Token | a |
Token ID | 257 |
Feature activation | +5.49e-03 |
Pos logit contributions | |
nob | +0.015 |
dial | +0.014 |
peek | +0.014 |
Looking | +0.013 |
peaked | +0.013 |
Neg logit contributions | |
rier | -0.010 |
working | -0.010 |
istered | -0.009 |
bored | -0.008 |
nam | -0.008 |
Token | fun |
Token ID | 1257 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
working | +0.007 |
rier | +0.007 |
istered | +0.007 |
kicker | +0.006 |
runner | +0.006 |
Neg logit contributions | |
nob | -0.009 |
ad | -0.009 |
dial | -0.009 |
ered | -0.008 |
Looking | -0.008 |
Token | ride |
Token ID | 6594 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
Looking | +0.045 |
ran | +0.045 |
nob | +0.045 |
dial | +0.044 |
ad | +0.044 |
Neg logit contributions | |
working | -0.027 |
runner | -0.023 |
guys | -0.021 |
kicker | -0.021 |
rainy | -0.019 |
Token | at |
Token ID | 379 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
istered | +0.040 |
rier | +0.040 |
ishy | +0.037 |
working | +0.036 |
ivery | +0.036 |
Neg logit contributions | |
peek | -0.013 |
nob | -0.012 |
called | -0.011 |
dial | -0.011 |
ad | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
rier | +0.023 |
istered | +0.023 |
working | +0.022 |
ishy | +0.021 |
attered | +0.020 |
Neg logit contributions | |
peek | -0.012 |
dial | -0.011 |
nob | -0.011 |
ad | -0.010 |
turned | -0.010 |
Token | carn |
Token ID | 18466 |
Feature activation | +8.14e-02 |
Pos logit contributions | |
rier | +0.031 |
istered | +0.031 |
working | +0.030 |
ishy | +0.028 |
ivery | +0.027 |
Neg logit contributions | |
peek | -0.018 |
dial | -0.018 |
turned | -0.016 |
nob | -0.016 |
ad | -0.015 |
Token | ival |
Token ID | 2473 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
rier | +0.125 |
working | +0.119 |
istered | +0.117 |
ishy | +0.106 |
goals | +0.101 |
Neg logit contributions | |
peek | -0.130 |
nob | -0.122 |
dial | -0.119 |
ad | -0.114 |
Looking | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
istered | +0.063 |
rier | +0.062 |
ishy | +0.060 |
working | +0.058 |
ivery | +0.058 |
Neg logit contributions | |
peek | -0.024 |
dial | -0.022 |
nob | -0.021 |
turned | -0.019 |
called | -0.019 |
Token | Twe |
Token ID | 24205 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
nob | +0.012 |
dial | +0.012 |
peek | +0.011 |
ad | +0.011 |
fore | +0.010 |
Neg logit contributions | |
rier | -0.019 |
working | -0.019 |
istered | -0.018 |
bored | -0.017 |
ishy | -0.017 |
Token | ety |
Token ID | 2963 |
Feature activation | +5.44e-03 |
Pos logit contributions | |
istered | +0.026 |
rier | +0.026 |
working | +0.026 |
ishy | +0.024 |
attered | +0.022 |
Neg logit contributions | |
dial | -0.034 |
peek | -0.033 |
nob | -0.032 |
turned | -0.031 |
Looking | -0.031 |
Token | thought |
Token ID | 1807 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
working | +0.011 |
rier | +0.011 |
istered | +0.011 |
ishy | +0.010 |
goals | +0.010 |
Neg logit contributions | |
dial | -0.005 |
peek | -0.005 |
nob | -0.005 |
ad | -0.005 |
Looking | -0.004 |
Token | She |
Token ID | 1375 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
rier | +0.023 |
working | +0.023 |
istered | +0.023 |
ishy | +0.020 |
goals | +0.020 |
Neg logit contributions | |
peek | -0.019 |
dial | -0.019 |
nob | -0.018 |
turned | -0.018 |
Looking | -0.018 |
Token | loved |
Token ID | 6151 |
Feature activation | +5.05e-03 |
Pos logit contributions | |
dial | +0.021 |
nob | +0.021 |
peek | +0.021 |
ad | +0.019 |
Looking | +0.019 |
Neg logit contributions | |
working | -0.020 |
istered | -0.019 |
rier | -0.019 |
ishy | -0.017 |
goals | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | +8.61e-03 |
Pos logit contributions | |
rier | +0.011 |
istered | +0.011 |
working | +0.010 |
ishy | +0.010 |
goals | +0.009 |
Neg logit contributions | |
dial | -0.005 |
peek | -0.005 |
nob | -0.004 |
turned | -0.004 |
ad | -0.004 |
Token | eat |
Token ID | 4483 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
working | +0.015 |
istered | +0.014 |
rier | +0.014 |
ishy | +0.013 |
goals | +0.013 |
Neg logit contributions | |
dial | -0.011 |
nob | -0.010 |
peek | -0.010 |
Looking | -0.010 |
Turning | -0.010 |
Token | av |
Token ID | 1196 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
rier | +0.030 |
istered | +0.030 |
working | +0.029 |
kicker | +0.026 |
ishy | +0.025 |
Neg logit contributions | |
peek | -0.031 |
dial | -0.031 |
turned | -0.029 |
nob | -0.029 |
ad | -0.028 |
Token | oc |
Token ID | 420 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
rier | +0.086 |
istered | +0.083 |
working | +0.080 |
ishy | +0.073 |
sweaty | +0.072 |
Neg logit contributions | |
dial | -0.055 |
peek | -0.055 |
nob | -0.054 |
ad | -0.051 |
turned | -0.048 |
Token | ados |
Token ID | 22484 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
istered | +0.089 |
rier | +0.078 |
working | +0.075 |
ishy | +0.072 |
ivery | +0.071 |
Neg logit contributions | |
ad | -0.152 |
dial | -0.147 |
nob | -0.144 |
peek | -0.139 |
p | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | +9.02e-03 |
Pos logit contributions | |
istered | +0.029 |
rier | +0.029 |
working | +0.027 |
ishy | +0.026 |
goals | +0.025 |
Neg logit contributions | |
peek | -0.017 |
dial | -0.017 |
nob | -0.016 |
turned | -0.015 |
ad | -0.015 |
Token | One |
Token ID | 1881 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
rier | +0.016 |
working | +0.016 |
istered | +0.016 |
ishy | +0.014 |
goals | +0.014 |
Neg logit contributions | |
dial | -0.011 |
peek | -0.011 |
nob | -0.011 |
turned | -0.010 |
ad | -0.010 |
Token | day |
Token ID | 1110 |
Feature activation | +1.54e-03 |
Pos logit contributions | |
rier | +0.024 |
istered | +0.023 |
working | +0.022 |
goals | +0.020 |
ishy | +0.020 |
Neg logit contributions | |
peek | -0.021 |
dial | -0.021 |
nob | -0.020 |
turned | -0.020 |
ad | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +6.96e-03 |
Pos logit contributions | |
rier | +0.003 |
istered | +0.003 |
working | +0.003 |
ishy | +0.003 |
goals | +0.003 |
Neg logit contributions | |
peek | -0.002 |
dial | -0.002 |
nob | -0.001 |
turned | -0.001 |
Looking | -0.001 |
Token | shape |
Token ID | 5485 |
Feature activation | -8.52e-03 |
Pos logit contributions | |
dial | +0.022 |
nob | +0.022 |
peek | +0.021 |
Looking | +0.020 |
ad | +0.020 |
Neg logit contributions | |
rier | -0.026 |
working | -0.026 |
istered | -0.024 |
ishy | -0.023 |
goals | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
dial | +0.010 |
peek | +0.010 |
nob | +0.010 |
Looking | +0.009 |
turned | +0.009 |
Neg logit contributions | |
rier | -0.016 |
working | -0.015 |
istered | -0.015 |
ishy | -0.014 |
goals | -0.014 |
Token | |
Token ID | 220 |
Feature activation | +4.46e-03 |
Pos logit contributions | |
rier | +0.045 |
working | +0.043 |
istered | +0.042 |
goals | +0.039 |
nam | +0.037 |
Neg logit contributions | |
dial | -0.038 |
peek | -0.038 |
nob | -0.037 |
ad | -0.036 |
turned | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
rier | +0.011 |
istered | +0.010 |
working | +0.010 |
goals | +0.010 |
ishy | +0.010 |
Neg logit contributions | |
dial | -0.003 |
peek | -0.003 |
nob | -0.003 |
turned | -0.002 |
ad | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
rier | +0.036 |
working | +0.036 |
goals | +0.033 |
istered | +0.033 |
kicker | +0.032 |
Neg logit contributions | |
dial | -0.017 |
nob | -0.016 |
turned | -0.016 |
peek | -0.015 |
ad | -0.015 |
Token | C |
Token ID | 34 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
rier | +0.036 |
working | +0.036 |
istered | +0.034 |
goals | +0.032 |
kicker | +0.031 |
Neg logit contributions | |
dial | -0.025 |
peek | -0.025 |
nob | -0.025 |
turned | -0.024 |
ad | -0.023 |
Token | urious |
Token ID | 16421 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
rier | +0.109 |
istered | +0.105 |
working | +0.102 |
ishy | +0.091 |
sweaty | +0.089 |
Neg logit contributions | |
peek | -0.130 |
dial | -0.129 |
nob | -0.125 |
ad | -0.121 |
turned | -0.119 |
Token | , |
Token ID | 11 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
rier | +0.027 |
istered | +0.027 |
working | +0.027 |
ishy | +0.025 |
goals | +0.024 |
Neg logit contributions | |
peek | -0.015 |
dial | -0.015 |
nob | -0.014 |
turned | -0.013 |
ad | -0.013 |
Token | Amelia |
Token ID | 49769 |
Feature activation | -2.00e-04 |
Pos logit contributions | |
rier | +0.018 |
working | +0.018 |
istered | +0.018 |
ishy | +0.016 |
goals | +0.016 |
Neg logit contributions | |
dial | -0.018 |
peek | -0.018 |
nob | -0.017 |
turned | -0.016 |
ad | -0.016 |
Token | tried |
Token ID | 3088 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
nob | +0.000 |
dial | +0.000 |
Looking | +0.000 |
ad | +0.000 |
Turning | +0.000 |
Neg logit contributions | |
working | -0.000 |
istered | -0.000 |
rier | -0.000 |
ishy | -0.000 |
goals | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
dial | +0.028 |
nob | +0.028 |
izer | +0.027 |
ad | +0.026 |
Looking | +0.026 |
Neg logit contributions | |
working | -0.034 |
bored | -0.031 |
goals | -0.031 |
rier | -0.030 |
istered | -0.029 |
Token | She |
Token ID | 1375 |
Feature activation | -7.12e-03 |
Pos logit contributions | |
rier | +0.020 |
working | +0.019 |
istered | +0.019 |
ishy | +0.017 |
goals | +0.017 |
Neg logit contributions | |
peek | -0.016 |
dial | -0.016 |
nob | -0.015 |
turned | -0.015 |
Looking | -0.015 |
Token | loved |
Token ID | 6151 |
Feature activation | +7.25e-03 |
Pos logit contributions | |
dial | +0.011 |
nob | +0.011 |
peek | +0.011 |
ad | +0.010 |
Looking | +0.010 |
Neg logit contributions | |
working | -0.010 |
istered | -0.010 |
rier | -0.010 |
ishy | -0.009 |
goals | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | +8.24e-03 |
Pos logit contributions | |
rier | +0.016 |
istered | +0.015 |
working | +0.015 |
ishy | +0.014 |
goals | +0.014 |
Neg logit contributions | |
dial | -0.007 |
peek | -0.006 |
nob | -0.006 |
turned | -0.006 |
ad | -0.005 |
Token | eat |
Token ID | 4483 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
working | +0.014 |
istered | +0.012 |
rier | +0.012 |
ishy | +0.012 |
goals | +0.012 |
Neg logit contributions | |
dial | -0.011 |
nob | -0.011 |
peek | -0.010 |
Looking | -0.010 |
Turning | -0.010 |
Token | av |
Token ID | 1196 |
Feature activation | +4.44e-02 |
Pos logit contributions | |
rier | +0.030 |
istered | +0.029 |
working | +0.029 |
kicker | +0.025 |
ivery | +0.025 |
Neg logit contributions | |
peek | -0.030 |
dial | -0.030 |
turned | -0.028 |
nob | -0.028 |
ad | -0.027 |
Token | oc |
Token ID | 420 |
Feature activation | +7.61e-02 |
Pos logit contributions | |
rier | +0.083 |
istered | +0.080 |
working | +0.076 |
ishy | +0.070 |
sweaty | +0.069 |
Neg logit contributions | |
dial | -0.053 |
peek | -0.053 |
nob | -0.052 |
ad | -0.049 |
turned | -0.047 |
Token | ados |
Token ID | 22484 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
istered | +0.088 |
rier | +0.077 |
working | +0.075 |
ishy | +0.071 |
ivery | +0.070 |
Neg logit contributions | |
ad | -0.148 |
dial | -0.143 |
nob | -0.140 |
peek | -0.135 |
p | -0.131 |
Token | because |
Token ID | 780 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
istered | +0.034 |
rier | +0.034 |
working | +0.032 |
ishy | +0.030 |
ivery | +0.029 |
Neg logit contributions | |
peek | -0.019 |
dial | -0.019 |
nob | -0.018 |
turned | -0.017 |
ad | -0.017 |
Token | they |
Token ID | 484 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
istered | +0.033 |
ishy | +0.028 |
rier | +0.028 |
ivery | +0.028 |
goals | +0.027 |
Neg logit contributions | |
turned | -0.016 |
opened | -0.015 |
dial | -0.015 |
Inside | -0.014 |
called | -0.014 |
Token | were |
Token ID | 547 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
rier | +0.067 |
istered | +0.063 |
goals | +0.057 |
working | +0.057 |
ishy | +0.057 |
Neg logit contributions | |
peek | -0.039 |
turned | -0.039 |
dial | -0.038 |
opened | -0.037 |
ad | -0.034 |
Token | y |
Token ID | 331 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
istered | +0.045 |
rier | +0.043 |
working | +0.042 |
ishy | +0.041 |
ivery | +0.041 |
Neg logit contributions | |
dial | -0.034 |
peek | -0.033 |
turned | -0.032 |
opened | -0.032 |
called | -0.032 |
Token | who |
Token ID | 508 |
Feature activation | +5.35e-03 |
Pos logit contributions | |
dial | +0.020 |
nob | +0.019 |
peek | +0.018 |
Looking | +0.018 |
ad | +0.016 |
Neg logit contributions | |
working | -0.054 |
istered | -0.052 |
rier | -0.051 |
ishy | -0.048 |
bored | -0.048 |
Token | liked |
Token ID | 8288 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
rier | +0.009 |
istered | +0.008 |
working | +0.008 |
goals | +0.007 |
ishy | +0.007 |
Neg logit contributions | |
peek | -0.008 |
dial | -0.008 |
nob | -0.008 |
turned | -0.008 |
opened | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +6.67e-03 |
Pos logit contributions | |
rier | +0.034 |
istered | +0.033 |
working | +0.032 |
ishy | +0.031 |
goals | +0.030 |
Neg logit contributions | |
peek | -0.011 |
dial | -0.010 |
nob | -0.009 |
turned | -0.009 |
ad | -0.008 |
Token | collect |
Token ID | 2824 |
Feature activation | +1.56e-03 |
Pos logit contributions | |
working | +0.011 |
istered | +0.010 |
rier | +0.010 |
ishy | +0.009 |
sweaty | +0.009 |
Neg logit contributions | |
dial | -0.009 |
nob | -0.008 |
Looking | -0.008 |
ran | -0.008 |
fore | -0.008 |
Token | or |
Token ID | 393 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
rier | +0.002 |
working | +0.002 |
istered | +0.002 |
ishy | +0.002 |
goals | +0.002 |
Neg logit contributions | |
dial | -0.003 |
nob | -0.003 |
peek | -0.003 |
Looking | -0.002 |
turned | -0.002 |
Token | nam |
Token ID | 7402 |
Feature activation | +7.61e-02 |
Pos logit contributions | |
rier | +0.021 |
istered | +0.021 |
working | +0.018 |
ishy | +0.016 |
ivery | +0.014 |
Neg logit contributions | |
peek | -0.069 |
dial | -0.067 |
turned | -0.064 |
opened | -0.063 |
nob | -0.062 |
Token | ents |
Token ID | 658 |
Feature activation | +5.43e-03 |
Pos logit contributions | |
rier | +0.129 |
working | +0.127 |
istered | +0.125 |
ishy | +0.116 |
obb | +0.113 |
Neg logit contributions | |
peek | -0.099 |
dial | -0.097 |
ad | -0.094 |
nob | -0.090 |
opened | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
istered | +0.012 |
rier | +0.011 |
working | +0.011 |
ishy | +0.011 |
goals | +0.010 |
Neg logit contributions | |
peek | -0.005 |
dial | -0.005 |
nob | -0.005 |
turned | -0.004 |
ad | -0.004 |
Token | Everywhere |
Token ID | 48201 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
rier | +0.016 |
working | +0.016 |
istered | +0.016 |
goals | +0.014 |
ishy | +0.014 |
Neg logit contributions | |
dial | -0.019 |
peek | -0.019 |
nob | -0.018 |
ad | -0.017 |
turned | -0.017 |
Token | she |
Token ID | 673 |
Feature activation | +3.76e-05 |
Pos logit contributions | |
rier | +0.033 |
istered | +0.032 |
ivery | +0.028 |
working | +0.027 |
ishy | +0.027 |
Neg logit contributions | |
peek | -0.027 |
turned | -0.027 |
opened | -0.027 |
dial | -0.026 |
called | -0.025 |
Token | went |
Token ID | 1816 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
working | +0.000 |
rier | +0.000 |
istered | +0.000 |
ishy | +0.000 |
attered | +0.000 |
Neg logit contributions | |
dial | -0.000 |
nob | -0.000 |
peek | -0.000 |
ad | -0.000 |
Looking | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +1.89e-03 |
Pos logit contributions | |
dial | +0.012 |
peek | +0.012 |
nob | +0.011 |
turned | +0.010 |
ad | +0.010 |
Neg logit contributions | |
rier | -0.027 |
working | -0.026 |
istered | -0.026 |
ishy | -0.024 |
goals | -0.024 |
Token | they |
Token ID | 484 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
working | +0.004 |
rier | +0.004 |
istered | +0.004 |
nam | +0.003 |
bored | +0.003 |
Neg logit contributions | |
nob | -0.002 |
dial | -0.002 |
peek | -0.002 |
ad | -0.002 |
Looking | -0.002 |
Token | walked |
Token ID | 6807 |
Feature activation | +9.18e-03 |
Pos logit contributions | |
rier | +0.026 |
istered | +0.025 |
working | +0.024 |
ishy | +0.023 |
goals | +0.022 |
Neg logit contributions | |
peek | -0.014 |
dial | -0.014 |
turned | -0.013 |
nob | -0.012 |
opened | -0.012 |
Token | through |
Token ID | 832 |
Feature activation | +4.67e-03 |
Pos logit contributions | |
rier | +0.016 |
working | +0.016 |
istered | +0.016 |
bored | +0.014 |
goals | +0.014 |
Neg logit contributions | |
dial | -0.011 |
nob | -0.011 |
peek | -0.011 |
ad | -0.010 |
Looking | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +3.95e-04 |
Pos logit contributions | |
rier | +0.010 |
istered | +0.010 |
working | +0.010 |
ishy | +0.009 |
ivery | +0.009 |
Neg logit contributions | |
dial | -0.005 |
peek | -0.005 |
nob | -0.004 |
turned | -0.004 |
ad | -0.004 |
Token | carn |
Token ID | 18466 |
Feature activation | +7.56e-02 |
Pos logit contributions | |
working | +0.001 |
istered | +0.001 |
rier | +0.001 |
goals | +0.001 |
bored | +0.001 |
Neg logit contributions | |
nob | -0.001 |
dial | -0.001 |
peek | -0.001 |
pat | -0.001 |
p | -0.001 |
Token | ival |
Token ID | 2473 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
rier | +0.112 |
working | +0.108 |
istered | +0.105 |
ishy | +0.096 |
goals | +0.090 |
Neg logit contributions | |
peek | -0.122 |
dial | -0.117 |
nob | -0.116 |
turned | -0.110 |
Looking | -0.109 |
Token | gates |
Token ID | 17435 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
istered | +0.058 |
rier | +0.057 |
ishy | +0.054 |
working | +0.054 |
ivery | +0.052 |
Neg logit contributions | |
peek | -0.028 |
dial | -0.027 |
turned | -0.023 |
nob | -0.023 |
opened | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
rier | +0.023 |
istered | +0.022 |
working | +0.022 |
ishy | +0.021 |
goals | +0.020 |
Neg logit contributions | |
peek | -0.012 |
dial | -0.012 |
nob | -0.012 |
turned | -0.011 |
ad | -0.011 |
Token | while |
Token ID | 981 |
Feature activation | +8.51e-03 |
Pos logit contributions | |
nob | +0.024 |
dial | +0.023 |
ad | +0.022 |
peek | +0.022 |
Looking | +0.021 |
Neg logit contributions | |
rier | -0.038 |
working | -0.038 |
istered | -0.037 |
bored | -0.035 |
sweaty | -0.035 |
Token | Max |
Token ID | 5436 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
rier | +0.015 |
ishy | +0.014 |
istered | +0.014 |
working | +0.013 |
attered | +0.013 |
Neg logit contributions | |
dial | -0.012 |
peek | -0.012 |
turned | -0.011 |
opened | -0.011 |
called | -0.011 |
Token | went |
Token ID | 1816 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
working | +0.000 |
rier | +0.000 |
istered | +0.000 |
ishy | +0.000 |
attered | +0.000 |
Neg logit contributions | |
dial | -0.000 |
nob | -0.000 |
peek | -0.000 |
ad | -0.000 |
Looking | -0.000 |
Token | she |
Token ID | 673 |
Feature activation | +6.69e-03 |
Pos logit contributions | |
rier | +0.041 |
istered | +0.041 |
working | +0.039 |
ishy | +0.037 |
goals | +0.036 |
Neg logit contributions | |
peek | -0.026 |
dial | -0.025 |
nob | -0.024 |
turned | -0.023 |
ad | -0.023 |
Token | searched |
Token ID | 16499 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
rier | +0.014 |
istered | +0.013 |
working | +0.013 |
ishy | +0.012 |
goals | +0.012 |
Neg logit contributions | |
peek | -0.007 |
dial | -0.007 |
nob | -0.006 |
turned | -0.006 |
ad | -0.006 |
Token | for |
Token ID | 329 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
rier | +0.031 |
istered | +0.030 |
working | +0.030 |
ishy | +0.027 |
goals | +0.026 |
Neg logit contributions | |
peek | -0.019 |
dial | -0.019 |
nob | -0.018 |
turned | -0.018 |
ad | -0.017 |
Token | or |
Token ID | 393 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
rier | +0.037 |
working | +0.036 |
istered | +0.036 |
ishy | +0.032 |
goals | +0.031 |
Neg logit contributions | |
dial | -0.039 |
peek | -0.039 |
nob | -0.038 |
turned | -0.036 |
ad | -0.035 |
Token | nam |
Token ID | 7402 |
Feature activation | +7.54e-02 |
Pos logit contributions | |
rier | +0.018 |
istered | +0.016 |
working | +0.015 |
ishy | +0.011 |
goals | +0.009 |
Neg logit contributions | |
peek | -0.091 |
dial | -0.091 |
nob | -0.088 |
turned | -0.087 |
opened | -0.085 |
Token | ents |
Token ID | 658 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
rier | +0.131 |
working | +0.129 |
istered | +0.126 |
ishy | +0.118 |
obb | +0.115 |
Neg logit contributions | |
peek | -0.095 |
dial | -0.092 |
ad | -0.089 |
nob | -0.087 |
opened | -0.085 |
Token | , |
Token ID | 11 |
Feature activation | +9.85e-03 |
Pos logit contributions | |
istered | +0.035 |
rier | +0.035 |
working | +0.033 |
ishy | +0.032 |
ivery | +0.031 |
Neg logit contributions | |
peek | -0.013 |
nob | -0.012 |
dial | -0.012 |
turned | -0.012 |
opened | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
rier | +0.017 |
istered | +0.017 |
working | +0.016 |
ishy | +0.015 |
goals | +0.014 |
Neg logit contributions | |
peek | -0.013 |
dial | -0.013 |
nob | -0.013 |
turned | -0.012 |
ad | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
istered | +0.018 |
rier | +0.018 |
ishy | +0.016 |
working | +0.015 |
ivery | +0.015 |
Neg logit contributions | |
peek | -0.016 |
dial | -0.016 |
turned | -0.015 |
opened | -0.015 |
nob | -0.015 |
Token | more |
Token ID | 517 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
istered | +0.055 |
working | +0.052 |
rier | +0.049 |
ishy | +0.049 |
attered | +0.047 |
Neg logit contributions | |
peek | -0.045 |
dial | -0.045 |
turned | -0.041 |
nob | -0.041 |
opened | -0.040 |
Token | , |
Token ID | 11 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
rier | +0.015 |
istered | +0.014 |
working | +0.014 |
ishy | +0.013 |
goals | +0.013 |
Neg logit contributions | |
peek | -0.007 |
dial | -0.006 |
turned | -0.006 |
nob | -0.006 |
opened | -0.006 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.09e-03 |
Pos logit contributions | |
rier | +0.023 |
istered | +0.022 |
working | +0.021 |
goals | +0.020 |
ishy | +0.019 |
Neg logit contributions | |
peek | -0.016 |
dial | -0.015 |
turned | -0.015 |
opened | -0.014 |
called | -0.013 |
Token | went |
Token ID | 1816 |
Feature activation | -6.74e-03 |
Pos logit contributions | |
rier | +0.008 |
working | +0.008 |
istered | +0.008 |
ishy | +0.007 |
goals | +0.007 |
Neg logit contributions | |
dial | -0.004 |
peek | -0.004 |
nob | -0.004 |
turned | -0.004 |
ad | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +1.35e-03 |
Pos logit contributions | |
ad | +0.010 |
dial | +0.010 |
Looking | +0.009 |
nob | +0.009 |
ran | +0.009 |
Neg logit contributions | |
working | -0.011 |
bored | -0.009 |
rier | -0.008 |
istered | -0.008 |
tired | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
working | +0.002 |
nam | +0.002 |
bored | +0.002 |
kicker | +0.001 |
sweaty | +0.001 |
Neg logit contributions | |
nob | -0.002 |
ran | -0.002 |
ad | -0.002 |
fore | -0.002 |
Looking | -0.002 |
Token | carn |
Token ID | 18466 |
Feature activation | +7.51e-02 |
Pos logit contributions | |
rier | +0.052 |
working | +0.052 |
istered | +0.051 |
ishy | +0.045 |
goals | +0.045 |
Neg logit contributions | |
dial | -0.051 |
peek | -0.050 |
nob | -0.049 |
ad | -0.046 |
turned | -0.046 |
Token | ival |
Token ID | 2473 |
Feature activation | +5.49e-03 |
Pos logit contributions | |
rier | +0.098 |
working | +0.095 |
istered | +0.094 |
ishy | +0.082 |
goals | +0.080 |
Neg logit contributions | |
peek | -0.130 |
dial | -0.129 |
nob | -0.126 |
turned | -0.120 |
ad | -0.119 |
Token | with |
Token ID | 351 |
Feature activation | +6.07e-03 |
Pos logit contributions | |
rier | +0.012 |
working | +0.012 |
istered | +0.012 |
goals | +0.011 |
ishy | +0.011 |
Neg logit contributions | |
dial | -0.005 |
peek | -0.004 |
nob | -0.004 |
turned | -0.004 |
ad | -0.004 |
Token | her |
Token ID | 607 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
rier | +0.011 |
istered | +0.011 |
working | +0.011 |
ishy | +0.010 |
ivery | +0.010 |
Neg logit contributions | |
peek | -0.008 |
dial | -0.008 |
turned | -0.008 |
ad | -0.007 |
nob | -0.007 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
rier | +0.016 |
working | +0.015 |
istered | +0.015 |
ishy | +0.014 |
goals | +0.013 |
Neg logit contributions | |
dial | -0.017 |
peek | -0.017 |
nob | -0.016 |
turned | -0.016 |
ad | -0.016 |
Token | my |
Token ID | 1820 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
nob | +0.038 |
Looking | +0.037 |
peek | +0.036 |
dial | +0.036 |
AAAA | +0.035 |
Neg logit contributions | |
working | -0.049 |
istered | -0.042 |
rier | -0.042 |
bored | -0.042 |
ishy | -0.040 |
Token | , |
Token ID | 11 |
Feature activation | +4.69e-03 |
Pos logit contributions | |
dial | +0.001 |
peek | +0.001 |
nob | +0.000 |
turned | +0.000 |
ad | +0.000 |
Neg logit contributions | |
rier | -0.001 |
istered | -0.001 |
working | -0.001 |
ishy | -0.001 |
goals | -0.001 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.11e-03 |
Pos logit contributions | |
rier | +0.008 |
istered | +0.008 |
working | +0.007 |
ishy | +0.007 |
goals | +0.007 |
Neg logit contributions | |
peek | -0.007 |
dial | -0.007 |
nob | -0.006 |
turned | -0.006 |
opened | -0.006 |
Token | went |
Token ID | 1816 |
Feature activation | -8.15e-03 |
Pos logit contributions | |
peek | +0.002 |
dial | +0.002 |
nob | +0.002 |
turned | +0.002 |
opened | +0.002 |
Neg logit contributions | |
rier | -0.004 |
istered | -0.004 |
working | -0.004 |
ishy | -0.004 |
goals | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +1.09e-03 |
Pos logit contributions | |
dial | +0.012 |
ad | +0.012 |
Looking | +0.012 |
nob | +0.012 |
AAAA | +0.011 |
Neg logit contributions | |
working | -0.012 |
bored | -0.011 |
rier | -0.010 |
forward | -0.010 |
tired | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
working | +0.001 |
bored | +0.001 |
nam | +0.001 |
kicker | +0.001 |
goals | +0.001 |
Neg logit contributions | |
nob | -0.002 |
fore | -0.002 |
ad | -0.002 |
ran | -0.002 |
dial | -0.002 |
Token | carn |
Token ID | 18466 |
Feature activation | +7.44e-02 |
Pos logit contributions | |
rier | +0.040 |
working | +0.040 |
istered | +0.039 |
ishy | +0.035 |
goals | +0.034 |
Neg logit contributions | |
dial | -0.041 |
nob | -0.040 |
peek | -0.040 |
ad | -0.038 |
turned | -0.037 |
Token | ival |
Token ID | 2473 |
Feature activation | +3.00e-03 |
Pos logit contributions | |
rier | +0.095 |
working | +0.092 |
istered | +0.091 |
ishy | +0.079 |
goals | +0.078 |
Neg logit contributions | |
peek | -0.131 |
dial | -0.130 |
nob | -0.127 |
turned | -0.121 |
ad | -0.120 |
Token | with |
Token ID | 351 |
Feature activation | +3.63e-03 |
Pos logit contributions | |
rier | +0.006 |
working | +0.006 |
istered | +0.006 |
ishy | +0.006 |
goals | +0.006 |
Neg logit contributions | |
dial | -0.003 |
peek | -0.003 |
nob | -0.002 |
turned | -0.002 |
ad | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | +6.70e-03 |
Pos logit contributions | |
rier | +0.007 |
istered | +0.007 |
working | +0.007 |
ishy | +0.007 |
ivery | +0.006 |
Neg logit contributions | |
peek | -0.005 |
dial | -0.005 |
ad | -0.004 |
turned | -0.004 |
opened | -0.004 |
Token | family |
Token ID | 1641 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
rier | +0.009 |
working | +0.009 |
istered | +0.009 |
ishy | +0.008 |
goals | +0.008 |
Neg logit contributions | |
dial | -0.011 |
peek | -0.011 |
nob | -0.011 |
turned | -0.010 |
ad | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +2.48e-03 |
Pos logit contributions | |
dial | +0.016 |
nob | +0.016 |
peek | +0.015 |
Looking | +0.014 |
turned | +0.014 |
Neg logit contributions | |
working | -0.023 |
rier | -0.023 |
istered | -0.022 |
goals | -0.020 |
ishy | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
nob | +0.022 |
dial | +0.020 |
peek | +0.019 |
ad | +0.018 |
question | +0.017 |
Neg logit contributions | |
rier | -0.029 |
istered | -0.026 |
working | -0.026 |
goals | -0.025 |
bored | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -5.02e-03 |
Pos logit contributions | |
nob | +0.013 |
dial | +0.013 |
peek | +0.012 |
fore | +0.011 |
turned | +0.011 |
Neg logit contributions | |
rier | -0.018 |
working | -0.018 |
istered | -0.018 |
goals | -0.017 |
opy | -0.016 |
Token | So |
Token ID | 2396 |
Feature activation | -5.91e-03 |
Pos logit contributions | |
dial | +0.006 |
peek | +0.006 |
nob | +0.006 |
turned | +0.006 |
ad | +0.005 |
Neg logit contributions | |
rier | -0.009 |
istered | -0.009 |
working | -0.009 |
goals | -0.008 |
ishy | -0.008 |
Token | Jack |
Token ID | 3619 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
dial | +0.007 |
peek | +0.007 |
nob | +0.007 |
turned | +0.007 |
ad | +0.007 |
Neg logit contributions | |
rier | -0.011 |
istered | -0.011 |
working | -0.010 |
ishy | -0.009 |
goals | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
nob | +0.013 |
dial | +0.013 |
peek | +0.013 |
Looking | +0.012 |
ad | +0.012 |
Neg logit contributions | |
working | -0.026 |
rier | -0.025 |
istered | -0.024 |
ishy | -0.024 |
goals | -0.023 |
Token | St |
Token ID | 520 |
Feature activation | +7.36e-02 |
Pos logit contributions | |
working | +0.023 |
rier | +0.023 |
istered | +0.022 |
ishy | +0.020 |
goals | +0.020 |
Neg logit contributions | |
nob | -0.017 |
dial | -0.017 |
peek | -0.017 |
ad | -0.015 |
turned | -0.015 |
Token | itch |
Token ID | 2007 |
Feature activation | +2.02e-03 |
Pos logit contributions | |
rier | +0.090 |
working | +0.081 |
istered | +0.078 |
nam | +0.071 |
sweaty | +0.069 |
Neg logit contributions | |
dial | -0.137 |
ad | -0.136 |
peek | -0.135 |
turned | -0.131 |
nob | -0.129 |
Token | started |
Token ID | 2067 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
rier | +0.004 |
istered | +0.004 |
working | +0.004 |
ishy | +0.004 |
goals | +0.003 |
Neg logit contributions | |
dial | -0.002 |
peek | -0.002 |
nob | -0.002 |
turned | -0.002 |
ad | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
nob | +0.039 |
dial | +0.037 |
peek | +0.036 |
ad | +0.036 |
Looking | +0.035 |
Neg logit contributions | |
working | -0.043 |
istered | -0.039 |
bored | -0.038 |
rier | -0.038 |
sweaty | -0.037 |
Token | play |
Token ID | 711 |
Feature activation | -1.96e-03 |
Pos logit contributions | |
nob | +0.017 |
dial | +0.017 |
peek | +0.016 |
Looking | +0.016 |
ad | +0.016 |
Neg logit contributions | |
working | -0.021 |
istered | -0.019 |
rier | -0.019 |
ishy | -0.018 |
goals | -0.018 |
Token | together |
Token ID | 1978 |
Feature activation | -5.14e-04 |
Pos logit contributions | |
peek | +0.002 |
dial | +0.002 |
nob | +0.002 |
turned | +0.002 |
ad | +0.002 |
Neg logit contributions | |
rier | -0.004 |
istered | -0.004 |
working | -0.004 |
ishy | -0.004 |
attered | -0.003 |
Token | bitter |
Token ID | 12922 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
rier | +0.043 |
istered | +0.040 |
working | +0.039 |
ivery | +0.037 |
onel | +0.036 |
Neg logit contributions | |
turned | -0.031 |
peek | -0.031 |
opened | -0.029 |
dial | -0.029 |
called | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
istered | +0.041 |
working | +0.039 |
rier | +0.039 |
nam | +0.038 |
ivery | +0.037 |
Neg logit contributions | |
peek | -0.013 |
turned | -0.013 |
dial | -0.013 |
nob | -0.012 |
called | -0.011 |
Token | made |
Token ID | 925 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
rier | +0.063 |
istered | +0.058 |
ivery | +0.057 |
working | +0.057 |
runner | +0.054 |
Neg logit contributions | |
turned | -0.045 |
peek | -0.043 |
opened | -0.042 |
called | -0.040 |
dial | -0.040 |
Token | her |
Token ID | 607 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
working | +0.058 |
istered | +0.057 |
rier | +0.056 |
ishy | +0.052 |
ivery | +0.051 |
Neg logit contributions | |
peek | -0.038 |
dial | -0.038 |
turned | -0.036 |
ad | -0.036 |
nob | -0.034 |
Token | face |
Token ID | 1986 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
working | +0.043 |
rier | +0.042 |
istered | +0.040 |
ivery | +0.037 |
attered | +0.036 |
Neg logit contributions | |
peek | -0.034 |
dial | -0.031 |
turned | -0.030 |
pat | -0.030 |
call | -0.029 |
Token | scr |
Token ID | 6040 |
Feature activation | +7.34e-02 |
Pos logit contributions | |
working | +0.037 |
rier | +0.036 |
istered | +0.035 |
kicker | +0.034 |
runner | +0.033 |
Neg logit contributions | |
peek | -0.031 |
turned | -0.030 |
dial | -0.030 |
nob | -0.028 |
opened | -0.027 |
Token | unch |
Token ID | 3316 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
working | +0.114 |
rier | +0.107 |
nam | +0.102 |
istered | +0.098 |
happiest | +0.097 |
Neg logit contributions | |
ad | -0.107 |
peek | -0.100 |
p | -0.098 |
nob | -0.092 |
turned | -0.091 |
Token | up |
Token ID | 510 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
istered | +0.064 |
working | +0.062 |
Majesty | +0.061 |
nam | +0.061 |
kicker | +0.060 |
Neg logit contributions | |
p | -0.024 |
turned | -0.024 |
nob | -0.022 |
peek | -0.020 |
ad | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +7.27e-03 |
Pos logit contributions | |
working | +0.021 |
istered | +0.021 |
rier | +0.021 |
kicker | +0.020 |
ivery | +0.019 |
Neg logit contributions | |
peek | -0.009 |
turned | -0.008 |
dial | -0.008 |
nob | -0.008 |
opened | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e-03 |
Pos logit contributions | |
rier | +0.016 |
working | +0.016 |
istered | +0.016 |
ishy | +0.015 |
goals | +0.015 |
Neg logit contributions | |
dial | -0.006 |
peek | -0.006 |
nob | -0.005 |
turned | -0.005 |
ad | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +4.63e-03 |
Pos logit contributions | |
dial | +0.001 |
nob | +0.001 |
peek | +0.001 |
turned | +0.001 |
opened | +0.001 |
Neg logit contributions | |
working | -0.002 |
istered | -0.002 |
rier | -0.002 |
goals | -0.002 |
kicker | -0.002 |
Token | ongo |
Token ID | 25162 |
Feature activation | +9.08e-03 |
Pos logit contributions | |
rier | +0.023 |
working | +0.022 |
istered | +0.022 |
ishy | +0.019 |
goals | +0.018 |
Neg logit contributions | |
dial | -0.038 |
peek | -0.038 |
nob | -0.037 |
turned | -0.036 |
ad | -0.035 |
Token | loved |
Token ID | 6151 |
Feature activation | +5.68e-03 |
Pos logit contributions | |
working | +0.014 |
rier | +0.014 |
istered | +0.013 |
ishy | +0.012 |
goals | +0.012 |
Neg logit contributions | |
dial | -0.014 |
peek | -0.013 |
nob | -0.013 |
ad | -0.012 |
turned | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
working | +0.011 |
rier | +0.009 |
bored | +0.009 |
goals | +0.009 |
nam | +0.009 |
Neg logit contributions | |
nob | -0.007 |
dial | -0.006 |
peek | -0.006 |
Looking | -0.006 |
Turning | -0.006 |
Token | eat |
Token ID | 4483 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
working | +0.018 |
rier | +0.015 |
istered | +0.015 |
goals | +0.015 |
ishy | +0.015 |
Neg logit contributions | |
Looking | -0.015 |
dial | -0.015 |
nob | -0.014 |
fore | -0.014 |
Turning | -0.014 |
Token | av |
Token ID | 1196 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
rier | +0.027 |
working | +0.026 |
istered | +0.025 |
ishy | +0.022 |
goals | +0.022 |
Neg logit contributions | |
dial | -0.035 |
peek | -0.034 |
nob | -0.034 |
turned | -0.032 |
ad | -0.032 |
Token | oc |
Token ID | 420 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
rier | +0.082 |
istered | +0.077 |
working | +0.075 |
ishy | +0.069 |
sweaty | +0.067 |
Neg logit contributions | |
peek | -0.051 |
dial | -0.051 |
nob | -0.050 |
ad | -0.048 |
turned | -0.045 |
Token | ados |
Token ID | 22484 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
istered | +0.077 |
rier | +0.069 |
working | +0.068 |
ishy | +0.062 |
sweaty | +0.060 |
Neg logit contributions | |
ad | -0.148 |
dial | -0.143 |
nob | -0.141 |
peek | -0.139 |
p | -0.134 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
rier | +0.044 |
istered | +0.044 |
working | +0.041 |
ishy | +0.040 |
ivery | +0.039 |
Neg logit contributions | |
peek | -0.019 |
dial | -0.018 |
nob | -0.017 |
turned | -0.016 |
called | -0.016 |
Token | One |
Token ID | 1881 |
Feature activation | +9.54e-03 |
Pos logit contributions | |
dial | +0.022 |
peek | +0.021 |
nob | +0.021 |
ad | +0.020 |
turned | +0.020 |
Neg logit contributions | |
rier | -0.024 |
working | -0.024 |
istered | -0.022 |
ishy | -0.021 |
goals | -0.021 |
Token | day |
Token ID | 1110 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
working | +0.013 |
rier | +0.012 |
istered | +0.012 |
ishy | +0.011 |
goals | +0.010 |
Neg logit contributions | |
dial | -0.016 |
peek | -0.016 |
nob | -0.016 |
ad | -0.015 |
Looking | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
dial | +0.017 |
peek | +0.017 |
nob | +0.016 |
turned | +0.015 |
ad | +0.015 |
Neg logit contributions | |
rier | -0.028 |
working | -0.027 |
istered | -0.027 |
ishy | -0.025 |
goals | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
peek | +0.006 |
dial | +0.005 |
nob | +0.005 |
turned | +0.005 |
opened | +0.005 |
Neg logit contributions | |
rier | -0.013 |
istered | -0.013 |
working | -0.013 |
ishy | -0.012 |
goals | -0.012 |
Token | play |
Token ID | 711 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
rier | +0.034 |
istered | +0.032 |
working | +0.032 |
ishy | +0.030 |
ivery | +0.029 |
Neg logit contributions | |
peek | -0.020 |
dial | -0.019 |
turned | -0.018 |
opened | -0.017 |
nob | -0.017 |
Token | with |
Token ID | 351 |
Feature activation | +2.80e-03 |
Pos logit contributions | |
working | +0.050 |
istered | +0.050 |
rier | +0.048 |
ishy | +0.047 |
attered | +0.046 |
Neg logit contributions | |
peek | -0.015 |
nob | -0.014 |
dial | -0.012 |
called | -0.010 |
call | -0.010 |
Token | your |
Token ID | 534 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
working | +0.006 |
rier | +0.006 |
istered | +0.006 |
ishy | +0.006 |
ivery | +0.006 |
Neg logit contributions | |
peek | -0.003 |
dial | -0.003 |
nob | -0.002 |
turned | -0.002 |
called | -0.002 |
Token | or |
Token ID | 393 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
rier | +0.030 |
working | +0.030 |
istered | +0.029 |
ishy | +0.026 |
ivery | +0.025 |
Neg logit contributions | |
peek | -0.030 |
dial | -0.030 |
nob | -0.029 |
turned | -0.028 |
ad | -0.027 |
Token | nam |
Token ID | 7402 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
rier | +0.062 |
istered | +0.061 |
working | +0.058 |
ishy | +0.056 |
ivery | +0.054 |
Neg logit contributions | |
peek | -0.043 |
dial | -0.041 |
turned | -0.037 |
call | -0.036 |
nob | -0.036 |
Token | ents |
Token ID | 658 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
rier | +0.128 |
working | +0.124 |
istered | +0.121 |
ishy | +0.114 |
obb | +0.113 |
Neg logit contributions | |
peek | -0.090 |
dial | -0.088 |
ad | -0.086 |
nob | -0.083 |
opened | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
istered | +0.049 |
rier | +0.047 |
ishy | +0.045 |
working | +0.045 |
ivery | +0.043 |
Neg logit contributions | |
peek | -0.016 |
nob | -0.015 |
turned | -0.014 |
opened | -0.014 |
dial | -0.014 |
Token | Maybe |
Token ID | 6674 |
Feature activation | +1.72e-03 |
Pos logit contributions | |
rier | +0.020 |
istered | +0.020 |
working | +0.019 |
goals | +0.018 |
ishy | +0.018 |
Neg logit contributions | |
dial | -0.016 |
peek | -0.016 |
nob | -0.015 |
turned | -0.015 |
ad | -0.014 |
Token | you |
Token ID | 345 |
Feature activation | +6.86e-03 |
Pos logit contributions | |
rier | +0.002 |
working | +0.002 |
istered | +0.002 |
nam | +0.002 |
goals | +0.002 |
Neg logit contributions | |
nob | -0.003 |
dial | -0.003 |
peek | -0.003 |
Looking | -0.003 |
ad | -0.003 |
Token | can |
Token ID | 460 |
Feature activation | +4.90e-03 |
Pos logit contributions | |
rier | +0.014 |
istered | +0.013 |
working | +0.013 |
ishy | +0.012 |
ivery | +0.011 |
Neg logit contributions | |
peek | -0.008 |
dial | -0.008 |
nob | -0.007 |
turned | -0.007 |
opened | -0.007 |
Token | Max |
Token ID | 5436 |
Feature activation | +9.01e-04 |
Pos logit contributions | |
rier | +0.012 |
working | +0.012 |
istered | +0.012 |
ishy | +0.011 |
goals | +0.011 |
Neg logit contributions | |
dial | -0.005 |
peek | -0.005 |
nob | -0.005 |
turned | -0.004 |
Looking | -0.004 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
rier | +0.002 |
istered | +0.002 |
working | +0.002 |
ishy | +0.002 |
goals | +0.002 |
Neg logit contributions | |
dial | -0.001 |
peek | -0.001 |
nob | -0.001 |
turned | -0.001 |
ad | -0.001 |
Token | being |
Token ID | 852 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
working | +0.006 |
rier | +0.005 |
istered | +0.005 |
goals | +0.005 |
bored | +0.005 |
Neg logit contributions | |
nob | -0.004 |
peek | -0.004 |
dial | -0.004 |
Looking | -0.004 |
ad | -0.004 |
Token | at |
Token ID | 379 |
Feature activation | +9.00e-04 |
Pos logit contributions | |
rier | +0.020 |
istered | +0.020 |
working | +0.019 |
ishy | +0.018 |
goals | +0.018 |
Neg logit contributions | |
dial | -0.013 |
peek | -0.013 |
nob | -0.011 |
turned | -0.011 |
opened | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
rier | +0.002 |
working | +0.002 |
istered | +0.002 |
ishy | +0.001 |
goals | +0.001 |
Neg logit contributions | |
dial | -0.001 |
peek | -0.001 |
nob | -0.001 |
turned | -0.001 |
ad | -0.001 |
Token | carn |
Token ID | 18466 |
Feature activation | +7.25e-02 |
Pos logit contributions | |
rier | +0.024 |
istered | +0.024 |
working | +0.023 |
ishy | +0.022 |
ivery | +0.021 |
Neg logit contributions | |
dial | -0.016 |
peek | -0.016 |
nob | -0.014 |
turned | -0.014 |
ad | -0.014 |
Token | ival |
Token ID | 2473 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
rier | +0.118 |
istered | +0.114 |
working | +0.110 |
ishy | +0.102 |
nam | +0.098 |
Neg logit contributions | |
peek | -0.107 |
nob | -0.100 |
dial | -0.098 |
ad | -0.095 |
Turning | -0.093 |
Token | so |
Token ID | 523 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
rier | +0.038 |
istered | +0.038 |
working | +0.036 |
ishy | +0.035 |
ivery | +0.034 |
Neg logit contributions | |
dial | -0.017 |
peek | -0.017 |
nob | -0.016 |
turned | -0.015 |
ad | -0.014 |
Token | much |
Token ID | 881 |
Feature activation | +3.96e-03 |
Pos logit contributions | |
rier | +0.035 |
ivery | +0.033 |
istered | +0.032 |
ishy | +0.030 |
working | +0.029 |
Neg logit contributions | |
peek | -0.030 |
ad | -0.030 |
turned | -0.029 |
dial | -0.028 |
opened | -0.027 |
Token | that |
Token ID | 326 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
istered | +0.008 |
rier | +0.008 |
working | +0.008 |
ishy | +0.008 |
ivery | +0.008 |
Neg logit contributions | |
dial | -0.004 |
peek | -0.004 |
turned | -0.003 |
nob | -0.003 |
ad | -0.003 |
Token | he |
Token ID | 339 |
Feature activation | +5.98e-03 |
Pos logit contributions | |
istered | +0.020 |
rier | +0.019 |
ivery | +0.018 |
ishy | +0.018 |
working | +0.017 |
Neg logit contributions | |
peek | -0.014 |
turned | -0.014 |
dial | -0.014 |
opened | -0.013 |
called | -0.013 |
Token | ummy |
Token ID | 13513 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
rier | +0.029 |
istered | +0.028 |
working | +0.028 |
nam | +0.022 |
kicker | +0.022 |
Neg logit contributions | |
peek | -0.056 |
ad | -0.055 |
dial | -0.055 |
turned | -0.053 |
nob | -0.053 |
Token | cotton |
Token ID | 15985 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
nob | +0.045 |
peek | +0.044 |
dial | +0.044 |
Looking | +0.043 |
turned | +0.041 |
Neg logit contributions | |
working | -0.032 |
rier | -0.032 |
istered | -0.029 |
ishy | -0.028 |
goals | -0.028 |
Token | candy |
Token ID | 18550 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
rier | +0.019 |
istered | +0.019 |
working | +0.019 |
ivery | +0.017 |
ishy | +0.017 |
Neg logit contributions | |
dial | -0.018 |
peek | -0.018 |
turned | -0.017 |
nob | -0.016 |
ad | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +5.99e-03 |
Pos logit contributions | |
istered | +0.033 |
rier | +0.032 |
ivery | +0.031 |
working | +0.031 |
ishy | +0.030 |
Neg logit contributions | |
peek | -0.008 |
dial | -0.008 |
turned | -0.007 |
nob | -0.007 |
opened | -0.007 |
Token | play |
Token ID | 711 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
rier | +0.011 |
istered | +0.011 |
working | +0.010 |
ishy | +0.010 |
ivery | +0.010 |
Neg logit contributions | |
dial | -0.008 |
peek | -0.008 |
turned | -0.007 |
opened | -0.007 |
nob | -0.007 |
Token | carn |
Token ID | 18466 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
istered | +0.045 |
working | +0.045 |
rier | +0.045 |
ishy | +0.042 |
attered | +0.041 |
Neg logit contributions | |
peek | -0.015 |
dial | -0.014 |
nob | -0.013 |
ad | -0.011 |
turned | -0.011 |
Token | ival |
Token ID | 2473 |
Feature activation | +7.81e-03 |
Pos logit contributions | |
rier | +0.106 |
working | +0.103 |
istered | +0.103 |
ishy | +0.092 |
goals | +0.088 |
Neg logit contributions | |
peek | -0.114 |
nob | -0.110 |
dial | -0.109 |
ad | -0.103 |
Looking | -0.102 |
Token | games |
Token ID | 1830 |
Feature activation | +8.34e-03 |
Pos logit contributions | |
istered | +0.018 |
rier | +0.018 |
ishy | +0.017 |
working | +0.017 |
ivery | +0.017 |
Neg logit contributions | |
peek | -0.006 |
dial | -0.006 |
nob | -0.006 |
ad | -0.005 |
turned | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -8.45e-04 |
Pos logit contributions | |
rier | +0.018 |
istered | +0.018 |
working | +0.017 |
ishy | +0.017 |
ivery | +0.017 |
Neg logit contributions | |
peek | -0.008 |
dial | -0.007 |
nob | -0.007 |
turned | -0.006 |
ad | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -6.75e-03 |
Pos logit contributions | |
peek | +0.001 |
dial | +0.001 |
Looking | +0.001 |
nob | +0.001 |
turned | +0.001 |
Neg logit contributions | |
rier | -0.002 |
istered | -0.002 |
working | -0.002 |
ishy | -0.002 |
goals | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -1.34e-03 |
Pos logit contributions | |
dial | +0.005 |
peek | +0.005 |
nob | +0.005 |
ad | +0.004 |
turned | +0.004 |
Neg logit contributions | |
working | -0.015 |
istered | -0.015 |
rier | -0.015 |
goals | -0.014 |
kicker | -0.014 |
Token | lived |
Token ID | 5615 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
dial | +0.005 |
nob | +0.005 |
peek | +0.005 |
Looking | +0.005 |
ad | +0.005 |
Neg logit contributions | |
working | -0.003 |
istered | -0.003 |
rier | -0.003 |
goals | -0.003 |
ishy | -0.003 |
Token | in |
Token ID | 287 |
Feature activation | +9.28e-05 |
Pos logit contributions | |
rier | +0.032 |
working | +0.032 |
istered | +0.031 |
ishy | +0.029 |
goals | +0.029 |
Neg logit contributions | |
dial | -0.016 |
peek | -0.015 |
nob | -0.015 |
turned | -0.014 |
ad | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
rier | +0.000 |
working | +0.000 |
istered | +0.000 |
ishy | +0.000 |
goals | +0.000 |
Neg logit contributions | |
dial | -0.000 |
peek | -0.000 |
nob | -0.000 |
turned | -0.000 |
ad | -0.000 |
Token | warm |
Token ID | 5814 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
rier | +0.106 |
working | +0.097 |
istered | +0.097 |
ier | +0.096 |
der | +0.094 |
Neg logit contributions | |
peek | -0.050 |
dial | -0.038 |
turned | -0.037 |
story | -0.037 |
opened | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +4.92e-03 |
Pos logit contributions | |
rier | +0.021 |
working | +0.021 |
istered | +0.020 |
ishy | +0.019 |
goals | +0.018 |
Neg logit contributions | |
dial | -0.017 |
peek | -0.017 |
nob | -0.017 |
ad | -0.016 |
turned | -0.016 |
Token | c |
Token ID | 269 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
rier | +0.009 |
istered | +0.008 |
working | +0.007 |
ivery | +0.007 |
ishy | +0.007 |
Neg logit contributions | |
peek | -0.007 |
turned | -0.007 |
dial | -0.007 |
opened | -0.007 |
nob | -0.006 |
Token | udd |
Token ID | 4185 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
nam | +0.113 |
working | +0.106 |
rier | +0.101 |
neither | +0.100 |
istered | +0.100 |
Neg logit contributions | |
peek | -0.100 |
turned | -0.093 |
ad | -0.091 |
yl | -0.090 |
story | -0.088 |
Token | ly |
Token ID | 306 |
Feature activation | +9.91e-03 |
Pos logit contributions | |
istered | +0.065 |
nam | +0.054 |
working | +0.051 |
kicker | +0.051 |
goals | +0.051 |
Neg logit contributions | |
peek | -0.059 |
dial | -0.058 |
turned | -0.056 |
ad | -0.056 |
nob | -0.055 |
Token | home |
Token ID | 1363 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
istered | +0.019 |
rier | +0.018 |
working | +0.017 |
ishy | +0.017 |
sweaty | +0.016 |
Neg logit contributions | |
peek | -0.012 |
dial | -0.012 |
turned | -0.011 |
called | -0.010 |
call | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
rier | +0.033 |
istered | +0.033 |
working | +0.032 |
ishy | +0.031 |
goals | +0.030 |
Neg logit contributions | |
peek | -0.014 |
dial | -0.014 |
nob | -0.013 |
turned | -0.013 |
opened | -0.012 |
Token | Every |
Token ID | 3887 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
working | +0.025 |
goals | +0.023 |
rier | +0.023 |
istered | +0.022 |
bored | +0.021 |
Neg logit contributions | |
dial | -0.026 |
nob | -0.025 |
peek | -0.024 |
fore | -0.024 |
ran | -0.024 |
Token | has |
Token ID | 468 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
rier | +0.016 |
istered | +0.016 |
working | +0.016 |
goals | +0.015 |
ishy | +0.014 |
Neg logit contributions | |
dial | -0.007 |
peek | -0.007 |
nob | -0.006 |
Looking | -0.006 |
ad | -0.006 |
Token | lettuce |
Token ID | 39406 |
Feature activation | +6.37e-03 |
Pos logit contributions | |
rier | +0.026 |
working | +0.026 |
istered | +0.026 |
ishy | +0.023 |
ivery | +0.023 |
Neg logit contributions | |
peek | -0.017 |
dial | -0.017 |
nob | -0.016 |
turned | -0.016 |
ad | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
working | +0.016 |
rier | +0.015 |
istered | +0.015 |
ishy | +0.014 |
ivery | +0.014 |
Neg logit contributions | |
peek | -0.004 |
dial | -0.003 |
turned | -0.003 |
nob | -0.003 |
opened | -0.003 |
Token | tomatoes |
Token ID | 23972 |
Feature activation | +6.41e-03 |
Pos logit contributions | |
istered | +0.022 |
rier | +0.022 |
working | +0.021 |
ivery | +0.020 |
ishy | +0.019 |
Neg logit contributions | |
peek | -0.020 |
dial | -0.020 |
turned | -0.019 |
opened | -0.018 |
nob | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
rier | +0.014 |
working | +0.014 |
istered | +0.014 |
ishy | +0.012 |
ivery | +0.012 |
Neg logit contributions | |
peek | -0.006 |
dial | -0.005 |
nob | -0.005 |
turned | -0.005 |
opened | -0.005 |
Token | cuc |
Token ID | 38421 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
rier | +0.025 |
istered | +0.025 |
working | +0.025 |
ivery | +0.023 |
ishy | +0.022 |
Neg logit contributions | |
peek | -0.025 |
dial | -0.024 |
turned | -0.023 |
opened | -0.022 |
nob | -0.022 |
Token | umbers |
Token ID | 17024 |
Feature activation | -3.52e-03 |
Pos logit contributions | |
working | +0.083 |
istered | +0.075 |
sweaty | +0.072 |
rier | +0.071 |
bored | +0.068 |
Neg logit contributions | |
peek | -0.126 |
dial | -0.123 |
ad | -0.123 |
p | -0.119 |
nob | -0.117 |
Token | . |
Token ID | 13 |
Feature activation | +6.08e-03 |
Pos logit contributions | |
peek | +0.003 |
dial | +0.003 |
turned | +0.003 |
nob | +0.003 |
ad | +0.003 |
Neg logit contributions | |
istered | -0.007 |
working | -0.007 |
rier | -0.007 |
ishy | -0.007 |
ivery | -0.007 |
Token | And |
Token ID | 843 |
Feature activation | +1.52e-03 |
Pos logit contributions | |
rier | +0.010 |
istered | +0.009 |
working | +0.009 |
goals | +0.009 |
ishy | +0.008 |
Neg logit contributions | |
dial | -0.009 |
nob | -0.009 |
peek | -0.009 |
ad | -0.008 |
turned | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | +9.06e-03 |
Pos logit contributions | |
rier | +0.003 |
working | +0.003 |
istered | +0.003 |
ivery | +0.002 |
ishy | +0.002 |
Neg logit contributions | |
peek | -0.002 |
dial | -0.002 |
turned | -0.002 |
opened | -0.002 |
nob | -0.002 |
Token | has |
Token ID | 468 |
Feature activation | +8.36e-03 |
Pos logit contributions | |
rier | +0.020 |
working | +0.019 |
istered | +0.019 |
ishy | +0.018 |
goals | +0.018 |
Neg logit contributions | |
dial | -0.008 |
peek | -0.008 |
nob | -0.007 |
ad | -0.006 |
Looking | -0.006 |
Token | John |
Token ID | 1757 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
rier | +0.039 |
istered | +0.037 |
working | +0.037 |
ishy | +0.034 |
goals | +0.033 |
Neg logit contributions | |
dial | -0.029 |
peek | -0.029 |
nob | -0.027 |
turned | -0.026 |
ad | -0.026 |
Token | chose |
Token ID | 7690 |
Feature activation | -2.47e-03 |
Pos logit contributions | |
rier | +0.032 |
working | +0.032 |
istered | +0.031 |
ishy | +0.029 |
goals | +0.028 |
Neg logit contributions | |
dial | -0.016 |
peek | -0.015 |
nob | -0.015 |
turned | -0.013 |
ad | -0.013 |
Token | some |
Token ID | 617 |
Feature activation | -2.24e-03 |
Pos logit contributions | |
dial | +0.003 |
peek | +0.003 |
nob | +0.003 |
turned | +0.002 |
ad | +0.002 |
Neg logit contributions | |
rier | -0.005 |
working | -0.005 |
istered | -0.005 |
ishy | -0.004 |
goals | -0.004 |
Token | carrots |
Token ID | 34397 |
Feature activation | +6.50e-03 |
Pos logit contributions | |
dial | +0.003 |
peek | +0.003 |
nob | +0.003 |
turned | +0.003 |
ad | +0.003 |
Neg logit contributions | |
rier | -0.003 |
working | -0.003 |
istered | -0.003 |
ishy | -0.003 |
goals | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
istered | +0.015 |
rier | +0.014 |
working | +0.014 |
ishy | +0.013 |
ivery | +0.013 |
Neg logit contributions | |
peek | -0.005 |
dial | -0.005 |
nob | -0.005 |
turned | -0.005 |
ad | -0.004 |
Token | cuc |
Token ID | 38421 |
Feature activation | +7.19e-02 |
Pos logit contributions | |
istered | +0.023 |
rier | +0.023 |
ivery | +0.021 |
working | +0.021 |
ishy | +0.020 |
Neg logit contributions | |
peek | -0.016 |
dial | -0.016 |
turned | -0.015 |
opened | -0.015 |
pat | -0.014 |
Token | umbers |
Token ID | 17024 |
Feature activation | +6.66e-03 |
Pos logit contributions | |
istered | +0.081 |
working | +0.078 |
sweaty | +0.073 |
rier | +0.072 |
bored | +0.071 |
Neg logit contributions | |
dial | -0.124 |
peek | -0.124 |
ad | -0.122 |
p | -0.120 |
nob | -0.118 |
Token | to |
Token ID | 284 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
istered | +0.013 |
working | +0.013 |
rier | +0.013 |
ishy | +0.012 |
goals | +0.012 |
Neg logit contributions | |
peek | -0.007 |
dial | -0.006 |
turned | -0.006 |
nob | -0.006 |
ad | -0.006 |
Token | take |
Token ID | 1011 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
rier | +0.062 |
istered | +0.062 |
working | +0.059 |
ishy | +0.056 |
goals | +0.055 |
Neg logit contributions | |
peek | -0.030 |
dial | -0.030 |
nob | -0.027 |
turned | -0.026 |
ad | -0.026 |
Token | home |
Token ID | 1363 |
Feature activation | -3.43e-03 |
Pos logit contributions | |
rier | +0.035 |
istered | +0.035 |
working | +0.034 |
ishy | +0.031 |
ivery | +0.031 |
Neg logit contributions | |
dial | -0.024 |
peek | -0.024 |
nob | -0.023 |
turned | -0.022 |
ad | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +9.12e-03 |
Pos logit contributions | |
dial | +0.003 |
peek | +0.003 |
nob | +0.003 |
turned | +0.003 |
ad | +0.003 |
Neg logit contributions | |
istered | -0.007 |
rier | -0.007 |
working | -0.007 |
ishy | -0.007 |
kicker | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
dial | +0.014 |
peek | +0.014 |
nob | +0.013 |
turned | +0.013 |
ad | +0.012 |
Neg logit contributions | |
rier | -0.012 |
istered | -0.012 |
working | -0.012 |
ishy | -0.011 |
goals | -0.010 |
Token | Max |
Token ID | 5436 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
peek | +0.027 |
dial | +0.027 |
nob | +0.026 |
turned | +0.025 |
opened | +0.024 |
Neg logit contributions | |
rier | -0.030 |
istered | -0.029 |
working | -0.029 |
ishy | -0.025 |
goals | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
dial | +0.014 |
peek | +0.013 |
nob | +0.013 |
ad | +0.012 |
turned | +0.012 |
Neg logit contributions | |
rier | -0.030 |
working | -0.030 |
istered | -0.029 |
ishy | -0.027 |
goals | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
dial | +0.013 |
peek | +0.013 |
nob | +0.012 |
turned | +0.011 |
ad | +0.011 |
Neg logit contributions | |
rier | -0.026 |
working | -0.025 |
istered | -0.025 |
ishy | -0.023 |
goals | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
working | +0.027 |
rier | +0.025 |
istered | +0.024 |
goals | +0.024 |
kicker | +0.024 |
Neg logit contributions | |
nob | -0.019 |
dial | -0.019 |
peek | -0.018 |
Looking | -0.017 |
turned | -0.017 |
Token | carn |
Token ID | 18466 |
Feature activation | +7.18e-02 |
Pos logit contributions | |
rier | +0.044 |
istered | +0.043 |
working | +0.043 |
ishy | +0.040 |
goals | +0.037 |
Neg logit contributions | |
peek | -0.029 |
dial | -0.029 |
nob | -0.026 |
turned | -0.025 |
ad | -0.025 |
Token | ival |
Token ID | 2473 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
rier | +0.102 |
working | +0.098 |
istered | +0.097 |
ishy | +0.087 |
goals | +0.083 |
Neg logit contributions | |
peek | -0.119 |
dial | -0.116 |
nob | -0.112 |
turned | -0.107 |
ad | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +4.24e-03 |
Pos logit contributions | |
rier | +0.026 |
istered | +0.026 |
working | +0.026 |
ishy | +0.024 |
goals | +0.023 |
Neg logit contributions | |
peek | -0.014 |
dial | -0.014 |
nob | -0.013 |
turned | -0.012 |
ad | -0.012 |
Token | Everywhere |
Token ID | 48201 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
rier | +0.007 |
istered | +0.007 |
working | +0.007 |
ishy | +0.006 |
goals | +0.006 |
Neg logit contributions | |
dial | -0.006 |
peek | -0.006 |
nob | -0.006 |
turned | -0.005 |
Looking | -0.005 |
Token | they |
Token ID | 484 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
rier | +0.049 |
istered | +0.047 |
ivery | +0.045 |
onel | +0.043 |
ishy | +0.042 |
Neg logit contributions | |
dial | -0.024 |
opened | -0.024 |
was | -0.023 |
turned | -0.023 |
peek | -0.023 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.56e-04 |
Pos logit contributions | |
rier | +0.028 |
istered | +0.027 |
ivery | +0.024 |
goals | +0.024 |
working | +0.024 |
Neg logit contributions | |
peek | -0.009 |
dial | -0.008 |
turned | -0.008 |
opened | -0.008 |
called | -0.006 |
Token | One |
Token ID | 3198 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
dial | +0.004 |
peek | +0.004 |
nob | +0.004 |
turned | +0.004 |
ad | +0.004 |
Neg logit contributions | |
rier | -0.007 |
working | -0.007 |
istered | -0.007 |
ishy | -0.006 |
goals | -0.006 |
Token | day |
Token ID | 1110 |
Feature activation | +1.01e-03 |
Pos logit contributions | |
working | +0.016 |
rier | +0.016 |
istered | +0.015 |
ishy | +0.014 |
goals | +0.013 |
Neg logit contributions | |
dial | -0.021 |
nob | -0.020 |
peek | -0.020 |
ad | -0.019 |
Looking | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -2.80e-03 |
Pos logit contributions | |
rier | +0.002 |
istered | +0.002 |
working | +0.002 |
ishy | +0.002 |
goals | +0.002 |
Neg logit contributions | |
dial | -0.001 |
peek | -0.001 |
nob | -0.001 |
turned | -0.001 |
ad | -0.001 |
Token | Tommy |
Token ID | 19919 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
dial | +0.004 |
peek | +0.004 |
turned | +0.004 |
opened | +0.004 |
Looking | +0.003 |
Neg logit contributions | |
rier | -0.005 |
istered | -0.005 |
working | -0.004 |
ishy | -0.004 |
ivery | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +1.36e-03 |
Pos logit contributions | |
working | +0.028 |
istered | +0.027 |
rier | +0.027 |
bored | +0.025 |
ishy | +0.025 |
Neg logit contributions | |
dial | -0.016 |
nob | -0.016 |
peek | -0.015 |
ad | -0.015 |
Looking | -0.015 |
Token | St |
Token ID | 520 |
Feature activation | +7.17e-02 |
Pos logit contributions | |
working | +0.002 |
rier | +0.002 |
istered | +0.002 |
goals | +0.002 |
bored | +0.002 |
Neg logit contributions | |
nob | -0.002 |
peek | -0.002 |
dial | -0.002 |
ad | -0.002 |
Looking | -0.002 |
Token | itch |
Token ID | 2007 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
rier | +0.092 |
working | +0.084 |
nam | +0.080 |
istered | +0.078 |
chard | +0.076 |
Neg logit contributions | |
dial | -0.126 |
ad | -0.122 |
peek | -0.122 |
turned | -0.120 |
Looking | -0.113 |
Token | went |
Token ID | 1816 |
Feature activation | +2.42e-03 |
Pos logit contributions | |
rier | +0.056 |
istered | +0.052 |
working | +0.052 |
ishy | +0.048 |
goals | +0.047 |
Neg logit contributions | |
peek | -0.025 |
dial | -0.024 |
turned | -0.023 |
nob | -0.021 |
opened | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
working | +0.004 |
rier | +0.004 |
istered | +0.004 |
bored | +0.004 |
goals | +0.004 |
Neg logit contributions | |
dial | -0.003 |
nob | -0.003 |
peek | -0.003 |
ad | -0.003 |
Looking | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
rier | +0.019 |
working | +0.019 |
istered | +0.019 |
ishy | +0.017 |
goals | +0.017 |
Neg logit contributions | |
dial | -0.012 |
peek | -0.011 |
nob | -0.011 |
turned | -0.010 |
ad | -0.010 |
Token | park |
Token ID | 3952 |
Feature activation | -1.15e-03 |
Pos logit contributions | |
rier | +0.048 |
istered | +0.046 |
working | +0.046 |
ishy | +0.043 |
ivery | +0.041 |
Neg logit contributions | |
peek | -0.028 |
dial | -0.027 |
nob | -0.025 |
turned | -0.024 |
call | -0.023 |
Token | ." |
Token ID | 526 |
Feature activation | -5.71e-03 |
Pos logit contributions | |
istered | +0.007 |
rier | +0.007 |
working | +0.007 |
ishy | +0.007 |
nam | +0.006 |
Neg logit contributions | |
dial | -0.006 |
peek | -0.006 |
turned | -0.005 |
call | -0.005 |
ad | -0.005 |
Token | She |
Token ID | 1375 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
dial | +0.008 |
peek | +0.008 |
nob | +0.007 |
turned | +0.007 |
ad | +0.007 |
Neg logit contributions | |
rier | -0.010 |
istered | -0.009 |
working | -0.009 |
goals | -0.008 |
ishy | -0.008 |
Token | gives |
Token ID | 3607 |
Feature activation | +1.57e-03 |
Pos logit contributions | |
nob | +0.027 |
peek | +0.025 |
ad | +0.024 |
eway | +0.024 |
peaked | +0.024 |
Neg logit contributions | |
goals | -0.042 |
rier | -0.040 |
istered | -0.040 |
working | -0.039 |
bored | -0.038 |
Token | them |
Token ID | 606 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
istered | +0.003 |
working | +0.003 |
rier | +0.003 |
ishy | +0.003 |
ivery | +0.003 |
Neg logit contributions | |
dial | -0.002 |
peek | -0.002 |
turned | -0.002 |
ad | -0.002 |
nob | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
peek | +0.022 |
dial | +0.021 |
nob | +0.020 |
turned | +0.019 |
ad | +0.019 |
Neg logit contributions | |
rier | -0.032 |
working | -0.031 |
istered | -0.030 |
ishy | -0.027 |
ivery | -0.027 |
Token | new |
Token ID | 649 |
Feature activation | -8.58e-02 |
Pos logit contributions | |
nob | +0.037 |
ad | +0.036 |
ered | +0.032 |
p | +0.032 |
ran | +0.032 |
Neg logit contributions | |
istered | -0.044 |
working | -0.043 |
rier | -0.041 |
bored | -0.039 |
ishy | -0.039 |
Token | tap |
Token ID | 9814 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
nob | +0.135 |
ad | +0.127 |
Looking | +0.120 |
unlocked | +0.119 |
ran | +0.118 |
Neg logit contributions | |
working | -0.121 |
rier | -0.108 |
kicker | -0.106 |
favorite | -0.103 |
ishy | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | +9.67e-03 |
Pos logit contributions | |
dial | +0.028 |
peek | +0.027 |
nob | +0.026 |
Looking | +0.025 |
ad | +0.023 |
Neg logit contributions | |
rier | -0.055 |
working | -0.053 |
istered | -0.053 |
ishy | -0.049 |
goals | -0.049 |
Token | The |
Token ID | 383 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
rier | +0.016 |
working | +0.016 |
istered | +0.016 |
ishy | +0.014 |
goals | +0.014 |
Neg logit contributions | |
dial | -0.013 |
peek | -0.013 |
nob | -0.013 |
turned | -0.012 |
ad | -0.012 |
Token | tap |
Token ID | 9814 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
nob | +0.054 |
dial | +0.052 |
peek | +0.051 |
ad | +0.048 |
Looking | +0.047 |
Neg logit contributions | |
rier | -0.078 |
working | -0.076 |
istered | -0.074 |
goals | -0.070 |
ishy | -0.070 |
Token | is |
Token ID | 318 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
nob | +0.019 |
dial | +0.019 |
peek | +0.018 |
Looking | +0.017 |
ad | +0.016 |
Neg logit contributions | |
rier | -0.037 |
working | -0.036 |
istered | -0.036 |
goals | -0.035 |
ishy | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
peek | +0.000 |
dial | +0.000 |
turned | +0.000 |
nob | +0.000 |
opened | +0.000 |
Neg logit contributions | |
working | -0.001 |
istered | -0.001 |
rier | -0.001 |
ishy | -0.001 |
goals | -0.001 |
Token | said |
Token ID | 531 |
Feature activation | +4.37e-03 |
Pos logit contributions | |
nob | +0.018 |
dial | +0.016 |
peek | +0.016 |
ad | +0.015 |
Looking | +0.014 |
Neg logit contributions | |
working | -0.034 |
rier | -0.033 |
istered | -0.033 |
ishy | -0.031 |
sweaty | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
istered | +0.010 |
working | +0.010 |
rier | +0.010 |
ivery | +0.010 |
ishy | +0.009 |
Neg logit contributions | |
peek | -0.003 |
dial | -0.003 |
turned | -0.003 |
opened | -0.003 |
nob | -0.003 |
Token | " |
Token ID | 366 |
Feature activation | -6.65e-03 |
Pos logit contributions | |
dial | +0.034 |
nob | +0.034 |
peek | +0.034 |
ad | +0.032 |
turned | +0.030 |
Neg logit contributions | |
rier | -0.040 |
working | -0.037 |
istered | -0.036 |
bored | -0.032 |
ishy | -0.032 |
Token | Let |
Token ID | 5756 |
Feature activation | -2.78e-03 |
Pos logit contributions | |
dial | +0.009 |
peek | +0.009 |
nob | +0.008 |
turned | +0.008 |
ad | +0.008 |
Neg logit contributions | |
rier | -0.011 |
working | -0.011 |
istered | -0.011 |
ishy | -0.010 |
kicker | -0.010 |
Token | me |
Token ID | 502 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
peek | +0.003 |
dial | +0.003 |
nob | +0.003 |
turned | +0.003 |
ad | +0.003 |
Neg logit contributions | |
rier | -0.006 |
istered | -0.005 |
working | -0.005 |
ishy | -0.005 |
goals | -0.005 |
Token | help |
Token ID | 1037 |
Feature activation | +1.53e-03 |
Pos logit contributions | |
nob | +0.108 |
dial | +0.102 |
Looking | +0.099 |
ad | +0.095 |
Turning | +0.093 |
Neg logit contributions | |
working | -0.142 |
istered | -0.130 |
ishy | -0.130 |
rier | -0.129 |
bored | -0.120 |
Token | you |
Token ID | 345 |
Feature activation | -4.97e-02 |
Pos logit contributions | |
rier | +0.002 |
istered | +0.002 |
working | +0.002 |
ishy | +0.002 |
goals | +0.002 |
Neg logit contributions | |
peek | -0.002 |
dial | -0.002 |
nob | -0.002 |
turned | -0.002 |
ad | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -2.15e-03 |
Pos logit contributions | |
peek | +0.060 |
dial | +0.059 |
nob | +0.056 |
turned | +0.054 |
ad | +0.052 |
Neg logit contributions | |
rier | -0.092 |
istered | -0.091 |
working | -0.090 |
ishy | -0.082 |
goals | -0.080 |
Token | I |
Token ID | 314 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
dial | +0.003 |
nob | +0.003 |
peek | +0.003 |
ad | +0.003 |
turned | +0.003 |
Neg logit contributions | |
rier | -0.003 |
working | -0.003 |
istered | -0.003 |
ishy | -0.003 |
goals | -0.003 |
Token | will |
Token ID | 481 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
nob | +0.073 |
dial | +0.072 |
ad | +0.068 |
peek | +0.068 |
Looking | +0.067 |
Neg logit contributions | |
working | -0.069 |
istered | -0.068 |
rier | -0.065 |
ishy | -0.064 |
kicker | -0.058 |
Token | she |
Token ID | 673 |
Feature activation | -5.87e-03 |
Pos logit contributions | |
peek | +0.010 |
dial | +0.010 |
turned | +0.009 |
nob | +0.009 |
opened | +0.009 |
Neg logit contributions | |
rier | -0.010 |
istered | -0.010 |
working | -0.010 |
ishy | -0.009 |
goals | -0.009 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
nob | +0.006 |
dial | +0.006 |
peek | +0.006 |
ad | +0.005 |
Looking | +0.005 |
Neg logit contributions | |
rier | -0.011 |
istered | -0.011 |
working | -0.011 |
ishy | -0.010 |
bored | -0.010 |
Token | Teddy |
Token ID | 29345 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
istered | +0.024 |
rier | +0.024 |
working | +0.022 |
ishy | +0.021 |
goals | +0.021 |
Neg logit contributions | |
dial | -0.022 |
peek | -0.021 |
nob | -0.021 |
turned | -0.020 |
ad | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
dial | +0.032 |
peek | +0.032 |
nob | +0.031 |
turned | +0.028 |
ad | +0.028 |
Neg logit contributions | |
rier | -0.073 |
istered | -0.071 |
working | -0.070 |
ishy | -0.065 |
goals | -0.064 |
Token | brand |
Token ID | 4508 |
Feature activation | +3.84e-03 |
Pos logit contributions | |
nob | +0.063 |
ad | +0.061 |
dial | +0.057 |
ran | +0.057 |
p | +0.057 |
Neg logit contributions | |
working | -0.047 |
istered | -0.047 |
rier | -0.042 |
ishy | -0.042 |
sweaty | -0.040 |
Token | new |
Token ID | 649 |
Feature activation | -8.23e-02 |
Pos logit contributions | |
rier | +0.005 |
working | +0.004 |
istered | +0.004 |
ishy | +0.004 |
goals | +0.004 |
Neg logit contributions | |
nob | -0.007 |
dial | -0.007 |
peek | -0.007 |
ad | -0.006 |
opened | -0.006 |
Token | sc |
Token ID | 629 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
nob | +0.120 |
Agent | +0.111 |
r | +0.110 |
dial | +0.108 |
p | +0.107 |
Neg logit contributions | |
working | -0.123 |
rier | -0.108 |
favorite | -0.103 |
runner | -0.102 |
kicker | -0.100 |
Token | ooter |
Token ID | 25141 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
dial | +0.027 |
peek | +0.026 |
nob | +0.026 |
turned | +0.024 |
Looking | +0.023 |
Neg logit contributions | |
rier | -0.034 |
working | -0.033 |
istered | -0.032 |
ishy | -0.030 |
goals | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -7.29e-03 |
Pos logit contributions | |
peek | +0.015 |
dial | +0.014 |
nob | +0.014 |
turned | +0.013 |
ad | +0.013 |
Neg logit contributions | |
rier | -0.031 |
istered | -0.030 |
working | -0.029 |
ishy | -0.028 |
goals | -0.028 |
Token | He |
Token ID | 679 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
dial | +0.011 |
peek | +0.011 |
nob | +0.010 |
Looking | +0.010 |
turned | +0.010 |
Neg logit contributions | |
rier | -0.012 |
working | -0.011 |
istered | -0.011 |
ishy | -0.010 |
goals | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | +1.91e-03 |
Pos logit contributions | |
nob | +0.020 |
dial | +0.019 |
peek | +0.018 |
ad | +0.018 |
Looking | +0.018 |
Neg logit contributions | |
working | -0.042 |
istered | -0.041 |
rier | -0.040 |
ishy | -0.037 |
kicker | -0.037 |
Token | mess |
Token ID | 2085 |
Feature activation | -2.03e-03 |
Pos logit contributions | |
nob | +0.059 |
dial | +0.059 |
ad | +0.059 |
Looking | +0.058 |
peek | +0.057 |
Neg logit contributions | |
istered | -0.052 |
rier | -0.050 |
working | -0.049 |
ishy | -0.046 |
sweaty | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +7.42e-03 |
Pos logit contributions | |
dial | +0.001 |
peek | +0.001 |
nob | +0.001 |
turned | +0.001 |
ad | +0.001 |
Neg logit contributions | |
rier | -0.005 |
working | -0.005 |
istered | -0.005 |
ishy | -0.005 |
kicker | -0.005 |
Token | gave |
Token ID | 2921 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
rier | +0.015 |
working | +0.015 |
istered | +0.014 |
bored | +0.013 |
goals | +0.013 |
Neg logit contributions | |
nob | -0.007 |
dial | -0.007 |
peek | -0.007 |
ad | -0.007 |
Looking | -0.006 |
Token | her |
Token ID | 607 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
istered | +0.047 |
working | +0.046 |
rier | +0.046 |
ivery | +0.045 |
ishy | +0.042 |
Neg logit contributions | |
dial | -0.026 |
peek | -0.024 |
turned | -0.022 |
nob | -0.022 |
ad | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
dial | +0.036 |
peek | +0.035 |
nob | +0.034 |
turned | +0.032 |
ad | +0.031 |
Neg logit contributions | |
rier | -0.065 |
working | -0.063 |
istered | -0.063 |
ishy | -0.058 |
goals | -0.057 |
Token | new |
Token ID | 649 |
Feature activation | -8.14e-02 |
Pos logit contributions | |
nob | +0.036 |
ad | +0.034 |
ered | +0.031 |
ran | +0.031 |
p | +0.031 |
Neg logit contributions | |
istered | -0.040 |
working | -0.039 |
ishy | -0.036 |
rier | -0.036 |
bored | -0.035 |
Token | cup |
Token ID | 6508 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
nob | +0.143 |
ran | +0.134 |
ad | +0.133 |
Looking | +0.129 |
dial | +0.127 |
Neg logit contributions | |
istered | -0.092 |
working | -0.091 |
ishy | -0.090 |
favorite | -0.089 |
rier | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | +5.94e-03 |
Pos logit contributions | |
dial | +0.015 |
peek | +0.015 |
nob | +0.014 |
Looking | +0.013 |
turned | +0.013 |
Neg logit contributions | |
rier | -0.023 |
istered | -0.022 |
working | -0.022 |
ishy | -0.020 |
goals | -0.020 |
Token | Sara |
Token ID | 24799 |
Feature activation | +3.80e-03 |
Pos logit contributions | |
rier | +0.009 |
istered | +0.009 |
working | +0.009 |
ishy | +0.008 |
goals | +0.008 |
Neg logit contributions | |
dial | -0.009 |
peek | -0.009 |
nob | -0.008 |
ad | -0.008 |
turned | -0.008 |
Token | thanked |
Token ID | 26280 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
istered | +0.008 |
rier | +0.007 |
working | +0.007 |
goals | +0.007 |
ishy | +0.007 |
Neg logit contributions | |
nob | -0.003 |
dial | -0.003 |
peek | -0.003 |
Looking | -0.003 |
ad | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
rier | +0.019 |
working | +0.018 |
istered | +0.018 |
goals | +0.016 |
ishy | +0.016 |
Neg logit contributions | |
peek | -0.021 |
dial | -0.021 |
nob | -0.021 |
turned | -0.020 |
Looking | -0.020 |
Token | he |
Token ID | 339 |
Feature activation | -5.18e-03 |
Pos logit contributions | |
dial | +0.001 |
peek | +0.001 |
turned | +0.001 |
ad | +0.001 |
nob | +0.001 |
Neg logit contributions | |
istered | -0.001 |
rier | -0.001 |
working | -0.001 |
ishy | -0.001 |
ivery | -0.001 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
nob | +0.006 |
dial | +0.005 |
peek | +0.005 |
ad | +0.005 |
Looking | +0.005 |
Neg logit contributions | |
rier | -0.010 |
working | -0.009 |
istered | -0.009 |
ishy | -0.009 |
kicker | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
rier | +0.021 |
working | +0.021 |
istered | +0.021 |
ishy | +0.019 |
goals | +0.019 |
Neg logit contributions | |
dial | -0.013 |
peek | -0.013 |
nob | -0.013 |
turned | -0.012 |
ad | -0.012 |
Token | boy |
Token ID | 2933 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
nob | +0.080 |
dial | +0.075 |
ad | +0.073 |
peek | +0.072 |
Looking | +0.072 |
Neg logit contributions | |
rier | -0.064 |
working | -0.061 |
istered | -0.055 |
bored | -0.054 |
ishy | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
dial | +0.038 |
peek | +0.037 |
nob | +0.036 |
turned | +0.033 |
ad | +0.033 |
Neg logit contributions | |
rier | -0.070 |
working | -0.068 |
istered | -0.068 |
ishy | -0.062 |
goals | -0.061 |
Token | biggest |
Token ID | 4094 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
nob | +0.093 |
dial | +0.088 |
ad | +0.086 |
Looking | +0.083 |
peek | +0.082 |
Neg logit contributions | |
rier | -0.087 |
working | -0.080 |
istered | -0.079 |
ishy | -0.078 |
sweaty | -0.076 |
Token | hug |
Token ID | 16225 |
Feature activation | -6.30e-03 |
Pos logit contributions | |
nob | +0.124 |
dial | +0.118 |
Looking | +0.113 |
p | +0.109 |
Agent | +0.109 |
Neg logit contributions | |
rier | -0.108 |
istered | -0.102 |
working | -0.102 |
ishy | -0.101 |
sweaty | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
peek | +0.005 |
dial | +0.005 |
nob | +0.005 |
turned | +0.005 |
ad | +0.005 |
Neg logit contributions | |
rier | -0.014 |
istered | -0.014 |
working | -0.014 |
ishy | -0.013 |
goals | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
dial | +0.012 |
nob | +0.012 |
peek | +0.012 |
ad | +0.010 |
turned | +0.010 |
Neg logit contributions | |
rier | -0.021 |
working | -0.020 |
istered | -0.020 |
goals | -0.019 |
kicker | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -4.41e-03 |
Pos logit contributions | |
dial | +0.021 |
nob | +0.019 |
peek | +0.018 |
turned | +0.016 |
bul | +0.016 |
Neg logit contributions | |
istered | -0.024 |
goals | -0.024 |
working | -0.024 |
rier | -0.024 |
dd | -0.022 |
Token | Together |
Token ID | 41631 |
Feature activation | -2.00e-03 |
Pos logit contributions | |
dial | +0.006 |
peek | +0.006 |
nob | +0.006 |
turned | +0.005 |
call | +0.005 |
Neg logit contributions | |
rier | -0.007 |
istered | -0.007 |
working | -0.007 |
goals | -0.007 |
kicker | -0.006 |
Token | � |
Token ID | 95 |
Feature activation | +3.60e-03 |
Pos logit contributions | |
rier | +0.037 |
working | +0.037 |
istered | +0.036 |
ishy | +0.032 |
goals | +0.031 |
Neg logit contributions | |
dial | -0.035 |
peek | -0.035 |
nob | -0.034 |
turned | -0.033 |
ad | -0.032 |
Token | € |
Token ID | 26391 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
working | +0.006 |
istered | +0.006 |
rier | +0.005 |
ivery | +0.005 |
ishy | +0.005 |
Neg logit contributions | |
peek | -0.005 |
dial | -0.005 |
turned | -0.004 |
opened | -0.004 |
pat | -0.004 |
Token | � |
Token ID | 129 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
dial | +0.017 |
peek | +0.016 |
nob | +0.016 |
turned | +0.015 |
ad | +0.015 |
Neg logit contributions | |
rier | -0.019 |
working | -0.018 |
istered | -0.018 |
ishy | -0.016 |
goals | -0.016 |
Token | � |
Token ID | 241 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
rier | +0.022 |
working | +0.021 |
istered | +0.021 |
ishy | +0.018 |
goals | +0.017 |
Neg logit contributions | |
dial | -0.030 |
peek | -0.030 |
nob | -0.029 |
turned | -0.028 |
ad | -0.028 |
Token | Yes |
Token ID | 5297 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
dial | +0.059 |
nob | +0.058 |
peek | +0.057 |
turned | +0.052 |
ad | +0.051 |
Neg logit contributions | |
working | -0.076 |
rier | -0.072 |
istered | -0.072 |
ishy | -0.065 |
kicker | -0.064 |
Token | , |
Token ID | 11 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
dial | +0.015 |
peek | +0.015 |
nob | +0.014 |
turned | +0.013 |
ad | +0.013 |
Neg logit contributions | |
rier | -0.025 |
working | -0.024 |
istered | -0.024 |
ishy | -0.022 |
goals | -0.022 |
Token | but |
Token ID | 475 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
nob | +0.124 |
dial | +0.109 |
peek | +0.102 |
Looking | +0.100 |
Turning | +0.098 |
Neg logit contributions | |
working | -0.112 |
rier | -0.106 |
guys | -0.096 |
nam | -0.095 |
sweaty | -0.095 |
Token | be |
Token ID | 307 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
nob | +0.057 |
dial | +0.053 |
Looking | +0.049 |
ad | +0.049 |
peek | +0.048 |
Neg logit contributions | |
working | -0.048 |
rier | -0.045 |
nam | -0.041 |
neither | -0.040 |
none | -0.040 |
Token | very |
Token ID | 845 |
Feature activation | +5.50e-03 |
Pos logit contributions | |
dial | +0.033 |
peek | +0.033 |
nob | +0.032 |
ad | +0.031 |
turned | +0.031 |
Neg logit contributions | |
rier | -0.014 |
istered | -0.014 |
working | -0.013 |
ishy | -0.011 |
goals | -0.011 |
Token | careful |
Token ID | 8161 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
rier | +0.008 |
istered | +0.008 |
working | +0.008 |
ishy | +0.007 |
goals | +0.007 |
Neg logit contributions | |
peek | -0.009 |
dial | -0.009 |
turned | -0.008 |
nob | -0.008 |
opened | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
nob | +0.012 |
peek | +0.012 |
dial | +0.012 |
ad | +0.011 |
turned | +0.011 |
Neg logit contributions | |
istered | -0.030 |
rier | -0.029 |
working | -0.029 |
ishy | -0.028 |
goals | -0.026 |
Token | taught |
Token ID | 7817 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
nob | +0.009 |
eway | +0.009 |
Looking | +0.008 |
Turning | +0.008 |
izer | +0.008 |
Neg logit contributions | |
istered | -0.012 |
working | -0.011 |
bored | -0.011 |
sweaty | -0.011 |
rier | -0.011 |
Token | her |
Token ID | 607 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
rier | +0.020 |
working | +0.019 |
istered | +0.019 |
goals | +0.018 |
ishy | +0.017 |
Neg logit contributions | |
dial | -0.027 |
nob | -0.027 |
peek | -0.027 |
Looking | -0.025 |
peaked | -0.025 |
Token | how |
Token ID | 703 |
Feature activation | +5.19e-03 |
Pos logit contributions | |
dial | +0.044 |
peek | +0.043 |
nob | +0.041 |
turned | +0.038 |
ad | +0.038 |
Neg logit contributions | |
rier | -0.072 |
working | -0.071 |
istered | -0.070 |
goals | -0.064 |
ishy | -0.064 |
Token | to |
Token ID | 284 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
rier | +0.012 |
istered | +0.011 |
working | +0.011 |
ishy | +0.011 |
goals | +0.010 |
Neg logit contributions | |
peek | -0.004 |
dial | -0.004 |
nob | -0.004 |
turned | -0.004 |
ad | -0.003 |
Token | use |
Token ID | 779 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
Looking | +0.039 |
nob | +0.036 |
dial | +0.036 |
Isn | +0.035 |
Turning | +0.035 |
Neg logit contributions | |
working | -0.059 |
istered | -0.053 |
ishy | -0.051 |
ivery | -0.050 |
rier | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
nob | +0.065 |
dial | +0.064 |
peek | +0.064 |
Looking | +0.063 |
Turning | +0.060 |
Neg logit contributions | |
rier | -0.074 |
working | -0.073 |
istered | -0.070 |
goals | -0.069 |
ishy | -0.065 |
Token | f |
Token ID | 277 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
nob | +0.127 |
ad | +0.112 |
unlocked | +0.109 |
Looking | +0.108 |
Ell | +0.108 |
Neg logit contributions | |
rier | -0.112 |
bored | -0.104 |
ishy | -0.100 |
obb | -0.099 |
working | -0.098 |
Token | auc |
Token ID | 14272 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
dial | +0.089 |
peek | +0.088 |
nob | +0.087 |
Looking | +0.082 |
opened | +0.081 |
Neg logit contributions | |
rier | -0.069 |
istered | -0.063 |
working | -0.063 |
ishy | -0.063 |
attered | -0.057 |
Token | et |
Token ID | 316 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
dial | +0.051 |
peek | +0.050 |
nob | +0.047 |
Looking | +0.047 |
opened | +0.046 |
Neg logit contributions | |
rier | -0.042 |
working | -0.041 |
ishy | -0.040 |
istered | -0.040 |
goals | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +8.10e-03 |
Pos logit contributions | |
dial | +0.062 |
peek | +0.060 |
nob | +0.058 |
Looking | +0.055 |
ad | +0.052 |
Neg logit contributions | |
rier | -0.097 |
working | -0.096 |
istered | -0.091 |
ishy | -0.086 |
goals | -0.085 |
Token | Now |
Token ID | 2735 |
Feature activation | +5.98e-03 |
Pos logit contributions | |
rier | +0.013 |
istered | +0.013 |
working | +0.013 |
ishy | +0.011 |
goals | +0.011 |
Neg logit contributions | |
dial | -0.012 |
peek | -0.011 |
nob | -0.011 |
ad | -0.010 |
turned | -0.010 |
Token | can |
Token ID | 460 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
nob | +0.060 |
dial | +0.060 |
peek | +0.055 |
ad | +0.055 |
Looking | +0.054 |
Neg logit contributions | |
working | -0.057 |
istered | -0.057 |
rier | -0.054 |
ishy | -0.052 |
kicker | -0.049 |
Token | help |
Token ID | 1037 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
peek | +0.013 |
dial | +0.013 |
nob | +0.012 |
turned | +0.011 |
ad | +0.011 |
Neg logit contributions | |
rier | -0.028 |
istered | -0.028 |
working | -0.028 |
ishy | -0.026 |
goals | -0.025 |
Token | ." |
Token ID | 526 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
peek | +0.017 |
dial | +0.017 |
nob | +0.017 |
turned | +0.016 |
ad | +0.016 |
Neg logit contributions | |
rier | -0.014 |
istered | -0.014 |
working | -0.014 |
ishy | -0.012 |
goals | -0.012 |
Token | He |
Token ID | 679 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
nob | +0.080 |
dial | +0.074 |
spell | +0.074 |
flap | +0.073 |
grant | +0.070 |
Neg logit contributions | |
rier | -0.069 |
sweaty | -0.064 |
working | -0.063 |
attered | -0.063 |
bored | -0.062 |
Token | took |
Token ID | 1718 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
nob | +0.038 |
ad | +0.034 |
ey | +0.032 |
Looking | +0.031 |
Turning | +0.030 |
Neg logit contributions | |
working | -0.047 |
rier | -0.045 |
kicker | -0.042 |
istered | -0.041 |
der | -0.041 |
Token | out |
Token ID | 503 |
Feature activation | -7.79e-02 |
Pos logit contributions | |
nob | +0.048 |
dial | +0.047 |
peek | +0.046 |
Turning | +0.045 |
story | +0.044 |
Neg logit contributions | |
working | -0.054 |
rier | -0.053 |
ishy | -0.049 |
istered | -0.049 |
bored | -0.045 |
Token | a |
Token ID | 257 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
nob | +0.120 |
dial | +0.108 |
Turning | +0.107 |
peek | +0.107 |
Looking | +0.106 |
Neg logit contributions | |
working | -0.104 |
rier | -0.101 |
istered | -0.098 |
nam | -0.094 |
bored | -0.093 |
Token | fishing |
Token ID | 12478 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
nob | +0.078 |
ad | +0.076 |
dial | +0.070 |
p | +0.067 |
Looking | +0.067 |
Neg logit contributions | |
working | -0.068 |
rier | -0.067 |
istered | -0.066 |
ishy | -0.060 |
sweaty | -0.058 |
Token | rod |
Token ID | 15299 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
dial | +0.047 |
nob | +0.047 |
Looking | +0.045 |
peek | +0.044 |
ad | +0.043 |
Neg logit contributions | |
rier | -0.038 |
working | -0.035 |
goals | -0.032 |
istered | -0.031 |
kicker | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
dial | +0.012 |
peek | +0.012 |
nob | +0.011 |
turned | +0.010 |
ad | +0.010 |
Neg logit contributions | |
rier | -0.040 |
working | -0.039 |
istered | -0.039 |
ishy | -0.036 |
goals | -0.036 |
Token | swoop |
Token ID | 38527 |
Feature activation | -1.84e-03 |
Pos logit contributions | |
dial | +0.014 |
peek | +0.014 |
nob | +0.014 |
ad | +0.013 |
Looking | +0.013 |
Neg logit contributions | |
rier | -0.023 |
working | -0.023 |
istered | -0.022 |
ishy | -0.020 |
goals | -0.020 |
Token | said |
Token ID | 531 |
Feature activation | +8.85e-03 |
Pos logit contributions | |
rier | +0.020 |
working | +0.020 |
istered | +0.019 |
ishy | +0.018 |
kicker | +0.018 |
Neg logit contributions | |
nob | -0.010 |
dial | -0.010 |
peek | -0.010 |
ad | -0.009 |
Looking | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
istered | +0.018 |
working | +0.018 |
rier | +0.017 |
ishy | +0.016 |
goals | +0.016 |
Neg logit contributions | |
peek | -0.009 |
dial | -0.009 |
nob | -0.009 |
turned | -0.008 |
ad | -0.008 |
Token | " |
Token ID | 366 |
Feature activation | +7.85e-03 |
Pos logit contributions | |
nob | +0.022 |
dial | +0.021 |
peek | +0.021 |
ad | +0.021 |
turned | +0.019 |
Neg logit contributions | |
rier | -0.023 |
working | -0.021 |
istered | -0.020 |
bored | -0.018 |
der | -0.018 |
Token | Please |
Token ID | 5492 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
rier | +0.013 |
working | +0.013 |
istered | +0.013 |
ishy | +0.012 |
goals | +0.011 |
Neg logit contributions | |
dial | -0.011 |
peek | -0.010 |
nob | -0.010 |
turned | -0.010 |
ad | -0.009 |
Token | let |
Token ID | 1309 |
Feature activation | -3.58e-03 |
Pos logit contributions | |
dial | +0.029 |
nob | +0.028 |
peek | +0.028 |
Looking | +0.026 |
ad | +0.026 |
Neg logit contributions | |
rier | -0.032 |
working | -0.031 |
istered | -0.031 |
goals | -0.027 |
ishy | -0.027 |
Token | me |
Token ID | 502 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
peek | +0.003 |
dial | +0.003 |
nob | +0.003 |
turned | +0.003 |
ad | +0.003 |
Neg logit contributions | |
rier | -0.008 |
istered | -0.008 |
working | -0.008 |
ishy | -0.007 |
nam | -0.007 |
Token | march |
Token ID | 9960 |
Feature activation | -7.54e-03 |
Pos logit contributions | |
nob | +0.098 |
dial | +0.095 |
Looking | +0.090 |
ad | +0.088 |
peek | +0.084 |
Neg logit contributions | |
working | -0.132 |
rier | -0.124 |
istered | -0.119 |
ishy | -0.116 |
bored | -0.115 |
Token | , |
Token ID | 11 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
peek | +0.008 |
dial | +0.008 |
nob | +0.008 |
turned | +0.008 |
ad | +0.007 |
Neg logit contributions | |
istered | -0.015 |
rier | -0.014 |
working | -0.014 |
ishy | -0.013 |
goals | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
dial | +0.032 |
peek | +0.031 |
nob | +0.031 |
turned | +0.029 |
ad | +0.029 |
Neg logit contributions | |
rier | -0.037 |
working | -0.036 |
istered | -0.036 |
ishy | -0.033 |
goals | -0.032 |
Token | is |
Token ID | 318 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
dial | +0.032 |
peek | +0.032 |
nob | +0.030 |
turned | +0.029 |
ad | +0.028 |
Neg logit contributions | |
rier | -0.044 |
working | -0.043 |
istered | -0.043 |
ishy | -0.039 |
goals | -0.038 |
Token | so |
Token ID | 523 |
Feature activation | -7.33e-04 |
Pos logit contributions | |
dial | +0.040 |
peek | +0.040 |
nob | +0.039 |
turned | +0.037 |
ad | +0.037 |
Neg logit contributions | |
rier | -0.043 |
working | -0.042 |
istered | -0.041 |
ishy | -0.037 |
goals | -0.037 |
Token | He |
Token ID | 1544 |
Feature activation | -7.92e-03 |
Pos logit contributions | |
dial | +0.003 |
peek | +0.003 |
nob | +0.003 |
turned | +0.003 |
opened | +0.003 |
Neg logit contributions | |
working | -0.004 |
rier | -0.004 |
istered | -0.004 |
goals | -0.004 |
kicker | -0.003 |
Token | would |
Token ID | 561 |
Feature activation | +4.60e-03 |
Pos logit contributions | |
nob | +0.010 |
dial | +0.009 |
peek | +0.009 |
Looking | +0.009 |
ad | +0.009 |
Neg logit contributions | |
working | -0.013 |
rier | -0.013 |
istered | -0.013 |
goals | -0.012 |
ishy | -0.012 |
Token | still |
Token ID | 991 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
working | +0.008 |
rier | +0.008 |
istered | +0.008 |
goals | +0.007 |
ishy | +0.007 |
Neg logit contributions | |
dial | -0.006 |
nob | -0.005 |
peek | -0.005 |
Looking | -0.005 |
turned | -0.005 |
Token | take |
Token ID | 1011 |
Feature activation | +5.33e-03 |
Pos logit contributions | |
dial | +0.023 |
nob | +0.023 |
Looking | +0.021 |
peek | +0.020 |
Turning | +0.020 |
Neg logit contributions | |
working | -0.032 |
rier | -0.031 |
istered | -0.031 |
goals | -0.029 |
ishy | -0.029 |
Token | his |
Token ID | 465 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
rier | +0.009 |
working | +0.008 |
istered | +0.008 |
goals | +0.008 |
ishy | +0.007 |
Neg logit contributions | |
dial | -0.007 |
peek | -0.007 |
nob | -0.007 |
Looking | -0.007 |
turned | -0.006 |
Token | long |
Token ID | 890 |
Feature activation | -7.68e-02 |
Pos logit contributions | |
nob | +0.081 |
dial | +0.076 |
ad | +0.075 |
Looking | +0.072 |
peek | +0.070 |
Neg logit contributions | |
working | -0.093 |
ishy | -0.082 |
istered | -0.081 |
rier | -0.079 |
bored | -0.078 |
Token | stroll |
Token ID | 41360 |
Feature activation | -2.82e-03 |
Pos logit contributions | |
nob | +0.108 |
dial | +0.108 |
Looking | +0.105 |
Turning | +0.102 |
Turn | +0.100 |
Neg logit contributions | |
working | -0.126 |
ishy | -0.105 |
guys | -0.100 |
muscles | -0.097 |
rainy | -0.095 |
Token | s |
Token ID | 82 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
peek | +0.003 |
dial | +0.002 |
turned | +0.002 |
nob | +0.002 |
opened | +0.002 |
Neg logit contributions | |
istered | -0.006 |
rier | -0.006 |
working | -0.006 |
ivery | -0.006 |
ishy | -0.005 |
Token | through |
Token ID | 832 |
Feature activation | -5.15e-03 |
Pos logit contributions | |
dial | +0.025 |
nob | +0.024 |
peek | +0.024 |
Looking | +0.022 |
ad | +0.022 |
Neg logit contributions | |
rier | -0.040 |
working | -0.039 |
istered | -0.038 |
goals | -0.036 |
ishy | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -7.56e-03 |
Pos logit contributions | |
dial | +0.006 |
peek | +0.006 |
nob | +0.006 |
turned | +0.005 |
ad | +0.005 |
Neg logit contributions | |
rier | -0.010 |
working | -0.010 |
istered | -0.009 |
goals | -0.009 |
ishy | -0.009 |
Token | park |
Token ID | 3952 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
nob | +0.012 |
dial | +0.012 |
peek | +0.011 |
ad | +0.010 |
Looking | +0.010 |
Neg logit contributions | |
working | -0.011 |
goals | -0.010 |
rier | -0.010 |
istered | -0.010 |
ishy | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
istered | +0.019 |
working | +0.018 |
rier | +0.018 |
ivery | +0.018 |
ishy | +0.017 |
Neg logit contributions | |
peek | -0.007 |
opened | -0.006 |
turned | -0.006 |
: | -0.006 |
dial | -0.005 |
Token | bear |
Token ID | 6842 |
Feature activation | -6.05e-03 |
Pos logit contributions | |
dial | +0.041 |
peek | +0.040 |
nob | +0.038 |
turned | +0.037 |
ad | +0.036 |
Neg logit contributions | |
rier | -0.053 |
istered | -0.051 |
working | -0.050 |
ishy | -0.045 |
goals | -0.045 |
Token | showed |
Token ID | 3751 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
dial | +0.006 |
peek | +0.006 |
nob | +0.005 |
ad | +0.005 |
turned | +0.005 |
Neg logit contributions | |
rier | -0.012 |
working | -0.012 |
istered | -0.012 |
ishy | -0.011 |
goals | -0.011 |
Token | everyone |
Token ID | 2506 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
istered | +0.040 |
working | +0.038 |
rier | +0.037 |
ivery | +0.036 |
ishy | +0.035 |
Neg logit contributions | |
dial | -0.036 |
peek | -0.036 |
ad | -0.034 |
nob | -0.033 |
turned | -0.033 |
Token | his |
Token ID | 465 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
dial | +0.039 |
peek | +0.039 |
nob | +0.037 |
turned | +0.035 |
ad | +0.034 |
Neg logit contributions | |
rier | -0.063 |
working | -0.061 |
istered | -0.061 |
ishy | -0.055 |
goals | -0.055 |
Token | new |
Token ID | 649 |
Feature activation | -7.56e-02 |
Pos logit contributions | |
dial | +0.080 |
peek | +0.079 |
nob | +0.079 |
Looking | +0.074 |
ad | +0.074 |
Neg logit contributions | |
working | -0.079 |
rier | -0.078 |
istered | -0.078 |
ishy | -0.072 |
goals | -0.069 |
Token | boots |
Token ID | 14412 |
Feature activation | -1.86e-03 |
Pos logit contributions | |
dial | +0.112 |
nob | +0.112 |
peek | +0.109 |
Looking | +0.107 |
Turning | +0.105 |
Neg logit contributions | |
working | -0.108 |
rier | -0.104 |
istered | -0.100 |
ishy | -0.096 |
goals | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | +9.18e-03 |
Pos logit contributions | |
peek | +0.001 |
dial | +0.001 |
nob | +0.001 |
turned | +0.001 |
opened | +0.001 |
Neg logit contributions | |
rier | -0.004 |
istered | -0.004 |
working | -0.004 |
ishy | -0.004 |
goals | -0.004 |
Token | Everyone |
Token ID | 11075 |
Feature activation | -7.27e-04 |
Pos logit contributions | |
rier | +0.016 |
working | +0.015 |
istered | +0.015 |
ishy | +0.014 |
goals | +0.014 |
Neg logit contributions | |
dial | -0.012 |
peek | -0.012 |
nob | -0.012 |
turned | -0.011 |
ad | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | -1.03e-03 |
Pos logit contributions | |
nob | +0.001 |
dial | +0.001 |
peek | +0.001 |
ad | +0.001 |
Looking | +0.001 |
Neg logit contributions | |
working | -0.001 |
rier | -0.001 |
istered | -0.001 |
bored | -0.001 |
goals | -0.001 |
Token | so |
Token ID | 523 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
dial | +0.001 |
peek | +0.001 |
nob | +0.001 |
turned | +0.001 |
ad | +0.001 |
Neg logit contributions | |
rier | -0.002 |
working | -0.002 |
istered | -0.002 |
ishy | -0.001 |
goals | -0.001 |
Token | park |
Token ID | 3952 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
rier | +0.005 |
working | +0.005 |
istered | +0.005 |
ishy | +0.004 |
ivery | +0.004 |
Neg logit contributions | |
peek | -0.003 |
dial | -0.003 |
turned | -0.003 |
nob | -0.003 |
ad | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
istered | +0.028 |
rier | +0.028 |
working | +0.028 |
ishy | +0.026 |
ivery | +0.026 |
Neg logit contributions | |
peek | -0.015 |
dial | -0.014 |
turned | -0.013 |
nob | -0.013 |
ad | -0.013 |
Token | Timothy |
Token ID | 22283 |
Feature activation | -2.30e-03 |
Pos logit contributions | |
nob | +0.027 |
dial | +0.027 |
peek | +0.027 |
spell | +0.024 |
ad | +0.024 |
Neg logit contributions | |
working | -0.032 |
rier | -0.032 |
istered | -0.032 |
goals | -0.030 |
attered | -0.029 |
Token | gave |
Token ID | 2921 |
Feature activation | +8.73e-03 |
Pos logit contributions | |
dial | +0.002 |
peek | +0.002 |
nob | +0.002 |
ad | +0.002 |
Looking | +0.002 |
Neg logit contributions | |
working | -0.005 |
rier | -0.005 |
istered | -0.005 |
ishy | -0.004 |
goals | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
rier | +0.014 |
working | +0.013 |
istered | +0.013 |
ishy | +0.012 |
goals | +0.011 |
Neg logit contributions | |
dial | -0.013 |
peek | -0.013 |
nob | -0.013 |
turned | -0.012 |
ad | -0.012 |
Token | final |
Token ID | 2457 |
Feature activation | -7.52e-02 |
Pos logit contributions | |
nob | +0.081 |
ad | +0.073 |
dial | +0.073 |
ran | +0.067 |
Looking | +0.067 |
Neg logit contributions | |
working | -0.058 |
istered | -0.057 |
rier | -0.056 |
ishy | -0.053 |
kicker | -0.051 |
Token | loud |
Token ID | 7812 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
nob | +0.108 |
dial | +0.092 |
ran | +0.086 |
unlocks | +0.086 |
ad | +0.084 |
Neg logit contributions | |
working | -0.121 |
ishy | -0.112 |
rier | -0.111 |
kicker | -0.108 |
istered | -0.106 |
Token | blow |
Token ID | 6611 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
nob | +0.071 |
dial | +0.066 |
peek | +0.059 |
ran | +0.057 |
unlocked | +0.057 |
Neg logit contributions | |
rier | -0.092 |
working | -0.087 |
ishy | -0.077 |
istered | -0.076 |
kicker | -0.074 |
Token | on |
Token ID | 319 |
Feature activation | +3.43e-04 |
Pos logit contributions | |
dial | +0.036 |
nob | +0.035 |
peek | +0.035 |
Looking | +0.031 |
turned | +0.031 |
Neg logit contributions | |
rier | -0.052 |
working | -0.048 |
istered | -0.047 |
ishy | -0.045 |
goals | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
istered | +0.001 |
rier | +0.001 |
working | +0.001 |
ishy | +0.001 |
ivery | +0.001 |
Neg logit contributions | |
dial | -0.000 |
peek | -0.000 |
turned | -0.000 |
nob | -0.000 |
ad | -0.000 |
Token | whistle |
Token ID | 16121 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
peek | +0.017 |
dial | +0.017 |
nob | +0.016 |
turned | +0.015 |
ad | +0.015 |
Neg logit contributions | |
rier | -0.023 |
working | -0.023 |
istered | -0.023 |
ishy | -0.020 |
goals | -0.020 |
Token | 's |
Token ID | 338 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
dial | +0.003 |
peek | +0.003 |
nob | +0.003 |
ad | +0.003 |
turned | +0.003 |
Neg logit contributions | |
rier | -0.006 |
working | -0.006 |
istered | -0.006 |
ishy | -0.006 |
goals | -0.006 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
nob | +0.031 |
dial | +0.030 |
peek | +0.029 |
Looking | +0.028 |
ad | +0.028 |
Neg logit contributions | |
rier | -0.023 |
istered | -0.022 |
working | -0.021 |
ishy | -0.019 |
goals | -0.019 |
Token | said |
Token ID | 531 |
Feature activation | +4.24e-03 |
Pos logit contributions | |
nob | +0.018 |
dial | +0.017 |
Looking | +0.016 |
ad | +0.016 |
peek | +0.016 |
Neg logit contributions | |
istered | -0.031 |
working | -0.031 |
rier | -0.029 |
bored | -0.029 |
sweaty | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
istered | +0.010 |
working | +0.010 |
rier | +0.010 |
ivery | +0.009 |
ishy | +0.009 |
Neg logit contributions | |
peek | -0.003 |
dial | -0.003 |
turned | -0.003 |
opened | -0.003 |
ad | -0.003 |
Token | " |
Token ID | 366 |
Feature activation | -2.62e-03 |
Pos logit contributions | |
nob | +0.026 |
dial | +0.026 |
peek | +0.026 |
ad | +0.025 |
turned | +0.023 |
Neg logit contributions | |
rier | -0.027 |
working | -0.026 |
istered | -0.025 |
bored | -0.022 |
goals | -0.022 |
Token | This |
Token ID | 1212 |
Feature activation | -7.48e-02 |
Pos logit contributions | |
dial | +0.004 |
nob | +0.004 |
peek | +0.004 |
turned | +0.003 |
call | +0.003 |
Neg logit contributions | |
working | -0.004 |
rier | -0.004 |
istered | -0.004 |
attered | -0.004 |
ishy | -0.004 |
Token | is |
Token ID | 318 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
nob | +0.091 |
pat | +0.082 |
wit | +0.082 |
dial | +0.082 |
Looking | +0.080 |
Neg logit contributions | |
rier | -0.119 |
working | -0.117 |
bored | -0.109 |
ishy | -0.107 |
guys | -0.107 |
Token | a |
Token ID | 257 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
peek | +0.041 |
dial | +0.041 |
nob | +0.040 |
ad | +0.038 |
Looking | +0.037 |
Neg logit contributions | |
working | -0.047 |
rier | -0.046 |
istered | -0.045 |
bored | -0.042 |
goals | -0.041 |
Token | magic |
Token ID | 5536 |
Feature activation | -6.19e-02 |
Pos logit contributions | |
nob | +0.055 |
dial | +0.051 |
ad | +0.051 |
peek | +0.048 |
Looking | +0.048 |
Neg logit contributions | |
rier | -0.045 |
working | -0.044 |
istered | -0.044 |
bored | -0.040 |
ishy | -0.039 |
Token | f |
Token ID | 277 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
nob | +0.084 |
Looking | +0.079 |
ad | +0.076 |
dial | +0.076 |
peek | +0.074 |
Neg logit contributions | |
rier | -0.106 |
working | -0.103 |
istered | -0.087 |
ishy | -0.087 |
guys | -0.086 |
Token | auc |
Token ID | 14272 |
Feature activation | -6.74e-03 |
Pos logit contributions | |
dial | +0.024 |
peek | +0.024 |
nob | +0.023 |
turned | +0.022 |
Looking | +0.022 |
Neg logit contributions | |
rier | -0.023 |
working | -0.022 |
istered | -0.022 |
ishy | -0.020 |
attered | -0.019 |
Token | � |
Token ID | 129 |
Feature activation | +3.39e-03 |
Pos logit contributions | |
peek | +0.031 |
dial | +0.030 |
nob | +0.029 |
turned | +0.028 |
Looking | +0.027 |
Neg logit contributions | |
rier | -0.042 |
working | -0.039 |
istered | -0.038 |
ishy | -0.036 |
goals | -0.035 |
Token | � |
Token ID | 241 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
rier | +0.004 |
istered | +0.004 |
working | +0.004 |
nam | +0.004 |
ishy | +0.004 |
Neg logit contributions | |
dial | -0.006 |
peek | -0.006 |
nob | -0.006 |
ad | -0.006 |
turned | -0.006 |
Token | If |
Token ID | 1532 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
dial | +0.024 |
peek | +0.023 |
nob | +0.023 |
turned | +0.021 |
ad | +0.021 |
Neg logit contributions | |
working | -0.032 |
rier | -0.032 |
istered | -0.031 |
ishy | -0.028 |
goals | -0.028 |
Token | we |
Token ID | 356 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
dial | +0.028 |
peek | +0.027 |
nob | +0.027 |
turned | +0.025 |
Looking | +0.025 |
Neg logit contributions | |
rier | -0.021 |
working | -0.021 |
istered | -0.020 |
goals | -0.017 |
ishy | -0.017 |
Token | use |
Token ID | 779 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
peek | +0.031 |
dial | +0.030 |
nob | +0.029 |
turned | +0.028 |
ad | +0.027 |
Neg logit contributions | |
rier | -0.044 |
working | -0.042 |
istered | -0.042 |
ishy | -0.037 |
goals | -0.037 |
Token | this |
Token ID | 428 |
Feature activation | -7.46e-02 |
Pos logit contributions | |
peek | +0.066 |
dial | +0.065 |
nob | +0.065 |
ad | +0.060 |
Looking | +0.060 |
Neg logit contributions | |
rier | -0.084 |
working | -0.082 |
istered | -0.082 |
goals | -0.075 |
ishy | -0.074 |
Token | oil |
Token ID | 3056 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
peek | +0.107 |
ad | +0.106 |
nob | +0.106 |
dial | +0.105 |
Looking | +0.101 |
Neg logit contributions | |
working | -0.113 |
rier | -0.110 |
istered | -0.107 |
ishy | -0.102 |
goals | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
peek | +0.010 |
dial | +0.010 |
nob | +0.010 |
turned | +0.010 |
ad | +0.009 |
Neg logit contributions | |
rier | -0.023 |
istered | -0.023 |
working | -0.022 |
ishy | -0.021 |
ivery | -0.021 |
Token | we |
Token ID | 356 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
dial | +0.063 |
peek | +0.063 |
nob | +0.061 |
ad | +0.058 |
turned | +0.058 |
Neg logit contributions | |
rier | -0.061 |
working | -0.059 |
istered | -0.058 |
ishy | -0.052 |
goals | -0.051 |
Token | can |
Token ID | 460 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
peek | +0.023 |
dial | +0.023 |
nob | +0.021 |
turned | +0.021 |
ad | +0.020 |
Neg logit contributions | |
rier | -0.034 |
istered | -0.032 |
working | -0.032 |
ishy | -0.029 |
goals | -0.029 |
Token | make |
Token ID | 787 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
peek | +0.022 |
dial | +0.022 |
nob | +0.020 |
turned | +0.019 |
ad | +0.019 |
Neg logit contributions | |
rier | -0.045 |
working | -0.043 |
istered | -0.043 |
ishy | -0.040 |
goals | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | +4.53e-03 |
Pos logit contributions | |
dial | +0.028 |
ad | +0.028 |
Looking | +0.027 |
nob | +0.027 |
peek | +0.026 |
Neg logit contributions | |
working | -0.040 |
bored | -0.035 |
rier | -0.035 |
istered | -0.034 |
ishy | -0.032 |
Token | his |
Token ID | 465 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
rier | +0.007 |
working | +0.007 |
istered | +0.006 |
bored | +0.006 |
sweaty | +0.006 |
Neg logit contributions | |
nob | -0.006 |
dial | -0.006 |
Looking | -0.006 |
peek | -0.006 |
Turning | -0.006 |
Token | work |
Token ID | 670 |
Feature activation | -5.75e-04 |
Pos logit contributions | |
dial | +0.027 |
nob | +0.026 |
peek | +0.026 |
ad | +0.025 |
Looking | +0.025 |
Neg logit contributions | |
rier | -0.028 |
working | -0.028 |
istered | -0.028 |
goals | -0.025 |
ishy | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
dial | +0.001 |
peek | +0.001 |
nob | +0.001 |
Looking | +0.000 |
ad | +0.000 |
Neg logit contributions | |
rier | -0.001 |
working | -0.001 |
istered | -0.001 |
goals | -0.001 |
ishy | -0.001 |
Token | removes |
Token ID | 20694 |
Feature activation | -3.67e-03 |
Pos logit contributions | |
nob | +0.011 |
ad | +0.011 |
dial | +0.011 |
peek | +0.011 |
Looking | +0.010 |
Neg logit contributions | |
working | -0.022 |
rier | -0.022 |
istered | -0.022 |
bored | -0.021 |
goals | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | -7.43e-02 |
Pos logit contributions | |
dial | +0.004 |
peek | +0.004 |
nob | +0.004 |
turned | +0.004 |
ad | +0.004 |
Neg logit contributions | |
rier | -0.007 |
working | -0.007 |
istered | -0.006 |
ishy | -0.006 |
goals | -0.006 |
Token | tube |
Token ID | 12403 |
Feature activation | -8.58e-03 |
Pos logit contributions | |
ad | +0.100 |
nob | +0.099 |
dial | +0.098 |
Looking | +0.095 |
peek | +0.095 |
Neg logit contributions | |
rier | -0.119 |
working | -0.116 |
istered | -0.113 |
ishy | -0.108 |
bored | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +3.66e-03 |
Pos logit contributions | |
dial | +0.009 |
peek | +0.009 |
nob | +0.008 |
ad | +0.008 |
Looking | +0.008 |
Neg logit contributions | |
rier | -0.017 |
working | -0.017 |
istered | -0.017 |
ishy | -0.015 |
goals | -0.015 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
rier | +0.005 |
working | +0.005 |
istered | +0.005 |
goals | +0.005 |
ishy | +0.004 |
Neg logit contributions | |
dial | -0.006 |
peek | -0.006 |
nob | -0.006 |
ad | -0.005 |
turned | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
peek | +0.014 |
dial | +0.014 |
nob | +0.014 |
Looking | +0.013 |
ad | +0.013 |
Neg logit contributions | |
working | -0.032 |
rier | -0.031 |
istered | -0.031 |
goals | -0.030 |
ishy | -0.029 |
Token | Mia |
Token ID | 32189 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
dial | +0.015 |
nob | +0.015 |
peek | +0.014 |
Looking | +0.013 |
Turning | +0.013 |
Neg logit contributions | |
working | -0.025 |
rier | -0.025 |
istered | -0.024 |
goals | -0.024 |
kicker | -0.023 |
Token | " |
Token ID | 366 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
nob | +0.032 |
dial | +0.032 |
peek | +0.031 |
ad | +0.031 |
turned | +0.028 |
Neg logit contributions | |
rier | -0.036 |
working | -0.033 |
istered | -0.032 |
bored | -0.029 |
goals | -0.029 |
Token | Mom |
Token ID | 29252 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
rier | +0.022 |
working | +0.022 |
istered | +0.022 |
attered | +0.020 |
ishy | +0.020 |
Neg logit contributions | |
dial | -0.020 |
peek | -0.019 |
nob | -0.018 |
ad | -0.018 |
turned | -0.018 |
Token | my |
Token ID | 1820 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
dial | +0.031 |
nob | +0.031 |
peek | +0.030 |
Looking | +0.028 |
ad | +0.027 |
Neg logit contributions | |
working | -0.052 |
rier | -0.051 |
istered | -0.050 |
ishy | -0.046 |
kicker | -0.045 |
Token | , |
Token ID | 11 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
dial | +0.020 |
peek | +0.020 |
nob | +0.019 |
turned | +0.018 |
ad | +0.018 |
Neg logit contributions | |
rier | -0.035 |
working | -0.034 |
istered | -0.034 |
ishy | -0.031 |
goals | -0.031 |
Token | let |
Token ID | 1309 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
nob | +0.038 |
dial | +0.038 |
ad | +0.036 |
peek | +0.036 |
Looking | +0.035 |
Neg logit contributions | |
rier | -0.048 |
working | -0.045 |
istered | -0.045 |
ishy | -0.041 |
goals | -0.041 |
Token | me |
Token ID | 502 |
Feature activation | -7.40e-02 |
Pos logit contributions | |
peek | +0.013 |
dial | +0.013 |
nob | +0.012 |
turned | +0.011 |
call | +0.011 |
Neg logit contributions | |
rier | -0.025 |
working | -0.025 |
istered | -0.024 |
ishy | -0.023 |
goals | -0.022 |
Token | support |
Token ID | 1104 |
Feature activation | -5.99e-03 |
Pos logit contributions | |
nob | +0.084 |
dial | +0.083 |
ad | +0.080 |
Looking | +0.077 |
fore | +0.075 |
Neg logit contributions | |
working | -0.131 |
istered | -0.126 |
rier | -0.126 |
ishy | -0.120 |
kicker | -0.115 |
Token | you |
Token ID | 345 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
dial | +0.009 |
peek | +0.009 |
nob | +0.009 |
turned | +0.009 |
ad | +0.009 |
Neg logit contributions | |
rier | -0.009 |
working | -0.009 |
istered | -0.008 |
ishy | -0.008 |
goals | -0.007 |
Token | so |
Token ID | 523 |
Feature activation | +7.15e-03 |
Pos logit contributions | |
peek | +0.039 |
dial | +0.039 |
nob | +0.037 |
turned | +0.034 |
call | +0.033 |
Neg logit contributions | |
istered | -0.063 |
rier | -0.062 |
working | -0.061 |
ishy | -0.058 |
goals | -0.056 |
Token | you |
Token ID | 345 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
istered | +0.015 |
ishy | +0.014 |
rier | +0.014 |
working | +0.014 |
wy | +0.013 |
Neg logit contributions | |
peek | -0.008 |
be | -0.007 |
turned | -0.007 |
called | -0.006 |
opened | -0.006 |
Token | don |
Token ID | 836 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
peek | +0.033 |
dial | +0.032 |
turned | +0.030 |
nob | +0.030 |
call | +0.029 |
Neg logit contributions | |
rier | -0.051 |
istered | -0.050 |
working | -0.049 |
nam | -0.045 |
ishy | -0.044 |
Token | understand |
Token ID | 1833 |
Feature activation | +2.47e-03 |
Pos logit contributions | |
rier | +0.002 |
working | +0.002 |
istered | +0.002 |
nam | +0.002 |
goals | +0.002 |
Neg logit contributions | |
peek | -0.001 |
dial | -0.001 |
turned | -0.001 |
call | -0.001 |
nob | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
working | +0.006 |
rier | +0.006 |
istered | +0.005 |
ishy | +0.005 |
attered | +0.005 |
Neg logit contributions | |
dial | -0.002 |
peek | -0.002 |
turned | -0.002 |
nob | -0.002 |
ad | -0.002 |
Token | We |
Token ID | 775 |
Feature activation | +1.48e-04 |
Pos logit contributions | |
rier | +0.034 |
working | +0.033 |
istered | +0.033 |
ishy | +0.030 |
goals | +0.029 |
Neg logit contributions | |
dial | -0.029 |
peek | -0.029 |
nob | -0.027 |
turned | -0.026 |
ad | -0.026 |
Token | will |
Token ID | 481 |
Feature activation | -5.10e-03 |
Pos logit contributions | |
rier | +0.000 |
working | +0.000 |
istered | +0.000 |
ishy | +0.000 |
goals | +0.000 |
Neg logit contributions | |
peek | -0.000 |
dial | -0.000 |
turned | -0.000 |
nob | -0.000 |
call | -0.000 |
Token | share |
Token ID | 2648 |
Feature activation | +9.23e-04 |
Pos logit contributions | |
dial | +0.005 |
nob | +0.005 |
peek | +0.005 |
Looking | +0.005 |
ad | +0.005 |
Neg logit contributions | |
rier | -0.010 |
working | -0.010 |
istered | -0.010 |
goals | -0.009 |
ishy | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -7.37e-02 |
Pos logit contributions | |
working | +0.002 |
rier | +0.002 |
istered | +0.002 |
ishy | +0.002 |
ivery | +0.002 |
Neg logit contributions | |
dial | -0.001 |
peek | -0.001 |
nob | -0.001 |
turned | -0.001 |
ad | -0.001 |
Token | tube |
Token ID | 12403 |
Feature activation | -1.02e-03 |
Pos logit contributions | |
nob | +0.109 |
dial | +0.107 |
p | +0.103 |
peek | +0.102 |
pat | +0.102 |
Neg logit contributions | |
goals | -0.099 |
working | -0.096 |
rier | -0.096 |
bored | -0.094 |
istered | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
peek | +0.001 |
nob | +0.001 |
dial | +0.001 |
turned | +0.000 |
ad | +0.000 |
Neg logit contributions | |
istered | -0.002 |
working | -0.002 |
rier | -0.002 |
ivery | -0.002 |
ishy | -0.002 |
Token | We |
Token ID | 775 |
Feature activation | +1.17e-03 |
Pos logit contributions | |
rier | +0.042 |
working | +0.040 |
istered | +0.040 |
ishy | +0.036 |
goals | +0.036 |
Neg logit contributions | |
dial | -0.034 |
peek | -0.034 |
nob | -0.033 |
turned | -0.031 |
ad | -0.031 |
Token | will |
Token ID | 481 |
Feature activation | -4.37e-03 |
Pos logit contributions | |
rier | +0.002 |
istered | +0.002 |
working | +0.002 |
nam | +0.002 |
ishy | +0.002 |
Neg logit contributions | |
peek | -0.001 |
dial | -0.001 |
turned | -0.001 |
nob | -0.001 |
call | -0.001 |
Token | be |
Token ID | 307 |
Feature activation | +2.93e-03 |
Pos logit contributions | |
dial | +0.004 |
nob | +0.004 |
peek | +0.004 |
Looking | +0.004 |
ad | +0.004 |
Neg logit contributions | |
rier | -0.009 |
working | -0.009 |
istered | -0.008 |
goals | -0.008 |
ishy | -0.008 |
Token | ran |
Token ID | 4966 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
rier | +0.000 |
istered | +0.000 |
working | +0.000 |
ishy | +0.000 |
goals | +0.000 |
Neg logit contributions | |
peek | -0.000 |
dial | -0.000 |
nob | -0.000 |
turned | -0.000 |
ad | -0.000 |
Token | off |
Token ID | 572 |
Feature activation | -3.71e-04 |
Pos logit contributions | |
rier | +0.017 |
working | +0.016 |
istered | +0.016 |
ishy | +0.014 |
goals | +0.014 |
Neg logit contributions | |
dial | -0.015 |
peek | -0.015 |
nob | -0.015 |
turned | -0.014 |
ad | -0.014 |
Token | with |
Token ID | 351 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
dial | +0.000 |
peek | +0.000 |
nob | +0.000 |
turned | +0.000 |
ad | +0.000 |
Neg logit contributions | |
rier | -0.001 |
working | -0.001 |
istered | -0.001 |
ishy | -0.001 |
goals | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
working | +0.029 |
rier | +0.029 |
istered | +0.028 |
attered | +0.027 |
ishy | +0.026 |
Neg logit contributions | |
peek | -0.012 |
ad | -0.011 |
dial | -0.011 |
turned | -0.010 |
call | -0.010 |
Token | shiny |
Token ID | 22441 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
nob | +0.038 |
dial | +0.038 |
peek | +0.037 |
Looking | +0.035 |
ad | +0.035 |
Neg logit contributions | |
working | -0.037 |
rier | -0.037 |
istered | -0.036 |
goals | -0.032 |
ishy | -0.032 |
Token | new |
Token ID | 649 |
Feature activation | -7.35e-02 |
Pos logit contributions | |
nob | +0.079 |
dial | +0.077 |
peek | +0.075 |
Looking | +0.073 |
Turn | +0.068 |
Neg logit contributions | |
working | -0.099 |
rier | -0.094 |
ishy | -0.091 |
istered | -0.085 |
goals | -0.083 |
Token | pipe |
Token ID | 12656 |
Feature activation | +7.04e-03 |
Pos logit contributions | |
nob | +0.108 |
dial | +0.102 |
Looking | +0.094 |
peek | +0.094 |
grant | +0.094 |
Neg logit contributions | |
working | -0.116 |
ishy | -0.105 |
rier | -0.101 |
istered | -0.099 |
goals | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | -6.87e-03 |
Pos logit contributions | |
rier | +0.014 |
istered | +0.014 |
working | +0.014 |
ishy | +0.013 |
goals | +0.012 |
Neg logit contributions | |
peek | -0.007 |
dial | -0.007 |
nob | -0.007 |
turned | -0.007 |
ad | -0.006 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
dial | +0.006 |
peek | +0.006 |
nob | +0.005 |
turned | +0.005 |
ad | +0.005 |
Neg logit contributions | |
rier | -0.015 |
working | -0.015 |
istered | -0.015 |
goals | -0.013 |
ishy | -0.013 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
dial | +0.015 |
peek | +0.013 |
nob | +0.013 |
call | +0.012 |
ad | +0.011 |
Neg logit contributions | |
working | -0.021 |
rier | -0.020 |
goals | -0.020 |
istered | -0.020 |
bored | -0.019 |
Token | upon |
Token ID | 2402 |
Feature activation | +4.79e-03 |
Pos logit contributions | |
rier | +0.024 |
istered | +0.023 |
working | +0.022 |
ishy | +0.021 |
ivery | +0.021 |
Neg logit contributions | |
peek | -0.009 |
turned | -0.008 |
opened | -0.008 |
dial | -0.007 |
ad | -0.007 |
Token | 's |
Token ID | 338 |
Feature activation | -2.68e-03 |
Pos logit contributions | |
peek | +0.006 |
dial | +0.006 |
turned | +0.005 |
nob | +0.005 |
opened | +0.005 |
Neg logit contributions | |
rier | -0.012 |
istered | -0.011 |
working | -0.011 |
ishy | -0.010 |
goals | -0.010 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
nob | +0.005 |
dial | +0.004 |
peek | +0.004 |
Looking | +0.004 |
ad | +0.004 |
Neg logit contributions | |
rier | -0.004 |
working | -0.003 |
istered | -0.003 |
ishy | -0.003 |
goals | -0.003 |
Token | gave |
Token ID | 2921 |
Feature activation | +4.33e-03 |
Pos logit contributions | |
nob | +0.024 |
Looking | +0.023 |
dial | +0.022 |
ad | +0.022 |
peek | +0.021 |
Neg logit contributions | |
working | -0.039 |
istered | -0.035 |
bored | -0.034 |
rier | -0.034 |
sweaty | -0.033 |
Token | her |
Token ID | 607 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
rier | +0.007 |
working | +0.007 |
istered | +0.007 |
ishy | +0.006 |
goals | +0.006 |
Neg logit contributions | |
dial | -0.006 |
peek | -0.006 |
nob | -0.006 |
turned | -0.006 |
ad | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
dial | +0.056 |
nob | +0.056 |
peek | +0.054 |
ad | +0.052 |
Looking | +0.052 |
Neg logit contributions | |
working | -0.082 |
rier | -0.080 |
istered | -0.079 |
ishy | -0.075 |
goals | -0.074 |
Token | new |
Token ID | 649 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
nob | +0.065 |
ad | +0.062 |
ran | +0.058 |
p | +0.056 |
dial | +0.055 |
Neg logit contributions | |
working | -0.042 |
istered | -0.041 |
rier | -0.038 |
ishy | -0.038 |
handsome | -0.037 |
Token | toy |
Token ID | 13373 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
nob | +0.122 |
Looking | +0.110 |
ran | +0.109 |
dial | +0.109 |
Turning | +0.107 |
Neg logit contributions | |
working | -0.098 |
rier | -0.091 |
favorite | -0.087 |
kicker | -0.086 |
ishy | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | -6.81e-03 |
Pos logit contributions | |
dial | +0.025 |
nob | +0.025 |
peek | +0.024 |
Looking | +0.023 |
ad | +0.022 |
Neg logit contributions | |
rier | -0.034 |
working | -0.033 |
istered | -0.032 |
ishy | -0.029 |
goals | -0.029 |
Token | It |
Token ID | 632 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
dial | +0.010 |
nob | +0.010 |
ad | +0.010 |
peek | +0.010 |
p | +0.009 |
Neg logit contributions | |
rier | -0.010 |
istered | -0.010 |
working | -0.010 |
ishy | -0.009 |
goals | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
dial | +0.017 |
nob | +0.016 |
peek | +0.015 |
ad | +0.015 |
Looking | +0.014 |
Neg logit contributions | |
istered | -0.047 |
working | -0.047 |
rier | -0.046 |
ishy | -0.043 |
bored | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
dial | +0.027 |
nob | +0.027 |
peek | +0.026 |
Looking | +0.025 |
Turning | +0.024 |
Neg logit contributions | |
istered | -0.028 |
rier | -0.028 |
working | -0.028 |
bored | -0.025 |
goals | -0.025 |
Token | saf |
Token ID | 1932 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
dial | +0.020 |
Looking | +0.020 |
nob | +0.020 |
Turning | +0.020 |
Turn | +0.019 |
Neg logit contributions | |
working | -0.018 |
guys | -0.016 |
goals | -0.015 |
rier | -0.014 |
favorite | -0.014 |
Token | ari |
Token ID | 2743 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
rier | +0.042 |
working | +0.042 |
istered | +0.040 |
goals | +0.037 |
ishy | +0.036 |
Neg logit contributions | |
dial | -0.032 |
peek | -0.031 |
nob | -0.030 |
turned | -0.029 |
Looking | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
istered | +0.024 |
rier | +0.022 |
working | +0.022 |
ishy | +0.021 |
ivery | +0.021 |
Neg logit contributions | |
peek | -0.013 |
dial | -0.012 |
nob | -0.012 |
called | -0.011 |
call | -0.011 |
Token | Please |
Token ID | 4222 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
rier | +0.025 |
working | +0.024 |
istered | +0.024 |
goals | +0.021 |
ishy | +0.021 |
Neg logit contributions | |
dial | -0.024 |
peek | -0.024 |
nob | -0.024 |
ad | -0.023 |
turned | -0.023 |
Token | let |
Token ID | 1309 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
dial | +0.014 |
peek | +0.013 |
nob | +0.013 |
Looking | +0.012 |
ad | +0.012 |
Neg logit contributions | |
rier | -0.026 |
working | -0.026 |
istered | -0.026 |
ishy | -0.024 |
goals | -0.023 |
Token | me |
Token ID | 502 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
istered | +0.029 |
rier | +0.028 |
ishy | +0.027 |
working | +0.027 |
nam | +0.026 |
Neg logit contributions | |
peek | -0.009 |
dial | -0.008 |
nob | -0.007 |
call | -0.007 |
turned | -0.007 |
Token | go |
Token ID | 467 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
nob | +0.089 |
dial | +0.087 |
Looking | +0.084 |
ad | +0.082 |
Turning | +0.078 |
Neg logit contributions | |
working | -0.127 |
rier | -0.121 |
istered | -0.116 |
ishy | -0.113 |
bored | -0.111 |
Token | , |
Token ID | 11 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
dial | +0.041 |
Looking | +0.040 |
nob | +0.040 |
peek | +0.039 |
ad | +0.038 |
Neg logit contributions | |
working | -0.040 |
rier | -0.038 |
istered | -0.034 |
bored | -0.033 |
ishy | -0.033 |
Token | mum |
Token ID | 25682 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
peek | +0.016 |
dial | +0.016 |
nob | +0.015 |
turned | +0.015 |
ad | +0.014 |
Neg logit contributions | |
rier | -0.022 |
working | -0.022 |
istered | -0.022 |
ishy | -0.020 |
goals | -0.019 |
Token | ." |
Token ID | 526 |
Feature activation | +4.65e-03 |
Pos logit contributions | |
dial | +0.044 |
peek | +0.044 |
nob | +0.042 |
turned | +0.040 |
ad | +0.040 |
Neg logit contributions | |
rier | -0.048 |
working | -0.047 |
istered | -0.046 |
ishy | -0.042 |
goals | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -5.37e-03 |
Pos logit contributions | |
istered | +0.009 |
working | +0.009 |
rier | +0.009 |
goals | +0.008 |
bored | +0.008 |
Neg logit contributions | |
nob | -0.004 |
peek | -0.004 |
dial | -0.004 |
ad | -0.004 |
p | -0.004 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
dial | +0.034 |
nob | +0.033 |
peek | +0.033 |
ad | +0.032 |
turned | +0.031 |
Neg logit contributions | |
rier | -0.035 |
istered | -0.034 |
working | -0.034 |
ishy | -0.030 |
goals | -0.029 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
nob | +0.060 |
dial | +0.060 |
peek | +0.058 |
ad | +0.055 |
Looking | +0.054 |
Neg logit contributions | |
working | -0.065 |
rier | -0.062 |
istered | -0.057 |
ishy | -0.055 |
kicker | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -3.41e-03 |
Pos logit contributions | |
dial | +0.029 |
peek | +0.029 |
nob | +0.028 |
ad | +0.026 |
turned | +0.026 |
Neg logit contributions | |
rier | -0.038 |
working | -0.037 |
istered | -0.037 |
ishy | -0.034 |
goals | -0.033 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.98e-03 |
Pos logit contributions | |
dial | +0.005 |
peek | +0.004 |
nob | +0.004 |
turned | +0.004 |
Looking | +0.004 |
Neg logit contributions | |
rier | -0.006 |
working | -0.006 |
istered | -0.006 |
ishy | -0.005 |
goals | -0.005 |
Token | felt |
Token ID | 2936 |
Feature activation | -4.60e-03 |
Pos logit contributions | |
nob | +0.006 |
dial | +0.006 |
peek | +0.006 |
ad | +0.005 |
Looking | +0.005 |
Neg logit contributions | |
working | -0.011 |
istered | -0.011 |
rier | -0.011 |
goals | -0.010 |
ishy | -0.010 |
Token | en |
Token ID | 551 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
peek | +0.007 |
turned | +0.006 |
dial | +0.006 |
Looking | +0.006 |
opened | +0.006 |
Neg logit contributions | |
working | -0.008 |
rier | -0.007 |
istered | -0.007 |
ishy | -0.007 |
kicker | -0.007 |
Token | vious |
Token ID | 1442 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
rier | +0.038 |
istered | +0.036 |
working | +0.036 |
goals | +0.028 |
ishy | +0.028 |
Neg logit contributions | |
peek | -0.083 |
dial | -0.083 |
nob | -0.080 |
turned | -0.079 |
ad | -0.078 |
Token | and |
Token ID | 290 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
rier | +0.029 |
istered | +0.029 |
working | +0.028 |
nam | +0.026 |
ishy | +0.026 |
Neg logit contributions | |
peek | -0.010 |
dial | -0.010 |
turned | -0.010 |
ad | -0.009 |
nob | -0.009 |
Token | wanted |
Token ID | 2227 |
Feature activation | +7.71e-03 |
Pos logit contributions | |
rier | +0.048 |
istered | +0.045 |
working | +0.043 |
avored | +0.042 |
umm | +0.040 |
Neg logit contributions | |
turned | -0.024 |
peek | -0.024 |
opened | -0.023 |
dial | -0.022 |
called | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
working | +0.015 |
rier | +0.015 |
istered | +0.014 |
goals | +0.013 |
bored | +0.013 |
Neg logit contributions | |
peek | -0.008 |
dial | -0.008 |
nob | -0.008 |
Looking | -0.007 |
ad | -0.007 |
Token | toy |
Token ID | 13373 |
Feature activation | +5.89e-03 |
Pos logit contributions | |
dial | +0.025 |
peek | +0.025 |
nob | +0.024 |
ad | +0.023 |
turned | +0.023 |
Neg logit contributions | |
rier | -0.026 |
working | -0.025 |
istered | -0.025 |
ishy | -0.022 |
goals | -0.022 |
Token | replied |
Token ID | 8712 |
Feature activation | +6.83e-03 |
Pos logit contributions | |
nob | +0.007 |
dial | +0.006 |
peek | +0.006 |
Looking | +0.006 |
AAAA | +0.005 |
Neg logit contributions | |
working | -0.011 |
rier | -0.010 |
istered | -0.010 |
bored | -0.010 |
ishy | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
istered | +0.016 |
working | +0.016 |
rier | +0.015 |
ishy | +0.015 |
goals | +0.014 |
Neg logit contributions | |
peek | -0.005 |
dial | -0.005 |
turned | -0.005 |
nob | -0.004 |
opened | -0.004 |
Token | " |
Token ID | 366 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
dial | +0.024 |
peek | +0.024 |
nob | +0.024 |
ad | +0.022 |
turned | +0.022 |
Neg logit contributions | |
rier | -0.031 |
working | -0.030 |
istered | -0.029 |
ishy | -0.026 |
goals | -0.026 |
Token | It |
Token ID | 1026 |
Feature activation | -6.31e-03 |
Pos logit contributions | |
rier | +0.031 |
working | +0.031 |
istered | +0.029 |
ishy | +0.027 |
goals | +0.027 |
Neg logit contributions | |
dial | -0.025 |
peek | -0.025 |
nob | -0.024 |
turned | -0.023 |
Looking | -0.022 |
Token | 's |
Token ID | 338 |
Feature activation | -1.38e-03 |
Pos logit contributions | |
dial | +0.006 |
peek | +0.006 |
nob | +0.006 |
turned | +0.005 |
ad | +0.005 |
Neg logit contributions | |
rier | -0.013 |
working | -0.013 |
istered | -0.013 |
ishy | -0.012 |
goals | -0.011 |
Token | c |
Token ID | 269 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
dial | +0.002 |
peek | +0.002 |
nob | +0.002 |
Looking | +0.002 |
ad | +0.002 |
Neg logit contributions | |
rier | -0.002 |
working | -0.002 |
istered | -0.002 |
ishy | -0.002 |
goals | -0.002 |
Token | heddar |
Token ID | 44937 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
nam | +0.050 |
working | +0.049 |
rier | +0.049 |
istered | +0.047 |
kicker | +0.044 |
Neg logit contributions | |
nob | -0.069 |
peek | -0.069 |
ad | -0.068 |
dial | -0.068 |
turned | -0.067 |
Token | cheese |
Token ID | 9891 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
rier | +0.033 |
istered | +0.033 |
working | +0.032 |
ivery | +0.030 |
nam | +0.030 |
Neg logit contributions | |
peek | -0.030 |
dial | -0.030 |
nob | -0.028 |
turned | -0.028 |
ad | -0.027 |
Token | ." |
Token ID | 526 |
Feature activation | -3.83e-03 |
Pos logit contributions | |
istered | +0.066 |
rier | +0.066 |
working | +0.063 |
ishy | +0.061 |
nam | +0.060 |
Neg logit contributions | |
peek | -0.029 |
dial | -0.028 |
nob | -0.028 |
turned | -0.026 |
opened | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -2.17e-03 |
Pos logit contributions | |
dial | +0.005 |
peek | +0.005 |
nob | +0.005 |
turned | +0.004 |
ad | +0.004 |
Neg logit contributions | |
rier | -0.007 |
working | -0.007 |
istered | -0.007 |
ishy | -0.006 |
goals | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +7.90e-03 |
Pos logit contributions | |
dial | +0.002 |
nob | +0.002 |
peek | +0.002 |
turned | +0.001 |
ad | +0.001 |
Neg logit contributions | |
istered | -0.005 |
rier | -0.005 |
working | -0.005 |
goals | -0.004 |
ishy | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
peek | +0.003 |
dial | +0.002 |
nob | +0.002 |
turned | +0.002 |
ad | +0.002 |
Neg logit contributions | |
rier | -0.008 |
istered | -0.008 |
working | -0.007 |
ishy | -0.007 |
goals | -0.007 |
Token | big |
Token ID | 1263 |
Feature activation | +3.05e-03 |
Pos logit contributions | |
rier | +0.040 |
istered | +0.039 |
working | +0.038 |
goals | +0.036 |
attered | +0.036 |
Neg logit contributions | |
peek | -0.023 |
dial | -0.022 |
turned | -0.020 |
call | -0.019 |
opened | -0.018 |
Token | chamber |
Token ID | 11847 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
rier | +0.006 |
istered | +0.006 |
working | +0.005 |
ishy | +0.005 |
ivery | +0.005 |
Neg logit contributions | |
peek | -0.004 |
dial | -0.004 |
turned | -0.003 |
ad | -0.003 |
nob | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +4.98e-03 |
Pos logit contributions | |
working | +0.017 |
rier | +0.016 |
istered | +0.015 |
bored | +0.014 |
kicker | +0.014 |
Neg logit contributions | |
dial | -0.014 |
nob | -0.013 |
Looking | -0.012 |
peek | -0.012 |
ad | -0.012 |
Token | Inside |
Token ID | 14384 |
Feature activation | -5.38e-03 |
Pos logit contributions | |
working | +0.008 |
istered | +0.007 |
rier | +0.007 |
goals | +0.006 |
kicker | +0.006 |
Neg logit contributions | |
dial | -0.007 |
fore | -0.006 |
peek | -0.006 |
ran | -0.006 |
nob | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
peek | +0.005 |
dial | +0.005 |
nob | +0.004 |
turned | +0.004 |
ad | +0.004 |
Neg logit contributions | |
rier | -0.012 |
istered | -0.012 |
working | -0.011 |
ishy | -0.011 |
goals | -0.010 |
Token | chamber |
Token ID | 11847 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
rier | +0.086 |
istered | +0.085 |
working | +0.084 |
ishy | +0.079 |
kicker | +0.077 |
Neg logit contributions | |
peek | -0.049 |
nob | -0.044 |
dial | -0.043 |
opened | -0.042 |
turned | -0.040 |
Token | there |
Token ID | 612 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
rier | +0.029 |
working | +0.029 |
istered | +0.028 |
goals | +0.026 |
ishy | +0.026 |
Neg logit contributions | |
dial | -0.015 |
peek | -0.014 |
nob | -0.014 |
ad | -0.013 |
Looking | -0.012 |
Token | were |
Token ID | 547 |
Feature activation | +8.42e-03 |
Pos logit contributions | |
rier | +0.026 |
working | +0.026 |
istered | +0.025 |
ishy | +0.024 |
goals | +0.023 |
Neg logit contributions | |
dial | -0.010 |
peek | -0.010 |
nob | -0.010 |
turned | -0.009 |
ad | -0.009 |
Token | many |
Token ID | 867 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
rier | +0.011 |
working | +0.010 |
istered | +0.010 |
ishy | +0.009 |
ivery | +0.008 |
Neg logit contributions | |
peek | -0.016 |
dial | -0.015 |
turned | -0.015 |
nob | -0.015 |
opened | -0.015 |
Token | things |
Token ID | 1243 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
rier | +0.038 |
istered | +0.037 |
working | +0.037 |
ivery | +0.034 |
ishy | +0.033 |
Neg logit contributions | |
peek | -0.030 |
dial | -0.030 |
opened | -0.028 |
turned | -0.027 |
called | -0.027 |
Token | invite |
Token ID | 14037 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
dial | +0.001 |
nob | +0.001 |
Looking | +0.001 |
peek | +0.001 |
Turning | +0.001 |
Neg logit contributions | |
working | -0.001 |
rier | -0.001 |
istered | -0.001 |
goals | -0.001 |
kicker | -0.001 |
Token | all |
Token ID | 477 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
rier | +0.025 |
working | +0.024 |
istered | +0.023 |
goals | +0.022 |
ishy | +0.021 |
Neg logit contributions | |
dial | -0.022 |
peek | -0.022 |
nob | -0.022 |
Looking | -0.020 |
turned | -0.020 |
Token | his |
Token ID | 465 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
dial | +0.051 |
nob | +0.049 |
peek | +0.049 |
Looking | +0.046 |
Turning | +0.046 |
Neg logit contributions | |
working | -0.057 |
rier | -0.055 |
istered | -0.051 |
goals | -0.051 |
ishy | -0.049 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
rier | +0.019 |
istered | +0.018 |
working | +0.018 |
ishy | +0.016 |
goals | +0.015 |
Neg logit contributions | |
dial | -0.026 |
peek | -0.026 |
nob | -0.025 |
turned | -0.024 |
ad | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
dial | +0.011 |
peek | +0.011 |
nob | +0.010 |
turned | +0.010 |
ad | +0.009 |
Neg logit contributions | |
rier | -0.022 |
istered | -0.021 |
working | -0.021 |
ishy | -0.019 |
goals | -0.019 |
Token | his |
Token ID | 465 |
Feature activation | +5.17e-02 |
Pos logit contributions | |
working | +0.030 |
rier | +0.030 |
istered | +0.028 |
goals | +0.027 |
kicker | +0.026 |
Neg logit contributions | |
dial | -0.018 |
nob | -0.018 |
peek | -0.018 |
Looking | -0.017 |
turned | -0.016 |
Token | palace |
Token ID | 20562 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
rier | +0.089 |
istered | +0.087 |
working | +0.083 |
ishy | +0.080 |
ivery | +0.077 |
Neg logit contributions | |
peek | -0.076 |
dial | -0.073 |
turned | -0.066 |
nob | -0.066 |
call | -0.066 |
Token | for |
Token ID | 329 |
Feature activation | +4.38e-03 |
Pos logit contributions | |
rier | +0.033 |
istered | +0.032 |
working | +0.032 |
ishy | +0.030 |
goals | +0.029 |
Neg logit contributions | |
peek | -0.017 |
dial | -0.017 |
nob | -0.016 |
turned | -0.015 |
ad | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
rier | +0.008 |
working | +0.008 |
istered | +0.008 |
goals | +0.007 |
ishy | +0.007 |
Neg logit contributions | |
dial | -0.005 |
peek | -0.005 |
nob | -0.005 |
turned | -0.005 |
Looking | -0.004 |
Token | big |
Token ID | 1263 |
Feature activation | +1.20e-03 |
Pos logit contributions | |
working | +0.026 |
rier | +0.025 |
istered | +0.024 |
bored | +0.021 |
ishy | +0.021 |
Neg logit contributions | |
nob | -0.031 |
dial | -0.031 |
ad | -0.030 |
peek | -0.029 |
Looking | -0.029 |
Token | party |
Token ID | 2151 |
Feature activation | +6.34e-03 |
Pos logit contributions | |
working | +0.002 |
rier | +0.001 |
istered | +0.001 |
ishy | +0.001 |
goals | +0.001 |
Neg logit contributions | |
dial | -0.002 |
nob | -0.002 |
peek | -0.002 |
Looking | -0.002 |
turned | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
rier | +0.005 |
istered | +0.005 |
ishy | +0.005 |
working | +0.005 |
attered | +0.005 |
Neg logit contributions | |
dial | -0.001 |
peek | -0.001 |
turned | -0.001 |
called | -0.001 |
call | -0.001 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.05e-03 |
Pos logit contributions | |
rier | +0.035 |
working | +0.034 |
istered | +0.034 |
ishy | +0.031 |
goals | +0.030 |
Neg logit contributions | |
dial | -0.018 |
peek | -0.018 |
nob | -0.017 |
turned | -0.016 |
Looking | -0.016 |
Token | L |
Token ID | 43 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
dial | +0.002 |
nob | +0.002 |
peek | +0.002 |
turned | +0.002 |
call | +0.002 |
Neg logit contributions | |
working | -0.004 |
rier | -0.004 |
istered | -0.004 |
goals | -0.004 |
kicker | -0.003 |
Token | ila |
Token ID | 10102 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
rier | +0.017 |
working | +0.017 |
istered | +0.017 |
ishy | +0.015 |
goals | +0.015 |
Neg logit contributions | |
dial | -0.014 |
peek | -0.014 |
nob | -0.014 |
turned | -0.013 |
ad | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
dial | +0.019 |
peek | +0.019 |
nob | +0.018 |
turned | +0.016 |
ad | +0.016 |
Neg logit contributions | |
rier | -0.062 |
working | -0.062 |
istered | -0.061 |
ishy | -0.057 |
goals | -0.056 |
Token | a |
Token ID | 257 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
rier | +0.051 |
working | +0.050 |
istered | +0.050 |
ishy | +0.044 |
goals | +0.043 |
Neg logit contributions | |
dial | -0.056 |
peek | -0.056 |
nob | -0.054 |
turned | -0.052 |
ad | -0.051 |
Token | little |
Token ID | 1310 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
istered | +0.084 |
rier | +0.083 |
ivery | +0.075 |
attered | +0.075 |
goals | +0.074 |
Neg logit contributions | |
peek | -0.065 |
dial | -0.058 |
turned | -0.056 |
call | -0.054 |
called | -0.053 |
Token | girl |
Token ID | 2576 |
Feature activation | -7.87e-03 |
Pos logit contributions | |
istered | +0.076 |
rier | +0.069 |
responsibly | +0.068 |
ivery | +0.068 |
ops | +0.067 |
Neg logit contributions | |
peek | -0.050 |
dial | -0.045 |
called | -0.044 |
pe | -0.044 |
call | -0.042 |
Token | who |
Token ID | 508 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
dial | +0.009 |
peek | +0.008 |
nob | +0.008 |
turned | +0.008 |
ad | +0.008 |
Neg logit contributions | |
rier | -0.015 |
working | -0.015 |
istered | -0.015 |
ishy | -0.013 |
goals | -0.013 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
rier | +0.031 |
istered | +0.027 |
working | +0.026 |
goals | +0.024 |
ishy | +0.024 |
Neg logit contributions | |
peek | -0.026 |
dial | -0.025 |
turned | -0.024 |
nob | -0.024 |
opened | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +4.65e-03 |
Pos logit contributions | |
rier | +0.043 |
istered | +0.042 |
working | +0.041 |
ishy | +0.038 |
ivery | +0.037 |
Neg logit contributions | |
dial | -0.018 |
peek | -0.017 |
nob | -0.016 |
turned | -0.015 |
ad | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
rier | +0.050 |
working | +0.048 |
istered | +0.047 |
ishy | +0.044 |
goals | +0.044 |
Neg logit contributions | |
peek | -0.023 |
dial | -0.022 |
turned | -0.021 |
nob | -0.020 |
opened | -0.020 |
Token | bored |
Token ID | 17533 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
rier | +0.027 |
working | +0.027 |
istered | +0.026 |
ishy | +0.025 |
goals | +0.024 |
Neg logit contributions | |
peek | -0.021 |
dial | -0.021 |
turned | -0.020 |
nob | -0.020 |
opened | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
rier | +0.036 |
working | +0.035 |
istered | +0.035 |
ishy | +0.033 |
ivery | +0.032 |
Neg logit contributions | |
peek | -0.015 |
dial | -0.015 |
nob | -0.014 |
turned | -0.014 |
opened | -0.013 |
Token | restless |
Token ID | 42537 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
rier | +0.055 |
umm | +0.052 |
istered | +0.051 |
avored | +0.049 |
kicker | +0.049 |
Neg logit contributions | |
turned | -0.055 |
peek | -0.055 |
opened | -0.055 |
called | -0.051 |
dial | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
istered | +0.055 |
rier | +0.054 |
working | +0.054 |
ishy | +0.051 |
ivery | +0.051 |
Neg logit contributions | |
peek | -0.018 |
dial | -0.017 |
nob | -0.017 |
turned | -0.016 |
opened | -0.015 |
Token | She |
Token ID | 1375 |
Feature activation | +3.62e-03 |
Pos logit contributions | |
rier | +0.046 |
working | +0.045 |
istered | +0.044 |
ishy | +0.040 |
goals | +0.039 |
Neg logit contributions | |
dial | -0.037 |
peek | -0.037 |
nob | -0.036 |
turned | -0.034 |
ad | -0.033 |
Token | f |
Token ID | 277 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
rier | +0.006 |
working | +0.006 |
istered | +0.006 |
ishy | +0.006 |
goals | +0.006 |
Neg logit contributions | |
dial | -0.004 |
peek | -0.004 |
nob | -0.004 |
ad | -0.004 |
Looking | -0.004 |
Token | idget |
Token ID | 17484 |
Feature activation | -3.38e-03 |
Pos logit contributions | |
rier | +0.034 |
istered | +0.033 |
working | +0.033 |
ishy | +0.030 |
goals | +0.028 |
Neg logit contributions | |
dial | -0.036 |
peek | -0.036 |
nob | -0.034 |
turned | -0.033 |
ad | -0.033 |
Token | ed |
Token ID | 276 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
peek | +0.005 |
dial | +0.005 |
turned | +0.005 |
opened | +0.005 |
nob | +0.004 |
Neg logit contributions | |
rier | -0.006 |
istered | -0.005 |
working | -0.005 |
goals | -0.005 |
ivery | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +5.84e-03 |
Pos logit contributions | |
working | +0.032 |
rier | +0.032 |
istered | +0.032 |
ishy | +0.030 |
kicker | +0.029 |
Neg logit contributions | |
peek | -0.011 |
dial | -0.011 |
turned | -0.010 |
nob | -0.010 |
opened | -0.009 |
Token | wh |
Token ID | 348 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
rier | +0.013 |
working | +0.012 |
istered | +0.012 |
ishy | +0.011 |
goals | +0.011 |
Neg logit contributions | |
dial | -0.006 |
peek | -0.006 |
turned | -0.005 |
nob | -0.005 |
opened | -0.005 |
Token | together |
Token ID | 1978 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
rier | +0.002 |
working | +0.002 |
istered | +0.002 |
ishy | +0.002 |
goals | +0.002 |
Neg logit contributions | |
peek | -0.002 |
dial | -0.002 |
nob | -0.002 |
turned | -0.002 |
ad | -0.002 |
Token | in |
Token ID | 287 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
rier | +0.007 |
istered | +0.007 |
working | +0.007 |
ishy | +0.007 |
ivery | +0.006 |
Neg logit contributions | |
peek | -0.003 |
dial | -0.003 |
nob | -0.002 |
turned | -0.002 |
opened | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +8.22e-03 |
Pos logit contributions | |
peek | +0.019 |
dial | +0.019 |
nob | +0.018 |
turned | +0.017 |
ad | +0.017 |
Neg logit contributions | |
rier | -0.039 |
istered | -0.037 |
working | -0.037 |
ishy | -0.034 |
goals | -0.033 |
Token | park |
Token ID | 3952 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
rier | +0.018 |
istered | +0.017 |
working | +0.016 |
nam | +0.015 |
der | +0.014 |
Neg logit contributions | |
peek | -0.010 |
turned | -0.008 |
nob | -0.008 |
dial | -0.008 |
ad | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -3.86e-03 |
Pos logit contributions | |
rier | +0.026 |
istered | +0.026 |
ishy | +0.024 |
working | +0.024 |
ivery | +0.023 |
Neg logit contributions | |
peek | -0.009 |
turned | -0.007 |
dial | -0.007 |
nob | -0.007 |
called | -0.007 |
Token | running |
Token ID | 2491 |
Feature activation | +3.57e-05 |
Pos logit contributions | |
peek | +0.004 |
dial | +0.004 |
nob | +0.004 |
turned | +0.004 |
opened | +0.004 |
Neg logit contributions | |
rier | -0.008 |
istered | -0.007 |
working | -0.007 |
ishy | -0.007 |
goals | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +3.82e-03 |
Pos logit contributions | |
rier | +0.000 |
istered | +0.000 |
working | +0.000 |
ivery | +0.000 |
ishy | +0.000 |
Neg logit contributions | |
peek | -0.000 |
dial | -0.000 |
nob | -0.000 |
turned | -0.000 |
opened | -0.000 |
Token | laughing |
Token ID | 14376 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
rier | +0.008 |
istered | +0.007 |
working | +0.007 |
ishy | +0.007 |
goals | +0.007 |
Neg logit contributions | |
peek | -0.005 |
dial | -0.004 |
nob | -0.004 |
turned | -0.004 |
opened | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
istered | +0.076 |
rier | +0.074 |
working | +0.073 |
ivery | +0.072 |
ishy | +0.070 |
Neg logit contributions | |
and | -0.014 |
peek | -0.014 |
turned | -0.012 |
dial | -0.011 |
called | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
dial | +0.012 |
peek | +0.012 |
nob | +0.011 |
ad | +0.010 |
turned | +0.010 |
Neg logit contributions | |
rier | -0.034 |
working | -0.034 |
istered | -0.034 |
ishy | -0.031 |
goals | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
dial | +0.013 |
nob | +0.013 |
peek | +0.012 |
ad | +0.011 |
turned | +0.011 |
Neg logit contributions | |
working | -0.032 |
istered | -0.032 |
rier | -0.032 |
goals | -0.030 |
bored | -0.029 |
Token | space |
Token ID | 2272 |
Feature activation | +1.34e-03 |
Pos logit contributions | |
dial | +0.005 |
peek | +0.005 |
nob | +0.005 |
Looking | +0.005 |
ad | +0.005 |
Neg logit contributions | |
rier | -0.009 |
working | -0.009 |
istered | -0.009 |
goals | -0.008 |
ishy | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
istered | +0.003 |
rier | +0.003 |
working | +0.003 |
ishy | +0.003 |
ivery | +0.003 |
Neg logit contributions | |
peek | -0.001 |
dial | -0.001 |
nob | -0.001 |
turned | -0.001 |
ad | -0.001 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.74e-03 |
Pos logit contributions | |
dial | +0.009 |
peek | +0.009 |
nob | +0.008 |
turned | +0.007 |
ad | +0.007 |
Neg logit contributions | |
rier | -0.024 |
working | -0.024 |
istered | -0.024 |
goals | -0.022 |
ishy | -0.022 |
Token | Once |
Token ID | 7454 |
Feature activation | +9.13e-03 |
Pos logit contributions | |
dial | +0.012 |
peek | +0.010 |
nob | +0.010 |
call | +0.010 |
knock | +0.010 |
Neg logit contributions | |
working | -0.015 |
goals | -0.015 |
rier | -0.015 |
istered | -0.014 |
bored | -0.014 |
Token | there |
Token ID | 612 |
Feature activation | +2.35e-03 |
Pos logit contributions | |
rier | +0.020 |
istered | +0.020 |
working | +0.019 |
ishy | +0.018 |
ivery | +0.018 |
Neg logit contributions | |
peek | -0.008 |
turned | -0.008 |
opened | -0.007 |
dial | -0.007 |
ad | -0.007 |
Token | were |
Token ID | 547 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
rier | +0.006 |
working | +0.005 |
istered | +0.005 |
ishy | +0.005 |
goals | +0.005 |
Neg logit contributions | |
dial | -0.001 |
peek | -0.001 |
nob | -0.001 |
ad | -0.001 |
turned | -0.001 |
Token | two |
Token ID | 734 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
rier | +0.025 |
istered | +0.023 |
working | +0.023 |
ivery | +0.020 |
ishy | +0.020 |
Neg logit contributions | |
peek | -0.013 |
ad | -0.012 |
turned | -0.012 |
opened | -0.012 |
dial | -0.012 |
Token | kids |
Token ID | 3988 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
rier | +0.024 |
working | +0.024 |
istered | +0.022 |
ivery | +0.020 |
ishy | +0.020 |
Neg logit contributions | |
peek | -0.024 |
dial | -0.023 |
turned | -0.023 |
ad | -0.022 |
call | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
dial | +0.028 |
nob | +0.027 |
peek | +0.026 |
Looking | +0.025 |
turned | +0.024 |
Neg logit contributions | |
working | -0.039 |
istered | -0.037 |
rier | -0.037 |
bored | -0.034 |
ishy | -0.034 |
Token | Sam |
Token ID | 3409 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
nob | +0.047 |
dial | +0.045 |
peek | +0.042 |
ad | +0.041 |
Looking | +0.041 |
Neg logit contributions | |
working | -0.053 |
istered | -0.050 |
rier | -0.048 |
nam | -0.048 |
bored | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +1.55e-03 |
Pos logit contributions | |
dial | +0.036 |
nob | +0.036 |
Looking | +0.034 |
peek | +0.033 |
peaked | +0.031 |
Neg logit contributions | |
working | -0.065 |
rier | -0.060 |
istered | -0.059 |
ishy | -0.057 |
goals | -0.056 |
Token | some |
Token ID | 617 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
istered | +0.006 |
working | +0.006 |
rier | +0.006 |
ivery | +0.006 |
ishy | +0.006 |
Neg logit contributions | |
peek | -0.005 |
dial | -0.004 |
nob | -0.004 |
turned | -0.004 |
ad | -0.004 |
Token | pe |
Token ID | 613 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
nob | +0.022 |
dial | +0.021 |
Looking | +0.021 |
peek | +0.020 |
ad | +0.019 |
Neg logit contributions | |
rier | -0.036 |
working | -0.033 |
istered | -0.032 |
goals | -0.032 |
ishy | -0.032 |
Token | aches |
Token ID | 3694 |
Feature activation | -7.17e-03 |
Pos logit contributions | |
dial | +0.025 |
peek | +0.025 |
Looking | +0.025 |
nob | +0.024 |
call | +0.023 |
Neg logit contributions | |
rier | -0.025 |
working | -0.022 |
istered | -0.022 |
ishy | -0.021 |
goals | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
peek | +0.005 |
dial | +0.005 |
turned | +0.005 |
nob | +0.005 |
opened | +0.005 |
Neg logit contributions | |
istered | -0.016 |
working | -0.016 |
rier | -0.015 |
ivery | -0.015 |
kicker | -0.015 |
Token | They |
Token ID | 1119 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
dial | +0.017 |
nob | +0.017 |
ad | +0.016 |
peek | +0.016 |
god | +0.015 |
Neg logit contributions | |
rier | -0.014 |
working | -0.014 |
istered | -0.013 |
goals | -0.013 |
ishy | -0.013 |
Token | were |
Token ID | 547 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
rier | +0.034 |
istered | +0.031 |
working | +0.031 |
goals | +0.029 |
ivery | +0.029 |
Neg logit contributions | |
peek | -0.016 |
turned | -0.015 |
dial | -0.015 |
opened | -0.015 |
nob | -0.013 |
Token | juicy |
Token ID | 35075 |
Feature activation | -5.26e-03 |
Pos logit contributions | |
working | +0.038 |
rier | +0.037 |
istered | +0.036 |
ishy | +0.032 |
ivery | +0.032 |
Neg logit contributions | |
peek | -0.034 |
dial | -0.033 |
turned | -0.033 |
opened | -0.032 |
nob | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
peek | +0.003 |
dial | +0.003 |
nob | +0.003 |
turned | +0.003 |
ad | +0.002 |
Neg logit contributions | |
working | -0.013 |
rier | -0.013 |
istered | -0.013 |
ishy | -0.012 |
ivery | -0.012 |
Token | sweet |
Token ID | 6029 |
Feature activation | +4.81e-03 |
Pos logit contributions | |
rier | +0.025 |
working | +0.024 |
istered | +0.023 |
goals | +0.021 |
ivery | +0.021 |
Neg logit contributions | |
peek | -0.023 |
dial | -0.022 |
turned | -0.022 |
opened | -0.021 |
nob | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
rier | +0.010 |
istered | +0.010 |
working | +0.010 |
ishy | +0.009 |
ivery | +0.009 |
Neg logit contributions | |
peek | -0.005 |
dial | -0.005 |
nob | -0.004 |
turned | -0.004 |
ad | -0.004 |
Token | They |
Token ID | 1119 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
nob | +0.032 |
dial | +0.031 |
ad | +0.029 |
peek | +0.029 |
god | +0.028 |
Neg logit contributions | |
working | -0.021 |
rier | -0.021 |
goals | -0.020 |
istered | -0.020 |
bored | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -4.13e-03 |
Pos logit contributions | |
peek | +0.001 |
dial | +0.001 |
turned | +0.001 |
nob | +0.001 |
opened | +0.001 |
Neg logit contributions | |
rier | -0.002 |
istered | -0.002 |
working | -0.002 |
ishy | -0.002 |
ivery | -0.002 |
Token | milk |
Token ID | 7545 |
Feature activation | -3.15e-03 |
Pos logit contributions | |
dial | +0.006 |
peek | +0.006 |
turned | +0.005 |
nob | +0.005 |
opened | +0.005 |
Neg logit contributions | |
rier | -0.007 |
istered | -0.007 |
working | -0.007 |
ishy | -0.006 |
ivery | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
dial | +0.003 |
peek | +0.003 |
turned | +0.003 |
nob | +0.002 |
call | +0.002 |
Neg logit contributions | |
istered | -0.007 |
working | -0.007 |
rier | -0.007 |
ishy | -0.006 |
nam | -0.006 |
Token | You |
Token ID | 921 |
Feature activation | +7.74e-03 |
Pos logit contributions | |
rier | +0.049 |
working | +0.048 |
istered | +0.047 |
ishy | +0.043 |
goals | +0.042 |
Neg logit contributions | |
dial | -0.034 |
peek | -0.034 |
nob | -0.032 |
turned | -0.031 |
Looking | -0.030 |
Token | will |
Token ID | 481 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
rier | +0.014 |
istered | +0.014 |
working | +0.014 |
ishy | +0.012 |
attered | +0.012 |
Neg logit contributions | |
peek | -0.010 |
dial | -0.010 |
nob | -0.009 |
turned | -0.009 |
opened | -0.009 |
Token | feel |
Token ID | 1254 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
rier | +0.006 |
working | +0.006 |
istered | +0.006 |
ishy | +0.006 |
goals | +0.006 |
Neg logit contributions | |
dial | -0.004 |
peek | -0.004 |
nob | -0.003 |
turned | -0.003 |
Looking | -0.003 |
Token | safe |
Token ID | 3338 |
Feature activation | -5.28e-03 |
Pos logit contributions | |
working | +0.034 |
istered | +0.032 |
rier | +0.031 |
attered | +0.030 |
ops | +0.029 |
Neg logit contributions | |
peek | -0.020 |
turned | -0.020 |
dial | -0.019 |
called | -0.019 |
pat | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
and | +0.003 |
peek | +0.003 |
nob | +0.002 |
called | +0.002 |
dial | +0.002 |
Neg logit contributions | |
istered | -0.013 |
working | -0.013 |
ishy | -0.012 |
attered | -0.012 |
nam | -0.012 |
Token | warm |
Token ID | 5814 |
Feature activation | -6.07e-03 |
Pos logit contributions | |
rier | +0.048 |
working | +0.047 |
istered | +0.046 |
attered | +0.043 |
kicker | +0.041 |
Neg logit contributions | |
peek | -0.048 |
turned | -0.044 |
dial | -0.043 |
opened | -0.043 |
call | -0.042 |
Token | ." |
Token ID | 526 |
Feature activation | -5.62e-03 |
Pos logit contributions | |
peek | +0.005 |
nob | +0.005 |
dial | +0.005 |
turned | +0.005 |
p | +0.005 |
Neg logit contributions | |
istered | -0.014 |
working | -0.013 |
rier | -0.012 |
kicker | -0.012 |
goals | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -6.31e-03 |
Pos logit contributions | |
dial | +0.004 |
nob | +0.004 |
peek | +0.004 |
ad | +0.003 |
turned | +0.003 |
Neg logit contributions | |
rier | -0.013 |
istered | -0.013 |
working | -0.013 |
goals | -0.012 |
ishy | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e-03 |
Pos logit contributions | |
rier | +0.006 |
working | +0.006 |
istered | +0.006 |
ishy | +0.005 |
goals | +0.005 |
Neg logit contributions | |
dial | -0.004 |
peek | -0.004 |
nob | -0.004 |
turned | -0.003 |
ad | -0.003 |
Token | She |
Token ID | 1375 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
rier | +0.003 |
working | +0.003 |
istered | +0.003 |
ishy | +0.002 |
goals | +0.002 |
Neg logit contributions | |
dial | -0.001 |
peek | -0.001 |
nob | -0.001 |
ad | -0.001 |
turned | -0.001 |
Token | smiled |
Token ID | 13541 |
Feature activation | +5.88e-03 |
Pos logit contributions | |
nob | +0.021 |
dial | +0.019 |
peek | +0.018 |
ad | +0.018 |
Looking | +0.017 |
Neg logit contributions | |
working | -0.041 |
rier | -0.039 |
istered | -0.039 |
kicker | -0.037 |
bored | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -4.78e-03 |
Pos logit contributions | |
istered | +0.015 |
working | +0.015 |
rier | +0.014 |
ishy | +0.014 |
ivery | +0.014 |
Neg logit contributions | |
peek | -0.003 |
dial | -0.003 |
turned | -0.003 |
opened | -0.003 |
Looking | -0.002 |
Token | waved |
Token ID | 26834 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
dial | +0.004 |
peek | +0.004 |
nob | +0.004 |
turned | +0.003 |
ad | +0.003 |
Neg logit contributions | |
rier | -0.010 |
working | -0.010 |
istered | -0.010 |
ishy | -0.009 |
goals | -0.009 |
Token | her |
Token ID | 607 |
Feature activation | -8.93e-03 |
Pos logit contributions | |
working | +0.038 |
istered | +0.038 |
ivery | +0.037 |
rier | +0.036 |
kicker | +0.034 |
Neg logit contributions | |
peek | -0.024 |
dial | -0.023 |
turned | -0.022 |
opened | -0.021 |
pat | -0.021 |
Token | arms |
Token ID | 5101 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
dial | +0.012 |
nob | +0.012 |
peek | +0.012 |
ad | +0.012 |
Looking | +0.011 |
Neg logit contributions | |
rier | -0.015 |
working | -0.014 |
istered | -0.014 |
goals | -0.013 |
ishy | -0.013 |
Token | around |
Token ID | 1088 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
istered | +0.032 |
rier | +0.031 |
working | +0.030 |
kicker | +0.029 |
ishy | +0.029 |
Neg logit contributions | |
peek | -0.015 |
dial | -0.015 |
turned | -0.014 |
opened | -0.013 |
nob | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
rier | +0.022 |
working | +0.022 |
istered | +0.022 |
ishy | +0.020 |
goals | +0.019 |
Neg logit contributions | |
dial | -0.017 |
peek | -0.016 |
nob | -0.016 |
turned | -0.015 |
ad | -0.015 |
Token | classroom |
Token ID | 15806 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
rier | +0.028 |
istered | +0.027 |
working | +0.027 |
ishy | +0.025 |
goals | +0.024 |
Neg logit contributions | |
peek | -0.020 |
dial | -0.019 |
nob | -0.018 |
turned | -0.018 |
ad | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +3.15e-03 |
Pos logit contributions | |
rier | +0.039 |
istered | +0.038 |
working | +0.038 |
ishy | +0.036 |
ivery | +0.034 |
Neg logit contributions | |
peek | -0.020 |
dial | -0.019 |
nob | -0.018 |
turned | -0.017 |
opened | -0.017 |
Token | wanted |
Token ID | 2227 |
Feature activation | +3.16e-03 |
Pos logit contributions | |
rier | +0.030 |
istered | +0.029 |
working | +0.029 |
ishy | +0.026 |
goals | +0.026 |
Neg logit contributions | |
peek | -0.020 |
dial | -0.019 |
turned | -0.018 |
nob | -0.018 |
opened | -0.017 |
Token | some |
Token ID | 617 |
Feature activation | +2.84e-03 |
Pos logit contributions | |
rier | +0.006 |
working | +0.006 |
istered | +0.006 |
bored | +0.005 |
goals | +0.005 |
Neg logit contributions | |
dial | -0.004 |
peek | -0.003 |
nob | -0.003 |
Looking | -0.003 |
ad | -0.003 |
Token | seeds |
Token ID | 11904 |
Feature activation | +6.29e-03 |
Pos logit contributions | |
rier | +0.004 |
working | +0.004 |
istered | +0.004 |
ishy | +0.003 |
goals | +0.003 |
Neg logit contributions | |
dial | -0.005 |
peek | -0.005 |
nob | -0.005 |
Looking | -0.004 |
ad | -0.004 |
Token | in |
Token ID | 287 |
Feature activation | -3.31e-03 |
Pos logit contributions | |
rier | +0.014 |
istered | +0.014 |
working | +0.014 |
ishy | +0.013 |
goals | +0.013 |
Neg logit contributions | |
peek | -0.005 |
nob | -0.004 |
dial | -0.004 |
turned | -0.004 |
opened | -0.004 |
Token | return |
Token ID | 1441 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
dial | +0.004 |
peek | +0.004 |
nob | +0.004 |
Looking | +0.004 |
turned | +0.004 |
Neg logit contributions | |
rier | -0.006 |
working | -0.006 |
istered | -0.006 |
goals | -0.005 |
ishy | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -4.47e-03 |
Pos logit contributions | |
istered | +0.042 |
working | +0.041 |
rier | +0.041 |
ishy | +0.038 |
ivery | +0.038 |
Neg logit contributions | |
peek | -0.017 |
nob | -0.016 |
dial | -0.016 |
turned | -0.015 |
ad | -0.015 |
Token | Twe |
Token ID | 24205 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
dial | +0.007 |
peek | +0.007 |
nob | +0.007 |
ad | +0.007 |
turned | +0.006 |
Neg logit contributions | |
rier | -0.007 |
working | -0.006 |
istered | -0.006 |
ishy | -0.006 |
goals | -0.005 |
Token | ety |
Token ID | 2963 |
Feature activation | +5.71e-03 |
Pos logit contributions | |
working | +0.014 |
rier | +0.014 |
istered | +0.014 |
ishy | +0.013 |
attered | +0.012 |
Neg logit contributions | |
dial | -0.023 |
peek | -0.022 |
nob | -0.021 |
turned | -0.021 |
Looking | -0.020 |
Token | agreed |
Token ID | 4987 |
Feature activation | +3.48e-03 |
Pos logit contributions | |
rier | +0.012 |
istered | +0.011 |
working | +0.011 |
ishy | +0.010 |
goals | +0.010 |
Neg logit contributions | |
dial | -0.006 |
peek | -0.006 |
nob | -0.005 |
turned | -0.005 |
ad | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
rier | +0.007 |
working | +0.007 |
istered | +0.007 |
goals | +0.006 |
bored | +0.006 |
Neg logit contributions | |
dial | -0.003 |
nob | -0.003 |
peek | -0.003 |
ad | -0.003 |
Looking | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -6.31e-03 |
Pos logit contributions | |
rier | +0.023 |
istered | +0.023 |
working | +0.022 |
ishy | +0.021 |
goals | +0.021 |
Neg logit contributions | |
peek | -0.013 |
dial | -0.013 |
turned | -0.012 |
nob | -0.011 |
opened | -0.011 |
Token | down |
Token ID | 866 |
Feature activation | -6.09e-03 |
Pos logit contributions | |
dial | +0.011 |
peek | +0.011 |
nob | +0.011 |
turned | +0.011 |
ad | +0.010 |
Neg logit contributions | |
rier | -0.012 |
working | -0.012 |
istered | -0.012 |
ishy | -0.010 |
goals | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
dial | +0.007 |
nob | +0.007 |
peek | +0.007 |
ad | +0.006 |
Looking | +0.006 |
Neg logit contributions | |
rier | -0.012 |
working | -0.011 |
istered | -0.011 |
goals | -0.010 |
ishy | -0.010 |
Token | When |
Token ID | 1649 |
Feature activation | -6.50e-03 |
Pos logit contributions | |
nob | +0.015 |
dial | +0.015 |
peek | +0.015 |
ad | +0.014 |
p | +0.013 |
Neg logit contributions | |
rier | -0.015 |
working | -0.014 |
istered | -0.014 |
goals | -0.013 |
attered | -0.013 |
Token | he |
Token ID | 339 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
peek | +0.010 |
dial | +0.010 |
turned | +0.010 |
nob | +0.010 |
ad | +0.009 |
Neg logit contributions | |
rier | -0.010 |
istered | -0.009 |
working | -0.009 |
ishy | -0.008 |
goals | -0.008 |
Token | got |
Token ID | 1392 |
Feature activation | -3.11e-03 |
Pos logit contributions | |
nob | +0.018 |
dial | +0.016 |
ad | +0.015 |
p | +0.015 |
Turning | +0.015 |
Neg logit contributions | |
working | -0.021 |
istered | -0.020 |
bored | -0.019 |
sweaty | -0.019 |
ishy | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
dial | +0.004 |
peek | +0.004 |
nob | +0.004 |
ad | +0.003 |
turned | +0.003 |
Neg logit contributions | |
rier | -0.006 |
working | -0.006 |
istered | -0.005 |
ishy | -0.005 |
goals | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -4.13e-03 |
Pos logit contributions | |
nob | +0.023 |
dial | +0.022 |
peek | +0.021 |
Turning | +0.021 |
Looking | +0.021 |
Neg logit contributions | |
working | -0.027 |
rier | -0.026 |
istered | -0.025 |
goals | -0.025 |
sweaty | -0.024 |
Token | beach |
Token ID | 10481 |
Feature activation | -2.81e-03 |
Pos logit contributions | |
peek | +0.006 |
dial | +0.005 |
turned | +0.005 |
call | +0.005 |
opened | +0.005 |
Neg logit contributions | |
istered | -0.008 |
working | -0.007 |
rier | -0.007 |
ishy | -0.007 |
ivery | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
peek | +0.003 |
dial | +0.003 |
turned | +0.003 |
nob | +0.002 |
opened | +0.002 |
Neg logit contributions | |
istered | -0.006 |
rier | -0.006 |
working | -0.006 |
ishy | -0.005 |
ivery | -0.005 |
Token | he |
Token ID | 339 |
Feature activation | -2.37e-03 |
Pos logit contributions | |
peek | +0.016 |
dial | +0.016 |
turned | +0.015 |
opened | +0.014 |
Looking | +0.014 |
Neg logit contributions | |
istered | -0.017 |
rier | -0.017 |
working | -0.016 |
ishy | -0.015 |
ivery | -0.015 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
dial | +0.002 |
peek | +0.002 |
nob | +0.002 |
ad | +0.002 |
turned | +0.002 |
Neg logit contributions | |
rier | -0.005 |
working | -0.005 |
istered | -0.005 |
ishy | -0.004 |
goals | -0.004 |
Token | you |
Token ID | 345 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
peek | +0.015 |
dial | +0.014 |
turned | +0.014 |
opened | +0.013 |
call | +0.013 |
Neg logit contributions | |
istered | -0.023 |
rier | -0.022 |
working | -0.022 |
ishy | -0.021 |
attered | -0.020 |
Token | learn |
Token ID | 2193 |
Feature activation | +5.12e-03 |
Pos logit contributions | |
rier | +0.018 |
istered | +0.017 |
working | +0.017 |
ishy | +0.015 |
nam | +0.015 |
Neg logit contributions | |
peek | -0.015 |
dial | -0.015 |
turned | -0.014 |
nob | -0.014 |
ad | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | -8.68e-04 |
Pos logit contributions | |
rier | +0.010 |
working | +0.009 |
istered | +0.009 |
ishy | +0.009 |
goals | +0.008 |
Neg logit contributions | |
dial | -0.006 |
peek | -0.006 |
nob | -0.006 |
turned | -0.005 |
ad | -0.005 |
Token | be |
Token ID | 307 |
Feature activation | +5.25e-03 |
Pos logit contributions | |
dial | +0.001 |
peek | +0.001 |
nob | +0.001 |
turned | +0.001 |
ad | +0.001 |
Neg logit contributions | |
rier | -0.002 |
working | -0.002 |
istered | -0.002 |
ishy | -0.002 |
goals | -0.002 |
Token | polite |
Token ID | 23507 |
Feature activation | +3.98e-03 |
Pos logit contributions | |
rier | +0.008 |
istered | +0.008 |
working | +0.008 |
ishy | +0.007 |
umm | +0.007 |
Neg logit contributions | |
dial | -0.008 |
peek | -0.008 |
turned | -0.008 |
called | -0.007 |
opened | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
istered | +0.008 |
rier | +0.008 |
working | +0.008 |
ishy | +0.007 |
ivery | +0.007 |
Neg logit contributions | |
dial | -0.004 |
peek | -0.004 |
nob | -0.004 |
turned | -0.004 |
called | -0.003 |
Token | â |
Token ID | 22940 |
Feature activation | +8.41e-03 |
Pos logit contributions | |
nob | +0.024 |
peek | +0.023 |
dial | +0.023 |
ad | +0.023 |
turned | +0.022 |
Neg logit contributions | |
rier | -0.021 |
istered | -0.019 |
working | -0.019 |
goals | -0.018 |
bored | -0.017 |
Token | € |
Token ID | 26391 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
working | +0.012 |
rier | +0.010 |
ivery | +0.010 |
ishy | +0.010 |
istered | +0.009 |
Neg logit contributions | |
dial | -0.015 |
peek | -0.014 |
turned | -0.013 |
pat | -0.013 |
opened | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
peek | +0.016 |
dial | +0.016 |
nob | +0.015 |
turned | +0.014 |
Turning | +0.014 |
Neg logit contributions | |
rier | -0.020 |
working | -0.018 |
istered | -0.018 |
ishy | -0.017 |
nam | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
dial | +0.028 |
bul | +0.025 |
nob | +0.024 |
peek | +0.023 |
extingu | +0.023 |
Neg logit contributions | |
working | -0.026 |
istered | -0.025 |
goals | -0.025 |
dd | -0.024 |
ivery | -0.023 |
Token | Mo |
Token ID | 16632 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
dial | +0.056 |
nob | +0.055 |
peek | +0.054 |
opens | +0.053 |
pat | +0.052 |
Neg logit contributions | |
dd | -0.041 |
working | -0.040 |
istered | -0.038 |
ivery | -0.037 |
goals | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
working | +0.030 |
istered | +0.029 |
rier | +0.028 |
ivery | +0.027 |
ishy | +0.027 |
Neg logit contributions | |
peek | -0.016 |
turned | -0.014 |
dial | -0.014 |
pat | -0.014 |
opened | -0.013 |
Token | |
Token ID | 220 |
Feature activation | +5.74e-03 |
Pos logit contributions | |
dial | +0.025 |
nob | +0.024 |
peek | +0.023 |
ad | +0.021 |
opened | +0.020 |
Neg logit contributions | |
rier | -0.024 |
working | -0.024 |
istered | -0.023 |
goals | -0.023 |
ishy | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
rier | +0.013 |
istered | +0.013 |
working | +0.013 |
goals | +0.012 |
ishy | +0.012 |
Neg logit contributions | |
dial | -0.004 |
peek | -0.004 |
nob | -0.004 |
ad | -0.003 |
turned | -0.003 |
Token | They |
Token ID | 2990 |
Feature activation | +4.39e-03 |
Pos logit contributions | |
dial | +0.022 |
nob | +0.020 |
peek | +0.020 |
call | +0.018 |
turned | +0.018 |
Neg logit contributions | |
working | -0.028 |
rier | -0.028 |
istered | -0.028 |
goals | -0.026 |
ivery | -0.026 |
Token | decided |
Token ID | 3066 |
Feature activation | -4.98e-03 |
Pos logit contributions | |
rier | +0.010 |
istered | +0.009 |
working | +0.009 |
ishy | +0.008 |
ivery | +0.008 |
Neg logit contributions | |
peek | -0.005 |
turned | -0.005 |
dial | -0.005 |
opened | -0.005 |
ad | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
dial | +0.004 |
peek | +0.004 |
nob | +0.004 |
turned | +0.004 |
ad | +0.004 |
Neg logit contributions | |
rier | -0.011 |
working | -0.010 |
istered | -0.010 |
ishy | -0.010 |
goals | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
dial | +0.007 |
peek | +0.007 |
nob | +0.007 |
turned | +0.006 |
ad | +0.006 |
Neg logit contributions | |
rier | -0.024 |
working | -0.024 |
istered | -0.024 |
ishy | -0.022 |
goals | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | +8.74e-03 |
Pos logit contributions | |
dial | +0.016 |
peek | +0.016 |
nob | +0.015 |
turned | +0.014 |
ad | +0.014 |
Neg logit contributions | |
rier | -0.029 |
istered | -0.028 |
working | -0.028 |
ishy | -0.026 |
goals | -0.025 |
Token | good |
Token ID | 922 |
Feature activation | -2.99e-03 |
Pos logit contributions | |
rier | +0.012 |
working | +0.012 |
istered | +0.011 |
ishy | +0.010 |
goals | +0.010 |
Neg logit contributions | |
dial | -0.015 |
peek | -0.015 |
nob | -0.014 |
ad | -0.014 |
turned | -0.014 |
Token | idea |
Token ID | 2126 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
dial | +0.004 |
peek | +0.004 |
nob | +0.004 |
ad | +0.004 |
turned | +0.004 |
Neg logit contributions | |
working | -0.005 |
rier | -0.005 |
istered | -0.005 |
ishy | -0.004 |
goals | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +8.95e-03 |
Pos logit contributions | |
dial | +0.047 |
peek | +0.043 |
nob | +0.043 |
Looking | +0.042 |
Turning | +0.040 |
Neg logit contributions | |
working | -0.054 |
rier | -0.053 |
istered | -0.051 |
goals | -0.046 |
anymore | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
dial | +0.011 |
peek | +0.011 |
nob | +0.010 |
turned | +0.010 |
ad | +0.010 |
Neg logit contributions | |
rier | -0.016 |
working | -0.016 |
istered | -0.016 |
ishy | -0.015 |
goals | -0.014 |
Token | what |
Token ID | 644 |
Feature activation | +4.88e-03 |
Pos logit contributions | |
dial | +0.034 |
peek | +0.034 |
nob | +0.032 |
turned | +0.031 |
ad | +0.031 |
Neg logit contributions | |
rier | -0.038 |
working | -0.037 |
istered | -0.037 |
ishy | -0.033 |
goals | -0.032 |
Token | shall |
Token ID | 2236 |
Feature activation | -9.18e-03 |
Pos logit contributions | |
rier | +0.010 |
istered | +0.009 |
working | +0.009 |
ishy | +0.009 |
ivery | +0.009 |
Neg logit contributions | |
peek | -0.006 |
dial | -0.005 |
turned | -0.005 |
nob | -0.005 |
opened | -0.005 |
Token | you |
Token ID | 345 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
peek | +0.015 |
dial | +0.015 |
nob | +0.014 |
turned | +0.014 |
ad | +0.014 |
Neg logit contributions | |
rier | -0.013 |
istered | -0.013 |
working | -0.013 |
ishy | -0.011 |
goals | -0.011 |
Token | do |
Token ID | 466 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
dial | +0.057 |
nob | +0.055 |
peek | +0.053 |
Looking | +0.051 |
turned | +0.050 |
Neg logit contributions | |
working | -0.070 |
rier | -0.068 |
istered | -0.064 |
ishy | -0.060 |
ivery | -0.059 |
Token | now |
Token ID | 783 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
dial | +0.016 |
peek | +0.016 |
nob | +0.015 |
turned | +0.014 |
Looking | +0.014 |
Neg logit contributions | |
rier | -0.030 |
working | -0.030 |
istered | -0.029 |
ishy | -0.027 |
goals | -0.026 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.47e-03 |
Pos logit contributions | |
dial | +0.047 |
peek | +0.046 |
nob | +0.045 |
Looking | +0.041 |
turned | +0.041 |
Neg logit contributions | |
rier | -0.087 |
working | -0.087 |
istered | -0.082 |
ishy | -0.077 |
goals | -0.076 |
Token | one |
Token ID | 530 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
dial | +0.002 |
peek | +0.002 |
nob | +0.002 |
ad | +0.002 |
turned | +0.002 |
Neg logit contributions | |
rier | -0.003 |
working | -0.003 |
istered | -0.003 |
ishy | -0.002 |
goals | -0.002 |
Token | of |
Token ID | 286 |
Feature activation | -7.54e-03 |
Pos logit contributions | |
rier | +0.019 |
working | +0.018 |
istered | +0.017 |
ishy | +0.016 |
kicker | +0.016 |
Neg logit contributions | |
nob | -0.015 |
dial | -0.015 |
peek | -0.013 |
Looking | -0.013 |
ad | -0.013 |
Token | them |
Token ID | 606 |
Feature activation | +3.92e-03 |
Pos logit contributions | |
dial | +0.010 |
peek | +0.009 |
nob | +0.009 |
turned | +0.009 |
ad | +0.008 |
Neg logit contributions | |
rier | -0.013 |
working | -0.013 |
istered | -0.013 |
ishy | -0.012 |
goals | -0.012 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
rier | +0.008 |
working | +0.008 |
istered | +0.008 |
ishy | +0.008 |
goals | +0.008 |
Neg logit contributions | |
dial | -0.003 |
peek | -0.003 |
nob | -0.003 |
ad | -0.003 |
turned | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -7.15e-03 |
Pos logit contributions | |
rier | +0.027 |
working | +0.027 |
istered | +0.027 |
ishy | +0.025 |
goals | +0.025 |
Neg logit contributions | |
dial | -0.007 |
peek | -0.006 |
nob | -0.006 |
turned | -0.005 |
ad | -0.005 |
Token | " |
Token ID | 1 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
dial | +0.008 |
peek | +0.007 |
turned | +0.007 |
nob | +0.007 |
call | +0.006 |
Neg logit contributions | |
rier | -0.013 |
istered | -0.013 |
working | -0.013 |
ivery | -0.012 |
goals | -0.012 |
Token | Let |
Token ID | 5756 |
Feature activation | -3.33e-03 |
Pos logit contributions | |
working | +0.014 |
rier | +0.014 |
istered | +0.013 |
kicker | +0.012 |
nam | +0.012 |
Neg logit contributions | |
dial | -0.014 |
nob | -0.014 |
peek | -0.014 |
call | -0.013 |
turned | -0.013 |
Token | 's |
Token ID | 338 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
peek | +0.004 |
dial | +0.004 |
nob | +0.004 |
turned | +0.004 |
ad | +0.003 |
Neg logit contributions | |
rier | -0.006 |
istered | -0.006 |
working | -0.006 |
ishy | -0.006 |
goals | -0.006 |
Token | take |
Token ID | 1011 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
dial | +0.020 |
nob | +0.020 |
peek | +0.020 |
Looking | +0.018 |
turned | +0.018 |
Neg logit contributions | |
rier | -0.034 |
working | -0.034 |
istered | -0.033 |
ishy | -0.031 |
goals | -0.030 |
Token | this |
Token ID | 428 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
nob | +0.021 |
dial | +0.021 |
peek | +0.021 |
Looking | +0.019 |
turned | +0.019 |
Neg logit contributions | |
working | -0.029 |
rier | -0.029 |
istered | -0.027 |
ishy | -0.026 |
goals | -0.026 |
Token | ash |
Token ID | 12530 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
dial | +0.074 |
Looking | +0.074 |
ad | +0.073 |
Agent | +0.071 |
Turn | +0.068 |
Neg logit contributions | |
rier | -0.065 |
working | -0.065 |
ishy | -0.058 |
istered | -0.056 |
bored | -0.054 |
Token | and |
Token ID | 290 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
dial | +0.025 |
peek | +0.024 |
Looking | +0.023 |
ad | +0.023 |
nob | +0.022 |
Neg logit contributions | |
rier | -0.046 |
working | -0.045 |
istered | -0.043 |
ishy | -0.040 |
goals | -0.039 |
Token | build |
Token ID | 1382 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
dial | +0.016 |
peek | +0.015 |
nob | +0.015 |
turned | +0.013 |
ad | +0.013 |
Neg logit contributions | |
rier | -0.034 |
working | -0.033 |
istered | -0.033 |
ishy | -0.031 |
goals | -0.030 |
Token | something |
Token ID | 1223 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
dial | +0.015 |
peek | +0.015 |
nob | +0.014 |
turned | +0.013 |
ad | +0.013 |
Neg logit contributions | |
rier | -0.023 |
working | -0.022 |
istered | -0.022 |
ishy | -0.020 |
goals | -0.020 |
Token | with |
Token ID | 351 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
working | +0.030 |
rier | +0.029 |
istered | +0.029 |
ishy | +0.026 |
ivery | +0.026 |
Neg logit contributions | |
peek | -0.026 |
dial | -0.026 |
nob | -0.024 |
turned | -0.024 |
opened | -0.023 |
Token | set |
Token ID | 900 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
rier | +0.007 |
working | +0.007 |
istered | +0.006 |
goals | +0.006 |
ishy | +0.006 |
Neg logit contributions | |
dial | -0.004 |
peek | -0.004 |
Looking | -0.004 |
nob | -0.004 |
Turning | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
dial | +0.020 |
peek | +0.020 |
nob | +0.019 |
Looking | +0.019 |
Turning | +0.018 |
Neg logit contributions | |
rier | -0.025 |
working | -0.023 |
istered | -0.022 |
goals | -0.020 |
ishy | -0.020 |
Token | with |
Token ID | 351 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
dial | +0.027 |
nob | +0.027 |
peek | +0.027 |
ad | +0.025 |
Looking | +0.025 |
Neg logit contributions | |
rier | -0.048 |
working | -0.047 |
istered | -0.046 |
goals | -0.042 |
ishy | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
dial | +0.004 |
peek | +0.004 |
turned | +0.004 |
ad | +0.004 |
opened | +0.004 |
Neg logit contributions | |
rier | -0.009 |
istered | -0.009 |
working | -0.009 |
ivery | -0.009 |
ishy | -0.008 |
Token | train |
Token ID | 4512 |
Feature activation | -6.82e-03 |
Pos logit contributions | |
nob | +0.031 |
dial | +0.030 |
peek | +0.029 |
ad | +0.029 |
Looking | +0.028 |
Neg logit contributions | |
rier | -0.036 |
istered | -0.035 |
working | -0.035 |
ishy | -0.031 |
goals | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
dial | +0.007 |
peek | +0.007 |
nob | +0.007 |
Looking | +0.007 |
turned | +0.006 |
Neg logit contributions | |
rier | -0.013 |
working | -0.013 |
istered | -0.013 |
goals | -0.012 |
ishy | -0.011 |
Token | tracks |
Token ID | 8339 |
Feature activation | -2.97e-03 |
Pos logit contributions | |
nob | +0.049 |
peek | +0.047 |
dial | +0.046 |
Looking | +0.046 |
ad | +0.044 |
Neg logit contributions | |
rier | -0.083 |
working | -0.082 |
istered | -0.078 |
goals | -0.075 |
kicker | -0.073 |
Token | , |
Token ID | 11 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
dial | +0.003 |
peek | +0.003 |
nob | +0.003 |
turned | +0.003 |
ad | +0.003 |
Neg logit contributions | |
rier | -0.006 |
istered | -0.006 |
working | -0.006 |
ishy | -0.006 |
goals | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
nob | +0.038 |
Looking | +0.037 |
peek | +0.036 |
dial | +0.035 |
ad | +0.034 |
Neg logit contributions | |
rier | -0.083 |
working | -0.082 |
istered | -0.077 |
goals | -0.075 |
bored | -0.073 |
Token | stations |
Token ID | 8985 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
dial | +0.034 |
peek | +0.034 |
nob | +0.032 |
turned | +0.031 |
ad | +0.030 |
Neg logit contributions | |
rier | -0.046 |
working | -0.045 |
istered | -0.045 |
ishy | -0.041 |
goals | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
istered | +0.027 |
rier | +0.027 |
working | +0.026 |
ishy | +0.025 |
ivery | +0.024 |
Neg logit contributions | |
dial | -0.013 |
peek | -0.013 |
nob | -0.012 |
turned | -0.012 |
ad | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +3.38e-03 |
Pos logit contributions | |
rier | +0.016 |
working | +0.016 |
istered | +0.016 |
ishy | +0.015 |
goals | +0.014 |
Neg logit contributions | |
dial | -0.007 |
peek | -0.007 |
nob | -0.006 |
turned | -0.006 |
ad | -0.006 |
Token | He |
Token ID | 679 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
rier | +0.006 |
working | +0.005 |
istered | +0.005 |
goals | +0.005 |
ishy | +0.005 |
Neg logit contributions | |
dial | -0.005 |
nob | -0.004 |
peek | -0.004 |
ad | -0.004 |
turned | -0.004 |
Token | gave |
Token ID | 2921 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
nob | +0.021 |
dial | +0.020 |
peek | +0.018 |
ad | +0.018 |
Looking | +0.018 |
Neg logit contributions | |
working | -0.034 |
rier | -0.033 |
istered | -0.033 |
goals | -0.031 |
ishy | -0.031 |
Token | Bob |
Token ID | 5811 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
rier | +0.035 |
working | +0.034 |
istered | +0.034 |
ishy | +0.031 |
goals | +0.030 |
Neg logit contributions | |
dial | -0.031 |
peek | -0.031 |
nob | -0.030 |
turned | -0.028 |
ad | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
dial | +0.045 |
peek | +0.044 |
nob | +0.042 |
turned | +0.039 |
ad | +0.039 |
Neg logit contributions | |
rier | -0.086 |
working | -0.085 |
istered | -0.084 |
ishy | -0.077 |
goals | -0.076 |
Token | big |
Token ID | 1263 |
Feature activation | -5.15e-02 |
Pos logit contributions | |
nob | +0.071 |
ad | +0.067 |
dial | +0.065 |
ran | +0.062 |
peek | +0.060 |
Neg logit contributions | |
working | -0.053 |
istered | -0.051 |
rier | -0.050 |
ishy | -0.048 |
kicker | -0.046 |
Token | hug |
Token ID | 16225 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
nob | +0.060 |
dial | +0.056 |
peek | +0.050 |
ad | +0.048 |
Looking | +0.048 |
Neg logit contributions | |
rier | -0.095 |
working | -0.094 |
istered | -0.090 |
ishy | -0.090 |
kicker | -0.088 |
Token | and |
Token ID | 290 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
dial | +0.014 |
peek | +0.014 |
nob | +0.013 |
turned | +0.011 |
ad | +0.011 |
Neg logit contributions | |
rier | -0.044 |
working | -0.043 |
istered | -0.043 |
ishy | -0.040 |
goals | -0.039 |
Token | said |
Token ID | 531 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
dial | +0.012 |
nob | +0.012 |
peek | +0.012 |
ad | +0.011 |
turned | +0.010 |
Neg logit contributions | |
rier | -0.025 |
working | -0.024 |
istered | -0.024 |
ishy | -0.022 |
goals | -0.022 |
Token | " |
Token ID | 366 |
Feature activation | +1.28e-03 |
Pos logit contributions | |
istered | +0.036 |
working | +0.035 |
rier | +0.034 |
ishy | +0.033 |
ivery | +0.032 |
Neg logit contributions | |
peek | -0.018 |
dial | -0.017 |
turned | -0.016 |
nob | -0.016 |
ad | -0.016 |
Token | Well |
Token ID | 5779 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
working | +0.002 |
rier | +0.002 |
istered | +0.002 |
ishy | +0.002 |
kicker | +0.002 |
Neg logit contributions | |
dial | -0.002 |
peek | -0.002 |
nob | -0.002 |
turned | -0.002 |
Looking | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
nob | +0.044 |
dial | +0.044 |
izer | +0.041 |
x | +0.041 |
p | +0.039 |
Neg logit contributions | |
goals | -0.050 |
rier | -0.046 |
whenever | -0.046 |
forward | -0.043 |
bored | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
dial | +0.022 |
nob | +0.021 |
peek | +0.018 |
extingu | +0.018 |
bul | +0.018 |
Neg logit contributions | |
goals | -0.027 |
working | -0.024 |
istered | -0.024 |
rier | -0.024 |
both | -0.023 |
Token | Her |
Token ID | 9360 |
Feature activation | -8.28e-03 |
Pos logit contributions | |
dial | +0.016 |
nob | +0.016 |
peek | +0.015 |
pat | +0.015 |
call | +0.014 |
Neg logit contributions | |
goals | -0.024 |
rier | -0.023 |
working | -0.023 |
istered | -0.023 |
kicker | -0.021 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
nob | +0.017 |
ad | +0.016 |
p | +0.016 |
peek | +0.016 |
Looking | +0.016 |
Neg logit contributions | |
istered | -0.008 |
rier | -0.007 |
working | -0.007 |
goals | -0.007 |
ishy | -0.006 |
Token | showed |
Token ID | 3751 |
Feature activation | -2.60e-04 |
Pos logit contributions | |
nob | +0.012 |
AAAA | +0.012 |
eway | +0.012 |
Looking | +0.012 |
izer | +0.011 |
Neg logit contributions | |
working | -0.018 |
kicker | -0.016 |
rier | -0.015 |
ishy | -0.015 |
istered | -0.015 |
Token | Julie |
Token ID | 21946 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
dial | +0.000 |
peek | +0.000 |
nob | +0.000 |
Looking | +0.000 |
turned | +0.000 |
Neg logit contributions | |
rier | -0.000 |
working | -0.000 |
istered | -0.000 |
goals | -0.000 |
ishy | -0.000 |
Token | how |
Token ID | 703 |
Feature activation | +5.83e-03 |
Pos logit contributions | |
dial | +0.060 |
peek | +0.059 |
nob | +0.058 |
ad | +0.053 |
turned | +0.053 |
Neg logit contributions | |
rier | -0.100 |
working | -0.098 |
istered | -0.096 |
goals | -0.090 |
ishy | -0.088 |
Token | to |
Token ID | 284 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
rier | +0.015 |
istered | +0.014 |
working | +0.014 |
ivery | +0.013 |
ishy | +0.013 |
Neg logit contributions | |
peek | -0.004 |
dial | -0.003 |
turned | -0.003 |
nob | -0.003 |
opened | -0.003 |
Token | use |
Token ID | 779 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
Looking | +0.019 |
nob | +0.019 |
fore | +0.017 |
dial | +0.017 |
Turning | +0.017 |
Neg logit contributions | |
working | -0.029 |
rier | -0.026 |
istered | -0.026 |
ishy | -0.025 |
ivery | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
peek | +0.037 |
dial | +0.036 |
nob | +0.036 |
Looking | +0.034 |
ad | +0.033 |
Neg logit contributions | |
rier | -0.049 |
working | -0.048 |
istered | -0.047 |
goals | -0.043 |
ishy | -0.043 |
Token | comb |
Token ID | 1974 |
Feature activation | -1.12e-03 |
Pos logit contributions | |
Looking | +0.034 |
nob | +0.034 |
ad | +0.034 |
peek | +0.033 |
dial | +0.032 |
Neg logit contributions | |
rier | -0.040 |
working | -0.038 |
istered | -0.035 |
goals | -0.034 |
ishy | -0.033 |