | |
dd | -0.004 |
ime | -0.004 |
ug | -0.003 |
happened | -0.003 |
endings | -0.003 |
Token | knocks |
Token ID | 36539 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
grate | +0.014 |
Flor | +0.014 |
violet | +0.013 |
| +0.013 |
Deal | +0.013 |
Neg logit contributions | |
happened | -0.012 |
dd | -0.011 |
aw | -0.011 |
ime | -0.010 |
endings | -0.010 |
Token | Anna |
Token ID | 11735 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
Flor | +0.039 |
grate | +0.036 |
| +0.034 |
violet | +0.033 |
cheer | +0.032 |
Neg logit contributions | |
dd | -0.089 |
ime | -0.083 |
ug | -0.079 |
endings | -0.079 |
happened | -0.079 |
Token | and |
Token ID | 290 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
Flor | +0.128 |
grate | +0.120 |
violet | +0.116 |
cheer | +0.115 |
| +0.111 |
Neg logit contributions | |
dd | -0.162 |
ime | -0.150 |
happened | -0.143 |
ug | -0.140 |
aw | -0.138 |
Token | Ben |
Token ID | 3932 |
Feature activation | +6.44e-01 |
Pos logit contributions | |
dd | +0.014 |
happened | +0.014 |
ime | +0.013 |
endings | +0.013 |
ug | +0.012 |
Neg logit contributions | |
grate | -0.013 |
Flor | -0.012 |
violet | -0.012 |
promise | -0.012 |
cheer | -0.012 |
Token | down |
Token ID | 866 |
Feature activation | -9.67e-01 |
Pos logit contributions | |
Flor | +0.019 |
grate | +0.017 |
violet | +0.017 |
| +0.016 |
cheer | +0.016 |
Neg logit contributions | |
dd | -0.027 |
ime | -0.025 |
ug | -0.024 |
happened | -0.023 |
endings | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
dd | +0.037 |
ime | +0.035 |
happened | +0.034 |
ug | +0.033 |
aw | +0.032 |
Neg logit contributions | |
Flor | -0.030 |
violet | -0.028 |
grate | -0.027 |
| -0.026 |
Av | -0.025 |
Token | They |
Token ID | 1119 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
dd | +0.023 |
ime | +0.021 |
happened | +0.020 |
ug | +0.020 |
endings | +0.020 |
Neg logit contributions | |
Flor | -0.014 |
grate | -0.014 |
violet | -0.013 |
cheer | -0.013 |
| -0.013 |
Token | asked |
Token ID | 1965 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
Flor | +0.035 |
grate | +0.033 |
violet | +0.031 |
cheer | +0.031 |
| +0.031 |
Neg logit contributions | |
dd | -0.047 |
ime | -0.043 |
happened | -0.040 |
ug | -0.040 |
endings | -0.040 |
Token | her |
Token ID | 607 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
Flor | +0.007 |
grate | +0.006 |
| +0.006 |
cheer | +0.006 |
Av | +0.006 |
Neg logit contributions | |
dd | -0.017 |
ime | -0.016 |
ug | -0.015 |
endings | -0.014 |
happened | -0.014 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
grate | +0.049 |
Flor | +0.049 |
| +0.048 |
violet | +0.047 |
cheer | +0.046 |
Neg logit contributions | |
dd | -0.056 |
happened | -0.053 |
ime | -0.050 |
endings | -0.048 |
emergency | -0.048 |
Token | why |
Token ID | 1521 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
Flor | +0.045 |
violet | +0.043 |
grate | +0.042 |
refreshed | +0.040 |
cheer | +0.040 |
Neg logit contributions | |
ime | -0.043 |
happened | -0.043 |
dd | -0.042 |
ly | -0.041 |
emergency | -0.040 |
Token | it |
Token ID | 340 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Flor | +0.022 |
grate | +0.020 |
| +0.020 |
violet | +0.019 |
Av | +0.019 |
Neg logit contributions | |
dd | -0.044 |
ime | -0.041 |
ug | -0.040 |
endings | -0.039 |
happened | -0.037 |
Token | tasted |
Token ID | 29187 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
Flor | +0.036 |
grate | +0.035 |
cheer | +0.033 |
violet | +0.033 |
Deal | +0.031 |
Neg logit contributions | |
dd | -0.035 |
happened | -0.035 |
ime | -0.032 |
endings | -0.028 |
emergency | -0.027 |
Token | funny |
Token ID | 8258 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
Flor | +0.124 |
grate | +0.117 |
violet | +0.111 |
cheer | +0.111 |
| +0.110 |
Neg logit contributions | |
dd | -0.169 |
ime | -0.156 |
happened | -0.148 |
endings | -0.145 |
ug | -0.145 |
Token | and |
Token ID | 290 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
Flor | +0.044 |
grate | +0.042 |
violet | +0.040 |
| +0.039 |
cheer | +0.039 |
Neg logit contributions | |
dd | -0.068 |
ime | -0.063 |
happened | -0.061 |
endings | -0.060 |
ug | -0.059 |
Token | her |
Token ID | 607 |
Feature activation | +5.17e-03 |
Pos logit contributions | |
Flor | +0.017 |
grate | +0.016 |
violet | +0.015 |
cheer | +0.015 |
Deal | +0.015 |
Neg logit contributions | |
dd | -0.022 |
happened | -0.021 |
ime | -0.018 |
endings | -0.017 |
aw | -0.017 |
Token | mom |
Token ID | 1995 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
Flor | +0.000 |
grate | +0.000 |
violet | +0.000 |
cheer | +0.000 |
join | +0.000 |
Neg logit contributions | |
happened | -0.000 |
dd | -0.000 |
ime | -0.000 |
endings | -0.000 |
emergency | -0.000 |
Token | said |
Token ID | 531 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
Flor | +0.019 |
violet | +0.017 |
grate | +0.016 |
Av | +0.016 |
Deal | +0.015 |
Neg logit contributions | |
dd | -0.025 |
happened | -0.024 |
ime | -0.023 |
aw | -0.021 |
ug | -0.020 |
Token | it |
Token ID | 340 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
Flor | +0.027 |
grate | +0.026 |
violet | +0.025 |
cheer | +0.024 |
refreshed | +0.024 |
Neg logit contributions | |
dd | -0.037 |
ime | -0.034 |
ug | -0.033 |
happened | -0.033 |
endings | -0.032 |
Token | hopped |
Token ID | 45230 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
dd | +0.051 |
happened | +0.044 |
ime | +0.042 |
ug | +0.041 |
aw | +0.039 |
Neg logit contributions | |
Flor | -0.042 |
| -0.039 |
grate | -0.039 |
violet | -0.038 |
cheer | -0.038 |
Token | out |
Token ID | 503 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
dd | +0.001 |
ug | +0.001 |
ime | +0.001 |
ly | +0.001 |
happened | +0.001 |
Neg logit contributions | |
Flor | -0.001 |
forget | -0.001 |
violet | -0.001 |
promise | -0.001 |
bugs | -0.001 |
Token | of |
Token ID | 286 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
ime | +0.030 |
ug | +0.028 |
dd | +0.028 |
happened | +0.024 |
ruck | +0.022 |
Neg logit contributions | |
Flor | -0.046 |
violet | -0.044 |
grate | -0.043 |
promise | -0.042 |
| -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
grate | +0.015 |
Flor | +0.015 |
promise | +0.014 |
violet | +0.014 |
forget | +0.013 |
Neg logit contributions | |
dd | -0.014 |
ime | -0.014 |
happened | -0.013 |
ug | -0.013 |
ruck | -0.012 |
Token | te |
Token ID | 573 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
grate | +0.004 |
Flor | +0.004 |
violet | +0.004 |
| +0.003 |
cheer | +0.003 |
Neg logit contributions | |
ime | -0.005 |
dd | -0.005 |
happened | -0.005 |
emergency | -0.005 |
endings | -0.005 |
Token | ap |
Token ID | 499 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
| +0.140 |
Flor | +0.136 |
refreshed | +0.131 |
applaud | +0.128 |
closer | +0.127 |
Neg logit contributions | |
ime | -0.144 |
ug | -0.140 |
dd | -0.137 |
endings | -0.133 |
aw | -0.127 |
Token | ot |
Token ID | 313 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
Flor | +0.104 |
grate | +0.098 |
| +0.095 |
violet | +0.094 |
refreshed | +0.093 |
Neg logit contributions | |
dd | -0.089 |
ime | -0.086 |
ug | -0.081 |
happened | -0.075 |
endings | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
dd | +0.049 |
ime | +0.045 |
ug | +0.043 |
happened | +0.042 |
endings | +0.042 |
Neg logit contributions | |
Flor | -0.033 |
grate | -0.031 |
violet | -0.029 |
| -0.029 |
cheer | -0.029 |
Token | began |
Token ID | 2540 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
Deal | +0.001 |
Flor | +0.001 |
promise | +0.001 |
grate | +0.001 |
violet | +0.001 |
Neg logit contributions | |
dd | -0.001 |
happened | -0.001 |
ug | -0.001 |
ime | -0.001 |
aw | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
dd | +0.048 |
ime | +0.046 |
happened | +0.043 |
ug | +0.042 |
endings | +0.039 |
Neg logit contributions | |
Flor | -0.056 |
grate | -0.054 |
Deal | -0.053 |
promise | -0.052 |
violet | -0.051 |
Token | He |
Token ID | 1544 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
Flor | +0.003 |
grate | +0.003 |
violet | +0.002 |
cheer | +0.002 |
| +0.002 |
Neg logit contributions | |
dd | -0.003 |
ime | -0.002 |
ug | -0.002 |
aw | -0.002 |
endings | -0.002 |
Token | tries |
Token ID | 8404 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
dd | +0.027 |
happened | +0.025 |
ime | +0.025 |
aw | +0.024 |
ug | +0.023 |
Neg logit contributions | |
Flor | -0.024 |
grate | -0.023 |
| -0.023 |
refreshed | -0.022 |
cheer | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Flor | +0.023 |
grate | +0.022 |
violet | +0.022 |
refreshed | +0.022 |
promise | +0.021 |
Neg logit contributions | |
happened | -0.014 |
dd | -0.014 |
ime | -0.013 |
anked | -0.012 |
endings | -0.011 |
Token | grab |
Token ID | 5552 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
Flor | +0.030 |
refreshed | +0.030 |
| +0.029 |
Together | +0.029 |
violet | +0.028 |
Neg logit contributions | |
dd | -0.040 |
ime | -0.039 |
happened | -0.035 |
ug | -0.035 |
ly | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
Flor | +0.043 |
grate | +0.042 |
violet | +0.040 |
cheer | +0.039 |
refreshed | +0.039 |
Neg logit contributions | |
dd | -0.059 |
ime | -0.055 |
ug | -0.052 |
happened | -0.051 |
aw | -0.050 |
Token | doll |
Token ID | 3654 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
Flor | +0.038 |
grate | +0.037 |
refreshed | +0.034 |
violet | +0.033 |
| +0.033 |
Neg logit contributions | |
dd | -0.047 |
ime | -0.045 |
happened | -0.044 |
ug | -0.041 |
endings | -0.040 |
Token | from |
Token ID | 422 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
refreshed | +0.151 |
Flor | +0.143 |
violet | +0.141 |
Deal | +0.137 |
applaud | +0.135 |
Neg logit contributions | |
ime | -0.109 |
ly | -0.108 |
ug | -0.107 |
dd | -0.106 |
ors | -0.103 |
Token | Mia |
Token ID | 32189 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
Flor | +0.022 |
grate | +0.022 |
refreshed | +0.021 |
violet | +0.021 |
promise | +0.020 |
Neg logit contributions | |
dd | -0.034 |
ime | -0.031 |
happened | -0.030 |
ug | -0.030 |
endings | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +6.39e-01 |
Pos logit contributions | |
Flor | +0.064 |
grate | +0.061 |
violet | +0.059 |
refreshed | +0.058 |
cheer | +0.057 |
Neg logit contributions | |
dd | -0.091 |
ime | -0.090 |
ug | -0.084 |
happened | -0.083 |
aw | -0.081 |
Token | but |
Token ID | 475 |
Feature activation | -7.00e-01 |
Pos logit contributions | |
Flor | +0.019 |
grate | +0.018 |
violet | +0.017 |
Av | +0.017 |
refreshed | +0.017 |
Neg logit contributions | |
dd | -0.024 |
ime | -0.022 |
happened | -0.021 |
ug | -0.021 |
emergency | -0.020 |
Token | Mia |
Token ID | 32189 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
dd | +0.027 |
happened | +0.025 |
ime | +0.024 |
ug | +0.023 |
aw | +0.022 |
Neg logit contributions | |
Flor | -0.020 |
grate | -0.020 |
violet | -0.019 |
applaud | -0.019 |
cheer | -0.018 |
Token | You |
Token ID | 921 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
dd | +0.008 |
ime | +0.007 |
happened | +0.007 |
endings | +0.007 |
ug | +0.007 |
Neg logit contributions | |
grate | -0.005 |
Flor | -0.005 |
cheer | -0.004 |
refreshed | -0.004 |
violet | -0.004 |
Token | can |
Token ID | 460 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
dd | +0.009 |
ime | +0.008 |
endings | +0.008 |
ug | +0.008 |
happened | +0.008 |
Neg logit contributions | |
Flor | -0.004 |
grate | -0.003 |
cheer | -0.003 |
| -0.003 |
promise | -0.003 |
Token | 't |
Token ID | 470 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
Flor | +0.009 |
violet | +0.008 |
grate | +0.008 |
refreshed | +0.008 |
wider | +0.008 |
Neg logit contributions | |
dd | -0.011 |
ime | -0.010 |
happened | -0.010 |
ug | -0.009 |
endings | -0.009 |
Token | replace |
Token ID | 6330 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
Flor | +0.001 |
grate | +0.001 |
violet | +0.001 |
| +0.001 |
cheer | +0.001 |
Neg logit contributions | |
dd | -0.003 |
ime | -0.002 |
ug | -0.002 |
happened | -0.002 |
endings | -0.002 |
Token | my |
Token ID | 616 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
grate | +0.051 |
Flor | +0.051 |
violet | +0.048 |
refreshed | +0.047 |
cheer | +0.047 |
Neg logit contributions | |
dd | -0.071 |
ime | -0.064 |
happened | -0.062 |
endings | -0.061 |
ug | -0.061 |
Token | doll |
Token ID | 3654 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
grate | +0.047 |
Flor | +0.046 |
refreshed | +0.044 |
violet | +0.044 |
| +0.044 |
Neg logit contributions | |
dd | -0.042 |
ime | -0.041 |
happened | -0.040 |
emergency | -0.039 |
endings | -0.039 |
Token | with |
Token ID | 351 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
refreshed | +0.142 |
violet | +0.135 |
Flor | +0.134 |
applaud | +0.128 |
cheer | +0.127 |
Neg logit contributions | |
ors | -0.125 |
aw | -0.119 |
ly | -0.118 |
dd | -0.118 |
ime | -0.117 |
Token | that |
Token ID | 326 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
Flor | +0.042 |
grate | +0.039 |
| +0.037 |
violet | +0.037 |
cheer | +0.037 |
Neg logit contributions | |
dd | -0.061 |
ime | -0.056 |
ug | -0.053 |
happened | -0.053 |
endings | -0.053 |
Token | one |
Token ID | 530 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
grate | +0.032 |
Flor | +0.031 |
violet | +0.031 |
refreshed | +0.029 |
cheer | +0.028 |
Neg logit contributions | |
dd | -0.040 |
happened | -0.039 |
endings | -0.037 |
ime | -0.036 |
emergency | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
Flor | +0.030 |
grate | +0.029 |
violet | +0.028 |
refreshed | +0.027 |
cheer | +0.027 |
Neg logit contributions | |
dd | -0.041 |
ime | -0.040 |
happened | -0.039 |
endings | -0.037 |
ug | -0.037 |
Token | Give |
Token ID | 13786 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
dd | +0.013 |
endings | +0.012 |
ug | +0.011 |
ime | +0.011 |
happened | +0.011 |
Neg logit contributions | |
refreshed | -0.014 |
grate | -0.013 |
Flor | -0.013 |
cheer | -0.013 |
applaud | -0.013 |
Token | I |
Token ID | 314 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
Flor | +0.010 |
grate | +0.010 |
| +0.009 |
cheer | +0.009 |
refreshed | +0.009 |
Neg logit contributions | |
dd | -0.017 |
ime | -0.016 |
happened | -0.016 |
ug | -0.015 |
endings | -0.015 |
Token | want |
Token ID | 765 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
Flor | +0.042 |
grate | +0.039 |
violet | +0.037 |
cheer | +0.035 |
| +0.034 |
Neg logit contributions | |
dd | -0.066 |
ime | -0.060 |
happened | -0.060 |
ug | -0.056 |
endings | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
dd | +0.007 |
ime | +0.006 |
happened | +0.006 |
endings | +0.006 |
ug | +0.006 |
Neg logit contributions | |
Flor | -0.006 |
grate | -0.006 |
violet | -0.006 |
cheer | -0.006 |
refreshed | -0.006 |
Token | pull |
Token ID | 2834 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
Flor | +0.019 |
refreshed | +0.018 |
grate | +0.018 |
violet | +0.018 |
Alexa | +0.017 |
Neg logit contributions | |
ime | -0.028 |
dd | -0.028 |
happened | -0.027 |
ug | -0.026 |
endings | -0.025 |
Token | your |
Token ID | 534 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Flor | +0.020 |
grate | +0.019 |
| +0.018 |
cheer | +0.018 |
violet | +0.017 |
Neg logit contributions | |
dd | -0.026 |
ime | -0.024 |
happened | -0.023 |
endings | -0.023 |
ug | -0.023 |
Token | doll |
Token ID | 3654 |
Feature activation | +4.10e+00 |
Pos logit contributions | |
Flor | +0.036 |
grate | +0.035 |
refreshed | +0.035 |
violet | +0.033 |
Av | +0.033 |
Neg logit contributions | |
dd | -0.036 |
ime | -0.035 |
happened | -0.034 |
emergency | -0.034 |
ug | -0.033 |
Token | with |
Token ID | 351 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
Flor | +0.131 |
violet | +0.128 |
refreshed | +0.127 |
grate | +0.120 |
| +0.115 |
Neg logit contributions | |
dd | -0.138 |
ime | -0.134 |
ug | -0.132 |
ors | -0.128 |
aw | -0.127 |
Token | my |
Token ID | 616 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
Flor | +0.052 |
grate | +0.049 |
violet | +0.046 |
| +0.045 |
cheer | +0.045 |
Neg logit contributions | |
dd | -0.077 |
ime | -0.071 |
ug | -0.067 |
happened | -0.067 |
endings | -0.066 |
Token | rope |
Token ID | 17182 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
Flor | +0.022 |
grate | +0.021 |
refreshed | +0.021 |
violet | +0.021 |
Av | +0.020 |
Neg logit contributions | |
ime | -0.022 |
dd | -0.022 |
happened | -0.021 |
emergency | -0.021 |
ug | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -2.20e-03 |
Pos logit contributions | |
dd | +0.019 |
ime | +0.018 |
ug | +0.017 |
happened | +0.017 |
endings | +0.016 |
Neg logit contributions | |
Flor | -0.016 |
violet | -0.015 |
grate | -0.015 |
refreshed | -0.014 |
cheer | -0.014 |
Token | It |
Token ID | 632 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
dd | +0.000 |
ug | +0.000 |
ime | +0.000 |
aw | +0.000 |
endings | +0.000 |
Neg logit contributions | |
Flor | -0.000 |
refreshed | -0.000 |
grate | -0.000 |
cheer | -0.000 |
happier | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
dd | +0.036 |
ime | +0.033 |
happened | +0.031 |
ug | +0.031 |
endings | +0.031 |
Neg logit contributions | |
Flor | -0.026 |
grate | -0.024 |
violet | -0.023 |
cheer | -0.023 |
| -0.023 |
Token | Daddy |
Token ID | 26926 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
dd | +0.014 |
ime | +0.012 |
happened | +0.012 |
ug | +0.012 |
endings | +0.011 |
Neg logit contributions | |
Flor | -0.010 |
grate | -0.009 |
violet | -0.009 |
cheer | -0.009 |
| -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
dd | +0.060 |
ime | +0.056 |
ug | +0.053 |
happened | +0.053 |
endings | +0.052 |
Neg logit contributions | |
Flor | -0.036 |
grate | -0.034 |
| -0.032 |
violet | -0.032 |
cheer | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
dd | +0.021 |
ime | +0.019 |
happened | +0.018 |
ug | +0.018 |
endings | +0.018 |
Neg logit contributions | |
Flor | -0.014 |
grate | -0.014 |
violet | -0.013 |
cheer | -0.013 |
| -0.013 |
Token | three |
Token ID | 1115 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
dd | +0.076 |
ime | +0.071 |
ug | +0.067 |
endings | +0.067 |
happened | +0.067 |
Neg logit contributions | |
Flor | -0.050 |
grate | -0.047 |
| -0.044 |
cheer | -0.044 |
violet | -0.044 |
Token | - |
Token ID | 12 |
Feature activation | -6.30e+00 |
Pos logit contributions | |
dd | +0.079 |
ime | +0.070 |
endings | +0.069 |
happened | +0.068 |
ug | +0.067 |
Neg logit contributions | |
grate | -0.065 |
Flor | -0.065 |
| -0.064 |
cheer | -0.059 |
violet | -0.059 |
Token | year |
Token ID | 1941 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
dd | +0.236 |
ime | +0.223 |
endings | +0.220 |
happened | +0.215 |
ug | +0.204 |
Neg logit contributions | |
Flor | -0.196 |
grate | -0.185 |
cheer | -0.175 |
| -0.174 |
violet | -0.171 |
Token | - |
Token ID | 12 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
dd | +0.093 |
happened | +0.089 |
endings | +0.087 |
ime | +0.087 |
Stop | +0.082 |
Neg logit contributions | |
grate | -0.142 |
Flor | -0.135 |
cheer | -0.125 |
Av | -0.121 |
violet | -0.120 |
Token | old |
Token ID | 727 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
dd | +0.152 |
ime | +0.141 |
endings | +0.135 |
happened | +0.134 |
ug | +0.125 |
Neg logit contributions | |
Flor | -0.217 |
grate | -0.208 |
cheer | -0.200 |
| -0.197 |
Av | -0.194 |
Token | little |
Token ID | 1310 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
dd | +0.084 |
ime | +0.075 |
ug | +0.071 |
happened | +0.071 |
endings | +0.070 |
Neg logit contributions | |
Flor | -0.075 |
grate | -0.071 |
| -0.067 |
cheer | -0.067 |
violet | -0.066 |
Token | Sus |
Token ID | 8932 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
dd | +0.063 |
ime | +0.055 |
ug | +0.053 |
happened | +0.053 |
endings | +0.052 |
Neg logit contributions | |
Flor | -0.056 |
grate | -0.051 |
cheer | -0.050 |
| -0.049 |
Av | -0.048 |
Token | uncle |
Token ID | 7711 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
dd | +0.077 |
ime | +0.067 |
ug | +0.065 |
happened | +0.065 |
endings | +0.062 |
Neg logit contributions | |
Flor | -0.082 |
grate | -0.077 |
violet | -0.075 |
| -0.074 |
cheer | -0.073 |
Token | ever |
Token ID | 1683 |
Feature activation | -7.63e-01 |
Pos logit contributions | |
dd | +0.105 |
ime | +0.096 |
happened | +0.093 |
ug | +0.091 |
endings | +0.090 |
Neg logit contributions | |
Flor | -0.072 |
grate | -0.069 |
violet | -0.065 |
| -0.064 |
cheer | -0.064 |
Token | ." |
Token ID | 526 |
Feature activation | +6.15e-01 |
Pos logit contributions | |
dd | +0.035 |
happened | +0.032 |
ime | +0.031 |
ug | +0.030 |
endings | +0.029 |
Neg logit contributions | |
Flor | -0.020 |
grate | -0.019 |
violet | -0.018 |
cheer | -0.017 |
| -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
grate | +0.017 |
Flor | +0.016 |
violet | +0.016 |
cheer | +0.016 |
promise | +0.015 |
Neg logit contributions | |
dd | -0.025 |
ime | -0.023 |
happened | -0.023 |
endings | -0.022 |
aw | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
dd | +0.046 |
ime | +0.043 |
ug | +0.041 |
endings | +0.041 |
happened | +0.041 |
Neg logit contributions | |
Flor | -0.019 |
grate | -0.017 |
| -0.016 |
cheer | -0.016 |
violet | -0.015 |
Token | Un |
Token ID | 3118 |
Feature activation | -6.26e+00 |
Pos logit contributions | |
dd | +0.017 |
ime | +0.015 |
happened | +0.014 |
ug | +0.014 |
endings | +0.014 |
Neg logit contributions | |
Flor | -0.021 |
grate | -0.021 |
| -0.020 |
violet | -0.020 |
cheer | -0.020 |
Token | cle |
Token ID | 2375 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
dd | +0.185 |
ime | +0.167 |
endings | +0.153 |
happened | +0.153 |
ug | +0.151 |
Neg logit contributions | |
grate | -0.251 |
Flor | -0.251 |
violet | -0.239 |
| -0.238 |
cheer | -0.237 |
Token | Ben |
Token ID | 3932 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
dd | +0.086 |
ime | +0.078 |
happened | +0.076 |
aw | +0.072 |
ug | +0.071 |
Neg logit contributions | |
Flor | -0.093 |
grate | -0.089 |
violet | -0.086 |
cheer | -0.085 |
| -0.082 |
Token | hugged |
Token ID | 41130 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
dd | +0.030 |
happened | +0.027 |
ime | +0.027 |
aw | +0.026 |
ug | +0.025 |
Neg logit contributions | |
Flor | -0.020 |
violet | -0.018 |
grate | -0.018 |
cheer | -0.017 |
| -0.017 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
Flor | +0.010 |
grate | +0.009 |
| +0.008 |
cheer | +0.008 |
Av | +0.008 |
Neg logit contributions | |
dd | -0.025 |
ime | -0.023 |
ug | -0.022 |
happened | -0.022 |
endings | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
violet | +0.083 |
Flor | +0.082 |
applaud | +0.073 |
Av | +0.072 |
invite | +0.071 |
Neg logit contributions | |
dd | -0.063 |
aw | -0.061 |
ime | -0.061 |
ly | -0.061 |
ug | -0.056 |
Token | welcome |
Token ID | 7062 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
dd | +0.008 |
ime | +0.007 |
endings | +0.007 |
ug | +0.007 |
happened | +0.007 |
Neg logit contributions | |
Flor | -0.005 |
grate | -0.005 |
cheer | -0.005 |
applaud | -0.004 |
| -0.004 |
Token | here |
Token ID | 994 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
dd | +0.028 |
ime | +0.026 |
ug | +0.024 |
happened | +0.024 |
endings | +0.024 |
Neg logit contributions | |
Flor | -0.015 |
grate | -0.014 |
| -0.013 |
cheer | -0.013 |
violet | -0.012 |
Token | ." |
Token ID | 526 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
dd | +0.012 |
ime | +0.011 |
happened | +0.010 |
ug | +0.010 |
endings | +0.010 |
Neg logit contributions | |
Flor | -0.007 |
grate | -0.006 |
violet | -0.006 |
| -0.006 |
cheer | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
Flor | +0.022 |
grate | +0.021 |
violet | +0.019 |
cheer | +0.019 |
| +0.019 |
Neg logit contributions | |
dd | -0.052 |
ime | -0.048 |
ug | -0.046 |
happened | -0.046 |
endings | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
dd | +0.028 |
ime | +0.026 |
ug | +0.026 |
endings | +0.026 |
happened | +0.025 |
Neg logit contributions | |
| -0.008 |
Flor | -0.007 |
grate | -0.006 |
cheer | -0.005 |
refreshed | -0.005 |
Token | Un |
Token ID | 3118 |
Feature activation | -6.01e+00 |
Pos logit contributions | |
dd | +0.005 |
ug | +0.005 |
ime | +0.005 |
happened | +0.005 |
endings | +0.005 |
Neg logit contributions | |
| -0.007 |
grate | -0.006 |
Flor | -0.006 |
cheer | -0.006 |
Ms | -0.006 |
Token | cle |
Token ID | 2375 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
dd | +0.195 |
ime | +0.180 |
endings | +0.174 |
happened | +0.167 |
ug | +0.164 |
Neg logit contributions | |
grate | -0.221 |
| -0.207 |
Flor | -0.204 |
promise | -0.202 |
violet | -0.202 |
Token | Jack |
Token ID | 3619 |
Feature activation | -9.24e-01 |
Pos logit contributions | |
dd | +0.104 |
ime | +0.095 |
ug | +0.091 |
endings | +0.090 |
happened | +0.088 |
Neg logit contributions | |
Flor | -0.076 |
grate | -0.073 |
| -0.069 |
cheer | -0.068 |
violet | -0.067 |
Token | gr |
Token ID | 1036 |
Feature activation | -8.32e-01 |
Pos logit contributions | |
dd | +0.041 |
ime | +0.038 |
happened | +0.036 |
ug | +0.036 |
endings | +0.035 |
Neg logit contributions | |
Flor | -0.024 |
grate | -0.023 |
violet | -0.021 |
cheer | -0.021 |
| -0.021 |
Token | umbled |
Token ID | 11137 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
dd | +0.020 |
happened | +0.018 |
endings | +0.017 |
unnecessary | +0.017 |
missiles | +0.017 |
Neg logit contributions | |
grate | -0.032 |
Together | -0.030 |
| -0.030 |
wider | -0.029 |
sill | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +9.71e-02 |
Pos logit contributions | |
dd | +0.062 |
ime | +0.057 |
ug | +0.054 |
endings | +0.054 |
happened | +0.053 |
Neg logit contributions | |
Flor | -0.031 |
grate | -0.031 |
| -0.029 |
cheer | -0.029 |
violet | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
dd | +0.048 |
ime | +0.044 |
ug | +0.043 |
endings | +0.042 |
happened | +0.042 |
Neg logit contributions | |
Flor | -0.025 |
grate | -0.025 |
| -0.023 |
cheer | -0.023 |
join | -0.021 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
dd | +0.014 |
ime | +0.013 |
ug | +0.012 |
endings | +0.012 |
happened | +0.012 |
Neg logit contributions | |
Flor | -0.009 |
violet | -0.009 |
grate | -0.009 |
cheer | -0.009 |
count | -0.008 |
Token | Jimmy |
Token ID | 40335 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
dd | +0.024 |
ime | +0.022 |
happened | +0.021 |
ug | +0.021 |
endings | +0.020 |
Neg logit contributions | |
Flor | -0.016 |
grate | -0.015 |
violet | -0.014 |
cheer | -0.014 |
| -0.013 |
Token | was |
Token ID | 373 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
dd | +0.008 |
ime | +0.007 |
ug | +0.007 |
endings | +0.007 |
happened | +0.007 |
Neg logit contributions | |
Flor | -0.005 |
grate | -0.005 |
| -0.005 |
cheer | -0.005 |
violet | -0.005 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
dd | +0.057 |
ime | +0.052 |
ug | +0.050 |
happened | +0.049 |
aw | +0.046 |
Neg logit contributions | |
Flor | -0.040 |
grate | -0.038 |
violet | -0.037 |
cheer | -0.035 |
Av | -0.034 |
Token | - |
Token ID | 12 |
Feature activation | -5.97e+00 |
Pos logit contributions | |
dd | +0.056 |
ime | +0.050 |
happened | +0.048 |
endings | +0.047 |
ug | +0.047 |
Neg logit contributions | |
Flor | -0.055 |
grate | -0.054 |
cheer | -0.051 |
| -0.051 |
violet | -0.049 |
Token | years |
Token ID | 19002 |
Feature activation | -4.01e+00 |
Pos logit contributions | |
dd | +0.233 |
ime | +0.213 |
happened | +0.210 |
endings | +0.209 |
ug | +0.200 |
Neg logit contributions | |
Flor | -0.185 |
grate | -0.179 |
cheer | -0.169 |
| -0.166 |
violet | -0.160 |
Token | - |
Token ID | 12 |
Feature activation | -5.31e+00 |
Pos logit contributions | |
dd | +0.107 |
happened | +0.097 |
endings | +0.092 |
ime | +0.086 |
Stop | +0.080 |
Neg logit contributions | |
grate | -0.175 |
Flor | -0.168 |
cheer | -0.155 |
Av | -0.153 |
wider | -0.151 |
Token | old |
Token ID | 727 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
dd | +0.149 |
happened | +0.134 |
endings | +0.132 |
ime | +0.131 |
ug | +0.120 |
Neg logit contributions | |
Flor | -0.219 |
grate | -0.217 |
cheer | -0.206 |
| -0.203 |
Av | -0.200 |
Token | and |
Token ID | 290 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
dd | +0.066 |
ime | +0.061 |
ug | +0.058 |
happened | +0.057 |
endings | +0.057 |
Neg logit contributions | |
Flor | -0.044 |
grate | -0.042 |
violet | -0.040 |
cheer | -0.039 |
| -0.039 |
Token | he |
Token ID | 339 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
dd | +0.059 |
ime | +0.052 |
happened | +0.051 |
ug | +0.048 |
endings | +0.047 |
Neg logit contributions | |
Flor | -0.060 |
grate | -0.056 |
violet | -0.056 |
cheer | -0.053 |
Av | -0.051 |
Token | shrink |
Token ID | 22085 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
dd | +0.025 |
ime | +0.023 |
ug | +0.022 |
happened | +0.022 |
endings | +0.022 |
Neg logit contributions | |
Flor | -0.014 |
grate | -0.013 |
violet | -0.012 |
| -0.012 |
cheer | -0.012 |
Token | me |
Token ID | 502 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
Flor | +0.019 |
grate | +0.017 |
violet | +0.016 |
cheer | +0.016 |
| +0.016 |
Neg logit contributions | |
dd | -0.035 |
ime | -0.032 |
happened | -0.030 |
ug | -0.030 |
endings | -0.030 |
Token | !" |
Token ID | 2474 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
dd | +0.042 |
ime | +0.040 |
ug | +0.038 |
happened | +0.036 |
endings | +0.036 |
Neg logit contributions | |
Flor | -0.027 |
grate | -0.026 |
violet | -0.025 |
| -0.025 |
cheer | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
Flor | +0.012 |
grate | +0.012 |
violet | +0.011 |
cheer | +0.011 |
| +0.011 |
Neg logit contributions | |
dd | -0.023 |
ime | -0.021 |
ug | -0.020 |
happened | -0.020 |
endings | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
dd | +0.025 |
ime | +0.023 |
endings | +0.022 |
ug | +0.022 |
happened | +0.022 |
Neg logit contributions | |
Flor | -0.008 |
| -0.007 |
grate | -0.007 |
cheer | -0.006 |
violet | -0.006 |
Token | Un |
Token ID | 3118 |
Feature activation | -5.91e+00 |
Pos logit contributions | |
dd | +0.003 |
ime | +0.003 |
happened | +0.003 |
ug | +0.003 |
endings | +0.002 |
Neg logit contributions | |
Flor | -0.005 |
grate | -0.005 |
| -0.005 |
cheer | -0.005 |
violet | -0.005 |
Token | cle |
Token ID | 2375 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
dd | +0.181 |
ime | +0.165 |
endings | +0.154 |
happened | +0.154 |
ug | +0.148 |
Neg logit contributions | |
grate | -0.230 |
Flor | -0.226 |
violet | -0.218 |
| -0.216 |
cheer | -0.215 |
Token | Tom |
Token ID | 4186 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
dd | +0.090 |
ime | +0.083 |
ug | +0.078 |
happened | +0.078 |
endings | +0.077 |
Neg logit contributions | |
Flor | -0.068 |
grate | -0.064 |
violet | -0.061 |
| -0.061 |
cheer | -0.060 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
dd | +0.032 |
ime | +0.030 |
happened | +0.029 |
ug | +0.028 |
endings | +0.028 |
Neg logit contributions | |
Flor | -0.019 |
grate | -0.018 |
violet | -0.017 |
| -0.017 |
cheer | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
Flor | +0.048 |
violet | +0.044 |
grate | +0.042 |
cheer | +0.042 |
| +0.042 |
Neg logit contributions | |
dd | -0.045 |
ime | -0.041 |
happened | -0.039 |
aw | -0.038 |
ug | -0.038 |
Token | put |
Token ID | 1234 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
dd | +0.018 |
happened | +0.016 |
aw | +0.015 |
ime | +0.015 |
ug | +0.015 |
Neg logit contributions | |
Flor | -0.012 |
Deal | -0.011 |
promise | -0.011 |
grate | -0.011 |
| -0.011 |
Token | ement |
Token ID | 972 |
Feature activation | -7.54e-01 |
Pos logit contributions | |
dd | +0.082 |
endings | +0.075 |
oyd | +0.075 |
opter | +0.074 |
ug | +0.073 |
Neg logit contributions | |
grate | -0.043 |
wider | -0.036 |
Flor | -0.035 |
too | -0.033 |
cheer | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
dd | +0.038 |
ime | +0.035 |
endings | +0.034 |
happened | +0.033 |
ug | +0.033 |
Neg logit contributions | |
Flor | -0.016 |
grate | -0.015 |
| -0.013 |
cheer | -0.013 |
Av | -0.012 |
Token | |
Token ID | 220 |
Feature activation | -7.86e-01 |
Pos logit contributions | |
dd | +0.020 |
ime | +0.018 |
happened | +0.018 |
ug | +0.017 |
endings | +0.017 |
Neg logit contributions | |
Flor | -0.016 |
grate | -0.015 |
cheer | -0.015 |
promise | -0.015 |
violet | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
dd | +0.041 |
ime | +0.038 |
endings | +0.036 |
ug | +0.036 |
happened | +0.036 |
Neg logit contributions | |
Flor | -0.015 |
| -0.014 |
grate | -0.014 |
cheer | -0.012 |
violet | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
dd | +0.021 |
ime | +0.019 |
ug | +0.018 |
happened | +0.018 |
endings | +0.018 |
Neg logit contributions | |
Flor | -0.011 |
grate | -0.010 |
violet | -0.009 |
cheer | -0.009 |
Av | -0.009 |
Token | Un |
Token ID | 3118 |
Feature activation | -5.89e+00 |
Pos logit contributions | |
dd | +0.002 |
ime | +0.002 |
happened | +0.002 |
endings | +0.002 |
ug | +0.002 |
Neg logit contributions | |
Flor | -0.002 |
grate | -0.002 |
| -0.002 |
cheer | -0.002 |
violet | -0.002 |
Token | able |
Token ID | 540 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
dd | +0.191 |
ime | +0.178 |
happened | +0.176 |
endings | +0.175 |
ug | +0.163 |
Neg logit contributions | |
grate | -0.211 |
Flor | -0.205 |
cheer | -0.200 |
violet | -0.199 |
| -0.197 |
Token | to |
Token ID | 284 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
dd | +0.046 |
endings | +0.041 |
ime | +0.041 |
happened | +0.039 |
ug | +0.039 |
Neg logit contributions | |
Flor | -0.035 |
grate | -0.033 |
cheer | -0.031 |
Av | -0.031 |
| -0.030 |
Token | believe |
Token ID | 1975 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
Flor | +0.016 |
grate | +0.015 |
violet | +0.014 |
cheer | +0.014 |
| +0.014 |
Neg logit contributions | |
dd | -0.016 |
ime | -0.015 |
happened | -0.014 |
ug | -0.014 |
endings | -0.013 |
Token | his |
Token ID | 465 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
grate | +0.023 |
Flor | +0.023 |
cheer | +0.020 |
closer | +0.019 |
| +0.019 |
Neg logit contributions | |
ime | -0.033 |
dd | -0.033 |
endings | -0.032 |
opter | -0.030 |
ruck | -0.029 |
Token | luck |
Token ID | 8458 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
dd | +0.048 |
ime | +0.043 |
endings | +0.041 |
ug | +0.041 |
happened | +0.040 |
Neg logit contributions | |
Flor | -0.033 |
grate | -0.032 |
cheer | -0.029 |
violet | -0.029 |
| -0.029 |
Token | great |
Token ID | 1049 |
Feature activation | -7.26e-01 |
Pos logit contributions | |
dd | +0.021 |
happened | +0.019 |
ime | +0.018 |
emergency | +0.018 |
endings | +0.017 |
Neg logit contributions | |
Flor | -0.020 |
grate | -0.019 |
| -0.018 |
Together | -0.018 |
violet | -0.017 |
Token | cook |
Token ID | 4255 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
dd | +0.027 |
happened | +0.025 |
ime | +0.025 |
endings | +0.023 |
emergency | +0.022 |
Neg logit contributions | |
Flor | -0.024 |
grate | -0.022 |
| -0.021 |
violet | -0.021 |
refreshed | -0.020 |
Token | ." |
Token ID | 526 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
Flor | +0.010 |
grate | +0.009 |
violet | +0.009 |
| +0.009 |
cheer | +0.009 |
Neg logit contributions | |
dd | -0.014 |
ime | -0.013 |
happened | -0.013 |
aw | -0.012 |
ug | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
Flor | +0.027 |
grate | +0.026 |
cheer | +0.024 |
violet | +0.024 |
| +0.023 |
Neg logit contributions | |
dd | -0.043 |
ime | -0.040 |
happened | -0.038 |
ug | -0.038 |
endings | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
dd | +0.028 |
ime | +0.027 |
endings | +0.026 |
happened | +0.025 |
ug | +0.025 |
Neg logit contributions | |
Flor | -0.011 |
| -0.010 |
grate | -0.010 |
cheer | -0.009 |
violet | -0.009 |
Token | Un |
Token ID | 3118 |
Feature activation | -5.89e+00 |
Pos logit contributions | |
dd | +0.007 |
ime | +0.007 |
endings | +0.007 |
happened | +0.007 |
ug | +0.007 |
Neg logit contributions | |
grate | -0.007 |
Flor | -0.007 |
| -0.007 |
violet | -0.007 |
cheer | -0.007 |
Token | cle |
Token ID | 2375 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
dd | +0.173 |
ime | +0.155 |
endings | +0.147 |
happened | +0.144 |
ug | +0.142 |
Neg logit contributions | |
grate | -0.235 |
Flor | -0.234 |
violet | -0.225 |
| -0.224 |
cheer | -0.222 |
Token | Tom |
Token ID | 4186 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
dd | +0.079 |
ime | +0.072 |
happened | +0.069 |
ug | +0.066 |
endings | +0.066 |
Neg logit contributions | |
Flor | -0.076 |
grate | -0.073 |
cheer | -0.070 |
violet | -0.069 |
| -0.069 |
Token | took |
Token ID | 1718 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
dd | +0.020 |
happened | +0.018 |
ime | +0.018 |
aw | +0.017 |
ug | +0.016 |
Neg logit contributions | |
Flor | -0.019 |
grate | -0.018 |
violet | -0.017 |
cheer | -0.017 |
| -0.017 |
Token | a |
Token ID | 257 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
Flor | +0.001 |
grate | +0.001 |
| +0.001 |
cheer | +0.001 |
violet | +0.001 |
Neg logit contributions | |
dd | -0.002 |
ime | -0.002 |
endings | -0.002 |
ug | -0.002 |
happened | -0.002 |
Token | bite |
Token ID | 13197 |
Feature activation | +8.48e-01 |
Pos logit contributions | |
Flor | +0.030 |
grate | +0.028 |
violet | +0.025 |
cheer | +0.025 |
| +0.025 |
Neg logit contributions | |
dd | -0.052 |
ime | -0.049 |
happened | -0.047 |
ug | -0.045 |
endings | -0.044 |
Token | down |
Token ID | 866 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
Flor | +0.033 |
Deal | +0.032 |
promise | +0.031 |
violet | +0.031 |
grate | +0.030 |
Neg logit contributions | |
dd | -0.037 |
happened | -0.034 |
ime | -0.032 |
ug | -0.029 |
aw | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
dd | +0.096 |
ime | +0.089 |
happened | +0.085 |
ug | +0.084 |
endings | +0.083 |
Neg logit contributions | |
Flor | -0.062 |
grate | -0.059 |
violet | -0.056 |
| -0.055 |
cheer | -0.055 |
Token | |
Token ID | 220 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
dd | +0.032 |
ime | +0.030 |
happened | +0.029 |
endings | +0.028 |
ug | +0.028 |
Neg logit contributions | |
Flor | -0.019 |
grate | -0.018 |
cheer | -0.017 |
promise | -0.016 |
violet | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
dd | +0.045 |
ime | +0.041 |
endings | +0.040 |
ug | +0.040 |
happened | +0.040 |
Neg logit contributions | |
Flor | -0.017 |
grate | -0.016 |
| -0.015 |
cheer | -0.014 |
violet | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -7.23e-02 |
Pos logit contributions | |
dd | +0.023 |
ime | +0.021 |
ug | +0.020 |
happened | +0.020 |
endings | +0.020 |
Neg logit contributions | |
Flor | -0.010 |
grate | -0.010 |
| -0.009 |
violet | -0.009 |
cheer | -0.009 |
Token | Un |
Token ID | 3118 |
Feature activation | -5.86e+00 |
Pos logit contributions | |
dd | +0.003 |
ime | +0.003 |
happened | +0.002 |
endings | +0.002 |
ug | +0.002 |
Neg logit contributions | |
Flor | -0.002 |
grate | -0.002 |
| -0.002 |
violet | -0.002 |
cheer | -0.002 |
Token | cle |
Token ID | 2375 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
dd | +0.183 |
ime | +0.165 |
happened | +0.161 |
endings | +0.161 |
ug | +0.154 |
Neg logit contributions | |
grate | -0.218 |
violet | -0.210 |
Flor | -0.208 |
cheer | -0.205 |
| -0.204 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
dd | +0.103 |
ime | +0.095 |
happened | +0.090 |
ug | +0.090 |
endings | +0.089 |
Neg logit contributions | |
Flor | -0.062 |
grate | -0.058 |
violet | -0.054 |
cheer | -0.054 |
| -0.054 |
Token | at |
Token ID | 379 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
Flor | +0.039 |
violet | +0.038 |
grate | +0.036 |
count | +0.035 |
cheer | +0.035 |
Neg logit contributions | |
dd | -0.037 |
ime | -0.034 |
happened | -0.032 |
aw | -0.032 |
ug | -0.031 |
Token | her |
Token ID | 607 |
Feature activation | +7.99e-01 |
Pos logit contributions | |
Flor | +0.013 |
grate | +0.011 |
| +0.011 |
cheer | +0.011 |
Av | +0.010 |
Neg logit contributions | |
dd | -0.027 |
ime | -0.025 |
endings | -0.024 |
ug | -0.024 |
happened | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
Flor | +0.024 |
grate | +0.022 |
violet | +0.022 |
| +0.021 |
Av | +0.020 |
Neg logit contributions | |
dd | -0.031 |
ime | -0.030 |
happened | -0.028 |
ug | -0.027 |
endings | -0.027 |
Token | than |
Token ID | 621 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
dd | +0.045 |
ime | +0.040 |
happened | +0.038 |
ug | +0.038 |
endings | +0.038 |
Neg logit contributions | |
Flor | -0.040 |
grate | -0.038 |
| -0.037 |
violet | -0.036 |
cheer | -0.036 |
Token | Uncle |
Token ID | 23169 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
dd | +0.143 |
ime | +0.130 |
ug | +0.124 |
endings | +0.123 |
happened | +0.122 |
Neg logit contributions | |
Flor | -0.102 |
grate | -0.097 |
| -0.091 |
cheer | -0.090 |
violet | -0.089 |
Token | ! |
Token ID | 0 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
dd | +0.100 |
ime | +0.092 |
ug | +0.087 |
happened | +0.086 |
endings | +0.086 |
Neg logit contributions | |
Flor | -0.074 |
grate | -0.070 |
| -0.066 |
violet | -0.066 |
cheer | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -4.66e-01 |
Pos logit contributions | |
dd | +0.064 |
ime | +0.059 |
ug | +0.056 |
happened | +0.056 |
endings | +0.055 |
Neg logit contributions | |
Flor | -0.042 |
grate | -0.040 |
violet | -0.037 |
cheer | -0.037 |
| -0.037 |
Token |
|
Token ID | 198 |
Feature activation | -8.95e-02 |
Pos logit contributions | |
dd | +0.025 |
ime | +0.023 |
endings | +0.023 |
happened | +0.023 |
ug | +0.022 |
Neg logit contributions | |
| -0.008 |
Flor | -0.008 |
grate | -0.007 |
cheer | -0.007 |
violet | -0.006 |
Token | Un |
Token ID | 3118 |
Feature activation | -5.84e+00 |
Pos logit contributions | |
dd | +0.003 |
ime | +0.003 |
happened | +0.003 |
endings | +0.003 |
ug | +0.003 |
Neg logit contributions | |
| -0.003 |
grate | -0.003 |
Flor | -0.003 |
violet | -0.003 |
cheer | -0.003 |
Token | cle |
Token ID | 2375 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
dd | +0.180 |
ime | +0.164 |
endings | +0.159 |
happened | +0.157 |
ug | +0.156 |
Neg logit contributions | |
grate | -0.216 |
violet | -0.212 |
Flor | -0.210 |
| -0.205 |
cheer | -0.202 |
Token | was |
Token ID | 373 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
dd | +0.094 |
ime | +0.086 |
ug | +0.082 |
endings | +0.081 |
happened | +0.081 |
Neg logit contributions | |
Flor | -0.061 |
grate | -0.057 |
| -0.054 |
violet | -0.053 |
cheer | -0.053 |
Token | so |
Token ID | 523 |
Feature activation | +6.52e-01 |
Pos logit contributions | |
dd | +0.012 |
ime | +0.012 |
happened | +0.011 |
aw | +0.011 |
ug | +0.010 |
Neg logit contributions | |
Flor | -0.007 |
grate | -0.007 |
violet | -0.007 |
promise | -0.006 |
kiss | -0.006 |
Token | happy |
Token ID | 3772 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
Flor | +0.018 |
violet | +0.016 |
grate | +0.016 |
Sparkle | +0.015 |
Ruby | +0.015 |
Neg logit contributions | |
dd | -0.028 |
ime | -0.026 |
happened | -0.024 |
ug | -0.023 |
aw | -0.022 |
Token | he |
Token ID | 339 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
Flor | +0.020 |
grate | +0.019 |
| +0.018 |
cheer | +0.018 |
violet | +0.017 |
Neg logit contributions | |
dd | -0.028 |
ime | -0.026 |
ug | -0.024 |
happened | -0.024 |
endings | -0.024 |
Token | Jane |
Token ID | 12091 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
grate | +0.049 |
cheer | +0.042 |
Flor | +0.041 |
refreshed | +0.041 |
wiser | +0.041 |
Neg logit contributions | |
dd | -0.050 |
happened | -0.048 |
ime | -0.048 |
ug | -0.044 |
emergency | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -5.17e-01 |
Pos logit contributions | |
Flor | +0.010 |
grate | +0.009 |
| +0.009 |
cheer | +0.008 |
violet | +0.008 |
Neg logit contributions | |
dd | -0.024 |
ime | -0.021 |
endings | -0.021 |
ug | -0.021 |
happened | -0.020 |
Token | She |
Token ID | 1375 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
ime | +0.018 |
happened | +0.018 |
dd | +0.018 |
endings | +0.017 |
aw | +0.016 |
Neg logit contributions | |
grate | -0.016 |
Flor | -0.016 |
promise | -0.015 |
cheer | -0.014 |
violet | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
dd | +0.006 |
ug | +0.006 |
ime | +0.006 |
endings | +0.005 |
anked | +0.005 |
Neg logit contributions | |
| -0.003 |
Flor | -0.003 |
grate | -0.003 |
violet | -0.003 |
also | -0.003 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
dd | +0.045 |
ime | +0.042 |
ug | +0.040 |
happened | +0.040 |
emergency | +0.038 |
Neg logit contributions | |
grate | -0.027 |
Flor | -0.026 |
violet | -0.025 |
cheer | -0.024 |
| -0.024 |
Token | - |
Token ID | 12 |
Feature activation | -5.82e+00 |
Pos logit contributions | |
dd | +0.060 |
ime | +0.053 |
happened | +0.052 |
endings | +0.051 |
ug | +0.050 |
Neg logit contributions | |
grate | -0.052 |
Flor | -0.052 |
| -0.050 |
cheer | -0.049 |
violet | -0.046 |
Token | years |
Token ID | 19002 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
dd | +0.218 |
ime | +0.199 |
happened | +0.190 |
endings | +0.189 |
ug | +0.186 |
Neg logit contributions | |
Flor | -0.193 |
grate | -0.184 |
cheer | -0.174 |
violet | -0.173 |
| -0.173 |
Token | - |
Token ID | 12 |
Feature activation | -4.91e+00 |
Pos logit contributions | |
dd | +0.105 |
happened | +0.091 |
endings | +0.088 |
ime | +0.084 |
Stop | +0.078 |
Neg logit contributions | |
grate | -0.164 |
Flor | -0.159 |
cheer | -0.146 |
promise | -0.144 |
Av | -0.144 |
Token | old |
Token ID | 727 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
dd | +0.125 |
ime | +0.110 |
happened | +0.105 |
endings | +0.105 |
ug | +0.099 |
Neg logit contributions | |
Flor | -0.221 |
grate | -0.214 |
cheer | -0.205 |
| -0.203 |
violet | -0.202 |
Token | . |
Token ID | 13 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
dd | +0.047 |
ime | +0.043 |
happened | +0.040 |
ug | +0.040 |
endings | +0.040 |
Neg logit contributions | |
Flor | -0.035 |
grate | -0.033 |
violet | -0.032 |
cheer | -0.031 |
| -0.031 |
Token | Every |
Token ID | 3887 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
dd | +0.006 |
ime | +0.006 |
happened | +0.006 |
endings | +0.005 |
aw | +0.005 |
Neg logit contributions | |
grate | -0.004 |
Flor | -0.004 |
cheer | -0.004 |
promise | -0.004 |
violet | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
Flor | +0.017 |
grate | +0.016 |
| +0.015 |
cheer | +0.015 |
violet | +0.015 |
Neg logit contributions | |
dd | -0.024 |
ime | -0.022 |
ug | -0.021 |
happened | -0.020 |
endings | -0.020 |
Token | cake |
Token ID | 12187 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
Flor | +0.005 |
grate | +0.005 |
violet | +0.005 |
| +0.005 |
cheer | +0.005 |
Neg logit contributions | |
dd | -0.008 |
ime | -0.007 |
happened | -0.007 |
ug | -0.007 |
endings | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -6.91e-01 |
Pos logit contributions | |
Flor | +0.026 |
Deal | +0.025 |
violet | +0.025 |
| +0.023 |
Av | +0.023 |
Neg logit contributions | |
dd | -0.025 |
ime | -0.023 |
happened | -0.023 |
aw | -0.023 |
ly | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
dd | +0.030 |
ime | +0.028 |
ug | +0.026 |
happened | +0.026 |
endings | +0.026 |
Neg logit contributions | |
Flor | -0.020 |
| -0.018 |
grate | -0.018 |
violet | -0.017 |
cheer | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
dd | +0.030 |
ime | +0.028 |
endings | +0.028 |
ug | +0.027 |
happened | +0.027 |
Neg logit contributions | |
Flor | -0.010 |
| -0.009 |
grate | -0.009 |
cheer | -0.008 |
violet | -0.007 |
Token | Un |
Token ID | 3118 |
Feature activation | -5.82e+00 |
Pos logit contributions | |
ime | +0.004 |
endings | +0.004 |
dd | +0.004 |
ug | +0.004 |
happened | +0.004 |
Neg logit contributions | |
| -0.007 |
grate | -0.006 |
Flor | -0.006 |
cheer | -0.006 |
Ms | -0.006 |
Token | cle |
Token ID | 2375 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
dd | +0.199 |
ime | +0.187 |
endings | +0.176 |
ug | +0.173 |
happened | +0.172 |
Neg logit contributions | |
grate | -0.199 |
| -0.188 |
Flor | -0.185 |
count | -0.183 |
violet | -0.182 |
Token | Bob |
Token ID | 5811 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
dd | +0.095 |
ime | +0.087 |
ug | +0.084 |
endings | +0.082 |
happened | +0.078 |
Neg logit contributions | |
Flor | -0.076 |
grate | -0.071 |
| -0.069 |
violet | -0.067 |
cheer | -0.067 |
Token | was |
Token ID | 373 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
dd | +0.028 |
ime | +0.026 |
happened | +0.025 |
ug | +0.024 |
endings | +0.024 |
Neg logit contributions | |
Flor | -0.020 |
grate | -0.019 |
violet | -0.018 |
cheer | -0.018 |
| -0.018 |
Token | very |
Token ID | 845 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
dd | +0.017 |
ime | +0.016 |
happened | +0.015 |
aw | +0.015 |
ug | +0.014 |
Neg logit contributions | |
Flor | -0.012 |
grate | -0.011 |
violet | -0.011 |
Av | -0.010 |
promise | -0.010 |
Token | shocked |
Token ID | 11472 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
Flor | +0.007 |
violet | +0.007 |
grate | +0.006 |
| +0.006 |
Av | +0.006 |
Neg logit contributions | |
dd | -0.011 |
ime | -0.010 |
happened | -0.010 |
ug | -0.010 |
aw | -0.009 |
Token | classroom |
Token ID | 15806 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
dd | +0.010 |
ime | +0.008 |
happened | +0.008 |
ug | +0.008 |
endings | +0.007 |
Neg logit contributions | |
Flor | -0.014 |
grate | -0.014 |
| -0.013 |
cheer | -0.013 |
violet | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
dd | +0.038 |
ime | +0.035 |
happened | +0.034 |
ug | +0.033 |
endings | +0.033 |
Neg logit contributions | |
Flor | -0.021 |
grate | -0.020 |
violet | -0.018 |
cheer | -0.018 |
| -0.018 |
Token | Jake |
Token ID | 14757 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
dd | +0.001 |
ime | +0.001 |
ug | +0.001 |
happened | +0.001 |
endings | +0.001 |
Neg logit contributions | |
Flor | -0.001 |
grate | -0.001 |
| -0.001 |
violet | -0.001 |
cheer | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
dd | +0.032 |
endings | +0.029 |
ug | +0.029 |
ime | +0.029 |
anked | +0.027 |
Neg logit contributions | |
| -0.012 |
Flor | -0.012 |
count | -0.011 |
grate | -0.011 |
Av | -0.010 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
dd | +0.063 |
ime | +0.060 |
happened | +0.059 |
ug | +0.058 |
emergency | +0.055 |
Neg logit contributions | |
Flor | -0.031 |
grate | -0.031 |
violet | -0.028 |
cheer | -0.026 |
promise | -0.026 |
Token | - |
Token ID | 12 |
Feature activation | -5.80e+00 |
Pos logit contributions | |
dd | +0.052 |
ime | +0.046 |
happened | +0.045 |
endings | +0.044 |
ug | +0.043 |
Neg logit contributions | |
grate | -0.051 |
Flor | -0.051 |
| -0.048 |
cheer | -0.048 |
violet | -0.045 |
Token | years |
Token ID | 19002 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
dd | +0.221 |
ime | +0.203 |
happened | +0.197 |
endings | +0.196 |
ug | +0.188 |
Neg logit contributions | |
Flor | -0.186 |
grate | -0.179 |
cheer | -0.169 |
| -0.166 |
violet | -0.166 |
Token | - |
Token ID | 12 |
Feature activation | -4.98e+00 |
Pos logit contributions | |
dd | +0.101 |
happened | +0.085 |
endings | +0.083 |
ime | +0.081 |
ug | +0.071 |
Neg logit contributions | |
grate | -0.169 |
Flor | -0.165 |
cheer | -0.153 |
violet | -0.151 |
Av | -0.151 |
Token | old |
Token ID | 727 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
dd | +0.130 |
ime | +0.114 |
happened | +0.110 |
endings | +0.109 |
ug | +0.102 |
Neg logit contributions | |
Flor | -0.220 |
grate | -0.214 |
cheer | -0.205 |
| -0.203 |
violet | -0.201 |
Token | and |
Token ID | 290 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
dd | +0.045 |
ime | +0.042 |
ug | +0.039 |
happened | +0.039 |
endings | +0.039 |
Neg logit contributions | |
Flor | -0.031 |
grate | -0.030 |
violet | -0.028 |
cheer | -0.028 |
| -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
dd | +0.043 |
ime | +0.038 |
happened | +0.037 |
ug | +0.036 |
endings | +0.034 |
Neg logit contributions | |
Flor | -0.048 |
grate | -0.046 |
violet | -0.045 |
cheer | -0.043 |
promise | -0.042 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
Flor | +0.008 |
violet | +0.007 |
grate | +0.007 |
| +0.007 |
cheer | +0.007 |
Neg logit contributions | |
dd | -0.008 |
ime | -0.007 |
endings | -0.007 |
ug | -0.007 |
happened | -0.006 |
Token | my |
Token ID | 1820 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
dd | +0.019 |
ime | +0.017 |
endings | +0.017 |
ug | +0.017 |
happened | +0.016 |
Neg logit contributions | |
Flor | -0.006 |
| -0.006 |
grate | -0.006 |
cheer | -0.005 |
violet | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
endings | +0.054 |
dd | +0.053 |
ug | +0.051 |
ime | +0.050 |
pounds | +0.048 |
Neg logit contributions | |
| -0.021 |
say | -0.020 |
tell | -0.019 |
invite | -0.019 |
cheer | -0.019 |
Token | her |
Token ID | 607 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
dd | +0.064 |
ime | +0.063 |
ug | +0.062 |
endings | +0.060 |
ors | +0.060 |
Neg logit contributions | |
| -0.019 |
cheer | -0.017 |
Flor | -0.017 |
say | -0.017 |
talk | -0.016 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
dd | +0.047 |
ug | +0.041 |
ime | +0.041 |
endings | +0.040 |
ors | +0.040 |
Neg logit contributions | |
Flor | -0.032 |
kiss | -0.031 |
hug | -0.031 |
promise | -0.030 |
| -0.030 |
Token | - |
Token ID | 12 |
Feature activation | -5.76e+00 |
Pos logit contributions | |
dd | +0.049 |
endings | +0.042 |
ime | +0.040 |
ug | +0.040 |
happened | +0.040 |
Neg logit contributions | |
| -0.046 |
grate | -0.043 |
promise | -0.043 |
cheer | -0.043 |
Flor | -0.043 |
Token | year |
Token ID | 1941 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
dd | +0.218 |
ime | +0.199 |
endings | +0.193 |
happened | +0.192 |
ug | +0.184 |
Neg logit contributions | |
Flor | -0.186 |
grate | -0.178 |
cheer | -0.171 |
| -0.170 |
violet | -0.162 |
Token | - |
Token ID | 12 |
Feature activation | -4.84e+00 |
Pos logit contributions | |
dd | +0.080 |
happened | +0.074 |
endings | +0.073 |
ime | +0.072 |
Stop | +0.071 |
Neg logit contributions | |
grate | -0.118 |
Flor | -0.109 |
cheer | -0.105 |
promise | -0.103 |
hug | -0.102 |
Token | old |
Token ID | 727 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
dd | +0.142 |
ime | +0.126 |
endings | +0.121 |
happened | +0.121 |
ug | +0.115 |
Neg logit contributions | |
Flor | -0.192 |
grate | -0.186 |
| -0.183 |
cheer | -0.181 |
Av | -0.174 |
Token | daughter |
Token ID | 4957 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
dd | +0.061 |
ime | +0.053 |
ug | +0.052 |
endings | +0.050 |
anked | +0.049 |
Neg logit contributions | |
Flor | -0.036 |
| -0.033 |
cheer | -0.032 |
Av | -0.032 |
grate | -0.031 |
Token | were |
Token ID | 547 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
dd | +0.015 |
ime | +0.013 |
ug | +0.013 |
endings | +0.013 |
ors | +0.012 |
Neg logit contributions | |
Flor | -0.009 |
| -0.008 |
grate | -0.008 |
cheer | -0.008 |
Av | -0.007 |
Token | Summer |
Token ID | 10216 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
grate | +0.042 |
cheer | +0.036 |
refreshed | +0.035 |
wiser | +0.035 |
wider | +0.035 |
Neg logit contributions | |
dd | -0.048 |
ime | -0.046 |
happened | -0.045 |
ug | -0.043 |
endings | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
dd | +0.001 |
ime | +0.001 |
ug | +0.001 |
happened | +0.001 |
endings | +0.001 |
Neg logit contributions | |
Flor | -0.000 |
grate | -0.000 |
violet | -0.000 |
cheer | -0.000 |
| -0.000 |
Token | Summer |
Token ID | 10216 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
dd | +0.018 |
ime | +0.017 |
happened | +0.016 |
ug | +0.016 |
endings | +0.016 |
Neg logit contributions | |
grate | -0.012 |
Flor | -0.012 |
cheer | -0.011 |
promise | -0.010 |
Av | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | -9.46e-01 |
Pos logit contributions | |
Flor | +0.000 |
| +0.000 |
grate | +0.000 |
violet | +0.000 |
cheer | +0.000 |
Neg logit contributions | |
dd | -0.001 |
ime | -0.001 |
ug | -0.001 |
endings | -0.001 |
emergency | -0.001 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
dd | +0.041 |
ime | +0.039 |
happened | +0.037 |
ug | +0.037 |
endings | +0.035 |
Neg logit contributions | |
Flor | -0.025 |
grate | -0.024 |
violet | -0.022 |
cheer | -0.022 |
| -0.022 |
Token | - |
Token ID | 12 |
Feature activation | -5.71e+00 |
Pos logit contributions | |
dd | +0.058 |
ime | +0.051 |
happened | +0.051 |
endings | +0.050 |
ug | +0.048 |
Neg logit contributions | |
Flor | -0.057 |
grate | -0.056 |
| -0.054 |
cheer | -0.053 |
violet | -0.051 |
Token | years |
Token ID | 19002 |
Feature activation | -3.77e+00 |
Pos logit contributions | |
dd | +0.202 |
ime | +0.183 |
happened | +0.173 |
endings | +0.172 |
ug | +0.170 |
Neg logit contributions | |
Flor | -0.202 |
grate | -0.193 |
cheer | -0.184 |
violet | -0.183 |
| -0.183 |
Token | - |
Token ID | 12 |
Feature activation | -4.87e+00 |
Pos logit contributions | |
dd | +0.100 |
happened | +0.086 |
endings | +0.081 |
ime | +0.080 |
Stop | +0.070 |
Neg logit contributions | |
grate | -0.167 |
Flor | -0.163 |
cheer | -0.151 |
violet | -0.149 |
Av | -0.148 |
Token | old |
Token ID | 727 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
dd | +0.118 |
ime | +0.102 |
happened | +0.096 |
endings | +0.095 |
ug | +0.091 |
Neg logit contributions | |
Flor | -0.226 |
grate | -0.218 |
cheer | -0.210 |
| -0.209 |
violet | -0.209 |
Token | and |
Token ID | 290 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
dd | +0.046 |
ime | +0.043 |
ug | +0.041 |
aw | +0.040 |
happened | +0.040 |
Neg logit contributions | |
Flor | -0.038 |
grate | -0.036 |
violet | -0.036 |
refreshed | -0.035 |
cheer | -0.035 |
Token | she |
Token ID | 673 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
dd | +0.029 |
happened | +0.024 |
ime | +0.024 |
ug | +0.022 |
endings | +0.021 |
Neg logit contributions | |
Flor | -0.042 |
grate | -0.040 |
violet | -0.039 |
cheer | -0.038 |
promise | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
Flor | +0.008 |
grate | +0.007 |
| +0.007 |
cheer | +0.007 |
better | +0.007 |
Neg logit contributions | |
dd | -0.015 |
happened | -0.013 |
ime | -0.013 |
endings | -0.013 |
ug | -0.013 |
Token | coat |
Token ID | 13209 |
Feature activation | -9.21e-01 |
Pos logit contributions | |
Flor | +0.002 |
grate | +0.002 |
| +0.002 |
cheer | +0.002 |
violet | +0.002 |
Neg logit contributions | |
dd | -0.003 |
ime | -0.002 |
ug | -0.002 |
happened | -0.002 |
endings | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
dd | +0.041 |
ime | +0.038 |
ug | +0.036 |
happened | +0.036 |
endings | +0.035 |
Neg logit contributions | |
Flor | -0.024 |
grate | -0.022 |
violet | -0.022 |
cheer | -0.021 |
| -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
dd | +0.007 |
ime | +0.006 |
ug | +0.006 |
happened | +0.006 |
endings | +0.006 |
Neg logit contributions | |
Flor | -0.003 |
| -0.003 |
grate | -0.003 |
violet | -0.002 |
cheer | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
dd | +0.015 |
ime | +0.015 |
endings | +0.014 |
happened | +0.014 |
ug | +0.014 |
Neg logit contributions | |
| -0.005 |
Flor | -0.005 |
grate | -0.005 |
cheer | -0.004 |
violet | -0.004 |
Token | Un |
Token ID | 3118 |
Feature activation | -5.69e+00 |
Pos logit contributions | |
| +0.001 |
grate | +0.001 |
Flor | +0.001 |
cheer | +0.001 |
violet | +0.001 |
Neg logit contributions | |
dd | -0.002 |
ime | -0.001 |
endings | -0.001 |
happened | -0.001 |
ug | -0.001 |
Token | cle |
Token ID | 2375 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
dd | +0.184 |
ime | +0.172 |
endings | +0.172 |
happened | +0.169 |
ug | +0.161 |
Neg logit contributions | |
grate | -0.200 |
violet | -0.186 |
| -0.184 |
Flor | -0.183 |
cheer | -0.180 |
Token | Mike |
Token ID | 4995 |
Feature activation | -4.88e-01 |
Pos logit contributions | |
dd | +0.092 |
ime | +0.085 |
endings | +0.083 |
ug | +0.082 |
happened | +0.078 |
Neg logit contributions | |
Flor | -0.062 |
grate | -0.059 |
| -0.058 |
Av | -0.054 |
cheer | -0.054 |
Token | said |
Token ID | 531 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
dd | +0.022 |
ime | +0.020 |
endings | +0.020 |
ug | +0.020 |
happened | +0.019 |
Neg logit contributions | |
Flor | -0.012 |
grate | -0.012 |
| -0.011 |
cheer | -0.011 |
promise | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
dd | +0.015 |
ime | +0.014 |
ug | +0.013 |
endings | +0.013 |
aw | +0.013 |
Neg logit contributions | |
Flor | -0.007 |
grate | -0.007 |
| -0.007 |
cheer | -0.006 |
Av | -0.006 |
Token | himself |
Token ID | 2241 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
Flor | +0.012 |
| +0.012 |
grate | +0.012 |
cheer | +0.011 |
tell | +0.011 |
Neg logit contributions | |
dd | -0.016 |
ime | -0.014 |
happened | -0.014 |
endings | -0.014 |
ug | -0.013 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -7.27e-01 |
Pos logit contributions | |
dd | +0.010 |
ime | +0.010 |
ug | +0.009 |
endings | +0.009 |
happened | +0.009 |
Neg logit contributions | |
grate | -0.010 |
Flor | -0.009 |
violet | -0.009 |
count | -0.009 |
wider | -0.009 |
Token | One |
Token ID | 3198 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
dd | +0.030 |
ime | +0.028 |
happened | +0.027 |
ug | +0.026 |
endings | +0.026 |
Neg logit contributions | |
Flor | -0.021 |
grate | -0.020 |
violet | -0.019 |
cheer | -0.019 |
| -0.018 |
Token | day |
Token ID | 1110 |
Feature activation | -9.10e-02 |
Pos logit contributions | |
Flor | +0.066 |
grate | +0.063 |
violet | +0.060 |
cheer | +0.059 |
| +0.059 |
Neg logit contributions | |
dd | -0.068 |
ime | -0.062 |
happened | -0.058 |
ug | -0.058 |
endings | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
dd | +0.004 |
ime | +0.004 |
ug | +0.004 |
endings | +0.004 |
aw | +0.004 |
Neg logit contributions | |
Flor | -0.002 |
| -0.002 |
grate | -0.002 |
Av | -0.002 |
Molly | -0.001 |
Token | three |
Token ID | 1115 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
dd | +0.017 |
ime | +0.016 |
ug | +0.016 |
endings | +0.016 |
emergency | +0.015 |
Neg logit contributions | |
| -0.008 |
Flor | -0.008 |
Av | -0.006 |
grate | -0.006 |
Molly | -0.006 |
Token | - |
Token ID | 12 |
Feature activation | -5.67e+00 |
Pos logit contributions | |
dd | +0.047 |
endings | +0.044 |
ime | +0.042 |
happened | +0.042 |
ug | +0.041 |
Neg logit contributions | |
| -0.037 |
Flor | -0.036 |
grate | -0.036 |
violet | -0.032 |
w | -0.032 |
Token | year |
Token ID | 1941 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
dd | +0.190 |
ime | +0.173 |
happened | +0.167 |
endings | +0.167 |
ug | +0.160 |
Neg logit contributions | |
Flor | -0.208 |
grate | -0.197 |
| -0.191 |
cheer | -0.190 |
violet | -0.186 |
Token | - |
Token ID | 12 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
endings | +0.077 |
happened | +0.076 |
Stop | +0.074 |
dd | +0.073 |
ime | +0.073 |
Neg logit contributions | |
grate | -0.105 |
Flor | -0.102 |
Av | -0.092 |
wider | -0.089 |
violet | -0.089 |
Token | old |
Token ID | 727 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
dd | +0.131 |
ime | +0.116 |
happened | +0.111 |
endings | +0.110 |
ug | +0.107 |
Neg logit contributions | |
Flor | -0.225 |
grate | -0.214 |
| -0.211 |
cheer | -0.208 |
Av | -0.205 |
Token | Sam |
Token ID | 3409 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
dd | +0.084 |
ime | +0.079 |
ug | +0.077 |
endings | +0.076 |
happened | +0.073 |
Neg logit contributions | |
Flor | -0.038 |
| -0.033 |
grate | -0.032 |
Av | -0.032 |
cheer | -0.030 |
Token | was |
Token ID | 373 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
dd | +0.113 |
ug | +0.106 |
ime | +0.103 |
endings | +0.101 |
opter | +0.100 |
Neg logit contributions | |
| -0.051 |
Flor | -0.048 |
cheer | -0.043 |
grate | -0.043 |
say | -0.043 |
Token | Max |
Token ID | 5436 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
Flor | +0.030 |
grate | +0.029 |
violet | +0.027 |
cheer | +0.027 |
| +0.027 |
Neg logit contributions | |
dd | -0.038 |
ime | -0.035 |
happened | -0.033 |
ug | -0.033 |
endings | -0.032 |
Token | said |
Token ID | 531 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
dd | +0.028 |
happened | +0.026 |
ime | +0.025 |
aw | +0.024 |
endings | +0.023 |
Neg logit contributions | |
Flor | -0.020 |
violet | -0.018 |
cheer | -0.018 |
grate | -0.018 |
refreshed | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
dd | +0.045 |
ime | +0.042 |
happened | +0.042 |
ug | +0.042 |
ly | +0.039 |
Neg logit contributions | |
violet | -0.038 |
Flor | -0.038 |
forget | -0.034 |
applaud | -0.033 |
grate | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
dd | +0.004 |
ime | +0.004 |
happened | +0.003 |
ug | +0.003 |
endings | +0.003 |
Neg logit contributions | |
Flor | -0.002 |
grate | -0.002 |
violet | -0.002 |
cheer | -0.002 |
| -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
dd | +0.014 |
ime | +0.013 |
happened | +0.013 |
ug | +0.013 |
endings | +0.013 |
Neg logit contributions | |
Flor | -0.006 |
grate | -0.005 |
| -0.005 |
cheer | -0.005 |
violet | -0.005 |
Token | Un |
Token ID | 3118 |
Feature activation | -5.64e+00 |
Pos logit contributions | |
dd | +0.002 |
ime | +0.002 |
happened | +0.002 |
ug | +0.002 |
endings | +0.002 |
Neg logit contributions | |
Flor | -0.002 |
grate | -0.002 |
| -0.002 |
cheer | -0.002 |
violet | -0.002 |
Token | cle |
Token ID | 2375 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
dd | +0.183 |
ime | +0.171 |
endings | +0.160 |
happened | +0.158 |
ug | +0.153 |
Neg logit contributions | |
grate | -0.207 |
Flor | -0.199 |
violet | -0.195 |
| -0.193 |
cheer | -0.190 |
Token | Bob |
Token ID | 5811 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
dd | +0.084 |
ime | +0.077 |
ug | +0.074 |
endings | +0.073 |
happened | +0.070 |
Neg logit contributions | |
Flor | -0.066 |
grate | -0.062 |
| -0.060 |
cheer | -0.059 |
violet | -0.058 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
dd | +0.018 |
ime | +0.017 |
happened | +0.016 |
ug | +0.016 |
endings | +0.016 |
Neg logit contributions | |
Flor | -0.011 |
grate | -0.010 |
violet | -0.010 |
cheer | -0.010 |
| -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
Flor | +0.047 |
violet | +0.044 |
grate | +0.041 |
cheer | +0.041 |
Av | +0.041 |
Neg logit contributions | |
dd | -0.050 |
ime | -0.045 |
happened | -0.044 |
aw | -0.042 |
ug | -0.042 |
Token | put |
Token ID | 1234 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
dd | +0.020 |
happened | +0.017 |
ime | +0.017 |
aw | +0.016 |
ug | +0.016 |
Neg logit contributions | |
Flor | -0.012 |
grate | -0.011 |
cheer | -0.011 |
violet | -0.010 |
| -0.010 |
Token | books |
Token ID | 3835 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
dd | +0.026 |
ime | +0.024 |
endings | +0.024 |
happened | +0.024 |
ug | +0.023 |
Neg logit contributions | |
Flor | -0.020 |
grate | -0.019 |
violet | -0.018 |
| -0.018 |
cheer | -0.018 |
Token | first |
Token ID | 717 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
dd | +0.025 |
ime | +0.023 |
happened | +0.022 |
ug | +0.022 |
aw | +0.021 |
Neg logit contributions | |
Flor | -0.019 |
grate | -0.018 |
violet | -0.017 |
cheer | -0.017 |
| -0.016 |
Token | ?" |
Token ID | 1701 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
dd | +0.049 |
ime | +0.046 |
happened | +0.044 |
ug | +0.044 |
endings | +0.043 |
Neg logit contributions | |
Flor | -0.031 |
grate | -0.029 |
violet | -0.027 |
cheer | -0.027 |
| -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -5.28e-02 |
Pos logit contributions | |
Flor | +0.020 |
grate | +0.019 |
cheer | +0.018 |
violet | +0.018 |
promise | +0.016 |
Neg logit contributions | |
dd | -0.032 |
ime | -0.030 |
happened | -0.029 |
endings | -0.028 |
ug | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
dd | +0.003 |
ime | +0.003 |
endings | +0.003 |
ug | +0.003 |
happened | +0.003 |
Neg logit contributions | |
Flor | -0.001 |
| -0.001 |
grate | -0.001 |
cheer | -0.001 |
violet | -0.001 |
Token | Un |
Token ID | 3118 |
Feature activation | -5.64e+00 |
Pos logit contributions | |
Flor | +0.006 |
grate | +0.006 |
| +0.006 |
cheer | +0.006 |
violet | +0.006 |
Neg logit contributions | |
dd | -0.004 |
ime | -0.004 |
happened | -0.003 |
ug | -0.003 |
endings | -0.003 |
Token | cle |
Token ID | 2375 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
dd | +0.166 |
ime | +0.150 |
endings | +0.139 |
happened | +0.138 |
ug | +0.136 |
Neg logit contributions | |
grate | -0.226 |
Flor | -0.225 |
violet | -0.215 |
| -0.214 |
cheer | -0.212 |
Token | Tom |
Token ID | 4186 |
Feature activation | -5.33e-01 |
Pos logit contributions | |
dd | +0.079 |
ime | +0.072 |
ug | +0.068 |
happened | +0.068 |
endings | +0.067 |
Neg logit contributions | |
Flor | -0.070 |
grate | -0.067 |
| -0.063 |
violet | -0.063 |
cheer | -0.063 |
Token | laughed |
Token ID | 13818 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
dd | +0.024 |
ime | +0.022 |
happened | +0.021 |
ug | +0.021 |
aw | +0.020 |
Neg logit contributions | |
Flor | -0.014 |
grate | -0.012 |
violet | -0.012 |
cheer | -0.012 |
| -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
Flor | +0.045 |
violet | +0.042 |
grate | +0.041 |
cheer | +0.039 |
Av | +0.039 |
Neg logit contributions | |
dd | -0.048 |
ime | -0.043 |
happened | -0.041 |
ug | -0.041 |
aw | -0.040 |
Token | said |
Token ID | 531 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
Flor | +0.001 |
grate | +0.001 |
Deal | +0.001 |
promise | +0.001 |
violet | +0.001 |
Neg logit contributions | |
dd | -0.001 |
happened | -0.001 |
aw | -0.001 |
ime | -0.001 |
ug | -0.001 |
Token | wise |
Token ID | 10787 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
dd | +0.011 |
ime | +0.010 |
ug | +0.009 |
endings | +0.009 |
happened | +0.009 |
Neg logit contributions | |
Flor | -0.013 |
grate | -0.012 |
| -0.012 |
cheer | -0.012 |
violet | -0.012 |
Token | pony |
Token ID | 26902 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
dd | +0.051 |
ime | +0.046 |
endings | +0.045 |
ug | +0.044 |
happened | +0.043 |
Neg logit contributions | |
Flor | -0.037 |
grate | -0.034 |
| -0.033 |
cheer | -0.033 |
promise | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Flor | +0.003 |
grate | +0.003 |
violet | +0.003 |
cheer | +0.002 |
| +0.002 |
Neg logit contributions | |
dd | -0.004 |
ime | -0.003 |
happened | -0.003 |
ug | -0.003 |
endings | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -9.49e-01 |
Pos logit contributions | |
dd | +0.042 |
ime | +0.039 |
happened | +0.037 |
ug | +0.036 |
endings | +0.036 |
Neg logit contributions | |
Flor | -0.032 |
grate | -0.031 |
violet | -0.030 |
| -0.029 |
cheer | -0.029 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
dd | +0.034 |
ime | +0.031 |
ug | +0.029 |
happened | +0.028 |
endings | +0.028 |
Neg logit contributions | |
Flor | -0.034 |
grate | -0.032 |
| -0.031 |
cheer | -0.031 |
violet | -0.031 |
Token | - |
Token ID | 12 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
dd | +0.069 |
endings | +0.063 |
ug | +0.061 |
happened | +0.060 |
anked | +0.060 |
Neg logit contributions | |
| -0.060 |
grate | -0.057 |
cheer | -0.057 |
Flor | -0.057 |
promise | -0.056 |
Token | year |
Token ID | 1941 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
dd | +0.199 |
ime | +0.181 |
endings | +0.176 |
happened | +0.174 |
ug | +0.167 |
Neg logit contributions | |
Flor | -0.189 |
grate | -0.182 |
cheer | -0.173 |
| -0.172 |
violet | -0.171 |
Token | old |
Token ID | 1468 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
dd | +0.084 |
happened | +0.076 |
ime | +0.074 |
endings | +0.069 |
Stop | +0.063 |
Neg logit contributions | |
grate | -0.140 |
Flor | -0.133 |
cheer | -0.130 |
promise | -0.126 |
wider | -0.123 |
Token | girl |
Token ID | 2576 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
dd | +0.088 |
ime | +0.080 |
ug | +0.077 |
endings | +0.075 |
happened | +0.074 |
Neg logit contributions | |
Flor | -0.068 |
grate | -0.065 |
cheer | -0.064 |
| -0.064 |
violet | -0.060 |
Token | played |
Token ID | 2826 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
Flor | +0.017 |
grate | +0.016 |
violet | +0.015 |
Av | +0.015 |
| +0.015 |
Neg logit contributions | |
dd | -0.019 |
ime | -0.017 |
happened | -0.016 |
ug | -0.016 |
endings | -0.015 |
Token | together |
Token ID | 1978 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
Flor | +0.007 |
grate | +0.007 |
violet | +0.006 |
| +0.006 |
cheer | +0.006 |
Neg logit contributions | |
dd | -0.017 |
ime | -0.015 |
ug | -0.015 |
happened | -0.015 |
endings | -0.015 |
Token | 's |
Token ID | 338 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
Flor | +0.010 |
grate | +0.009 |
violet | +0.009 |
cheer | +0.008 |
| +0.008 |
Neg logit contributions | |
dd | -0.019 |
ime | -0.018 |
happened | -0.017 |
endings | -0.017 |
ug | -0.017 |
Token | silly |
Token ID | 14397 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
Flor | +0.040 |
grate | +0.038 |
refreshed | +0.036 |
cheer | +0.036 |
violet | +0.035 |
Neg logit contributions | |
dd | -0.045 |
ime | -0.040 |
happened | -0.040 |
ug | -0.038 |
endings | -0.038 |
Token | !" |
Token ID | 2474 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
Flor | +0.016 |
grate | +0.015 |
cheer | +0.014 |
| +0.014 |
violet | +0.014 |
Neg logit contributions | |
dd | -0.029 |
ime | -0.027 |
ug | -0.025 |
happened | -0.025 |
endings | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
Flor | +0.023 |
grate | +0.021 |
violet | +0.020 |
cheer | +0.020 |
promise | +0.019 |
Neg logit contributions | |
dd | -0.037 |
ime | -0.035 |
happened | -0.033 |
ug | -0.033 |
aw | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -8.53e-03 |
Pos logit contributions | |
dd | +0.013 |
ime | +0.012 |
ug | +0.011 |
happened | +0.011 |
endings | +0.011 |
Neg logit contributions | |
Flor | -0.005 |
grate | -0.005 |
| -0.004 |
violet | -0.004 |
cheer | -0.004 |
Token | Un |
Token ID | 3118 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
dd | +0.000 |
ime | +0.000 |
happened | +0.000 |
ug | +0.000 |
endings | +0.000 |
Neg logit contributions | |
Flor | -0.000 |
grate | -0.000 |
| -0.000 |
violet | -0.000 |
cheer | -0.000 |
Token | cle |
Token ID | 2375 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
dd | +0.162 |
ime | +0.145 |
endings | +0.135 |
happened | +0.135 |
ug | +0.132 |
Neg logit contributions | |
grate | -0.222 |
Flor | -0.222 |
violet | -0.214 |
| -0.212 |
cheer | -0.211 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
dd | +0.084 |
ime | +0.077 |
happened | +0.073 |
ug | +0.073 |
endings | +0.071 |
Neg logit contributions | |
Flor | -0.056 |
grate | -0.053 |
violet | -0.050 |
cheer | -0.050 |
| -0.049 |
Token | and |
Token ID | 290 |
Feature activation | -6.06e-01 |
Pos logit contributions | |
Flor | +0.041 |
violet | +0.038 |
grate | +0.036 |
cheer | +0.036 |
| +0.036 |
Neg logit contributions | |
dd | -0.048 |
ime | -0.044 |
happened | -0.042 |
ug | -0.041 |
aw | -0.041 |
Token | said |
Token ID | 531 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
dd | +0.027 |
happened | +0.024 |
ime | +0.023 |
ug | +0.022 |
aw | +0.022 |
Neg logit contributions | |
Flor | -0.016 |
grate | -0.014 |
violet | -0.014 |
cheer | -0.014 |
promise | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
dd | +0.034 |
ime | +0.031 |
ug | +0.030 |
endings | +0.030 |
happened | +0.029 |
Neg logit contributions | |
Flor | -0.017 |
grate | -0.016 |
| -0.015 |
cheer | -0.015 |
violet | -0.014 |
Token | some |
Token ID | 617 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
Flor | +0.055 |
grate | +0.054 |
cheer | +0.050 |
violet | +0.049 |
| +0.048 |
Neg logit contributions | |
dd | -0.087 |
ime | -0.078 |
happened | -0.076 |
endings | -0.074 |
ug | -0.074 |
Token | cake |
Token ID | 12187 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Flor | +0.003 |
grate | +0.003 |
violet | +0.003 |
| +0.003 |
cheer | +0.003 |
Neg logit contributions | |
dd | -0.004 |
ime | -0.004 |
happened | -0.004 |
ug | -0.004 |
aw | -0.003 |
Token | ?" |
Token ID | 1701 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
Flor | +0.038 |
grate | +0.035 |
violet | +0.035 |
refreshed | +0.034 |
cheer | +0.034 |
Neg logit contributions | |
dd | -0.037 |
happened | -0.035 |
ime | -0.034 |
ug | -0.031 |
aw | -0.030 |
Token | Tom |
Token ID | 4186 |
Feature activation | -9.46e-01 |
Pos logit contributions | |
grate | +0.015 |
Flor | +0.015 |
violet | +0.015 |
cheer | +0.014 |
promise | +0.014 |
Neg logit contributions | |
dd | -0.025 |
ime | -0.023 |
happened | -0.023 |
ug | -0.022 |
endings | -0.021 |
Token | asked |
Token ID | 1965 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
dd | +0.034 |
ime | +0.033 |
happened | +0.031 |
aw | +0.029 |
ug | +0.029 |
Neg logit contributions | |
Flor | -0.031 |
grate | -0.028 |
violet | -0.028 |
cheer | -0.028 |
refreshed | -0.028 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
dd | +0.029 |
ime | +0.027 |
ug | +0.026 |
happened | +0.026 |
endings | +0.025 |
Neg logit contributions | |
Flor | -0.019 |
violet | -0.018 |
grate | -0.018 |
cheer | -0.017 |
refreshed | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
Flor | +0.097 |
violet | +0.097 |
Deal | +0.090 |
refreshed | +0.088 |
forget | +0.086 |
Neg logit contributions | |
ime | -0.068 |
ly | -0.066 |
dd | -0.064 |
happened | -0.060 |
ug | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | -7.06e-02 |
Pos logit contributions | |
forget | +0.008 |
promise | +0.008 |
Flor | +0.008 |
refreshed | +0.008 |
grate | +0.008 |
Neg logit contributions | |
happened | -0.008 |
dd | -0.008 |
ime | -0.007 |
ug | -0.007 |
aw | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
dd | +0.003 |
ime | +0.003 |
ug | +0.003 |
happened | +0.003 |
endings | +0.003 |
Neg logit contributions | |
Flor | -0.002 |
grate | -0.002 |
violet | -0.001 |
cheer | -0.001 |
| -0.001 |
Token | " |
Token ID | 1 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
Flor | +0.006 |
grate | +0.005 |
violet | +0.005 |
refreshed | +0.005 |
cheer | +0.005 |
Neg logit contributions | |
dd | -0.006 |
ime | -0.005 |
ug | -0.005 |
happened | -0.005 |
endings | -0.005 |
Token | No |
Token ID | 2949 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
Flor | +0.011 |
grate | +0.011 |
cheer | +0.010 |
| +0.010 |
refreshed | +0.010 |
Neg logit contributions | |
dd | -0.014 |
ime | -0.012 |
happened | -0.012 |
endings | -0.012 |
ug | -0.011 |
Token | please |
Token ID | 3387 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
Flor | +0.063 |
grate | +0.057 |
violet | +0.054 |
refreshed | +0.053 |
Av | +0.051 |
Neg logit contributions | |
dd | -0.078 |
ime | -0.076 |
happened | -0.075 |
ug | -0.070 |
emergency | -0.064 |
Token | go |
Token ID | 467 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
Flor | +0.004 |
violet | +0.003 |
grate | +0.003 |
refreshed | +0.003 |
Alexa | +0.003 |
Neg logit contributions | |
ime | -0.004 |
happened | -0.004 |
dd | -0.004 |
ug | -0.004 |
emergency | -0.003 |
Token | look |
Token ID | 804 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Flor | +0.008 |
violet | +0.008 |
grate | +0.007 |
Deal | +0.007 |
Av | +0.007 |
Neg logit contributions | |
dd | -0.006 |
ime | -0.005 |
ug | -0.005 |
happened | -0.005 |
ly | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
Flor | +0.064 |
grate | +0.060 |
violet | +0.058 |
cheer | +0.057 |
| +0.057 |
Neg logit contributions | |
dd | -0.078 |
ime | -0.070 |
happened | -0.067 |
ug | -0.066 |
endings | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
Flor | +0.048 |
grate | +0.046 |
violet | +0.044 |
cheer | +0.042 |
| +0.042 |
Neg logit contributions | |
dd | -0.065 |
ime | -0.060 |
ug | -0.057 |
happened | -0.057 |
endings | -0.056 |
Token | living |
Token ID | 2877 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
ime | +0.002 |
dd | +0.002 |
happened | +0.002 |
emergency | +0.002 |
endings | +0.002 |
Neg logit contributions | |
Flor | -0.002 |
grate | -0.002 |
violet | -0.002 |
cheer | -0.002 |
Av | -0.002 |
Token | room |
Token ID | 2119 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Flor | +0.079 |
grate | +0.075 |
violet | +0.071 |
cheer | +0.067 |
| +0.067 |
Neg logit contributions | |
dd | -0.103 |
ime | -0.101 |
happened | -0.094 |
ug | -0.093 |
endings | -0.093 |
Token | ?" |
Token ID | 1701 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
Flor | +0.039 |
violet | +0.036 |
grate | +0.035 |
cheer | +0.033 |
| +0.033 |
Neg logit contributions | |
dd | -0.045 |
ime | -0.044 |
ug | -0.041 |
happened | -0.040 |
aw | -0.039 |
Token | She |
Token ID | 1375 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
Flor | +0.022 |
grate | +0.021 |
violet | +0.021 |
cheer | +0.020 |
refreshed | +0.019 |
Neg logit contributions | |
dd | -0.020 |
ime | -0.018 |
happened | -0.017 |
ug | -0.017 |
endings | -0.017 |
Token | nodded |
Token ID | 14464 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
Flor | +0.022 |
grate | +0.020 |
cheer | +0.020 |
violet | +0.020 |
join | +0.019 |
Neg logit contributions | |
dd | -0.034 |
happened | -0.030 |
ime | -0.030 |
ug | -0.028 |
aw | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
dd | +0.005 |
ime | +0.005 |
happened | +0.005 |
ug | +0.004 |
ly | +0.004 |
Neg logit contributions | |
Flor | -0.005 |
violet | -0.005 |
forget | -0.004 |
Av | -0.004 |
grate | -0.004 |
Token | as |
Token ID | 355 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Flor | +0.005 |
grate | +0.004 |
violet | +0.004 |
cheer | +0.004 |
| +0.004 |
Neg logit contributions | |
dd | -0.007 |
ime | -0.007 |
happened | -0.007 |
ug | -0.006 |
endings | -0.006 |
Token | they |
Token ID | 484 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
dd | +0.068 |
happened | +0.062 |
ime | +0.059 |
ug | +0.056 |
endings | +0.054 |
Neg logit contributions | |
Flor | -0.034 |
violet | -0.033 |
grate | -0.033 |
cheer | -0.031 |
join | -0.029 |
Token | were |
Token ID | 547 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
dd | +0.024 |
ime | +0.022 |
ug | +0.021 |
endings | +0.021 |
happened | +0.021 |
Neg logit contributions | |
Flor | -0.016 |
grate | -0.015 |
| -0.014 |
cheer | -0.014 |
violet | -0.014 |
Token | drinking |
Token ID | 7722 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
Flor | +0.000 |
grate | +0.000 |
cheer | +0.000 |
violet | +0.000 |
| +0.000 |
Neg logit contributions | |
dd | -0.001 |
ime | -0.001 |
happened | -0.001 |
ug | -0.001 |
endings | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
Flor | +0.037 |
grate | +0.035 |
violet | +0.035 |
| +0.032 |
cheer | +0.032 |
Neg logit contributions | |
dd | -0.049 |
ime | -0.045 |
happened | -0.044 |
ug | -0.042 |
endings | -0.041 |
Token | tea |
Token ID | 8887 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
grate | +0.017 |
Flor | +0.016 |
| +0.014 |
Together | +0.014 |
violet | +0.014 |
Neg logit contributions | |
dd | -0.033 |
happened | -0.032 |
ime | -0.031 |
endings | -0.029 |
ly | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
Flor | +0.081 |
grate | +0.075 |
violet | +0.071 |
cheer | +0.068 |
| +0.067 |
Neg logit contributions | |
dd | -0.110 |
ime | -0.104 |
happened | -0.104 |
endings | -0.096 |
ug | -0.095 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
Flor | +0.005 |
grate | +0.004 |
violet | +0.004 |
cheer | +0.004 |
| +0.004 |
Neg logit contributions | |
dd | -0.008 |
ime | -0.007 |
happened | -0.007 |
ug | -0.007 |
endings | -0.007 |
Token | accidentally |
Token ID | 14716 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
Flor | +0.047 |
grate | +0.045 |
| +0.043 |
cheer | +0.042 |
violet | +0.042 |
Neg logit contributions | |
dd | -0.076 |
ime | -0.070 |
ug | -0.067 |
endings | -0.066 |
happened | -0.065 |
Token | spilled |
Token ID | 31896 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
dd | +0.076 |
ime | +0.069 |
ug | +0.066 |
endings | +0.066 |
happened | +0.062 |
Neg logit contributions | |
Flor | -0.075 |
grate | -0.072 |
| -0.069 |
cheer | -0.068 |
violet | -0.068 |
Token | it |
Token ID | 340 |
Feature activation | -6.31e-02 |
Pos logit contributions | |
Flor | +0.003 |
grate | +0.003 |
violet | +0.003 |
cheer | +0.003 |
| +0.003 |
Neg logit contributions | |
dd | -0.005 |
ime | -0.004 |
happened | -0.004 |
ug | -0.004 |
endings | -0.004 |
Token | for |
Token ID | 329 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
dd | +0.099 |
ime | +0.091 |
happened | +0.089 |
ug | +0.087 |
endings | +0.086 |
Neg logit contributions | |
Flor | -0.065 |
violet | -0.062 |
grate | -0.060 |
| -0.057 |
cheer | -0.056 |
Token | Emma |
Token ID | 18966 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
dd | +0.076 |
ime | +0.069 |
happened | +0.066 |
ug | +0.065 |
aw | +0.065 |
Neg logit contributions | |
Flor | -0.053 |
grate | -0.050 |
| -0.048 |
cheer | -0.047 |
Av | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
Flor | +0.033 |
violet | +0.031 |
grate | +0.030 |
cheer | +0.029 |
refreshed | +0.028 |
Neg logit contributions | |
dd | -0.044 |
ime | -0.041 |
happened | -0.040 |
endings | -0.038 |
ug | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
dd | +0.012 |
ime | +0.012 |
endings | +0.011 |
happened | +0.011 |
John | +0.011 |
Neg logit contributions | |
cheer | -0.010 |
grate | -0.010 |
violet | -0.009 |
promise | -0.009 |
Flor | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
dd | +0.014 |
ime | +0.013 |
ug | +0.013 |
happened | +0.013 |
endings | +0.012 |
Neg logit contributions | |
Flor | -0.006 |
grate | -0.006 |
violet | -0.005 |
cheer | -0.005 |
| -0.005 |
Token | One |
Token ID | 3198 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
Flor | +0.001 |
grate | +0.001 |
| +0.001 |
violet | +0.001 |
cheer | +0.001 |
Neg logit contributions | |
dd | -0.002 |
ime | -0.002 |
ug | -0.002 |
happened | -0.002 |
endings | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
Flor | +0.092 |
grate | +0.088 |
violet | +0.084 |
| +0.083 |
cheer | +0.083 |
Neg logit contributions | |
dd | -0.093 |
ime | -0.084 |
happened | -0.079 |
ug | -0.078 |
endings | -0.076 |
Token | Tom |
Token ID | 4186 |
Feature activation | -8.14e-01 |
Pos logit contributions | |
Flor | +0.006 |
| +0.006 |
grate | +0.006 |
Av | +0.005 |
cheer | +0.005 |
Neg logit contributions | |
dd | -0.015 |
ime | -0.014 |
ug | -0.013 |
endings | -0.013 |
aw | -0.013 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
dd | +0.036 |
ime | +0.033 |
ug | +0.032 |
endings | +0.031 |
happened | +0.031 |
Neg logit contributions | |
Flor | -0.022 |
grate | -0.020 |
| -0.019 |
cheer | -0.019 |
violet | -0.019 |
Token | Emma |
Token ID | 18966 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
Flor | +0.044 |
grate | +0.042 |
violet | +0.041 |
cheer | +0.040 |
| +0.038 |
Neg logit contributions | |
dd | -0.073 |
ime | -0.066 |
happened | -0.064 |
ug | -0.063 |
endings | -0.062 |
Token | a |
Token ID | 257 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
Flor | +0.057 |
grate | +0.054 |
violet | +0.053 |
cheer | +0.052 |
| +0.050 |
Neg logit contributions | |
dd | -0.076 |
ime | -0.070 |
happened | -0.066 |
endings | -0.065 |
ug | -0.064 |
Token | it |
Token ID | 340 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
dd | +0.047 |
ime | +0.043 |
ug | +0.041 |
happened | +0.040 |
endings | +0.040 |
Neg logit contributions | |
Flor | -0.037 |
grate | -0.036 |
| -0.034 |
cheer | -0.034 |
violet | -0.033 |
Token | again |
Token ID | 757 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
dd | +0.020 |
happened | +0.019 |
ime | +0.018 |
endings | +0.018 |
ug | +0.018 |
Neg logit contributions | |
Flor | -0.016 |
violet | -0.015 |
cheer | -0.014 |
grate | -0.014 |
Av | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
Flor | +0.031 |
grate | +0.028 |
violet | +0.028 |
cheer | +0.027 |
| +0.026 |
Neg logit contributions | |
dd | -0.044 |
ime | -0.041 |
happened | -0.038 |
ug | -0.038 |
endings | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
dd | +0.008 |
ime | +0.007 |
happened | +0.007 |
ug | +0.007 |
endings | +0.007 |
Neg logit contributions | |
Flor | -0.005 |
grate | -0.005 |
cheer | -0.004 |
violet | -0.004 |
promise | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
Flor | +0.008 |
| +0.008 |
grate | +0.007 |
cheer | +0.007 |
violet | +0.007 |
Neg logit contributions | |
dd | -0.023 |
ime | -0.022 |
endings | -0.021 |
ug | -0.021 |
happened | -0.021 |
Token | One |
Token ID | 3198 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
Flor | +0.014 |
| +0.014 |
grate | +0.013 |
violet | +0.012 |
cheer | +0.012 |
Neg logit contributions | |
dd | -0.039 |
ime | -0.037 |
ug | -0.035 |
happened | -0.035 |
endings | -0.035 |
Token | day |
Token ID | 1110 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
Flor | +0.080 |
grate | +0.076 |
| +0.074 |
cheer | +0.073 |
violet | +0.072 |
Neg logit contributions | |
dd | -0.122 |
ime | -0.110 |
endings | -0.109 |
ug | -0.107 |
happened | -0.104 |
Token | , |
Token ID | 11 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
| +0.006 |
Flor | +0.006 |
grate | +0.005 |
join | +0.004 |
count | +0.004 |
Neg logit contributions | |
dd | -0.031 |
endings | -0.030 |
ime | -0.030 |
ug | -0.029 |
ors | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
dd | +0.007 |
ime | +0.007 |
endings | +0.007 |
ug | +0.007 |
happened | +0.006 |
Neg logit contributions | |
Flor | -0.003 |
grate | -0.003 |
| -0.003 |
cheer | -0.003 |
join | -0.003 |
Token | boy |
Token ID | 2933 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
dd | +0.002 |
ime | +0.002 |
ug | +0.002 |
endings | +0.002 |
ors | +0.002 |
Neg logit contributions | |
| -0.002 |
cheer | -0.002 |
better | -0.002 |
Flor | -0.002 |
grate | -0.002 |
Token | couldn |
Token ID | 3521 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
Flor | +0.010 |
grate | +0.010 |
| +0.009 |
count | +0.009 |
invite | +0.009 |
Neg logit contributions | |
dd | -0.021 |
ime | -0.019 |
ug | -0.019 |
endings | -0.018 |
opter | -0.018 |
Token | her |
Token ID | 607 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
dd | +0.013 |
ime | +0.012 |
ug | +0.011 |
happened | +0.011 |
endings | +0.011 |
Neg logit contributions | |
Flor | -0.010 |
grate | -0.010 |
violet | -0.009 |
| -0.009 |
cheer | -0.009 |
Token | dog |
Token ID | 3290 |
Feature activation | +8.85e-01 |
Pos logit contributions | |
dd | +0.010 |
ime | +0.009 |
ug | +0.008 |
happened | +0.008 |
endings | +0.008 |
Neg logit contributions | |
Flor | -0.006 |
grate | -0.006 |
cheer | -0.005 |
| -0.005 |
violet | -0.005 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
Flor | +0.025 |
grate | +0.024 |
violet | +0.023 |
cheer | +0.022 |
| +0.022 |
Neg logit contributions | |
dd | -0.036 |
ime | -0.034 |
happened | -0.033 |
ug | -0.032 |
endings | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
dd | +0.077 |
ug | +0.068 |
ime | +0.067 |
endings | +0.066 |
opter | +0.065 |
Neg logit contributions | |
Flor | -0.031 |
| -0.029 |
grate | -0.028 |
cheer | -0.027 |
Ms | -0.025 |
Token | children |
Token ID | 1751 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
Flor | +0.010 |
grate | +0.010 |
cheer | +0.009 |
violet | +0.009 |
| +0.009 |
Neg logit contributions | |
dd | -0.010 |
ime | -0.009 |
ug | -0.008 |
happened | -0.008 |
endings | -0.008 |
Token | fighting |
Token ID | 4330 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
dd | +0.006 |
ime | +0.006 |
happened | +0.005 |
ug | +0.005 |
endings | +0.005 |
Neg logit contributions | |
Flor | -0.005 |
grate | -0.004 |
violet | -0.004 |
Av | -0.004 |
cheer | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
Flor | +0.005 |
grate | +0.005 |
cheer | +0.004 |
| +0.004 |
violet | +0.004 |
Neg logit contributions | |
dd | -0.010 |
ime | -0.009 |
ug | -0.009 |
endings | -0.008 |
happened | -0.008 |
Token | She |
Token ID | 1375 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
grate | +0.010 |
Flor | +0.010 |
promise | +0.010 |
cheer | +0.010 |
violet | +0.010 |
Neg logit contributions | |
dd | -0.016 |
ime | -0.015 |
happened | -0.014 |
ug | -0.014 |
aw | -0.014 |
Token | came |
Token ID | 1625 |
Feature activation | -8.74e-01 |
Pos logit contributions | |
Flor | +0.022 |
grate | +0.020 |
violet | +0.019 |
| +0.019 |
cheer | +0.019 |
Neg logit contributions | |
dd | -0.036 |
ime | -0.033 |
happened | -0.031 |
ug | -0.031 |
endings | -0.031 |
Token | over |
Token ID | 625 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
dd | +0.038 |
ime | +0.037 |
ug | +0.035 |
happened | +0.034 |
aw | +0.032 |
Neg logit contributions | |
Flor | -0.022 |
violet | -0.020 |
grate | -0.020 |
refreshed | -0.019 |
promise | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +9.95e-01 |
Pos logit contributions | |
Flor | +0.013 |
grate | +0.012 |
| +0.012 |
cheer | +0.012 |
violet | +0.011 |
Neg logit contributions | |
dd | -0.020 |
ime | -0.018 |
ug | -0.017 |
endings | -0.017 |
happened | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
Flor | +0.003 |
grate | +0.003 |
violet | +0.003 |
| +0.003 |
cheer | +0.003 |
Neg logit contributions | |
dd | -0.005 |
ime | -0.005 |
happened | -0.004 |
ug | -0.004 |
endings | -0.004 |
Token | Her |
Token ID | 2332 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
grate | +0.001 |
Flor | +0.001 |
cheer | +0.001 |
refreshed | +0.001 |
promise | +0.001 |
Neg logit contributions | |
dd | -0.002 |
ime | -0.002 |
happened | -0.002 |
endings | -0.002 |
ug | -0.002 |
Token | mom |
Token ID | 1995 |
Feature activation | +7.38e-01 |
Pos logit contributions | |
Flor | +0.028 |
grate | +0.026 |
violet | +0.025 |
cheer | +0.025 |
Av | +0.024 |
Neg logit contributions | |
dd | -0.043 |
ime | -0.040 |
happened | -0.040 |
endings | -0.038 |
ug | -0.038 |
Token | told |
Token ID | 1297 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
Flor | +0.021 |
violet | +0.018 |
Av | +0.018 |
grate | +0.018 |
refreshed | +0.017 |
Neg logit contributions | |
dd | -0.032 |
ime | -0.029 |
happened | -0.029 |
ug | -0.027 |
aw | -0.026 |
Token | her |
Token ID | 607 |
Feature activation | +8.42e-01 |
Pos logit contributions | |
Flor | +0.005 |
grate | +0.004 |
cheer | +0.004 |
| +0.004 |
Av | +0.004 |
Neg logit contributions | |
dd | -0.014 |
ime | -0.012 |
ug | -0.012 |
endings | -0.012 |
opter | -0.012 |
Token | that |
Token ID | 326 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
Flor | +0.027 |
grate | +0.026 |
| +0.025 |
cheer | +0.025 |
violet | +0.025 |
Neg logit contributions | |
dd | -0.032 |
ime | -0.029 |
ug | -0.028 |
endings | -0.027 |
happened | -0.027 |
Token | sometimes |
Token ID | 3360 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
Flor | +0.044 |
grate | +0.042 |
violet | +0.039 |
| +0.039 |
cheer | +0.039 |
Neg logit contributions | |
dd | -0.064 |
ime | -0.059 |
happened | -0.056 |
ug | -0.056 |
endings | -0.055 |
Token | a |
Token ID | 257 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
Flor | +0.085 |
grate | +0.080 |
cheer | +0.078 |
| +0.078 |
violet | +0.075 |
Neg logit contributions | |
dd | -0.089 |
ime | -0.079 |
ug | -0.075 |
happened | -0.071 |
endings | -0.071 |
Token | goose |
Token ID | 37246 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
Flor | +0.019 |
grate | +0.019 |
violet | +0.018 |
cheer | +0.018 |
| +0.017 |
Neg logit contributions | |
dd | -0.020 |
ime | -0.019 |
happened | -0.018 |
endings | -0.017 |
ug | -0.017 |
Token | needs |
Token ID | 2476 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
Flor | +0.015 |
grate | +0.015 |
| +0.014 |
cheer | +0.014 |
violet | +0.013 |
Neg logit contributions | |
dd | -0.026 |
ime | -0.024 |
ug | -0.022 |
endings | -0.022 |
happened | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +6.80e-02 |
Pos logit contributions | |
dd | +0.010 |
ime | +0.009 |
ug | +0.008 |
happened | +0.008 |
endings | +0.008 |
Neg logit contributions | |
Flor | -0.006 |
grate | -0.006 |
| -0.006 |
cheer | -0.006 |
violet | -0.006 |
Token | forg |
Token ID | 5784 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
Flor | +0.008 |
grate | +0.007 |
violet | +0.007 |
| +0.007 |
cheer | +0.007 |
Neg logit contributions | |
dd | -0.015 |
ime | -0.014 |
happened | -0.014 |
ug | -0.014 |
endings | -0.013 |
Token | ave |
Token ID | 1015 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
grate | +0.014 |
Flor | +0.013 |
cheer | +0.013 |
kiss | +0.013 |
| +0.013 |
Neg logit contributions | |
dd | -0.014 |
ime | -0.013 |
emergency | -0.012 |
endings | -0.012 |
ug | -0.012 |
Token | them |
Token ID | 606 |
Feature activation | -9.46e-03 |
Pos logit contributions | |
dd | +0.024 |
ime | +0.020 |
happened | +0.020 |
endings | +0.020 |
opter | +0.020 |
Neg logit contributions | |
Flor | -0.013 |
| -0.012 |
cheer | -0.011 |
grate | -0.011 |
Av | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
dd | +0.000 |
ime | +0.000 |
ug | +0.000 |
ly | +0.000 |
aw | +0.000 |
Neg logit contributions | |
Flor | -0.000 |
violet | -0.000 |
Av | -0.000 |
grate | -0.000 |
refreshed | -0.000 |
Token | She |
Token ID | 1375 |
Feature activation | +9.03e-01 |
Pos logit contributions | |
Flor | +0.001 |
grate | +0.001 |
cheer | +0.001 |
violet | +0.001 |
promise | +0.001 |
Neg logit contributions | |
dd | -0.001 |
ime | -0.001 |
endings | -0.001 |
happened | -0.001 |
ug | -0.001 |
Token | said |
Token ID | 531 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
Flor | +0.026 |
violet | +0.024 |
Deal | +0.024 |
grate | +0.023 |
| +0.023 |
Neg logit contributions | |
dd | -0.035 |
happened | -0.034 |
aw | -0.031 |
ime | -0.030 |
ug | -0.029 |
Token | she |
Token ID | 673 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
dd | +0.019 |
ime | +0.018 |
ug | +0.017 |
endings | +0.017 |
happened | +0.017 |
Neg logit contributions | |
Flor | -0.008 |
grate | -0.007 |
| -0.006 |
cheer | -0.006 |
violet | -0.006 |
Token | loved |
Token ID | 6151 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
Flor | +0.050 |
grate | +0.047 |
violet | +0.045 |
cheer | +0.045 |
| +0.044 |
Neg logit contributions | |
dd | -0.058 |
ime | -0.052 |
happened | -0.050 |
ug | -0.049 |
endings | -0.049 |
Token | them |
Token ID | 606 |
Feature activation | -7.39e-01 |
Pos logit contributions | |
dd | +0.020 |
ime | +0.018 |
ug | +0.018 |
endings | +0.017 |
opter | +0.017 |
Neg logit contributions | |
Flor | -0.012 |
grate | -0.011 |
cheer | -0.010 |
| -0.010 |
Av | -0.010 |
Token | too |
Token ID | 1165 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
aw | +0.026 |
dd | +0.025 |
ime | +0.025 |
ug | +0.024 |
endings | +0.024 |
Neg logit contributions | |
violet | -0.022 |
Flor | -0.022 |
grate | -0.018 |
Av | -0.018 |
Alexa | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
Flor | +0.018 |
grate | +0.017 |
violet | +0.016 |
cheer | +0.016 |
| +0.016 |
Neg logit contributions | |
dd | -0.031 |
ime | -0.028 |
happened | -0.027 |
ug | -0.027 |
endings | -0.027 |
Token | animals |
Token ID | 4695 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
Flor | +0.013 |
grate | +0.012 |
cheer | +0.012 |
violet | +0.012 |
| +0.012 |
Neg logit contributions | |
dd | -0.015 |
ime | -0.013 |
ug | -0.013 |
happened | -0.013 |
endings | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
Flor | +0.014 |
violet | +0.013 |
grate | +0.013 |
Av | +0.012 |
| +0.012 |
Neg logit contributions | |
dd | -0.019 |
ime | -0.018 |
ug | -0.017 |
happened | -0.017 |
endings | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
Flor | +0.005 |
grate | +0.004 |
violet | +0.004 |
| +0.004 |
cheer | +0.004 |
Neg logit contributions | |
dd | -0.007 |
ime | -0.006 |
ug | -0.006 |
happened | -0.006 |
endings | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
Flor | +0.000 |
| +0.000 |
grate | +0.000 |
cheer | +0.000 |
violet | +0.000 |
Neg logit contributions | |
dd | -0.001 |
ime | -0.001 |
endings | -0.001 |
happened | -0.001 |
ug | -0.001 |
Token | But |
Token ID | 1537 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
| +0.009 |
grate | +0.008 |
Flor | +0.008 |
cheer | +0.008 |
violet | +0.008 |
Neg logit contributions | |
dd | -0.013 |
happened | -0.013 |
ime | -0.012 |
endings | -0.012 |
ug | -0.012 |
Token | then |
Token ID | 788 |
Feature activation | +4.98e-01 |
Pos logit contributions | |
Flor | +0.009 |
grate | +0.009 |
violet | +0.008 |
cheer | +0.008 |
| +0.008 |
Neg logit contributions | |
dd | -0.019 |
ime | -0.017 |
happened | -0.017 |
ug | -0.017 |
endings | -0.016 |
Token | something |
Token ID | 1223 |
Feature activation | -9.42e-01 |
Pos logit contributions | |
Flor | +0.013 |
grate | +0.012 |
| +0.011 |
cheer | +0.011 |
violet | +0.011 |
Neg logit contributions | |
dd | -0.023 |
ime | -0.021 |
ug | -0.020 |
endings | -0.020 |
happened | -0.020 |
Token | amazing |
Token ID | 4998 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
dd | +0.031 |
ime | +0.030 |
ug | +0.029 |
endings | +0.027 |
ors | +0.026 |
Neg logit contributions | |
Flor | -0.034 |
grate | -0.032 |
better | -0.032 |
cheer | -0.032 |
closer | -0.031 |
Token | happened |
Token ID | 3022 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
ime | +0.000 |
dd | +0.000 |
ug | +0.000 |
endings | +0.000 |
ud | +0.000 |
Neg logit contributions | |
grate | -0.001 |
too | -0.001 |
Flor | -0.001 |
closer | -0.001 |
say | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
dd | +0.045 |
ime | +0.041 |
happened | +0.041 |
ug | +0.039 |
endings | +0.039 |
Neg logit contributions | |
Flor | -0.029 |
grate | -0.027 |
violet | -0.026 |
cheer | -0.025 |
| -0.024 |
Token | He |
Token ID | 679 |
Feature activation | -7.74e-01 |
Pos logit contributions | |
Flor | +0.005 |
| +0.005 |
grate | +0.005 |
Av | +0.005 |
violet | +0.004 |
Neg logit contributions | |
dd | -0.009 |
ime | -0.009 |
ug | -0.008 |
endings | -0.008 |
happened | -0.008 |
Token | he |
Token ID | 339 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
Flor | +0.028 |
grate | +0.028 |
| +0.027 |
cheer | +0.026 |
count | +0.024 |
Neg logit contributions | |
dd | -0.048 |
ime | -0.046 |
ug | -0.043 |
endings | -0.042 |
happened | -0.042 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
dd | +0.026 |
ime | +0.024 |
happened | +0.022 |
ug | +0.022 |
endings | +0.022 |
Neg logit contributions | |
Flor | -0.019 |
grate | -0.017 |
violet | -0.017 |
| -0.016 |
cheer | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
Flor | +0.044 |
grate | +0.040 |
violet | +0.040 |
cheer | +0.039 |
| +0.039 |
Neg logit contributions | |
dd | -0.060 |
ime | -0.055 |
happened | -0.052 |
ug | -0.052 |
endings | -0.051 |
Token | slowly |
Token ID | 6364 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
dd | +0.018 |
happened | +0.016 |
ime | +0.015 |
ug | +0.015 |
aw | +0.015 |
Neg logit contributions | |
Flor | -0.012 |
violet | -0.011 |
grate | -0.011 |
cheer | -0.010 |
Av | -0.010 |
Token | began |
Token ID | 2540 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
dd | +0.092 |
ime | +0.086 |
endings | +0.086 |
ug | +0.082 |
happened | +0.080 |
Neg logit contributions | |
grate | -0.057 |
Flor | -0.056 |
| -0.053 |
cheer | -0.052 |
wider | -0.050 |
Token | to |
Token ID | 284 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
dd | +0.048 |
ime | +0.045 |
happened | +0.042 |
ug | +0.042 |
endings | +0.041 |
Neg logit contributions | |
Flor | -0.041 |
grate | -0.039 |
violet | -0.037 |
cheer | -0.036 |
| -0.036 |
Token | pedal |
Token ID | 26667 |
Feature activation | -8.59e-01 |
Pos logit contributions | |
dd | +0.019 |
ime | +0.018 |
happened | +0.017 |
ug | +0.017 |
endings | +0.016 |
Neg logit contributions | |
Flor | -0.012 |
violet | -0.011 |
grate | -0.011 |
Av | -0.010 |
mary | -0.010 |
Token | his |
Token ID | 465 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
dd | +0.043 |
ime | +0.040 |
ug | +0.038 |
happened | +0.038 |
endings | +0.038 |
Neg logit contributions | |
Flor | -0.018 |
grate | -0.017 |
| -0.016 |
cheer | -0.016 |
violet | -0.015 |
Token | way |
Token ID | 835 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
dd | +0.041 |
ime | +0.035 |
ug | +0.033 |
happened | +0.032 |
endings | +0.032 |
Neg logit contributions | |
promise | -0.032 |
Flor | -0.032 |
cheer | -0.031 |
grate | -0.031 |
| -0.031 |
Token | back |
Token ID | 736 |
Feature activation | -9.35e-01 |
Pos logit contributions | |
Flor | +0.001 |
grate | +0.001 |
violet | +0.001 |
cheer | +0.001 |
| +0.001 |
Neg logit contributions | |
dd | -0.001 |
ime | -0.001 |
ug | -0.001 |
happened | -0.001 |
endings | -0.001 |
Token | home |
Token ID | 1363 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
dd | +0.038 |
ime | +0.036 |
ug | +0.034 |
happened | +0.033 |
endings | +0.033 |
Neg logit contributions | |
Flor | -0.028 |
grate | -0.025 |
violet | -0.025 |
Av | -0.024 |
cheer | -0.023 |
Token | you |
Token ID | 345 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
Flor | +0.025 |
grate | +0.023 |
| +0.023 |
cheer | +0.023 |
Av | +0.022 |
Neg logit contributions | |
dd | -0.025 |
ime | -0.022 |
ug | -0.021 |
happened | -0.021 |
endings | -0.020 |
Token | feel |
Token ID | 1254 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
violet | +0.015 |
Flor | +0.014 |
grate | +0.013 |
Deal | +0.012 |
Alexa | +0.012 |
Neg logit contributions | |
dd | -0.024 |
ime | -0.024 |
happened | -0.023 |
endings | -0.022 |
ug | -0.022 |
Token | good |
Token ID | 922 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
violet | +0.026 |
Flor | +0.026 |
grate | +0.022 |
count | +0.022 |
bug | +0.021 |
Neg logit contributions | |
dd | -0.053 |
happened | -0.048 |
ime | -0.048 |
endings | -0.046 |
ug | -0.045 |
Token | too |
Token ID | 1165 |
Feature activation | +6.21e-01 |
Pos logit contributions | |
dd | +0.004 |
ime | +0.003 |
happened | +0.003 |
endings | +0.003 |
ug | +0.003 |
Neg logit contributions | |
Flor | -0.002 |
grate | -0.002 |
violet | -0.002 |
| -0.002 |
cheer | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
Flor | +0.016 |
grate | +0.015 |
violet | +0.014 |
cheer | +0.014 |
| +0.013 |
Neg logit contributions | |
dd | -0.028 |
ime | -0.026 |
happened | -0.025 |
ug | -0.025 |
endings | -0.025 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
dd | +0.007 |
ime | +0.007 |
ug | +0.006 |
endings | +0.006 |
happened | +0.006 |
Neg logit contributions | |
grate | -0.006 |
forget | -0.005 |
cheer | -0.005 |
promise | -0.005 |
Flor | -0.005 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
dd | +0.037 |
ime | +0.035 |
happened | +0.034 |
ug | +0.033 |
endings | +0.032 |
Neg logit contributions | |
Flor | -0.021 |
grate | -0.021 |
violet | -0.019 |
cheer | -0.019 |
promise | -0.018 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
Flor | +0.011 |
grate | +0.010 |
violet | +0.010 |
cheer | +0.009 |
Av | +0.008 |
Neg logit contributions | |
dd | -0.022 |
ime | -0.021 |
happened | -0.020 |
ug | -0.020 |
endings | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
dd | +0.058 |
ime | +0.053 |
happened | +0.051 |
endings | +0.050 |
ug | +0.050 |
Neg logit contributions | |
Flor | -0.037 |
grate | -0.036 |
violet | -0.033 |
cheer | -0.033 |
| -0.032 |
Token | time |
Token ID | 640 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
Flor | +0.010 |
grate | +0.009 |
| +0.009 |
violet | +0.009 |
cheer | +0.009 |
Neg logit contributions | |
dd | -0.007 |
ime | -0.006 |
ug | -0.006 |
endings | -0.005 |
happened | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
dd | +0.016 |
ime | +0.015 |
ug | +0.014 |
endings | +0.014 |
happened | +0.014 |
Neg logit contributions | |
Flor | -0.009 |
grate | -0.008 |
| -0.008 |
cheer | -0.007 |
violet | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
Flor | +0.023 |
violet | +0.022 |
grate | +0.022 |
cheer | +0.021 |
Deal | +0.020 |
Neg logit contributions | |
dd | -0.028 |
happened | -0.026 |
ime | -0.025 |
ug | -0.023 |
aw | -0.023 |
Token | goodbye |
Token ID | 24829 |
Feature activation | +9.85e-01 |
Pos logit contributions | |
dd | +0.013 |
ime | +0.012 |
ug | +0.012 |
endings | +0.012 |
happened | +0.012 |
Neg logit contributions | |
Flor | -0.007 |
grate | -0.007 |
| -0.006 |
cheer | -0.006 |
violet | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
Flor | +0.032 |
grate | +0.030 |
violet | +0.029 |
cheer | +0.029 |
| +0.028 |
Neg logit contributions | |
dd | -0.038 |
ime | -0.034 |
ug | -0.032 |
happened | -0.032 |
endings | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
grate | +0.014 |
Flor | +0.014 |
violet | +0.014 |
wider | +0.013 |
Together | +0.013 |
Neg logit contributions | |
dd | -0.015 |
ime | -0.014 |
happened | -0.013 |
ug | -0.013 |
endings | -0.012 |
Token | frog |
Token ID | 21264 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
grate | +0.001 |
Flor | +0.001 |
| +0.001 |
violet | +0.001 |
Together | +0.001 |
Neg logit contributions | |
dd | -0.001 |
ime | -0.001 |
happened | -0.001 |
endings | -0.001 |
emergency | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
Flor | +0.077 |
violet | +0.075 |
Deal | +0.073 |
grate | +0.071 |
cheer | +0.070 |
Neg logit contributions | |
dd | -0.069 |
ime | -0.063 |
ug | -0.060 |
happened | -0.059 |
aw | -0.057 |
Token | promised |
Token ID | 8072 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
dd | +0.020 |
happened | +0.018 |
ime | +0.017 |
ug | +0.017 |
aw | +0.016 |
Neg logit contributions | |
Flor | -0.017 |
violet | -0.017 |
grate | -0.016 |
Deal | -0.016 |
| -0.015 |
Token | to |
Token ID | 284 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
dd | +0.048 |
ime | +0.044 |
ug | +0.042 |
happened | +0.041 |
endings | +0.041 |
Neg logit contributions | |
Flor | -0.034 |
grate | -0.032 |
| -0.031 |
cheer | -0.030 |
violet | -0.030 |
Token | come |
Token ID | 1282 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
dd | +0.018 |
ime | +0.017 |
ug | +0.016 |
endings | +0.015 |
happened | +0.015 |
Neg logit contributions | |
Flor | -0.012 |
violet | -0.011 |
Alexa | -0.010 |
wider | -0.010 |
grate | -0.010 |
Token | back |
Token ID | 736 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
dd | +0.013 |
ug | +0.011 |
away | +0.011 |
ly | +0.011 |
ime | +0.011 |
Neg logit contributions | |
violet | -0.016 |
Alexa | -0.015 |
Flor | -0.015 |
Ms | -0.015 |
Deal | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
dd | +0.015 |
ime | +0.014 |
happened | +0.014 |
ug | +0.013 |
endings | +0.013 |
Neg logit contributions | |
violet | -0.012 |
Flor | -0.011 |
grate | -0.010 |
cheer | -0.010 |
Deal | -0.010 |
Token | pond |
Token ID | 16723 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
dd | +0.006 |
ime | +0.006 |
emergency | +0.006 |
aw | +0.006 |
endings | +0.006 |
Neg logit contributions | |
grate | -0.006 |
Flor | -0.006 |
violet | -0.006 |
| -0.006 |
cheer | -0.006 |
Token | with |
Token ID | 351 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
Flor | +0.060 |
Deal | +0.059 |
violet | +0.057 |
Alexa | +0.052 |
grate | +0.052 |
Neg logit contributions | |
dd | -0.049 |
ime | -0.045 |
ug | -0.044 |
aw | -0.044 |
happened | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
Flor | +0.042 |
violet | +0.040 |
grate | +0.040 |
cheer | +0.037 |
promise | +0.037 |
Neg logit contributions | |
dd | -0.038 |
ime | -0.034 |
happened | -0.032 |
ug | -0.031 |
endings | -0.030 |
Token | bucket |
Token ID | 19236 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
Flor | +0.005 |
grate | +0.005 |
Av | +0.005 |
violet | +0.005 |
Alexa | +0.005 |
Neg logit contributions | |
dd | -0.005 |
ime | -0.004 |
happened | -0.004 |
opter | -0.004 |
ug | -0.004 |
Token | on |
Token ID | 319 |
Feature activation | -5.19e-01 |
Pos logit contributions | |
dd | +0.034 |
ime | +0.031 |
ug | +0.030 |
happened | +0.030 |
endings | +0.028 |
Neg logit contributions | |
Flor | -0.029 |
grate | -0.026 |
violet | -0.025 |
cheer | -0.025 |
| -0.025 |
Token | his |
Token ID | 465 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
dd | +0.021 |
ime | +0.020 |
ug | +0.019 |
happened | +0.019 |
endings | +0.018 |
Neg logit contributions | |
Flor | -0.015 |
grate | -0.014 |
violet | -0.014 |
cheer | -0.014 |
| -0.013 |
Token | back |
Token ID | 736 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
dd | +0.059 |
ime | +0.054 |
ug | +0.051 |
endings | +0.051 |
happened | +0.051 |
Neg logit contributions | |
Flor | -0.043 |
grate | -0.040 |
violet | -0.038 |
| -0.038 |
cheer | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
ime | +0.017 |
dd | +0.016 |
ug | +0.015 |
endings | +0.014 |
happened | +0.014 |
Neg logit contributions | |
Flor | -0.016 |
grate | -0.015 |
violet | -0.015 |
Av | -0.014 |
cheer | -0.013 |
Token | tried |
Token ID | 3088 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
Flor | +0.022 |
violet | +0.021 |
Deal | +0.020 |
grate | +0.020 |
Av | +0.018 |
Neg logit contributions | |
dd | -0.025 |
happened | -0.023 |
ime | -0.021 |
endings | -0.021 |
ug | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
Flor | +0.005 |
grate | +0.004 |
violet | +0.004 |
cheer | +0.004 |
promise | +0.004 |
Neg logit contributions | |
dd | -0.004 |
ime | -0.004 |
happened | -0.004 |
ug | -0.003 |
endings | -0.003 |
Token | carry |
Token ID | 3283 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
Flor | +0.010 |
violet | +0.009 |
grate | +0.009 |
| +0.009 |
Av | +0.009 |
Neg logit contributions | |
dd | -0.017 |
ime | -0.016 |
ug | -0.015 |
happened | -0.014 |
endings | -0.014 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Flor | +0.009 |
grate | +0.009 |
violet | +0.008 |
cheer | +0.008 |
promise | +0.008 |
Neg logit contributions | |
dd | -0.015 |
ime | -0.013 |
happened | -0.013 |
endings | -0.013 |
ug | -0.013 |
Token | said |
Token ID | 531 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Flor | +0.030 |
violet | +0.028 |
grate | +0.028 |
cheer | +0.026 |
refreshed | +0.026 |
Neg logit contributions | |
dd | -0.042 |
ime | -0.039 |
happened | -0.038 |
ug | -0.036 |
aw | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
dd | +0.052 |
ime | +0.048 |
happened | +0.046 |
ug | +0.046 |
endings | +0.044 |
Neg logit contributions | |
Flor | -0.032 |
violet | -0.030 |
grate | -0.030 |
cheer | -0.028 |
| -0.027 |
Token | She |
Token ID | 1375 |
Feature activation | +9.99e-01 |
Pos logit contributions | |
dd | +0.002 |
ime | +0.001 |
happened | +0.001 |
ug | +0.001 |
endings | +0.001 |
Neg logit contributions | |
grate | -0.001 |
Flor | -0.001 |
promise | -0.001 |
forget | -0.001 |
refreshed | -0.001 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
Flor | +0.030 |
grate | +0.028 |
violet | +0.026 |
cheer | +0.026 |
| +0.026 |
Neg logit contributions | |
dd | -0.041 |
ime | -0.038 |
happened | -0.036 |
ug | -0.035 |
endings | -0.035 |
Token | something |
Token ID | 1223 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
dd | +0.090 |
ime | +0.082 |
ug | +0.079 |
endings | +0.078 |
happened | +0.077 |
Neg logit contributions | |
Flor | -0.048 |
grate | -0.041 |
| -0.041 |
cheer | -0.040 |
violet | -0.039 |
Token | shiny |
Token ID | 22441 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
dd | +0.026 |
ime | +0.023 |
happened | +0.023 |
ug | +0.022 |
endings | +0.022 |
Neg logit contributions | |
Flor | -0.022 |
grate | -0.021 |
| -0.020 |
violet | -0.020 |
cheer | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
dd | +0.081 |
ime | +0.076 |
ug | +0.073 |
endings | +0.072 |
opter | +0.068 |
Neg logit contributions | |
Flor | -0.042 |
grate | -0.041 |
violet | -0.038 |
cheer | -0.037 |
closer | -0.037 |
Token | round |
Token ID | 2835 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
Flor | +0.039 |
grate | +0.037 |
join | +0.035 |
violet | +0.035 |
cheer | +0.034 |
Neg logit contributions | |
dd | -0.034 |
happened | -0.032 |
ime | -0.030 |
endings | -0.028 |
aw | -0.027 |
Token | in |
Token ID | 287 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
dd | +0.093 |
ime | +0.086 |
ug | +0.083 |
endings | +0.081 |
happened | +0.080 |
Neg logit contributions | |
Flor | -0.045 |
grate | -0.043 |
violet | -0.039 |
| -0.038 |
cheer | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
Flor | +0.028 |
grate | +0.027 |
violet | +0.025 |
cheer | +0.025 |
| +0.024 |
Neg logit contributions | |
dd | -0.042 |
ime | -0.039 |
happened | -0.037 |
ug | -0.036 |
endings | -0.036 |
Token | From |
Token ID | 3574 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
Flor | +0.004 |
grate | +0.004 |
cheer | +0.004 |
violet | +0.004 |
promise | +0.004 |
Neg logit contributions | |
dd | -0.005 |
ime | -0.005 |
endings | -0.005 |
happened | -0.005 |
ug | -0.004 |
Token | that |
Token ID | 326 |
Feature activation | +4.85e-01 |
Pos logit contributions | |
Flor | +0.012 |
grate | +0.011 |
| +0.010 |
cheer | +0.010 |
violet | +0.010 |
Neg logit contributions | |
dd | -0.023 |
ime | -0.023 |
endings | -0.022 |
ug | -0.021 |
happened | -0.021 |
Token | day |
Token ID | 1110 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
Flor | +0.018 |
grate | +0.017 |
violet | +0.016 |
cheer | +0.016 |
Av | +0.016 |
Neg logit contributions | |
dd | -0.017 |
ime | -0.015 |
happened | -0.014 |
ug | -0.014 |
endings | -0.014 |
Token | on |
Token ID | 319 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
dd | +0.020 |
endings | +0.019 |
ime | +0.019 |
happened | +0.018 |
ug | +0.017 |
Neg logit contributions | |
| -0.009 |
grate | -0.009 |
Flor | -0.008 |
cheer | -0.008 |
better | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
dd | +0.032 |
ime | +0.029 |
endings | +0.028 |
ug | +0.028 |
happened | +0.027 |
Neg logit contributions | |
| -0.011 |
Flor | -0.011 |
grate | -0.010 |
cheer | -0.009 |
Molly | -0.009 |
Token | Tom |
Token ID | 4186 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Flor | +0.004 |
grate | +0.004 |
| +0.004 |
cheer | +0.004 |
violet | +0.004 |
Neg logit contributions | |
dd | -0.004 |
ime | -0.004 |
ug | -0.004 |
happened | -0.004 |
endings | -0.004 |
Token | steered |
Token ID | 49300 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
dd | +0.044 |
ime | +0.041 |
ug | +0.039 |
endings | +0.039 |
happened | +0.038 |
Neg logit contributions | |
Flor | -0.030 |
grate | -0.028 |
| -0.026 |
violet | -0.026 |
cheer | -0.026 |
Token | clear |
Token ID | 1598 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
Flor | +0.010 |
grate | +0.009 |
cheer | +0.009 |
| +0.009 |
violet | +0.009 |
Neg logit contributions | |
dd | -0.013 |
ime | -0.011 |
happened | -0.011 |
endings | -0.011 |
ug | -0.011 |
Token | of |
Token ID | 286 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
dd | +0.002 |
ime | +0.002 |
happened | +0.002 |
endings | +0.002 |
opter | +0.002 |
Neg logit contributions | |
Flor | -0.001 |
grate | -0.001 |
cheer | -0.001 |
| -0.001 |
join | -0.001 |
Token | p |
Token ID | 279 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
Flor | +0.002 |
grate | +0.002 |
cheer | +0.002 |
| +0.002 |
violet | +0.002 |
Neg logit contributions | |
dd | -0.003 |
ime | -0.002 |
ug | -0.002 |
endings | -0.002 |
happened | -0.002 |
Token | udd |
Token ID | 4185 |
Feature activation | -6.86e-01 |
Pos logit contributions | |
Flor | +0.036 |
grate | +0.034 |
violet | +0.033 |
cheer | +0.033 |
| +0.033 |
Neg logit contributions | |
dd | -0.025 |
ime | -0.022 |
ug | -0.021 |
happened | -0.020 |
endings | -0.020 |
Token | playing |
Token ID | 2712 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
dd | +0.061 |
ime | +0.058 |
ug | +0.056 |
opter | +0.055 |
endings | +0.054 |
Neg logit contributions | |
Flor | -0.022 |
| -0.021 |
talk | -0.020 |
applaud | -0.020 |
tell | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
dd | +0.037 |
ime | +0.036 |
endings | +0.034 |
happened | +0.034 |
ug | +0.033 |
Neg logit contributions | |
grate | -0.015 |
Flor | -0.015 |
cheer | -0.014 |
join | -0.013 |
| -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
Flor | +0.005 |
| +0.005 |
cheer | +0.005 |
grate | +0.004 |
violet | +0.004 |
Neg logit contributions | |
dd | -0.010 |
ime | -0.009 |
ug | -0.009 |
endings | -0.009 |
happened | -0.009 |
Token | mud |
Token ID | 17492 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
dd | +0.051 |
ug | +0.048 |
ime | +0.046 |
endings | +0.045 |
happened | +0.044 |
Neg logit contributions | |
| -0.023 |
Flor | -0.021 |
refreshed | -0.020 |
better | -0.020 |
violet | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
dd | +0.041 |
ime | +0.038 |
ug | +0.037 |
endings | +0.036 |
happened | +0.036 |
Neg logit contributions | |
Flor | -0.033 |
grate | -0.030 |
violet | -0.030 |
cheer | -0.029 |
| -0.028 |
Token | dirt |
Token ID | 13647 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
dd | +0.004 |
ime | +0.004 |
happened | +0.004 |
ug | +0.004 |
endings | +0.004 |
Neg logit contributions | |
Flor | -0.003 |
grate | -0.003 |
violet | -0.003 |
Av | -0.003 |
cheer | -0.003 |
Token | all |
Token ID | 477 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
dd | +0.181 |
ime | +0.167 |
ug | +0.159 |
happened | +0.158 |
endings | +0.158 |
Neg logit contributions | |
Flor | -0.097 |
grate | -0.092 |
| -0.085 |
cheer | -0.084 |
violet | -0.084 |
Token | day |
Token ID | 1110 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
dd | +0.044 |
ime | +0.040 |
happened | +0.038 |
ug | +0.038 |
endings | +0.036 |
Neg logit contributions | |
Flor | -0.039 |
grate | -0.037 |
violet | -0.035 |
cheer | -0.034 |
| -0.034 |
Token | long |
Token ID | 890 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
dd | +0.060 |
ime | +0.056 |
endings | +0.054 |
ug | +0.053 |
happened | +0.052 |
Neg logit contributions | |
Flor | -0.047 |
grate | -0.047 |
| -0.046 |
refreshed | -0.043 |
violet | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
dd | +0.208 |
ime | +0.197 |
ug | +0.188 |
opter | +0.187 |
endings | +0.185 |
Neg logit contributions | |
Flor | -0.055 |
| -0.055 |
grate | -0.054 |
cheer | -0.042 |
too | -0.040 |
Token | He |
Token ID | 679 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
dd | +0.002 |
ime | +0.002 |
happened | +0.002 |
ug | +0.002 |
endings | +0.002 |
Neg logit contributions | |
Flor | -0.001 |
grate | -0.001 |
cheer | -0.001 |
violet | -0.001 |
Av | -0.001 |
Token | it |
Token ID | 340 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
Flor | +0.032 |
grate | +0.031 |
violet | +0.029 |
cheer | +0.028 |
| +0.028 |
Neg logit contributions | |
dd | -0.044 |
ime | -0.041 |
happened | -0.039 |
ug | -0.039 |
endings | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
Flor | +0.015 |
violet | +0.013 |
grate | +0.013 |
cheer | +0.013 |
Av | +0.013 |
Neg logit contributions | |
dd | -0.018 |
ime | -0.016 |
ug | -0.016 |
happened | -0.016 |
endings | -0.015 |
Token | covered |
Token ID | 5017 |
Feature activation | -7.40e-02 |
Pos logit contributions | |
Flor | +0.011 |
grate | +0.010 |
violet | +0.010 |
| +0.010 |
Av | +0.010 |
Neg logit contributions | |
dd | -0.015 |
happened | -0.014 |
ime | -0.013 |
endings | -0.013 |
ug | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
dd | +0.003 |
ime | +0.003 |
happened | +0.003 |
ug | +0.002 |
endings | +0.002 |
Neg logit contributions | |
Flor | -0.002 |
grate | -0.002 |
| -0.002 |
cheer | -0.002 |
violet | -0.002 |
Token | with |
Token ID | 351 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
dd | +0.011 |
ime | +0.010 |
ly | +0.010 |
ug | +0.009 |
endings | +0.009 |
Neg logit contributions | |
violet | -0.011 |
Flor | -0.011 |
Deal | -0.010 |
Ms | -0.010 |
Alicia | -0.010 |
Token | dirt |
Token ID | 13647 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
Flor | +0.045 |
grate | +0.043 |
cheer | +0.041 |
| +0.041 |
violet | +0.040 |
Neg logit contributions | |
dd | -0.056 |
ime | -0.050 |
ug | -0.048 |
happened | -0.047 |
endings | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
dd | +0.149 |
ime | +0.136 |
happened | +0.131 |
ug | +0.131 |
endings | +0.128 |
Neg logit contributions | |
Flor | -0.097 |
grate | -0.089 |
violet | -0.086 |
| -0.085 |
cheer | -0.084 |
Token | They |
Token ID | 1119 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
dd | +0.024 |
ime | +0.022 |
ug | +0.022 |
happened | +0.021 |
endings | +0.021 |
Neg logit contributions | |
Flor | -0.014 |
grate | -0.014 |
cheer | -0.013 |
violet | -0.013 |
promise | -0.013 |
Token | were |
Token ID | 547 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
Flor | +0.007 |
violet | +0.007 |
grate | +0.007 |
| +0.007 |
Av | +0.006 |
Neg logit contributions | |
dd | -0.009 |
happened | -0.008 |
ime | -0.008 |
ug | -0.007 |
aw | -0.007 |
Token | happy |
Token ID | 3772 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
dd | +0.027 |
happened | +0.023 |
ime | +0.023 |
ug | +0.023 |
aw | +0.022 |
Neg logit contributions | |
Flor | -0.016 |
violet | -0.015 |
Av | -0.015 |
grate | -0.014 |
| -0.014 |
Token | with |
Token ID | 351 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
Flor | +0.029 |
grate | +0.027 |
violet | +0.027 |
| +0.026 |
cheer | +0.025 |
Neg logit contributions | |
dd | -0.038 |
ime | -0.035 |
happened | -0.035 |
endings | -0.034 |
ug | -0.033 |
Token | was |
Token ID | 373 |
Feature activation | -9.90e-01 |
Pos logit contributions | |
Flor | +0.001 |
| +0.001 |
grate | +0.001 |
violet | +0.001 |
Av | +0.001 |
Neg logit contributions | |
dd | -0.001 |
ime | -0.001 |
ug | -0.001 |
endings | -0.001 |
anked | -0.001 |
Token | just |
Token ID | 655 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
dd | +0.041 |
ime | +0.038 |
happened | +0.037 |
ug | +0.037 |
emergency | +0.035 |
Neg logit contributions | |
grate | -0.026 |
Flor | -0.025 |
promise | -0.023 |
violet | -0.023 |
cheer | -0.023 |
Token | three |
Token ID | 1115 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
grate | +0.030 |
Flor | +0.030 |
| +0.028 |
cheer | +0.027 |
violet | +0.027 |
Neg logit contributions | |
dd | -0.040 |
ime | -0.037 |
happened | -0.036 |
ug | -0.035 |
endings | -0.034 |
Token | - |
Token ID | 12 |
Feature activation | -4.85e+00 |
Pos logit contributions | |
dd | +0.025 |
ime | +0.022 |
happened | +0.021 |
endings | +0.021 |
ug | +0.021 |
Neg logit contributions | |
Flor | -0.024 |
grate | -0.023 |
| -0.022 |
cheer | -0.022 |
violet | -0.021 |
Token | years |
Token ID | 19002 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
dd | +0.187 |
ime | +0.175 |
happened | +0.168 |
endings | +0.167 |
ug | +0.162 |
Neg logit contributions | |
Flor | -0.154 |
grate | -0.146 |
violet | -0.138 |
cheer | -0.137 |
| -0.135 |
Token | - |
Token ID | 12 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
dd | +0.097 |
happened | +0.083 |
endings | +0.077 |
ime | +0.077 |
Stop | +0.070 |
Neg logit contributions | |
grate | -0.151 |
Flor | -0.146 |
cheer | -0.133 |
Av | -0.133 |
promise | -0.132 |
Token | old |
Token ID | 727 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
dd | +0.107 |
ime | +0.095 |
happened | +0.090 |
endings | +0.088 |
ug | +0.085 |
Neg logit contributions | |
Flor | -0.195 |
grate | -0.188 |
violet | -0.179 |
cheer | -0.179 |
| -0.179 |
Token | and |
Token ID | 290 |
Feature activation | -6.85e-01 |
Pos logit contributions | |
dd | +0.033 |
ime | +0.030 |
ug | +0.029 |
happened | +0.029 |
endings | +0.029 |
Neg logit contributions | |
Flor | -0.024 |
grate | -0.023 |
violet | -0.022 |
cheer | -0.021 |
| -0.021 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
dd | +0.019 |
happened | +0.016 |
ime | +0.015 |
ug | +0.014 |
endings | +0.014 |
Neg logit contributions | |
Flor | -0.029 |
grate | -0.028 |
violet | -0.027 |
cheer | -0.026 |
wider | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
dd | +0.051 |
ime | +0.047 |
ug | +0.045 |
endings | +0.043 |
happened | +0.043 |
Neg logit contributions | |
Flor | -0.029 |
grate | -0.027 |
| -0.026 |
violet | -0.026 |
cheer | -0.025 |
Token | explore |
Token ID | 7301 |
Feature activation | -8.14e-01 |
Pos logit contributions | |
dd | +0.041 |
ime | +0.039 |
happened | +0.037 |
endings | +0.036 |
ug | +0.036 |
Neg logit contributions | |
Flor | -0.019 |
grate | -0.017 |
violet | -0.015 |
| -0.015 |
cheer | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
dd | +0.020 |
ime | +0.019 |
ug | +0.018 |
endings | +0.018 |
opter | +0.016 |
Neg logit contributions | |
Flor | -0.013 |
| -0.012 |
grate | -0.012 |
cheer | -0.011 |
violet | -0.011 |
Token | very |
Token ID | 845 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
dd | +0.053 |
ime | +0.049 |
endings | +0.047 |
ug | +0.047 |
happened | +0.047 |
Neg logit contributions | |
Flor | -0.025 |
grate | -0.023 |
| -0.022 |
violet | -0.021 |
cheer | -0.021 |
Token | busy |
Token ID | 8179 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
dd | +0.003 |
ime | +0.002 |
endings | +0.002 |
happened | +0.002 |
ug | +0.002 |
Neg logit contributions | |
Flor | -0.002 |
grate | -0.001 |
| -0.001 |
cheer | -0.001 |
violet | -0.001 |
Token | all |
Token ID | 477 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
dd | +0.094 |
ime | +0.086 |
endings | +0.083 |
happened | +0.082 |
opter | +0.081 |
Neg logit contributions | |
Flor | -0.039 |
grate | -0.038 |
| -0.035 |
cheer | -0.034 |
count | -0.034 |
Token | day |
Token ID | 1110 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
dd | +0.051 |
ime | +0.046 |
endings | +0.045 |
ug | +0.044 |
happened | +0.044 |
Neg logit contributions | |
Flor | -0.036 |
grate | -0.035 |
| -0.034 |
cheer | -0.034 |
violet | -0.032 |
Token | long |
Token ID | 890 |
Feature activation | -4.28e+00 |
Pos logit contributions | |
dd | +0.049 |
ime | +0.048 |
endings | +0.047 |
ug | +0.044 |
happened | +0.044 |
Neg logit contributions | |
| -0.032 |
grate | -0.032 |
Flor | -0.031 |
refreshed | -0.028 |
violet | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
dd | +0.234 |
ime | +0.225 |
opter | +0.215 |
endings | +0.210 |
ug | +0.207 |
Neg logit contributions | |
| -0.061 |
Flor | -0.058 |
grate | -0.058 |
too | -0.049 |
cheer | -0.043 |
Token | He |
Token ID | 679 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
dd | +0.007 |
ime | +0.006 |
happened | +0.006 |
ug | +0.006 |
endings | +0.006 |
Neg logit contributions | |
Flor | -0.006 |
grate | -0.005 |
cheer | -0.005 |
violet | -0.005 |
Av | -0.005 |
Token | would |
Token ID | 561 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
ime | +0.043 |
dd | +0.043 |
endings | +0.042 |
ug | +0.040 |
anked | +0.040 |
Neg logit contributions | |
Flor | -0.025 |
grate | -0.024 |
| -0.024 |
count | -0.022 |
also | -0.022 |
Token | run |
Token ID | 1057 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
dd | +0.039 |
ime | +0.036 |
ug | +0.034 |
happened | +0.034 |
endings | +0.033 |
Neg logit contributions | |
Flor | -0.023 |
grate | -0.021 |
violet | -0.020 |
| -0.020 |
Av | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
Flor | +0.003 |
grate | +0.003 |
| +0.003 |
cheer | +0.003 |
violet | +0.002 |
Neg logit contributions | |
dd | -0.005 |
ime | -0.004 |
endings | -0.004 |
happened | -0.004 |
ug | -0.004 |
Token | got |
Token ID | 1392 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
dd | +0.006 |
happened | +0.006 |
ime | +0.005 |
ug | +0.005 |
endings | +0.005 |
Neg logit contributions | |
Flor | -0.005 |
violet | -0.005 |
grate | -0.004 |
Deal | -0.004 |
cheer | -0.004 |
Token | his |
Token ID | 465 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
dd | +0.028 |
ime | +0.026 |
happened | +0.025 |
ug | +0.024 |
endings | +0.024 |
Neg logit contributions | |
Flor | -0.022 |
violet | -0.020 |
grate | -0.020 |
cheer | -0.018 |
| -0.018 |
Token | journal |
Token ID | 3989 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
dd | +0.055 |
ime | +0.052 |
happened | +0.050 |
endings | +0.049 |
ug | +0.048 |
Neg logit contributions | |
Flor | -0.043 |
grate | -0.041 |
violet | -0.038 |
cheer | -0.037 |
| -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
dd | +0.028 |
ime | +0.026 |
happened | +0.026 |
ug | +0.025 |
endings | +0.025 |
Neg logit contributions | |
Flor | -0.020 |
violet | -0.018 |
grate | -0.017 |
| -0.016 |
Av | -0.016 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
dd | +0.011 |
happened | +0.011 |
ime | +0.010 |
endings | +0.010 |
ug | +0.009 |
Neg logit contributions | |
Flor | -0.010 |
violet | -0.009 |
grate | -0.009 |
cheer | -0.008 |
Av | -0.008 |
Token | ay |
Token ID | 323 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
dd | +0.046 |
endings | +0.041 |
happened | +0.040 |
emergency | +0.038 |
strangers | +0.037 |
Neg logit contributions | |
grate | -0.055 |
wider | -0.048 |
sill | -0.048 |
Together | -0.047 |
cheer | -0.047 |
Token | ons |
Token ID | 684 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
dd | +0.112 |
ime | +0.095 |
endings | +0.092 |
ug | +0.089 |
emergency | +0.084 |
Neg logit contributions | |
| -0.128 |
grate | -0.128 |
Flor | -0.124 |
cheer | -0.124 |
kiss | -0.121 |
Token | . |
Token ID | 13 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
dd | +0.072 |
ime | +0.066 |
ug | +0.063 |
happened | +0.063 |
endings | +0.062 |
Neg logit contributions | |
Flor | -0.040 |
grate | -0.038 |
| -0.035 |
cheer | -0.035 |
violet | -0.035 |
Token | They |
Token ID | 1119 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
| +0.006 |
grate | +0.005 |
Together | +0.005 |
Flor | +0.005 |
Av | +0.005 |
Neg logit contributions | |
dd | -0.011 |
ime | -0.010 |
ug | -0.009 |
happened | -0.009 |
endings | -0.009 |
Token | drew |
Token ID | 9859 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
Flor | +0.009 |
grate | +0.009 |
violet | +0.008 |
cheer | +0.008 |
| +0.008 |
Neg logit contributions | |
dd | -0.012 |
ime | -0.011 |
happened | -0.011 |
ug | -0.010 |
endings | -0.010 |
Token | pictures |
Token ID | 5986 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
Flor | +0.023 |
grate | +0.022 |
| +0.021 |
cheer | +0.021 |
violet | +0.020 |
Neg logit contributions | |
dd | -0.030 |
ime | -0.027 |
ug | -0.025 |
endings | -0.025 |
happened | -0.024 |